close

Вход

Забыли?

вход по аккаунту

?

ELEMENTAL ANALYSIS BY MASS SPECTROMETRY USING AN ATMOSPHERIC PRESSURE MICROWAVE INDUCED PLASMA AS AN ION SOURCE

код для вставкиСкачать
INFORMATION TO USERS
While the most advanced technology has been used to
photograph and reproduce this manuscript, the quality of
the reproduction is heavily dependent upon the quality of
the material submitted. For example:
•
Manuscript pages may have indistinct print. In such
cases, the best available copy has been filmed.
•
M anuscripts may not always be com plete. In such
cases, a note w ill indicate that it is not possible to
obtain m issing pages.
•
Copyrighted material may have been removed from
the manuscript. In such cases, a note w ill indicate the
deletion.
Oversize materials (e.g., maps, drawings, and charts) are
photographed by sectioning the original, beginning at the
upper left-hand corner and continuing from left to right in
equal sections with small overlaps. Each oversize page is
also film ed as one exposure and is a v a ila b le, for an
additional charge, as a standard 35mm slide or as a 17”x 23”
black and white photographic print.
Most photographs reproduce acceptably on p o sitiv e
microfilm or microfiche but lack the clarity on xerographic
copies made from the microfilm. For an additional charge,
35mm slides of 6”x 9” black and white photographic prints
are available for any photographs or illustrations that
cannot be reproduced satisfactorily by xerography.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8708151
Satzger, R. Duane
ELEMENTAL ANALYSIS BY MASS SPECTROMETRY USING AN
ATMOSPHERIC PRESSURE MICROWAVE INDUCED PLASMA AS AN ION
SOURCE
Ph.D.
University of Cincinnati
University
Microfilms
International
W:
.
.
1986
300 N. Zeeb Road, Ann Arbcr, Ml 48105
.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy.
Problems encountered with this document have been identified herewith a check mark V .
1.
Glossy photographs cr pages____
2.
Colored illustrations, paper or print_____
3.
Photographs with dark background____
4.
Illustrations are poor copy______
5.
Pages with black marks, not original copy_____
6. Print shows through as there is text on
both sides of page____
7. Indistinct, broken or small print on several pages
^
8. Print exceeds margin requirements_____
9.
Tightly bound copy with print lost in spine______
10.
Computer printout pages with indistinct print_____
11.
Page(s)__________ lacking when material received, and not available from school or
author.
12.
Rage(s)
13.
Two pages numbered
seem to be missing in numbering only as text follows.
. Text follows.
14. Curling and wrinkled pages______
15. Dissertation contains pages with print at a slant, filmed as received________
16.
Other
_________________ ________________________________________
University
Microfilms
I
ii
hoi
a +S A
/A !
i icuiui ia i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ELEMENTAL ANALYSIS BY MASS SPECTROMETRY
USING AN ATMOSPHERIC PRESSURE MICROWAVE
INDUCED PLASMA AS AN ION SOURCE
A d is s e rta tio n submitted to the
D iv is io n o f Graduate Studies and Research
o f the U n ive rsity o f C incinnati
in p a rtia l f u lf illm e n t o f the
requirements fo r the degree o f
DOCTOR OF PHILOSOPHY
in the Department o f Chemistry
o f the College o f Arts and Sciences
1986
by
R. Duane Satzger
B .S ., U n iv e rs ity o f C in c in n a ti, 1977
M.S., U n ive rsity o f C in c in n a ti, 1979
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF CINCINNATI
December 4 , - 1 9
86
I hereby recommend that the thesis prepared under my
Supervision by
R. Duane Sat.zger____________________
ELEMENTAL ANALYSIS BY MASS SPECTROMETRY
entitled____________ ____________________________
_____________ USING AN ATMOSPHERIC PRESSURE MICROWAVE
____________
INDUCEDPLASMAASAN ION SOURCE________
be accepted asfidfiSBng this part o f the requirementsfo r the
degree o f
DOCTOR OF PHILOSOPHY_______________________
Approved by:
A :
£■
\
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
An instrument was designed and b u ilt which is capable o f performing
ra p id ,
sequential
m ulti-elem ent analyses.
The instrum ent u tiliz e s
a
microwave induced plasma as a source o f s in g ly charged monatomic p o s itiv e
io n s .
Ions generated in the plasma are sampled by an in te rfa c e which
produces an atom ic beam which is tra n s m itte d and focused by using
e le c tro s ta tic
s p e c tro m e te r.
ion
lenses
and
f ilt e r e d
using
a
quadrupole
mass
D e te c tio n is accom plished in e it h e r analog o r pulse
counting modes depending on the current le v e l o f the response which is
re la te d to the concentration.
The
microwave c a v ity
enables atmospheric
s u s ta in e d in e it h e r argon o r h e liu m .
in tro d u c tio n o f s o lu tio n
n e bu liza tio n
techniques.
pressure
plasmas to
be
An Ar plasma enables d ir e c t
in to the mass spectrometer using conventional
Detection
lim its
fo r most elements
using the argon plasma are in the ng/mL range.
studied
Background interferences
such as molecular oxides and c r it ic a l operating parameters are discussed.
A He plasma was inve stig a te d as an ion source f o r elements o f higher
io n iz a tio n p o te n tia l.
B r, Cl and I were introduced as halomethanes in
He.
th e in te r fa c e are discussed which reduce th e
Improvements in
molecular
background
by
reduction
of
entrained
atmospheric
gases.
D etection lim its obtained fo r the halogens using the He plasma are in the
pg/s range, in d ic a tin g a high p o te n tia l as an element s e le c tiv e detector
fo r gas chromatography.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
The author would lik e to o ffe r a sincere thank you to the fo llo w in g
in d iv id u a ls :
to
h is w ife
and now th re e c h ild re n fo r th e ir constant
encouragement over the long haul; to the Food and Drug A dm inistration fo r
fu n d in g much o f t h is
re se arch
e ffo rt
and to
Fred F ric k e
f o r h is
co n fide n ce to th e a u th o r and th e p r o je c t; to th e members o f FDA's
Elemental A n a ly s is Research C enter whose fr ie n d s h ip made t h is work
proceed more smoothly; to Joe Caruso fo r h is patience during the design
and construction phases o f th is research p ro je c t; to the many generations
o f group members who were responsible fo r moral support; to Dr. Climaco
Metral and Professor Sam Houk fo r p ro vid ing te ch n ica l assistance; to Gary
Pauley, B ill Brauntz and Jim Lindahl fo r t h e ir tim e ly maching and design
assistance; and f in a lly to Marianne A lle n fo r her assistance in wrapping
up th is manuscript.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
TABLE OF CONTENTS
Page
CHAPTER 1
INTRODUCTION
CHAPTER 2
SYSTEM DESCRIPTION
Sample In tro d u c tio n
r*
12
Molecular Beam Sampling In te rfa c e
18
Vacuum System - Design Considerations
29
Ion Optics
36
Mass F ilt e r
42
Ion D etector/Signal Processing
44
PLASMA/RESULTS AND DISCUSSION
45
Ar
CHAPTER 4
HE PLASMA
.
9
Microwave Plasma - Ion Source
CHAPTER 3
In tro d u c tio n
73
Source M o d ifica tio n s
74
Results and Discussion
76
SUMMARY AND FUTURE WORK
CHAPTER 5
1
..
!r .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
vi
LIST OF TABLES
Table
Number
Page
1
System Components and S p e c ific a tio n
13
2
Vacuum System
31
3
Compromise Lens Voltages
39
4
Background Ions fo r Water and 5% HNOg
48
5
A d d itio n a l Background Ions Observed A risin g
51
from Inorganic Acids
Normalized Response
64
7
Detection L im its a t Two Power Levels
65
8
Detection L im its a t Low Power w ith 0.10 cmSkimmer
67
9
Background Ions Present in Unshielded HePlasma
79
10
Background Ions in Shielded H e /^ Plasma
82
11
Shielded He Plasma w ith 8 ppm CH^Br, CH^Cl
88
12
Detection L im its and Background Count Rates Obtained
94
6
in S ingle Ion Mode
T
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Page
Figure
Number
1
MIP-MS System Schematic
11
2
Designs o f M odified Microwave C avitie s Used
16
as Ion Sources
3
In te rfa ce Designs:
3A Sampler Mount/Expansion Stage
20
3B Skimmer Mount/Face o f Second Vacuum Stage
22
3C Sampler, Skimmer Cones
24
4
Molecular Beam Sampling In te rfa c e fo r MIP-MS
26
5
Ion Lens Stack
37
6
Tuning Curves o f Ton Lens Elements
41
7
Major and Minor Background Ions in D is t ille d
47
H o t
7 q H
U C l V I I I ^ C U
U
0
« , 2 ' W*
8
Radial View o f Ar Plasma
52
9
Spectra o f 50 ppm Y in 0.2% HC1
54
10
Scans o f 1% HNO^ and 500 ppm Ba in 1% HNOg
56
I llu s t r a t in g Presence o f BaO and BaOH
11
57
Scans o f 1% HNO^ and 10 ppm Co, Sc, Ti
O
I llu s t r a t in g Predominance o f Molecular Oxides
12
Spectra Demonstrate Iso b a ric In te rfe re n c e ,
Mo0+ on
59
13
E ffe ct o f Sampling Distance on ^ C u + Count
Rate
61
CO
14
E ffe c t o f Plasma Gas Flow Rate on
+
Cu
Rate
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
Page
Figure
Number
15
C a lib ra tio n Curves fo r 52 Cr+ , 6 3 Cu+ , 6 3 Zn+, 2 08 Pb+
68
at 225 W
16
E ffe c t o f Na on Io n iz a tio n o f 10 mg/mL Cu
69
17
Major and Minor Background Ions in
71
D is t ille d , Deionized Water
18
He Plasma - Sampling In te rfa c e
75
19
P rin cip a l Ions Sampled from 270 WHe Microwave
78
Induced Plasma
20
R elative In te n s itie s o f Major and Minor Background
84
Ions in the Shielded He Plasma
21
Comparison o f Spectra Obtained fo r
7Q
Br
x
01
and
X
Br
85
During Intense and D iffu se E le c tric a l In te ra c tio n
Between the He Plasma and the Sampler
22
Scan o f 8 ppm CH^Br, CH^Cl in He
90
23
E ffe c t o f He Flow on Ion Count Rate
92
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
1
CHAPTER 1
INTRODUCTION
"The ro le o f the a n a ly tic a l chemist is to develop and apply methods
to determine the nature and q u a n titie s o f chemical substances, whether
they are present in a specimen as s in g le , pure components or in the form
o f m ixtu re s ." ( 1 )
The above d e fin itio n
a n a ly tic a l ch e m is ts .
describes the
challenge which faces
today's
They must be able to a n tic ip a te th e problems
associated w ith a p p lic a tio n o f a p a rtic u la r method o f a n a lysis, which may
re s u lt in the m o d ifica tio n or development o f new a n a ly tic a l methodology.
Some have re fe rre d to th is
(2 ).
as the
"two sides o f a n a ly tic a l
chemistry"
However, few a n a ly tic a l chemists operate in both fundamental and
applied categories due to the degree o f s p e c ia liz a tio n required in order
to
be p ro fic ie n t
in
a p a rtic u la r area o f a n a ly tic a l
techniques a v a ila b le to the e a rly
'g e n e ra lis t'
s p e c ific
a n a ly tic a l
analysis were developed.
required
info rm a tio n ,
more
The
a n a ly tic a l chemist were
based on physical p ro pe rtie s or wet chemical methods.
more
chem istry.
With the need fo r
sophisticated
methods
of
Many o f these were instrumental methods and
a greater degree o f s p e c ia liz a tio n
by the modern a n a ly tic a l
chemist.
Development o f instrum entation and methodology fo r tra ce elemental
analysis is a small subcategory o f a n a ly tic a l chem istry.
Trace elements
are important from both n u tritio n a l and to x ic o lo g ic a l standpoints.
n a tu ra lly
occurring
le v e ls
of
trace
elements
c o n tin u a lly disturbed by today's lif e s t y le .
in
The
our environment are
The products we manufacture
and consume and th e fu e ls we use a ll a id in s h if t in g these n a tu ra l
le v e ls .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
Many instrumental techniques capable o f performing elemental analysis
at tra ce leve ls are a v a ila b le to the a n a ly s t.
lim ita tio n s
However, a ll have ce rta in
such as the number o f elements to which the technique is
a p p lic a b le ,
th e
typ e
of
sample m a trix
which can be handled by a
p a rtic u la r technique, the detection lim its re q uire d, the amount o f sample
necessary to perform the analysis and the time required to perform the
a n a ly s is , which u ltim a te ly a ffe c ts the cost per analysis and equipment
co st.
A ll o f the techniques, electrochem ical, spectroscopic and nuclear
have found t h e i r n ic h e .
No s in g le in s tru m e n ta l te c h n iq u e has been
capable of providing a ll o f the answers.
Since the f i r s t
reports dealing w ith the sampling o f ions d ir e c tly
from an atmospheric pressure plasma in to a quadruple mass spectrometer,
in te re s t in th is p o te n tia lly powerful new a n a ly tic a l technique has been
high
(3-14).
Never before had ions been extracted from a plasma fo r
a n a ly tic a l purposes.
a
means
of
An atmospheric pressure plasma ion source provided
performing
rapid
elemental
and
is o to p ic
analysis
n o n -v o la tile species in aqueous s o lu tio n by mass spectrom etry.
of
Samples
could be introduced in to the plasma source w ithout having to p h y s ic a lly
get the sample in to the vacuum chamber, as is the case w ith conventional
ion
sources
(15-20).
Equally
a ttr a c tiv e
to
the
ease
of
sample
in tro d u c tio n were reports o f sub p a r t- p e r - b illio n d etection lim its which
are ty p ic a l
o f mass spectrom etric techniques.
those whose a p plica tio n s
s p e c ific a lly
were tr a d itio n a lly
the dc c a p illa ry
arc
in
The plasmas used were
o p tic a l
plasma, the r f
emission work,
in d u c tiv e ly
coupled
plasma and the microwave induced plasma.
The
f e a s ib ilit y
of
e x tra c tin g
ions
from
an atmospheric
pressure
plasma in to which a sample so lu tio n was nebulized, was f i r s t demonstrated
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
w ith the dc c a p illa ry arc plasma (CAP)(3-5).
s in g ly charged, monatomic, p o s itiv e io n s.
Elements were detected as
The CAP sustained in argon
demonstrated good detection lim its w ith standard s o lu tio n s , few isob a ric
inte rfe ren ce s
generated
by
background
peaks which
arose
e ith e r
from
ion-molecule reactions in the afterglow or in the sampling process, and
few double charged species.
to
it s
However, the CAP had major drawbacks.
low gas temperature the to ta l
percentage of low io n iz a tio n
Due
ion population contained a large
p o te n tia l
molecular oxides.
The analyte
response depended on the io n iz a tio n p o te n tia l o f the element w ith respect
to the io n iz a tio n p o te n tia l o f the molecular oxide.
obtained fo r those elements whose f i r s t
approximately
8 eV.
However,
as
Good response was
io n iz a tio n p o te n tia l was below
the
analyte
io n iz a tio n
p o te n tia l
approached th a t o f NO (ca 9.25 eV) which was the base peak in the CAP
spectrum, the response qu ickly degraded.
entrained in the plasma.
the
The N0+ o rig in a te d
from a ir
The analyte response was also suppressed due to
presence o f large amounts o f other
ionized species in the plasma,
which o rig in a te d from the sample. E a sily ionized species such as sodium,
resulted in major suppression o f the analyte response. Another drawback
was th a t sample intro du ctio n could not be confined to the 'h o t' region of
the
plasma.
'r e a l'
These problems rendered the CAP-MS unusable in the world of
sample analysis, where the analyte e x is ts
in
a fa ir ly
complex
m a trix .
The in d u c tiv e ly coupled plasma (ICP) w ith i t s
central
channel
in to which the sample is
high gas temperature
introduced, re la tiv e
freedom
from m atrix interferences such as io n iz a tio n suppression, and the high
number density of ions from trace elements introduced to the plasma (as
demonstrated by op tical
emission techniques)
was the next atmospheric
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
pressure
ion
s e n s itiv ity
source
fo r
inve stig a te d
most
elements
(6 -1 0 ).
(10^
The ICP demonstrated good
counts/sec
per
um ole/L),
good
lin e a r it y (fo u r orders o f magnitude) and good detection lim its because of
low background noise.
the
io n iz a tio n
CAP.
The ICP source also demonstrated a reduction o f
suppression
inte rfe re n ce s
experienced e a r lie r w ith the
The major peak in the mass spectrum was no longer N0+ because the
plasma was shielded from entrained a ir by the high v e lo c ity o f the Ar
c o o la n t flo w .
A r+ was now th e p r in c ip a l io n in th e mass spectrum .
However, major problems existed in the ion sampling process which took
place in th e in te r fa c e
between th e
Because o f the small sampling o r if ic e
ICP and th e mass s p e c tro m e te r.
(<70 urn) used by both Houk and
Gray, ions were sampled from a cooler boundary la y e r.
Boundary layer
sampling provided a source o f ions w ith a narrow energy spread fo r mass
analysis which resulted in good mass re s o lu tio n .
However, th is sampling
mode perm itted the form ation o f both io n ic and neutral m olecular species.
Another problem due to the small sampling o r if ic e was the condensation o f
s o lid s , from those dissolved in the nebulized s o lu tio n .
This resulted in
poor p re cisio n and degradation o f d etection lim it s .
The next major c o n trib u tio n
in the development o f the atmospheric
pressure ion source fo r mass spectrom etric d etection o f trace elements in
s o lu tio n was by Douglas and co-workers (1 1 ).
a
200W microwave
induced
plasma
(MIP)
In th a t work, they u tiliz e d
source.
T h e ir
work
again
demonstrated a system response which varied w ith the io n iz a tio n p o te n tia l
o f th e
element o f i n t e r e s t .
S im ila r to
th e CAP, as th e
a n a ly te
io n iz a tio n p o te n tia l approached the io n iz a tio n p o te n tia l o f n i t r i c oxide
(9 .2 5 e V ), th e predom inant io n in th e MIP source mass spectrum , the
r e la tiv e s e n s itiv ity was reduced by several orders o f magnitude.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
When
5
Douglas et al
compared the MIP and the
ICP, the MIP source showed a
greater dependence on sample m atrix than did the ICP source (1 2 ), but a
b e tte r response per equivalent concentration o f a n alyte .
The m ajor advantage o f the work by Douglas was the use o f m olecular
beam sampling techniques which enabled continuum sampling, as opposed to
boundary la y e r sa m p lin g, by p e r m ittin g
o r ific e s
(410 urn) fo r
sampling the
th e use o f la r g e r d ia m e te r
plasma.
Molecular
beam sampling
involved the a d d itio n o f a 1 t o r r d if f e r e n t ia lly pumped region in which
the free je t core o f the plasma was skimmed a fte r passage through the
sampling
cone.
Molecular
beam
sampling
reduced
the
form ation
of
molecular (oxide and hydroxide) species associated w ith boundary layer
sa m p lin g.
The use o f a la r g e r sam pling o r i f i c e
increased s e n s itiv ity (10^ counts/sec per um ole/L).
a ls o
re s u lte d
in
One drawback to th is
sampling mode was the increased energy spread o f the ions reaching the
mass f i l t e r which resulted in broader peaks and poorer re s o lu tio n .
A ll succeeding work has used a molecular beam region, which operates
at a pressure o f approximately 1 t o r r , w ith continuous sampling o f the
ICP (21-46).
The f i r s t commercial
Isotopes, appeared in 1983.
ICP/MS instrum ents, by Sciex and VG
As the technique developed and s e n s itiv ity
continued to improve, w ell defined peaks were v is ib le in the background
spectrum.
These peaks were due to small amounts o f molecular oxides,
hydroxides, a d d itio n a l m olecular species which were dependent upon the
sample m atrix (e g ., type o f acid used in d ilu tio n ) as well as some double
ionized species.
M aterials used in the co n stru ction o f the in te rfa c e
also became evident in the background spectra.
observed were dependent on the instrum ental
The background masses
parameters such as plasma
sampling depth, nebulizer gas flo w , s o lu tio n uptake ra te , plasma power
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and e le c tro s ta tic
ion lens p o te n tia ls
(35-37).
A dditional experiments
demonstrated th a t suppression o f the analyte signal by other ion iza ble
species
present in
dealing w ith 'r e a l'
S im ila r to the
decade
e a r lie r ,
the
sample m atrix was s t i l l
samples (33,
a consideration
45, 46).
development o f the ICP o p tic a l
in v e s tig a to rs
when
inte rested
in
emission
elemental
systems a
analysis
were
presented w ith a m u ltie le m e n t te c h n iq u e which o ffe re d a tremendous
improvement in s e n s it iv ity , by as much as three orders o f magnitude fo r
some
elements.
c a p a b ilitie s
o p tic a l
of
This
new
obtaining
interferences
technique
isotope
r a tio
offered
info rm a tio n ,
and reduction o f m a trix
samples could be g r e a tly d ilu t e d .
in te rfe re n c e
also
in th is
the
a d d itio n a l
e lim in a tio n
interferences
of
since the
Even th e most d i f f i c u l t type o f
technique, the is o b a ric
in te rfe re n c e , could many
times be elim inated by changing experimental parameters, the most obvious
o f which would be to choose an a lte rn a tiv e
which are not m onoisotopic).
c a p a b ility
of
performing
isotope ( fo r those elements
The plasma MS technique also o ffe re d the
sequential
associated w ith simultaneous analyses.
a n a ly tic a l m eetings th a t t h is
m ultielem ent
There was even ta lk
at
speeds
a t recent
new te c h n iq u e m ight be regarded as a
'panacea', making o p tic a l methods obsolete (4 7 ).
the a n a ly tic a l
analyses
Could the inventory o f
laboratory performing elemental analysis be reduced to a
sin g le instrument?
"rapid growth and
A promising new technique goes through a period o f
exaggerated expectations before i t
s e ttle s to a more
balanced p o s itio n in the a n a ly tic a l community" (48).
At the 1986 w in te r conference on plasma spectrochem istry, a major
p o rtio n o f the program was dedicated to ICP-MS.
two th ird s
of
those
papers
However, approximately
were presented by researchers
associated
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
e ith e r
w ith
loudest
(short
u n iv e rs itie s
complaint
re g iste re d
and long te rm ).
about th is
the
instrum ent manufacturers.
by actual
'u s e rs '
The
was o f poor precision
As a greater wealth o f knowledge is
obtained
new technique a more r e a lis t ic stance is being taken by the
a n a ly tic a l community.
real
or w ith
The c a p a b ilitie s and lim ita tio n s in dealing w ith
sample m atrices are in the process o f being defined.
As was the
case in the development o f the ICP-OES technique, th is extremely powerful
system has exposed an even greater need fo r competence and in s ig h t in 'up
fr o n t' chemical methods.
T h is work in v e s tig a te s the p o te n tia l
u s e fu ln e ss o f an improved
microwave induced plasma as an ion source fo r mass spectrometry (54-56).
The moderate
power microwave
induced
argon
plasma has an e x c ita tio n
temperature in the center o f the discharge of 14,200 K and an electron
d e n s ity o f 2 x 10
1^
e le c tr o n s /c c ( 5 6 ).
T his source p e rm its d ir e c t
s o lu tio n n e bu liza tio n in to an argon plasma which is v is u a lly s im ila r to
the ICP, i . e . ,
at 250 W the plasma appears as a centered plasma donut
when viewed a x ia lly .
P o tential
advantages o f using a moderate power
microwave plasma source are:
1.
The
microwave
a lte rn a tiv e
c a v ity
gases to
w ill
allow
be sustained
a n a ly tic a l
(49-55).
plasmas
in
several
A helium plasma, fo r
example, w ill perm it determ ination o f the halogens as p o s itiv e ions.
The use o f a lt e r n a t iv e
background
spectra,
support gases may r e s u lt in
which
would
enable,
fo r
s im p lifie d
example,
the
determ ination o f CaT, Fe+ and Se+ by using th e ir major isotopes.
2.
Reduced gas and power consumption re s u lt in lower operating costs.
3.
Reduced demands fo r
compared w ith the
heat d is s ip a tio n
w ith
the sm aller MIP (when
ICP) may decrease the ra te
of sampling o r if ic e
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
d e te rio ra tio n .
4.
The microwave plasma reduces the e le c tro n ic problems associated w ith
r f noise from the ICP.
j~
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
CHAPTER 2
SYSTEM DESCRIPTION
The plasma-MS system used in th is work w ill be described in the order
o f th e sequence o f events th a t occur in
an a n a ly s is .
introduced to the system in so lu tio n form.
desolvated, atomized and io n ize d .
A sample is
The s o lu tio n is nebulized,
Ions are extracted from the plasma
in to a vacuum chamber which freezes the form o f the ion by reducing the
p ro b a b ility
chamber.
of
a c o llis io n
w ith
another
ion
w hile
insid e
the
vacuum
These ions are transm itted and focused in to a quadrupole mass
filte r .
A fte r a separation based on mass to charge r a tio , s p e c ific ions
are detected by generation o f a current upon c o llis io n w ith an electron
m u lt ip lie r .
A schematic diagram o f the plasma-MS system is
shown in
Figure 1.
Sample In tro d u ctio n
There are several methods o f sample in tro d u c tio n c u rre n tly a va ila b le
to the atomic sp e c tro s c o p is t, a ll o f which should be applicable to plasma
mass spectroscopy.
These include s o lu tio n n e b u liz a tio n , electrotherm al
v a p o riz a tio n , la se r a b la tio n , generation o f gaseous hydrides and s o lid
sample in tro d u c tio n .
ways
(57,
58).
A ll o f these can be accomplished in a v a rie ty o f
However,
so lu tio n
ne bu liza tio n
is
most commonly used
because o f the r e la tiv e ease o f presenting a homogeneous sample to the
source, whether the source is viewed o p tic a lly or has ions extracted fo r
mass
filte r in g .
S olution
n e bu liza tio n
generally
presents
the
in v e s tig a to r w ith a continuous s ig n a l, when there is s u ffic ie n t sample,
which
relaxes
the
necessity
fo r
so p histicated
signal
processing
techniques.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
Figure I .
MIP-MS system schematic: (A) s ta in le s s steel sampler,
(B) s ta in le s s steel skimmer, (C) expansion stage,
(D) tr a n s itio n flo w stage, (E) e x tra c tio n , einzel lenses,
(F) energy analyzer, (G) turbom olecular pump in le t ,
(H) quadrupole mass analyzer, ( I ) electron m u lt ip lie r ,
(J) analog c u rre n t a m p lifie r to computer, (K) pulse
counting p re a m p lifie r / d is c rim in a to r to ratemeter
r
.
"
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
Sample in tro d u c tio n in th is work was accomplished by using pneumatic
n e b u liz a tio n .
In itia l
experiments were conducted using a glass f r i t
n e bu lize r because o f i t s high tra n s p o rt e ffic ie n c e y (5 9 ),
This nebulizer
reduced the volume of standard so lu tio n s u tiliz e d during the e a rly stages
o f system development which were la rg e ly spent attem pting to obtain ion
transm ission.
However, due to the excessively long time period required
fo r wash-out, the glass f r i t
(60, 61).
was replaced with a concentric nebulizer
This was a less e f fic ie n t nebulizer but the wash-out time was
reduced to approximately one minute.
aided in
d ro p le t size
e ith e r
nebulizer
Nebulizer/plasma
c o n tr o lle r .
re d uctio n .
was
support
An impacter in the spray chamber
The rate of sample in tro d u c tio n to
c o n tro lle d
gas
flow
was
with
a
regulated
p e r is t a lt ic
w ith
a
mass
pump.
flow
System components and operating parameters are lis te d
in
Table 1.
Microwave Plasma - Ion Source
During the past decade, atomic spectroscopy has undergone a s h if t in
emphasis from the use o f flames as e x c ita tio n
plasmas sustained by electrom agnetic fie ld s .
sources to the use o f
The most popular o f the
plasmas which are used a n a ly tic a lly is the radio frequency in d u c tiv e ly
coupled plasma (ICP). This plasma has proven to be an extremely durable
source capable o f generating excited state atoms and ions fo r o p tic a l
emission work.
The ICP has more recently generated tremendous in te re s t
as an ion source fo r mass spectrometry (29-31).
This work u t i l i z e s
source.
a microwave induced plasma (MIP) as an ion
The MIP, w hile i t
does not command the same a tte n tio n as the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
T a b le 1 . — System components and sp ecifications
System Component
S p e c if ic a t io n s /O p e r a t in g P a ram ete rs
N e b u liz e r :
40 p si in p u t p re s s u re , solution
M e in h a rd co n c e n tric ,
Type A
Mass flow c o n tro lle r:
T y la n
u p ta k e 0 . 5 - 1 . 2 m L/m in re g u la te d
b y p e r is ta lt ic pump
Maximum flo w s:
2 SLPM A r
20 SLPM N £
10 SLPM He
M icro w ave g e n e r a to r:
200-500 W
M ic ro -N o w , 420-2
M icro w ave c a v ity :
Tm010’ internaUy tuned>
Modified f o r ion sampling
w a t e r cooled
AlgOg containment tube
Vacuum chamber:
See table 2 f o r cham ber p r e s s u r e s
L a b o r a t o r y d esign ed and
b u ilt
Vacuum pumps:
See table 2 f o r pu m p in g speeds
V ane p u m p s /A lc a te l,
L e y bold H eraeu s,
E d w a rd s
D iffu s io n p u m p /V a r ia n
T u r b o molecular p u m p /B la z e r
Io n lenses an d p o w er s u p p ly :
Lenses la b o r to ry d esig n ed
See table 3 f o r lens voltag es
an d b u ilt
S u p p ly - S ta n d a r d
0 - 2 5 0 V , 0 .0 5 A
P o w e r, I n c .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
T a b le 1 . — System components and specifications— C o ntin ued
S ystem Component
S p e c if ic a tio n s /O p e r a tin g Param eters
Mass f i l t e r :
Mass ra n g e :
H e w le tt P a c k a r d , 59930
10 to 800 amu; w ith
u n it resolution
Scan ra te s :
9 0 .6 to 725 am u/s
Mass f i l t e r :
q u a d ru p le , 203 mm
long h y p e rb o lic rods
D e te c t o r :
Galileo 4872 continuous
d yno de electro n
P ow er:
Up to 3 k V u n d e r com puter
control
L in e a r ra n g e :
1 MHz
m u ltip lie r
S ig n a l processin g:
A n alo g mode:
D ata acquisition and sto rage
H e w l e t t - P a c k a r d 1000 E
co m pu ter
Pulse c o u n tin g mode:
M od ern In s tru m e n ta tio n
O v e r 100 MHz in p u t f r e q u e n c y
T e c h n o lo g y , F100T
p r e a m p lifie r ,
d is c rim in a to r
PRM 100 r a t e meter
Maximum in p u t pulse re p e titio n
r a te :
100 MHz w ith 10 ns
resolution
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
ICP, is none the le s s , another widely used e x c ita tio n source fo r atomic
spectroscopy (51, 52).
MIP's vary in the types o f devices which are used
to s u s ta in an a n a ly tic a l plasma.
There are a v a r ie ty o f microwave
generators a v a ila b le ranging in power from 100 W to several k ilo w a tts ,
w ith the cost increasing w ith power o u tp u t.
There are also a number of
ca v ity designs which are used to couple the power o f the generator to the
plasma.
In the present work a TMq^q resonant c a v ity was used due to it s
a b ilit y to to le ra te the d ire c t in tro d u c tio n o f analyte aerosols in to the
plasma (49, 53-55).
M o d ifica tio n s to the Beenakker c a v ity incorporate a
double-stub tuner fo r in te rn a l
impedance matching enabling operation a t
powers up to 500 W, which is the maximum output o f the magnetron.
The in te r n a lly tuned c a vity required fu rth e r m o d ifications which are
shown in F ig u re 2 .
In th e o p tic a l
em ission w ork, the end o f th e
containment tube from which the plasma extended was on the same side o f
the c a v ity as the tuning stubs, which would prevent ion e x tra c tio n from
the plasma.
When the containment tube entered the c a v ity from the tuner
s id e , ions could be extracted from a cooler p o s itio n in the plasma plume
(a fte rg lo w ).
However, sampling at a distance closer than approximatley 2
cm was p h y s ic a lly im possible.
Therefore, a section o f the water cooled
side o f the c a v ity was removed in order to relocate the tuning stubs to
the
backside o f the
c a v ity .
The demountable coupling loop was then
grounded to the c a v ity fa ce p la te .
The e n tire assembly (c a v ity A) which
was made of copper was then plated w ith a 1.3 urn laye r o f s ilv e r followed
by a 0.25 urn la y e r o f gold.
This maintained a h ig h ly conductive surface
and elim inated problems associated w ith o xid a tio n o f the Cu.
o f the tu n ing /cou p lin g
Relocation
loop assembly perm itted deeper sampling o f the
plasma as w ell as more e ffic ie n t cooling o f the tuning stubs.
With th is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
Figure 2.
Designs o f m odified microwave c a v itie s
used as ion sources.
A djustable
sV de
fZ
mi....
Jiii."
1 i::
,ii
___ iA j-
BE----
G
Coupling loop
—
11
- 1— I
ii
1
I
C>
i
i
i
i
il_
i
i
1
Discharge tube
Face p la te
Low power (250 W) c a v ity
H20 ja c k e t
.V i
r
1—
A
1i
' ^ 1
m
Tuni no screws
U
High power (300 - 500 W) c a v ity
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
c o n fig u ra tio n , a stable centered plasma could be sustained up to 250 W.
A second c a v ity was used to sustain higher power centered Ar plasmas
(400-500 W).
The inner diameter o f th is c a v ity was reduced from 86.5 mm
to 84.8 mm, the tubing stubs were again located on the backside o f the
c a v ity ,
and
two
tuning
screws
separated
by
15.3
mm
(the
same
co n fig u ra tio n as the s lid e tuners) were added to the c a v ity fa ce p la te .
At 400-500 W the tubing
stubs
required
a d d itio n a l
cooling which was
accomplished by the a d ditio n of a water ja c k e t as shown in Figure 2.
The plasma was in itia te d in an aluminum oxide containment tube (8 mm
o .d . x 5 mm i . d . ) by holding a piece o f nichrome tape in the center of
the discharge tube.
Power to the c a v ity was increased which subsequently
heated the nichrome and p ro vid e d a source o f seed e le c tr o n s .
nichrome tape was withdrawn
a fte r
s o lu tio n
Ar,
nebulization
w ith
a
in it ia t io n
stable
o f the
centered
plasma.
plasma
The
During
could
be
m a in tain e d at powers up to a p p ro x im a te ly 250 W ( c a v ity A) w ith 0 W
re fle c te d (as measured at the generator) at an Ar flo w rate o f about 0.75
L/m in.
A fte r i n i t i a l tuning at th is power le v e l, l i t t l e
impedance matching was necessary.
Above th is
or no a d d itio n a l
power le v e l,
a plasma
spindle would form along the insid e o f the discharge tube and resulted in
a decreased discharge tube life tim e .
The decreased life tim e was due to
the thermal shock experienced by the AlgOg discharge tube.
At 250 W the
Ar plasma was approximately 3 cm in len g th , extending in both d ire c tio n s
from the c a v ity .
The above parameters allow a residence time of 47 ms
which is based on the Ar flo w rate and the volume o f the plasma.
ro u g h ly 20 tim es the residence tim e in th e ICP.
This is
A microwave power
d ensity in the plasma core is approximatley 500 W/cc in the 250 W plasma
and 1000 W/cc in the 500 W plasma which is comparable to th a t o f an ICP.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
MIP power readings were taken at the generator and the th e re fo re do not
include an transm ission losses.
The plasma is sampled from the t i p o f the cone formed by the plasma
donut
(c e n tra l
channel) as shown in Figure 8 .
plasma appears to be e l e c t r i c a l l y
At th is
'a tta c h e d '
to
p o s itio n , the
th e sam pler cone.
Sampling distance is g e ne ra lly between 3 mm and 8 iron from the end o f the
discharge tube and is determined by the nebulizer/plasm a gas flo w rate
and the plasma power. The n e bulizer and c a v ity assembly are mounted on
o p tic a l
This
tra n s la tio n
allows
stages to permit movement in
ions to
be extracted from p o in ts
x, y ,
o ff
a x is ;
z d ire c tio n s .
however, the
center lin e o f the plasma is generally sampled.
M olecular Beam Sampling In te rfa c e
Since the microwave plasma source operates at atmospheric pressure
-6
and the mass spectrometer at approximately 10 "
t o r r , marriage o f the
source w ith the d e te cto r requires an in te rfa c e which has the c a p a b ility
o f sampling species d ir e c tly from the plasma which are representative o f
events ta kin g place in the plasma source and not in the sampling process
or in t e r f a c e .
In o rd e r f o r t h is to be th e case, sampled ion s must
ra p id ly achieve a c o llis io n le s s s ta te .
This is accomplished through the
use o f an in te rfa c e which produces a high in te n s ity m olecular, or in th is
case, atomic beam.
lit e r a t u r e
The molecular beam in te rfa c e f i r s t appeared in the
in the 1930's
(6 4 ).
Recently, m olecular beam sampling has
been used to m onitor dynamic systems which involved a q u is itio n o f k in e tic
data during the observation o f short liv e d species (62, 64).
Molecular
beam sampling has also been used to sample species present in atmospheric
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
3A.
Sampler mount / expansion stage
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
CO
▼
U
t
1.242
t
.750
to
L
------
S
o
FT
I
I
J i i ____
1:
i tf C*O
K-0>—
CO
CO
LU
-I
z
.
UJ
CD
se
^8
<
hCO
U1
Oo
ID
ST
.
.
.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
3B.
Skimmer mount / face o f second vacuum stage
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SKIMMER
PLATE / STAINLESS
STEEL
22
ST•
~
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
3C.
Sampler, skimmer cones
f. ■ .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
SKIMMER
SAMPLER
.65 mm HOLE ■(.025)
►
1.498
2533’
.015
.703
31 2’
1.498
.630
g f0 .7 mm HOLE (.028)
iK
.750
t
.015
STAINLESS STEEL
.125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
.eto
25
flames (6 3 ), and as an in te rfa c e between a hollow cathode discharge and a
quadrupole mass spectrometer (65).
The design o f the in te rfa c e used in th is work is
illu s tr a te d
in
Figure 3 and is s im ila r to those designs used in most ICP-MS instruments
in use today (3 2 ), and is based on recent studies in v o lv in g generation of
high density molecular beams w ith narrow v e lo c ity spreads ( 6 6 , 67).
in te rfa c e minimizes the in t e r a c t io n
This
between the background gas in the
expansion stage, w ith the expanding j e t by creation o f a " fr e e - je t zone
o f s ile n c e ".
This is a region unaffected by the background gas, where
the supersonic expansion operates as i f
it
entered a p e rfe c t vacuum as
shown in Figure 4.
The in te rfa c e consists o f a dual cone system w ith a vacuum stage
between the two cones.
The sampling cone illu s tr a te d
in Figure 3c is
made o f sta in le ss ste e l w ith a 117° external angle and a 90° in te rn a l
a n g le .
The e x te rn a l
angle o f th e sam pler is
designed to
p re ven t
form ation o f a stagnant boundary la ye r which would in te rfe re w ith ion
e x tr a c tio n from th e plasma w h ile th e in te r n a l angle is designed to
m aintain a high gas conductance or pumping speed in the expansion stage.
The sampling cone used in th is work had a 0.70 mm diameter o r if ic e and a
length/diam eter r a tio of 0 .5 .
were determined
s ta n d a rd .
s ta in le s s
by
using
a L e itz
The skimmer a ls o
steel
w ith
O rific e sizes o f the sampler/skimmer cones
stereomicroscope and a c a lib ra tio n
i ll u s t r a t e d
a 62° external
in F ig u re 3c was made o f
angle and a 51° in te rn a l
angle.
Recent molecular beam work has shown the optimal skimmer angle to be 50°.
The skimmer angle is
defined as the mean o f the in te rn a l
angles (6 7 ).
The skimmer angle in th is work was 56.5°.
skimmer
designed
is
to
e x tra c t
the
fre e
je t
and external
The sharp lipped
core
w ith
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
lit tle
C
O
o
CD C
O
T~_
Fiaure
4.
X
Molecular
beam sampling
interface
for
MIP - MS
26
~
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
in te rfe re n c e .
A fte r gas passes through the sampler o r if ic e i t expands as
a su p e rso n ic f r e e je t in t o th e expansion chamber.
A p o r tio n o f th e
ce ntral core o f th is j e t (zone o f sile n c e ) which has had no in te rra c tio n
w ith
the
sampler
or
the
background
gas
is
e xtra cte d .
Skimmer
in te rfe re n c e problems are reduced by skimming at a Knudsen number near
u n it y .
At t h is
p o in t , th e mean fre e path and th e skimmer o r i f i c e
diameter would be equal.
fre e
The skimmer Knudsen number, Kng is the mean
path length at the skimmer o r if ic e
o r if ic e
diameter
(Ds ) ,
(A ) divided by the skimmer
and can be determined by using the
fo llo w in g
expression:
K»s = a /Ds
where,
\
n
=
(16/5) v
D nm(27T kT/m) 2
m
and,
v
= v is c o s ity c o e ffic ie n t fo r argon, 0 .6 x 10
n
= number density (P^/RT)
P-
= impact pressure at the skimmer o r if ic e
Pi
= 0.6595
Dq
= sampler o r if ic e diameter
xg
= sampler/skimmer separation, 0 .8 cm
_3
poise
(D q/ X s ) 2 PQ
PQ = plasma pressure, 1013 mbar ( lx 105 kg m_1 s ' 2)
k
= Boltzman constant (1.38 x 10” 2^kg m2 s ” 2 K“ ^ molecule"''')
m
= mean molecular weight o f the gas, 39.948 g/mole
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
T
= loca l gas temperature at the skimmer, 1000K
An estimated thermal temperature o f 1000 K at the skimmer o r if ic e was
used fo r the Kng c a lc u la tio n .
This temperature is considerably higher
than th a t observed in the expanding j e t o f neutral molecular beam studies
and was based on the d is c o lo ra tio n observed in the form o f rings on the
skimmer cone.
These rings are due to the presence o f a thermal gradient
th a t e x is ts between the skimmer top and i t s
combinations were used in th is work.
base.
Two sampler/skimmer
A 0.07 cm sampler was used w ith a
0.10 cm skimmer and 0.07 cm sampler was used w ith a 0.065 cm skimmer.
Dq = 0.07 cm and Dg = 0.10 cm, then Kns = 0.069.
If
This Kn$ value is
approximately 15 times below what is recommended by supersonic m olecular
beam studies fo r m inim izing skimmer in te rfe re n ce in the j e t expansion.
However, th is Kn$ value is comparable to those encountered in atmosphere
pressure in d u c tiv e ly coupled plasma sampling mass spectrom etric systems.
The length o f the skimmer has also been shown to be im portant in
reducing skimmer in te rfe re n ce e ffe c ts by reduction o f the end wall e ffe c t
which is due to a shock wave produced by the skimmer mount (6 7 ).
The
height o f the skimmers used in th is work was 19 mm.
F in a lly ,
in
order to optimize beam in te n s ity
and reduce skimmer
in te rfe re n ce e ffe c ts , the sampler/skimmer separation must be optim ized.
Conditions recommended fo r maximum beam in te n s ity are shown in Figure 4.
This distance is dependent on source and expansion stage pressures, and
on sampler diam eter.
The in te ra c tio n o f the supersonic gas flow through
th e sam pler w ith th e am bient background gas produces a shock wave
re fe rre d to as the Mach d is k , XM.
This shock wave forms at a distance
from the o r if ic e given by the fo llo w in g equation:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
XM ' ° - 67 0 0 <Po/P l >1 /2
where PQ Is the MIP pressure, P- is the pressure o f the background gas in
the expansion stage, and Dq is
the diameter o f the sampling o r if ic e .
Based on the work by Campargue (6 7 ), the optimum separation is then,
(Xs) max = 0.75 XM
where (Xg) max is the optimum sampler/skimmer separation.
This could be
varied by the addition/rem oval of Teflon gaskets which were also used to
e le c t r ic a lly
is o la te
the
skimmer cone when used in
Teflon bushings on s ta in le s s steel mounting screws.
conjunction w ith
A negative p o te n tia l
applied to the skimmer serves to increase the cross sectional area o f the
ion beam which is sampled as well as to decrease the number o f electrons
which enter the skimmer.
count ra te , very l i t t l e
Although a negative bias did increase the net
work was done in th is mode.
A ll data presented
here were acquired using a grounded skimmer.
Vacuum System - Design Considerations
The
m olecular
beam
sampling
in te rfa c e
described
above
places
s trin g e n t demands on the design o f a vacuum system which is capable o f
reducing the pressure from th a t found in the microwave induced plasma to
the
pressure at which the
filt e r e d and detected.
sampled ions can be tra n s m itte d ,
focused,
This pressure reduction is accomplished by using
a th re e stage vacuum system .
In d e sig n in g a vacuum chamber i t
is
im portant to keep in mind th a t the e ffic ie n c y o f ion transmission through
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
the system is dependent on the mean fre e path o f the p a r tic le which is
tra v e lin g through the chamber.
This average distance which a p a r tic le
tra v e ls
w ith
before
it
c o llid e s
pro po rtio n al to the pressure.
another
p a r tic le
is
inve rse ly
To obtain good transm ission e ffic ie n c y the
mean fre e path must be longer than the dimensions o f the vacuum chamber.
When th is
is the case, the p a r tic le w ill
have a greater p ro b a b ility o f
experiencing a c o llis io n w ith the w alls o f the vacuum vessel than w ith
another p a r tic le .
This
requires a high pumping speed at the chamber
which is o f course dependent on the pumping speed o f the vacuum pump, but
i t is ju s t as dependent on the conductance o f the tubing which is used in
connecting the vacuum pumps to the chamber.
The la rg e r the diameter of
the tu b in g , the greater the p ro b a b ility o f p a rtic le s bouncing downhill
(in the d ire c tio n
o f lower pressure) toward the vacuum pump and being
expelled.
Since th is
is a three stage pressure reduction system, d iffe r e n t
types o f vacuum pumps are required, the e ffic ie n c y o f which are dependent
on the pressure region in which they are being operated.
Table 2 lis t s
the pump ty p e , the pumping speed at the in le t to the pump and the stage
in which the pump is operated.
The size o f the pumps which are required
is dependent upon the o r if ic e sizes in each o f the three stages.
The
size o f the o r if ic e regulates the gas flow or throughput, Q, o f gas in to
the system.
For every stage which has achieved an e q u ilib riu m pressure, the flow
through the o r if ic e must equal the ra te at which the pump evacuates the
system (68-71).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T a b le 2 . — Vacuum system
S tage
P r e s s u re (m b a r ) *
Pump T y p e
Pump Speed ( L / s )
1
1 .3
2 - R o t a r y V ane
Mechanical ( R V )
5,
2
3 . 7 X 10~4
Oil D iffu s io n , R V
15402 , 13
3
2 .1 X 10~5
T u rb o m o le c u la r, R V
1702 , 1
6. 7
* P re s s u re s obtained w ith 0 .0 7 cm sam pler o r ific e an d 0 .1 0 cm skim m er
o r if ic e .
2
Pum ping speed fo r A r .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
Q.jn is also dependent on the pressure region in which the system is
operating which determines the type o f flo w through a co n du it.
Three
types o f flow through e ith e r tubes or o r ific e s which are encountered in
th is system are viscous, tra n s itio n a l and m olecular.
The c r it e r ia used
to describe each flo w regime is the Knudsen number.
This is defined as
the r a tio o f the mean fre e path o f a molecule (^) to the diameter (d) or
length o f the o r if ic e or tube through which the gas is flo w in g .
Knudsen
number lim its assigned ( 6 8 ) to d e fine the type o f flow are as fo llo w s .
When the reciprocal o f the Knudsen number, dA> 100, the flo w is viscous.
When d/^< 1.00, the flo w is molecular and when 100 < dA l.O O , the flow is
tr a n s itio n a l.
Viscous flow is encountered through the sampler o r if ic e
fir s t
(expansion) stage.
and in the
T ra n s itio n flo w is encountered in the region
surrounding the skimmer o r if ic e and m olecular flow is encountered in the
d if fe r e n t ia l pumping o r if ic e .
The gas flow in to the vacuum system during e x tra c tio n o f ions from
the plasma can be estimated by determ ining the viscous flo w conductance
(32) o f the sampler o r if ic e , C
c .
sp
SLSH M ^,
(2)
4(M R T0 ) 2
where
Dq
= sampler o r if ic e diam eter, 7.0 x 10
N
= Avogadro's number
-4
m
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
PQ
= MIP pressure 1.0 x 10^ kg m _1 s " 2
M
= mean molecular weight o f the gas (kg/mole)
R
= gas constant = 8.31 kg m2 K" 1 sec- 2 mole" 1 ( i . e . , JK" 1 mole -1 )
Tq
= gas temperature o f the MIP, 3000K (72)
t
= s p e c ific heat r a tio o f A r, (Cp/Cv) = 1.67
f
(t)
=t
1 /2
[ 2/( r + l) ]
('i'+D / 2 (-r-l) =
C = 5.3 x 10
sp
0>73
on
molecules/s
or
Cgp = 8.9 x 10" 4 moles/s
Converting from moles/s to L/'s,
where,
PV = nRT
and,
V/s = nRT/P0s
(3)
Csp = 2.2 x 10" 1 L/s
The gas flow through the skimmer o r if ic e can be estimated by equation
4 which relates equation 2 to the conductance o f the skimmer (C ,).
l
csk = csp f <T > < V xs>2
<4)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
where,
Dg
= skimmer o r if ic e diam eter, 1 .0 mm
X$
= sampler to skimmer distance, 8.0 mm
Csk = 2.5 x 10" 3 L/s
At
the
d iffe r e n tia l
pumping
o r if ic e
(DPO),
the
molecular
flow
conductance, CqPq, can be calculated by using equation 5 ( 6 8 ) .
27Td
DPO " 3 1' C l+ ( 8 d /l
(5)
where,
d
= DPO diameter, 0.4 cm
11
= DPO length, 1.3 cm
T
= temperature o f the gas,300K
R
= gas constant
M
= mean molecular weight o f the gas
CDpn = 1.2 L/s fo r argon
Note th a t
region.
Cgpg is
independent o f pressure
in
the m olecular flow
Cgpg is also in v e rs e ly proportional to the length of the tube,
so th a t a pre ssure d i f f e r e n t i a l
car be m a in ta in e d i f
enlarged by increasing the length o f th a t o r if ic e (tu b e ).
an o r i f i c e
is
A lso, C$D, Csk
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
anc* S pO are
mass £,ePenclent
so th a t a t higher mass, the conductances w ill
be low er.
Therefore, the vacuum system w ill
have greater d if f i c u lt y
pumping He as opposed to A r.
The e ffe c tiv e pumping speed at the th ir d
stage in le t (DPO) can be
estimated as fo llo w s .
^ in
=
%P0
=
PDP0 CDP0
^
where Qopg is the throughput (mbar - lit e r s / s e c ) , Pgpg is the pressure
drop across the tube (P2 -P 3 ) > and
<W
=
P3S3
<7>
from equation 1, i t fo llo w s combining equations 6 and 7 th a t
S3
=
PDP0 CDP0 ^P3
^
and
S3
= 20 L/S
Since S3 is less than the rated pumping speed of the turbom olecular
pump fo r Ar at the in le t to the pump, there is a reduction in e ffe c tiv e
pumping
speed
at
the
o r if ic e
due
to
re s tric tio n s
in
conductance.
Pressures used in determ ining the pressure drop across the o r if ic e (tube)
are dependent on the lo c a tio n o f the pressure sensor in the vacuum system
w ith respect to the o r if ic e , so th a t calculated e ffe c tiv e pumping speeds
are approxim ations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
Ion Optics
A series o f e le c tro s ta tic lens elements is used to transm it and focus
the ion beam which was formed a fte r passing through the skimmer (73-81).
This section o f the instrument is co nstantly being m odified, so th a t the
ion o p tic a l system which is now in use w ill probably be obsolete by the
time th is document is
p rin te d .
The cu rre n t lens stack consists o f 10
elem ents as shown in
F ig u re 5 .
T h is design is based p r im a r ily on
experimental evolution w ith signal to noise being the p rin c ip a l c r it e r ia
in the evaluation o f a p a rtic u la r design.
th is
lens
system
to
determine
A mathematical treatm ent o f
p o te n tia l
d is tr ib u tio n s ,
spherical
aberration c o e ffic ie n ts , or even focal le n g th s, has not been undertaken.
A drawout e le c tro d e
is
lo c a te d
in th e expansion stage and is
constructed o f s ta in le s s steel mesh screen mounted in a s ta in le s s steel
d is c .
This
lens
permits
high conductance in
the removal
o f neutral
species, o f which the beam e x itin g the skimmer is la rg e ly comprised (3 2 ).
A negative p o te n tia l
and a c c e le ra te s
ions
applied to the drawout electrode repels electrons
in the d ir e c t io n
o f th e quadrupole (7 4 , 7 5 ).
Immediately fo llo w in g the drawout electrode is
a t r i p le
aperture lens
which is used as an einzel lens (76-78), w ith the two outer apertures
held at the same p o te n tia l.
d iffe r e n tia l
pumping
o r if ic e
Ions e x itin g th is lens are focused on the
which
is
the
fir s t
element
in
an
e le c tr o s ta tic energy analyzer, and is re fe rre d to as a Bessel Box (6 5 ).
This is
a second three element lensstack which incorporates a central
metal stop to which a p o te n tia l is a p p lie d .
The stop is necessary fo r
the removal o f beam-like neutral species and to prevent photons from the
plasma from reaching the electron m u ltip lie r which re s u lts in a reduced
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5.
Ion lens stack
rr
1- e x tr a c to r ; 2 -e in ze l le n s , o u te r element;
3 -e in ze l le n s , center element; 4-energy a nalyzer,
o u te r elements; 5 -photon/neutral stop;
6-energy a n alyze r, center element; 7-entrancelens
38
number o f background counts.
Ions e x itin g the Bessel Box encounter a
leaky d ie le c tr ic tube before entry in to the quadrupole.
This tube acts
as a conductor to dc f ie ld s , excluding them from the in te r io r o f the
tube.
By preventing the ions from experiencing the dc frin g e f ie ld s , but
a llow ing the ion to experience the r f fie ld s (7 3 ), ion in je c tio n in to the
quadrupole is improved.
The fo llo w in g procedure is used to obtain ion transm ission when a
newly designed ion lens stack is in s ta lle d .
The quadrupole is set up to
scan from 10 to 80 amu w ith the to ta l ion signal and as many as 6 major
spectral
components displayed on the CRT.
below ground p o te n tia l,
generally -10 V.
A ll
lens voltages are set
While sampling the plasma,
scanning is in itia te d w ith the channel e le ctron m u ltip lie r at -1400 V.
low
in itia l
m u lt ip lie r .
operating
voltage
is
used
to
prevent
to
the
The m u lt ip lie r voltage is slow ly boosted u n til a to ta l ion
response is obtained.
The to ta l ion response is optimized by varying the
voltage on one lens w h ile holding the others constant.
to ta l
damage
A
Several d iffe r e n t
ion maxima were observed so th a t many d iffe r e n t combinations o f
lens voltages could be used to obtain ion transm ission.
signal
As the to ta l ion
increases, the m u ltip lie r voltage was decreased accordingly and
tuning was continued.
A fte r i n i t i a l
tuning was accomplished, the lens
stack was optimized fo r the s p e c ific mass or mass region o f in te r e s t.
Current compromise co n ditio n s are lis te d Table 3.
The
focusing
actio n
of
the
in d iv id u a l
lens
elements
can
be
visu a lize d by p lo ttin g the ion count rate fo r Ar versus the in d iv id u a l
lens element voltages as seen in Figure 6.
two ten second count c o lle c tio n
fo llo w s :
periods.
Each data point represents
The tuning sequence was as
Bessel Box ( 1 ,3 ) , stop , Bessel ce n te r, drawout, einzel center,
ET.V- .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
T a b le 3 . — Compromise lens voltages
P o ten tio m eter
V
D e s c rip tio n
1
E xtrac to r
2
E in z e l ( 1 , 3 )
0
3
E in zel C e n t e r ( 2 )
0
4
E n e r g y A n a ly z e r ( 1 , 3 )
0
-170
(B e s s e l B o x )
5
P h o t o n /N e u tr a l Stop
-5 4 .9
6
E n e r g y A n a ly z e r ,
-4 4 .0
C e n te r ( 2 )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FIGURE 6.
Tuning curves o f ion lens elements fo r
^ A r T w ith the e le ctro n m u ltip lie r voltage
a t -1800 V, n e bu lize r flo w a t 750 mL/min
plasma power = 285 W, sampling distance
= 14.3 mm.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
CO
<M
O
CVJ
CO
Q
U l
( /)
U
L iJ
O
VO
Voltage
CVJ
CO
o
CO
a;
fO
S-
LO
CO
Cvl
c
3
o
+
S<£
o
■d-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
einzel
( 1 ,3 ) .
A ll 3 elements o f the Bessel Box appear to be stro n g ly
focusing, w ith ion transm ission cut o f f as the lens voltage approaches
ground p o te n tia l
in the p o s itiv e d ir e c tio n .
However, the einzel
elements th a t are located in the second stage show l i t t l e
large p o te n tia l ranges.
ion transm ission.
lens
a ffe c t over
Their p rin c ip a l fu n ctio n appears to be th a t of
A lso, the pressure in the second stage may be too high
fo r any focusing to take place.
The greatest count rate obtained fo r Ar
was w ith the three einzel elements at ground p o te n tia l.
The e x tra c tio n
o r drawout lens functioned as expected, increasing the count rate at more
negative p o te n tia ls .
A more systematic in v e s tig a tio n o f e le c tro s ta tic ion lenses should
prove to
be a worthwhile endeavor in the evolution o f th is
plasma-MS
system.
Mass Spectrometer
The quadrupole mass analyzer used in th is work was borrowed from a
GC/MS system.
The mass range of the quadrupole is 10 to 800 amu at scan
rates from 90 to 725 amu/sec w ith a re so lu tio n of 1 amu.
d riv in g and signal
A ll
o f the
processing e le c tro n ic s associated w ith the o rig in a l
instrument are s t i l l in use, w ith the quadrupole now located in the three
stage vacuum system described e a r lie r .
The only m o d ifica tio n s to the analyzer section o f the instrument
have been in th e removal o f th e a x ia l e le c tro n im pact (E l) so urce,
in clu d in g
le n s.
the
re p e lle r,
fila m e n ts ,
ele ctron
focus
plates
and drawout
This required a minor hardware change in the ion source control
board (A5) to disable the El source emission fa u lt detection system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The
43
entrance lens mounted on a m odified source block is the la s t remnant of
the o rig in a l electron impact ion source.
The advantages o f a quadrupole instrument in th is a p p lic a tio n have
been described elsewhere (6 ).
The advantages o f th is system l i e in the
associated softw are, which is
used to d riv e the quadrupole.
The mass
a n a ly z e r can be r a p id ly scanned over the s e le c te d mass range w ith
spectral data stored on d is c .
and graphic d is p la y .
These data are a v a ila b le fo r q u a n tita tio n
Mass spectral
data can also be acquired in the
selected ion mode where the analyzer jumps ra p id ly to preselected masses
w hile sampling the ele ctron m u ltip lie r current at each mass.
This mode
re s u lts in increased s e n s itiv ity i f the spectral components are known and
can be performed at data a c q u is itio n speeds which approach simultaneous
a c q u is itio n .
the
These data are stored in disc f i l e s
sampling period
fo r in te g ra tio n over
and these sampling period areas corresponding to
selected isotopes f a c ilit a t e isotope r a tio determ inations.
There are two major disadvantages o f th is instrum ent.
system
is
computer c o n tro lle d
and was designed
d e te cto r, the system is very in f le x ib le .
Because th is
as a chromatographic
For example, v a ria tio n o f the
quadrupole center p o te n tia l to monitor it s e ffe c t on ion transmission or
the use o f an external ramp generator to enable simultaneous ramping of
the quadrupole and a m ulti-channel analyzer fo r data c o lle c tio n in the
pulse
counting/scanning
re s tric te d
mode,
system such as t h is .
are
more
d if fic u lt
w ith
a
software
Another disadvantage ty p ic a l
o f a ll
quadrupole based instruments is th a t chem ically d iffe r e n t species which
are unresolved w ith u n it mass re so lu tio n could be d iffe re n tia te d by using
a mass f i l t e r w ith greater resolving power (82).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
Ion Detection - Signal Processing
Ion detection is accomplished by using a channel e le ctron m u ltip lie r
(CEM).
The m u lt ip l ie r s u p p lie d w ith th e quadrupole in s tru m e n t was
re p laced w ith a high gain CEM which p ro vid e d b e tte r d is c r im in a tio n
against background noise when operated in the pulse-counting mode (8 3 ).
T his CEM is mounted o f f - a x is
in a ceram ic housing w ith a d e fle c to r
electrode which is used to reduce the count ra te from photons, electrons
and n e utrals.
Signal processing is accomplished in e ith e r analog or pulse counting
modes.
In the analog mode current pulses from the CEM are converted to a
0 to 10 v o lt signal w ith
a log a m p lifie r.
Analog data c o lle c tio n is
computer c o n tro lle d enabling rapid scanning o r selected ion m onitoring,
peak in te g ra tio n and isotope r a tio c a lc u la tio n s .
In the pulse counting
mode, a d is c rim in a to r is used to e lim ina te noise pulses which re s u lts in
increased s e n s itiv ity
by improving the signal
to
noise
r a tio .
Pulse
frequencies are displayed on a rate meter and a hardcopy is generated by
manually recording the to ta l number of counts accummulated in a 10 second
period.
Five to te n , 10 second counting periods are averaged fo r each
data p o in t.
A ll data co lle cte d in the pulse counting mode were acquired
w hile m onitoring a sin g le io n .
The CEM is
lim ite d to a count rate o f approximatley 10 MHz where
pulses would not be resolved.
A n o n -lin e a r count rate is experienced
above approximately
1MHz. The p re a m p lifie r d is c rim in a to r could be used
at count rates up to
50MHz. The lin e a r
range o f the a n a ly tic a l signal is
lim ite d at the lower end bybackground
noise and on the upper end by the
lin e a r dynamic range o f the CEM.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
CHAPTER 3 ARGON PLASMA - RESULTS AND DISCUSSION
Since the in te r n a lly tuned TMq1q microwave c a v ity was capable o f
operating at several hundred watts (300-500) which is considerably higher
than the power level
used in the e a rly MIP-MS work (11,
improvements were expected.
tube,
12) several
The m odified c a v ity w ith an AlgOg discharge
5.0 mm I.D . x 8 mm O.D. did enable d ir e c t s o lu tio n n e b u liz a tio n ;
however, the expected reduction o f molecular oxides by an e le va tio n in
the thermal temperature was not re a liz e d .
This was immediately evident
when a background scan w hile introducing d is t ille d deionized water (DDW),
revealed th a t 14 N1 ®0 + and ^ 0 + were the p rin c ip le ions in the spectra as
seen in Figure 7.
Figure 7 also illu s tr a te s
r e la tiv e
in te n s itie s of
major and minor background peaks in both DDW and 5% HNO^.
N0+ is the
base peak (29.2% o f the to ta l abundance was due to N0+ the s in g le scan
d isp la ye d ).
In the lower d is p la y , the responses are m u ltip lie d by a
fa c to r o f 50 revealing ad ditio n al background species.
is
predominantly due to the presence o f a ir
a fte rg low
p r io r
to
sampling.
The
N0+ in the plasma
entrained
presence
of
in
the plasma
molecular
species
containing N, 0 and Ar re s u lt in a complicated background spectra which
is illu s tr a te d in Table 4.
sampling d istance.
The plasma power was 300 W w ith a 14.3 mm
The elemental
species
in whose determ ination the
background ions would in te rfe re and also lis t e d in the ta b le .
in itia l
advantages o f plasma-MS was the s im p lic ity
spectra.
number
However, th is
of
background
One o f the
o f the background
is not the case as can be seen by the large
interferences
in
the
MIP-MS.
In
fa c t,
the
s im p lic ity o f the spectra turns out to be a disadvantage because o f the
lim ite d
number
of
a lte rn a tiv e
isotopes
a v a ila b le
fo r
use
in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the
46
Fig. 7. Major and minor background ions in d is t ille d
deionized water a t 430 W w ith an Ar plasma
gas flo w ra te o f 750 mL/min.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PK/ 'PEUHD:
30 .1 /’
11429
47
39130.
EASE
\D
nEUHD=
Lf>
TOT
vD
\D
i*i*i
n
OJ
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T a b le 4 . — Background, ions* f o r w a t e r a n d 5% HhJO^
tnfZ
Most P ro b a b le I o n ( s )
I n t e r f e r e s W ith ( A b u n d a n c e )
14
14N+
N (99.63)
15
14NH+
N (0.37)
16
160 +
0 ( 99.76)
17
160H+
O (0.037)
18
16° H2+
0
19
16° H *
F (100)
23
9? +
**Na
Na (100)
27
27A t
A1 (100)
28
14N 14N +
Si ( 9 2 . 1 8 )
29
14N 14N H +
Si ( 4 . 7 1 )
30
14N 160 +
Si ( 3 . 1 2 )
31
14N16OH+
P (100)
32
160 160 +
S (9 5 .0 2 )
33
160 160 H +
S (0 .7 5 )
40
40A r +
41
41 ArH+
42
14N 3+
(0 -2 0 4 )
Ar
( 9 9 . 6 ) , Ca ( 9 6 . 9 2 )
K (6.91)
Ca ( 0 . 6 4 ) )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
T a b le 4 . — B a c k g r o u n d io n s * f o r w a t e r an d 5 i
m /Z
Most P ro b a b le I o n ( s )
I n t e r f e r e s With (A b u n d a n c e )
43
27A 1 160 +
Ca ( 0 .1 3 )
44
12 c 16 o 2+ , 14 n 2 16 o +
Ca (2 .1 3 )
45
12 c 16 o 2 h + , 14N 2 16OH+
Sc ( 1 0 0 )
46
14 ?W
Ti (7 .9 5 ),
52
3 6 . 14..+
Ar N
C r (8 3 .7 6 )
54
40A r 14N +
Fe ( 5 . 9 0 ) ,
56
40A r 16O+
Fe (9 1 .5 2 )
59
27 a i 16 o 2+
Co (1 0 0 )
68
4 0 . 1 4 ..14..+
Ar N N
Zn ( 1 8 . 5 6 )
69
40A r 14N 14N H +
Ga ( 6 0 . 1 6 )
in
76
40 A 36 . +
Ar Ar
SO
40A r 40A r *
Se ( 4 9 . 9 6 ) ,
40A r 40A r H +
B r (4 9 .4 3 )
A r plasma, p o w er = 300
r .
r-- •
— C o n tin u e d
.
.
Ca ( 0 . 0 0 3 )
C r (2 .3 8 )
K r (2 .2 7 )
W.
.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
—
r~.
50
determ ination o f an element w ith a d ire c t sp e ctra l o ve rla p .
For example,
the determ inations o f S i, P, S, Ca, Se and Fe e x h ib it e ith e r reduced
s e n s it iv ity , or they may not be able to be determined at a ll
by th is
technique
spectra
due
to
molecular
in te rfe re n c e s .
The
background
obtained w hile n e bulizing 5% HNOg are very s im ila r to th a t obtained fo r
DDW.
U t iliz a t io n
o f a lte rn a tiv e
inorganic acids
re s u lts
in
a d d itio n a l
background ions and th e re fo re a d d itio n a l background in te rfe re n c e s .
If,
f o r exam ple, HgSO^ o r HC1 is used in th e d ig e s tio n , e x t r a c t io n , o r
d ilu tio n
of
containing
a
sample
species w ill
fo r
a n a ly s is ,
in te rfe re
in
m olecular
s u lfu r
and
the determ ination
c h lo rin e
o f a d d itio n a l
(o th e r than those already lis te d in Table 4) isotopes o f Ca, T i , Cr and
Zn.
These are lis te d in Table 5.
are also lis t e d .
Several other m olecular in te rfe re n ce s
Is o b a ric inte rfe ren ce s such as those lis te d
g re a tly
increase the com plexity and reduce the p re c is io n o f isotope r a tio
and
th e re fo re isotope d ilu tio n measurements.
T
«
x n
/
>
*
■
*
r
**
'» f/xK'haTrt
vsiu C i
vvs
ssLsvCtin
f ha
vu C
r-.r.f^m
:!m sn
a
T
tz+
o
v/^svittiuiii
u i i u i jr u C
n
a
m
i i s i i
•'Q
C
n
A
n
fQ
i
}
tha
v n C
m
i r r'A
U
ipW
Q
tmv>i w n u v C
c a v ity was mounted on tra n s la tio n sleds which enabled three dimensional
p r o filin g the plasma across the sampling o r if ic e .
When the plasma is
viewed a x ia lly , there appears to be a plasma donut in the discharge tube.
This forms a cone when viewed r a d ia lly as illu s tr a te d in Figure 8 .
50 ppm Y in 0.2% HC1 was introduced a t 0.70 mL/min in to a 360 W Ar
plasma, w ith a plasma gas flow rate of 750 mL/min in order to view the
d iffe re n t a n a ly tic a l
zones in
the plasma.
There was no v is ib le blue
emission due to the presence o f Y+ and no intense red due to the presence
o f excited Y0+ emission.
However, a b rig h t yellow-orange was v is ib le in
th e area between th e two cones i ll u s t r a t e d
r .
.
-
in F ig u re 8 .
This was
.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
T a b le 5 . — A d d itio n a l b a c k g r o u n d ions*' o b s e rv e d a r is in g from in org an ic
ac id s 2
A c id
I o n , m /Z
HCl
35
2SC l+
Cl ( 7 5 . 4 )
37
37C l+
Cl ( 2 4 . 6 )
51
25C1160 +
V (9 9 .7 6 )
53
37C1160 +
C r (9 .5 5 )
H 2S 0 4
32
Most P ro b a b le Io n
I n t e r f e r e s With (A b u n d a n c e )
22S+
S (9 5 .0 1 8 )
33
33S+
S (0 .7 5 )
34
34S+
S (4 .2 1 5 )
48
32S160 +
Ca ( 0 . 1 7 9 )
50
34S 160 +
Ti ( 5 .3 4 ),
64
32S 160 2 + ,
66
SW
68
34S2
,
2 A r plasma, 300 W.
F iv e p r e c e n t H C l o r 5%
32S2+
34S i S 0 *
V (0 .2 4 ),
C r (4 .3 1 )
Zn ( 4 8 . 8 9 )
Zn ( 2 7 . 8 1 )
Zn ( 1 8 . 5 6 )
b y volume in d is tille d deionized w a te r.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
Discharge tube
Figure 8 . Radial view o f
Ar plasma,
illu s t r a t in g non-homogeniety
o f analyte ion (X '} d is tr ib u tio n .
r.*
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
presumed to
be excited
YO emission based on the
v is ib le
wavelengths
(5972, 6132) and emission in te n s itie s a t these wavelengths (8 4 ).
When
th is plasma was sampled a t th re e d iffe r e n t sampling depths as shown in
Figure 9, i t
became evident th a t most o f the Y was present as ^Y*®0+ .
At 15.8 mm, the Y0+/Y+ r a tio was approxim ately 90.
I t also appeared as
though the bulk o f analyte ions were located in the region ju s t outside
o f the inner cone as illu s tr a te d in Figure 8 .
This may be ra tio n a liz e d since the TMq^q c a v ity has a transverse
magnetic f ie ld which accelerates ions outward toward the w alls of the
discharge tube.
This observation may also be explained by a process
termed ambipolar d iffu s io n (85, 8 6 ) , whereby e le c tro n s , which possess the
same k in e tic
energy as io n s,
are moving at a higher v e lo c ity due to
diffe re nce s in th e ir respective masses and th e re fo re d iffu s e at a greater
rate from a region o f higher d e n s ity , the region o f maximum e le c tr ic
f ie ld
strength (5 0 ), to a region o f lower d e n s ity .
th is process occurs is c o n tro lle d by the
e le c tro n e u tr a lity , which is dependent on
the ele ctron s.
The ra te at which
need fo r the plasma to maintain
the rate at which
ions fo llo w
Both of these processes might explain the above results.
The sh o rte r the sampling d is ta n c e , o r the deeper the sampler was
moved in to the plasma, the lower the in te n s ity o f the ^ y i 6 g+ reSp0 nse.
At a sampling distance o f 12.8 mm, the greatest response fo r
hence the best Y0+/Y+ r a tio
(2.7)
89 +
Y and
was observed, however, the to ta l
response (*^Y+ and ^ 9 Y16 0+ ) was at a minimum.
Y
A ll subsequent sampling
was performed outside o f th is inn e r cone.
Much o f the ICP-MS work which has
problem has been performed using Ba.
io n iz a tio n p o te n tia l
been done to evaluate the oxide
Ba is also a gauge o f the apparent
o f the plasma, w ith the lowest second io n iz a tio n
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
Acid blank, 0.2% HCl
85
LIN=153
83
101
L I N =9 2
1 09
50 ppm Y
15.8 mm
89y+
89yl 6o+
\
85
LIN=316
93
101
LIN=28000
14.8 mm
I I 89„16„+
1 u
89y+
jt I
j
85
LIN=477
93
101
\
L ! N= 7 S4 3
109
89Y160+
89y+
1 2 .8
1 09
mm
J
85
L I N :- 8 9 S
93
101
LIN'=2423
169
Fig. 9. Spectra o f 50 ppm Y in 0.2% HCl a t 350 W,
illu s t r a t in g Y+ , Y0+ response dependence
on sampling distance.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
p o te n tia l
(10.001
eV)
of
a ll
illu s t r a t e s the oxide problem.
4*
to m onitor BaO
elements
(8 7 ).
Figure
10
fu rth e r
Three mass ranges ware scanned in order
+
isotopes, Ba
isotopes and Ba
background scan o f 1% HNQg - illu s t r a t e s
background ion present in th is range.
2+
isotopes.
The upper m/z
th a t 4 ^ A r* 4 N2+ is th e o n ly
In tro d u c tio n o f a 500 ppm s o lu tio n
o f Ba in 1% HNOg (lower scan) illu s tr a te s a nearly 1:1:1 r a tio o f
12 ^Ba16 0+
and 12 ®Ba*6 0H+.
Again, the
138
4*
Ba ,
region o f the plasma which was
sampled was between the two cones (Figure 8 ) , w ith a sampling distance o f
14.8 mm, 2 mm o f f ce n te r.
The same power, plasma gas flo w ra te and
s o lu tio n d e liv e ry ra te were used.
There was no evidence o f
138
24*
Ba
at
m/z = 59, however, the 500 ppm Ba s o lu tio n did appear to cause a s h if t in
the background, as there was an increase in the leve l o f
in the plasma.
40
Ar
14 +
N present
Since * 4 N ^0 + was the major ion in the plasma w ith an
io n iz a tio n
p o te n tia l
o f 9.25 eV fo r NO, l i t t l e
doubly ionized Ba was
expected.
Again, as the sampler was moved deeper in to the plasma, the
Ba+ , Ba0+ and Ba0H+ signals decreased.
The in v e s tig a tio n o f oxide in te rfe re n c e problems in the Ar MIP/MS
work was extended to a d d itio n a l elements.
illu s t r a t e s
plasm a.
The upper scan in Figure 11
th e background ions p re s e n t in 1% HNOg w ith a 225 W Ar
Background ions appear a t m/z 44 ( * 2 C ^ 0 2+ , ~4 N2 ^ 0 + ) , 45
( 1 2 C1 6 02 H+ , 14 N21 6 GH+ ) , 45 ( 1 4 N1 6 02+ ) and 59 ( 2 7 A116 02+) .
In the lower
scan, a 1% HN03 s o lu tio n containing 10 ppm o f Ca, Mg, Sc, T i, A1 and Co
was introduced.
a sm all
in c ra s e
The lower scan in d ica te s l i t t l e
over background.
45
4*
Sc p resent, w ith only
(There appeared to
be a s lig h t
suppression o f the background ions when looking at the lin e a r in te n s itie s
reported fo r the maximum ion in the m/z window during a simple analog
scan.)
r...
The response fo r
48
Ti
*4
was also poor.
However, the lower rig h t
-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
500 ppm Ba in 1% HNOg
Response
, 138 Ba+
I I38Ba+
k.
M
40. 14., ♦
■'A.r^No'
|-"i
2
l
\
j
___________ .
6 5 ' * L 1N = 1 3 7 s '
" 73 "
—
Fig
’V
j
v '*
134
138 B
a 'V
T’>.
I \
l 3 8 Ba16 0H
i
A
\
V.-'V-_________________
LIH=1211 ~
m/ z
142
150
L1N=1123
-_
158
^
10. Scans o f 1% HNOg and 500 ppm Ra in 1% HN03 illu s t r a t in g the presence o f
BaO and BaOH. The lin e a r in te n s ity (LIN) is displayed fo r the major ion
in each mass range.
43
L I N= 1 6 5 0
49
57
65
LIN =154
10 ppm Co, Sc, T i, in 1% HNOg
k
4i f
A
I ' i
/
,
Sc160 +
\
f\
5v
\
j
43
LIN =1507
/
A 1'
49
57
y
LIN =7988
65
m /z
Fig. 11. Scans o f 1% HN03 and 10 ppm Co, Sc, Ti illu s t r a t in g
predominance o f m olecular oxides. The lin e a r in te n s ity
is displayed fo r the major ion in each mass range.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
scan illu s tr a te s th a t most o f the Sc and Ti were present as
4 8 jil6 o + ^
o f Ni (
gl
59q0+ .jS a-jSQ y-jSj b i e>
.
^
Ni ; 1.25%) and TiO
45
Sc
16 +
0 and
$co+ in te rfe re s w ith a minor isotope
isotopes in te rfe re in the determ ination o f
Ni ( 6 2 Ni+, 3.66%; 6 4 Ni+, 1.16%), Cu ( 6 3 Cu+, 69.09%; 65 Cu+, 30.91%) and Zn
( 6 4 Zn, 48.89%; 6 6 Zn, 27.81%).
F igure 12 illu s t r a t e s
an extreme case o f an is o b a ric in te rfe re n ce
due to the presence o f molecular oxides o f Mo.
Three sets o f scans are
illu s tr a te d w ith each m/z range in c lu d in g the isotopes o f Mo and Cd.
upper set of scans illu s tr a te s a 1% HNOg background.
The
In the second set
o f 10 ppm Cd standard was run, and most o f the Cd isotopes are v is ib le .
^®Cd was not in c lu d e d
in th e scan window and ^®Cd w ith a 0.89%
abundance was not detected in the s in g le scan, analog mode.
In the lower
scan, a m ultielem ent standard containing 10 ppm o f Mo and Cd in 1% HNO-j
was in tro d u c e d
p r in c ip a lly
is o to p e s
in t o
th e
plasma.
as th e o x id e , w ith
o f Cd.
The o n ly
In t h is
a ll
is o to p e
extreme case, Mo e x is ts
o f th e
rg
o f Cd w ith
in te rfe re n ce is *^C d which is only 1.22% abundant.
Cd in
isot-opes o v e rla p p in g
which th e re
is
no
The determ ination o f
the presence o f Mo0+ would be performed using th is
isotope and
would re s u lt in a tremendous reduction o f the s e n s it iv ity fo r Cd.
The response per given amount o f analyte as well as the background
count
rate
were dependent
on ion
source
(MIP) operating
parameters.
These parameters which have the greatest a ffe c t on the ion signal are the
plasma power, the sampling distance and the plasma gas (n e b u liz e r) flow
ra te .
The plasma power could not be e a s ily set to a s p e c ific le v e l.
Powers were chosen whereby a sta b le centered plasma could be sustained
w ith
lit tle
re fle c te d
power.
The higher the power, the greater the
extension o f the plasma a fte rg low from the end o f the discharge tube.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
ENT LENS CMV-'AMU)
XRAY
(V)
EM VOLTAGE
(V)
50
3 9 . 7S
3000
AMU GAIN
CD)1 7 4
AMU OFFSET
CD) 6 2
MASS A X I S GAIN
1.00226
MASS A X IS O FF SE T
-.9 34631
1% HNO3
wWfvVvvWtv
91
101
107
U ?
10 ppm Cd in 1% HN03
112
114
A
A
111
lie
110 U ! )|
Ai
'V '
ti
J' -A'\ 1
V
■vAW ’1
91
1 01
107
J. 1 7
10 ppm Mo, Cd in 1% HN03
112
114
n o r lf flL lp
92 Ko160 +
9V
V
91
U
J
V
116
i II
k
1
J ^ j.j
p
101
107
117
m/z
+
Fig. 12. Spectra demonstrate is o b a ric in te rfe re n c e , MoO on
Cd; a ll isotopes o f Mo overlap w ith Cd isotopes.
Computer c o n tro lle d mass f i l t e r parameters during
spectral a c q u is itio n lis te d above.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Once 3 S t a b l e Ar p l a S m S W3S e s t a b l i s h e d , t h e S amp 1 1Hy d ' S t a i l C c WaS
optim ized fo r a p a r tic u la r flo w ra te .
fo r ion sampling is
The optimum region o f the plasma
at th e edge o f the in n e r cone.
At th is p o in t, a
d iffu s e e le c tr ic a l in te ra c tio n e x is ts between the plasma and the sampler.
The response degraded very q u ic k ly as th e sampling distance was e ith e r
decreased o r increased from th is p o in t as illu s tr a te d in Figure 13 fo r a
1% HNOg s o lu tio n o f 10 ppm Cu.
a c q u ire d w ith a q u a rtz
entrained a ir .
(The data in Figures 13 and 14 were
bonnet added to
th e sam pler which reduced
E ffe c ts o f the bonnet are discussed la t e r . )
The to ta l
distance tra ve le d in th is p lo t was only 0.8 mm.
Because the response is
so s e n s itiv e to
any changes in
s t a b i li t y
e ith e r
changes in
by
sampling d is ta n c e ,
arcing
to
the
containment
tube
or
plasma
from
the
in tro d u c tio n o f a sample, may re s u lt in a change in the plasma geometry
and a flu c tu a tio n in the ion count ra te .
The s o lu tio n uptake ra te had no
a ffe c t on the ""Cu+ count ra te above 0.5 mL/min.
Slower s o lu tio n uptake
rates caused arcing o f the plasma to the A^Og tube w a ll.
F ig u re 14 illu s t r a t e s the e ffe c t o f a change in the plasma gas flow
ra te on the
changes
geometry.
in
63
Cu
count ra te . The response again changes a b ru p tly w ith
Ar flo w
For
ra te
example,
because th e flo w
ra te a ffe c t s th e plasma
plasma
from
extension
the
discharge
tube
increases w ith increased flo w ra te , re s u ltin g in a necessary decrease in
the sampling distance.
The analyte response s e n s itiv ity is dependent on several a d d itio n a l
fa c to rs .
1.
These include:
The f i r s t
io n iz a tio n
p o te n tia l
(I.P .)
o f th e elem ent and th e
apparent io n iz a tio n p o te n tia l o f the plasma.
2.
The atomic weight o f the element under in v e s tig a tio n .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
- Quartz bonnet #5, 7.3 mm
Response fo r
sampling distance
10 ppm 6 3 Cu+
125
- microwave power = 310 W
- s o lu tio n in tro d u c tio n ra te
(kHz)
= 0 .7 0 mL/min
100
75
cr>
50
25
7.0
Sampling d is ta n c e , (mm)
Figure 13.
t.
9.0
8 .0
E ffe c t o f sampling distance on
/TO
Cu
count ra te .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 14. E ffe c t o f plasma gas flo w ra te , (mL/min)
- Quartz bonnet #5, 7.3 mm
Response fo r
i
10 ppm
(kHz)
sampling distance
- microwave power = 310 W
125
- s o lu tio n in tro d u c tio n ra te =
0.55 mL/min
100
600
700
Ar plasma gas flo w r a te ,
V
900
800
(mL/min)
3.
The isotope abundance.
As th e f i r s t
p rin c ip a l
ion in
decreases.
6 4 Zn+
I.P .
o f an analyte element approaches th a t o f the
the plasma, the amount o f io n iz a tio n
o f the analyte
This may account fo r the decreased count ra te observed fo r
when compared to ^C u + as illu s tr a te d in Table 6 .
The f i r s t IP o f
Zn is 9.391 eV, ju s t above th a t o f NO at 9.25eV.
There are several manners in which the atomic weight influences the
response.
There is a d is c rim in a tio n against lig h t ions in the m olecular
beam sam pling in t e r f a c e .
The expanding j e t w i l l e x e r t a g re a te r
influ e n ce on lig h te r io n s .
Heavier ions w ill have a tendency to remain
tra v e lin g in the d ire c tio n o f the mass f ilt e r / d e t e c t o r because o f t h e ir
greater momentum.
Balancing th is e ffe c t is the quadrupole transm ission
e ffic ie n c y , which becomes poorer as the mass increases.
atomic
weight
influ e n ce s
the
response
depending on
preparation o f standard so lu tio n s fo r a n a ly s is .
O bviously, the
the
manner
of
For example, Cr and Pb
standards o f equivalent concentration when prepared on a ppm basis w ill
c o n ta in
4
normalized
tim es
as many Cr atoms as Pb atoms.
responses which take the
Table 6 p re sen ts
isotope abundance and the atomic
weight in to c o n sid e ra tio n .
Table 7 illu s t r a t e s the detection lim its obtained a t two d iffe r e n t
power le v e ls w ith a 0.07 cm sampler and a 0.065 cm skimmer.
When the
power was reduced from 350W to 225W, the sampling distance was reduced
from 14.3 mm to 13.5 mm in order to obtain the optimum net count ra te .
For every element stu d ie d , the detection lim it s improved at I o a r power.
There was
lit tle
improvement in
the
background
count ra te waslower and the standard
w ith the background counts was reduced.
net count r a te ;
however, th e
d e via tio n associated
D etection lim its are reported as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
T ab le 6 . — Normalized re sp o n se1
Element
Abundance
1st I . P .
(eV)
N e t C o u n t s /s
ppm
N e t C o u n ts /s
M
6 . 8 X 108
52
Cr
8 3 .7 6
6 .7 6
1 1,000
63
Cu
6 9 .0 8
7 .7 2
4 ,3 0 0
3 .9 X
o00
m /Z
64
Zn
4 8 .8 9
8 .3 9
940
1 .2 X
10s
114
Cd
28.81
8 .9 9
1 ,2 0 0
208
Pb
52 .3 8
7.41
890
4 . 7 X 10*
Co
I
c> ;1
3 .5 X
* Microwave po w er (m e asu red at g e n e r a t o r ) = 350
distance = 1 4 .3 mm.
W;
sampling
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
Table 7. — Detection limits* at two power levels
ng/ml
M/Z
Element
@ 225 W2
§ 350 W3
52
Cr
22
47
63
Cu
7
85
64
Zn
70
320
114
Cd
17
238
208
Pb
33
94
* Detection limit defined as concentration equivalent to three times
the standard deviation of the background count rate; based on mean
of ten 10-second measurements of background and standard, with
standard response in linear portion of response curve; 0.07 cm
sampler orifice, 0.065 cm skimmer orifice.
2
3
Sampling distance = 13.5 mm at 225 W.
Sampling distance = 14.3 mm at 350 W.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
the
concentration whose count ra te
is
equivalent to
three times the
standard deviation o f th e blank count ra te .
Table 8 l i s t s the d etection lim its obtained a t 225 W using a 0.07 cm
sampler and a 0.10 cm skimmer.
Several a d d itio n a l elements are included.
L i t t l e o ve rall improvement was observed w ith the la rg e r skimmer.
In most
cases, n e ith e r background count rates nor net analyte count rates were
a ffe c te d .
Improvements th a t were observed were probably due to a more
sta b le plasma source on the p a rtic u la r day th a t these data were acquired.
The noticeably poorer d e te ctio n lim its
h ig h
f lu c tu a tin g
background counts
obtained fo r ®®Mn+ were due to
a t m/z = 55 where th e re
is
a
c o n trib u tio n
from both ^ A r ^ N + and ^ A r*® 0 + which are not adequately
resolved from
55 +
Mn .
C a lib r a tio n curves fo r Cu and Cr are illu s tr a te d in Figure 15.
The
lin e a r range is lim ite d to three orders o f magnitude, and is lim ite d at
the low concentrations by background noise and a t the high concentrations
by a nonlinear response o f the electron m u ltip lie r at count rates much
above 1 x 10 ® counts/s.
The a ffe c t o f the presence o f an e a s ily ionized element, in th is
case Na, on the io n iz a tio n o f Cu is
were
acquired
at two
illu s tr a te d in Figure
d iffe r e n t power
le v e ls using
16. These data
the concentric
n e b u liz e r w ith a s o lu tio n uptake ra te o f 0 .7 0 m L/min.
sampler and 0.065
sampled from 2
cm skimmer were
mmo f f a x is .
used in th is work.
The 0.07 cm
The plasma was
Sampling distance was again
accommodate changes in the microwave power.
changed to
At both 360 W and 420 W, Na
concentration o f 100 ppm resulted in a complete loss o f response fo r 10
ppm Cu.
This ionizaton suppression occurs much e a r lie r w ith the MIP than
is reported w ith the ICP.
The decrease in the response fo r Cu was not a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
Table 8 .—Detection limits at low power with 0.10 cm skimmer
M/Z
Element
52
Cr
55
Mn
58
Ni
59
Co
Abundance (%)
83.76
100
67.76
100
ng/ml
25
670
23
6.0
63
Cu
69.09
12
64
Zn
48.89
38
114
Cd
28.81
16
208
Pb
52.38
6.7
Detection limit defined as concentration equivalent to three times
the standard deviation of the background count rate; based on mean
of ten 10-second measurements of background and standard with
standard response on linear portion of response curve; single ion
mode; 0.07 cm sampler; microwave power = 225 W.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
log concentration
(ppm)
68
I*-■ ■
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
Figure 16.
6 3 Cu+
E ffe c t o f Na on io n iz a tio n o f 10 ug/mL Cu.
count ra te
a
P
360 W Ar plasma, z = 14.3 mm
O 420 W Ar plasma, z = 15.3 mm
(c o u n ts/s) x 10 "
-
s o lu tio n in tro d u c tio n ra te =
0.70 mL/min
-A2.0
\
tr
\
\
\
\
\
\
'
N
\
\
\
'
\
1.0
1
' \
\ \
u
w
\I
\
0 added Na
log
Na
ug/mL
70
Figure 17.
Major and minor background ions in
d is t ille d , deionized water in a 300 W
Ar plasma w ith a quartz bonnet; plasma
wj
gas flo w ra te = 750 mL/min.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
r e s u lt o f o r if ic e plugging as n e bu liza tio n o f a 10 pm Cu standard w ithout
added Na restored the i n i t i a l count ra te .
B e tte r s e n s it iv ity fo r Cu was
observed at the lower power (360 W).
In an attem pt to s ta b liz e the Ar plasma, a quartz bonnet was added
to the face o f the f i r s t vacuum stage as illu s tr a te d in Figure 18.
The
bonnet was simply a quartz rin g which reduced the amount o f a ir entrained
in the plasma.
The torch shown was not used in these experiments.
With
the bonnet in place, a ir currents in the room no longer a ffe c te d the
plasma and the o ve ra ll plasma s t a b ilit y was improved.
The p rin c ip a l ion
in the plasma s h ifte d from *^ N ^ 0 + to ^ A r + , as illu s tr a te d in Figure 17
which should help improve the s e n s itiv ity fo r higher io n iz a tio n p o te n tia l
e le m e n ts.
The net count r a te s , th e s ig n a l to
background and the
d etection lim its were unchanged w ith the a d d itio n o f the quartz bonnet.
A lso , the molecular oxide problem was not improved.
A scan o f 10 ppm Mo
in 1% HN03 indicated th a t the isotopes o f Mo0+ were again th e p rin c ip a l
species present.
Sc was also present p rim a rily as ScO+ .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
CHAPTER 4 HELIUM PLASMA
In tro d u ctio n
Most plasma mass spectrom etric work has been performed w ith an argon
in d u c tiv e ly coupled plasma (ICP) source (29-31).
A few researchers have
used the argon microwave induced plasma (MIP) source (11-12).
In e ith e r
case, the Ar plasma source has demonstrated e xce lle nt detection lim its
f o r most elements w ith a sim p le background sp e ctra when u t i l i z i n g
deionized HgO or HNOg so lu tio n s (3 8 ).
gas
has
some
lim ita tio n s .
Major
However, the use o f Ar as a plasma
Ar
background
ions
preclude
the
determ ination of ^C a +, ^ F e + and ^ S e + , which are the most abundant ions
o f Ca, Fe and Se.
The Ar plasma does not generate su b sta n tial q u a n titie s
o f Br+ and especially Cl+ which are only 5% and 0.9% ionized re sp e ctive ly
(8 8 ) .
Plasma mass spectrom etric work in v o lv in g the determine o f Br and
Cl has been performed in the negative ion mode which to date is less
s e n s itiv e
( 1 1 ) , or in the p o s itiv e
ion mode w ith
re s u ltin g
detection
lim its in the nanogram per second range ( 8 8 ) .
The moderate power Ar MIP-MS s tu d ie s presented in
e xh ib ite d
problems
flu c tu a tio n s
in the
w ith
plasma
ion
s ig n a ls .
s t a b ilit y
M atrix
which
resulted
and io n iz a tio n
Chapter 3,
in
large
suppression
e ffe c ts have been more pronounced w ith the Ar MIP than w ith the Ar ICP
(1 2 ).
In the MIP, d i f f i c u l t y
is encountered in io n iz in g species w ith
io n iz a tio n potentials greater than th a t o f NO (9.25 eV), which is the
p rin c ip a l ion in the Ar MIP mass spectrum when using a s in g le plasma gas
flow
sampled conventionally
(11-12).
(th a t
is ,
w ithout
a sheathing
flow
gas)
Large amounts of N0+ in the plasma lim it the plasma's a b ilit y
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
to io n ize the halogens.
Helium microwave plasmas w ith o p tic a l emission detection have been
used e xte n sive ly as element s e le c tiv e detectors fo r gas chromatography
(8 9 -9 1 ).
nonmetal
The advantage o f using the He plasma is i t s a b ilit y to e xcite
atomic em ission.
A He ICP has re ce n tly been reported as an
emission source fo r determ ination of aqueous Br and Cl (9 2 ).
P relim inary
re s u lts presented here were obtained w ith an atmospheric-pressure He MIP
at moderate powers as an ion source fo r the de te ctio n o f B r, Cl and I .
M o d ifica tio n s to the source/sampler in te rfa c e have improved the s t a b ilit y
o f the plasma and have also reduced the le v e l o f N0+ found in the plasma.
Source M o d ifications
A ta n g e n tia l sheath flo w o f Ng was introduced by the co n stru ctio n of
a demountable to rch to reduce the entrainment o f a ir in the plasma plume
(Figure 18, not to s c a le ).
The torch consisted o f an 8.25 cm quartz tube
(1.1 cm O.D. x O.S cm I. B .) w ith a ta n g e n tia l sheath flow in l e t , and 3
ind entations 2.5 cm from the fr o n t end o f the torch which are used fo r
centering a 10.2 cm long A^Og plasma tube (0.8 cm O.D. x 0.5 cm I. D . ) .
The quartz and AlgOg tubes were adjoined a t the rear o f the torch w ith a
Teflon base.
A quartz bonnet was added to the sampler to fu rth e r reduce entrained
atmospheric gases.
rin g .
The bonnet was mounted over the sampler re ta in in g
A series o f d iffe r e n t bonnet heights ranging from 1.32 cm to 0.89
cm enabled in v e s tig a tio n o f several sampling distances where the bonnet
was in contact w ith the c a v ity face and the face o f the vacuum chamber.
This generated an area o f elevated temperature w ith a p o s itiv e pressure
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
e
s
9
©
O
X
Fi g.
18.
He
pl asma
- s a mp l i n g
Interface
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
r e la tiv e to atmospheric which g re a tly reduced the in flu x o f atmospheric
gases in to the plasma.
The m a jo rity o f the plasma/sheath gases is not
sampled and e ith e r pass around the bonnet or back through the microwave
c a v ity .
No arcing in the c a v ity due to th is m ixture o f excess gases was
observed when the bonnet was in place.
C e rtifie d custom gas mixtures (Matheson Gas Products) o f 8 ppm CHgBr
and CHgCl in u lt r a
high p u r it y
(UHP) He and 88 ppm CH^I in UHP He
provided a continuous source o f halogen ions fo r th is in v e s tig a tio n .
flows and d ilu tio n
Gas
o f the above standards are regulated by using mass
flow c o n tro lle rs (T yla n ).
Lens p o te n tia ls were set at the compromise voltages lis te d in Table
3.
In order to optim ize the count ra te fo r a selected mass, only the
stop and the center o f the energy analyzer were adjusted.
Results and Discussion
P
i njnuri, L
a
i in
1j.0
ed
hn
/w
v »n
i ed G
a cj iin n liov. j
cv
^.U
P m rt-f
demonstrates the e ffe c t
14 N16 0+
eolortoH
j\. iw
w
v.u
maec
m
G
d
d
rannpc
i G
m
^
C
O
artH
G
iiu
of the a d d itio n o f the quartz bonnet to the
sample cone re ta in in g rin g .
p r in c ip a l
+<
K
r»
ov
o
1
1\
-
The spectra in Figure 19A in d ic a te th a t the
ion s in th e He plasma b e fo re a d d itio n o f th e bonnet were
and ^ 02 + .
(The p rin c ip a l ion is probably 4 He+ , however, th is
quadrupole does not operate w ell below 10 amu.
Below 10 amu, re s o lu tio n
is
These ions
poor and th e
c a lib r a t io n
is
s u s p e c t.)
are due to
entrainment o f a ir in the He plasma plume (a fte rg low ) p r io r to or during
the sampling process.
Table 9 l is t s a d d itio n a l background ions present
in the unshielded He plasma.
used in th is scan.
Reagent grade He (99.995%) was the sole gas
The plasma power was 360 W, w ith a 1.0 L/min He flo w
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
Figure 19. P rin c ip a l ions sampled from a 270 W He
microwave induced plasma, (A) unshielded plasma,
1.0 L/min He plasma gas, (B) shielded plasma,
1.0 L/min He plasma gas, 0.83 L/min Ng sheath
gas, quartz bonnet, sampling distance = 3.4 mm.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
r' ■■
-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
Table 9 .— Background ions present in unshielded He plasma1
m/Z
Probable Identity
14
JV
16
160 +
17
160H+
18
16OH2+
19
16° H3
28
29
30
31
14
+
N2
14
+
N2H
14N160 +
14
N
1P
OH
32
160 *
33
10° 2H"
40
40Ar+
42
+
14«, +
«3
44
12C160 2+
45
12C160 2H+
46
14N160 2+
54
40Aar
„ +
N , 54Fe
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
Table 9 .—Background ions present in unshielded He plasma1— Continued
m/Z
Probable Identity
56
40A r16O+, 56Fe+
58
40A r16OH *
68
40A r14N2+
80
40A r40Ar+ , 80Kr+
82
82K r+
83
83Kr+
84
84K r +
86
86Kr+
1 He plasma, 360 VJ; 1.0 L/m He (99.995%); sampling distance = 5 mm.
V.
~
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
ra te and a sampling distance o f 5 mm.
The plasma background is ric h in
molecular N, 0 and Ar containing species.
Also evident are isotopes o f
K r, a contaminant in the c y lin d e r o f He gas.
The combination o f the
bonnet and the nitrogen sheath gas e ffe c tiv e ly shielded the plasma from
atmospheric gases and increased the apparent io n iz a tio n p o te n tia l o f the
plasma source above th a t o f NO which is approximately 9.25eV.
Neither
the sheath gas nor the bonnet, i f used s e p a ra te ly, su cce ssfu lly reduced
NQ+ form ation.
The p rin c ip a l
ions in the shielded plasma (Figure 19B)
14 + 14 +
14 +
N ,
N2 and
N^ which o rig in a te from the N2 used as the sheath
are
gas.
He and Ar could not be used as sheath gases because a plasma was
sustained
in
the
quartz
d e te rio ra tio n o f the to rch
sheathgas
containment
tube
and
caused
assembly. N2was probably more e ffe c tiv e than
He a t reducing entrained a ir because o f i t s
higher mass.
The re la tiv e
peak heights o f N2+ and N0+ are shown in Figure 19B w ith the to rch bonnet
combination.
Since ^ N 2+ is the p rin c ip a l ion in the mass spectrum w ith
the io n iz a tio n p o te n tia l o f ^ N 2 being 15.6 eV, greater s e n s it iv ity is
observed f o r B r,
Cl
andI as p o s it iv e
p o te n tia ls o f 11.84, 13.01
Table 10 id e n t i f i e s
io n s w ith
fir s t
io n iz a tio n
and 10.45eV, re s p e c tiv e ly .
species p re s e n t in th e background s p e c tra
obtained fo r u ltr a high p u rity (UHP) He in the shielded plasma.
20
Figure
illu s tr a te s the re la tiv e in te n s itie s o f the major and minor background
ions in the shielded plasma.
m u ltip lie d by a fa c to r o f 50.
sampler erosion.
In the lower scan, the response has been
Fe ions can probably be a ttrib u te d to
The use o f a n itro ge n sheath gas reduced oxygen and
argon containing molecular species to
16 0 2+
a le ve l
where only ^ N ^ 0 + and
were detected in the scanning mode using analog data a c q u is itio n .
Gas flows during a c q u is itio n o f these data were 0.83 L/min o f N2 sheath
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 10.—Background ions in shielded He/N
m/Z
Probable Identity
2
plasma*
Interferes With (abundance)
14
14N+
N (99.63)
16
160 +
O (99.759)
28
14N ^
Si (92.21)
30
14N160 +
Si (3.09)
32
160 2+
S (95.02)
40
40Ar+
A r (9 9 .6 ),
42
14N3+
Ca (0.64)
54
54Fe+
Fe (5 .8 2 ),
56
56Fe+
Fe (91.66)
* UHP He, 1.0 L/mim 0.83 L/min
distance = 3.4 mm.
Ca (97)
Cr (2.38)
; 270 W plasma; sampling
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 20.
R elative in te n s itie s o f the major and
minor background ions in the shielded
plasma; lower scan response m u ltip lie d
by fa c to r o f 50.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
84
"t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
gas and 1.0 L/min o f He plasma gas.
I n i t i a l detection o f Br+ was accomplished w hile intro d u cin g 1 L/min
o f UHP He containing 8 ppm (mole to mole) o f CHgBr and CHgCl.
sheath flo w was 0.83 L/m in.
The shielded plasma appeared transparent in
the v is ib le wavelength region w ith no e le c tric a l
between the sampler and the plasma.
had
not
previously
e le c tr ic a l
been
in te ra c tio n
obtained
(d ischarge).
in te ra c tio n
observed
A s ig n ific a n t analyte ion count rate
on
th is
instrument
However, in th is
w ithout
quartz bonnet.
«L
Br .
and the
The sampling d ista n ce , defined as the distance between the face
o f the c a v ity and the t i p
Q1
th is
case, a 1.32 cm
quartz bonnet formed a chamber between the microwave c a v ity
sampler.
The ^
o f the sampler, was 5.9 mm when using th is
The most abundant ions in the spectra were
79 +
Br
and
However, at th is sampling distance, s e n s itiv ity fo r ch lo rin e ions
OC
was poor.
The
0 7
Cl
and
.j.
Cl
count rates were increased by reducing the
plasma sampling distance u n til a d iffu s e e le c tric a l discharge was v is ib le
in the center o f a He plasma plume, which was attached to the sampler
o r if ic e . This required replacement o f the previous quartz bonnet w ith one
which perm itted an optimum sampling distance o f 3.4 mm.
count rates also improved at th is d istance.
e le c t r ic a l
Bromine ion
I f the in te n s ity o f the
in t e r a c t io n were incre a se d by a f u r th e r decrease in th e
sampling distance, an increase in power, or an increase in the He flow
r a te , B r+ and C l+ peaks were broad and t a i l i n g ,
r e s o lu tio n
r e s u lt in g
as the quadrupole scanned from high to
in poor
low mass.
The
e le c tr ic a l discharge probably resulted in a large ion energy spread in
the plasma source which degraded the spectral
re s o lu tio n
(6 5 ).
Peak
shapes could not be improved by v a ry in g th e e le c t r o s t a t ic ion lens
p o te n tia ls .
Figure 21 shows a comparison of spectra obtained fo r Br+
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 21.
79
+
Comparison o f spectra obtained fo r
Br and
81
Br during intense and d iffu s e e le c tr ic a l
in te ra c tio n between the He plasma and the
sampler.
Current
response
76
84
84
m /
Z
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
isotopes w ith an intense e le c tr ic a l discharge and a d iffu s e discharge.
The o r if ic e o f the sampler became enlarged and p itte d due to s p u tte rin g
o f the s ta in le s s steel surface during the intense discharge.
Work done
w ith another sampler/skimmer combination (1.0 mm sampler/0.65 mm skimmer
demonstrated
e le c tr ic a l
e xce lle nt
re s o lu tio n
as
w ell
as
a
reduction
in te ra c tio n between the sampler and the plasma.
widths at h a lf height were on the order o f f 0.5 amu.
in
the
The peak
However, w ith the
la rg e r diameter sampler, there was an increase in the amount o f N0+
present in the mass spectrum.
of the plasma.
This was due to sampling a la rg e r p o rtio n
Because of the large increase in the background N0+, the
0.7 mm diameter sampler was used fo r a ll a d d itio n a l experiments.
The background ions obtained w hile intro du cin g 1.0 L/min o f UHP He
containing 8 ppm (mole to mole) o f CHgCl and CHgBr are lis te d in Table
11.
The
a d d itio n a l
background
peaks
lis te d
halogen-containing species and are possibly due to
nitrogen a t some point in the sampling process.
isotopes o f Br and Cl are c le a rly v is ib le
shows a sin g le
in
the
ta b le
are
recombination w ith
The atomic and molecular
in the spectra.
analog scan o f CHgBr and CHgCl in He.
Figure 22
R ela tive peak
heights fo r ^ C l + , ^ B r +, ^ C 1 ^ N + and ^ C 1 ^ N + are also v is ib le in the
spectra.
Assignments were not made fo r m/z values o f 69 and 71, the
ra tio o f which appeared to in d ic a te a Cl containing molecular species.
However, th is r a tio could also in d ic a te the presence o f Ga, a contaminant
in A1, which may o rig in a te from the AlgOg discharge tube.
Iodine was determined by intro du cin g 12.7 mL/min o f a separate gas
m ixture containing 88 ppm o f CHgl (mole to mole) in UHP He to a plasma
support flow o f 0.91 L/min.
Fluorine was introduced as 1 ppm CHgF in UHP
1 Q
He.
Attempts to detect F as
X
F
were not successful.
This is not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
Table 11 .—Shielded He plasma* with 8 ppm CH^Br, CH^Cl
m/Z
Probable Identity
12
12C+
14
2
V
16
160 +
24
I 2 C;
28
14N2+
30
14N160 +
32
32°2
35
35Cl+
37
3?Cl+
An
40
40 .
42
1V
49
35Cl14N+
51
37Cl14N+
54
r A
54j-,+
56
56Fe+
Ar +
Fe
69
71
?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
Table 11.
•Shielded He plasma* with 8 ppm CH^Br, CH^Cl— Continued
m/Z
Probable Identity
79
79Br+
81
81Br+
93
79B r14N+
95
81 Br14N+
* UHP He, 1.0 L/min; 0.83 L/min N ; 270 IV plasma; sampling
distance = 3.4 mm.
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
su rp ris in g since the f i r s t io n iz a tio n p o te n tia l o f F is 17.42 eV, w ell
above th a t o f Ng which is used as the sheath gas.
Plasma parameters which have the g reatest a ffe c t on the analyte
signal are sampling depth, plasma power and flo w rates of He and N2.
As
in the previous Ar MIP-MS work (Chapter 3 ), the zone in the He plasma o f
greatest a n a ly tic a l u t i l i t y
is small so th a t, fo r example, movement of
the plasma away from the sampling o r if ic e along the system axis by only 1
mm from an optim ized sampling p o s itio n may re s u lt in a 90% reduction in
ion in te n s ity .
The signal may be recovered by increasing the flo w and/or
increasing the power, both o f which re s u lt in extension o f the plasma
plume.
F ig u re
23 is
a p lo t o f lin e a r in t e n s it y
obtained, in the analog mode) observed fo r
( a r b it r a r y
u n its
127 +
I versus the He flo w ra te .
The sampling distance was 3.4 mm and the plasma power was 270 W.
The
sharpness o f the curve illu s tr a te s the importance o f a stable source.
s im ila r
curve
flu c tu a tio n s
is
in
obtained
by
gas flows or in
varying
the
plasma
generator s t a b ilit y
power.
A
Small
which a ffe c t the
plasma power w ill have a large e ffe c t on the ion count ra te .
Detection lim its fo r Br+, Cl+ and I + were obtained w hile m onitoring
a sin g le ion in the pulse counting mode using a sampling distance o f 3.4
mm.
This was the optimum sampling distance fo r m onitoring a ll o f the
above io n s .
An e lectron m u ltip lie r voltage o f -2400 V y ie ld e d count
rates o f approximately 1 x 10^ counts s~* fo r Br and I , w ith the Cl count
rates approximately an order o f magnitude low er.
The 8 ppm CHgBr/CHgCl
in He was d ilu te d by m ix in g w ith UHP He, re d ucin g i t s
mL/min.
flo w to 83.8
The 88 ppm CHgl was introduced a t a ra te o f 12.7 mL/min.
plasma gas flo w (standard and UHP He was 1.04 L/m in.
Total
Detection lim its
were based on the mean o f te n , 10 -s counting periods of the sample and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
127
+
12,000
270 WHe Plasma
8 x 5 mm Al?0o tube
Sampling distance = 3.4 mm
EM = 2000 V
CHqI = 12.7 mL/min (88 ppm)
10,000
Linear
Intensity
8,000
6,000
4,000
2,000
0.4
0.5
0.6
0.7
He
Fig.
flow
2 3. E f f e c t
of
0.8
0.9
1.0
(L/mln)
He f l o w
on i o n c o u n t
rate.
93
background.
Background counts were acquired w ith the quadrupole set at
an adjacent mass because o f the length o f time required to elim inate
CHgBr/CHgCl/CHjI memory e ffe c ts
experiments w il l
incorporate
from the
s ta in le ss
Teflon
steel
count rates decreased at higher m/z values.
tu b in g .
tu b in g .
Subsequent
The background
Background count rates and
d etection lim its determined a t three times the standard de via tio n o f the
background are reported
o p tic a l
emission
in
(86)
Table 12.
and
recent
These values are compared w ith
ICP/MS
(84)
detection
le v e ls .
P relim inary MIP-MS d etection lim its fo r B r, Cl and I compare w ell w ith
those reported by He plasm a/optical emission spectrometry and are b e tte r
than the detection lim its
These r e s u lts
reported by Ar plasma/ICP-MS fo r Br and C l.
suggests a h ig h p o te n tia l
f o r elem ent s e le c tiv e gas
chromatographic d etection o f halogen containing species at u ltr a tra ce
le v e ls .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
Table 12.—Detection limits and background count rates obtained in
single ion mode*
Background
MIP/MS
( Pd/s)
Element
Isotope
( counts s
Br
79B r+
406 + 10
Cl
35Cl+
495 + 29
I
r
127j+
19„+
r
72 ± 6
)
1.2
21
1.8
MIP/OES2
ICP/MS3
( pa/ s)
( PQ/.S.)
62
100
40
40,000
—
1
•1
1
~ Detection limits calculated from the mean of 1G X IG-s counting
periods of the standard and of the background; based on three times
the standard deviation of the background at an adjacent mass.
2
3
Reference 89.
Reference 88.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
CHAPTER 5 SUMMARY AND FUTURE WORK
The b a s ic o b je c tiv e s
system/molecular
pressure
beam
sampling
of
o f th is
sampling
ions
p r o je c t ,
in te rfa c e ,
generated
in
induced plasma, have been accomplished.
Ar-MIP w ith d ire c t s o lu tio n
w ith
changes in
which
enables
atmospheric
the moderate power microwave
However, a t th is
n e b u liza tio n
source fo r elemental a n a ly s is .
th e design o f a vacuum
is
p o in t, the
not recommended as an ion
Because the response is so sharply peaked
sam pling d is ta n c e
o r plasma gas flo w
d i f f i c u l t to m aintain a sta b le analyte response.
ra te ,
it
is
A lso, plasma s t a b ilit y
is poor and the sample m a trix has a large e ffe c t on the ion count ra te .
W ith no c o n tro l over how th e sample is
in je c te d in th e plasm a, the
optimum lo c a tio n fo r ion sampling is a t the periphery o f the plasma where
th e
therm al te m p e ratu re is low er and m o le c u la r oxid e fo rm a tio n
favored.
On the other hand, the helium plasma source appears to
is
be fa r
more prom ising.
The combination o f a N£ sheath gas w ith a quartz bonnet
has e ffe c tiv e ly
decreased the
re s u ltin g
in
g re a tly
amount o f
improved s e n s it iv ity
compared w ith the unshielded plasma.
entrained
a ir
in
the
plasma
fo r B r, Cl and I ions when
Of p a r tic u la r importance is the
re la tiv e s im p lic ity o f the dry He background spectra when compared to the
Ar plasma w ith s o lu tio n n e b u liz a tio n .
Future
work
w ith
the Ar-MIP
w ill
involve
desolvation
o f the
nebulized sample, w ith in je c tio n o f the dry sample in to the center o f the
d is c h a rg e .
D e s o lv a tio n should im prove io n iz a tio n in th e plasma by
reducing solvent loading.
To prevent the form ation o f plasma sp in d les, a
m ixture o f He and Ar as plasma gases may be used.
A d d itio n a l work w ith the He-MIP w ill involve a study o f the e ffe c t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
o f discharge tube diameter on s e n s itiv ity
o f Br, Cl and I .
Further
attempts to determine F ions must u t iliz e He as a sheath gas to increase
th e apparent io n iz a tio n p o te n tia l o f th e plasma source.
T h is w i l l
require m o d ific a tio n o f the present to rch to prevent an a d d itio n a l plasma
from being sustained between concentric tubes o f the torch which leads to
rapid to rch degradation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
BIBLIOGRAPHY
1.
Bram, L. L ., P h illip s , R. S ., Dickey, N. H. Funk and Wagnalls New
Encyclopedia, 1979, V ol. 5, p. 402.
2.
H ie ftje , G. M., Anal. Chem., 1985, 57, 256A.
3.
Gray,
A. L ., Anal. Chem.,
4.
Gray,
A. L ., Dynamic
Mass Spectrom., 1976, 4,
153.
5.
Gray,
A. L ., Dynamic
Mass Spectrom., 1979, 5,
106.
6.
Houk,
R. S ., Thesis,
Iowa State U n iv e rs ity , Ames, Iowa, 1980,
1975, 47, 600.
I S - I -98
7.
Houk, R. S ., Fassel, V. A ., Flesch, G. D ., Svec, H. J . , Gray, A. L .,
T a ylo r, C. E ., Anal. Chem. 1980, 52, 2283.
8.
Date, A. R ., Gray, A. L ., A nalyst, 1981, 106, 1255.
9.
Houk, R. S. Fassel, V. A ., Svec, H. J . , Dynamic Mass Spectrom.,
1981, 6 , 234.
10.
Gray, A. L ., Date, A. R ., Dynamic Mass Spectrom.,
1981, 6 , 252.
11.
Douglas, D. J . , French, J. B ., 1981, 53, 37.
12.
Douglas, D. J . , Quan, E. S. K ., Smith, R. G.,Spectrochim.
A c ta .,
1983, 38B, 39.
13.
Houk, R. S ., Fassel, V. A ., Svec, H. J . , Org. Mass Spectrom., 1982,
5, 240.
14.
Houk, R. S ., Thompson, J . J . , Amer. M ine ra l, 1982, 67, 238.
15.
Schulten, H. R ., Monkhouse, P. B ., Amer. Lab., 1983, March, 44.
16.
Lehmann, W. D ., Schulten, H. R ., Anal. Chem., 1977, 49, 1744.
17.
Schulten, H. R ., Monkhouse, P. B ., M u lle r, R ., Anal. Chem., 1982,
54, 654.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
18.
Beske, H. E ., In t. J . Mass Spectrom. Ion Phys., 1982, 45, 173.
19.
Koppenaal, D. W., L e tt, R. G., Brown, F. R. Manahan, S. E ., Anal.
Chem., 1980, 52, 44.
20.
Knab, H. J . , Hintenberger, H ., Anal. Chem., 1980, 52, 390.
21.
Date, A. R ., Gray, A.
L ., Analyst, 1983, 108, 159.
22.
Gray, A. L ., Date, A.
R ., A nalyst, 1983, 108, 1033.
23.
Date, A. R ., Gray, A.
L ., Spectrochim. A cta ., 1983, 38B,
24.
Houk, R. S ., Thompson, J . J . , Biomed. Mass Spectrom.,1983, 10, 107.
25.
Gray, A. L ., Date, A. R ., I n t . J . Mass Spectrom. Ion Phys., 1983,
29.
46, 7.
26.
Date, A. R ., Gray, A. L ., I n t . J . Mass Spectrom. Ion Phys., 1983,
48, 357.
27.
Date,
A.
R ., Trends Anal. Chem., 1983, 2, 225.
28.
Houk,
R.
S .. submitted fo r p u b lic a tio n in PureAppl.
29.
Chem.,1985.
Douglas, D. J . , Houk, R. S ., Prog. Anal. Atomic S pectrosc., 1985, 8 ,
1.
30.
Gray, A. L ., Spectrochim. A c ta ., 1985, 40B, 1525.
31.
Houk, R. S ., Anal. Chem., 1986, 58, 97A.
32.
O liv a re s , J . A .,
Houk, R. S .,
Anal. Chem., 1985, 57, 2674.
33.
O livare s, J. A .,
Houk, R. S .,
Anal. Chem., 1986,
34.
O liva re s, J . A .,
Houk, R. S .,
Appl. S pectrosc., 1985, 39, 1070.
35.
H o rlik , G., Tan, S. H ., Vaughan, M. A ., Rose, C. A ., Spectrochim.
58, 20.
A c ta ., 1985, 40B, 1555.
36.
Zhu, G., Browner, R. F ., submitted to A ppl. S pectrosc., 1986.
37.
Vaughan,
38.
Tan, S. H ., H o rlik , G ., Appl. S pectrosc., 1986, 40,
39.
Gray, A. L ., Spectrochim. A c ta ., 1986, 41B, 151.
M. A ., H o rlik , G., Appl. Spectrosc., 1986,
40, 434.
445.
jr. •
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
40.
McLeod, C. W., Date, A. R ., Cheung, Y. Y ., Spectrochim. A c ta ., 1986,
41B, 169.
41.
Boomer, D. W., Powell, M., Sing, R. L. A ., S a lin , E. D ., Anal.
Chem., 58, 975.
42.
Ting, B. T. G., Janghorbani, M., Anal. Chem., 1986, 58, 1334.
43.
Thompson, J . J ., Houk, R. S ., Anal. Chem., 1986, 58, 2541.
44.
Date, A. R ., Hutchison, D ., Spectrochim. A c ta ., 1986, 41B, 175.
45.
McLaren, J. Ws, M ykytuik, A. P ., W illie , S. N., Berman, S. S ., Anal.
Chem., 1985, 57, 2907.
46.
P ic k fo rd , C. J . , Brown, R. M., Spectrochim. A cta ., 1986, 41B, 183.
47.
Barnes, R. M., Anal. Chem., 1983,
48.
Galan, L ., Anal. Chem., 1986, 58,
49.
Beenakker, C. I . M., Spectrochim. A c ta ., 1977, 32B, 173.
50.
Beenakker, C. I . M., Spectrochim. A c ta ., 1976, 31B, 483.
51.
Goode, S. R ., Baughman, K. W., Appl. Spectrosc., 1984, 38, 755.
55, 611A.
697A.
52. Dahmen, J . , ICP Inform . News, 1985, 11.
53. BoIIo-Kamara, A ., Codding, E. G., Spectrochim. A cta ., 1981, 36B,
973.
54.
Haas, D. L ., Carnahan, J. W., Caruso, J . A ., Appl. S pectrosc., 1983,
37, 82.
55.
Haas, D. L ., Caruso, <3. A ., Anal. Chem., 1984, 56, 2014.
56.
Brown, P. G., Workman,
J . M., Haas, D. L ., F le itz , P. A ., M ille r , D.
C ., S e lis k a r, C. J . , Caruso, J. A ., Appl. Spectrosc., 1986, 40,477.
57.
Browner, R. F ., Boorn,
A. W., Anal. Chem., 1984, 56, 786A.
58.
Browner, R. F ., Boorn,
A. W., Anal. Chem., 1984, 56, 875A.
59.
Layman, L. R., L ic h te , F ., Anal. Chem., 1982, 54, 638.
60.
Meinhard, J . E ., "A pplications o f Plasma Emission Spectrochem istry,"
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
Heyden, P h ila d e lp h ia , 1979, 1.
61.
Meinhard, J . E ., ICP Inform . News, 1976, 2, 163.
62.
Brown, G. R ., P e r te l, R ., I n t . J . Mass Spectrom. Ion Phys., 1975,
16, 53.
63.
Serauskas, R. V ., Brown, G. R ., P e rte l, R ., I n t . J . Mass Spectrom.
Ion Phys., 1975, 16, 69.
64.
P e rte l, R ., I n t . J . Mass Spectrom. Ion Phys., 1975, 16, 39.
65.
Bentz, B. L ., Bruhn, C. G., H arrison, W. W., I n t . J. Mass Spectrom.
Ion Phys., 1978, 28, 409.
66.
Dankert, C ., Legge, H. presented as Paper 173, Tw elfth I n t . Sym. on
Rarefied Gas Dynamics, C h a rlo tte s v ille , Va., 1980, 882.
67.
Campargue, R ., J . Phys. Chem., 1984, 88, 4466.
68.
Gudzinowicz, "Fundamentals o f Integrated GCMS," Part 3,
Chromatographic Science S e rie s, V ol. 7, Marcel Decker, N. Y ., 1977,
Chapter 7, 1.
69.
"Vacuum Technology:
I t ' s Foundations, Formulae and T a b le s," Product
and Vacuum Technology Reference Book, Leybold-Heraeus, In c ., 1986.
70.
B arrington, A. E ., "High Vacuum Engineering," P re n tice -H a l1, N. J . ,
1963, 47.
71.
Peterson, J . F ., S teinherz, H. A ., S o lid State Technology, Dec,
1981.
72.
Zander, A. T ., A nal. Chem. 1986, 58, 1139A.
73.
F ite , W. L ., "Ion O p tic s ," paper presented at 31st Annual Conference
on Mass Spectrometry and A llie d Topics, 1983.
74.
Kumar, S. V. K ., Siddagangappa, M. C ., Venkatasubramanian, V. S ., J.
Phys. E .: S c i. In s tru m ., 1982, 15.
75.
Aston, G., W ilbu r, P. J . , J . Appl. Phys., 1981, 52, 2614.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
76.
Read, F.
H ., J . Phys. E .: S c i. Instrum. 1971, 4, 562.
77.
Read, F.
H ., J . S c i. Instrum .: J . Phys. E ., 1969, 2, 679.
78.
Adams, A ., Read, F. H ., J . Phys, E.: S ci. In stru m ., 1972, 5, 150.
79.
Read, F. H ., J . Phys. E .: S ci. Instrum ., 1970, 3, 127.
80.
Adams, A ., Read, F. H ., J . Phys. E.: S c i. In stru m ., 1972, 5, 156.
81.
M artinez, G., Sancho, M., J . Phys. E.: S ci. In stru m ., 1983, 16, 626,
82.
McDowell, C. A ., "Mass Spectrometry," K rieg e r, New York, 1979.
83.
Kurz, E.
84.
Herman, R. and Alkemade, C. T. 0 ., "Chemical Analysis by Flame
A ., Amer. Lab., 1979, March issue.
Photometry," V ol. XIV, Chemical Analysis Series, In te rs c ie n c e , N.
Y ., 1963, 407.
85.
Chen, F. F ., "An In tro d u c tio n to Plasma Physics and C ontrolled
Fusion, Vol. 1, Plenum Press, N. Y., 1984.
86.
87.
Workman, J . , personal communication.
"CRC Handbook o f Chemistry and Physics," 53rd Ed., 1972-1973T, E-56
88. Chong, N. S. and Houk, R. S ., Appl. Spectrosc., in press.
89. Bruce, M. L. and Caruso, J. A ., Appl. Spectrosc., 1985, 39,
942.
90.
Haas, D. L. and Caruso, J . A ., Anal. Chem., 1985, 57, 846.
91.
Mohamad, A. H. and Caruso, J . A ., "Advances in Chromatography," Vol
26, Ch. 5, Marcel Dekker, N. Y ., 1986.
92.
Chan, S. C ., Van Hoven, R. L ., and Montaser, A ., Anal. Chem., 58,
2342 (1986).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
3 016 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа