Spectroscopic Tricks
-
Upload
others
-
View
13
-
Download
0
Embed Size (px)
Citation preview
2.pdfVolume 1: 1959-1965
Volume 2: 1966-1969
Volume 3: 1970-1973
The Cmholic University of America Washington, D. C.
Springer Science+Business Media, LLC
Library of Congress Cataloging in Publication Data
May, Leopold, comp.
Spectroscopic tricks.
Articles from Tricks and notes in Applied Spectroscopy selected
from the periods: 1959-65, 1966-69, 1970-73,
Includes bibliographical references. 1. Spectrum analysis. 1.
Applied spectroscopy. II. Title.
QC450.M38 53 5'.84 67-17377
The material contained in this volume originally appeared in
AppJied SpectroscoP'Y from 1970 through 1973, and is reprinted here
by
permission of the Sodety for Applied Spectroscopy.
ISBN 978-1-4684-2744-8 ISBN 978-1-4684-2742-4 (eBook) DOI
10.1007/978-1-4684-2742-4
©1970-1973 Sodety for Applied Spectroscopy
© 1974 Springer Science+Business Media New York Originally
published by Plenum Press, New York in 1974
AH rights reserved
No pact of this book may be reproduced, stored in a retrieval
system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, microfilming,
recording, or otherwise, without written permission from the
Publisher
FOREWORD
This is the third volume of the collection of new devices,
modifications of existing equipment, and other items of interest of
this nature published in the journal Applied Spectroscopy. These
tricks have proved of value since they first appeared in the
journal in 1959. They give solutions to many problems of workers in
the var ious fields of spectroscopy. For the novice, the use of
ali three vol umes may provide insight into the improvements that
have been made in the instruments and techniques that he is
currently using. The novice may be saved the necessity of
discovering some shortcut that many experienced spectroscopists are
already using.
The contributions in this third volume are selected from the years
1970 through 1973. The subject arrangement is the same as in
Volumes 1 and 2 according to the area of spectroscopy. Those tricks
concerned with the same device are placed together so that the
reader can easily compare them. To maintain the advantages in
herent in a single collection of contributions, the subject index
for this volume is cumulative including the tricks in the previous
vol umes. Both author and journal indices are provided for this
vol ume, the latter citing the original Applied Spectroscopy
citation.
The use of the contributions has been approved by the So ciety for
Applied Spectroscopy, whose cooperation in this matter is
gratefully acknowledged.
Leopold May
EMISSION ANO ATOMIC ABSORPTION SPECTROSCOPY 1.1 Rapid and
Inexpensive Sampling Technique for Emission
Spectroscopic Analysis of Thin Films, 1. Dieleman, A. W. Witmer, J.
C. M. A. Ponsioen, and C. P. T. M. Damen ...... .
1.2 A Computer-Controlled Sampler for Atomic Flame Spec troscopy,
W. Sunderland, R. S. Hodge, W. G. Boyle, and E. Fisher . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .. . .. . . 5
1.3 The Preparation of Metal Ingots for Use as Chemical and
Spectrographic Standards, S. L. Odess and G. S. Golden . . . . . .
9
1.4 Qualitative Analysis of Precipitates by Graphite Filter Meth-
ods, M. S. Wang . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . 13
1.5 An Improved Spectrographic Evaporating Dish, R. E. Rainford . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
15
1.6 A Rotating-Disk Sample Holder for the Sparking of Flat
Metal-Disk Samples, P. E. Walters and T. Monaci .... . . . .
16
1.7 Vented Cupped Electrodes, L. Toft and G. A. Roworth
.................................... 22
1.8 Suggestions and Comments on: "Vented Cup Electrodes." J. B.
Marling . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
23 Reply to Dr. Marling, L. Toft and G. A. Roworth . . . . . . . .
. .. 24
1.9 A Cylindrical Sector Driven by Either Water or Air, J. W.
Mellichamp and L. L. Wilcox . . . . . . . . . . . . . . .. . . ..
25
1.10 A Symmetrical Cylindrical Rotating Step Sector, H. G. Yuster .
........... . . . . . . . . . . . ............... 28
1.11 Prevention of Laser Microprobe Staining of Analyzed Metals, H.
N. Barton and J. Benallo ....... -. . . . . . . . . . . .. . . ..
35
1.12 A Simple Multiport Atomic Absorption Burner Head, M. S. Wang
.... . . . . . . . . . . . . . . . . . . . . . . . . . . . .......
37
vii
viii CONTENTS
1.13 Modification of a Commercial Carbon Rod Flameless Atom izer
to Accept Graphite Tubes, R. W. Morrow and R. J. McElhaney
................................. 39
1.14 Tuning Stubs as an Aid to Coupling RF Energy to Electrode
less Discharge Lamps, W. G. Schrenk, S. E. Valente, and K. E. Smith
.................................... 44
1.15 A Compact Gas Jet for Optica1 Emission Spectroscopy, K. J.
Curry and E. F. Cooley. . . . . . . • . • • . . . . . . . . . . . .
. .. 52
1.16 Electrode Heater, P. B. Adams, E. C. Goodrich, and J. S.
Sterlace . . . . . . . . . . . . . . . . . . . . . . . . . . . . •
. . . . . .. 58
1.17 A Simple Modification of a Flame Photometer for Routine Trace
Potassium Analysis, W. R. Knolle ..... . . . . . . . . .. 59
1.18 Mounting for New Safety Door for Perkin-Elmer Model 303 Atomic
Absroption Spectrophotometer, L. T. Sennello ..... 62
1.19 Selective Filtration in Optica1 Emission Spectroscopy, A.
Szule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . .. .. 65
1.20 Simple Inexpensive Method of Time Resolved Spectroscopy, R. A.
Koehler and F. J. Morgan. . . . . . . . . . . . . . . . . . . . ..
69
1.21 Photoelectric Time Differentiation in Laser Microprobe Optical
Emission Spectroscopy, W. J. Treyt1, J. B. Orenberg, K. W. Marich,
and D. Glick . . . . . . . . . . . . . . . . . . . . . . . .
74
1.22 A Photographic Plate Processing System, T. B. Griswold, W. H.
Dennen, and W. H. Blackburn. . . . . . . . . . . . . . . . ..
81
1.23 A Microphotometer Digital Readout System, R. E. Mason
.................................... 86
INFRARED SPECTROSCOPY 2.1 Microsampling for Infrared and Emission
Analyses, P. W.
H. Schuessler ..•.••...•••....••••....... . . .. 93 2.2 Cold
Pressing Solid Samples in a Wax Disk for Far Infrared
Analysis, M. E. Peterkin . . . . . . . . . . . . . . . . . . . . .
. . . .. 95 2.3 A Manual Rectangu1ar KBr Pellet Press, M. Van
Swaay
and E. M. Winkler . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . .. 97 2.4 An Improved Infrared Microcell, E. C.
Sunas, J. F. Williams,
C. Walker, and D. Kidd .............................. 104 2.5
Infrared Cells for Salt Solutions, W. F. Edgell ............. 108
2.6 Far Infrared Sealed Liquid Cell with Polyethylene
Windows,
A. T. Tsatsas and W. M. Risen, Jr ...................... 111 2.7 An
Inert Infrared Cell for Measuring Quantitative Solution
Spectra of Carbonium Ions and Other Reactive Species, T. J.
Broxton, J. Chippindall, L. W. Deady, and R. Topsom
......................................... 115
2.8 A Simple Evacuable, Double-Beam Infrared Hot Cell As- sembly,
H. W. Wilson ............................... 118
CONTENTS ix
2.9 A Novel Infrared Gas Cell, A. B. Harvey, F. E. Saalfeld, and C.
W. Sink ........................................ 124
2.10 A Diamond-Window Infrared Short Path Length Cell for Corrosive
Liquids, H. H. Hyman, T. Surles, L. A. Quarterman, and A. 1. Popov
................................... 127
2.11 A New Gasketing Technique for Studies with the High-Pres sure
Diamond Anvil Cell, J. R. Ferraro and A. Quat- trochi
........................................... 130
2.12 The Application of the Quartz Crystal Microbalance for
Monitoring Rates of Deposition of High Temperature Species in
Matrix Isolation Infrared and Raman Spectroscopy, M. Moskovits and
G. A. Ozin ........................... 133
2.13 Internal Reflectance Spectroscopy. III. Micro Sampling, A. C.
Gilby, J. Cassels, and P. A. Wilks, Jr. . ............. 135
2.14 Infrared Spectra of Deuterated Solvents, N. L. McNiven and R.
Court ........................................ 148
2.15 Measurement of Aqueous Solution Temperatures in Infra- red
Spectroscopy, M. Cormier and J. L. Thompson ......... 159
2.16 Ultrahigh Sensitivity Detection System for Far Infrared
Spectrophotometers, W. M. Poteet and R. D. Feltham . . . . . . . .
. . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . ..
162
2.17 Derivative Traces in Infrared Fourier Transform Spectros-
copy, M. J. D. Low and H. Mark ...................... 167
2.18 On Resolution Enhancement af Line Spectra by Decan- volution,
A. Goldman and P. Alon ..................... 173
2.19 Negative Skin Sensitization Text with KRS-5, R. P. Oertel and
E. A. Newmann ................................ 177
MASS SPECTROSCOPY 3.1 Trapping Volatiles from GLC for Injection
into a Mass
Spectrometer, M. G. Moshonas and P. E. Shaw ........•.... 181 3.2 A
Simple System for Transferring Air-Sensitive Compounds
into Capillaries from Schlenk Tubes, W. G. Eggerman ........ 184
3.3 Construction of a Leak-Inlet System for the LKB 9000 Gas
Chromatograph-Mass Spectrometer, R. E. Hawk and R. W. Jennings
......................................... 186
NUCLEAR MAGNETIC RESONANCE 4.1 A New NMR Microtechnique, L. V.
Haynes and C. D.
Sazavsky . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . 191 4.2 A Nonbreakable Nuclear Magnetic
Resonance Sample Con-
tainer for Radioactive Materials, L. R. Crisler .............. 198
4.3 A Convenient Device for Removing Dissolved Oxygen from
NMR Samples, N. Mandava ...................•...... 202
x CONTENTS
4.4 A Method for Capping Nuclear Magnetic Resonance Tubes, R.
Foester ..•...............•..•.................. 20S
4.S Nuclear Magnetic Resonance Tube Washer, D. W. Mastbrook and E.
A. Hansen .................................. 207
RAMAN SPECTROSCOPY S.l SampHng Techniques for Raman Spectroscopy of
Minerals,
L. E. Makovsky .................................... 211 5.2
Aluminum Metaphosphate as a Hydrofluoric Acid Resistant
Raman Cell Materials, J. E. Griffiths .............•....... 21S S.3
A CeH for Resonance Rarnan Excitation with Lasers in Liq
uids, W. Kiefer and H. J. Bemstein •.................... 219 S.4
MultipleSampHng RamanCold Cell,J. B. Bates ....•....... 223 5.S A
Windowless Cell for Laser-Raman Spectroscopy of Molten
Fluorides, A. S. Quist ............................... 226 5.6 A
Laser-Raman Cell for Pressurized Corrosive Gas and Liq-
uids, J. C. Cornut and P. V. Huong ..................... 232 S.7
Thermostating Capillary Cells for a Laser-Raman Spectro
photometer, G. J. Thomas, Jr. and J. R. Barylski ...•.... 236 S.8
Low Temperature CeH for Measurement of Raman Spectra,
1. Stokr and B. Schneider ............................ 239 5.9
Variable Temperature Sample Holder for Raman Spectros-
copy, F. A. MiIler and B. M. Haney ........ '" .•........ 243 5.10
A Fumace for Molten Salt Rarnan Spectroscopy to 800°C,
A. S. Quist ....................................... 24S 5.11 A
Simple Furnace for Obtaining High Temperature Rarnan
Spectra, G. M. B~gun .............................. 252 5.12
Modification of a Commercial Argon Ion Laser for Enhance
ment of Gas Phase Raman Scattering, G. O. Neely, L. Y. Nelson, and
A. B. Harvey .....•.•.•...........•...... 256
S.13 Polarized Raman Scattering from Small Singie Crystals, B. 1.
Swanson •...••..•.......................•........ 262
S.14 On "Scrambler Plates" Used to Depolarize Visible Radiation, L.
A. Rabn, P. A. Temple, and C. E. Hathaway ............. 269
5.1S On "Scrambler Plates" Used to Depolarize Visible Radiation P.
R. Reed and D. O. Landon . . . . . . • • . . . . . . . . . • . . .
. . .. 276
S.16 AConstantSpectraiSHtWidthServo,C.D.Allemand ....... 278 5.17 A
Method for EHminating Resonance Fluorescence Ef
fects in Raman Studies of Some High Temperature Vapors: Raman
Spectra of BiG3 from 450 to 800°C, P. T. Cun ningham and V. A.
Maroni .....................•..... 283
5.18 Computer Time Averaging of Laser Raman Spectra for
Matrix-Isolated Species, D. A. Hatzenbuhler, R. R. Smard zewski,
and L. Andrews . . . . . • . . . . . . . . . . . . . . . . . . . .
. .. 287
CONTENTS
ULTRAVIOLET ANO VISIBLE SPECTROSCOPY 6.i Construction and Use of
Reflecting Multiple-Pass Absorp
tion Cells for the Ultraviolet, Visible, and Near Infrared,
xi
J. H. Gould ...................................... 293 6.2 A Long
Path Length, Low Temperature Multiple Traversal
CeH, A. Biernacki, D. C. Moule, and J. L. Neale . . . . . . . . . .
. .. 302 6.3 Microspectrophotometer CeHs of Fused Construction, W.
T.
Camall and P. R. Fields .............................. 307 6.4 An
Investigational Technique for the Behavior of a Con
taminated Optical Surface in the Near Ultraviolet-Visible Near
Infrared, W. W. Moore, Jr., P. W. Tashbar, and G. L. Bums
......................................... 310
6.5 Optimum Reference Wavelength Selection in Multi-Wave length
Spectrophotometry of Turbid Media, J. E. Stewart . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
317
6.6 Visible Spectroscopy of the Aging Process in Passive N2-C02
-
He-Xe Laser CeHs, J. W. Mellichamp and J. C. Bickart .........
323
X-RA Y SPECTROSCOPY 7.1 A Simple, Fast Technique for the Sample
Preparation of
Composite Metal Powders for Analysis by X-Ray Fluo- rescence, B.
Brachfeld .............................. 329
7.2 Modified Micro Sample Support for X-Ray Emission Spec-
trography, D. A. Nickey and J. O. Rice .................. 331
7.3 An Improved Liquid CeH Cap for X-Ray Fluorescence Analysis, S.
Bonfiglio ................................ 333
7.4 Adaptation of the X-Ray Milliprobe for the Examination of Small
Single Crystals Obtained from Lunar Samples, H. T. Evans, Jr. and
R. P. Christian ......................... 334
7.5 Selected Area X-Ray Luminescence Spectroscopy with the X-Ray
Milliprobe, S. E. Sommer ....................... 339
MISCELLANEOUS 8.1 A Dissolving Technique for Thin Platelet
Preparation from
Bulk Single Crystals, A. J. Fischinger .................... 343 8.2
A Simple, Inexpensive, Versatile Optical Bench for Spectro
scopic Research, V. Svoboda, W. P. Townsend. and J. D. Winefordner
...................................... 349
8.3 Reduction of Grating Spectrograms, T. Goto, M. S. Gautam, and
Y. N. Joshi ................................... 351
8.4 A Simple Method for Reducing Astigmatism from Off Axis Concave
Spherical Mirrors, D. W. Steinhaus and B. Brixner
.......................................... 356
xii CONTENTS
8.6 Pen Adaptor for Recording Spectrometers, J. P. Luongo
......................................... 367
8.7 A Convenient Method for Vacuum Deoxygenation of Elec tron Spin
Resonance Samples, K. Tanaka, R. P. Quirk, G. D. Blyholder, and D.
A. Johnson ....................... 370
8.8 An Internal Standard for Electron Spectroscopy for Chem ical
Analysis Studies of Supported Catalysts, J. L. Ogilvie and A.
Wolberg .................................... 372
Applied Spectroscopy Reference Index ....................... 379
Author Index ........................................... 381
Cumulative Subject Index .•...............................
385
SECTION 1
Rapid and Inexpensive Sampling J .1 Technique for Emission
Spectroscopic Analysis of Thin Films
J. Dieleman, A. W. Witmer, J.C.M.A. Ponsioen,* and C. P. T. M.
Damen
Philips Research Laboratories, Eindhoven N etherlands
Methods of analyzing thin solid films have recently beenreviewed by
Pliskin and Zanin.l Their review shows that a good method for
obtaining a fairly rapid survey analysis of a large number of
impurities is emission spectroscopic analysis. When used for
samples of which a suffici<,nt quantity is available for
analysis, this method yields excellent sensitivities for most of
the more common impurities. }<'or example, Addink2 de-
* Present address: Central Laboratory Light Division, N. V.
Philips' Gloeilampenfabrieken. Eindhoven -N etherlands.
1. W. A. Pliskin and S. J. Zanin, in Halldbook of Thill Film
Technology, L. J. Maissel and R. Glang, Eds. (McGraw Hill, New
York, 1970), Chap. Il.
2. N. W. H. Addink, D.e. Arc AllalY8i8 (MacMillan, London,
1971).
2 SECTION 1
scribed a m<,thod in which six 1O-mg exposures are made of a
material to determine the concentration of 70 ele ments with a
limit of detection varying from 0.1 to 100 ppm by w<,ight
depending on both the nature of the samplc and the element
considered. In analyzing thin films, on<, is mostly
confront<,d with the problem that less matprial is available
than needed for reaching this sensitivity. Of course, a film with
an area of 1 cm2, a thickn('ss of 10 JI., and a denslty of 10 g
cm-3 would furnish thp rpquired 10 mg per exposure, but of ten the
films havp thicknesses of 1 JI. or less and their area and density
is lPHs than in this example. Knowll ways to increasc thc amount of
matl'rial nel'dcd to improve the sensibvity to thp lpvel mentionro
above are to prepare thin films of largi' an'a or to di'posit thick
films under virtualIy thp samp conditions as used for the thin
films. In both casps thc film is removed from the substrate and
transfcrroo to the graphite electrodcs used for emisslOn
spectroscopy. This is of ten a h'dious and time-consum ing
procpss, which almost illevitably causes loss of film material
and/or contamination with impurities from thc substratp. To
circumvent interference from the sub stratp matprial, the use of a
graphite substrate has been introducpd bccause this is easily
obtained in purity gradcH comparablp to those of the graphite
electrodes.3 , 4
Although this may certainly be a step in the right direc tion, it
dops not sol ve the problems of time-consuming removal procedurps
and the risk of contamination dur ing this pIOccdurp.
Thpsc considprations promptffi us to investigate the possibility of
pxtending the graphite-substrate technique to introduction of the
graphite electrodes, used for emission spectroscopy, as substrates
for collection of samples for analysis. Obviously, such a procedure
would avoid alI the disadvantages mentioned above. As re gards
sensitivity, \Ve chose three different approaches.
3. L. D. Shubin and J. H. Chaudet, Appl. Spectrosc. 18, 137
(1964).
4. J. D. Nohe, Appl. Spectrosc. 21, 364 (1967).
EMISSION ANO ATOMIC ABSORPTION 3
Whcn the scnsitivity was sufficient for the impurities of
intcl'<'St with thc film thickness used in practice, there was,
of coursc, no problem. In the case of insufficient sensitivity we
cither collected thin films on the same elcetrodes ovcr as many
dcposition runs as necessary to get the rcquired film thickness or,
if appropriate, pre parcd a thICk film of the rt'Quircd thickness
in one deposition.
The tcchniquc was applied to investigate the depend ence of the
purity of evaporatcd gallium selenide layers on the material used
for constructing the evaporation sources. Addink2 described in his
book a general method of sp('Ctrochemical analysis. The same
principles were applied to the analysis of gallium selenide. The
results obtained in this way were checkcd by means of wet chemical
analysis, after which the method was adapted for our special
purpose. Three different sources were used: (1) silica, Pursil
quality from Quartz et Silice, Francej (2) boron nitride, boralloy
quality from Union Carbide, U .S.A. j and (3) glassy carbon from
Vitreous Carbon, England.
Emission spectroscopic graphite electrodes used for sampling and
analysis were R.W. 0871 electrodes from Ringdorff, Germany, having
a price of about one D.M. per electrode (see Ref. 2). Three of
these electrodes were used in each run. They were fixed in holes in
a glassy carbon holder in such a way that only their top face was
exposed. Their positions were such that they were representative
samples of the films we were interested in.
Before each evaporation run the sources and the glassy carbon
holder were cleaned by boiling in nitric acid (p.a. quality of
Merck, Germany) in silica beakers, rinsed with demineralized water,
and dricd. The graph ite E'lectrodes did not need cleaning as they
were taken directly from the package and did not touch anything
el8O. Mounting of the various parts was performcd using clean nylon
gloves and a pair of clean Teflon tweezers.
Impurities in the gallium selenide to be evaporated were at most
at, or below the ppm range, except 10 ppm
4 SECTION 1
Element Detection Silica Glassy Boralloy limit source carbon source
source
B 10 10 Si 15 150 Hg 15 Sn 20 Al .5 Fe 15 Mo 5 Cu 0.5 Ag 0.3
• All numbers in ppm by weight.
of carbon and oxygen. Evaporation was performed in a Balzers BA510
evaporation equipment using dc heating of a molybdenum wire-wound
fumace enclosed in a silica envelope.
The coated electrodes were stored and transferred in dust-free,
clean Teflon boxes.
Table 1 shows some results of the emission spec troscopic analyses
for elements of interest to us. As the diameter of the electrodes
was 0.7 cm and the density of the layers was about 5 g cm-a, the
amount of material present on each of the three electrodes was 1.4
mg. The detection limits given in Table 1 were obtained by means of
a method based on thc principles described by Boumans.5
The results presented in this table clearly show the power of the
new technique: quite satisfactory sensi tivities, no need for
development and use of expensive film removal techniques, and thus
no interference due to impurities introduced during film
removal.
5. P. W. J. M. Boumans, Z. Anal. Chem. 220, No. 4, 241 (1966); 225,
No. 2,98 (1967).
EMISSION ANO ATOMIC ABSORPTlON 5
A COInputer-Controlled Sampler for J .2 Atomic Flame
Spectroscopy
William Sunderland, Robert S. Hodge, Walter G. Boyle, and Eugene
Fisher
Lawrence Livermore Laboratory, University of California, Livermore,
California 94550
An on-line computer can be very valuable for data acquisition and
reduction in atomic Hame spectroscopy (AFS). When the computer is
also used to control the sampling process, this total system offers
the advan tages of ease of sample handling, repeated standardiza
tions, and multiple analyses with averaging and randomization of
the sampling sequence.1 Therefore, a solution sampler that is
capable of rapid response and dependable operation and can be
simply programmed for computer control has been designed and
constructed.
The fundamental mode of operation of the sampler is the rotation of
the sampling tube itself instead of the usual method of moving the
sample wheel holding the containers. Thus, the sampler can be
simply and pre cisely controlled by the computer with only four
com mands, which raise or lower the sampling arm and rotate the
stepping motor clockwise or counterclockwise.
The principal components of this sampling device (Fig. 1) are a 200
steps per revolution stepping motor, a double action air cylinder,
a four-way electric air valve, a nylon sampling arm, a TeHon
aspirating tube, and a sample holder with containers arranged in a
circle around the pivot point of the sampler arm.
The stepping motor, upon receiving the proper com mand from the
computer, moves the sampling tube to the desired sample cup. The
sampling tube is raised or lowered by first energizing a
computer-controlled relay. Power is thereby switched to the
electric air valve,
1. W. G. Boyle and W. Sunderland, Anal. Chem.42,l403 (19iO).
6 SECTION 1
FIG. 1. Automatic sampling device.
which injects air from a compressed air source of about 40 psig
into the air cylinder to push the piston either up or down.
In order to have both the up and down and turning motions originate
from the center of the sample holder, a double action driving
mechanism is employed. This driving mechanism consists of two
shafts, one inside the other, both of which turn together by rneans
of a key and keyway. The inside shaft, which conveys the up and
down motion, moves along the key.
The inside shaft is fabricated from a 1O.5-in. length of 0.25-in.
diam stainless steel rod. A keyway 3 in. long, 0.0625 in. wide, and
0.1875 in. deep is machined into the rod, starting 1 in. from the
top. A horizontal bar (1.1875 X 0.5 X 0.1875 in.) is used to
connect the piston rod of the air cylinder to the inside shaft. One
end of
EMISSION ANO ATOMIC ABSORPTlON 7
the bar is attached firmly to the piston rod with a nut and lock
washer, while the other end is attached to the inside shaft through
the use of a %0 screw and a nylon bushing, so that the shaft will
be free to turn.
The outside shaft (Fig. 2) is fabricated from a 6-in. length of
l-in. diam nylon rod which, with the excep tion of the top 0.75
in., has been machined to a diam eter of 0.5 in. A hole, 0.2656
in. in diameter, is drilled through the center of the entire length
of the rod in order to accommodate the inside shaft. As shown in
detail B in Fig. 2, 0.375 in. is milled off one side of the l-in.
diam section. The part shown as detail A in Fig. 2 is joined to the
milled portion by four %0 X 0.5-in. cap screws. This part, which
also has been machined from
FIG. 2. Nylon ahaft.
FIG. 3. Wiring diagram of electrical components,.
l-in. diam nylon rod, incorporates the key that tits into the
keyway in the stainless steel inside shaft. As illus trated in
Fig. 1, the entire shaft assembly rotates within two nylon hubs.
One is embedded in the center of the sample holder, while the other
is attached to the alumi num shelf. The endplay is taken up
through the use of a nylon washer placed between the aluminum top
and the gear attached to the nylon shaft. For precise posi
tioning, a gear train is used to transmit torque from the stepping
motor to the shaft assembly.
A wiring diagram of the various electrical compo nents is shown in
Fig. 3. The electric air valve has its own self-contained power
supply while the stepping motor receives power from a
computer-controlled driver. When switch 8 2 is placed in the "Local
Control" position, the sampler arm can be raised or lowered with
switch 8 3• When switch 82 is in the "Computer Con-
EMISSION ANO ATOMIC ABSORPTlON 9
trol" position, switch 83 is isolated from the circuit and the
sampler arm can only be operated by the com puter-controlled
relay. The stepping "motor wiU func tion only when switch 8 2 is
in the "Computer-Control" position and when at the same time a
microswitch located at the top of the air cylinder's stroke is in
the normally open position. With this electrical interlock, the
stepping motor functions only when the sampler arm is raised.
The sampling arm cun be rotated at a rate of 0.02 sec per step
without missing a step. About 0.6 sec should be allowed for the arm
to rise and clear the sampling cups before step commands are sent
to the motor. With 200 steps per 300° available, it is easy to
program the computer 150 that the aspirating tube wiU be able to
sample solutions in containers of any size set out in a
circle.
ACKNOWLEDGMENT
This work was prepared under the auspices of the U. 8. Atomic
Energy Commission.
The Preparation of Metal Ingots for Use '.3 as Chemical and
Spectrographic Standards
S. L. Odess* Pratt and Whitney Diyision, United Aircralt
Corporation
G. S. Golden United Aircralt Research Laboratories, East Hartlord,
Connecficut 06108
The preparation of small amounts of alloys for use as
spectrographic standards by melting the constit uents in an arc
button melting furnace is a common
* Deceased.
Element Percent
Ni Balance Or 8.0 00 10.0 Ti 1.0 Al 6.0 Mo 6.0 Ta 4.3 B 0.015 Zr
0.07 C 0.11
practice. Larger amounts for use in round robins or as long term
chemi cal standards are prepared by normal foundary techniques. In
both cases the levels of trace or contaminant elements obtained are
difficult to predict, particularly when they are volatile at the
meIt temperature or do Bot alloy with the matrix.
This paper describes the procedure used to prepare 30 lb ingots of
the nickel base alloy B-1900 (PW A 663), whose nominal composition
is given in Table 1, to which were to be added 1, 5, and 10 ppm of
bismuth, lead, tellurium, tin, and zinc. This corresponds to 13.6,
68, and 136 mg, respectively, of each element in the ingots. These
elements generally have low melting points «500°C) and high vapor
pressures at about 1455°C where B-1900 is molten. In addition, some
of these are not soluble in nickel. Therefore, large melt losses
are expected to occur. In fact, a standard practice to de crease
the content of these elements is to hold the alloy at melt
temperature in vacuum. l
To prepare full-size heats, analyze, and remelt with additions to
compensate for losses would be both time consuming and
uneconomical.
The procedure adapted was to melt milligram quantities of the
elements with 50 g of commercially pure nickel in an arc button
melting furnace under an
1. D. R. Wood and R. M. Oook, Metallurgia 109 (1963).
EMISSION ANO ATOMIC ABSORPTION 11
argon atmosphere. To determine melt losses, the entire button was
dissolved and the contaminant concentrations (in the tenth of a
percent range) determined by emission spectrographic and atomic
absorption techniques. This was done twice using 75 and 150 mg
charges of the added elements. The results obtained are shown in
Table II. The higher retention for the second sample was possibly
due to the enclosure of the additions in nickel foiI before
melting.
The hypothesis was then made that these metal concentrations of a
few thousand parts per million in nickel would not be appreciably
decreased when added to a B-1900 melt. Accordingly, buttons were
prepared by arc melting nickel with nickel foiI contain ing the
elements at a level to give the desired con centrations after
correcting for the melt losses found previousIy.
Thirty pound ingots of a heat of B-1900 selected to contain low
levels of these contaminants were melted in a vacuum fumace, heated
to 1530°C, and held at a pressure of < 5 Il for at least 10 min
to further reduce volatile impurities. The temperature was reduced
to 1455°C and the fuma ce backfilled with 500 mm of argon. The
doped nickel buttons were added from a loading arm into the melts.
After allowing 2 min for thermal mixing, the heats were poured into
copper molds 2i in. in diameter.
The results of the analyses of these ingots are given in Table III.
Bismuth, lead, and tin were determined
Table II. Retention of element. in nickel button.
% Retained Element A B
Bi 28 33 Pb 44 50 Te 60 90 Sn 70 100 Zn 10 13
12 SECTION 1
Found Ingot Added Bi Pb Sn Te Zn
Z O <O.S 2±1 7±3 2 4 A 1 1.8±0.6 3±2 7±3 3 10 B S 7±1 7±2 1O±4 6
16 C 10 13±3 12±2 14±S IS 16
by a carrier distillation emission spectrographic technique,2
tellurium by the concentration x-ray fluorescence method of Burke
et al.,a and zinc by the extraction-spectrophotometric procedure of
Ott et al.4
The first three were determined by seven laboratories on separate
slices from each ingot. The standard deviations are similar to
those obtained on replicates of the same portion of the ingot, by a
single laboratory, thus indicating lack of gross segregation.
These results were quite encouraging, showing good correlation with
the aim concentrations particularly in view of the present state of
analytical reproduc ibility and accuracy at these levels. Zinc,
which showed the highest melt losses during button preparation, had
the largest deviations from nominal, but even here there was
sufficient gradation for use as standards.
The technique described should be adaptable for the addition of
other elements to nickel base and other high temperature alloys for
Use as trace level standards.
The aid of R. A. Smith in preparing the doped buttons, R. Ishkanian
in the casting of the ingots, and M. G. Atwell in the performance
of the majority of the analyses is gratefully acknowledged.
2. M. G. Atwell and G. S. Golden, Appl. Spectrosc. 24, 362 (1970).
3. K. E. Burke, M. M. Yanak, and C. H. Albright, Anal. Chem.
39, 14 (1967). 4. W. J. Ott, H. R. McMillen, and W. R. Hatch, Anal.
Chem.
36, 363 (1964).
Qualitative Analysis of Precipitates by '.4 Graphite Filter
Methods
M. S. Wang
Electronic Products & Controls Division, Monsanto Company, St.
Louis, Missour; 63166
In many chemical processes there is a need to quickly identify
unexpected precipitates in a solution. If x-ray fluorescence
equipment is available, the precipitates can be collected by a
Millipore filter and easily analyzed qualitatively. When an x-ray
fluo rescence spectrometer is not available, or the suspected
elements are not amenable to the technique, the methods described
here are helpful .
Sparking or arcing the Millipore filter directly is not an
efficient way of excitation, but graphite material may be used as a
substitute filter. As an added ad vantage, a graphite or carbon
filter is resistant to practicalIy alI corrosive solutions or
solvents.
Most of the porous cup electrodes are tin. outer diameter but quite
different in shape and capacity. Any of these electrodes can be
used to collect precipi tates by connecting both ends to Tygon
tubing. The solution with the precipitates is poured into the top
of the Tygon tubing with the closed end of the porous cup pointing
upward. A vacuum is applied to the Tygon tubing connected to open
end of the porous cup electrode. The precipitate is collected on
the tip of the porous cup electrode and is ready for arc or spark
excitation. The system is illustrated in Fig. 1.
Porous graphite (or carbon) cut to substitute for the filter paper
in the Millipore filter holder can also collect precipitate from a
solution. One which has been used successfully is a disk 1 in. in
diameter and t-156 in. in thickness made from National Carbon's
grade 60 porous carbon. This material has an average pore
14 SECTION 1
Tygon Tubmg ----;001]
1 Ta Vacuum
FIG. 1. Porous cup electrode used as a filt·er.
diameter to 0.0013 in. In order to avoid leaking of solution or
air, tin. wide Teflon tape is wrapped around the side of the carbon
filter disk. Then a small amount of graphite powder (- 200 mesh)
suspended in water or other suita bie liquid is filtered so that
the graphite particles will cover some of the relatively big pores
in the disk. The precipitates are then filtered in the usual way.
The disk is usuaIly brought to Uni-arc or spark excitation.
EMISSION ANO ATOMIC ABSORPTlON 15
Washing of precipitates by solvent or water is practiced whenever
necessary. Quantitation was not tried because the amount of
precipitate is not known.
An Improved Spectrographic Evaporating Dish
R. E. Rainford IBM, Componenfs Division, Essex Junction, Vermonf
05452
Teflon evaporating dishes have been available for some time.
However, their use in emission spectros copy has been limited
because of poor recovery of powdered samples due to electrostatic
charging of the dishes.
An experimental lot of polytetrafluorethylene (TFE) was prepared
containing 15% carbon fiUer by weight. The fiUer prevents the
Teflon from acquir ing a large static charge.
One hundred milliliter evaporating dishes were molded from the
carbon-filled Teflon by a commercial Teflon molding company.
Recovery experiments, using evaporated acid and water residues
adsorbed on graphite powder, showed the dish superior to dishes
made of standard Teflon, quartz, and platinum.
The carbon-filled Teflon evaporating dishes have replaced Pt dishes
in our laboratory for applicatiollS 110t requiring temperatures in
excess of 260°C.
The dishes may be cleaned with aqua regia and commercial scouring
pads for stubborn stains. Dishes should be localized to individual
matrices because regardless of the cleaning, the matrix would be
re tained at a trace level by the Teflon.
1.5
16 SECTION 1
J .6 A Rotating-Disk Sample Holder for the Sparking of
Flat-Metal-Disk Samples
P. E. Walters and T. Monaci Department of Physics, University of
Stellenbosch, South Africa
Interelement effects are of ten experienced between the various
constituents of metal alloys, when analyzed in spark or arc
discharges. Investigating the effects due to varying composition of
the analytical material in the discharge column, by time-resolved
spectros copy, to better explain interelement effects, pre
cautions must be taken to safeguard against the effects of
selective evaporization, which are not in terelement effects in
the true sense. Introducing fresh sample into the analytical gap
for each dis charge offers a means to overcome the effects of
selective evaporization.
~ \
EMISSION ANO ATOMIC ABSORPTlON 17
FIG. 2. Sectional front view of the rotating-disk sample
holder.
The design produces a pattern of the craters to be equally spaced
on an Archimedian spiral.
Figure 1 gives a schematic representation of the system's
kinematics. The sample supporting table A has two degrees of
freedom, i.e., one to rotate about B in bearing Hand one of
translation perpendicular to the axis in the plane of projection.
The sample sup porting table is driven on a point fixed with
reference to the counter-graphite electrode C, by drive D, rotating
at constant angular speed. On rotating, the supporting table drives
wheel E at a constant radius from B. The threaded end of shaft F,
fixed to wheel E, is guided in a fixed nut G. Shaft F is also
connected to the bearing H by means of bearing J, thus forming a
system which drags the table along as the shaft screws in and out
of nut G. The net result is a trans latory movement of the
supporting table perpendicular to the optical axis, which is
perpendicular to the plane of projection. Rotation and translation
of the
18 SECTION 1
FIG. 3. Sectional side view of the rotating-disk sample
holder.
sample supporting table A wiII make aU points that pass under
electrode C lie on an Archimedian spiral. Since the table is driven
at constant angular speed under C by D, the spiral is traced out at
constant velocity. As a result, sparks occurring at equal time
intervals will be equally spaced along the spiral.
Detailed sketches of the rotating-disk sample holder are shown in
Figs. 2 and 3. The sample sup porting table (1) is made to rotate
by the friction drive (2), of which the worm gear (3) is an
integral part. The worm (4) is driven via an insulated shaft (5)
bya reversible motor (6).
To achieve the required horizontal translation of the sample
supporting table a threaded shaft (7) is made to rotate by the
table itself by friction wheel (8). The threaded shaft screws
itself into nut (9), thus
EMISSION ANO ATOMIC ABSORPTlON 19
transporting the table supporting T piece (10), the two being
connected by baU bearings (11) on the left arm of the T piece, the
right arm of its sliding through the ball bearing (12). The shaft
of the supporting table (13) moves in a mercury-filled vessel (14)
and (15), ensuring good electrical contact between the revolving
table and the baseplate. The fork (16) keeps the sample supporting
table straight. The sample (17) is screwed on the supporting table
by the center screw (18).
Using a reversible electric motor the Archimedian spiral can be
traced inwards or outwards alterna tively. The motor is controUed
manua1ly.
To aUow for height adjustment of the spark gap the whole system is
attached to the lower part of a Hilger spark stand by means of an
electrically in sulating block (18).
The ratio of the velocity at which the spiral is traced out and the
translation of the sample is constant. With a repetition rate of 50
sparks/sec the spacing between successive craters on the sample
surface is about 1.3 mm. The distance between spirals is a constant
2 mm, while the spacing of the craters along the spiral can be
adjusted by controlling the speed of the motor.
Some typical results of time-resolved spectra ob tained with this
sample holder are shown in Fig. 4. A multisource unit, i.e., a
low-voltage spark source with an underdamped circuit, was used.
This results in an oscillatory current which is responsible for the
corresponsing time varying intensity obtained.
Marked differences in the cathodic and anodic half-cycles of the
discharge have been observed. A strong dependence of the asymmetry
in the half cycles on Zn concentration was also found.
With the rotating sample of low Zn content, the line intensities
during the anodic half-cycles, i.e., when the sample is positive,
is very much Iower than
20 SECTION 1
the intensities emitted during the cathodic half cycles, as can be
seen with both Si 2882 and Al 3057 in Fig. 4(a). When the Zn
content increases there is a corresponding increase of the line
intensity during the anodic half-cycles, as shown in Fig.
4(b).
With a stationary sample the line intensities from anodic and
cathodic half-cycles are almost the same,
,. 1- <ii z UJ 1-
~
<!)
<!) o .J
'\ r.
, / \ / \
I \ I \ f \ I
EMISSION ANO ATOMIC ABSORPTION 21
independent of Zn concentration. The effect of in creasing Zn is
not visible at aU becoming swamped in the case of a stationary
sample.
In view of the results obtained it is clear that these effects, if
present, could only be observed by using the rotating sample or
similar method. The possible interpretation will be discussed in
another paper.
(\DATIMG
" ___ -.. r, I , / '~~ I \ \ /'"" f \ / \ \~""",,""''''' I v \J '(
f \ n f \/ \
f'j \ j I \ f I I \ I \
250 1250 "o. 119 HOO
TIMEll'soc)
(b)
22 SECTION 1
r.7 Vented Cupped Electrodes
L. T oft and G. A. Roworth Chemical Inspecforafe, Royal Arsenal
East, Woolwich, London S.E.lB, England
When arcing powder samples in graphite cupped electrodes using the
dc arc the charge is sometimes ejected from the cup due to the
sudden evolution of vapor produced by the rapid heating. The
practice of venting the packed electrode by pricking into the
charge with a pin is not particularly effective in preventing
ejection. A more effective method which overcomes the difficulty is
to vent the cup at the bottom before filling. This is carried out
by drilling two small holes through the cup wall, the holes being
diametrically opposed and just above the floor of the cup.
:No difficulties arise when filling or arcing the elec trodes and
no differences have been observed between spectra derived from
vented and satisfactorily arced unvented cups. The diagram shows
the location and size of vent hole which has proyed satisfactory
with a cupped electrode made from tin. diam graphite rod. 11"
r-64~ H H
1'1'I" B '.1
Ifs
1 (NQ 54 DRILL) (0.055" DIAM ) ~B B~fENT HOLES
CUPPED ELECTRODE
FIG. 1.
Suggestions and Comments on: "Vented CUp Electrodes"
John B. Marling Saird Atomic, Incorporatetl, Sedford, Massachusetts
07730
Referring to the Spectrographic Techniques section of ApPLIED
SPECTROSCOPY, 1 would like to go back to "Vented Cupped Electrodes"
by Toft and Roworth.1 There are several suggestions and comments
that 1 have on this article.
It is suggested that the venting of the charge is to prevent
ejection. While this may be true, at least for the carrier analysis
of U 308, venting also improves sensitivity and is essential for
this technique. At Los Alamos, back in the forties, we checked this
and found the sensitivity improved with increasing the vent holes.
N ormally, one is adequate and the proba bility of uranium passing
into the discharge increases markedly with more than one vent
hole.
Ejection of the charge is prevented by drying the loaded electrodes
for 1-2 h at 200°C in an oven. What Toft and Roworth have missed is
that the base vent ing of the crater that they suggest also
improves sensitivity and burn characteristics. Prior to their
publication, Rossi and DeGregorio of Euratom, Ispra, Italy,
published an article2 on this subject in 1969. 1 believe they
deserve credit for this technique.
1. L. Toft and G. A. Roworth, Appl. Spectrosc. 24, 132 (1970). 2.
G. R.oMi and P. De Gregorio, Met. ItaI. 61, 375 (1969).
J .8
Reply to Dr. Marling
L. T oft and G. A. Roworth Quality Assurance Directorate
(MateriaIsJ, Headquarters Building, Royal Arsenal East, Woolwich,
London SE 18, England
In repIy to the points raised by Dr. Marling we have the following
comments to make.
(a) We do not accept that we missed the point of vented electrodes
improving sensitivity since in our experience no increase has been
observed. This was stated in our note: "N o differences have been
ob served between spectra derived from vented and satis factorily
arced unvented cups." Our main usage of the vented electrode for
quantitative work has been the determination of 20 or so elements
in a matrix of lithium sulfate and certainIy, in this particular
ap plication, no increase in sensitivity resuIts from vent ing.
We would not, however, dispute that an increase can result when it
is used with other materials.
(b) With regard to improvement in burn character istics, it
naturally follows that the elimination of any tendencies toward
ejection will be beneficial in this respect.
(c) We formerly used oven drying as a means of reducing the number
of ejections but found that aIthough this treatment improved
matters, it did not, as implied by Dr. Marling, fully overcome the
prob Iem, hence our use of vented electrodes.
(d) To our knowledge, up to the time we submitted our note for
publication, no mention had been made in the EngIish Ianguage
scientific press of the use of vented electrodes for the purpose
recommended in the note. We would thank Dr. Marling for drawing our
attention to the paper by Rossi and DiGregorio.
EMISSION ANO ATOMIC ABSORPTlON
A Cylindrical Sector Driven by Either Water or Air
J. W. Mellichamp and L. L. Wilcox Institute lor Exploratory
Research, U. S. Army Electronics Command, Fort Monmouth, New Jersey
07703
25
The cylindrical sector designed by Yuster1 is more compact and
simpler in construction than the disk sector. A further development
is a cylindrical sector driven by either water or compressed air in
place of the usual electrical motor. It is designed to operate in
either the water system for cooling the electrode holders, or in
the compressed air line circulated from a centrallocation in the
laboratory building. The term sector defines the rotating disk or
cylinder that con trols the gradient of the light illuminating the
spec trographic slit. Each gradient desired (step, log func tion,
intensity reduction, etc.) requires a different sector design. A
small air turbine driven step sector is described elsewhere.2
The design of the device is shown in Fig. 1. Mate rials used in
the construction are nylon for the housing, brass for the rotor,
and Bakelite for the sector portion. Reveral O-rings are llsed to
make the housing water tight. The sector shown [Fig. 1 (b)J is for
intensity re duction and consists of an inner and an outer
cylinder with one-half of each circumference removed in sym
metrical, one-eighth segments. Marks at the top of each cylinder
index the fraction of the light to be passed"one-half in' the open
position and lesser frac tions in closed positions. While neutral
filters can be used for intensity reduction, the sector method is
preferred because of greater uniformity throughout
1. H. G. Yuster, Appl. Spectrosc. 24, 365 (1970). 2. H. S. Bennett,
W. E. Quinton, R. Othberg, and G. W. Reis.
Appl. Spectrosc. 7, 129 (1953).
1.9
26
cm
SECTION 1
FIG. 1. Design of cylindrical sector driven by either water or
compressed air. (a) Cross section of top view showing inlet with
respect to rotor blades and outlet. (b) Cross sect.ion of side view
showing sector used for intensity reduction. Marks at the top of
the inner and outer cylinder index the fractioll of the light to be
passed. In the drawing the cylinders are one-half closed to permit
25% of the light to pass.
the spectral region. The sector portion of the device is
interchangeable with other cylindrical sectors de signed for other
desired gradients.
The device is designed to operate in the return line of the water
system normally used to cool the electrode holders (Spex
Industries, Incorporated). The wa.ter pump which is submerged in
the reservior has enough power to both recycle the water and at the
same time drive the sector at a sufficient speed. The nozzle
directs the water jet so that it will strike the rotor blades at
peak posltion and continue uninterrupted
EMISSION ANO ATOMIC ABSORPTION 27
through the outlet [Fig. 1 (a)], thus minimizing turbu lence and a
back pressure. An air intake is needed at the water outlet to
prevent a partial vacuum at that point. When properly designed, the
water is expelled from the rotor compartment permitting the rotor
to spin in essentialIy alI air. Water wiU not back up and cause
leakage at the air intake.
Relatively constant speeds of around 3000 rpm are obtained with
water. II desired, speeds can be reduced by the use of a clamp that
controls the water flow. II the air intake at the water outlet is
shut off, the rotor compartment is flooded and the speed reduced in
half. Speeds exceeding 5000 rpm are easily reached with compressed
air. A device to be used with air only would be relatively simple
and consist of only the sector portion, a rotor with bushing, and
an air nozzle with no necessity for a return line or a watertight
compartment. Details of the device made (such as exact dimensions,
method of assembling, etc.) are not essential because other
variations would probably give equal or possibly better
results.
The device is compact and free from vibrations, and can be mounted
in any convenient location in the light path when used with a
sector for intensity reduction. However with other cylindrical
sector designs, posi tioning at the slit only may be required. The
device is simple and inexpensive to construct, it operates from
existing water or air lines, and wiU effectively control light
intensity for emission spectroscopy.
28 SECTION 1
H. G. Yuster
New Brunswick Laboratory, United States Atomic Energy Commission,
New Brunswick, New Jersey 08903
INTRODUCTION
The controlled reduction of light intensity illumi nating a height
of spectrographic slit to give a graded exposure for plate
calibration and quantitative imple mentation may be accomplished
by several methods. Numerous literature references are listed by
Meggers and Scribner 1 and include step slits, step filters, step
wedges, rotating log sectors, and rotating step sectors. Harrison
2,3 gives a detailed critical discussion of the methods and a
convenient summary has been made by Ahrens and Taylor.4
Practically all the commercially available sectoring devices are
now either of the step filter or the disk step sector type. The
former requires skill in the thin film deposition of metals on
quartz and sophisticated high vacuum equipment, while the latter
requires the services of a trained machinist on both lathe and
mill ing machine. The disk type of rotating step sector also
requires, due to its geometry, strong materials of con struction
such as brass or aluminum and subsequent
1. W. F. Meggers and B. F. Scribner, Index to the Literature on
Spectrochemical Analysis, Pts. I-IV (1920-1955).
2. G. R. Harrison, J. Opt. Soc. Amer. 24, 60 (1934). 3. G. R.
Harrison, R. C. Lord, and J. R. Loofbourow, Practical
Spectroscopy (Prentice--Hall, Inc., New York, 1948), Chap. 13, pp.
331-342.
4. L. H. Ahrens and S. R. Taylor, Spectrochemical Analysis
(Addison-Wesley Publishing Co., Ine., Reading, Mass., 1961), 2nd
ed., Chap. 11, pp. 150-151.
EMISSION ANO ATOMIC ABSORPTlON 29
anodizing, or careful painting to reduce reflection from critical
surfaces.
The cylindrical rotating step sector to be described is easily
machined, and because of its geometric con figuration it can be
made sufficiently strong and rigid from black nonreflecting
plastic. In addition, the sym metrical construction produces a
balanced sector rela tively free of vibration at high speeds. The
sector tested was constructed from black Tenite rod.
1. CONSTRUCTION
~'
30 SECTION 1
FIG. :!. Cross sectioll of second sh'l"
Full exposure is clear of the sector and will be called the first
step. The second step is one half expo sure and is shown in cross
section in Fig. 2. Solid lines show the actual sector and dotted
lines show the theory for construction.
To determine the diameter of the blank rod that is necessary,
consider the following:
arc AB = arc BC
and chord AB = chord BC = X,
where X is equal to the size of the milling tool.
X = 2R sin 45°,
2R = D = 1.414 X,
where D is equal to the diameter of the blank.
EMISSION ANO ATOMIC ABSORPTlON 31
An example may be given. If the milling tool is i in. then
D = 1.414 (0.750) = 1.0605 in.
For the five step sector, the size of the end mills are ~, ~, 136'
and 332 in., respectively, for steps 2, 3, 4, and 5. A cross
section of the sector in Fig. la is shown in Fig. 3 with dimensions
in inches.
The sector blank is first machined in a lathe with care in making
the top portion conform to the calcu lated diameter. The bottom
half can be indentpd to decrease the weight of the sector. Care
shoulJ be
_------ 1.0605
32 SECTION 1
taken in centering the sector blank relative to the largest end
mill in the milling machine. A pass 0.080 in. in depth is then made
with this end mill, and the process repeated with the smaller end
mills in de cr'cHsing ordcr using f'qual depth settings. Finally,
the piece, as shown in Fig. la, is transferred to a Iathe and the
steps bored out (Fig. lb) to give a wan thick lIess of 0.030
in.
II. EVALUATION
FlIll length slit exposures, without the SCCtOl' in place, were
made in the 2200-3400 A regiOll using a high preeision iron spark.
Densitometric measure lIH'llts of line segments showed uniform
slit illumina tiOll. "Tith the sector in place, exposures were
takell and dl'nsitometry was applied within a 100 A span ol the
:noo.\ regiOl!. A preliminary curve using the
z o ... ...
EMISSION ANO ATOMIC ABSORPTION 33
two-step plotting- lIIethod 5 was used to smooth the data. The
sllloothed data from this cun·e were used to plot the calibratioll
cUrYe ShOWll in Fig. 4. The data were further treated on a Kaiser
calculating board anel a straight line plot resulted when a
transforma tion constant of 0.6 was used. A final mathematical
treatmcnt was made 011 the data. Using the modified l'(!Uatioll 6
of the Seidel functio11 7 the straight line eune ShOW11 in Fig. 5
was obtained. In this plot
.:l = [log (lOO/T) + lo)!' (lOO/T - 1) ]/2.
o
FIG. 5. Plate calihTation with modified Seidel cquation.
5. J. R. Churchill, Anal. Chem. 16,653 (1944). 6. M.
Honerjager·Sahn and H. Kaiser, Spectrochim. Acta 2,
396 (1944). 7. H. Kaiser, Spectrochim. Acta 2,1 (1941).
34 SECTION 1
'fhe ('xtr(,llU' upper and lower poillts were transmis sion yalues
of 7.5 and 98, respectiYely.
III. DISCUSSION 'fhe data were taken with a 3.4 m Jarrel-Ash
Ebert
spectrograph. This instrument is stigmatic and gave sharp sectoring
even though the sector had appre ciable depth compared to the disk
type. The motor for driving the sector was mounted on the optical
bench of the spectrograph with its shaft in an upward ver tical
position. Its position was as close as possible to the slit lens to
stiH allow rotation of the sector. It was found that care was
necessary in mounting the sector relative to the entrance slit and
the spark discharge. Alignment with. step 5 was critical and
required care fuI adjustment. Measurements on a completed sector
made with new end mills showed an undersize error of approximately
1 % in each step. This was no doubt due to plastic memory as
diameter measurements after construction showed a slight taper from
top to bottom.
It is evident that other sizes of end mills can be used. If it is
desired to change the step ratio from two to other values the
machining complexity would be increased.
ACKNOWLEDGMENT The author wishes to thank Austin Padgett, of
the
instrument section at the New Brunswick Laboratory, for his helpful
suggestions and machining of the sector.
EMISSION ANO ATOMIC ABSORPTION
Prevention of Laser Microprobe Staining of Ânalyzed Metals*
H. N. Barton and J. Benallo The Dow Chem;cal Company, Roclcy Flats
D;y;s;on, Golden, Colorado 80401
35
The laser microprobe is a useful tool for the analysis of specific
sample areas of approximately 50 J.I. diam eter. The auxiliary
spark excitation of Brech and Cross1 increases spectrum intensity
but results in a staining of metal sample surfaces in the area of
analysis.2 Sample surface detail is thus obscured and selection of
analysis sites for subsequent determina tions is extremely
difficult. The capability of the laser microprobe to selectively
sample closely adjacent areas is thereby nullified. Deposition of
the stain can be restricted to the dimensions of a central hole in
a paper shield; however, placement of a shield is difficult even
for plane surface samples.
The application of a thin coating of collodion3
(40 g/liter nitrocellulose in 75 vol% ether, 2.5 vol% alcohol) to
the sample surface prior to analysis pro vides a rapid drying
'transparent surface on which the stain does not deposit.
The stain deposited on an unprotected coin by a single laser
microprobe analysis with auxiliary spn.rk excitation is shown in
Fig. 1. The dark stain has
,.. Work performed under U. S. Atomic Energy Commission con tract
AT(29-1)-1l06.
1. F. Brech and L. Cross, Abstracts of Xth Colloquium Spec
troscopium Internationale and First Meeting of the Society for
Applied Spectroscopy, Appl. Spectrosc. 16, 59 (1962).
2. S. D. Rasberry, B. F. Scribner and M. Margoshes, Appl. Opt. 6,
81 (1967).
3. Collodion, U. S. P., J. T. Baker 9209, J. T. Baker Co.,
Phillips burg, N. J.
loJI
FIG. 2. Collodioll protected coin.
EMISSION ANO ATOMIC ABSORPTlON 37
obliterated surface detail within 2 mm of the analysis site. The
result of a similar laser microprobe analysis of a collodion
protected coin is shown in Fig. 2. No significant reduction of
surface detail has been pro duced by the collodion layer nor is
staining detectable. An irregular area of collodion approximately
30 j.J. in diameter h:1s been broken from the surface at the ,jO
j.J.
crater. N o contaminants were detectable from the brush applied
collodion layer. N o effect on the an:l.lysis of aluminum, brass,
copper, gold, iron, manganese, silver, stainless steel, tantalum,
titanium, and tungsten metals was detectable from application of
the col lodion layer.
A Simple Multiport Atomic Absorption J. J 2 Burner Head
M. S, Wang
Electronic Products and Control Division, M on sanio Company, Si.
Louis, Missouri 63166
Since Boling1 described his multiple slot burner, application of
this design has been popular in the field. The multiport burner
head reported here is easy to fabricate, and the cost is only a
fraction of the cost of the product that is available commercially.
It can be made out of a single slot old burner hcad or stainless
steel tubing of a suitable size.
The burner head shown in Fig. 1 was made from an old Perkin-Elmer
burner head by drilling 32 0.052-in. diameter holes on each side of
the slot. The width of the burner slot is 0.015 in. and the holes
are on 0.125-in. centers, 0.0685 in. from the centerline of the
slot.
The burner performs almost exactly as the Boling burner. Because of
the smaller flame outlet, flashback
1. E. A. Boling, f':ipectrochim. Acta 22, 425 (1966).
38 SECTION 1
FIG. 1. A simple multiport atomic absorption burner head with
aluminum radiator fins.
caused by a more oxidizing flame has never been ob served using
this burner head. The aluminum radiator fins have a stabilizing
effect on the flame and also serve as a cooling mechanism. This
burner head is in the 7th year of daily operation and continues to
per form satisfactorily.
ACKNOWLEDGMENT
The author wishes to express his sincere appreciation to Mr. Larry
L. WiIliams, the present operator of the equipment, for his
critical opinion of this burner.
EMISSION ANO ATOMIC ABSORPTION 39
Modification of a Commercial Carbon J • J 3 Rod Flameless A tomizer
to Accept Graphite Tubes *
R. W. Morrow and R. J. McElhaney
Oak Ridge Y-12 Plant,t P. O. Box Y Oak Ridge, Tennessee 37830
Flameless atomization is becoming accepted as a com plementary
technique to conventional flames for use in atomic absorption
spectroscopy. Commercial nonflame atomizers have become available
and depend upon electrothermal heating of a graphite rod,! a
graphite tube,2.3 or a tantalum ribbon4 to atomize the sample.
Detection limits in the microgram per liter range can be expected
with these devices. A recent review article dis cussed three of
the flameless atomizers that are commer cially available. 5
A Varian-Tectron6 model 61 carbon rod atomizer was acquired for use
in determining metals in environmental water samples. The graphite
rod used in this device can
* Presented in part at the 24th Southeastern Regional Meeting of
the American Chemical Society, Birmingham, Alabama, 2-4 N ovember
1972.
t Operated for the U. S. Atomic Energy Commission by Union Carbide
Corporation, Nuclear Division, under Contract W -7405-eng-26.
1. J. P. Matousek, Am. Lab. 3,45 (June 1971). 2. D. C. Manning and
F. Fernandez, At. Abs. Newsletter 9,
65 (1970). 3. F. J. Fernandez and D. C. Manning, At. Abs.
Newsletter
10, 65 (1971). 4. J. Y. Hwang, P. A. Ullucci, and S. B. Smith, Am.
Lab. 3,
41 (August 1971). 5. M. D. Amos, Am. Lab. 4, 57 (August 1972). 6.
Reference to a company or product name does not imply
approval or recommendation of the product by Union Carbide
Corporation or the U. S. Atomic Energy Commis sion to the exclus
ion of others that may meet specifica tions.
40 SECTION 1
accept a 1-,1.11 sample and has a 5-mm absorption path length. This
flameless atomi zer was found to have sev eraI shortcomings that
made it somewhat difficult to use. The small dlameter (1.5 mm) of
the transverse hole through the rod would transmit only 20 to 30 %
of the spectral beam. Blocking so much of the sample spectral beam
resulted in severe noise levels and erratic behavior when the
device was mounted in a double beam instru ment. The precision of
the carbon rod atomizer was found to be poor and was attributed to
difficu1ty in re producing the 1-,1.11 injections. The small
sample size and the short residence time of the atomic vapor in the
spec tral beam also contributed to poor volume sensltivity for
some elements. W ork was undertaken to convert the model 61
workhead to a graphite tube fumace atomizer that would possess
improved precision and sensitivity and be usable with double beam
instruments.
The carbon rod workhead was modified to accept graphite tubes of
such geometry that the power supply obtained with the device would
be sufficient to atomize most metals. The brass mounting posts of
the workhead were scaled up to 0.75 in. X 1.12 in. X 1.12 in. while
maintaining the original basic design. The center holes were
drilled large enough to accept the graphite tube. The base of the
carbon rod workhead was used without modification.
/
r-•••• W~_Llniection Port J 0.93 cm •••• _______ ..., ••• ~ O 63
cm
L-L 1.. 3.0 cm ----J _Il f-o-R------- 6.7 cm -----...... -~.
FIG. 1. A cross sectional view of the graphite tube.
EMISSION ANO ATOMIC ABSORPTION 41
FIG. 2. The modified carbon rod work head with a graphite tube
mounted.
ohm-cm resistivity. The thin walls of the center portion of the
cell provide sufficient resistance to be heated to near maximum
temperature with 4 to 7 sec of power. The thick regions at each end
can withstand the tight clamping necessary for good electrical
contact. The atomizing currents necessary range from 90 A for zinc
to 250 A for aluminum. A 2-mrn diameter hole is drilled into the
center to allow introduction of a microliter pipet tip. Two l-mrn
diameter holes are drilled 0.25 in. off center and 90° from the
injection port to facilitate argon purging of the ceH interior.
Depending upon the atomizing current, a tube can be used for 150 to
200 firings. Infrared thermometry was used to measure the surface
tempera ture of the graphite tube as a function of current and
time. With an atomizing current of 250 A applied for 7 sec a
temperature of 2700°C was reached.
In Fig. 2 the modified carbon rod workhead is shown with a graphite
tube mounted and the injection hole visible. The knurled knobs on
each post used for clamp-
42 SECTION 1
ing the ceH in place are seen. The electrical power cables are
visible as are the polyethylene tubes for water cool ing the base.
Argon for purging the fumace is admitted through a slot in the
base.
A cover for the fumace that permits purging the air from the space
around the ceH and from the ceH interior was fabricated from
0.02-in. tantalum sheet. The di mensions of the cover are 1.0 in.
X 1.0 in. X 1.5 in., and it is inserted between the Tefton
insulators shown in Fig. 2. (The Tefton was later replaced WIth
ceramic in sulators.) A glass stopper is used to close the
injection port in the cover. The graphite fumace atomizer can
accept up to a 50-JoII sample although 25 ,ul is the op timum
size. Injections are easily made with push button microliter pipets
with disposable plastic tips.
The over-aH Slze of the graphite fumace atomi zer ex cluding the
mounting post is 2.5 in. X 1.12 in. X 1.75 in. It can be mounted in
place of the bumer head, and the existing bumer controls can be
used to position the fumace relative to the spectral beam.
Approximately 70 to 80 % of the spectral beam \\ ill be
transmitted.
TABLE I. Detection limits and sensitivities obtained with the
graphite furnace atomizer.
Element Wavelength (A)
Ag 3281 As 1937 Be 2349 Cd 2288 Co 2407 Cr 3579 Cu 3247 Mn 2795 Ni
2320 Pb 2171 Zn 2139
Detection limit
(mg/liter)a
0.0003 0.02 0.00008 0.0001 0.002 0.0004 0.0006 0.0002 0.006 0.0008
0.00001
a Computed using a 25-1-<1 sample.
Volume sensitivity (mg/liter)a
0.0004 0.02 0.0002 0.0001 O.OO-! 0.001 0.002 0.001 0.02 0.0009
0.00002
Flame atomic absorption sensitivity (mg/liter)
0.05 0.3 0.02 0.02 0.07 0.06 O.O-! 0.03 0.10 0.15 0.015
EMISSION ANO ATOMIC ABSORPTION 43
This fumace has been successfully used on a Perkin Elmer 303, a
Varian-Techtron AA5, and a Jarrell-Ash 810 atomic absorption
instrument.
The detection limits and sensitivities of a number of elements
determined using the graphite fumace ato mizer and a Jarrell-Ash
810 instrument are shown in Table 1. The corresponding flame atomic
absorption sensitivities are given for comparison purposes.7 The
ab sorption signal for a given standard was found to be a function
of atomizing voltage; i.e., increasing the ato mization rate
results in an increased absorption peak. The data in Table 1 were
taken using the minimum voltage that would quantitatively remove
the metals from the cell. The sample size \Vas maintained at 25
j.tI.
The detection limits were dl'termined using the method proposed by
Slavin and co-workers.8 Standards containing 0.5 to 50 j.tgjliter
were prepared and used to establish the sensitivity. Sensitivity is
defined in the usual manner as the concentration necessary to
produce a 1 % absorption signaI. The standard was then diluted to a
concentration that would yield a 1 to 2.5 % absorp tion signal,
and the sensitivity ,vas again determined to establish linearity. A
minimum of six injections of this standard were made, and the mean
percent absorption and standard deviation were computed. The
detection limit was then taken as the concentration necessary to
produce an absorption signal equivalent to twice the standard
deviation. The sl'nsitivity and detection limit values determined
compare favorably with those re ported for the Perkin-Elmer HGA-70
heated graphite atomizer.2 ,3
7. Atomic Absorption Analylical Melhods (Jarrell-Ash Division,
Fisher Scientific Ca., Waltham, Mass., 1972).
8. S. Slavin, W. B. Barnett, and H. L. Kahn, At. Abs. News letter
lI, 2 (1972).
44 SECTION 1
J • J 4 Tuning Stubs as an Aid to Coupling RF Energy to
Electrodeless Discharge Lamps*t
w. G. Schrenk, S. E. Valente, and K. E. Smith
Chemistry Department, Kansas Agricultural Experiment 5tation,
Manhattan, Kansas 66502
Efficient coupling of rf energy to an rf excited electrodeless
discharge lamp is an important procedure frequently overlooked by
spectroscopists. If incorrect coupling of the rf energy (via the
usual coaxial transmission line) occurs, several undesirable condi
tions result: (1) The energy transfer is inefficient and the
coaxial transmission line wiU dissipate some of the energy as heat,
and (2) some of the rf energy will be refiected back to the rf
generator with possible damage to the generator, particularly the
magnetron osciUator tube.
Most efficient transfer of rf energy from an rf generator to a load
wiU occur when the output impe dance of the generator, the
characteristic impedance of the coaxial cable, and the load
impedance are equal (matched). In the case of the electrodeless
discharge lamp the load includes the antenna or microwave cavity
and the lamp. When aU impedances are matched, there are no standing
waves on the coaxial transmission line and the line is said to be
fiat.
Matching the rf generator to a coaxial transmission line is not a
problem since most generators are con structed with a 50-n output
impedance and several varieties of coaxial cables are available
",ith a 50-n
* Contribution No. 655, Department of Chemistry, KAES, KSU,
Manhattan, Kans. 66502.
t Supported by N.S.F. Grant GP-9579. t Present address: Chemistry
Department, Regis College,
Denver, Colo.
EMISSION ANO ATOMIC ABSORPTION 45
characteristic impedance. Antennas and microwave cavities can be
constructed to produce a load impe dance of approximately 5011;
however, when any object is placed in the rf field close to the
antenna or in a cavity the impedance is changed, sometimes
radically. Standing waves thus are established in the line to
reflect power back to the rf generator and lower the efficiency of
energy transfer.
In our research dealing with microwave excited electrodeless
discharge lamps as atomic absorption spectroscopy sources we found
the problem of standing waves (mismatch between the line and the
load) on the coaxial transmission cable to be adversely affected by
a number of factors. These included such items as size of the
discharge lamp, nature of the fill gas, nature of the active
substance in the lamp (metal, metal salt, etc.), temperature of the
discharge tube, use of jacketing materials, intensity of the
discharge, and probably others. It therefore became important to
have independent means for minimizing the power reflected back to
the rf generator.
Since the impedance of the transmission line is fixed as is that of
the load, an impedance matching trans former between the coaxial
line and the load is re quired. At low rf frequencies impedance
matching can be accomplished using an inductance-capacitance
network. At microwave frequencies (in our case 2450 MHz) tuning
stubs are used and are more efficient. Also necessary is some
device which indicates when a condition of minimum reflected power
is obtained. A standing wave ratio meter (or a reflected power
meter) should be used for this purpose. Such meters are built into
some rf generator units and can also be purchased separately.
The wavelength corresponding to 2450 MHz is small enough to permit
the construction of a one fourth to one wavelength tuning stub or
stubs for impedance matching. The design of shorted tuning stubs
that meets these requirements is described herein.
46 SECTION 1
1. DESIGN CONSIDERATIONS
The tunable, double-stub, impedance-matching transformer consists
of a length of coaxial transmission line with two branching lines
in which the inner con ductor is shorted to the outer conductor.
The position of the short in each branch line is adjustable.
Branch ing lines are positioned approximately one-half wave
length apart on the transmission line.
Because the tuning device is part of the coaxial transmission line,
the following considerations are important.
A. Natural Impedance of Coaxial Line
Impedance of the coaxial line composing the impe dance matching
device is determined by the ratio of the inside diameter D of the
outer conductor, to the outside diameter d of the inner conductor
according to the relationship:
Z = 138IogD/d,
B. Surface Resistance
The Ohmic resistance of the inside surface of the tuning device
should be minimized. At microwave frequencies, most of the energy
is transmitted along the surface of the conductor. Plating the
inside of the tuning device with a metal of high electrical conduc
tivity significantly improves transmission efficiency.
C. Contact Resistance
Ohmic resistance of aU electrical contacts carrying microwave
energy should be minimized. Particular attention must be paid to
contacts which are required to move while the microwave power is
being trans mitted. Wherever Ohmic resistance occurs, a power loss
wiU result, causing local heating and reduced transmission
efficiency.
EMISSION ANO ATOMIC ABSORPTION
1- 3/32 - inch brass - .. rod
47
11"'~f------------ 61/2" ----------->1.,
II. CONSTRUCTION DETAILS
Figure 1 is a diagram of the stub assembly; Fig. 2 shows details of
the stub design. The figures help clarify details of the
construction.
To maintain good electrical contact between the inner and outer
conductors at the short, each stub is made of two concentric
tubular sections (Secs. A and B, Fig. 2) with expandable ends. The
inner section is threaded into the outer section, allowing the
tapered ends of each section to be forced against one another,
which contracts the inner section and expands the outer
section.
48
Knurled
SECTION 1
Soldered jOint
A. Movable Shorting Stubs
, . Outer Section
Section B is fabricated from two parts, both machined from t-in.
brass rod (Fig. 2). The knurled
EMISSION ANO ATOMIC ABSORPTlON 49
handled is positioned so it limits penetration of the shorting stub
into the branching line to the length of the outer conductor of the
branching line. The 60-deg taper on the insi de surface of the end
of the section is formed by drilling a shallow concentric hole
(approx. tin. deep) in a short length of brass rod (turned to
176-in. o.d.) using center drill with a 60-deg shoulder. That end
piece is th:m silver soldered onto the main body of the outer
section. To align the two parts and obtain a strong bond between
them, (necessary because the end of the section is later split
axially) close-fitting complimentary ridges are cut into the
contacting surfaces. The surfaces to be joined are first tinned
with No. 46 silver solder, then reheated and pressed together to
form a strong joint. Excess solder is trimmed smooth from both
sides of the joint. The insi de and outside of the tip of the outer
section are cleaned with steel wool and silver plated by immersing
the brass in a solution of approximately 5% AgNOa• The end of the
outside section is split axially (for approximately 35 mm) into six
wedge-shaped sections by means of a jewelers' saw (Fig. 2).
2. I nner Secfion
The inner section (Sec. A, Fig. 2) is fabricated from I-in. brass
rod. The rod is tapered 60 deg to 6t into the lower end of Sec. B
to provide maximum electrical contact between sections. A
longitudinal hole tin. in diameter is drilled to within ! in. of
the lower end of Sec. A. A 3\-in. diameter hole is extended through
the remainder of Sec. A. The end of Sec. A is split axially into
four wedge-shaped sections with a jewelers' saw. The section is
then silver plated as previously described, producing a
low-resistance con nection between the outer and inner conductor
when Sec. A is tightened into Sec. B.
The threads for tightening Sec. A into Sec. Bare 1-24. The top ends
of both sections are knurled for convenience in sliding or
tightening the shorting stubs.
50 SECTION 1
B. Outer Conductor
The outer conductor is fabricated from !-in. o.d. hard-copper
refrigeration tubing. A 0.500-in. si de cutting milling tool is
used to mill a semicircular end on each of two 3-l in. lengths of
tubing. Two 29/64-in. holes are drilled in the side of a 5 in.
length of the tubing. (The center of one hole is positioned 1 in.
from one end; the center of the second, 2 lin. from the first.) Two
tee joints are formed from the three tubing sections by clamping
the respective sections in place and joining them with silver
solder. A square flange suitable for centering and connecting a
type N bulk head connector to the assembly is made from i-in.
brass, and silver soldered to one end of the transmis sion line
section of the outer conductor. The rear collar of a type N coaxial
connector is drilled to l-in. i.d., slipped over the opposite end
of the transmission line section, and silver soldered to the tubing
in such a position that a snug fit between the connector and line
is produced by turning the connector onto its collar. All traces-of
solder flux are removed by soaking in hot water, and the inside
surface of the outer con ductor assembly is polished with steel
wool. The assembly is silver plated as described.
C. Inner Conductor
The inner conductor is fabricated from -A-in. brass rod. The
transmission line is fabricated and placed into position
temporarily to mark the intersection points of the branching lines.
The intersection points are lJarefully marked by placing the
movable stubs in the branching lines and passing i2-in. rod through
the stubs, contacting, and scratching the transmission line. The
transmission line is then removed and flat areas, hin. square,
filed in the line at the points of inter section. The centers of
each flat are drilled and 0--80 threads tapped. One end of each
branching line is drilled axially and a short brass 0--80 set screw
soldered into place, allowing sufficient threaded length
extend-
EMISSION ANO ATOMIC ABSORPTlON 51
ing to just fi11 the threaded holes in the transmission line. The
transmission line is then permanentIy joined to the type N
connectors by soft solder, and a11 excess solder removed from the
brass rod. The rods are cleaned with steel wool and silver plated
as described. An alignment sleeve of Tefton (or other suitable
insulator) is fabricated and positioned in the trans mission line
between the two branching lines.
Our stubs required type N connectors to be com pati bie with our
transmission line and antenna fittings. Other connectors can be
used as necessary provided insulation is effective at the rf
frequencies employed.
III. OPERATION
The tuning stubs should be mounted as close to the antenna or
microwave cavity as possible. Preferably there should be no coaxial
cable between the tuning stubs and the out put device.
Operation consists of applying low-rf power to the system,
observing the standing wave ratio with the meter and adjusting
first one stub and then the other until the standing wave ratio
reaches minimum. Fu11 power can then be applied, slight retuning
may be needed when fulI power is applied.
The stubs function welI and are easily adjusted. We have found no
experimental conditions in our work that could not be adjusted to a
satisfactory standing wave ratio (1.5 or less) with the
stubs.
ACKNOWLEDGMENT The assistance of Dr. D. H. Lenhert of the
Kansas
State University Department of Electrical Engineer ing is
gratefully acknowledged.
52 SECTION 1
Spectroscopy *
u.s. Geo/oflical SUf'fey, Den'fer, Co/orado 80225
Significant improvement in reducing background build-up and a
decrease in burn time have been achieved with a new compact gas jet
developed by the U. S. Geological Survey. :\[oving-plate studies on
geologic materials revealed a considerable increase in the rate of
controlled vaporization of the elements as compared with other jets
under the same excitation conditions. The use of the jet produces a
stabilized arc, as well as the depression of the cyanogen bands, in
a nitrogen-fl'ee atmosphere. The simplicity of the gas jet
electrode holder simplifies the change of the electrodes between
samples.
The jet was made in the U. S. Geological Survey's machine shop in
Denvel' under the direction of W. H. W ood and C. Carollo. The jet
parts and dimensions are shown in Fig. 1. The over-all dimension of
the jet (Fig. 1) is 1! X 1 in. The main body is made from l-in.
brass bar stock and is hollowed out to form a water jacket around
the gas chamber. The water inlet and outlet are silver soldered to
the water chamber to give it a U-shaped appearance. The gas inlet
is silver soldered at a tangent to the base of the i-in. diam gas
chamber. The gas inlet is ma de from a5z-in. brass tubing and the
water connections are made from l-in. brass tubing. A 156-in. hole
was drilled at the center of the jet base to alIo\\" the electrode
holder to pass freely. The gas vent plate is made from i-in.
• Publica.tiolla.uthorized by the Director, U. S. Geologica.l
Survey.
EMISSION ANO ATOMIC ABSORPTlON 53
1·
hall-moon OI*'illQ
stainless steel bar stock. The plate is 1 in. thick, and a -fi-in.
hole is drilled in the center. The plate's outel· perimeter is
machined slightly so it will tit inside the gas chamber. The outer
edge of the plate has nine h-in.-diam half-moon openings to allow
the gas to flow upward. It is held in place by a set screw.
The
54 SECTION 1
upper part of the chamber has l-in. threads on which a conical
ceramic cap is screwed. The ceramic cap (the same type Helz l used
for his gas jet) is aNo. 6 Linde Hiliarc cup that has the top
shortened to enable the cylindrical and conical parts of the cup to
meet.
The over-all length of the stainless-steel electrode holder is li
in. (Fig. 1). The base is ! in. in diameter and i in. long, and the
upper l-in. portion is -h in. in diameter. The top 1 in. is
hollowed out to accom modate a standard O.242-in. electrode. The
clearance between the electrode and the opening at the top of the
ceramic cap is approximately l6 in., which allows enough space for
change of the electrode from the top of the jet.
The jet ia compact and may be adapted to any arc stand that can
provide a space of li in. between the lower electrode jaws and the
optical axis of the spectrograph (Fig. 2). The jet is held in
position "by a clamping device soldered onto a piece of i-in. brass
stock. The opposite end of the brass plate is clamped onto an
electrically insulated post which allows move ment up aud down or
from side to side. The electrode holder is clamped in the lower
jaws and its movement is independent of the gas jet. This makes it
possible to keep the electrode separation constant throughout the
burn time. The sample electrode is positioned to extend about lin.
above the ceramic cap.
The gas is brought in at a tangent to the base of the jet to
produce a swirlin.g motion in the water-cooled gas chamber (Fig.
1). One portion of the gas passes through the space between the
electrode and the stainless steel gas vent plate and out into the
atmo sphere. The other portion of gas passes up the chamber
through the several small machined openings in the gas vent plate,
along the brass wall of the water jacket, and into the tapered part
of the jet and then
1. A. W. Helz, U. S. Geological Survey Professional Paper 475-D,
Paper No. 159, D176-D178 (1964).
r- ~~
O
56 SECTION 1
Table 1. Moving plate study showing the complete vaporization of
elements at a concentration of 500 ppm [Conditions: atmosphere
(argon 70%, oxygen 30%) de arc, 12 Al.
Jet A Compact jet Element (sec) (sec)
Zn ... 60 30 Pb ... 70 30 Mn··· 100 60 As 20 10 Au 90 50 Fe 100 70
Sb 70 30 Cu 130 80 Bi 80 30 Mo ... 130 60 Ag ... 130 70
finally escapes from the conically shaped ceramic cap surrounding
the electrode. A ceramic cap was selected over other heat-resistant
materials to avoid possible contamination of the elements
sought.
A moving plate study was made using a rock standard to show the
complete vaporizations of the elements in seconds (Table 1). The
standard was composed of zinc, lead, arsenic. manganese, gold,
iron, antimony, copper, bismuth, molybdenum, and silver at a
concentration of 500 ppm. Ten milligrams of the rock standard was
mixed with 20 mg of graphite, and the resulting mixture was burned
for 1 min in a 12-A dc arc. The controlled gas was a mixture of 70%
argon and 30% oxygen set at a flow of 15 ft3 jh. The compact jet
produces a considerable decrease in burn time as compared to other
gas jets under the same excitation conditions.
To show the effectiveness of the jet in the cyanogen band region of
the spectrum, a comparison was made with the Helz1 and the
Margoshes'2 gas jets (Fig. 3). Ten milligrams of G-l rock standard
was mixed with 20 mg of graphite and the resulting mixture
was
2. B. F. Scribner aud Marvin Margoshes, Appl. Spectrosc. 17, 142
(1963).
CY AN
OG EN
H EA
D BA
ND S
.3 5
9 0
a t m o ~ p h e r e o
f 7
58 SECTION 1
burned for 1 min in a 12-A dc arc under different conditions. The
spectra was recorded on a spectrum analysis No. 1, 35-mm film. The
jet shows depression of the cyanogen bands, comparable to the other
gas jets.
The relative compactness, effectiveness in reducing background, and
shorter burn time make this jet effective for providing a
controlled atmosphere for research or for routine high
production.
, • , 6 Electrode Heater
P. B. Adams, E. C. Goodrich, and J. S. Sterlace
Corning Glass Works, Corn;ng, New York 14830
Trace spectrochemical methods of ten involve evapo ration of a
solvent on the tip of a graphite electrode to concentrate the
sample. An ir lamp is commonly employed. We have devised an
alternate method.
HIGH SILICA GLASS TUSE
FIG. 1. Electrode heater.
EMISSION ANO ATOMIC ABSORPTlON 59
Figure 1 shows an electrode heater. The inner 96% silica glass tube
must fit snugly around the electrode to provide good heat transfer.
The outer 96% silica glass tube encloses the Nichrome wire heater
wind ings. A slot has been cut to allow the leads to pass through.
Both glass tubes have been sealed at the bottom. A variable