w~———-—--v——..___,-__~ _ vi" ’ —— ’— _————- ———— é _——-— ;.——— _———; —'——— __—A_; _.—— _—— __—Q— 4——-— ’ AN APPARATUS DESIGNEB FOR A STUDY OF THE MEAN LIFE OF FHE 1294(5‘?‘ NUCLEAR ENERGY STATE OF [R 191 Thesis ‘or @319 Degree of M. S. MECHEGM STATE UNEVERSETY Ronaid Ames Hill 1958. --— IN‘I'EWIE‘IEMMLWLW AN'APPARLTUS DESIGNED FOR A STUDY OF THE MEAN LIFE OF THE 129-KEV NUCLEAR ENERGY STATE OF IR 191 3! RONKLD AIRS HILL A THESIS submitted to the College of Science and Arte of Michigan State university of Agriculture and Applied Science in partial fulfillment of the requirements for the degree or ‘IASTER OF SCIENCE Department of Physics (1958) RONALD ARES HILL ABSTRACT An apparatus is described which is to be used in the determination of the mean life or the lag-key gamma tran- sition in iridium 191 by nuclear resonance fluorescence.. The apparatus consists of a magnetically suspended, magnetically driven, high speed rotor upon which a source of iridium.191 gamma rays (Os 191) is to be attached in order that the resonance condition be restored through the DOppler effect. A gating system is employed to turn on a scintillation counter when the source is approaching the scatterer to improve the signal to noise ratio. Peripheral velocities of the order of 4 x lO4 cm/see for a one inch diameter steel rotor have been obtained, well above the 2.2 x lo4 cm/sec required for the Ir 191 experi- ment. A test of the gating circuit indicates that the scin- tillation counter is turned on for a larger portion of the rotor's cycle for faster rotor velocities. This effect will need to be compensated for in the Ir experiment. ACKNOWLEDGEMENTS I wish to express my sincere appreciation to Dr. W. H. Kelly and Dr. G. B. Beard for their help and encouragement during the progress of this work. I am gratefhl to Dr. 3. Handler for his help and sug- gestions in the develOpment of the apparatus and to.lr. u. Chapman of the General Motors Technical Center for his assistance in the design of the apparatus and for the drive coils so kindly furnished. I wish to thank ur. a. Perkins of New Departure for furnishing the steel ball bearings, the Messrs. C. ningston, R. Hoskins and R. Salemka for their help and suggestions in the construction of the apparatus, and the Air Force Office of Scientific Research and the Physics Department of Michigan State University for making this work financially possible through a research assistant- Ship. TABLE OF CONTENTS Page INTRODUCTION 1 IAGNBTICALLY‘SUSPENDID ROTOR AND Assocxlrsn APPARATUS 9 Historical 9 Magnetic Suspension Apparatus 10 Magnetic Drive Circuit 13 The Rotor 15 Nuclear Counting Equipment 15 PERFORMDNCE OF THE APPARATUS 22 Magnetic Suspension system. 33 Iagnetic Drive Circuit 24 nuclear Counting Equipment 25 APPENDIX ' 27 BIBLIOGRAPHY 33 LIST OF FIGURES Figure: Page: I. Decay Scheme of Ca 191. 2 2. magnetic Suspension Apparatus. ll 3. magnetic Suspension Circuit. 12 4. magnetic Driving Circuit. 14 5. Photomultiplier Arrangement. 16 6.(a) Block Diagram.of Detection System. 19 (b) 931A Pro-Amplifier Circuit. 20 (0) Gate Circuit. 21 7. Gate Width as a Function of Rotor Frequency. 26 1. INTRODUCTION The investigation of electromagnetic transitions be- tween nuclear states together with theoretical calculations on internal conversion coefficients, angular correlation and Coulomb excitation have provided much useful data for the study of the prOperties of nuclear energy levels. Some of the quantities characterizing these nuclear states are energy, total angular momentum, parity, magnetic moment, electric quadrupole moment, partial level widths, and mean lifetimes of excited states. or particular interest in this investi- gation is the apparatus to be used in the determination of the mean lifetime of the 129-kev gamma transition in iridium 191 by nuclear resonance fluorescence. Resonant scattering of gamma rays can occur when an incident photon has an energy corresponding to the energy difference between the ground state and an excited energy level of the scattering nucleus. In the experiment under consideration, one uses a radioactive source decaying to the element under study as the source of the exciting radiation. This is accomplished by using osmium 191 which decays by D- emission to iridium.191 in the excited state as shown in figure 1. 2. Os [9! 3/2- 74 K l4 Hour - ev 9/2— ' l5 Day l43-Kev ll/Z— 4.9 Sec 42-KQV 40 5/2+ mlo Sec l29-Kev 3/2+ ‘ Ir 19: (STable) 1 fig. 1: Decay scheme of Os 191 . Remover, due to the recoil of the decaying nucleus, the energy of the emitted radiation is less than the energy difference between the two levels corresponding to the transition. If E is the energy of the emitted radiation, 3n the energy of the recoiling nucleus and lo the energy of the excitation level, then by energy conservation E =hY=Eo-En (I) where h is Planck's constant and 7 the frequency of the radiation. Row by the conservation of momentum, 2 2 E ::-—EbE—::-—jiL—:: 01‘0 ::-J;E:2’ (2) “ 2m 2m ch2 ch where %:Qare the momenta of the recoil nucleus and gamma ray respectively and m.is the mess of the emitting nucleus. Combining equations \1) and (2) one obtains 2 — ___E__ hV_E° chZ The same amount of kinetic energy is transferred to the 3. scattering nucleus. Thus the total energy difference is 2 2 AE : E /mc (3) and the system is out of resonance by AB. Irom electrodynamical theory, the resonant scattering cross-section is given by the Brcit-Iigner relationz, 2 2 A F at =9 2' 2 (4) |+2J 811’ (E- E0) + (r72) where g:—————£a J5 and J are the excited and ground state I+2J 3 spins respectgvely, )\ is the wavelength of the incident radiation, E is the gamma ray transition probability for the absorption and re-emission transition and F'is the total width of the level. The two possible modes of de- excitation for the 129-kev transition in Ir 191 are by emission of a gamma ray photon or by internal conversion. If He is the number of conversion electrons and fix is the number of photons emitted in the same time interval, the total internal conversion coefficient is N VV (x; z: e ::-—£L— (5) N 5 WX where‘Ix is the probability per unit time for emission of a photon and I; is the probability per unit time that the same nuclear field will transfer its energy to any bound electron of its own atom. The total transition probability, i.e., the probability of the nucleus undergoing any transi- tion is sze+wtz '/"5 (6) where 131s the mean life of the transition. By the uncer- 4o tainty principle, P12?) and thus Fzfifl+m) (fl Combining equations (4) and (7), one obtains 2 >‘2 (I +Pd- ) (8) 8“ (e - a)2+(ryz)2 It may be observed that for a large (in in comparison with O":g the level width I", the cross section becomes unobservably small. In practice, level widths of the order of lO'G-ev and larger can be measured which correspond to mean life- times of the order of 10’9 seconds and shorter. Several methods have been employed to restore the resonance condition; among them are the recoil due to pre- ceding beta or gamma emission3, thermal motion by increasing the temperature of the source4, and through mechanical mo- tion5’5. All the methods require that the source be moving toward the scatterer with a velocity of the order of uzn/me so that the energy shift may be compensated by the Dappler- effect. For the 129-kev transition under study, us:2.2 x 104 cm/sec, a velocity which can be obtained at the edge of a high speed rotor. An expression for the resonant scattering cross-section which takes into account the mechanical velocity u of the gamma source and the thermal velocities of the source and scattering nuclei has been derived byiloon5’6’7. Consider that the thermal velocities of the nuclei are distributed 5. as in a gas at a temperature ‘1'. Then the probability that a source and scattering nucleus have thermal velocity com- ponents that differ by v cm/sec in the gamma-ray direction 13 _ mv2 1 m— P(v> dv : m e dv (9) ‘1TrK_T Since the Dappler-effect adds an energy (u + v)n/c to the game ray, (E—Eo) is replaced by E (u+v)— ——-—2— c m C in equation (8) and there results 12 (1+1) E :9? (‘E'WUW‘ me)2 + ("'21)2 llultiplying by P(v) and integrating, +00 2 P _ A2 P(v) (Ti-T} dv “— _ m at + v- a? + (—92 The integrand has a sharp maximum around v :R/(mc)-u, and so little error results in replacing P(v) by Pm? - u). 6. Thus 2 22 we) «no _ 9)\ F m 4KT dv O‘_—-——-—8 2 —— 2 2 2 Tr(l+ot) 41TKT E U+V__§_) + _r_ _ a me 2) and so ~ 2 E 2 2 1 _ m U“ mc) 0“ 3 2 e ('0) where >\ is replaced by hc/L’. At resonance, the rotor speed u: R/mc and the resonant scattering cross section becomes 2 2 ‘ h c r' 2 0V : 9 3 2 m c (H) 4 E (I +09 41“” 8 For the 129-kev transition in Ir 191, on: 2.4 , Jo: 5/2, 18: 3/2 and thus g:3/2. Assuming a source and scatterer temperature of SOOQK, o~:l.7 x [0.49 r cmZ/ev (l2) Since P is of the order of 10-5 evl, 0"~l.7 barns. For 'a rotor speed of u=0 cm/sec, the resonance scattering cross-section depends upon E 2 mi ‘ Fn'E') 4 K T o Calculating this factor using T=300 , E2 mc24 KT 2 : .405 Thus for this off resonance condition, 0": .69 barns. 7. In the case for a source at a temperature T1 moving toward a scatterer which is at a temperature Ta;é T1, the average temperature is used in equation (10), i.e., T is replaced by (Tl+-T§)/2. Thus by lowering the temperature of the scatterer, one should expect an enhancement of the resonance scattering cross-section as the rotor speed approaches the resonance velocity u. Competing scattering processes which make the obser- vation of the nuclear resonance fluorescence effect diffi- cult are Compton scattering,.Rayleigh scattering, Thomson scattering, and Delbruck scattering. for incident radia- tion of 129-kev, the quanta scattered inelasticelly at a scattering angle of 90° are of lOS-kev and thus an energy selective gamma detector may be used to discriminate against them. Rayleigh scattering is due to the interaction of quanta with the bound electrons, Thomson scattering is due to the interaction of quanta with the nuclear charge and Delbruck scattering is the result of virtual pair production in the nuclear Coulomb field. Rayleigh scattering predomi- nates in the Xeray region, the cross-section in this exper- iment being of the order of .34 barns. Thomson scattering is negligible at 129-kev in comparison to Rayleigh scatter- ing while Delbruck scattering predominates at high energies. The resonant scattering experiment on the lBO-kev transition in Ir 191 is carried out by attaching Os 191 to a solid steel high speed rotor which can be spun to vel- 8. ocities of 4.0 x 104 cm/sec, well above the 2.2 x 104cm/sec required for the resonance condition. A scintillation counter is employed to detect the Rayleigh and resonance scattering from an iridium scatterer and the Rayleigh scat- tering from a lead scatterer. By choosing the lead scat- terer to be of thickness ‘3pr 2“ :Ir tIr ('3) PD FR) where glis the density and z the atomic number of the ele- ment indicated by the subscribt, a scattering cross-section- al area containing the same number of electrons is obtained for bOth scatterers. Thus the Rayleigh scattering will be approximately the same from both scatterers. One can then plot the ratio of iridium scattering to lead scattering versus the rotor speed to Observe the resonance scattering effect. The data obtained is used to determine the reac- nant scattering cross-section which is used in equation (12) to determine the width of the level and subsequently the mean life of the nuclear state. 9. IAGNETIGALLY SUSPENDED ROTOR AND ASSOCIATED APPARATUS Historical: According to HacHBttieg, speeds in excess of several thousand revolutions per second were first obtained in 1925 by Henroit and Huguenard who reported spinning a top-like conical rotor to 11,000 r.p.s. The rotor rode on a cushion of air in the conical cup of a stator, and was spun by a circular row of obliquely directed air Jets. An improved form of this apparatus was first used by Noon6 to examine the resonance fluorescence effect in Hg 198. Later, netzger10 employed an air supported and air driven rotor copied from a design of Beamsn to study the resonance fluorescence in fig 198, Pr 141 and T1 203. In 1941, lbcl-‘Iattie9 developed an apparatus for magnet- ically rotating rod and spherical rotors. .ladHattie suc- ceeded in driving a 3/32" diameter rotor to 110,000 r.p.s. and in exploding a 3/16" diameter rotor at 49,000 r.p.s. In 1946, Beams, Young and Moorelz, with improvements in MacHattie's apparatus, exploded rotors of 3.9? mm diameter at 77,000 r.p.s. and .795 mm.diameter at 386,000 r.p.s. The maximum peripheral speed in all cases was about 105 cm/sec, well above the 2.2 x 104 cm/sec velocity required for the 129-kev transition in Ir 191. 10. Ilagnetic Suspension Apparatus: The apparatus described here is patterned after a model being used at the General meters Technical Center.13 The rotor R (Figure 2) is freely suspended in the glass vacuum chamber V by the axial magnetic field of the solenoid S. The horizontal stability of the rotor is maintained by the symmetrically diverging magnetic field of the solenoid and through viscous damping of both the solenoid's iron core C and the 1/8" diameter iron rod positioned beneath the rotor. The vertical stability of the rotor is maintained by the sensing coil P‘which is part of the grid circuit of a partially neutralized tuned grid-tuned plate radiofrequency oscillator in the support circuit shown in Figure 3. Small variations in the height of the rotor change the ”Q" of the grid circuit through the slight change in inductance and capacitance of the pickeup coil. Consider a virtual upward displacement of the rotor. This will increase the amplitude of oscillation in the grid circuit of the oscillator v1 and thus in the plate circuit as the oscillator is partially neutralized. This results in an increase in potential across the cathode follower v3 which causes a higher poten- tial or error signal to be applied to the control grid of v The resulting decrease in potential on the grids of 30 the parallel 5881's, v4, v5 and v6 causes a decrease in cur- rent through the support solenoid which restores the rotor to its original position through a decrease in the lifting 11. FIGURE 2 MAGN ETIC SUSPENSION APPARATUS Lucite .Ol8”STeel Tube Wire 51 a Cooan _ Coils GE 2202862 PhenoHc <—-To Pump Silicone Drive Coils Fluid N p .l25 Iron Pod ork II Scale: ,._'___| 12. ~30 L3 - x 20‘ E «NU. .ItSk k I\ 53036 :00 .UnO \ w I. A A .3 _ kl V: 3.. 2.32.1 \ +4. nor-3.581 Zo_mzmam3m Bamzoflz _ v33 % tom J _ a Moo _ Q . nuo i 4.x cm. _ _l - kit w . Llésppfi \30 hiatus“ . ‘69.? 03... , em 30% — l m mmaol 13. force. Any vertical oscillations (hunting) are eliminated by the differentiating circuit on the output of the cathode follower v2 while the 003 voltage regulators tend to stabi- lize the voltage across the solenoid. The Iagnetic Drive circuit: The rotor is spun by a rotating magnetic field produced by the four drive coils D shown in Figure 4a. These coils are mounted symmetrically about the rotor with Opposite coils connected in parallel. The output of an audio oscil- lator is amplified by a 32 watt power amplifier and passed to a phase splitting circuit (Figure 4b). This circuit con- sists of two capacitors in parallel, each in series with a pair of drive coils, the capacity being determined by the desired driving frequency. The phase splitting is accomp plished by making one of the resonant circuits more capaci- tive so that the currents in the two pairs of coils differ in phase by 90°. The currents in each pair are equalized by slightly detuning the resonant circuits, i.e., the fre- quency is adjusted to a point midway between the resonant frequencies of the two resonant circuits. Thus the solid steel rotor is actually the armature of an induction motor as the rotor is spun by the interaction of the eddy currents set up in the rotor with the rotating field. The "slip" is very high in starting the rotor but gradually decreases as the rotor accelerates. 14. FIGURE 4 MAGNETIC DPlVlNG CIPCUIT vacuum Chamber (a) CA *1: ¢-9o° “mi ‘C\" Wu D AUDlO 32 WATT A»-.. 7 +1 133.7“? OSClLLATOP AMPLIFIER C. (b) 15. The coils each consist of 1000 turns of #26 Formex cop» per wire wound with an inside diameter of 5/4", outside dia- meter of 2", a width of one inch and are of approximately 52 millihenrys inductance. The coils are mounted on a 5/8" thick Phenolic board as the peak to peak voltage across them; is of the order of 1000 volts. The Rotor: The rotor is prepared by grinding two diametrically opposite flats 15 mils deep on the rotor's surface through which will pass the axis of rotation. Two small grooves are placed diametrically apposite on the equator of the rotor, one groove to be used for the source, the other for a coun- terbalance. A l" i.d., 53/52" o.d. x 1/8" wide ring turned from.7075-T5 alumdnum is pressed over the equator of the rotor to hold the source and counterbalance in place in the afore mentioned groves. The ring is of sufficient strength to be spun at a velocity of approximately 4.2 x 104 cm/sec without exploding (See Appendix). Nuclear Counting Equipment: A scintillation counter consisting of a NaI(T1) crystal and a 6“ length of 1.68” diameter lucite light piper mount-‘ ed on a 6298 photomultiplier is used as the gamma detector (Figure 5). The light piper is employed to obtain a large distance between the 6292 photomultiplier and the suspension solenoid as the gain of the photomultiplier is adversely affected by the presence of a strong magnetic field. 16. FIGURE 5 PHOTOMULTIPLIER ARRANGEMENT 93|/\ Dre-Amp 6292 I Scale. ,_|_, 17. The scatterer assembly consists of a .020" thick x 1.5“ diameter iridium disk and a .037" thick x 1.5“ diameter lead disk mounted on a .75" thick 1 1.5" diameter aluminum rod. The thickness of the lead scatterer is chosen by equation (13). The scatterer assembly is mounted on the suspension apparatus so that in rotating the scatterer through 180°, the iridium or lead disk will face the source through the hole in one of the drive coils (See Figure 5). Three inches of lead is placed between the source and the Rs! crystal of the gamma detector to reduce the intensity of the direct 129-kev radiation of 15 day 0a 191 and the SBO-kev radiation of 95 day 0a 185 which are both present in the source. The 0100 129-kev component is reduced by 1 and the BBQ-kev com- ponent is reduced by 103. A gating system is employed so that the gamma detector is turned on only when the source is approaching the scat- terer. In this manner, the signal to noise ratio is greater than that obtained if the source had been distributed about the circumference of the rotor. The gating is accomplished by darkening all of the rotor with black "Flo-Master"‘ ink except for a narrow vertical strip over the radioactive source. The light from an incandescent lamp (Figure 5) is focused on the spinning rotor and the resulting scattered light focused on s 951A photomultiplier. The signal is fed to a scalar where a frequency measurement of the rotor is ‘ Trademark, Cushman and Denison Ifg. 00., New Yark. 18. obtained. The shaped pulses of the frequency detecting scalar are fed to a gate circuit (Figure 6a) along with the output signal of the single channel pulse height analyzer of the gamma detector. The ease inverter tube (Figure 6c) of the gate circuit inverts the gamma signal which is then fed to one control grid of the GREG. The shaped pulse of the frequency detecting scalar is fed directly to the 2nd con- 'trol grid of the CBNG which is biased to conduct only when both signals are present simultaneously, i.e., the square pulse drives the tube near conduction and any gamma signals incident during this time will cause the tube to conduct. The output of the gate is fed to a 2nd scalar which records the number of quanta scattered into the gamma detector while the source is approaching the scatterer. The width of the gate, i.e., the width of the shaped signal of the frequency detecting scalar, is controlled by the width of the reflec- ting line on the rotor. 19. CHER; 3?..ng guises? ,_ d2m zoCumsmo LO §00m: > oomH mmIZQEAulmma 3mm so mmaol 21. zmm > cow CDUQAU whvzm30mmm QOHOWA mm mm VN ON 9 NA w A» Om fl A A a J A A A A A A A A A A A \\e\\ m @\e \ 9 xx 9 x Enos Bugs 5 m: H2300 920 @lG'OlGlQIelal : >Uzm30mmm EOHOQ LO 202.023; 4 m4 IHQA>> wh