close

Вход

Забыли?

вход по аккаунту

?

ANALYTICAL BEHAVIOR OF HALOGENATED COMPOUNDS WITH MICROWAVE INDUCED PLASMA EMISSION SPECTROMETRY (CHLORINE, BROMINE, TRIS DIBROMOPROPYL PHOSPHATE)

код для вставкиСкачать
INFORMATION TO USERS
This reproduction was made from a copy of a document sent to us for microfilming.
While the most advanced technology has been used to photograph and reproduce
tills document, the quality of the reproduction is heavily dependent upon the
quality of the material submitted.
The following explanation of techniques is provided to help clarify markings or
notations which may appear on this reproduction.
1.The sign or “target” for pages apparently lacking from the document
photographed is “Missing Page(s)”. If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages. This
may have necessitated cutting through an image and duplicating adjacent pages
to assure complete continuity.
2. When an image on the film is obliterated with a round black mark, it is an
indication of either blurred copy because of movement during exposure,
duplicate copy, or copyrighted materials that should not have been filmed. For
blurred pages, a good image of the page can be found in the adjacent frame. If
copyrighted materials were deleted, a target note will appear listing the pages in
the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photographed,
a definite method of “sectioning” the material has been followed. It is
customary to begin filming at the upper left hand comer of a large sheet and to
continue from left to right in equal sections with small overlaps. If necessary,
sectioning is continued again—beginning below the first row and continuing on
until complete.
4. For illustrations that cannot be satisfactorily reproduced by xerographic
means, photographic prints can be purchased at additional cost and inserted
into your xerographic copy. These prints are available upon request from the
Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases the best
available copy has been filmed.
University
Micrajlms
International
300 N. Zeeb Road
Ann Arbor, Ml 48106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8313491
Carnahan, Jon Winston
ANALYTICAL BEHAVIOR OF HALOGENATED COMPOUNDS WITH
MICROWAVE INDUCED PLASMA EMISSION SPECTROMETRY
PH.D. 1983
University o f Cincinnati
University
Microfilms
International
300 N. Zeeb Road, Ann Arbor, MI 48106
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 here with a check mark V .
1.
Glossy photographs or 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 spins______
10.
Computer printout pages with indistinct print.
11.
Page(s)___________ lacking when material received, and not available from school or
author.
12.
Page(s)___________ seem to be missing in numbering only as text follows.
13.
Two pages numbered___________. Text follows.
14.
15.
iS
iX'
Curling and wrinkled pages______
Other__________________________________________________________________
University
Microfilms
International
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ANALYTICAL BEHAVIOR OF HALOGENATED COMPOUNDS
WITH MICROWAVE INDUCED PLASMA EMISSION SPECTROMETRY
A dissertation subm itted to the
Division of Graduate Studies and Research
of the University of Cincinnati
in partial fulfillm ent of the
requirem ents for the degree of
DOCTOR OF PHILOSOPHY
in th e D epartm ent of Chem istry
of the College of A rts and Sciences
1983
by
3on W. Carnahan
B.S., Southern Illinois University a t Carbondale
M.S., Southern Illinois University a t Carbondale
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF CINCINNATI
March 14,
qg 83
I hereby recommend that the thesis prepared under my
supervision hy
entitled
Jon w -
Carnahan_______________
A nalytical Behavior o f H alogenated Compounds w ith Microwave
_______________Induced Plasm a Emission Spectrom etry_________________________
be accepted as fu lfillin g this p a rt of the requirements for the
degree o f
D octor of Philosophy________________________________________
Approved by:
o>.
i
*
.L
I
/)
/
Form 668—G rad. School—HA—1 2 -6 9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT OF DISSERTATION
Studies dealing with determ ination of halogens using the atm ospheric
pressure helium microwave induced plasma (MIP) a re presented.
Two systems
are constructed to extend th e application of this system and enhance minimum
detectable levels of analyte.
The first investigation entails a rapid method for the determ ination of
halogens. E lectrotherm al vaporization is used as th e sample introduction mode.
System optim ization yielded a detection lim it of 8 ng for both chlorine and
bromine. Calibration curves are present and applications to real sample analysis
is made.
Matrix e ffe cts were prominant with complex samples.
The fire
retard an t tris(2,3 dibromopropyl) phosphate is ex tracted from garm ents and
determ ined.
The second portion of this work deals with th e developm ent of a m oderate
power (200 to 500 W) MIP system and its use as a d etec to r for gas chrom atogra­
phy.
Initial investigations deal with dissipation of h eat generated within th e
system. Upon development this system was used as a halogen selective d etecto r
for chromatography.
O ptim ization gave rise to d etection lim its comparable to
lite ra tu re values. Studies of signal dependence on flow rates, power applied and
scavenger gases are presented.
Minimum d etectab le amounts are reduced by
utilization of a background corrected system.
i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWKEDGEMENTS
To th e group of emission specialists I owe a g re a t debt for my sanity. I
would like to especially thank "Scoop" Brueggemeyer, "Positive Feedback"
Eckhoff, Z.M. Mantay, "P.M." M cCarthy, M. Mesmer, "Wild Irish" Mulligan and
M. Zerezghi for th eir insights as well as th eir wine and song.
For th e ir technical assistance and high level of patience, g reat apprecia­
tion is extended to Ed Younginger, Dr. T.H. Ridgeway and others of th e "Silicon
Valley Speak" fram e of mind on th e second floor.
For th e suggestion of th e
research problem and encouragem ent while following th e dusty tra il, I direct
gratitude and respect to my landlord. Also deserving ,a hearty p a t on th e back
for their helpfulness and friendship are Judy and Dr. Joseph A. Caruso.
Thanks go to Dr. T.W. G ilbert and Dr. D.H. McDaniel fo r serving on my
com m ittee.
The N ational In stitu te for Occupational Safety and H ealth, the
U niversity R esearch Council and th e G raduate School a re acknowledged for th eir
assistance during this endeavor.
My m other and fath er deserve a g reat deal of cred it fo r th eir support and
encouragem ent throughout the years. Thanks to my wife's parents for th eir help
and understanding. Above all, I would like to thank my im m ediate fam ily. My
wive, Nancy, made th e good tim es b e tte r and th e frustrations more tolerable
with her love, understanding and ability to see th e fo rest in th e th ick et of life.
Lastly, I would like to thank Betsy and Leslie for providing fulfillm ent and joy to
Nancy and I.
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
PAGE
I.
II.
III.
INTRODUCTION
A.
The Present Status of Selective D etection for Halogens
B.
History and Development of Elem ental Emission Spectrom e­
try as an A nalytical Technique for Halogen Containing Sub­
stances
C.
Purpose of D issertation
1
13
DETERMINATION OF HALOGENS BY ELECTROTHERMAL
VAPORIZATION INTO A HELIUM MICROWAVE INDUCED
PLASMA AT ATMOSPHERIC PRESSURE
A.
Introduction
14
B.
Experim ental
15
C.
R esults and Discussion
20
D.
Conclusion
36
REQUIREMENTS FOR MAINTENANCE OF A MODERATE
POWER MICROWAVE-INDUCED PLASMA
A.
Introduction
39
B.
Experim ental and Instrum entation
40
C.
Results and Discussion
41
D.
Conclusion
47
IV. ' MODERATE POWER MICROWAVE INDUCED PLASMA GAS
CHROMATOGRAPHIC DETECTOR FOR HALOGENS
A.
Introduction
51
B.
Experim ental
52
C.
Results and Discussion
61
D.
Conclusion
91
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PAGE
V.
SUGGESTIONS FOR FUTURE WORK
97
VI.
LITERATURE CITED
98
VII.
APPENDIX
103
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
FIGURE NO.
PAGE
1-1
Energy Levels of Various Species.
I-2
Q ualitative Depiction of the Plasma.
10
II-l
ETV/MIP System Diagram.
18
II—
2
Signal for 1 Ug Citex BC-26.
22
II-3
Dependence of Vaporization Voltage on Peak Height.
25
II-4
Dependence of Flow R ate on Peak Height.
27
II—
5
ETV/MIP Peak Shape.
29
II-6
C itex BC-26 Calibration Curves.
33
II-7
Response of Various Viscous Solvents.
35
III-l
Dependence of Power Applied upon Lead Peak Height
with E lectrotherm al Vaporization.
43
III-2
Dependence of Stub Tuner Tem perature upon Lead Signal.
46
III-3
Lead Signal with Q uartz and Alumina Containm ent Tubes.
49
IV-1
Moderate Power GC/MIP Interface.
54
IV-2
Overall System Diagram for GC/MIP.
56
IV-3
Pyrex Cooling Device.
60
IV-4
Dependence of Power Applied upon Normalized Signal and
Background for Chlorine, Bromine and Lead.
66
Dependence of Power Applied upon Normalized Signal and
Background for Chlorine, Bromine and Lead.
67
Dependence of Power Applied upon Normalized Signal and
Background for Chlorine, Bromine and Lead.
68
IV-5
IV-6
v
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
7
FIGURE NO.
IV-7
PAGE
Dependence of Flow R ate upon Normalized Signal for
Chlorine, Bromine and Lead.
71
C alibration Curves for Chlorine using Axial and L ateral
Viewing of the Plasma.
75
Calibration Curves for Chlorine and Bromine.
77
IV-10
E ffect of Oxygen on Chlorine Signal.
82
IV-11
E ffe ct of Hydrogen on Chlorine, Bromine and Lead Signal.
85
IV-12
Chromatogram of 20 pg Chlorine Illustrating Background
IV-8
IV-9
D rift.
88
IV-13
E ffe ct of Slit Width upon Signal and Background.
90
IV-14
Chromatograms of Chlorine a t th e 63, 6.3 and 3.2 ng
Level with the Background C orrected System.
93
Calibration Curves for Chlorine and Bromine with th e
Background C orrected System.
95
VI-15
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
TABLE NO.
I-1
PAGE
Halogen D etection Limits by Plasm a Methods
12
II-l
Equipment Specifications
19
II—
2
D eterm ination of TRIS in Spiked Methanol Samples
31
II—
3
Selectivity of Chloride
37
IV-1
Equipment Specifications
57
IV-2
Plasma Centering Flows
62
IV-3
E ffect of Power on Helium Line Intensities
64
IV-4
Reproducibility
79
IV-5
D etection Lim its
80
IV-6
Comparison of Minimum D etectable Levels
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
I.
INTRODUCTION
A.
THE PRESENT STATUS OF SELECTIVE DETECTION FOR HALOGENS
Compound and elem ent specific detection has been an elusive pursuit of
th e analytical chem ist.
Truly specific d etection implies response to only the
sample constituent of in terest.
Interferences have been observed w ith every
analytical technique. Even detecto rs as discrim inatory as gamma emission may
yield erroneous data due to th e interference of another species in the sample (1).
Thus, m ost workers have endeavored to develop system s which a re as selective
to the analyte as possible w ith minimal response to other m atrix components.
Selective detection of halogenated compounds has been studied by many
workers.
The necessity to monitor halogens has arisen from man's increasing
production of halogenated organic compounds. In recent years these compounds
have become im portant industrially, agriculturally and environmentally. In many
cases, it is im portant to monitor these compounds a t the u ltra-trac e level.
Several detection system s are currently in use which respond selectively to
halogenated compounds.
Perhaps the most widely used selective d etecto r for halogenated molecules
is the electron capture d etecto r (ECD) coupled with gas chromatographic
separation. Response is seen for molecules exhibiting electron affinity in th e gas
phase.
Compounds exhibiting sensitive response include alkylated m etals and
molecules with electron withdrawing groups attached to an arom atic ring. While
not necessarily a halocarbon specific d etecto r, the ECD generally responds well
to these compounds as they are generally efficient electron capturing species.
Thus, a good many of these compounds are detectable a t low levels. The ECD
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
has shortcomings as a halogen detecto r since response is highly dependent upon
molecular structure.
For instance, hexachlorobenzene gives a response 7000
tim es th a t of chlorobenzene (2) while chlorinated paraffins show very poor
response (3).
Another principle exploited in th e detection of halogens is solution
conductance. D etectors based on this mode generally operate by catalytically
decomposing the molecule. The gaseous products are then placed in co n tact with
a solution which solvates the halide. A change in conductance is then indicative
of halides. Initially a technique designated as microcoulometry was used w ith a
silver ion solution (4,5). Following a decrease in conductance caused by silver
chloride form ation (Cl" originating from the analyte), the solution was backtitra te d to the initial conductance.
Knowledge of th e am ount of the titra n t
necessary yielded th e analytical inform ation. While very sensitive and selective
for the halogen family, microcoulometry was found to be very tim e consuming.
Hall (6) m iniaturized this system , utilized dynamic flow and simply m easured
conductance to develop an efficien t, sensitive and selective detection system.
The Hall electrolytic conductivity d etecto r (HECD) exhibits chlorine selectivity
with respect to carbon as high as 10^. Varying molar chlorine response is seen
with molecular structure due to incomplete combusion e ffe cts (7).
As with
microcoulometry, th e HECD responds to all halogens as well as nitrogen and
sulfur. This gives rise to some loss in selectivity.
Mass spectrom etry is now being investigated for elem ental analysis (8-10).
At present sample introduction from "hard" sources such as the plasma are being
investigated to determ ine whether com plete analyte fragm entation occurs. This
technique exhibits great promise for th e determ ination of halogens; however,
problems still exist with the interface (9,10).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
As early as 1965, McCormick, Tong and Cooke (11) used atom ic emission to
monitor iodinated chrom atographic effluents a t the 2062
R
atom ic iodine line.
Several papers on th e determ ination of halogens by emission spectrom etry have
subsequently appeared. However, most of these publications to date have dealt
with the determ ination of m etals. Sensitivities in the p art per billion range and
linearity of response over several orders of magnitude in concentration may be
attained for many m etals and m etalloids.
Selectivity can be enhanced by
utilizing one of several available analytical lines for each elem ent.
avoid spectral overlap and background fluctuation problems.
This may
E fforts are now
being made to extend the usefulness of the technique to the routine determ ina­
tion of nonmetals.
B.
HISTORY AND
DEVELOPMENT
OF ELEMENTAL EMISSION SPEC­
TROMETRY AS AN ANALYTICAL TECHNIQUE FOR HALOGEN CON­
TAINING SUBSTANCES
The concept of using emission spectrom etry as a universal means of
selective elem ental detection is rapidly becoming a reality.
R ecent growth in
this area may be attributed to the developm ent of various sources which serve to
transfer energy to analyte atom s and ions.
While some elem ental emission
sources have reached apparent m aturity others are still in the infancy stage and
their capabilities are not y et fully realized.
As has been th e case with the
development of each major emission source in the past, the sources currently
under investigation are extending the range of elem ents amenable to analytical
elem ental emission spectrom etry.
Following the observations of Bunsen and Kirchhoff (12) in which the
visible emission of a salt in a flam e was found to be ch aracteristic of the m etal,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
analytical flam e spectrom etry has been well characterized . The flam e has been
a dependable source for the determ ination of alkali and alkaline earth m etals for
many years. However with flam e tem peratures in th e range of 2000 to 3200°C
(13) the flam e does not possess sufficient energy to ex cite many elem ents of the
transition series to analytically useful levels. For this reason atom ic absorption
is generally th e flam e method of choice when analysis of elem ents in groups
other than I, II and III A is required.
The arc or spark produced when a high voltage is applied across graphite or
carbon electrodes yields excitation tem peratures in th e range of 4000-8000° K.
These system s yield intense emission for many m etals and metalloids.
While
much analytical inform ation is available, arc and spark wander degrade quanti­
fication capabilities a t tra c e levels. As a resu lt reproducible solution introduc­
tion into these system s is a difficult task. Molecular cyanogen bands and other
complex background in terferen ts drastically obscure considerable analytical
inform ation. Arc wander and cyanogen bands may be reduced by sheathing the
electrodes with argon or an argon/oxygen m ixture. N evertheless, work with arc
and spark has been generally restricted to the analysis of solids, powders and
residues above tra c e levels.
The most recent emission source to gain widespread popularity is the
analytical plasm a. The combination of stability com parable to the flam e and
tem peratures exceeding the a rc makes the plasm a an excellent spectroscopic
source. The use of the plasm a has been extended to th e analysis of solutions (1417) gas and liquid chrom atographic eluates (18,19) and chemically (20,21) and
therm ally vaporized samples (22-24).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
Various modes now exist to sustain analytical plasm as.
By far the most
popular is the inductively coupled plasma (ICP). The present configuration most
commonly uses argon as th e plasma support gas. O perational frequencies of 2k
to 50 MHz with powers of 1 to 2.5 kW are generally used.
The d irect curren t
plasma (DCP) also uses argon as a plasma support gas and operates a t a power of
approximately 750 W. The third common plasma configuration is the microwave
induced plasma (MIP). Frequencies of 912, 1000 and 2450 MHz have been used
with the la tte r the most common. A t 2450 MHz most workers operate w ith less
than 100 W of power.
A major a ttra c tio n of th e MIP is its ability to support
atm ospheric pressure plasmas of both argon and helium. This ch aracteristic has
extended the application of emission spectrom etry to virtually every elem ent of
the periodic table.
In order to b e tte r understand the major differences betw een th e plasmas
sustained with different support gases, it is useful to consider the plasm a in
g reater detail. In Figure I—1, th e major en tities of the plasma sta te are depicted.
N eutral and excited atom s exist within the plasm a. An im portant subclass of the
excited atom group is th e m etastable sta te .
M etastables (i.e. - trip let and
singlet 2S helium states) exist a t high energy levels for relatively long periods of
tim e (10 us) (75). Collisions of m etastables w ith analyte atom s are thought to
play an im portant p a rt in excitation processes (25). Also within the plasma are
ionized molecules and their corresponding excited states. The recombination of
these ions with non-quantized, unbound electrons gives rise to a continuum of
radiation described by Equation 1. Energy also may be dissipated as black body
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1-1.
Energy Levels of Various Species.
Upper energy levels of plasma support gas
m etastable sta te s and support gas ionization potentials.
Energies for other
atom s are the upper energy levels of th e transition observed a t th e given
wavelength.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
551
He++
50
Ar'
25
He+
■S 20
^em
OJ
1
111 15
Ar*
C l(n ) 4795A
Br(IX) 4786^
Arm
10
P b (ll) 5 6 0 8 A
P b (I) 4058A
N a (I) 5890A
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
e~ + G+
where
-* ■
G* + hv (continuum )
(1)
e is an electron
G is th e plasm a support gas
h is Plancks' constant
v is the frequency of th e radiation
* indicates an excited sta te
radiation.
Significantly analyte and non-analyte atom s also give rise to
background radiation according to th e equation:
e" + M+ -*■ M* + h v (continuum)
where
(2)
M is an analyte atom or other foreign atom within th e plasm a
This plays an im portant role in term s of selectivity as a change in background
continuum a t th e analytical wavelength may be m istakenly in terp reted as
analytical signal when sample loading within the plasm a is significant.
At this
point it is im portant to note the observation of previous workers (26) th a t an
increase in to ta l background radiation is concom itant w ith an increase in
absolute fluctuation of the radiation. This facto r, in turn, a ffe c ts ultim ately the
signal to noise ra tio seen by th e analyst.
A significant energy difference is noted betw een analogous s ta te s of argon
and helium.
Collisional processes involving analyte ex citatio n d ic ta te th a t
available energy of the exciting species be equal to or exceed th a t necessary to
a tta in the upper energy s ta te of the analyte for the emission line observed.
Examination of Figure 1-2 reveals th e basis of the above c ited differences with
respect to various elem ents determ ined by elem ental emission spectrom etry (13,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F igure 1-2
Q ualitative Depiction of th e Plasma.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without
PLASMA
Excited Atoms
recom bination
Excited Ions
& Molecules
M etastable
S tates
Stable Atoms
& Molecules
' Ions
rae om blnatlo n
11
27-33).
It is apparent th a t for transitions commonly observed a t the listed
wavelengths the upper energy s ta te of m etals lies well below the m etastable
energies and ionization potentials of helium and argon.
Consequently, both
plasm a gases a re capable of producing elem ental emission of an analytically
useful intensity. The ion lines of m etals have much higher upper energy states
due to the ionization potential "offset". It follows th a t less emission is generally
seen for ion lines due to the necessity to prom ote them to these higher energy
states. This e ffe c t is even more pronounced w ith elem ents of g reater ionization
potential, such as the halogens.
Consequently, in the argon plasma, atom ic
halogen lines in the near-infrared region of the spectrum are the most intense
(29, 30).
Ion lines in the visible region are easily observed with the helium
plasm a (31) and are more intense than th e atom ic lines em itted from th e argon
plasm a. This is illustrated by examination of d etection lim its for various sources
in Table 1-1.
The im portant difference lies presumably in the energy levels
required to produce emission from th e halogens. The energy levels depicted in
Figure 1-2 of helium all lie above the excited sta te s of the halogens. The excited
s tates of th e halogens lie between the high energy argon states.
Unless a
significant concentration of a sta te with higher energy than the first argon
m etastable can be obtained, it is unlikely th a t argon will ever be the plasma
support gas of choice for producing the most intense halogen elem ental emission.
While this discussion has focused on the excited particles within the plasma, it is
im portant to note th a t other workers have observed tran sfer of energy via high
energy photons of the support gas high energy transitions (76).
The initial detailed study of the halogens was done by Bache and Lisk (37).
The work lists five atom ic emission lines for both chlorine and bromine and nine
for iodine. D ata for over tw enty compounds w ere given with sensitivities for the
m ost intense emission lines.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1-1
Halogen D etection Lim its by Plasm a Methods,* ng
Elem ent
Ar-ICP
Ar-MIP
He-MIP
He-DCP
I
24(28)
22.4(34)
0.056(35)
4.5(36)
Br
50(30)
159(34)
0.106(35)
45(36)
Cl
50(30)
4(34)
0.155(35)
45(36)
0.064(35)
4.5(36)
F
1000(29)
-----
* Numbers in parenthesis indicate reference number.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13-
Halogen-containing pesticides have been successfully m onitored with the
helium MIP d etec to r (38, 39).
Subsequently, Quimby e t al. (40) determ ined
trihalom ethanes in drinking w ater.
D etection lim its of less than 1 ppb and
selectivities g re a te r than 10 were obtained. Mulligan e t al. (41) also have used
GC/MIP to m onitor polybrominated biphenyl and related compounds with com­
parable results.
The most advanced GC/MIP system to d ate in term s o f chromatography
was applied by Quimby e t al. (42) for the determ ination of aqueous chlorination
products of humic substances. A capillary GC column effluent was routed into a
TMqiq mirowave cavity.
The detection lim it of pentachlorophenol was esti­
m ated a t 40 ppt (ngL“*). Application of a similar system was made by th e same
research group (35) and over 30 elem ents including the halogens w ere monitored.
It should be noted th a t Tanabe, e t al. have applied a very sim ilar system and
obtained detection lim its 10 fold b e tte r than previously reported (43). This work
has not y e t been duplicated by other workers.
C.
PURPOSE OF DISSERATION.
As has been discussed in two previous sections the need for an elem ental
selective halogen d etecto r has been noted and the helium MIP appears to w arrant
further investigation. The purpose of this dissertation is to study th e helium MIP
for this reason.
The first study is the developm ent of a sample introduction
technique for the rapid determ ination of to ta l halogen in a sample. Application
of the technique for the determ ination of tris(2,3-dibromopropyl) phosphate is
shown. The second investigation deals with e ffo rts to enhance th e sensitivity of
th e MIP for halogens by increasing the power utilized for plasm a m aintenance.
Plasma to rch design and the in terface to a gas chromatograph are illustrated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
II.
DETERMINATION OF HALOGENS BY ELECTROTHERMAL VAPORIZA­
TION INTO A HELIUM MICROWAVE INDUCED PLASMA AT ATMOS­
PHERIC PRESSURE
A.
INTRODUCTION
The necessity to monitor halogenated organic compounds cannot be over­
emphasized.
Many such compounds a re known to be carcinogenic and a
m ultitude of others a re suspect.
The toxicities of compounds such as te tra -
chloridibenzo-p-dioxins (44-46), polychlorinated biphenyls (47) and tris(2,3-dibromopropyl) phosphate (TRIS) (48) have been well documented.
Of th e 114 non-
m etallic priority pollutants listed by th e United States Environmental Protection
Agency, 72 fall into th e halocarbon category (49).
In some cases it is useful to determ ine the to ta l amount of halogen in a
sample. This is worthwhile if th e sample is known to contain th e analyte as the
only halogenated compound or if speciation of the various halogen containing
componds is unnecessary. A number of workers have pursued this task (50, 51).
Analyte vaporization has played an im portant role as an analytical spectrom etric technique. Major advantages which may lead to b e tte r detection lim its
w ith this technique are:
1)
High concentrations of th e analyte a re produced a t th e d etecto r by
confining the analyte to a small volume prior to and during vaporization.
This is achieved by rapidly vaporizing th e analyte into a flowing stream of
gas thus confining the analyte to a sm all volume as it reaches the d etecto r.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
Prior preconcentration also is possible by absorption of analyte vapor onto
a stationary phase. Again, rapid revaporization produces a high instanta­
neous concentration of analyte a t th e d etecto r.
2)
Reduction of m atrix interferences by spatial separation of th e analyte
from other substances in th e sample which may eith er enhance or reduce
the signal. This may be accomplished by selectively vaporizing, trapping or
chromatographing the analyte.
Several workers have utilized desolvation prior to electro th erm al vaporiza­
tion (22,23,52).
Conversion of the analyte to a volatile species has also been
successful (20,21,53).
The microwave induced plasma has been utilized in the
rapid quantification of inorganic halides by generation of HC1 (20) and
gases from solution.
(21)
D etection lim its (analyte signal corresponding to 2 tim es
the standard deviation of th e baseline noise) of
6
ng and 25 ng respectively were
obtained with linear responses of over 3 orders of magnitude.
The present work describes a method intended for rapid and accurate
determ ination of to ta l halogenated organic compounds utilizing an electro ­
therm al vaporization device to generate analyte vapor for the MIP. Application
of the technique to the determ ination of th e fire retard an t TRIS in garm ents is
dem onstrated.
B.
EXPERIMENTAL
Reagents and Instrum entation
The solvents were reagent grade.
Tris(2,3-dibrompropyl)phosphate was
obtained from P faltz and Bauer (Stanford, CT). C itex BC-26 was obtained from
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
the United States Food and Drug Administration.
All chem icals were used
w ithout further purification.
The instrum ental configuration is depicted in Figure II—1.
The electro ­
therm al vaporizer is based on a design by Nixon e t al. (22) as modified by Rose e t
al. (23,54). The remaining components are detailed in Table II—1.
Procedure
Dry helium (passed through a D ri-Rite laboratory gas-drying unit) is routed
to the electrotherm al vaporizer and the auxiliary flow is m aintained a t 200 mL
min- ^. A sample of the analyte in a volatile solvent (either methanol or hexanes
with boiling range 65-67°C) is placed on the vaporization device (the carbon cup
or the tantalum boat) through the sample port with valve A open and valve B
closed. The solvent is allowed to evaporate through the sample p o rt a t am bient
tem perature. A fter th e bulk of th e solvent has escaped from th e vaporizer (ca.
60 s) the sample port is capped, valve B is opened, and valve A is closed. This
procedure is required so th e solvent flow does not extinguish th e plasm a. The
residual solvent remaining in the dome causes an increase in background emission
as it is passed into the plasm a. A fter the dome has been purged of this residual
solvent (ca.
120
s), a steady baseline is obtained and th e carbon cup or tantalum
boat is resistively heated to vaporize the analyte. The support gas sweeps th e
vaporized analyte into th e plasma and emission is measured and recorded.
The method described above was applied to the determ ination of e x tra c table tris(2,3-dibromopropyl)phosphate in garm ents. The compound was extracted
by a procedure similar to th a t of Smith and Wheiihan (55). Approxmimately 0.6 g
samples of the tre a ted garm ent obtained from the Consumer Protection Agency
were placed in 5 ml of methanol in capped 2-dram vials.
The samples were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F igure II-1.
ETV/MIP System Diagram.
A and B, shut off valve; C, drying tube; D,
flow m eters; E, resonator cavity; F, 3-stub tuner; G, microwave power supply; H,
focusing lens; I, monochromator; J, photom ultiplier tube; K, cu rren t to voltage
converter/variable gain am plifier; L, strip c h art recorder; M, vaporization device
base; N, pyrex done w ith sampling port; P, vaporization power supply.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
Table II-l
Equipment Specifications
Microwave generator
Kiva Instruments (Rockville, MD). Capable
of producing 1 2 0 W with reflected power
m eter (operated a t 70 W forward and 1 W
reflected).
Microwave power transm ission
Model 1878B-3 stub tuner from Maury
Microwave Corp. (Cucamonga, CA) and a
30-in. 39809 cable w ith 44SW-N type
connectors from Andrew Corp. (Orland
Park, IL).
Beenakker TMq^q cavity
Similar to th a t described by Mulligan, e t
al. (41) with cavity depth of 8 . 0 mm an
diam eter of 91.5 mm.
Monochromator
0.5-m 3arrell-Ash E bert scanning mono­
chrom ator with 50 pm entrance and exit
slits, grating blazed a t 330 nm, f / 6 . 8
Optics
5.0-cm diam eter, 5.2-cm focal length lens.
Photom ultiplier tube
Hamamatsu R212.
C hart recorder
Hew lett-Packard Model 7101B.
Sample application
10-
Carbon cup
Varian 99-100224 cups ground to an ap­
proximately 1 0 p i inner volume and pyrolytically coated (54).
Tantalum boat
Barnes Analytical 0006-527.
Electrotherm al vaporization
power supply
Varian Model 63 used in th e dry mode.
1
w iretrol pipettes.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
20
manually agitated interm ittently and a fte r 1.5 h aliquots of the solutions were
placed in the vaporizer for analysis.
Either standard calibration curves or
standard additions were used. Spiked synthetic samples w ere also processed by
comparison with a calibration curve.
C.
RESULTS AND DISCUSSION
Wavelength Selection
Because th e major in terferen ce of this technique is due to a general shift
in th e background, and because th e most useful Cl and Br lines in th e visible
region are within
12
nm of each other, the most intense sp ectral lines were
chosen. Continuous sample introduction in a method sim ilar to th a t of Mulligan
e t al. (41) yielded maximum emission for chlorine a t th e 479.5 nm line and
bromine a t the 478.6 nm line for CCl^ and CBr^, respectively.
Selection of Vaporization Filam ent
Both the carbon cup and th e tantalum boat w ere examined for th eir
suitability as th e vaporization device with organic analytes. Although the carbon
cup is inherently more durable, background problems precluded its use. Figure
II-2a shows recorder tracings using the carbon cup. A change in baseline on dry
firing of the cup resulted in a substantial background signal., which may be
attrib u ted to a change in flow ra te upon heating th e plasm a support gas. Similar
behavior was noted with the tantalum boat. With solvent (10 yL of hexanes) only
the baseline signal increased, indicating probable adsorption of the solvent to the
cup. The increase in the baseline is probably due to band emission of th e solvent.
Upon increasing the drying tem perature sufficiently to rid the cup of solvent, the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure II—
2.
Signal for 1 y g C itex BC-26.
2A, using modified graphite cup.
2B, using
tantalum boat; a, 1 yg BC-26 in 10 yL hexanes; b, 10 y L solvent; c, dry firing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
f CD
CM
CO
GO
<
CVI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
analyte also was lost.
The detection lim it (signal equivalent to 2 tim es the
standard deviation of th e baseline noise) using the carbon cup was calculated to
be 40 ng.
The tantalum boat proved more suitable for this technique.
As seen in
Figure II-2b, the background is much sm aller than with th e carbon cup.
Also,
because th e backgrounds on firings of the dry boat with and w ithout solvent are
th e sam e, it appears th a t th e solvent adsorption to the tantalum boat is
negligible. The detection lim it is improved to
8
ng of C itex BC-26.
System O ptim ization
The voltage applied to th e tantalum boat was varied to determ ine the
maximum obtainable peak height, with results shown in Figure II—
3.
At
potentials less than 0.2 V, th e boat did not reach a tem p eratu re sufficient to
vaporize th e analyte. With increased voltage, the peak height greatly increased
and reached a maximum a t approxim ately 0.31 V. A t voltages above 0.31 V, the
decrease in peak height is probably due to g reater analyte dispersion as it is
vaporized.
Varying flow rates yielded a maximum signal a t 400 (±20) ml min~*. At
higher flow rates, the reduced peak heights are attrib u ted to decreased residence
tim e of th e analyte in the plasm a. This is shown in Figure II-4. A t lower flow
ra te s, the heating of th e helium by th e tantalum boat induces a g reater
fractional change in the flow ra te .
This e ffe c t causes the plasm a to be more
greatly detuned and subsequently less robust. Therefore, the response is reduced
because less halogen is excited.
The peak shape a fte r system optim ization is shown in Figure II-5. Although
some tailing does occur, the peak was sufficiently sym m etrical th a t linear
working curves were obtained betw een 100 ng and 20 lag of BC-26 when peak
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure II-3
Dependence of Vaporization Voltage on Peak Height.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CO
o
CD
<
■
O
o
o
CO
H
_ ]
o>
CM
o
CVI
oO
3SNOdS3U
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F igure II-4
Dependence of Flow R ate on Peak Height.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
o
00
o iLU
o <—
^ "
1
IT
o
_ l
LL
o
CO
o
CM
o
T-
3SN0dS3ld
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure II-5.
ETV/MIP Peak Shape.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
^A1ISN3±N!
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
heights w ere measured. Evidence of self-absorption is seen a t th e chlorine line.
Typical calibration curves a re shown in Figure II-6 .
D eterm ination of Tris(2.3-dibromopropyl)phosphate in Garm ents
Calibration curves of TRIS in methanol were linear over 2 orders of
magnitude when determ ined a t th e 478.6 nm Br(II) line.
R elative standard
deviation a t th e 1 V g level was 4.4%. Five samples spiked with TRIS in th e range
of th e working curve produced relative errors of 9.1 to - 2 . 0 %. These d a ta are
seen in Table II-2. C orrelation coefficients using this method ranged from 0.994
to 0.999996. As day to day recalibration is generally required w ith MIP systems,
it was found more convenient to use standard additions for th e determ ination of
TRIS in children's sleepwear. Actual extractions yielded an average relative
standard deviation of 9.1% for garm ents with TRIS concentrations ranging from
1.55 to 28.4 mg/gm garm ent.
Investigation of Various O ther Samples
Studies were initiated to determ ine the applicability of th e described
technique to other types of samples.
These included organic sam ples with
complex m atrices and synthetic aqueous inorganic samples.
The determ ination of low level polychlorinated biphenyls in transform er
fluids is presently an analytical problem (56). The method described above was
investigated for this application.
Typical transform er fluid bases (silicone oil
and paraffin oil) w ere found to alte r the background substantially as shown in
Figure II-7. Given are recorder tracings of 10 V
of 4% (v/v) paraffin oil and
silicone oil in hexanes and 200 ng C itex BC-26.
Clearly th e background shift
caused by the m atrix is substantial. A ttainm ent of an equivalent signal by PCBs
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
Table II- 2
D eterm ination of TRIS in Spiked Methanol Samples
Amount TRIS ( yg)
Amount TRIS Found (y g)
% RSD
% Error
8.39
9.15
0.14
1.5
+9.1
5.58
5.49
0.25
4.6
- 1 .6
5.24
5.21
0.12
2.3
-0 .6
3.49
3.42
0.29
8.5
-2 .0
2 .1 0
2.17
0.11
5.1
+3.3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F igure II-6
C itex BC-26 Calibration Curves, a, Bromine a t th e 4786
teh 4795
R
R
line; b, chlorine a t
line.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RESPONSE
1.0
1.0
10
MASS CITEX BC-26 (ug)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure II-7
Response of Various Viscous Solvents.
10 U L samples in hexanes, oils diluted
1:25.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
200ng
„
BC' 26
EPA
Paraffin
Oil
UC
Paraffin
Oil
UC
Silicone
Oil
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
in transform er oil would require th a t th e minimum chlorine content of th e
samples be 0.1 to 1.0%.
While this technique may be applicable to the
determ ination of percentage amounts of PCBs, low level determ ination is not
possible. Experim ents perform ed w ith e x tracts of methoxychlor and captan from
Ortho Tomato and Vegetable Dust yielded similar problems due to th e m atrix.
Analysis of aqueous samples for inorganic chlorides was examined.
chloride as sodium chloride,
10
For
ng was found to yield a substantial signal with
detection lim its (2 s.d.) of 600 pg. Selectivities of chloride with respect to some
common anions were determ ined by adding a
100
fold excess (w/w) of th e
interfering compound to 10 ng of chloride sample. The selectivities, which are
listed in Table II—
3 range from 370:1 for G f:N O j“ to 540:1 for Cl_:CO^- .
It
should be noted th a t a c e ta te and n itra te enhanced th e chlorine signal while
carbonate and sulfate caused signal depression.
D.
CONCLUSION
The potential of electrotherm al vaporization into th e microwave induced
plasm a for th e determ ination of to ta l halogens has been illustrated.
Although
only bromine and chlorine were investigated, the method should be applicable to
other nonm etals such as iodine and sulfur (27).
Successful application has been made for the rapid and simple determ ina­
tion of tris(2,3-dibromopropyl) phosphate by monitoring th e bromine emission
line a t 478.6 nm.
Selectivities indicate th a t this method holds promise for
monitoring chloride a t environm ental levels.
Matrix problems a re prevalent in
complex samples such as tom ato dust e x tra c t and transform er oil. Circumventing
these problems will necessitate some separation prior to introduction into the
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
37
Table II-3
Selectivity of Chloride
_______ Sample_____________
R elative Response
Selectivity
lOng Cl-
1.00
-----
lOng Cl-:1000ng CH3 COO-
1.20
500:1
lOng Cl-:1000ng CO§~
0.83
590:1
lOng Cl~:1000ng NO- 3
1.27
370:1
lOng Cl":1000sng SO?f
0.80
500:1
* Selectivity = mass of interferant x response due to analyte alone
mass of analyte
response due to analyte alone response of analyte and in terferan t
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
38
plasma. A lternatively, or in conjunction, rapid background correction techniques
such as those described by Estes, Uden and Barnes (35) should greatly decrease
the minimum detectable amount for halogens with samples of this nature.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
III.
REQUIREMENTS FOR MAINTENANCE OF A MODERATE POWER
MICROWAVE-INDUCED PLASMA
A.
INTRODUCTION
Several investigators have noted the e ffe c t of applied power on emission
produced by the microwave-induced plasma (11,52,57,58). Generally an increase
in analytical signal is noted with increasing power. This phenomenon has been
shown for elem ental emission of Cl, Br, I, S, C, H and N (11,57,58) and
molecular bands of C 2 (58) and CN (11) with gas phase sample introduction.
These trends w ere noted in studies done a t powers of <200W since available
microwave generators were generally capable of maximum powers of either
120W or 200W.
Investigations of d irect solution analysis w ith 200W or less MIP systems
show th a t prior desolvation (59,60), low sample uptake ra te s (61) or ultrasonic
nebulization (67) are necessary.
Results obtained by Burman and Bostrom (63)
utilizing a 600W-2450 MHz plasma are promising. Their system was capable of
stable operation with a 1.2 mL/min aqueous sample uptake ra te w ith a European
Applied Research Laboratories plasma torch. In this com parative study the ICP
was preferred due to its low susceptability to m atrix e ffe cts. However, the MIP
system was and still has not been com pletely optim ized. It is clear th a t the need
has been dem onstrated for further investigations of the use of higher powers
with th e MIP.
The use of higher powers with presently available MIP system s presents
additional experim ental difficulties.
Plasma containm ent tube degradation be­
comes more significant with increasing power,
van Dalen and co-workers (58)
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
40
have noted a substantial increase in emission a t th e silicon line with increasing
power from 50 to 175W using a helium plasm a contained within a 2mm i.d. quartz
tube.
Spectra of helium plasm as taken by E stes, Uden and Barnes (37,38) have
shown intense silicon emission w ith powers as low as 54W and a 0.5 mm i.d.
tube.
D eutsch and H ieftje (64) have noted th a t qu artz will m elt a t powers
exceeding 190W unless provision is made for cooling.
From th e preceeding discussion, it is apparent th a t the problems encoun­
te re d with th e MIP are analogous to those encountered in the early work w ith the
ICP (17). Therm al stablity is mandatory for the m aintenance of a rugged stable
m oderate power* MIP system which is capable of day to day reproducibility a t
acceptable levels. With such a system it may be possible to improve minimum
d e tectab le am ounts of th e nonm etals as well as enhance the capabilities of the
MIP for solution analysis. The purpose of this study is to examine some of the
problems encountered and to indicate c rite ria necessary for the m aintenance of
th e m oderate power MIP.
B.
EXPERIMENTAL AND INSTRUMENTATION
The lead chloride (Baker) was reagent grade and used w ithout fu rth er
purification.
Laboratory distilled w ater was deionized by passing through ion
exchange columns (Illinois W ater T reatem ent Co., Rockford, Illinois).
The instrum ental configuration is described in C hapter II of this disserta­
tion with th e following exceptions. Unless otherw ise stated , the generator was a
Micro-Now Model 420 (Chicago) capable of producing up to 500W forward power.
♦H ereafter, m oderate power refers to operational powers betw een 150W and
500W as measured a t th e generator.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
An Andrews type FSJ4-50A microwave transmission cable of a 3 foot length with
445W type end connectors was utilized. Plasma containm ent tubes consisted of
quartz (2.5 mm i.d.,
6
mm o.d. Heraeus Amersil, Sayreville, New Jersey) and
aluminum oxide (4 mm i.d., 7 mm o.d., A lfa Products, Danvers, Massachusetts).
G raphite cups ground to a 15 y L inner volume and pyrolytically coated were
utilized (24). Aqueous lead samples were placed in th e carbon cup. The sample
was dried for 1 minute a t approxim ately 90°C with th e sample port opened, the
p ort was capped' and th e sample was heated to a sufficient tem p eratu re to
vaporize th e lead. Lead was m onitored a t th e m ost intense line (3639.6 R ) in
the helium plasma (65).
C.
RESULTS AND DISCUSSION
E ffect of Power on Lead Signal
Utilizing th e quartz containm ent tube, 50 ng lead samples were vaporized
into th e MIP. The helium flow ra te was 550 mL/min. The power supplied to the
plasma was varied betw een 60 W and 227W. The e ffe c t of applied power on lead
signal is seen in Figure III—1. The peak height due to lead increases from 60W to
160W.
This increase in signal levels o ff from 160W to 225W. Leveling off is
presumably caused by degradation of the plasma containm ent tube.
A t powers
greater than 190W th e quartz tube fuses because of th e higher therm al
tem perature of the helium plasm a. A t power levels of 225W this occurred a fte r
about 15 minutes of operation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure III—1
Dependence of Power Applied Upon Lead Peak Height with E lectrotherm al
Vaporization.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CM
&
CM
s
i
i
£
O
CO
o
CM
O
r—
asuodsay eA^ey
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
E ffect of System Cooling on Signal Intensity
It was found th a t w ater cooling as suggested by Mulligan e t al. (36)
m aintained the cavity a t a relatively cool tem p eratu re. Moreover th e tem pera­
tu re rem ained relatively constant.
O peration of th e plasma a t power levels above 70W for extended periods
caused excessive heating of th e power tra n sfe r cable and 3-s.tub tuner.
These
high tem peratures caused the teflon d ielectric inserts to decompose. Passing a
stream of am bient tem perature air or nitrogen over th e 3-stub tuner did not
elim inate the problem. In another a tte m p t a t cooling, the tuner was encased in a
2"
x 6 " x 8 " aluminum box insulated with foam padding held in place by a slightly
larger aluminum box.
A stream of nitrogen was cooled by passing through a 1
.m eter length of 1/8 *1 o.d. teflon tubing immersed in liquid N j.
allowed tem peratures of
This cooling
0°C to be established for the tuner. While this method
avoided therm al degradation of th e tuner, th e analytical signal was highly
tem perature dependent. This is illu strated in Figure III-2. The analytical signal
for 1 ng of lead w ith the tuner a t -5°C is approxim ately 10 tim es th a t seen a t
+5°C.
It is obvious th a t precise tem p eratu re regulation will be necessary to
operate th e plasm a in this configuration with acceptable analytical reproduci­
bility.
E ffect of C ontainm ent Tube Composition on Tuning and A nalytical Signal
With a 93.0 mm i.d. Beenakker cavity and a Kiva microwave generator (01 2 0 W)
it was possible to m aintain a helium plasm a in both quartz and aluminum
oxide tubes. However, with the Al^O^ plasm a containm ent tube plasma ignition
was somewhat difficult, tuning was e rra tic and long term stability was poor, van
Dalen e t al. (6 6 ) have noted th a t quartz has markedly less e ffe c t on th e
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Figure III-2
Dependence of Stub Tuner T em perature Upon Lead Signal.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
■Q
CL
c
O)
c
CO
A
0
o
1C
1
o
o
lO
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
resonance frequency of Beenakker type cavities than
dielectric constant of the la tte r. It is likely th a t the
due to the high
induced reduction in
cavity resonance frequency exceeded the tuning range of th e system. The ability
of this experim ental system to m aintain a stable plasma is further decreased by
heating of the containm ent tube.
In some cases th e A ^O ^ tube would crack
causing th e plasm a to be extinguished.
A nalytical signals obtained for 50 ng of electrotherm ally vaporized lead
with quartz and A ^O ^ containm ent tubes are shown in Figure III-3. Each plasm a
was operated a t 70W. A number of factors may contribute to th e difference in
the signals. Among these factors are: larger internal volume with A ^O ^ tube,
differences in the reflected powers (0W, quartz; 15W A ^ O j) and source realign­
m ent which was required upon changing of containm ent tubes.
Although
somewhat inconclusive, these data indicate th a t th e analytical signal is not
greatly a ffe cted by the plasma containm ent tube composition.
D.
>
CONCLUSION
Experim ents have been conducted to determ ine th e requirem ents for the
m aintenance of a microwave-induced plasma systems a t m oderate power levels.
The -most im portant difficulty to deal with is th e excess h eat generated
throughout th e system.
While th e com m ercial coaxial cable utilized was
sufficient to operate a t these power levels (only
6%
power attenuation a t 2450
MHz) the 3-stub tuner requires the dissipation of large amounts of h eat with
precise tem perature regulation.
This facto r would be greatly reduced if
impedance matching were accomplished with the therm ally stable w ater cooled
cavity. This has been achieved by th e attach m en t of a 54-wave Evenson'cavity to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure III-3
Lead Signed w ith Q uartz and Alumina Containm ent Tubes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I
50 ng Pb
5 ng Pb
Blank
Quartz
AI2O3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
the face of a Beenakker resonance stru ctu re (6 6 ). An arrangem ent of this nature
would allow dissipation of heat by therm ai conduction from the tuner to the
resonant cavity to the coolant w ater. Another problem which m ust be minimized
is therm ai degradation of the plasm a containm ent tube. It has been shown th a t
A ^ O j is a viable alternative to quartz as a plasm a containm ent tube. A lteration
of th e Beenakker cavity internal diam eter is necessary to com pensate for the
high dielectric constant of h X ^O y
The use of A ^O ^ would be advantageous
since its melting point is 2045°C, 435 degrees higher than quartz (67).
O ther
possibilities which would be helpful include centering of the plasma to avoid
co n tact with containm ent tube walls in the manner of Bolla-Kamara and Codding
(6 8 ) and addition of coolant gas as is utilized w ith the ICP torch.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
IV.
MODERATE POWER MICROWAVE INDUCED PLASMA GAS CHROMA­
TOGRAPHIC DETECTOR FOR HALOGENS
A.
INTRODUCTION
A microwave resonant cavity capable of stable operation a t m oderate
powers has been constructed by Haas e t al. (69).
This piasma m aintenance
device utilizes a design similar to the TMq ^q structure as described by
Beenakker (70). It was modified so th a t the internal diam eter was reduced to
86.5 mm to fa c ilita te th e use of A ^O ^ containm ent tubes. The cavity was w ater
cooled in the manner of Mulligan, Fricke and Caruso (41), and the impedance was
m atched by adjustm ent of two shorting stubs attach ed directly to th e coupling
loop. This "cavity-tuner" combination allows stable operation of argon or helium
plasmas w ith 100-500W forward power and 0W reflected as measured a t the
generator. Prelim inary results with direct solution nebulization are promising.
As discussed in C hapter III of this dissertation, halogens show more intense
emission with an increase in applied microwave power. With this facto r in mind,
an investigation of the e ffe c t of higher power levels on the emission of halogens
is in order. This chapter describes the e ffe c t of power, flow ra te and scavenger
gases on halogen and lead emission.
Also, an air cooled plasma torch/gas
chrom atographic in terface is presented. This torch centers the plasm a within an
easily interchangeable containm ent tube while maintaining m oderate support gas
flow rates.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
52
B.
EXPERIMENTAL
Reagents
Tetraethyllead (TEL 80%) in toluene was obtained from Ethyl Corporation.
O ther reagents were ACS reagent grade.
All chem icals were used without
further purification.
Instrum entation
The GC/MIP system is depicted in Figure IV-1 and a diagram of the
interface is shown in Figure IV-2. This equipment is described in detail in Table
IV-1.
The gas chrom atograph utilized was a modified F and M Scientific Model
700. Heating of th e oven and injection port a re as originally provided. A h eater
control circuit was deisgned and utilized to regulate th e tem perature of a 2.5
cm x 5 cm x
8
cm aluminum block interposed betw een this end of the column and
th e discharge tube as it enters th e cavity.
A stainless steel needle guide was
inserted into th e injection port to accom m odate injection into th e GC column.
The chrom atographic packing was conditioned by heating to 200°C for 24 hrs.
with He passing through the column.
Eluents were routed from the column
through a 1/4” to 1/10” stainless steel reducing union and into 1/16" stainless
steel tubing.
This tubing connected with a low dead volume 4-way switching
valve. This enabled routing of the flow either to vent or another 1/16" stainless
steel tube term inating 1 .5-2.0 mm from the helium plasma.
The microwave resonant cavity is th a t described by Haas, e t al. (69).
Plasma containm ent tubes consisted of quartz or A ^ O y
These tubes extended
from 5 mm beyond th e cavity faceplate to within the GC oven. Inside the oven
the tubes were fitte d snugly into a T -fitting bored to the proper diam eter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig u re IV-1
Moderate Power GC/MIP Interface.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A uxiliary flew
QC
MX
Q uartz
Block
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure IV-2
Overall System Diagram fo r GC/MIP.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
JB
Monochromator
W
«
•o
o
o
«
ee
e
CO
o
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
T able IV-1
Equipment Specifications
Microwave generator
0-500 W, Micro-Now Model 420, Chicago
0-120W, Kiva Instrum ents, Rockville, MD.
Power tra n sfe r cable
FS34-50A (3 ft.) with 445W end con­
n ectors, Andrews, Orland Park, IL.
Resonant cavity
Described by Haas, e t al (69).
Single Channel system
Monochrometer
0.5m 3arrell-Ash E bert scanning mono­
chrom eter with 50 p m en tran ce slits,
grating blazed a t 300 nm, f / 6 . 8 .
Optics
5.0 cm diam eter, 15 cm focal length glass
plano-convex lens.
i/V converter-am plifier
Described by 3. P. M cCarthy (72) with a
low pass filte r follower (RC=0.047 to 4.7
seconds adjustable with a 10 M resistor).
Strip c h a rt recorder
Model B-5117-5 Omniscribe
Microcomputer
University of Cincinnati Intel 8080 based;
d a ta acquisition program described in
Appendix I.
Background C orrected
polychrom eter system
Described by M. A. Eckhoff (71).
Gas chrom atographic system
Modified F and M Scientific Model 700
Switching valve
4-way, high tem perature (300°C), zero
dead volume, Valeo, Houston
Chromatographic column
6
H eated tran sfer block circuitry
Described in National Semiconductor
C atalog, 1980.
Plasma Torch
ft., 1/4" o.d., 1/8" i.d. glass column
packed with 3% OV-17 on 100/10 mesh
chromosorb 750.
Described in te x t
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
Teflon ferrules provided a gas tig h t fit while avoiding containm ent tube
cracking. Auxiliary gas a t flow ra te s of 0.5 to 4.0 L/m in passed through a
6
ft.
coil of 1/4" aluminum tubing within th e oven and into the side-arm of a T -fittin g .
The gas then flowed through the containm ent tube and through the threads of a
threaded brass insert. This tangental flow arrangem ent produced a helical flow
pattern within the containm ent tube a t the point where th e plasm a was main­
tained.
Also, it provided plasm a centering in th e manner sim ilar to th a t
described by Bollo-Kamara and Codding (6 8 ).
A t m oderate powers it was
necessary to provide a means of cooling for th e containm ent tube.
This was
accomplished by passing filtered laboratory line air into a pyrex device which
fits concentrically around th e discharge tube and directs cooling flow around the
containm ent tube surface. The pyrex concentric cham ber and placem ent within
th e cavity is shown in Figure IV-3.
Both a m onochrom eter and a polychrom eter w ere used to isolate the
emission lines of in terest.
Unless otherw ise stated , th e d a ta taken in this
investigation w ere obtained w ith th e monochrometer system . The polychrom eter
d ata acquisition system has been described previously (71). The m onochrom eter
instrum entation system consisted of a glass focusing lens, a cu rren t to voltage
converter/am plifier as described by McCarthy (72), followed by an operational
am plifier variable frequency low pass filte r. D ata w ere taken on a strip ch art
recorder and/or an Intel 8080 based m icrocom puter system .
for chromatogHraphy which averaged
1000
A BASIC program
d ata acquisitions/data point was
utilized. The program provided a 25 point Savitzky-Golay (73) d a ta smoothing
routine and provision for calculation of peak height, width and area, baseline
noise and detection lim its (defined as
2
tim es the std. dev. of the baseline noise).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure IV-3
Pyrex Cooling Device
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
<
e
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
O ther details of both system s are provided in Table IV-1.
The BASIC
program "NUGNOMER" and th e h e ate r block circuitry are described in Appen­
dices I and II respectively.
C.
RESULTS AND DISCUSSION
Plasma Torch
The use of three d ifferen t containm ent tubes was examined. The first was
a 5 mm i.d., 7mm o.d. aluminum oxide tube and th e la tte r two were 2.5 mm i.d.,
6 .0
mm o.d. and *f.O mm i.d.,
8
mm o.d. quartz tubes, respectively.
Initially the torch was ch aracterized using A ^O ^ containm ent tubes.
Threaded brass inserts>of pitches ranging from 3)4 to 20 threads/inch with 1 to 4
thread cuts were constructed. Only slight differences were noted in th e ability
of each to sustain plasma centering with th e inserts a t a distance of 2 to 5 mm
from the plasm a. D ifferences were more easily seen with the inserts withdrawn
11 mm from th e plasma. These results are listed in Table IV-2. In all cases the
single threaded inserts required higher flows to cen ter th e plasm a.
O ften the
plasma was removed from th e wall but off-cen ter causing non-uniform heating of
the containm ent tube. The double and quadruple threaded inserts required the
least flow. Of th e three inserts with similar minimum flow rates, th e choice was
made to use th e double threaded, 5Yi threads/inch insert as it provided suitable
experim ental situation and was less difficult to construct than th e quadruple
thread.
During actu al chrom atographic runs, the insert was positioned 1-2 mm
from the plasma. Q uartz containm ent tubes w ere utilized since th e A ^O ^ tubes
had a tendency to crack causing substantial changes in background intensity,
background noise and plasm a tuning characeristics. Of the quartz tubes, th e 2.5
mm i.d. tube yielded the best detection lim its for all elem ents studied.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
Table IV-2
Plasm a Centering Flows
Threading
P itch (threads/inch)
Minimum Flow Required (mL/min)
Single
20
4560
Single
11
3120
Single
5.5
3220
Double (180° offset)
11
3480
Double (180° offset)
5.5
2650
Double (180° offset)
3.25
2620
Quadruple (90° offset)
5.75
2650
Quadruple (90° offset)
3.25
3100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
E ffect of Power on Plasm a C haracteristics and A nalytical Signal
The e ffe c t of power on analytical and plasm a ch aracteristics was investi­
gated. High flow rates (3.1 L/min) allowed th e plasm a to be m aintained over a
range of powers while keeping other variables constant.
M easurements of
chrom atographic peak height and background signal were made in th e range of
180 to 500 W forward power.
Helium spectra from 3600 to 6000
R
mm were
taken over the same power range.
The helium sp ectra revealed
8
atom ic lines in the range analyzed.
The
wavelengths, the term symbols for th e upper and lower states, th e excitation
energies and the e ffe c t of increasing power on intensity a re listed for each
oberved transition in Table IV-3. It is apparent th a t while th e population of the
excited sta te s of 23.7 eV and higher increase as th e amount of power is increased
th e population of excited s ta te s of energy less than 23.7 eV rem ained unchanged.
The only exception is th e unresolved He(I) doublet a t 4471
R
in which th e 4d D
upper s ta te population is unaffected by changes in applied power. However, it is
interesting to note th a t th e transitions a t this excitation potential whose
intensity increases w ith power are singlet sta te s while the 4471
R
doublet is
composed of trip le t states.
Analytically, the general trend for chlorine and bromine was an increase in
emission intensity as the power was increased. The Pb signal a t both th e 4058
nm atom ic line and th e 5069
power.
R
R
nm ion line rem ained constant with changes in
In all cases th e background continuum increased significantly.
results a re shown in Figures IV-4, IV-5 and IV-6 .
These
Values for signal and
background are normalized to th e values obtained a t the lowest power level.
The differences in th e signal dependence with respect to power for
halogens and Pb may be explained in term s of the upper energy level of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
Table IV-3
*
E ffects of Power on Helium Line Intensity
#
Upper Energy (eV)
24.04
W avelensth (R)
Transition
Int (470 W)
Int (210 W)
4026.36
5d 3 D-2p3P
3.07
4026.19
24.04
4387.93
5d 1 D-2p1P
2.50
23.74
4921.93
4d 1 D-2p1P
2.52
23.74
3964.73
4p 1 P-2s1S
2.09
23.73
4471.68
4d 3 D-2p3P
1.05
4471.48
23.67
5047.74
4s 1 S-2p1P
2.5
23.09
5015.68
Sp^W s
1.05
23.07
5875.62
3d 3 D-2p3P
1.03
5875.97
*
Inform ation regarding atom ic transitions was obtained from reference 74.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure IV-4, 5, 6
Dependence of Power Applied Upon Normalized Signed and Background for
Chlorine, Bromine and Lead. All signals normalized to 1.0 fo r th e low est power
setting utilized.
Figures k and 5; triangles indicate analytical signal and
diamonds indicate background intensity. Figure
6
; diamonds indicate signal for
lead ion, triangles indicate signal for atom ic lead and circles indicate background
intensity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Normalized Intensity
2 .5
2.0
1.5
1.0
200
300
400
500
Power(Watts)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5 .0
4 .0
6i
CO
c
4
CO
E
zo
2.0
1.0
200
300
400
500
Power(Watts)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.5
>
*55 2.0
c
c.
£
TJ
r
CO
E 1.5
o
Z
1.0
200
300
400
500
Power(Watts)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
analytes.
As shown in Figure 1-2 th e m etastable energies of He a re near in
energy to the upper energy sta te s required for halogen emission. This work has
shown an increase in high energy transitions a t higher powers with th e helium
plasma.
The increase in halogen emission with increased power suggests th at
halogen emission is lim ited by the number of exciting species available. This is
not the case with lead atom ic and ionic lines. That a change in signal is not seen
with a change in power indicates th a t a sufficient number o f exciting species are
available to excite lead with lower power plasmas.
The increase in background a t higher powers yields g re a ter noise. Since
the background a t the lead lines increased a t higher powers with th e signal
remaining constant, a decreasing signal/background ratio was noted as th e power
was increased.
It was therefore analytically detrim ental to use higher powers
for the determ ination of lead.
Halogen signal did increase a t higher powers.
However, th e magnitude of the noise increased more rapidly than th e halogen
signal again indicating a decrease in S/B. This indicates th a t no advantages were
to be obtained a t higher powers in th e range studied for halogens unless steps are
taken to e x tra c t this larger signal from the noise.
E ffe ct of Flow R ate
The e ffe c t of flow ra te of th e auxiliary support gas with respect to
analytical signal was studied. The power was m aintained a t 250W in a ll cases to
maintain plasma integrity a t low flow rates.
The e ffe c t of flow ra te on chrom atographic peak height is shown in Figure
IV-7. The lead signal a t the atom ic line shows little change with respect to flow.
Lead(II) show a doubling of signal a t lower flow ra te s. The dependency of flow
ra te on signal is more pronounced with the halogen transitions investigated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig u re IV-7
Dependence of Flow R ate upon Normalized Signal for Chlorine, Bromine and
Lead. All signals normalized to a maximum response of 1.00.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ci(n)
4795*
>*
“»
'CO
, .
■g
4786A
§ Br(n^
S
P b (I
6608
1500
2000
2500
3000
3500
Flow(mL/min)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
With chlorine and bromine th e largest signals a re obtained a t lower flow rates
followed by an exponential increase in signal as flows are decreased to less than
2.0 L/min.
In all cases a change in flow ra te caused insignificant changes in
intensity of th e plasm a background.
This indicates th a t increased analyte
residence tim e is less im portant for transitions with excited s ta te s of a much
lower energy than th e helium m etastable. With analyte transitions in th e energy
range of th e helium m etastable energy, residence tim e becom es more im portant.
This may possibly be a ttrib u te d to th e relatively low concentrations of th e high
energy species necessary. In a ll cases the low est gas flow ra te possible in term s
of consumption and analytical signal.
Placing th e e ffe c t of power and flow ra te in perspective, th e b est
signal/background ratios w ere obtained using a relatively low power plasm a and
th e lowest possible flow ra te which kept th e plasm a cen tered . As the minimum
flow required becam e g re a te r with increasing power, it was found best to
operate a t 200-250W.
L ateral and Axial Viewing of th e Plasma
A comparision was made of axial and la te ral viewing of th e plasm a. Axial
viewing is observing th e plasm a tube end-on while la te ral viewing observes th e
plasma through the vertically mounted discharge tube. The axial configuration is
depicted in Figure IV-2. The system was modified to fa c ilita te la te ral viewing
by substituting a 54" stainless steel tube for the containm ent tube a t th e o u tlet of
th e T -fitting. This tube was extended beyond th e heated tran sfer block and th e
gas was directed upward by means of 54" elbow.
The containm ent tube was
inserted into th e elbow and passed through th e cavity which was ro tated 90°
upward in. the plane of th e paper.
As before, the insert and stainless stee l
chromatographic tran sfer line term inated 1 to 2 mm from th e plasma. A 10 mm
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
hole was drilled into th e wall of the resonant stru ctu re to fa c ilita te optical
monitoring through th e quartz containm ent tube.
The image was focused
vertically upon th e entrance slit of the monochrometer and extended approxi­
mately 5 mm above and below the slit.
O ptical distances identical to those
described in Table IV-1 are used.
Calibration curves obtained for chlorine as dichlorobenzene a t the 479.5
nm line a re shown in Figure IV-S. The log peak height axis is arbitrary and the
axial viewing d ata have been displaced by one unit for clarification. Although
th e linear ranges and lowest detectable amounts were sim ilar in both cases the
correlation coefficient is much b e tte r when the plasma is viewed laterally. This
may be due to improved focusing of the image or, more likely, th e absence of
noise from th e afterglow region when viewing in this manner. This arrangem ent
was used for all subsequent studies in this work.
Linear Range, Reproducibility and D etection Lim its
With chlorine and bromine linear dynamic ranges exceeded 3 orders of
magnitude. Calibration curves were constructed in the range of 19 ng to 19 u g
chlorine and 36 ng to 36 vg bromine.
The lower end of these ranges was
extended with th e dynamically background corrected system and is discussed in a
la te r section of this chapter.
With lead, th e concentration lim its covered a
smaller range as a glass lens was used.
intense lead line.
This precluded th e use of the most
Also, the 560.9 nm ion line of lead is within a region of
approximately 40% of the maximum response with the Hammatsu R212 photo­
m ultiplier tube. The upper lim it of the lead calibration curves is lim ited by the
amount of analyte entering the plasma. A physical disturbance of the plasma is
visible a t the point where the calibration curves deviate from linearity.
working curves are shown in Figure IV-9.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
These
F igure IV-8
C alibration Curves for Chlorine using Axial and L ateral Viewing of the Plasm a.
Triangles indicate la te ral Viewing and diamonds indicate axial viewing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Intensity
Log
Log Mass Cl (gm )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig u re IV-9
Calibration Curves for Chlorine and Bromine. Triangle indicates chlorine signal
and diamonds indicate bromine signal.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
-8
-7
-6
"5
-4
Log Mass(gm)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
The reproducibility of each elem ent was determ ined. These values and the
level a t which they were determ ined a re shown in Table IV-4.
In all cases
acceptable values w ere obtained.
D etection lim its are shown in Table IV-5. D etection lim its for the halogens
a re comparable to those in th e literatu re (35). Atomic lead detection lim its are
considerably worse than those in th e lite ra tu re . The reason fo r this is not fully
understood a t present but may be due to a combination of th e previously cited
reasons, plus problems with tetraeth y llead degradation within th e stainless steel
transfer line. No comparison is available for detection lim its a t th e lead ion line
with a helium MIP.
E ffect of Oxygen and Hydrogen on A nalytical Signal Intensities
Many workers have utilized small percentages of added gases such as
oxygen, nitrogen and hydrogen to scrub containm ent tube walls of deposits and to
enhance analytical signals (35, 38). From past work it is unclear what caused
signal enhancem ent. In order to further investigate these particular phenomena,
as applied to the m oderate power He MIP, a system atic study of various gases on
elem ents of different excitation potential was pursued.
While oxygen has been shown to strip carbon deposits from the walls of
containm ent tubes, these studies indicate th a t th e chlorine ion signal is de­
pressed by the addition of oxygen even in small amounts as shown in Figure IV10.
A small amount of hydrogen added to th e helium support gas served to
alm ost double the analytical signal of chlorine with no measurable change in
background noise.
With bromine signal enhancem ent was less pronounced and
lead emission was unaffected by hydrogen addition.
Plots of emission versus
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
Table IV-4
Reproducibility
Elem ent
A nalytical Wavelength(ff)
% R elative Standard Deviation
Br(II)
4785.5
5.9
Cl(II)
-<$734.5
4.2
Pb(II)
5608.8
5.9
Pb(I)
4057.8
5.8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
Table IV-5
D etection Limits
*
D etection Lim it (pg/sec)
L iteratu re
D etection Lim it (pg/sec)**
Br(II)
30
34(35)
C1(II)
78
45(35)
Element
46*
Pb(II)
2610
-
Pb(I)
346
2.3(35)
* axial viewing of th e plasma
** Numbers in parenthesis indicate referen ce number.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Figure IV-10
E ffect of Oxygen on Chlorine Signal.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
p u r e
H e
0.6% 0
1.5% O
E ffect of O xygen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
hydrogen flow ra te are seen in Figure IV-11.
Chlorine and bromine signai
enhancem ent was seen only a t relatively low H2 flows and signal depression was
seen with 1 ^ flows in excess of 4 mL/min (He-2000 mL/min).
The reason for th e differences in signal enhancem ent for th e various
elem ents is unclear a t present. Each of th e elem ents under consideration form
volatile species with hydrogen (HC1, HBr and PbH^). A tradeoff betw een analyte
volatility and depletion of helium high energy sta te s with added hydrogen may be
taking place. This would explain the decrease in halogen emission a t higher H2
concentrations since the high energy He molecules may be involved in halogen
emission a t the analytical lines utilized. A lternatively, increased production of
high energy helium sta te s exciting the halogens a t low hydrogen flows may be a
possible cause. As w ith the first suggestion, the decrease in halogen emission a t
higher H2 concentrations may be due to a depletion in th e number of high energy
helium sta te s due to reactions with the hydrogen species. Regardless, th a t these
possible e ffe c ts do not a lte r the emission of th e lower energy lead transitions
indicates differences in excitation mechanisms for lead and th e halogens if an
im portant change in plasma characteristics is taking place.
Application of Background C orrection Techniques for th e Reduction of
Minimum D etectable Amounts of Halogen
The d ata discussed to this point have all involved single channel monitor­
ing. Although these d ata yielded relatively low detection lim its, th e minimum
detectable am ounts have been g reater than w hat might be expected. The reason
for this was instability in background emission.
While noise calculated over a
large tim e interval (16 pts. - 3.2 sec.) is relatively small, d rift in continuum
intensity on the tim e scale of the chrom atographic peaks (10 sec.) presented
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig u re IV-i 1
E ffe ct of Hydrogen on Chlorine, Bromine and Lead Signal.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.5
>
CO
c
<D
.
*
1.0
c
T3
N
CD
E
o
z
Pb(
0 .5
•iBr
Cl
0
5
10
15
20
25
H2 Flow (mL/min)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
difficulties in resolving the analytical signal from the baseline noise.
An
example of this is shown in Figure IV-12 where the detection lim it is calculated
to be 63 pg/sec of Cl (-0 .7 ng) but th e sm allest amount which could possible be
seen is approxim ately 3 ng providing the retention tim e was precisely known.
For this reason offline dynamic background correction was investigated to
reduce the e ffe c t of th e varying background.
A system designed in this laboratory provides spectral background correc­
tion by rotating a re fracto r plate mounted a fte r the en tran ce slit to provide
background correction. By rotating the re fracto r plate 45° from perpendicular
to the light beam , approximately 2 R of spectrum could be scanned (71). A data
point a t the analytical wavelength was then taken by rotating th e refracto r plate
perpendicular to the light path.
Following th e conversion of each signal to a
digital representation, th e background was subtracted from th e analytical signal.
This dynamically background corrected signal was proportional to the analytical
signal and independant of continuum intensity changes within the tim e con­
strain ts of the system (1 sec/plotted pt.). D ata could be altern ately taken a t the
ra te of 2 or 3 points/second.
For more details on this system , the reader is
referred to reference 71.
The optical configuration precluded la te ral viewing of th e plasma. With an
image enlargem ent of approximately 10X a t the entrance slit, light throughput
was poor causing th e signal to be shot noise lim ited.
B etter results were
obtained viewing the plasma axially. The shot noise lim itation was overcome by
using a 150 urn entrance slit. Noise and signal with slit widths of 25, 100 and 150
y m are plotted in Figure IV-13.
The signal increased linearly with light
throughput while noise did not increase until the slit width was increased to 150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure IV-12
Chromatogram of 20 ng Chlorine Illustrating Background D rift.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
•8
•K
2
^ :><d
:£
F
•s
A)!SU3)U|
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F igure IV-13
E ffect of Slit Width upon Signal and Background. Triangles indicate analytical
signal diamonds indicate level of noise.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o
CO
Std Dev Noise
o
CM
o
O
O
o
Slit Width ( u rn )
in
o
in
o
o
o
co
o
o
oCM
O
O
O
B 0 JV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
pm indicating a transition to source noise lim itation of th e system .
For this
reason, the widest slit was utilized.
Enhanced minimum detectab le am ounts were seen with th e background
corrected system as shown in chrom atogram s of 3.1 and 9 A ng of chlorine in
Figure IV-l'f. A comparison of minimum detectable levels is seen in Table IV-6 .
Similar results were obtained w ith bromine. Calibration curves for Br and Cl are
shown in Figure IV-15.
D.
CONCLUSION
A system for the operation of a He-MIP a t m oderate powers has been
designed and investigated for use as a gas chromatographic d e tec to r. The most
im portant aspect of operation was dissipation of heat throughout th e system .
The dependance of analytical signal with changes in applied power and flow
ra te was shown for atom ic lead, ionic lead, chlorine and bromine.
These
param eters were most im portant w ith th e halogens and th e e ffe c t of changes in
flow ra te on analytical signal th e m ost pronounced. Similarly, an investigation
of th e e ffe c t of added hydrogen to th e plasma support gas produced enhanced
signal for the halogens and no e ffe c t upon the lead signals. D ata in both cases
suggest a possible alteration in He high energy states within th e plasma.
Baseline drift on the tim e scale of the chrom atographic peak widths
increased minimum detectable am ounts. This problem was substantially reduced
by the application of background correction.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F igure IV-14
Chromatograms of Chlorine a t the 63,
6
.3 and 3.2 ng Level w ith th e Background
C orrected System. Scaling for 6.3 and 3.2 ng signals is am plified 10 x.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A jisu aju i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F igure IV-15
Calibration Curves for Chlorine and Bromine with th e Background C orrected
System. Triangles indicate chlorine signal diamonds indicate bromine signal.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
4
CD
c
3
i
to
2
&
1
0
9
8
7
-6
-5
■4
Log Mass(gm)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table IV-6
Comparison of Minimum D etectable Levels (ng)
Element
D etection
Limit (pg/sec)
Continuous Wavelength
System (ng)
Background C orrected
System (ng)
Cl
78
20.7
3.1
Br
30
18.9
7.1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
V.
SUGGESTIONS FOR FUTURE WORK
Electrotherm al analyte vaporization into a He-MIP has shown promise as a
rapid method for the determ ination of halogens. However, instabilities created
during the vaporization stage are significant.
Enhanced results would be
obtained if a rapid method of background subtraction were employed.
The moderate power MIP is relatively new and more ch aracterizatio n of
this plasma source needs to be done. Solution nebulization for th e determ ination
of halogens should be considered since the m oderate power plasmas show g reater
tolerance to sample loading. Also, the spatial ch aracteristics of th e plasm a are
le ft unexplored. This should be investigated since evidence indicates a change in
plasma characteristics with variation of power.
Finally, a project to be pursued is th e im provem ent of optics w ith the
background corrected system . Improvements of th e spectrom eter f-num ber by
employing a larger grating would allow g reater light throughput.
More rapid
background correction may reduce even more the susceptability of the system to
baseline d rift.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
VI.
1.
LITERATURE CITED
B.A. Rhodes and B.Y. C ro ft, "Basics of Radiopharmacy", C.V. Mosby, St.
Louis, 1978.
2.
3.C . Giddings and R.A. K eller, "Advances in Chrom atography", Vol.
6
,
M arcel Dekker, New York, 1968.
3.
3.1. Hallies, D.F. Pinnington and A.3. Handley, Anal. Chim. A cta, 111 201
(1979).
4.
D. Friedm an and P. Lombardo, J.A.O.A.C., 58 703 (1975).
5.
V. Zitco, 3. Chrom atogr., 81^ 152 (1973).
6
.
R.C. Hall, 3. Chrom atogr. Sci., 12 152 (1974).
7.
B.E. Pope, D.H. Rodgers and T.C. Flynn, 3. Chrom atogr.,
134 1 (1977).
8.
D.3. Douglas and 3.B. French, Anal. Chem., 53 37 (1981).
9.
R.S. Houk, V.A. Fascel, G.D. Flesch, H.3. Svec, A.L. Gray and C.E. Taylor,
Anal. Chem., 52 2283 (1980).
10. A.L. Gray, Analyst, 100 289 (1975).
11. A.3. McCormick, S.C. Tony and W.D. Cooke, Anal. Chem., 37 1470 (1965).
12.
C.K. Mann, T.3. Vickers and W.M. Gulick, "Instrum ental Analysis", Harper
and Row, New York (1974).
13.
H.A. Strobel, "Chem ical Instrum entation:
A System atic Approach", p.
397, Addison-West Publishing Co., Reading, M assachusetts.
14.
R.M. Barnes, CRC C rit. Rev. Anal. Chem., 7 203 (1978) .
15.
S. Greenfield, H.M.D. McGeachin and P.B. Smith, T alanta, 23 1 (1976).
16.
V.A. Fassel and R.N. Kniseley, Anal. Chem., 46 1110A (1974).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
17. V.A. Fassel and R.N. Kniseley, Anal. Chem., 46 11.55A (1974).
18. I.S. Krull and S.J. Jordan, Am. Lab., 12 (10) 21 (1980).
19. J.W. Carnahan, K .J. Mulligan and J.A. Caruso, Anal. Chim. A cta, 130 227
(1981).
20.
J.F . Alder, Q. Jin and R.D. Snook, Anal. Chim. A cta, 123 329 (1981).
21.
J.F . Alder, Q. Jin and R.D. Snook, Anal. Chim. A cta, 120 147 (1980).
22.
D.E. Nixon, V.A. Fassel and R.N. Kniseley, Anal. Chem ., 46 210 (1976).
23.
O. Rose, W.R. Heineman, J.A . Caruso and F.L. Fricke, Analyst, 103 113
(1978).
24.
J.W. Carnahan and J.A . Caruso, Anal. Chim. A cta, 136 261 (1982).
25.
A.T. Zander and G.M. Hief je, Appl. Spectrosc., 35 357 (1981).
26.
H. Kawajuchi, T. Ito and A. Mizuike, Anal. Chim. A cta, 122 75 (1980).
27.
K. Tanabe, H. Haraguchi and K. Fuwa, Spectrochim. A cta, 36B 119 (1981).
28.
D.L. Windsor and M.B. Denton, J . Chrom. Sci., 17 492 (1979).
29.
R.C. Fry, S.J. Northway, R.M. Brown and S.K. Hughes, Anal. Chem., 52
1716 (1980).
30.
S.K. Hughes and R .C. Fry, Anal. Chem., 53 1111 (1981).
31.
C.I.M Beenakker, Spectrochim . A cta, 32B 173 (1977).
32. A.W. Zaidel, V.R. Prokof'ev, S.M. Raickii, V.A. Slavyi and
E.Y.Shreider,
"Tables of Spectral Lines", Plenum Press, New York, (1970).
33.
R. Mavrodineanu and H. Boiteax, "Flame Spectroscopy", J . Wiley, New
York (1965).
34.
L.E. Boos and J.D . Winefordner, Anal. Chem., 44 1020 (1972).
35.
S.A. Estes, P.C. Uden and R.M. Barnes, Anal. Chem., 53 1829 (1981).
36.
D.T. Bostock and Y .J. Talmi, J. Chrom. Sci., L5 164 (1977).
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
100
37. C.A. Bache and D .J. Lisk, Anal. Chem., 39 786 (1967).
38. C.A. Bache and D .J. Lisk, 3.A.O.A.C., 50 1246 (1967).
39. C.A. Bache and D .J. Lisk,, J . Gas Chromatogr.,
40.
6
301 (1968).
B.D. Quimby, M.F. Delaney, P.C. Uden and R.M. Barnes, Anal. Chem., 51.
875 (1979).
41.
K .J. Mulligan, F.L. Fricke and J.A . Caruso, Analyst, 105 1060(1980).
42.
B.D Quimby, M.F. Delaney, P.C . Uden and R.M. Barnes,Anal.
Chem., 52
259 (1980).
43. K. Tanabe, H. Hariguchi and K. Fuwa, Spectrochim. A cta, 36B 633(1981).
44. A.P. Gray, C .P. Cepa, I.J. Solomon and D. Aniline, J . Org. Chem., 4_1 2435
(1976).
45.
A.S. Kende, J .J . Wade, D. Ridge and A. Poland, J . Org, Chem., 39 931
(1974).
46.
B.A. Schwetz, J.M. Norris, G.L. Sparschu, V.K. Rowe, P .J. Gehring, J.L.
Emerson and C.G. Gerbig, Adv. Chem. Ser., 120 55 (1973).
47.
L .J. C arter, Science, 192 240 (1976).
48.
A.F. K erst, JF F /F ire R etardant Chem istry, _1 205 (1974).
49.
B. Rouson, Am. Lab. 12 49 (1980)._______ ____
50.
R.K. Small, Total Organic Halide:
O ccurrence, Stability and Process
Control in Drinking W aters, presented a t the 181st A.C.S. National
Meeting, 1981.
51.
a) J.E . Henderson, R.R. Whitney, C.K. Tanaka and S.C. Madden, A Rapid
Method for the Analysis of Halogenated Volatile Organic Compounds in
Soil, presented a t the 181st A.C.S. National Meeting, 1981.
b) U.S. Environmental Protection Agency Method 450.1, Total Organic
Halide, 1980.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
52.
C.I.M. Beenakker, P.W.3.M. Boumans and P.3. Rommers, Philips Technical
Review, 39 65 (1980).
53. W.B. Robbins and 3.A. Caruso, Anal. Chem. 51. 889A (1979).
54. O. Rose, Ph.D. Thesis, University of Cincinnati, OH, 1977.
55. T.L. Smith and B.N. Whelihan, Text. Chem. Lab., 102 35 (1978).
56.
T. Bellar, private communication, United S tates Environmental P rotection
Agency, Cincinnati.
57.
R.M. Dagnall, T.S. West and P. Whitehead, Anal. Chim. A cta, 60 25 (1972).
58.
3.P.3. van Dalen, P.A. de Legenne Coulander and L. de Galen, Anal. Chim.
A cta, 94 1 (1977).
59. K.R. Fallgalter, V. Svoboda and 3.P. Winefordner, Appl. Spectrosc., 25 347
(1971).
60.
R.K. Skogerboe and G.N. Coleman, Anal. Chem., 48 611A (1976).
61.
L.G. Matus, C.B. Boss and A.W. Riddle, Coupling of Microwave Energy to a
Resonant Cavity, presented a t th e 33rd Pittsburgh Conference, 1982.
62.
H. Kawaguchi, M. Hasegawa and A. Mizuike, Spectrochim. A cta, 27B 205
(1972).
63.
3. Burman and K. Bostrom, Anal. Chem., 51 516 (1979).
64.
R.D. Deutsch and G.M. H ieftje, "Instrum ental and O perational C h aracter­
istics of an Atmospheric Pressure Microwave Induced Plasma", paper #222,
8
th FACSS Meeting, Philadelphia, 1981.
65. M.A. Eckhoff, private communication, Pfiser, Groton, C onnecticut.
66.
3.P.3. van Dalen, P.A. de Legenne Coulander and L. de Galen, Spectrochim .
A cta, 33B 545 (1978).
67. R.C. Weast, editor, "CRC Handbook of Chem istry and Physics", 33rd
edition, CRC Press, Cleveland, 1974.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
68.
A. Bolla-Kamara and E. G. Codding, Spectrochim. A cta, 36B 973 (1981).
69.
D.L. Haas, 3.W. Carnahan and 3.A. Caruso, Appl. Spectrosc., 37 82 (1983).
70.
C.I.M. Beenakker, Spectrochim. A cta, 31B 483 (1976).
71.
M.A. Eckhoff, Ph.D. Thesis, University of Cincinnati, 1982.
72.
3.P. M cCarthy, Ph.D. Thesis, University of Cincinnati, 1982.
73.
A. Savitzky and M.3.E. Golary, Anal. Chem., 36 1627 (1964).
74.
A.R. Striganov and N.S. Sventitskii, "Tables of Spectral Lines of N eutral
and Ionized Atoms", Plenum Press, New York, 1968.
75.
C.3.Seliskar,privatecom m unication,U niversityofCincinnati,Cincinnati,O hio.
76.
V. Bohmer and K. Worsdorfer, Appi. Spectrosc., 31. 334 (1977).
*
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX I
NUGNOMER is a program w ritten for th e UC/UNC 8080 m icrocom puter.
The program is designed to take chrom atographic d ata a t a user specified tim e
and ra te .
Provision is made for a 25 point Golay smooth, integration, peak
height and width determ ination for chrom atographic peaks.
The program also
calculates noise, signal to background ratios and d etection lim its. D ata may be
presented on eith e r a printer or a plotter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16 DIKRK1B)
28 DI?m<£^3)jDUS2(SB88)jMtB3(2ee8)
38 01851(38)
43 Gl(8)=5i75jGK 1>=-S53;GK2>=-lSS;GlfS5=-33;G!<4>=€5
56 G1(5)=14?;G1(6>=££2;G1C8>=£87;5i(3>=3c2;51C 16?=S7
53 G1<li )=42£;GK 12>=447;G1< 13>=462jGK 14 >=46?;G1( 15 >=462
54 5Kl£)=44?;5KI?>=422jG!(18>=®7;GK19>=2£2;Gl(28>=28?
56 GK21 :j=£££;GK£E'>=147jbl<£3>=62;Gi(34>=-33;GK £5>=-13S
58 Gl<£6>=-253
68 ERfiSCl)
78 PRINT*MUT: # OF POMS'
88 INPUTS
% tm n
188 PRIKFiHPUT;
TIE SPfiN (SECf
116 WUTft?fT
126 B=(T/S^?*.888ii5-.0243S?>/.88412i*.5
138 M TH EH PRIH Tm TRY M W i'jE ®
146 0*5(1} .
156 PRIH TW ?S£ YOU INJECTING fiHD H3» f&SKNG) ?*
168 IHFiiTas,X
176 N=13:I1=8;1£=8
188 PRiNPTRES SPRCE Bfi* TO BEGIN ORTft OyiSniOH*
198 J=?EEK(2>
m IF J= t£7T i£N t%
218 IFJ=32TEH£38
£26 5OT0138
238 MHT-YOO'E CTTIH5 HNWI *
£46 FO?I=iT05
258 DOIT<GIPHTfYK I ),B7 >
268 RK8=1TG8
£?6 ESTR
288 HEXTI
m . w y o u 'r e doe e h i e n v
3K6=£5
318 r o i K a m ^ G K e i ^ Y i a ) , ® ^ } )
328 FLtI=l3T£S-12
338 S2CI)=-G2(I)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34? o n
358 K=S-£4
366 D0n(SEJSCHIG3(13)JK1Ll^l)
378 L&tl-Sl
38? F0Ri=!3T6(S-!2>
398 S£D=52U>-51
460 tEXTI
41BEHISC1)
438 CttFIRPLOT OHSCR)jCOHFIGCPRIHT OKSCR).
438 PL0T<8,.1,.4>
448 PLQK1,.9,.4)
m PLGTCl,.?,!)
460 PL0TC1,.!,:)
47s PL07(],.!,.4)
475 P*JT<8,.43,.34>
488 ?LOr<l,.43,.34Vl)r ril£ <SE£>*
490 F9SHT04
506 J=?*I/4
5® PUJKBf.0BHv5f.3?>
518PLOT(i,.0BH/5f.37XlM
5EB HEXTI
53? FCfi!=8T03
548 PUir(0,.l+W ,.41)
558 PL0r«,.l+I/5f.43>
5SS r^XTI
578 Fl£I=8TQ4
555 PLQTC8,8,.45*1*. i£5 5
588 PL0T<i,8,.45*!*.iE5X!M<LM/4>
5ft rSSTI
m F«!=§704
616 PUSR8,.!1,.45HM25>
m FLOTC13.13,.45*1*.i25)
636ffiul
646 PLuT<6,.!+16,4ft/=^.5iG 3(18i/l2)
6 a?
666 P l P C 1 , . W I * .3*;,.45».5*Sl!M 2>
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
678 HEXTI !
688 IFy=iTHEKS38
698 a iR S im ,. 1 ^ . 8 3 , .4g*.5*&13)42) '
708 PLOT( 1,a,.£ > "S ^ie;i^I5H !ljL = L F T ljI= L ^(?H T jR = i^iiT "
718 J=PEEK<83
728 IFJ=1£7THEN718
738 CURSOKB,.i<H*.8^i .48t.5»G3(HVLS)
748 IFJ3BTHEMHM
753 IFJ=83TIS#Mft-18
758 IFJ=76TS»H4-1
770 O ^ O R \I,.m 8 ^ ,.4 S + .5 ^ (H )/v 2 )
788 IF>73TfSei®
788 I?J=X7©«68
8 s 5070718
818 I?F>6MH7ie
826 P=s
m P L 0 T (8 ,.H P * .^.4 4 )
835 PL0T(i1.W * .8 '5 ,.4 4 )“ s
848 IFY=17HEh876
m SjTu710
858 8=K
878 PL0rce,.i+®*.8£,.4O
875 PL0rU,.i4e*.8/S,.44)£f*
888 IFYslTHEHil96
888
HM£‘
988 FUn<8J. i J,£>E£==bTE?lSj^I^:i_=E7^i5l=a«i5HT|G=®cSfKTii
918 FC£I=P700
888 IFS3C1M21ffiH=Ij M
838 NEm
840 FCm=P7QH
m iF<63CK H SCP3K ^:B:i>^?>i7c-fc=:j:=?!
888m i
mm tem m
888 IRG 3(H )^Q ))=>£^G 8aH 38:i:'X ;^-=:\:=8
990 E f f l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1008 «5=(HE-H1)
1618 PL0T<1,.81#. 3 > W WPTK = •JB /P W S V
% ® *T/S/btT
1020 Pi.OT(! ,. 81,. £4)' IrFUT EVEN § OF POINTS TO H R m
BHJIP1
IS48 F iO T C U .e i,.^ .) '! .^ EVEN # OF POIHFS TO RIGHT8
1858
im\m
m PLflr<e,.ei,.3mF «i&ih=\»/nnsv aj&us /
sef
107b Fierce, j ! ,. £ 4 > “itpur even# o f points to l e f t 8
m
PL0T(8,8,.2), ?s
18ft Ft_0T<8,.81,.2>Pi
l i f t Pi.OT(0„01,,16il!I?PJT EVEN t OF POIKTS TO RIGHT*
1110 ?LAT<M,.i>*?
l i f t PLOT(8„01,.1)81
l i f t FftI=-PiTOCPl-2>Sit?2 .
m ii=ii-ht3<PtI^)/pi
nag f e n
1150 *TOH&T0Ctt-£35SB»
1170 I2 = I£ *5 3 ( f ir j /d > /0 i
l i f t E X IT
l i f t P IM 8 ,.H P * .3 /S I .« * .5 * IM £ >
12ft f ftG K l,.! - ^ .f tS J.45*.5*I24i>
1210 IF Y = lT © Ii? 4 8
l i f t f-3
1238 B = « H 2 M P ^>
1240 B1=I1-W
125? F0RI=-PITOP1-^MP2
12ft fi=P-M/2
12ft Vs¥+(i3CflHWI-Bi>KS3CfiH*fi-B!>
lift HEm
1296 FKI=-81TO01-2)ftSP2
1 ® F -ftL ft
me
1320 tfcXTI
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1338 Sl=S^CV/(Pl+ei-l))
1348 fi=8
1350.
)70rrHH2-H)
1368 feftKGKIHM-Bl>T/S
1376 NEXJI
i m R=S*T/S
1356 P£=P*T7S
1488 ffi=8*T/S
1418 H3=H1*T/S
1428 H4=t£*PS
1438 W5*T«&£
1446 iKSCSHUfrBl
1458 H1=H-'S1
1468 H2=BAPSD
1476 D1=2S1*S/H
i4® m > m
1458 D4M«SWE^
1588 D3=D4*S
1518 8101(8,8,,84)
m
FtOTCI,.8,.84)
1536 F 1 6 T C 3)
1548 2,01(1,8,.3)
m
pmrciA.64)
____________________
15S8 FtOTC6,.39,. 145)
1576 p um i,.ei,.145>
15BB F10T(t,.35,,£o)K,E Mb *,W*
15® F10K1,=15I.24)£H=* .HsEffiC"
m PtMiI.5,.£4)cy=s,M/®:c'
1616 pun,( iI.i5 ,.^ ),,eKBcoREds e e,?2 , w , e , Es x te s E
1636 -PLGTf l,.15,.£)*ffiLF ICISflS m ^ W ^ / S E - a B S *
16® F10TC1,.4 5 ,. 1>ES4f
DUNGi
M M S?
1648 PLGT(l,.83,.®rEI5iT!;H,Efu ;B
1658 R M 1 ,.4 £ ,.6 3 M
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m
PlCT(l,.6f.83»l
1678 ?U sKi,.7?,M W
1688 PLBTa,.05,.g7)-f^c Sfi/flK^SEC1698 FLOT(1,.4£,.07V£
1788 PJjTCl,.6,.87X?3
1718 FL0T(i,.77,.87X>4
1730 Ft-OTc i , . 1 5 ,. 81 ) BR=RD:tHR;F=PLGT :L=LIST; H=S¥FLE;V=¥f£r¥”
1748 Y=8
1758 J=PEEKC£>
1768 IFJ=8£TfEN18£B
1778 IfJ=88IffiN1836
1788 IFM Sf© 088»
171 !FJ=?8T®il8 ®
•
188? IFJ=86T®I187E
isle m t m
m p=ej{4=!SiiMji£=e.jS704i¥
1838 Y=i
1848 CeFISCPLDT Oh’ PLOT)
1858 0 * 1 0 4 ^ 1 ^ 8 P=g{S?i:i >15010178
1878 F=8|ft=13jIl=8|I£^;S}7DK
1888 CfflFIKPMHr ONTTY)
1898 FRMX/HG%!*
1988 P£IKTsEISffT=* .H. E®C CflMTS1
1918 HfflPfi©=Sft/FjDS*StC*
IS® PRINTSI0TR=5,y , “SOUNDS*
• 1938 f lE H r n O B S lK * £0F?£C 12) fiT‘ , F 2 , W , 8 2 , * £ C 0 ® S ‘
1948 f m m s {Oiiirs hTSBS/M^FM/SECfTOS’
19® PHHHEIfflT 541 M > SL<H5'5), ,N11MJ^
19® PRIHTBff£fi
m
m DLCNS) DL<H54S»Mg,D3J>4
COSFIKPRMtt*SC?>
19® 8801758
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
2 722 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа