Spectroscopic Tricks

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
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-
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).
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.
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.
~ \
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
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
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
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.
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.
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 ... ...
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.
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.
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.
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
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.
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
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-
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.
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.
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

Spectroscopic Tricks

Documents

Transcript of Spectroscopic Tricks