The Radioactivity of Argon-371,2 - Journal of the American Chemical

The Radioactivity of Argon-371,2. Robert W. Kiser, and W. H. Johnston. J. Am. Chem. Soc. , 1959, 81 (8), pp 1810–1813. DOI: 10.1021/ja01517a009. Pub...
0 downloads 0 Views 458KB Size
ROBERTU-.KISERA K D

IS10 [COiVTRIBCTIOSFROM

THE

I\-.H. JOHSSTOX

RICHARD BESBRIDCE \VETHERILL LABORATORY OF CHE>IISTKY, PURDCE

Vol. I 1 CSIVERSITY]

The Radioactivity of k g 0 n - 3 7 ~ 1 ~ B Y ROBERTlv. &SER3 A N D 55'. H.JOHKSTON RECEIVED OCTOBER 14, 1958 Proportional counter spectroscopy has been utilized in the investigation of the electron capture processes of radioactive argon-37. From quantitative measurements of the K- and L-capture of argon-37 in various counters using a number of different gases, the L/K-capture ratio of argon-37 has been determined to be 0.102 =t0.008. Recent results which might iiidicate that the argon-37 L/K-capture ratio might be significantly greater than 0.1U are discussed. Measurements show that double K-vacancies occur to the extent of (3.7 i 0.9) X lo-' of I(-capture events. The half-life of argon-37 against dtcay was observed to be 31.30 f 0.14 days.

Introduction Argon-37 was first shown to decay by the process orbital electron capture by IYeimer, Kurbatov and P o o ~ . ~the^',^ , ~ gave a half-life against decay uf 34.1 days. More recent determinations by AIiskel and Perlman6 using ionization chamber and proportional counter techniques give the half-life as 35.5 days and 35.0 f 0.4 days, respectively. A I a r ~ h a k , ~Segreg J and Bouchez, Daudel, Daudel and Jluxart'O predicted from theoretical bases the existence of the phenomenon of L-capture, and it was first shown by Kirkwood, Pontecorvo and Hanna" that such L-capture could be observed experimentally in argon-37. Pontecorvo, Kirkwood and Hanna12 found the ratio of L-capture to K-capture to be 0.08 to 0.09, using proportional counter spectrometry. Lange\-in and Radvanyi,I3 also using proportional spectrometry, determined the L/K-ratio to be 0.092 0.010 or -0.005. However, Rubenstein and PnyderI4 pointed out that the charge spectrum of the recoil chlorine-37 ions resulting from the orbital electron capture in argon, as measured by Koioed-Hansen, l b was not consistent with the measured L/K-capture ratio of 0.09. 'ItTinther'6 attempted to show t h a t the charge spectrum of recoil chlorine-37 ions could be explained if about L'5s/c of the orbital electron capture was L-capture. Primakoff and Porter'' first calculated from theoretical bases the probability of orbital electron capture accompanied by atomic excitation and ( I f

+

( 1 1 T h i s work was supported in p a r t by t h e United States Atomic E n e r g y Commission under Contract No. A t ( l l - 1 ) - 1 6 6 with Purdue University. Presented a t the 134th hleeting of the American Chemi-

cal Society, Chicago, Ill., September 7-12, 1958. ( 2 ) Taken in part from a thesis submitted by Robert W. Kiser t o the Graduate School of Purdue University in partial fulfillment of t h e rerluirements for the degree of Doctor of Philosophy, January, 1958. ( 3 ) NOWa t Department of Chemistry, Kansas S t a t e College. M a n h a t t a n . Kansas. , Reprint requests should be addressed t o this author. (-11 P. K . Weimer, J. D. Kurbatov and h l . L. Pool, P h y s . R e x , 60, 4tio (1941). ( 5 ) P. R. N'eimer, J. D. Kurbatov and A l . L. Pool, ibid., 66, 209 (lY44). ( t i ) J . A. lliskel and A!, L. Perlman, ibid , 87, 543 (1952). ( 7 ) R. E. Marshak, ibid., 61, 388 (1942). ( 8 ) R. E. hlarshak. i b i d . , 61, 431 (1942). (9) E . Segre. i b i d . , 71, 274 (1947). 110) R . Bouchez. R . Daudel, P. Daudel and R . hluxart. J . phys. > o , I ~ i t w[,8 ] 8 , 336 (1947). f 1 1 ) D . H . IV. Kirkwood, B. Pontecorvo and G C. Hanna, Phys. R r f ' . 74, -107 (19-18). 112) R.Pontccorvo, D . H . W. Kirkwood and G. C.H a n n a , ibid., 'IS, 982 f lTl-1Q). : 1 3 ) A!. 1,angevin and P . Radvanyi. ConiPL. rend., 241, 33 (1955). (1.1) R . A. Rubenstein and J . N . Snyder, P h y s . Reu., 99, 189 (1955). (12) 0 Kofoed-Hansen. ibid.. 96, 1045 (1954). f l i i ) A IVinthcr. J . phys. radium, [SI 16, 562 (1955). (17) H . Primakoff and F. T. Porter, P h y s . Rev., 89, 930 (1953).

ionization. These authors gave the probability for a double K-vacancy per K-capture event in argon-37 to be 2.S X 1OF4. lYolisberg** later showed that by inclusion of terms neglected by Primakoff and Porter, the theoretical value is 3.9 X Levingerlg also calculated theoretically the probability of double K-vacancies in argon-37, arriving a t the value of 4.9 x 1 0 F . lliskel and Perlman20J1are the only workers to have experimentally observed these double I;vacancies in argon-37, although L a n g e ~ i n ? ~ , ? ~ has experimentally determined the double Kvacancy occurrence in germanium-T 1. Miskel and Perlman gave (3.9 i O.T) X as the occurrence per K-capture event in the decay of argon37. In view of the apparently contradictory results of the experimentally-determined L/K-capture ratios, the relatively large error in the reported values of the half-life, and that only one deterniination has been made of the double K-vacancies in argon-37 decay, we subjected argon-37 to further investigation in our laboratories. Making use of the techniques of proportional counter spectrometry and using a number of different counters and counting gases, we have obtained a value for the L/K-capture ratio of argon-37 of 0.102 f 0.008. The details of the investigation and the results are given below and an explanation is offered for the apparent conflict with recoil charge spectrum measurements. The occurrence of double K vacancies per K-capture event was determined to and the half-life was found be (3.7 f 0.9) X to be 322.30 f 0.14 days. Experimental The techniques of proportional counter spectrometry have been employed in this investigation. T h e proportional counter, integrating the fast cascade of X-rays and Auger K-electrons which result from each L-capture event, gives a sharp "K-peak" in the pulse height-distribution spectrum. .\ny K-capture event which is followed by escape of the chlorine K-X-rays from the sensitive volume, thereby producing L-radiation, as well as the capture of the Lr-electron and the Auger L-electrons, gives rise to a n "Lpeak. " In our determination of the L/K-capture ratios, the intensities of the L-peak, L, and of the IC-peak, K, were measured by integral counting using a 100-channel pulse height anall-zer. Letting F be the K-fluorescence yield of chlorine and E the escape probability for the K-X-ray, we obtain (L/K)rneaa - EFII $- (L/'K)measl (L/K)true (18) M . Wolfsberg, i b i d . , 96, 1712 (1954). (19) J . S. Levinger, i b i d . , DO, 11 (1953). (20) J . A. Miskel and 14. L. Perlman, i b i d . , 94, I683 (1954). ( 2 1 ) J . A. hliskel and h l . L.Perlman, i b i d . , 96, 612 (1954). (22) M . Langevin, Compf. rend.. 246, 664 (1957). (23) M. Langevin, J . phys. r a d i u m , [SI 19, 34 (1958).

The assumption made here (since no reliably accurate information could be found in the literature), that the fraction of Ko-X-rays in the I;-series is unity, will not seriously affect the L/I;-ratio results; a change of 10ycin this value affects the results by approximately the error which is quoted, and i t should be noted t h a t such a change mould cause the value of ( L / K ) t r u e to become larger and not smaller. Variation of the counter sizes and the counting gases will cause variations in the amount of escape of the chlorine Xray. Since the L-X-rays and the L-Auger electrons are completely absorbed, the variations in the counter size and the counter gas will not affect the L-capture portion of the “L-peak.” The correction for E;-X-ray escape in the “L-peak” may be made from a knowledge of the counter size, counting gas, filling pressure, S - r a y absorption coelficients and E;-fluorescence yield. These are two ways t h a t elimination of some uncertainty in the calculation of the amount of escape may be accomplished: ( a ) allow no escape or ( b ) allow 1007, escape. It will be noted in the tabular presentation of the results t h a t method ( b ) has been used in all b u t three cases. T h e three cases which were not 1007, escape, but which had escape factors ranging between zero and unity mere carried out t o check the validity of the method of escape calculations for values between zero and unity; the method appears satisfactory. Lye have used the value of 0.108 for the K-fluorescence yield of chlorine, as given by Haas.“ (See also Broyles, Thoiuas and H a y n e ~ . ~ By ~ ) means of the tables and “universal” functions given by Henke, White and Lundberg,*‘ we have obtained, for use in the above-mentioned corrections, the following mass absorption coefficients for a 2.82 kev. X-ray: carbon, 102 =t 2 cm.*/g., oxygen, 260 =t 4 cm.*/g., and argon, 208 2= 2 cm.Z/g. In the calculation of the path lengths to be used in calculating the escape probabilities, we have used the model designed by Lind.27 The path length d is related t o the radius R of a spherical volume by the equation d = 0.75R This is taken t o hold, as s h o r n by Lind, as approximately true for cylindrical volumes. In this case, however, the volume of the cylinder is first converted t o a spherical volume and then R is calculated. Since the validity of the approximation varies, depending roughly on the length t o diameter ratio, slightly different values of the constant, selected on such bases, are given along with the values of the L/K-capture ratios for individual counters. An error of 15Yc in the d / R ratio will affect the results by approximately 1% when the escape probability approaches unity, thus making the technique of measurement of the L/K-capture ratio at or near a n escape probability of unity even more attractive. T h e escape probabilities were qalculated according to the equation l n ( 1 l E ) = (Cc/p)pd where ( p / p ) is the mass absorption coefficient and p is the density. T h e double K-vacancies should exhibit a “peak” at approximately 5.7 kev. By integration of the peak between 4.0 and 9.0 kev. and by subtraction of the background, any double E;-vacancies may be observed, after correction for chance coincidences and counter dead time. For the purposes of this investigation, we had need for a stable high voltage supply. T h e high voltage for operation of the proportional counters was obtained from a Fluke model 400-BDA power supply.28 This supply had an outp u t variable from 500 to 5100 v. d.c., positive or negative, a t a current drain of 0 t o 1 ma., and the stability was better than O . O l ~ o per hour. T h e ripple was less than 5 mv. at a n y output voltage and current, and the voltage resolution was 50 mv. at a n y output voltage.

IVe employed a Bay a n ~ p l i f i e r ?for ~ amplification of the proportional pulses. The unit consisted of a preamplifier stage close to the counter and the final amplification stages removed a short distance from the counter. The rnasirnurn electronic gain attainable was 215,000. Escellent linearity of response of the amplifier for the proportional pulses \vas established experimentally. A 100-channel pulse height analyzer, RIDL model 3300,a was used for measuring and recording the pulse heiglitdistribution spectra. The escellent linearity of response of the analyzer was shown both by pulse-generator calibrations and by proportional counter calibrations involving the use of argon-37 and iron-55. The physical cfiaracteristics of the counters eniployed in this investigation are given in Table I . Details of the effects of changes of counter parameters on counter characteristics will appear shortly in publication. I t will be noted, however, t h a t counters I11 and SI’ are hiaze-typejl counters.

(24) M . Haas, A n n . Physik. (51 18, 473 (1933). (25) C.D . Broyles, D . A. Thomas and S. K . Haynes, PhyE. Rev.,89, 715 (1953). (26) B. L. Henke, R . White and B. Lundberg, J . A p p l . Phyr., 98, 98 (1957). (27) S. C . Lind. “The Chemical Effects of Alpha Particles and ElecLrons,’’A.C.S. Monograph. No. 2, Chemical Catalog Co., New York, N.Y , 1928. pp. 92-94. (28) John Fluke Mannfacturing C o . , Inc., 1111 W. Nickerson Street, Seattle 99, Washington.

Fig. 1.-K-Capture

TABLE I PHYSICAL CHARACTERISTICS OF COUSTERS Counter no.

Shell

Cathode

1”tirial shell Center radius wire (in.)

Center wire Counter radius length (in.) (in.)b

Pyres External“ W 0 . 5 1 0.0010 8 . 3 SV Soda Esternal“ W 1.00 ,0010 15.0 Brass Pt 2182 Brass 0.94 ,0125 1 0 . 5 b‘ 0 . 9 4 ,0010 15.6 3151 Copper Copper A silver layer placed on the external wall of the counter shell. The silver layer was prepared by painting the shell surface with Silver Print Paint (General Cement ManuLength of actual body facturing Co., Rockford, Illinois). of the counters. I11

All gases used were obtained from h f a t h e ~ o n . ~hlethane ? was C.P. grade, stated t o be 99.0% pure; ethane was stated t o be 957, purity; and argon was stated t o have a minimum purity of 99.97,. The carbon dioxide used was of “Bone dry” grade and was stated to have a minimum purity of 99.8yo. These gases were not further purified when withdrawn from the cylinders in which they were furnished.

Results and Discussion Typical proportional counter spectra of argon-37 are shown in Figs. 1 and 2 . It will be noted that

01 1.0

W

30

40

ENERGY (MEV).

distribution in the decay of argon-37.

the resolution, 1/2 rtI2,is 12.670 for the “K-peak” and is 50y0 for the “L-peak.” It may be further (29) Bay Engineering Company, Chicago, Illinois. (30) Radiation Instrument Development Laboratory, 5 i 3 7 South Halstead Street, Chicago, Illinois. (31) R hfaze, J . p h s s r n d i a m , [SI7 , 164 (1946). (32) The hlatheson Company. East Rutherford. h7ew Jersey.

iiients It may be seen from this that if the theoretical calculations also include some sniall contributions from double E=-vacancies and by use of the revised value of the LjR-capture ratio ~i 0.102 =tO.O&S, agrecnicnt between our proportiori:il counter spectrometric measurements of the L E;capture ratio of argon-37 and of the charge distribution measurements of the chlorine-37 recoil ions made by Snell and Pleasonton is quite good. From an examination of Table IT1 it is seen that one of the possibilities included by Ruberisteiii and SnyderIf was that of 75yo K-capture. Sy LI-capture and 17% LIT-and LIII-capture. From

008

TAnLE CII~IR~7l 1 :~ I S ~ R i l l l ~ l 0I1~' JCSI I I A R I N E - r ? i I O N S _-__ ~- Theoretical" Cli.ir1, (l.liILIi,rIi 0.0SL

111

RESCLTINC FROM

,