Table II.
Sample Wt., Grams Purified grade lithium 0 4123 0 5786 Reagent grade lithium 0.5726 1,0989 0.8913
Analysis
of Lithium Wt. Li" Based on
yo Li" Based on
Sample Wt. Based on Titration, Grams
Absorbed, Grams
Absorbed, Grams
0 4093 0 5772
0 5269 0 7452
0 405!1 0 5741
98 45 99 23
99 17
0.56%
0.7352 1.4293 1.1568
0.5664 1.1012 0.8913
98.92 100.20 99 9'3
100.11 100.03 99.68
1,1008 0.8941
Wt. H20
Wt. H20
Sample Wt. Gravimetric Volumetric 89 4T 250cc.SISTlLLlNG
/ '
FLASK
~
1 ,
*
supply. The former are a normal source of contamination in bottled oxygen. Data for the system blank are presented in Table I. The motivation for deriving the method of analysis as described n as the determination of lithium dispersed in mineral oil. Hoir-ever, bccnuse of the presence of the impurities as shown above, correlation of Li" with sample weight WAS impossible. Therefore. in order to rstablish the validity of the method, lithium metal obtained from the Fisher Scientific Co was analyzed. A piece, approximately 0.5 t o 1.0 gram, was trimmed so that fresh metal surface was exposed, weighed in hesane, and then dried rapidly and placed in the reaction chamber. After hydrolysis the residue in the form of a lithium hydrouide solution was titrated ~ i t 2-Y h hydrochloric acid. Because oxidation of the Li" had to be avoided, no attempt n a s made to remove the protective hexane coating completely. This is the reason why the analysis values based on the gravimetric n eights of the samples shon
~
~
+50Wc. ERLENMEYER F L A S K
/
a simple titration technique for the estimation of Li" was precluded. The NhlR samples, ca. 0.5 t o 1.0 gram of lithium dispersed in 4 ml. of mineral oil, were contained in sealed glass tubes. Because of the inhomogeneity of the mixture, i t was necessary t o analyze the entire sample. The rather large quantity of lithium present required controlled reaction times on the order of 75 minutes and flush-out times of around 20 hours. The system blank n-as determined under exact esperimental conditions, including distillation, in the absence of lithium. On the basis of 10 such "dry" runs, the blank was 1.7 1.2 mg. per 20 hours. This value, small enough to be ignored, was achieved only after a n oxygen prepurification system identical to the catalytic chamber was introduced into the reaction train. Without this precaution, nonreproducible values as high as 82 mg. per 20 hours Lvere observed. This is believed attributed to the presence of combustible hydrocarbons or possibly hydrogen in the oxygen
~
I
Figure 2. Water distillation and reaction chamber
greater deviation than the vducs computed on the basis of volumetrically determined weights listrd in Table 11. Although developed for the determination of the metallic lithium content of mineral oil dispersions, the procedure is suitable for analysis of the other alkali metals and alkali inrtnl ion miytures. LITE RATU RE CITED
(1) Hecht, R., Redfield, A . G., Bull. Am. Phys. SOC. 4, 240 (1959).
MELVINBERKESBLIT f l R S O L D REISMAS Watson Laboratories International Business Machines Corp. Columbia Universitv New York 25, N . T. RECEIVED for review December 18, 1959. Accepted March 9, 1960.
Quantitative Radioassay of Paper Chromatograms by Liquid Scintillation Counting Application to Carbon-14-La beled Salicylic Acid SIR: I n a study of the metabolic fate of C14 carbonyl salicylamide ( I ) , it became necessary to devise an assay procedure for small amounts of low specific activity metabolites separated from tissues and body fluids by paper chromatography. Elution techniques were rejected because substantial losses of activity would be expected. An alternate procedure wherein the paper itself was immersed in the liquid scintillator was investigated. Factors to be investigated were the effect of the paper on the sample counting rate, the most appropriate solvent system for use, and the linearity between ap722
a
ANALYTICAL CHEMISTRY
plied activity and observed counting rate. T o various area circles of N h a t m a n Xo. 1 filter paper were added 1250
d.p.m. (disintegrations per minute) of Ci4 carboxylsalicylic acid. The papers were air-dried, placed in 15 ml. of a scintillator solution composed of 2,5-diphenyloxazole (PPO) (0.4%) and 1,4bis [2-(5-phenyloxazoyl)] benzene (POPOP) (0.05%) in .toluene, and counted in a Tri-Carb liquid scintillation spectrometer (Packard Instrument Co., LaGrange, Ill.). KO attempt n-as made to maintain reproducible geometry of the papers in the vials. The observed counting rates (Table I) indi-
cate that in this scintillator an interaction had occurred giving rise to spurious counts, Background counts taken with the papers present in the vials were no higher than in their absence. S o relationship was apparent between counting rate and size of the paper. The spurious counts observed did not decal\\ith time over a 48-hour period when the samples were stored in the dark a t
-6"
c.
The spurious counting rates observed may be due t o a phosphorescence coniponent mith a long decay time. Herberg (2) has shown acidification of the
scintillator to rliminate pliosphorescence observed in the liquid scintillation counting of certain proteinaceous materials. Accordingly, a scintillator system n-as prepared cont'aining xylene (500 nil.)l dioxane (500 nil.), ethyl alcohol (300 ml.) resublimed naphthalene (102 g r a m ) , PPO (6.5 grams), anc-l POPOP 10.13 gram), which will dissolve aqueous acid. To test this system a chromatogram was run b y the ascending technique with five spots of Cl4-1abeled salicylic acid ranging in activity from 313 to 7400 d.p.m. The solvent system used was henzeiie-glacial acetic acid (8 to 2). Upon spot visualization by spra>-ing with ferric nitrate solution. a circle 6.2 sq. cm. in area n-as inscribed about each spot and the spot was cut out and transferred to a vial cont,:hing 15 ml. of the sylcne-dioxane-ethj.1 alcohol scintillator. Counting of these samples before acidification again showed abnorinally high count rates (141 to 2100/, of theoretical), whicli did not decrease with time under the storage conditions mentioned above. Upon the addition of 25 pl. of concentrated hydroc,hloric acid to each jample. the counting rates ~
recorded (Table 11) indicate an essentially linear relationship betn-een applied and observed activities.
Table I. Count Rate Data on Ci4 on Filter Paper in Toluene-Based Scintilla tor I'aper Area, C.P.M. Sq. Cm. Recorded % Theoretical" 2.3 7, 705 616 2.8 3.8 4.6
7.5
19,900 3,940 2,700 1,620
1580 314 216 129
1250 d.p.m. represents 100yo activity.
Using R h a t m a n S o . 1 filter paper and the scintillator system described, the procedure outlined is valid for the quantitative radioassay of C I4-labeled salicylic acid on paper chromatograms. Acidification of the scintillator eliminates the s p r i o u s count rates observed in nonacid media. TF'orkers using different types and grades of paper or different scintillator systems should evaluate their procedure individually to
Table II. Count Rate Data on CI4 on Filter Paper in Acidified XyleneDioxane-Ethyl Alcohol-Based Scintillator Bctivity Activity Applieda Observedb 7 0 Theoretical 313 1480 2960 4440 7400
*
74
23.6 21.4 23.3 24.1 26.0
316 688 1069 1922
D.p.m CI4 activity. C.p.m. corrected for background.
establish counting efficiencies under specific conditions. LITERATURE CITED
Bousquet, W. F., Christian, ,J. E., J . A m . Pharm. Assoc., Sei. Ed., in press. ( 2 ) Herberg, R. J., Science 128, 189
( 1)
(1958).
F. BOL-SQEET JOHN 11:. CHRISTIAS
WILLIani
Bionucleonics Department Purdue University Lafayett,e, Ind.
RECEIVED for review January 28; l!)CiO. Accepted Llarch 7, 1960.
Gas Chromatographic Retention Time of Formaldehyde SIR: I n studying the carbonyl components in smoke from cigar tobacco, it was found convenient to separate and fractionate these component's through the medium of their 2,4-(1initrophenylhytlrazonts ( D X P H ' d , For identification purposes, the flash exchange gas chromatographic method of Ralls ( 2 ) was used. X series of known DXPH's was run , t o establish retention times. Among these samples as formaldehydeDW'H. d formaldehyde peak could not be. seen when a short (&foot Craig polyester-succinate) column was used, because of its proximity to the large carbon dioxide peak. (The carbon dioside arises from decarboxylation of the a-ketoglutaric acid eschanger.) Hon-ever, when a longer column (10foot' Carbonax 20x1) was used, a formaldehyde peak was easily detected. Until this time, apparently no one
had been able to report a retention time for formaldehyde ( 1 ) . Presuniabl?- this was due to the difficulty of maintaining the monomer in liquid form for injection into a gas column by yringe. Therefore, the method of Ralls is offeied as a very simple and convenient n a y to obtain this retention time. Although our o n n e-,pericnce ha. been limited, it is felt that any pas chromatography column capable of resolving carbonyls could be used, p r o d e d it is long enough. I n obtaining a retention time for formaldeh) de of 2.30 minutes, our operating conditions IT erp : Column. Carbon ax 2011 on Chroniosorb (35/80 mesh), 10 feet, in stainless steel ('/a-inch). Detector. 4-filament thermal coilductivity cell. Detector current. 300 ma. Temperature. 90" i 0.3" C.
Helium flow. 32 ml. per minute. Recorder. 2.5 mv., 3-second pen speed, 30 inches per hour. LITERATURE CITED
(1) Gager, F. L., Jr., Philip IIorris, IIIC., Richmond, T'a., private communication. ( 2 ) Ralls. J. W.,,4s.4~.CHEM.32, 33'7 (1960).'
A . I. &!HEPkRTZ P. E. ~ ~ C D O U - E I . L
Eastern Utilization Research and Development Division hgricultural Research Service United States Department of Agriculture Philadelphia 18, Pa. I ~ E C E I Vfor E Dreview February 4, lY(i0. -1ccepted March 7, 1960. Work supported in part by funds made available by the Cigar hlanufacturers Association of America, Inc. Atention of a specific commercial product does not constitute endorsement by the United States Department of Agriculture over others not named.
VOL. 32, NO. 6, M A Y 1960
723
