110
ANALYTICAL CHEMISTRY
(43) Portevin. h., Chaudron, G., and Aloreau, L., Compt. rend. 204, 1252 (1937). (44) Post, C. B., and Schoffstall, D. G., Trans. Am. Inst. Miniiig Met. Engrs. 162, 390 (1945). (45) Rooney, T. E., and Barr, G., J . Iron Steel Inst. London 119, 575 (1929). (46) Rummel, T., Esmarch, W., and Beuther, K., Metallwirtschaft 19. 1029 (1940). (47) Scafe. R. Ri., Tjans. Am. Insf. Mining Met. Engrs. 162, 375 (1945). (48) Schmid, E., and Schweinita, H. D. von, Aluminium 21, 772 (1939). (49) Schwarte, H. rl., and Guiler, G. M., Trans. Am. Foundrymen’s Assoc. 47, 742 (1939-40). (50) Scott, W. W.. “Standard Methods of Chemical Analysis,” vol. I, p. 443, Van Nostrand, New York, 1939. (51) Shields, B. AI., Chipman, J., Grant, N. J., and Carney, D. J., Materials & Methods 40 (No. l ) ,220 (1954). (52) Sloman, H. A., Harvy, C. A , , and Kubaschewski, O., J . Inst. Metals 80. 391 (1951-2). (53) Smith, D. P., “Hydrogen’in Metals,” p. 160, University of Chicago Press, Chicago 1948. (54) Smithells, C. J., “Metals Reference Book,” p . 380, Interscience, Sew York, 1949.
Sterner-Rainer, R., Z. Metallkunde 23, 274 (1932). (50) Swain, ii. J., J . Inst. Metals 75, 863 (1949). 157) Tammann, G . , and Schneider, J., Z . Q?WQ. u. allgem. C‘hem. 172, 43 (1928). (58) Thompson, J. G., Trans. A m . litst. Minino Met. Engrs. 162, 369 (1945:. (59) Titanium Metals Corp. of -Iinerica, New York, “Handbook on Titanium Metal,” pp. 66, 75, 89, 1952. (60) Veinberg, G. Ya., and Proshutinskii, 9. I., Zavodskaya Lab. 6, 422 (1937). 161) Walter, D. I., A x . 4 ~ CHEx . 22, 297 (1950). (62) Wells, J. E., and Barraclough, K. C., J . Iron Steel Inst. Loudon 155, 27 (1947). 163) Willems, F., Aluminium 23, 337 (1941). (64) Yao, Y . L., and hfilliken, K. S., ANAL. CHEY. 25, 363 (1953). (65) Yavoiskii, V. I., Zavodskaya Lab. 11, 406 (1945). (66) Ibid., 13, 262 (1947). (67) Yeaton, R. A., Vacuum 2, 115 (1952). (68) Yenson, T. D., and McGeary, R. K., Trans. Am. Inst. Mining Met. Engrs. 162, 355 (1945). 155)
R E C E I V E Dfor review June 4. 1955. Accepted September 21, 1955.
liquid Scintillation Counter for Carbon-1 4 Employing Automatic Sample Alternation ARTHUR J. WEINBERGER, JACKSON B. DAVIDSON’,
and
GUS A. ROPP
Chemistry Division, O a k Ridge N a t i o n a l Laboratory, O a k Ridge, Tenn.
A scintillation counter for carbon-14, which uses solution phosphors, has been described. This instrument has been used to measure the activity ratio of a pair of carbon-14-containing solutions with an average deviation of approximately0.5% from the mean and from the calculated value. It employs automatic alternate counting of three cells, one of which is usually a standard and one a background. The alternating feature was designed to reduce possible effects of instrument drift and changing background on the results from long counting periods. A series of experiments to test this feature is reported. The counter is simple, electronically. The mechanical parts are not difficult nor expensive to build where adequate shop facilities are available.
T
HE vibrating reed electrometer when used with a Borkowskitype ionization chamber is a standard instrument for the assay of carbon-14 compounds. The organic compound is burned to form carbon-14 dioxide, which, when introduced into the ionization chamber, produces a current which is measured by the electrometer (6, 6). Scintillation counting (8, 3) of carbon14 compounds dissolved in organic solvents containing scintillators such as terphenyl has the advantage that the compound need not be burned. Several instruments have been described for this type counting. The present paper discusses the design and testing of one such instrument. The principal objective was to obtain an instrument M hich would measure precisely and reproducibly the ratio of activities contained in two or more samples of the same compound, n ithout regard for whether or not the individual measurements of the activity of any sample varied from time to time. It is well known that for most tracer studies ratios of activities of several samples are needed, whereas the absolute activity of any one sample is not needed. For this reason an automatic sample changer was designed to allow three samples (one to measure the 1 Present address, Instrumentation and Controls Division, Oak Ridge Sational Laboratory, Oak Ridge, Tenn.
background which could be subtiacted from each of the other two before calculation of their ratio) to be counted alternately for short intervals with the integrated counts for each sample stored on a eeparate register. It was felt that such an alternating system could reduce the effect of instrument drift and probably permit ratios of activities of several samples t o be measured with greater precision than that obtained with instruments not employing sample alternation. The use of refrigerated systems employing two photomultipliers, two amplifiers, and a coincidence circuit to reduce background has been described by several writers (1, 4), and a commercially available unit has been recently introduced ( 7 ) . These dual channel systems are characterized by high efficiencies for carbon-14 (values as high as 75% have been reported) and by low backgrounds (100 to 200 counts per minute). However, they are bulky, since refrigerators and rack mounting are required, and the electronic circuitry is fairly complex. These systems also require that the counting cell and contents be refrigerated to 0” C. or belolT, which in general decreases the solubility of the scintillator and the sample in the solvent. Finally, they are expensive. For the present purpose it seemed that a counter using one refrigerated phototube and one amplifier would be more stable, easier to maintain, and, incidentally, less expensive than one using two such units and a coincidence circuit plus a large refrigerator. The amplifier and scaling unit employed in the counter described here are those of a commonly used proportional counter; thus it may appeal to those who already possess such equipment DESIGV
Detector. The detector is a lP21 photomultiplier which is operated a t liquid nitrogen temperature. The tube rests with its base u p on the bottom of a 500-ml. Deyar flask with its cathode facing a 1.25 x 2 inch unsilvered window in the side of the flask. The base of the phototube is partially removed by chucking the base end in a lathe and cutting away the upper part of the base until there is no mechanical connection with the glass envelope, except that of the wire leads. This is done in order t o eliminate the strains on the envelope and on the soldered connections to the prongs of the base, which are caused by the drastic cooling which occurs when liquid nitrogen is poured over the phototube.
111
V O L U M E 28, NO. 1, J A N U A R Y 1 9 5 6 I
system. It consists of a carriage operated by a yoke driven by s Bodine motor a t 10 r.p.m. Inside the carriage housing (Figure 2, top view) are located 3 double pole, double throw micro switches which stop the motion of the carriage, position the cell being counted in front of the window, switch in the appropriate register. and start the timer. At the end of the timing period, nThich may be from 1 to 20 minutes, the carriage moves the next sample in place and the timing cycle is repeated. Tn order to Dosition the cells accuratelv
I
I
I
l
l
+
-r Figure 1.
Schematic diagram of counting system
Although part of the base of the tube is removed, it remains rigid and is easily plugged into a standard socket around which are wired the voltage-divider resistors. During operation these resistors are also at liquid-nitrogen temperature. The high voltage and &gal leads to the phototube, which serve to keep the phototube in place in the Dewar flask, are brought out to conneotors mounted on the side of the sheet hrsss housing. The phototube output is developed across a 15,000-ohm load resistor and is fed directly to the input of the amplifier through an 8-inch coaxial cable. Sample Changer. An alternating sample changer was included which would count, in short cycles, the cell series: sample, hackground, and standard. Figure 1 is a schematic diagram of the
and makes i t 'light tight, egcept for the opening seen by the phototube. Glass plate lids were used. The dimensions of the window in the cell holder are several millimeters larger than those of the opening seen by the phototube, 60 that the geometry is essentially constant, The changer may be manually operated hy setting the timer for a short interval, say 5 seconds, and bringing the desired sample into position. The timer is then turned off.
Table I. Mcasured Ratios of Suoeessive Daily Pairs of Samples Having a Calculated Ratio of 2.000 (Samples were mounted dternstdu f o r t h e interval shown. Total counting time for a given interval was 4 hours1 Inter9% "SI De".' Av. from Ti,,,e, Sample Pair Number for Caled. aiinUtes I 2 3 4 3 0 Interval Ratio 1 1.098 1 . ~ 3 -0 4 5 1.081 1.973 1.077 -1.1
in
1.080
20
GO 120
1.975 1.070
1.078
2.002
1 . ~ 5 2 1.970 1.934 1.906
1.070 1.975 1.965
-1.0 -1.5
-1.8 -1.8
F i g u r e 2.
Assomhled eountii
Table 11. Results Obtained ,me Pair of Samples Over Period.of 2 2 Lyu.m z-a.-L.Aute, l-IIour, and %Hour Intervals 2.0001 Ratio of C ounts 10 one-minute 2 one-hour i"ter"dS intervals
(Calculated ratio 4-H,our Period from W I i i C h Ratio W a s Taken
Av. de". of mean
9% de". of mean
from o d d value
=
2 two-hour
i"tenixls
fn.008
*n.oii
*0.017
-0.1;
-0.8
-1.0
Amplifier and Scaler. The counter proper IS a IUucleitr I n s t n inent and Chemical Co Model 162 proportional counter, which had been modified by the Chemistry Division Instrument Group for improved sigpal to noise ratio and dynamic range when used with a beta proportional counter. The modified circuit was retained except for the diode load a! the input. This NllS replaced hy a 15,000-ohm resistor. The high voltage supply was rebuilt according t o Drawing Q 11441 of Oak Ridge Nation21 Laboratory. This circuit was changed slightly t o give 700 to 1400 volts and was connected as a positive supply, It possesseti excellent regulation and is free from the tube selection which was necessary in the original Model 162 supply. The 0 t o 1 ma. meter in the counter was replaced with a 0 to 200 pa. meter, and provision was made for monitoring the phototube voltage with a potentiometer. [Circuit diagrams may be obtained from one of the authors (J. B. D.).] A switch operated by the sliding door of the cell
ANALYTICAL CHEMISTRY
112
_-
~~
run the solutions were discarded and the cells and their plate glass tops were washed thoroughly with xylene folIoTved by acetone. They were then dried. N o difference within experimental error was found when the cells, holders, and sample p o s i t i o n s Tvere interchanged.
Table 111. Sample of Data for Sample Pairs 1 and 5 from Table I Sample Pair 1
5
b
Alternation
R Total unning
Interval. Minutes 1 120
Time, Hours 4 4
Register Net Counting Counts/Rlin. Rate hcciimulated Register Countsn Gross Baok- Net Ratio Start of run E n d of run A B ground d B B / A A B d B 6 , 1 1 3 12,054 128 . ? . 9 8 i 1 1 , 9 2 0 1 Q93 91 184 84 170 1 0 , 2 3 4 19,645 14Oa 1 0 . 0 4 4 15,505 1 534 81 I(i5 81 160
D a t a from 32-Hour Run Showing Drop in Counting Rate Counting Rate Time Register Counts/Xlin. after Start A B 1 rnin. 92 182 10 min. 88 177 4 hr. 83.3 165.6 16 hr. 85.5 170.7 32 hr. 81.6 163,s Register counts X 128 = actual counts. Estimated from short counts a t beginning and end of 2-hour run.
R E S U L T S AYD DISCUSSIOY
Tables I and I1 shoiv the precision obtained when the ratio of the activity above background of two benzoic acid samples was determined. __ Table I11 gives a sample of the data from which the ratios in Tables I and I1 nere calculated. The shorter counting intervals gave ratios which deviated somewhat less from the calculated value and from the mean. The superiority of short time intervals is a rough measure of the value of the alternating feature of the instrument. The counting efficiency for carbon-14 for the results in the tables was approximately 3%. ~
housing removes the voltage from the phototnbc when the door is opened. The scaler section of the counter was unchanged except for the mounting of the three registers on the front panel and the inclusion of a scaling factor switch. The scaling factor switch allows the selection of the factor which stores the maximum number of counts on the registers without exceeding their speed limitations, (Figure 2 shows the assembled system.) EXPERIMENTAL
Preparation of Solutions for Test Runs. il stock solution ( I ) of 8.5 grams of recrystallized terphenyl in 2 liters of c. P. xylene was prepared. Each cell used for counting background contained 15 ml. of this solution. Stock solution I1 was obtained by dissolving 20 mg. (20 pc.) of benzoic-a-carbon-14 acid in 1 liter of stock solution I. Stock solution I11 was obtained by mixing equal volumes of solution I and solution 11. Thus, all three stock solutions contained the same concentration of terphenyl, but different concentrations of labeled benzoic acid. Fifteen milliliters of solution TI (0.30 pc.) were pipetted into a clean cell, and 15 ml. of solution I11 (0.15 pc.) were pipetted into a second clean cell in order to obtain a pair of cells with 2 to 1 ratio of activities. After each
LITERATURE CITED
Arnold, J., Science 119, 155 (1954). Farmer, E. C., and Berstein, J. d.,Ibid., 115, 460 (1952). Hayes, F. K.,Hiebert, R. D., and Schuch, R. L., I b i d . , 116, 140 (1952). Hiebert, R. D., and Watts, R . J., Sucleonics 11, 38 (1953). Neville, 0. K., J . A m . Chem. SOC.70, 3501 (1948). Raaen, 1’.F., and Ropp, G . A , , d s . 4 ~CHEM. . 25, 174 (1953). Tracerlab, Inc., Boston, Mass., Tracerlog, KO.59 (April 1954). RECEIVEDfor review April 25, 1955. Arcepted Septeniber 26, 1955. Based upon work performed under Contract Nnniber W-7405-eng-26 for Atomic Energy Commission a t Oak Ridge National Laboratory, Oak Ridpe. Tenn.
Report on Recommended Specifications for Microchemical Apparatus Alkoxy1
I
N PREVIOUS reports ( I , 9, 13-11) of the Committee on
Microchemical Apparatus, recommended specifications were published for pieces of apparatus which were either the most widely used for the work in question, or shown to be an improvement over the more widely used apparatus through tests made by the members of the committee and other cooperating chemists. In this report specifications are suggested for the semimicro alkoxyl apparatus, which has been selected on the basis of being the most widely used. Recommended specifications for an apparatus for the determination of alkoxyl groups were delayed until a collaborative study (IO, 11, 1 8 ) of methods for this determination had been conducted by the Association of Official Agricultural Chemists. I n this study, compounds representing ethyl and methyl esters and ethers were submitted to practicing microchemists, who had expressed their willingness t o cooperate. These individuals were asked t o analyze the samples by whatever methods they were using in their own laboratories and t o furnish detailed information on the procedure and apparatus. Where enough collaborators used a particular procedure or apparatus to permit the results t o be treated statistically, calculations were made t o determine which of these or their variations appeared to give the best accuracy and/or precision.
I
F L A S K -CAPACITY O f BULB 5 0 M L . APPROX
