UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Solid source lead isotope studies with application to rock samples from the Superior geological province Ozard, John Malcolm 1970

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1970_A1 O93.pdf [ 4.88MB ]
Metadata
JSON: 831-1.0053415.json
JSON-LD: 831-1.0053415-ld.json
RDF/XML (Pretty): 831-1.0053415-rdf.xml
RDF/JSON: 831-1.0053415-rdf.json
Turtle: 831-1.0053415-turtle.txt
N-Triples: 831-1.0053415-rdf-ntriples.txt
Original Record: 831-1.0053415-source.json
Full Text
831-1.0053415-fulltext.txt
Citation
831-1.0053415.ris

Full Text

SOLID SOURCE LEAD ISOTOPE STUDIES WITH APPLICATION TO ROCK SAMPLES FROM THE SUPERIOR GEOLOGICAL PROVINCE by JOHN MALCOLM OZARD B.Sc, The U n i v e r s i t y of Western Ontario, 1963. A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of GEOPHYSICS We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January, 1970. In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r equ i r emen t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h C o l u m b i a , I ag ree tha t 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 f o r r e f e r e n c e and s t u d y . I f u r t h e r agree tha p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y pu rposes may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d tha t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l owed w i thou t my w r i t t e n p e r m i s s i o n . Department o f The U n i v e r s i t y o f l$ritTsh Co lumbia Vancouve r 8. Canada Date ^ > u > . ABSTRACT Data of good q u a l i t y i s necessary to further the study of rock lead i s o t o p i c i n t e r p r e t a t i o n s . An i n t e r n a l standard (double-spike) was employed to correct f o r f r a c t i o n a t i o n accompanying the s i n g l e filament analyses, and lead isotope r a t i o s with a standard deviation of 0.15% were obtained. Lead sulphide mounted on tantalum was employed f o r the lead analyses. Discrimination i n the analyses using tantalum filaments was consistent with the t h e o r e t i c a l f r a c t i o n a t i o n laws. This was not found to be the case f o r rhenium. Lead and uranium analyses from the Vogt-Hobbs area near Lake Timagami, Ontario and from the Rice Lake-Beresford Lake Area, Manitoba revealed v a r i a t i o n s from a simple two stage model that could not be accounted f o r by experimental e r r o r . I n t e r p r e t a t i o n of the data y i e l d s a three stage h i s t o r y f o r both of these regions, marked by major events at approximately 3400 and 2600 my ago. This e a r l i e r c r u s t a l h i s t o r y i s supported by ore lead data from the v i c i n i t y . Both of the suites have a lower average value of 238 204 (U /Pb ) f o r the second stage, of t h e i r three stage h i s t o r y , than i s c h a r a c t e r i s t i c of the source of s i n g l e stage leads. Variations from closed system behaviour f o r the Ontario samples i s a t t r i b u t e d to lead r e m o b i l i z a t i o n at the time of the G r e n v i l l e event. Remobilization of the lead tends to mask the 2600 my old event. The Vogt-Hobbs area i s characterized by more a c i d i c igneous rocks and more radiogenic lead than the Rice Lake-Beresford Lake area. These differences are thought to represent the environment 2600 my ago or e a r l i e r . The Manitouwadge lead analysed by Osti c resembles the lead ~ 2600 my ago i n the Rice Lake-Beresford Lake rock samples. The Rice Lak'e-Beresford Lake rock samples are shown according to the i n t e r p r e t a t i o n presented, to have had a c r u s t a l h i s t o r y p r i o r to 2600 my ago. This lends weight to arguments that exclude Manitouwadge from the class of s i n g l e stage leads used i n c a l c u l a -tions of the age of the earth. i i i TABLE OF CONTENTS Page ABSTRACT i TABLE OF CONTENTS -. i i i LIST OF FIGURES . v LIST OF TABLES v i i ACKNOWLEDGEMENTS ... v i i i CHAPTER I. INTRODUCTION 1 •Perspective 1 Delineation of the Present Study 3 CHAPTER II. METHOD OF LEAD AND URANIUM EXTRACTION 7 Introduction 7 Rock Preparation 7 Volatilization 8 Extraction of Uranium . 13 Separation of Uranium and Lead 13 CHAPTER III. MASS SPECTROMETRY 17 Sample Ionization Technique 17 Lead and Uranium Isotopic Analysis 18 Instrumentation 21 CHAPTER IV. DISCRIMINATION IN LEAD ISOTOPE ABUNDANCE MEASUREMENT . 24 Introduction 24 Theory of Fractionation 25 i v Page Lead Oxalate Analyses 27 Discussion 32 Theory of Double Spiking 36 CHAPTER V. SAMPLE LOCATIONS AND DESCRIPTIONS 45 Vogt-Hobbs Area, Ontario 45 Rice Lake-Beresford Lake, Manitoba 48 New Rock Lead Data 53 Uranium and Lead Concentrations Determinations ... 59 CHAPTER VI. INTERPRETATION OF ROCK LEAD DATA 62 Lead-Lead Interpretations 62 Uranium-Lead Interpretations 67 Comparison of Uranium-Lead and Rubidium-Strontium Systems 85 Lead M o b i l i z a t i o n i n the Timagami Samples 86 Conclusions 88 BIBLIOGRAPHY 92 APPENDIX 100 Chemical Procedures 100 Column Preparation 101 Lead Preparation . 101 N i t r a t e Resin Column f o r Uranium 102 Uranium Filament Heating Pattern 102 Concentration Determinations 103 V LIST OF FIGURES ' Page FIGURE 2-1 DIAGRAM OF ENVELOPE FOR R.F. INDUCTION HEATING 10 FIGURE 2-2 FLOW CHART FOR CHEMICAL SEPARATION 14 FIGURE 3-1 FRACTIONATED LEAD OXALATE DATA 29 FIGURE 3-2 FRACTIONATED STRONTIUM DATA 30 FIGURE 5-1 REGIONAL AND SIMPLIFIED GEOLOGIC MAPS VOGT-HOBBS AREA 47 FIGURE 5-2 REGIONAL AND SIMPLIFIED GEOLOGIC MAPS RICE LAKE-BERESFORD LAKE AREA 49 FIGURE 6-1 ISOTOPIC ABUNDANCES RICE LAKE-BERESFORD LAKE AREA 63 FIGURE 6-2 ISOTOPIC ABUNDANCES VOGT-HOBBS AREA 64 FIGURE 6-3 URANIUM-235 - LEAD-207 PLOT RICE LAKE-BERESFORD LAKE AREA 69 FIGURE 6-4 URANIUM-238 - LEAD-206 PLOT RICE LAKE-BERESFORD LAKE AREA 70 FIGURE 6-5 URANIUM-235 - LEAD-207 PLOT VOGT-HOBBS AREA . 71 FIGURE 6-6 URANIUM-238 - LEAD-206 PLOT VOGT-HOBBS AREA . 72 FIGURE 6-7 MODIFIED CONCORDIA PLOT RICE LAKE-BERESFORD LAKE AREA 75 FIGURE 6-8 MODIFIED CONCORDIA PLOT VOGT-HOBBS AREA 76 FIGURE 6-9 ISOTOPIC ABUNDANCES OF SUITES STUDIED AND RELATED DATA . 78 v i ! Page FIGURE 6-10 PREFERRED INTERPRETATIONS MANITOBA SAMPLES ... 82 FIGURE 6-11 PREFERRED INTERPRETATION ONTARIO SAMPLES 84 FIGURE A - l OPTIMUM SPIKING FOR LEAD CONCENTRATION DETERMINATION 106 v i i LIST OF TABLES Page TABLE 2-1 CONCENTRATION MEASUREMENTS FOR STANDARD GRANODIORITE GSP-1 . . 12 TABLE 4-1 LEAD OXALATE ANALYSES 28 TALBE 4-2a LOGARITHMIC SLOPES FOR FRACTIONATED DATA 35 TABLE 4-2b LOGARITHMIC FRACTIONATION SLOPES FOR RHENIUM AND TANTALUM 41 TABLE 4-3 FRACTIONATED DATA FOR SPIKE CALIBRATION 42 TABLE 4-4 SPIKE, EQUAL ATOM AND BROKEN HILL #1 COMPOSITION , 43 TABLE 4-5 VALUES AFTER EACH ITERATION FOR FRACTIONATION CORRECTION 44 TABLE 5-1 . UNCORRECTED SINGLE FILAMENT ROCK LEAD ANALYSES 50 TABLE 5-2 UNCORRECTED DOUBLE-SPIKED SAMPLE ANALYSES ... 51 TABLE 5-3 ANALYSES OF ROCK LEAD SAMPLES CORRECTED FOR FRACTIONATION 52 TABLE 5-4 ERROR PROPAGATION FOR BROKEN HILL #1 54 TABLE 5-5 URANIUM AND LEAD CONCENTRATIONS 60 TABLE 6-1 SYMBOLS AND CONSTANTS USED IN AGE DETERMINA-TIONS 66 TABLE 6-2 INTERPRETIVE AGES 74 TABLE 6-3 OBSERVED AND CALCULATED y VALUES FOR TWO STAGE MODEL 80 v i i i ACKNOWLEDGEMENTS The work necessary to produce the data reported i n t h i s t h e s i s was g r e a t l y a s s i s t e d by the help of a number of people. It i s to the c r e d i t of a l l those people mentioned and many others as well that every reasonable request was c a r r i e d out w i l l i n g l y . Dr. R.D. Rus s e l l bore with the writer i n h i s e f f o r t s to accumulate the required experimental techniques, and provided much valuable advice both during the research and i n the formulation of the t h e s i s . Dr. W.F. Slawson, Dr. T.J. Ulrych and Mr. J . Blenkinsop held thought provoking discussions throughout the period of the research. Many hours were shared with Dr. Russell and J . Blenkinsop during the development of the necessary e l e c t r o n i c equipment. The time spent by these two people i n r e w r i t i n g the computer programs f o r the IBM 360/67 w i l l not be forgotten. Two v i s i t o r s to the laboratory also deserve thanks. Dr. P. Reynolds, j u s t back a f t e r h i s studies at the A u s t r a l i a n National U n i v e r s i t y , provided h e l p f u l d i s c u s s i o n of present day r e s u l t s dealing with analysis of lead and uranium. Dr. M. Oziroa, of the U n i v e r s i t y of Tokyo, d i d a number of uranium analyses during h i s v i s i t to U.B.C. His speculative ideas provided a pleasant contrast. During t h i s research A. Loveless designed a thick lens i o n source and focussing shims as well as constructing an e l e c t r o n i c sequencing device; these s i g n i f i c a n t l y improved the ix performance of the mass spectrometer. For a two month period Mr. R. Curtis conscientiously assisted i n sample preparation. Because t h i s thesis required the development of new techniques and i n many cases new equipment the technical work done by Mr. K.D. Schreiber, Mr. M. Haines, and Mr. E. B e l l i s were of no l i t t l e importance. Mr. J . Lees helped design new glassware and both J . Lees and E. Williams s k i l f u l l y constructed and repaired glassware. Drs. W.H. White and R.E. Delavault, of the Department of Geology, kindly loaned sample heating, rock crushing and atomic absorption equipment during the course of t h i s study. Chemical separations were c a r r i e d out i n a laboratory provided by Dr. J.V. Ross. Dr. A. Turek kindly c o l l e c t e d and shipped rock samples from the Rice Lake-Beresford Lake Area. The writer c o l l e c t e d the Timagami samples with Dr. J . Grant without whom the task would have been considerably more d i f f i c u l t . During the e a r l i e r part of the writers stay Dr. J.A. Jacobs provided enjoyable and stimulating lectures i n geophysics. Dr. Russell and Dr. R.M. E l l i s introduced the writer to some aspects of d i g i t a l processing that have provided a useful background. A l l of the members of the committee are thanked f o r t h e i r consideration of the t h e s i s . Those others including Mr.R . Culbert, X Mr. J . Blenkinsop and Dr. T.J. Ulrych who have also read the th e s i s during i t s preparation are also thanked. The thesis was c a r e f u l l y typed by Miss V. Ormerod. F i n a n c i a l support f o r t h i s research came p r i m a r i l y from the National Research Council of Canada, e i t h e r through grants to Dr. R.D. R u s s e l l , Dr. W.F. Slawson and Dr. T.J. Ulrych, or as a studentship. One year of personal support was provided by the Un i v e r s i t y of B r i t i s h Columbia and some equipment was financed by the National Science Foundation of the United States. 1 CHAPTER I INTRODUCTION Perspective There have been many studies of variations in lead isotope abundances in natural minerals. The resulting data have led to numerous unanswered questions pertaining to the solid earth. The majority of precise analyses have been those of ore bodies containing at least a few percent of lead. However, any understanding of lead isotope abundance variations in the crust is incomplete unt i l more i s known about lead isotope variations i n crustal rocks. A clue as to the possible significance of rock lead can be had when one realizes that the lead contained i n a typical ore body could have been extracted from about 10 cubic kilometers of rock (Sinclair, 1965). In an effort to c l a r i f y the understanding of available data on lead extracted from whole rocks, Ulrych and Reynolds (1966) presented an interpretation analogous to rubidium-strontium whole rock studies (Nicolaysen, 1961). In their model the f i r s t stage, called by them the "common lead environment", can be likened to the history of some common leads and may correspond to more than one lead-uranium system. The second stage, the "radiogenic lead environment", i s the one in which the uranium-lead system presently observed i s responsible for the last enrichment of the radiogenic lead component through the decay of uranium and thorium. Details of this interpretation w i l l be elaborated in Chapter 5. Ulrych and Reynolds examined Zartman's data from the Llano U p l i f t , Texas, i n the light of their interpretation. Although the 2 interpretation was applied to the best data available at the time, ages and isotope ratios determined were very imprecise. Of the scatter exhibited by the data i t was not clear what proportion arose from open system behaviour of the rocks and what proportion arose from experimental errors. More recently Ulrych has presented an alternative scheme for extracting information from rocks having a two stage history (Ulrych, 1967). This was a modification of a model by Gerling and Shukolyukov, 1957, which extracts more directly the time at which the f i r s t and second stages commenced. These two ages correspond to the age of the earth and time of formation of the second stage. Clearly the assumption of only two stages i s more restrictive than the interpretation described in the earlier paragraphs, but i t may be appropriate i n special cases. One objective of this thesis i s to contribute new data of sufficient quality to further the understanding of such lead isotope interpretations. A second objective of this study is to compare the behaviour of U-Pb and Rb-Sr systems in the same region. A recent study by Wanless et. a l . (in press) of samples taken from the v i c i n i t y of the Grenville Front near Chibougameau, Quebec, has shown that the response of the Rb-Sr system, to metamorphism, is i n marked contrast to that found for the K-Ar system. Most notable is the anomalously high K-Ar ages possible and the low apparent Rb-Sr ages for the mineral bi o t i t e . This sort of anamolous behaviour leaves the significance of K-Ar and Rb-Sr biotite ages open to question unless supporting evidence is available. 3 A f i n a l o b j e c t i v e i s t o draw any i n f e r e n c e s t h a t may be p o s s i b l e c o n c e r n i n g t h e g e o c h r o n o l o g i c h i s t o r y o f t h e r e g i o n s s t u d i e d ; o r k n o w i n g t h e g e o l o g i c h i s t o r y t o draw c o n c l u s i o n s t h a t may be g e n e r a l i t i e s f o r U - P b s y s t e m s . To draw p o s s i b l e i n f e r e n c e s c o n c e r n i n g t h e g e o c h r o n o l o g i c h i s t o r y o f t h e r e g i o n s s t u d i e d i s d e f i n i t e l y a s e c o n d a r y o b j e c t i v e . I t i s r e a l i z e d t h a t t h e s a m p l i n g s o f t h i s s t u d y a r e l e s s t h a n i d e a l f o r i n t e r p r e t a t i v e s t u d i e s . D e l i n e a t i o n o f t h e P r e s e n t S t u d y C l e a r l y , d a t a o f good a c c u r a c y i s r e q u i r e d t o meet t h e o b j e c t i v e s s t a t e d a b o v e . Much o f t h e p u b l i s h e d d a t a f o r r o c k l e a d i s o t o p e abundances i s much t o o i m p r e c i s e f o r t h e p u r p o s e . The f i r s t o b j e c t i v e was t o o b t a i n d a t a o f a c c e p t a b l e q u a l i t y . When t h i s s t u d y was b e i n g i n v e s t i g a t e d two p r o m i s i n g t e c h n i q u e s were b e i n g d e v e l o p e d . C a t a n z a r o (1967) had s u c c e s s f u l l y a n a l y s e d 500 m i c r o - g r a m r e f e r e n c e s a m p l e s w i t h a s t a n d a r d d e v i a t i o n o f 0.03% i n P b ^ ^ / P b 2 ^ . I n t h i s t r i p l e f i l a m e n t t e c h n i q u e t h e s amp le i s e v a p o r a t e d f rom two r h e n i u m s i d e f i l a m e n t s and i o n i z e d on a c e n t e r p l a t i n u m f i l a m e n t a t a p p r o x i m a t e l y 1500°C. No f r a c t i o n a t i o n o f t h e l e a d was d e t e c t a b l e f rom r a t i o s o b t a i n e d i n t h e f i r s t s t u d y , 208 20 b u t f r a c t i o n a t i o n o f abou t 0.03% was l a t e r r e p o r t e d i n t h e Pb / P b r a t i o ( C a t a n z a r o , 1968). T h i s work has b e e n r e c o g n i z e d as an o u t s t a n d i n g a c h i e v e m e n t . The a l t e r n a t i v e t e c h n i q u e , u s u a l l y c a l l e d d o u b l e s p i k i n g (Compston and O v e r s b y , 1969), r e q u i r e s two a n a l y s e s . ^ The f i r s t i s a c o n v e n t i o n a l s o l i d s o u r c e s i n g l e f i l a m e n t a n a l y s i s w i t h a t t e n d a n t f r a c t i o n a t i o n o f up t o 2%. Then a s e c o n d 4 analysis i s required to establish the magnitude of the frac t i o n a t i o n . In simple terms, t h i s i s done by adding large quantities of 2 isotopes of known r a t i o and measuring the composition of the mixture. The second analysis enables one to correct for fr a c t i o n a t i o n . A smaller t o t a l sample i s required for t h i s technique; the present studies were made with lOOu gm samples. Both techniques were used i n the present investigation; but the double spike technique was found to be far more sat i s f a c t o r y for rock lead analyses because of better s e n s i t i v i t y . The mass spectrometer available for t h i s study was a 12 inch, single focussing, 90 degree, gas source analyser, which was under construction when the research began. The f i r s t part of the project consisted of converting the instrument to s o l i d source, assembling with the help of R.D. Russell the necessary electronic supplies, and the i n s t a l l a t i o n of improved pumping systems for the source and c o l l e c t o r regions. A.Loveless designed and supervised the i n s t a l l a t i o n of a new thick lens source and magnet shims to achieve second order focussing, which resulted i n greater s e n s i t i v i t y and a more s i m p l i f i e d operation than we had before (Ozard and Russell, 1969; Loveless and Russell, i n press). Since the double spiking technique assumes the form of f r a c t i o n a t i o n patterns, i t was thought necessary to observe such patterns for our machines. The results of t h i s study w i l l be reported i n chapter 3 (see also Ozard and Russell, i n press). Various techniques were used to extract the lead and uranium i n forms suitable for mass spectrometric analysis. Although not new, 5 most of these techniques had not been used previously at t h i s laboratory. These included the oxalate technique of Cooper and Richards (1966), the sulphide technique of Patterson (Mair, 1958), the v o l a t i l i z a t i o n technique of Masuda (1962) and Tatsumoto (1966), and the d i s s o l u t i o n method of Doe (1967). De t a i l s of t h i s work w i l l be found i n the following chapters. In order to contribute meaningfully to i n t e r p r e t a t i o n s of whole rock U-Pb analyses, i t i s desirable to study rocks for which the i n t e r p r e t a b l e e f f e c t s are large. The present day uranium and lead concentrations, and only these can be measured, are probably representa-t i v e of the l a s t stage of development of the lead. In the l i g h t of t h i s consideration the oldest known rocks are l i k e l y to be the most s u i t a b l e . To meet the second objective of the research (a comparison of Rb-Sr and U-Pb systems i n the same region) i t would be necessary to have Rb-Sr analyses of samples taken from the same region. Although analyses of sample s p l i t s would be preferable f o r a d e t a i l e d geological under-standing, analyses of the same bodies of rock should provide a reasonable comparison on a broad s c a l e . An obvious choice i n Canada i s the Superior geological province. Two regions were selected, both of which had been the subject of extensive geochronological i n v e s t i g a t i o n s . One of these i s the Rice Lake-Beresford Lake area of south-eastern Manitoba, which i s well within the Superior province. Turek and Peterman (1968) found ages of 2500 to 2700 my i n t h i s region. The second area was on the G r e n v i l l e side of the G r e n v i l l e front i n an area studied by Grant (1964a, 1964b). The importance of understanding the complicated phenomena i n the region of the' front has 6 already been mentioned. The w r i t e r accompanied Dr. Grant to the Timagami region and c o l l e c t e d , with him, the samples f o r the present study. This i s one of the f i r s t studies to achieve lead i s o t o p i c composition measurements of t h i s p r e c i s i o n and accuracy f o r whole rocks. E a r l i e r analyses of t h i s p r e c i s i o n have been c a r r i e d put f o r ores or minerals containing larger concentrations of lead or uranium. In many of the e a r l i e r studies the concentrations of lead or uranium were one or more orders of magnitude la r g e r . . For cases i n which whole rocks were analysed p r i o r to 1969 the p r e c i s i o n or accuracy were d e f i n i t e l y poorer. 7 CHAPTER II METHOD OF URANIUM AND LEAD EXTRACTION Introduction The objective in extracting lead and uranium is to obtain pure samples of these elements that are representative of the whole rock. Although isotopic dilution was used to determine concentrations, quantitative recovery is required to ensure there is no significant discrimination between common and radiogenic lead. Throughout the uranium and lead extraction procedures care must be taken to ensure that the samples are not contaminated. As isotope ratios correct at the 0.1% level were the primary objective of the analytical techniques, contamination errors should be kept below this level. To this end f i l t e r e d a i r was supplied to keep the laboratory at a positive pressure. Reagents were purified as described in the appendix. Rock Preparation Samples from Ontario were collected by the writer and J. Grant. When necessary, dynamite and d r i l l i n g were employed to ensure that the samples were fresh. The samples from Manitoba were collected by A. Turek. Only the largest and least weathered of these samples were used. The whole rocks, which weighed about 20 kg in total, were s p l i t into pieces of approximately 100 g and the weathered fragments were removed. Contamination of whole rocks during crushing is much more c r i t i c a l than for mineral separates, as the latter are usually leached with acid to remove any lead or uranium held on the grain surfaces. 8 Unweathered pieces were crushed to pass approximately 40 mesh si,ze. The f i r s t half through the crusher was used to flush out any rock dust not removed by cleaning. A 250 gm s p l i t from the second portion was further crushed by hand to pass 100 mesh. A l l polyethylene containers used to store the samples were soaked in 15% n i t r i c acid for 24 hours, and the t i n soldered stainless steel sieves were cleaned of particles of rock with a fine soft brush. Volatilization of Lead Extraction of lead by v o l a t i l i z a t i o n was chosen in preference to a perchloric-hydrofluoric acid .attack. The pyrochemical technique is less hazardous and less liable to contaminate the samples, and partial separation of the elements is achieved by the d i s t i l l a t i o n . Furthermore i t i s impractical to dissolve .40 g samples while i t is quite practical to extract lead from this quantity by v o l a t i l i z a t i o n , with no additional effort. At f i r s t , samples of U.S. Geological Survey analysed granodiorite (GSP-1) were heated in a tube resistance furnace at 1000°C to 1100°C in a hydrogen stream at atmospheric pressure. As only 80 to 90% recovery was achieved, the need for higher temperatures was indicated. The importance of sufficiently high temperatures has been noted by other researchers (Masuda, 1962; Tatsumoto, 1966; Welke, 1968) in the case of basic rocks. Tatsumoto reports at least 95% recovery at 1300°C, while Welke (1968) considers temperatures above 1150°C essential for 90% recovery. The results of this study, to be presented later, show that similar temperatures are appropriate for granodiorites. 9 In order to obtain higher temperatures, radio frequency induction heating was used. The apparatus is shown in Figure 2-1. The sample, in the form of 100-mesh powder, was mixed thoroughly with one f i f t h of i t s weight of purified graphite powder. For concentration determinations, lead-206 spike was added at this stage. The graphite crucible was then placed inside the envelope and the chamber evacuated. The evacuation had to be performed slowly to prevent loss of sample. -3 After about one hour the pressure reached 10 mm of mercury and the diffusion pump was switched on. To heat the crucible, the voltage applied to the work c o i l was raised i n 20 equal steps over a two hour period so that the f i n a l brightness temperature was 1350°C. This temperature was maintained for one hour, then the crucible was allowed to cool and, for ratio determinations, a second loading was placed inside the envelope and heated. The mirror was then dissolved i n 50 ml of 1.5N hydrochloric acid. Although the mirror dissolved with a violent reaction, the acid was allowed to remain in contact with the envelope overnight. Between lead volatilizations, the graphite was purified by heating to 1450°C for one hour. After heating, the Vycor insert was rinsed with 50% hydrofluoric acid for two minutes, then soaked in n i t r i c acid for twelve hours, and f i n a l l y rinsed with t r i p l e - d i s t i l l e d water. Lead blanks were from 0.1 to 0.5u g. These measurements include contamination from chemical processing and are therefore an upper limit for the contamination during the v o l a t i l i z a t i o n step. The blank represents about 0.2% of the sample lead for concentration runs and 0.05% of the sample for ratio determinations. The contaminant isotope ratios 10 FIG. 2-1 DIAGRAM OF ENVELOPE FOR RF. INDUCTION HEATING WATER OUT MOLYBDENUM STAND BOROSILICATE CASE TO VACUUM PUMP 11 were measured and found to be approximately those of modern lead. As the sample and contaminant r a t i o s d i f f e r by only 50% at most the maximum error introduced i n a r a t i o determination i s 0.025%. A fusion technique i s often considered necessary to extract lead f o r concentration determinations. Doe compared a borax fusion method with h i s h y d r o f l u o r i c - p e r c h l o r i c acid attack and found i d e n t i c a l values f o r the two procedures (Doe, 1967). Therefore the.acid attack may be considered as s a t i s f a c t o r y as the borax fusion f o r extracting the lead from a granodiorite. The a c i d attack i s often c r i t i c i z e d because i t i s thought to leave accessory minerals such as sphene and z i r c o n undissolved. However a sphene separate was e n t i r e l y dissolved at t h i s laboratory by using an acid attack s i m i l a r to that of Doe (R. Culbert, personal communication). Furthermore, the high temperatures required f o r the borax fusion technique may be responsible f o r lead loss by v o l a t i l i z a t i o n or by a l l o y i n g with the c r u c i b l e , which e f f e c t s are less l i k e l y with the acid attack. (See also Tatsumoto, 1966) Analysed granodiorite, GSP-1, was analysed under a v a r i e t y of conditions at t h i s laboratory to determine the lead concentration. The r e s u l t s are l i s t e d i n Table 2-1. Heating a mixture of rock and graphite at 1350°C, f o r at least one hour, gives lead concentrations consistent with the a c i d attack. It i s concluded that the v o l a t i l i z a t i o n technique, used i n t h i s study, may be employed i n determining lead concentrations by isotope d i l u t i o n . Doe reports a value of 58.7 ppm, by weight, for the lead concentration of t h i s rock. However the standard i s divided i n t o quart scoops then s p l i t f o r d i s t r i b u t i o n (Doe, personal communication) so that such a di f f e r e n c e f o r the lead concentration i s quite p o s s i b l e . 12 TABLE 2-1 CONCENTRATION MEASUREMENTS FOR STANDARD GRANODIORITE GSP-1. ANALYSIS METHOD TEMPERATURE CONCENTRATIONS 104 Tube furnace in hydrogen 1000°C 45.8 105 II II it 1100°C 52.1 107 Induction heating i n vacuum 1350°C 55.3 115 II II ti II 55.3 117 II II II II 55.1 119 Perchloric-hydrofluoric acid 55.2 Concentrations are in ppm by weight. 13 Extraction of Uranium Uranium was extracted for concentration determinations. The dissolution procedure was that described by Doe et. a l . (1967). The sample was dissolved in hydrofluoric-perchloric acid and the uranium concentration was determined by isotope dilution. As four times Doe's quantity of rock (and hence acid) was employed, and as the surface area for evaporation was only slighly greater, the dissolution took three days, a time considered to be long enough to guarantee complete dissolution. No evidence of undissolved minerals was found when the residue was dissolved in 1.5N hydrochloric acid at 80°C. An ammonium hydroxide precipitation was used at a pH of 7 to precipitate insoluble metallic hydroxides. After centrifuging, the precipitate was taken up in 1.5N hydrochloric acid, and was ready for the separation of other elements. The blank for the above procedure was from 0.02 to 0.05u g of uranium and represented less than 0.5% of the sample. Separation and Purification of Uranium and Lead By using Dowex 1 x 8 resin, n i t r i c acid, hydrochloric acid, and d i s t i l l e d water i t was possible to separate and purify lead and uranium sufficiently for mass spectrometric analysis. In the mass range in the v i c i n i t y of singly charged lead no contaminant ions were found. Thorium was usually observed in the analysis of uranium, but no attempt was made to remove i t . Some analysts have spiked for thorium and run the thorium analysis after uranium. Solid source mass spectrometer analysis of lead is very inefficient (typically one ion from 10^ atoms) because of the high F I G U R E 2 - 2 FLOW CHART FOR C H E M I C A L S E P A R A T I O N V O L A T I L I Z A T I O N O F L E A D P E R C H L O R I C - H O D R O F L U O R I C A C I D D I S S O L U T I O N S A M P L E ADDED TO C H L O R I D E R E S I N COLUMN D I S S O L V E D I N 1 . 5 N H Y D R O C H L O R I C A C I D A F T E R URANIUM E L U T I O N L E A D IS E L U T E D WITH D I S T I L L E D .WATER R E P E A T F I R S T COLUMN FOR LOW C O N C E N T R A T I O N S A M P L E S SOME URANIUM P A S S E S THROUGH T H E COLUMN T H E R E M A I N D E R IS E L U T E D WITH H Y D R O C H L O R I C A C I D URANIUM E N T E R S N I T R A T E R E S I N COLUMN D I S S O L V E D IN 7N N I T R I C A C I D L E A D E N T E R S S M A L L C H L O R I D E R E S I N COLUMN D I S S O L V E D IN 1 . 5 N 'HYDROCHLORIC A C ID "COLUMN IS WASHED W I T H " H Y D R O C H L O R I C A C I D v E L U T E L E A D WITH 6N H Y D R O C H L O R I C A C I D P R E C I P I T A T E L E A D WITH HYDROGEN S U L P H I D E COLUMN I S WASHED WITH 7N N I T R I C A C I D 1 URANIUM IS E L U T E D WITH D I S T I L L E D WATER FOLLOWED BY 1 . 5 N H Y D R O C H L O R I C A C I D N I T R A T E R E S I N COLUMN PROCEDURE R E P E A T E D WITH S M A L L COLUMN I L E A D S U L P H I D E U R A N Y L N I T R A T E 15 ionization potential of this element. In order to have measurable ion beams, therefore, tens to hundreds of micrograms of lead are required for an analysis. Because i t i s impractical to extract larger quantities from granitic rock containing about 10 ppm of lead, i t is essential to achieve maximum ionization efficiency. The chemical separation should be designed to remove materials of low ionization potential which tend to neutralize lead ions (Mair, 1958). In particular, every effort should be made to keep the concentration of sodium and potassium as low as possible. These are the most abundant ions commonly present during the analysis of lead in the mass spectrometer. The purification adopted for lead was based on a study by Kraus and Nelson (1955). The separation depends on the fact that lead is held on Dowex 1 x 8 (anion exchange resin chloride form) in hydro-chloric acid solutions between OM and 6M. No other ion in their study showed a peak at 1.5 molar hydrochloric acid with desorption at OM and 6M molar hydrochloric acid. A brief description of the column procedure follows; the details are to be found in the appendix. The f i r s t column procedure was obtained by personal communica-tion from the U.S.G.S. i n Denver, and serves to separate the uranium and purify the lead. The sample from the vo l a t i l i z a t i o n step (for lead) was dissolved in 1.5N hydrochloric acid, and was added to the prepared resin column. The resin was washed with 1.5N hydrochloric acid. Triple d i s t i l l e d water was used to elute the lead. At this stage the major contaminants of the lead are probably iron, zinc, cadmium, indium, t i n , and antimony. A second small column (Catanzaro and Gast, 1960) may be used to remove the major part of a l l 16 of these elements. The lead i s added in 1.5N hydrochloric acid to the second prepared column. The column is washed with 1.5N acid and the lead eluted with 6N hydrochloric acid. This completes the lead purification. Three columns were used to obtain sufficiently pure uranium for mass spectrometer analysis. The f i r s t column was used to separate the uranium from the lead when both are being extracted from the same s p l i t . The residue from the hydroxide precipitation containing the spiked uranium in 1.5N hydrochloric acid was added to the prepared resin column. Then resin was washed with 1.5N hydrochloric acid. The acid passing through the column w i l l contain a l l the uranium. Uranium from the f i r s t column remains to be purified. A column of Dowex 1 x 8 resin, nitrate form, was used to purify i t . The uranium was added to the column in 7N n i t r i c acid, then the resin was washed with more 7N n i t r i c acid. The uranium (and the thorium) were eluted with.water followed by 6N hydrochloric acid. A similar procedure using a smaller volume of resin was used to improve the purification of the uranium. The uranium was sufficiently pure to produce large stable beams from 1 gm of rock containing 1 ppm of uranium. 17 CHAPTER III MASS SPECTROMETRY Sample Ionization Technique The solid source single filament lead sulphide loading adopted for the greater part of this study was f i r s t described by Mair (1958) and attributed to CC. Patterson. The oxalate technique of Cooper and Richards (1966) and the t r i p l e filament method of Catanzaro (1967) were also employed. Several variants on these methods were tried, some a number of times, with l i t t l e success. When adequate quantities of lead are available the t r i p l e -filament method of Catanzaro can be employed. In this method lead hydroxide i s precipitated on two rhenium side filaments. These filaments are heated by passing a current through them to evaporate the sample. The centre platinum ionizing filament i s heated to 1500°C. It was possible to produce either stable ion beams of 10 ^  amperes that lasted an hour, or higher sensitivity for a shorter period, from 500u g samples. However only about one analysis i n two was successful, and when lead extracted from rocks was analysed lower sensitivity was usually obtained. On account of the time required to extract 500u g quantities of lead from rocks this method was considered unsuitable. A more sensitive technique that gave absolute ratios was needed. Comparison of sensitivities reported in the literature showed that the oxalate and sulphide single filament loadings are ten times as sensitive as that reported for the t r i p l e filament technique. This was confirmed by analyses performed in this study. The paper by Compston and Oversby (1969) 18 indicated that errors caused by fractionation accompanying single filament lead analyses can be adequately corrected by using double spiking. The gain in effective sensitivity is about a factor of five as two runs in the mass spectrometer are required for the double spiking correction. Both of the single filament loadings give sufficient sensitivity. On account of the simpler sample pre-treatment for filament loading the sulphide was used for the rock lead analyses. Lead sulphide and ammonium nitrate solution were carefully dried on outgassed tantalum filaments. Stable total lead beams of 5 x 10 amps were routinely achieved from 75u g loadings. This method proves very reliable i f the technique described in the appendix is followed exactly. It i s most important to dry the filament loading slowly to prevent the sample from flaking off the filament. Furthermore the correct pH for precipitation of the sulphide and the gradual beam building during the analysis are prerequisites for good s t a b i l i t y and sensitivity. Lead and Uranium Isotopic Analysis The filament heating pattern employed for the lead isotopic analysis was based on that described by Cooper and Richards for the oxalate technique (personal communication). The filament temperature was raised uniformly to 750°C over a period of an hour and l e f t at this temperature for two hours. In this way i t is possible to v o l a t i l i z e the lower melting point materials preferentially, and achieve a further purification of the lead in the mass spectrometer. Residual a l k a l i beams were comparable to those reported by Cooper and Richards (1 x 10 A), and tended.to die away during the analysis. Over the 19 next hour the temperature was gently raised to 850°C at which temp-erature adequate sensitivity was achieved. Data taking for a lead analysis consisted of peak hopping, followed by scanning. The baseline just below mass 204 was f i r s t recorded on a l l shunts, then a peakhop set consisting of ten sets of the sequence: baseline, 204, 206, 207, and 208 peak tops. Baselines were again taken on a l l shunts and the spectrum scanned twice. For a l l analyses of rock lead reported the d i g i t a l voltmeter readings were recorded manually. On account of the high sensitivity and s t a b i l i t y any operator bias i n this operation is much less than 0.1%. This method was preferred to measuring each ratio 204/206, 207/206, 208/206 in turn, because i t ensures the same fractionation conditions for a l l ratios. The baseline values obtained from the scanned spectrum pair were used to correct for pressure scattering. For the pressures _7 encountered, about 1 x 10 mm of mercury, the corrections to a peak were about 0.05% of i t s height. Peak hopping enables one to monitor four peaks, and the baseline in l i t t l e more than one minute. For such short periods a linear extrapolation may be employed between successive values of a peak top for ratio determinations. The maximum systematic error incurred from such an approximation was 0.01%. The small size of this error is a consequence of the small and approximately linear growth in peak height over short periods. 20 For uranium isotopic analysis a triple-filament rhenium-ribbon source was used. The heating pattern employed was developed by the United States National Bureau of Standards (1966). Fractionation is inherent in t r i p l e filament uranium analysis, but the variation of fractionation can be minimized by employing identical filament condi-tions. The acidity of the uranyl nitrate solution, sample size and temperature of the filament were shown to be important factors in the study performed by the National Bureau of Standards. The data reported i n this Thesis were a l l obtained from low temperature analyses of approximately lOu g samples. By extrapolating the data presented by the National Bureau of Standards fractionation . of about -0.4% in the U238/U235.ratio would be expected. A l l fractionated data presented by the N.B.S. shows an enrichment of the lighter isotope. The extremely reproducible sample behaviour during the analysis i s considered to be an indication that filament condi-tions were successfully reproduced from one analysis to another. Consequently fractionation should provide a bias for a l l uranium data but no significant between-run error at the 1% level. Further details of the analytical procedure are to be found i n the appendix. Data taking was similar to that for lead analysis except that only two peak tops were monitored. Pressure scattering, although corrected for, was less than 0.1% as the poorer peak shape and small peak separation per mass unit are offset by the three mass unit separation between peaks. 21 Instrumentation • Analyses were carried out with a 90° sector, 12 inch radius, single stage solid source mass spectrometer. The instrument was designed principally by R.D. Russell, and incorporates a focussing thick lens ion source (Loveless and Russell, 1969) and second order focussing shims. When focussed for singly ionized lead atoms, i t has a resolution of about 400 at mass 200. The electronic supplies are almost entirely solid state. For precise analyses, adequate beam intensity is essential. This was partly achieved by using an ion source that transmits ions e f f i c i e n t l y and partly by using a solid source, for which f a i r l y efficient ionization techniques exist. By analysing lead samples in the thin lens previously used by us and with the new thick lens source, i t i s estimated that a factor of about three was gained in transmission efficiency. This design permitted precise analyses of lOOu g quantities of lead on a routine basis. The measuring system, filament and high voltage supplies were designed by Russell and constructed before the rock lead analyses were performed. The s o l i d state measuring system proved extremely reliable; i t incorporated peak selection in conjunction with the magnet current supply. A third aspect of the measuring system is the provision of peak height, shunt and scan direction information for recording on magnetic tape. An intercessor was b u i l t by Loveless to sequence the output of the d i g i t a l voltmeter and measuring system for input to the seven track tape recorder. However the greater part of the rock lead analyses were recorded manually as the computer reduction was set up 22 for scanned rather than peak-switched data. • The measuring system incorporates a ladder attenuator which determines the shunt ratios. To calibrate the ladder attenuator a precision decade divider (Electro-scientific Instruments, Model 622 A. maximum deviation from linearity one part in 10^ and input resistance 10"' ohms) and null detector were employed in a bridge. The calibrated shunt values were found to be within 0.1% of the nominal values as specified by the manufacturer (also Electro-scientific Instruments). As a more thorough calibration of the measuring system a voltage source was inserted in the feedback lead. This voltage was supplied by the decade divider just described. The voltage, applied in the feedback lead, to produce a constant output above the baseline for different attenuations, provides a calibration of the measuring system including the d i g i t a l voltmeter. The values for each shunt ratio for the two different methods agreed to better than 0.1%, and were used in a l l calculations. A stable ion beam enables one to make precise ratio determina-tions more easily. As the ion beam strength is an exponential function of temperature, the current used to heat the filament must be well regulated. A portion of the filament current was fed through the filament of a temperature limited diode. The anode current was used for providing feedback, and with this arrangement fluctuations of the ion beam caused by filament current variation were less than 0.1% over a period of one minute. Before converting the mass spectrometer to a solid ionization source the writer had investigated the po s s i b i l i t y of improving the 23 efficiency of the gas ionization source. A gas source was designed •on the basis of a mathematical theorem presented by Naidu (1966). This theorem specifies the electrostatic f i e l d to guide charged particles along a prescribed set of paraxial paths. The gas source was designed to produce damped oscillatory trajectories. The new gas source was found to have an overall efficiency the same as the modified Nier source that had been used previously (Ozard and Russell, 1969). Computations by Loveless verified that the failure to improve efficiency was a direct consequence of the i n i t i a l conditions assigned by Naidu. Loveless then designed a solid source with the aid of a computer program to calculate ion trajectories. This new solid source has fewer electrodes than the conventional solid source but s t i l l employed damped oscillatory trajectories to focus the ions. As the ion source produces ions with a larger angular divergence, the magnet of the mass spectrometer was f i t t e d with shims to provide second order focussing. The collector also had to be modified to accept ions with a larger angular convergence. 24 CHAPTER IV DISCRIMINATION IN LEAD ISOTOPE ABUNDANCE MEASUREMENT* Introduction The t r i p l e filament technique developed by Catanzaro (1967) was shown by him to give isotopic ratios agreeing closely with the known values for synthetic mixtures of separated lead isotopes. A double spike has been used by Compston and Oversby (1969) and by Cooper, Reynolds and Richards (1969) to obtain unfractionated values for lead isotope ratios. Double spiking depends on calibra-tion against the NBS equal-atom standard for an absolute reference, and hence on Catanzaro's results. For routine isotopic analysis of lead from rocks, double spiking i s much more attractive, mainly because of greater sensitivity. However, reduction of double spiked data assumes that a l l discrimination effects are due to mass dependent fractionation processes, and hence assumes particular numerical values for the relative fractionation of each isotope ratio. Therefore i t i s approp-riate to investigate the discrimination properties of the mass spectrometer used i n this study. Fractionation of lead isotopes through physical-chemical processes has been of long standing interest. The vapour pressure and a f f i n i t y of isotopes of lead was related to the physical properties as early as 1919 by Lindemann. The d i f f i c u l t i e s of fractionating lead _ The f i r s t half of this chapter was presented i n Zurich at the Conference on Geochronology of Phanerozoic Orogenic Belts, and has been submitted for publication. 25 isotopes are recounted by Richards et. a l . (1926). Duchylard et. a l . (1953) also searched for f r a c t i o n a t i o n , i n multistage Grignard reactions which are involved i n the preparation of lead tetramethyl for mass spectrometer analysis. Senftle and Bracken (1954) calculated isotopic f r a c t i o n a t i o n data for lead d i f f u s i o n processes i n nature. Clearly i t i s desirable to know as much as possible about instrumental discrimination i n order to t r y to better control and correct for i t . Quantitative corrections have generally been based on a f r a c t i o n a t i o n process, often, considered to derive from physical-chemical mass dependent processes (for example, Doe, 1967; Dodson, 1963). In the following sections data i s presented which appears to be s u f f i c i e n t l y precise to show that t h i s assumption, though a good approximation, may actually be inadequate for precise double spiking corrections where the f r a c t i o n a t i o n i s large. Theory of Fractionation Since the effect of greatest importance i s the behaviour of the mass spectrometer as a whole, rather than a s p e c i f i c process such as f r a c t i o n a t i o n on the filament, i t i s useful to consider the theory i n i t s most general form. Dodson (1963) suggests a derivation that i s independent of the a n a l y t i c a l form of the fractionation law, and assumes only that i t i s a smooth function of mass. To predict the expected f r a c t i o n a t i o n the following derivation has been used which, though equivalent to Dodson 1s, i s more compact and refers s p e c i f i c -a l l y to lead. 26 Define the isotope ratios D,206 / D,204 m 207 y m204 D,208 / D,204 x = Pb /Pb ; y = Pb /Pb ; z = Pb /Pb where the chemical symbols represent the intensity of the measured ion beam at each mass number. Now for a small change in ion beam intensity fix x 6 Pb 206 6 Pb 204 Pb 206 Pb 204 and similarly for — and — y z X 6Pb 2 0 7/Pb 2 0 7 - 6Pb 2 0 4/Pb 2 0 4 6Pb 2 0 6/Pb 2 0 6 - SPb 2 0 4/Pb 2 0 4 C4 -1 ) 207 207 Let us assume that the ratios 6Pb /Pb , etc., are functions only of mass number in the sense that the ratios vary smoothly with mass for a given chemical element, regardless of the mechanism of the fractionation. M M Then 6Pb /Pb = f(M), M = mass number, and & - L x ff(207) - f(204)1 \f(206) - f(2Q4)J (4-2) Similarly 6z 3 7 If(208) - f(204)\ |f(206) - f(204)J (4-3) 27 If f(M) i s a smooth function of M, then we may expand about f(204), using a Taylor's series. Thus f(207) = f(204) + 3f'(204) + |f"(204) ... , and f(207) - f(204) r 3f'(204). Similarly £(206) - £(204) - 2f'(204), (4-4) f(208) - f(204) - 4f'(204). Substituting relationship (4) into (2) and (3) gives 6y/6x = 1.5 y/x (4-5) 6z/Sx = 2.0 z/x (4-6) The error made by dropping the terms in f"(M) is very small in common cases. Where f(M) <* l/v'ftf, for example, ratio 6y/6x should be 1.49 y/x and for 6z/6x, should be 1.97 z/x. This analysis is general enough to include discrimination resulting from vapour pressure, ionization potential and diffusion constant differences, from multiplier discrimination, ionization processes generally and from other purely mass or momentum dependent phenomena. It does not include non linearities of collector or measuring system (which are purely amplitude dependent), geometrical masking of parts of the ion beam (which does not vary smoothly with mass) and pressure scattering (which varies with mass and amplitude). Lead Oxalate Analyses In this part of the study the single filament technique employed followed that of Cooper and Richards (1966). The sample 28 TABLE 4-1 LEAD OXALATE ANALYSES P b 2 0 6 P b 2 0 7 P b 2 0 8 P b 2 0 4 P b 2 0 4 P b 2 0 4 19.046 ± 0.02 16.000 ± 0.02 39.164 ± 0.04 19.124 ± 0.02 16.094 ± 0.02 39.471 ± 0.04 19.106 ±0.01 16.071 ± 0.01 39.421 ± 0.02 19.205 ± 0.01 16.176 ± 0.01 39.721 ± 0.02 18.700 ± 0.04* 15.590 ± 0.07 38.140 ± 0.17 18.690 ± 0.04* 15.580 ± 0.07 38.150 ± 0.15 18.710 ± 0.05* 15.660 ± 0.04 38.190 ± 0.24 New analyses of a single sample with the standard deviation of the mean reported. An asterisk indicates triple-filament analyses. 29a FIG. 3-1 FRACTIONATED LEAD OXALATE DATA Pb 16.3 207 Pb 204 16.2 REPLICATE ANALYSES OF REAGENT LEAD h ERROR BARS ONE STANDARD DEVIATION; 16.1 16.0 15.9 15.8 15.7 15.6 15.5 SINGLE FILAMENT 1 SLOPE = 1.11 ±0.03 TRIPLE FILAMENT 1 1 18.4 18.6 18.8 19.0 19.2 19.4 206, Pb*U7pb204 29b Pb Pb FIG. 3-1 FRACTIONATED LEAD OXALATE DATA 40.81 208 I REPLICATE ANALYSES OF REAGENT LEAD ERROR BARS ONE STANDARD DEVIATION 204 40.4 40.0 39.6 39.2 38.8 38.4 38.0 37.6 SINGLE FILAMENT SLOPE = 3.15 ± 0.1 TRIPLE FILAMENT 1 1 1 18.4 18.6 18.8 19.0 19.2 19.4 Pb 2° 6/p b204 30 FIG. 3 -2 FRACTIONATED STRONTIUM DATA 87 c 86 0.712 0.711 0.710 0.709 SLOPE = 0.042 ±0.0026 TRIPLE-FILAMENT ANALYSIS OF STRONTIUM REAGENT ERROR BARS ONE STANDARD DEVIATION I I _1 I 8.32 8.34 8.36 8.38 8.40 S r 8 8 / S r 86 8.50 DATA FROM BLENKINSOP (UNPUBLISHED) 31 material was reagent lead n i t r a t e , subsequently washed free from a l k a l i s with ethyl alcohol. The lead was converted to lead oxalate i n ammonium n i t r a t e solution, and then dried on an outgassed filament of rhenium. The t r i p l e filament technique as described by Catanzaro (1967) was also used. The sample, i n the form of lead hydroxide was loaded on rhenium side filaments. Ionization was effected by a platinum center filament at 1450°C "brightness temperature". The single filament spectra were scanned magnetically, and the output recorded on d i g i t a l magnetic tape. The computation of the isotope r a t i o s was done e n t i r e l y by computer using an adapta t i o n of the procedure of Weichert et.. a l . (1967) . In terms of the inter-analysis variations obtained, the ion beam measurement introduced i n s i g n i f i c a n t noise. The t r i p l e filament measurements were obtained using the peak-switching technique, described e a r l i e r Minor variations i n technique seemed not to affect the r a t i o s obtained. The analyses are shown i n Table 4-1 and plotted i n Figures 3-1 and 3-2. The straight lines i n the figures are the least squares slopes, calculated a f t e r York (1966). For comparison with theory i t i s convenient to integrate equations (5) and (6) to give log y = 1.5 log x + c log z = 2.0 log x + c' (4-7) (4-8) 32 Least square li n e s were f i t t e d to the logarithms of the isotopic r a t i o s . From the theory, the slopes on logarithmic plots should be 1.5 and 2.0, respectively. Table 4-2 shows the observed slopes for the data. For the y/x graph the experimental res u l t f a l l s below the th e o r e t i c a l r e s u l t by s i x standard deviations and for the z/x graph, by nine standard deviations. Therefore the mass spectrometer gives f r a c t i o n a t i o n slopes d i f f e r i n g from the theory at a confidence l e v e l well above 99.5%. Discussion Because i t i s very probable that a fra c t i o n a t i n g process would be a smooth function of mass, i . e . , one for which second order terms are n e g l i g i b l e , i t i s hard to believe that the theory derived can be inadequate. To check further on t h i s point quantitative checks for s p e c i f i c f r a c t i o n a t i o n processes were made. The ch a r a c t e r i s t i c s of the fr a c t i o n a t i o n , themselves, suggest a complex process. S p e c i f i c a l l y : (1) While i t i s comparatively easy to obtain constant isotope abundances during the analysis of a single filament loading (Doe, 1967), i t i s much more d i f f i c u l t to reproduce analyses loaded separately. . (2) The fra c t i o n a t i o n process depletes the l i g h t e r isotope i n preference to the heavier, which suggests that the fractionating step precedes the i o n i z a t i o n . 33 (3) The magnitude of the f r a c t i o n a t i o n , up to 2%, i s much larger than would be expected for a single-stage process f o r such a heavy element as lead. / 2 A fractionating law of the form M ' predicts slopes lower than 3/2. The expected cha r a c t e r i s t i c s of the residue from a Rayleigh d i s t i l l a t i o n can produce a range of slopes including the observed values. However, the magnitude of the fr a c t i o n a t i o n could not be reconciled with the observed slopes. Studies of s o l i d - s o l i d d i f f u s i o n and thermal d i f f u s i o n ended i n s i m i l a r r e s u l t s , although the l a t t e r examinations were quite s u p e r f i c i a l . In view of the d i f f i c u l t i e s i n understanding the r e s u l t s , i t i s appropriate to look for an explanation i n terms of experimental errors i n the measurements. The lead-204 peak i s the smallest and therefore the most susceptible to error. Lead-204 measurement uncertainties are of the order of 0.1% and therefore can only make a minor contribution to the change i n slope. By i t s e l f , lead-204 error would give slopes on logarithmic plots of unity, rather than 1.5 and 2. I f the two effects combine to give the observed l i n e s , they would have to. be correlated, which seems u n l i k e l y . For pressure scattering, i n the simplest case where scattering i s symmetrical and a fixed proportion of each peak appears under i t s immediate neighbours, the slope of the Pb /Pb vs 207 206 1 Pb /Pb (logarithmic) graph should be about - (It i s more d i f f i c u l t to predict the slope for the other r a t i o s , because of the greater separation of lead-204 from the rest of the spectrum, and 34 because of i t s different abundance.) c m 2 ° 8 , n u 2 0 6 nu 207 / m206 , For the Pb /Pb vs Pb /Pb graph, the slope predicted by the theory given is 2.0 for the logarithmic plot. The data presented in this paper gives slopes of 1.42 ± 0.2. The 204 discrepancy cannot therefore be attributed to Pb measurement error, and is unlikely to be due to pressure scattering. There remains to question whether this effect is particular to our mass spectrometers, or whether i t is a more general phenomenon. Some measurements by J. Blenkinsop shown in Figure ^ ''(unpublished data) can be used as a check of the validity of the theory for strontium analyses with the same instruments. The predicted slope on log-log plots i s 0.50, while the measured slope is 0.49. To provide some additional basis for comparison, we have also used data by Ulrych (unpublished), Doe (set 1, Doe et. a l . 1967) and Cooper and Richards (1966). The results of the calculations are found i n Table 4-2. It is d i f f i c u l t to reach conclusions about the individual sets, but the weighted mean of a l l data (1/weight a square of standard deviation) does not-change the conclusion. However, i t is d i f f i c u l t to evaluate the significance of this result, because i t is biased strongly towards data that was conveniently accessible and probably not representative. For example, Doe et. a l . (1967) report that newer data (for which numerical isotope ratios were not tabulated) do seem to f i t the fractionation model. It must be concluded from the available data that isotopic fractionation i s the dominant source of error in single filament 35 TABLE 4-2 a Logarithmic Slopes for Fractionated Data P b 2 0 7 P b 2 0 6 P b 2 0 8 P b 2 0 6 vs. vs. P b 2 0 4 P b 2 0 4 P b 2 0 4 P b 2 0 4 Calculated slope for fractionation (log-log plot) This paper This paper* Ulrych* Doe (Set 1) Cooper and Richards Weighted mean of last four values 1.318 ± 0.03 1.642 ±0.11 * These include t r i p l e filament as well as single-filament data. Slopes obtained from a least squares f i t on a log-log plot with the standard deviation of the mean reported. A l l data sets include single filament analyses, an asterisk indicates t r i p l e filament analyses also employed. Ulrych's data i s taken from a personal communication. For source of remaining data see DOE et. a l . (1967) and COOPER and RICHARDS (1966). 1.5 1.29 ± 0.04 1.32 ± 0.03 1.30 ± 0.04 1.27 ± 0.20 1.66 ± 0.18 2.0 1.59 ±0.11 1.53 ±0.05 1.84 ±0.09 1.73 ±0.27 2.14 ±0.16 36 lead isotope measurement by solid source mass spectrometry. In addition, some analyses show discrimination patterns that di f f e r significantly from the elementary theory. Clearly i t is advisable for laboratories using the double spiking technique to question the discrimination patterns of their instruments. Theory of Double Spiking A sample i s said to be double-spiked when there are added to i t two enriched isotopes in a fixed (known) proportion. Two analyses are required to obtain"fractionation free data by double spiking. Fi r s t the lead sample is analysed and data with fractionation of unknown magnitude is obtained. Then an analysis of the sample mixed with the double spike i s performed to determine the magnitude of the fractionation; the data obtained in this analysis also includes fractionation. The fractionated data may then be corrected on the basis of an appropriate fractionation law, provided the absolute ratios of the spike are known. The ratio of the added isotopes in the doubly-spiked sample w i l l be dominated by the known ratio in the double spike that has been added. Any difference between the known and measured ratio of the enriched isotopes in the mixture may be attributed to fractionation. This measure of fractionation may be used to correct the ratio of the two unspiked isotopes in the mixture. These two isotopes in the mixture are dominated by rthe sample, so that one of the ratios of the sample is now known. Since the magnitude of the 37 fra c t i o n a t i o n i s known a l l isotope r a t i o s of the unspiked analysis may be corrected for fra c t i o n a t i o n . In p r i n c i p l e , the spiked mixture has one isotope r a t i o determined by the double-spike and another independent r a t i o , by the sample. In practice, a l l isotopes are influenced by spike and sample compositions. Since, at the beginning of the correction c a l c u l a t i o n , the true i s o t o p i c composition of the sample i s unknown, an i t e r a t i v e procedure i s used. Convergence i s very rapid. Which isotopes to add i s determined by the least abundant isotopes i n the sample. Increasing the abundance of the least abundant isotope enables one to measure precise isotopic r a t i o s more e a s i l y . Lead-204 and lead-207 are the least abundant isotopes i n most lead samples encountered. The choice of lead-207 i s to be preferred to lead-206 as any pressure scattering effect on the r a t i o of the remaining p a i r of isotopes i s l i k e l y to produce a smaller error. C l e a r l y , the i d e a l double spike contributes only two isotopes so that the remaining isotopes are representative of the sample. However contributions to unspiked isotopes of ten to twenty percent can be e a s i l y allowed f o r without introducing errors i n the f i n a l sample composition, provided the composition of the spike i s exactly known. The equation on which the fr a c t i o n a t i o n correction was 206 based w i l l now be derived. The symbol Pb w i l l represent the measured ion beam in t e n s i t y corrected for errors that assume the 38 form of the f r a c t i o n a t i o n law, and free of any other error at the 0.1% l e v e l . The s u f f i x e s m, s, and sp w i l l i n d i c a t e mix, sample and spike r e s p e c t i v e l y . Ion beam i n t e n s i t i e s corrected f o r f r a c t i o n a t i o n , which w i l l be c a l l e d ' t r u e 1 , are r e l a t e d by the expression: (Pb 1)sp + (Pb 1)s m 'Pb 1 1 ^Pb 207 , n, 207. mi. 2 0 7-, (Pb )sp + (Pb )s which may be modified to read: PbJ 207 Pb , p b207. W h e r e ( ^ = fp.207 * (Pb )sp m fpb 1 ] n + [Pb 1 I U 2 0 7 J s L 2 0 7 J S £ 1 + Q (7) I f i = 204 t h i s expression gives the r a t i o Pb 204 Pb 207 m i n terms of the known value of t h i s r a t i o i n the spike, and the true 204> value of Pb *>Pb 207 , f o r which an approximate value i s known. Q may be estimated by s u b s t i t u t i n g i = 206 or 208 i n (7). The value of P b 2 0 4 l . i s then calculated from these q u a n t i t i e s . In t h i s way m Pb 207 f r a c t i o n a t i o n i n the mixture data i s reduced to one f i f t h of the f r a c t i o n a t i o n i n the sample data. (The f a c t o r of one f i f t h i s a funct of Q and the numerical value i s s p e c i f i c to the composition and r e l a t i v e q u a n t i t i e s of spike and sample used i n t h i s study.) 39 The difference between the observed and calculated values of Pb 204 lPb 207 are used as estimates of the fractionation in the mixture. m Then by using the relationship for fractionation, e.g. equations (5) and (6), the mixture analysis may be corrected for fractionation. The sample composition may be expressed in terms of the mixture and spike compositions. The most useful relationship i s : Pb 208 vPb J s P b 2 0 8 - P b 2 0 8  m sp P b 2 0 6 - P b 2 0 6 m sp which can be written f p b 2 0 8 l [Pb 2 0 8 1 [Pb 2 0 8] [Pb 2 0 8) l P b206j s U 2 0 6 J m U 2 0 6 J m U 2 0 6 J sp where • fpb 2 0 4 l fpb 2 0 8] ' s U 2 0 6 J sp U 2 0 6 J m obtained from (7) had fractiona-tion of about one f i f t h of the sample analysis. Since the second and third terms are small, are approximately known, and almost cancel, Pb 2 0 8) calculated has the same percentage of fraction-s the value of ation as Pb Pb 208 206 206 Fractionation effects are thus reduced by a factor m i n one iteration. of five Pb In the computer reduction seven iterations were employed although convergence was usually achieved after three. From the equations i t is easily shown that the uncorrected fractionation after a single iteration is directly proportional to the value of Q. This is a property of the sample and spike compositions in this particular 40 study. i Two enriched isotopes of lead were purchased for the double spiking, the one 89% lead-204 the other 92% lead-207. These were weighed out and dissolved to make stock solutions of 30 and 300 ppm of lead by weight. A mixture of these two, the 7/4 spike, was 207 204 prepared with a Pb /Pb ratio of ten. As attempts by Compston and Oversby (1969) and Cooper, Reynolds and Richards (1969) to determine the composition gravimetrically failed, this method was not employed. Calibration was performed by analysing the spike alone and a mixture containing the 7/4 spike and NBS equal atom standard. The analyses were repeated to improve r e l i a b i l i t y . The value of the standard reported by Catanzaro (1968), and the fractionated values of mixture and spike analyses were employed in the iterative procedure described above. The data before correction i s presented in Table 4-3. Corrected values and values obtained by the t r i p l e filament technique are presented i n Table 4-4. The results obtained by double spiking d i f f e r from the mean of the triple-filament values by about 0.1%. This agreement i s considered quite satisfactory for the purposes of this study. Table 4-5 shows the values of double-spiked and sample composition after each iteration. For the values lis t e d in Table 4-4 the theoretical fractionation slopes were employed. When the observed slopes, presented earlier i n this chapter, were used in the fractionation correction the agreement was definitely poorer. The calibration therefore 4 1 TABLE 4-2 b Logarithmic Fractionation Slopes for Rhenium and Tantalum P b 2 0 7 P b 2 0 6 P b 2 0 8 P b 2 0 6 P b 2 0 4 V S P b 2 0 4 P b 2 0 4 ^ P b 2 0 4 Calculated slope for fractionation (log-log plot) 1.5 2.0 Doe, Tantalum 1.47 ± 0 . 1 0 * 2.10 ± 0.20 Doe, Rhenium 1.35 ± 0.09 2.39 ± 0.30 * one standard deviation of the mean quoted favours the theoretical slope, i.e. the tantalum analyses of the two standards are fractionated according to the theoretical slope. The theoretical value was used throughout the study. Doe recently reduced his unpublished single filament data. The logarithmic slopes (personal communications) are given in Table 4-2b. The results provided independent evidence which indicates that rhenium filaments produce discrimination that does not f i t well the theoretical fractionation slope. Tantalum filaments do however produce slopes corresponding to the fractionation law. The use of the theoretical slope for tantalum is thus j u s t i f i e d by independent evidence. 42 TABLE 4-3 FRACTIONATED DATA FOR SPIKE CALIBRATION P b 2 0 6 P b 2 0 4 P b 2 0 7 P b 2 0 4 P b 2 0 8 P b 2 0 4 7/4 Spike 0.2866 10.095 0.6343 it 0.2854 10.076 • 0.6348 7/4 Spike and Equal Atom 3.1490 10.624 3.4780 II 3.1950 10.626 3.5260 7/4 Spike and Broken H i l l #1 1.6885 10.585 3.7840 Equal Atom 36.959 17.305 37.242 Broken H i l l #1 16.038 15.478 35.998 II 16.040 15.452 35.865 A l l data obtained by single filament sulphide technique, one standard deviation of mean less than 0.1%. , 43 TABLE 4-4 SPIKE, EQUAL ATOM AND BROKEN HILL #1 COMPOSITION P b 2 0 6 P b 2 0 7 P b 2 0 8 P b 2 0 4 P b 2 0 4 P b 2 0 4 *7/4 Spike 0.2841 10.013 0.6296 *Equal Atom 36.702 17.125 36.725 *Broken H i l l #1 15.995 15.386 35.662 •Broken H i l l #1 Triple Filament 16.002 15.384 35.609 11 16.007 15.452 35.723 " 16.015 15.403 35.750 *Equal Atom Triple Filament 36.738 17.160 36.745 * Compositions from Table 4-2 corrected for fractionation * Triple filament analyses this study, one standard deviation of mean less than 0.1%. * Catanzaro's t r i p l e filament value used i n 7/4 spike calibration calculation. 44 TABLE 4-5 VALUES AFTER EACH ITERATION FOR FRACTIONATION CORRECTION BROKEN HILL #1 BROKEN HILL #1 AND 7/4 SPIKE P b 2 0 6 P b 2 0 7 P b 2 0 8 P b 2 0 6 P b 2 0 7 P b 2 0 8 P b 2 0 4 P b 2 0 4 P b 2 0 4 P b 2 0 4 P b 2 0 4 P b 2 0 4 16.040 15.452 35.865 1.688 10.585 3.784 15.999 15.393 35.682 1.679 10.494 3.741 15.995 15.387 35.664 1.678 10.490 3.739 15.995 15.386 35.662 1.678 10.490 3.739 15.995 15.386 35.662 1.678 10.490 3.739 Each l i n e of data represents one complete i t e r a t i o n . Three further i t e r a t i o n s produce no further change. 45 CHAPTER V SAMPLE LOCATIONS AND DESCRIPTIONS Vogt-Hobbs Area, Ontario This area i s a s t r i d e the G r e n v i l l e Front near Lake Timagami. Rocks from t h i s v i c i n i t y are p a r t i c u l a r l y s u i t a b l e f o r a study of the behaviour of uranium-lead systems because of the p o s s i b i l i t y of observing the e f f e c t of the increasing metamorphism towards the south. The G r e n v i l l e Front i s defined here p r i n c i p a l l y by the t r a n s i t i o n from a metamorphic grade of the greenschist f a c i e s i n the north to the amphibolite f a c i e s i n the south. Grant was able to trace rock units across t h i s boundary and locate equivalents on the basis of l i t h o l o g y . It i s the leucogranite to the north and some of i t s equivalents to the south that were sampled. For t h i s f i r s t study, rocks expected to contain large quantities of uranium and lead are to be preferred. Examination of the l i t e r a t u r e shows that acid rocks generally contain more lead and uranium than more basic rocks. A more s p e c i f i c i n v e s t i g a t i o n by Doe (1967) indicates that lead content depends on feldspar type and abundance. Potassium feldspars are found to contain from 20 to 60 ppm t y p i c a l l y , while c o e x i s t i n g sodium-calcium feldspars contain about one t h i r d the concentrations. For t h i s reason the more access-i b l e a c i d i c rocks were c o l l e c t e d . A regional and a s i m p l i f i e d geologic r map (Figure 5-1) show the general l o c a t i o n of the region and serve to i d e n t i f y the sample l o c a t i o n s . 46 The samples were a l l collected from outcrops close to the logging road shown in Grant's more detailed map (Grant, 1964). As he personally supervised the collection of the Ontario samples studied i n this Thesis, i t was possible to sample the same sites as were used for his Rb-Sr study. Portions of the following sample descriptions are taken from Grant, 1964. Gl was taken from the south side of a leucrocratic albite stock two miles i n diameter called the Vogt granite. This pink, medium-grained, albite granite contains less than 5% green bio t i t e . The potassium feldspar, microcline microperthite, may form phenocrysts. Accessory minerals include muscovite, chlorite, epidote, sphene, apatite, zircon and opaque minerals. This is the only sample from north of the Grenville Front. A similar granite forms a belt about four miles long and a half mile wide centered on the south end of Sinton Lake and astride the Grenville Front. Sample G2 is representative of the pink leucocratic albite granite phase of this body while G3 represents the quartz monzonite to granodiorite phase. CSCH was taken from a vein of pegmatite rich i n potassium feldspar, and included small traces of the surrounding quartz-biotite-plagioclase schist. The sample from the vein was collected because of i t s potentially high lead content. The surrounding rock is described as migmatite and occupies much of this region. The remaining sample G5 was also thought to be derived from the same original granite, shows cataclasis and is also a quartz monzonite. i F I G U R E 5-1 R E G I O N A L AND S I M P L I F I E D G E O L O G I C MAPS V O G T IIOBBS A R E A MAPS T A K E N FROM GRANT ( 1 9 6 4 ) x X.-C71 • — — • ' • / - " H i-5] Oiobose 1. '. II Huronion sediments f* 1^ Gronite and equivalent gneiss G1 Sample locality Fault I / — •• — Mctamorphic N tront I*. *.\ QuorU diorite r~rr-\ Kccwotin-type rocks and ^ equivalent schist EL3 Migmotite 48 Rice Lake-Beresford Lake Area, Manitoba ' This area has been of particular interest to geologists since 1911 on account of the gold discovery at that time. A map showing the location and general geology of the area is presented in Figure 5-2 (based on Stockwell and taken from Turek, 1968) . The region consists of a potassic granite in the north and a gneissic belt in the south. The Rice Lake group lies between these two and has been intruded by basic dykes and s i l l s which have in turn been intruded by quartz diorite plutons. Samples analysed from this area were from the northern potassic granite and the sample locations are marked on Figure 5-2. Each of the hand specimens was stained with sodium cobaltinitrate to determine the potassium feldspar content. The rocks were then class-i f i e d according to their quartz and potassium feldspar content. Samples 29 and 47 may be approximately classified as quartz diorite while 78, 207 and 214 are granodiorites. A l l of the Manitoba samples are di s t i n c t l y more gneissic i n structure than the Ontario samples, and show considerable inhomogeniety. There are also indications of replacement of potassium feldspars or potassium feld-spars replacing other feldspars. On the average the Manitoba samples contain approximately 10 to 15% potassium feldspar» this is slightly less than one half of the potassium feldspar content of the Ontario samples. FIGURE 5-2 REGIONAL AND SIMPLIFIED GEOLOGIC MAPS RICE LAKE-BERESFORD LAKE AREA A on L L G L N D Q«ort, <i>wif* Oobbro. d loboM, diorit* V»lco< R*. Lot.* ' c U Soni|)J« location » ftu*bar , v - ' \ ..." c MAP TAKEN FROM TUREK 1968 50 TABLE 5-1 UNCORRECTED SINGLE FILAMENT ROCK LEAD ANALYSES P b 2 0 6 P b 2 0 7 P b 2 0 8 P b 2 0 4 P b 2 0 4 P b 2 0 4 29 18.766 15.656 36.211 47 15.266 14.934 34.863 78 16.169 15.158 35.412 207 16.697 15.225 38.806 214 16.933 15.402 37.464 Gl 23.399 16.564 40.831 G2 27.411 17.137 39.863 G3 19.834 15.938 39.413 CSCH 21.655 16.343 36.826 G5 14.832 15.128 35.192 New analyses of rock lead samples, standard deviation of mean less than 0.1%. The analyses were carried out with the single filament sulphide technique. This is the data before the double spiking corrections. 51 TABLE 5-2 UNCORRECTED DOUBLE SPIKED SAMPLE ANALYSES P b 2 0 6 P b 2 0 7 P b 2 0 8 P b 2 0 4 P b 2 0 4 P b 2 0 4 29 3.0700 10.907 5.983 47 1.6880 10.522 3.846 78 1.9146 10.604 4.192 207 2.3002 10.711 5.324 214 1.9847 10.652 4.390 Gl 1.9810 10.586 3.597 G2 2.9840 10.773 4.542 G3 2.2996 10.675 4,624 CSCH 3.5970 11.048 6.247 G5 1.6469 10.542 3.863 New analyses of rock lead samples mixed with double-spike, standard deviation of the mean less than 0.1%. Analyses carried out by single filament sulphide technique and not corrected for fractionation. 52 1 TABLE 5-3 ANALYSES OF ROCK LEAD SAMPLES CORRECTED FOR FRACTIONATION P b 2 0 6 P b 2 0 4 P b 2 0 7 P b 2 0 4 P b 2 0 8 P b 2 0 4 29 18.651 15.512 35.768 47 15.247 14.906 34.776 78 16.040 14.977 34.848 207 16.634 15.140 38.516 214 16.793 15.211 36.844 Gl 23.343 16.504 40.635 G2 27.345 17.075 39.670 G3 19.723 15.804 38.972 CSCH 21.579 16.257 36.569 G5 14.746 14.996 34.783 The above data were obtained by correcting the data of Table 4-1 for fractionation with data from Table 4-2. A l l Manitoba samples have entirely numerical labels. 53 New Rock Lead Data A l l isotopic composition analyses for rock lead data were obtained by using the single filament sulphide technique. The raw sample and double-spiked sample analyses are reported in Tables 5-1 and 5-2. The rock lead compositions were corrected for fractionation by assuming a fractionation law of the form; ^ = L S I and & = 2.0^ dx x dx x The standard deviation of the mean, for the raw data, was less than 0.1% in a l l cases and typically 0.05%. Errors in the three sample ratios, the three double spiked ratios and the three spike ratios, a l l contribute towards the errors in the fractionation-corrected data. A numerical experiment was performed to determine the effect of errors in the sample, spike and mixture ratios. Perturbations of + 0.1% were applied to each of the nine ratios in turn, the other eight ratios being free of this perturbation. A l l nine ratios were chosen to incorporate fractionation of 0.13% per mass unit. A second study was also performed in which the perturbation was + 1.0% and the fractionation 1% per mass unit. The percentage change in a sample ratio divided by the percentage perturbation is called the error magnification for that ratio. The error magnifications were found to be the same for the two cases studied and are given in Table 5-4. 54 TABLE 5-4 ERROR PROPAGATION FOR BROKEN HILL #1 ^^.ERROR MAGNIFICATIONS ^ \ F O R + PERTURBATION P b 2 0 6 P b 2 0 4 P b 2 0 7 P b 2 0 4 P b 2 0 8 P b 2 0 4 *SAMPLE RATIO RAISED: 206/204 + 2.1 .+ i-7 + 2.2 207/204 + 0.1 + 1.2 + 0.3 208/204 - 1.1 - 1.7 - 1.2 DOUBLE-SPIKED RATIO RAISED: 206/204 - 1.3 - 2.0 - 2.7 207/204 - 0.9 - 1.5 - 1.9 208/204 + 1.3 + 2.0 + 2.6 SPIKE RATIO RAISED: 206/204 + 0.2 +. 0.3 + 0.3 207/204 + 0.8 + 1.3 + 1.5 208/204 - 0.2 - 0.3 - 0.4 f The error magnifications are tabulated under the reduced sample ratios to which they refer. * A l l ratios incorporated fractionation and only one ratio was raised at a time. 55 The fact that the error magnification was not found to be a function of the magnitude of the f r a c t i o n a t i o n , indicates that large f r a c t i o n a t i o n corrections can probably be made as precisely as small corrections. However i n t h i s study the use of tantalum filaments was preferred because of the small magnitude of the f r a c t i o n a t i o n and the reduced l i k e l i h o o d of large errors from correc-tions f o r f r a c t i o n a t i o n . The source of the errors i n the reduced data can be divided i n t o two parts. The error introduced by the spike r a t i o s determines the absolute c a l i b r a t i o n of the f i n a l data and i s a constant value for a l l of the analyses. This error arises when the spike i s calibrated. The agreement between the double spiked measurements and the t r i p l e filament values of the standards (see Table 4-2) indicates that error from t h i s source i s probably less than 0.1%. The second portion of the error varies from analysis to analysis and determines the r e p r o d u c i b i l i t y . A numerical estimate of the standard deviation f o r the reduced data may be obtained by considering any reduced isotope r a t i o , R^ , to be a function of the s i x measured isotope r a t i o s R^, R2, ... R^ , employed i n the f r a c t i o n -ation correction: = f(R^ j R^). The standard deviation, a, of such a compound quantity i s a function of the standard deviations, a.., ... a, of the s i x r a t i o s : 56 I f the quantities 3f 3R. 1 are approximated by the error magnifications 3f 3R. 1 l i s t e d i n Table 5-4 then for a standard deviation a. of 0.05% for I each r a t i o the standard deviation, a, for the reduced data w i l l be 0.22%. Clearly t h i s i s an oversimplication as the quantities were i d e n t i f i e d with the value of t h i s quantity when a l l but one r a t i o i s correct. I t i s i n s t r u c t i v e to consider a special case to show how errors i n the s i x r a t i o s may actually i n t e r a c t . The most common form of error i n lead isotope r a t i o determinations results from measurement errors f o r the lead-204 peak. I f such an error occurs then a l l r a t i o s to lead-204 experience the same percentage change. I f the mixture analysis incorporates lead-204 error, then a p a r t i a l cancellation of errors could be anticipated (see Table 5-4). This effect was v e r i f i e d by numerical t e s t s . A s i m i l a r s i t u a t i o n arises fo r the sample analysis. Thus i f error of measurement of the lead-204 peak i s the only source of error i n the sample and double-spiked analysis then equation 5-1 i s e f f e c t i v e l y reduced to two terms. In t h i s case, the estimated standard deviation f o r the f i n a l data, from an i n i t i a l standard deviation for the raw r a t i o s of 0.05%, i s 0.1%. Numerical reductions, incorporating lead-204 error, gave errors i n the reduced sample r a t i o s i n agreement with the predicted errors. There are two checks of the r e p r o d u c i b i l i t y of the data. 208 206 The f i r s t of these results from the measurement of the Pb /Pb i n the mixture, which should relate simply to the same r a t i o i n the sample. During the f r a c t i o n a t i o n correction the difference between 57 208 206 the Pb /Pb ratio i n the corrected mixture ratios and the uncorrected sample ratios is used indirectly as a measure of 208 206 fractionation. If the Pb /Pb ratio were not reproducible then there would result a much larger range and average magnitude of fractionation. The average fractionation of 0.215% per mass unit i s in close agreement with fractionations of 0.13% and 0.15% per mass unit obtained by Doe (1967) and Zartman (1969) for lead sulphide loadings on tantalum (the standards analysed by these two workers have been analysed on t r i p l e filament by Catanzaro, 1969). The magnitude of the average fractionation in this study is higher than the reported fractionation for the standards. However i t was found during this study that the analyses of rock leads generally showed greater fractionation than the analyses of standards. This is probably a function of the purity of the samples. The second reproducibility check arose because concentrations were measured separately by using radiogenic lead, rather than combining the concentration determination with the double spiking correction. Two concentrations were determined, one i s based on the i D l 208 / m206 , « . u ' i m 207 / m206 sample Pb /Pb ratio and the other on the sample Pb /Pb ratio. The concentrations calculated from the two ratios agreed to within the precision of the measurements. This implies that the 208 207 sample Pb /Pb ratios were reproducible to about 0.2% or better. The fact that the double spiked analysis was not used to determine lead concentrations also l e f t one unused equation in the fractionation correction. This equation was used to obtain a second 58 estimate of the proportion of double spike and sample i n the mixture. The two estimates were averaged i n the reduction and were equal at the end of the reduction. C l e a r l y t h i s increases the r e l i a b i l i t y of the correction. Compston and Oversby (1969) claim that f r a c t i o n a t i o n may be corrected by double-spiking to give a sample composition whose 95% confidence l i m i t s are ±0.1%. In that study, one sample analysis was corrected for f r a c t i o n a t i o n by using r e p l i c a t e double spiked analyses. Their uncertainty therefore only r e f l e c t s errors a r i s i n g from the double spiked analysis, whereas sample analyses tend to have a larger standard deviation. Consequently the expected 95% confidence l i m i t s of the reduced data f o r pairs of sample and double-spiked analyses, each contributing errors, would be about double t h i s value, or ± 0.2%. In conclusion the technique used i n t h i s thesis i s believed to be as precise as that of Compston and Oversby (1969), but studies of propagation of errors y i e l d a more pessimistic view of the pr e c i s i o n available from data of t h i s q u a l i t y . * For interpretations of t h i s data a r e p r o d u c i b i l i t y of ± 0.3% (95% confidence level) w i l l be assumed. In addition the uncertainty of the composition of the double spike adds a possible bias of ± 0.1% which i s not s i g n i f i c a n t f o r the interpretations. Note that Compston and Oversby's P b 2 0 6 / P b 2 ^ 4 r a t i o s and Pb 2 (^ 8/ 204 Pb r a t i o s each d i f f e r from the t r i p l e filament values by 0.2%. 59 Uranium and Lead Concentration Determinations Lead concentrations were calculated from the ratios obtained from samples spiked with radiogenic lead; uranium-235 was added to determine uranium concentrations. The details of the calculation procedure are given in the Appendix. Optimum spiking curves are also found there. It is quite easy to perform precise mass spectrometric analyses for the lead concentration determinations, as only the abundant isotopes, lead-206, lead-207 and lead-208, need be measured. Because sensitivity i s not a serious problem i t was possible to do these analyses by t r i p l e filament. Before discussing the precision of the concentration measurements a comment w i l l be made on the precision required. The quantities of greatest importance that are derived from the concentra-*u , r 235 / n u204 , ..238/m 204 tions are the ratios U /Pb and U /Pb . The ratio 238 204 U /Pb varies from 0.4 to 20 whereas the lead isotope ratio 207 204 Pb /Pb varies from 15 to 17. Consequently i f the relative errors in the concentrations are ten times the relative errors in the isotope ratios, then the relative errors in the variable portions of these quantities are about the same. In view of the precision and accuracy of the ratio determinations, concentrations measurements correct to 1% were attempted. Reproducibility of lead concentrations can be estimated from Table 2-1. The total variation i s less than 0.5% for the conditions employed in this study. Concentrations of lead and uranium for the two suites analysed in this study are listed in Table 5-5. 60 TABLE 5-5 URANIUM AND LEAD CONCENTRATIONS 29 10.49 1.992 47 7.79 0.469 78 17.19 1.456 1.457 207 10.74 1.050 214 15.12 1.456 1.460 Gl 29.95 11.78 G2 37.12 2.118 G3 21.337 5.626 CSCH 23.89 6.467 G5 38.88 0.225 GSP-1 2.207 2.206 Lead and uranium concentrations are total lead and total uranium in parts per million by weight. One standard deviation i s 0.25% for a concentration determination. 61 Uranium concentration determinations show equally good r e p r o d u c i b i l i t y as can be seen for the three duplicate analyses. The uranium concentration determined for GSP-1 i s about 8% less than the preferred value given by Doe et. a l . (1967). However as his values vary from 2.2 to 3.2 ppm an 8% discrepancy i s of doubtful si g n i f i c a n c e . Clearly the present r e p r o d u c i b i l i t y i s s i g n i f i c a n t l y better than Doe et. a l . (1967). Uranium isotopic r a t i o determinations are subject to frac t i o n a t i o n errors even when carried out by a t r i p l e filament method. Unlike lead, t h i s f r a c t i o n a t i o n i s reported to lead to an enrichment i n the l i g h t e r isotope (NBS Technical Note, 1966). A correction was made i n the concentration calculation corresponding to an enrichment of 0.4% i n the l i g h t e r isotope. Although t h i s correction i s only an estimate on the basis of reported fra c t i o n a t i o n t h i s correction should reduce the magnitude of any such error to n e g l i g i b l e s i z e . Regulations did not permit the purchase of the large quantities of uranium supplied for standards suitable for measuring f r a c t i o n a t i o n . For the purposes of computation the 95% confidence l i m i t s for a concentration determination are ± 0.5%. This includes the error magnification and spike c a l i b r a t i o n . 62 - CHAPTER VI INTERPRETATION OF ROCK LEAD DATA In order to complete the f i r s t objective of this research the data w i l l be interpreted i n the light of the lead-lead and lead-uranium models of Russell and Farquhar (1960), Ulrych and Reynolds (1966), Ulrych (1967) and Russell et. a l . (1968). It i s then possible, by using other published data, to compare the results for the uranium-lead and rubidium-strontium systems for the two loc a l i t i e s investigated. Finally, i t is possible to look at the interaction of geologic events and uranium-lead systems. Lead-Lead Interpretations The rock samples analysed are equivalent to common leads mineralized at the present. For common leads i t i s found most • * n ^ n l 207 / m 204 m 206 / m204 . , - ,^ instructive to plot Pb /Pb vs Pb /Pb as is done for the Manitoba and Ontario data i n Figures 6-1 and 6-2. Russell and Farquhar (1956, 1960) have shown that, for samples whose radiogenic component was produced between t^ and t^, and which contain the same common lead component, a linear relationship results on the 207 204 n l 206 / n l 204 , _ „ n , , ^  . Pb /Pb vs Pb /Pb plot. The slope, R, on such a plot i s given by: 'exp (X't^ - exp (Xt 2)' exp (Atp - exp (Xt 2) (All new symbols i n this chapter are defined in Table 6-1.) a (6-1) FIG. 6-1 ISOTOPIC ABUNDANCES RICE LAKE-BERESFORD LAKE AREA Pb 42 208 Pb Pb Pb 207 204 I6h 204 38 34 + 214+ 207 K » 3 4 . 3 * + ' + 4 7 , + 7 8 15 1 16 17 2000 my 29 , 8 f9 P b 2 0 6 / P b 2 0 4 /JL - 8.79 Omy + 29 SLOPE* 0.I86±0.0IS a s 14. 13 14 15 16 17 18 19 20 P b 2 0 S / P 5 2 0 4 21 FIG. 6-2 ISOTOPIC ABUNDANCES VOGT-HOBBS AREA Pb 42 2 0 8 Pb Pb Pb 2 0 7 2 0 4 1 8 16 2 0 4 38 34 / + Gl + G3 / K-34.3 + CSCH I + G2 14 18 fj.» 8.79 2000my J O O O ™ O m £ 22 26 P b 2 0 6 / P b 2 0 4 30 * G2 SLOPE" O.I68±0.008 14 14 16 18 20 22 24 26 p u 2 0 6 P b /Pb 28 204 30 65 For leads mineralized at the present, = 0. The points shown were f i t t e d with least squares straight lines as described by York (1966). Even though rather pessimistic figures for the 95% confidence limits of the data points are shown in Figures 6-1 and 6-2, ' Ontario data s t i l l appear to differ significantly from a straight line. Although the number of points i s very small, a numerical estimate of the goodness of f i t can be made. Assuming 95% confidence limits of 0.3% for the data, the chi-square test indicates that there is a 25% probability of exceeding the value of chi-squared for the Manitoba suite and a probability of 1% of exceeding the value of chi-squared for the Ontario suite. This signifies that the Manitoba suite approximates the Russell-Farquhar model. The Ontario data cannot reasonably be considered to be even a poor s t a t i s t i c a l sample from a population that f i t s the Russell-Farquhar model. It w i l l be shown later, however, that the data does admit interpretation. r For the Manitoba potassic granites the inferred age, based on the observed slope, is 2750 ± 150 my, where the error limits are one standard deviation (as elsewhere in this thesis unless otherwise stated). The standard deviation quoted is based on York's estimate and represents the f i t of the data points to the line,and not the standard deviation of the slope based on the precision of the data points (Williamson, 1968). As York's estimate i s larger than Williamson's this is a more pessimistic (though in fact more re a l i s t i c ) estimate of the standard deviation of the age. This procedure is used throughout the remainder of this chapter. 66 TABLE 6-1 Symbols and Constants Used i n Age Determination Isotope At present At time At time Ratio t = 0 t t o P b 2 0 6 p b204 a x ao P b 2 0 7 P b 2 0 4 P b 2 0 8 P b 2 0 4 P b 2 0 4 b o p b204 ^ * Co u 2 3 8 Xt Xt ye ye o _ _ _ y/o y / a e X , t y e X , t o P b 2 0 4 232 T h ^ v • • X» • „ X"t Ke t Ke o - 9 - 1 Decay Constants Value i n 10 (years) Parent Atom X 0.1537** U 2 3 8 X' 0.9722 U 2 3 5 X" 0.0499 T h 2 3 2 *a = U 2 3 8 : U 2 3 5 = 137.8 : 1 (Inghram, 1947) ** Kovarik, A.F. and Adams, N.E., J r . , Phys. Rev. 98, 46, 1955. Fleming, E.H., J r . , Ghiorso, A., and Cunningham, B.B. Phys. Rev. 88, 642, 1952. P i c c i o t t o , E. and Wilgain, S. Nuovo Cim. Ser. 10, 4, 1525, 1956. 67 The slope for the Ontario suite gives an age of 2,580 ± i 80 my. The smaller standard deviation i n the slope of the l i n e i s not an in d i c a t i o n that the Ontario data f i t s the assumptions of the model better than the Manitoba data; as has already been mentioned the converse i s true. The smaller standard deviation i n the Ontario data i s a consequence of a larger range of isotope r a t i o s i n the Ontario s u i t e . More can be said about the lead-lead p l o t when an i n i t i a l composition f o r the common lead i s available for these suites. This matter w i l l be discussed further when the common lead component has been obtained from the uranium-lead interpretations. Uranium-Lead Interpretations Ulrych and Reynolds (1966) have shown that whole rock uranium-lead data may be interpreted i n a manner analogous to rubidium-strontium whole rock interpretations. For rocks s t a r t i n g from a common i n i t i a l P b 2 ^ / P b 2 ^ 4 r a t i o , and for which the rock has remained closed to the transfer of lead and uranium, the present m 206, m 204 . . , . Pb /Pb r a t i o i s given by the expression: f P b 2 0 6 lPb204 \ f P b 2 0 6 ] f u 2 3 8 1 J p U 2 0 4 J i t U2 0 4 J ( e X + - l ) , (6-2) 207 204 with a s i m i l a r expression for Pb /Pb . I f a suite of rocks s a t i s f i e s the assumptions then a l i n e a r relationship results with a slope, R, which depends only on t ; t can be obtained from: t = j- In (1 + R) . (6-3) 68 When the Manitoba data, Figures 6-3 and 6-4, are i n t e r -preted i n t h i s manner the ages from the lead-2 0 7 - uranium-235-lead-206 - uranium-238 plots are 2,626 ± 200 my and 2,238 ± 205 my respectively. As the standard deviation of the age based on the f i t of the points to the l i n e i s greater than the standard deviation based on the precision of the data, the model i s only considered to be an approximation to the history of the samples. Similar plots of the Ontario data show that the data as a whole cannot be approximated by the single stage model. The slopes, ignoring G2, correspond to ages of 2,230 ± 350 my and 1,874 ± 400 my respectively for the lead-207 and lead-206 p l o t s . I f these samples are treated as uranium minerals then a 207 206 Pb /Pb age may also be determined. This age i s not independent of the two lead uranium ages just determined. However to make such an age determination i t i s necessary to correct for common lead. This has already been done i n the Russell-Farquhar model by assuming the common lead-component to be i d e n t i c a l for a l l samples. Another method of correcting for the common lead i s employed i n the modified concordia p l o t . The samples are assumed to have a two stage h i s t o r y of which the f i r s t stage i s represented by the common lead present and the second stage by radiogenic addi-tions from uranium now found i n the rock. I t i s assumed i n the 207 206 model that the Pb /Pb r a t i o of the common lead incorporated i n each rock i s the same for a l l samples. Ulrych (1967) (see also Russell et. a l . 1967) have shown that i f the quantities: 69 FIG. 6-3 URANIUM-235- LEAD-207 PLOT Pb 16.0 207 Pb 204 15.8 14.6 14.4 0 RICE LAKE-BERESFORD LAKE AREA SLOPE = !l .85±2.74 INTERCEPT = 14.62 1 0.02 0.04 0.06 0.08 U 2 3 5 / P b 2 0 4 0.10 70 FIG. 6-4 URANIUM-238 — LEAD-206 PLOT RICE LAKE - BERESFORD LAKE AREA Pb 19.0 206 Pb 204 18.0 17.0 16.0 15.0 * — 29 — 214 g / *- / 207 SLOPE = 0.411 ±0.047 INTERCEPT = 14.09 / IB 47, / 1 1 1 1 2.0 4.0 6.0 8.0 10.0 12.0 U 2 3 8 / P b 2 0 4 72 FIG. 6-6 URANIUM-238 ~ LEAD-206 PLOT VOGT-HOBBS AREA Pb 30.0 206 Pb 204 26.0 24.0 22.0 20.0 18.0 16.0 14.0 (*G 2) — if/ f G 1 CSCH / * / SLOPE =0.3341.052 INTERCEPT = 14.6 — — / / G 3 ^G 5 1 1 1 1 0 10 20 30 40 U 2 3 8 / P b 2 0 4 50 73 Pb 206* ,238 Pb 206 Pb 204 f p b 2 0 4 ] J i u 2 3 8 v. J 207* [pb2°i are p l o t t e d as absissa and ordinate r e s p e c t i v e l y , a l i n e a r r e l a t i o n r e s u l t s , i f the assumptions of the model are j u s t i f i e d f o r the s u i t e of rocks." The Ontario and Manitoba data was p l o t t e d i n t h i s manner and least squares s t r a i g h t l i n e s f i t t e d (Figures 6-7 and 6-8.) . The time value f o r the upper intercept, of the l i n e a r r e l a t i o n with the concordia, i s i n t e r p r e t e d as the age of the earth and the lower intercept time value as the time of commencement of the l a s t stage. The s c a t t e r of the data about these l i n e s indicates that the lead uranium system does not s a t i s f y the two stage model very w e l l . The i n f e r r e d commencement of the l a s t stage f o r Manitoba and Ontario suites are 2430 ± 576 my ago and 1640 ± 65 my ago r e s p e c t i v e l y . A summary of the i n f e r r e d ages i s given i n Table 6-2. The data from Manitoba shows the pattern lead-lead age > uranium 235-lead 207 age > uranium 238-lead 206 age . This f i t s the pattern observed f o r loss of radiogenic lead. The lead-lead age i s known to o be least a f f e c t e d . (Lead Isotopes i n Geology, 1960). The Manitoba data did show a larger s c a t t e r on the P b 2 0 6 / P b 2 0 4 207 204 versus Pb /Pb p l o t than would be expected f o r the p r e c i s i o n of the data. This could be the consequence of d i f f e r i n g compositions 74 TABLE 6-2 Interpretative Ages MODEL ONTARIO MANITOBA Russell-Farquhar 2580 ± 80 my 2750 ± 150 my Modified Concordia 1640 ± 65 my 2430 ± 576 my Uranium-Lead u 2 3 5 - P b 2 0 7 u 2 3 8 - P b 2 0 6 2230 ± 350 my 1874 ± 400 my 2630 ± 200 my 2240 + 200 my Time limits correspond to one standard deviation for f i t of data. 77 for the common components incorporated at the beginning of the last stage. Differing i n i t i a l components for the last stage are best explained as the result of a three stage history. If this i s in fact the case, then the modified concordia plot is probably not suitable for representing these samples. The Manitoba data is then consistent with a three stage history. It remains to c l a r i f y that history. In order to discover more about the second stage i t is desirable to obtain better estimates of the i n i t i a l composition for the last stage. The intercepts on the uranium-235 - lead-207 and uranium-238-lead-206 plots are included on Figure 6-9. These however are poor estimates because the lead-uranium slopes have been reduced by lead remobilization. If instead intercepts are calculated on the basis of the lead-lead age and the centroid of the data on the uranium-lead plot lower intercepts result. These are also shown in Figure 6-9 . The nearest ore lead data is also plotted on 6-9 from Manitouwadge (Ostic et. a l . , 1967). Clearly the Manitouwadge data, including the Geco vein, f i t the anomalous lead line very well. (The analyses of Ostic were corrected for sample-line fractionation). The preferred model for these samples is depicted in Figure 6-10. The time of formation of the second-stage is greater than 3000 my and a best estimate based on an anomalous lead line passing through ordinary lead 2500 my old and the Manitouwadge like composition is 3500 my. The value of 2500 my is the youngest event known to have FIG. 6-9 ISOTOPIC ABUNDANCES OF SUITS STUDIED AND RELATED DATA Pb 18 207 Pb 2 0 4 17 16 15 13 * 78 300C ^ 4 7 ^MAN ITOUWADGE GECO SLOPE « 0.168+0.008 .Gl. fj. = 8.79 1000my O m v — +" 2000my ^---~~^>^^" v"29 CSCH 0.187 ± 0 . 0 0 4 o INITIAL COMPOSITION FROM URANIUM-LEAD INTERCEPT X BEST ESTIMATE OF INITIAL COMPOSITION © COBALT -NORANDA ORES 13 15 17 19 23 25 27 ~/Pb: 29 Pb 2 0 6/r^,. 204 co 79 affected the northern granite (Turek et. a l . , 1968). If the common lead incorporated in the Manitoba suite is to be produced as recently as possible, then the formation of the crust in this region and the production of the common lead component may have taken place at 3000 my. This is the minimum age for this event. These interpretations imply that the Manitouwadge ore leads were derived from material with a crustal history. This interpretation is in contrast to the arguments of Tilton and Steiger (1965) who employ Manitouwadge and related leads, to calculate an age for the earth of 4700 my. In their argument, they assume that Manitouwadge i s a 2700 my old primary lead that has not been contaminated by crustal material. j The slope of the anomalous lead line, including the Rice Lake-Beresford Lake samples, the Manitouwadge sample and the Geco vein sample, is not significantly different from the value without the ore lead sample. If the last stage in the history of the ore lead samples i s genetically as well as isotopically related to the rock lead samples, then the Geco sample was emplaced quite recently in the Fox Creek fault. It is possible that the Geco lead has been extracted recently from the rock lead adjacent to the fault. This is a closer approximation to the f i r s t argument put forward by c Ostic (1963). The older age determined for the composite anomalous lead line, t^, is then 2760 ± 40 my, for t 2 = 0. There i s abundant evidence for an event at this time in this area (e.g. Turek and Peterman, 1968; Goldich et. a l . , 1962). 80 TABLE 6-3 Observed and Calculated y Values For Two Stage Model y 2 y j Value - ,,238 n, 206 from U -Pb u Value from U 2^ 5-Pb 2 0 7 MANITOBA 29 11.51 6.94 7.647 47 3.39 7.399 7.574 78 4.83 7.615 7.507 207 5.95 6.245 7.669 214 5.75 8.205 7.813 ONTARIO Gl 27.386 7.923 7.990 G2 4.149 21.653 9.786 G3 18.366 6.591 7.421 CSCH 17.437 9.334 8.153 G5 0.3675 6.820 6.891 y 1 and u2 are the values of ,.238/nu204 _ ... U /Pb for the f i r s t and second stages respectively extrapolated to the present. 81 Emplacement of the Geco lead at an earlier time would imply-that the Geco lead, though isotopically related, is not genetically related to the rock leads. This situation, which seems improbable isotopically, does not alter the argument put forward for the Rice Lake-Beresford Lake region. The i n i t i a l lead isotope composition of.the Ontario suite, for the last stage, was also calculated on the basis of the uranium-235-lead-207 and uranium-238-lead-206 plots, in the manner used for Manitoba. It i s questionable whether this procedure is valid for the Ontario suite because of inadequate sampling for rocks that have experienced lead remobilization. From the Figure (6-9), and in view of the low uranium content, G5 may be considered a close approximation to the common lead incorporated at the beginning of the last stage. The fact that the i n i t i a l lead isotope ratios f a l l below the growth curve can best be explained by a crustal history for the common lead component prior to the commencement of the last stage. The greater part of this i n i t i a l lead was probably derived prior to 2300 my ago although the Grenville event may have caused small recent additions. Kanesewich and Farquhar (1965) have produced evidence from lead isotopic analyses of ores emplaced in nearby rocks of the Cobalt-Noranda area, that shows that a crust existed \ r) there prior to 3200 my ago. The ore deposits they examined formed a chord between 2300 ± 150 my ago and 3250 ± 150 my ago. Kanesewich and Farquhar were of the opinion that the ore leads entered the c-82 FIGURE 6-10 PREFERRED INTERPRETATION MANITOBA SAMPLES PRESENT - 0 my GRENVILLE EVENT? GECO EMPLACED? LEAD REMOBILIZED? 900-1100 my? SUPERIOR EVENT PRESENT LEAD-URANIUM SYSTEM ISOLATED FIRST CRUSTAL MATERIAL FORMED IN THIS REGION - ~ 2700 my 3400 my AGE OF EARTH EARTH AND METEORITES SAME ISOTOPIC COMPOSITION - 4580 ± 50 my 83 older country rock through f i s s u r e s . This older country rock i n the southernmost region, i n the v i c i n i t y of Cobalt, i s known as the Bruce and Cobalt series. Grant (1964) i d e n t i f i e s the meta-morphic rocks north of the Grenville Front and i n the area studied as belonging to the Bruce and Cobalt series. An unconformity separates the overlying Bruce and Cobalt series from the older underlying granite north of the front, studied here. I t may be concluded that the common lead incorporated i n the Timagami samples was derived from cr u s t a l material e x i s t i n g p r i o r to 2300 my ago. Individual samples analysed i n t h i s study were probably p a r t i a l l y derived from rocks with a quite s i m i l a r common lead component 3500 to 3200 my ago, and have been subjected to an event at 2300 my ago which effected homogenization of the lead isotopic compositions that had developed up to t h i s time. I t i s not possible to d i s t i n g u i s h a spread i n i n i t i a l lead is o t o p i c composition through incomplete homogenization at 2300 my ago from the effect of a p a r t i a l r e d i s t r i b u t i o n of the common lead component i n these rocks at the time of the Grenville event. That such a spread exists i s confirmed by the deviations on the lead-lead plot f o r the Ontario data. Evidence supporting lead remobilization i s discussed l a t e r i n t h i s chapter. 84 FIGURE 6-11 PREFERRED INTERPRETATION ONTARIO SAMPLES PRESENT 0 my GRENVILLE EVENT LEAD REMOBILIZED 900-1100 my Rb-Sr SUPERIOR EVENT PRESENT LEAD-URANIUM SYSTEM ISOLATED -2600 my 2350 Rb-Sr 2580 Pb-Pb 2700 y.-Pb Zircons FIRST CRUSTAL MATERIAL FORMED IN THIS REGION ~ 3300 my AGE OF THE EARTH EARTH AND METEORITES SAME LEAD ISOTOPIC COMPOSITION 4580 ± 50 my 85 Comparison of Uranium-Lead and Rubidium-Strontium Systems. Turek and Peterman (1968) have carried out a very thorough rubidium-strontium study of the Rice Lake-Beresford Lake area. Their investigation included an age determination for the northern potassic granite studied in this thesis. They obtained 2550 ± 40 my (A = 1.39 x 1 0 - 1 1 y r - 1 ) for this body and an i n i t i a l 0"7 O A ratio for Sx /Sr of 0.7031 ± 0.0021. The i n i t i a l ratio is consistent with a previous history i n a system of moderate 87 86 Rb /Si? ratio. That the potassic granite is reworked material is supported by i t s strongly gneissic structure and the presence of numerous xenoliths. An investigation by D. Culbert and J. Blenkinsop has shown that a large body of crustal material may exist for several hundred million years in a low rubidium to strontium environment. Their analyses were carried out on the Coast Range Batholith, British Columbia. (Culbert and Blenkinsop, to be published). The rubidium-strontium results for the Timagami region are consistent with the model presented earlier, and have indeed been used to construct the model. J.A. Grant investigated whole rock and mineral rubidium-strontium systems for the Grenville Front near Lake Timagami, Ontario. He determined an average primary age, from the whole rocks, 0 7 O f ; of 2350 ± 150 my with i n i t i a l Sr /Sr of 0.703. The isochron plot showed some deviations from linearity, possibly as a result of open system behaviour. Biotite-whole rock joins yielded secondary ages of 920 my for G5, 1270 my for G2, and 1360 my for Gl. Apatite-86 whole rock joins generally gave greater values than the biotite-whole rock joins. These mineral ages are thought to be indicative of the Grenville event and approach the accepted Grenville age as one progresses south, i.e. in the direction of increasing meta-morphism. Clearly the lead-lead determination of 2580 ± 80 my is consistent with Grant's age from the rubidium-strontium system. It has been shown clearly that some event, probably the Grenville metamorphism, has disturbed the uranium-lead whole rock system. This was manifested by the deviations for the rubidium-strontium whole rock system from a straight line isochron plot. Furthermore the i n i t i a l ratio obtained from the strontium data does not preclude an early period in an environment of low to moderate 87 86 Rb /Sn ratio. It is also possible that this ratio is set by strontium additions at the same time as the uranium-lead system is formed for the last stage. Lead Mobilization in the Timagami Samples Steiger and Wasserburg (1969) have analysed zircons from sample G5. Their data supports a primary age of 2689 ± 12 my. The zircons are discordant, in general, and i t is thought'that lead loss may have been effected by the Grenville metamorphism. That lead was lost by some of the zircons, is well established. It was probably this lead which migrated during the Grenville metamorphism and is responsible for the lead migration inferred 87(a) from the age determinations discussed e a r l i e r i n t h i s chapter. It i s proposed here that the lead i s r e d i s t r i b u t e d on a scale dependent on the d i s t u r b i n g force and conditions p r e v a i l i n g . I f some of the radiogenic lead from the southernmost rocks has migrated north during the metamorphism, then some of the samples should have a higher lead/uranium r a t i o than i s consistent with t h e i r age; G2 i s j u s t such a sample. There i s some support f o r such a suggestion. F i r s t l y , i t i s well known that the majority of the uranium i n a rock i s found i n accessory minerals and i n t e r s t i t i a l l y . The concentrations of uranium i n the zircons, and the percentage of zirc o n i n G5 (Steiger and Wasserburg, 1969), indicates that a s i g n i f i c a n t f r a c t i o n (~ 30%) of the uranium f o r G5 i s found i n the recovered zircons. A su b s t a n t i a l f r a c t i o n of the radiogenic lead produced i n these zircons i s l o s t . I n t e r s t i t i a l lead w i l l be remobilized much more r e a d i l y . Lead i n some accessory minerals, such as sphenes, w i l l be more r e s i s t a n t to a l t e r a t i o n ; there i s no evidence that such minerals e x i s t i n important q u a n t i t i e s . The potassium feldspars i n the area have incorporated excess radiogenic lead i n t o t h e i r structure subsequent to t h e i r formation. On balance the assumption of remo b i l i z a t i o n of lead i n t h i s area i s amply j u s t i f i e d . I t i s the scale rather than the existence of migration of radiogenic lead thVt may be questioned. Evidence f o r the migration of lead over s i g n i f i c a n t distances has been presented before (Slawson, et. a l . , 1962). I t i s pos s i b l e that the lead i n t h i s thes 87(b) study area may have migrated over a distance of several miles. Open system behaviour of a uranium-lead system may result from remobilization of lead and/or uranium. Without evidence i n addition to the lead isotopic ratios and the lead and uranium concentrations i t is not possible to say which of the lead and uranium have been mobile. Mobility of the uranium in the Ontario suite, at the present, would provide a simple explanation for the open system behaviour as exhibited by the uranium-lead plot. However i t is more l i k e l y that the daughter of a radioactive decay is subject to mobility, i.e. the lead, because i t w i l l reside in a crystal latice with which i t is not compatible. It i s for this reason and the reasons given in the preceding paragraphs that remobilization of lead was preferred to remobilization of uranium. 88 Conclusions The f i r s t objective of the thesis was to provide data of sufficient quality to further the understanding of lead isotope interpretations. A standard deviation of 0.15% was obtained for the lead isotope ratios and 0.25% for the lead and uranium concentra tions. To achieve this precision double-spiking was employed to correct for fractionation that accompanies single filament lead isotope analyses. A study of the discrimination exhibited by the mass spectrometer was made. It was found that theoretical fractiona tion laws could not completely explain the discrimination pattern found for rhenium filaments. It was also found, when double-spiking was employed to correct for fractionation on tantalum filaments, that the theoretical fractionation laws give better agreement than the discrimination pattern observed for rhenium. Independent data obtained by Doe, on rhenium and tantalum filaments, has confirmed that the discrimination pattern for rhenium filaments does not f i t the theoretical fractionation law while the fractionation law i s quite adequate for the tantalum filaments. 0 The double spiking technique was employed by P. Reynolds to obtain interpretable patterns of lead isotopes. His analyses of rocks from the vici n i t y of Broken H i l l , Australia, were the f i r s t truly interpretable analyses of lead isotopes from whole rocks. 89 In this thesis the lead isotope ratios for samples from the Vogt-Hobbs area near Lake Timagami revealed variations from a simple two stage model that could not be accounted for by experi-mental error. Comparison of the uranium-lead ages with lead-lead ages, for two suites studied, revealed open system behaviour i n the last, stage, which was shown to be a consequence of lead migration. The experimental uncertainties in these uranium-lead and lead-lead ages were sufficiently small that the two could be clearly distinguished. The migration of lead was found to be more pronounced for the Timagami samples than for the Rice Lake-Beresford Lake samples, which is to be expected from the known and severe metamorphic event i n the Timagami region. Open system behaviour has been suggested by P. Reynolds in another precise study of rock leads (Seminar held at U.B.C, 1969). The history of the Manitoba samples, as determined primarily from the rock lead analyses and supported by other dating schemes, may be summarized as follows. The lead now observed experienced a minimum of three stages during the past 4580 my. The f i r s t stage, from 4580 my ago to approximately 3500 my ago, was spent i n a primary system, with respect to uranium and lead, having a present U 2 3 8/Pb 2 0 4 ratio of 8.8 to 8.9. The f i r s t stage of the sample's history i s the most remote and therefore the least accurately established. Between 3500 my and 2600 my ago the lead developed in a crustal environment and about 2600 my ago a second major event accompanied by metamorphism and intrusion produced the 90 common lead component incorporated i n the potassic granites. This 2600 my old event probably lasted several hundred m i l l i o n years. Since that time the granites have remained r e l a t i v e l y undisturbed. There i s some evidence i n adjacent regions (Ostic, 1963) that suggests that some lead remobilization may have been associated with the Grenville event 900 my ago, however further investigation i s required before the possible effect of t h i s event can be checked i n the Rice Lake-Beresford Lake area. The Vogt-Hobbs area near Timagami i n Ontario experienced a very s i m i l a r h i s t o r y to the Rice Lake-Beresford Lake area i n Manitoba. In fact i t may be said with some j u s t i f i c a t i o n that these two regions, f o r events that have been investigated, show only a d i f f e r e n t severity of the same events. The lead now observed was incorporated into rocks i n that v i c i n i t y more than 3250 my ago. After another 1000 my t h i s c r u s t a l material underwent a major event, approximately 2500 my ago, which may have spanned several hundred m i l l i o n years. The l a s t major event i n Grenville times produced the rocks as they ex i s t today. I t i s very probable that the g r a n i t i c rocks were not much changed from t h e i r form and composition at the time of the 2500 my event. These rocks are indeed f a r more a c i d i c than the Manitoba suite and probably represent a d i f f e r e n t environment 2500 my ago or e a r l i e r . The severity of the Grenville event was such as to p a r t i a l l y r e d i s t r i b u t e the lead, and also the rubidium and strontium, so that the 2500 my old event i s somewhat masked. 91 The study of the two suites indicates that the associated ' 238 204 cr u s t a l systems have a lower average present U /Pb r a t i o than the source of single stage ore deposits (Ostic et. a l . , 1967; Stacey et. a l . , 1969; Cooper et. a l . , 1969). This has been suggested as a widespread phenomena by Reynolds (1967). The h i s t o r y presented f o r the Manitoba samples suggests that Manitouwadge i s not representative of a single stage lead. This would contradict the assumption on which T i l t o n and Steiger (1965) calculated an age of 4750 my for the earth. The two independent ages for the earth calculated on the basis of the suites studied here were 4700 and 4640 my. These are based on e s s e n t i a l l y the same, inadequate assumptions, and are believed to be s i m i l a r l y i n error. I t i s considered that the age recently calculated f o r the earth by Cooper et. a l . (1969) i s the best estimate available. Although t h i s study has added much new information to the understanding of lead isotopes i n rocks, there i s s t i l l considerable scope for further work. This study, l i k e those preceding i t , shows that there i s much information to be obtained from studying lead isotope r a t i o s , but that the benefit can be obtained only by doing many analyses of the highest q u a l i t y . Over 100 isotopic analyses were required to obtain the data presented i n t h i s thesis. 92 BIBLIOGRAPHY Catanzaro, E.J., and Gast, P.W. (1959). Isotopic composition of lead i n pegmatitic feldspars. Geochem. Cosmochim. Acta. 19, 113-126. Catanzaro, E.J. (1967). T r i p l e filament method f o r s o l i d -sample lead isotope analysis. J . Geophys. Res. 72, 1325-1327. Catanzaro, E.J. (1968). Absolute isotopic abundance r a t i o s of three common lead reference samples. Earth Planet S c i . Lett. 3, 343-346. Catanzaro, E.J., Murphy, T.J., Shields, W.R., and Garner, E.L. (1968). Absolute is o t o p i c abundance ra t i o s of common, equal atom and radiogenic lead standards. J . Res. Nat. Bur. Std. A. 72-A, 261-267. Catanzaro, E.J., Murphy, T.J., Shields, W.R., and Garner, E.L. (1968). Absolute is o t o p i c standards (abstract). Trans. Am. Geophys. Union 49, 348. Compston, W., and Oversby, V.M. (1969). Lead i s o t o p i c analysis using a double spike. J . Geophys. Res. 74, 4338-4348. Cooper, J.A., Reynolds, P.H., and Richards, J.A. (1969). Double-spike c a l i b r a t i o n of the Broken H i l l standard lead. Earth Planet. S c i . Lett. 6_, 467-478. Cooper, J.A., and Richards, J.R. (1966). Solid-source lead isotope measurements and isotopic f r a c t i o n a t i o n . Earth Planet. S c i . Lett, 1, 58-64. 93 Dodson, M.H. (1963). A theoretical study of the use of in t e r n a l standards for precise isotopic analysis by the surface i o n i z a t i o n technique: Part I. J . S c i . Instr. 40, 289-295. Doe, B.R., and T i l l i n g , R.I. (1967). The d i s t r i b u t i o n of lead between co-existing K-feldspar and plagioclase. Am. Mineralogist, 52, 805-816. Doe, B.R., Tatsumoto, M., Delevaux, M.H., and Peterman, Z.E. (1967). Isotope d i l u t i o n determination of f i v e elements i n G-2 (granite), with a discussion of the analysis of lead. U.S. Geol. Surv. Profess. Papers, 575-B, B170-177. Duchylard, G., Lazard, B., and Roth, E. (1953). Isotopic analysis of lead by the mass spectrometer.J. Chem. Phys. 50, 497-500. Farquhar, R.M., and Russell, R.D. (1957). Anomalous leads from the Upper Great Lakes Region of Ontario. Tran. Amm. Geophys. Union, 38, 552-565. Farquhar, R.M., and Russell, R.D. (1963). A nomograph for the inter p r e t a t i o n of anomalous lead isotope abundances. Geochem. Cosmochim. Acta. 2_7, 1143-1148. Goldich, S.S., Nier, A.O., Baadsgaard, H., Hoffman, J.H., and Krueger, H.W. (1961). The Precambrian geology and geochronology of Minnesota. Minnesota Geol. Survey, B u l l . 41. Grant, J.A. (1964). Rubidium-strontium isochron study of the Grenvil l e Front near Lake Timagami, Ontario. Science 146, 1049-1053. 94 Grant, J.A. (1964). Geology of the Vogt-Hobbs Area District of Nipissing. Ontario Department of Mines Geological Report No. 22. Jacobs, J.A., Russell, R.D., and Wilson, J.T. (1959). Physics and Geology. McGraw-Hill Book Co. Inc., New York. Kanasewich, E.R., and Farquhar, R.M. (1965). Lead isotope ratios from the Cobalt-Noranda Area, Canada. Can. J. Earth. Sci. 2, 361-384. Kraus, K.A., and Nelson (1955). Anion exchange studies of the fission products. Proceedings of the International Conference on the peaceful uses of atomic energy 7, 113-125. Lindeman, F.A. and Aston, F.W. (1919). The possibility of separating isotopes. Phil. Mag. 37_, 523-534. Long, L.E. (1966). Isotope dilution analysis of common and radio-84 genie strontium using Sr-enriched spike. Loveless, A.J., and Russell, R.D. A strong-focussing lens for mass spectrometer ion sources. (To be published in International Journal of Mass Spectrometry and Ion Physics.) Mair, J.A. (1958). Solid source mass spectrometry with application to the age determination of rocks. (Unpublished Ph.D. thesis, University of Toronto). v Masuda, A. (1962). Experimental method for determination of isotopic composition of lead in volcanic rock. Journal of Earth Science, Nagoya University, 10, 117-124. Naidu, P.S., and Westphal, K.O. (1966). The inverse problem of particle motion and i t s application. Appl. Sci. Res. B12, 435-450. 95 National Bureau of Standards (1966). Technical Note 277, 53-65. Nicolaysen, L.O. (1961). Graphic interpretation of discordant age measurements of metamorphic rocks. Am. N.Y. Acad. Sci., 91, 198-206. Ostic, R.G., Russell, R.D., and Stanton, R.L. (1967). Additional measurements of the isotopic composition of lead from stratiform deposits. Can. J. Earth Sci. 4_, 245-269. Oversby, V.M. The isotopic composition of lead in iron meteorites. Geochem. Cosmochim. Acta. (Submitted, 1969). Ozard, J.M., and Russell, R.D. Efficiency of an electro-s t a t i c a l l y focussed electron impact ion source. Appl. Sci. Res. B20, 55-60. Peterman, Z.E., Doe, B.R., and Bartel, A. (1967). Data on the rock GSP-1 (granodiorite) and the isotope-dilution method of analysis for Rb and Sr. U.S. Geol. Surv. Profess. Papers. 575-B, B181-B186. Reynolds, P.H. (1967). A lead isotope study of ores and adjacent rocks. (Unpublished, Ph.D. thesis). Richards, L.W., King, H.S., and Hall, L.P. (1926). Attempts to fractionate mixed isotopes of lead, and the atomic weight of this metal. J. Am. Chem. Soc. 48, 1530-1543. 96 Russell, R.D. (1956). Interpretation of lead isotope abund-ances: Proceedings Second Conference on Nuclear Processes in Geological Settings, 68-78. Russell, R.D., and Farquhar, R.M. (1960). Dating galenas by means of their isotopic constitution II. Geochim et. Cosmochim. Acta, 1_9, 41-52. Russell, R.D., and Farquhar, R.M. (1960). Lead isotopes in geology: Interscience Pub. Inc., New York. Russell, R.D. (1963). Some recent researches on lead isotope abundances. Earth Science and Meteorites Ch3, 44-73. Russell, R.D., Kanasewich, E.R., and Ozard, J.M. (1966). Isotopic Abundances of lead from a "frequently-mixed" source. Earth Planet. Sci. Lett., j_, 85-88. Russell, R.D., Slawson, W.F., Ulrych, T.J., and Reynolds, P.H. (1968). Further applications of concordia plots of rock lead isotope abundances. Earth Planet. Sci. Lett., J3, 284-288. Senftle, F.E., and Bracken, J.T. (1954). Theoretical effect of diffusion on isotopic abundance ratios in rocks and associated fluids. Geochem. Cosmochim. Acta 7_, 61-76. Sinclair, A.J. (1965). Volume of source rocks of the radiogenic component of multistage anomalous leads. Econ. Geol 60, 1709-1717. Sinclair, A.J. (1965). Oceanic lead isotopes and ore genesis. Econ. Geol. 60, 1533-1542. 97 Slawson, W.F., and Austin, CF. (1960). Anomalous leads from a selected geological environment in West Central New Mexico, Nature, 187, 400. Slawson, W.F. and Austin, C F . (1962). A lead isotope study defines a geological structure. Econ. Geol. 57, 21-62. Stacey, J.S., Delevaux, M. and Ulrych, T.J. (1969). Some t r i p l e filament lead isotope analyses and absolute parameters for single stage leads. Earth. Planet. Sci. Lett. 6, 15-25. Steiger, R.'H., and Wasserburg, G.J. (1969). Comparative U-Th-Pb 9 systematics in 2.7 x 10 yr plutons of different geologic histories. Geochem. Cosmochim. Acta. 33, 1213-1232. Tatsumoto, M. (1966). Genetic relations of oceanic basalts as indicated by lead isotopes. Science 153, 1094-1101. Tilton, G.R. (1951). The distribution of trece quantities of uranium in nature. Atomic Energy Commission Report AECD 3182. Tilton, G.R., Patterson, C.E., Brown, H., Inghram, M., Hayden, R., Hess, D., and Larsen, E., J r . (1955). Isotopic composition and distribution of lead uranium and thorium in a r precambrian granite. Bull. Geo. Soc. Am. 62, 1131-1148. Tilton, G.R., and Steiger, R.H. (1965). Lead isotopes and the age of the earth. Science 150, 1805-1808. Turek, A., and Peterman, Z.E. (1968). Preliminary Rb-Sr geochronology of the Rice-Lake-Beresford Lake area, south-eastern Manitoba. Can. J. Earth Sci. 5, 1373-1380. 98 Ulrych, T.J., and Reynolds, P.H. (1966). Whole-rock and mineral leads from the llano Uplift, Texas. J. Geophys. Res. 71, 3089-3094. Ulrych. T.J. (1967). Oceanic basalt leads: A new interpretation and an independent age for the earth. Science 158, 252-256. Ulrych, T.J. (1968). Oceanic basalt leads and the age of the earth. Science 162, 928-929. Wanless, R.K., Stevens, R.D., and Loveridge, W.D. Excess radiogenic argon in biotites. (To be published.) Wasserburg, G.J. (1963). Diffusion processes in lead-uranium systems. J. Geophys. Res. 68_, 4823-4846. Weichert, D.H., and Russell, R.D. (1968). The deconvolution of a mass spectrum. Can. J. Phys. 46, 1443-1448. Weichert, D.H., Russell, R.D., and Blenkinsop, J. (1967). A method for d i g i t a l recording for mass spectra. Can. J. Phys. 45_, 2609-2619. c Welke, H., Moorbath, S., dimming, G.L., and Sigurdsson, H. Lead isotope studies on igneous rocks from Iceland. Earth Planet. Sci. Lett. 4_, 221-231. Wetherill, G.W. (1963). Discordant uranium-lead ages II. J. Geophys. Res. 6S_, 2957-2965. Williamson, J.H. (1968). Least squares f i t t i n g of a straight line. Can. J. Phys. 46, 1843-1847. York, D. (1966). Least-squares f i t t i n g of a straight line. Can. J. Phys. 44, 1079-1086. 99 York, D. (1968). Least squares f i t t i n g of a straight line with correlated errors. Earth. Planet. Sci. Lett. 5, 320-324. Zartman, R.E. (1965). The isotopic constitution of lead in microclines from the llano Uplift, Texas. J. Geophys. Res. 70, 965-975. Zartman, R.E. (1969). The isotopic composition of lead in potassium feldspars from some 1.0 b.y. old North American igneous rocks. Geochem. Cosmochim. Acta. 33, 901-942. L 100 APPENDIX i Chemical Procedures Air-borne Contamination: To reduce air-borne contamination a p o s i t i v e pressure of f i l t e r e d a i r was maintained i n the laboratory and when-ever possible samples being processed were covered. Apparatus: Pyrex glassware and Teflon receptacles were allowed to stand f o r twelve hours i n hot detergent solution then rinsed. The apparatus was then stored i n warm 15% n i t r i c acid for at least twelve hours and remained there t i l l required. A l l receptacles were rinsed three times with d i s t i l l e d , deionized water before use. D i s t i l l e d Water: D i s t i l l e d water from a Barnstead s t i l l was deionized and r e d i s t i l l e d i n a b o r o s i l i c a t e s t i l l . The lead content was 0.0001 ppm by weight. (Lead concentrations were determined by isotope d i l u t i o n i n a l l cases.) Hydrochloric Acid: Reagent grade dilu t e d to 20% concentration by weight, and d i s t i l l e d twice. Lead content was approximately 0.0001 ppm by weight. N i t r i c Acid: D i s t i l l e d twice at 70% concentration by weight. Lead content 0.0002 ppm by weight. Ammonium Hydroxide: Where possible ammonia gas was used. Ammonium hydroxide of s p e c i f i c gravity 0.9 was prepared by bubbling ammonia gas into ice cold water. 101 Perchloric Acid: Perchloric acid was available commercially, double vacuum d i s t i l l e d i n vycor and shipped i n vycor. The lead content was measured to be less than 0.05 ppm. Hydrofluoric Acid: Baker analysed reagent was used, the lead content was less than 0.05 ppm lead. Column Preparation: 1. S t i r 50 gm of Dowex 1 x 8 r e s i n with 100 ml of 1.5N hydrochloric acid and decant to remove f i n e s ; repeat twice. 2. Pour s l u r r y into columns to make one column 28 cm long by 1.1 cm inside diameter and one 10 cm long by 0.8 cm inside diameter. 3. Se t t l e columns with a long glass rod and adjust flow rate to 0.8 and 0.5 ml per minute respectively. 4. Wash columns with f i v e column volumes of d i s t i l l e d water. 5. Convert to chloride form by passing through three column volumes of 1.5N hydrochloric acid. 6. For nitrate-form columns use three column volumes of 7N n i t r i c acid. Lead P u r i f i c a t i o n : 1. Centrifuge lead concentrate dissolved i n 1.5N hydrochloric acid; discard residue and add sample to column. 2. Wash column with 60 ml of 1.5N hydrochloric acid. 3. Elute lead with 60 ml of d i s t i l l e d water. Discard f i r s t 10 ml from column and save remainder. 102 4. Evaporate eluate to dryness and redissolve i n 15 ml of 1.5N ' hydrochloric acid. 5. Repeat steps one to four using one quarter the volumes of reagents and the small column. Nitr a t e Resin Column for Uranium 1. Add uranium dissolved i n 80 ml 7N n i t r i c acid to the large, nitrate-form column. 2. Wash column with three 20 ml portions of 7N n i t r i c acid. 3. To elute the uranium use 10 ml portions of water. Discard a l l eluate t i l l the orange colour i s a few centimeters from the bottom of the column. Add additional water t i l l the r e s i n i s quite orange, c o l l e c t i n g the eluate. Add 6N hydrochloric acid u n t i l the r e s i n i s yellow again, continue to c o l l e c t the eluate. 4. Repeat the procedure with a small column using one quarter the volume of reagents. Uranium Filament Heating Pattern 1. Raise the filament temperature to a brightness temperature of 2060°C. After f i v e minutes locate and focus the rhenium-187 signal at which time the ion current w i l l be 0.6 x 10~^a. The rhenium-187 ion current remains at t h i s value at t h i s temperature. 2. Set the sample filament current at 1.5 a and re-focus the rhenium-187 s i g n a l . Turn up the side filament current u n t i l the uranium signal i s 0,1 x 10 ^ a. 103 3. After 10 minutes t o t a l time from the s t a r t , a l l filaments are i turned o f f . _7 4. When the pressure i n the analyser tube i s below 2 x 10 rai s e the center filament to 2060°C and maintain at t h i s value throughout the analysis. Then set the side filament current at 1.5 a and re-focus on rhenium-187. 5. After another f i v e minutes adjust the sample filament current to give a uranium signal of 0.6 x 10 ^  a. 6. Leave f i v e minutes, then ra i s e the uranium signal to 1.2 x 10 a. 7. F i n a l l y , a fter a further f i v e minutes, increase the uranium signal to 1.8 x 10 a. When the ion current i s s u f f i c i e n t l y stable commence taking data. Concentration Determinations Lead-206 (SRM 983) was employed for stable isotope d i l u t i o n determinations of lead concentrations. The concentration of lead-206 i n the spike solution was calibrated against "Specpure" lead solutions by mass spectrometric analysis of pipetted mixtures of these solutions. For these analyses the t r i p l e filament method of Catanzaro was employed. Pipettes used were calibrated by weighing the volumes delivered and applying evaporation corrections. Concentrations obtained gravimetrically agreed with the mass spectrometric determina-tions to within 0.2%. 104 For uranium concentration determinations uranium enriched to 99.4% uranium-235 was used. Solutions were c a l i b r a t e d against "Specpure" uranium. As the theory of isotope d i l u t i o n i s quite s i m i l a r f o r both concentration determinations, only one case need be discussed. The explanation i s f a c i l i t a t e d by using a s p e c i f i c case so the discussion i s based on the P b 2 ^ / P b 2 ^ 8 r a t i o . Let P = ^ C* represent the /.Da n .th . Pb' f r a c t i o n of the i isotope, i n atoms per atom, i n the sample, and C 1 the concentration of the i^1 isotope i n the spike, sp Then i n a mixture of spike and sample: P = r206,. „206 c C N + C S n sp C 2 0 8 N + C 2 0 8 S n sp Where N and S represent the number of moles of t o t a l lead contributed to the mixture by the sample and the spike r e s p e c t i v e l y . This may be written: N = S PC 208 'sp ,206 .206 - PC 1 208 n By d i f f e r e n t i a t i n g and rearranging an expression f o r the f r a c t i o n a l error i n N r e s u l t i n g from the error i n P may be obtained. 9N.3P. N V P ,208 Jsp ,208 Jn p c208 _ c206 I SP p c208 _ c206 n n 105 This r a t i o of r e l a t i v e errors was calculated for C 2 0 6 = 0.92, C 2 0 8 = 0.013, C 2 0 6 = 0.24, and C 2 0 8 =0.54. Figure A-sp SP n n shows the optimum spiking curve. By using values of P between one and twenty, the error magnification through this effect was kept below 1.5. FIG. A- l OPTIMUM SPIKING FOR LEAD CONCENTRATION P b 2 0 6 / P b 2 0 8 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

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.831.1-0053415/manifest

Comment

Related Items