Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Isotopic composition of gadolinium, samarium and europium in the Abee meteorite Loveless, Arthur John 1970

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1970_A1 L68.pdf [ 6.16MB ]
[if-you-see-this-DO-NOT-CLICK]
Metadata
JSON: 1.0053418.json
JSON-LD: 1.0053418+ld.json
RDF/XML (Pretty): 1.0053418.xml
RDF/JSON: 1.0053418+rdf.json
Turtle: 1.0053418+rdf-turtle.txt
N-Triples: 1.0053418+rdf-ntriples.txt
Original Record: 1.0053418 +original-record.json
Full Text
1.0053418.txt
Citation
1.0053418.ris

Full Text

ISOTOPIC COMPOSITION OF GADOLINIUM, SAMARIUM AND EUROPIUM IN THE ABEE METEORITE by ARTHUR JOHN LOVELESS B.Sc, The University of. Toronto, 1966 M.Sc, The University of Toronto, 1967 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF s DOCTOR OF PHILOSOPHY i n the Department of GEOPHYSICS We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1970 In p resen t ing t h i s t hes i s in p a r t i a l f u l f i l m e n t of the requirements fo r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e fo r re ference and Study. I f u r t h e r agree that permiss ion fo r ex tens i ve copying of t h i s t hes i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h is r e p r e s e n t a t i v e s . It is understood that copying or p u b l i c a t i o n of t h i s t hes i s fo r f i n a n c i a l gain s h a l l not be a l lowed wi thout my w r i t t e n pe rm i s s i on . Department of The U n i v e r s i t y of B r i t i s h Columbia Vancouver 8, Canada Date A B S T R A C T The w r i t e r has measured the isotopic composition of gadolinium, europium and samarium in the Abee meteorite and in two t e r r e s t r i a l ores. Gd, Eu and Sm have large thermal neutron capture cross sections; they may therefore be used to r e v e a l differences in the i r r a d i a t i o n h i s t o r i e s of samples contain- ing trace amounts of these elements. It is widely believed that the contracting protosun passed through a phase of high energy par t i c l e radiation during the e a r l y h i s t o r y of the solar system. The interaction of these p a r t i c l e s with the m a t e r i a l of the solar nebula may have produced a large thermal neutron flux. The Abee enstatite chondrite is representative of the most highly reduced class of chondritic meteorites. The chondrites are .stones which are thought to be very p r i m i t i v e m a t e r i a l of the solar system. Mason, M i y a s h i r o and others have proposed that variations i n the oxidation state of chondrites re s u l t from their formation in different regions of the nebula: the enstatite chondrites represent m a t e r i a l which was derived from the hot inner region of the solar nebula, while the highly oxidized carbona- ceous chondrites had their o r i g i n i n the cooler, outer region. On the basis of this theory, the earth was probably derived from an intermediate region of the nebula. Isotopic analyses were performed on mi c r o g r a m quantities of Gd, E u and Sm. A pr e c i s i o n of 0.02-0.2% (at the 9 5 % confidence level) was achieved for a l l isotopic ratios of interest. P r i o r to 1970 the best published isotopic analyses of these elements had a p r e c i s i o n of 1-2%. The higher p r e c i s i o n reported here was made possible through improvements in the mass spectrometer ion optics and the use of di g i t a l recording of mass spectra. This work is comparable to recent meteorite analyses on Gd by Eugster, Tera, Burnett and Wasserburg at the C a l i f o r n i a Institute of Technology, although their lunar analyses were of superior quality and had a pr e c i s i o n of 0.01%. No significant Gd, Eu or Sm isotopic anomalies were 157 15 8 observed for the Abee and t e r r e s t r i a l samples. The Gd /Gd rat i o , which is the most sensitive to neutron i r r a d i a t i o n , was identica l for the three samples studied within 0.1%. This places 15 2 an upper l i m i t of 3 x 10 neutrons/cm for the di f f e r e n t i a l i r r a d i a t i o n of the source m a t e r i a l from which the earth and the Abee meteorite were derived. A s i m i l a r conclusion was reported by the Caltech group. The absence of any anomalies may be attributed to uniform i r r a d i a t i o n and dilution of the source m a t e r i a l , efficient shielding inside large planetary bodies during the i r r a d i a - tion phase, the absence of an intense i r r a d i a t i o n phase, or a common spatial o r i g i n within the solar nebula for chondritic and t e r r e s t r i a l matter. T A B L E OF CONTENTS A B S T R A C T T A B L E O F CONTENTS LIST OF FIGURES LIST O F T A B L E S A C K N O W L E D G E M E N T S C H A P T E R 1. INTRODUCTION ., - 1.1 Chondrites - Remnants of P r i m i t i v e Planetary M a t e r i a l 1. 2 Quantization of the Oxidation State of Chondrites 1 . 3 High Energy P a r t i c l e Irradiation and its Consequences 1.4 Delineation of the Present Study C H A P T E R 2. A REVIEW OF THE E V I D E N C E CON- CERNING IRRADIATION ANOMALIES 2.1 Synthesis of the Elements and High Energy P a r t i c l e I r r a d i a t i o n 2 . 2 A Review of the Search for Ir r a d i a t i o n . . -Anomalies 2 . 3 The Effect of Recent Cosmic Irradiation. C H A P T E R 3. INSTRUMENTATION 3. 1 Operational P r i n c i p l e s of a Mass Spectrometer 3. 2 Ion Optics 3 . 3 The Ionization P r o c e s s V 3.4 D i g i t a l Recording of Mass Spectra 42 3.5 Noise Rejection Using a Low-Pass F i l t e r 45 C H A P T E R 4. P R E P A R A T I O N OF S A M P L E S 4 7 4.1 Separation of the Rare E a r t h Elements 4 7 4. 2 Preparation of Filaments 55 4. 3 Production of Ion Spectra with Minimum Interference 5 7 C H A P T E R 5. COMPUTER REDUCTION OF C O M P L E X S P E C T R A 59 5. 1 Evaluation of Peak Heights 59 5. 2.Least Squares Reduction of a Set of Scans 61 5. 3 Calculation of Means and Standard Deviations of Isotopic Ratios for a Set of Scans 68 5.4 Calculation of Fractionation - Corrected Parameters 71 5.5 C o r r e c t i o n for the Isotopic Composition of Oxygen 76 C H A P T E R 6. I N T E R P R E T A T I O N OF M E T E O R I T I C & T E R R E S T R I A L ISOTOPIC RATIOS 78 6.1 T e r r e s t r i a l Gadolinium 78 6. 2 Gadolinium in the Abee Meteorite 84 6. 3 Do Abee and T e r r e s t r i a l Gadolinium have S i m i l a r Composition? ' 90 6.4 Interpretation of Europium Data 93 i6. 5 Interpretation of Samarium Data 95 6.6 Conclusions 99 v i A P P E N D I X I. E F F E C T OF NEUTRON C A P T U R E ON GD, SM AND E U 102 A P P E N D I X II. SECOND ORDER FOCUSSING SHIMS 109 A P P E N D I X III. M E ASURED ISOTOPIC RATIOS FOR A L L GADOLINIUM A N A L Y S E S 119 A P P E N D I X IV. S L O P E OF T H E O R E T I C A L C O R R E L A T I O N LINES FOR NEUTRON C A P T U R E 129 B I B L I O G R A P H Y 131 \ \ LIST OF FIGURES FIGURE 1-1 ABUNDANCES OF THE M E T A L L I C E L E M E N T S FIGURE 1-2 OXIDIZED AND R E DUCED IRON IN CHONDRITES FIGURE 1-3 F R E Q U E N C Y OF O C C U R R E N C E VS OXIDATION S T A T E FIGURE 1-4 T H E R M A L NEUTRON C A P T U R E IN LUNAR GADOLINIUM FIGURE 2-1 M O D E L FOR WHICH 5% OF PRIMITIVE M A T E R I A L WAS IRRADIATED FIGURE 2-2 UNIFORM IRRADIATION M O D E L FIGURE 3-1 CONTINUOUS SCANNING OVER THE GADOLINIUM S P E C T R U M FIGURE 3-2 ION B E A M PRODUCED B Y NEW ION SOURCE FIGURE 3-3 DISCONTINUOUS SCANNING OVER THE GADOLINIUM S P E C T R U M FIGURE 3-4 F R E Q U E N C Y RESPONSE OF DIGITAL F I L T E R F IGURE 4-1 E L U T I O N CURVES FOR I B , GD, E U A N D SM FIGURE 5-1 SLOW DISCONTINUOUS SCANNING FIGURE 5-2 P O L Y N O M I A L R E P R E S E N T A T I O N OF ION B E A M INTENSITY FIGURE 5-3 MASS S P E C T R O M E T E R FR A C T I O N A T I O N FIGURE 6-1 ISOTOPIC A N A L Y S E S ON T E R R E S T R I A L GADOLINIUM FIGURE 6-2 C O R R E L A T I O N B E T W E E N B A N D A F O R T E R R E S T R I A L G D FIGURE 6-3 C O R R E L A T I O N O F GD ISOTOPES IN M E T E O R I T E S FIGURE 6-4 C O R R E L A T I O N B E T W E E N A A N D B IN M E T E O R I T E S FIGURE II-l MODIFICATION O F MAIN E L E C T R O M A G N E T T O A C C O M M O D A T E FOCUSSING SHIMS FIGURE II-2 FRINGING F I E L D O F MAIN M A G N E T FIGURE II-3 C U R V A T U R E O F S E C O N D ORDER FOCUSSING SHIMS F I G U R E II-4 O P T I M U M C U R V A T U R E O F SHIMS AS A F U N C T I O N O F B E A M D I V E R G E N C E F I G U R E II-5 P E A K S H A P E F O R TWO V A L U E S O F T H E A N G U L A R D I V E R G E N C E F I G U R E II -6 B E A M SPREADING A T T H E C O L L E C T O R F O R FIRST A N D S E C O N D O R D E R ;FO CUSSING i x LIST OF T A B L E S Page T A B L E 1-1 T H E R M A L NEUTRON C A P T U R E CROSS SECTIONS 8 T A B L E 2-1 M A X I M U M D I F F E R E N T I A L IRRADIATION OF P R E - T E R R E S T R I A L AND PRE- M E T E O R I T I C M A T E R I A L 28 T A B L E 4-1 DESCRIPTION OF S A M P L E S STUDIED 48 T A B L E 4-2 I N T E R F E R I N G IONS NEAR GD, E U AND SM S P E C T R A . 50 T A B L E 4-3 B E A M INTENSITIES FOR S A M P L E S CONTAINING A L L R A R E EARTHS 51 T A B L E 4-4 E X T R A C T I O N OF GD, E U AND SM F R O M A B E E 53 T A B L E 5-1 F R A C T I O N A T I O N C O R R E C T E D P A R A M E T E R S 75 T A B L E 6-1 ISOTOPIC RATIOS OF GD IN T E R R E S T R I A L S A M P L E S 79 T A B L E 6-2 ADDITIONAL RESULTS FOR T E R R E S T R I A L GADOLINIUM 83 T A B L E 6-3 ISOTOPIC COMPOSITION .OF.GD IN THE A B E E M E T E O R I T E 87 T A B L E 6-4 ISOTOPIC COMPOSITION OF T E R R E S T R I A L AND A B E E E U 94 T A B L E 6-5 ISOTOPIC COMPOSITION OF SMIN T E R R E S T R I A L S A M P L E S ' 96 T A B L E 6-6 ISOTOPIC COMPOSITION OF SAMARIUM IN THE A B E E M E T E O R I T E 97 T A B L E S III -1 to III-8 GD ISOTOPIC RATIOS FOR A L L T E R R E S T R I A L AND A B E E A N A L Y S E S 121 X A C K N O W L E D G E M E N T S The work reported in this thesis received considerable support from several persons in two different institutions. The study of the Abee enstatite chondrite was f i r s t proposed by Dr. H. Mabuchi of the Department of Chemistry, University of Tokyo. He and Mr. S. Yanagita assumed re s p o n s i b i l i t y for separating gadolinium, europium and samarium from a sample of the Abee meteorite. Their major contribution to this study is gratefully acknowledged. The w r i t e r is indebted to Dr. M. Ozima of the Geophysi- c a l Institute, U n i v e r s i t y of Tokyo, for his assistance with the mass spectrometry during the i n i t i a l phase of this research. Dr. Ozima also coordinated the work of the two laboratories and was a v i s i t i n g professor at U. B.C. from May to October, 1969. During this period the w r i t e r benefited from many profitable discussions with h i i r K Dr. R u s s e l l shared much of his experience i n mass spec- trometry with the w r i t e r and provided able supervision through- out this research. The w r i t e r is p a r t i c u l a r l y grateful for Dr. Russell's assistance in solving many of the problems related to mass spectrometer instrumentation and data processing. Without the generous cooperation of other graduate students and faculty members this work would not have been possible. In p a r t i c u l a r , the close cooperation of John Ozard, John Blenkinsop and L i l y Lee was a r e a l asset. The excellent services of the Computing Center at U. B. C. deserve a special word of thanks. Their f a c i l i t i e s were used extensively and were always r e l i a b l e . F i n a n c i a l support came p r i m a r i l y from the National Research Council of Canada in the form of faculty grants and a studentship to the w r i t e r . A 40 gram sample of the Abee meteorite was provided by the Geological Survey of Canada. 1 C H A P T E R 1 INTRODUCTION 1. 1 Chondrites - Remnants of P r i m i t i v e Planetary M a t e r i a l The study of meteorites has led to a vast amount of experimental data without much unanimity of interpretation. Yet, in many respects, the meteorites contain the best r e c o r d available to us for events which occurred during the h i s t o r y of the solar system. Meteorites consist p r i m a r i l y of s i l i c a t e , m e t a l l i c and sulphide phases; the three phases often occur together, in undifferentiated form. There are three main classes of meteor- ites which are based on the proportions of the s i l i c a t e and me t a l l i c phases: stones, irons and stony-irons. The stones are further subdivided into chondrites (meteorites containing round s i l i c a t e grains called chondrules in a m a t r i x of s i l i c a t e s and metals) and achondrites (meteorites lacking chondrules). The achon- drites resemble t e r r e s t r i a l igneous rocks, while the irons are thought to be s i m i l a r to the m a t e r i a l of the earth's core. It is widely believed that the achondrites and.irons are broken f r a g - ments of one (or more than one) parent body having a diameter in excess of 10 km. Since the extent of differentiation into m e t a l l i c and s i l i c a t e fractions has not been f i r m l y established, the stony- irons may represent either m a t e r i a l from the v i c i n i t y of a 2 core-mantle boundary or fragments derived from the contact surfaces between met a l l i c inclusions of meteoritic size and a stony matrix. The remaining class of meteorites, the chondrites, comprises 85% of a l l meteorites, based on the number of recovered meteorites which were observed to f a l l . Detailed studies of their mineralogy and chemistry have led most researchers to believe that the chondrites represent the most p r i m i t i v e major class of meteorites (Anders, 1964; Wood, 1968). It is therefore this class which provides the c r u c i a l test of any theories for the o r i g i n of meteorites. The bulk chemical composition of the chondrites closely resembles the solar chemical composition, neglecting volatile elements. (Figure 1-1). This fact, in addition to the unique chondritic texture, strongly supports the belief that these are very p r i m i t i v e objects. 1. 2 Quantization of the Oxidation State of Chondrites Although the bulk chemical composition of chondrites is s u r p r i s i n g l y uniform, there exist marked differences in their oxidation states. In fact, the oxidation states based on elemental and oxidized i r o n abundances appear to be quantized. Mason (1962) has plotted weight per cent i r o n (as metal or as FeS) against weight per cent oxidized i r o n (as s i l i c a t e s ) for s e v e r a l precise analyses of chondrites (Figure 1-2). Six distinct groups are 3 FIGURE 1-1. ABUNDANCES OF THE METALLIC ELEMENTS Atmosphere of the sun 10~ 2 I O - 1 1 10 1 0 2 1 0 3 10^ I O 5 10 Ordin a r y c h o n d r i t e s E l e m e n t a l abundances i n the atmosphere of the sun ( r e l a t i v e t o 10" S i atoms) are compared wi t h those i n o r d i n a r y (hypersthene and b r o n z i t e ) c h o n d r i t e s . Elements with low b o i l i n g p o i n t s have been excluded. Note the high abundance of L i i n c h o n d r i t e s when compared with the sun. The diagram was reproduced from M e t e o r i t e s and the O r i g i n o f P l a n e t s by John A. Wood, 196W. 4 evident, ranging from f u l l y reduced enstatite chondrites to f u l l y oxidized carbonaceous chondrites. Mason (1963) proposed that differences in oxidation . state arose in the planetary nebula. The enstatite chondrites represent highly reduced m a t e r i a l from the hot inner regions, while the carbonaceous chondrites represent more p r i m i t i v e , highly oxidized m a t e r i a l which accreted in the cold outer regions of the nebula. The quantization of the oxidation states may re s u l t from the accretion of m a t e r i a l in the nebula to form planetary bodies, each of which possessed p h y s i c a l and chemical properties which were representative of the region from which it was derived. The chondritic meteorites may then represent a biased sample of c o l l i s i o n fragments from only four or five planetary bodies. The reduction of iron-magnesium.; s i l i c a t e s proceeds by a decrease i n the FeO content of the s i l i c a t e s and a correspond- ing increase i n the amount of Fe in the m e t a l l i c phase (Anders, 1964; Ringwood, 1966; M i y a s h i r o , 1968). The MgO i n the s i l i c a t e s remains essent i a l l y unaltered un t i l extreme reduction has occurred. Therefore, the mole fr a c t i o n f—FeO/(MgO+FeO) in s i l i c a t e s is a useful index of the degree of oxidation. By plotting frequency of occurrence against f for chondrites, M i y a s h i r o (1968) drew attention to the large hiatus between enstatite chond- ri t e s and the other groups (Figure 1-3). He offered two possible 5 FIGURE 1-2. OXIDIZED AND REDUCED IRON IN CHONDRITES Weight 35 i E - Enstatite chondrites per cent 30 « « 1 • E B - Bronzite chondrites i r o n i n | H - Hypersthene chondrites metal 25 .i 4 I " X w V A - Amphoteric chondrites and FeS 20 15 10 V * • • • H P - C - Pigeonite chondrites Carbonaceous chondrites 5 X A a X °p P B - ° C 0 • • i c ) 5 10 15 20 25 30 Weight per cent oxidized i r o n Relationship between reduced i r o n (metal + FeS) and oxidized i r o n (FeO) i n chondrites. The diagram was adapted from Mason (1962) and Anders (196^). FIGURE 1-3. FREQUENCY OF OCCURRENCE VS OXIDATION STATE 4 B H FeO FeO + MgO Oxidation state of s i l i c a t e s The frequency of occurrence i s plotted as a function of the f-value (oxidation state) f o r chondrites (after Miyashiro, 1967). See Figure 1-2 for i d e n t i f i c a t i o n of the chondrite classes. 6 explanations for this gap: (a) Since carbon and hydrogen are the pr i n c i p l e reducing agents, the large gap between the enstatite chondrites and the ordinary chondrites represents a marked difference in the reducing power of these two elements. Carbon was the p r i n c i p l e reducing agent at higher temperatures close to the protosun while hydrogen reduction dominated in the cooler, outer regions. (b) An alternative explanation is that the gap i n the region f =0.01-0.14 corresponds to the oxidation state of the p r i m i t i v e m a t e r i a l from which the earth accreted. B i r c h (1964) reported an FeO content for the mantle of 10-2 weight per cent. This im p l i e s that the amounts of oxidized and reduced i r o n for the earth as a whole (assuming the core contains 9 0 % m e t a l l i c iron) are approximately 4% and 23% respectively. On this b a s i s , the oxidation state of the earth is intermediate between groups E and B i n Figure 1-2. The f-value of the earth is less certain because of the unknown MgO content of the mantle. If either of Miyashiro's explanations is v a l i d , then v a r i a - tions in the oxidation state of chondrites result from differences in the .amount of solar i r r a d i a t i o n incident on the p r i m i t i v e m a t e r i a l from which they were formed. It is generally accepted that the con- tracting protosun passed through a phase when its luminosity was 7 s e v e r a l hundred time.s its present-day value (Hayashi, 1966; Hoyle et a l , 1968). The temperature distribution in .the solar nebula has not been established, however. On the basis of Miyashiro's hypothesis, one would expect the enstatite chondrites to show other evidence of their close proximity to the sun in their e a r l y history. Such evidence may be revealed through a comparison of the i r r a d i a t i o n h i s t o r i e s of meteor- i t i c and t e r r e s t r i a l m a t e r i a l . The earth provides a convenient standard of reference against which the o r b i t a l positions of other planetary bodies might be compared. Several theories for the formation of the solar system postulate the existence of a period of high energy (^v 500 MeV) par t i c l e i r r a d i a t i o n , of solar o r i g i n , during the collapse of the protosun and the transfer of angular momentum to the surrounding nebula v i a a magnetic coupling. The high energy p a r t i c l e flux would resemble the present-day p r i m a r y cosmic ray flux, although its intensity was probably s e v e r a l orders of magnitude greater. When these p a r t i c l e s (mainly protons and other light nuclei stripped of their electrons) bombard other nuclei they chip, or spall, nucleons from them, generating a large number of secondary p a r t i c l e s . In p a r t i c u l a r , the number of neutrons produced by these spallation reactions would be comparable to the number of incident p r i m a r y p a r t i c l e s (Fowler et a l , 19 62). Both the p r i m a r y 8 spallation reactions and secondary neutron capture processes are capable of altering the isotopic composition of theelements ; i n the i r r a d i a t e d m a t erial. 1.3 High Energy P a r t i c l e I r r a d i a t i o n and its Consequences Gadolinium, samarium and europium are known enormously high thermal neutron capture cross sections m a y b e attributed to a few specific isotopes (Table 1-1). Isotope Cross Section 242,000 t 4000 barns _:58,000 ± 3000 40,800 t 900 7, 800 t 200 450.± 20 T A B L E 1-1 The r m a l (20° C) neutron capture cross sections for (n,V ) reactions (reported i n Cook et a l , 1968). these may serve as highly sensitive neutron flux indicators. An examination of the isotopic compositions of Gd, Sm and Eu in meteorites and in t e r r e s t r i a l samples would r e v e a l any significant differences in their i r r a d i a t i o n h i s t o r i e s . In p a r t i c u l a r , a study to have which Hence G d " 7 G d 1 " Sm 1 4? E u 1 5 1 E u 1 5 3 9 of these elements in enstatite chondrites might provide a c r i t i c a l test of the hypothesis of Miyashiro. P r i o r to 1969, no n a t u r a l l y - o c c u r r i n g isotopic anomalies had been detected for these elements, either in t e r r e s t r i a l or in meteoritic samples. No enstatite chondrites had been investigated, however. Recent advances in instrumentation have made it possible to measure isotopic ratios for m i c r o g r a m samples of Gd, Sm and Eu within 0.1% standard deviation. Using very precise tech- niques Eugster et a l (1970a, 1970b) were able to detect a deficiency 1 ^ 7 1 of Gd i n lunar rock and s o i l samples, as w e l l as in the Norton County achondrite. The deficiency of Gdl57 relative to t e r r e s t r i a l Gdl57, and the corresponding enhancement in G d 1 5 8 (Figure 1-4) may c l e a r l y be attributed to the Gdl57 (n,X )Gdl58 reaction. The neutrons required for this process were produced by spallation reactions o c c u r r i n g i n the surface lay e r s of the moon and the parent meteorite fragment, due to bombardment by cosmic rays. Cosmic rays in the 500 MeV range are capable of pene- trating approximately lOOg/crri^ of matter. This is the mean free path for the p r i m a r y p a r t i c l e s which induce such spallation reactions as F e 5 6 + H ^ C l 3 6 + H 3+ ZHS + He 3 + 3rf + 4 neutrons T Both Eugster et a l (1970b) and Lugmair (1970) reported e r r o r s of 0.01% (two standard deviations of the mean) for lunar Gd analyses. 10 0.712 0.713 0.714 0.715 0.716 * « 1 ' 1 J 1 ! 1 1 I I I . . . . » 15 10 ' 5 o Integrated thermal neutron f l u x (10 D neutrons/cm ) FIGURE 1-4. THERMAL NEUTRON CAPTURE IN LUNAR GADOLINIUM G d 1 5 8 / G d 1 6 0 vs G d ^ / G d 1 6 0 (normalized to G d ^ / G d 1 6 0 = .9361) f o r several lunar samples from Apollo 11. Lunar samples include s i x rocks, one "breccia, two core samples and s o i l . The Gd isotopic r a t i o s are c l e a r l y s h i f t e d ( r e l a t i v e to t e r r e s t r i a l Gd) due to neutron capture. Neutrons are continuously produced during bombardment of the lunar surface by cosmic rays. The co r r e l a t i o n diagram and a l l data were published by Eugster et a l (1970b). 11 High energy neutrons are a product of these reactions. In the presence of light nuclei such as hydrogen, the neutrons are readily thermalized. ^ The thermalization process involves slowing down high energy pa r t i c l e s by multiple c o l l i s i o n s with other n u c l e i , until energies in the thermal, range (^20 C) have been reached. Since reaction cross sections are much greater for slow neutrons, the majority of the neutrons w i l l be captured only after reaching thermal v e l o c i t i e s . In low density media, a significant fraction of the neutrons may undergo radioactive decay into protons and electrons, since the h a l f - l i f e of a neutron i s only 11 minutes. Neutron decay w i l l be dominant only in dispersed media having mean densities ~ 10-7 g / c m 3 (Fowler et a l , 1962). The neutron t h e r m a l - ization and capture processes occur over a range of approximately 30 g/cm^. It is therefore evident that the secondary neutrons stray only a short distance from the p r i m a r y cosmic ray path and are confined to the volume i r r a d i a t e d by the high energy p a r t i c l e s . Although the penetration depth of cosmic rays v a r i e s with the energy of the p r i m a r y p a r t i c l e s and the composition of the i r r a d i a t e d m a t e r i a l , a range of 1 to 4 meters is probable for *>r 500 MeV par t i c l e s penetrating materials comparable to ordinary rock with a density of 2 to 4 g/cm^. The surface of the earth i t s e l f is l a r g e l y shielded from cosmic rays by the 10^ g/cm^ of a i r i n its atmosphere. 12 Outside the earth's atmosphere the present-day flux of p r i m a r y cosmic rays having energies 200 MeV is 1 to 10 p a r t i c l e s / 2 ' r cm sec. The bulk of this flux is of galactic o r i g i n and appears to have been of uniform intensity throughout the previous histo r y of our solar system. Only a few per cent of the flux i s derived from our sun. This fraction is of variable intensity, with maxima oc c u r r i n g during periods of intense solar f l a r e activity. Although the present-day high energy pa r t i c l e flux from the sun is s m a l l relative to the galactic flux, there is considerable evidence to suggest that this has not always been the case. Several theories for the o r i g i n of the solar system propose the existence of an intense high energy p a r t i c l e flux during its e a r l y h i s t o r y (Burbidge et a l , 1957; Fowler et a l , 1962; Cameron, 1965; Hayashi, 1966). Although the i r r a d i a t i o n times are short (10 to 10 years) the proposed values of the integrated high energy pa r t i c l e flux range from 1 0 ^ to 1 0 ^ protons/cm^. This may be 17 d i r e c t l y compared with the integrated cosmic ray flux of 10 to 10^ 8 protons/cm^ since the formation of the earth and meteorite parent bodies approximately 4. 5 b i l l i o n years ago. If an intense high energy proton flux was emitted from the protosun during the period p r i o r to the formation of the major planetary bodies, then one would expect varying degrees of i r r a d - iation of the m a t e r i a l of the nebula depending upon its distance from the protosun. The high energy p a r t i c l e s would induce spallation 13 reactions in the m a t e r i a l of the nebula which, in turn, would gen- erate neutrons. Where l o c a l condensation of the m a t e r i a l had c occurred, a large fraction of the neutrons would be thermalized and captured; the remainder would decay. The majority of the captured neutrons would react with H.O,Fe,Ni,S and Si nuclei whichhave high absolute abundances, although their cross sections are s m a l l (< 5 barns). These elements are not i d e a l l y suited for subsequent detection of the i r r a d i a t i o n effects since the resulting changes in isotopic composition are s m a l l (except for H). The elements which show the largest memory effect after i r r a d i a t i o n are those having isotopes with very high thermal neutron capture cross sections (e. g. Gd, Sm, Eu) or those having isotopes of low i n i t i a l abundance (e. g. K, V). The isotopic abundances of s e v e r a l of the lighter elements (H,He,Li, B,C) are also very sensitive to particle i r r a d i a t i o n , although the nuclear reactions i n which they participate are much more dif f i c u l t to resolve. 1.4 Delineation of the Present Study The investigation of Gd, Sm and E u in enstatite chondrites was f i r s t proposed to the w r i t e r by Dr. M. Ozima of the Geophysical Institute, U n i v e r s i t y of Tokyo. Dr. Ozima was v i s i t i n g professor in the Department of Geophysics at the University of B r i t i s h Columbia during the period May to October, 1969. He and Dr. Mabuchi 14 of the Department of Chemistry, U n i v e r s i t y of Tokyo, initiated r e s e a r c h into this problem. The present w r i t e r was invited to assume r e s p o n s i b i l i t y for the isotopic analyses of Gd, Sm and Eu because of his experience i n mass spectrometry. The chemistry of the r a r e earth element separations was undertaken by Mr. Yanagita under the direc t i o n of Dr. Mabuchi. This r e s e a r c h is therefore a joint project between the U n i v e r s i t y of Tokyo and our own laboratory. The previous r e s e a r c h of the w r i t e r was i n the f i e l d of ion optics at the Uni v e r s i t y of Toronto. A computer program was written to calculate ion t r a j e c t o r i e s through two-dimensional elect r o s t a t i c lenses. This method was sufficiently general to be applied to any r e a l i s t i c lens configuration encountered in mass spectrometer ion sources operating at low beam intensities. Ion beams were computed for three conventional ion sources which were shown to have low t r a n s m i s s i o n efficiencies (Loveless, 1967). The f i r s t year of re s e a r c h at the Un i v e r s i t y of B r i t i s h Columbia was devoted to the design of a more efficient ion source. This led to a clear definition of the fundamental l i m i t a t i o n to the degree of focussing which may be achieved in an ion source. In p a r t i c u l a r , it was possible to estimate the maximum t r a n s m i s s i o n efficiency of an ide a l ion source i n terms of basic parameters of the mass spectrometer and the ionization process (Loveless and R u s s e l l , 1969). This had not been reported before in the mass spectrometer l i t e r a t u r e . 15 A simple ion source having approximately one half the optimum t r a n s m i s s i o n efficiency was then designed and built. This gave a factor of five increase in sensitivity (compared with the ion source previously in use) when tested on the available mass spectrometer (with'.a 12.in radius .90 de_g. sector, single stage mass analyzer). A further factor of two gain in sensi t i v i t y was achieved by attaching second order focussing shims to the analyzer magnet, so that a beam of higher angular divergence could be transmitted through the analyzer without loss of resolution. These fundamental improvements in instrument s e n s i t i v i t y , combined with the implementation of di g i t a l recording of mass spectra, were significant contributions by the w r i t e r in the f i e l d of mass spectro- meter instrumentation. These modifications substantially increased the p r e c i s i o n of isotopic analyses on the rare earth elements. A 40 gram sample of the Abee enstatite chondrite was obtained from Dr. Douglas of the Geological Survey of Canada. This sample was part of specimen 13b taken from an equatorial s l i c e through the meteorite. It was located 9-10 cm from the nearest fusion crust surface. (The meteorite was observed to f a l l on June 9, 19 52 and was found five days later in a wheat f i e l d , near Abee, just north of Edmonton, A l b e r t a ( M i l l m a n , 1953). It had a recovery mass of 107 kg and a specific gravity of 3. 5. The major constituents of the meteorite are enstatite (MgSiC^) - 1 6 48%, kamacite-taenite ( n i c k e l - i r o n alloys) - 22%, and t r o i l i t e (FeS) - 13%. Dawson et al(19 6 0 ) report further details of the mineralogy, chemistry and petrology of the Abee meteorite. ) It is evident that several factors combined favourably to make this research p r a c t i c a l . Although our laboratory has had considerable experience i n the isotopic analysis of elements in the higher mass range (Rb, Sr, Pb, U), the rare earth elements had never been investigated previously at this institution. This challenge, in addition to the significance of the as t r o p h y s i c a l problem i t s e l f , provided the interest and motivation for the writer's contributions to this study. The specific contributions of the w r i t e r to the combined r e s e a r c h project are: (a) Improvements in instrumentation. An order of magnitude increase in sensit i v i t y was achievedbn the available mass spectrometer through improvements to the ion optics. The instrument was also interfaced with a magnetic tape recorder for d i g i t a l recording of mass spectra. (b) The development of procedures.for analyzing the isotopic composition of 1 microgram quantities of Gd, Sm and Eu within a standard deviation of 0. 1% or less. This includes experimental and computational techniques for removing the effects of trace quantities of inte r f e r i n g spectra from neigh- bouring elements, due to incomplete chemical separation. 17 (c) The precise isotopic analysis of Gd, Sm and E u in t e r r e s t r i a l ores derived from two different geographic locations (USA and B r a z i l ) and in the Abee enstatite chondrite. The analysis of the above samples would re v e a l any isotopic anomalies as s m a l l as 0. 1 - 0. 3% due to variations in previous exposure to thermal neutrons. The two t e r r e s t r i a l ores were studied to establish the consistency of the t e r r e s t r i a l r a t i o s , and to provide a r e l i a b l e standard against which the meteoritic ratios could be compared. The p r i m a r y objective of this research is to search for an i r r a d i a t i o n anomaly in the Abee enstatite chondrite. There is evidence which suggests that the protosun was a strong emitter of high energy pa r t i c l e s during the e a r l y history of the solar system, p r i o r to the formation of the planets and parent meteorite bodies. The existence of an i r r a d i a t i o n anomaly for Abee would give strong support to the hypotheses of Mason and Miyashiro: that variations in the oxidation state of chondritic meteorites result from distinct differences i n the mean distance from the sun of the p r i m i t i v e . m a t e r i a l from which the chondrites were derived. 18 C H A P T E R 2 A REVIEW OF THE E V I D E N C E CONCERNING IRRADIATION A N O M A L I E S 2. 1 Synthesis of the Elements and High Energy P a r t i c l e Irradiation Burbidge, Burbidge, Fowler and Hoyle (1957) successfully explained the synthesis of most elements on the basis of chains of thermal nuclear reactions ocurring in s t e l l a r i n t e r i o r s . Their theory is based on the assumption that only hydrogen is p r i m e v a l . Successive stages of hydrogen burning, helium burning, and more complex nuclear processes are capable of synthesizing most elements in the i n t e r i o r of stars. However, the high temperature and pressure which play the essential role i n their theory would destroy deuterium (D), lithium ( L i ) , b e r y l l i u m (Be) and boron (B) rather than synthesizing them. Even at moderately low temperatures these elements would be rapidly converted to helium by proton bombardment. I n order to explain the existence of these elements in meteorites and the earth (note the L i abundance i n Fi g u r e 1-1) Fowler, Greenstein and Hoyle (FGH, 1962) proposed that the m a t e r i a l of the solar nebula was subjected to high energy p a r t i c l e i r r a d i a t i o n from the protosun p r i o r to the formation of large planetary bodies. The most important interactions would be those of high energy (*v 500 Mev) protons and alpha-particles, with abundant nuclei such as 0^, M g ^ , ~28 56 S i and Fe . The elements D, L i , Be and B would be products 1 9 of such spallation reactions. The isotopic composition of some of these elements should differ from the t e r r e s t r i a l values, however, if they were produced by spallation alone. In p a r t i c u l a r , the t e r r e s - • t r i a l values of L i / L i ^ and B^/B * are 0.08 and 0.23 respectively, whereas the predicted spallation yields should give ratios close to unity. ~i "Fowler et a l (1962) drew attention to the large thermal neutron cross sections for the reactions L i ^ ( n , 0< and B^(n,C<)Li^. They showed quantitatively that an integrated thermal neutron flux of 4 x lO^ 1 neutrons/cm^ would be capable of producing the observed isotopic ratios for L i and B. The assumption that thermal neutrons existed is e n t i r e l y l o g i c a l , since neutrons are direct products of spallation reactions, and th e r m a l i z a t i o n of neutrons w i l l occur p r i o r to capture as long as there is a moderate excess of hydrogen in the i r r a d i a t e d m a t e r i a l . ( F o w l e r et al(1962) suggested a composition for the i r r a d i a t e d m a t e r i a l of I-^Oand the oxides of Mg, Si and Fe i n the ratio two to one by volume.) The t e r r e s t r i a l D^/H ra t i o of 1.5 x 10 4 was also explained by the neutron capture process H^(n, e" )D^, by imposing the additional condition that only 10% of t e r r e s t r i a l matter was exposed to the i r r a d i a t i o n . Burnett, Fowler and Hoyle (1965) late r r e v i s e d this estimate to 5%. The f r a c t i o n of t e r r e s t r i a l m a t e r i a l which was i r r a d i a t e d was not relevant to the discussion of L i and B, since these elements were 20 assumed to have negligible abundances p r i o r to i r r a d i a t i o n . Their production was attributed e n t i r e l y to i r r a d i a t i o n processes, whereas H was very abundant before i r r a d i a t i o n . In order to shield most of the t e r r e s t r i a l matter from the i r r a d i a t i o n , Fowler et a l (1962) proposed that the m a t e r i a l of the nebula had condensed into s o l i d planetesimals of dimensions from 1 to 50 metres. The high-energy p a r t i c l e s could then penetrate the near-surface m a t e r i a l to a depth of about one metre. The tempera- ture of the planetesimals was estimated to be in the range 130-200° K. The F G H theory for the synthesis of D, L i , Be, and B is supported by the observed high L i abundance in young (T Tauri) stars. On the other hand, and contrary to prediction, very few variations i n isotopic composition have yet been observed, between meteorites and the earth, for those nuclides which have very high neutron capture cross sections, low abundance ratios,.- or high spallation yields. A review of the search for such anomalies w i l l be given i n Section 2. 2. F o r the moment, however, we., suggest that one of the following alternatives must be true: (a) The F G H process was operative but there are no significant i r r a d i a t i o n anomalies between t e r r e s t r i a l and meteoritic samples because of uniform i r r a d i a t i o n and dilution (Burnett et a l , 1965). (b) The F G H process was not operative; D, L i , Be and 2 1 B were not formed p r i m a r i l y by high energy pa r t i c l e i r r a d i a t i o n during the e a r l y h i s t o r y of the solar system, or (c) The search for anomalies has not been sufficiently • thorough. A n alternative to the F G H theory for the synthesis of D, L i , Be and B has been given by Cameron (1962, 1965). He suggests that these elements are products of galactic nucleosynthesis and were already present in the i n t e r s t e l l a r medium from which the solar system was formed. Although Cameron dismi s s e s the p o s s i b i l i t y of intense i r r a d i a t i o n of the protoplanetary m a t e r i a l as required by the F G H theory, he does postulate the existence of a moderate flux for the production of enough radioactive A l ^ t Q rnelt bodies of as t e r o i d a l size (Cameron, 1965). Evidence of ' f o s s i l ' M g ^ , f r o m the .72 m. y. decay of A l ^ , has recently been reported for s e v e r a l meteorites by Clarke et a l (1970). 2. 2 A Review of the Search for Irr a d i a t i o n Anomalies The search for i r r a d i a t i o n anomalies has been r e s t r i c t e d to t e r r e s t r i a l and meteoritic samples, although lunar rocks have become available within the past year. The only well-established anomalies which have been attributed to spallation reactions during (or p r i o r to) the formation of the solar system were observed in the noble gases xenon and krypton (e.g. Reynolds, 1963; Merihue, 1963). Although these anomalies are large in a relative sense (enrichments • -• 2 2 of up to 600%), they are s m a l l in an absolute sense (10 ̂  to 10 ^ ppb of the meteorite mass). Murthy (1960,1962) reported anomalous isotopic compositions for s i l v e r (2 to 4%) and molybdenum (up to 7%) in certain i r o n meteor- ites, and an enrichment (< 2%) in the light isotopes of barium was reported by Umemoto (1962). These apparent anomalies are a l l large in an absolute sense (0. 1 to 170 ppb) but s m a l l in a relative sense. Recent attempts to v e r i f y these anomalies have l a r g e l y discredited the e a r l i e r observations on the basis of mass spectro- meter fractionation (Eugster et a l , 1969; Chakraburtty et a l , 19 64). Anomalies among the lighter elements have also been reported (references are given by Anders, 1964). Their interpretation is ambiguous, however, because of their susceptibility .to chemical fractionation, addition of m a t e r i a l from the solar wind, or bulk transfer in volatile gases. Among the non-volatile elements, the isotopic ratios L i ? / L i 6 , K 4 0 / K 4 1 , V 5 0 / V 5 1 , G d 1 5 7 / G d 1 5 8 , G d ^ / G d ^ , S m ^ / S m 1 ^ and E u ^ l / E u ^ ^ a r e p a r t i c u l a r l y sensitive to i r r a d i a t i o n . The isotopes of L i , K and V are direct products of spallation reactions, although their abundances are significantly altered by thermal neutron capture as well. F o r the rare earth elements (Gd, Sm and Eu) the isotopic abundances are much more strongly influenced by thermal neutrons than by direct spallation. Since these elements are of p r i m a r y 23 interest i n the present research, the results of investigations on L i , K and V w i l l be interpreted in terms of the thermal neutron flux, rather than the p r i m a r y high energy pa r t i c l e flux. This interpreta- t i o n is based on the assumption of approximately a one to one c o r r e s - pondence between the thermal neutron flux and the p r i m a r y flux > 200 Mev (after Fowler et a l , 1962). F i g u r e 2-1 shows the relative se n s i t i v i t y of the Gd, Sm and E u isotopic ratios to d i f f e r e n t i a l i r r a d i a t i o n of meteoritic and t e r r e s t r i a l m a t e r i a l . In this diagram 1^n^[ and ~)f/nj? represent the integrated thermal neutron flux i r r a d i a t i n g meteoritic and t e r r e s - t r i a l m a t e r i a l respectively. The curves il l u s t r a t e the range of values of J ( ^ n M - 7//nE)/Y/nE|and"YnE f o r w h i c h a anomaly would be observed in the relevant isotopic ratio. The area above each of these curves represents an anomaly >0.1%. (The procedure for computing the curves is outlined in Appendix I. ) Figure 2-2 i s a s i m i l a r diagram except that only the most sensitive r a t i o , ^ 157 . 158 , Gd /Gd , has been i l l u s t r a t e d . F i g u r e 2-1 i l l u s t r a t e s a model for which only 5% of the p r i m i t i v e m a t e r i a l was i r r a d i a t e d , while the remainder was shielded in the i n t e r i o r of planetesimals s e v e r a l meters in radius. This model is consistent with the F G H hypothesis when""^ ̂ , ; 4 x 10^ 2 neutrons/cm . It does not assume that a l l planete simals had the same dimensions, but rather that the distribution of planetesimal 24 FIGURE 2-1. MODEL FOR WHICH 5 % OF PRIMITIVE MATERIAL D i f f e r e n t i a l i r r a d i a t i o n \ WAS IRRADIATED n nM ^nE nE 0.1 0.01 0.001 WMmmm l l l l l l l l l l l l l •:•:->:• Region of V ^ i n f l u e n c e of \ SSrecent cosmic\ Sray exposure \ (for Abee) ^ » Region excluded * because of the \absence of a K \ isotopic y anomaly (for Abee) Flux predicted FGH hypothesis •IS mm i 10 16 10 18 10 20 10 22 nE Time-integrated neutron flux which i r r a d i a t e d primor- d i a l material from which the earth was formed (n/cm^). Curves show the s e n s i t i v i t i e s of the Gd, Eu and Sm isotopic r a t i o s to a difference i n the thermal neutron fluxes which i r r a d i a t e d primitive meteoritic (M) and t e r r e s t r i a l (E) matter p r i o r to the formation of large planetary bodies. The area above each curve represents >0,1% isot o p i c anomaly. 25 FIGURE 2-2. UNIFORM IRRADIATION MODEL D i f f e r e n t i a l i r r a d i a t i o n -if nE nE O.Ol F e a s i b l e jv-:-x-Upper l i m i t imposed by- anomaly f o r £:V:;x the absence of a K 0.1 ~ 0.001 Time-integrated neutron f l u x which i r r a d i a t e d primor- d i a l m a t e r i a l from which the earth was formed (n/cm 2) The r e g i o n above the s o l i d l i n e represents a Gd^-^/Gd^^ anomaly ^0.1% due to d i f f e r e n t i a l i r r a d i a t i o n of p r i m i t i v e m e t e o r i t i c (M) and t e r r e s t r i a l (E) matter. The other r a t i o s are l e s s s e n s i t i v e to neutron capture (see Figure 2-1). 26 sizes and shapes was the same for both the pre-meteoritic and p r e - t e r r e s t r i a l m a t e r i a l . Presumably, the planetesimals later accreted to form l a r g e r planetary bodies in which mixing of the ir r a d i a t e d and non-irradiated fractions could occur. The second model (Figure 2-2) assumes uniform i r r a d - iation of p r i m i t i v e material. This corresponds to planetesimals having a maximum radius of approximately one metre. F o r this model an upper l i m i t for ")^nj£ of 2 x 10^ neutrons / cm ^ can be set on the basis of the present-day G d " 7 / G d " ^ ratio, A higher flux I C Q would imply a negative Gd abundance i n i t i a l l y . F i g u r e s 2-1 and 2-2 are representative of several possible models; another important p o s s i b i l i t y is that the fraction i r r a d i a t e d was different for p r i m o r d i a l meteoritic and t e r r e s t r i a l m a t e r i a l . estimated that a given percentage va r i a t i o n in the integrated neutron flux would produce approximately the same percentage va r i a t i o n in 7 6 L i / L i . The fact that Krankowsky et a l (1964, 1967) observed no variations in this ratio in several meteorites (including the Abee enstatite chondrite) within an experimental uncertainty of 2% was therefore interpreted as placing an upper l i m i t of 2% on the value of the t e r r e s t r i a l L i was produced by spallation within the solar On the basis of the F G H hypothes i s , Burnett et al (1965) of This i s only true, however, if most system, neutrons/cm . 27 The l i m i t s on the d i f f e r e n t i a l i r r a d i a t i o n h i s t o r y based on a 5% i r r a d i a t i o n model (Figure 2-1) and a uniform i r r a d i a t i o n model (Figure 2-2) are given in Table 2-1. An investigation of t h e . V ^ / V ^ ratio i n s e v e r a l meteor- it e s , including the Abee enstatite chondrite, revealed no anomalies within an experimental uncertainty of 2% (Balsiger et a l , 1969). Burnett et a l (1966) searched for an anomaly i n the K 4 ^ / K 4 1 r a t i o i n many classes of meteorites, again including Abee, They observed no anomalies, within an experimental uncertainty of 0.5%,which could be attributed to anomalous i r r a d i a t i o n during the e a r l y h i s t o r y of the solar system. Small anomalies in the Norton County meteorite (a highly reduced achondrite), an iro n meteorite, and a stony-iron, were attributed to recent cosmic ray exposure p r i o r to capture by the earth. Estimates of the upper l i m i t to the d i f f e r e n t i a l p a r t i c l e flux associated with a 0.5% anomaly for K 4 0 / K 4 1 are given in Table 2-1 (after Burnett et a l , 1969). The absence of an i r r a d i a t i o n anomaly for K i n the Abee enstatite chondrite places an upper l i m i t on the d i f f e r e n t i a l i r r a d i a t i o n J t ^ n ^ " "Xf" ̂  | of 1 x 10 ̂  neutrons/cm^ for the 18 7 5% i r r a d i a t i o n model and * 5 x 10 neutrons/cm for the uniform i r r a d i a t i o n model. These are order of magnitude estimates only. The shaded regions on the right side of Figures 2-1 and 2-2 indicate unreasonable values of ^ n y i and " V ^ E based on the above l i m i t s for 28 TABLE 2-1. MAXIMUM DIFFERENTIAL IRRADIATION OF PRE- TERRESTRIAL AND PRE-METEORITIC MATERIAL U n c e r t a i n t y i n p u b l i s h e d Uniform experimental Energy range 5fo i r r a d , i r r a d . I s o t o p i c abundance of p a r t i c l e model model r a t i o measurements f l u x (part/cm 2) (part/cm ) L i 7 / I ' i 6 a 2% >?5 Mev 2 x 10 1 9 1 x 10 1 8 K^ U/K^ 1 *b o.5-l% £ 500 Mev 6 x 10 1 9 3 x 10 1 8 Thermal neut. 1 x 10 2 0 5 x 10 1 8 2-20 Mev neut. 6 x 10 1 9 3 x 10 1 8 v50/v51 *cd 2fo > 50 Mev 8 x 10 1 8 4 x 10 1 7 G d l 5 7 / G d 1 5 8 ^ e Vfo Thermal neut. 4 x 10 l 6 Gd157/Gdl58 #f Otlf0 Thermal neut. 3x 1015 * I n c l u d i n g Abee e n s t a t i t e c h o n d r i t e . - # E x c l u d i n g a l l e n s t a t i t e c h o n d r i t e s , a Krankowsky e t a l (1964). b Burnett e t a l (1966). c B a l s i g e r e t a l (1969). e Murthy e t a l (1963). f E u g s t e r e t a l (1970a). 29 I "VnM " "l/ nE I a n C^' * n t^ i e c a s e °^ t n e uniform i r r a d i a t i o n model, on the maximum value of "ty'-j- consistent with the present-day „ .157 ,„ ,158 Gd /Gd ratio. Murthy and Schmitt (1963) investigated Gd, Sm andEu isotopic ratios for a recent Hawaiin basalt, three ordinary chondrites (i . e . hypersthene and bronzite chondrites), and one carbonaceous chondrite. They found no significant v a r i a t i o n in the isotopic ratios within an experimental uncertainty of 1%. Although their observations suggest the absence of any large anomalies among the chondrites, they do not rule out the p o s s i b i l i t y of finding a significant anomaly for the Abee enstatite chondrite, since this class of chondrites was not investigated. More recently, Eugster et a l (1970) have measured the isotopic composition of Gd in s e v e r a l meteorites with a p r e c i s i o n of 0.1%. T h e i r results show agreement with t e r r e s t r i a l Gd within the experimental uncertainty, except for the Norton County achondrite 157, 158 . which showed a decrease in the Gd /Gd r a t i o of (0.27 i .04)% . They attributed this anomaly to the very large (230 m. y. ) cosmic ray exposure age (Begemann et a l , 1957) of the Norton County meteorite. Again, it is of significance to the present study that no enstatite chondrites were included i n their investigations. 2. 3 The Effect of Recent Cosmic Ir r a d i a t i o n Any interpretation of i r r a d i a t i o n anomalies in meteorites 30 requires a knowledge of the extent of recent exposure to cosmic rays, p r i o r to capture by the earth. Several methods have been developed for determining the cosmic ray exposure age of meteor- ites. Most methods employ two cosmogenic nuclides, one radioactive and one stable {e.g. A 3 9 - A 3 8 , C l 3 6 - N e 2 1 , A l 2 6 - N e 2 1 and K 4 0 - K 4 1 ) . Spallation reactions generate the nuclides at relative rates which can be estimated from controlled experiments with cyclotron beams. Assuming a constant cosmic ray flux, the rate of production of the radioactive nuclide w i l l approach its decay rate after a few half- li v e s of exposure. If this steady state had been reached p r i o r to capture by the earth, then the present-day abundances of the nuclides provide an estimate of the total exposure age. A comparison of the apparent exposure ages using different nuclide p a i r s provides a check on the accuracy of the method and the long term uniformity of the cosmic ray flux. The cosmic ray exposure" ages of meteorites are short 9 (< 10 years) compared to the s o l i d i f i c a t i o n and gas retention ages of the meteorite, parent bodies. Most stone meteorites have cosmic ray exposure ages less than 60 m.y. (Anders, 1964). Evidently, the meteorites have spent most of their existence (after codling, but before exposure to cosmic radiation) inside cold, inert bodies, or sizeable c o l l i s i o n fragments of these bodies. The Abee enstatite chondrite has a cosmic ray exposure 31 age of 13 m. y. (Begemann et a l , 1959). If we assume a mean thermal neutron production rate of one neutron/cm^sec (after Eugster et a l , 1970a, 1970b) then the integrated thermal neutron flux generated in Abee during recent exposure to cosmic radiation is 14 2 \/ 4 x 10 neutrons/cm . This flux would induce (n, 5 ) reactions in meteoritic m a t e r i a l which would be indistinguishable from i r r a d - iation effects during the early h i s t o r y of the solar system. Since the.earth i t s e l f is shielded from such processes, we may conclude that a di f f e r e n t i a l i r r a d i a t i o n I ~\L , , - ~\f „ I less than 4 x 10^ J Y n M " nE J neutrons/cm , between the Abee meteorite and the earth could not be interpreted in terms of the hypotheses of Mason and Miyashiro. This l i m i t a t i o n is i l l u s t r a t e d in Figures 2-1 and 2-2. 32 C H A P T E R 3 INSTRUMENTATION The aim of the present r e s e a r c h is to p r e c i s e l y determine the isotopic composition of Gd, Sm, and Eu in a few selected samples. P a r t i c u l a r attention w i l l be focussed on those isotopes whose abund- i i i i ii i- ^ 155 ,156 ances would be changed by neutron i r r a d i a t i o n : Gd , Gd , r - A 1 5 1 c 149 c 150 „ 151 ,-,153 Gd , Gd , Sm , Sm , Eu , Eu 3.1 Ope rat i o n a l P r i n c i p l e s of a Mass Spectrometer The most common method of determining the isotopic composition of the m e t a l l i c elements in general employs a s o l i d - source mass spectrometer. The basic operational p r i n c i p l e s of this instrument are: (a) The element to be studied is concentrated into a few drops of solution and deposited on one or two ribbon filaments, depending upon whether a single or tr i p l e filament configuration is to be used. The filaments are t y p i c a l l y made of high work- function metals such as tungsten, rhenium, tantalum or platinum. The drops are evaporated to dryness by passing a current through ^ each filament while exposed to the atmosphere. This leaves a salt, containing the element of interest, on the surface of the filament(s). (b) The filament(s) are then positioned in a mass 33 spectrometer and the system is evacuated to a low pressure ( £ IO" 7 t o r r ). (c) The sample is slowly evaporated and p a r t i a l l y ionized . at high filament temperatures. If a single filament is used, both processes occur on the same surface. In the case of the t r i p l e filament configuration (after Inghram et a l , 19 53) the sample is evaporated from two side filaments and ionized mainly by a t h i r d (center) filament which is at a much higher temperature. The type of ions produced, and their relative abundances, w i l l depend upon the composition of the salt, the filament temperatures (and work function) , . the ionization potentials of the components and the total geometry. (d) An electrostatic lens is used to accelerate the resulting positive ions through a potential difference, which is t y p i c a l l y a few k i l o v o l t s . The lens also focusses and collimates the beam into a ribbon-shaped stream of monoenergetic ions. The combined filament assembly and electrostatic lens is commonly r e f e r r e d to as an ion source. (e) The monoenergetic ion beam is transmitted through a magnetic f i e l d which separates the beam into s e v e r a l components with different charge-to-mass ratios. F o r a suitable value of the magnetic f i e l d each component beam can be fdcussed indep- endently onto a current sensing device (e.g. a Faraday cup or electron m u l t i p l i e r ) . 34 (f) By slowly varying the accelerating voltage or the magnetic f i e l d intensity, the relative intensities of the component ion beams can be monitored. Figure 3-1 shows a sample spectrum for G d + ions (omitting the mass 152 peak). In the absence •A ;,of mass d i s c r i m i n a t i o n effects, or interference by other spectra, the relative peak heights are d i r e c t l y related to the relative isotopic abundance values in the o r i g i n a l sample. It is often ; _ necessary to apply a c o r r e c t i o n for the growth (or decay) of the ion beam intensity as a function of time. This is sometimes accomplished by using a polynomial representation of the total beam intensity and transforming a l l measured peak heights to their effective values at a common point i n time. The above discussion outlines the basic operational p r i n c i p l e s of a single-stage mass spectrometer. As with any experimental technique there are many p r a c t i c a l d i f f i c u l t i e s associated with each new application. There are also fundamental limitations of the instrument i t s e l f regardless of the problem under investigation. Some of the problems which received p a r t i c u l a r attention during the present r e s e a r c h w i l l now be reviewed. 3. 2 Ion Optics "The ion source is c l e a r l y the 'heart' of the (mass) spectrometer and i s , as might be expected, the part exhibiting the greatest complexity of action. " (Barnard, 1953, p. 47) 160 L 100 A t i m e ( s e c ) - r 7 5 158 157 156 155 50 25 154 155 156 j u 157 -158 n 160 FIGURE 3-1. CONTINUOUS SCANNING OVER THE GADOLINIUM S P E C T R U M 36 P r i o r to investigating the rare earth elements, the writer's interests were directed toward the ion optical properties of ion sources, i.e. their a b i l i t y to focus and transmit ions. To study this problem, ion t r a j e c t o r i e s were calculated through several two-dimensional ion sources. Laplace's equation was solved by a finite-difference method and ion paths were computed by n u m e r i c a l integration of the trajectory equation (Loveless, 1967). These investigations led to the design of the lens system shown in F i g u r e 3-2. The tr a j e c t o r i e s i l l u s t r a t e d were calculated assuming an i n i t i a l ion energy kT = .015 eV (cor responding to T = 1750* K on the center ionizing filament - .001 x .030 inch ribbon) and angles of inclination to the lens axis from 0 to ̂  75 deg. This lens was shown to be capable of transmitting 70% of the ions produced at the center filament (for the most c r i t i c a l focussing plane i l l u s t r a t e d in Figure 3-2). Loveless and R u s s e l l (1969) compared the properties of this lens with those of an ideal lens, using the concepts of beam emmittance and instrument acceptance which had not been applied before to mass spectrometer ion sources. Through the use of these concepts it was possible to c l e a r l y define the l i m i t a t i o n to beam focussing and tra n s m i s s i o n efficiency in terms of basic instrument parameters. An understanding of this l i m i t a t i o n w i l l be of considerable value in the future design of mass spectrometers, and w i l l avoid wasted effort in attempting to achieve what is theoret i c a l l y impossible. 37 5000 V * y 5000 V F I G U R E 3-2. ION B E A M PRODUCED B Y NEW ION SOURCE 38 The ion source in Figure 3-2 is a simple design, with strong focussing properties (because of the use of thick electrodes) which has proved its effectiveness in practice and has consistently- provided satisfactory peak shape and resolution when the focussing electrodes are tuned to give maximum beam intensity. It replaces a complex stack of thin s l i t s which were acting essentially as a collimating system, with m i n i m a l focussing capability. The new source gave a factor of five increase in sen s i t i v i t y , with compar- able resolution. This achievement was coupled with a factor of two increase in s e n s i t i v i t y through the use of second order focussing shims which enabled a beam of higher angular divergence to be used. This method of achieving second order focussing i s attractive because it uses only a single pair of shims on one side of the analyzer magnet, and requires no modification of the basic magnet. It is discussed i n Appendix II. The increased sensi t i v i t y resulting from fundamental improvements in the ion optics of the available mass spectrometer represents a significant contribution by the present w r i t e r . Although the analysis of the rare earth elements would s t i l l have been possible without these modifications, the p r e c i s i o n of the results would have been somewhat poorer. 3.3 The Ionization Process 39 The most inefficient process occurring in the mass spectrometer was the conversion of neutral atoms into p o s i t i v e l y - charged ions which can be accelerated and injected into the analyzer magnet. With the t r i p l e (rhenium) filament arrange- ment, only about 15% of the evaporated atoms succeeded in hitting the hot ionizing filament; Of these, only .1% or less were emitted from the center filament as G d + ions. The ionization efficiency was greater for Sm and Eu. Three possible ways of improving the ionization efficiency w i l l be b r i e f l y reviewed. (a) The single filament technique has been shown, at least in c e r tain configurations, to produce an ion beam which is an order of magnitude more intense than that produced by the tr i p l e filament method (Eugster et a l , 1970a). Both Eugster et a l (1970a, 1970b) and Lugmair (1970) have extensively used single, zone-refined (high purity) rhenium filaments for the isotopic analysis of Gd. Since their work has been published, the present w r i t e r has used the single filament method for s e v e r a l Gd analyses, including Abee., Although comparable sen s i t i v i t y was achieved for both the single and t r i p l e filament techniques, the former did not appear to be significantly better, even when zone-refined rhenium filaments were used. Since very few single filament analyses were performed, however, 40 this may refle c t a lack of experience with this technique. A l t e r n a t i v e l y , the poor t r i p l e filament se n s i t i v i t y observed by Eugster et a l (1970) may result from low t r a n s m i s s i o n - efficiency in their ion source when using this filament confi- guration. One minor complication with the single filament technique is the fact that G d O + ions are much more abundant than G d + ions. F o r this reason the GdO* spectrum was analyzed, and a c o r r e c t i o n was made for the isotopic composition of oxygen. This actually proved to be an advantage in the analys of the Abee sample which produced in t e r f e r i n g L a O + ions in the G d + spectrum but no significant interference in the GdO^ spectrum. The magnitude of mass fractionation is generally greater for single filament analyses, making this method less precise for the determination of absolute isotopic abundances. (Mass fractionation, or d i s c r i m i n a t i o n , commonly occurs on hot filament surfaces due to pr e f e r e n t i a l evaporation of the lighter isotopes of an element. ) (b) The use of a porous filament surface (e. g. G o r i s , 1962) or the retention of the sample in a s i l i c a gel (Cameron et a l , 1969) are promising ways of increasing the y i e l d of positive ions from a single filament. The w r i t e r has not 41 investigated these techniques because of the highly specialized filament treatment or sample preparation required. (c) A t h i r d possible improvement involves the use of two p a r t i a l l y - c l o s e d (canoe-shaped) sample filaments with s l i t openings aimed d i r e c t l y at the centre ionizing filament. This method should greatly increase the probability that an evaporated atom w i l l strike the centre filament. Several attempts were made by the w r i t e r to use this technique, but no significant improvement was evident. Apparently, this method has promise but requires close control during each stage of preparation: making a V-shaped crease along the length of the filament, shaping and mounting it on posts, outgassing it without loss of duc t i l i t y , depositing and evaporating the sample inside the 'canoe', closing the filament sides without loss of sample, and c a r e f u l l y alligning them with respect to the centre filament. It i s believed that fractionation would not be enhanced by this technique. Although the ionization process is recognized as being the least efficient process occurring in the instrument, the w r i t e r has not made significant progress on this problem. The tr i p l e filament technique was used for most analyses because an elemental Gd"*" ion beam of sufficient intensity could be produced by this method, and mass di s c r i m i n a t i o n effects were expected to be less. 42 3.4 D i g i t a l Recording of Mass Spectra No modifications were made to the existing collector system of the mass spectrometer. The electronic c i r c u i t r y and the elec- trode configurations were designed by other workers in our lab- oratory ( R . D. R u s s e l l , J. Blenkinsop, J.S. Stacey, and J.M. Ozard} The entrance s l i t of the collector had a width of 0. 020 inch. A standard Faraday cup and 1 0 ^ ohm r e s i s t o r served as an ion detector. The voltage developed across the r e s i s t o r was . amplified by an adjustable gain (5 ranges), hybrid d. c. a m p l i f i e r , and fed both to a chart recorder and to a di g i t a l voltmeter. The chart recorder provided a v i s u a l display of the beam intensity at the collector while the digit a l voltmeter was used p r i m a r i l y to output data in a di g i t a l (binary-coded decimal) format. An interfacing device was designed and built by the w r i t e r to enable continuous recording of spectra on a seven track magnetic tape recorder. The data could then be processed by computer (see Section 3. 5 and Chapter 5). The output from the mass spectrometer measuring system was integrated over A* ̂  second intervals by the digi t a l voltmeter and tr a n s f e r r e d t o m a g n e t i c tape at the rate of 7 data/sec. Each datum consisted of 6 characters: 3^ binary- coded decimal digits from the digit a l voltmeter, 1 control character _ giving shunt information and scan direction (up, down or stationary), and an end-of-word character. 43 + The Gd spectrum of Figure 3-1 was taken by continuous magnetic scanning over the mass ranges 160-154 and 154-160 in succession. However, because of the steady decay of the G d + beam as a function of time, this does not represent an efficient way of u t i l i z i n g the available beam. L i t t l e time is spent measur- ing the peak intensities, while much time is lost between the peaks measuring baselines which are r e l a t i v e l y stable. More efficient use of the ion beam is i l l u s t r a t e d in Figure 3-3. In this case the baselines are measured for compu- tational purposes at positions \ mass above and \ mass below the highest and lowest masses respectively. This is done for both the down-mass and up-mass portions of the scan. Between peaks the magnet current is advanced manually. Slow continuous scan- ning may then be employed over the peaks only. This scanning technique, hereafter r e f e r r e d to as discontinuous slow scanning, was used for a l l analyses reported here. E a ch scan comprises the following records: (a) Measurement of baselines on each shunt for a minimum of 5 seconds at a position \ mass above the high-mass peak in the spectrum to be recorded. (Scan direction: stationary). (b) Discontinuous, slow scan down the spectrum to a position fiS \ mass below the low-mass peak. (Scan direction: down). (c) Measurement of baselines on each shunt for a minimum 44 time (sec) <~T \ 1 - l j 100 75 50 25 o FIGURE 3-3. DISCONTINUOUS SCANNING OVER THE GADOLINIUM SPECTRUM The diagram represents one scan of the Gd + spectrum. .In t h i s t h e s i s a scan i s defined as one down-mass sweep of the spectrum plus one up-mass sweep.. Baselines are measured above -the high-mass peak of the spectrum and - below the low-mass peak. 45 of 5 seconds. (Scan direction: stationary). (d) Discontinuous slow scan up the spectrum to a position *ss \ mass above the high-mass peak. (Scan direction: up). v. (e) Same as (a). During subsequent reduction of the data, baselines were recognized by the fact that the scan direction was stationary. The effective peak intensity and the time of measurement could r e a d i l y be deduced from the recorded data (see Chapter V). 3.5 Noise Rejection Using a Low-Pass F i l t e r It is evident from Figure 3-3 that high frequency noise is superimposed on the mass spectrum. Since we are interested in computing peak heights, rather than the area under each peak, it is p a r t i c u l a r l y important to smooth out the effect of noise by using a suitable f i l t e r . The l o c a l maxima of the f i l t e r e d spectrum w i l l then be a tr u e r representation of the signal intensity of the sp e c t r a l peaks. A low-pass f i l t e r was designed for this specific application by R. D. R u s s e l l and J. Blenkinsop. The frequency response function of the f i l t e r is i l l u s t r a t e d in F i g u r e 3-4. The data window has a width of 2. 4 sec, which is consistent with the minimum width of peak-tops of 3 sec. The f i l t e r e d data consists of - data points/sec. 6 The f i l t e r i n g process was only the f i r s t stage in the reduction of the recorded data. A l l processing was performed on an I B M 360 computer through the f a c i l i t i e s of the Computing Centre at U. B. C. Frequency response 1.0 - 0.8 0.6 0.4 0.2 0.0 0.7 F i l t e r i n g sequence (a) Convolve with 7-point box-car f i l t e r . (.5 1.0 1.0 1.0 1.0 1.0 .5) (b) Reject every second point. (c) Convolve with 3-point box-car f i l t e r . (.5 1.0 .5) (d) Convolve with 5-point box-car f i l t e r . (.5 1.0 1.0 1.0 .5) r (e) Reject 2 out of 3 points. Width of data window 2.4 sec. Nyquist frequency 3.5 cps. 1.4 2.1 Frequency (cps) 2.8 3.5 FIGURE 3-4. F R E Q U E N C Y RESPONSE OF DIGITAL F I L T E R The f i l t e r was designed by R. D. Ru s s e l l and J. B. B.lenkinsop to reject high frequency noise from the recorded data. 4 7 C H A P T E R 4 • P R E P A R A T I O N OF S A M P L E S T e r r e s t r i a l ore samples were obtained in the form of reagent Gd ? 0 , > Sm a , and E u O, from Matheson Coleman and B e l l , Cincinnati, Ohio, and the Shinetsu Kagaku Co. , Tokyo, Japan. A l l reagents had a specified purity of 99-9%. A sample name was assigned to each of the reagents for convenience of reference (see Table 4-1). A basalt sample from Kitamatsu-ura, Japan (JB1) was used to gain experience in the chemical separation and mass spec- trometry of the r a r e earth elements. Although the experimental techniques were tested on this sample, no precise analyses were performed to determine its isotopic composition. The concentration of the rare earths was an order of magnitude greater in the basalt than in the Abee meteorite. In order to consistently use microgram samples of Gd in a l l mass spectrometer analyses approximately 10 grams of the meteorite were processed for analysis, whereas only 1 gram samples of the basalt were used. 4. 1 Separation of the Rare E a r t h Elements The extraction of Gd, Sm and E u from the basalt and meteorite samples was performed by S. Yanagida and Dr. H. Mabuchi 48 TABLE 4-1. DESCRIPTION OF SAMPLES STUDIED Sample S p e c i f i c a t i o n s \ Gd- US Gd 20^ a Eu- US E u 2 0 3 a Sm- US Sm203a Gd- J Gd 20^ a Eu- J Eu 20^ a Sm- J Sm20-^a 4 ppm Gd >y JB1 ^ l b ppm Eu 5° ppm Sm ) 0 . 3 ° ppm Gd \ Abee J 0.06° ppm Eu Source Matheson Coleman & B e l l , U.S.A - derived from Humphrey's Mine F o l k s t o n , Georgia, U.S.A. Shinetsu Kagaku Co., Japan - derived from a.mine near Rio de Ja n e i r o , B r a z i l . Kitamatsu-ura, Japan G e o l o g i c a l Survey of Canada > 0 . 2 C ppm Sm i p a r t of specimen 13b a A l l t e r r e s t r i a l samples were i n the form of reagents having a s p e c i f i e d p u r i t y of 9 9 . 9 $ . b Abundances correspond to t e r r e s t r i a l mean values f o r diabases. c Based on analyses of the Abee meteorite by Shima and Honda (196?). 49 of the Department of Chemistry, University of Tokyo. The mass spectrometry was undertaken by the present w r i t e r in the Department of Geophysics, University of B r i t i s h Columbia. The rare earth elements in the lanthanide series ( L a, Ce, P r , Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, E r , Tm, Yb, Lu) have s i m i l a r chemical properties, making the separation of one element, from the remainder of the elements in the series, dif f i c u l t . In order to assess whether separation of these elements was essential, the f i r s t analyses on Gd, Eu and Sm were performed on samples containing the complete lanthanide series of elements (samples JB1-1 and JBl-2). Although Gd, Sm and Eu ion beams were successfully produced in the mass spectrometer, the abundance of i n t e r f e r i n g ion spectra inhibited r e l i a b l e interpretation of the isotopic composition of these elements. Tables 4-2 and 4-3 indicate some of the ions which were identified during the mass spectro- meter analyses. Apart from the rare earth elements and their oxides, the most c r i t i c a l contaminant was barium which produced Ba , B a F and B a C l ion beams. The abundance of these ions was attributed to the low ionization potential of barium and its compounds. Even s m a l l traces of barium were sufficientto generate intense ion beams. Several modifications were made to the separation techniques until a three-stage ion exchange process was adopted. This technique 5 0 TABLE 4-2. INTERFERING IONS NEAR GD, EU AND SM SPECTRA . Mass Ions 143 Nd(12.17) 144 Sm(3.09),Nd(23.85) . . 145 Nd(8.20) 146 Nd(17.22),BaO(.11) 147 Sm(l4.97) 148 Sm(11.24),Nd(5-73),BaO(.10) 149 ' Sm(13.83),BaF(.11) 150 Sm(7.44),Nd(5.62),BaO(2.4l) 151 Eu(47.82),BaO(6.58),BaF(.10) 152 Sm(26.72),Gd(.20)fCeO(.19),BaO(7.84) 153 Eu(52.l8),BaO(11.22),BaF(2.42) 154 Sm(22.?l),Gd(2.15),LaO(.09),Ceo(.25),BaO(71.55),BaF(6. 59) 155 . Gd(l4.73),LaO(99.67),BaO(.05),BaF(7.85) 156 Gd(20.47) ,Dy( .05) ,Ce'0(88.27) ,LaO( .04) ,BaO(.15) ,Ba F ( l l . 23) 157 Gd(15.68),PrO(99.76),CeO(.03),LaO(.20),BaF(71.70) 158 Gd(24.87),Dy(.09),Ce0(11.23),Nd0(27.04),Pr0(.04) 159 Tb(100.00),Nd0(12.15),Pr0(.20) 160 Gd(2l.90),Dy(2.29),Nd0(23.85),Sm0(3.08),Ce0(.02) 161 Dy(l8.88),NdO(8.3D Each type of ion i s l i s t e d with i t s percentage abundance as given i n the Handbook of Chemistry and Physics (1970). Barium abundances were taken from Eugster et a l (1969). A l l oxygen isotopes were considered when c a l c u l a t i n g the abundances of the oxide.ions. Some of the abundances may be i n error by as much as 1-2%. Sample & F i l . + Filament Current B a Configuration (amps) 138 Ba F 157 + LaO 155 CeO 156 + Eu 153 Sm 152 + Gd 158 Other J B l - l a 2.1 tantalum 3.0 closed canoe 4.0 .05 .04 .002 <!.005 .05 .05 GdO (174)~ .07 J B l - l b tantalum flat 1.4 2.6 2.8 2.0 >10 .'• < .003 .1 .02 .08 .04 10. .02 .02 <£.003 .1 .08 Nd +(144) = .01 JB l - 2 a 1.3 tantalum 2.6 part, closed 3.0 .02 .2 .001 .04 .06 .01 .03 «.l Tb (159) < .001 JBl-2b rhenium flat 1.1 1.5 1.8 1.9 2.2 .2 .2 .2 >10. >10. .001 <.0005 1. .008 .01 .02 .1 <.0005 <.003 Cs +(133) = .05 GdO +(174) = .01 Dy +(l62) = .03 GdO +(174) < .005:. 4. T A B L E 4-3. Examples of beam intensities observed (using variations of the tr i p l e filament technique) when Gd, Eu and Sm were not separated from the other rare earth elements. Beam intensities are specified in units of 10"^ amps. Filament currents cannot be di r e c t l y compared between analyses because of different configurations and sample histories. Blank entries in the table do not imply the absence of an ion beam. 52 attempted complete separation of Gd, Eu and Sm into three different fractions for independent analysis in the mass spectrometer. The analy t i c a l scheme is summarized in Table 4-4. The 9.5 gram Abee sample was completely dissolved in HF and HCIQ^, evaporated to dryness, and redissolved in 0.5 N HC1. The solution was loaded onto a cation exchange column containing Diaion SK#1 r e s i n . Most of the rare earths were separated from all.major constituents by elution with 2 N HC1 followed by 6 N HC1. A second column was then used to separate Gd, Eu and Sm from the other r a r e earths using 0.5 M 2-methyl la c t i c acid. The rare earth elements were eluted in order of decreasing atomic number (or increasing ionic radius). C a l i b r a t i o n of the exchange column was performed using a solution containing a r t i f i c i a l l y - produced radioactive isotopes of the r a r e earth elements. The activity of the eluate was monitored to y i e l d the elution curves in Figure 4-1. . - A f t e r separation of Gd, E u and Sm into three separate fra c t i o n s , a t h i r d ion exchange process ( s i m i l a r to the f i r s t one) was used to further reduce the concentration of a l k a l i s and alkaline earths in the individual fractions. The separated Gd, E u and Sm fractions were then converted into perchlorate salts in preparation for loading in the mass spectrometer. Although the separation technique succeeded, the elution 53 T A B L E 4-4. E X T R A C T I O N OF GD, E U AND SM F R O M A B E E 1. Dissolve 9.5 gram Abee sample in 300 ml HF and 20 m l HCIO4 and evaporate to dryness. 2. Dissolve in 20 m l 6 N HC1 and evaporate to dryness. 3. Dissolve in 400 m l 0.5 N HC1 and store in polyethylene bottle. 4. Pack r e s i n in 45 x 1.6 cm column and precondition with 1500 m l 6 N HC1 and 1500.ml H O to a-height-of 43 cm. 5. Load 400 m l solution (containing sample) onto column. 6. Elute with 840 m l 2 N HC1 to remove major constituents. 7. Elute with 150 m l 6 N HC1 and collect eluate containing a l l r are earth elements. 8. Evaporate to dryness and dissolve in a few drops of HNO3. 9. Repeat procedure 8. 10. Dissolve in 2.5 m l H2O in preparation for loading on second ion -exchange column. 11. Precondition r e s i n to remove traces of rare earth elements and barium. Wash with approximately 1000 m l 6 N HC1 and then remove acid with 3x d i s t i l l e d H£0. Convert r e s i n to NH^"^ form by adding 500 m l 7.5 N NH^OH. Wash with H zO. Preconditioning gives a r e s i n height of 22 cm. 12. The sample solution (from 10 above) was loaded onto the second (25 x 0.4 cm) cation exchange column. 13. Elute with 0.5 M 2-methyl lactic acid (pH of 3.20). 14. Collect the Gd, E u and Sm fractions from the eluate as specified in F i g u r e 4-1. 15. Add 1 N HC1 to each fraction to lower the pH to 2.5 ; then dilute (by a factor of two) by adding an equal volume of H2O. 16. Load each fraction onto a t h i r d column to remove the 2-methyl lactate from the solutions. F o r each fraction in turn, precondi- tion r e s i n with 100 m l 6 N HC1 and 50 m l H2O to a resinheight of 10 cm in a 15 x 0.4 cm cation exchange column. Load the fra c t i o n onto the column and wash with 5 m l 2 N HC1. Elute with 10 m l 6 N HC1 to release the rare earth content of the column. 17. Evaporate each fraction to dryness and redissolve in a drop of H C I O 4 . Evaporate to dryness in a teflon beaker. Gadolinium, europium and samarium were separated from the Abee sample by ion exchange, using a three column operation. Diaion SK#1, 100-200 mesh r e s i n was used in a l l columns; the cation exchange columns were made of a c r i l . The separation was performed by S. Yanagita, Department of Chemistry, U n i v e r s i t y of Tokyo. 54 LOG(ACTIVITY) (arb i t r a r y units) i l 1 i I ! I ^. 50 1 0 0 1 5 0 2 0 0 2 5 0 ml ELUATE FIGURE 4-1. ELUTION CURVES FOR TB, GD, EU AND SM A r t i f i c i a l l y - p r o d u c e d radioactive isotopes were used to cal i b r a t e the second ion exchange process (see Table 4-4). The e l u t i o n curves were determined by monitoring the a c t i v i t y of the eluate and i d e n t i f y i n g i n d i v i d u a l radio- active isotopes of Tb, Gd, Eu and Sm. During the subsequent Abee separation, the Gd-AB, Eu-AB and Sm-AB fraction s correspond to the 40 - 7 0 , ? 0 - l 4 0 and 140-240 ml fractions of the eluate. The c a l i b r a t i o n was performed by S. Yanagita, Department of Chemistry, University of Tokyo. 55 curves for the meteorite sample and the ca l i b r a t i o n run were not identical, possibly because of different concentrations of the rare earth elements i n the two solutions. The three fractions Gd-AB, Eu-AB and Sm-AB were selected from the eluate to coincide with the calibrated elution curves for Gd, Eu and Sm respectively (see F i g u r e 4-1). However, subsequent analyses in the mass spectrometer showed that most of the Gd from the meteorite was collected in the fraction Eu-AB, and a large portion of the E u was in the fraction Sm-AB. The Gd-AB fraction contained some Gd (estimated to be <• .1 j*g) as wel l as Tb. Smal l quantities of Eu, Sm, Nd, Ce, L a and Ba were observed in a l l three fractions. Special care was required during the mass spectrometer analyses to minimize the effect of in t e r - fering elements on the spectra of Gd, E u and Sm. 4. 2 Preparation of Filaments Although tantalum filaments were used for some of the p r e l i m i n a r y analyses l i s t e d in Table 4-3, a l l other analyses were performed with rhenium filaments. The rhenium metal had a specified purity of 99-9%. Two single filament analyses ( G d - J Z l and GE-AB) made use of zone-refined rhenium of very high purity. F o r these two analyses, G dO + ions were observed in the mass spec- c trometer at much lower temperatures ( 1350 C) than was observed when the other (99.9% purity) rhenium was used (<*•• 1700* C). 56 The most c r i t i c a l i m p urities in the rhenium metal were Ba, L a and Ce. These were reduced to tolerable levels by outgassing a l l filaments at 2000° C for one hour. This eliminated L a and Ce completely while Ba was reduced to a contamination le v e l which was only a s m a l l fraction of the Ba contamination in the samples themselves. The purity of a l l filaments used for analyses on the Abee meteorite was checked in the mass spectrometer, at temperatures w e l l above operating conditions, p r i o r to loading the samples. A l l reagent analyses were performed with m i c r o g r a m quantities of Gd, Sm or Eu. The elements were studied separately, except for Sm and E u which were occasionally analyzed together because of the lack of isobars for these elements. The samples were deposited on outgassed filaments as chlorides, and then con- verted to perchlorate salts on the filaments. F o r Gd single filament preparations, the filaments were subsequently heated in a i r to a dull red glow ( in a dark room) to convert the salt to the oxide form, thereby increasing the production of G dO + ions in the mass spec- trometer. The Abee fractions were taken up in a drop of 3x d i s t i l l e d water, deposited on outgassed filaments, and evaporated to dryness. The complete Sm-AB fra c t i o n was loaded for one t r i p l e filament analysis of Sm. Most of the Eu-AB fra c t i o n was loaded for one t r i p l e filament analysis of Eu (at low side filament temperatures), 57 followed by an analysis of Gd in the same fraction (at higher temperatures). Approximately one half of the Gd-AB fraction was loaded for one tr i p l e filament analysis of Gd. The remainder of fractions Eu-AB and Gd-AB was combined for one single filament analysis of Gd. This latter analysis w i l l be r e f e r r e d to as GE-AB. 4. 3 Production of Ion Spectra with Minimum Interference Operating pressures of 10 t o r r were maintained for a l l -- _ g mass spectrometer analyses. P r e s s u r e s approaching 2 x 10 t o r r were achieved for many analyses on the Abee fractions by thoroughly outgassing the instrument, and by allowing s e v e r a l h o u r s for barium and other contaminants to be driven out of each sample at low filament temperatures. There was no substitute for slow heating of the filaments to increase the purity of the Gd, Sm and Eu in each sample. None of the contaminants in the reagent samples caused significant s p e c t r a l interference which could not be eliminated by careful heating of the sample filaments in the mass spectrometer. F o r the meteorite analyses, however, it was not always possible to eliminate a l l i n t e r f e r i n g ions in this way. LaO and Nd~*" ions interfered with the measurement of the S m + spectrum during analysis Sm-AB, and LaO"^ ions interfered with the Gd~^ spectrum during analyses Eu-AB and Gd-AB. Fortunately, it was possible to monitor the intensity of the Nd and L a O + ion beams 58 a t m a s s e s 146 a n d 155 r e s p e c t i v e l y , s o t h a t a c o r r e c t i o n c o u l d b e m a d e t o s e p a r a t e t h e s e s p e c t r a f r o m t h e S m s p e c t r u m c o m p u t a - t i o n a l l y . T h e m e t h o d o f a p p l y i n g t h e s e c o r r e c t i o n s i s d i s c u s s e d • i n C h a p t e r 5. A s i m i l a r c o r r e c t i o n f o r L a O + i n t e r f e r e n c e ( a t m a s s 155) w i t h the G d s p e c t r u m c o u l d n o t be m a d e b e c a u s e of t h e s i m i l a r m a s s r a n g e s f o r b o t h i o n s . T h e o n l y w a y i n w h i c h t h e 155 G d i s o t o p i c a b u n d a n c e c o u l d b e d e t e r m i n e d f o r t h e m e t e o r i t e s a m p l e w a s b y u s i n g t h e G d ; 0 + s p e c t r u m p r o d u c e d d u r i n g t h e s i n g l e f i l a m e n t a n a l y s i s G E - A B . 59 C H A P T E R 5 C O M P U T E R REDUCTION OF C O M P L E X S P E C T R A 5.1 Evaluation of Peak Heights The method of recording spectra, and the definition of a scan (the unit r e c o r d on magnetic tape) were discussed in Section 3. 4. A representative scan of the Gd* spectrum was shown in Figure 3-3. A combined E u * and S m + spectrum is i l l u s t r a t e d in Figure 5-1. Data was recorded in di g i t a l form and a low-pass f i l t e r was used to remove high frequency noise from the spectrum (see Section 3. 5). After f i l t e r i n g the data, a moving 5-point average was computed for the purpose of locating the time coordinates of the various peaks. At a position where the 5-point average was a maximum, the associated peak height (without the baseline removed) was chosen to be the largest value of the five points comprising the moving average. The true peak height was then computed by- subtracting the recorded baseline which was represented, for each shunt in turn, by the best least-squares straight line fitting a l l (baseline) data (within a single scan) which was recorded while the scan di r e c t i o n was stationary. The corrected peak height and its associated time coordinate were subsequently used to represent the intensity and time of measurement of the. component ion beam 60 152 153 100 "T" 75 time ( s e c ) - — r — — 50 25 0 144 Sm 147 •149 148 150 i > i ! .151 ;Sm Sm Sm Sm Eu Sm 154 i i Eu Sm FIGURE 5-1. SLOW DISCONTINUOUS SCANNING Down-mass p o r t i o n o f a sc a n showing the Eu and Sm s p e c t r a . 61 at a p a r t i c u l a r mass number. The reduction of data from a single scan gave two measurements of the intensity of each peak in the spectrum. B y retaining a running time coordinate it was possible to combine measurements derived from a set of s e v e r a l successive scans and apply a c o r r e c t i o n for the continuous growth or decay of each type of ion during the set of scans. 5. 2 Least Squares Reduction of a Set of Scans Past experience in our mass spectrometer laboratory has shown that a t h i r d order polynomial provides a f a i r l y good representation of the intensity of a single peak as a function of time t , i . e. 2 3 I(t) = a i + a 2 t + a 3 t + a 4 t (5-1) In order to minimize the e r r o r s introduced by this type of approx- imation it is desirable to make use of the fact that the isotopic composition of each type of ion remains essentially constant during a set of scans at fixed filament temperatures. As an example, 151 .153 the intensities 1̂  and I^ of the Eu and E u component ion beams may be represented in the form 62 2 3 I (t) = r (b + b t + b t + b t ) 1 l i 2 3 4 ' I 2 ( t ) = r 2 ( b + b 2 t + b 3 t 2 + b 4 t 3 (5-2) where r^ and r ^ are proportional to the relative isotopic abund- ance values (one of which must be a r b i t r a r i l y fixed, e. g. r^ sr 1. 0). These parameters are constant for the duration of a set of scans at fixed filament temperatures, except for m i n i m a l fractionation effects which are generally s m a l l ( < 0.05% per unit mass difference) compared with s t a t i s t i c a l e r r o r s . The term in brackets is a polynomial representation of the growth or decay 151 153 c h a r a c t e r i s t i c s of both the Eu and E u ion beams. Repres- entation (5-2) requires the determination of only 5 parameters whereas 8 parameters would have been required if the constancy of the isotopic ratio had not been used as a constraint. A suitable method for determining the optimum values of the parameters r ^ , b^, b_>, b^, b^ is one which minimizes the variance between the measured peak heights and their polynomial representation in equations (5-2). If p ^ a n d t ^ are the peak 151 th height and time of measurement of the Eu peak during the k measurement of that peak, and p , t are s i m i l a r quantities for 2k 2k 153 the E u peak, then the function tobe minimized may be represented in the form 63 k = l,2n I L where n is the number of scans in the set. There are se v e r a l possible sources of e r r o r when experimentally measuring any single peak intensity on the mass magnet current, making an e r r o r in shunt selection, sudden line transients, etc. These can readily be recognized on the chart r e c o r d during an analysis. P r o v i s i o n has therefore been made for eliminating faulty peak measurements from the right hand side of equation (5-3). Such e r r o r s were, however, infrequent in practice. spectrometer: overshooting a peak when manually advancing the The conditions which must be satisfied in order for g(r ,b ,b ,b ,b ) to be a minimum are: ^ r 2~ * b l " * b 3 " " ^ b 4 ~ (5-4) These give r i s e to five non-linear equations in the variables A solution is possible by a perturbation method provided a good f i r s t approximation may be obtained. 64 This is possible since the isotopic ratio r ^ / r ^ is known to within a few per cent. If r ^ is i n i t i a l l y treated as a constant =: r , then four linear equations may be obtained in the variables : b^, b^, b^, b^. These can e a s i l y be solved d i r e c t l y to give good estimates of b^, b , b^, b^. At this stage, equations (5-4) may be l i n e a r i z e d using the new variables A r ^ , Ab^, ^b^, A ^ 3 ' where r 2 = r 2 +. A r 2 b 3 = b 3 + A b 3 b j S b j + A.bj . b 4 = b 4 + A b 4 (5-5) b 2 = b 2 + A b 2 The l i n e a r i z e d equations (5-4) may be solved to get an improved estimate of r 2 , b ^ b 2 > b^, b 4 . Several iterations may be performed in this way until r 2 > b^, b 2 , b^, b 4 are a l l known with sufficient p r e c i s i o n ( < .01% was considered more than adequate).. Figure 5-2 shows a direct comparison between the measured peak heights and the solution polynomials I^(t) and ^ ( t ) for a set of scans of the E u spectrum. The quality of the polynomial fit is quite good, even when the beam intensity shows strong growth and/or decay c h a r a c t e r i s t i c s . In p r a c t i c e , however, the beam intensity usually changed monotonically during a set of scans, with a total v a r i a t i o n of 20% or l e s s . 4.0 3.5 3.0 2.5 Ion beam i n t e n s i t y (10" 1 2amp) Upper c u r v e : I ^ a r ^ b j + b 2t + b^t2 + b 4 t 3 ) 2 3 L o w e r curve: I (t) = r (b + b t + b t + b _t ) 1 1 1 2 3 4 ; r = 1.0 r _ , b, , b , b_, and b . were d e t e r m i n e d 2 1 , 2 3 4 by m i n i m i z i n g the fu n c t i o n g(r2»b) = j k = I, 12 k= 1,12 ( P z k - V z k ) ) ' * 4 1 x • * • . . . x • I JL 153 E u s c a n 6 x. x* .x E u 151 E u r l 151 E u 153 M e a s u r e d peak height p j ^ at time t ; 10 15 L e a s t s q u a r e s p o l y n o m i a l fit- I .—I ^ T i m e (min) 20 25 F I G U R E 5-2. P O L Y N O M I A L R E P R E S E N T A T I O N O F ION B E A M I N T E N S I T Y The E u 1 5 1 and E u " 3 i o n beams have been r e p r e s e n t e d b y p o l y n o m i a l s of degree 3 f o r a set of scans. 66 The above example i l l u s t r a t e s the basic numerical proce- dure employed in this r e s e a r c h to compute the least squares poly- nomial representation of the various peak intensities associated with ions of a single element. F o r the majority of analyses it was not necessary to apply a c o r r e c t i o n for the effect of interfering ion spectra. However, whenever spe c t r a l interference was observed, care was taken to record additional peaks i f possible, so that the growth or decay c h a r a c t e r i s t i c s of each additional ion species could be determined and subsequently removed. Where s e v e r a l ion spectra occur in a set of scans the equations to be solved are more complex. A l s o , not a l l isotopic ratios should be treated as variab l e s . Where inte r f e r i n g ions form only a s m a l l percentage of the total beam, it is more reli a b l e to f i x their isotopic composition at published values than to treat them as additional v a r i a b l e s i n the least squares solution. In the general case, where there are m elements and q different peaks, the function which replaces (5-3) as the sum of squares to be mi n i m i z e d is (5-6) where p ^ is the k**1 measurement of the j ^ 1 peak at time 67 t. •; r . i s the relative isotopic abundance associated with peak j and element i; and b^, b ^ ' b ^ , b ^ a r e the coefficients of the growth polynomial for element i . The vector b represents a l l b. (i = l,m; w = 1, 4) 1W while the vector r represents only those abundance values which are to be treated as variables in the least squares solution. F o r convenience, a new l o g i c a l function f^j may be defined such that f-- is false only when r.. is to be treated as a variable. This means that r represents a l l r ^ (i s: l,m; j = l,q) for which f — is false. (Note that equation (5-6) includes the common situation where isotopes of more than one element exist at a given mass number The function g(r,b) is a minimum when 14-= 0 i = l.m; w = 1, 4 «b. 1W (5-7) and \_g_ _ o i ~ \ , m ; j = l , q ; = false j r . - i j These equations reduce to (P.k- SL p.. SL b. F-1)T .ty,-1 = 0 j = 1, q k = 1, 2n i = 1, m w = 1 ";4 u =: 1, m; v ; 1,4 S L ( P v k - ^ b. t ^ X ^ b t\\=0 1 T -> v k . 1 iv r~ . iw vk , , uw vk k = 1, 2n i = l , m w = l,4 w = 1, 4 u = l , m ; v = l , q ; f = false (5-8) uv v ' 68 Equations (5-8) may be l i n e a r i z e d and solved by a perturbation method s i m i l a r to the previous example, in order to obtain solution vectors r and b. 5. 3 Calculation of Means and Standard Deviations of Isotopic Ratios for a Set of Scans It is evident from the discussion in the previous section that isotopic ratios resulted d i r e c t l y from the solution of equations (5-8). In the case of the example given e a r l i e r one could readily calculate the isotopic ratio Eu*^/Eu == r ^ / r ^ . However, there is no easy way to estimate the e r r o r in this ratio apart from com- paring results from s e v e r a l sets of scans. This was considered inadequate because of large variations in the stability of an ion beam from one set of scans to another. Differences in sample purity and abundance, and variations in the operating c h a r a c t e r i s t i c s of the mass spectrometer also contribute to r e a l variations in the p r e c i s i o n achieved. (When isotopic ratios and standard deviations were calculated by the method to be described in this section, a test for homogeneity of variance from one set of scans to another showed convincingly that there were r e a l - v a r i a t i o n s in the variance. ) A method was therefore devised for estimating the standard deviation of each isotopic ratio for a set of scans. The polynomial representation of the intensity of peak j as a function of time is 69 I.(t) = <^ r.. > _ -b. t J i = l,m 1 Jw = l , 4 1 W (5-9) Suppose the dominant component in this peak belongs to the element represented by subscript i = s. The isotopic abundance coefficient associated with peak j and element s is r g j - If w e assume that, on the k**1 measurement of peak j , the difference between the measurec as a peak height p., and the polynomial value I.(t. ) may be regarded J J-- jk semi-independent measurement of r s j > then we can transform equation (5-9) into the form p.,. = r'(s,j,k)5>"~ b t ^ " 1 + J> r.. b. t w ' * jk ^-—r- sw jk ~ i j ^ aw jk w = 1,4 i = l , m w = 1, 4 J i jd s r ' ( s , j , k ) = ( P -21 r £Z__ b. b J k i = l,m 1 J w = l , 4 1 W J k w = l , 4 s w J k i ptf s (5-10) where r'(s, j,k) is a semi-independent measurement of the coefficient r g j at time tj^ , . Consider the case where we are interested in measuring the isotopic ratio R =: r . / r . , the ratio of isotopes of element s S J1 S J 2 o c c u r r i n g at peaks j and j . Because of the symmetry of spectra within a single scan it is reasonable to compute one estimate of R for each scan. Let R^ (i == 1, 2, ... n) be the computed estimate of 70 R during scan i . Then ' r H s . j p l ) + r ' ( s , j , 2) r 1 ( s , j 2 , l ) + r ' ( s , j 2 > 2) r'(s,j ,3) + r ' ( s J j 1 , 4 ) : R z = : r ' ( s , j 2 , 3 ) + r ' ( s 3 j 2 , 4 ) (5-11) etc. This method of computing semi-independent estimates of each isotopic ratio was applied to a l l ratios of interest for the three elements Eu, Sm and Gd. The combined data, for each isotopic r a t i o , was then presented in the form i = 1, n ( ^ ~ ( R i - R ) 2 / ( n - l ) ) ^ i = 1, n qr = <r~ / f n (5-12) where R is the mean value of the isotopic r a t i o , ̂ ~ is an estimate of the standard deviation of each measurement R^, and P̂" is an estimate of the standard deviation of the mean. 71 5. 4 Calculation of Fr a c t i o n a t i o n - C o r r e c t e d Parameters In the precise ̂ determination of isotopic abundances, care must be taken to account for any processes which might cause mass d i s c r i m i n a t i o n in the sample at any stage p r i o r to, or during, the measurement of an ion beam in the mass spectrometer. In the present research, mass fractionation was detected by comparing isotopic ratios for s e v e r a l sets of scans obtained at one or more filament temperatures. The largest fractionation effect observed was 0.3% per unit mass difference. This represents the change in isotopic composition of the ion beam over the extremes of sample filament temperatures for which a measureable ion beam could be produced. F o r a l l three elements, Eu, Sm, and Gd, the process proceded toward the enrichment of the heavier isotopes in the ion beam (or the depletion of the lighter isotopes) with increasing filament temperatures and time. In agreement with simple theory (see Ozard and R u s s e l l , 1970) and common experience, the fractionation effect for a par- t i c u l a r element was observed to satisfy the equations X = X N ( 1 + ( m 2 - m 1 ) ^ ) Y = Y N ( 1 + ( m 4 - m 3 ) ^ ) (5-13) where X and Y are measured isotopic ratios (for the same element) 72 corresponding to the mass ratios m,/m and m /m . - 1. 2 3 4 respectively; X^T and Y_ are normalized (or fractionation-N N corrected) isotopic ratios; ^ is the fractional change in the i s o - topic ratios per unit mass difference (due to fractionation). ^ «-l. The quantity ^ may be eliminated from equations (5-13) to give Y N ( m 4 - m 3 ) ^ , Y = Y N + : U-X N) (5-14) X N { m 2 " m l ) F i g u r e 5-3 shows the close agreement between the slope predicted by equation (5-14) and the measured isotopic ratios. The s o l i d line i l l u s t r a t e d in this plot was forced to pass through the point (.9361, .6769) adopted by Eugster et a l (1970a) for t e r r e s t r i a l gadolinium. The published standard deviation of the latter coordinate is 0.01% while the ratio G d 1 5 6 / G d l 6 ° = .9361 is the value which they used for normalization of a l l other Gd isotopic ratios. The normalization process involves measuring two isotopic r a t i o s , X and Y, assigning a specific value to 156 ; 160 X-^ (e.g. X^j — Gd /Gd rr .9361) and solving equation (5-14) f o r Y N . An alternative method may be used to remove the effect of fractionation from measured isotopic ratios. Equation (5-14) may be rewritten in the form 7 3 FIGURE 5-3. MASS S P E C T R O M E T E R F R A C T I O N A T I O N .155 Y = .688 .684 .680 .676 ~ .672 .930 .9 34 .938 .942 .946 The data points represent measured isotopic ratios for individual sets of scans (usually 6 scans per set). Data are shown for one single filament analysis (after c o r r e c t i o n for the isotopic composition of oxygen) and for one t r i p l e filament analysis on the same sample. The s o l i d line has the theoretical fractionation slope and passes through the point (.9361, .6769). 74 X-X Y N ( y- ) = l + * ( £-) Y N X N N X N Y ^— = = Y = constant (5-15) X X N where <X = (m^-m^J/^m^-m^). By measuring appropriate isotopic ratios for the elements Gd and Sm it is possible to evaluate parameters which are corrected for fractionation and are also very sensitive to ^ ^ 5 5 , v ^,156 ,-.,157, v , , the neutron capture processes Gd (n, $ )Gd , Gd (n, <j ) G d ^ ^ and S r a ^ V i ^ )Sm Three parameters which satisfy these conditions are given in Table 5-1. - Each parameter is sen- sitive to only one capture process. The value of a l l three parameters (A,B and C) w i l l be greater for samples which have been exposed to a l a r g e r neutron flux. They should be constant, however, for a l l analyses performed on the same sample. Since there are only two isotopes of Eu, no c o r r e c t i o n for fractionation was possible. The method of computing a normalized ratio Y^. or 75 TABLE 5-1. FRACTIONATION-CORRECTED PARAMETERS c< = ± \ = Constant m2-m.1 Gd 156 g d 1 6 0 1 _ G d ^ ( G * £ £ * \ Gdl55 G d156 4 G d155 G d l 6 0 G d ^ ! G d ^ 1 R G d 1 ^ 8 r G d ^ 8 | G d15? G d158 2 a G d i 5 ? ^ G d l 6 0 ; . S m 1 5 0 S m 1 ^ 1 • S m 1 ^ Sml50 | Sm^9 Sml50 2 S m l 4 9 ^ S m152 j Each of the f r a c t i o n a t i o n - c o r r e c t e d parameters A,B and C i s s e n s i t i v e to a p a r t i c u l a r neutron capture p r o c e s s . The s e n s i t i v i t i e s of the parameters are somewhat g r e a t e r than the simple r a t i o s (X i n above t a b l e ) from which they are d e r i v e d . The s e n s i t i v i t i e s r e l a t i v e to the i s o t o p i c r a t i o s G d 1 5 5 ^ 1 ^ 6 , G d l 5 7 / G d 1 5 8 and l4Q 150 Q *5 5 Sm /Sm are ̂  , ̂  and ̂  r e s p e c t i v e l y . The i s o t o p e s Gd^^° and Sm^^2 are not s i g n i f i c a n t l y changed by neutron i r r a d i a t i o n . 76 afractionation-corrected parameter Y w i l l now be given. Let and be the appropriate isotopic ratios computed for j scan i (i = i , 2,. . . n) from equations (5-11). The corresponding normalized ratio Yj^. (or parameter \ .) was computed from equation (5-14) (or equation (5-15)). The mean Y^ (or ^ ) and standard deviations \f~ and C~ were then computed from equations (5-12). 5.5 Co r r e c t i o n for the Isotopic Composition of Oxygen A l l t r i p l e filament analyses were performed on elemental E u , Sm or Gd ions. A few single filament analyses were performed using the GdO spectrum, and a subsequent 17 c o r r e c t i o n was required for the s m a l l contribution of the O 18 and O isotopes. The corrections required for a l l isotopes 16 + at masses where GdO ions were measured are given by the equations - (168/176) r 1.00232 x (168/176) c o r r meas ((170/176) = 1.00213 x (170/176) c o r r meas (171/176) - 1.00226 x (171/176) v 'corr - 'meas . (172/176) =1.00183 x (172/176) c o r r meas (173/176) =0.99990x (173/176) c o r r meas (174/176) = 1.00040 x (174/176) v 'corr v meas (5-16) 77 The ratios indicated by the subscript 'meas' are the measured ratios (uncorrected for mass fractionation) while those indicated by the subscript 'corr' are the ratios which would have been measured if oxygen were monoisotopic (i.e. only G d O ^ + ions present). The corrections in equations (5-16) are based on the isotopic ratios 0 1 8 / 0 1 6 = .00204 and 0 1 7 / 0 1 6 = .00037 given by Nie r (1950). Eugster et a l (1970a) have estimated that the maximum e r r o r in any of the c o r r e c t i o n factors in equations (5-16) is less than 0. 005%, which is negligible compared to the experimental uncertainty. 78 C H A P T E R 6 I N T E R P R E T A T I O N OF M E T E O R I T I C AND T E R R E S T R I A L ISOTOPIC RATIOS 6. 1. T e r r e s t r i a l Gadolinium The results of several isotopic analyses on t e r r e s t r i a l samples Gd-J and Gd-US are summarized in Table 6-1 and Figures 6-1 and 6-2. (Complete data for individual scans is given in Appendix III). Within the experimental uncertainty of the measured isotopic ratios (shown as two standard deviations of the mean) there is no significant difference between the t e r r e s t r i a l samples derived from two different geographic locations (eastern US and Rio de Janeiro). Furthermore, the good agreement with the t e r r e s t r i a l ratios obtained by Eugster et al (1970a) supports the view that t e r r e s t r i a l Gd is of f a i r l y uniform isotopic composition when samples representing large segments of the earth are compared. It is interesting to note that the displacement between the mean t e r r e s t r i a l values obtained by Eugster et a l (1970a) and those obtained by the present w r i t e r (Figures 6-1 and 6-2) is s i m i l a r to that which would be produced by a difference in exposure to thermal neutrons. (The calculation of the slope of each of the theoretical c o r r e l a t i o n lines for thermal neutron capture is given in Appendix IV. ) In the writer's opinion, no significance can be attached to this apparent 79 TABLE 6-1. ISOTOPIC RATIOS OF GD IN TERRESTRIAL SAMPLES Analyses G d 1 ^ * G d 1 6 0 G d 1 ^ * G d l 6 U G d ^ 8 * G d i 6 ° A - B Gd-Jl,j4,J5 t r i p l e f i l . .67686 ±.00015 .71620 ±.00016 1.13620 ±.00044 1.36030 ±.00027 1.69083 ±.00067 Gd-j6 t r i p l e f i l . .67685 +.00026 .71593 +.00014 1.13566 +.00021 1.36029 +.00054 I.69038 +.00054 Gd-JSl,JS2 s i n g l e f i l . .67693 +.00022 .71575 +.00029 1.13586 +.00063 1.36001 +.00042 1.6910 1.0013 Gd-JZl s i n g l e f i l . z o n e-ref. Re .67702 ±.00025 .71591 . ±.00023 I.I357O ±.00018 1.35999 ±.00050 1.69070 ±.00069 Gd-US t r i p l e f i l . > .67682 ±.00015 .71639 ±.00048 1.13595 ±.00034 1.36026 ±.00027 1.6900 ±.0016 Average ( t h i s work) .67688 +.00008 .71601 +.00009 1.13576 +.00012 1.36021 +.00016 I.6906O +.00034 Eugster e t a l (1970a) s i n g l e f i l . z o n e-ref. Re .67692 ±.00009 •71589 ±.00004 1.13590 ±.00009 1.36024 ±.00018 1.69108 ±.00031 C o l l i n s e t a l (1956) .683 .717 1.124- 1.348 1.662 * F r a c t i o n a t i o n c o r r e c t i o n normalized t o Gd156//G(jl60 _ # 0/361,- A l l s i n g l e f i l a m e n t analyses used the Gd,0+ spectrum,and a c o r r e c t i o n was made f o r the i s o t o p i c composition of oxygen. The f r a c t i o n a t i o n - c o r r e c t e d , i r r a d i a t i o n - s e n s i t i v e parameters A and B have the va l u e s G dl55 Gdi°o B G d157 G dl60 Only s t a t i s t i c a l e r r o r s are shown. They correspond to two standard d e v i a t i o n s of the mean. 80 FIGURE 6 - 1 . ISOTOPIC ANALYSES ON TERRESTRIAL GADOLINIUM , 1 5 8 * Gd" A G d 1 6 0 1 . 1 3 6 5 1 . 1 3 6 0 1 . 1 3 5 5 1 . 1 3 5 0 Gd-JSl,JS2 ( s i n g l e f i l - ament) Gd-JZl ( s i n g l e f i l - ament; zone- r e f i n e d Re) I I I T I G d - J l , J 4 , J 5 ( t r i p l e f i l a m e n t ) Value adopted by Eugster et a l (19?Oa) Gd-US ( t r i p l e f i l a m e n t ) Weighted mean ( a l l analyses) \ Gd - J 6 ( t r i p l e f i l a m e n T h e o r e t i c a l c o r r e l a t i o n l i n e f o r thermal neutron, capture (see Appendix IV) 1 I '. L_ Gd 157 . 7 1 5 0 . 7 1 5 5 . 7 1 6 0 . 7 1 6 5 . 7 1 7 0 Gd T7JO G d 1 5 8 / G d l 6 ° *" vs G d 1 5 ? / G d l 6 ° ' * f o r a l l t e r r e s t r i a l analyses. *. F r a c t i o n a t i o n c o r r e c t i o n normalized to G d ^ ^ / c d 1 ^ ^ . 9 3 6 1 . 81 F I G U R E 6-2. C O R R E L A T I O N B E T W E E N B A N D A F O R T E R R E S T R I A L GD /ft 1.693 1.692 1.691 I.690 1.689 1.688 1.687 B = Gd158 158 I G ^ 3 ? ( H d i ^ o ) G d - J S l , J S 2 ( s i n g l e f i l a m e n t ) V a l u e c o m p u t e d f r o m d a t a o f E u g s t e r e t a l (1970a) G d - J Z l ( s i n g l e f i l a m e n t ; z o n e - r e f i n e d R e ) W e i g h t e d m e a n ( a l l a n a l y s e s ) G d - J l ,J4,J5 ( t r i p l e f i l a m e n t ) \ ^ G d - j 6 ( t r i p l e f i l a m e n t ) G d - U S ( t r i p l e f i l a m e n t ) T h e o r e t i c a l c o r r e l a t i o n l i n e f o r t h e r m a l n e u t r o n c a p t u r e ( s e e A p p e n d i x I V ) G d 156 nA156 i G d G d " 155 v G dl60 ) 1.3590 1.3594 1.3598 1.3602 1.3606 C o r r e l a t i o n " b e t w e e n i r r a d i a t i o n - s e n s i t i v e p a r a m e t e r s B a n d A f o r a l l t e r r e s t r i a l a n a l y s e s . 82 c o r r e l a t i o n since a l l the isotopic ratios agree within the 95% confidence l i m i t s shown. A l s o , the specified e r r o r s correspond to s t a t i s t i c a l e r r o r s only. The analyses were performed on different mass spectrometers using significantly different tech- niques for scanning, recording and reducing mass spectra. The good agreement for a l l isotopic ratios suggests that any external e r r o r s must be s m a l l ; however, the existence of any bias in measure- ment on either instrument could only increase the absolute e r r o r of the isotopic ratios. and meteoritic samples, t e r r e s t r i a l Gd w i l l be assumed to have the isotopic composition given by the average values in Table 6-1. F o r the purpose of comparison between Gd in t e r r e s t r i a l The mean r and standard deviation were computed, for each column in turn, from the equations 2 i s 1, n 2 i = 1, n i (6-1) where x- and ̂ P. are the mean and standard deviation for the i th entry in the column. 8 3 TABLE 6-2. ADDITIONAL RESULTS FOR TERRESTRIAL GADOLINIUM Gd 152 *+ Gd 160 Gd 154 *+ Gd 16*0 G d 1 ^ 0 meas This work Eu g s t e r e t a l (1970a) Value c u r r e n t l y used f o r normali- z a t i o n of a l l other r a t i o s .00918 b i.00005 .00928 ±.00002 .09961 ±.0000? .09975 1.00002 .943 d ±.004 e .9361 1 (adopted a r b i t r a r i l y ) Per cent abundances of t e r r e s t r i a l Gd Gd 152 Gd 154 Gd 155 Gd 156 Gd 157 Gd 158 Gd 160 T h i s work # .203 2.191 14.86 20.52 15.67 24.80 21.76 E u g s t e r e t .2029 2.1809 14.800 20.466 15.652 24.835 21.863 a l (1970a) C o l l i n s e t a l (1956) .205 '2.23 15.1 20.6 15.7 24.5 21.6 * F r a c t i o n a t i o n c o r r e c t i o n normalized t o G d 1^^/Gd 1^°= .9361. + E r r o r s r e p r e s e n t two standard d e v i a t i o n s of the mean. # Normalized to G d 1 5 6 / G d l 6 ° = .943 . a Based on a l l t r i p l e f i l a m e n t analyses on t e r r e s t r i a l samples Gd-J and Gd-US. b Average o f three data s e t s ( r e p r e s e n t i n g three d i f f e r e n t a n a l y s e s ) c Average of e i g h t data s e t s (more than one a n a l y s i s ) , d Simple average f o r 23 data s e t s ( s e v e r a l a n a l y s e s ) , e. One s t a n d a r d d e v i a t i o n of a s e t (not of the mean), f Based on s i n g l e f i l a m e n t a n a l y s e s . 84 Analyses on the t e r r e s t r i a l sample Gd-J also showed the absence of any detectable bias between data obtained by the t r i p l e filament method (using the G d + spectrum) and single filament data (using the GdO + spectrum and correcting for the isotopic composition of oxygen). Furthermore, there was no significant inter-analysis e r r o r which could be attributed to loading the same sample (Gd-J in solution form) onto different filaments. There are two additional Gd isotopes which were seldom 152 154 measured: Gd and Gd . Their t e r r e s t r i a l abundances are reported in Table 6-2. A l s o , the mean value of the unnormalized G d ^ ^ / G d 1 ^ r a t i o is reported in Table 6-2 for a l l t r i p l e filament analyses on t e r r e s t r i a l Gd. This value is 0.7% higher than the value currently used for normalization of a l l other ratios (after Murthy et al,1970 , and Eugster et a l , 1970); the current value was determined from single filament analyses only. The t r i p l e f i l a - ment value is undoubtedly closer to the absolute abundance ra t i o , and should therefore be used when calculating the absolute abundance of the Gd isotopes (see Table 6-2). 6. 2 Gadolinium in the Abee Meteorite Three separate analyses were performed on Gd from the Abee meteorite. Two of these used the t r i p l e filament con- figuration and the Gd^ spectrum. Because of L a O + interference at mass 155 (see Tables III - 6 and III - 7 of Appendix III) it was not 85 possible to measure the Gd abundance. Consequently, only «-,157 ,160 * _ ,158/>~ J60 * , 0 , Gd /Gd , Gd /Gd and the parameter B couldbe deter- mined. The t h i r d analysis was performed by the single filament technique. Because the G dO + spectrum was employed, there was no interference from LaO* ions, and it was possible to determine 155 160 * both the Gd /Gd ratio and the parameter A. The average values for the three analyses were computed from equations (6-1), and are shown in Table 6-3. It i s apparent that none of the Gd isotopic ratios differs by more than 0. 06% from the t e r r e s t r i a l value. This also applies to the i r r a d i a t i o n - s e n s i t i v e parameters A and B. A plot of r,157,_,l60 * 158.„,160* ,_. , , +, . ' Gd /Gd vs. Gd /Gd (Figure 6-3) shows that the 9 5 % confidence l i m i t s for the Abee and t e r r e s t r i a l data overlap, and the slope of the line passing through the two points is not the same as the slope of the theoretical c o r r e l a t i o n line for t e r m a l neutron capture. On a plot of B vs A, the data points for Abee and for t e r r e s t r i a l Gd appear to be dist i n c t l y different on the basis of the 9 5 % confidence l i m i t s . However, the slope of the line through the two points is almost perpendicular to the> theoretical c o r r e l a t i o n line for thermal neutron capture. Within the experimental uncertainty, there is no s i g n i f i - cant thermal neutron i r r a d i a t i o n anomaly for Gd in the Abee sample ..(when compared with the earth). This result is disappointing because of its implications for the class of enstatite chondrites as a whole. 86 The failure to observe an anomaly means that there is no definitive experimental evidence to support the theories of Miya- shiro and Mason: that differences in the oxidation state of chondrites result from differences in their mean distances from the sun during the e a r l y h i s t o r y of the solar system. At the same time, their theories have not been disproved since other essential conditions must also have been satisfied in order to produce a detectable i r r a d i a t i o n anomaly. These w i l l be discussed in Section 6.6.. P r e c i s e analyses on meteorites have also been performed by Eugster et a l (1970a). Their results are plotted in Figures 6-3 and 6-4. (The data for Pasamonte and Norton County are weighted means of s e v e r a l . published analyses. ) Only the Norton County achondrite contains Gd of significantly different isotopic composition from t e r r e s t r i a l Gd. It lies along the theoretical c o r r e l a t i o n line for thermal neutron capture. An i r r a d i a t i o n anomaly c l e a r l y exists; it has been attributed to the long cosmic ray exposure age of the Norton County meteorite. The Weekeroo Station i r o n has an even longer cosmic ray exposure age, but its s m a l l e r mass and i n f e r i o r moderating properties presumably resulted in the production of very few thermal neutrons (Eugster et a l 1970a). The other meteorites include another i r o n (Copiapo), an achondrite (Pasamonte) and a bronzite chondrite (Forest City). 87 TABLE 6 - 3 . ISOTOPIC COMPOSITION OF GD IN THE ABEE METEORITE Chemical f r a c t i o n Gd 155 * Gd I"&~o Gd 157 * Gd Gd 158 * Gd T6~0 B Eu-AB t r i p l e f i l . .71614 +.00024 1.13533 ±.00035 1.68987 +.00065 Gd-AB t r i p l e f i l . ,..71607 1.00032 1.1351 3 . 0 0 1 1 1.6878 ±.0036 GE-AB .67652 .71598 s i n g l e f i l . +.00024 +.00012 zone-ref. Re 1.13557 + .OOO31. 1.36092 +.00044 1.68991 +.00058 Average .67652 .71602 1.13545 I .36092 1.68986 ( t h i s work) +.00024 +.00010 +.00023 +.00044 +.00043 T e r r e s t r i a l .67688 .71601 1.13576 I .36021 I .69060 ( t h i s work) +.00008 +.00009 +.00012 +.00016 +.00034 * F r a c t i o n a t i o n c o r r e c t i o n normalized to Gd^"^/Gd"^° = .9361. a The f r a c t i o n Eu-AB contained most of the Gd as w e l l as the Eu from the o r i g i n a l 9 .5 g Abee sample. There was no s p e c t r a l i n t e r f e r e n c e between Eu+ and Gd+ i o n s . b The f r a c t i o n Gd-AB contained v e r y l i t t l e Gd (estimated to be < 0.1p«.g on the b a s i s of the observed Gd+ beam i n t e n - s i t y ) . c Not a l l o f the Eu-AB and Gd-AB__fractions were loaded f o r the f i r s t two analyses above. The remainder, approximately 20% of f r a c t i o n Eu-AB and 30% of f r a c t i o n Gd-AB, was combined on a s i n g l e ( z o n e - r e f i n e d ) rhenium f i l a m e n t f o r a n a l y s i s GE-AB. See the f o o t n o t e s to Table 6-1 f o r d e f i n i t i o n s of A and B. 88 FIGURE 6-3. CORRELATION. OF GD ISOTOPES IN METEORITES Gd Gd 158 * 150 1.1370 1.1365 1.1360 1.1355 1.1350 1.1345 .7140 F o r e s t C i t y Pasamonte Weekeroo S t a t i o n T e r r e s t r i a l ( t h i s work) Copiapo T h e o r e t i c a l c o r r e l a t i o n l i n e f o r thermal neutron capture (see Appendix IV) ± G d ^ .7150 7160 .7170 .7180 G d 1 5 8 / G d l 6 ° * vs G d l 5 7 / G d l 6 ° * for.Abee ( t h i s work) and pr e v i o u s m e t e o r i t e analyses (Eugster e t a l , 1970a). * F r a c t i o n a t i o n c o r r e c t i o n n o r m a l i z e d to G d ^ ^ / G d ^ ^ ^361, 89 FIGURE 6-4. CORRELATION BETWEEN A AND B IN METEORITES A Gdl57 l G dl60 J I.696 1.694 1.692 1.690 1.688 1.686 1.684 F o r e s t C i t y Weekeroo S t a t i o n Norton County I - Pasamonte 1 -a— Abee T e r r e s t r i a l ( t h i s work) Copiapo T h e o r e t i c a l c o r r e l a t i o n l i n e f o r thermal neutron capture (see Appendix IV) A - G d 1 5 6 , G d 1 5 6 , — \ — i 6 o ) JL J _ G d 1 5 5 Gd -1 > 1.358 1.359 1.360 1.361 1.362 C o r r e l a t i o n between i r r a d i a t i o n - s e n s i t i v e parameters B and A f o r Abee ( t h i s work) and p r e v i o u s m e t e o r i t e analyses (Eugster e t a l , 1970a). 90 6.3 Do Abee and T e r r e s t r i a l Gadolinium have S i m i l a r Composition? The apparent difference between the values of A and B for the Abee and t e r r e s t r i a l samples may be significant. It cannot be explained by thermal neutron capture alone. Nor can it be the result of mass fractionation during the chemical preparation of the Abee sample, provided the fractionation process was governed by a simple mass dependence. It is possible that the Abee meteorite was derived from a region where the relative abundance of the nuclides was slightly dif- ferent f rom that of p r e - t e r r e s t r i a l matter. Apart from meteorite analyses, the only other Gd isotopic studies on e x t r a - t e r r e s t r i a l samples are the lunar analyses of Eugster et al (1970b) and Lugmair (1970). Their lunar and t e r r e s t r i a l data c l e a r l y lie along the c o r r e - lation line for thermal neutron capture (see Figure 1-4). This is to be expected if the earth and moon are genetically related, and were produced from the same pool of nuclides. The meteorites may, however, have been produced from slightly different source m a t e r i a l . A l t e r n a t i v e l y , the apparent difference between the t e r r e s t - r i a l and Abee values of A and B may be the result of poor s t a t i s t i c a l estimates of the standard deviations for the Abee data, or they may result from additional sources of e r r o r . There are three possible sources of e r r o r which may have introduced a bias in the measurement of Abee Gd relative to t e r r e s t r i a l 9 1 Gd.- Trace amounts of one or more int e r f e r i n g ions may have been.superimposed upon the Gd or GdO spectra. The most probable ions are B a F + , B a C l + , L a C l + and LaC>2+ . None of these ' alone would give the observed anomaly, however, and no systematic 155 15 8 lAf) '* change in any of the Gd /Gd ratios was observed. Nevertheless, the p o s s i b i l i t y of ion interference cannotbe ruled out completely. A second p o s s i b i l i t y is that the baseline was not constant over the mass range 155-160 (or 171-176 for G d O + ions). Since baselines were measured only at the ends of each spectrum (see Section 3.4), the existence of any baseline i r r e g u l a r i t y within the sp e c t r a l range would not have been detected. Baselines were routinely checked between peaks at the beginning of each analysis, but distortion of up to 0.1 to Q..2% would not have been detected during a v i s u a l reading of the d i g i t a l voltmeter. The most probable causes of baseline distortion are secondary electrons and/or scattered ions (most l i k e l y Ba^) in the v i c i n i t y of the ion detector. Baseline distortion was a major source of e r r o r for Sm* analyses (see Section 6. 5); it was attributed to B a + ions. However, the mass range of Sm is much c l o s e r to that of Ba; dis t o r t i o n from B a + ions should be significantly less in the higher Gd mass range. A l s o , Gd was vapourized at higher filament temperatures than Sm; . i t was therefore possible to burn off more of the r e s i d u a l Ba before 9 2 producing a G d + ion beam. Although baseline distortion may be a source of e r r o r during Gd analyses, it does not explain the s i m i l a r isotopic ratios obtained for Abee when using the Gd* and GdO* spectra. A t h i r d source of e r r o r is pressure scattering in the mass spectrometer. This creates a slight t a i l on either side of the spectral peaks. D i r e c t measurement of peak prof i l e s showed that the height of the t a i l at a position \ mass unit above or below a spectral peak.was < 0.1% of the peak height for operating pressures < 1.0 x 10 t o r r . The corresponding t a i l height 1 mass unit away was < 0. 03%. These e r r o r s represent a significant bias in measurement which v a r i e d with the operating pressure. The maximum e r r o r s , which could be attributed to pressure scattering, in the ratios G d 1 5 5 / G d l 6 ° * , G d 1 5 7 / G d 1 6 0 * , G d 1 5 8 / G d l 6 ° * , A and B were 0.01%, 0.04%,0.Q3%, 0.05% and 0.01% respectively. The slope of the e r r o r line associated with pressure scattering would be s i m i l a r to that of a line passing through the Abee and Copiapo data points i n Figures 6-3 and 6-4. The intersections of these e r r o r lines with the theoretical c o r r e l a t i o n lines for thermal neutron capture are below the t e r r e s t r i a l points. This would suggest, if anything, that Abee was exposed to fewer thermal neutrons during its h i s t o r y than the earth was. However, the most sensitive parameter B would not change by more than 0.01%, and the difference 9 3 between the mean t e r r e s t r i a l and Abee values would not exceed 0.06%. 6.4 Interpretation of Europium Data When the present study of the rare earths was initiated, the extent of mass spectrometer fractionation was not known for the tri p l e filament technique. The magnitude of the fractionation effect (i.e. the deviation from the absolute ratios) was less than that which was observed with the single filament technique, but the va r i a t i o n during an analysis was the same for both techniques: approximately 0.3% v a r i a t i o n per unit mass difference (see Figur e 5-3). Since Eu has only two isotopes, a simple normalization process could not be used. Table 6-4 shows the mean values obtained for a l l analyses on t e r r e s t r i a l (Eu-J and Eu-US) and Abee (Eu-AB) samples. Two selective methods of combining the data decreased the d i s p e r s i o n due to fractionation, but it was not completely e l i m i - nated. The f i r s t method combined data which was obtained for the i n i t i a l 10-20% of the sample (based on the time-integrated beam intensity). The second method used only the results of analyses for which the data-taking intervals were uniformly distributed over the lif e t i m e of the E u * ion beam. Recording data for the f u l l duration of the sample minimized bias due to fractionation and provided the most r e l i a b l e estimate of the absolute isotopic ratio for each sample. It is apparent that there is no gross difference between " t e r r e s t r i a l and Abee Eu. The slightly higher ratio for the Abee sample TABLE 6-4. ISOTOPIC COMPOSITION OF TERRESTRIAL AND ABEE EU Average Ev?~^/Eu^^ i s o t o p i c r a t i o D a t a Eu-J Eu-US Eu-AB s e l e c t i o n ( B r a z i l ore) (U.S. ore) (Abee) Average f o r a l l .91? (20) .914 (20) .918 (38) an a l y s e s 1.001 1.004 1.002 Average f o r s e t s .9182 (5) .9183 (4) .9195 (24) obtained w i t h i n i t i a l 1.0006 1.0010 1.0004 10-20% of sample on l y # Average f o r analyses .9155 (9) .9160 (9) .9173 (11) f o r which data was 1.0009 1.0018 1.0022 obtained f o r the f u l l d u r a t i o n of the sample E r r o r s correspond to one standard d e v i a t i o n of a s e t . The standard d e v i a t i o n of the mean would not be a s u i t a b l e r e p r e s e n t a t i o n o f the e r r o r because i t would not account f o r r e a l v a r i a t i o n s i n the i s o t o p i c r a t i o i n the E u + i o n beam due to mass f r a c t i o n a t i o n . The numbers i n br a c k e t s r e p r e s e n t the t o t a l number of s e t s ( u s u a l l y 6 scans per s e t ) used to c a l c u l a t e the means and standard d e v i a t i o n s . Murthy e t a l (1970) found t h a t E u 1 5 1 / E u 1 5 3 = .9147 f o r s i n g l e f i l a m e n t a n a l y s e s on t e r r e s t r i a l europium. # The a n a l y s i s of the Eu-AB f r a c t i o n extended over three days. Since the mass spectrometer was turned o f f over- n i g h t , sample em i s s i o n was not continuous. This appears to have i n t r o d u c e d an e r r o r i n de t e r m i n i n g the sample d e p l e t i o n from the t i m e - i n t e g r a t e d beam i n t e n s i t y . 95 probably reflects a bias in measurement: most data sets for analysis Eu-AB were obtained with the i n i t i a l portion of the sample. In view of the large va r i a t i o n caused by fractionation (0.6%) during the analyses, no significance can be assigned to the apparent difference of 0.14 - 0.17% between t e r r e s t r i a l and Abee Eu. 6. 5 Interpretation of Samarium Data The combined results of several analyses of samples Sm-J and Sm-US are recorded in Table 6-5. Each entry is the weighted mean of a minimum of six sets of scans of the S m + spectrum. The two t e r r e s t r i a l samples have the same isotopic composition within the experimental uncertainty (shown as two standard deviations of the mean). The average t e r r e s t r i a l value was calculated for each ratio i n turn from equations (6-1). A comparison between t e r r e s t r i a l and Abee Sm is shown in Table 6-6- Although the i r r a d i a t i o n - s e n s i t i v e parameter D appears to have the same value for both samples, s e v e r a l of the isotopic ratios appear to differ. The apparent difference cannot be attributed to thermal neutron capture. The discrepancy between the t e r r e s t r i a l and Abee results is almost cer t a i n l y due to baseline distortion. A depression of the baseline in the mass range 144 - 147, and a moderate enhancement at higher masses, was c l e a r l y observed in association with an intense B a * ionbeam. However, the c h a r a c t e r i s t i c shape and T A B L E 6-5, ISOTOPIC COMPOSITION OF SM IN T E R R E S T R I A L S A M P L E S Sample S r n ^ S m 1 4 8 * Sm 1 4? * S m 1 5 0 * S m 1 ^ * Sm^_° S m 1 5 ° * S m 1 5 2 S m 1 5 2 ,Sm152< S m 1 5 2 S m 1 5 2 . ° " s m 1 4 ? ( S m 1 5 2 > # Sm-J .11647 .42323 .51943 .27725 .84836 .28106 ( B r a z i l ore) 1.00015 + 00013 t.00013 t. 00009 + 00026 + 00014 Sm-US # .11689 .42322 .51936 .27728 .84852 .28096 (U.S. ore) -.00031 +.00025 +.00031 +.00023 +.00080 +.00040 # Average .11655 .42323 .51942 .27725 .84838 .28105 +.00014 +.00012 . +.00012 +.00008 + 00025 +.00013 Murthy et .118 .426 .522 .280 .858 .283 a l (1963) 147 152 x- Fractionation correction normalized to Sm /Sm = .5650 (mean value for a l l t e r r e s t r i a l analyses for this work). # Baseline distortion was recognized to be a source of bias in measurement for a l l of these analyses. This was substantiated by comparing the values obtained for the parameter D when using the two spectral ranges 149-152 and 144-154. In both cases the baseline was measured at positions \ mass above the high mass peak and \ mass below the low mass peak only. No subsequent correction for the distortion was possible without having measured baselines between intermediate peaks. T A B L E 6-6,. ISOTOPIC COMPOSITION OF SAMARIUM IN THE A B E E M E T E O R I T E Sm 1 4 4 * S m 1 4 8 * Sm 1 4? * S m 1 5 0 * S m 1 5 4 * S m 1 5 0 S m 1 5 0 * , c 152 _ 152 _ 152 „ 152 c 152 D = _ 149 ( c 152 ) Sample bm Sm om Sm Sm Sm Sm A b e e * .11585 a .42284 a .51871 .27709 a .84976 b .28116 a (Sm-AB) 00009 t. 00016 +.00022 +.00017 +.00061 +.00023 T e r r e s t r i a l * .11655 .42323 .51942 .27725 .84838 .28105 +.00014 +.00012 +.00012 +.00008 +.00025 +.00013 jfc Fractionation correction normalized to Sm /Sm = . 5650 . 138 # Baseline distortion (mainly secondary electrons from Ba ions) is the probable cause of the major discrepancies between Abee and t e r r e s t r i a l samples. The distortion was greatest in the lower mass range 144-147. a After computationally subtracting the Nd+ spectrum which was monitored at mass 146. (Nd+/Sm+< 0.02). b Af t e r computationally subtracting the LaO+ spectrum which was monitored at mass 155. (LaO+/Sm + < 0.005). 98 magnitude of the dis t o r t i o n were highly variable from one analysis to another, so that a c o r r e c t i o n could not be made without having measured baselines between individual peaks. The good agreement • obtained for the two t e r r e s t r i a l samples is probably a r e s u l t of their s i m i l a r chemical composition. Baseline dis t o r t i o n introduced a bias in measurement, but it was approximately the same for both samples. The existence of baseline distortion was proven by analyzing t e r r e s t r i a l Sm samples using only the shorter mass range 149-152, and computing the parameter D. This was observed to have the values .28038 + .00019 for Sm-J and .28036 + .00010 for Sm-US. The difference between these values, which are undoubtedly c l o s e r to the true value, and the average value in Table 6-6 is significant. It can only be explained by the fact that the baselines were measured at different places for the two s p e c t r a l ranges. The existence of baseline distortion would lead to a further source of e r r o r for the Abee analyses. A c o r r e c t i o n for the Nd + beam was made by monitoring the intensity of the peak at mass 146. Since baseline distortion was greatest in the v i c i n i t y of this peak, the intensity of the N d + ion beam would not have been c o r r e c t l y determined, and a subsequent e r r o r would result wherever isobars of Nd and Sm occurred. The largest e r r o r for Sm would occur at mass 144. . 99 6. 6 Conclusions An investigation of Gd, E u and Sm in the Abee enstatite chondrite, and in two t e r r e s t r i a l ores, has shown that there are no significant isotopic anomalies for Abee which could be attributed 155 to one or more of the thermal neutron capture processes: Gd / X \^^ 1 5 6 r-J51i )C \ r ^ l 5 8 c 149, y xc 150 „ 151, y . (n, 0 )Gd , Gd (n, 0 )Gd , Sm '(n, ( )Sm , E u (n, fl ) ^ 152 „ 153, Y \ i r 154 . . . „ ,155,-,156 Eu , Eu (n, I )Eu . The isotopic ratios Gd /Gd , G d 1 5 7 / G d 1 5 8 , S m 1 4 9 / S m 1 5 0 and E u 1 5 1 / E u 1 5 3 are identical for the three samples studied within maximum experimental uncertainties of 0.15%, 0.10%, 0.3% and 0.3% respectively. These estimates include s t a t i s t i c a l e r r o r s , as we l l as e r r o r s introduced by possible sources of bias in measurement (e.g. baseline dist o r t i o n , pressure scattering and/or uncorrected fractionation). On the basis of the G d ^ 7 / G d ^ 8 r a t i o , we may conclude that the maximum di f f e r e n t i a l thermal neutron i r r a d i a t i o n between the t e r r e s t r i a l and Abee samples is 3 x 10 _ neutrons/cm for the uniform i r r a d i a t i o n model of Fi g u r e 2-2. It is reasonable to assume that this conclusion applies to the Abee meteorite as a whole, when compared with the earth. In terms of the i r r a d i a t i o n models discussed in Chapter 2, we may exclude the regions above the curves in Figures 2-1 and 2-2 corresponding to a G d ^ 7 / G d ^ 8 anomaly ^ 0.1% from the family of possible solutions for the i r r a d i a t i o n ... hist o r i e s of the Abee meteorite and the earth. 100 If the time - integrated thermal neutron flux associated with the i r r a d i a t i o n phase was as large as the F G H hypothesis pre- 21 2 157 158 dieted ( i . e. \L r 4 x 10 neutrons/cm ), then the Gd /Gd nE ratio is more sensitive to the fraction of the m a t e r i a l which was i r r a d i a t e d than to the di f f e r e n t i a l i r r a d i a t i o n of the p r i m i t i v e chondritic and t e r r e s t r i a l matter (Fowler et a l , 1962; Burnett et a l , 1966; Murthy et a l , 1963). The absence of a Gd anomaly >• 0.1% implies that the i r r a d i a t e d fractions of the Abee and t e r r - e s t r i a l source m a t e r i a l were identical within 2%. This applies to the F G H model only, where it has been assumed that approxi- mately 5% of the p r i m i t i v e t e r r e s t r i a l matter was irradiated. Apart from uniform i r r a d i a t i o n and/or dilution of the source m a t e r i a l , alternative explanations could account for the 157 158 absence of significant Gd /Gd anomalies: efficient shielding of most planetary m a t e r i a l inside large bodies having diameters » 50 meters; the lack of sufficient hydrogen to thermalize spallation-produced neutrons; or the absence of an i r r a d i a t i o n phase of sufficient intensity to generate detectable anomalies. Apart from the highly reduced state of the Abee meteor- ite, there is no clear evidence to suggest that this e x t r a - t e r r e s t r i a l object originated in the inner region of the solar system. If any- thing, the isotopic composition of Gd in the meteorite suggests that Abee received less i r r a d i a t i o n than the earth during its previous 101 history. This evidence would therefore favour the view that Abee had its o r i g i n at a greater distance from the sun, possibly in the asteroid belt. Since the asteroid belt is a natural source of meteorites, any explanation of the gradation in the oxidation states of chondrites would receive stronger support from the available evidence if it assumed an o r i g i n i n this region of the solar system. One i n t e r - esting p o s s i b i l i t y which has received considerable attention recently (see L a r i m e r et a l , 1970) is that variations in the oxidation state of chondrites result from differences in their spatial distribution during gravitational settling of p r i m o r d i a l m a t e r i a l toward the o r b i t a l plane of the nebula. 102 A P P E N D I X I E F F E C T OF NEUTRON C A P T U R E ON GD, SM AND E U The probability of a nuclear process is generally expressed in terms of a cross section \T~ which has the dimen- sions of an area. The unit commonly used is the barn, which is - 2 4 2 defined as 10 cm . A simple definition of the thermal neutron capture cross 157 section of a specific nuclide (e. g. Gd ) may be deduced from c l a s s i c a l theory. Consider a b eam of thermal neutrons s t r i k i n g a thin target in which the beam is attenuated only i n f i n i t e s i m a l l y . If R^ is the number of capture processes o c c u r r i n g in the target per unit time for nuclide i then the corresponding reaction cross section is defined by the equation R. zz In. q ~ d 1 1 i where I is the number of incident thermal neutrons per unit time n^ is the number of target nuclei of nuclide i per unit volume <jr" is the thermal neutron capture cross section for nuclide i d is the target thickness. In this thesis we are interested in thermal neutron capture processes o c c u r r i n g in the p r i m i t i v e m a t e r i a l of the solar system. The size of the planetesimals, the duration of the assumed i r r a d i a t i o n period, and the magnitude of the neutron flux are a l l unknown. 103 However, for heavy isotopes with large thermal neutron capture cross sections it is possible to establish a clear relationship between isotopic composition and the time-integrated thermalneutron flux. Let N Q be the i n i t i a l atom abundance of a pa r t i c u l a r nuclide in p r i m o r d i a l m a t e r i a l which later accreted to form a planetary- body (e. g. the earth or the parent body of the Abee meteorite). The quantity N Q represents the total number of atoms of the nuclide p r i o r to i r r a d i a t i o n . If thermal neutron capture is the dominant process causing a change in the abundance of the nuclide during i r r a d i a t i o n , then the atom abundance INL after i r r a d i a t i o n by n neutrons may be deduced from the di f f e r e n t i a l equation (after Fowler et a l , 1962) (1-1) D N N A A where is the thermal neutron capture cross section for the nuclide under consideration f is the fr a c t i o n of the incident neutrons which is captured by the i r r a d i a t e d m a t e r i a l . The remaining fraction ( l - f r ) includes those neutrons which escape from the i r r a d i a t e d m a t e r i a l or undergo ^ decay. ^ CTT N represents the total cross sectional target area presented A A by a l l nuclides. 104 L e t n t o t a 2 be the total number of neutrons produced during i r r a d i a t i o n . Then, integration of equation (I-l) yields N f - N o e x p or A n r total A A A (1-2) where N.̂  is the f i n a l atom abundance of the nuclide i which has been depleted by neutron capture. So f a r , no mention has been made of the energy distribution of the neutrons. In the thermal range, the neutron capture cross sections of most nuclides are inv e r s e l y proportional to the velocity of the incident neutrons. Hence, it is evident from equation (1-2) that the velocity terms would cancel out, leaving essentially independent of the neutron energy distribution, provided it is in the thermal range. F o r the purposes of this thesis, however, we have chosen to separate theexponent in equation (1-2) into two factors C . and"!/' . We have chosen to be the cross section for ^ i 7 n i neutron capture, assuming a thermal neutron velocity of 2200 m/sec (or 20°C). The parameter is defined as f n r total A A A 20°C (1-3) This is simply the time-integrated thermal neutron flux (at 20°C) 105 for a l l neutrons which undergo capture. We may now write equation (1-2) in the form N - N = - N 1 o o £ l-exp(- ^~ Ifn J (1-4) By introducing another parameter, the dilution factor (after Fowler et a l , 1962), we may account for the p o s s i b i l i t y that only a fraction of the p r i m o r d i a l m a t e r i a l was irradiated. The remainder, was effectively shielded within planetesimals having dimensions of the order of one meter or greater. Subsequent mixing of the two fractions during accretion to form a l a r g e r planetary body would y i e l d N A N = N,-N i o " d - > [ l - e * P ( - ^ . V>n>] ^ Equation (1-5) gives the net decrease in the abundance of a nuclide undergoing neutron capture. F o r (n, \ ) reactions (which includes a l l reactions on Gd, Sm and Eu under consideration here), equation (1-5) also gives the net increase in the neighbouring higher mass nuclide resulting from this process. Let R q and R be the i n t i t i a l and f i n a l isotopic abundance ratios for masses m and m + Then 106 R r tz> R = N N _ ( N IT ) f l - e x p ( - \ b ) "1 i m om om d «- m f n ' J N f , m + 1 N o ^ + l+^rn^L 1 -^"^^] F d V R o L^-T^VnO F d + R 0[l-exp ( -^ m y n ) J d-6) 155 Equation (1-6) may be applied to each of the isotopic ratios Gd / . 156 157 158 149 150 Gd , Gd /Gd and Sm /Sm . Two independent (n, g ) 151 153 reactions may contribute to a change in the Eu /Eu ratio (see Table 1-1). Therefore, for this ratio we have F d V R o I>exp(- ^ T s i ^ l F d - [ l - e x p ( - <r i 5 3V„>] ( I" 7 ) It is evident from equations (1-6) and(T-7)that each of the present- day isotopic ratios mentioned above may be regarded as a function of R , F and "\U . The cross sections are a l l known. Let o d R g and R ^ be the present-day values of aparticular ratio in t e r r e s - t r i a l ores and in meteoritic m a t e r i a l respectively. If these materials were exposed to previous i r r a d i a t i o n by integrated neutron fluxes oi^)U and "7/' .where the dilution factors were F and F " n E H n M dE dM respectively, then 107 R(R ,F ) = R o dE ' nE E (1-8) R(R ,F , 7 / : ) = R o dM * nM M (1-9) In equations (1-8) and (1-9) it has been assumed that the i n i t i a l isotopic ratios were the same for both p r e - t e r r e s t r i a l and pre- meteoritic m a t e r i a l s . Consider the case where it is just possible to detect a difference ofo.l% between the t e r r e s t r i a l and meteoritic isotopic ratios R^. and R ^ respectively. The lower l i m i t of detection is given by R -R M E R E o dM nM R(R , F o dE VnE> = .001 ( r: 0.1% detectable difference) (1-10) Figures 2-1 and 2-2 il l u s t r a t e p a r t i c u l a r solutions of equations (1-8) and (1-10) for specific values of the dilution factors. The curves for Figur e 2-1 represent the solution of the equations E (1-11) ( = .632 for the G d 1 5 7 / G d 1 5 8 ratio) 108 and R ( R ,20, 1/ X J - R ( R ,20, " l / .„) o « n M o • n £ R ( R o ' 2 0 ' V n E > = .001 (1-12) The parameter R q was eliminated from equations (I-11) and (1-12), and the solution curve for *^ n£ and Y n M w a s P i t t e d in the form | 'VnM-VnEi 'VnEl V S YnE ' 157 158 The sensi t i v i t y of the Gd /Gd ratio to di f f e r e n t i a l i r r a d i a t i o n of pre-meteoritic and p r e - t e r r e s t r i a l m a t e r i a l is i l l u s t r a t e d for the case F • •= F ,, , r: 1 in Figure 2-2. The dE dM 5 requirement that ^ 0*^ '0, combined with the present-day ratio R =r .632, places an upper l i m i t on the value of 'Xf' which w i l l E * nE satisfy the equation R(R ,1, 1> ) = .632 (1-13) o 1 n E This is i l l u s t r a t e d in Figure 2-2. 109 A P P E N D I X II SECOND ORDER FOCUSSING SHIMS Increased se n s i t i v i t y , without loss of resolution m aybe achieved by ."imp:roving the focussing properties of the analyzer magnet to permit the use of an ion beam with l a r g e r angular divergence. F i r s t order focussing results from using plane surfaces at the pole faces where the ion beam enters and leaves the magnet (at normal incidence and emergence). By suitably contouring one or both of these surfaces it is possible to produce a sharper focus at the collec t o r . A variety of methods for achieving second order focussing have been reviewed by B a r n a r d (1953). One simple method involves the use of c y l i n d r i c a l rather than plane surfaces for the magnet pole faces. F o r a symmetric 90 deg sector magnet, s i m i l a r to the one used by the present w r i t e r , the ideal radius of curvature of the pole faces is identical to the or b i t a l radius of the central ion tra j e c t o r y within the magnet. An alternative procedure adopted by the w r i t e r is to use one plane surface and one c y l i n d r i c a l surface, where the latter has a radius equal to one half the o r b i t a l radius (Figure II-1). Two c y l i n d r i c a l shims were designed to fit onto the existing pole face on the collector side of the main electromagnet. The shims were positioned on either side of the analyzer tube, and provision was made for adjusting them to enable precise focussing * 1 a=34 . l i.2 ' > COLLECTOR ( m e a s u r e d ) ^ s m c E APEX ; t " a - .9±.2 ' s p e c i f i c a t i o n ) #/ i o n t r a j e c t o r y w i t h i n i t i a l angle . l ^ V • . a t o the c e n t r a l r a y ioi.01 / ' / • , M y / J^\> 10- 01 (computed) . - /TV antral rav . s f \ ,' \ . . . (̂ computed) shim p o s i t i o n i n g \ (compute 3' +- . .» ' X NOT DRAWN TO SCALE screw FIG. I I - l MODIFICATION OF MAIN ELECTROMAGNET TO ACCOMMODATE FOCUSSING SHIMS The d o t t e d c o n t o u r shows the o r i g i n a l magnet p o s i t i o n . By moving the main magnet, and ad d i n g two c y l i n d r i c a l shims (7.0 i n c h r a d i u s ) the pa t h of the c e n t r a l r a y remains unchanged. A l l d i s t a n c e s are s p e c i f i e d i n i n c h e s . I l l experimentally. No alteration of the main magnet was necessary except for repositioning to compensate for the additional thickness of the magnet shims. In order to apply a c o r r e c t i o n for the effect of the fringing magnetic fields on both sides of the magnet, the f i e l d intensity was determined in the median plane along the path of the central ray. The theoretical method of Coggeshell (1947) was used to calculate the f i e l d close to the magnet, and measured f i e l d values were employed at distant points. The resulting contour of the fr i n g i n g f i e l d is shown in Figure II-2 . While the magnet was in the position of best focus for f i r s t order focussing, the parameters a and d (see Figure II-1) were measured. The large e r r o r of .2 inch in each measurement reflects the uncertainty in determining the effective positions of minimum beam width at the source and collector. Taking into account the uncertainties in a and d, the trajectory equation was integrated through the known magnetic f i e l d , and the radius of cur- vature of the central ray was computed to be r = 12.1 -.1 inch. T r a j e c t o r i e s were computed for several ions having i n i t i a l angles c inclination to the central ray in the range -4° <.©<'< 4°. F o r eac tr a j e c t o r y the required change i n path length (within the magnetic field) which would bring the traj e c t o r y to perfect focus at the co l l e c t o r , was computed. This led to the determination of the 112 Relative magnetic f i e l d intensity 1.0 .3 0. 0 T h e o r e t i c a l f i e l d values Smooth tra n s i t i o n between theoretical and experimental f i e l d values Magnet pole face Experimental f i e l d values JL JL_1 JL 0 10 2 4 6 8 Distance from pole face (inches) F IGURE I I - 2 . FRINGING F I E L D OF MAIN MAGNET 12 113 contour for the magnet surface which would give perfect focus at the collector (see Figure II-3). A least squares c i r c l e (in the sense of min i m i z i n g T ~ ^ t - A t , J ) was computed through the surface points to determine the best radius associated with ion beams of varying angular divergence ot' (Figure II-4). F o r values of the m six angular divergence exceeding 2 deg it is apparent that 7.0 inches is the ideal curvature for the magnet shims. This may be compared with the value of 6.0 inches where the effect of fringing fields is neglected (see Figure II-4). The magnet shims were designed to have a c y l i n d r i c a l surface of 7.0 inch radius. They were accurately machined from grade 1010 hot r o l l e d carbon steel plate obtained fromDr. Auld of the T R I U M F group at U. B. C. P r o v i s i o n was made for holding the shims r i g i d l y in position on the main magnet pole faces, although adjustment of their position along the pole face was possible A f t e r mounting the shims the magnet was positioned as shown in Figure II-1. P r e c i s e allignment was then achieved ex p e r i - mentally by adjusting the shim position and the v e r t i c a l position of the magnet to give optimum peak shape. Figure II-5 shows the peak quality obtained for beams of angular divergence 1.9 and 3.0 deg respectively. Although the resolution was excellent (400-450), the higher _ angular divergence, coupled with wider source s l i t s , gave rounded Required reduction in path length (in the magnetic field) to give perfect focus at the collector A A t (inches) FIGURE II-3. CURVATURE OF SECOND ORDER FOCUSSING SHIMS Ideal curvature of shims assuming (i) no correction for the effect of fringing fields outside the main magnet (represented by dots) or (ii) a correction applied to account for the fringing fields (represented by crosses). Note the different scales used for the two axes. When A t and A z are plotted on the same scale the least squares ci r c l e s f i t t i n g the two sets of points have r a d i i of 6.1 and 7.0 inches respectively (using data for = 0, + . 2, t .4, ... ± 4.0°). 115 Optimum radius of curvature (inches) t 7.0 - 6.5 6.0 X / / / -Jj C o r r e c t i o n applied for fringing f i e l d No c o r r e c t i o n applied max Maximum angular divergence of ion beam (degrees) F IGURE II-4. O P T I M U M C U R V A T U R E OF SHIMS AS A FUNCTION OF B E A M D I V E R G E N C E The scatter of the data points is due to truncation e r r o r s in the calculations and the use of r e l a t i v e l y few tr a j e c t o r i e s . Each point represents the radius of curvature of the least squares c i r c l e fitting through a l l values of A t and d.z corresponding to tra j e c t o r i e s with i n i t i a l angles of inclination to the lens axis of < = 0 , +0.2, + 0 . 4 , +0.6, . . . + max 206 208 207 n 208 li L n FIGURE II-5. P E A K SHAPE FOR TWO V A L U E S OF THE ANGULAR DIVERGENCE 206 207 n i—J 208 206 ri 207 A u (b) CK =• 3.0 des max Part of the Pb spectrum is shown for each value o f ^ ^ , The second order focussing shims were used to obtain the above spectra. 1 1 7 peak tops and a slight assymmetry to the peaks. The lower divergence (corresponding to two 0.010 inch source collimating s l i t s separated by 0.31 inch) was adopted in the present research. •This is s t i l l a factor of two greater than the divergence used with only f i r s t order focussing. The effect of angular divergence on beam width at the collector is shown for f i r s t and second order focussing i n Figure II -6. \ 1 1 8 FIGURE I I - 6 . B E A M SPREADING AT THE C O L L E C T O R FOR FIRST AND SECOND ORDER FOCUSSING Beam spreading at the collector due to beam angular divergence at ion source max 0 . 1 2 3 4 Beam divergence at exit s l i t of ion source (degrees) The effect of angular divergence ( ̂ m a x ) at the ion source exit s l i t on the beam width at the collector is shown for both f i r s t and second order focussing. The scatter of data points for the case of second order focussing reflects the s m a l l number of tra j e c t o r i e s used in the calculations ( = 0 , + 0.2, + 0.4, .. .+o^ ). J x ' ' . max' 1 1 9 A P P E N D I X III M E A SURED ISOTOPIC RATIOS FOR A L L GADOLINIUM A N A L Y S E S The experimental values of the isotopic ratios obtained for each set of scans of the G d + (or GdO +) spectrum are l i s t e d i n Tables I I I - l to III-8. The isotopic ratios G d 1 5 5 / G d l 6 ° , G d 1 5 7 / r ^ 1 6 0 A ̂ j ! 5 8 / r , ,160 , • n u i - A ^ i 1 5 6 / ^ ^ 1 6 0 Gd and Gd /Gd have a l l been normalized to Gd /Gd = .9361 to eliminate the effect of mass dis c r i m i n a t i o n . A co r r e c t i o n was made for the isotopic composition of oxygen in a l l analyses which used the G d O + spectrum (Tables 111-3, III-4 and III-8). The procedure for making this c o r r e c t i o n is given in Section 5. 5. The estimated e r r o r s i n Tables III-1 to III-8 correspond to two standard deviations of the mean. F o r each set of scans the isotopic ratios and the parameters A and B were computed by the method discussed in Sections 5. 3 and 5. 4. The weighted mean and its standard deviation were computed for the data in each column of Tables III-1 to III-8 from the equations i = l , n <H i = l,n CH' i n l . n * Y (- - ^ ) 2- \ (n - 1) i = l , n (III-l) 1 2 0 In these equations, and Ĵ". represent the mean value of a p a r t i c u l a r isotopic ratio (or parameter), and the estimated standar deviation of the mean, for set i .. These computations were performed using one more significant digit than has been shown in the tables. The entries in the bottom row of each of the Tables I I I - l to III-8 are the values of JA_ and 2 given by equations (III-l). These quantities are the data which appear in Tables 6-1 and 6-2. TABLE I I I - l . TRIPLE FILAMENT ANALYSES ON SAMPLE Gd-J A n a l y s i s Set No. of scans G d 1 5 5 * G d 1 6 0 G d 1 5 7 * G d 1 6 0 G d ^ 8 * G d 1 6 0 A B Gd-Jl 1 6 . 6 ? 6 l +.0008 .7161 ±.0009 1.1363 ±.0026 1.3615 ±.0012 1.6916 ±.0038 Gd - j 4 # 1 7 .6767 +.0004 .7162 ±.0003 1.1367 ±.0017 1.3606 +.0009 1.6922 +.0034 2 6 .6764 ±.0007 .7166 +.0008 1.1358 +.0014 1.3611 ±.0013 1.6893 +.0019 3 6 .6777 ±.0021 .7154 ±.0015 1.1348 ±.0012 1.3586 ±.0042 1.6898 ±.0026 Gd-J5 1 6 .6?67 ±.0013 .7161 ±.0010 1.1355 ±.0017 1.3606 ±.0026 1.6898 ±.0032 2 6 .6769 ±.0002 .7161 ±.0003 1.1364 ±.0005 1.3602 ±.0003 1.6917 ±.0008 3 6 , .6767 _.ooo6 .7165 ±.0005 1.1365 +.0012 1.3606 +.0013 1.6911 ±.0023 4 6 1.6902 ±.0010 Average .67686 .71620 I . 13620 I .36030 I . 6 9 O 8 3 ±.00015 ±.00016 ±.00044 ±.00027 ±.00067 * F r a c t i o n a t i o n c o r r e c t i o n c o r r e c t e d to Gd?-^^/Gd}-^= .9361 # Analyses Gd-J2 and Gd -J3 were r e j e c t e d because of B a F + i n t e r f e r e n c e w i t h the Gd+ spectrum. This was caused by r a i s i n g the s i d e f i l a m e n t temperature too q u i c k l y without a l l o w i n g s u f f i c i e n t time f o r Ba to decay a t lower tempera- t u r e s . The parameters A and B are d e f i n e d i n Table 6-1 . 1 2 2 TABLE I I I - 2 . TRIPLE FILAMENT ANALYSIS Gd-j6 Gd-J6 Set No. of scans G d 1 5 5 * G d 1 6 0 G d 1 5 7 * G d i 6 ° G d 1 5 8 * G d 1 6 0 A B 1 6 .6782 +.0011 .7172 +.0009 1.1361 +.0004 1.3573 +.0021 1.6884 1,0023 2 5 .6?68 +.0013 .7165 ±.0008 1.1358 ±.0012 1.3603 ±.0025 1.6894 ±.0018 3 6 .6770 +.0009 .7165 +.0006 ' 1.1349 +.0006 1.3598 +.0018 1.6873 +.0024 4 6 .6758 ±.0011 .7157 ±.0010 1.1356 + .000? " I.3623 ±.0021 1.6910 +.0021 5 6 .6772 +.0004 .7160 +.0002 1.1358 +.0003 1.3596 +.0008 1.6907 +.0008 6 6 .6770 +.0004 .7159 +.0004 1.1354 +.0004 1.3600 ±.0009 1.6899 +.0010 7 6 .6769 +.0003 .7158 +.0002 1.1355 +.0006 1.3601 +.0006 1.6904 ±.0013 8 6 .6766 +.0005 .7159 ±.0004 1.1359 ±.0006 1.3608 ±.0010 1.6910 ±.0011 9 10 10 6 .6771 ±.0010 .7156 +.0006 1.1354 ±.0019 1.3598 ±.0020 1.6906 ±7 0029 6 .6763 +.0004 .7155 +.0005 1.1355 +.0005 1.3614 +.0007 1.6910 ±.0011 Average .67685 .71593 1.13566 I.36029 1.69038 ±.00026 ±.00014 ±.00021 ±.00054 ±.00054 * F r a c t i o n a t i o n c o r r e c t i o n n ormalized to Gd 15^/Gd 1^°= .9361 The parameters A and B are d e f i n e d i n Table 6-1. Only s t a t i s t i c a l e r r o r s are shown. They correspond to -two standard d e v i a t i o n s of the mean. 123 TABLE I I I - 3 . SINGLE FILAMENT ANALYSES ON SAMPLE Gd-J 155 * 157 * 158 * No. of G d\~. n G d t / G \ ° A n a l y s i s Set scans G d l f e 0 G d 1 6 0 G d l f e Q A B Gd-JSl 1 5 . 6 7 7 0 . 7 1 5 ^ 1 . 1 3 4 3 1 . 3 5 9 8 1 . 6 8 8 8 ±.0008 ± .0010 ±.0018 ± .0017 ± .0030 1 6 . 6 7 7 1 . 7 1 5 8 1 . 1 3 6 1 1 . 3 5 9 7 1 . 6 9 1 5 ± .0015 ± . 0 0 0 6 , + .0006 ±.0029 ± .0013 2 6 . 6 7 6 7 . 7 1 5 9 1 . 1 3 5 9 1 . 3 6 0 4 1 . 6 9 0 9 ± .0010 ±.0012 ±.0008 ±.0020 ±.0028 . 6 ? 6 9 3 . 7 1 5 7 5 1 . 1 3 5 8 6 I.36OOI 1 . 6 9 1 0 + .00022 ±.00029 ±.00063 ±.00042 ± .0013 Gd-JS2 Average * F r a c t i o n a t i o n c o r r e c t i o n normalized to Gd^^/Gd^^ 0= . 9 3 6 1 A l l a n a l y s es were performed u s i n g the GdO + spectrum. The r a t i o s have been c o r r e c t e d f o r the i s o t o p i c composition of oxygen. Rhenium having a s p e c i f i e d p u r i t y of 99.9% was used. Zone- r e f i n e d rhenium was not used f o r these two a n a l y s e s . The parameters A and B are d e f i n e d i n Table 6-1. 124 TABLE I I I - 4 . • SINGLE FILAMENT ANALYSIS Gd-JZl A n a l y s i s Set No. of scans Gd*55 * G d 1 6 ° G d 1 " • G d 1 6 0 M 1* 8 * G d 1 6 6 A B Gd-JZl 1 6 .6774 ±.0005 .7162 ±.0005 1.1351 ±.0015 1.3592 ±.0011 1.6885 ±.0033 2 6 .677'+ ±.0008 .7162 ±.0008 1.1356 ±.0012 1.3592 ±.0016 1.6897 ±.0022 3 6 .6767 ±.0003 • .7153 ±.0007 1.1356 ±.0010 1.360? ±.0006 1.6917 ±.0024 4 6 .6768 ±,0012 .7156 ±.0004 1.1360 ±.0008 1.3606 ±.0024 • 1.6920 '±.0018 5 5 .6774 ±.0004 .7160 ±.0003 1.1357 ±.0004 1.3593 ±.0007 1.6904 ±.0013 6 6 .6769 ±.0011 .7162 ±.0007 1.1355 ±.0006 1.3603 ±.0022 1.6895 ±.0020 7 3 .6772 ±.0010 .7161 ±.0005 1.1361 ±.0018 1.3597 ±.0021 1.6909 ±.0047 8 5 .6?70 ±.0013 ' .7163 ±.0009 I.I363 ±.0011 1.3601 ±.0026 1.6911 ±.0013 9 6 .6766 ±.0016 .7150 ±.0010 1.1364 ±.0033 1.3609 ±.0031 1.6942 ±.0079 Average .67702 ±.00025 .71591 ±.00023 1.13570 ±.00018 1.35999 ±.00050 1.69070 ±.00069 * F r a c t i o n a t i o n c o r r e c t i o n normalized to Gd^^/Gd 1^ 0--. 9361 The GdO + spectrum was used throughout. The r a t i o s have been c o r r e c t e d f o r the i s o t o p i c composition of oxygen. High p u r i t y z o n e - r e f i n e d rhenium was used f o r t h i s a n a l y s i s , The parameters A and B are d e f i n e d i n Table 6-1. 125 TABLE I I I - 5 . TRIPLE FILAMENT ANALYSES ON SAMPLE.Gd-US N o . o f Gd 1** « G d « 7 * G d158 * A n a l y s i s Set scans G d 1 6 0 G d 1 6 0 G d 1 6 0 A B Gd-us4 1 6 .6764 .7153 1.13^7 I .3609 I . 6 8 9 8 1.0009 - . 0 0 1 5 +.0011 1.0018 +.0035 2 8 .6768 .7163 1.1357 I . 3 6 0 3 1.6897 +.0004 +.0003 +.0006 +.000? +.0011 3 10 .6776 .7160 1.1369 1.3589 1.6931 ±.0015 +.0009 +.0017 +.OO31 +.0028 Gd-US5 1 6 .6768 .7159 1.1363 1.3602 1.6919 +.0003 +.0006 +.0004 +.0006 +.0014 Gd-US6 1 6 . 6 ? 6 l .7169 1.1360 I . 3 6 I 8 1.6890 +.0018 +.0008 +.0020 +.OO36 +.OO33 Gd-US7 1 8 .6770 .7176 1.1358 I .3601 1.6868 ±.000? +.000? +.0004 +.0014 ±.0019 Average .67682 , ? l 6 3 9 1.13595 1.36026 I . 6 9 0 0 +.00015 ±.ooo48-+.00034 ±.00027 ±.0016 •• F r a c t i o n a t i o n c o r r e c t i o n n ormalized to G d 1 ^ 6 / G d l 6 ° = . 9 3 6 1 The parameters A and B are d e f i n e d i n Table 6 - 1 . 126 TABLE I I I - 6 . ANALYSIS Eu-AB USING TRIPLE FILAMENTS 15^ * 1^7 * 1^8 * „ G d X 3 ) Gd±:>' 9A~1_ No. of 160 160 A n a l y s i s Set scans- Gd Gd Gd A_ B Eu-AB 1 6 2 6 3 6 4 6 5 6 6 4 7 6 ,6800 # .7160 1.1357 1 . 3 5 4 o # I . 6 9 0 3 1.001? 1 .0005 1.0013 1 .0034 1 .0025 .6828 # .7162 1.1362 1.3486 # 1.6911 1.0010 1.0010 1.0019 1.0020 1 .0027 . 6 8 1 6 # .7160 I . 1345 1 . 3 5 1 0 # 1.6877 1 .0016 1.0013 1 .000? I . 0 0 3 1 1.0041 .6820 # .7160 1.1356 1 . 3 5 0 1 # 1.6901 1.0013 1 .0005 1.0005 1 .0026 1.0010 . 6 8 0 8 # .7161 1.1361 1 . 3 5 2 5 # 1.6911 1.0008 1 .000? 1.0015 1.0015 1.0028 .6808^ ..7169 1.1352 1 . 3 5 2 5 # I .6872 1.0020 1.0007 1 . 0 0 2 1 +.0040 1 . 0 0 5 2 . 6 8 1 5 # .7162 1.1353 1 . 3 5 1 0 # I . 6 8 9 2 ±.000? ±.0006 ±.0005 ±.0014 ±.0013 Average. .6815 . 7 l 6 l 4 1.13533 1.3512* 1.68987 ±.0006 ±.00024 ±.00035 ±.0012 ±.00065 * F r a c t i o n a t i o n c o r r e c t i o n n o rmalized to Gd^^/Gd^^.9 3 6 1 # These r a t i o s show evidence of L a O + contamination a t mass 155. No c o r r e c t i o n was p o s s i b l e . The parameters A and B are d e f i n e d i n Table 6-1 . 127 TABLE III-7. ANALYSIS Gd-AB USING TRIPLE FILAMENTS A n a l y s i s Set No. of scans 155 * Gd X D : > G d 1 6 0 G d 1 5 7 * G d 1 6 0 G d 1 5 8 * G d 1 6 0 A B Gd-AB 1 6 .6968# +.0078 .7160 ±.0005 1.1355 ±.0023 1.322# ±.015 I.690 ±..005 2 6 .6867# +.0015 .7161 ±.0017 1.1335 ±.0046 l . 3 ^ - l # ±.003 1.685 + .00? 3 6 .6808 # ±.0055 .7184 ±.0051 1.1354 ±.0083 1.352 # ±.011 1.684 ±.009 Average .6867# ±.0033 .71607 ±£00032 1.1351 ±.0011 1.34l # ±.007 I.6878 ±.0036 * F r a c t i o n a t i o n c o r r e c t i o n n ormalized to Gd 1^^/Gd 1^°=.9361 # These r a t i o s show evidence of LaO + contamination a t massj:155. No c o r r e c t i o n was p o s s i b l e . The parameters A and B are d e f i n e d i n Table 6-1. 128 TABLE III-8. ANALYSIS GE-AB USING A ZONE-REFINED RHENIUM SINGLE FILAMENT Analysis Set No. of scans G d 1 ^ * G d 1 6 0 G d 1 5 7 * G d 1 6 0 G d 1 5 8 * G d 1 6 0 •A B GE-AB 1 6 .6769 ±.0013 .7169 ±.0011 I.I367 ±.0016 1.3601 ±.0026 1.6904 ±.0045 2 6 .6760 ±.0005 .7161 ±.0007 1.1357 ±.0005 1.3619 ±.0011 1.6900 ±.0019 3 6 ' .6766 +.0006 .7160 ±.0004 1.1351 ±.0005 1.3608 ±.0011 1.6890 ±.0011 4 6 .6767 ±.0004 .7160 +.0002 1.1355 ±.0005 1.3606 ±.0009 1.6900 + . 00.08 5 6 .6767 ±.0012 .7160 ±.0004 1.1363 ±.0007 1.3606 ±.0024 1.6918 ±.0017 6 6 .6766 ±.0010 .7158 ±.0005 1.1352 ±.0010 1.3609 ±.0021 1.6897 ±.0012 7 6 .6?66 ±.0009 .7157 ±.0009 1.1355 ±.0024 1.3608 ±.0018 1.6906 ±.0043 Average .67652 ±.00024 .71598 ±.00012 1.13557 ±.00031 1.36092 ±.ooo44 1.68991 ±.00058 * Fractionation correction normalized to Gd?-^^/Gd^^°~ .$361 A l l r a t i o s were obtained using the GdO+ spectrum. The ra t i o s have been corrected f o r the isotopic composition of oxygen. The parameters A and B are defined i n Table 6-1. 129 A P P E N D I X IV S L O P E OF T H E O R E T I C A L C O R R E L A T I O N LINES FOR NEUTRON C A P T U R E 157 160 * Let X and Y be the values of the Gd /Gd 158 160 * and Gd /Gd isotopic ratios when normalized to 156, 160 Gd /Gd .9361. Then ,156 X = 160 j 1 - -75(^T6u- - 9 3 6 1 ) / i G d 156,^ f Gd \ 1^160/ Y-G^P Gdl60 ( l - - 5 0 ( g T o O - - 9 3 6 l ) / ( g ^ o ) j (IV-1) where the isotopic ratios in equations (IV-1) are the measured r a t i o s , without normalization. Consideration of the thermal neutron capture processes leads to the following d i f f e r e n t i a l equations d ( G d 1 5 7 / G d 1 6 0 ) d ( G d 1 5 8 / G d 1 6 0 ) d ( G d 1 5 6 / G d 1 6 0 ) ^ 5 5 / G d 1 5 5 ' d ( G d 1 5 8 / G d 1 6 0 ) ^157 x Gd i J'' (cont. over) 130 d ( G d 1 5 5 / G d 1 6 0 ) d ( G d 1 5 8 / G d l 6 D ) ^ 5 5 / G d 1 5 5 5 / G d i J J \ „ v „ ,157 ' ^157 ' G d 1 5 7 (IV-2) F r o m equations (IV-1) and (IV-2) it can be shown that dY dX I- S{ Qu{ (IV-3) F o r s m a l l changes in isotopic composition we may- assume that Gd^^/Gd^^ = .9361, which greatly s i m p l i f i e s equation (IV-3), and leads to a slope for the c o r r e l a t i o n line for thermal neutron capture (see Figures 6-1 and 6-3) of dY/dX = - .763 near t e r r e s t r i a l Gd (using data from Tables 1-1 and 6-1). S i m i l a r l y , the slope of the c o r r e l a t i o n line for Figures 6-2 and 6-4 is given by dB dA <T B 157 ^155 1.5 l G d 1 5 8 ' + 1 H S 3 + 1 - 5.30 (IV-4) 131 BIB L IOGRAPHY Albee, A. L. , Burnett, D. S. , Chodos, A. A. , Eugster, O. J. , Huneke, J.C. , Papanastassiou, D. A. , Podosek, F. A. , Russ, G. P. , Sanz, H. G. , Tera, F. , and Wasserburg, G. J. (1970). Ages, i r r a d i a t i o n h i s t o r y , and chemical composi- tion of lunar rocks from the Sea of Tranquility. Science 167 , 463-466. Anders, E. (1964). O r i g i n , age and composition of meteorites. Space Sci. Rev. 3_> 583-714. B a l s i g e r , H. , G e i s s , J. , and Lipschutz, M.E. (1969). Vana- dium isotopic composition in meteoritic and t e r r e s t r i a l matter. E a r t h Planet. Sci. Lett. 6, 117-122. Barnard, G. P. ,(1953). Modern Mass Spectrometry, London Institute of P h y s i c s . Begemann, F. , Geiss, J. , and Hess, D.C. (1957). Radiation age of a meteorite from cosmic - ray-produced He^ and H^. Phys. Rev. 107, 540-542. Begemann, F. , Eberhardt, P. , and Hess, D. C. (1959). He - -strahlungsalter eines steinmeteoriten. Zs. f. Natur- forsch. 14a , 500-503. B i r c h , F. (1964). Density and composition of mantle and core. J. Geophys. Res. 69, 4377-4388. Burbidge, E. M. , Burbidge, G.R. , Fowler, W. A. , and Hoyle, F. (1957). Synthesis of the elements in the stars. Rev. Mod. Phys. 29, 548-650. Burnett, D. S. , Fowler, W. A. , and Hoyle, F. (1965). Nucleo- synthesis in the e a r l y h i s t o r y of the solar system. Geochim. et Cosmochim. Acta ^9, 1209-1241. Burnett, D. S. , Lippolt, H. J. , and Wasserburg, G. J. (1966). The relative isotopic abundance of K 4^ in t e r r e s t r i a l and meteoritic samples. J. Geophys. Res. 71, 1249-1269 and 3609. Cameron, A. G. W. (1962). The formation of the sun and planets. Icarus 1 , 13-69. 132 Cameron, A. G. W. (1965). Introduction to Space Science, W.N. Hess ed. , Gordon and Breach, New York. Cameron, A.E. , Smith, D. H, , and Walker, R.L. (1969). Mass spectrometry of nanogram - size samples of lead. Anal. Chem. 41, 525-526. Chakraburtty, A. K. , Stevens, C. M. , Rushing, H. C. , and Anders, E. (1964). Isotopic composition of s i l v e r in ir o n meteorites. J. Geophys. Res. 6<5, 505-520. Chemical Rubber Co. :(1970). Handbook of Chemistry and P h y s i c s , The Chemical Rubber Co. , Cleveland. Clarke, W. B. , de Laeter, J.R. , Schwarcz, H. P. , and Shane, K. C. (1970). Aluminum 26-magnesium 26 dating of feldspar in meteorites. J. Geophys.. Res. 75, 448-462. Coggeshell, N. D. (1947). F r i n g i n g flux corrections for magnetic focussing devices. J. Appl. Phys. 18, 855-861. C o l l i n s , T. L. , Rourke, F. M. , and White, F. A. (1956). Mass spectrometric investigation of the rare earth elements for the existence of new stable isotopes. Phys. Rev. 101, 196-197. Cook, J. L. and Wall, A. L. (1968). The s t a t i s t i c a l estimation of thermal-neutron cross sections. Nucl. Sci. Eng. 31, 234-240. Dawson, K. R. , Maxwell, J. A. and Parsons, D.E. (I960). A description of the meteorite which f e l l near Abee, A l b e r t a , Canada. Geochim. et Cosmochim. Acta 21, 127-144. Eugster, O., Tera, F. , and Wasserburg, G.J. (1969). Isotopic analyses of barium in meteorites and in t e r r e s t r i a l samples. J. Geophys. Rev. 74, 3897-3908. Eugster, O. , Tera, F. , Burnett, D. S. , and V/asserburg, G. J. (1970a). Isotopic composition of gadolinium and neutron- capture effects in some meteorites. J. Geophys. Res. 75, 2753-2768. Eugster, O. , Tera, F. , Burnett, D. S. , and Wasserburg, G.J. (1970b). The isotopic composition of Gd and the neutron capture effects in samples from Apollo 11. E a r t h Planet. Sci. Lett. 8, 20-30. 133 Fowler, W. A. , Greenstein, J. L. , and Hoyle, F. (1962). Nucleosynthesis during the early history of the solar system. Geophys. J. Roy. Astron. Soc. 6, 148-220. G o r i s , P. (1962). Improved bonding and emission with tantalum surface ionization filaments. ID0-14590, U.S. Atomic Energy Commission Research and Development Report TTD-4500, Ed. 17. Hayashi, C . (1966). Evolution of protostars. Annual Rev. Astron. Astrophys. 4, 171-192. Hoyle, F. , and Wickramasinghe, N. C. (1968). Condensation of the planets. Nature 217 , 415-418. Inghram, M.G. , and Chupka, W. A. (1953). Surface ionization source using multiple filaments. Rev. Sci. Instr. 24, 518-520. Krankowsky, D., and M u l l e r , O. (1964). Isotopenhaufigkeit und konzentration des lithiums in steinmeteoriten. Geochim. et Cosmochim. Acta 28, 1625-1630. Krankowsky, D. and M u l l e r , O. (1967). Isotopic composition and abundance of lithium in meteoritic matter. Geochim. et Cosmochim. Acta 31_, 1833-1842. L a r i m e r , J.W. , and Anders, E. (1970). Chemical fractionations in meteorites-III (Major element: fractionations in chondrites). Geochim. et Cosmochim. Acta 34, 367-387. Lo v e l e s s , A . J . (1967). Ion t r a j e c t o r i e s in mass spectrometer ion sources. Unpublished M. Sc. thesis, U n i v e r s i t y of Toronto. L o v e l e s s , A.J. , and R u s s e l l , R. D. (1969). A strong-focussing lens for mass spectrometer ion sources. Int. J. Mass Spectrom. Ion Phys. 3, 257-266. Lugmair, G.W. (1970). Neutron capture effects in lunar m a t e r i a l . A paper presented at the eighteenth Annual Conference on Mass Spectrometry and A l l i e d Topics in San F r a n c i s c o . Mason, B. (1962). Meteorites , John Wiley and Sons, New York. 134 Mason, B. (1963). Olivine composition in meteorites. Geochim. et Cosmochim. Acta 27_, 1011-1023. Merrihue, C M . (1963). Excess xenon 129 in chondrules from the Bruderheim meteorite. J. Geophys. Res. 68, 325-330. M i l l m a n , P. M. (1953). A cataloque of Canadian meteorites. R.A. S. C. Jour. 47, 29-33. M i y a s h i r o , A. (1968). Effect of the high-luminosity stage of the protosun on the composition of planets and meteorites. Chemie der Erde 27, 252-259. M i y a s h i r o , A. (1967 ). O r i g i n of planets and meteorites. Kagaku : -• 3J_, 520-527. Murthy, V. R. (I960). Isotopic composition of s i l v e r in an i r o n meteorite. Phys. Rev. Letters 5, 539-540. Murthy, V. R. (1962). Isotopic anomalies of molybdenum in some iro n meteorites. J. Geophys. Res. 67, 905-907. Murthy, V. R. , and Schmitt, R. A. (1963). Isotopic abundances of r a r e - e a r t h elements in meteorites. J. Geophys. Res. 68, 911-917. Murthy, V. R. , Schmitt, R.A. , and Rey, P. (1970). Rubidium- strontium age and elemental and isotopic abundances of some trace elements in lunar samples. Science 167 , 476-479. Ozard, J. M. , and R u s s e l l , R. D. (1970). D i s c r i m i n a t i o n in sol i d source lead isotope abundance measurement. E a r t h Planet. Sci. Lett. 8, 331-336. Reynolds, J.H. (1963). Xenology. J. Geophys. Res. 68, 2939-2956 Ringwood, A. E. (1966). Chemical evolution of the t e r r e s t r i a l planets. Geochim. et Cosmochim. Acta 30, 41-104. Umemoto, S. (1962). Isotopic composition of barium and cerium in stone meteorites'. J. Geophys. Res. 67, 375-379. Wood, J. A. (1968). Meteorites and the O r i g i n of Planets , McGraw H i l l Inc. , New York.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 11 0
United States 9 0
Japan 3 0
Germany 2 5
Netherlands 2 0
Poland 1 0
City Views Downloads
Beijing 8 0
Ashburn 7 0
Tokyo 3 0
Shenzhen 3 0
Zittau 2 0
Amsterdam 2 0
Mountain View 1 0
Unknown 1 5
Xenia 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0053418/manifest

Comment

Related Items