hf-w chronology of planetary accretion and
TRANSCRIPT
Research Collection
Doctoral Thesis
Hf-W chronology of planetary accretion and differentiation at thedawn of solar system history
Author(s): Kruijer, Thomas
Publication Date: 2013
Permanent Link: https://doi.org/10.3929/ethz-a-010053881
Rights / License: In Copyright - Non-Commercial Use Permitted
This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.
ETH Library
DISS. ETH NO. 21487
Hf-W CHRONOLOGY OF PLANETARY ACCRETION AND DIFFERENTIATION AT
THE DAWN OF SOLAR SYSTEM HISTORY
A thesis submitted to attain the degree of
DOCTOR OF SCIENCES of ETH ZÜRICH
(Dr. sc. ETH Zurich)
presented by
THOMAS KRUIJER
MSc., VU University Amsterdam
born on 06.03.1986
citizen of the
Kingdom of the Netherlands
accepted on the recommendation of
Prof. Dr. Rainer Wieler
Prof. Dr. Thorsten Kleine
Prof. Dr. Maria Schönbächler
Prof. Dr. Alexander N. Halliday
PD Dr. Ingo Leya
2013
II
III
We’re Extremely Fortunate
We’re extremely fortunate not to know precisely
the kind of world we live in.
One would have to live a long, long time, unquestionably longer
than the world itself.
Get to know other worlds, if only for comparison.
Rise above the flesh,
which only really knows how to obstruct
and make trouble.
For the sake of research, the big picture
and definitive conclusions, one would have to transcend time,
in which everything scurries and whirls (…)
Wis!awa Szymborska
(‘The End and the Beginning’, 1993)
IV
V
Table of contents
Abstract 1
Zusammenfassung 3
Chapter 1: General introduction 5
1.1 Origin of the solar system and planet formation 6 1.1.1 The solar system at present 6
1.1.2 Evolution of the protoplanetary disk and accretion of planetary bodies 6
1.1.2.1 Molecular cloud core collapse and formation of the solar nebula 6 1.1.2.2 Time-zero as defined in cosmochemistry 8 1.1.2.3 Planet and planetesimal formation: What is (un)known? 8
1.2 Early solar system chronology as recorded in meteorites 9 1.2.1 Elemental and isotopic ratios in meteorites 9
1.2.2 Short-lived radionuclides in meteorites 10
1.2.3 Isotope chronometry of meteorites 10
1.2.3.1 Short-lived vs. long-lived decay systems 11 1.2.4 Tungsten isotopes and the 182Hf-182W chronometer 11
1.2.4.1 W isotope systematics in meteorites 13 1.2.4.2 Hf-W chronometry of metal samples 14 1.2.4.3 Hf-W chronometry of silicate rock samples 15
1.3 Aim of this thesis 15
PART A: QUANTIFICATION OF NEUTRON CAPTURE EFFECTS AND THE Hf-W CHRONOLOGY OF IRON METEORITES
Chapter 2: Introduction to part A 20
2.1 Core formation in planetary bodies 21 2.2 Iron meteorites as remnants of planetesimal cores 21
2.2.1 Classification and characteristics 21
2.2.1.1 Chronology of iron meteorites 22 2.2.1.2 Hf-W studies of iron meteorites and
neutron capture-induced W isotope variations 22 2.3 Cosmic ray exposure and secondary neutron capture effects
in iron meteoroids 23 2.4 Aim and outline of Part A 24
VI
Chapter 3: Hf!W chronology of core formation in planetesimals inferred from weakly irradiated iron meteorites 28
3.1 Introduction 29 3.2 Theory and Approach 30 3.3 Analytical methods 31
3.3.1 Sample selection 31
3.3.2 Noble gas measurements 32
3.3.3 Tungsten isotope analyses 34
3.3.3.1 Sample preparation and chemical separation of W 34 3.3.3.2 W isotope measurements 36
3.4 Results 38 3.4.1 Cosmogenic noble gases 38
3.4.2 Tungsten isotopes 38
3.4.2.1 Terrestrial standard 38 3.4.2.2 Iron meteorites 41
3.5 Discussion 45 3.5.1 Identifying iron meteorite samples with
very low fluences of (epi)thermal neutrons 45
3.5.2 Core formation ages for samples with minimal cosmic ray effects 49
3.5.3 Tungsten isotope anomalies and Hf!W chronometry of ungrouped iron meteorites 52
3.6 Conclusions 52
Chapter 4: Neutron capture on Pt isotopes and the HF-W chronology of core formation in planetesimals 59
4.1 Introduction 60 4.2 Sample preparation and model calculations 61
4.2.1 Sample selection 61
4.2.2 Analytical techniques for Pt and W isotope measurements 62
4.2.3 Model calculations 63
4.3 Results 65 4.3.1 Pt isotopes 65
4.3.2 W isotopes 65
4.4 Discussion 66 4.4.1 Origin of Pt isotope variations in iron meteorites 66
4.4.2 Nucleosynthetic vs. radiogenic and cosmogenic W isotope variations 67
4.4.3 Combined Pt and W isotope systematics 69
4.4.3.1 Pre-exposure "182W for IVB, IID and IVA iron meteorites 70 4.4.3.2 Comparison to combined Pt-W model results 71 4.4.3.3 Comparison to previous studies 72
VII
4.4.4 Chronology of metal segregation and age differences between iron meteorites and chondrites 73
4.5 Conclusions 74 4.6 Appendix : Supplementary text and figures 79
4.6.1 Sample preparation and chemical separation of Pt 79
4.6.2 Pt isotope measurements 79
4.6.2.1 Instrumentation and data acquisition protocol 79 4.6.2.2 Os interferences 80 4.6.2.3 Tailing from Ir 80 4.6.2.4 Uncertainty of Pt isotope measurements 82
4.6.3 W isotope measurements 82
4.6.4 Additional tables and figures 86
Chapter 5: Accretion and metal segregation timescales of protoplanets from small 182W variations among iron meteorites 91
5.1 Introduction 92 5.2 Results: Pt and W isotope compositions 93 5.3 Discussion 94
5.3.1 Pt isotope systematics 94
5.3.2 W isotope systematics of iron meteorites 95
5.3.3 Combined Pt-W isotope systematics 96
5.3.4 Interpretation of variable pre-exposure "182W 99
5.3.4.1 Summary of pre-exposure "182W values 99 5.3.4.2 Observed correlations with volatile elements Ga, Ge, and S 100 5.3.4.3 Distinct melting temperatures for iron meteorite parent bodies and interpretation of
"182W vs. S correlation 101 5.3.4.4 Thermal model for post-accretional heating of iron meteorite parent bodies 102 5.3.4.5 SDistinct times of metal segregation and concurrent accretion of iron meteorite
parent bodies 102 5.3.4.6 Suprachondritic Hf/W for precusor material of IID and IVB parent bodies 103
5.4 Conclusions 104 5.5 Appendix: Supplementary text, figures and tables 108
5.5.1 Materials and Methods 108
5.5.1.1 Iron meteorite samples 108 5.5.1.2 Sample preparation and chemical separation of Pt and W
for MC-ICPMS analyses 108 5.5.1.3 Mass spectrometry 111
5.5.2 Data tables 113
Chapter 6: Abundance and isotopic composition of Cd in iron meteorites 118
6.1 Introduction 119 6.2 Analytical methods 120
6.2.1 Sample preparation and chemical separation of Cd 120
VIII
6.2.2 Cadmium concentration determination by isotope dilution (ID) 121
6.2.3 Cadmium isotope composition measurements 122
6.2.4 Pt and W isotope measurements 123
6.3 Results 124 6.3.1 Cd concentrations 124
6.3.2 Cd, Pt and W isotope compositions 126
6.4 Discussion 127 6.4.1 Cadmium abundances in the parent bodies of iron meteorites 127
6.4.2 Cd isotopes as a monitor for neutron capture in iron meteorites 128
6.4.3 Secondary neutron capture in iron meteorites: Importance of target chemistry 129
6.5 Conclusions 131
PART B: NUCLEOSYNTHETIC W ISOTOPE HETEROGENEITY AND THE
IMPLICATIONS FOR EARLY SOLAR SYSTEM CHRONOLOGY
Chapter7: Introduction to Part B 135
7.1 Nucleosynthesis and nucleosynthetic isotope anomalies in meteorites 136 7.2 CAI: Relicts of ‘time zero’ in solar system history 137
7.2.1 Introduction 137
7.2.2 Nucleosynthetic isotope anomalies in CAI 138
7.2.3 Chronology of CAI 138
7.2.3.1 Al-Mg systematics of CAI 139 7.2.3.2 Pb-Pb chronometry of CAI 140
7.3 Aim and outline of Part B 140
Chapter 8: Nucleosynthetic W isotope anomalies and Hf-W chronometry of Ca-Al-rich inclusions 145
8.1 Introduction 146 8.2 Samples and analytical methods 147
8.2.1 Sample preparation and chemical separation 147 8.2.2 Isotope measurements 148
8.3 Results 151 8.4 Nucleosynthetic W isotope anomalies in CAI 152 8.5 182Hf-182W systematics of bulk CAI 153
8.5.1 Effects of nucleosynthetic W isotope anomalies on Hf-W systematics 153 8.5.2 Hf-W isochron for bulk CAI 156 8.5.3 Initial 182W/184W and 182Hf/180Hf of the solar system 158
IX
8.6 Implications for the chronology of the early solar system 159 8.6.1 Iron meteorites 159 8.6.2 Comparison to 26Al-26Mg ages and evidence for 26Al homogeneity 159
8.7 Appendix: Supplementary text, figures, and tables 164 8.7.1 Petrographic and chemical classification of CAI 164
8.7.1.1 Petrographic classification 164 8.7.1.2 REE concentration analyses by ICPMS 168
8.7.2 Additional data tables 171 8.7.3 Additional figures 173
Chapter 9: 180W Anomalies in Iron Meteorites: Implications for p-Process Heterogeneity in the Solar Nebula 174
9.1 Introduction 175 9.2 Methods 175
9.2.1 Samples and Preparation 175 9.2.2 Separation of Tungsten 176 9.2.3 Mass Spectrometry 176
9.3 Results 176 9.3.1 Terrestrial Samples 177
9.3.1.1 183W Deficits 177 9.3.1.2 180W Measurements 177
9.3.2 Meteoritic Samples 178 9.4 Discussion 182
9.4.1 Anomalies in 180W 182 9.4.1.1 Accuracy of 180W Measurements 184 9.4.1.2 180W Isotope Anomalies in Iron Meteorites 184
9.4.2 Effects from an s-Deficit 186 9.4.3 Neutron Capture Effects 187 9.4.4 Radiogenic Effects 188 9.4.5 Possible Spallation Effects 191
9.5 Conclusions 192
Chapter 10: Synthesis and Outlook 196
Acknowledgements 200
Curriculum Vitae 202
X
1
Abstract The accretion and differentiation of km-sized planetesimals and larger planetary embryos during the first few million years (Myr) of solar system history represents a fundamental step towards building terrestrial planets like the Earth. The extinct 182Hf-182W chronometer (t1/2 = 8.9 Myr) is ideally suited to investigate the time-scales of planetary accretion and metal-silicate differentiation during the earliest stages of solar system history.
The principle aim of this dissertation is to investigate the time interval over which planetesimals and planetary embryos accreted and segregated their metal cores in the solar protoplanetary disk using Hf-W chronometry. This main objective can be addressed through high precision W isotope measurements in iron meteorites, but also requires a solid solution to two fundamental issues in Hf-W chronometry: (A) the presence of superimposed cosmic ray-induced neutron capture effects in iron meteorites which bias the 182W/184W and thus the inferred Hf-W ages to erroneously low values, and (B) the presence of nucleosynthetic W isotope anomalies in Ca-Al-rich inclusions (CAI), which significantly bias the inferred initial 182W/184W composition of the solar system as given by CAI. The first issue requires the development of a direct neutron dosimeter for iron meteorites (Part A), while the second issue warrants a detailed study of the magnitude and nature of nucleosynthetic W isotope anomalies in bulk CAI (Part B).
Part A of this thesis comprises the Hf-W chronometry of iron meteorites. The quantification of cosmic ray-induced shifts on W isotope compositions in iron meteorites is accomplished using two different, but complementary approaches: (i) the identification of samples with minimal cosmic ray effects, and (ii) the correction of cosmic ray-induced W isotope variations using a direct neutron dosimeter.
We first use the former approach and apply combined noble gas and W isotope analyses to identify iron meteorite samples with minimal cosmic-ray effects (chapter 3). As such samples do not require any correction, they represent key samples to more firmly establish the Hf-W chronology of iron meteorites. The best approach to quantify neutron capture-induced shifts in iron meteorites is by measuring the neutron fluence directly using an independent neutron dose monitor. Platinum isotope anomalies in iron meteorites are due to secondary neutron capture and are correlated with 182W/184W variations (chapter 4). As such, Pt isotopes represent an excellent and sensitive neutron dosimeter for correcting cosmic ray-induced shifts in iron meteorites. Combined Pt-W isotope analyses provide a precise ‘pre-exposure’ 182W/184W composition for any iron meteorite group, which represents the W isotope composition that is unaffected by neutron capture and that can be reliably interpreted in terms of Hf-W chronology.
Combined Pt-W isotope analysis on all major groups of magmatic iron meteorites (chapter 5) reveals that resolvable, ~5-20 parts-per million (ppm) differences in pre-exposure 182W/184W exist between different iron meteorite parent bodies. The variations in pre-exposure 182W/184W variations, and correlations thereof with the abundances of moderately volatile elements like S, Ga and Ge. The variable pre-exposure 182W/184W indicate that there are resolvable differences in the time of core formation in the IIAB, IIIAB, IVA, IVB and IID iron meteorite parent bodies. These differences reflect variations in the average melting temperatures of the metal, which are controlled by the S contents of the metal cores. Regardless of differences in the time of core formation, all five parent bodies seem to have accreted at about the same time, most likely between ~0.3-0.7 Myr after CAI formation. Moreover, the elevated pre-exposure 182W/184W of the IID and IVB irons in part reflect that their parent bodies accreted from precursor material with elevated time-integrated Hf/W. This
2
implies that the precursor planetary bodies accreted and differentiated in chemically distinct nebular reservoirs.
To investigate neutron capture effects in iron meteorites in some more detail we report new Cd concentration data and combined Pt-Cd-W isotope data for several iron meteorites (chapter 6). Cadmium is strongly depleted in all iron meteorite groups, implying that the parent bodies accreted well above the low condensation temperature of Cd and thus incorporated only minimal amounts of highly volatile elements. No Cd isotope anomalies were found, whereas Pt and W isotope anomalies for the same iron meteorite samples indicate a significant fluence of epithermal and higher energetic neutrons. This observation demonstrates that neutron capture in iron meteorites mainly occurs at epithermal and higher energies, and that the relative magnitude of neutron capture-induced isotope anomalies are strongly affected by the chemical composition of the irradiated material. The resulting low fluence of thermal neutrons in iron meteorites and their very low Cd concentrations make Cd isotopes unsuitable as a neutron dosimeter for iron meteorites.
In Part B of this thesis we examine nucleosynthetic W isotope heterogeneities in CAI and bulk meteorites, and assess their effect on Hf-W chronology. The major objective is to quantify the extent and nature of nucleosynthetic W isotope variations in CAI in order to better constrain the initial 182Hf/180Hf and 182W/184W ratios at the start of solar system history (chapter 8). We report W isotope compositions for several fine- and coarse-grained CAI from several carbonaceous chondrites. The investigated fine-grained CAI exhibit large and variable anomalies in 183W/184W, extending to much larger values than previously observed, and reflecting variable abundances of s- and r-process W isotopes in CAI. Conversely, the coarse-grained (mostly type B) inclusions show only small (if any) nucleosynthetic W isotope anomalies. The investigated CAI define a precise empirical correlation between 183W/184W and initial (i.e., 182Hf decay corrected) 182W/184W which provides a direct empirical means to correct the 182W/184W of any CAI for nucleosynthetic isotope anomalies. The corrected 182W/184W vs. 180Hf/184W define a bulk CAI isochron that dates CAI formation and provides a precise solar system initial 182Hf/180Hf and 182W/184W ratios. The Hf-W and Al-Mg systems exhibit concordant formation intervals between CAI and angrites, providing new supportive evidence for a homogeneous distribution of 26Al in the inner solar system. Finally, with respect to the major iron meteorite groups, only the IID and IVB irons show small nucleosynthetic W isotope anomalies, which are indicative of small-scale heterogeneities in the abundance of s- or r-process W isotopes in the primitive solar nebula (chapters 3-5). In addition, we confirm that large 180W excesses (up to ~0.06 %) occur in magmatic iron meteorites (chapter 9). However, significant within-group variability in 180W signatures among IIAB and IVB irons are not consistent with the inheritance of an initial widespread p-process heterogeneity of W isotopes in the solar nebula.
Collectively, the Hf-W results from this dissertation demonstrate that the parent bodies of iron meteorites accreted several 100 kyr after the start of solar system history as defined by the formation of CAI. The iron meteorite parent bodies subsequently segregated their cores at distinct times within a time interval of ~1-2 Myr after solar system formation.
3
Zusammenfassung
Für die Entstehung von terrestrischen Planeten wie die Erde ist die Akkretion und Differentiation von km-grossen Planetesimalen und grösseren planetaren Embryonen während der ersten Millionen von Jahren (Mio. J.) in der Geschichte des Sonnensystems von fundamentaler Bedeutung. Das ausgestorbene radiometrische 182Hf-182W System (t1/2 = 8.9 Mio. J.) ist ein ideales Werkzeug um die Zeitskalen von planetarer Akkretion und Metall-Silikat Differentiation während der Entstehungsphase des Sonnensystems zu studieren.
Das Ziel dieser Dissertation ist es, den Zeitabschnitt, in dem Planetesimale und planetare Embryonen in der protoplanetaren Scheibe akkretiert und ihre Metallkerne segregiert sind, mittels dem Hf-W Chronometer zu studieren. Dazu werden einerseits hochpräzise W-Isotopenmessungen an Eisenmeteoriten benötigt. Andererseits müssen zwei fundamentale Probleme des Hf-W Chronometers gelöst werden: (A) der durch kosmische Strahlung induzierte Neutroneneinfang modifiziert die W-Isotopenzusammensetzung von Eisenmeteoriten, welche speziell das 182W/184W Isotopenverhältnis betrifft und zu zu niedrig berechneten Hf-W Altern führt, sowie (B) nukleosynthetische W-Isotopenanomalien in Ca-Al-reichen Einschlüssen (CAI), welche die in den CAI enthaltene initiale 182W/184W Zusammensetzung des Sonnensystems überlagern. Das erste Problem erfordert die Entwicklung eines Neutronendosimeters für Eisenmeteoriten (Teil A). Das zweite Problem verlangt eine detaillierte Studie des Ursprungs und der Magnitude von nukleosynthetischen W-Isotopenanomalien in CAI (Teil B).
Teil A dieser Thesis umfasst die Hf-W Chronometrie von Eisenmeteoriten. Die Quantifizierung von W-Isotopenanomalien in Eisenmeteoriten, welche durch kosmische Strahlung verursacht worden sind, wird durch zwei unterschiedliche und komplementäre Ansätze erreicht: (i) Die Bestimmung von Proben mit minimalen kosmischen Strahlungseffekten, sowie (ii) die Korrektur von durch kosmische Strahlung modifizierten W-Isotopenzusammensetzungen mittels eines direkten Neutronendosimeters.
Zunächst wird der erste Ansatz verfolgt und Eisenmeteoriten mit minimalen kosmischen Strahlungseffekten mittels Edelgas- und W-Isotopenanalysen bestimmt (Kapitel 3). Da solche Proben keine Korrektur erfordern, stellen sie Schlüsselproben für die Erstellung der Hf-W Chronologie von Eisenmeteoriten dar. Die vielversprechendste Methode zur Quantifizierung von Isotopeneffekten in Eisenmeteoriten durch Neutroneneinfang ist die direkte Messung der Neutronenfluenz mittels eines unabhängigen Neutronendosimeters. Platinisotopenanomalien in Eisenmeteoriten sind auf Neutroneneinfang zurückzuführen und korrelieren mit 182W/184W Schwankungen (Kapitel 4). Daher eigenen sich Pt-Isotope hervorragend als sensitives Neutronendosimeter, welches für die Korrektur von W-Isotopendaten in Eisenmeteoriten benutzt werden kann. Kombinierte Pt-W-Isotopenanalysen liefern präzise 182W/184W Zusammensetzungen für jede Meteoritengruppe für die Zeit vor der Exposition der Proben durch kosmische Strahlung (und folglich Neutroneneinfang). Die korrigierten 182W/184W Werte können demzufolge verlässlich für die Hf-W Chronologie verwendet werden.
Kombinierte Pt-W-Isotopenanalysen von magmatischen Eisenmeteoriten aller Hauptgruppen ergeben (Kapitel 5), dass auflösbare Unterschiede der 182W/184W Werte vor der Exposition durch kosmische Strahlung von ~5-20 Teilen pro Million (ppm) für unterschiedliche Ausgangskörper von Eisenmeteoriten existieren. Die Variationen der Prä-Expositions 182W/184W Werte zeigen, dass die Ausgangskörper der Eisenmeteorite ihre Kerne innerhalb von ~1-3 Mio. J. nach der Bildung von CAI segregiert haben müssen. Die Variablen Hf-W Alter deuten auf unterschiedliche Schmelztemperaturen in gleichzeitig akkretierten Planetesimalen hin. Daraus kann gefolgert werden, dass die Kernbildung von früh-akkretierten planetaren Körpern wahrscheinlich innerhalb von 2 Mio. J. stattgefunden hat. Darüberhinaus spiegeln erhöhte Prä-Expositions-Werte der 182W/184W Verhältnisse von IID und IVB Eisenmeteoriten teilweise den Umstand wieder, dass ihre Ausgangskörper von
4
einem Material mit erhöhtem, zeitlich integrierten Hf/W Verhältnis gebildet wurden. Dies impliziert, dass die Ausgangskörper in Regionen des solaren Nebels mit unterschiedlichen chemischen Zusammensetzungen akkretiert und differenziert sind, welche sich demzufolge innerhalb der ersten 1-2 Mio. J. nicht komplett gemischt haben können.
Um die Effekte von Neutroneneinfang in Eisenmeteoriten im Detail zu untersuchen, werden neue Cd-Konzentrations- und kombinierte Pt-Cd-W-Isotopendaten für mehrere Eisenmeteorite präsentiert (Kapitel 6). Cadmium ist in allen Eisenmeteoritengruppen stark verarmt, was darauf hindeutet, dass die Ausgangskörper oberhalb der Kondensationstemperatur von Cd akkretiert sein müssen und folglich nur geringe Mengen an hochvolatilen Elementen eingebaut haben. Während keine Cd-Isotopenanomalien nachgewiesen werden konnten, deuten Pt- und W-Isotopenanomalien derselben Eisenmeteoritenproben auf eine signifikante Fluenz von epithermischen und höherenergetischen Neutronen hin. Diese Beobachtung verdeutlicht, dass Neutroneneinfang in Eisenmeteoriten vordergründlich bei epithermischen und höherenergetischen Energien auftritt und dass das Ausmass der durch Neutroneneinfang hervorgerufenen Isotopenanomalien stark von der chemischen Zusammensetzung des bestrahlten Materials abhängt. Aufgrund der geringen Fluenz von thermischen Neutronen in Eisenmeteoriten sowie deren niedrige Cd-Konzentrationen kann gefolgert werden, dass Cd-Isotope sich nicht als Neutronendosimeter für Eisenmeteorite eignen. In Teil B dieser Arbeit werden nukleosynthetische W-Isotopenheterogenitäten in CAI und Gesamtgesteinsmeteoriten sowie deren Auswirkungen auf die Hf-W Chronologie untersucht. Das Ziel ist es, Ausmass und Art von nukelosynthetischen W-Isotopenvariationen in CAI zu quantifizieren, um die initialen 182Hf/180Hf und 182W/184W Verhältnisse bei der Entstehung des Sonnensystems genauer zu bestimmen (Kapitel 8). Dazu werden die W-Isotopenzusammensetzungen von verschiedenen fein- und grobkörnigen CAI aus kohligen Chondriten bestimmt. Die feinkörnigen CAI sind durch grosse und variable Anomalien der 183W/184W Verhältnisse gekennzeichnet. Das Ausmass der Anomalien ist dabei wesentlich grösser als bisher bekannt und reflektiert den variablen Ursprung der W-Isotope aus s- und r-Prozessen. Demgegenüber enthalten die grobkörnigen CAI (mehrheitlich Typ B) nur kleine (bzw. gar keine) nukleosynthetische W-Isotopenanomalien. Die untersuchten CAI sind durch eine präzise bestimmte Korrelation von 183W/184W Verhältnissen und initialen (und 182Hf-Zerfall korrigierten) 182W/184W Verhältnissen charakterisiert, was eine direkte Möglichkeit darstellt, nukleosynthetische Anomalien von 182W/184W Daten in jeglichen CAI zu korrigieren. Die korrigierten 182W/184W vs. 180Hf/184W Daten ergeben eine Isochrone, welche das Entstehungsalter der CAI datiert und präzise 182Hf/180Hf und 182W/184W Verhältnisse des initialen Sonnensystems liefert. Konkordante Entstehungsintervalle von CAI und Angriten basierend auf den Systemen von Hf-W und Al-Mg, liefern weitere Indizien dafür, dass das innere Sonnensystem eine homogene Verteilung von 26Al besass. Zuletzt wird gezeigt, dass von den Hauptgruppen der Eisenmeteorite lediglich die Gruppen IID und IVB kleine nukleosynthetische W-Isotopenanomalien haben, was auf kleinmassstäbliche Heterogenitäten von s- und r-Prozess stammenden W-Isotopen im primitiven solaren Nebel zurückzuführen ist (Kapitel 3-5). Zudem wird bestätigt, dass grosse Überschüsse an 180W (bis zu ~0.06 %) in magmatischen Eisenmeteoriten zu finden sind (Kapitel 9). Allerdings sind Schwankungen der 180W-Signaturen innerhalb der Gruppen IIAB und IVB nicht mit einer vererbten initialen p-Prozess Heterogenität der W Isotope im solaren Nebel im Einklang.
Zusammenfassend zeigen die Hf-W Ergebnisse in dieser Dissertation, dass die Ausgangskörper von Eisenmeteoriten mehrere hunderttausend Jahre nach der Entstehung des Sonnensystem akkretiert sind, welches durch die Entstehung der CAI datiert ist. Die Ausgangskörper der Eisenmeteorite segregierten in der Folge innerhalb von ~1-3 Mio. J. nach der Entstehung des Sonnensystems ihren Kern.
Chapter 1 5
Chapter 1
General introduction
6 Introduction
1.1 Origin of the solar system and planet formation The origin of the Solar System and the planetary system therein represents a fundamental scientific problem. An understanding of planet forming processes can be obtained through astronomical observations of other star forming regions, circumstellar disks and extrasolar planets, accompanied by astrophysical models of planetary formation. The most direct knowledge is provided by reconstructing the history of our own solar system, which can be directly accessed by studying the chemical and isotopic signatures of meteorites. Planetary systems most likely form in disks of gas and dust around young stars, but the exact mechanisms are still debated. A concise review of our current knowledge regarding the origin of the solar system and planet formation is given in the paragraphs below.
1.1.1 The solar system at present The current configuration of the solar system places powerful constraints on its history. After about ~4.6 billion years (Ga) of its history the solar system presently consists of a single central star, the Sun, surrounded by four major terrestrial (rocky) planets, four giant gas planets, and a multitude of smaller asteroids, moons, dwarf planets and comets. The orbits of the major planets are close to circular, nearly coplanar, in the equatorial plane of the Sun, and confined to heliocentric distances of !30 astronomical units (AU). The four terrestrial planets (Mercury, Venus, Earth, Mars) are mainly composed of rocky material and their orbits lie closest to the Sun. In contrast, the gas giant planets (Jupiter, Saturn, Uranus, and Neptune) primarily consist of lighter elements and orbit at larger heliocentric distances. The Sun makes up more than 99.8% of the Solar system’s mass, whereas the angular momentum in the solar system largely (>98 %) resides in the orbital motions of (mostly the giant) planets. A significant number (on the order of ~106) of the smaller planetary bodies with radii mostly ~1-100 km resides in the Asteroid Belt, which is located between Mars and Jupiter. Most meteorites are the remnants of asteroids (many of them being originally main belt asteroids) that made it to the Earth’s surface after breakup of their parent asteroids. A second population of smaller rocky bodies, called the Kuiper belt, is located beyond the orbit of Neptune (>30 AU). At even greater distances, a large, spherical collection of comets resides, known as the Oort cloud.
1.1.2 Evolution of the protoplanetary disk and accretion of planetary bodies
1.1.2.1 Molecular cloud core collapse and formation of the solar nebula Below follows a concise outline of the most likely formation scenario of the solar system, based both on evidence from astronomy, astrophysics and cosmochemistry. The history of the solar system most likely commenced with the partial gravitational collapse of a giant parental interstellar molecular cloud of gas and dust. This collapsing cloud core was (relatively) cold (T = 10-100 K) and dense (! = 103 - 106 particles/cm3) in comparison to the interstellar medium (ISM). Conversation of angular momentum depicts that the collapse advanced fastest near the rotational poles of the rotating protostellar cloud, thereby generating a mass centre that heated up due to contraction. Dust particles in the protoplanetary cloud lost their vertical velocity component by mutual collisions and settled into the equatorial plane of the (proto)Sun, while angular momentum was transported outward from the mass centre. This ultimately resulted in a flattened protoplanetary disk of gas and dust surrounding a T-Tauri star (i.e., the proto-Sun) (Adams, 2010) (Fig 1.1). Note that the above scenario explains both the current circular and coplanar orbits of the planets and the distribution of angular
Chapter 1 7
momentum in the solar system (Section 1.1.1). Collectively, the protoplanetary disk and the proto-Sun are referred to as the ‘primitive solar nebula’. The protoplanetary disk served to channelize the in-falling material that eventually accreted to the Sun, and also was the locus of processes leading to the accretion of planetary bodies. The solar nebula was likely characterized by a relatively smooth radial thermal gradient with the highest temperatures occurring relatively close to the Sun. This explains why the rocky planets accreted closest to the Sun (at high T), while the gas giants and comets accreted beyond the ‘Snow line’ (at lower T). This thesis is mostly dedicated to the earliest stages of terrestrial planet formation occurring at heliocentric radii inside the Snow line (Section 1.1.2.3).
The modern viewpoint, framed in the so-called ‘Nice model’, is that migration of the giant planets played an important role in the early dynamical evolution of the solar system (Gomes et al., 2005; Morbidelli et al., 2005; Tsiganis et al., 2005; Levison et al., 2011). The Nice model proposes that the four giant planets initially (at ~4.4 Ga) had closely packed, nearly circular orbits. Only after ~600-700 Myr of solar system evolution these orbits were dramatically disturbed, primarily due to resonances between the gas giants and long-term gravitational interactions with planetesimals. This invoked a period of major dynamical instability that involved inward and outward migration of the gas giants, and scattering of planetesimals and comets into highly eccentric and elliptical orbits. Eventually, the four gas giants finally settled into their present, slightly elliptical orbits. The Nice model explains both the existence and current location of the Oort cloud, the Kuiper belt, the asteroid belt, and also the hypothetical occurrence of the Late Heavy Bombardment (at ~3.8 Ga), a brief period of intense bombardment of the inner solar system.
Fig 1.1: Astronomical images of protoplanetary disks (or proplyds) surrounding young stars in the Orion nebula taken by the Hubble Space telescope. Credit: NASA/ESA and L. Ricci (ESO)
8 Introduction
1.1.2.2 Time-zero as defined in cosmochemistry From a cosmochemical perspective the beginning of solar system history is defined by the formation of Ca-Al-rich inclusions (CAI). These are refractory element-rich, cm-sized inclusions contained in carbonaceous chondrites. The mineralogical and chemical characteristics of CAI indicate that they formed under high temperature conditions, and in rare cases CAI might be direct condensates out of a nebular gas (Grossman, 1972; Tanaka and Masuda, 1973; Boynton, 1975; Davis and Grossman, 1979). Their Pb-Pb ages are the oldest of any known natural material that formed in the in the solar system. The absolute age of the solar system as defined by CAI corresponds to ~4567-4568 Ma (Amelin et al., 2010; Bouvier and Wadhwa, 2010). More details on the cosmochemistry of CAI are given in Chapter 7.
1.1.2.3 Planet and planetesimal formation: What is (un)known? The conventional theory of terrestrial planet formation in the protoplanetary disk is that it occurred in four (partly) overlapping stages (Fig. 1.2): (i) Collisional coagulation of dust grains and the subsequent accretion of kilometre sized planetesimals (mostly <10 km), (ii) Runaway growth of planetesimals into increasingly larger (up to r " 103 km) planetary embryos primarily as a result of gravitational focusing (time scales on the order of 105- 107 years), (iii) Slower growth when the larger planetary embryos accrete remaining planetesimals (oligarchic growth), and (iv) Chaotic growth characterized by large collisions (‘Giant impacts’) of planetary embryos due to crossing orbits of planetary embryos (up to Mars size). This stage lasts for ~100 Myr and culminates in the currently existing terrestrial planets (e.g., Chambers, 2004).
Fig. 1.2: Evolution of matter and planet formation processes in the inner regions of a protoplanetary disk around a young star. After Dauphas and Chaussidon (2011).
The main focus of this thesis is on the first two stages, the processing leading to the
accretion and differentiation of km-sized planetesimals and the larger planetary embryos, together essentially the building blocks of the terrestrial planets. At the smallest spatial scales (<1 cm) Van der Waals forces and chemical bonds promote coagulation of microscopic grains into larger objects (Weidenschilling, 1980), while at the largest spatial scales (>1 km) gravitational forces dominate. Growth of objects on these spatial scales is relatively well understood. However, an aspect of planet formation that is much harder to explain is the growth from small cm-sized dust grains towards km-sized planetesimals (see e.g., Chiang and Youdin, 2010 and references therein), commonly referred to as the radial drift and
Chapter 1 9
fragmentation barrier (or ‘meter size’ barrier). Two main issues prevent the accretion of larger planetesimals of >1 km in size (Brauer et al., 2008). First, the large velocity difference between cm-sized objects and m-sized boulders causes mutual erosion or fragmentation (Blum and Wurm, 2008). Second, m-sized objects have very short dynamical lifetimes. Orbiting objects of this size experience a strong headwind from the slower rotating gas, resulting in gas drag – which is maximal for a m-sized objects at 1 AU - and consequently, leads to rapid inward drift towards the central proto-Sun, in which these objects are then lost (e.g., Weidenschilling, 1977). In general, overcoming this problem requires a particle trapping or concentrating mechanism that locally increases the dust-to-gas ratio, promotes gravitational instability and which then leads to the rapid accretion of planetary bodies. In the last decade several mechanisms to overcome this problem have been proposed, all based on numerical models (e.g., Barge, 1995; Johansen et al., 2007; Kretke and Lin, 2007; Cuzzi et al., 2008; Birnstiel et al., 2013). The recent observation of an asymmetric dust trap in the disk around a young star (van der Marel et al., 2013) suggests that models invoking local pressure maxima to build relatively large planetary embryos (Barge, 1995; Birnstiel et al., 2013) might be correct.
1.2 Early solar system chronology as recorded in meteorites Meteorites are small extraterrestrial bodies that entered the Earth’s atmosphere, survived and landed on the Earth’s surface. Most meteorites probably originated as meteoroids or as part of larger asteroids within the asteroid belt between Jupiter and Mars (2-3.3 AU). These asteroids are believed to represent the remaining planetesimals that accreted during the earliest stages of solar system history. When main belt asteroids encounter a Kirkwood Gap (i.e., orbits in a resonance with the orbital period of Jupiter), they are excited into eccentric orbits, which causes them to move towards the inner solar system and sometimes to hit the Earth’s surface. The strong gravitational influence of Jupiter (as e.g., evidenced by the Kirkwood Gaps) might also be the reason that the planetesimals at 2-3.3 AU - a fraction of surviving bodies now being main belt asteroids - could not accrete to a planet beyond Mars. This likely is the very reason that most meteorites exist in the first place (with the exception of lunar and Martian meteorites).
Being the surviving remnants of planetesimals, meteorites provide a direct record of processes and events that occurred in the earliest stages of solar system history. As such they can provide important insights into processes related to the accretion and differentiation of planetesimals and planetary embryos. Investigating their chemical and isotopic signatures provides important information on the sequence of events in the first ~10 Myr of the solar system, and can also shed light on the dynamics of material transport and mixing in the solar protoplanetary disk. Therefore, cosmochemistry provides an essential complement to related disciplines like astronomy and theoretical astrophysics, and ultimately may also help to place constraints on fundamental issues such as the fragmentation and radial drift barrier (see above).
1.2.1 Elemental and isotopic ratios in meteorites Meteorites are grossly divided into (i) primitive (undifferentiated) chondrites which are essentially mechanical mixtures of components with different origins (e.g., chondrules, CAI, metal, matrix) and (ii) differentiated meteorites which derive from parent bodies that have undergone melting and differentiation at some stage (e.g., iron meteorites and achondrites). Undifferentiated meteorites, particularly CI chondrites, have elemental compositions that are
10 Introduction
almost identical to the Sun (within 10% in most cases, and except for the most volatile elements, noble gases and Li) (e.g., Anders and Grevesse, 1989; Lodders, 2003; Palme and Jones, 2003). In addition, the isotopic compositions of most meteorites, both undifferentiated and differentiated, and that of the terrestrial planets are almost identical (mostly within ~0.1%, excluding radioactive decay-induced and mass-dependent variations). These observations indicate that the bulk of the solar system matter overall was chemically and isotopically well mixed. However, smaller-scale isotopic anomalies are ubiquitous in many meteorites, and especially in the different components therein (e.g., CAI). These isotopic variations demonstrate that small chemical and isotopic heterogeneities existed in the primitive solar nebula. The magnitude of the isotopic variability generally is on the order of ~0.01%, but can be significantly larger in some cases (e.g., in pre-solar grains, or in the case of O isotopes).
1.2.2 Short-lived radionuclides in meteorites A fundamental and intriguing property of the solar system is the evidence for the former presence of extinct, short-lived radionuclides (e.g., 26Al, 53Mn, 56Fe, 182Hf, 107Pd,) during the first few Myr of solar system history, as inferred from isotopic anomalies of their decay products in meteorites. As these short-lived nuclides have relatively short half-lives (mostly <10 Myr), only a short time interval can have elapsed between their production and their incorporation in meteorites (Wasserburg et al., 2006). The exact mechanism or source that injected the short-lived radionuclides remains debated, but may have included a nearby supernova (Type II) explosion (e.g., Meyer and Clayton, 2000) or the stellar wind from an AGB star (e.g., Wasserburg et al., 2006). In any scenario the production of the short-lived isotopes must have occurred at spatial and temporal scales matching the birth environment of the solar system. As such they can place important constraints on the astrophysical context of solar system formation and the efficiency of large-scale mixing in the protoplanetary disk (Adams, 2010; Dauphas and Chaussidon, 2011). In addition, the decay of some of these radionuclides (in particular 26Al) most likely provided the heat source for melting and differentiation of the parent bodies of differentiated meteorites (e.g., Urey, 1955; Hevey and Sanders, 2006). Finally, some short-lived radioactive decay systems can be used to our advantage and provide relative ages for meteorites (see Sections 1.2.3-1.2.4).
1.2.3 Isotope chronometry of meteorites The accretion and differentiation of planetesimals and planetary embryos likely occurred within a very narrow time interval (~10 Myr) during the earliest stages of solar system history (see Section 1.1.2.3). Precisely reconstructing the sequence of events in the early solar system may help to better understand the conditions and mechanisms related to the accretion and differentiation of planetesimals or planetary embryos. Such a chronology of events in the early solar system can be obtained by dating meteorites and their components using radioisotope chronometers. Distinguishing early solar system events or processes clearly requires a very high temporal resolution. This necessitates high precision measurements of isotope ratios and the application of radioisotope chronometers that provide the required resolution for the relevant time window in the first few, to first few tens, of millions of years after the start of the solar system.
1.2.3.1 Short-lived vs. long-lived decay systems Radioactive decay systems used for dating meteorites can be subdivided into long-lived and short-lived chronometers. As their parent isotopes have been decaying until the present-day,
Chapter 1 11
long-lived decay systems provide absolute age information (Fig. 1.3). An important example is the 207Pb-206Pb chronometer, which provides relatively precise absolute ages for meteorites and their components (Amelin et al., 2002; Jacobsen et al., 2008; Amelin et al., 2010; Blichert-Toft et al., 2010; Bouvier and Wadhwa, 2010; Connelly et al., 2012). Short-lived radionuclides are unstable nuclides that existed at the beginning of solar system history but that have since then decayed. Hence, their long-gone presence can only be detected by studying the isotope composition of their daughter isotopes. The great potential of short-lived chronometers is that the abundance of the radioactive daughter isotopes decreased rapidly over relatively short time intervals (Fig. 1.3). Therefore, by measuring the isotopic composition their daughter isotopes, such short-lived chronometers can provide ages with a very high temporal resolution, which is a necessity for dating events within the first Myr of solar system history (e.g., Halliday and Kleine, 2006; Kleine and Rudge, 2011). An important assumption is that the abundance of the short-lived parent nuclides was homogeneously distributed in the early solar system. This assumption can be tested by cross-calibration of different long-lived and short-lived chronometers (Nyquist et al., 2009). In addition, this approach permits to anchor relative ages from short-lived chronometers to an absolute time scale. Important examples of short-lived chronometers that are highly relevant for early solar system chronology include the 26Al-26Mg (t1/2 " 0.7 Myr) and 182Hf-182W (t1/2 " 8.9 Myr) decay systems.
Fig. 1.3: Concept of dating with short-lived and long-live isotope chronometers, showing the abundance of long-lived and short-lived radioactive daughter isotopes plotted vs. time.
1.2.4 Tungsten isotopes and the 182Hf-182W chronometer Tungsten (Z=74) is a refractory element with a 50% condensation temperature of 1790 K and five (now) stable isotopes. The major W isotopes are produced by both the s- and r-process (180W, 182W, 183W, 184W and 186W), while the minor isotope 180W (0.13%) represents a p-process nuclide (see Chapter 7 for details on nucleosynthesis). The radioactive decay of now extinct 182Hf (t1/2 = 8.9 Myr; (Vockenhuber et al., 2004)) to 182W provides a chronometer for dating chemical and physical processes that fractionate Hf and W during the first ~60 Myr of solar system history (Kleine et al., 2009). This time interval is particularly suited for studying
0
50
100
4.24.34.44.5
!"#$"%&'%(&)#$%(*("+,Long-lived decay system:- Absolute ages- e.g., 238U (t1/2 = 4.47 Gyr)
Time before present (Ga)
Radi
oact
ive p
aren
t rem
anin
g (%
)
Short-lived decay system:- Only relative ages for narrow time interval- High time resolution- e.g., 182Hf (t1/2 = 8.9 Myr)
12 Introduction
the processes leading to the accretion and differentiation, and evolution of planetary bodies in the inner solar system.
The Hf-W chronometer can be used to study early solar system processes in two principle ways. First of all, the Hf-W system can provide model ages of core formation in a planetary body. This is because both Hf and W are refractory elements but have different geochemical affinities during metal-silicate separation. As W is moderately siderophile and Hf strongly lithophile, metal-silicate separation (i.e., core formation) in a planetary body fractionates the Hf/W ratio, resulting in high Hf/W in the silicate portion and Hf/W of essentially zero in the metal core (Fig. 1.4). Hence, the Hf-W system is ideally suited to investigate the timing of metal-silicate separation in planetary bodies that accreted during the earliest stages of solar system history (i.e., within the effective lifetime of 182Hf) (Kleine et al., 2009). As such, Hf-W chronometry of meteorite samples can place strong constraints on the accretion and differentiation history of planetesimals and planetary embryos (Chapters 2-6).
Second, the Hf-W chronometer can help to constrain the cooling or differentiation history of individual meteorite parent bodies (e.g., of ordinary chondrites or angrites, respectively) or Hf/W fractionation in the solar nebula (e.g., from CAI or carbonaceous chondrites) (e.g., Markowski et al., 2007; Burkhardt et al., 2008; Kleine et al., 2008; Touboul et al., 2009; Kleine et al., 2012). As the constituent mineral phases of meteorites and their components commonly have variable Hf/W (i.e., due to the incorporation of Hf and W in varying proportions), an internal Hf-W isochron of a sample precisely dates the last isotopic equilibration (i.e., closure of the Hf-W system). Due to its high closure temperature the Hf-W system is not as susceptible to resetting as most other chronometers, and hence can provide unique constraints on the high-temperature evolution of meteorite (parent bodies) and their components during the earliest stages of solar system evolution.
Fig. 1.4: Explanatory diagram showing 182W/184W vs. time with evolution lines for reservoirs with distinct 180Hf/184W. Upon metal-silicate separation (red star) in a planetary body at a given time after solar system formation (blue square), the silicate mantle (high Hf/W) follows a steeper trajectory than the chondritic reservoir, whereas the 182W/184W of the metal core (Hf/W ! 0) is invariant from that point in time onward.
1.2.4.1 W isotope systematics in meteorites The mass bias-corrected ratio of two W isotopes (iW/jW) measured for a sample is typically expressed relative to the isotopic composition of the terrestrial mantle, represented by a terrestrial standard reference material:
Chondritic reservoir
180Hf/184W 1.23
Metal core
Silic
ate m
antle
high H
f/W
Time [Myr]
182 W
/184 W
Solar system formation
CAI
Metal-silicate separation
Metal coreHf/W = 0
Silicate mantle
Hf/W >> 1
Chapter 1 13
Eq. 1.1
In addition to radiogenic contributions from 182Hf decay, there are two other processes
that can generate W isotope anomalies in meteorites (Fig. 1.5). These include (i) neutron capture effects (Section 2.3) induced during exposure of the meteoroids to galactic cosmic rays (Masarik, 1997), and (ii) nucleosynthetic isotope anomalies (Section 7.1) reflecting a heterogeneous distribution of presolar components in the early solar nebula. In contrast to cosmogenic and radiogenic effects, which only affect #182W, nucleosynthetic W isotope heterogeneities generate both anomalies in #182W and collateral #183W anomalies (Burkhardt et al., 2008; Qin et al., 2008; Burkhardt et al., 2012). [Note that the measured W isotope composition of a meteorite sample (after mass bias correction) represents the net effect of these processes on all isotopes involved (i.e., 182W, 183W, 184W, 186W)]. Thus, while #182W anomalies can result from a combination of radiogenic, cosmogenic and nucleosynthetic isotope effects, the #183W variations in CAI solely result from nucleosynthetic W isotope heterogeneities. From Fig. 1.5 it is clear that obtaining accurate Hf-W ages for meteorites requires the quantification of any neutron capture effects or nucleosynthetic W isotope heterogeneity that might be present.
Fig. 1.5: Explanatory diagram of "182W vs. "183W showing the possible anomalies on W isotope compositions in meteorites, which may derive from (i) radiogenic contributions from 182Hf decay, (ii) neutron capture effects induced during cosmic-ray exposure, and (iii) nucleosynthetic variability due to a heterogeneous distribution of s- and/or r-process W isotopes. The subscript ‘6/4’ refers to the normalisation ratio used for mass bias correction (i.e., 186W/184W).
1.2.4.2 Hf-W chronometry of metal samples Iron meteorites do not incorporate Hf and thus retained the 182W/184W at the time of core formation in their parent bodies (Fig. 1.4, 1.6). A two-stage model age of metal segregation relative to CAI formation can be calculated by comparing the measured 182W/184W of the iron
! iW =iW jWspliW jWstd
"1#$%
&%
'(%
)%*104
–1 –0.5 0 0.5 1–4
–2
0
2
4
ε183W (6/4)
ε182
W (6
/4)
182Hf decay
n-capture
nucleosynthetic variability
14 Introduction
meteorite to the evolution line of a reservoir with chondritic 180Hf/184W. This is expressed in the following equation:
!tCAI ="1l#
$%&
'()ln 1"
182W184W
#
$%
&
'(
Iron
"182W184W
#
$%
&
'(
CAI180Hf184W
#
$%
&
'(
cc
)182Hf180Hf
#
$%
&
'(
CAI
*
+,,
-
,,
.
/,,
0
,, Eq. 1.2
where 182W/184Wiron is the measured value of an iron meteorite, 182W/184WCAI and 182Hf/180HfCAI are the initial ratios of CAI, 180Hf/184W is the average value of ~1.23±0.15 obtained for carbonaceous chondrites (Kleine et al., 2004), and ! is the 182Hf decay constant of 0.078±0.002 Myr-1 (Vockenhuber et al., 2004).
Fig. 1.6: Illustration of two-stage Hf-W model ages for iron meteorites. The solid black line shows the evolution line of a reservoir with chondritic Hf/W. The measured "182W of iron meteorites (diamond symbols) can be directly compared to this curve to obtain a model age of metal segregation relative to the time of CAI formation (orange star).
Importantly, the Hf-W model ages of iron meteorites critically rely on the assumption
that the iron meteorite parent bodies accreted from a reservoir with chondritic Hf/W. As the primitive solar nebula was chemically (and isotopically) well mixed, especially with respect to refractory elements like Hf and W (Section 1.2.1), this appears to be a generally good assumption suited for most scenarios. However, metal-silicate separation in the solar nebula can also fractionate the Hf/W ratio (Kleine et al., 2008; Touboul et al., 2009), and therefore, as will be shown in Chapter 5, the assumption of a chondritic Hf/W might not necessarily be valid for iron meteorites.
1.2.4.3 Hf-W chronometry of silicate rock samples The Hf-W systematics of silicate rock samples can provide relative ages of metal-silicate separation. Conservation of mass depicts that the present-day 182W/184W of a sample is given by the following equation:
0 1 2 3 4 5–3.6
–3.4
–3.2
–3.0
ΔtCAI [Myr]
ε182
W
t0 (CAI)
t1
t2
t3
t3 − t0
t2 − t0
t1 − t0
Solar 180 Hf/1
84 W
Chapter 1 15
182W184W!
"#
$
%&0
=182W184W!
"#
$
%&i
+180Hf184W
!
"#
$
%& '
182Hf180Hf!
"#
$
%&i
Eq. 1.3
where the ratio at the present-day are indicated by subscript ‘0’ and the initial ratios by the subscript ‘i’. Eq. 1.3 demonstrates that a plot of present-day 182W/184W vs. 180Hf/184W defines an isochron whose slope provides the initial 182Hf/180Hf and its intercept the initial 182W/184W composition for a suite of samples (or for one particular sample in the case of internal mineral isochrons). The different initial 182Hf/180Hf (i.e., the isochron slope) of two samples can be used to calculate relative Hf-W ages using:
!t = " 1#$ ln
182Hf 180Hf( )1182Hf 180Hf( )2
%&'
('
)*'
+' Eq. 1.4
where $ is the decay constant of 182Hf of 0.078±0.002 Myr-1. The above principles will be applied to the Hf-W data obtained for bulk CAI in Chapter 8. Note again that for the Hf-W chronometry iron meteorites (Chapters 3-5) the above is not possible because only the low Hf/W (i.e., the metal) component is available for study. Hence, only two-stage model ages of metal segregation can be obtained (Section 1.2.4.2).
1.3 Aim of this thesis The principle aim of this dissertation is to investigate the time interval over which planetesimals and planetary embryos accreted and segregated their metal cores in the solar protoplanetary disk using Hf-W chronometry. This objective can be addressed through high precision W isotope measurements in iron meteorites, which allows obtaining precise Hf-W ages of core formation for their parent bodies. However, two fundamental issues in Hf-W chronometry currently preclude constraining an accurate chronology of core formation:
(A) The presence of superimposed, cosmic ray-induced W isotope shifts in iron
meteorites. These secondary neutron capture effects lower the 182W/184W of iron meteorites and bias the inferred Hf-W ages to erroneously old values.
(B) The presence of nucleosynthetic W isotope anomalies in CAI, which hamper obtaining accurate knowledge about the initial Hf and W isotope compositions of the solar system.
The main objective of this dissertation is to obtain a solid solution for these two problems.
The first issue requires the development of a direct neutron fluence dosimeter for iron meteorites and will be addressed in Part A of this thesis (Chapters 2-6). The second issue warrants a detailed study of the magnitude and nature of nucleosynthetic W isotope anomalies in CAI, as will be discussed in Part B of this thesis (Chapters 7-9). Once the effects of neutron capture and nucleosynthetic W isotope heterogeneity are quantified, the Hf-W system will provide fundamentally new insights into the accretion and differentiation history of planetesimals, and ultimately, may help to understand the processes leading to the formation of terrestrial planets like the Earth.
16 Introduction
References Adams F. C. (2010) The Birth Environment of the Solar System. Annual Review of
Astronomy and Astrophysics 48, 47-85. Amelin Y., Kaltenbach A., Iizuka T., Stirling C. H., Ireland T. R., Petaev M. and Jacobsen S.
B. (2010) U-Pb chronology of the Solar System's oldest solids with variable 238U/235U. Earth and Planetary Science Letters 300, 343-350.
Amelin Y., Krot A. N., Hutcheon I. D. and Ulyanov A. A. (2002) Lead isotopic ages of chondrules and calcium-aluminum-rich inclusions. Science 297, 1678-1683.
Anders E. and Grevesse N. (1989) Abundances of the elements: Meteoritic and solar. Geochimica et Cosmochimica Acta 53, 197-214.
Barge (1995) Did planet formation begin inside persistent gaseous vortices? Astronomy and Astrophysics 295, L1.
Birnstiel T., Dullemond C. P. and Pinilla P. (2013) Lopsided dust rings in transition disks. A&A 550, L8.
Blichert-Toft J., Moynier F., Lee C. T. A., Telouk P. and Albarede F. (2010) The early formation of the IVA iron meteorite parent body. Earth and Planetary Science Letters 296, 469-480.
Blum J. and Wurm G. (2008) The Growth Mechanisms of Macroscopic Bodies in Protoplanetary Disks. Annual Review of Astronomy and Astrophysics 46, 21-56.
Bouvier A. and Wadhwa M. (2010) The age of the Solar System redefined by the oldest Pb-Pb age of a meteoritic inclusion. Nature Geoscience 3, 637-641.
Boynton W. V. (1975) Fractionation in the solar nebula: condensation of yttrium and the rare earth elements. Geochimica et Cosmochimica Acta 39, 569-584.
Brauer F., Dullemond C. P. and Henning T. (2008) Coagulation, fragmentation and radial motion of solid particles in protoplanetary disks. A&A 480, 859-877.
Burkhardt C., Kleine T., Bourdon B., Palme H., Zipfel J., Friedrich J. M. and Ebel D. S. (2008) Hf-W mineral isochron for Ca,Al-rich inclusions: Age of the solar system and the timing of core formation in planetesimals. Geochimica et Cosmochimica Acta 72, 6177-6197.
Burkhardt C., Kleine T., Dauphas N. and Wieler R. (2012) Nucleosynthetic tungsten isotope anomalies in acid leachates of the Murchison chondrite: Implications for hafnium-tungsten chronometry. The Astrophysical Journal Letters 753, L6.
Chambers J. E. (2004) Planetary accretion in the inner Solar System. Earth and Planetary Science Letters 223, 241-252.
Chiang E. and Youdin A. N. (2010) Forming Planetesimals in Solar and Extrasolar Nebulae. Annual Review of Earth and Planetary Sciences 38, 493-522.
Connelly J. N., Bizzarro M., Krot A. N., Nordlund Å., Wielandt D. and Ivanova M. A. (2012) The Absolute Chronology and Thermal Processing of Solids in the Solar Protoplanetary Disk. Science 338, 651-655.
Cuzzi J. N., Hogan R. C. and Shariff K. (2008) Toward Planetesimals: Dense Chondrule Clumps in the Protoplanetary Nebula. Astrophysical Journal 687, 1432-1447.
Dauphas N. and Chaussidon M. (2011) A Perspective from Extinct Radionuclides on a Young Stellar Object: The Sun and Its Accretion Disk. Annual Review of Earth and Planetary Sciences 39, 351-386.
Davis A. M. and Grossman L. (1979) Condensation and fractionation of rare earths in the solar nebula. Geochimica et Cosmochimica Acta 43, 1611-1632.
Gomes R., Levison H. F., Tsiganis K. and Morbidelli A. (2005) Origin of the cataclysmic Late Heavy Bombardment period of the terrestrial planets. Nature 435, 466-469.
Grossman L. (1972) Condensation in Primitive Solar Nebula. Geochimica Et Cosmochimica Acta 36, 597-&.
Halliday A. N. and Kleine T., 2006. Meteorites and the Timing, Mechanisms, and Conditions of Terrestrial Planet Accretion and Early Differentiation. In: Lauretta, D. S. and McSween, H. Y. Eds.), Meteorites and the Early Solar System II. The University of Arizona Press.
Chapter 1 17
Hevey P. J. and Sanders I. S. (2006) A model for planetesimal meltdown by Al-26 and its implications for meteorite parent bodies. Meteoritics & Planetary Science 41, 95-106.
Jacobsen B., Yin Q. Z., Moynier F., Amelin Y., Krot A. N., Nagashima K., Hutcheon I. D. and Palme H. (2008) Al-26-Mg-26 and Pb-207-Pb-206 systematics of Allende CAIs: Canonical solar initial Al-26/Al-27 ratio reinstated. Earth and Planetary Science Letters 272, 353-364.
Johansen A., Oishi J. S., Mac Low M. M., Klahr H. and Henning T. (2007) Rapid planetesimal formation in turbulent circumstellar disks. Nature 448, 1022-1025.
Kleine T., Hans U., Irving A. J. and Bourdon B. (2012) Chronology of the angrite parent body and implications for core formation in protoplanets. Geochimica et Cosmochimica Acta 84, 186-203.
Kleine T., Mezger K., Munker C., Palme H. and Bischoff A. (2004) Hf-182-W-182 isotope systematics of chondrites, eucrites, and martian meteorites: Chronology of core formation and early mantle differentiation in Vesta and Mars. Geochimica Et Cosmochimica Acta 68, 2935-2946.
Kleine T. and Rudge J. F. (2011) Chronometry of Meteorites and the Formation of the Earth and Moon. Elements 7 41-46.
Kleine T., Touboul M., Bourdon B., Nimmo F., Mezger K., Palme H., Jacobsen S. B., Yin Q. Z. and Halliday A. N. (2009) Hf-W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochimica Et Cosmochimica Acta 73, 5150-5188.
Kleine T., Touboul M., Van Orman J. A., Bourdon B., Maden C., Mezger K. and Halliday A. N. (2008) Hf-W thermochronometry: Closure temperature and constraints on the accretion and cooling history of the H chondrite parent body. Earth and Planetary Science Letters 270, 106-118.
Kretke K. A. and Lin D. N. C. (2007) Grain Retention and Formation of Planetesimals near the Snow Line in MRI-driven Turbulent Protoplanetary Disks. The Astrophysical Journal Letters 664, L55.
Levison H. F., Morbidelli A., Tsiganis K., Nesvorny D., Gomes R. and Rodney G. (2011) Late Orbital Instabilities in the Outer Planets Induced by Interaction with a Self-gravitating Planetesimal Disk. The Astronomical Journal 142, 152.
Lodders K. (2003) Solar system abundances and condensation temperatures of the elements. The Astrophysical Journal 591, 1220-1247.
Markowski A., Quitte G., Kleine T., Halliday A. N., Bizzarro M. and Irving A. J. (2007) Hafnium-tungsten chronometry of angrites and the earliest evolution of planetary objects. Earth and Planetary Science Letters 262, 214-229.
Masarik J. (1997) Contribution of neutron-capture reactions to observed tungsten isotopic ratios. Earth and Planetary Science Letters 152, 181-185.
Meyer B. and Clayton D. (2000) Short-Lived Radioactivities and the Birth of the sun. Space Science Reviews 92, 133-152.
Morbidelli A., Levison H. F., Tsiganis K. and Gomes R. (2005) Chaotic capture of Jupiter's Trojan asteroids in the early Solar System. Nature 435, 462-465.
Nyquist L. E., Kleine T., Shih C. Y. and Reese Y. D. (2009) The distribution of short-lived radioisotopes in the early solar system and the chronology of asteroid accretion, differentiation, and secondary mineralization. Geochimica et Cosmochimica Acta 73, 5115-5136.
Palme H. and Jones A., 2003. 1.03 - Solar System Abundances of the Elements. In: Editors-in-Chief: Heinrich, D. H. and Karl, K. T. Eds.), Treatise on Geochemistry. Pergamon, Oxford.
Qin L. P., Dauphas N., Wadhwa M., Markowski A., Gallino R., Janney P. E. and Bouman C. (2008) Tungsten nuclear anomalies in planetesimal cores. Astrophysical Journal 674, 1234-1241.
Tanaka T. and Masuda A. (1973) Rare-earth elements in matrix, inclusions, and chondrules of the Allende meteorite. Icarus 19, 523-530.
18 Introduction
Touboul M., Kleine T., Bourdon B., Van Orman J. A., Maden C. and Zipfel J. (2009) Hf-W thermochronometry: II. Accretion and thermal history of the acapulcoite-lodranite parent body. Earth and Planetary Science Letters 284, 168-178.
Tsiganis K., Gomes R., Morbidelli A. and Levison H. F. (2005) Origin of the orbital architecture of the giant planets of the Solar System. Nature 435, 459-461.
Urey H. C. (1955) The cosmic abundances of potassium, uranium, and thorium and the heat balances of the Earth, the Moon, and Mars. Proc. Natl. Acad. Sci. U.S. 41, 127-144.
van der Marel N., van Dishoeck E. F., Bruderer S., Birnstiel T., Pinilla P., Dullemond C. P., van Kempen T. A., Schmalzl M., Brown J. M., Herczeg G. J., Mathews G. S. and Geers V. (2013) A Major Asymmetric Dust Trap in a Transition Disk. Science 340, 1199-1202.
Vockenhuber C., Oberli F., Bichler M., Ahmad I., Quitte G., Meier M., Halliday A. N., Lee D. C., Kutschera W., Steier P., Gehrke R. J. and Helmer R. G. (2004) New half-life measurement of Hf-182: Improved chronometer for the early solar system. Physical Review Letters 93.
Wasserburg G. J., Busso M., Gallino R. and Nollett K. M. (2006) Short-lived nuclei in the early Solar System: Possible AGB sources. Nuclear Physics A 777, 5-69.
Weidenschilling S. J. (1977) Aerodynamics of solid bodies in the solar nebula. Mon. Not. R. Astron. Soc. 180.
Weidenschilling S. J. (1980) Dust to planetesimals: Settling and coagulation in the solar nebula. Icarus 44, 172-189.
Part A Chapter 2 19
PART A
QUANTIFICATION OF NEUTRON CAPTURE EFFECTS AND THE Hf-W CHRONOLOGY OF IRON METEORITES
20 Introduction to Part A
Chapter 2
Introduction to part A
Part A Chapter 2 21
2.1 Core formation in planetary bodies Planetary differentiation into a silicate mantle and a metal core represents the most important differentiation even in the history of a planetary body (e.g., Rubie et al., 2007). Upon melting of an undifferentiated planetary body the metal portion – due to its high density – will segregate towards the centre of the object. As core formation requires substantial melting of the interior of a planetary body, an effective heat source is needed. Three different heat sources can provide the energy required for melting: (i) the decay of short-lived radionuclides, in particular of 26Al (ii) the kinetic energy delivered by impacts, and (iii) the reduction of gravitational potential energy. While the latter processes are most relevant for larger planetary bodies that accreted relatively late (e.g., Earth, Moon, Mars), the decay of short-lived 26Al (t1/2 ! 0.7 Myr) was the most effective heat source for melting and differentiation of planetesimals and planetary embryos that accreted within the first few Myr of solar system history (e.g., Urey, 1955; Walter and Trønnes, 2004; Hevey and Sanders, 2006; Rubie et al., 2007). Due to its rapid decay the potential of 26Al as heat source diminished rapidly after ~1.5 Myr of solar system evolution. As differentiation is unavoidable once melting has started, planetary accretion and core formation are intimately linked processes (Rubie et al., 2007). This is particularly true if accretion and core formation occur simultaneously as a series of discrete events over a longer period of time (i.e., in contrast to a single event of metal segregation after instantaneous accretion of a planetary body). Therefore, understanding core formation is essential for constraining the timescales and conditions of planetesimal accretion and the subsequent high-temperature evolution of such planetary bodies.
2.2 Iron meteorites as remnants of planetesimal cores
2.2.1 Classification and characteristics
Magmatic iron meteorites consist mainly of metallic iron or Fe-Ni alloys and are interpreted to be the fragments of metal cores of planetesimals (e.g., Scott, 1972; Scott and Wasson, 1975). Hence, magmatic irons provide a direct opportunity to study the differentiation, cooling and crystallisation history of the metallic cores of asteroid-sized parent bodies (Chabot and Haack, 2006). The trace element signatures of iron meteorites indicate that the metal cores formed after metal-silicate separation, and subsequent cooling and fractional crystallisation of metallic melt (e.g., Scott, 1972; Chabot and Drake, 1999; Cook et al., 2004; Campbell and Humayun, 2005; Chabot and Haack, 2006; Walker et al., 2008; Goldstein et al., 2009; McCoy et al., 2011). The magmatic iron meteorites are classified into eight different groups (IC, IIAB, IIC, IIG, IIIAB, IIIF, IVA, IVB) based on their chemical composition (Scott and Wasson, 1975), and each group is interpreted to sample a distinct parent body. Specifically, the different iron meteorite groups are distinguished using their distinct abundances of the moderately volatile elements Ge and Ga (with 50% condensation temperatures of 883 and 968 K, respectively). These variations likely result from variable degrees of volatile element depletion of the different iron meteorite parent bodies (e.g., Davis, 2006 and references therein). Therefore, chemical and isotopic analysis of iron meteorites can potentially also provide insightful information on processes in the solar protoplanetary disk that acted upon the precursor material of the iron meteorites. Finally, the non-magmatic iron meteorites (groups IAB, IIICD, IIE) have trace element patterns that are incompatible with fractional crystallisation and likely do not derive from planetary cores (Scott and Wasson,
22 Introduction to Part A
1975; Choi et al., 1995; Benedix et al., 2000; Wasson and Kallemeyn, 2002), and hence will not be further considered in this thesis.
2.2.1.1 Chronology of iron meteorites Dating iron meteorites may help to reconstruct the detailed sequence of events involved in the accretion, differentiation, and cooling history of their parent bodies. Iron meteorites have been dated using several long-lived and short-lived decay systems, including the 187Re-187Os, 207Pb-206Pb, 107Pd-107Ag, 53Mn-53Cr and 182Hf-182W chronometers (e.g., Chen and Wasserburg, 1990; Smoliar et al., 1996; Horan et al., 1998; Sugiura and Hoshino, 2003; Cook et al., 2004; Kleine et al., 2005; Qin et al., 2008; Blichert-Toft et al., 2010; Theis et al., 2013). The Hf-W chronometry of iron meteorites is a major focus of this thesis and is outlined in the paragraph below. Collectively, the available Pd-Ag, Re-Os and Mn-Cr ages suggest that the parent bodies cooled and crystallized within a few tens of Myr after the start of the solar system (e.g., Chen and Wasserburg, 1990; Smoliar et al., 1996; Sugiura and Hoshino, 2003; Cook et al., 2004). Specifically, Re-Os chronometry yields indistinguishable core crystallization ages of 4530±50, 4517±32 Ma, 4540±17 Ma, and 4579±34 Ma for the IIAB, IIIAB, IVA, and IVB iron meteorite parent bodies (Cook et al., 2004; Walker et al., 2008; McCoy et al., 2011). The Pd-Ag systematics of magmatic iron meteorites indicates that their cores cooled within ~10 Myr of each other and around ~10-15 Myr after the start of solar system history (Chen and Wasserburg, 1990; Schönbächler et al., 2008). A recent 207Pb-206Pb age for the IVA iron Muonionalusta even provides evidence that the IVA core crystallized within a few Myr years after CAI formation (Blichert-Toft et al., 2010).
2.2.1.2 Hf-W studies of iron meteorites and neutron capture-induced W isotope variations
The first Hf-W studies on iron meteorites (Lee and Halliday, 1995; Horan et al., 1998) revealed that iron meteorites exhibit strong deficits in 182W. Later studies showed with higher precision that 182W deficits are variable among different iron meteorites, and that many have 182W/184W indistinguishable from or even lower than the initial W isotope composition of CAI, suggesting that iron meteorites accreted and segregated their cores at the start of solar system history as defined by formation of CAI (Horan et al., 1998; Kleine et al., 2005; Markowski et al., 2006; Scherstén et al., 2006; Qin et al., 2008), i.e., predating the formation of chondrules and accretion of (undifferentiated) chondrite parent bodies (Kleine et al., 2005). Accretion of differentiated parent bodies (i.e., iron meteorites) before the primitive, undifferentiated parent bodies would be consistent with the abundance of 26Al at parent body accretion being the dominant parameter controlling the thermal history of planetesimals (Urey, 1955; Hevey and Sanders, 2006; Kleine et al., 2009; Kleine and Rudge, 2011).
However, a major problem when dating iron meteorites using Hf-W chronometry is that neutron capture effects induced during cosmic ray exposure of the iron meteoroids can significantly modify W isotope compositions (Masarik, 1997; Leya et al., 2000; 2003; Leya and Masarik, 2013). These neutron capture-induced shifts cause a net decrease 182W/184W (after mass bias correction) by up to ~1 "-unit, making inferred core formation ages older by up to ~10 Myr (Kleine et al., 2009). Given that planetesimal accretion and differentiation likely were rapid processes (Section 1.1.2.3), cosmic ray-induced shifts thus severely limit the ability of the Hf-W system to precisely constrain the timing of core formation in iron meteorite parent bodies. Therefore, a reliable method for quantifying the magnitude of cosmic ray-induced W isotope variations in iron meteorites is required.
Part A Chapter 2 23
2.3 Cosmic ray exposure and secondary neutron capture effects in iron meteoroids
As already briefly highlighted in Chapter 1, the irradiation of meteoroids by galactic cosmic rays and subsequent secondary neutron capture can cause measureable differences in the isotopic compositions of (mostly heavy) elements commonly used in early solar system chronology (e.g., Lingenfelter et al., 1972; Masarik, 1997; Leya et al., 2003; Leya and Masarik, 2013). Below follows a concise summary of cosmic ray effects in meteoroids, mostly focused on the processes leading to secondary neutron capture effects in iron meteorites.
Fig. 2.1: Illustration of cosmic ray effects in a spherical meteoroid. Shown are (i) primary cosmic ray particles hitting the meteoroid surface, (ii) the production of cosmogenic nuclides (and noble gases), and (iii) the production and moderation of secondary neutrons to lower energies with increasing distance from the meteoroid surface. Neutron capture probabilities at (epi)thermal energies are highest at some depth within the meteoroid (purple band).
Meteoroids are continuously exposed to galactic cosmic rays during their lifetime as
small objects in interplanetary space. With exposure ages on the order of several 100 Ma (and up to ~1500 Ma), the duration of cosmic ray exposure turns out to be particularly long for iron meteoroids (e.g., Wieler, 2002; Eugster, 2003 and references therein). The effects of cosmic ray exposure in an iron meteoroid are illustrated in Fig. 2.1. Primary cosmic ray particles are highly energetic (~0.1–~10 GeV) and mostly consist of protons and alpha particles. Interaction of primary cosmic ray particles with target atoms in the upper parts of meteoroids generates a cascade of nuclear reactions, producing - among other products - secondary high-energy protons and secondary neutrons in the energy range of a few to a few hundred MeV. Primary and secondary high-energy particles produce cosmogenic nuclides predominantly by spallation-type reactions, including cosmogenic noble gas isotopes whose concentrations are a measure for the fluence of medium- to high-energy particles in a meteorite sample (e.g., Wieler, 2002; Ammon et al., 2009; Leya and Masarik, 2009). Concurrently, the secondary cosmic ray particles are slowed down, either due to electronic stopping in the case of secondary protons or by elastic scattering in the case of secondary
Galactic cosmic raysPrimary protons
Epithermal energies0.025 eV - few keV
Thermal energies~0.025 eV
Production of secondary
neutrons
centre surface
24 Introduction to Part A
neutrons. The latter are moderated to lower-epithermal (~0.025 eV to ~10 keV) and eventually to thermal (~0.025 eV at 293 K) energies with increasing distance from the meteoroid surface (Lingenfelter et al., 1972; Leya and Masarik, 2013). The resulting neutron energy distribution depends on the radius of the meteoroid, the shielding depth of the sample therein, and also on the chemical composition of the irradiated object, referred to as the matrix effect (Masarik and Reedy, 1994; Kollár et al., 2006; Leya and Masarik, 2009; 2013). An important consequence of the matrix effect for iron meteoroids (i.e., Fe-dominated matrices) is that (i) the number of secondary particles iron meteoroids is proportionally higher than in silicate-rich targets, and (ii) that these are relatively efficiently moderated to lower energies. This, together with their long exposure ages, suggests that relatively strong neutron capture effects can be expected in iron meteorites.
Neutron capture cross-sections are a measure of the probability for a particular target nuclide to capture a neutron, and are expressed as a function of neutron energy in so-called excitation functions [#(n,$) in barns (10%24 cm) vs. neutron energy in MeV]. Two energy regions are of importance with respect to neutron capture cross-sections: (i) For low (thermal) neutron energies the capture cross-section scales inversely with the neutron velocity (# ~ 1/&), while (ii) at higher (epithermal) energies, capture resonances – i.e., narrow energy ranges at which the capture probability is largely enhanced - occur superimposed on the 1/v dependence. Therefore, the probability for a particular target atom to capture a neutron is strongly dependent on the neutron energy spectrum. In this context we note here that most of the heavy elements (including W) have relatively large neutron capture resonances, and that (ii) the major capture resonance of 56Fe dominates neutron capture in iron meteoroids (Leya and Masarik, 2013). The consequences of these observations for neutron capture effects in iron meteoroids are discussed in detail in Chapters 4 and 6. From all of the above it is also clear that the production of cosmogenic nuclides (including noble gases) and neutron capture on W isotopes are fundamentally distinct processes, occurring at different energies and depths in the iron meteoroid. This will be discussed in detail in Chapter 3. Finally, we note that neutron capture rates can be theoretically modelled using evaluated neutron capture cross section data files of target atoms and sophisticated particle spectra (i.e., as a function of meteoroid size and depth therein) (Leya and Masarik, 2013). The results of new model calculations by I. Leya (Univ. Bern) for neutron capture in iron meteoroids are presented in the Chapters 3, 4, and 6.
2.4 Aim and outline of Part A The major aim of Part A of this thesis is to quantify cosmic ray-induced shifts on W isotope compositions in iron meteorites. The quantification of neutron capture effects on W isotopes in iron meteorites will allow to precisely (re)define the chronology of metal segregation of their parent bodies. This in turn will provide new and firm constraints on the time scales of planetesimal accretion and differentiation in the solar protoplanetary disk. More specific objectives that will be addressed are: (i) assessing over what time interval (after CAI formation) the iron meteorite parent bodies segregated their cores, (ii) investigating whether different parent bodies segregated their cores concurrently or at distinct times, and to a lesser extent, (iii) examining whether each particular parent body accreted and segregated its core instantaneously, or alternatively, over a longer time interval that involved several events of metal-silicate separation.
The quantification of cosmic ray-induced shifts on W isotope compositions in iron meteorites can be accomplished using two different, but complementary approaches: (i) the
Part A Chapter 2 25
identification of samples with minimal cosmic ray effects, and (ii) the correction of cosmic ray-induced W isotope variations using a direct neutron dosimeter. An example of the former approach is described in Chapter 3, which reports on the results of a combined noble gas and W isotope study that is focused on identifying iron meteorite samples with minimal cosmic-ray effects. The best approach to quantify neutron capture-induced shifts in iron meteorites is by measuring the neutron fluence directly using an independent neutron dose monitor. In Chapter 4 it will be shown that Pt isotopes represent an excellent neutron dosimeter and that combined Pt-W isotope analyses can be used to obtain accurate core formation ages of iron meteorites. In Chapter 5, we examine the Hf-W chronology of different iron meteorite parent bodies in greater detail by obtaining combined Pt-W isotope data for each of the major iron meteorite groups. In Chapter 6, we present new Cd concentration data for iron meteorites and explore neutron capture effects more elaborately using combined Pt-Cd-W isotope data.
References Ammon K., Masarik J. and Leya I. (2009) New model calculations for the production rates of
cosmogenic nuclides in iron meteorites. Meteoritics & Planetary Science 44, 485-503.
Benedix G. K., McCoy T. J., Keil K. and Love S. G. (2000) A petrologic study of the IAB iron meteorites: Constraints on the formation of the IAB-Winonaite parent body. Meteoritics & Planetary Science 35, 1127-1141.
Blichert-Toft J., Moynier F., Lee C. T. A., Telouk P. and Albarede F. (2010) The early formation of the IVA iron meteorite parent body. Earth and Planetary Science Letters 296, 469-480.
Campbell A. J. and Humayun M. (2005) Compositions of group IVB iron meteorites and their parent melt. Geochimica Et Cosmochimica Acta 69, 4733-4744.
Chabot N. L. and Drake M. J. (1999) Crystallization of magmatic iron meteorites: The role of mixing in the molten core. Meteoritics & Planetary Science 34, 235-246.
Chabot N. L. and Haack H., 2006. Evolution of Asteroidal Cores. The University of Arizona Press.
Chen J. H. and Wasserburg G. J. (1990) The Isotopic Composition of Ag in Meteorites and the Presence of Pd-107 in Protoplanets. Geochimica Et Cosmochimica Acta 54, 1729-1743.
Choi B. G., Ouyang X. W. and Wasson J. T. (1995) Classification and Origin of Iab and Iiicd Iron-Meteorites. Geochimica Et Cosmochimica Acta 59, 593-612.
Cook D. L., Walker R. J., Horan M. F., Wasson J. T. and Morgan J. W. (2004) Pt-Re-Os systematics of group IIAB and IIIAB iron meteorites. Geochimica Et Cosmochimica Acta 68, 1413-1431.
Davis A. M., 2006. Volatile evolution and loss. In: Lauretta, D. S. and McSween, H. Y. Eds.), Meteorites and the early solar system II. The University of Arizona Press, Tuscon, Arizona.
Eugster O. (2003) Cosmic-ray exposure ages of meteorites and lunar rocks and their significance. Chemie Der Erde-Geochemistry 63, 3-30.
Goldstein J. I., Scott E. R. D. and Chabot N. L. (2009) Iron meteorites: Crystallization, thermal history, parent bodies, and origin. Chemie Der Erde-Geochemistry 69, 293-325.
Hevey P. J. and Sanders I. S. (2006) A model for planetesimal meltdown by Al-26 and its implications for meteorite parent bodies. Meteoritics & Planetary Science 41, 95-106.
Horan M. F., Smoliar M. I. and Walker R. J. (1998) W-182 and Re-187-Os-187 systematics of iron meteorites: Chronology for melting, differentiation, and crystallization in asteroids. (vol 62, pg 545, 1998). Geochimica et Cosmochimica Acta 62, 1653-1653.
Kleine T., Mezger K., Palme H., Scherer E. and Munker C. (2005) Early core formation in asteroids and late accretion of chondrite parent bodies: Evidence from Hf-182-W-182
26 Introduction to Part A
in CAIs, metal-rich chondrites, and iron meteorites. Geochimica Et Cosmochimica Acta 69, 5805-5818.
Kleine T. and Rudge J. F. (2011) Chronometry of Meteorites and the Formation of the Earth and Moon. Elements 7 41-46.
Kleine T., Touboul M., Bourdon B., Nimmo F., Mezger K., Palme H., Jacobsen S. B., Yin Q. Z. and Halliday A. N. (2009) Hf-W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochimica Et Cosmochimica Acta 73, 5150-5188.
Kollár D., Michel R. and Masarik J. (2006) Monte Carlo simulation of GCR neutron capture production of cosmogenic nuclides in stony meteorites and lunar surface. Meteoritics & Planetary Science 41, 375-389.
Lee D. C. and Halliday A. N. (1995) Hafnium-Tungsten Chronometry and the Timing of Terrestrial Core Formation. Nature 378, 771-774.
Leya I. and Masarik J. (2009) Cosmogenic nuclides in stony meteorites revisited. Meteoritics & Planetary Science 44, 1061-1086.
Leya I. and Masarik J. (2013) Thermal neutron capture effects in radioactive and stable nuclide systems. Meteoritics & Planetary Science, n/a-n/a.
Leya I., Wieler R. and Halliday A. N. (2000) Cosmic-ray production of tungsten isotopes in lunar samples and meteorites and its implications for Hf-W cosmochemistry. Earth and Planetary Science Letters 175, 1-12.
Leya I., Wieler R. and Halliday A. N. (2003) The influence of cosmic-ray production on extinct nuclide systems. Geochimica Et Cosmochimica Acta 67, 529-541.
Lingenfelter R. E., Canfield E. H. and Hampel V. E. (1972) The Lunar Neutron-Flux Revisited. Earth and Planetary Science Letters 16, 355-369.
Markowski A., Leya I., Quitte G., Ammon K., Halliday A. N. and Wieler R. (2006) Correlated helium-3 and tungsten isotopes in iron meteorites: Quantitative cosmogenic corrections and planetesimal formation times. Earth and Planetary Science Letters 250, 104-115.
Masarik J. (1997) Contribution of neutron-capture reactions to observed tungsten isotopic ratios. Earth and Planetary Science Letters 152, 181-185.
Masarik J. and Reedy R. C. (1994) Effects of bulk composition on nuclide production processes in meteorites. Geochimica et Cosmochimica Acta 58, 5307-5317.
McCoy T. J., Walker R. J., Goldstein J. I., Yang J., McDonough W. F., Rumble D., Chabot N. L., Ash R. D., Corrigan C. M., Michael J. R. and Kotula P. G. (2011) Group IVA irons: New constraints on the crystallization and cooling history of an asteroidal core with a complex history. Geochimica Et Cosmochimica Acta 75, 6821-6843.
Qin L. P., Dauphas N., Wadhwa M., Masarik J. and Janney P. E. (2008) Rapid accretion and differentiation of iron meteorite parent bodies inferred from Hf-182-W-182 chronometry and thermal modeling. Earth and Planetary Science Letters 273, 94-104.
Rubie D. C., Nimmo F. and Melosh H. J., 2007. Formation of Earth’s Core. In: Editor-in-Chief: Gerald, S. (Ed.), Treatise on Geophysics. Elsevier, Amsterdam.
Scherstén A., Elliott T., Hawkesworth C., Russell S. and Masarik J. (2006) Hf-W evidence for rapid differentiation of iron meteorite parent bodies. Earth and Planetary Science Letters 241, 530-542.
Schönbächler M., Carlson R. W., Horan M. F., Mock T. D. and Hauri E. H. (2008) Silver isotope variations in chondrites: Volatile depletion and the initial 107Pd abundance of the solar system. Geochimica et Cosmochimica Acta 72, 5330-5341.
Scott E. R. D. (1972) Chemical Fractionation in Iron-Meteorites and Its Interpretation. Geochimica Et Cosmochimica Acta 36, 1205-&.
Scott E. R. D. and Wasson J. T. (1975) Classification and Properties of Iron-Meteorites. Reviews of Geophysics 13, 527-546.
Smoliar M. I., Walker R. J. and Morgan J. W. (1996) Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271, 1099-1102.
Part A Chapter 2 27
Sugiura N. and Hoshino H. (2003) Mn-Cr chronology of five IIIAB iron meteorites. Meteoritics & Planetary Science 38, 117-143.
Theis K. J., Schönbächler M., Benedix G. K., Rehkämper M., Andreasen R. and Davies C. (2013) Palladium–silver chronology of IAB iron meteorites. Earth and Planetary Science Letters 361, 402-411.
Urey H. C. (1955) The cosmic abundances of potassium, uranium, and thorium and the heat balances of the Earth, the Moon, and Mars. Proc. Natl. Acad. Sci. U.S. 41, 127-144.
Walker R. J., McDonough W. F., Honesto J., Chabot N. L., McCoy T. J., Ash R. D. and Bellucci J. J. (2008) Modeling fractional crystallization of group IVB iron meteorites. Geochimica Et Cosmochimica Acta 72, 2198-2216.
Walter M. J. and Trønnes R. G. (2004) Early Earth differentiation. Earth and Planetary Science Letters 225, 253-269.
Wasson J. T. and Kallemeyn G. W. (2002) The IAB iron-meteorite complex: A group, five subgroups, numerous grouplets, closely related, mainly formed by crystal segregation in rapidly cooling melts. Geochimica Et Cosmochimica Acta 66, 2445-2473.
Wieler R. (2002) Cosmic-ray-produced noble gases in meteorites. Noble Gases in Geochemistry and Cosmochemistry 47, 125-170.
28 Hf-W chronology of weakly irradiated iron meteorites
Chapter 3
Hf!W chronology of core formation in planetesimals inferred from weakly irradiated iron meteorites
Kruijer, T.S.a,b, Sprung, P.,a,b Kleine, T.b, Leya, I.c, Burkhardt, C.a, Wieler, R.a
aETH Zürich, Institute of Geochemistry and Petrology, Zürich, Switzerland. bWestfälische Wilhelms-Universität Münster, Institut für Planetologie, Münster, Germany.
cUniversity of Bern, Space Research and Planetary Sciences, Bern, Switzerland.
Published in Geochimica et Cosmochimica Acta 99, 287-304 (2012)
Abstract The application of Hf!W chronometry to determine the timescales of core formation in the parent bodies of magmatic iron meteorites is severely hampered by 182W burnout during cosmic ray exposure of the parent meteoroids. Currently, no direct method exists to correct for the effects of 182W burnout, making the Hf!W ages for iron meteorites uncertain. Here we present noble gas and Hf!W isotope systematics of iron meteorite samples whose W isotopic compositions remained essentially unaffected by cosmic ray interactions. Most selected samples have concentrations of cosmogenic noble gases at or near the lowermost level observed in iron meteorites and, for iron meteorite standards, have very low noble gas and radionuclide based cosmic ray exposure ages (<60 Ma). In contrast to previous studies, no corrections of measured W isotope compositions are required for these iron meteorite samples. Their "182W values (parts per 104 deviations from the terrestrial value) are higher than those measured for most other iron meteorites and range from –3.42 to –3.31, slightly elevated compared to the initial 182W/184W of Ca!Al-rich Inclusions (CAI; "182W = –3.51±0.10). The new W isotopic data indicate that core formation in the parent bodies of the IIAB, IIIAB, and IVA iron meteorites occurred ~1–1.5 Myr after CAI formation (with an uncertainty of ~1 Myr), consistent with earlier conclusions that the accretion and differentiation of iron meteorite parent bodies predated the accretion of most chondrite parent bodies. One ungrouped iron meteorite (Chinga) exhibits small nucleosynthetic W isotope anomalies, but after correction for these anomalies its "182W value agrees with those of the other samples. Another ungrouped iron (Mbosi), however, has elevated "182W relative to the other investigated irons, indicating metal!silicate separation ~2–3 Myr later than in the parent bodies of the three major iron meteorite groups studied here.
Part A Chapter 3 29
3.1 Introduction Magmatic iron meteorites are interpreted as fragments from the metal cores of small planetary bodies (Scott, 1972; Scott and Wasson, 1975) that had formed early in solar system history (e.g., Chen and Wasserburg, 1990; Smoliar et al., 1996; Kleine et al., 2005a). Obtaining a precise chronology of iron meteorites, therefore, is of fundamental importance for constraining the early history of the solar system, and the accretion and differentiation history of some of the earliest planetesimals. The short-lived 182Hf-182W system has proven uniquely useful for determining the timescales of core formation in planetary bodies, because both Hf and W are refractory and exhibit very different geochemical behaviour during metal!silicate separation (e.g., Lee and Halliday, 1995; Harper and Jacobsen, 1996; Kleine et al., 2009). The first comprehensive Hf!W study on iron meteorites was performed by Horan et al. (1998), showing that the different iron meteorite parent bodies underwent core formation within ~5 Myr of each other. Later studies demonstrated that magmatic iron meteorites derive from planetesimals that segregated their cores very early (e.g., Kleine et al., 2005a; Markowski et al., 2006b; Scherstén et al., 2006; Qin et al., 2008b), at about the same time as Ca!Al-rich inclusions (CAI), the oldest yet dated objects that formed in the solar system (Gray et al., 1973; Amelin et al., 2010; Bouvier and Wadhwa, 2010). The chronological interpretation of the Hf!W data for iron meteorites is severely complicated, however, by 182W burnout due to capture of secondary thermal and epithermal neutrons produced during cosmic ray exposure of the iron meteoroids (Masarik, 1997; Leya et al., 2000, 2003). The most obvious manifestation of these neutron capture reactions is the decrease of 182W/184W ratios of iron meteorites with increasing cosmic ray exposure age, leading to spuriously old Hf!W model ages. Neutron capture on W isotopes may be responsible for generating 182W/184W ratios lower than the initial W isotope composition of CAI (Burkhardt et al., 2008, 2012) and may also be the sole cause of 182W variations observed within individual iron meteorite groups or even within a single iron meteorite. The reliable interpretation of W isotopic data for iron meteorites in terms of core formation timescales, therefore, requires the quantification of cosmic ray-induced shifts on W isotope compositions.
The interaction of iron meteoroids with cosmic rays did not only cause neutron capture reactions on W isotopes but also led to the production of cosmogenic noble gases. For this reason, a number of studies focused on cosmogenic noble gases and published cosmic ray exposure ages to correct measured W isotope compositions for cosmic ray-induced shifts (Kleine et al., 2005a; Markowski et al., 2006a; Scherstén et al., 2006; Qin et al., 2008b). For instance, Markowski et al. (2006a) used 3He abundances in conjunction with independently determined exposure ages to correct measured W isotope compositions in Carbo (IID) and Grant (IIIAB). However, the corrected 182W/184W of these two iron meteorites still are lower than the initial W isotope composition of CAI (Burkhardt et al., 2008, 2012). At face value this would indicate core formation of the respective parent bodies to predate CAI formation. However, more likely this strongly suggests that the correction procedure employing 3He did not fully account for cosmic ray-induced W isotope shifts. Qin et al. (2008b) estimated lower and upper bounds of 182W/184W ratios for several groups of magmatic iron meteorites and modelled a maximum expected cosmogenic W isotope effect for a given exposure age. For all iron meteorite groups the upper bounds on 182W/184W were found to be higher than the presently accepted initial 182W/184W of CAI (Burkhardt et al., 2012), solving the problem that some iron meteorites have measured 182W/184W below the CAI initial. However, the difference between the lower and upper bounds of 182W/184W ratios for individual iron meteorite groups remaining after this correction corresponds to apparent 2-5 Myr intervals of metal segregation
30 Hf-W chronology of weakly irradiated iron meteorites
(e.g., for IC, IIAB, IID, IIIF, IVA irons), reflecting the inherent uncertainty of current correction procedures employing noble gas systematics.
The major problem when using cosmogenic noble gases and/or exposure ages to correct cosmic ray-induced shifts in 182W/184W is that cosmogenic noble gas production rates reach their maximum at a shallower depth than that corresponding to the maximum fluence of thermal and epithermal secondary neutrons. Cosmogenic noble gases are thus not a perfect proxy for the fluence of slow neutrons. Obtaining a precise Hf!W chronometry of iron meteorites thus requires the development of a direct neutron dosimeter for iron meteorites, or the identification of specimens that remained largely unaffected by (epi)thermal neutrons. While recent studies have shown that Os (Walker and Touboul, 2012; Wittig et al., 2012) and Pt isotopes (Kruijer et al., 2012) may be suitable neutron dosimeters for iron meteorites, we here will focus on identifying iron meteorite specimens with minor to absent (epi)thermal neutron fluences. Such samples do not require any correction on measured W isotope ratios and thus present key samples for establishing a precise Hf!W chronology of iron meteorites.
Although cosmogenic noble gases do not allow a precise and accurate correction for cosmic ray-induced neutron capture effects on W isotopes, they do provide very useful constraints on the cosmic ray exposure history of meteorites. In some cases noble gas data allow recognising meteorites with a very low exposure age (for iron meteorite standards). Such meteorites will have experienced a very low (epi)thermal neutron fluence irrespective of the sample position within the meteoroid. Therefore, in this contribution we use concentrations of cosmogenic He, Ne, and Ar in samples of magmatic iron meteorites from different groups as major selection criteria to identify specimens with W isotopic compositions largely unaffected by cosmic ray effects. New precise W isotope measurements on these samples that define 182W/184W values devoid of cosmic ray effects are then used to obtain improved constraints on the timing and duration of core formation in iron meteorite parent bodies.
3.2 Theory and Approach Interactions of highly energetic (primary) cosmic ray particles (~0.1-~10 GeV) with target atoms in meteoroids generate a cascade of nuclear reactions. Among other products this yields secondary high-energy protons and neutrons in the energy range of a few to a few hundred MeV. The primary and secondary high-energy particles produce cosmogenic nuclides, mostly by spallation-type reactions. Important for this study is the production of cosmogenic noble gas isotopes, whose concentrations are a measure for the fluence of medium- to high-energy particles in the sample. At the same time, the secondary cosmic ray particles are slowed down, either due to electronic stopping in the case of secondary protons or by elastic scattering in the case of secondary neutrons. The latter are thus moderated to epithermal (~0.025 eV to a few keV) and eventually to thermal (~0.025 eV at 293 K) energies with increasing distance from the meteoroid surface (Lingenfelter et al., 1972; Leya et al., 2000). Tungsten isotopes are predominantly affected by neutron capture reactions at epithermal energies (Masarik, 1997; Leya et al., 2000; 2003), and possess different abilities to capture epithermal neutrons, expressed as distinct resonance integrals (e.g., at T = 300 K: 182W ~600 barns, 183W ~355 barns, 184W ~16 barns, 186W ~520 barns; ENDFB-VI.8 300K library, 0.5 eV to 1#105 eV). The maximum fluence of thermal and epithermal neutrons occurs at larger depth than the maximum production of noble gases. Therefore, cosmogenic noble gases are not a direct measure for neutron capture-induced shifts on W isotopes.
Part A Chapter 3 31
As the majority of iron meteorites have been exposed to cosmic rays for longer than 100 Myr (and up to 2 Gyr) (e.g., Voshage, 1978; 1984; Wieler, 2002; Eugster, 2003), the expected cosmic ray effects on W isotopes will be more pronounced than in stony meteorites (e.g., Leya et al., 2000; 2003). Moreover, the iron-dominated matrix promotes neutron energy spectra that are biased towards epithermal energies (Kollár et al., 2006; Sprung et al., 2010) at which W isotopes are most susceptible to neutron capture. In stony meteorites and lunar rocks W isotope ratios are predominantly affected by the neutron capture reaction 181Ta(n,!)182Ta ("-
)182W, resulting in elevated 182W/184W ratios (e.g., Leya et al., 2000; 2003). This reaction is responsible for generating large 182W excesses in lunar rocks (Lee et al., 2002; Kleine et al., 2005b). Since Ta is not present in Fe-Ni metal, neutron capture reactions that have W isotopes as their target (e.g., 182W(n,!)183W, 183W(n,!)184W, 184W(n,!)185W("!)185Re, 186W(n,!)187W("!)187Re) dominate in iron meteorites (Masarik, 1997; Leya et al., 2000). Neutron capture reactions in iron meteorites induce a negative shift in 182W/184W and a positive shift in 183W/184W. The correction for instrumental mass discrimination (using either 186W/183W or 186W/184W), however, tends to largely cancel out the effect on 183W/184W ratios, while magnifying the effect on 182W/184W. Internal corrections for cosmic ray-induced neutron capture effects (i.e., using W isotopes only) are thus not possible.
Given the difficulty in quantifying cosmic ray-induced neutron capture effects on W isotopes, the pre-exposure 182W/184W ratio of magmatic iron meteorites would ideally be determined on samples largely unaffected by such neutron capture reactions. Cosmogenic noble gases – although not suitable for correcting neutron capture effects on W isotopes – are useful to identify such samples. First, in some cases the combination of a noble gas and a radionuclide allows the determination of the production rate of the noble gas nuclide at the sample position and hence a reliable shielding-corrected exposure age of the meteorite. Two of the meteorites in this study (Braunau, IIAB, and Gibeon, IVA) were selected according to their exposure ages being below 60 Ma as determined by previous workers with this approach (e.g., Cobb et al. 1966; Chang and Wänke, 1969; Honda et al., 2009). As demonstrated in Section 3.5, such samples will have very small if any neutron capture effects on their W isotopic composition. A third meteorite studied here (Cape York, IIIAB) has a slightly higher exposure age of about 90 Ma (Mathew and Marti, 2009), but we will demonstrate that the W isotope compositions of the Cape York samples investigated here also remained essentially unaffected by cosmic ray effects. Overall, iron meteorites with shielding-corrected very low exposure ages are rare, however.
Second, noble gases in iron meteorites provide information about meteoroid size and sampling depth, as ratios like 4He/21Ne or 3He/4He depend on these parameters (e.g., Signer and Nier, 1960). In the present study, such parameters must be used with caution, however, because the very low noble gas concentrations (sometimes close to blank levels) often lead to large analytical uncertainties. Nevertheless, for most of the specimens studied here, the He, Ne, and Ar concentration and isotopic composition analyses permit semi-quantitative estimates on the pre-atmospheric size, sample location in the meteoroid and maximum possible cosmic ray exposure age.
3.3 Analytical methods
3.3.1 Sample selection Magmatic iron meteorites whose reported cosmogenic noble gas concentrations are consistently at the lower end of the range observed in iron meteorites (Schultz and Franke,
32 Hf-W chronology of weakly irradiated iron meteorites
2004) were selected for this work. The samples investigated here include iron meteorite specimens from four different magmatic groups (IIAB, IIIAB, IVA, IIG) and two ungrouped iron meteorites (Table 3.1 and 3.3). Noble gas concentrations were re-measured for all of the meteorite specimens investigated in this study in material that had been in immediate contact to that used for W isotope analysis.
3.3.2 Noble gas measurements Extraction and measurement of He, Ne and Ar concentrations and isotope compositions were performed at the University of Bern. Analytical procedures are outlined in Ammon et al. (2008, 2011). Iron meteorite samples (50-160 mg) were cut using a diamond saw, cleaned ultrasonically in ethanol (5 min.), and wrapped in commercial Ni-foil. Extraction of He, Ne and Ar was performed by heating the samples at 1700-1800 °C in a Mo-crucible for 35 minutes. A boron-nitride liner placed inside the Mo crucible helped to avoid corrosion of the Mo crucible. Extracted gases were cleaned using various Ti-getters and activated charcoal. The Ar fraction was separated from He and Ne using activated charcoal (at !196 °C). Helium and Ne were measured in a 90° sector field mass spectrometer and Ar in a tandem mass spectrometer with two 90° magnetic sector fields. Both non-commercial mass spectrometers have Nier-type ion sources and are equipped with a Faraday collector and an electron multiplier working in analogue mode. Since the noble gas amounts of the studied samples are low, only electron multiplier measurements were used. Noble gas concentrations were determined by peak height comparison to calibrated standard gases that are isotopically similar to atmospheric composition (only 3He is elevated relative to atmospheric composition). Calibration gases were measured before or after each sample analysis. All 20Ne signals were corrected for interferences from H2
18O. The sensitivity of the mass spectrometers used in this study varies non-linearly as a function of total gas pressure. The measured He and Ne amounts were corrected for this effect using relations between measured He and Ne isotope concentrations and gas amounts that have been determined in dilution series experiments.
Blanks were measured with the same extraction and purification protocol as that used for the samples. At least one blank was measured after each sample. Subtracted blank signals are the average of multiple blanks (n=25). For the He isotopes blanks were subtracted directly from raw signals. Measured Ne and Ar isotope signals were first corrected for trapped atmospheric noble gases using a two-component de-convolution based on assumed atmospheric and cosmogenic noble gas isotope ratios (20Ne/22NeCOS = 0.83; 21Ne/22NeCOS = 1; 36Ar/38ArCOS=0.65, 36Ar/38ArATM=5.32) (Wieler, 2002). Cosmogenic 21Ne and 38Ar amounts were subsequently corrected for average cosmogenic contributions in the blanks, stemming from previous samples.
Reported external uncertainties (1SD) of cosmogenic noble gas concentrations represent the combined uncertainties of calibration, blank and sample measurements, together with the uncertainties involved in correcting for non-linear effects. For most of the samples these uncertainties are ~5-10% for 4He, 5-10% for 3He, 4-25% for 21Nec, and 6-10% for 38Arc. However, some noble gas concentrations that barely exceed background levels, and thus required large relative blank subtractions, have much larger uncertainties approaching ~100% (Table 3.1). Therefore, 4He/21Ne ratios are only reported in Table 3.1 if their uncertainties do not exceed 30% (1$, s.d.).
Part A Chapter 3 33
Tab
le1
Cosm
ogenic
noble
gasconcentrationsofironmeteoritessamplesinvestigated
inthisstudy.
Meteorite
nam
eID
3He[±
1SD]
4He[±
1SD]
21Ne c
os[±
1SD]
38Ar cos[±
1SD]
4He/
21Ne c
os[±
1SD]
3He/
4He[±
1SD]
[10!
8cm
3STP/g]
[10!
8cm
3STP/g]
[10!
8cm
3STP/g]
[10!
8cm
3STP/g]
[10!
8cm
3STP/g]
[10!
8cm
3STP/g]
IIABirons
Edmonton(C
anad
a)A02
0.290±
0.029
3.25
±0.39
0.011±
0.006
0.057±
0.006
na
0.09
±0.01
Braunau
M03
2.41
±0.15
24.8
±1.5
0.17
±0.03
0.645±
0.048
150±
290.10
±0.01
Sikhote
Alin
‘SA01’
B01
133±
9561±
382.61
±0.20
14.5
±0.95
215±
220.24
±0.02
Sikhote
Alin
‘SA02’
G03a
12.7
±0.68
58.3
±3.1
0.22
±0.03
1.22
±0.079
265±
400.22
±0.02
Sikhote
Alin
‘SA02’
G03b
13.7
±0.76
63.6
±3.5
0.25
±0.04
1.12
±0.074
255±
450.22
±0.02
IIIA
Birons
Cap
eYork
‘CY01’
A01
0.574±
0.043
3.47
±0.31
0.006±
0.004
0.042±
0.004
na
0.17
±0.02
Cap
eYork
‘CY02’
G02a
0.229±
0.019
2.02
±0.17
0.005±
0.001
0.017±
0.011
410±
104
0.11
±0.01
Cap
eYork
‘CY02’
G02b
0.220±
0.021
1.57
±0.15
0.005±
0.001
0.025±
0.003
320±
910.14
±0.02
IVA
irons
Gibeon‘102’
N02
<Detection
<Detection
<Detection
0.0003
±0.0003
na
na
Muonionalusta
A04
<Detection
6.03
±0.50
<Detection
0.006±
0.001
na
na
Gibeon‘R
ailway’
A03
0.06
±0.01
0.58
±0.19
0.001±
0.001
0.002±
0.002
na
na
Gibeon‘Egg’
G01a
5.20
±0.29
24.2
±1.4
0.09
±0.02
0.454±
0.030
270±
490.21
±0.02
Gibeon‘Egg’
G01b
5.56
±0.37
26.0
±1.7
0.10
±0.021
0.434±
0.031
250±
550.21
±0.02
Gibeon‘99’
C06
1.61
±0.10
18.3
±1.2
0.06
±0.01
0.338±
0.028
300±
730.088±
0.008
IIG
irons
Tombigbee
River
K04
3.34
±0.36
21.4
±1.3
0.08
±0.02
0.421±
0.031
270±
540.16
±0.02
Twan
nberg
C01
4.90
±0.28
23.4
±1.4
0.09
±0.02
0.483±
0.035
270±
510.21
±0.02
Ungrouped
irons
Chinga
M01
5.23
±0.35
23.8
±1.6
0.08
±0.02
0.400±
0.032
310±
850.22
±0.02
Mbosi
M02
0.032±
0.007
0.2±
0.1
<Detection
0.005±
0.003
na
na
Tishomingo
C02
43.6
±2.5
198±
110.71
±0.10
na
280±
420.22
±0.02
290 T.S. Kruijer et al. /Geochimica et Cosmochimica Acta 99 (2012) 287–304
34 Hf-W chronology of weakly irradiated iron meteorites
3.3.3 Tungsten isotope analyses
3.3.3.1 Sample preparation and chemical separation of W Samples of ~0.5-1 g from all selected specimens were cut using a handsaw, thoroughly cleaned with abrasive paper followed by ethanol and de-ionised water in an ultrasonic bath. The outermost 15-25% of each sample were removed by leaching in 6 M HCl-0.06 M HF and a few drops of concentrated HNO3 on a hotplate (90-110 °C) for 20-30 minutes. Complete dissolution of the samples was accomplished in ~15 ml 6 M HCl–0.06 M HF in Savillex® vials at 130°C on a hotplate overnight. After complete dissolution, ~5% solution aliquots were taken to determine Hf and W concentrations by isotope dilution using a mixed 180Hf-183W isotope tracer (Kleine et al., 2004). As expected, all samples are virtually free of Hf and have 180Hf/184W lower than ~8#10-4. Tungsten was separated from the sample matrix using a modified two-stage anion exchange chromatography after Horan et al. (1998) and Kleine et al. (2002; 2005a). The samples were loaded onto pre-cleaned anion exchange columns (4 ml BioRad® AG1X8, 200-400 mesh size) in 75 ml 0.5 M HCl–0.5 M HF. Most of the Fe-Ni matrix was washed off using 0.5 M HCl–0.5 M HF, and W was eluted in 15 ml 6 M HCl–1 M HF. After the first chemical separation, samples were dried down in HNO3–H2O2 several times, and were then re-dissolved in 5 ml 0.5 M HCl–0.5 M HF. The second anion exchange chromatography is similar to the first-stage separation, but uses one instead of four ml anion exchange resin and includes an additional elution step (8 M HCl–0.01 M HF) that rinses off other trace elements (e.g., Ag). This step was found to be necessary when analysing the terrestrial metal standard (high-sulphur steel NIST 129c), which contains significant amounts of Ag. Silver, together with Ar and Cl, may form molecular interferences in the W mass range, as verified by isotope measurements of the W standard with admixed Ag. Magmatic iron meteorites contain too little Ag to significantly affect the W isotope analyses, however. Nevertheless, for direct comparison to the terrestrial standard, all samples were processed through the exact same chemical purification procedure. After the second chromatographic step, the W cuts were dried again, treated with HNO3–H2O2 several times, and finally dissolved in a running solution (0.56 M HNO3–0.24 M HF) yielding analyte W concentrations of 100 ppb (Nu Plasma) or 40-100 ppb (Neptune). The combined W yield for the two-stage chemistry was typically ~60-80%. Total procedural blanks varied between ~30 and 130 pg W and no blank corrections were necessary, given that ~150-400 ng W were analysed for each sample.
Part A Chapter 3 35
Tab
le2
Tungstenisotopecompositionsfortheterrestrialstan
dardsan
alysed
inthisstudy.
MC-ICPMS
Ne1
82/183W
e182/184W
e182/184W
e183W
e184W
e182/183W
e182/184W
e182/184W
(6/3) m
eas.a
(6/4) m
eas.a
(6/3) m
eas.a
(6/4) m
eas.a
(6/3) m
eas.a
(6/3) corr.a
(6/4) corr.a
(6/3) corr.a
±2r
±2r
±2r
±2r
±2r
NIST129c
b
A07
cNuPlasm
a5
0.11
±0.15
!0.05
±0.15
0.06
±0.15
!0.21
±0.10
!0.14
±0.07
!0.16
±0.21
!0.05
±0.15
!0.07
±0.17
B07
cNuPlasm
a5
0.11
±0.10
!0.11
±0.06
0.02
±0.04
!0.14
±0.05
!0.09
±0.03
!0.08
±0.12
!0.11
±0.06
!0.07
±0.05
C07
cNuPlasm
a5
0.15
±0.23
0.05
±0.09
0.09
±0.22
!0.07
±0.12
!0.05
±0.08
0.06
±0.28
0.05
±0.09
0.04
±0.23
D07
cNuPlasm
a5
0.16
±0.08
0.02
±0.08
0.09
±0.07
!0.10
±0.07
!0.07
±0.05
0.02
±0.13
0.02
±0.08
0.02
±0.09
E07
cNuPlasm
a4
0.27
±0.13
0.07
±0.07
0.17
±0.07
!0.16
±0.08
!0.10
±0.05
0.07
±0.16
0.07
±0.07
0.07
±0.09
F07
cNuPlasm
a4
0.16
±0.14
!0.05
±0.05
0.07
±0.09
!0.15
±0.09
!0.10
±0.06
!0.04
±0.19
!0.05
±0.05
!0.03
±0.11
MeanNuPlasm
a(±
2SD,n=
6)!0.01
±0.14
!0.02
±0.18
!0.01
±0.14
!0.01
±0.12
NIST129c
G07
cNeptunePlus
40.17
±0.07
0.04
±0.16
0.10
±0.10
!0.08
±0.13
!0.05
±0.09
0.06
±0.19
0.04
±0.16
0.05
±0.13
R09
NeptunePlus
2!0.02
±0.12
!0.03
±0.09
!0.01
±0.11
0.03
±0.03
0.02
±0.02
!0.02
±0.12
!0.03
±0.09
!0.01
±0.11
M09
NeptunePlus
50.06
±0.08
0.04
±0.07
0.05
±0.06
!0.01
±0.06
0.00
±0.04
0.06
±0.08
0.04
±0.07
0.05
±0.06
L05
cNeptunePlus
30.25
±0.09
!0.01
±0.11
0.10
±0.12
!0.18
±0.04
!0.12
±0.03
0.02
±0.11
!0.01
±0.11
!0.01
±0.12
L06
cNeptunePlus
40.22
±0.06
!0.04
±0.06
0.10
±0.09
!0.19
±0.03
!0.12
±0.02
!0.03
±0.07
!0.04
±0.06
!0.02
±0.09
O02
NeptunePlus
30.05
±0.10
0.08
±0.14
0.06
±0.12
0.01
±0.09
0.00
±0.06
0.05
±0.10
0.08
±0.14
0.06
±0.12
O03
cNeptunePlus
20.21
±0.06
0.04
±0.20
0.13
±0.15
!0.10
±0.21
!0.07
±0.14
0.07
±0.29
0.04
±0.20
0.06
±0.20
P09
cNeptunePlus
30.18
±0.12
0.05
±0.12
0.13
±0.12
!0.08
±0.04
!0.05
±0.03
0.08
±0.13
0.05
±0.12
0.08
±0.12
S04
NeptunePlus
50.02
±0.07
0.01
±0.05
0.01
±0.05
!0.01
±0.04
!0.01
±0.03
0.02
±0.07
0.01
±0.05
0.01
±0.05
MeanNeptunePlus(±
2SD,n=
9)0.02
±0.08
0.03
±0.08
0.02
±0.08
0.03
±0.08
MeanNIST129c
All(±
2SD,n=
15)
0.01
±0.10
0.01
±0.13
0.01
±0.10
0.01
±0.10
AlfaAesar
solutionstan
dardd
NeptunePlus
50.22
±0.06
!0.04
±0.06
0.10
±0.09
!0.19
±0.03
!0.12
±0.02
!0.03
±0.07
!0.04
±0.06
!0.02
±0.09
NIST3163
solutionstan
dard
NeptunePlus
50.01
±0.03
!0.02
±0.03
0.00
±0.03
!0.01
±0.03
!0.01
±0.02
0.01
±0.03
!0.02
±0.03
0.00
±0.03
aNorm
alized
to186W/184W
=0.92767(6/4)or186W/183W
=1.9859
(6/3)usingtheexponential
law.
bHighsulphursteel(N
IST129c)doped
withad
ditional
Wto
match
concentrationsofsamplesan
dsubsequentlyprocessed
through
fullchem
ical
separation.
cA!ectedbyan
dcorrectedformass-independente!ect(Section4.2).
dAlfaAesar
solutionstan
dardprocessed
through
fullchem
ical
separation.
T.S. Kruijer et al. /Geochimica et Cosmochimica Acta 99 (2012) 287–304 291
36 Hf-W chronology of weakly irradiated iron meteorites
3.3.3.2 W isotope measurements Tungsten isotope compositions were measured on a Nu Plasma MC-ICPMS at ETH Zürich following previously published procedures (Kleine et al., 2008) and on a ThermoScientific Neptune Plus MC-ICPMS at the Westfälische Wilhelms-Universität Münster. On both instruments, samples were introduced into the mass spectrometer using a Cetac Aridus II® desolvator. Standard Ni cones were used for all W isotope measurements (type B sampler + type A WA skimmer on the Nu Plasma; H cones on the Neptune Plus). Total ion beam intensities for W varied between 1 and 2#10-10 A at a 75 %L/min uptake rate (Nu Plasma, 100 ppb W) and 1.5 to 5#10-10 A at a 50-100 %L/min uptake rate (Neptune Plus, 40-100 ppb W). Small isobaric interferences of Os on masses 184 and 186 were corrected by monitoring interference-free 188Os. The Os interference corrections were generally smaller than 20 ppm and insignificant. Only Chinga (UNG) required a larger interference correction of ~3 "-units. Instrumental mass bias was corrected normalising to either 186W/183W = 1.9859 or 186W/184W = 0.92767 and using the exponential law.
The 182W/184W and 183W/184W of the samples were measured relative to a terrestrial solution standard prepared from pure W metal (Alfa Aesar; Kleine et al., 2002, 2004). Tungsten isotope composition measurements consisted of 40 cycles of 5s (Nu Plasma), or 200 cycles of 4.2s integration time (Neptune Plus). Each measurement was bracketed by analyses of the Alfa Aesar solution standard. Baselines were measured by deflecting the ion beam using the electrostatic analyser (ESA) for 60-120 s. For analyses with relatively low ion beam intensities, baselines were determined on-peak using an acid blank solution (0.56 M HNO3–0.24 M HF). Both methods proved to yield identical results. A 3-5 minute washout using 0.56 M HNO3–0.24 M HF between two successive measurements was sufficient to clean the sample introduction system.
Measured 182W/184W and 183W/184W ratios of samples are reported as "-unit (i.e., parts per 104) deviations relative to the bracketing standard analyses. Reported are mean "182W and "183W values of pooled solution replicates (n=2-9) at an external reproducibility of <0.2 "-units (95% confidence limits of the mean) in most cases. The accuracy of the measurements was investigated through analysis of a terrestrial standard (high sulphur steel NIST129c) that was processed through the same purification procedure as the iron meteorite samples. Also, alternative isotope ratios (182W/183W and 182W/184W) and normalisation ratios for instrumental mass bias correction (186W/183W, ‘6/3’ and 186W/184W, ‘6/4’) were monitored and are used to assess possible matrix effects or artefacts.
Part A Chapter 3 37
Tab
le3
Wisotopecompositionsofironmeteortes
investigated
inthisstudy.
Meteorite
IDSource
Nb
e182/183W
(6/3) m
eas.a
e182/184W
(6/4) m
eas.a
e182/184W
(6/3) m
eas.a
e183/184W
(6/4) m
eas.a
e183/184W
(6/3) m
eas.a
MC-ICPM
Sd
e182/183W
(6/3) corr.a
e182/184W
(6/4) corr.a
e182/184W
(6/3) corr.a
±2r
±2r
±2r
±2r
±2r
±2r
±2r
±2r
Low
exposure
age(<
60Myr)an
d/or
e!ective
shielding
IIAB
irons
Edmonton,Can
adac
A02
Univ.Alberta
10!3.29
±0.08
!3.38
±0.07
!3.34
±0.07
!0.08
±0.05
!0.05
±0.03
Nept.
!3.39
±0.10
!3.38
±0.07
!3.39
±0.08
Braunau
cM03
Vienna
9!3.22
±0.07
!3.40
±0.05
!3.31
±0.04
!0.13
±0.05
!0.09
±0.03
Nu
!3.40
±0.09
!3.40
±0.05
!3.40
±0.05
MeanIIAB
irons
!3.39
±0.08
!3.39
±0.08
!3.39
±0.08
!3.40
±0.08
IIIA
Birons
Cap
eYork
‘CY01’c
A01
UC
San
Diego
8!3.22
±0.11
!3.36
±0.12
!3.29
±0.11
!0.14
±0.04
!0.09
±0.03
Nu
!3.40
±0.12
!3.36
±0.12
!3.38
±0.11
Cap
eYork
‘CY02’c
G02
JNMC
Zurich
3!3.24
±0.05
!3.39
±0.17
!3.34
±0.06
!0.09
±0.14
!0.06
±0.09
Nept.
!3.36
±0.19
!3.39
±0.17
!3.40
±0.11
MeanIIIA
Birons
!3.37
±0.10
!3.38
±0.18
!3.37
±0.14
!3.39
±0.12
IVA
irons
Gibeon‘102’
C04/
N02
Tokyo
4!3.22
±0.15
!3.31
±0.08
!3.22
±0.10
!0.02
±0.10
!0.01
±0.07
Nept.
!3.22
±0.15
!3.31
±0.08
!3.22
±0.10
Gibeon‘R
ailway’c
A03
/S01
Senckenberg
5!3.19
±0.07
!3.42
±0.08
!3.30
±0.09
!0.17
±0.06
!0.11
±0.04
Nept.
!3.41
±0.11
!3.42
±0.08
!3.41
±0.10
Muonionalustac
A04
/S03
ETH
IR!16
5!3.23
±0.03
!3.33
±0.07
!3.29
±0.07
!0.08
±0.05
!0.05
±0.03
Nept.
!3.33
±0.07
!3.33
±0.07
!3.34
±0.07
MeanIV
Airons(excl.
Gibeon‘R
ailway’)
!3.32
±0.08
!3.28
±0.08
!3.32
±0.08
!3.28
±0.08
Ungrouped
irons
Chinga
c,e
M01
ETH
IR-54
4!3.21
±0.10
!2.86
±0.13
!3.04
±0.10
0.26
±0.09
0.17
±0.08
Nept.
!3.30
±0.10
!3.30
±0.17
!3.29
±0.14
Mbosi
M02
Senckenberg
4!3.05
±0.10
!3.09
±0.03
!3.06
±0.07
!0.04
±0.08
!0.03
±0.05
Nept.
!3.05
±0.10
!3.09
±0.03
!3.06
±0.07
Higherexposure
age(>
100Myr)an
d/or
lower
shielding
IIAB
irons
Sikhote
Alin
‘SA01’
B01
ETH
IR-27
7!3.69
±0.16
!3.58
±0.24
!3.61
±0.16
0.08
±0.28
0.06
±0.18
Nu
!3.69
±0.16
!3.58
±0.24
!3.61
±0.16
Sikhote
Alin
‘SA02’c
G03
JNMC
Zurich
2!3.71
±0.14
!3.80
±0.12
!3.76
±0.11
!0.05
±0.13
!0.03
±0.08
Nept.
!3.77
±0.21
!3.80
±0.12
!3.79
±0.14
IIG
irons
Tombigbee
River
cK04
AMNH
2!3.38
±0.08
!3.60
±0.08
!3.50
±0.09
!0.16
±0.04
!0.11
±0.02
Nept.
!3.60
±0.09
!3.60
±0.08
!3.61
±0.09
IVA
irons
Gibeon‘Egg’c
G01
/S02
JNMC
Zurich
5!3.38
±0.13
!3.44
±0.09
!3.40
±0.06
!0.05
±0.10
!0.04
±0.07
Nept.
!3.45
±0.19
!3.44
±0.09
!3.44
±0.09
Gibeon‘99’
cC06
Tokyo
6!3.50
±0.22
!3.63
±0.20
!3.59
±0.22
!0.04
±0.16
!0.03
±0.11
Nu
!3.56
±0.31
!3.63
±0.20
!3.62
±0.25
AllW
isotopecompositionsarereported
ase-unitdeviationsrelative
tothebracketingsolutionstan
dard(A
lfaAesar):((18i W
/18j W
) sample/(18i W
/18j W
) standard!
1)x104.
Uncertainties
represent95%
confidence
limitsofthemeanan
darecalculatedusing(SD
*t 0.95,N
!1)/pN
ifN
>4.
ForsampleswhereN
<4theuncertainty
was
estimated
usingtheaverage2S
Dofmultiple
analyses
oftheterrestrialmetal
stan
dard(N
IST129c)duringthesamesession.
aCorrectedforinstrumentalmassbiasthrough
internal
norm
alisationto
186W/183W
(‘6/3’)or186W/184W
(‘6/4’)usingtheexponential
law.
bN
=Numberofsolutionreplicates.
cA!ectedbyan
dcorrected(‘corr’)formass-independente!ect(Section5.2).
dMC-ICPMSused;‘N
u’:NuInstruments
Plasm
aMC-ICPMS(ETH
Zurich);‘N
ept.’:ThermoScientificNeptunePlusMC-ICPMS(U
niversity
ofMunster).
eCorrectedfors-process
deficits
usingArlan
diniet
al.(1999)
andtherelationsdescribed
inBurkhardtet
al.(2012).
390 Corrigendum /Geochimica et Cosmochimica Acta 112 (2013) 389–390
38 Hf-W chronology of weakly irradiated iron meteorites
3.4 Results
3.4.1 Cosmogenic noble gases The cosmogenic noble gas concentrations for the samples from this study are presented in Table 3.1. As anticipated, cosmogenic noble gas concentrations (index ‘c’) in most samples were found to be near the low end of the range observed in iron meteorites (Schultz and Franke, 2004). About half of the samples analysed here have 21Nec concentrations below < 0.01x10-8 cm3STP/g, while another group of samples shows slightly higher 21Nec values between 0.01-0.7 x10-8 cm3STP/g. In contrast, one of the Sikhote Alin samples (SA01) contains ~2.6 x10-8 cm3STP/g of 21Nec. Since cosmogenic 22Ne cannot be determined directly we cannot discuss cosmogenic 22Ne/21Ne ratios and, therefore, cannot correct the 21Nec concentrations for contributions from possible phosphorous and/or sulphur inclusions. This might in principle somewhat compromise the quality of the 21Nec data and the discussion of pre-atmospheric sizes and exposure ages. However, given the consistency of the dataset (see below), we consider any contributions to 21Nec from sulphur and/or phosphorus inclusions to be small for the meteorites studied here.
3.4.2 Tungsten isotopes
3.4.2.1 Terrestrial standard The W isotopic data for the terrestrial standard (NIST129c) are provided in Table 3.2 and Fig. 3.1 and 3.2. The "18iW values are given for different W isotope ratios (182W/183W, 182W/184W, 183W/184W) and for different normalising ratios (186W/183W or 186W/184W, subsequently marked as ‘(6/3)’ and ‘(6/4)’, respectively). All eight measurements for NIST 129c – each of which represents the average of 2-5 measurements of a single sample that was processed through the full chemical separation – show small but resolved positive "182/183W (6/3) (+0.02 to +0.27) and negative "183/184W (6/3) (+0.02 to !0.14) anomalies. These effects are observed for measurements performed both on the Neptune Plus (WWU Münster) and on the Nu Plasma MC-ICPMS (ETH Zürich). The average of all eight measurements of the NIST129c standard yields mean "182/183W (6/3) = +0.14±0.05, "182/184W (6/3) = +0.08±0.03, and "183/184W (6/4) = !0.10±0.04 (SE 95% conf., n=15). Thus, while both 182W/183W (6/3) and 182W/184W (6/3) values are elevated, the 183W/184W ratio is lower than that of the solution standard. The anomaly in "182/183W (6/3) is twice that in "183/184W (6/3), while the anomalies in "182/184W (6/3) and "183/184W (6/3) are identical. At the same time no anomaly is apparent in "182/184W (6/4) with an average value identical to zero ("182/184W (6/4) = 0.01±0.10, 2SD, n=15). Thus, only W isotope ratios involving 183W yield inaccurate results, while the 182W/184W ratio (6/4) is not affected. This indicates that the slight offset observed for the NIST129c data is caused by a mass-independent W isotope fractionation of the odd isotope 183W from the even isotopes 182W, 184W and 186W. The same negatively correlated shifts in "182/184W (6/3) and "183/184W (6/3) relative to the unprocessed solution standard have been observed by Willbold et al. (2011) for W isotope measurements on terrestrial silicate samples.
Part A Chapter 3 39
Fig. 3.1: Measured W isotope compositions for the terrestrial metal standard (NIST129c) relative to bracketing analyses of the solution standard (Alfa Aesar). (a) !182/184W (6/4) vs. !183/184W (6/4), (b) !182/183W (6/3) vs. !183/184W (6/3), and (c) !182/184W (6/3) vs. !183/184W (6/3), where ‘6/3’ indicates that measured W isotope ratios were corrected for instrumental mass bias by internal normalization using 186W/183W, and ‘6/4’ by normalization to 186W/184W. Also shown is the expected correlation line for a deficit in 183W. Each data point represents multiple solution replicates (n=2-5) of a digestion that was processed through the full chemical separation. Distinguished are samples measured with the Nu Plasma and ThermoScientific Neptune Plus MC-ICPMS. Also shown are the W isotope data for one aliquot of the Alfa Aesar solution standard that was processed through the full chemical separation (purple diamond), and for the NIST 3163 solution standard, which was not processed through the chemical separation (blue circle). Uncertainties are 95% conf. limits of the mean. Although the uncertainties of !182W and !183W (6/i) are correlated, error ellipses were omitted in this diagram for clarity. –0.4 –0.2 0.0 0.2 0.4
–0.4
–0.2
0.0
0.2
0.4
–0.4 –0.2 0.0 0.2 0.4
–0.4
–0.2
0.0
0.2
0.4
183/184W (6/4)
182/
184 W
(6/4
) 183W deficit
–0.4 –0.2 0.0 0.2 0.4–0.4
–0.2
0.0
0.2
0.4
183/184W (6/3)
182/
183 W
(6/3
)
(a)
(b)
(c)
182/
184 W
(6/3
)
183/184W (6/3)
183W deficit
183W deficit
Nu Plasma
Alfa Aesar (chem.)
NIST3163 (sol.)
NIST129c (chem.):Neptune Plus
40 Hf-W chronology of weakly irradiated iron meteorites
Fig. 3.2: External reproducibility of !182W for the terrestrial metal standard (NIST129c). (a) !182/183W (6/3), (b) !182/184W (6/3), and (c) !182/184W (6/4). Each data point represents multiple solution replicates (n=2-5) of a digestion that was processed through the full chemical separation. For normalizations involving 183W (i.e., those that are potentially affected by a mass-independent effect), both the corrected and uncorrected !182W values are shown. Error bars on individual data points are 95% conf. limits of the mean. The hashed area shows the external reproducibility of the NIST129c standard analyses (2SD).
–0.4
–0.2
0
0.2
0.4
–0.4
–0.2
0.0
0.2
0.4
–0.4
–0.2
0
0.2
0.4
–0.4
–0.2
0.0
0.2
0.4
–0.4
–0.2
0
0.2
0.4
–0.4
–0.2
0.0
0.2
0.4
182/
183 W
(6/3
)18
2/18
4 W (6
/3)
182/
184 W
(6/4
)
NIST129c:182/183W (6/3)corr. = +0.01±0.13
(2SD, n=15)
NIST129c:182/184W (6/3)corr. = +0.01±0.10
(2SD, n=15)
NIST129c:182/184W (6/4)meas. = +0.01±0.10
(2SD, n=15)
(a)
(b)
(c)
measured
correctedNu Plasma
Neptune Plus
Part A Chapter 3 41
The observed 183W isotope effect in the terrestrial metal standard may reflect an anomalous isotope composition of the bracketing solution standard (i.e., Alfa Aesar) relative to the true terrestrial composition. However, an aliquot of the Alfa Aesar solution standard passed through the full chemical separation produced identical anomalies to those observed for the metal standard (NIST129c) (Fig. 3.1; Table 3.2). Furthermore, another solution standard (NIST3163) yields a W isotope composition identical to that of the Alfa Aesar standard (Fig. 3.1). An anomalous W isotope composition of the Alfa Aesar solution standard, therefore, cannot be the cause of the mass-independent 183W effect observed for the NIST 129c metal standard. This effect must rather be related to processes during sample preparation and W purification and we concur with Willbold et al. (2011) that a mass-independent isotope fractionation between odd (i.e., 183W) and even W isotopes (i.e., 182W, 184W and 186W) associated with W-loss during re-dissolution of purified W in Savillex vials is causing the observed 183W deficits. The mass-independent 183W effect on terrestrial standards and meteorite samples can be corrected by using different normalisation schemes for the W isotope measurements. For example, the terrestrial metal standards analysed in this study plot on a line with a slope of ~ !2 that is predicted for 183W deficits in "182/183W (6/3) versus "183/184W (6/3) space (Fig. 3.1b). Likewise, the data plot on a line of slope ~ –1 in "182/184W (6/3) versus "183/184W (6/3) space, again consistent with the predicted effect of a 183W deficit (Fig. 3.1c). For all samples, measured "182/183W (6/3) and "182/184W (6/3) values can thus be corrected using the measured " 183/184W (6/3) and the relations "182/183W (6/3)corr. = "182/183W (6/3)meas. – (!2) # "183/184W (6/3) and "182/184W (6/3)corr. = "182/184W (6/3)meas. – (!1) # "183/184W (6/3) (Table 3.2). The corrected "182/183W (6/3) and "182/184W (6/3) values are identical to the measured "182/184W (6/4), indicating that the corrections are accurate (Fig. 3.2).
3.4.2.2 Iron meteorites Tungsten isotope compositions for the iron meteorite samples are displayed in Table 3.3 and Fig. 3.3 and 3.4. The co-variation of "182/184W (6/3) and "183/184W (6/3) measured for the iron meteorites (Fig. 3.3) is similar to the correlated effects observed for the terrestrial metal standard (NIST129c) (Fig. 3.1). While many of the samples analysed here have "183/184W indistinguishable from the terrestrial value (i.e., the Alfa Aesar solution standard), about half of the samples exhibit small "183/184W deviations that are resolved from the terrestrial value. These samples also show differences of similar magnitude in their "182W values calculated using different normalisation schemes (Fig. 3.3).
For all iron meteorite samples investigated here (except Chinga) the difference between "182/184W (6/3) and "182/184W (6/4) is identical to the negative anomaly observed for "183/184W (6/3) (Table 3.3). Similarly, the difference between "182/183W (6/3) and " 182/184W (6/4) corresponds to twice the anomaly observed for "183/184W (6/3). These systematic W isotope shifts are the same as those observed for the NIST129c standard, and are also in agreement with the predicted effects for a deficit in 183W (Fig. 3.3). This strongly suggests that the observed mass-independent 183W effects in the NIST129c data and the iron meteorite samples have the same origin. The iron meteorite data can thus be corrected using the same approach employed above for the NIST129c standard. The corrected "182/184W (6/3) and "182/183W (6/3) values are shown in Table 3.3, and are in excellent agreement with the "182/184W (6/4) values directly measured for the iron meteorites. This again demonstrates that, as for the NIST 129c data, the correction for the mass-independent 183W effect is accurate. We nevertheless emphasise that the main conclusions of our study are not compromised by the mass-independent 183W effect present in some of the sample measurements, because the chronological interpretation of the Hf!W data is entirely based on "182/184W (6/4) values (i.e.,
42 Hf-W chronology of weakly irradiated iron meteorites
the W isotope ratio not involving 183W), which do not show this effect and, hence, do not require any correction at all.
The iron meteorite samples from groups IIAB, IIIAB and IVA with the lowest noble gas concentrations display a narrow range of "182/184W (6/4) values, from !3.42 to !3.31 (Table 3.3), slightly higher than the current best estimate for the initial W isotope composition of CAI ["182W = !3.51±0.10; Burkhardt et al. (2012)]. In general, samples with higher noble gas concentrations tend to have lower "182W values (Table 3.1 and 3.3). Among these samples, Sikhote Alin (IIAB, sample SA02) stands out by having "182W significantly below the initial W isotope composition of CAI. The two ungrouped iron meteorites analysed here (Chinga and Mbosi) have "182W significantly higher (up to ~0.4 "-units) than those observed for samples from the major groups of investigated magmatic iron meteorites (Fig. 3.3 and 3.4; Table 3.3). One of the ungrouped iron meteorites (Chinga) stands out by also having a positive "183W anomaly resolved from zero.
Part A Chapter 3 43
Fig. 3.3: !182W-!183W variation for iron meteorite samples analysed in this study: (a) !182/184W (6/4) vs. !183/184W (6/4), (b) !182/183W (6/3) vs. !183/184W (6/3). Error bars indicate 95% conf. limits of the mean (n=4-9). Also shown are: (i) the expected correlation line for the mass-independent effect (i.e., a deficit in 183W); (ii) the expected correlation line for variations in s- and r-process W isotopes calculated using the stellar model from Arlandini et al. (1999); and (iii) model arrays (small grey squares) for neutron capture effects on W isotopes in iron meteoroids. The solid lines intersect at an ordinate value of !182W= "3.35, which is approximately the average value for the investigated iron meteorites groups. Although the uncertainties of !182/iW and !183/184W (6/i) are positively correlated, error ellipses were omitted in this diagram for clarity.
–0.4 –0.2 0.0 0.2 0.4–4.0
–3.8
–3.6
–3.4
–3.2
–3.0
–2.8
–2.6
–0.4 –0.2 0.0 0.2 0.4–4.0
–3.8
–3.6
–3.4
–3.2
–3.0
–2.8
–2.6
183/184W (6/4)
182/
184 W
(6/4
)
183W deficit
182/
183 W
(6/3
)
(a)
(b)
183/184W (6/3)
sde
ficit/r
exce
ss
n capture
183W deficit
s deficit/r excess
Mbosi
Mbosi
Chinga
Chinga
SA02 SA01
SA02
SA01
GB ’99’
GB ’99’
TR
TRGB ´Railway’
44 Hf-W chronology of weakly irradiated iron meteorites
Fig. 3.4: !182/184W (6/4) for the iron meteorite samples investigated in this study. Distinguished are samples with low exposure ages whose W isotope compositions are unaffected by cosmic ray effects (filled symbols) and samples which have longer exposure times and/or whose !182W was lowered as a result of cosmic ray interactions (open symbols). For Chinga (UNG) the !182W corrected for nucleosynthetic isotope anomalies is shown. The vertical hashed area represents the CAI initial of !182W = "3.51±0.10 (Burkhardt et al., 2008; 2012). Error bars indicate 95% conf. limits of the mean.
182W (6/4)
IIIAB irons
IIAB irons
IVA irons
CAI
initi
al
IIG irons
Ungrouped irons
Mbosi
Chinga
Tombigbee River
Gibeon ’99’
Gibeon ’Egg’
Muonionalusta
Gibeon ’Railway’
Gibeon ’102’
Cape York ’CY02’
Cape York ’CY01’
Sikhote Alin ’SA02’
Sikhote Alin ’SA01’
Braunau
Edmonton, Canada
–4.2 –4.0 –3.8 –3.6 –3.4 –3.2 –3.0 –2.8
Part A Chapter 3 45
3.5 Discussion
3.5.1 Identifying iron meteorite samples with very low fluences of (epi)thermal neutrons
Very low concentrations of cosmogenic noble gases in an iron meteorite sample may be the result of a low exposure age or of a heavily shielded location in a large meteoroid. In the first case, W isotope compositions will essentially be unaffected by cosmic rays irrespective of meteoroid size and sample position. A sample from a large meteoroid may also be unaffected by cosmic rays, regardless of its exposure age, if it was located deep enough in the pre-atmospheric body such that not only the high energy particle fluence but also the epi(thermal) neutron fluence were strongly attenuated. However, if a sample was irradiated at some intermediate depth where the high energy particle flux was already low but the (epi)thermal neutron flux was still high, noble gas concentrations in the sample may underestimate the modification of its W isotopic composition by neutron capture effects. Model calculations for neutron capture effects on W isotope compositions suggest that for a spherical iron meteoroid with an exposure age of 60 Ma, the maximum cosmic ray induced decrease in "182W would be ~0.09 "-units [calculated using the nuclide production model described for noble gases and several radionuclides by Ammon et al. (2009) and Leya and Masarik (2009)], which is close to our average analytical uncertainty (Table 3.2 and 3.3). This maximum cosmogenic effect would be reached in the centre of an iron meteoroid with a radius of 85 cm. As for all other meteoroid sizes and sample depths the correction would be – partly considerably – smaller than 0.09 "-units, we consider the measured W isotopic composition in meteorites with exposure ages <60 Ma unaffected by neutron capture effects. In the following paragraphs we therefore evaluate which of the meteorites studied here can be assigned exposure ages below 60 Ma (Table 3.4). We will, at this stage, assume that all samples were subjected to a single stage exposure history to cosmic rays, i.e., we assume that the entire fluence of (epi)thermal neutrons in any sample was acquired at the same shielding and over the same time interval as the entire fluence of the high energy particles producing cosmogenic noble gases. We will later discuss the validity of this assumption in light of the W isotope data. We furthermore assume a temporally constant primary cosmic ray flux (cf., Wieler et al. 2011).
Braunau (IIAB, 39 kg) has a shielding-corrected very low exposure age of approximately 8 Ma, determined with the cosmogenic nuclide pairs 39Ar-38Ar, 36Cl-10Be and 36Cl-36Ar, respectively (Cobb, 1966; Chang and Wänke, 1969). According to the model calculations by Ammon et al. (2009), the noble gas concentrations of the investigated sample (e.g., 21Nec = 0.17; 38Arc = 0.65#10-8 cm3STP/g) indicate production rates as expected for a meteoroid with a radius of &15 cm, consistent with the recovered mass of Braunau. The very low 4He/21Ne ratio of ~150, although not precise, also hints at a relatively small pre-atmospheric size. The low exposure age and small pre-atmospheric size indicate that the W isotope composition of Braunau certainly remained unaffected by cosmic ray effects. The maximum shift in "182W predicted by our neutron capture model for a meteorite having TCRE = 8 Ma corresponds to ~ !0.01 "-units (Table 3.4).
Cape York (IIIAB) is a large (>58’000 kg; pre-atmospheric radius >1.2m) and well-studied iron meteorite. Our three noble gas analyses on samples from two different specimens (CY01 & CY02) yield similar concentrations at the very low end of the range of values observed in iron meteorites (Table 3.1), in agreement with other Cape York analyses (Schultz and Franke, 2004). A shielding corrected 129I-129Xe exposure age of 82±7 Ma was determined by Mathew and Marti (2009), slightly higher than the 60 Ma limit mentioned above.
46 Hf-W chronology of weakly irradiated iron meteorites
However, according to the nuclide production model of Ammon et al. (2009) and Leya and Masarik (2009), in a r = 1.2 m meteoroid the maximum cosmic ray induced decrease in 182W/184W for a 82 Ma exposure age is ~0.08 "182W, even marginally lower than the limit of 0.09 "182W tolerated here. We therefore include Cape York into our set of ‘low exposure age meteorites’ (Table 3.4).
Gibeon (IVA) is a large iron meteorite (>26’000 kg; radius > 0.9 m). Reported noble gas concentrations are variable, yet again near the low end of the range observed for iron meteorites (Schultz and Franke, 2004). The Gibeon samples analysed here also have varying noble gas concentrations (Table 3.1). In a comprehensive study, Honda et al. (2009) report shielding-corrected 21Ne-10Be cosmic ray exposure ages for a large number of Gibeon specimens. These authors identified two groups of samples, one with only loosely defined but low (~10-50 Ma) and another with high (~300 Ma) exposure ages. Honda et al. (2009) conclude that Gibeon had a complex exposure history. The samples with low nominal exposure ages (<50 Ma) thus presumably were largely shielded from highly energetic cosmic rays prior to their second (meteoroid) exposure stage and thus likely also had seen a considerably lower (epi)thermal neutron fluence than the samples of the other group. The 21Ne analyses for Gibeon specimens ‘99’, ‘102’ and ‘Railway’ from Honda et al. (2009) are in agreement with our 21Ne results for exactly these samples (Table 3.1). Honda et al. (2009) interpreted the samples ‘Railway’ and ‘102’ to belong to the low exposure age group (<50 Ma). We thus expect cosmic ray effects on W isotopes to be minimal for these samples, assuming a low (epi)thermal neutron fluence for the two samples during the first exposure stage (i.e., for the low noble gas group the above assumption that the total fluences of high energy- and (epi)thermal particles were both acquired at the same constant shielding can be expected to be correct). The maximum negative shift in "182W predicted by the neutron capture model for TCRE = 50 Ma and r > 0.9 m corresponds to !0.07 "-units. As minor – albeit not resolved – variations might thus be expected, the sample with the most elevated "182W, i.e., Gibeon ‘102’, will be taken as representative for Gibeon before exposure to cosmic rays. The other two Gibeon samples analysed here (‘Egg’ and ‘99’) have higher noble gas concentrations than the ‘Railway’ and ‘102’ specimens. Although in absolute terms the concentrations of ‘Egg’ and ‘99’ are still low, these two specimens supposedly belong to the second group recognised by Honda et al. (2009) with an exposure age of approximately 300 Ma. The W isotope compositions of these two samples may thus have been modified by cosmic ray effects.
No shielding-corrected exposure ages based on pairs of a radioactive and a stable cosmogenic nuclide are available for Edmonton, Canada (IIAB, pre-atmospheric mass >7 kg), Muonionalusta (IVA, >230 kg), Tombigbee River (IIG, >43 kg), Chinga (UNG, >200 kg), and Mbosi (UNG, >16’000 kg). However, it seems likely that all these meteorites owe their low noble gas concentrations primarily to a low exposure age rather than a very high shielding. For Tombigbee River and Chinga a rough maximum pre-atmospheric radius of ~60 cm (Ammon et al., 2009) can be estimated from the upper limit of the 4He/21Ne ratio of ~300 (Table 3.1). The minimum 21Ne production rate in the centre of meteoroids of this radius is ~0.19#10-10 cm3STPg-1Ma-1 (Ammon et al., 2009). This yields maximum one stage cosmic ray exposure ages of ~40 Ma for both Tombigbee River and Chinga. For Edmonton, the uncertainty of the 4He/21Ne ratio is too large to obtain an estimate of the maximum pre-atmospheric radius. However, assuming that the Edmonton sample originated from close to the centre of an iron meteoroid with a pre-atmospheric radius of 120 cm, which is the largest object modelled by Ammon et al. (2009), yields an maximum possible one stage exposure age
Part A Chapter 3 47
of ~60 Ma. With the same assumption we arrive at upper limits of ~6 Ma for the exposure ages of Muonionalusta and Mbosi for which 21Necos was below detection limit, if we additionally assume these samples to have had a 21Necos concentration of 1#10-11 cm3STP/g, i.e., the lowest value we were able to measure in one of our samples (Gibeon ‘Railway’). Thus, for all these samples cosmic ray effects on W isotopes can be expected to be minimal and smaller than our analytical resolution for 182W/184W.
Sikhote Alin (27’000 kg; radius > 0.9m) has a 40K-K exposure age of 430 Ma (Voshage, 1984). Hence, Sikhote Alin is the only meteorite in this study with an unequivocally high exposure age. The samples investigated here (SA01 and SA02) presumably derive from rather near the surface of this large meteoroid, as suggested by their 4He/21Ne ratios of ~215 and 250 (Ammon et al., 2009). Also, sample SA01 has by far the highest noble gas concentrations observed in this study and even contains an order of magnitude more gas than the two SA02 samples. Significant cosmic ray effects on W isotopes can thus be anticipated in SA01 but possibly also in SA02.
48 Hf-W chronology of weakly irradiated iron meteorites
thevery
low
endoftherange
ofvalues
observed
iniron
meteorites(T
able
1),in
agreem
entwithother
Cap
eYork
analyses
(Schultzan
dFranke,
2004
).A
shieldingcorrected
129I–
129Xeexposure
ageof82
±7Mawas
determined
by
Mathew
andMarti(2009),slightlyhigher
than
the60
Ma
limitmentioned
above.However,accordingto
thenuclide
productionmodel
ofAmmonet
al.(2009)
andLeyaan
dMasarik
(2009),in
ar=
1.2m
meteoroid
themaxim
um
cosm
icray-induced
decreasein
182W/184W
fora
82Ma
exposure
ageis
!0.08
e182W,even
marginally
lower
than
thelim
itof0.09
e182W
toleratedhere.Wetherefore
include
Cap
eYork
into
oursetof‘lo
wexposure
agemeteorites’
(Tab
le4).
Gibeon(IVA)is
alargeironmeteorite
(>26,000
kg;
ra-
dius>0.9m).Reported
noble
gasconcentrationsarevari-
able,yetagainnearthelow
endoftherange
observed
for
iron
meteorites(Schultzan
dFranke,
2004).
TheGibeon
samplesan
alysed
herealso
havevaryingnoble
gasconcen-
trations(Tab
le1).In
acomprehensive
study,
Hondaet
al.
(2009)
report
shielding-corrected
21Ne–
10Be
cosm
icray
exposure
ages
foralargenumber
ofGibeon
specim
ens.
Theseau
thors
identified
twogroupsofsamples,
onewith
only
loosely
defined
butlow(!
10–50Ma)
andan
other
with
high(!
300Ma)
exposure
ages.Hondaet
al.(2009)
con-
cludethat
Gibeon
had
acomplexexposure
history.The
sampleswith
low
nominal
exposure
ages
(<50
Ma)
thus
presumab
lywerelargelyshielded
from
highly
energeticcos-
mic
rays
priorto
theirsecond(m
eteoroid)exposure
stage
andthuslik
elyalso
had
seen
aconsiderab
lylower
(epi)ther-
mal
neutronfluence
than
thesamplesoftheother
group.
The
21Nean
alyses
forGibeon
specim
ens‘99’,‘102’an
d‘R
ailway’from
Hondaet
al.(2009)
arein
agreem
entwith
our21Neresultsforexactlythesesamples(T
able1).Honda
etal.(2009)
interpretedthesamples‘R
ailway’an
d‘102’to
belongto
thelow
exposure
agegroup(<50
Ma).Wethus
expectcosm
icraye!ects
onW
isotopes
tobeminim
alfor
thesesamples,assumingalow
(epi)thermal
neutronfluence
forthetw
osamplesduringthefirstexposure
stage(i.e.,for
thelownoblegasgrouptheab
ove
assumptionthat
thetotal
fluencesofhigh
energy-an
d(epi)thermal
particles
were
both
acquired
atthesameconstan
tshieldingcan
beex-
pected
tobe
correct).The
maxim
um
negative
shiftin
e182W
predicted
by
the
neutron
capture
model
for
TCRE=
50Maan
dr>0.9m
correspondsto
"0.07
e-units.
Asminor–albeitnotresolved
–variationsmightthusbe
expected,thesample
with
themost
elevated
e182W,i.e.,
Gibeon‘102’,will
betaken
asrepresentative
forGibeonbe-
fore
exposure
tocosm
icrays.Theother
twoGibeonsam-
plesan
alysed
here(‘Egg’an
d‘99’)havehigher
noble
gas
concentrationsthan
the
‘Railway’an
d‘102’specim
ens.
Although
inab
solute
term
stheconcentrationsof‘Egg’
and‘99’
arestill
low,thesetw
ospecim
enssupposedly
be-
long
tothe
second
group
recogn
ised
by
Honda
etal.
(2009)
with
anexposure
ageofap
proximately
300Ma.
TheW
isotopecompositionsofthesetw
osamplesmay
thus
havebeenmodified
bycosm
icraye!ects.
Tab
le4
Summaryofironmeteoritesampleswithminim
alcosm
icraye!ects.
Meteorite
IDRad
iusa
Cosm
ic-ray
exposure
age(T
CRE)b
Max
.De1
82W
GCRc
Measured
e182W
Dt C
AI
[cm]
[Ma]
Method
(model)
[±95
%conf.]
[Myr]
IIABirons
Braunau
M03
615
!8
39Ar–
38Ar,
36Cl–
10Bean
d36Cl–
36Ar(1)
6"0.01
"3.40
±0.05
Edmonton,Can
ada
A02
<12
0<60
Productionrates(4)
6"0.09
"3.38
±0.07
MeanIIAB
"3.39
±0.08
#1:0#
1:1
"1:0
IIIA
Birons
Cap
eYork
(CY01
)A01
P12
082
±7
129I–
129Xe(2)
6"0.08
"3.36
±0.05
Cap
eYork
(CY02
)G02
aP
120
82±
7129I–
129Xe(2)
6"0.08
"3.39
±0.17
MeanIIIA
B"3.37
±0.14
#1:2#
1:5
"1:4
IVA
irons
Gibeon‘102
’C04
P90
<50
10Be–
21Ne(3)
6"0.07
"3.31
±0.08
Muonionalusta
A04
<12
0<6
Productionrates+
noble
gases(4)
6"0.01
"3.33
±0.07
Gibeon‘R
ailway
’A03
P90
<50
10Be–
21Ne(3)
6"0.07
"3.42
±0.08
MeanIV
A(excl.‘R
ailway
’)"3.32
±0.08
#1:6#
1:1
"1:0
Ungrouped
irons
Chinga
M01
660
<40
Productionrates+
noble
gases(4)
6"0.05
"3.30
±0.17
#1:8#
1:8
"1:7
Mbosi
M02
<12
0<6
Productionrates+
noble
gases(4)
6"0.01
"3.09
±0.03
#3:9#
0:9
"0:9
Tombigbee
River
(IIG
)isnotshownas
itlik
elyhad
acomplexexposure
history
anditsW
isotopecompositionwas
likelymodified
bycosm
icrays
(Section5.3).
aPre-atm
osphericradiusforaspherical
meteoroid,b
ased
oneither
thetotalrecoveredmass(lower
bound),ornoblega
ssystem
atics(upper
bound).
bCosm
ic-ray
exposure
ages
(orupper
andlower
boundsthereof)an
dthemethod(s)usedfordeterminingtheseag
es.References:(1)Cobb
(196
6);Chan
gan
dWan
ke(196
9);(2)Mathew
andMarti(200
9);(3)Hondaet
al.(200
9);(4)Ammonet
al.(200
9)/thisstudy.
cModel
prediction
formax
imum
cosm
ic-ray
e!ectone1
82W
forgivenTCRE(M
a),pre-atm
osphericradius,an
d4He/
21Ne(ifdetermined).
298
T.S.Kruijeret
al./Geochim
icaet
Cosm
ochim
icaActa99
(201
2)28
7–30
4
Part A Chapter 3 49
In summary, the samples with shielding-corrected exposure ages <60 Ma as deduced by pairs of stable and radioactive nuclides [Braunau and Gibeon (‘102’ & ‘Railway’)], as well as Cape York with its 82 Ma exposure age are expected to have a W isotopic composition not modified by epi(thermal) neutron capture reactions beyond the present analytical uncertainty of ~0.1 " 182W (95% conf.), and any corrections should mostly be lower than this value (Table 3.4). Five other meteorites (Edmonton Canada, Muonionalusta, Tombigbee River, Chinga, and Mbosi) very likely also have low noble gas based exposure ages and hence should be expected to have W isotope compositions that are unaffected by cosmic rays within our current analytical resolution. We will discuss below, however, that Tombigbee River may have to be excluded from this group, because our assumption of a single stage exposure history to cosmic rays does not seem to be valid for this sample. The W isotope compositions of Sikhote Alin (samples SA01 & SA02), and possibly also Gibeon (‘99’ and ‘Egg’) will likely have been modified more strongly by neutron capture. We emphasise that for the following discussion we use only measured 182W/184W, without any attempt to correct for neutron capture effects.
3.5.2 Core formation ages for samples with minimal cosmic ray effects A model time of metal!silicate separation (i.e., core formation) in iron meteorite parent bodies can be calculated as the time of Hf!W fractionation from an unfractionated, chondritic reservoir. The model age of core formation is conveniently expressed as the time elapsed since formation of CAI and is calculated using the relation
!tCAI = "1#ln
$182W( )Sample " $182W( )Chondrites$182W( )SSI " $182W( )Chondrites
%&'
('
)*'
+' Eq. 3.1
in which # is the 182Hf decay constant of 0.078±0.002 Myr-1 (Vockenhuber et al., 2004), "182WChondrites is the present-day "182W of chondrites ("182W = !1.9±0.1) (Kleine et al., 2002; Schönberg et al., 2002; Yin et al., 2002; Kleine et al., 2004), "182Wsample is the "182W value of an iron meteorite, and "182WSSI is the solar system initial W isotopic composition as determined from a Hf!W isochron for CAI (Burkhardt et al., 2008; 2012). The current best estimate for this value is "182WSSI = –3.51±0.10 (Burkhardt et al., 2012). Note that this value is lower than that used previously (Burkhardt et al., 2008), as a result of correcting the CAI data for nucleosynthetic isotope anomalies (Burkhardt et al., 2012).
As pointed out in Section 3.5.1 the fluence of (epi)thermal neutrons is expected to have been minimal for many of the investigated samples. This applies to the meteorites that have shielding-corrected known low exposure ages (i.e., Braunau, Gibeon ‘102’ and ‘Railway’, Cape York) as well as further samples with most likely low exposure ages (Edmonton Canada, Muonionalusta, Tombigbee River, Chinga, and Mbosi). The ungrouped irons Chinga and Mbosi will be discussed in Section 3.5.3. With the exception of Tombigbee River, all samples with low noble gas based exposure ages indeed show "182W values identical to each other within our analytical resolution (i.e., ~0.1 " units), and slightly higher than the initial W isotope composition of CAI (Fig. 3.4). In contrast, samples for which the noble gas systematics indicate exposure ages of several hundred Ma (i.e., Sikhote Alin, and the two Gibeon specimens ‘99’ and ‘Egg’ with a discernible long first exposure stage) display more negative and more variable "182W values. Sample SA02 from Sikhote Alin has a "182W value significantly lower than the initial W isotope composition in CAI. Thus, the low values of
50 Hf-W chronology of weakly irradiated iron meteorites
both Sikhote Alin and the two Gibeon specimens ‘99’ and ‘Egg’ are reasonably explained as the result of 182W burnout resulting from cosmic ray-induced neutron capture reactions.
The only data point in Fig. 3.4 not perfectly fitting the overall picture is that of Tombigbee River, whose "182W value is slightly lower than the CAI initial and significantly lower than values of any other sample for which the noble gas systematics indicate a low exposure age. Complex exposure histories are not uncommon among iron meteorites (e.g., Welten et al., 2008). One explanation for this discrepancy, therefore, is that – contrary to our assumption in Section 3.5.1 - Tombigbee River had a complex irradiation history, shielded in a first stage deeply enough to have seen only a relatively small fluence of high-energy cosmic ray particles but discernible amounts of (epi)thermal neutrons. However, for all other samples for which cosmic ray effects on W isotopes are expected to be minimal, the perfect consistency between W isotopic data on the one hand and cosmogenic noble gas or radionuclide data on the other strongly indicates that our basic assumption here is valid: samples identified as having a low cosmic ray exposure age have experienced a negligible fluence of (epi)thermal neutrons. Among our low exposure age group of meteorites, Tombigbee River thus appears to be the only meteorite with a complex exposure history.
The mean "182/184W (6/4) values of the investigated iron meteorite samples with low exposure ages and negligible cosmic ray effects are !3.39±0.08, !3.37±0.14, and !3.32±0.08 (±95% conf.) for the IIAB, IIIAB, and IVA iron meteorite groups (Table 3.3 and 3.4; Fig. 3.5). These values are in excellent agreement with results from a previous study that used theoretical models and noble gas systematics to correct for cosmic ray-induced effects (i.e., IIAB: !3.30±0.10; IIIAB: !3.34±0.06; IVA: !3.36±0.07; Fig. 3.5; Qin et al., 2008b), indicating that this procedure quantitatively corrected for the effects of 182W burnout in the IIAB, IIIAB and IVA iron meteorites. Nevertheless, in contrast to the samples investigated in previous studies, the weakly irradiated samples identified here require no correction for cosmic ray effects on their W isotopic compositions. As such, these samples are of paramount importance to establish a precise Hf!W chronometry of iron meteorites.
Core formation ages deduced from the measured "182W values of the IIAB, IIIAB, and IVA irons having negligible cosmic ray effects are displayed in Fig. 3.5. The core formation ages, calculated relative to a solar system initial "182W = !3.51±0.10 (Burkhardt et al., 2012), correspond to !!!!!!!!!!!! Myr for the IIAB, !!!!!!!!!!!! Myr for the IIIAB, and !!!!!!!!!!!! Myr for the IVA iron meteorite parent bodies. The uncertainty on these ages includes the uncertainty on the mean "182W for each iron group, and the uncertainties on the CAI initial and the present-day "182W of carbonaceous chondrites (Equation 3.1). The core formation ages for the IIAB, IIIAB and IVA iron meteorite parent bodies are indistinguishable from each other and indicate that metal segregation in these bodies occurred within less than ~1 Myr of each other. It is noteworthy that unlike the negative model ages reported in several previous studies, the core formation ages obtained here are positive. This does not only reflect the fact that the iron meteorite samples used here have only negligible cosmic ray effects, but is also due to the recent downward revision of the initial "182W of CAI (Burkhardt et al., 2012).
Part A Chapter 3 51
Fig. 3.5: Mean !182W for samples with low cosmic ray exposure ages for the IIAB, IIIAB, IVA iron meteorite groups. The upper axis shows the timing of core formation relative to the formation of CAI [i.e., !182W = "3.51±0.10; (Burkhardt et al., 2008; 2012)]. Shown for comparison are the corrected !182W values obtained in previous studies that employed cosmogenic noble gas systematics or theoretical models of more strongly irradiated meteorites to correct for cosmic ray effects (2: Markowski et al., 2006a; 1: Qin et al., 2008b). Note that the uncertainty on the initial W isotope composition of CAI is excluded from the error bars on the calculated ages, but instead is shown separately (hashed area).
The new Hf!W results from this study confirm earlier conclusions that accretion and core formation in the parent bodies of at least the IIAB, IIIAB, and IVA iron meteorites predated chondrule formation and the accretion of chondrite parent bodies (cf. Kleine et al., 2005a). From the new Hf!W ages and using the model for 26Al heating of planetesimals from Qin et al. (2008b), we can infer accretion time scales for the IIAB, IIIAB, and IVA iron meteorite parent bodies, indicating that these bodies most likely formed within less than ~1 Myr after CAI formation and no later than ~1.5 Myr after CAI formation. In contrast, the parent bodies of chondrites appear to have formed later, as constrained by age differences between CAI and chondrules based on U-Pb and Al-Mg isotope systematics (Kita et al., 2000; Amelin et al., 2002; Kunihiro et al., 2004; Rudraswami and Goswami, 2007; Kurahashi et al., 2008). Note that chondrule formation must have predated accretion of chondrite parent bodies, so that chondrule formation ages correspond to the earliest possible accretion time for a given chondrite parent body. Available Al-Mg ages for chondrules from ordinary chondrites (L, LL) as well as CO and CR chondrites (Kita et al., 2000; Kunihiro et al., 2004; Kurahashi et al., 2008; Hutcheon et al., 2009; Kita and Ushikubo, 2012) indicate that at least these chondrite parent bodies accreted more than ~2 Myr after CAI formation, i.e., ~1 Myr later than the IIAB, IIIAB and IVA iron meteorite parent bodies. It is noteworthy that some chondrules separated from the CV chondrite Allende appear to be as old as CAI (Connolly, 2011), suggesting that some chondrite parent bodies may have accreted as early as the iron meteorite parent bodies.
6-4 4
ε182Wpre-exposure
ΔtCAI [Ma]
IIAB
IIIAB
IVA
(1)
(2)
Previous studiesThis study
-2 0 2
(1)CAI
initi
al
(1)
–4.2 –4.0 –3.8 –3.6 –3.4 –3.2 –3.0 –2.8
52 Hf-W chronology of weakly irradiated iron meteorites
3.5.3 Tungsten isotope anomalies and Hf!W chronometry of ungrouped iron meteorites
Apart from cosmic ray-induced shifts, the W isotopic composition of iron meteorites may vary due to the presence of nucleosynthetic isotope anomalies. Nucleosynthetic heterogeneity among bulk iron meteorite samples has been reported for several elements like Ni, Ru and Mo (Dauphas et al., 2002; Regelous et al., 2008; Chen et al., 2010; Burkhardt et al., 2011). Bulk nucleosynthetic W isotope variations have so far only been identified in CAI from Allende (Burkhardt et al., 2008; 2012) and in some members of the IVB group and ungrouped magmatic iron meteorites (Qin et al., 2008a). These anomalies could either be explained by a deficit in s-process or an excess in r-process W isotopes. Such nucleosynthetic heterogeneity could significantly affect "182W and produces correlation lines in diagrams of "182W versus "183W, whose slopes depend on the normalisation used (Burkhardt et al., 2012) (Fig. 3.3).
The two ungrouped iron meteorites studied here (Chinga and Mbosi) show elevated "182W (up to ~0.4 "-units) in comparison with samples from the major groups of investigated magmatic iron meteorites (Fig. 3.3 and 3.4; Table 3.3). In addition, Chinga also displays elevated "183W of +0.26 (Fig. 3.3). The higher "182W in this meteorite may thus be the result of a nucleosynthetic W isotope anomaly rather than a relatively young core formation age of its parent body. The "182/184W and "183/184W (6/3) values of Chinga correlate as expected for s-process deficits (or r-process excesses) relative to the terrestrial composition (Fig. 3.3) and are consistent with nucleosynthetic effects in W isotopes observed for one other ungrouped iron meteorite (Deep Springs; Qin et al., 2008a). Correcting the measured "182/184W (6/4) of Chinga using the "182W- "183W relation for a variable distribution of s- and r-process isotopes, yields "182Wcorr = !3.30±0.17, in close agreement with the measured values of the IIAB, IIIAB and IVA iron meteorite groups (Fig. 3.4). Thus, there appears to be no resolvable age difference between Chinga and samples from the major magmatic iron meteorite groups (Table 3.4).
In contrast to Chinga, Mbosi does not show evidence for nucleosynthetic isotope variation as no anomalies are observed in "183W. The measured "182W of Mbosi, therefore, indicates that this iron meteorite is slightly younger ('tCAI
= +3.9±0.9 Myr) than the other meteorites studied here. If Mbosi derived from the metallic core of a differentiated parent body, then core formation in this object occurred ~2-3 Myr later than in the IIAB, IIIAB and IVB iron meteorite parent bodies. In this case, core formation in the iron meteorite parent bodies occurred over a longer time interval than given by the Hf!W ages for the major groups of magmatic irons. Alternatively, Mbosi may have formed by a more localized metal!silicate separation event akin to those recorded in the non-magmatic iron meteorites. The metal!silicate separation age for Mbosi would in that case be consistent with ages reported for some non-magmatic iron meteorites (e.g., Kleine et al., 2005a; Schulz et al., 2012).
3.6 Conclusions This study demonstrates that the cosmogenic noble gases He, Ne, and Ar are suitable for identifying iron meteorite specimens with likely minimal cosmic ray effects on W isotopes, especially if meteorites with low cosmic ray exposure ages (i.e., <60 Ma) can be identified. Based on noble gas data we selected several magmatic iron meteorite samples (including IIAB, IIIAB, IVA and IIG, as well as two ungrouped iron meteorites) that were expected to show no resolvable cosmic ray-induced effects on their W isotope compositions. With the exception of the IIG meteorite Tombigbee River all these samples have "182W values identical to each other within our analytical resolution of ~0.1 " units. All these values are slightly
Part A Chapter 3 53
higher than the CAI initial, reflecting radiogenic ingrowth of 182W (corresponding to ~0.1-0.2 " -units) in a time interval between CAI formation and core formation in iron meteorite parent bodies. In contrast, samples from two meteorites with exposure ages on the order of several hundred Ma (but still with low concentrations of cosmogenic noble gases) show "182W values scattering towards more negative values, some of which are lower than the CAI initial. This demonstrates that cosmic ray-induced neutron capture reactions on W isotopes even affected iron meteorite samples having low concentrations of cosmogenic noble gases, if these samples stem from relatively well shielded positions in large meteoroids with long exposure times.
The W isotopic compositions of iron meteorites with negligible cosmic ray effects indicate that core formation in the IIAB, IIIAB and IVA iron meteorite parent bodies occurred !!!!!!!!!!!! Myr, !!!!!!!!!!!! Myr, and !!!!!!!!!!!! Myr after CAI formation, respectively, and that metal segregation in these bodies occurred within a brief time interval of less than ~1 Myr of each other. In contrast, one ungrouped iron meteorite (Mbosi) derives from a parent body that underwent metal!silicate separation 2-3 Myr later than the major magmatic iron meteorite groups. The new Hf!W results are consistent with conclusions of earlier studies that accretion and core formation of the parent bodies of magmatic iron meteorites predated accretion of chondrite parent bodies. However, unlike the sometimes negative model ages obtained in some previous Hf!W studies on magmatic iron meteorites, the ages obtained here are all positive and consistent with the results of a previous study that used physical models to correct for cosmic ray effects on W isotope compositions of iron meteorites. The major advance of the present study, however, is the identification of samples that remained essentially unaffected by cosmic rays, and thus require no correction on their measured "182W at all.
While this study demonstrates that iron meteorites with low cosmic ray exposure ages are well suited to determine W isotopic compositions essentially unmodified by cosmic ray interactions, such samples are rare. Furthermore, Tombigbee River likely shows neutron capture effects on its W isotope composition in spite of very low concentrations of cosmogenic noble gases. In such cases, even a short second stage exposure derived from a low noble gas concentration (perhaps in conjunction with a cosmogenic radionuclide analysis) may not guarantee a negligible (epi)thermal neutron fluence during a first irradiation at larger shielding. The comprehensive application of Hf!W chronometry to all known groups of iron meteorites, therefore, requires the development of a direct neutron dosimeter that will permit the full quantification of neutron capture-induced W isotope shifts in iron meteorites with longer exposure ages (>100 Ma). Initial results suggest that Pt isotopes may be a suitable neutron dosimeter for iron meteorites and an appropriate proxy for neutron capture-induced shifts in W isotopes. Nevertheless, future studies employing such a direct neutron dosimeter will profit from analysis of the weakly irradiated samples identified in this study. As such samples essentially require no correction, they will be key samples to firmly establish the Hf!W chronology of iron meteorites.
Acknowledgements F. Brandstätter (Naturhistorisches Museum Wien), D. Ebel and J. Boesenberg (American Museum of Natural History, New York), B. Hofmann (Naturhistorisches Museum Bern), M. Honda (Tokyo), A. Locock and C. Herd (Univ. Alberta, Edmonton), K. Marti (Univ. California, San Diego), J. Nauber (JNMC, Zürich), and J. Zipfel (Forschungsinstitut und Naturmuseum Senckenberg, Frankfurt) are gratefully acknowledged for providing samples.
54 Hf-W chronology of weakly irradiated iron meteorites
Thomas Kruijer thanks F. Oberli, L. Huber, N. Dalcher and M. Cosarinsky for advice. The very detailed and constructive comments by Larry Nyquist are gratefully acknowledged. This paper greatly benefited from constructive reviews by Nicolas Dauphas and Larry Nyquist and the editorial efforts of Greg Herzog. This study was supported by the Schweizerische Nationalfonds (SNF grant PP00P2_123470 to T. Kleine).
References Amelin Y., Kaltenbach A., Iizuka T., Stirling C. H., Ireland T. R., Petaev M. and Jacobsen S.
B. (2010) U-Pb chronology of the Solar System's oldest solids with variable (238)U/(235)U. Earth and Planetary Science Letters 300, 343-350.
Amelin Y., Krot A. N., Hutcheon I. D. and Ulyanov A. A. (2002) Lead isotopic ages of chondrules and calcium-aluminum-rich inclusions. Science 297, 1678-1683.
Ammon K., Leya I. and Lin Y. (2011) Noble gases in the Xinjiang (Armanty) iron meteorite-A big object with a short cosmic-ray exposure age. Meteoritics & Planetary Science 46, 785-792.
Ammon K., Masarik J. and Leya I. (2008) Noble gases in Grant and Carbo and the influence of S- and P-rich mineral inclusions on the K-41-K-40 dating system. Meteoritics & Planetary Science 43, 685-699.
Ammon K., Masarik J. and Leya I. (2009) New model calculations for the production rates of cosmogenic nuclides in iron meteorites. Meteoritics & Planetary Science 44, 485-503.
Arlandini C., Kappeler F., Wisshak K., Gallino R., Lugaro M., Busso M. and Straniero O. (1999) Neutron capture in low-mass asymptotic giant branch stars: Cross Sections and abundance signatures. Astrophysical Journal 525, 886-900.
Bouvier A. and Wadhwa M. (2010) The age of the Solar System redefined by the oldest Pb-Pb age of a meteoritic inclusion. Nature Geoscience 3, 637-641.
Burkhardt C., Kleine T., Bourdon B., Palme H., Zipfel J., Friedrich J. M. and Ebel D. S. (2008) Hf!W mineral isochron for Ca,Al-rich inclusions: Age of the solar system and the timing of core formation in planetesimals. Geochimica et Cosmochimica Acta 72, 6177-6197.
Burkhardt C., Kleine T., Dauphas N. and Wieler R. (2012) Nucleosynthetic tungsten isotope anomalies in acid leachates of the Murchison chondrite: Implications for hafnium-tungsten chronometry. The Astrophysical Journal Letters 753, L6.
Burkhardt C., Kleine T., Oberli F., Pack A., Bourdon B. and Wieler R. (2011) Molybdenum isotope anomalies in meteorites: Constraints on solar nebula evolution and origin of the Earth. Earth and Planetary Science Letters 312, 390-400.
Chang C. and Wänke H., 1969. Beryllium-10 in iron meteorites, their cosmic-ray exposure and terrestrial ages. In: Millman, P. (Ed.), Meteorite research. Reidel, Dordrecht, The Netherlands.
Chen J. H., Papanastassiou D. A. and Wasserburg G. J. (2010) Ruthenium endemic isotope effects in chondrites and differentiated meteorites. Geochimica et Cosmochimica Acta 74, 3851-3862.
Chen J. H. and Wasserburg G. J. (1990) The Isotopic Composition of Ag in Meteorites and the Presence of Pd-107 in Protoplanets. Geochimica et Cosmochimica Acta 54, 1729-1743.
Cobb J. C. (1966) Iron Meteorites with Low Cosmic Ray Exposure Ages. Science 151, 1524.
Part A Chapter 3 55
Connolly H. C. e. a., 2011. Diffusion within the CAI Bocce Ball 1: The distribution of 26Mg* correlated with variations in Al/Mg within a type B2 inclusion from Allende.42nd Lunar and Planetary Science Conference #1858, Houston, Texas.
Dauphas N., Marty B. and Reisberg L. (2002) Molybdenum evidence for inherited planetary scale isotope heterogeneity of the protosolar nebula. Astrophysical Journal 565, 640-644.
Eugster O. (2003) Cosmic-ray exposure ages of meteorites and lunar rocks and their significance. Chemie Der Erde-Geochemistry 63, 3-30.
Gray C. M., Papanastassiou D. A. and Wasserburg G. J. (1973) Identification of Early Condensates from the Solar Nebula. Icarus 20, 213-239.
Harper C. L. and Jacobsen S. B. (1996) Evidence for Hf-182 in the early solar system and constraints on the timescale of terrestrial accretion and core formation. Geochimica Et Cosmochimica Acta 60, 1131-1153.
Honda A., Nagai H., Nagao K., Bajo K., Takaoka N., Oura Y. and Nishiizumi K. (2009) Irradiation Histories of Iron Meteorites. Proc. Int. Workshop Advances in Cosmic Ray Science 78, 12-17.
Horan M. F., Smoliar M. I. and Walker R. J. (1998) 182W and 187Re-187Os systematics of iron meteorites: Chronology for melting, differentiation, and crystallization in asteroids. Geochimica Et Cosmochimica Acta 62, 545-554.
Hutcheon I. D., Marhas K. K., Krot A. N., Goswami J. N. and Jones R. H. (2009) 26Al in plagioclase-rich chondrules in carbonaceous chondrites: Evidence for an extended duration of chondrule formation. Geochimica et Cosmochimica Acta 73, 5080-5099.
Kita N. T., Nagahara H., Togashi S. and Morishita Y. (2000) A short duration of chondrule formation in the solar nebula: evidence from 26Al in Semarkona ferromagnesian chondrules. Geochimica et Cosmochimica Acta 64, 3913-3922.
Kita N. T. and Ushikubo T. (2012) Evolution of protoplanetary disk inferred from 26Al chronology of individual chondrules. Meteoritics & Planetary Science 47, 1108-1119.
Kleine T., Mezger K., Munker C., Palme H. and Bischoff A. (2004) Hf-182-W-182 isotope systematics of chondrites, eucrites, and martian meteorites: Chronology of core formation and early mantle differentiation in Vesta and Mars. Geochimica et Cosmochimica Acta 68, 2935-2946.
Kleine T., Mezger K., Palme H., Scherer E. and Munker C. (2005a) Early core formation in asteroids and late accretion of chondrite parent bodies: Evidence from Hf-182-W-182 in CAIs, metal-rich chondrites, and iron meteorites. Geochimica et Cosmochimica Acta 69, 5805-5818.
Kleine T., Munker C., Mezger K. and Palme H. (2002) Rapid accretion and early core formation on asteroids and the terrestrial planets from Hf!W chronometry. Nature 418, 952-955.
Kleine T., Palme H., Mezger M. and Halliday A. N. (2005b) Hf!W chronometry of lunar metals and the age and early differentiation of the Moon. Science 310, 1671-1674.
Kleine T., Touboul M., Bourdon B., Nimmo F., Mezger K., Palme H., Jacobsen S. B., Yin Q. Z. and Halliday A. N. (2009) Hf!W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochimica et Cosmochimica Acta 73, 5150-5188.
Kleine T., Touboul M., Van Orman J. A., Bourdon B., Maden C., Mezger K. and Halliday A. N. (2008) Hf!W thermochronometry: Closure temperature and constraints on the accretion and cooling history of the H chondrite parent body. Earth and Planetary Science Letters 270, 106-118.
56 Hf-W chronology of weakly irradiated iron meteorites
Kollár D., Michel R. and Masarik J. (2006) Monte Carlo simulation of GCR neutron capture production of cosmogenic nuclides in stony meteorites and lunar surface. Meteoritics & Planetary Science 41, 375-389.
Kruijer T. S., Fischer-Gödde M., Sprung P., Leya I., Wieler R. and Kleine T., 2012. Neutron capture on platinum and tungsten isotopes in iron meteorites: Implications for Hf!W chronometry. 43rd Lunar and Planetary Science Conference #1529, Houston, Texas.
Kunihiro T., Rubin A. E., McKeegan K. D. and Wasson J. T. (2004) Initial Al-26/Al-27 in carbonaceous-chondrite chondrules: Too little Al-26 to melt asteroids. Geochimica Et Cosmochimica Acta 68, 2947-2957.
Kurahashi E., Kita N. T., Nagahara H. and Morishita Y. (2008) 26Al–26Mg systematics of chondrules in a primitive CO chondrite. Geochimica et Cosmochimica Acta 72, 3865-3882.
Lee D. C. and Halliday A. N. (1995) Hafnium-Tungsten Chronometry and the Timing of Terrestrial Core Formation. Nature 378, 771-774.
Lee D. C., Halliday A. N., Leya I., Wieler R. and Wiechert U. (2002) Cosmogenic tungsten and the origin and earliest differentiation of the moon. Earth and Planetary Science Letters 198, 167-175.
Leya I. and Masarik J. (2009) Cosmogenic nuclides in stony meteorites revisited. Meteoritics & Planetary Science 44, 1061-1086.
Leya I., Wieler R. and Halliday A. N. (2000) Cosmic-ray production of tungsten isotopes in lunar samples and meteorites and its implications for Hf!W cosmochemistry. Earth and Planetary Science Letters 175, 1-12.
Leya I., Wieler R. and Halliday A. N. (2003) The influence of cosmic-ray production on extinct nuclide systems. Geochimica Et Cosmochimica Acta 67, 529-541.
Lingenfelter R. E., Canfield E. H. and Hampel V. E. (1972) The Lunar Neutron-Flux Revisited. Earth and Planetary Science Letters 16, 355-369.
Markowski A., Leya I., Quitté G., Ammon K., Halliday A. N. and Wieler R. (2006a) Correlated helium-3 and tungsten isotopes in iron meteorites: Quantitative cosmogenic corrections and planetesimal formation times. Earth and Planetary Science Letters 250, 104-115.
Markowski A., Quitté G., Halliday A. N. and Kleine T. (2006b) Tungsten isotopic compositions of iron meteorites: Chronological constraints vs. cosmogenic effects. Earth and Planetary Science Letters 242, 1-15.
Masarik J. (1997) Contribution of neutron capture reactions to observed tungsten isotopic ratios. Earth and Planetary Science Letters 152, 181-185.
Mathew K. J. and Marti K. (2009) Galactic cosmic ray-produced (129)Xe and (131)Xe excesses in troilites of the Cape York iron meteorite. Meteoritics & Planetary Science 44, 107-114.
Qin L. P., Dauphas N., Wadhwa M., Markowski A., Gallino R., Janney P. E. and Bouman C. (2008a) Tungsten nuclear anomalies in planetesimal cores. Astrophysical Journal 674, 1234-1241.
Qin L. P., Dauphas N., Wadhwa M., Masarik J. and Janney P. E. (2008b) Rapid accretion and differentiation of iron meteorite parent bodies inferred from Hf-182-W-182 chronometry and thermal modeling. Earth and Planetary Science Letters 273, 94-104.
Regelous M., Elliott T. and Coath C. D. (2008) Nickel isotope heterogeneity in the early Solar System. Earth and Planetary Science Letters 272, 330-338.
Part A Chapter 3 57
Rudraswami N. G. and Goswami J. N. (2007) Al-26 in chondrules from unequilibrated L chondrites: Onset and duration of chondrule formation in the early solar system. Earth and Planetary Science Letters 257, 231-244.
Scherstén A., Elliott T., Hawkesworth C., Russell S. and Masarik J. (2006) Hf!W evidence for rapid differentiation of iron meteorite parent bodies. Earth and Planetary Science Letters 241, 530-542.
Schönberg R., Kamber B. S., Collerson K. D. and Eugster O. (2002) New W-isotope evidence for rapid terrestrial accretion and very early core formation. Geochimica et Cosmochimica Acta 66, 3151-3160.
Schultz L. and Franke L., 2004. Helium, Neon & Argon in Meteorites: A data collection. Max-Planck-Institut für Geochemie, Mainz (Germany).
Schulz T., Upadhyay D., Münker C. and Mezger K. (2012) Formation and exposure history of non-magmatic iron meteorites and winonaites: Clues from Sm and W isotopes. Geochimica et Cosmochimica Acta 85, 200-212.
Scott E. R. D. (1972) Chemical Fractionation in Iron-Meteorites and Its Interpretation. Geochimica et Cosmochimica Acta 36, 1205-1236.
Scott E. R. D. and Wasson J. T. (1975) Classification and Properties of Iron-Meteorites. Reviews of Geophysics 13, 527-546.
Signer P. and Nier A. O. (1960) The Distribution of Cosmic-Ray-Produced Rare Gases in Iron Meteorites. J. Geophys. Res. 65, 2947-2964.
Smoliar M. I., Walker R. J. and Morgan J. W. (1996) Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271, 1099-1102.
Sprung P., Scherer E. E., Upadhyay D., Leya I. and Mezger K. (2010) Non-nucleosynthetic heterogeneity in non-radiogenic stable Hf isotopes: Implications for early solar system chronology. Earth and Planetary Science Letters 295, 1-11.
Vockenhuber C., Oberli F., Bichler M., Ahmad I., Quitte G., Meier M., Halliday A. N., Lee D. C., Kutschera W., Steier P., Gehrke R. J. and Helmer R. G. (2004) New half-life measurement of Hf-182: Improved chronometer for the early solar system. Physical Review Letters 93.
Voshage H. (1978) Investigations on Cosmic-Ray-Produced Nuclides in Iron-Meteorites, .2. New Results on K-41-K-40-He-4-Ne-21 Exposure Ages and Interpretation of Age Distributions. Earth and Planetary Science Letters 40, 83-90.
Voshage H. (1984) Investigations of Cosmic-Ray-Produced Nuclides in Iron-Meteorites .6. The Signer-Nier Model and the History of the Cosmic-Radiation. Earth and Planetary Science Letters 71, 181-194.
Walker R. J. and Touboul M., 2012. Improved constraints on the relative timing of metal segregation in the early solar system using coupled W-Os isotopes. 43rd Lunar and Planetary Science Conference #1166, Houston, Texas.
Welten K. C., Nishiizumi K., Lavielle B., Caffee M. W., Hillegonds D. J., Finkel R. C., Kollar D. and Masarik J. (2008) The complex exposure histories of the Pitts and Horse Creek iron meteorites: Implications for meteorite delivery models. Meteoritics & Planetary Science 43, 1321-1332.
Wieler R. (2002) Cosmic-ray-produced noble gases in meteorites. Noble Gases in Geochemistry and Cosmochemistry 47, 125-170.
Wieler, R., Beer, J., Leya, I. (2011). The galactic cosmic ray intensity over the past 106 - 109 years as recorded by cosmogenic nuclides in meteorites and terrestrial samples. Space Sci.Rev., doi: 10.1007/s11214-11011-19769-11219.
58 Hf-W chronology of weakly irradiated iron meteorites
Willbold M., Elliott T. and Moorbath S. (2011) The tungsten isotopic composition of the Earth's mantle before the terminal bombardment. Nature 477, 195-U91.
Wittig N., Humayun M., Huang S. and Brandon A. D., 2012. Revised tungsten isotope chronology of IVB iron meteorites from W-Os systematics. 43rd Lunar and Planetary Science Conference #1482, Houston, Texas.
Yin Q. Z., Jacobsen S. B., Yamashita K., Blichert-Toft J., Telouk P. and Albarede F. (2002) A short timescale for terrestrial planet formation from Hf!W chronometry of meteorites. Nature 418, 949-952.
Part A Chapter 4 59
Chapter 4
Neutron capture on Pt isotopes and the Hf-W chronology of core formation in planetesimals
Kruijer, T.S.a,b, Fischer-Godde, M.b, Kleine, T.b, Sprung, P.a,b, Leya, I.c, and Wieler, R.a
aETH Zürich, Institute of Geochemistry and Petrology, Zürich, Switzerland. bWestfälische Wilhelms-Universität Münster, Institut für Planetologie, Münster, Germany.
cUniversity of Bern, Space Research and Planetary Sciences, Bern, Switzerland.
Published in Earth and Planetary Science Letters 361, 162-172 (2013)
Abstract The short-lived 182Hf-182W isotope system can provide powerful constraints on the timescales of planetary core formation, but its application to iron meteorites is hampered by neutron capture reactions on W isotopes resulting from exposure to galactic cosmic rays. Here we show that Pt isotopes in magmatic iron meteorites are also affected by capture of (epi)thermal neutrons and that the Pt isotope variations are correlated with variations in 182W/184W. This makes Pt isotopes a sensitive neutron dosimeter for correcting cosmic ray-induced W isotope shifts. The pre-exposure 182W/184W derived from the Pt-W isotope correlations of the IID, IVA and IVB iron meteorites are higher than most previous estimates and are more radiogenic than the initial 182W/184W of Ca-Al-rich inclusions (CAI). The Hf-W model ages for core formation range from +1.6±1.0 million years (Ma; for the IVA irons) to +2.7±1.3 Ma after CAI formation (for the IID irons), indicating that there was a time gap of at least ~1 Ma between CAI formation and metal segregation in the parent bodies of some iron meteorites. From the Hf-W ages a time limit of <1.5-2 Ma after the start of the solar system can be inferred for the accretion of the IID, IVA and IVB iron meteorite parent bodies, consistent with earlier conclusions that the accretion of differentiated planetesimals predated that of most chondrite parent bodies.
60 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
4.1 Introduction Magmatic iron meteorites are thought to represent fragments of the metal cores of small planetary bodies that formed by efficient metal-silicate separation and subsequent segregation and crystallization of metallic melt (e.g., Scott, 1972). Dating iron meteorites, therefore, can help to establish a precise chronology for the accretion, differentiation and cooling history of protoplanets that had formed early in solar system history. Several chronometers involving extinct and extant isotopic systems (including 53Mn-53Cr, 107Pd-107Ag, 182Hf-182W, 187Re-187Os, 207Pb-206Pb) have been applied to iron meteorites and indicate that metal segregation and core crystallization in their parent bodies occurred rapidly, within a few tens of Ma after the start of the solar system (e.g., Chen and Wasserburg, 1990; Smoliar et al., 1996; Kleine et al., 2005; Blichert-Toft et al., 2010). While most of these chronometers record the crystallization and cooling history of the metal cores, the 182Hf-182W system (t1/2 = 8.9 Ma) is particularly suited for constraining the timescale of metal segregation. The refractory nature of Hf and W and the strong fractionation of these two elements during metal-silicate separation render this chronometer uniquely useful to date core formation (e.g., Kleine et al., 2009).
Horan et al. (1998) conducted the first systematic W isotope study on iron meteorites and demonstrated that they are all characterized by strong 182W deficits relative to chondrites. Later studies showed that magmatic iron meteorites have variable 182W/184W, which for some samples appeared to be less radiogenic than the initial 182W/184W of the solar system as determined based on Ca,Al-rich inclusions (CAI) (Kleine et al., 2005; Markowski et al., 2006b; Scherstén et al., 2006; Burkhardt et al., 2008; Qin et al., 2008b). This observation is surprising, because CAI are the oldest yet dated objects recognized to have formed in the solar system (e.g., Gray et al., 1973; Amelin et al., 2010) and consequently should have the lowest 182W/184W of any meteoritic material. It has been argued that the initial 182W/184W obtained for CAI might be too high as a result of re-mobilization of radiogenic W during parent body alteration (Humayun et al., 2007), but the Hf-W systematics of CAI do not show any evidence for this (Burkhardt et al., 2008). More recently, small nucleosynthetic W isotope anomalies in CAI were identified that require a 23 ppm downward revision of the initial 182W/184W of CAI (Burkhardt et al., 2012). Nevertheless, some iron meteorites still have measured 182W/184W ratios that are several tens of ppm below the initial 182W/184W of CAI, by as much as ~70-100 ppm for some samples [e.g., Ainsworth (IIAB), Tlacotepec (IVB)].
The common interpretation of the low 182W/184W in iron meteorites is that they not only reflect radiogenic contributions from 182Hf-decay, but also superimposed neutron capture effects that result from exposure to galactic cosmic rays (Kleine et al., 2005; Markowski et al., 2006b; Scherstén et al., 2006; Qin et al., 2008b). The neutron capture reactions lower the 182W/184W ratios (Masarik, 1997; Leya et al., 2000; Leya et al., 2003), thus leading to spuriously old core formation ages, and are also a possible explanation for 182W/184W variations within a given group of iron meteorites. Cosmic ray-induced W isotope shifts, therefore, severely limit the ability of Hf-W isotope chronometry to precisely constrain the timing of metal segregation in iron meteorite parent bodies. It is neither possible to accurately determine the timing of metal segregation relative to the formation of CAI, nor to resolve differences in the relative timing of metal segregation in iron meteorite parent bodies. Recent multi-collector inductively-coupled plasma mass spectrometry (MC-ICPMS) (Willbold et al., 2011; Kruijer et al., 2012) and negative thermal ionization mass spectrometry (N-TIMS) studies (Touboul and Walker, 2012) have shown that W isotope compositions can be measured with a precision of ~5 ppm (2!), in principal allowing time differences of less than ~500,000 years to be resolved. However, to fully exploit these recent analytical advances to
Part A Chapter 4 61
resolve such small differences in the time of metal segregation requires the development of a reliable proxy for quantifying the magnitude of cosmic ray-induced W isotope variations in iron meteorites.
Currently, no accurate method exists for correcting shifts in the W isotopic composition in iron meteorites induced by neutron capture reactions. Previous studies have employed 3He abundances and/or exposure ages for this purpose (e.g., Markowski et al., 2006a; Qin et al., 2008b). However, cosmogenic noble gases are primarily produced by higher energy nuclear reactions that occur near the surface of a meteoroid, while W isotopes are predominantly affected by neutron-capture reactions at (epi)thermal energies occurring at larger depths (Masarik, 1997; Leya et al., 2003). Hence, noble gas-based proxies only provide an indirect measure of cosmic ray-induced effects on W isotopes, making them inherently model-dependent. It thus remains unclear as to whether noble gas-based correction procedures fully correct for cosmic ray effects on W isotopes. Nevertheless, cosmogenic noble gases are useful for identifying iron meteorite samples with minor to absent (epi)thermal neutron fluences, but such samples are rare (Kruijer et al., 2012).
This study aims to explore the potential of Pt isotopes as a neutron fluence monitor for iron meteorites. Neutron capture probabilities on Pt isotopes (e.g., ENDFB-VI.8 300K library) and Pt concentrations in iron meteorites are reasonably high (Walker et al., 2008; McCoy et al., 2011), making Pt isotopes a promising proxy for quantifying neutron capture effects on W isotopes. We present combined Pt and W isotope data for group IID, IVA, and IVB iron meteorites, supplemented with new model calculations for neutron capture effects on Pt and W isotopes in iron meteoroids. The correlation of Pt and W isotope compositions is used to determine the 'pre-exposure' W isotopic composition of the iron meteorites, that is, their 182W/184W before modification by cosmic ray interactions. The new, neutron capture-corrected W isotopic data are then used to obtain improved constraints on the timing of metal segregation in iron meteorite parent bodies.
4.2 Sample preparation and model calculations
4.2.1 Sample selection The sample selection was guided by previously reported W isotope composition or noble gas isotope systematics of iron meteorites, which indicated variable degrees of irradiation. The four selected IVA samples (Gibeon ‘Railway’, Gibeon ‘102’, Gibeon ‘Egg’ and Muonionalusta) are characterized by very low concentrations of cosmogenic noble gases and thus may have only small if any cosmic ray-induced W isotope effects (Kruijer et al., 2012). The IVB sample set consists of 11 different specimens that, based on their previously reported W isotope compositions (Kleine et al., 2005; Markowski et al., 2006b; Scherstén et al., 2006; Qin et al., 2008b), are likely to cover a wide range in degrees of irradiation. The IID sample suite includes six specimens from Carbo, a strongly irradiated meteorite displaying large cosmic ray-induced W isotope variations (Markowski et al., 2006a). The Carbo specimens studied here were sampled in close proximity to some of the samples investigated by Markowski et al. (2006a), including samples whose 182W/184W are among the lowest yet reported for iron meteorites. To compare the results for Carbo to another IID iron meteorite, a sample of Rodeo was analysed. For this iron meteorite no previous W isotope data are available, but its small recovered mass (~44 kg) suggests a pre-atmospheric size too small to sufficiently allow moderation of secondary neutrons to kinetic energies significant for neutron capture on W.
62 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
4.2.2 Analytical techniques for Pt and W isotope measurements The analytical techniques are described in detail in the Appendix (Section 4.6). In brief, iron meteorite samples (~700 mg) were cleaned by leaching in 6 M HCl and then dissolved in reverse aqua regia (HNO3-HCl 2:1) at 130°C. Tungsten was separated from a ~90% solution aliquot of the sample solution using a two-stage anion exchange chromatography in HCl-HF media (Kruijer et al., 2012), while Pt was extracted from the remaining ~10%. To avoid possible isobaric interference of 192Os on 192Pt, the purification of Pt first involved solvent extraction of Os from reverse aqua regia into CCl4 (Cohen and Waters, 1996), resulting in Os/Pt < 5x10-5. Platinum was then further purified using anion exchange chromatography (Rehkamper and Halliday, 1997).
The Pt and W isotope measurements were performed on the Thermo Scientific® Neptune Plus MC-ICPMS at the University of Münster. Platinum was introduced into the mass spectrometer using an ESI APEX-Q-IR or a Cetac® Aridus II desolvating system. Total ion beam intensities of ~3-5.5"10-10 A were obtained for a ~200 ppb Pt standard solution and a 50-60 #l/min uptake rate. Each isotope measurement consisted of 60 s baseline integrations (deflected beam) followed by 60-100 cycles of Pt isotope ratio measurements of 4.2s each. Instrumental mass bias was corrected relative to 198Pt/195Pt = 0.2145 (‘8/5’) or 196Pt/195Pt = 0.7464 (‘6/5’) using the exponential law. Osmium interference corrections on 192Pt varied between ~2 and 40 $-units, corresponding to ~5-150 ppt Os. Tests with Os-doped Pt standards showed that Os interferences at this level can accurately be corrected (see Appendix, Section 4.6). Significant amounts of Ir remain in the Pt analytes causing tailing effects on neighbouring Pt masses 192 and 194. The magnitude of these tailing effects were monitored and corrected for by repeatedly analysing Ir-doped Pt solution standards during each analytical session (see Appendix, Section 4.6).
Tungsten was introduced into the mass spectrometer using a Cetac® Aridus II desolvating system. Isotope measurements were performed at total ion beam intensities of ~1.5"10-10 A, which were obtained for ~100 ppb W at an uptake rates of ~50-60 µl/min. Each measurement consisted of 60 s baseline integrations (deflected beam) followed by 200 cycles of W isotope ratio measurements of 4.2 s each. Small isobaric interferences from 184Os and 186Os on W isotope ratios were corrected for by monitoring interference-free 189Os, and were generally negligible (<10 ppm). Instrumental mass bias was corrected by normalization to either 186W/183W = 1.9859 (‘6/3’) or 186W/184W = 0.92767 (‘6/4’) using the exponential law.
Both Pt and W isotope compositions are given relative to terrestrial solution standards (Alfa Aesar®) that were analysed bracketing the sample runs, and are reported as $-unit (i.e., parts per 104) deviations from the mean values of the bracketing standards. For W, the reported $iW represent the mean of pooled solution replicates (n=2-6). Reported uncertainties are 95% confidence limits of the mean (n%3), or twice the standard deviation obtained for the NIST129c metal standard (high S steel) that was analysed during the same session (n<3). This standard was doped with additional Pt and W before sample dissolution to match the concentrations of the investigated samples. This procedure enabled us to estimate the accuracy and external reproducibility of the Pt and W isotope measurements. For Pt isotopes the doped NIST129c standard yields $198Pt (6/5) = 0.04±0.23, $196Pt (8/5) -0.01±0.08 (2SD, n=13), and $192Pt (6/5) = 0.5±2.1 relative to the solution standards, demonstrating the accuracy of our analytical routine for Pt isotope analysis.
The majority of the analysed terrestrial metal standards (NIST129c), each of which represents a separate digestion processed through the full chemical separation, yield precise $iW values that are identical to the value of the solution standard (Fig. 4-A5, Table 4-A3).
Part A Chapter 4 63
Some NIST129c (and iron meteorite) analyses, however, yielded small but resolved positive shifts in $182/183W (6/3) (+0.05 to 0.27) and negative shifts in $183/184W (6/3) (0.00 to &0.12), while the normalization not involving 183W is unaffected [e.g., $182W (6/4) = 0.02±0.03 (95% conf., n=8), Fig. 4-A5, Table 4-A3]. This mass-independent effect has also been documented in recent high-precision MC-ICPMS studies for both terrestrial standards as well as silicate rock and iron meteorite samples, and likely relates to W loss during re-dissolution of the samples in Savillex beakers (Willbold et al., 2011; Kruijer et al., 2012). However, different normalization schemes, combined with measurements of the terrestrial metal standard (NIST129c), allow accurate correction of this effect for all $iW normalizations (Willbold et al., 2011; Kruijer et al., 2012).
4.2.3 Model calculations The model used is essentially the same as used previously for calculating production rates of cosmogenic nuclides in iron (Ammon et al., 2009) and stony meteorites (Leya and Masarik, 2009). Briefly, the capture rates were calculated by folding the flux densities for the projectiles with the excitation functions of the relevant reactions. The flux densities of primary protons, secondary protons, and especially secondary neutrons, which were calculated by following the primary and secondary particles using Monte-Carlo methods, are the same as used by Ammon et al., (2009) in their study of cosmogenic nuclides in iron meteorites. All thermal neutron capture rates were calculated using the cross sections from the JEFF-3.0/A database. In addition to the thermal neutron capture reactions, we also consider reactions induced by fast particles, i.e., by particles with energies above a few MeV, to fully cover all possible burnout and production effects. Since no experimentally determined cross sections are available for this energy range and for the reactions relevant for this study, all necessary excitation functions (up to energies of 240 MeV) were calculated using the TALYS code (Koning et al., 2004). At higher energies the TALYS predictions were extrapolated by fully considering the individual reaction mechanisms. Despite the fact that calculated cross-sections are relatively uncertain (by a factor of two, approximately) and the extrapolation procedure relatively crude, most of the effects relevant to this study are due to thermal neutron capture, and hence, any uncertainties resulting from high-energy contributions will only have a small effect on the final result.
64 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
Tabl
e 4.
1Tu
ngst
en a
nd p
latin
um is
otop
e co
mpo
sitio
ns fo
r iro
n m
eteo
rites
.M
eteo
rite
Sour
ceID
N!18
2/18
4 W (6
/4) m
eas.a
!183 W
(6/4
) mea
s.a!18
2/18
3 W (6
/3) m
eas.a
!184 W
(6/3
) mea
s.a!18
2/18
4 W (6
/4) co
rr.a.
b!18
2/18
3 W (6
/3) co
rr.a,
bN
!192 Pt
(8/5
)d!19
6 Pt (
8/5)
d
±2"
±2"
±2"
±2"
±2"
±2"
±2"
±2"
IVB
irons
Sant
a C
lara
c JN
MC
Zür
ich
M04
5-3
.48
±0.
050.
05±
0.03
-3.5
5±
0.06
-0.0
4±
0.02
-3.7
1±
0.05
-3.7
1±
0.07
421
.8±
2.2
0.43
±0.
08Ta
wal
lah
Valle
ycET
H IR
-29
M05
5-3
.29
±0.
120.
05±
0.07
-3.3
5±
0.05
-0.0
3±
0.04
-3.5
2±
0.11
-3.5
2±
0.11
28.
6±
2.2
0.26
±0.
08H
oba
AM
NH
#421
3Q
015
-3.1
5±
0.14
0.12
±0.
15-3
.36
±0.
13-0
.08
±0.
10-3
.38
±0.
13-3
.41
±0.
132
14.4
±2.
20.
12±
0.08
War
burto
n R
ange
A
MN
H#4
229
Q02
5-3
.22
±0.
060.
11±
0.05
-3.3
6±
0.07
-0.0
7±
0.03
-3.4
5±
0.06
-3.4
1±
0.07
27.
6±
2.2
0.19
±0.
08W
arbu
rton
Ran
ge
USN
M#5
884
R06
5-3
.31
±0.
060.
14±
0.06
-3.4
7±
0.03
-0.0
9±
0.04
-3.5
4±
0.06
-3.5
2±
0.04
26.
9±
2.2
0.21
±0.
08Iq
uiqu
eU
SNM
#123
0R
025
-3.4
5±
0.05
0.16
±0.
06-3
.66
±0.
06-0
.11
±0.
04-3
.68
±0.
05-3
.71
±0.
063
28.9
±2.
20.
42±
0.08
Dum
ont
USN
M#7
537
R03
5-3
.43
±0.
050.
16±
0.05
-3.6
4±
0.06
-0.1
1±
0.03
-3.6
6±
0.05
-3.6
9±
0.06
325
.0±
2.2
0.38
±0.
08Sk
ooku
m
AM
NH
#273
Q04
5-3
.20
±0.
060.
18±
0.05
-3.4
6±
0.07
-0.1
2±
0.03
-3.4
3±
0.06
-3.5
1±
0.07
36.
9±
2.2
0.13
±0.
10Sk
ooku
m
USN
M#5
362
R05
5-3
.30
±0.
040.
13±
0.04
-3.4
4±
0.04
-0.0
8±
0.03
-3.5
3±
0.05
-3.4
9±
0.04
26.
9±
2.2
0.18
±0.
08Tl
acot
epec
U
SNM
#872
R04
5-4
.06
±0.
040.
13±
0.05
-4.2
2±
0.05
-0.0
8±
0.03
-4.2
9±
0.05
-4.2
7±
0.05
258
.6±
2.2
0.88
±0.
10W
eave
r Mou
ntai
ns U
SNM
#142
7R
086
-3.2
6±
0.07
0.11
±0.
09-3
.40
±0.
08-0
.07
±0.
06-3
.49
±0.
07-3
.45
±0.
082
7.0
±2.
20.
19±
0.08
Mea
n IV
B iro
ns0.
14±
0.02
-0.0
9±
0.01
IID
iron
sR
odeo
M
ünst
erR
075
-3.0
1±
0.08
0.11
±0.
10-3
.16
±0.
06-0
.07
±0.
07-3
.23
±0.
14-3
.24
±0.
074
1.3
±2.
20.
01±
0.11
Rod
eoc
Mün
ster
S06
5-2
.96
±0.
090.
08±
0.06
-3.0
1±
0.10
-0.0
6±
0.04
-3.1
8±
0.14
-3.1
2±
0.13
1-0
.4±
2.2
-0.0
1±
0.08
Car
boc
ETH
IR-3
P01
5-3
.88
±0.
120.
06±
0.05
-3.9
6±
0.08
-0.0
4±
0.03
-4.1
0±
0.16
-4.1
1±
0.10
228
.5±
2.2
0.63
±0.
08C
arbo
cET
H IR
-3P0
24
-4.0
2±
0.08
-0.0
8±
0.08
-3.8
7±
0.07
0.05
±0.
06-4
.24
±0.
14-4
.20
±0.
132
33.1
±2.
20.
70±
0.08
Car
bo
ETH
IR-3
P03
6-4
.07
±0.
100.
13±
0.07
-4.2
7±
0.07
-0.0
9±
0.05
-4.2
9±
0.14
-4.3
2±
0.08
237
.5±
2.2
0.76
±0.
08C
arbo
cET
H IR
-3P0
45
-4.1
6±
0.13
-0.1
1±
0.06
-4.0
7±
0.05
0.07
±0.
04-4
.38
±0.
16-4
.44
±0.
102
39.1
±2.
20.
82±
0.08
Car
boc
ETH
IR-3
P05
5-4
.17
±0.
06-0
.01
±0.
06-4
.15
±0.
100.
00±
0.04
-4.3
9±
0.13
-4.3
8±
0.13
137
.7±
2.2
0.88
±0.
08C
arbo
cET
H IR
-3S0
55
-3.7
6±
0.06
-0.0
5±
0.06
-3.6
7±
0.08
0.04
±0.
04-3
.98
±0.
13-3
.96
±0.
121
25.1
±2.
20.
50±
0.08
Mea
n II
D ir
ons
0.12
±0.
06-0
.09
±0.
04
IVA
irons
Gib
eon
'102
'cTo
kyo
N02
4-3
.31
±0.
08-0
.02
±0.
10-3
.22
±0.
150.
01±
0.07
-3.3
1±
0.08
-3.2
2±
0.15
13.
5±
2.2
0.02
±0.
08G
ibeo
n 'R
ailw
ay'c
Senc
kenb
erg
S01
5-3
.42
±0.
08-0
.17
±0.
06-3
.19
±0.
070.
11±
0.04
-3.4
2±
0.08
-3.4
1±
0.11
10.
1±
2.2
0.06
±0.
08G
ibeo
n 'E
gg'c
JNM
C Z
üric
hS0
25
-3.4
4±
0.09
-0.0
5±
0.10
-3.3
8±
0.13
0.04
±0.
07-3
.44
±0.
09-3
.45
±0.
191
2.6
±2.
20.
08±
0.08
Muo
nion
alus
tac
ETH
IR-1
6S0
35
-3.3
3±
0.07
-0.0
8±
0.05
-3.2
3±
0.03
0.05
±0.
03-3
.33
±0.
07-3
.33
±0.
071
2.6
±2.
20.
02±
0.08
a Nor
mal
ized
to 1
86W
/184 W
= 0
.927
67 (6
/4) o
r 186 W
/183 W
= 1
.985
9 (6
/3) u
sing
the
expo
nent
ial l
aw. !18
i W =
((18
i W/18
j Wsa
mpl
e)/(18
i W/18
j Wst
d)#1)
x104 .
b Cor
rect
ed fo
r s-p
roce
ss d
efic
its u
sing
the
follo
win
g re
latio
ns (A
rland
ini e
t al.,
199
9; B
urkh
ardt
et a
l., 2
012)
: !
182/
184 W
(6/4
) corr =
!18
2/18
4 W(6
/4) m
eas –
[1.
686 $ !18
3 W (6
/4) Av
g IVB
or I
ID] a
nd !
182/
183 W
(6/3
) corr =
!18
2/18
3 W(6
/3) m
eas –
[–0.
524 $ !18
4 W(6
/3) Av
g IV
B o
r IID
]. R
epor
ted
unce
rtain
ties o
n !
i W c
orre
cted
for n
ucle
osyn
thet
ic e
ffect
s rep
rese
nt p
ropa
gate
d un
certa
intie
s fro
m a
ll va
riabl
es in
the
abov
e eq
uatio
ns.
The
unc
erta
intie
s on
fact
ors f
or s-
and
r- p
roce
ss v
arib
ility
from
Arla
ndin
i et a
l. (1
999)
are
ass
umed
to b
e 20
% (2")
. cA
ffect
ed b
y an
d co
rrec
ted
for m
ass-
inde
pend
ent e
ffect
(Sec
tion
4.2.
2). A
naly
ses a
ffect
ed b
y th
is e
ffect
wer
e ex
clud
ed fr
om m
ean
valu
es.
d Nor
mal
ized
to 19
8 Pt/19
5 Pt =
0.2
145
usin
g th
e ex
pone
ntia
l law
. !19
i Pt =
((19
i Pt/19
5 Ptsa
mpl
e)/(19
i Pt/19
5 Ptst
d)#1)
x104 .
Part A Chapter 4 65
4.3 Results
4.3.1 Pt isotopes The Pt isotope compositions of the magmatic iron meteorites studied here are summarized in Table 4.1 and displayed in Fig. 4.1 (full dataset given in the Appendix, Table 4-A1). All $iPt values reported in the following are normalized to 198Pt/195Pt (i.e., ‘8/5’). While the investigated IVA samples and Rodeo (IID) exhibit a narrow range of $iPt values that are all relatively close to the terrestrial value, all IVB iron meteorites and the six Carbo (IID) samples investigated here display positive anomalies in $192Pt (up to +58.58), $194Pt (up to +1.91) and $196Pt (up to +0.88).
Fig. 4.1: !192Pt (8/5) vs. !196Pt (8/5) for the IID, IVA and IVB iron meteorites analysed in this study. Error bars represent external uncertainties (2SD). Although the uncertainties are correlated, error ellipses were omitted from this diagram for clarity. Labels next to symbols indicate name and Ir/Pt ratios of the meteorite [IID from (Wasson and Huber, 2006), IVA from (McCoy et al., 2011), and IVB from (Walker et al., 2008) and (Campbell and Humayun, 2005)]. Shown as grey dots are modelled effects of neutron capture in iron meteorites for different Ir/Pt (see Section 4.4.1). Abbreviations of meteorite names: CB (Carbo), GB (Gibeon), DU (Dumont), HO (Hoba), IQ (Iquique), MU (Muonionalusta), RD (Rodeo), SC (Santa Clara), SK (Skookum), TL (Tlacotepec), TV (Tawallah Valley), WM (Weaver Mountains), WR (Warburton Range).
4.3.2 W isotopes The W isotope compositions of the investigated iron meteorite samples are reported in Table 4.1 and Fig. 4.2 and 4.3 (full dataset given in Appendix, Table 4-A2-A3). The IVB iron meteorites exhibit strong 182W deficits with $182/184W (6/4) ranging from &3.15 to &4.06, and also show an essentially uniform $183W (6/4) averaging at +0.14±0.02 (2SD, n=10, Table 4.1). The slightly lower $183W (6/4) values of +0.05 observed for Santa Clara and Tawallah Valley (Table 4.1) are attributed to the mass-independent W isotope effect also observed in measurements of the NIST129c standard (see section 4.2.2). All six Carbo (IID) samples display strong deficits in $182/184W (6/4) ranging from –3.88 to –4.17 and $183W (6/4) values varying from –0.11 to +0.13. In contrast, two subsamples of Rodeo (IID) have higher $182W (6/4) and slightly positive $183W (Table 4.1). The variation in measured $183W (6/4) values is caused by a mass-independent 183W fractionation during sample preparation, which lowers the
0 0.2 0.4 0.6 0.8 1.00
10
20
30
40
50
60
70
s-excess / r-deficit
ε196Pt
ε192
Pt
Ir/Pt=1.0
Ir/Pt=0.75
Ir/Pt=0.50
Ir/Pt=0.25HO, 1.01
CB, 0.64
TL, 1.00
IQ, 1.05
DU
MU, GB
RD, 0.66
IIDIVA
IVB
Ir/Pt=0.65
SC, 0.73
(a)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
0.0
0.4
0.8
1.2
1.6
2.0
s-excess / r-deficit
ε194Pt (6/5)
ε194
Pt (8
/5)
Ir/Pt=1.0
Ir/Pt=0.75
Ir/Pt=0.50
Ir/Pt=0.25
SC, 0.73
TL, 1.00
IQ, 1.05
DU
RD, 0.66
CB, 0.64
(b)
66 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
$183W (6/4) values (see section 4.2.2). The true $183W (6/4) of the IID irons thus most likely is slightly elevated relative to the terrestrial composition. The four investigated IVA specimens (three from Gibeon, one from Muoniounalusta) have $182W (6/4) values between –3.31 and –3.44, and $183W (6/4) values between –0.17 and –0.02. Thus, unlike the IVB and IID irons, none of the IVA specimens shows positive $183W (6/4) values, indicating that the true $183W (6/4) of the IVA irons is indistinguishable from the terrestrial composition.
4.4 Discussion
4.4.1 Origin of Pt isotope variations in iron meteorites Iron meteorites show small but resolvable mass-independent isotopic deviations from the terrestrial compositions in elements like Ni (Regelous et al., 2008; Steele et al., 2011), Ru (Chen et al., 2010), and Mo (Dauphas et al., 2002; Burkhardt et al., 2012), whereas such anomalies appear to be absent for Os (Walker, 2012). These anomalies were interpreted to reflect a heterogeneous distribution of isotopically distinct, presolar components at the bulk meteorite scale. However, two lines of evidence indicate that the Pt isotope anomalies observed in the iron meteorites are not of nucleosynthetic origin. First, nucleosynthetic isotope anomalies should be identical for samples from the same meteorite or meteorite group, given that iron meteorites derive from bodies that underwent large-scale melting and differentiation erasing any pre-existing isotope heterogeneity. The large within-group Pt isotope variations observed for the IVB and IID irons, therefore, indicate that the Pt isotope anomalies cannot be of nucleosynthetic origin. Second, the iron meteorites do not plot on the $196Pt-$192Pt and $194Pt (8/5)-$194Pt (6/5) correlation lines expected for varying proportions of s- and r-process Pt isotopes, but are entirely consistent with the effects predicted for neutron capture during cosmic ray exposure (Fig. 4.1). An important observation from Fig. 4.1b is that all the irons plot on correlation lines passing through the terrestrial value. Thus, the Pt isotope variations among the iron meteorites are entirely cosmogenic (i.e., they reflect capture
ε182W (6/4)
IID irons
IVA irons
IVB irons
CAI
initi
al
Tlacotepec
Carbo
Rodeo
Gibeon '102'
Muonionalusta
DumontHobaIquiqueSanta ClaraSkookum AMNH
Tawallah Valley
Warburton Range AMNH
Weaver Mountains
Gibeon 'Railway'Gibeon 'Egg'
Skookum USNM
Warburton Range USNM
–4.5 –4.0 –3.5 –3.0
Fig. 4.2: !182/184W (6/4) variation for the IVA, IID and IVB iron meteorites investigated in this study. Error bars indicate external uncertainties (95% conf.). The CAI initial of !182W = "3.51±0.10 from Burkhardt et al. (2012) is shown for comparison.
Part A Chapter 4 67
reactions with (epi)thermal neutrons during cosmic ray exposure of the meteoroids) and, at least for the IVA, IVB and IID irons, no nucleosynthetic Pt isotope anomalies are resolvable at the current level of analytical precision. This is consistent with the isotopic homogeneity reported for Os in iron meteorites (Walker, 2012).
Fig. 4.1 compares the $192Pt, $196Pt and $194Pt values measured in the iron meteorites to predictions obtained from model calculations for neutron capture effects in iron meteorites (shown as small grey dots). Each linear array of dots represents the model results for a specific Ir/Pt ratio and for all shielding depths in iron meteoroids with pre-atmospheric radii between 10 cm and 100 cm. The interpretation of neutron capture effects on ratios of 195Pt, 196Pt and 198Pt is relatively straightforward, because the resulting Pt isotope anomalies are independent of Ir/Pt. Since 195Pt captures thermal, epithermal and faster neutrons more efficiently than 196Pt and 198Pt (e.g., ENDFB-VI.8 300K library), the reaction 195Pt(n,')196Pt dominates neutron capture-induced effects on these isotope ratios. Due to mass bias correction, the net effect of the neutron capture reactions yields excesses in $196Pt (8/5), as observed for almost all the iron meteorites investigated here (Fig. 4.1a). Both 191Ir and 193Ir have large resonance integrals (e.g., ENDFB-VI.8 300K library) and thus efficiently capture neutrons at epithermal energies. The reactions 191Ir(n,')192Ir((-)192Pt and 193Ir(n,')194Ir((-)194Pt can thus generate anomalies on 192Pt and 194Pt, whose magnitude depends on the Ir/Pt ratio. Fig. 4.1a shows that the IVB irons plot on distinct model arrays for different Ir/Pt ratios, in each case consistent with the measured Ir/Pt of the respective meteorite. The IID irons Carbo and Rodeo have the same Ir/Pt ratio (Wasson and Huber, 2006) and, as expected, fall on a single $192Pt-$196Pt correlation line, again consistent with the model array for their Ir/Pt = 0.65 (Fig. 4.1a). The good agreement of the iron meteorite data with the modelled capture effects from (epi)thermal neutrons demonstrates that the Pt isotope variations are cosmogenic in origin.
4.4.2 Nucleosynthetic vs. radiogenic and cosmogenic W isotope variations Tungsten isotope variations in meteorites may in principle have three different origins. In addition to radiogenic (i.e., from 182Hf-decay) or cosmogenic (i.e., cosmic ray-induced) effects that essentially only affect $182W, there may also be coupled $182W and $183W variations that are nucleosynthetic in origin. Burkhardt et al. (2012) showed that nucleosynthetic isotope anomalies on $182W are ~1.7 times those in $183W. Given that in a chondritic reservoir $182W evolves by ~0.1 $/Ma, this implies that even small nucleosynthetic $183W anomalies can have a significant effect on the chronological interpretation of W isotopic data (Burkhardt et al., 2012).
Nucleosynthetic W isotope anomalies have previously been identified in CAI and some iron meteorites, as is evident from their non-terrestrial 183W/184W (Burkhardt et al., 2008; Qin et al., 2008a; Burkhardt et al., 2012; Kruijer et al., 2012). Most IVB iron meteorites investigated in this study display small but uniform excesses in $183W (Fig. 4.3; Table 4.1, 4-A2), averaging at $183W (6/4) = +0.14±0.02 or $184W (6/3) = –0.09±0.01 (95% conf., n=9), consistent with a previously reported $184W of -0.09±0.01 for the IVB irons (Qin et al., 2008a). The anomalous 183W/184W of the IVB irons is consistent with a slight deficit in s-process isotopes relative to the Earth (Qin et al., 2008a). This interpretation is supported by the systematic difference between $182W (6/4) and $182W (6/3) observed here for the IVB irons (Table 4.1). Because 184W has a larger s-process contribution than 183W, an s-deficit will lead to a difference between measured $182W (6/4) and $182W (6/3). Using the stellar model of s-process nucleosynthesis (Arlandini et al., 1999) and W isotopic data for acid leachates from
68 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
Murchison (Burkhardt et al., 2012) it can be shown that the difference between $182W (6/4) and $182W (6/3), )$182W, corresponds to 1.337"$183W. Thus, based on the average $183W = +0.14±0.02 for the IVB irons, their )$182W is predicted to be 0.19, which is exactly the observed difference between $182W (6/4) and $182W (6/3) (Table 4.1).
The interpretation of the IID iron meteorite data is more complicated, because the $183W values measured for the Rodeo and Carbo samples scatter and show both positive and negative $183W (Fig. 4.3). While the negative $183W values can be attributed to a mass-independent 183W fractionation introduced during sample preparation (see section 4.2.2), this effect cannot produce positive $183W. The true $183W of the IID irons, therefore, must also be slightly positive, as measured for Rodeo and one of the Carbo samples. Based on these samples the average $183W of the IID irons is +0.12±0.06 (2!), indistinguishable from that of the IVB iron meteorites.
In contrast to the IID and IVB irons, the IVA samples do not show resolvable $183W excesses, but most of the samples again show negative $183W values (Fig. 4.3). The IVA irons, therefore, do not seem to have elevated $183W compared to the terrestrial composition, but most likely are characterized by terrestrial 183W/184W.
The presence of small nucleosynthetic W isotope anomalies in the IVB and IID iron meteorites requires a correction of measured $182W values before these can be interpreted in terms of Hf-W chronometry. Based on W isotopic data for acid leachates for the Murchison chondrite, Burkhardt et al. (2012) showed that the co-variation of nucleosynthetic anomalies
Fig. 4.3: W isotope systematics for the iron meteorite samples investigated in this study. (a) !182/184W (6/4) vs. !183/184W (6/4), and (b) !182/183W (6/3) vs. !183/184W (6/3). Distinguished are samples that are affected by a mass-independent effect on 183W (described in Section 4.2.2 and shown as open symbols), and samples that appear unaffected (filled symbols). Also shown are (i) the modelled effect of neutron capture on W isotopes in iron meteorites (grey dots), (ii) the expected trends for deficits in 183W and (iii) the expected correlation line for variations in s- and r-process W isotope abundances calculated using the stellar model from (Arlandini et al., 1999). The solid lines are all plotted at an ordinate value of !182W= "3.23, which is approximately the average pre-exposure value for the investigated iron meteorites groups. Note, however, that the IVA irons probably have a slightly lower pre-exposure !182W and hence, the intersection of the 183W deficit and s-process/r-process lines should be at a slightly lower !182W than shown in the figure. Error bars represent 95% conf. limits around the mean of multiple solution replicates (if n>4) or the external 2# on the metal (NIST129c) standard analysed during the same session. Although the uncertainties are correlated, error ellipses were omitted from this diagram for clarity.
–0.3 –0.2 –0.1 0 0.1 0.2 0.3–4.6
–4.2
–3.8
–3.4
–3.0
Tlacotepec (IVB)
Rodeo (IID)
Neutron capture
ε183W (6/4)
ε182
W (6
/4)
IID
IVA
IVB
s-deficit / r-excess
183W deficit
–0.3 –0.2 –0.1 0 0.1 0.2 0.3–4.6
–4.2
–3.8
–3.4
–3.0183W deficit
s-deficit / r-excess
ε183W (6/3)
ε182
W (6
/3)
Neutron capture
Tlacotepec (IVB)
Rodeo (IID)
(a) (b)
Part A Chapter 4 69
in $183W and $182W follows the predictions of the stellar model for s-process nucleosynthesis from Arlandini et al. (1999). This model, therefore, can be used to correct the measured $182W for nucleosynthetic variations. After correction, the $182/184W (6/4) values of the IVB iron meteorites range from &3.38 to &4.29, and those of the IID iron meteorites from &3.18 to &4.39 (Table 4.1, 4-A2). Identical results are observed for the corrected $182/183W (6/3) values (Table 4.1).
4.4.3 Combined Pt and W isotope systematics
4.4.3.1 Pre-exposure !182W for IVB, IID and IVA iron meteorites The combined Pt and W isotope systematics of the investigated iron meteorites are presented in Fig. 4.4 and 4.5 (and Fig. 4-A7-A8). The IID and IVB iron meteorite groups display well defined correlation lines in $182W versus $iPt space (Fig. 4.4 and 4.5), demonstrating that Pt and W isotopes are most sensitive to neutron capture at the same spectrum of thermal, epithermal, and faster neutron energies. This observation is important given that the neutron energy spectrum in iron-dominated target compositions is shifted to epithermal and higher energies (Kollár et al., 2006; Sprung et al., 2010). Hence, Pt isotopes provide an excellent neutron dosimeter for correcting W isotope compositions of iron meteorites for cosmogenic effects, because the pre-exposure $182W of a given iron meteorite group can be obtained from the $182W–$iPt correlation lines and the intercept $182W value at $iPt = 0.
The $182/184W (6/4) versus $196Pt correlation defined by the IVB irons (MSWD = 0.56) (Fig. 4a) yields a pre-exposure $182/184W (6/4) of &3.26±0.06 (95% conf.). The same value is obtained when using $182/183W (6/3) instead of $182/184W (6/4) (Fig. 4-A7). In $182/184W (6/4) versus $192Pt space the measured data scatter significantly (MSWD=4.8), reflecting the different Ir/Pt of the samples (Fig. 4.4b). This problem can be overcome by normalizing the $192Pt values to a common Ir/Pt using the neutron capture model calculations for Pt isotopes. After normalization to the average Ir/Pt of the IVB irons of 0.682 (solid symbols in Fig. 4.4b), $192Pt correlates well with $182W (6/4) (MSWD=1.5), yielding a pre-exposure $182W of &3.19±0.05 (95% conf.) (Fig. 4.4b), consistent with the value obtained from the $182W (6/4) versus $196Pt correlation (Fig. 4.4a). Using instead the measured $192Pt in the regression results in a more negative pre-exposure $182/184W (6/4) of –3.35±0.08, indicating that the use of measured $192Pt anomalies alone can introduce a systematic bias in the determination of pre-exposure $182W values. We conclude that $182W = &3.26±0.06 is the best estimate for the pre-exposure $182W of the IVB irons, because this value is derived from a purely empirical Pt-W isotope correlation and does not involve any uncertainties introduced by the model calculations.
70 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
Fig. 4.4: Combined Pt and W isotope systematics for the IVB iron meteorites investigated in this study. (a) !182/184W (6/4) vs. !196Pt (8/5), and (b) !182/184W (6/4) vs. !192Pt (8/5). The !192Pt values used for the regression (large filled symbols) were obtained by normalizing the measured values (small open symbols) using their respective Ir/Pt and the neutron capture model for Pt isotopes to a common Ir/Pt of 0.658 (see Section 4.4.3). Also shown are best-fit regressions (York, 1966) for the data (solid lines) that were calculated using Isoplot (Ludwig, 2003), and the corresponding intercept !182Wpre-exposure
values. The regression line in Fig. 4.4b shows the best-fit for the Ir/Pt-normalized data. Error bars around symbols represent external uncertainties (2# for Pt and 95% conf. for W), and error envelopes around regression lines indicate 95% conf. limits. The !182W values shown in the figure are corrected for nucleosynthetic effects (Section 4.4.2). Also shown are modelled arrays (white dots) predicted by the neutron capture model for Pt and W, anchored at the ordinate intercept of the best-fit line through the measured data. See Fig. 4.1 for abbreviations of meteorites.
The IID irons Rodeo and Carbo also display a well-defined correlation in $182/184W (6/4)
versus $196Pt space (MSWD = 0.23), yielding a pre-exposure $182/184W (6/4) of –3.22±0.13 (95% conf.) (Fig. 4.5a). When using the $182/183W (6/3) data an indistinguishable value of –3.19±0.12 is obtained (Fig. 4-A8). Rodeo and Carbo have nearly identical Ir/Pt (Rodeo: Ir/Pt=0.66; Carbo: Ir/Pt=0.64; Wasson and Huber, 2006), and hence, in this specific case the correlation of $192Pt versus $182W can be used to obtain a reliable and precise pre-exposure $182/184W (6/4) of –3.20±0.10 (95% conf.) (Fig. 4.5b) (For $182/183W (6/3) an indistinguishable pre-exposure value of –3.18±0.08 (95% conf.) is obtained). It is noteworthy that Carbo and Rodeo have strikingly similar cosmogenic noble gas concentrations (Schultz and Franke, 2004), yet Carbo displays much larger Pt isotope anomalies than Rodeo (Fig. 4.5). This indicates a much lower total (epi)thermal neutron fluence for Rodeo compared to Carbo, consistent with the much smaller recovered mass of Rodeo. The observed systematic exemplifies that cosmogenic noble gases are produced by high-energy particles and thus in almost all cases cannot be used to reliably correct neutron capture effects on W isotopes.
0 0.2 0.4 0.6 0.8 1.0–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε196Pt
ε182
/184
W (6
/4)
IVB iron meteorites: ε182Wpre-exposure= -3.26±0.06 (95% conf.) slope = -1.13±0.17 MSWD = 0.56
SC
TV
HO
TL
IQDU
0 10 20 30 40 50 60–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε192Pt
ε182
/184
W (6
/4)
IVB iron meteorites ε182Wpre-exposure = -3.19±0.05 (95% conf.) slope = -0.027±0.003 MSWD = 1.5
SC
HO
TL
IQDU
(a)
(b)
Part A Chapter 4 71
Fig. 4.5: Combined Pt and W isotope systematics for the IID iron meteorites investigated in this study. (a) !182/184W (6/4) vs. !196Pt (8/5), and (b) !182/184W (6/4) vs. !192Pt (8/5). Also shown are best-fit regressions (York, 1966) through the data (solid lines) that were calculated using Isoplot (Ludwig, 2003), and the corresponding intercept !182Wpre-exposure
values. Error bars on symbols represent external uncertainties (2# for Pt and 95% conf. for W), and error envelopes around regressions indicate 95% conf. limits. The !182W values shown in this diagram are corrected for nucleosynthetic effects (Section 4.4.2). Also shown are modelled arrays (white dots) predicted by the neutron capture model for Pt and W, anchored at the ordinate intercept of the best-fit line through the measured data. See Fig. 4.1 for abbreviations of meteorites.
All IVA iron meteorite samples display no resolved Pt isotope anomalies, although
Gibeon 'Egg' and Gibeon Railway show hints of small 196Pt excesses (Table 4.1). These two samples also show slightly more negative $182W than the other two IVA irons, consistent with minor neutron capture effects. Owing to the limited spread in $196Pt, the pre-exposure $182W of the IVA irons cannot be derived from an empirical $182W–$196Pt correlation, but can be obtained by correcting the measured $182W using the measured $196Pt and the average slope of the $182W–$196Pt best-fit lines of the IVB and IID irons. This approach results in a pre-exposure $182W of –3.32±0.07 (2SD), indistinguishable from the average –3.32±0.05 (2SD) measured for Gibeon 102 and Muonionalusta. As these two samples have $196Pt~0 and very low cosmogenic noble gas concentrations (Kruijer et al., 2012), no correction of their $182W values is thus required.
4.4.3.2 Comparison to combined Pt-W model results To further assess the significance of the combined Pt-W isotope systematics, we calculated the production rates of cosmogenic Pt and W nuclides produced by capture of thermal, epithermal, and faster neutrons in spherical iron meteoroids as a function of cosmic ray exposure time (0-1500 Ma), parent body radius (5-120 cm) and depth within the parent body
0 0.2 0.4 0.6 0.8 1.0–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε182
/184
W (6
/4)
IID iron meteorites: ε182Wpre-exposure = -3.22±0.13 (95% conf.) slope = -1.40±0.20 MSWD = 0.23
Rodeo (IID)
Carbo (IID)
0 10 20 30 40–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε192Pt
ε182
/184
W (6
/4)
IID iron meteorites: ε182Wpre-exposure = -3.20±0.10 (95% conf.) slope = -0.031±0.004 MSWD = 0.18
Rodeo (IID)Ir/Pt = 0.67
Carbo (IID)(Ir/Pt=0.65)
ε196Pt
(a)
(b)
72 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
(see Section 4.2.3). The modelled effects on Pt and W isotopes are shown in Figs. 4.4 and 4.5 together with the isotopic data. For all exposure ages, meteoroid sizes and sample depths the modelled data points fall on (by approximation) linear arrays. The model in its essence reproduces the measured W and – in particular – Pt (Fig. 4.1) isotopic compositions in iron meteorites very well. The slopes of the $196Pt-$182W best-fit lines for the IID (-1.40±0.20; Fig. 4.5a) and IVB iron meteorites (-1.13±0.17; Fig. 4.4a) agree within uncertainty, but seem to be slightly shallower than the nominal slope predicted by the model (Fig. 4.4 and 4.5). Whether this difference is significant remains unclear, however, because the uncertainty on the slope of the modelled $196Pt-$182W correlation line is difficult to assess. Nevertheless, compositional differences of the meteorite matrices may have influenced the moderation of the secondary neutrons and thus the resultant energy spectrum. Such differences in neutron energy spectrum could have caused a slight difference in the relative neutron capture effects of Pt and W isotopes. In any case, the slight offset between the nominal slope obtained from the neutron capture model and the IID and IVB data underscores the necessity of either identifying iron meteorites with negligible neutron capture effects (as would be evident from $iPt = 0), or alternatively, of obtaining an empirical $182W–$iPt correlation (Figs. 4.4 and 4.5) to determine reliable pre-exposure $182W values for iron meteorites.
4.4.3.3 Comparison to previous studies The present study is among the first to utilize a direct neutron dosimeter for correction of neutron capture effects on W isotopes in iron meteorites. We argue that utilizing such a direct proxy is the most reliable and precise approach to determine pre-exposure $182W values, in particular when empirical Pt-W isotope correlations are used. Previous studies, in contrast, relied on cosmic ray exposure ages or concentrations of cosmogenic noble gases to quantify neutron capture-induced shifts on $182W (Kleine et al., 2005; Markowski et al., 2006a; 2006b; Scherstén et al., 2006; Qin et al., 2008b). The most remarkable outcome of the present study is that the pre-exposure $182W values determined from the combined Pt–W isotope systematics are less negative than previously obtained estimates utilizing noble gas isotope systematics (Fig. 4.6). For instance, the pre-exposure $182W of –3.26±0.06 inferred here for the IVB irons is substantially higher than –3.48±0.02 determined in a previous study (Qin et al., 2008b). For the IID iron meteorites the pre-exposure $182W = –3.20±0.10 is also higher than $182W = –3.70±0.21 reported from a correlation of 3He and $182W in different subsamples of Carbo (Markowski et al., 2006a). It only marginally overlaps with the previously reported range of –3.95 to –3.24 for the IID irons obtained after correcting for the 'maximum' cosmogenic effects (Qin et al., 2008b). For the IVA irons there is better agreement between earlier proposed pre-exposure $182W values and the results presented here, because all previous estimates were largely based on Gibeon, which shows only minor if any cosmic ray-induced W isotope effects.
The higher pre-exposure $182W values determined here compared to those obtained in previous studies utilizing noble gas isotope systematics highlight that Pt isotopes provide a direct neutron dosimeter for correcting cosmogenic shifts on W isotopes while noble gases do not. Thus, for strongly irradiated samples reliable pre-exposure $182W values can only be derived in combination with Pt isotope measurements. Cosmogenic noble gases, in contrast, cannot be used for a quantitative correction of cosmic ray-induced shifts on W isotopes, but are useful for identifying samples that remained largely unaffected by such effects (Kruijer et al., 2012). This is exemplified by the Pt and W isotopic data for Gibeon 102 and Muonionalusta reported here, which show no evidence for secondary neutron capture effects
Part A Chapter 4 73
and are also characterized by very low concentrations of cosmogenic noble gases (Kruijer et al., 2012).
4.4.4 Chronology of metal segregation and age differences between iron meteorites and chondrites
A model age of metal segregation in iron meteorite parent bodies relative to the formation of CAI can be calculated as the time of Hf-W fractionation from an unfractionated reservoir with chondritic Hf/W using the relation
where ($182W)sample is the W isotopic composition of an iron meteorite, ($182W)chondrites = –1.9±0.1 is the present-day W isotopic composition of carbonaceous chondrites (Kleine et al., 2002; Schönberg et al., 2002; Yin et al., 2002; Kleine et al., 2004), and ($182W)SSI = –3.51±0.10 is the solar system initial W isotopic composition as determined from Hf-W data for CAI (Burkhardt et al., 2012).
Fig. 4.6: Pre-exposure !182W values for the IID, IVA and IVB iron meteorite groups investigated in this study, and their associated model ages of core formation, relative to the CAI initial of !182W = "3.51±0.10 from (Burkhardt et al., 2012). Error bars indicate 95% confidence limits around the pre-exposure !182W values derived in this study. Shown for comparison are the !182W and core formation ages obtained in previous studies (2: Markowski et al., 2006b; 1: Qin et al., 2008b) that employed 3He, cosmic ray exposure ages, and/or theoretical models to correct for cosmic ray effects.
The pre-exposure $182W values derived here for the IVB, IVA and IID iron meteorites
are higher than most of the previously reported values for these irons (Fig. 4.6) and thus lead to younger Hf-W ages than calculated previously. The pre-exposure $182W of the IVB irons of –3.26±0.06 corresponds to a Hf-W model age of metal segregation of +2.2!1.0
+1.0 Ma after CAI formation, significantly younger than +0.2±0.8 Ma as calculated using a previously determined pre-exposure $182W of –3.48±0.02 for the IVB iron meteorites (Qin et al., 2008b)
6-4 4
ε182Wpre-exposure
ΔtCAI [Ma]
IID
IVB
IVA
(1)
(2)
Previous studiesThis study
-2 0 2
(1)
CAI
initi
al
(1)
–4.2 –4.0 –3.8 –3.6 –3.4 –3.2 –3.0 –2.8
!tCAI = "1#ln
$182W( )Sample " $182W( )chondrites$182W( )SSI " $182W( )chondrites
%&'
('
)*'
+'
74 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
and the new CAI initial of Burkhardt et al. (2012). Likewise, for the IID irons (pre-exposure $182W = &3.20±0.10) a Hf-W model age of +2.7!1.2
+1.3 Ma after CAI formation is calculated,
younger than ages of &0.7±2.8 Ma (Qin et al., 2008b) and !1.4!1.6+1.8 Ma (Markowski et al.,
2006a) obtained previously for this group of irons. The pre-exposure $182W of the IVA irons of –3.32±0.07 corresponds to a Hf-W model age of +1.6!1.0
+1.0 Ma after CAI formation, in good agreement with a previously reported age for this group of irons (Qin et al., 2008b).
The revised Hf-W ages indicate that core formation in the IID, IVA and IVB iron meteorite parent bodies was about coeval to chondrule formation as given by Al-Mg ages for chondrules from L and LL ordinary chondrites (Kita et al., 2000; Kunihiro et al., 2004; Rudraswami and Goswami, 2007). However, accretion of iron meteorite parent bodies nevertheless must have predated that of most chondrite parent bodies, because core formation occurred after accretion, while chondrule formation took place before accretion. Qin et al. (2008b) showed that core formation ages can be linked to an (instantaneous) accretion timescale using thermal modelling of planetesimals heated by decay of 26Al. Using this model, accretion ages of +1.1!0.7
+0.5 Ma (IVA), +1.4!0.6+0.4 Ma (IVB) and +1.6!0.7
+0.3 Ma (IID) after CAI formation are calculated. Collectively, these ages indicate that iron meteorite parent bodies accreted slightly earlier than the parent bodies of ordinary chondrites, which based on Al-Mg chronometry must have accreted more than ~2 Ma after CAI formation (Kita et al., 2000; Rudraswami and Goswami, 2007; Villeneuve et al., 2009). Chondrules from CR and CO chondrites seem to have formed more than ~3 Ma after CAI formation (Kunihiro et al., 2004; Kurahashi et al., 2008; Hutcheon et al., 2009; Kita and Ushikubo, 2012), indicating that their parent bodies accreted distinctly later than those of the iron meteorites. Thus, although the results in this study require an upward revision of the Hf-W ages for iron meteorites, they are consistent with the main conclusion of earlier Hf-W studies that the parent bodies of differentiated meteorites accreted earlier than those of most chondrites (Kleine et al., 2005; Markowski et al., 2006b; Scherstén et al., 2006; Qin et al., 2008b). Moreover, the difference in accretion timescales between iron meteorite and chondrite parent bodies is also consistent with earlier conclusions that the thermal history of planetesimals is predominantly controlled by the abundance of 26Al at the time of parent body accretion (Urey, 1955; Hevey and Sanders, 2006; Kleine and Rudge, 2011).
4.5 Conclusions Iron meteorites show large Pt isotope variations that result from neutron capture reactions induced during cosmic ray exposure of their parent meteoroids. The Pt isotope anomalies are inversely correlated with variations in 182W/184W ratios, making it possible to quantify the effects of secondary (epi)thermal neutron capture reactions on measured W isotope ratios in iron meteorites. The pre-exposure $182W values derived here from empirically determined Pt-W isotopic correlations are higher than those obtained in most previous studies, which relied on cosmogenic noble gas systematics to quantify neutron capture effects on W isotopes. The pre-exposure $182W values of the IVA, IVB and IID iron meteorites correspond to core formation ages of +1.6!1.0
+1.0 , +2.2!1.0+1.0 and +2.7!1.2
+1.3 Ma after CAI formation, indicating that there
was a time gap of at least ~1 Ma between the formation of CAI and metal segregation in some iron meteorite parent bodies. Furthermore, a time limit of <1.5-2 Ma after CAI formation can be deduced for the accretion of the IVA, IVB and IID iron meteorite parent bodies, which is slightly earlier than accretion of most chondrite parent bodies >2 Ma after CAI formation.
Part A Chapter 4 75
The new Hf-W results are thus consistent with earlier conclusions that accretion of the parent bodies of differentiated meteorites predated that of chondrites.
The results from this study show that the pre-exposure $182W of any given iron meteorite group can be determined to high precision. This will make it possible to identify small differences in the timing of metal segregation in iron meteorite parent bodies of as little as ~1 Ma. Obtaining this level of precision, however, will also require the quantification of any nucleosynthetic W isotope variability among iron meteorites, as these can have a significant effect on 182W abundances.
Acknowledgements The following institutions are gratefully acknowledged for providing the samples investigated in this study: The Field Museum of Natural History, Chicago (P. Heck / J. Holstein), American Museum of Natural History, New York (D. Ebel / J. Boesenberg), Smithsonian Institution, Washington DC (T. McCoy / L. Weizenbach), Senckenberg Forschungszentrum, Frankfurt (J. Zipfel), and M. Honda (Tokyo). We thank D. Cook for discussions, J. Mazarik for providing the particle spectra, and U. Heitmann for technical support. We gratefully acknowledge James Day and an anonymous referee for their constructive reviews, and Tim Elliott for his thoughtful comments and editorial efforts. This study was supported by the Swiss National Science Foundation.
76 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
References Amelin, Y., Kaltenbach, A., Iizuka, T., Stirling, C.H., Ireland, T.R., Petaev, M., Jacobsen,
S.B., 2010. U-Pb chronology of the Solar System's oldest solids with variable 238U/235U. Earth and Planetary Science Letters 300, 343-350.
Amelin, Y., Krot, A.N., Hutcheon, I.D., Ulyanov, A.A., 2002. Lead isotopic ages of chondrules and calcium-aluminum-rich inclusions. Science 297, 1678-1683.
Ammon, K., Masarik, J., Leya, I., 2009. New model calculations for the production rates of cosmogenic nuclides in iron meteorites. Meteoritics & Planetary Science 44, 485-503.
Arlandini, C., Kappeler, F., Wisshak, K., Gallino, R., Lugaro, M., Busso, M., Straniero, O., 1999. Neutron capture in low-mass asymptotic giant branch stars: Cross sections and abundance signatures. Astrophysical Journal 525, 886-900.
Blichert-Toft, J., Moynier, F., Lee, C.T.A., Telouk, P., Albarede, F., 2010. The early formation of the IVA iron meteorite parent body. Earth and Planetary Science Letters 296, 469-480.
Burkhardt, C., Kleine, T., Bourdon, B., Palme, H., Zipfel, J., Friedrich, J.M., Ebel, D.S., 2008. Hf-W mineral isochron for Ca,Al-rich inclusions: Age of the solar system and the timing of core formation in planetesimals. Geochimica et Cosmochimica Acta 72, 6177-6197.
Burkhardt, C., Kleine, T., Dauphas, N., Wieler, R., 2012. Nucleosynthetic tungsten isotope anomalies in acid leachates of the Murchison chondrite: Implications for hafnium-tungsten chronometry. The Astrophysical Journal Letters 753, L6.
Burkhardt, C., Kleine, T., Oberli, F., Pack, A., Bourdon, B., Wieler, R., 2011. Molybdenum isotope anomalies in meteorites: Constraints on solar nebula evolution and origin of the Earth. Earth and Planetary Science Letters 312, 390-400.
Campbell, A.J., Humayun, M., 2005. Compositions of group IVB iron meteorites and their parent melt. Geochimica et Cosmochimica Acta 69, 4733-4744.
Chen, J.H., Papanastassiou, D.A., Wasserburg, G.J., 2010. Ruthenium endemic isotope effects in chondrites and differentiated meteorites. Geochimica et Cosmochimica Acta 74, 3851-3862.
Chen, J.H., Wasserburg, G.J., 1990. The Isotopic Composition of Ag in Meteorites and the Presence of Pd-107 in Protoplanets. Geochimica et Cosmochimica Acta 54, 1729-1743.
Cohen, A.S., Waters, F.G., 1996. Separation of osmium from geological materials by solvent extraction for analysis by thermal ionisation mass spectrometry. Anal Chim Acta 332, 269-275.
Dauphas, N., Marty, B., Reisberg, L., 2002. Molybdenum evidence for inherited planetary scale isotope heterogeneity of the protosolar nebula. Astrophysical Journal 565, 640-644.
Hevey, P.J., Sanders, I.S., 2006. A model for planetesimal meltdown by Al-26 and its implications for meteorite parent bodies. Meteoritics & Planetary Science 41, 95-106.
Horan, M.F., Smoliar, M.I., Walker, R.J., 1998. 182W and 187Re-187Os systematics of iron meteorites: Chronology for melting, differentiation, and crystallization in asteroids. Geochimica et Cosmochimica Acta 62, 545-554.
Humayun, M., Simon, S.B., Grossman, L., 2007. Tungsten and hafnium distribution in calcium-aluminum inclusions (CAIs) from Allende and Efremovka. Geochimica et Cosmochimica Acta 71, 4609-4627.
Hutcheon, I.D., Marhas, K.K., Krot, A.N., Goswami, J.N., Jones, R.H., 2009. 26Al in plagioclase-rich chondrules in carbonaceous chondrites: Evidence for an extended duration of chondrule formation. Geochimica et Cosmochimica Acta 73, 5080-5099.
Kita, N.T., Nagahara, H., Togashi, S., Morishita, Y., 2000. A short duration of chondrule formation in the solar nebula: evidence from 26Al in Semarkona ferromagnesian chondrules. Geochimica et Cosmochimica Acta 64, 3913-3922.
Kita, N.T., Ushikubo, T., 2012. Evolution of protoplanetary disk inferred from 26Al chronology of individual chondrules. Meteoritics & Planetary Science 47, 1108-1119.
Kleine, T., Mezger, K., Munker, C., Palme, H., Bischoff, A., 2004. Hf-182-W-182 isotope systematics of chondrites, eucrites, and martian meteorites: Chronology of core
Part A Chapter 4 77
formation and early mantle differentiation in Vesta and Mars. Geochimica et Cosmochimica Acta 68, 2935-2946.
Kleine, T., Mezger, K., Palme, H., Scherer, E., Munker, C., 2005. Early core formation in asteroids and late accretion of chondrite parent bodies: Evidence from Hf-182-W-182 in CAIs, metal-rich chondrites, and iron meteorites. Geochimica et Cosmochimica Acta 69, 5805-5818.
Kleine, T., Munker, C., Mezger, K., Palme, H., 2002. Rapid accretion and early core formation on asteroids and the terrestrial planets from Hf-W chronometry. Nature 418, 952-955.
Kleine, T., Touboul, M., Bourdon, B., Nimmo, F., Mezger, K., Palme, H., Jacobsen, S.B., Yin, Q.Z., Halliday, A.N., 2009. Hf-W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochimica et Cosmochimica Acta 73, 5150-5188.
Kollár, D., Michel, R., Masarik, J., 2006. Monte Carlo simulation of GCR neutron capture production of cosmogenic nuclides in stony meteorites and lunar surface. Meteoritics & Planetary Science 41, 375-389.
Koning, A.J., Hilaire, S., Duijvestijn, M.C., 2004. TALYS: Comprehensive nuclear reaction modeling., Proceedings of the International Conference on Nuclear Data for Science and Technology. AIP, Santa Fe, USA. .
Kruijer, T.S., Sprung, P., Kleine, T., Leya, I., Burkhardt, C., Wieler, R., 2012. Core formation in the earliest planetesimals inferred from Hf-W chronometry on samples with minimal cosmic-ray effects. Geochimica et Cosmochimica Acta 99, 287-304.
Kunihiro, T., Rubin, A.E., McKeegan, K.D., Wasson, J.T., 2004. Initial Al-26/Al-27 in carbonaceous-chondrite chondrules: Too little Al-26 to melt asteroids. Geochimica et Cosmochimica Acta 68, 2947-2957.
Kurahashi, E., Kita, N.T., Nagahara, H., Morishita, Y., 2008. 26Al–26Mg systematics of chondrules in a primitive CO chondrite. Geochimica et Cosmochimica Acta 72, 3865-3882.
Lee, T., Papanastassiou, D.A., Wasserburg, G.J., 1976. Demonstration of Mg-26 Excess in Allende and Evidence for Al-26. Geophysical Research Letters 3, 109-112.
Leya, I., Lange, H.J., Neumann, S., Wieler, R., Michel, R., 2000. The production of cosmogenic nuclides in stony meteoroids by galactic cosmic-ray particles. Meteoritics & Planetary Science 35, 259-286.
Leya, I., Masarik, J., 2009. Cosmogenic nuclides in stony meteorites revisited. Meteoritics & Planetary Science 44, 1061-1086.
Leya, I., Wieler, R., Halliday, A.N., 2003. The influence of cosmic-ray production on extinct nuclide systems. Geochimica et Cosmochimica Acta 67, 529-541.
Ludwig, K.R., 2003. Isoplot 3.00: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center Special Publication 4, 70 pp.
Markowski, A., Leya, I., Quitte, G., Ammon, K., Halliday, A.N., Wieler, R., 2006a. Correlated helium-3 and tungsten isotopes in iron meteorites: Quantitative cosmogenic corrections and planetesimal formation times. Earth and Planetary Science Letters 250, 104-115.
Markowski, A., Quitte, G., Halliday, A.N., Kleine, T., 2006b. Tungsten isotopic compositions of iron meteorites: Chronological constraints vs. cosmogenic effects. Earth and Planetary Science Letters 242, 1-15.
Masarik, J., 1997. Contribution of neutron-capture reactions to observed tungsten isotopic ratios. Earth and Planetary Science Letters 152, 181-185.
McCoy, T.J., Walker, R.J., Goldstein, J.I., Yang, J., McDonough, W.F., Rumble, D., Chabot, N.L., Ash, R.D., Corrigan, C.M., Michael, J.R., Kotula, P.G., 2011. Group IVA irons: New constraints on the crystallization and cooling history of an asteroidal core with a complex history. Geochimica et Cosmochimica Acta 75, 6821-6843.
Qin, L.P., Dauphas, N., Wadhwa, M., Markowski, A., Gallino, R., Janney, P.E., Bouman, C., 2008a. Tungsten nuclear anomalies in planetesimal cores. Astrophysical Journal 674, 1234-1241.
78 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
Qin, L.P., Dauphas, N., Wadhwa, M., Masarik, J., Janney, P.E., 2008b. Rapid accretion and differentiation of iron meteorite parent bodies inferred from Hf-182-W-182 chronometry and thermal modeling. Earth and Planearyt Science Letters 273, 94-104.
Regelous, M., Elliott, T., Coath, C.D., 2008. Nickel isotope heterogeneity in the early Solar System. Earth and Planearyt Science Letters 272, 330-338.
Rehkamper, M., Halliday, A.N., 1997. Development and application of new ion-exchange techniques for the separation of the platinum group and other siderophile elements from geological samples. Talanta 44, 663-672.
Rudraswami, N.G., Goswami, J.N., 2007. Al-26 in chondrules from unequilibrated L chondrites: Onset and duration of chondrule formation in the early solar system. Earth and Planetary Science Letters 257, 231-244.
Scherstén, A., Elliott, T., Hawkesworth, C., Russell, S., Masarik, J., 2006. Hf-W evidence for rapid differentiation of iron meteorite parent bodies. Earth and Planetary Science Letters 241, 530-542.
Schönberg, R., Kamber, B.S., Collerson, K.D., Eugster, O., 2002. New W-isotope evidence for rapid terrestrial accretion and very early core formation. Geochimica et Cosmochimica Acta 66, 3151-3160.
Schultz, L., Franke, L., 2004. Helium, Neon & Argon in Meteorites: A data collection. Max-Planck-Institut für Geochemie, Mainz (Germany).
Scott, E.R.D., 1972. Chemical Fractionation in Iron-Meteorites and Its Interpretation. Geochimica et Cosmochimica Acta 36, 1205-1236.
Smoliar, M.I., Walker, R.J., Morgan, J.W., 1996. Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271, 1099-1102.
Sprung, P., Scherer, E.E., Upadhyay, D., Leya, I., Mezger, K., 2010. Non-nucleosynthetic heterogeneity in non-radiogenic stable Hf isotopes: Implications for early solar system chronology. Earth and Planetary Science Letters 295, 1-11.
Steele, R.C.J., Elliott, T., Coath, C., Regelous, M., 2011. Confirmation of mass-independent Ni isotopic variability in iron meteorites. Geochimica et Cosmochimica Acta 75, 7906-7925.
Touboul, M., Walker, R.J., 2012. High precision tungsten isotope measurement by thermal ionization mass spectrometry. Int J Mass Spectrom 309, 109-117.
Urey, H.C., 1955. The cosmic abundances of potassium, uranium, and thorium and the heat balances of the Earth, the Moon, and Mars. Proc. Natl. Acad. Sci. U.S. 41, 127-144.
Walker, R.J., 2012. Evidence for homogeneous distribution of osmium in the protosolar nebula. Earth and Planetary Science Letters 351-352, 36-44.
Walker, R.J., McDonough, W.F., Honesto, J., Chabot, N.L., McCoy, T.J., Ash, R.D., Bellucci, J.J., 2008. Modeling fractional crystallization of group IVB iron meteorites. Geochimica et Cosmochimica Acta 72, 2198-2216.
Wasson, J.T., Huber, H., 2006. Compositional trends among IID irons; their possible formation from the P-rich lower magma in a two-layer core. Geochimica et Cosmochimica Acta 70, 6153-6167.
Willbold, M., Elliott, T., Moorbath, S., 2011. The tungsten isotopic composition of the Earth's mantle before the terminal bombardment. Nature 477, 195-U191.
Yin, Q.Z., Jacobsen, S.B., Yamashita, K., Blichert-Toft, J., Telouk, P., Albarede, F., 2002. A short timescale for terrestrial planet formation from Hf-W chronometry of meteorites. Nature 418, 949-952.
York, D., 1966. Least-squares fitting of a straight line. Canadian Journal of Physics 44, 1079-1086.
Part A Chapter 4 79
4.6 APPENDIX : Supplementary text and figures
4.6.1 Sample preparation and chemical separation of Pt Metal samples (!700 mg) were cut using a diamond saw, polished with abrasives and then ultrasonically cleaned in acetone to remove any saw marks and dust. To circumvent any terrestrial contamination, the outermost parts (10%) of the samples were removed by leaching in 6 M HCl at 80-100 ºC for ±20 min. The samples were then dissolved in HNO3-HCl (2:1) on a hot plate (130 °C) for at least 24 hr. Upon complete dissolution, approximately 500 mg and 50-100 mg aliquots were taken for W and Pt isotope analyses. Tungsten was separated from the sample matrix using a two-stage anion exchange chromatography. Full details on the chemical separation of W are given in Kruijer et al. (2012).
The presence of any Os generates significant isobaric interference on 192Pt. Due to the low abundance of 192Pt (0.78%) and high abundance of 192Os (∼41%) even minute amounts of Os have a significant effect (e.g., ∼ 5 $192Pt units for Os/Pt !1x10-5). The purification of Pt, therefore, involved solvent extraction of Os from reverse aqua regia (HNO3-HCl 2:1) into CCl4 (Cohen and Waters, 1996), which significantly reduced Os amounts (5% of initial Os present), while 100% of the Pt was retained. Residual Os was removed by repeated evaporation in HCl-HNO3 (130 ºC), HClO4 (190-200 ºC), and HNO3-H2O2 (80 ºC). Final Pt cuts generally have Os/Pt < 5x10-5, resulting in interference corrections on 192Pt/195Pt between ~2 and ~40 $-units, corresponding to !5-150 ppt Os.
Separation of Pt from the sample matrix was performed using a modified anion exchange chromatography after Rehkamper and Halliday (1997). Quartz glass columns (dimensions: ±0.3 cm2 " 4 cm) were filled with ∼1.2 ml anion exchange resin (Biorad® AG1X8, 200-400 mesh). Samples were loaded onto the resin in 1 M HCl – 0.1 % bromine water and Pt and other highly siderophile elements were eluted as described in Rehkämper and Halliday (1997). Only the 0.8 M HNO3
elution step was discarded in the presented study as sufficiently clean Pt cuts are obtained without this step. Pure Pt cuts are obtained after collection in 12 ml 15.3 M HNO3
at the end of the elution sequence. Total procedural blanks of the Pt isotope measurements were < ~50 pg and hence negligible given the large amounts of Pt analysed. Total yields for Pt were ~50-60 %.
4.6.2 Pt isotope measurements
4.6.2.1 Instrumentation and data acquisition protocol Platinum isotope compositions were measured using a Thermo Scientific® Neptune Plus MC-ICPMS at the University of Münster. Samples were introduced into the mass spectrometer using either an ESI APEX- IR® or a Cetac Aridus II desolvating system. The 192Pt and 189Os masses were assigned to Faraday cups with 1012 * resistors, while all other masses were measured in Faraday cups connected to 1011 * resistors. The MC-ICPMS was equipped with standard Ni (H) sample and skimmer cones for the Pt isotope measurements. Total ion beam intensities of ~3-5.5 x10-10 A were obtained for a ∼200 ppb Pt standard solution and 50-60 #l/min uptake rates. Baselines were obtained by deflecting the beam using the electrostatic analyser for 60-120 s. Platinum isotope measurements consisted of 60-100 cycles of 4.2 s each. Isobaric interference on 192Pt from 192Os was corrected for by monitoring interference-free 189Os. Measured Pt isotope ratios were corrected for instrumental mass bias through
80 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
internal normalization to 198Pt/195Pt = 0.2145 (‘8/5’) or 196Pt/195Pt = 0.7464 (‘6/5’) using the exponential law.
4.6.2.2 Os interferences To investigate the effect of Os interferences on the 192Pt isotope measurements, Pt standard solutions admixed with varying amounts Os were analysed (Fig. 4-A1). For Os/Pt < 5x10-5 that are typical for the samples analysed in this study, accurate Os interference corrections can be made (Fig. 4-A1). Although not feasible for actual samples, even extreme Os/Pt of ∼1x10-2 can accurately be corrected for.
Fig. 4-A1: (a) !192Pt vs. 189Os/195Pt (6/5) for Pt solution standards (Alfa Aesar) admixed with varying amounts of Ir and Os. Solid lines depict measured Pt isotope compositions and correlations lines hence obtained. Dashed lines display !192Pt after correction for Os interference. Inset shows !192Pt for the high end of the investigated Os/Pt range. (b) Reproducibility of !192Pt for the same solution standards after Os interference correction. Error bars indicate the internal precision (2SE), while the hashed area shows the external reproducibility (2SD).
4.6.2.3 Tailing from Ir Separation of Ir from Pt using ion exchange chromatography was incomplete and significant amounts of Ir thus remained in the Pt separates after chemical purification. As demonstrated through mass scans after introduction of Ir standard solutions (Fig. 4-A2), large 191Ir (37.3%) and 193Ir (62.7%) peaks produce tailing on the adjacent relatively small 192Pt (0.78%) peak
10–6 10–5 10–4 10–3 10–2 10–1–5
–4
–3
–2
–1
0
1
2
3
4
5
0 0.0001 0.0002 0.0003 0.0004 0.0005
0
200
400
600
800
189Os/195Pt (6/5)
192 Pt
(6/5
)
0.00 0.02 0.04
0
2.104
4.104
(a)corrected
measured
Pt (Alfa Aesar) with admixed Os: 192PtOs corr. = +0.3±1.6 (2SD, n=8)
189Os/195Pt (6/5)
192 Pt
(6/5
)
(b)
Part A Chapter 4 81
and, to a smaller degree, on 194Pt (32.9%), leading to an overestimation of these isotope abundances. The observed tailing effects of Ir isotopes on 192Pt are as expected for the ~2 ppm abundance sensitivity of the Münster Neptune Plus, as determined using 197Au.
Fig. 4-A2: Tailing effect of Ir masses on 192Pt.
Fig. 4-A3: !192Pt (6/5) vs. 191Ir/195Pt (6/5) for Pt solution standards admixed with varying amounts of Ir. Due to the presence of Ir in the analyzed Pt solutions, mass bias and interference
corrected 192Pt/195Pt and 194Pt/195Pt must be corrected for tailing from the adjacent Ir beams. To evaluate and monitor the effects of Ir tailing on the Pt isotope measurements, Pt standard solutions with different amounts of admixed Ir were analysed throughout each session. The mass bias and interference corrected 192Pt/195Pt and 194Pt/195Pt ratios show excellent correlations (r2 = 0.99) with both (mass bias-corrected) 191Ir/195Pt and 193Ir/195Pt ratios (Fig. 4-A3). Hence, the slopes of these correlation lines were used to correct measured 192Pt/195Pt ratios of samples for tailing contributions from the Ir peaks. Because the magnitude of tailing on 192Pt varies as a function of the back-end vacuum conditions, the 19nPt/195Pt vs. 193Ir/195Pt correlations had to be re-measured during and throughout each measurement session. Each
191.0 191.5 192.0 192.5 193.0
0
1×104
2×104
3×104
4×104
Atomic mass [Amu]
Cou
nts
192Pt background (±1.5 mV)
Tailing from193Ir (±35 V)
0.0 0.5 1.0 1.5 2.0 2.50
5
10
15
20
191Ir/195Pt (6/5)
ε192
Pt (6
/5)
ε192Pt0 = 0.31±0.34 (2SD)slope = 7.70±0.37 (2SD)
82 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
sample measurement was, therefore, interspersed with analyses of Ir-doped standards. The Pt separates for iron meteorites analysed in this study generally showed Ir/Pt ranging from 0.25 to 1.5. Hence, tailing corrections varied from ~2 to ~15 $192Pt units, and were < 0.5 $194Pt units. To assess the accuracy of these tailing corrections, a terrestrial metal standard (NIST129c) doped with Ir and Pt that was processed through the full chemical separation. After tailing correction, a $192Pt of 0.5±2.2 and $194Pt of -0.02±0.09 (2SD, n=7) is obtained, demonstrating that the correction for tailing from Ir is accurate.
4.6.2.4 Uncertainty of Pt isotope measurements The accuracy and external reproducibility of the Pt isotope measurements were estimated by repeated analyses of a terrestrial metal standard (NIST129c, high S steel) that was doped with additional Pt (Alfa Aesar®) before sample dissolution to match the concentrations of the investigated samples. The NIST129c standard without admixed Ir yields $198Pt (6/5) = 0.04±0.23 and $196Pt (8/5) -0.01±0.08 (2SD, n=13), while the NIST129c standard to which both Ir and Pt were added before the chemical separation yields $196Pt (8/5) 0.02±0.07 (2SD, n=7), (Fig. 4-A4). Based on the NIST 129c standard measurements and measurements of Ir-doped standard solutions the external reproducibility on (tailing corrected) $192Pt and $194Pt are estimated to be ±2.2 and ±0.15 $-units (2 SD), respectively. For samples analysed once or twice, these uncertainties are used. For samples analysed more than twice, the standard deviation (2!) of the replicate sample measurements is taken as an estimate for the external reproducibility.
Fig. 4-A4: External reproducibility of !196Pt (8/5) for the terrestrial metal standard (NIST129c) that was processed through the full chemical separation. Error bars indicate internal uncertainties (2SE) of single measurements, while hashed areas represent the external reproducibility (2SD).
4.6.3 W isotope measurements The accuracy and reproducibility of the W isotope measurements were evaluated by repeated measurements of the terrestrial metal standard NIST129c, each of which represents a separate digestion processed through the full chemical separation. The measurements of the NIST 129c standard yield precise $iW values that are identical to the value of the solution standard, but for some samples yielded small but resolved positive shifts in $182/183W (6/3) (up to +0.27)
196 P
t
NIST129c
NIST129c + additional Ir
NIST129c: 196Pt = 0.01±0.08 (2SD, n=13)
NIST129c (+Ir): 196Pt = 0.02±0.07 (2SD, n=7)
–0.30
–0.20
–0.10
0
0.10
0.20
0.30
–0.30
–0.20
–0.10
0
0.10
0.20
0.30
Part A Chapter 4 83
and negative shifts in $183/184W (6/3) (down to &0.12) and $183/184W (6/4) (down to &0.18) (Fig. 4-A5). However, the normalization not involving 183W is unaffected and the mean $182/184W (6/4) for all NIST 129c analyses is 0.02±0.03 (95% conf., n=8) (Fig. 4-A5, Table 4-A3). Such $182W-$183W systematics have also been observed in recent high-precision MC-ICPMS studies for both terrestrial standards as well as silicate rock and iron meteorite samples (Willbold et al., 2011; Kruijer et al., 2012), and are related to a mass-independent W isotope fractionation only affecting 183W induced by W-loss during re-dissolution of the samples in Savillex beakers. Note that the magnitude of this mass-independent 183W effect is variable and that the effect does not occur for all samples.
The mass-independent 183W effect can be corrected by using the different normalization schemes for the W isotope measurements, combined with measurements of the terrestrial metal standard (NIST129c) (Willbold et al., 2011; Kruijer et al., 2012). For example, the terrestrial metal standards analysed in this study plot on a slope of ∼&2 that is predicted for 183W deficits in $182/183W (6/3) vs. $183/184W (6/3) space (Fig. 4-A5b). Likewise, the data plot on a line of slope ~ –1 "#$ $182/184W (6/3) vs. $183/184W (6/3) space, again consistent with the predicted effect of a 183W deficit (Fig. 4-A5c). For all standards and iron meteorites from this study that showed the mass-independent 183W effect, their measured $182/183W (6/3) and $182/184W (6/3) were thus corrected using the measured $183/184W (6/3) and the relations $182/183W (6/3)corr.
= $182/183W (6/3)meas. – (&2) " $183/184W (6/3) and $182/184W (6/3)corr. = $182/184W (6/3)meas. – (&1) " $183/184W (6/3) (Table 4-A2, 4-A3). The corrected $182/183W (6/3) and $182/184W (6/3) values are identical to the measured $182/184W (6/4), indicating that the corrections are accurate (Table 4-A2; Fig. 4-A6).$
84 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
Fig. 4-A5: Tungsten isotope compositions for the terrestrial metal standard (NIST129c) analysed in this study. (a) !182/184W (6/4) vs. !183/184W (6/4), (b) !182/183W (6/3) vs. !183/184W (6/3), and (c) !182/184W (6/3) vs. !183/184W (6/3). Also shown is the expected correlation line for a deficit in 183W. Distinguished are measured !182/183W (solid squares) and those corrected for a mass-independent effect on 183W (open squares). Each data point represents multiple solution replicates (n=2-5) of a digestion that was processed through the full chemical separation. Uncertainties are displayed as 95% conf. limits of the mean. Although the uncertainties of !182W and !183W (6/i) are correlated, error ellipses were omitted in this diagram for clarity.
–0.4 –0.2 0.0 0.2 0.4–0.4
–0.2
0.0
0.2
0.4
–0.4 –0.2 0.0 0.2 0.4
–0.4
–0.2
0.0
0.2
0.4
183/184W (6/4)
182/
184 W
(6/4
) 183W deficit
–0.4 –0.2 0.0 0.2 0.4–0.4
–0.2
0.0
0.2
0.4
183/184W (6/3)
182/
183 W
(6/3
)
(a)
(b)corrected
measured
(c)
182/
184 W
(6/3
)
183/184W (6/3)
183W deficit
Part A Chapter 4 85
Fig. 4-A6: External reproducibility of !182W for the terrestrial metal standard (NIST129c). (a) !182/183W (6/3), (b) !182/184W (6/3), and (c) !182/184W (6/4). Each data point represents multiple solution replicates (n=2-5) of a digestion that was processed through the full chemical separation. For normalizations involving 183W (i.e., those that are potentially affected by a mass-independent effect), both the corrected and uncorrected !182W values are shown. Error bars indicate external uncertainties (95% conf. of the mean). The hashed area shows the combined reproducibility of the NIST129c standards (2SD).
–0.4
–0.2
0.0
0.2
0.4
182/
183 W
(6/3
)18
2/18
4 W (6
/3)
–0.4
–0.2
0.0
0.2
0.4
182/
184 W
(6/4
)NIST129c:
182/183W (6/3)corr. = +0.03±0.08(2SD, n=8)
NIST129c:182/184W (6/3)corr. = +0.03±0.08
(2SD, n=8)
NIST129c:182/184W (6/4)meas. = +0.02±0.08
(2SD, n=8)
–0.4
–0.2
0.0
0.2
0.4
(a)
(b)
(c)
measuredcorrected
86 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
4.6.4 Additional tables and figures
Fig. 4-A7: Combined Pt and W isotope systematics for the IVB iron meteorites analysed in this study. (a) !182/184W (6/4) vs. !196Pt (8/5), (b) !182/183W (6/3) vs. !196Pt (8/5), and (c) !182/184W (6/4) vs. !192Pt (8/5). The !192Pt values used for the regression (large filled symbols) were obtained by normalizing the measured values (small open symbols) using their respective Ir/Pt and the neutron capture model for Pt isotopes to a common Ir/Pt of 0.682 (see Section 4.4.3). Also shown are best-fit regressions (York, 1966) through the data (solid lines) that were calculated using Isoplot (Ludwig, 2003), and the corresponding intercept !182Wpre-exposure
values. Error bars around symbols represent external uncertainties (2SD for Pt and 95% conf. for W), and error envelopes around regression lines indicate 95% conf. limits. The !182W values shown in the figure are corrected for nucleosynthetic effects (Section 4.4.2). Also shown are modelled arrays (white dots) predicted by the neutron capture model for Pt and W, anchored at the ordinate intercept of the best-fit line through the measured data. See Fig. 4.1 (main text) for abbreviations of meteorites.
0.0 0.2 0.4 0.6 0.8–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε196Pt
ε182
/183
W (6
/3)
IVB iron meteorites: ε182Wpre-irradiation = -3.26±0.07 (95% conf.) slope = -1.13±0.17 MSWD = 0.62
SC
TV
HO
TL
IQDU
(a)
(b)
(c)
0.0 0.2 0.4 0.6 0.8 1.0–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε196Pt
ε182
/184
W (6
/4)
IVB iron meteorites: ε182Wpre-exposure= -3.26±0.06 (95% conf.) slope = -1.13±0.17 MSWD = 0.56
SC
TV
HO
TL
IQDU
0 10 20 30 40 50 60–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε192Pt
ε182
/184
W (6
/4)
IVB iron meteorites ε182Wpre-exposure = -3.19±0.05 (95% conf.) slope = -0.027±0.003 MSWD = 1.5
SC
HO
TL
IQDU
Part A Chapter 4 87
Fig. 4-A8: Combined Pt and W isotope systematics for the IIDB iron meteorites analysed in this study. (a) !182/184W (6/4) vs. !192Pt (8/5), (b) !182/183W (6/3) vs. !196Pt (8/5), and (c) !182/184W (6/4) vs. !196Pt (8/5). The !192Pt values used for the regression (large filled symbols) were obtained by normalizing the measured values (small open symbols) using their respective Ir/Pt and the neutron capture model for Pt isotopes to a common Ir/Pt of 0.682 (see Section 4.4.3). Also shown are best-fit regressions (York, 1966) through the data (solid lines) that were calculated using Isoplot (Ludwig, 2003), and the corresponding intercept !182Wpre-exposure
values. Error bars on symbols represent external uncertainties (2SD for Pt and 95% conf. for W), and error envelopes around regressions indicate 95% conf. limits. The !182W values shown in this diagram are corrected for nucleosynthetic effects (Section 4.4.2). Also shown are modelled arrays (white dots) predicted by the neutron capture model for Pt and W, anchored at the ordinate intercept of the best-fit line through the measured data. See Fig. 4.1 (main text) for abbreviations of meteorites.
0 10 20 30 40–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε192Pt
ε182
/183
W (6
/3)
Rodeo (IID)Ir/Pt = 0.67
Carbo (IID)(Ir/Pt=0.65)
(a)
(b)
(c)
IID iron meteorites: ε182Wpre-exposure = -3.18±0.08 (95% conf.) slope = -0.031±0.003 MSWD = 0.33
0 10 20 30 40–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε192Pt
ε182
/184
W (6
/4)
IID iron meteorites: ε182Wpre-exposure = -3.20±0.10 (95% conf.) slope = -0.031±0.004 MSWD = 0.18
Rodeo (IID)Ir/Pt = 0.67
Carbo (IID)(Ir/Pt=0.65)
0.0 0.2 0.4 0.6 0.8 1.0–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
ε182
/184
W (6
/4)
IID iron meteorites: ε182Wpre-exposure = -3.22±0.13 (95% conf.) slope = -1.40±0.20 MSWD = 0.23
Rodeo (IID)
Carbo (IID)
ε196Pt
88 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
Tabl
e 4-
A1
Plat
inum
isot
ope
com
posi
tions
of i
ron
met
eorit
es.
Met
eorit
e ID
N!19
2 Pt (6
/5)a
!194 Pt
(6/5
)a!19
8 Pt (6
/5)a
!192 Pt
(8/5
)a!19
4 Pt (8
/5)a
!196 Pt
(8/
5)a
±2"
±2"
±2"
±2"
±2"
±2"
IVB
iron
sSa
nta
Cla
ra
M04
422
.9±
2.2
0.96
±0.
15-1
.36
±0.
2321
.8±
2.2
0.58
±0.
200.
43±
0.08
Taw
alla
h Va
lley
M05
29.
4±
2.2
0.51
±0.
15-0
.79
±0.
238.
6±
2.2
0.21
±0.
150.
26±
0.08
Hob
a Q
012
14.7
±2.
20.
43±
0.21
-0.3
6±
0.23
14.4
±2.
20.
32±
0.26
0.12
±0.
08W
arbu
rton
Ran
ge
Q02
28.
1±
2.2
0.45
±0.
37-0
.58
±0.
237.
6±
2.2
0.25
±0.
500.
19±
0.08
War
burto
n R
ange
R
062
7.6
±2.
20.
55±
0.15
-0.6
1±
0.23
6.9
±2.
20.
30±
0.15
0.21
±0.
08Iq
uiqu
eR
023
30.2
±2.
21.
73±
0.15
-1.2
7±
0.23
28.9
±2.
21.
32±
0.15
0.42
±0.
08D
umon
t R
033
26.2
±2.
21.
37±
0.28
-1.1
3±
0.23
25.0
±2.
21.
01±
0.19
0.38
±0.
08Sk
ooku
m
Q04
37.
3±
2.2
0.18
±0.
15-0
.39
±0.
306.
9±
2.2
0.08
±0.
150.
13±
0.10
Skoo
kum
R
052
7.4
±2.
20.
48±
0.15
-0.5
4±
0.23
6.9
±2.
20.
30±
0.15
0.18
±0.
08Tl
acot
epec
R
042
61.3
±2.
22.
76±
0.22
-2.6
3±
0.28
58.6
±2.
21.
91±
0.13
0.88
±0.
10W
eave
r Mou
ntai
ns
R08
27.
6±
2.2
0.56
±0.
44-0
.58
±0.
237.
0±
2.2
0.39
±0.
320.
19±
0.08
IID
iron
sR
odeo
R
074
1.3
±2.
20.
01±
0.16
-0.0
2±
0.33
1.3
±2.
20.
00±
0.20
0.01
±0.
11R
odeo
S0
61
-0.4
±2.
20.
01±
0.15
0.02
±0.
23-0
.4±
2.2
0.03
±0.
15-0
.01
±0.
08C
arbo
P0
12
30.4
±2.
21.
53±
0.22
-1.8
7±
0.23
28.5
±2.
20.
93±
0.15
0.63
±0.
08C
arbo
P0
22
35.2
±2.
21.
86±
0.18
-2.0
9±
0.23
33.1
±2.
21.
15±
0.15
0.70
±0.
08C
arbo
P0
32
39.8
±2.
21.
92±
0.15
-2.2
8±
0.23
37.5
±2.
21.
21±
0.15
0.76
±0.
08C
arbo
P0
42
41.6
±2.
22.
24±
0.15
-2.5
5±
0.23
39.1
±2.
21.
40±
0.15
0.82
±0.
08C
arbo
P0
51
40.5
±2.
22.
33±
0.15
-2.6
5±
0.23
37.7
±2.
21.
45±
0.15
0.88
±0.
08C
arbo
S0
51
26.6
±2.
21.
39±
0.15
-1.4
9±
0.23
25.1
±2.
20.
84±
0.15
0.50
±0.
08IV
A ir
ons
Gib
eon
'102
' N
021
nd0.
15±
0.15
-0.0
6±
0.23
nd0.
13±
0.15
0.02
±0.
08G
ibeo
n 'R
ailw
ay'
S01
10.
4±
2.2
0.19
±0.
15-0
.18
±0.
230.
1±
2.2
0.15
±0.
150.
06±
0.08
Gib
eon
'Egg
'S0
21
2.8
±2.
20.
21±
0.15
-0.2
3±
0.23
2.6
±2.
20.
14±
0.15
0.08
±0.
08M
uoni
onal
usta
S03
12.
7±
2.2
-0.0
2±
0.15
-0.0
5±
0.23
2.6
±2.
2-0
.07
±0.
150.
02±
0.08
Terr
estri
al st
anda
rds
NIS
T129
c st
d b
S04
130.
0±
1.2
-0.0
6±
0.10
0.04
±0.
230.
0±
1.2
-0.0
5±
0.10
-0.0
1±
0.08
NIS
T129
c st
d b,
cM
097
0.5
±2.
20.
02±
0.14
-0.0
4±
0.23
0.4
±2.
2-0
.02
±0.
090.
02±
0.07
a Nor
mal
ized
to 19
8 Pt/19
5 Pt =
0.2
145
(8/5
) or 19
6 Pt/19
5 Pt =
0.7
464
(6/5
) usi
ng th
e ex
pone
ntia
l law
. b T
erre
stria
l met
al st
anda
rd (N
IST1
29c)
dop
ed w
ith p
ure
Pt a
nd p
roce
ssed
thro
ugh
full
chem
ical
sepa
ratio
n an
d th
e as
soci
ated
2SD
unc
erta
intie
s.c P
ure
Ir st
anda
rd so
lutio
n ad
ded
befo
re p
roce
ssin
g th
roug
h ch
emic
al se
para
tion.
Part A Chapter 4 89
Tabl
e 4-
A2
Tung
sten
isot
ope
com
posi
tions
of i
ron
met
eorit
es.
Met
eorit
e ID
N!18
2/18
3 W (6
/3) m
eas.a
!182 W
(6/4
) mea
s.a!18
2/18
4 W (6
/3) m
eas.a
!183 W
(6/4
) mea
s.a!18
4 W (6
/3) m
eas.a
!183 W
(6/3
) mea
s.a!18
2/18
3 W (6
/3) co
rr.a,
b!18
2/18
4 W (6
/4) co
rr.a.
b!18
2/18
4 W (6
/3) co
rr.a,
b
±2"
±2"
±2"
±2"
±2"
±2"
±2"
±2"
±2"
IVB
iron
sSa
nta
Cla
rac
M04
5-3
.55
±0.
06-3
.48
±0.
05-3
.51
±0.
030.
05±
0.03
-0.0
4±
0.02
0.04
±0.
02-3
.71
±0.
07-3
.71
±0.
05-3
.70
±0.
05Ta
wal
lah
Valle
ycM
055
-3.3
5±
0.05
-3.2
9±
0.12
-3.3
0±
0.08
0.05
±0.
07-0
.03
±0.
040.
03±
0.04
-3.5
2±
0.11
-3.5
2±
0.11
-3.5
0±
0.10
Hob
a Q
015
-3.3
6±
0.13
-3.1
5±
0.14
-3.2
5±
0.04
0.12
±0.
15-0
.08
±0.
100.
08±
0.10
-3.4
1±
0.13
-3.3
8±
0.13
-3.3
9±
0.11
War
burto
n R
ange
Q
025
-3.3
6±
0.07
-3.2
2±
0.06
-3.3
0±
0.06
0.11
±0.
05-0
.07
±0.
030.
07±
0.03
-3.4
1±
0.07
-3.4
5±
0.06
-3.4
3±
0.08
War
burto
n R
ange
R
065
-3.4
7±
0.03
-3.3
1±
0.06
-3.3
9±
0.03
0.14
±0.
06-0
.09
±0.
040.
09±
0.04
-3.5
2±
0.04
-3.5
4±
0.06
-3.5
2±
0.06
Iqui
que
R02
5-3
.66
±0.
06-3
.45
±0.
05-3
.57
±0.
030.
16±
0.06
-0.1
1±
0.04
0.11
±0.
04-3
.71
±0.
06-3
.68
±0.
05-3
.70
±0.
06D
umon
t R
035
-3.6
4±
0.06
-3.4
3±
0.05
-3.5
3±
0.04
0.16
±0.
05-0
.11
±0.
030.
11±
0.03
-3.6
9±
0.06
-3.6
6±
0.05
-3.6
6±
0.06
Skoo
kum
Q
045
-3.4
6±
0.07
-3.2
0±
0.06
-3.3
3±
0.04
0.18
±0.
05-0
.12
±0.
030.
12±
0.03
-3.5
1±
0.07
-3.4
3±
0.06
-3.4
7±
0.06
Skoo
kum
R
055
-3.4
4±
0.04
-3.3
0±
0.04
-3.3
7±
0.03
0.13
±0.
04-0
.08
±0.
030.
08±
0.03
-3.4
9±
0.04
-3.5
3±
0.05
-3.5
0±
0.05
Tlac
otep
ec
R04
5-4
.22
±0.
05-4
.06
±0.
04-4
.13
±0.
030.
13±
0.05
-0.0
8±
0.03
0.08
±0.
03-4
.27
±0.
05-4
.29
±0.
05-4
.27
±0.
05W
eave
r Mou
ntai
ns
R08
6-3
.40
±0.
08-3
.26
±0.
07-3
.33
±0.
030.
11±
0.09
-0.0
7±
0.06
0.07
±0.
06-3
.45
±0.
08-3
.49
±0.
07-3
.46
±0.
08II
D ir
ons
Rod
eo
R07
5-3
.16
±0.
06-3
.01
±0.
08-3
.10
±0.
050.
11±
0.10
-0.0
7±
0.07
0.07
±0.
07-3
.24
±0.
07-3
.23
±0.
14-3
.23
±0.
06R
odeo
c S0
65
-3.0
1±
0.10
-2.9
6±
0.09
-2.9
9±
0.10
0.08
±0.
06-0
.06
±0.
040.
06±
0.04
-3.1
2±
0.13
-3.1
8±
0.14
-3.1
6±
0.11
Car
boc
P01
5-3
.96
±0.
08-3
.88
±0.
12-3
.92
±0.
100.
06±
0.05
-0.0
4±
0.03
0.04
±0.
03-4
.11
±0.
10-4
.10
±0.
16-4
.10
±0.
11C
arbo
cP0
24
-3.8
7±
0.07
-4.0
2±
0.08
-3.9
5±
0.03
-0.0
8±
0.08
0.05
±0.
06-0
.05
±0.
06-4
.20
±0.
13-4
.24
±0.
14-4
.22
±0.
07C
arbo
P0
36
-4.2
7±
0.07
-4.0
7±
0.10
-4.1
7±
0.07
0.13
±0.
07-0
.09
±0.
050.
09±
0.05
-4.3
2±
0.08
-4.2
9±
0.14
-4.3
0±
0.07
Car
boc
P04
5-4
.07
±0.
05-4
.16
±0.
13-4
.13
±0.
09-0
.11
±0.
060.
07±
0.04
-0.0
7±
0.04
-4.4
4±
0.10
-4.3
8±
0.16
-4.4
2±
0.10
Car
boc
P05
5-4
.15
±0.
10-4
.17
±0.
06-4
.15
±0.
05-0
.01
±0.
060.
00±
0.04
0.00
±0.
04-4
.38
±0.
13-4
.39
±0.
13-4
.38
±0.
05C
arbo
cS0
55
-3.6
7±
0.08
-3.7
6±
0.06
-3.7
3±
0.07
-0.0
5±
0.06
0.04
±0.
04-0
.04
±0.
04-3
.96
±0.
12-3
.98
±0.
13-3
.98
±0.
09IV
A ir
ons
Gib
eon
'102
'cN
024
-3.2
2±
0.15
-3.3
1±
0.08
-3.2
2±
0.10
-0.0
2±
0.10
0.01
±0.
07-0
.01
±0.
07-3
.22
±0.
15-3
.31
±0.
08-3
.22
±0.
10G
ibeo
n 'R
ailw
ay'c
S01
5-3
.19
±0.
07-3
.42
±0.
08-3
.30
±0.
09-0
.17
±0.
060.
11±
0.04
-0.1
1±
0.04
-3.4
1±
0.11
-3.4
2±
0.08
-3.4
1±
0.10
Gib
eon
'Egg
'cS0
25
-3.3
8±
0.13
-3.4
4±
0.09
-3.4
0±
0.06
-0.0
5±
0.10
0.04
±0.
07-0
.04
±0.
07-3
.45
±0.
19-3
.44
±0.
09-3
.44
±0.
09M
uoni
onal
usta
cS0
35
-3.2
3±
0.03
-3.3
3±
0.07
-3.2
9±
0.07
-0.0
8±
0.05
0.05
±0.
03-0
.05
±0.
03-3
.33
±0.
07-3
.33
±0.
07-3
.34
±0.
07a N
orm
aliz
ed to
186W
/184 W
= 0
.927
67 (6
/4) o
r 186 W
/183 W
= 1
.985
9 (6
/3) u
sing
the
expo
nent
ial l
aw. !18
i W =
((18
i W/18
j Wsa
mpl
e)/(18
i W/18
j Wst
d)-1)
x104 .
b Cor
rect
ed fo
r s-p
roce
ss d
efic
its u
sing
the
follo
win
g re
latio
ns (A
rland
ini e
t al.,
199
9; B
urkh
ardt
et a
l., 2
012)
: !
182/
184 W
(6/4
) corr =
!18
2/18
4 W(6
/4) m
eas –
[1.
686 # !18
3 W (6
/4) Av
g IVB
or I
ID] a
nd !
182/
183 W
(6/3
) corr =
!18
2/18
3 W(6
/3) m
eas –
[–0.
524 # !18
4 W(6
/3) Av
g IV
B o
r IID
]. R
epor
ted
unce
rtain
ties o
n !
i W c
orre
cted
for n
ucle
osyn
thet
ic e
ffect
s rep
rese
nt p
ropa
gate
d un
certa
intie
s fro
m a
ll va
riabl
es in
the
abov
e eq
uatio
ns.
The
unc
erta
intie
s on
fact
ors f
or s-
and
r- p
roce
ss v
arib
ility
from
Arla
ndin
i et a
l. (1
999)
are
ass
umed
to b
e 20
% (2")
. cA
ffect
ed b
y an
d co
rrec
ted
for m
ass-
inde
pend
ent e
ffect
(Sec
tion
4.2.
2).
90 Neutron capture on Pt isotopes and Hf-W chronology of iron meteorites
Tabl
e 4-
A3
Tung
sten
isoto
pe c
ompo
sitio
ns fo
r the
terre
stria
l met
al st
anda
rd (N
IST1
29c)
.
Stan
dard
N!18
2/18
3 W (6
/3) m
eas.a!18
2/18
4 W (6
/4) m
eas.a!18
2/18
4 W (6
/3) m
eas.a!18
3 W (6
/4) m
eas.a
!184 W
(6/3
) mea
s.a!18
3 W (6
/3) m
eas.a
!182/
183 W
(6/3
) corr.
a!18
2/18
4 W (6
/4) m
eas.a!18
2/18
4 W (6
/3) co
rr.a
±2"
±2"
±2"
±2"
±2"
±2"
±2"
±2"
±2"
NIS
T129
cb
R09
2-0
.02
±0.
12-0
.03
±0.
09-0
.01
±0.
110.
03±
0.03
-0.0
2±
0.02
0.02
±0.
02-0
.02
±0.
12-0
.03
±0.
09-0
.01
±0.
11M
095
0.06
±0.
080.
04±
0.07
0.05
±0.
06-0
.01
±0.
060.
00±
0.04
0.00
±0.
040.
06±
0.08
0.04
±0.
070.
05±
0.06
L05c
30.
25±
0.09
-0.0
1±
0.11
0.10
±0.
12-0
.18
±0.
040.
12±
0.03
-0.1
2±
0.03
0.02
±0.
10-0
.01
±0.
11-0
.01
±0.
13L0
6c4
0.22
±0.
06-0
.04
±0.
060.
10±
0.09
-0.1
9±
0.03
0.12
±0.
02-0
.12
±0.
02-0
.03
±0.
07-0
.04
±0.
06-0
.02
±0.
09O
023
0.05
±0.
100.
08±
0.14
0.06
±0.
120.
01±
0.09
0.00
±0.
060.
00±
0.06
0.05
±0.
100.
08±
0.14
0.06
±0.
12O
03c
20.
21±
0.06
0.04
±0.
200.
13±
0.15
-0.1
0±
0.21
0.07
±0.
14-0
.07
±0.
140.
07±
0.28
0.04
±0.
200.
06±
0.21
P09c
30.
18±
0.12
0.05
±0.
120.
13±
0.12
-0.0
8±
0.04
0.05
±0.
03-0
.05
±0.
030.
08±
0.13
0.05
±0.
120.
08±
0.12
S04
50.
02±
0.07
0.01
±0.
050.
01±
0.05
-0.0
1±
0.04
0.01
±0.
03-0
.01
±0.
030.
02±
0.07
0.01
±0.
050.
01±
0.05
Mea
n N
IST1
29c
0.02
±0.
03M
ean
0.03
±0.
080.
02±
0.08
0.03
±0.
08 (n
=8, ±
95%
con
f.)(2
SD, n
=8)
a Nor
mal
ized
to 1
86W
/184 W
= 0
.927
67 (6
/4) o
r 186 W
/183 W
= 1
.985
9 (6
/3) u
sing
the
expo
nent
ial l
aw.
b H
igh
sulp
hur s
teel
(NIS
T129
c) d
oped
with
add
ition
al W
to m
atch
con
cent
ratio
ns o
f sam
ples
. a
nd su
bseq
uent
ly p
roce
ssed
thro
ugh
full
chem
ical
sepa
ratio
n.c
Affe
cted
by
and
corre
cted
for m
ass-
inde
pend
ent e
ffect
(Sec
tion
4.2.
2).
Part A Chapter 5 91
Chapter 5
Accretion and metal segregation timescales of protoplanets from small 182W variations among iron
meteorites
T.S. Kruijer1,2, M. Touboul3, M. Fischer-Gödde1, K.R. Bermingham3, T. Kleine1, R.J. Walker3
1Westfälische Wilhelms-Universität Münster, Institut für Planetologie, Münster, Germany. 2ETH Zürich, Inst. of Geochemistry and Petrology, Zürich, Switzerland.
3Dept. of Geology, University of Maryland, College Park, Department of Geology, USA
A version of this chapter will be submitted to a peer-reviewed scientific journal
Abstract Dating planetary core formation in iron meteorite parent bodies using 182Hf-182W chronometry is possible now that superimposed cosmic ray effects on W isotopes can accurately be quantified through combined Pt and W isotope analyses. Here we report 5-20 parts-per-million (ppm) isotopic heterogeneities in 182W between different magmatic iron meteorite groups, which appear to be correlated with abundances of volatile elements including S, Ga and Ge. We interpret the heterogeneities in 182W to reflect variations in the average melting temperatures of the metal, which are controlled by the S contents of the metal cores. Irrespective of differences in the time of core formation, all five parent bodies accreted about concurrently, most likely between ~0.3-0.7 million years (Myr) after formation of Ca-Al-rich inclusions (CAI). In addition, the higher 182W/184W of the IVB and IID irons in part reflect that their precursor material derived from chemically distinct nebular or mantle material with suprachondritic Hf/W. Our dataset provides strong evidence that variable depletions of moderately volatile elements reflect spatial heterogeneities between different feeding zones in the solar protoplanetary disk that contributed material to the iron meteorite parent bodies. Furthermore, the Hf-W results are fully consistent with radioactive decay of 26Al to have been the principle heat source for differentiation of early-accreted planetary bodies.
92 Resolved metal segregation ages of magmatic iron meteorites
5.1 Introduction Magmatic iron meteorites are fragments of the solid metal cores of differentiated protoplanetary bodies that formed after metal-silicate separation and subsequent crystallization of metallic melt (Scott, 1972; Scott and Wasson, 1975). The different iron meteorite groups, defined by chemical characteristics, each represent metal from a distinct parent body. Extinct and extant radiogenic isotope chronometers indicate that metal segregation, core crystallization and cooling of iron meteorite parent bodies occurred within the first few to the first few tens of millions of years (Myr) of solar system evolution (Chen and Wasserburg, 1990; Smoliar et al., 1996; Cook et al., 2004; Kleine et al., 2005; Blichert-Toft et al., 2010). However, the precise time of parent body accretion as well as potential differences in core formation time scales among the different iron meteorite parent bodies have yet not been resolved. Obtaining such age information would be critical, however, to better understand the mechanisms of - and relation between - protoplanet accretion and core formation. For instance, whether accretion of protoplanets occurred rapidly or over a longer time interval (several Myr) is largely unknown. Another aspect yet to be assessed is a possible presence or absence of a relation between the accretion time and degree of volatile element depletion (Scott and Wasson, 1975; Davis, 2006) of iron meteorite parent bodies.
The extinct 182Hf-182W chronometer with t1/2 = 8.9 Myr (Vockenhuber et al., 2004) is ideally suited to constrain the accretion and differentiation timescales of planetary bodies (Kleine et al., 2009), but the application of Hf-W chronometry to precisely date iron meteorites is hampered by cosmic ray-induced neutron capture effects on W isotope compositions (Horan et al., 1998; Kleine et al., 2005; Markowski et al., 2006; Scherstén et al., 2006; Qin et al., 2008; Kruijer et al., 2012), making calculated core formation ages uncertain by as much as ~10 Myr in some cases. Quantification of neutron capture effects on W isotopes, and corrections for these effects, however, is possible now that it has been shown that Pt isotopes provide a reliable and direct neutron fluence dosimeter for iron meteorites (Kruijer et al., 2013a; Wittig et al., 2013). Combined Pt and W isotope analyses on samples from a particular iron meteorite group provide an empirical neutron capture-induced Pt-W isotope correlation line, whose intercept represents the pre-exposure !182W (where !182W is the deviation of 182W/184W in 0.01% relative to terrestrial reference standards) for the respective parent body, that is, the radiogenic 182W signature that is unbiased by galactic cosmic rays. We here obtained combined high-precision Pt and W isotope data for metal samples from the major groups of iron meteorites (IIAB, IIIAB, IVA, IVB, IID).
Part A Chapter 5 93
5.2 Results: Pt and W isotope compositions The analytical methods for Pt and W isotope analyses by MC-ICPMS (Pt and W) and N-TIMS (W) are described in the Appendix (Section 5.5.1). The Pt isotope compositions for the irons from the major groups, as determined by MC-ICPMS, are displayed in Fig. 5.1 and reported in Table 5-A3. The samples are characterized by variable excesses in !192Pt (up to ~61.3), !196Pt (up to ~1.04), !194Pt (up to ~1.9) (internally normalised to 198Pt/195Pt, denoted ‘8/5’). Variations among irons of a single group define linear trends, consistent with progressive, correlated effects among the different isotopes of Pt. The W isotope data obtained by MC-ICPMS for the IIAB and IIIAB irons are plotted as !182W vs. !184W (6/3) and !182W vs. !183W (6/4) in Fig. 5.2, and reported in Table 5-A4. The W isotope data obtained by TIMS are reported in Table 5-A5. Large within-group variations in !182W deficits are observed for both the IIAB irons (from ~ "3.4 to "4.8) and IIIAB irons (from ~ "3.3 to "4.0), in an extreme case extending down to !182W ~ "4.82 for Ainsworth (IIAB; Fig. 5.2). The !184W and !183W of the newly investigated IIAB and IIIAB irons are all indistinguishable from the terrestrial W isotope composition after correcting some of the samples for a small mass-independent effect on 183W (see Section 5.5.1.3.3). The W isotope data obtained by TIMS for several IIAB, IIIAB, IVA and IVB irons agree well with the results obtained by MC-ICPMS (Table 5-A4-A5).
Fig. 5.1: Platinum isotope compositions of iron meteorites investigated in this study. (A) !192Pt (8/5) vs. !196Pt (8/5), and (B) !194Pt (8/5) vs. !194Pt (6/5). Data for IID and (most) IVB irons are from Kruijer et al. (2013). Error bars indicate external uncertainties (2", s.d.), estimated from repeated analyses of several digestion replicates of a terrestrial metal standard (NIST129c; see Table 5-A1). Small white circles designate model results for neutron capture in iron meteoroids for different Ir/Pt (Kruijer et al., 2013; Leya and Masarik, 2013). Also plotted is the correlation line predicted for variability in s- and/or r-process Pt isotopes from Arlandini et al. (1999).
0 0.2 0.4 0.6 0.8 1.0 1.2
0
10
20
30
40
50
60
ε196Pt (8/5)
ε192
Pt (8
/5)
Ir/Pt = 0.005
Ir/Pt = 0.1
Ir/Pt =
1.0
IVB IVA
IIIAB
IID
IIAB
Tlacotepec(Ir/Pt = 1.0)
Ainsworth (Ir/Pt = 0.005)
(B)
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
0
0.5
1.0
1.5
2.0
2.5
ε194Pt (6/5)
ε194
Pt (8
/5)
Ir/Pt = 0.005
Ir/Pt =
1.0
Ir/Pt = 0.5
s-excess / r-deficitAinsworth (Ir/Pt = 0.005)
Tlacotepec(Ir/Pt = 1.0)
(A)
Ir/Pt = 0.65
Carbo(Ir/Pt = 0.65)
94 Resolved metal segregation ages of magmatic iron meteorites
5.3 Discussion
5.3.1 Pt isotope systematics Platinum isotope anomalies in iron meteorites are caused by secondary neutron capture (Kruijer et al., 2013a; Wittig et al., 2013). Since 195Pt captures thermal, epithermal and faster neutrons more efficiently than 196Pt and 198Pt (e.g.,ENDFB-VI.8 300K library), the reaction 195Pt(n,!)196Pt dominates neutron capture-induced effects on mass bias corrected 196Pt/195Pt ratios, i.e., internally normalized to 198Pt/195Pt and expressed as !196Pt (8/5) (Kruijer et al., 2013a). Hence, all investigated iron meteorites exhibit excesses in !196Pt (Fig. 5.1). Both 191Ir and 193Ir have large resonance integrals (e.g.,ENDFB-VI.8 300K library) and thus efficiently capture neutrons at epithermal energies. The reactions 191Ir(n,!)192Ir("-)192Pt and 193Ir(n,!)194Ir("-)194Pt can thus generate anomalies on 192Pt and 194Pt, whose magnitude depends on the Ir/Pt ratio (Kruijer et al., 2013a). Variable !192Pt anomalies are observed for the IID, IIIAB, IVA and IVB irons, where the magnitude for each sample anomaly is consistent with
Fig. 5 .2: Tungsten isotope compositions for the IIAB and IIIAB iron meteorites obtained using ICPMS (or TIMS) in this study, in case of some samples after correction for a mass-independent effect on 183W (Section 5.5.1.3.3). (A) !182W vs. !184W internally normalised to 186W/183W (denoted ‘6/3’), and (B) !182W (6/4) vs. !183W (denoted 6/4). Error bars indicate external uncertainties, represented by 95% conf. limits of the mean of multiple solution replicates. Small white circles denote modelled neutron capture effects on W isotopes in iron meteoroids (Leya and Masarik, 2013). The solid lines show, for illustration, approximate correlations predicted for (i) variability in s- and/or r-process W isotopes (Arlandini et al., 1999), and (ii) a deficit in 183W as described in Kruijer et al. (2012), and are plotted at an ordinate value of !182W = #3.32, which is identical to the pre-exposure !182W previously determined for IVA meteorites.
–0.4 –0.2 0 0.2 0.4–5.0
–4.5
–4.0
–3.5
–3.0
ε182
W (6
/3)
ε184W (6/3)
n-capture
s-deficit/r-excess
Residual 1
83 W deficit
IIIAB IIAB
(A)
–0.4 –0.2 0 0.2 0.4–5.0
–4.5
–4.0
–3.5
–3.0
ε182
W (6
/4)
ε183W (6/4)
n-capture
s-deficit/r-excess
Residual 183W deficit
IIIAB IIAB
(B)
Ainsworth
Ainsworth
Part A Chapter 5 95
their variable !196Pt and Ir/Pt (Kruijer et al., 2013a), as shown here for two extremes (Ainsworth, Ir/Pt 0.005 and Tlacotepec, Ir/Pt =1). In contrast, due to the relatively low Ir/Pt, only relatively small excesses in !192Pt are observed for most IIAB irons, despite very large !196Pt excesses for some samples (e.g.,~1.04 for Ainsworth). This observation illustrates that, although the !192Pt anomalies are generally larger, the !196Pt anomalies are most suitable for quantifying neutron capture effects, as they are independent of the Ir/Pt (Kruijer et al., 2013a).
Also plotted in Fig 5.1 are the results for model calculations for neutron capture on Pt isotopes in iron meteorites (Kruijer et al., 2013a). Each linear array (small open circles) represents the model results of !iPt for a specific Ir/Pt ratio and for relevant shielding depths in iron meteoroids with pre-atmospheric radii between 10 and 100 cm. The measured Pt isotope anomalies for iron meteorites plot on distinct model trends for different Ir/Pt, in each case consistent with the measured Ir/Pt of the respective meteorite (i.e, as shown for Ainsworth and Tlacotepec). The very good agreement of the iron meteorite data with the modelled secondary neutron capture effects confirms that the Pt isotope variations are cosmogenic in origin (Kruijer et al., 2013a). Conversely, a heterogeneous distribution of s- and r-process Pt isotopes would produce a distinctly different trend in !194Pt (8/5) vs. !194Pt (6/5) space to the well-defined correlation shown by the samples (Fig. 5.1B), precluding an interpretation of a nucleosynthetic origin. Instead, and confirming earlier conclusions (Kruijer et al., 2013a; Wittig et al., 2013), the Pt isotope anomalies are purely cosmogenic in origin, and thus, can be used to quantify neutron capture effects in iron meteorites.
5.3.2 W isotope systematics of iron meteorites Tungsten isotope anomalies in meteorites may originate from (i) radioactive decay of 182Hf (t1/2 = 8.9 Myr), (ii) neutron capture effects induced during cosmic ray exposure (Masarik, 1997), or (iii) nucleosynthetic variations which reflect a heterogeneous distribution of presolar s- and or r- process components (Qin et al., 2008; Burkhardt et al., 2012; Kruijer et al., 2013a). While radiogenic and cosmogenic effects can only generate sizeable !182W variations, a heterogeneous distribution of s- and r- process W isotopes yields coupled !182W and !183W anomalies. Nucleosynthetic W isotope anomalies can therefore be distinguished by their non-terrestrial excesses in !183W and disparate !182W (6/3) and !182W (6/4) anomalies (Qin et al., 2008; Burkhardt et al., 2012; Kruijer et al., 2012) Reliable chronological interpretation of !182W data requires careful quantification of !183W anomalies in order to correct for these effects because even small nucleosynthetic W isotope anomalies can have a significant effect on !182W (6/4) anomalies (Burkhardt et al., 2012; Kruijer et al., 2013b),
Small nucleosynthetic W isotope anomalies, as evident from !184W deficits, have previously been identified for the IVB irons (!184W = "0.09±0.01; (Qin et al., 2008; Kruijer et al., 2013a)) and the IID irons (!183W = "0.09±0.04; (Kruijer et al., 2013a)). The effect of nucleosynthetic heterogeneity on !182W can be quantified using the measured !183W and the linear relation between !182W vs. !183W. Until recently, this relationship could only be constrained using the stellar model for s-process nucleosynthesis from Arlandini et al. (1999), resulting in a !182W vs. !183W slope of ~1.686 (Burkhardt et al., 2012), which was used to correct the !182W values of the IID and IVB irons in Kruijer et al. (2013a). However, the !182W vs. !183W relationship was recently empirically quantified using Hf-W data for bulk CAI (Kruijer et al., 2013b), which yields a !182W-!183W (6/4) slope of +1.41±0.05, i.e., slightly shallower than the slopes predicted by Arlandini et al. (1999). Similarly, a very shallow !182W vs. !184W (6/3) slope of "0.10±0.07 was obtained for the normalisation to 186W/183W. For this reason, correcting the measured !182W data of the IID and IVB irons using
96 Resolved metal segregation ages of magmatic iron meteorites
the newly determined !182W vs. !183W (6/4) and !182W vs. !184W (6/4) slopes yields slightly more elevated 182W (Table 5-A4) than the values reported in Kruijer et al. (2013a). As the N-TIMS measurements require a double normalization procedure to correct for instrumental mass bias, no precise !183W data can be obtained in such measurements. Therefore, the presence of nucleosynthetic W isotope anomalies, as evidenced by non-terrestrial !183W, could not be assessed or quantified for the three newly analysed IVB irons. Nevertheless, both the !182W (6/4) and the !182W (6/3) anomalies, that is, the normalizations most strongly and least affected by nucleosynthetic W heterogeneity, show very good agreement with each other and with the results previously obtained by MC-ICPMS (Tables 5-A4-A5). This likely reflects that the nucleosynthetic W isotope anomalies are cancelled out by the double normalization procedure used for N-TIMS.
In contrast to the IID and IVB irons, neither the IIAB irons nor the IIIAB irons exhibit clearly resolved !183W anomalies (Fig. 5.2, Table 5-A4). In addition, most investigated irons have identical !182W (6/3) and !182W (6/4). A minority of the investigated samples and the terrestrial metal standard (NIST129c) show slight, barely resolved deficits in !183W (mostly <0.1 !-unit), and correspondingly different !182W (6/3) and !182W (6/4) (Table 5-A4). Such !182W-!183W systematics is explained by a mass-independent effect on 183W, likely induced during sample preparation, which can accurately be corrected for (Section 5.5.1.3.3). After correction for the mass-independent effect, the !182W (6/3) and !182W (6/4) values are indistinguishable for all investigated iron meteorite samples (Table 5-A4, Fig. 5.2). Thus, the IIAB, IIIAB irons [and the IVA irons, see Kruijer et al. (2013a)] are characterized by terrestrial !183W, and in contrast to the IID and IVB irons (Kruijer et al., 2013a), do not show evidence for nucleosynthetic W isotope variations. This implies that variability in !182W shown by the IIAB, IIIAB, and IVA irons results from radiogenic and cosmogenic effects alone.
5.3.3 Combined Pt-W isotope systematics The combined Pt-W isotope results for all major groups of iron meteorites are presented in Fig. 5.3. The IIAB irons display a well-defined correlation in !182W vs. !196Pt space (Fig. 5.3A) with negligible external scatter (MSWD = 0.80). A linear regression through the data yields a very precise intercept ‘pre-exposure’ !182W of "3.40±0.03 (95% conf., n=8). Ainsworth (IIAB) shows very low !182W (6/4) of "4.82±0.05, which is the largest !182W deficit obtained for any meteorite sample to date. The data point for Ainsworth lies very well on the regression defined by the other data points for IIAB irons. Note also that the W isotope data obtained by N-TIMS and by MC-ICPMS plot on the same correlation line, which demonstrates that both methods yield consistent results. The !182W vs. !196Pt correlation for the IIIAB irons also define a precise and well-defined empirical !182W vs. !196Pt correlation, although this exhibits slightly more scatter (MSWD = 1.4) and a more limited spread than the correlation obtained for the IIAB irons (Fig 5.3B). Nevertheless, the IIIAB irons define a precise intercept pre-exposure !182W of "3.35±0.03 (95% conf., n=13).
Combined Pt and W isotope data for two new IVA samples (Jamestown and Muonionalusta) supplement the dataset obtained previously (Table 5-A3-A4). Although the spread in Pt and W isotope anomalies is still limited, Jamestown shows a slightly larger !196Pt anomaly of ~0.22. Consequently, the six IVA irons exhibit a narrow but well-defined !182W-!196Pt correlation (MSWD = 0.19), yielding a pre-exposure !182W of "3.32±0.05 (95% conf.) (Fig. 5.3C).
Part A Chapter 5 97
The previously obtained IVB dataset from Kruijer et al. (2013a) was supplemented with three newly analysed IVB samples (from Hoba, Warburton Range and Tlacotepec) determined using N-TIMS (Table 5-A5). Their !182W and !196Pt plot on the regression line previously obtained (Fig. 5.3D), confirming the results previously obtained by (Kruijer et al., 2013a). Due to the newly established empirical !182W- !183W relation defined by nucleosynthetic W isotope anomalies in CAI (Kruijer et al., 2013b), the !182W of the IID and IVB irons corrected for nucleosynthetic W isotope heterogeneity are slightly more elevated than those reported in Kruijer et al. (2013a). Accordingly, regression of the newly corrected !182W vs. !196Pt values of the IVB irons yields pre-exposure !182W of "3.22±0.06 (95% conf. n=11), that is, slightly higher than the value of "3.26±0.06 reported in Kruijer et al. (2013a). Regression of all Pt-W isotope data obtained for the IVB irons, yields a revised pre-exposure !182W of "3.17±0.05 (95% conf. n=14) in excellent agreement with the value of "3.22±0.06 that is obtained when only MC-ICPMS data are used (Fig. 5.3D). Similarly, regressions of the !182W-!196Pt and !182W-!192Pt relationships of the IID irons results in revised pre-exposure !182W of "3.18±0.11 (Fig. 5.3E) and "3.15±0.07 (Fig. 5.3F) that are identical to each other, and slightly higher than the values of "3.22±0.13 and "3.20±0.10 reported in Kruijer et al. (2013a).
The regression-derived slopes of the !182W-!196Pt correlations obtained for the different iron meteorite groups are similar and identical within uncertainty (Fig 5.3, 5.4). The IVB irons previously seemed to plot on a line with a slightly shallower slope (Kruijer et al., 2013a), but with the new IVB data the slopes becomes comparable to that obtained for the other iron meteorite groups (Fig. 5.4). This observation highlights the consistency of the combined Pt and W isotope data obtained for different iron meteorite groups.
Finally, we note that the model results from Leya and Masarik (2013) for neutron capture effects on Pt and W isotopes are in good agreement with the empirical Pt-W isotope correlations obtained for all iron meteorite groups (Fig. 5.3), as previously demonstrated for the IID and IVB irons (Kruijer et al., 2013a). However, detailed examination reveals that the !182W-!196Pt slope predicted by the neutron capture model is slightly steeper than that shown by the data (Fig. 5.4). The reason for the disparity between the model and the data remains unclear (Leya and Masarik, 2013). Nevertheless, we emphasize that the general consistency between the model and the data should be considered very good, especially given that the uncertainties of the model calculations have not been taken into account.
98 Resolved metal segregation ages of magmatic iron meteorites
–0.1 0 0.1 0.2 0.3 0.4 0.5 0.6–4.1
–3.9
–3.7
–3.5
–3.3
–3.1
0 0.2 0.4 0.6 0.8 1.0–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
0 0.2 0.4 0.6 0.8 1.0–4.4
–4.2
–4.0
–3.8
–3.6
–3.4
–3.2
0 10 20 30 40–4.5
–4.2
–3.9
–3.6
–3.3
–3.0
0 0.2 0.4 0.6 0.8 1.0 1.2
–4.8
–4.4
–4.0
–3.6
–3.2
IVB iron meteorites: ε182Wpre-exposure= −3.17±0.05
slope = −1.25±0.12 (95% conf.) MSWD = 1.0, n = 14
IIAB iron meteorites:ε182Wpre-exposure = −3.40±0.03
slope = −1.40±0.11 (95% conf.)MSWD = 0.80, n = 8
IID iron meteorites: ε182Wpre-exposure = −3.18±0.11
slope = −1.40±0.17 (95% conf.) MSWD = 0.36, n = 8
ε196Pt (8/5)
ε182
W (6
/4)
Ainsworth
Sikhote AlinMt. Joy
(A)
Tlacotepec
ε196Pt (8/5)
ε182
W (6
/4)
Rodeo
Carbo
ε196Pt (8/5)
ε182
W (6
/4)
(F)
(D)
IIIAB iron meteorites:ε182Wpre-exposure = −3.35±0.03
slope = −1.24±0.15 (95% conf.)MSWD = 1.4, n = 13
ε196Pt (8/5)
ε182
W (6
/4)
(E)
(B)
IID iron meteorites: ε182Wpre-exposure = −3.15±0.07
slope = −0.031±0.003 (95% conf.) MSWD = 0.35, n = 8
ε192Pt (8/5)
ε182
W (6
/4)
Carbo
Rodeo
IVA iron meteorites:ε182Wpre-exposure = −3.32±0.05
slope = −1.18±0.54 (95% conf.)MSWD = 0.19, n = 6
–0.1 0 0.1 0.2 0.3–3.7
–3.6
–3.5
–3.4
–3.3
–3.2(C)
ε196Pt (8/5)
ε182
W (6
/4)
Jamestown
Gibeon
Muonionalusta
Gibeon
Fig. 5.3. !182W vs. !iPt for the major magmatic iron meteorite groups. (A) IIAB irons, (B) IIIAB irons, (C) IVA irons, (D) IVB irons, and (E+F) IID irons. !182W (6/4) and !iPt (8/5) are 0.01% deviations from the terrestrial 182W/184W ratios (internally normalized to 186W/184W, denoted ‘6/4’) and iPt/195Pt ratios (normalized to 198Pt/195Pt, denoted ‘8/5’). Shown as solid lines are best-fit regressions (York, 1966) through the data with their 95% conf. envelopes (dashed lines) and pre-exposure !182W intersecting the ordinate at !iPt = 0. Tungsten isotope analyses performed on MC-ICPMS are shown as closed symbols, and those on TIMS as open symbols. Error bars on symbols represent external uncertainties (2" s.d. for Pt and 95% conf. for W). Data for IID and IVB irons are from Kruijer et al. (2013a). Model predictions for neutron capture effects on Pt and W isotopes in iron meteoroids are shown as thin dashed lines (anchored to the ordinate intercepts of the regressions through the measured data; in Fig. 5.3F shown for an Ir/Pt of 0.65).
Part A Chapter 5 99
5.3.4 Interpretation of variable pre-exposure !182W
5.3.4.1 Summary of pre-exposure !182W values The Pt and W isotope anomalies for all iron meteorite groups exhibit well-defined, empirical !182W vs. !196Pt correlations (Fig. 5.3), which provide strong supportive evidence for a cosmogenic origin of the variable 196Pt and 182W isotope anomalies within each particular group. The intercept values provide pre-exposure !182W values for each of the investigated iron meteorite groups and are summarized in Fig. 5.5 and Table 5.1. Our results demonstrate the presence of small, ~5-20 ppm differences in pre-exposure !182W between different groups of magmatic iron meteorites with the IVB irons having a pre-exposure !182W of "3.17±0.05 (±95% conf.) that is distinctly higher than the value of "3.40±0.03 obtained for the IIAB irons. Model ages of metal segregation in the iron meteorite parent bodies relative to the formation of CAI can be calculated as the time of Hf/W fractionation from an unfractionated reservoir with chondritic Hf/W of ~1.23 (Kleine et al., 2004). At face value, the distinct pre-exposure !182W yield resolved and very precise Hf-W ages spanning a total range of ~1 to ~3.5 Myr after CAI formation (Fig. 5.5).
Table 5.1
Summary of !182W vs. !196Pt- regressions for the major iron meteorite groups.!182Wpre-exposure (intercept) !182W vs. !196Pt slope MSWD N
Group (±95% conf.) (±95% conf.)
IIAB irons "3.40 ± 0.03 "1.40 ± 0.11 0.80 8IID ironsa "3.16 ± 0.07 "1.40 ± 0.17 0.36 8IIIAB irons "3.35 ± 0.03 "1.24 ± 0.15 1.4 13IVA irons "3.32 ± 0.05 "1.18 ± 0.54 0.19 6IVB irons "3.17 ± 0.05 "1.25 ± 0.12 1.0 14
a Pre-exposure !182W of IID irons represents mean from !182W vs. !192Pt and !182W vs. !196Pt regressions.
Fig. 5.4. Summary of the regression derived !182W vs. !196Pt slopes for the major iron meteorite groups. The solid grey line shows average slope for the investigated iron meteorite group and the hatched area the associated uncertainties (2s.d.). The red dashed line shows the slope predicted by the neutron capture model from Leya and Masarik (2013).
slop
e (ε
182 W
vs.
ε19
6 Pt)
IIAB IID
IVAIVBIIIAB
neutron capture model
–2.0
–1.6
–1.2
–0.8
–0.4
100 Resolved metal segregation ages of magmatic iron meteorites
5.3.4.2 Observed correlations of !182W with volatile elements Ga, Ge, and S The pre-exposure !182W exhibit inverse correlations with the initial Ga/Ni ratios (and Ge/Ni) and S concentrations inferred for the bulk cores of the investigated iron meteorite groups (Fig. 5.6). The abundances of the moderately volatile elements Ga and Ge [with 50% condensation temperatures, T50%, of 968 K and 883 K] in iron meteorites are primarily governed by volatility related processes, and relatively unaffected by fractional crystallization (Scott and Wasson, 1975; Davis, 2006). As such, they are commonly used as an indicator for the degree of volatile element depletion for each parent body. While within-group variations are relatively small, CI-normalized Ge and Ga concentrations show strong variations between different iron meteorite groups (Scott and Wasson, 1975). For example, the IVB irons are most strongly depleted in volatile elements with (Ga/Ni)CI ~ 0.001, while the IIAB parent body has near-chondritic Ga/Ni. The broad inverse ε182W vs. (Ga/Ni)CI correlation, therefore, suggests that the degree of volatile element depletion exerted some control on the time scale of core formation in iron meteorite parent bodies. Sulphur also is a volatile element (T50% = 664 K) showing large variations among the iron meteorite groups (e.g.,Wasson, 1999;
Fig. 5.5: Pre-exposure !182W for the major iron meteorite groups and their associated two-stage Hf-W model ages, assuming 180Hf/184W = 1.23 for a chondritic reservoir (Kleine et al., 2004). Data for the IID and IVB irons are (in part) from Kruijer et al. (2013a) and the CAI initial is from Burkhardt et al. (2012) and Kruijer et al. (2013b). Model ages of metal segregation in the iron meteorite parent bodies relative to the formation of CAI were calculated as the time of Hf-W fractionation from an unfractionated reservoir with chondritic Hf/W using the relation:
where 182W/184Wiron is the pre-exposure composition of an iron meteorite group, 182W/184WCAI and 182Hf/180HfCAI are the initial ratios of CAI (Kruijer et al., 2013b) [i.e., converted ratios from !-unit notation], 180Hf/184W is the average value of ~1.23±0.15 obtained for carbonaceous chondrites (Kleine et al., 2004), and $ is the 182Hf decay constant of 0.078±0.002 Myr-1 (Vockenhuber et al., 2004).
ε182Wpre-exposure
IID
IVB
IVA
CAI
initi
al
IIIAB
IIAB
ΔtCAI [Myr]0 1 2 3 4 5−1
–3.6 –3.5 –3.4 –3.3 –3.2 –3.1 –3
Part A Chapter 5 101
Chabot, 2004; Wasson and Huber, 2006; McCoy et al., 2011). Because of its low solubility in solid Fe-Ni metal, the S abundances of the parent body cores cannot be directly measured in iron meteorites. Instead, the S concentrations of the cores are calculated either using experimental solid/melt partition coefficients and modelled fractional crystallisation trends within magmatic iron meteorite groups (e.g.,Chabot, 2004), or by modal analysis of large samples (e.g.,Wasson, 1999). Consequently, the uncertainties on these estimates are relatively large. The S concentrations adopted here (Chabot, 2004; Wasson and Huber, 2006) range from ~17 wt% for the IIAB irons, to about zero for the IVB irons. Although estimates from the literature vary considerably, the order of inferred S abundances is generally consistent for the different iron meteorite groups (e.g.,Haack and McCoy, 2003; Chabot, 2004). The IID irons are the only group not fitting the !182W vs. Ga/Ni correlation, and also plot slightly off-trend in the !182W vs. S diagram (Fig. 5.6B). The latter may reflect the inherently large uncertainties on the inferred S concentrations of the metal cores.
5.3.4.3 Distinct melting temperatures for iron meteorite parent bodies and interpretation of !182W vs. S correlation
The S concentration of Fe-Ni metal strongly affects its melting (liquidus) temperature (e.g.,Fei et al., 1997; Stewart et al., 2007), which can range between the Fe-FeS eutectic up to the melting temperature of pure Fe. Estimates of the melting temperatures of Fe(Ni)-S metal can be obtained by plotting the inferred initial S concentrations of the iron meteorite cores in the Fe-FeS phase diagram (e.g.,Fei et al., 1997). At atmospheric pressure (1 ATM) the inferred melting temperatures of the metal cores of the iron meteorite parent bodies, obtained using the estimated S contents from Chabot (2004), range from ~1330 °C for the IIAB irons to ~1600 °C for the IVB irons. At higher pressures, which are more appropriate for larger bodies of ~1000 km in size, the associated higher internal pressures would result in a larger range of inferred melting temperatures for the same range in S concentrations (Fei et al., 1997) with the inferred melting temperatures ranging from 1300 °C (IIABs) to 1820 °C (IVBs). After accretion the (still) undifferentiated iron meteorite parent bodies likely consisted of an un-equilibrated mix of metallic Fe, FeS and silicate components. Upon heating, the high S components (FeS) would likely melt first and relatively close to the
Fig. 5.6: (A) Pre-exposure !182W vs. CI-normalized Ga/Ni, (B) Pre-exposure !182W vs. the inferred S concentration (wt%) in the core. Error bars on !182W represent 95% confidence limits of the mean. The Ge, Ga and Ni concentrations are from (Haack and McCoy, 2003) and references therein, and S concentrations of IIAB, IIIAB, IVA, and IVB irons from (Chabot, 2004), and for the IID irons from Wasson and Huber (2006).
0.0010.010.11–3.6
–3.5
–3.4
–3.3
–3.2
–3.1
0 5 10 15 20–3.6
–3.5
–3.4
–3.3
–3.2
–3.1
S [wt%]
ε182
Wpr
e-ex
posu
re
IIAB
IVA
IVB
IIIAB
CAI initial
IID
(A) (B)
(Ga/Ni)CI norm.
ε182
Wpr
e-ex
posu
re
IIAB
IVA
IVB
IIIAB
CAI initial
IID
102 Resolved metal segregation ages of magmatic iron meteorites
eutectic temperature of ~1260 K determined for the Fe-FeS system (Fei et al., 1997), i.e., below the silicate solidus. The FeS melts would likely almost immediately segregate to the core of the parent body (Yoshino et al., 2003). The silicate mineral phases would be the next to melt, followed by pure metallic Fe at ~1850 K (Fei et al., 1997). The differences in pre-exposure !182W among the different iron meteorite groups, and the inverse !182W vs. S correlation, could then be explained by varying proportions of early (having low !182W) and late (having higher !182W) segregated metal fractions in the core.
5.3.4.4 Thermal model for post-accretional heating of iron meteorite parent bodies We tested the qualitative interpretation presented above using a simple thermal model for post-accretional heating of planetesimals by 26Al decay (Carslaw and Jaeger, 1959; Miyamoto et al., 1981; Kleine et al., 2008; Touboul et al., 2009) (Fig. 5.7). The thermal model assumes instantaneous accretion and calculates the temperature-time-depth relations for an undifferentiated spherical body that is heated by 26Al decay. The model accounts for heat production by radioactive decay and for heat loss by conduction and diffusion. The thermal model presented here is an oversimplification, because it does neither include the effects of an insulating regolith nor local and temporal variations in thermal parameters due to sintering. The presence of an insulating regolith would cause more rapid heating and, hence, earlier melting, but this effect is only significant for small parent body sizes (Akridge et al., 1998). The main effect of sintering is that melting can occur in small, km-sized bodies (not possible for compact bodies) and that cooling is more protracted (Henke et al., 2012). The effects of sintering on the initial heating of the body are not large, however. Thus, to a first order, the model used here helps to assess whether the distinct metal segregation ages reflect different metal melting temperatures within the iron meteorite parent bodies.
5.3.4.5 Distinct times of metal-segregation and concurrent accretion of iron meteorite parent bodies
The radii of iron meteorite parent bodies are not well constrained and current estimates range from ~3 to ~1000 km (e.g.,Haack et al., 1990; Chabot and Haack, 2006; Yang et al., 2007; Yang et al., 2008). We, therefore, performed calculations using parent body radii of 40 and 1000 km. In both cases the resulting heating curves are very similar, but the difference in inferred melting temperatures for a given S content of the metal cores becomes significantly larger with increasing parent body size and, hence, higher pressure (see above) (Fig. 5.7B,D). In general, the model results show better agreement with the observed differences in core formation ages for the IIAB, IIIAB and IVA iron meteorite parent bodies for a larger range in melting temperatures and, hence, for larger parent body sizes. However, regardless of parent body size, the model results show that the observed time difference in core formation in the IIAB and IVA parent bodies of ~1 Myr is consistent with the time required to raise the temperature in the middle of the parent body from the melting temperature of the IIAB to that inferred for the IVA irons (Fig. 5.7A,B). Thus, the variable #182W of the IIAB, IIIAB, IVA irons most likely reflect different core segregation times as a result of the slightly different melting temperatures of the parent bodies. Our model predicts that in spite of their different core formation ages these three iron meteorite parent bodies all accreted at about the same time, between ~0.3 and ~0.7 Myr after formation of CAI.
Part A Chapter 5 103
5.3.4.6 Suprachondritic Hf/W for precursor material of IID and IVB parent bodies In contrast, the IID and IVB irons plot to the right of the heating curves for accretion times that are consistent with core formation ages obtained for the IIAB, IIIAB and IVA irons (Fig. 5.7A,B). Although it cannot be excluded a priori that accretion and metal segregation in the IVB and IID iron meteorite parent bodies occurred later than in the other bodies, several lines of evidence suggest an alternative explanation, namely that the precursor material of the IVB
Fig. 5.7: Temperature-time diagrams showing model results for post-accretional internal heating by 26Al decay of spherical protoplanets with radii of 40 km (a,c) and 1000 km (b,d). The two solid curves show the temperature evolution at half the radius (i.e, at 20 or 500 km depth) in planetesimals accreted at 0.3 and 0.5 Myr after CAI formation and for an Al concentration of 0.87 wt % as determined for CI chondrites (Lodders and Fegley, 1998). The dashed curves show the temperature evolution at half the radius in planetesimals accreted at 0.5, 0.7, and 0.9 Myr after CAI formation and for an Al concentration of 1.7 wt % inferred for CV chondrites (Lodders and Fegley, 1998). The parameters used for the model are: thermal conductivity K = 2.1 Wm-1K-1 (Hevey and Sanders, 2006), thermal diffusivity ! = 5.0%10-7 m2s-1, solar system initial 26Al/27Al = 5%10-
5, Al concentration in CI chondrites = 8.6 wt%, heat production A = 8.6%(26Al/27Al) Wm-3, decay constant of 26Al = 9.8%10-7 yr-1 (Norris et al., 1983), density of planetesimal " = 3.2 gcm-3, emissivity h = 1.0 m-1, the assumed ambient temperature T0 = 250 K (Hevey and Sanders, 2006). The 26Al/27Al at accretion times of 0.3, 0.5, 0.7, 0.9 Myr after CAI formation correspond to 3.72%10-5, 3.06%10-5, 2.51%10-5, and 2.06%10-5
respectively. Solid symbols show the model ages of metal segregation inferred for the iron meteorite parent bodies, which were calculated using the pre-exposure !182W from this study and assuming parent body accretion in an unfractionated reservoir with chondritic 180Hf/184W of 1.23±0.15 (Kleine et al., 2004). The average ‘melting’ temperatures for the different iron meteorite parent bodies – as inferred using estimates of the S concentrations in their cores (Chabot, 2004; Wasson and Huber, 2006) and the Fe-FeS phase diagram at two different pressures (1 ATM, 10 GPa) from Fei et al. (1997) – correspond to: IIAB (1330 °C, 1300 °C), IID (1450 °C, 1750 °C), IIIAB (1400 °C, 1550 °C), IVA (1450 °C, 1750 °C), IVB (1610 °C, 1820 °C).
(A)
(C)
0 1 2 3 40
500
1000
1500
2000
0 1 2 3 40
500
1000
1500
20000.3 Myr
∆tCAI (Myr)
T (°
C)
IIAB IIIABIVA
IVB
IID
r = 40 km
0.5 Myr
ε182Wpre-exposure−3.5 −3.4 −3.3 −3.2 −3.1
Fe-FeS eutectic
Pure Fe metal
0.3 Myr
∆tCAI (Myr)
T (°
C)
IIAB IIIABIVA
IVB (Hf/W = 1.9)
IID
r = 40 km
0.5 Myr
Fe-FeS eutectic
Pure Fe metal
suprachondritic Hf/W?
0.9
Myr
0.5
Myr
0.7
Myr
0.9
Myr
0.5
Myr
0.7
Myr
1.7 wt% Al
0.87 wt% Al
0 1 2 3 40
500
1000
1500
2000
0 1 2 3 40
500
1000
1500
2000
∆tCAI (Myr)
T (°
C)
IIAB
IIIAB
IVA
IVB
IID
ε182Wpre-exposure−3.5 −3.4 −3.3 −3.2 −3.1
Fe-FeS eutectic
Pure Fe metal
∆tCAI (Myr)
T (°
C)
IIAB
IIIAB
IVA
IVB (Hf/W = 1.9)
IID
r = 1000 km
Fe-FeS eutectic
Pure Fe metal
suprachondritic Hf/W?
r = 1000 km
(B)
(D)
0.3
Myr
0.5
Myr
0.3
Myr
0.5
Myr
0.5
Myr
0.7
Myr
0.9
Myr
0.5
Myr
0.7
Myr
0.9
Myr
104 Resolved metal segregation ages of magmatic iron meteorites
and IID irons had higher-than-chondritic Hf/W and, hence, evolved to higher pre-exposure !182W. The inferred bulk composition of the IVB parental melt is strongly fractionated, indicating substantial high temperature processing of the precursor material in the solar nebula (Campbell and Humayun, 2005; Walker et al., 2008). Most notably, relative to other refractory siderophile elements W is depleted, indicating either core formation under relatively oxidized conditions—where W becomes less siderophile—or sub-chondritic W abundances in the IVB parent body. Assuming a metal-silicate partition coefficient for W of ~100 (Cottrell et al., 2010)—which is appropriate for core formation at fO2 of IW = "1 as inferred for the IVB parent body (Campbell and Humayun, 2005)—and assuming that the IVB parent body had chondritic proportions of other refractory elements (Os, Hf), we infer by mass balance a Hf/W of ~1.9 for the bulk IVB parent body, significantly higher than the chondritic Hf/W of ~1.23 (Kleine et al., 2004). Using a Hf/W ratio of 1.9 for the IVB parent body provides a Hf-W model age of metal segregation of 2.0±0.6 Myr after CAI formation. This age is in good agreement with the co-variation of Hf-W age and melting temperatures obtained for the IIAB, IIIAB and IVA irons (Fig. 5.7C,D), strongly suggesting that the IVB iron meteorite parent body accreted at about the same time as those of the IIAB, IIIAB and IVA irons.
Wasson and Huber (2006) proposed that core formation in the IID iron meteorite parent body was a two-step process, including one segregation step of mostly FeS phases at relatively low (e.g.,eutectic) temperature, and a second segregation step of primarily Fe metal at higher temperature. In this model, the IID irons derive from the latter segregation step, while the early segregated metals have not been sampled yet. If we assume that the early segregated fraction of the IID irons occurred at the eutectic temperature (1280 K) and at #tCAI $ 1.0 Myr, i.e., similar to that inferred for the IIAB irons, and that the second segregation occurred ~1 Myr later (i.e., when the IVA core formed) we can infer that the residual IID mantle would need to have had an only slightly higher-than-chondritic Hf/W of ~3 in order to evolve to the elevated pre-exposure !182W of the IID irons. Such a modest Hf/W fractionation is likely, given that the early segregated, S-rich metal almost certainly contained some W, such that segregation of that metal left behind a residual mantle—from which the IID irons later formed—with higher-than-chondritic Hf/W. We conclude that the metal segregation event that gave rise to the IID irons may well have occurred at about the same time as core formation in the IVA iron meteorite parent body, in which case the IID irons would plot on the same heating curves as the other iron meteorite groups (Fig. 5.7C,D).
5.4 Conclusions The Hf-W results indicate that there are resolvable differences in the time of core formation in the IIAB, IIIAB, IVA, IVB and IID iron meteorite parent bodies. These differences reflect variations in the average melting temperatures of the metal, which are controlled by the S contents of the metal cores. Regardless of differences in the time of core formation, all five parent bodies seem to have accreted at about the same time, most likely between ~0.3-0.7 Myr after CAI formation. A major implication of this conclusion is that the variable depletions of moderately volatile elements in the iron meteorite parent bodies cannot be due to accretion at different times in an evolving solar nebula. Instead, they seem to reflect spatial heterogeneities between different feeding zones in the solar protoplanetary disk that contributed material to the iron meteorite parent bodies. Our results confirm that the radioactive decay of 26Al most likely was the principle heat source for differentiation of early-accreted planetary bodies (Urey, 1955; Lee et al., 1976; Hevey and Sanders, 2006).
Part A Chapter 5 105
Acknowledgments We thank R. Wieler, J. Wasson, P. Sprung, D. Cook, R. Hin, F. Nimmo, and W. van Westrenen for discussions, and U. Heitmann for technical support during sample preparation. This study was partly supported by a Förderungsprofessor to T. Kleine of the Swiss National Science Foundation (Grant no. PP00P2_123470).
References Akridge G., Benoit P. H. and Sears D. W. G. (1998) Regolith and Megaregolith Formation of
H-Chondrites: Thermal Constraints on the Parent Body. Icarus 132, 185-195. Arlandini C., Kappeler F., Wisshak K., Gallino R., Lugaro M., Busso M. and Straniero O.
(1999) Neutron capture in low-mass asymptotic giant branch stars: Cross sections and abundance signatures. Astrophysical Journal 525, 886-900.
Blichert-Toft J., Moynier F., Lee C. T. A., Telouk P. and Albarede F. (2010) The early formation of the IVA iron meteorite parent body. Earth and Planetary Science Letters 296, 469-480.
Burkhardt C., Kleine T., Dauphas N. and Wieler R. (2012) Nucleosynthetic tungsten isotope anomalies in acid leachates of the Murchison chondrite: Implications for hafnium-tungsten chronometry. The Astrophysical Journal Letters 753, L6.
Campbell A. J. and Humayun M. (2005) Compositions of group IVB iron meteorites and their parent melt. Geochimica Et Cosmochimica Acta 69, 4733-4744.
Carslaw H. S. and Jaeger J. C., 1959. Conduction of heat in solids. Oxford University Press. Chabot N. L. (2004) Sulfur contents of the parental metallic cores of magmatic iron
meteorites. Geochimica Et Cosmochimica Acta 68, 3607-3618. Chabot N. L. and Haack H., 2006. Evolution of Asteroidal Cores. The University of Arizona
Press. Chen J. H. and Wasserburg G. J. (1990) The Isotopic Composition of Ag in Meteorites and
the Presence of Pd-107 in Protoplanets. Geochimica Et Cosmochimica Acta 54, 1729-1743.
Cook D. L., Walker R. J., Horan M. F., Wasson J. T. and Morgan J. W. (2004) Pt-Re-Os systematics of group IIAB and IIIAB iron meteorites. Geochimica Et Cosmochimica Acta 68, 1413-1431.
Cottrell E., Walter M. J. and Walker D. (2010) Erratum to “Metal–silicate partitioning of tungsten at high pressure and temperature: Implications for equilibrium core formation in Earth” [Earth and Planetary Science Letters 281 (2009) 275–287]. Earth and Planetary Science Letters 289, 631-634.
Davis A. M., 2006. Volatile evolution and loss. In: Lauretta, D. S. and McSween, H. Y. Eds.), Meteorites and the early solar system II. The University of Arizona Press, Tuscon, Arizona.
Fei Y., Bertka C. M. and Finger L. W. (1997) High-Pressure Iron-Sulfur Compound, Fe3S2, and Melting Relations in the Fe-FeS System. Science 275, 1621-1623.
Haack H. and McCoy T. J., 2003. Iron and stony-iron meteorites. In: Davis, A. M. (Ed.), Treatise on Geochemistry, Vol. 1: Meteorites, Comets, and Planets, Elsevier, Amsterdam.
Haack H., Rasmussen K. L. and Warren P. H. (1990) Effects of regolith/megaregolith insulation on the cooling histories of differentiated asteroids. Journal of Geophysical Research: Solid Earth 95, 5111-5124.
106 Resolved metal segregation ages of magmatic iron meteorites
Henke S., Gail H.-P., Trieloff M., Schwarz W. H. and Kleine T. (2012) Thermal evolution and sintering of chondritic planetesimals. A&A 537, A45.
Hevey P. J. and Sanders I. S. (2006) A model for planetesimal meltdown by Al-26 and its implications for meteorite parent bodies. Meteoritics & Planetary Science 41, 95-106.
Horan M. F., Smoliar M. I. and Walker R. J. (1998) W-182 and Re-187-Os-187 systematics of iron meteorites: Chronology for melting, differentiation, and crystallization in asteroids. (vol 62, pg 545, 1998). Geochimica et Cosmochimica Acta 62, 1653-1653.
Kleine T., Mezger K., Munker C., Palme H. and Bischoff A. (2004) Hf-182-W-182 isotope systematics of chondrites, eucrites, and martian meteorites: Chronology of core formation and early mantle differentiation in Vesta and Mars. Geochimica Et Cosmochimica Acta 68, 2935-2946.
Kleine T., Mezger K., Palme H., Scherer E. and Munker C. (2005) Early core formation in asteroids and late accretion of chondrite parent bodies: Evidence from Hf-182-W-182 in CAIs, metal-rich chondrites, and iron meteorites. Geochimica et Cosmochimica Acta 69, 5805-5818.
Kleine T., Munker C., Mezger K. and Palme H. (2002) Rapid accretion and early core formation on asteroids and the terrestrial planets from Hf-W chronometry. Nature 418, 952-955.
Kleine T., Touboul M., Bourdon B., Nimmo F., Mezger K., Palme H., Jacobsen S. B., Yin Q. Z. and Halliday A. N. (2009) Hf-W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochimica Et Cosmochimica Acta 73, 5150-5188.
Kleine T., Touboul M., Van Orman J. A., Bourdon B., Maden C., Mezger K. and Halliday A. N. (2008) Hf-W thermochronometry: Closure temperature and constraints on the accretion and cooling history of the H chondrite parent body. Earth and Planetary Science Letters 270, 106-118.
Kruijer T. S., Fischer-Gödde M., Kleine T., Sprung P., Leya I. and Wieler R. (2013a) Neutron capture on Pt isotopes in iron meteorites and the Hf–W chronology of core formation in planetesimals. Earth and Planetary Science Letters 361, 162-172.
Kruijer T. S., Kleine T., Fischer-Gödde M., Burkhardt C. and Wieler R. (2013b) Nucleosynthetic W isotope anomalies in Ca-Al-rich inclusions: implications for Hf-W chronometry. Earth and Planetary Science Letters, in revision.
Kruijer T. S., Sprung P., Kleine T., Leya I., Burkhardt C. and Wieler R. (2012) Hf–W chronometry of core formation in planetesimals inferred from weakly irradiated iron meteorites. Geochimica et Cosmochimica Acta 99, 287-304.
Lee T., Papanastassiou D. A. and Wasserburg G. J. (1976) Demonstration of Mg-26 Excess in Allende and Evidence for Al-26. Geophysical Research Letters 3, 109-112.
Leya I. and Masarik J. (2013) Thermal neutron capture effects in radioactive and stable nuclide systems. Meteoritics & Planetary Science, n/a-n/a.
Lodders K. and Fegley B., 1998. The Planetary Scientist's Companion. Oxford University Press.
Markowski A., Quitte G., Halliday A. N. and Kleine T. (2006) Tungsten isotopic compositions of iron meteorites: Chronological constraints vs. cosmogenic effects. Earth and Planetary Science Letters 242, 1-15.
Masarik J. (1997) Contribution of neutron-capture reactions to observed tungsten isotopic ratios. Earth and Planetary Science Letters 152, 181-185.
McCoy T. J., Walker R. J., Goldstein J. I., Yang J., McDonough W. F., Rumble D., Chabot N. L., Ash R. D., Corrigan C. M., Michael J. R. and Kotula P. G. (2011) Group IVA irons:
Part A Chapter 5 107
New constraints on the crystallization and cooling history of an asteroidal core with a complex history. Geochimica Et Cosmochimica Acta 75, 6821-6843.
Miyamoto M., Fujii N. and Takeda H. (1981) Ordinary chondrite parent body: An internal heating model. Proc. Lunar Planet. Sci. 12B, 1145–1152.
Norris T. L., Gancarz A. J., Rokop D. J. and Thomas K. W. (1983) Half-life of 26Al. Journal of Geophysical Research: Solid Earth 88, B331-B333.
Qin L. P., Dauphas N., Wadhwa M., Markowski A., Gallino R., Janney P. E. and Bouman C. (2008) Tungsten nuclear anomalies in planetesimal cores. Astrophysical Journal 674, 1234-1241.
Rehkamper M. and Halliday A. N. (1997) Development and application of new ion-exchange techniques for the separation of the platinum group and other siderophile elements from geological samples. Talanta 44, 663-672.
Scherstén A., Elliott T., Hawkesworth C., Russell S. and Masarik J. (2006) Hf-W evidence for rapid differentiation of iron meteorite parent bodies. Earth and Planetary Science Letters 241, 530-542.
Schultz L. and Franke L., 2004. Helium, Neon & Argon in Meteorites: A data collection. Max-Planck-Institut für Geochemie, Mainz (Germany).
Scott E. R. D. (1972) Chemical Fractionation in Iron-Meteorites and Its Interpretation. Geochimica Et Cosmochimica Acta 36, 1205-1236.
Scott E. R. D. and Wasson J. T. (1975) Classification and Properties of Iron-Meteorites. Reviews of Geophysics 13, 527-546.
Smoliar M. I., Walker R. J. and Morgan J. W. (1996) Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271, 1099-1102.
Stewart A. J., Schmidt M. W., van Westrenen W. and Liebske C. (2007) Mars: A New Core-Crystallization Regime. Science 316, 1323-1325.
Touboul M., Kleine T., Bourdon B., Van Orman J. A., Maden C. and Zipfel J. (2009) Hf-W thermochronometry: II. Accretion and thermal history of the acapulcoite-lodranite parent body. Earth and Planetary Science Letters 284, 168-178.
Touboul M. and Walker R. J. (2012) High precision tungsten isotope measurement by thermal ionization mass spectrometry. International Journal of Mass Spectrometry 309, 109-117.
Urey H. C. (1955) The cosmic abundances of potassium, uranium, and thorium and the heat balances of the Earth, the Moon, and Mars. Proc. Natl. Acad. Sci. U.S. 41, 127-144.
Vockenhuber C., Oberli F., Bichler M., Ahmad I., Quitte G., Meier M., Halliday A. N., Lee D. C., Kutschera W., Steier P., Gehrke R. J. and Helmer R. G. (2004) New half-life measurement of Hf-182: Improved chronometer for the early solar system. Physical Review Letters 93.
Walker R. J., McDonough W. F., Honesto J., Chabot N. L., McCoy T. J., Ash R. D. and Bellucci J. J. (2008) Modeling fractional crystallization of group IVB iron meteorites. Geochimica Et Cosmochimica Acta 72, 2198-2216.
Wasson J. T. (1999) Trapped melt in IIIAB irons; solid/liquid elemental partitioning during the fractionation of the IIIAB magma. Geochimica et Cosmochimica Acta 63, 2875-2889.
Wasson J. T. and Huber H. (2006) Compositional trends among IID irons; their possible formation from the P-rich lower magma in a two-layer core. Geochimica Et Cosmochimica Acta 70, 6153-6167.
Willbold M., Elliott T. and Moorbath S. (2011) The tungsten isotopic composition of the Earth's mantle before the terminal bombardment. Nature 477, 195-U91.
108 Resolved metal segregation ages of magmatic iron meteorites
Wittig N., Humayun M., Brandon A. D., Huang S. and Leya I. (2013) Coupled W–Os–Pt isotope systematics in IVB iron meteorites: In situ neutron dosimetry for W isotope chronology. Earth and Planetary Science Letters 361, 152-161.
Yang J., Goldstein J. I. and Scott E. R. D. (2008) Metallographic cooling rates and origin of IVA iron meteorites. Geochimica Et Cosmochimica Acta 72, 3043-3061.
Yang J. J., Goldstein J. I. and Scott E. R. D. (2007) Iron meteorite evidence for early formation and catastrophic disruption of protoplanets. Nature 446, 888-891.
York D. (1966) Least-squares fitting of a straight line. Canadian Journal of Physics 44, 1079-1086.
Yoshino T., Walter M. J. and Katsura T. (2003) Core formation in planetesimals triggered by permeable flow. Nature 422, 154-157.
5.5 Appendix: Supplementary text, figures and tables
5.5.1 Materials and Methods
5.5.1.1 Iron meteorite samples Several samples from the IIAB, IIIAB, IVA and IVB iron meteorite groups were selected for this work. We analysed 8 IIAB irons, 14 IIIAB irons, 2 IVA irons and 3 IVB irons for their Pt and W isotope compositions. The data are supplemented by previously published Pt and W isotope data for the IID, IVA and IVB irons (Kruijer et al., 2013a). Care was taken to select samples with varying degrees of irradiation as suggested by previously reported cosmogenic noble gas concentrations (Schultz and Franke, 2004) and 182W/184W compositions (Kruijer et al., 2012; Kruijer et al., 2013a). A wide spread in neutron capture induced isotope variations is an important prerequisite to obtain well-defined empirical Pt-W isotope correlation lines. The IIAB samples Braunau and Edmonton, the IIIAB samples Cape York, and the IVA samples Gibeon and Muonionalusta are characterized by very low concentrations of noble gases and/or very low exposure ages and thus may have only very small neutron capture-induced W isotope variations (Kruijer et al., 2012). As these samples essentially require no correction, they are of great importance to establish the pre-exposure 182W/184W composition of a particular iron meteorite group.
5.5.1.2 Sample preparation and chemical separation of Pt and W for MC-ICPMS analyses
The new metal samples (!0.7-1.1 g) from this study were cut using a diamond saw, polished with abrasives (SiC) and subsequently ultrasonically cleaned in ethanol or acetone to remove any saw marks and adhering dust. To prevent any (additional) terrestrial contamination, the outermost parts (5-15%) of the samples were removed by leaching in 6 M HCl (+trace HNO3) at 80 ºC for 10-15 min. All samples were digested in concentrated HNO3-HCl (2:1) on a hot plate (130 °C) for at least 24hr. Upon complete dissolution, 500-1000 mg and 50 mg aliquots were taken for W and Pt isotope analyses.
Tungsten was separated from the sample matrix using a two-stage anion exchange chromatography in HCl-HF media, described in detail in Kruijer et al. (2012) and largely based on previously developed procedures (Horan et al., 1998; Kleine et al., 2002; 2004). Platinum was separated following procedures described in Kruijer et al. (2013a), which involves an anion chromatography step modified after Rehkamper and Halliday (1997). Note that solvent extraction to remove Os – which may generate an isobaric interference on 192Pt –
Part A Chapter 5 109
was abandoned in the present study. Instead, the final Pt cuts were evaporated to dryness in concentrated HClO4 at 200°C several times. This procedure resulted in pure Pt cuts with Os/Pt ranging from ~1%10-8–3%10-6, corresponding to <0.3 pg Os/measurement and resulting in interference corrections on 192Pt/195Pt of <150 ppm, i.e., mostly within the external uncertainty of the measurements. About 15% of the Pt cuts showed Os/Pt exceeding this range (up to ~1%10-4), corresponding to <10 pg Os/measurement and resulting in larger interference corrections on 192Pt/195Pt up to ~60 !-units. However, analyses of Pt solution standards with varying amounts of admixed Os showed that Os interferences of this magnitude can accurately be corrected (Kruijer et al., 2013a). Total procedure blanks correspond to ~50-100 pg Pt and ~50 pg W, which was negligible given the large amounts of analysed Pt and W. Total yields were ~60-90% for Pt and ~70-90% for W.
5.5.1.3 Mass spectrometry
5.5.1.3.1 Procedures for Pt and W isotope measurements by MC-ICPMS and TIMS Isotope analyses (Pt and W) were performed on a ThermoScientific® Neptune Plus MC-ICPMS at the University of Münster. The analytical procedures for Pt and W isotope analyses by MC-ICPMS are described in detail in Kruijer et al. (2012) and Kruijer et al. (2013a). In addition, some of the W isotope analyses were performed by negative thermal ionization mass spectrometry on a ThermoFischer® Triton TIMS at the University of Maryland, College Park. Detailed procedures for W isotope analyses by N-TIMS are described in Touboul and Walker (2012). A brief summary of the Pt and W isotope analyses by MC-ICPMS is given below.
Samples and standards for both Pt and W isotope analyses were introduced into the MC-ICPMS using an ESI® self-aspirating PFA nebulizer (50-60 &L/min) connected to a Cetac® Aridus II desolvator. For all Pt isotope measurements standard Ni (H) sample and skimmer cones were used. Total ion beam intensities of ~2.5-3.5 %10-10 A were obtained for a ~200 ppb Pt standard solution at uptake rates of 50-60 &l/min. Baselines were obtained by deflecting the beam using the electrostatic analyser for 60-90s. An individual Pt isotope measurement comprised 100 cycles of 4.2s each. Raw 192Pt/195Pt ratios were corrected for isobaric interference from 192Os by monitoring interference-free 189Os. Measured Pt isotope ratios were corrected for instrumental mass bias through internal normalization to 198Pt/195Pt = 0.2145 (denoted ‘8/5’) or 196Pt/195Pt = 0.7464 (‘6/5’) using the exponential law. For the W isotope analyses, standard (H) cones were used during the initial stages of this study. Using this cone design total ion beam intensities of ~2-2.5 %10-10 A were generally obtained for a ~100 ppb W standard solution at uptake rates of 50-60 &l/min. After careful accuracy tests (Section 5.3.5.3.3), a combination of Jet sampler and X-skimmer cones was used during the later stages of this study. The Jet sampler / X skimmer cone set up provides a significant increase in sensitivity to ~1100-1600 V/ppm relative to the standard cones (400-450 V/ppm), resulting in total ion beams of ~3-5 %10-10 obtained for a ~50 ppb W standard solution at uptake rates of 50-60 &l/min. Baselines were obtained by deflecting the beam using the electrostatic analyser for 60s. All samples were measured 4-5 times and an individual W isotope measurement comprised 200 cycles of 4.2s each. Small isobaric interferences from 184Os and 186Os on W isotope ratios were corrected by monitoring interference-free 189Os, and were generally negligible (<10 ppm). Instrumental mass bias was corrected by normalization to 186W/183W = 1.9859 (denoted ‘6/3’) or 186W/184W = 0.92767 (denoted ‘6/4’) using the exponential law.
The sample analyses for both the Pt and W isotope were bracketed by measurements of terrestrial solution standards (Alfa Aesar®) and results are reported as !-unit (i.e., 0.01%)
110 Resolved metal segregation ages of magmatic iron meteorites
deviations from the mean values of the bracketing standards, together with their associated external uncertainties (Sections 5.5.1.3.3). The reported !iW and !iPt represent the mean of pooled solution replicates (n=4-6 for W and n=1-5 for Pt).
5.5.1.3.2 Accuracy and reproducibility of Pt isotope measurements To assess the accuracy and reproducibility of the Pt isotope analyses, in each round of measurements, a terrestrial metal standard (NIST129c) was repeatedly analysed. Additional amounts of highly siderophile elements from solution standards (Alfa Aesar®), whose amounts and proportions matched those of iron meteorites (e.g, ~4 &g Pt/g of NIST129c), were added before digestion of the metal standard. The standard was consecutively digested, processed through the full chemical separation and analysed together with the iron meteorite samples. The !196Pt (8/5) and !192Pt (8/5) results obtained for the terrestrial metal standard are presented in Fig. 5.8. The long-term external reproducibility of this standard was estimated from single measurements of several digestion replicates (n=7) which yield mean !196Pt of 0.00±0.07 and !192Pt of 0.0±1.3 (±2' s.d., n=62). Hence, within uncertainty all Pt isotope analyses of the terrestrial standard are identical to the terrestrial value, indicating that the analyses are accurate. For samples analysed 1-3 times, the uncertainties quoted above were used. For samples analysed >3 times, the 95% conf. limits of the mean as given by of the replicate sample measurements is taken as an estimate for the external reproducibility. The external precision of !196Pt (±0.07!, 2' s.d.) is identical to the typical internal precision obtained for single measurement (±0.07!, 2' s.e.). This indicates that external scatter – which could be induced during digestion and/or chemical separation - is largely absent. Conversely, the !192Pt measurements exhibit an external precision (±1.3 !, 2' s.d.) that is significantly larger than the associated within-run measurement precision (±0.5!, 2' s.e.). This observation most likely reflects the fact that 192Pt is a minor isotope (0.78% ab.) and as such, more susceptible to slight impurities from the sample matrix or introduction system, which can induce isobaric (i.e., Os) or molecular interferences. Finally, we note that the use of the Aridus instead of the ESI® APEX introduction system strongly reduced a tailing effect on 192Pt from neighbouring Ir isotopes (see Kruijer et al., 2013a). The tailing corrections on 192Pt/195Pt were ~1.5 !-units for measurement solutions having Ir/Pt $ 1. Depending on the Ir/Pt of the sample and quality of the chemical separation of Ir from Pt, the Ir/Pt of the measurement solutions was generally much smaller than 1. Hence, the magnitude of the tailing corrections did not exceed the measurement uncertainty on 192Pt/195Pt (±1.3 !, 2' s.d.).
Part A Chapter 5 111
5.5.1.3.3 Accuracy and reproducibility of W isotope measurements by MC-ICPMS The accuracy and reproducibility of the W isotope analyses by MC-ICPMS were evaluated by repeated measurements of a terrestrial metal standard (NIST129c) that was doped with additional W before dissolution to match the W concentrations of samples. In each round of measurements, a separate digestion of this standard (~500 mg) was processed through the full chemical separation procedure and analysed together with the iron meteorite samples. Each metal standard was measured 3-5 times and each single measurement consisted of 200 cycles of 4.2s each. The measurements of the metal standard yield precise !iW values that are mostly identical to the value of the terrestrial solution standard (Fig. 5.9), both for standard H cones as well as for high-sensitivity Jet sampler and X skimmer cones (Fig. 5.9). For instance, the mean value obtained for the NIST129c analyses using Jet sampler and X-skimmer cones corresponds to !182/184W (6/4) of 0.01±0.05 (2' s.d., n=7) (Fig. 5.9; Table 5-A2), demonstrating the high level of accuracy of the W isotope measurements. Reported uncertainties for iron meteorites represent 95% confidence limits of the mean obtained for pooled solution replicates (n=4-6).
Fig. 5.8: External reproducibility of (A) !196Pt (8/5) and (B) !192Pt (8/5) for the terrestrial metal standard (NIST129c with admixed HSE) that was processed through the full chemical separation. Each symbol represents a single Pt isotope measurement (100 cycles) measured at 200 ppb, while the different colours denote separate digestions of the metal standard (50 mg each). Error bars indicate the typical internal uncertainties (2", s.e.) of single measurements (~0.07 for !196Pt, ~0.5 for !192Pt), hatched areas represent the external reproducibility (2" s.d.), and grey area the associated 95% conf. limits of the mean.
ε196
Pt (8
/5)
Terrestrial metal standard (NIST129c):ε196Pt (mean, n=62)
"= 0.00±0.07 (2σ, s.d.)"= 0.00±0.01 (95% conf.)
ε192
Pt (8
/5)
Terrestrial metal standard (NIST129c):ε192Pt (mean, n=62)
"= 0.0±1.3 (2σ, s.d.)"= 0.0±0.2 (95% conf.)
(A)
(B)
–2
0
2
–0.2
–0.1
0.0
0.1
112 Resolved metal segregation ages of magmatic iron meteorites
Some standard (and sample) analyses showed small anomalies for normalizations involving 183W, including excesses in !182/183W (6/3) (up to +0.25) and in !184W (6/3) (up to +0.12) (Table 5-A2). These coupled !182W-!183W systematics have previously been observed in recent high-precision MC-ICPMS studies for terrestrial standards as well as silicate rock and iron meteorite samples (Willbold et al., 2011; Kruijer et al., 2012; 2013a), and are attributed to a mass-independent W isotope fractionation only affecting 183W that is induced by W-loss during re-dissolution of the samples in Savillex beakers. Repeatedly drying the samples in hot (200 °C) concentrated HClO4 after each chemical separation step appeared an effective means to strongly diminish the effect. Consequently, the mass-independent effect is only observed for a minority of the standards and samples from the present study (Tables S5-A2, 5-A4), and mostly for those samples that were analysed using H cones at the beginning of our analytical campaign (e.g.,the IIAB irons). The W isotope compositions of iron meteorite samples can be corrected for the mass-independent effect on 183W using the different normalization schemes for the W isotope measurements and the results obtained for the terrestrial standard (see Kruijer et al., 2013a; 2012; Willbold et al., 2011). Accordingly, the measured !182W (6/3), !183W (6/4), and !184W (6/3) of the iron meteorites from this study showing the mass independent effect were corrected using the mean !iW values obtained for the NIST129c standard (Tables 5-A2) for H cones (in case of the IIAB irons) and for Jet sampler and X-skimmer cones (in case of the IIIAB irons), and the associated uncertainties of the NIST129c data (95% conf.) were propagated into the W isotope data reported in Table 5-A2. The corrected !182/183W (6/3) values are indistinguishable from the measured !182/184W (6/4), indicating that the corrections are accurate (Tables 5-A2, 5-A4).
Fig. 5.9: External reproducibility of !182W (6/4) based on several digestion replicates of the terrestrial metal standard (NIST129c) analysed by MC-ICPMS. Each symbol represents the average of several solution replicates (i.e, 3-5 % 200 cycles) and their error bars show the associated 95% conf. limits of the mean. Identical results are obtained for different cone designs (i.e., standard H vs. Jet sampler + X skimmer). Hatched areas represent the long-term external reproducibility (2", s.d.) and filled areas show the 95% confidence limits of the mean.
ε182
W (6
/4)
H cones:ε182W (mean, n= 8)
"= 0.02±0.07 (2σ, s.d.)"= 0.02±0.03 (95% conf.)
Jet / X cones:ε182W (mean, n= 7)
"= 0.01±0.05 (2σ, s.d.)"= 0.01±0.03 (95% conf.)
Terrestrial metal standard (NIST129c)
–0.3
–0.2
–0.1
0.0
0.1
0.2
0.3
Part A Chapter 5 113
5.5.2 Data tables
Table 5-A1Platinum isotope compositions of the terrestrial metal standard (NIST129c) determined by MC-ICPMS.
ID !192Pt (6/5)a !194Pt (6/5)a !198Pt (6/5)a !192Pt (8/5)a !194Pt (8/5)a !196Pt (8/5)a
(±2" s.e.) (±2" s.e.) (±2" s.e.) (±2" s.e.) (±2" s.e.) (±2" s.e.)
February - June 2012S04_1 -1.5 ± 0.5 -0.13 ± 0.11 0.02 ± 0.20 -1.5 ± 0.5 -0.13 ± 0.08 -0.01 ± 0.07S04_2 0.0 ± 0.5 -0.05 ± 0.11 -0.15 ± 0.20 -0.2 ± 0.5 -0.09 ± 0.08 0.05 ± 0.07S04_3 0.7 ± 0.5 -0.01 ± 0.11 0.09 ± 0.20 0.8 ± 0.5 -0.01 ± 0.08 -0.03 ± 0.07S04_4 0.1 ± 0.5 -0.10 ± 0.11 0.18 ± 0.20 0.3 ± 0.5 -0.08 ± 0.08 -0.06 ± 0.07S04_5 -0.1 ± 0.5 -0.04 ± 0.11 0.09 ± 0.20 0.1 ± 0.5 0.00 ± 0.08 -0.03 ± 0.07S04_6 -0.3 ± 0.5 -0.12 ± 0.11 0.16 ± 0.20 -0.1 ± 0.5 -0.04 ± 0.08 -0.05 ± 0.07S04_7 0.6 ± 0.5 0.03 ± 0.11 -0.04 ± 0.20 0.6 ± 0.5 0.01 ± 0.08 0.01 ± 0.07S04_8 0.1 ± 0.5 -0.01 ± 0.11 0.07 ± 0.20 0.1 ± 0.5 -0.01 ± 0.08 -0.02 ± 0.07S04_9 -0.1 ± 0.5 -0.12 ± 0.11 0.09 ± 0.20 0.0 ± 0.5 -0.05 ± 0.08 -0.03 ± 0.07S04_10 1.0 ± 0.5 -0.02 ± 0.11 0.02 ± 0.20 0.9 ± 0.5 -0.03 ± 0.08 -0.01 ± 0.07S04_11 -0.3 ± 0.5 -0.10 ± 0.11 0.16 ± 0.20 -0.2 ± 0.5 -0.04 ± 0.08 -0.05 ± 0.07S04_12 0.1 ± 0.5 -0.09 ± 0.11 0.00 ± 0.20 0.1 ± 0.5 -0.06 ± 0.08 0.00 ± 0.07S04_13 -0.2 ± 0.5 -0.04 ± 0.11 -0.21 ± 0.20 -0.5 ± 0.5 -0.15 ± 0.08 0.07 ± 0.07
M09 0.6 ± 0.5 0.10 ± 0.11 -0.14 ± 0.20 0.2 ± 0.5 0.00 ± 0.08 0.05 ± 0.07M09 0.3 ± 0.5 0.05 ± 0.11 -0.07 ± 0.20 0.3 ± 0.5 0.04 ± 0.08 0.02 ± 0.07M09 -0.8 ± 0.5 -0.01 ± 0.11 -0.01 ± 0.20 -0.8 ± 0.5 -0.02 ± 0.08 0.00 ± 0.07M09 0.2 ± 0.5 -0.08 ± 0.11 0.09 ± 0.20 0.3 ± 0.5 -0.08 ± 0.08 0.01 ± 0.07M09 2.3 ± 0.5 0.04 ± 0.11 -0.21 ± 0.20 2.2 ± 0.5 -0.05 ± 0.08 0.07 ± 0.07
September 2012 - May 2013W09b_1 -1.4 ± 0.5 0.07 ± 0.11 0.08 ± 0.20 -1.3 ± 0.5 0.10 ± 0.08 -0.03 ± 0.07W09b_2 0.0 ± 0.5 -0.04 ± 0.11 0.14 ± 0.20 0.2 ± 0.5 0.00 ± 0.08 -0.05 ± 0.07W09b_3 0.1 ± 0.5 0.10 ± 0.11 -0.08 ± 0.20 0.1 ± 0.5 0.07 ± 0.08 0.03 ± 0.07W09b_4 -1.4 ± 0.5 0.03 ± 0.11 -0.02 ± 0.20 -1.4 ± 0.5 0.02 ± 0.08 0.01 ± 0.07W09b_5 -0.2 ± 0.5 0.03 ± 0.11 0.04 ± 0.20 -0.1 ± 0.5 0.04 ± 0.08 -0.01 ± 0.07W09b_6 -0.5 ± 0.5 0.05 ± 0.11 -0.04 ± 0.20 -0.5 ± 0.5 0.03 ± 0.08 0.01 ± 0.07W09b_7 -0.6 ± 0.5 0.08 ± 0.11 -0.10 ± 0.20 -0.6 ± 0.5 0.04 ± 0.08 0.03 ± 0.07
W09a_1 -0.3 ± 0.5 0.01 ± 0.11 0.06 ± 0.20 -0.2 ± 0.5 0.03 ± 0.08 -0.02 ± 0.07W09a_2 0.1 ± 0.5 -0.08 ± 0.11 0.19 ± 0.20 0.3 ± 0.5 -0.02 ± 0.08 -0.06 ± 0.07W09a_3 -0.5 ± 0.5 0.02 ± 0.11 -0.06 ± 0.20 -0.6 ± 0.5 -0.01 ± 0.08 0.02 ± 0.07W09a_4 -0.2 ± 0.5 0.00 ± 0.11 -0.08 ± 0.20 -0.3 ± 0.5 -0.03 ± 0.08 0.03 ± 0.07W09a_5 0.8 ± 0.5 -0.01 ± 0.11 0.11 ± 0.20 1.0 ± 0.5 0.03 ± 0.08 -0.04 ± 0.07W09a_6 0.3 ± 0.5 0.05 ± 0.11 -0.07 ± 0.20 0.3 ± 0.5 0.03 ± 0.08 0.02 ± 0.07W09a_7 -0.3 ± 0.5 -0.09 ± 0.11 0.02 ± 0.20 -0.2 ± 0.5 -0.08 ± 0.08 -0.01 ± 0.07
U07b_1 0.1 ± 0.5 0.05 ± 0.11 -0.04 ± 0.20 0.1 ± 0.5 0.04 ± 0.08 0.01 ± 0.07U07b_2 0.2 ± 0.5 0.06 ± 0.11 -0.07 ± 0.20 0.1 ± 0.5 0.03 ± 0.08 0.03 ± 0.07U07b_3 0.0 ± 0.5 0.08 ± 0.11 -0.06 ± 0.20 0.0 ± 0.5 0.06 ± 0.08 0.02 ± 0.07U07b_4 -0.2 ± 0.5 0.01 ± 0.11 0.16 ± 0.20 0.0 ± 0.5 0.07 ± 0.08 -0.05 ± 0.07U07b_5 -0.3 ± 0.5 0.00 ± 0.11 -0.07 ± 0.20 -0.3 ± 0.5 -0.02 ± 0.08 0.02 ± 0.07U07b_6 0.8 ± 0.5 0.08 ± 0.11 -0.13 ± 0.20 0.7 ± 0.5 0.04 ± 0.08 0.05 ± 0.07U07b_7 -0.2 ± 0.5 0.01 ± 0.11 0.03 ± 0.20 -0.2 ± 0.5 0.02 ± 0.08 -0.01 ± 0.07U07b_8 -0.1 ± 0.5 -0.02 ± 0.11 -0.15 ± 0.20 -0.3 ± 0.5 -0.07 ± 0.08 0.05 ± 0.07U07b_9 1.1 ± 0.5 0.02 ± 0.11 -0.04 ± 0.20 1.1 ± 0.5 0.01 ± 0.08 0.01 ± 0.07U07b_10 0.7 ± 0.5 0.07 ± 0.11 -0.12 ± 0.20 0.6 ± 0.5 0.03 ± 0.08 0.04 ± 0.07U07b_11 0.4 ± 0.5 -0.05 ± 0.11 0.08 ± 0.20 0.5 ± 0.5 -0.02 ± 0.08 -0.03 ± 0.07U07b_12 -0.8 ± 0.5 -0.14 ± 0.11 0.05 ± 0.20 -0.7 ± 0.5 -0.12 ± 0.08 -0.02 ± 0.07U07b_13 0.9 ± 0.5 0.11 ± 0.11 -0.14 ± 0.20 0.8 ± 0.5 0.07 ± 0.08 0.05 ± 0.07
U07a_1 -0.9 ± 0.5 -0.04 ± 0.11 -0.07 ± 0.20 -0.9 ± 0.5 -0.07 ± 0.08 0.02 ± 0.07U07a_2 0.4 ± 0.5 0.01 ± 0.11 -0.08 ± 0.20 0.3 ± 0.5 -0.01 ± 0.08 0.03 ± 0.07U07a_3 0.5 ± 0.5 -0.02 ± 0.11 -0.18 ± 0.20 0.3 ± 0.5 -0.08 ± 0.08 0.06 ± 0.07U07a_4 0.6 ± 0.5 -0.08 ± 0.11 -0.06 ± 0.20 0.6 ± 0.5 -0.10 ± 0.08 0.02 ± 0.07U07a_5 0.2 ± 0.5 -0.09 ± 0.11 0.07 ± 0.20 0.3 ± 0.5 -0.07 ± 0.08 -0.02 ± 0.07U07a_6 -0.2 ± 0.5 -0.06 ± 0.11 0.15 ± 0.20 0.0 ± 0.5 -0.01 ± 0.08 -0.05 ± 0.07U07a_7 0.7 ± 0.5 -0.05 ± 0.11 0.19 ± 0.20 0.9 ± 0.5 0.01 ± 0.08 -0.06 ± 0.07U07a_8 -0.5 ± 0.5 0.02 ± 0.11 -0.16 ± 0.20 -0.7 ± 0.5 -0.03 ± 0.08 0.05 ± 0.07U07a_9 -0.2 ± 0.5 -0.03 ± 0.11 0.12 ± 0.20 -0.1 ± 0.5 0.01 ± 0.08 -0.04 ± 0.07U07a_10 0.0 ± 0.5 -0.02 ± 0.11 0.02 ± 0.20 0.0 ± 0.5 -0.01 ± 0.08 -0.01 ± 0.07U07a_11 -0.8 ± 0.5 -0.03 ± 0.11 0.14 ± 0.20 -0.6 ± 0.5 0.02 ± 0.08 -0.05 ± 0.07U07a_12 -0.6 ± 0.5 -0.08 ± 0.11 0.13 ± 0.20 -0.5 ± 0.5 -0.04 ± 0.08 -0.04 ± 0.07U07a_13 0.5 ± 0.5 0.05 ± 0.11 -0.05 ± 0.20 0.4 ± 0.5 0.03 ± 0.08 0.02 ± 0.07U07a_14 -0.1 ± 0.5 0.06 ± 0.11 0.03 ± 0.20 0.0 ± 0.5 0.07 ± 0.08 -0.01 ± 0.07
AG10_1 -1.6 ± 0.5 -0.04 ± 0.11 -0.17 ± 0.20 -1.8 ± 0.5 -0.14 ± 0.08 0.06 ± 0.07AG10_2 -0.4 ± 0.5 -0.09 ± 0.11 -0.11 ± 0.20 -0.5 ± 0.5 -0.12 ± 0.08 0.04 ± 0.07AG10_3 0.0 ± 0.5 0.07 ± 0.11 -0.11 ± 0.20 0.0 ± 0.5 -0.01 ± 0.08 0.04 ± 0.07
Mean all (±2" s.d.) 0.0 ± 1.3 -0.01 ± 0.13 0.00 ± 0.22 0.0 ± 1.3 -0.01 ± 0.11 0.00 ± 0.07(±95% conf.) 0.2 0.02 0.03 0.2 0.01 0.01
Shown are Pt isotope data for several digestions of the terrestrial metal standard that was processed and analyzed together with the samples.Each line represents a single measurement consisting of 100 cycles (4.2s each). Quoted uncertainties represent within-run precisions, expressed as two standard error of the mean (2s.e.).Note that the long term external precision (±2" s.d.) is almost identical to the within-run precision (except for !192Pt).aNormalized to 198Pt/195Pt = 0.2145 (8/5) or 196Pt/195Pt = 0.7464 (6/5) using the exponential law.
114 Resolved metal segregation ages of magmatic iron meteorites
Tabl
e 5-
A2
Tung
sten
isoto
pe c
ompo
sitio
ns o
f the
terre
stria
l met
al st
anda
rd (N
IST1
29c)
.ID
N!18
2/18
3 W (6
/3) m
eas.a
!182/
184 W
(6/4
) mea
s.a!18
3 W (6
/4) m
eas.a
!184 W
(6/3
) mea
s.a!18
2/18
3 W (6
/3) co
rr.a,
b!18
2/18
4 W (6
/3) co
rr.a,
b
(±95
% c
onf.)
(±95
% c
onf.)
(±95
% c
onf.)
(±95
% c
onf.)
(±95
% c
onf.)
(±95
% c
onf.)
Stan
dard
H c
ones
L05
30.
25±
0.09
-0.0
1±
0.11
-0.1
8±
0.04
0.12
±0.
030.
02±
0.09
-0.0
1±
0.11
L06
40.
22±
0.06
-0.0
4±
0.06
-0.1
9±
0.03
0.12
±0.
02-0
.03
±0.
06-0
.04
±0.
06O
023
0.05
±0.
100.
08±
0.14
0.01
±0.
090.
00±
0.06
0.06
±0.
100.
08±
0.14
O03
20.
21±
0.06
0.04
±0.
20-0
.10
±0.
210.
07±
0.14
0.07
±0.
060.
04±
0.20
SO4b
(H-c
ones
)5
-0.0
2±
0.07
0.03
±0.
060.
01±
0.08
-0.0
1±
0.05
0.00
±0.
070.
03±
0.06
T07a
50.
02±
0.03
0.02
±0.
04-0
.01
±0.
050.
01±
0.04
0.00
±0.
030.
02±
0.04
T07b
5-0
.02
±0.
090.
03±
0.11
0.03
±0.
10-0
.02
±0.
070.
02±
0.09
0.03
±0.
11M
095
0.06
±0.
080.
04±
0.07
-0.0
1±
0.06
0.00
±0.
040.
05±
0.08
0.04
±0.
07
H c
ones
(mea
n; ±
2" s.
d.)
80.
10±
0.22
0.02
±0.
07-0
.05
±0.
180.
040.
120.
020.
070.
020.
07 (±
95%
con
f.)±
0.09
±0.
03±
0.07
0.05
0.03
0.03
Jet s
ampl
er /
X sk
imm
er c
ones
Z08a
40.
04±
0.03
0.01
±0.
04-0
.02
±0.
040.
01±
0.03
0.01
±0.
030.
01±
0.04
Z08b
50.
04±
0.06
0.01
±0.
08-0
.02
±0.
050.
02±
0.03
0.01
±0.
060.
01±
0.08
S04a
(Jet
/X-c
ones
)5
0.03
±0.
040.
02±
0.04
-0.0
1±
0.05
0.01
±0.
030.
01±
0.04
0.02
±0.
04S0
4a (J
et/X
-con
es; 1
00 c
ycle
s))
7-0
.05
±0.
04-0
.01
±0.
07-0
.01
±0.
060.
01±
0.04
-0.0
6±
0.04
-0.0
1±
0.07
T07c
50.
02±
0.03
0.03
±0.
040.
01±
0.04
-0.0
1±
0.03
0.04
±0.
030.
03±
0.04
R09
5-0
.02
±0.
12-0
.03
±0.
090.
03±
0.03
-0.0
2±
0.02
0.01
±0.
12-0
.03
±0.
09A
G04
50.
05±
0.08
0.04
±0.
05-0
.01
±0.
080.
01±
0.05
0.04
±0.
080.
04±
0.05
X c
ones
(mea
n; ±
2" s.
d.)
70.
02±
0.07
0.01
±0.
05-0
.01
±0.
040.
000.
020.
010.
070.
010.
05 (±
95%
con
f.)±
0.03
±0.
02±
0.02
0.01
0.03
0.02
All
(mea
n; ±
2" s.
d.)
150.
06±
0.18
0.02
±0.
06-0
.03
±0.
140.
020.
090.
020.
070.
020.
06 (±
95%
con
f.)±
0.05
±0.
020.
040.
030.
020.
02H
igh
sulp
hur s
teel
(NIS
T129
c) d
oped
with
add
ition
al W
to m
atch
con
cent
ratio
ns o
f sam
ples
and
subs
eque
ntly
pro
cess
ed th
roug
h fu
ll ch
emic
al se
para
tion.
a Nor
mal
ized
to 1
86W
/184 W
= 0
.927
67 (6
/4) o
r 186 W
/183 W
= 1
.985
9 (6
/3) u
sing
the
expo
nent
ial l
aw.
b Tu
ngste
n iso
tope
com
posit
ions
cor
rect
ed fo
r mas
s-in
depe
nden
t effe
ct o
n 18
3 W (S
ectio
n 5.
5.1.
3.3)
.
Part A Chapter 5 115
Table 5-A3Platinum isotope compositions of iron meteorites determined by MC-ICPMS.
Meteorite ID N !192Pt (6/5)a !194Pt (6/5)a !198Pt (6/5)a !192Pt (8/5)a !194Pt (8/5)a !196Pt (8/5)a
(±2SD) (±2SD) (±2SD) (±2SD) (±2SD) (±2SD)
IIAB ironsSikhote Alin T01 2 0.7 ± 1.3 0.33 ± 0.12 -0.80 ± 0.21 -0.1 ± 1.3 0.06 ± 0.10 0.27 ± 0.07Edmonton, Canada T02 5 0.0 ± 0.7 0.11 ± 0.05 -0.03 ± 0.10 0.0 ± 0.7 0.10 ± 0.04 0.01 ± 0.03Braunau T03 4 0.0 ± 1.4 0.20 ± 0.05 -0.10 ± 0.16 -0.1 ± 1.2 0.17 ± 0.03 0.03 ± 0.05Ainsworth T04 2 4.7 ± 1.3 1.71 ± 0.12 -3.11 ± 0.21 1.5 ± 1.3 0.66 ± 0.10 1.04 ± 0.07Bennett County T13 4 7.4 ± 0.6 0.32 ± 0.02 -0.19 ± 0.08 7.2 ± 0.6 0.26 ± 0.04 0.07 ± 0.03Filomena T10 4 0.6 ± 0.2 0.15 ± 0.05 -0.20 ± 0.07 0.4 ± 0.2 0.09 ± 0.07 0.07 ± 0.02Negrillos T11 4 0.0 ± 0.9 0.10 ± 0.07 0.09 ± 0.16 0.1 ± 0.9 0.13 ± 0.05 -0.03 ± 0.05Mt. Joy Z01 5 2.1 ± 0.3 0.57 ± 0.06 -0.95 ± 0.05 1.2 ± 0.4 0.25 ± 0.05 0.32 ± 0.02
IID ironsRodeob R07 4 1.3 ± 2.2 0.01 ± 0.16 -0.02 ± 0.33 1.3 ± 1.3 0.00 ± 0.20 0.01 ± 0.11Rodeob S06 1 -0.4 ± 2.2 0.01 ± 0.15 0.02 ± 0.23 -0.4 ± 1.3 0.03 ± 0.15 -0.01 ± 0.08Carbob P01 2 30.4 ± 2.2 1.53 ± 0.22 -1.87 ± 0.23 28.5 ± 1.3 0.93 ± 0.15 0.63 ± 0.08Carbob P02 2 35.2 ± 2.2 1.86 ± 0.18 -2.09 ± 0.23 33.1 ± 1.3 1.15 ± 0.15 0.70 ± 0.08Carbob P03 2 39.8 ± 2.2 1.92 ± 0.15 -2.28 ± 0.23 37.5 ± 1.3 1.21 ± 0.15 0.76 ± 0.08Carbob P04 2 41.6 ± 2.2 2.24 ± 0.15 -2.55 ± 0.23 39.1 ± 1.3 1.40 ± 0.15 0.82 ± 0.08Carbob P05 1 40.5 ± 2.2 2.33 ± 0.15 -2.65 ± 0.23 37.7 ± 1.3 1.45 ± 0.15 0.88 ± 0.08Carbob S05 1 26.6 ± 2.2 1.39 ± 0.15 -1.49 ± 0.23 25.1 ± 1.3 0.84 ± 0.15 0.50 ± 0.08
IIIAB ironsFairview W05 4 7.5 ± 1.7 0.48 ± 0.12 -0.47 ± 0.27 7.1 ± 1.7 0.32 ± 0.04 0.16 ± 0.09Cape York (CY01) T05 3 0.4 ± 1.3 0.06 ± 0.12 0.03 ± 0.21 0.4 ± 1.3 0.08 ± 0.10 -0.01 ± 0.07Cape York( CY02) T06 4 0.3 ± 0.3 0.08 ± 0.10 -0.06 ± 0.08 0.2 ± 0.3 0.06 ± 0.07 0.02 ± 0.03Costilla Peak T09 4 0.7 ± 0.6 0.11 ± 0.14 0.02 ± 0.18 0.7 ± 0.4 0.12 ± 0.11 -0.01 ± 0.06Charcas Z02 4 6.4 ± 0.4 0.86 ± 0.07 -1.29 ± 0.10 5.1 ± 0.4 0.43 ± 0.06 0.43 ± 0.03Boxhole Z06 3 21.7 ± 1.3 1.24 ± 0.12 -1.34 ± 0.21 20.5 ± 1.3 0.79 ± 0.10 0.45 ± 0.07Willamette W04 4 -0.2 ± 0.8 0.00 ± 0.12 0.10 ± 0.11 -0.1 ± 1.0 0.04 ± 0.11 -0.03 ± 0.04Willamette (replicate) AG01 3 1.4 ± 1.7 0.09 ± 0.12 0.03 ± 0.21 1.4 ± 1.9 0.10 ± 0.11 -0.01 ± 0.07
Willamette (Mean) W04/AG01 7 0.6 ± 0.9 0.04 ± 0.08 0.06 ± 0.12 0.7 ± 1.1 0.07 ± 0.08 -0.02 ± 0.04Turtle River W02 1 0.4 ± 1.3 0.16 ± 0.12 -0.30 ± 0.21 0.1 ± 1.3 0.06 ± 0.10 0.10 ± 0.07Grant W03 2 1.1 ± 1.3 0.49 ± 0.12 -0.65 ± 0.21 0.5 ± 1.3 0.27 ± 0.10 0.22 ± 0.07
Henbury T12 4 18.4 ± 1.4 0.95 ± 0.10 -1.00 ± 0.11 17.4 ± 1.4 0.61 ± 0.08 0.34 ± 0.04Youanmi Z04 3 4.3 ± 1.3 0.50 ± 0.12 -0.74 ± 0.21 3.6 ± 1.3 0.26 ± 0.10 0.25 ± 0.07Dejbel-in-Azzene Z05 2 1.7 ± 1.3 0.16 ± 0.12 -0.32 ± 0.21 1.6 ± 1.3 -0.02 ± 0.10 0.11 ± 0.07
IVA ironsGibeon '102'b N02 1 nd ± 0.15 ± 0.15 -0.06 ± 0.23 nd ± 0.13 ± 0.15 0.02 ± 0.08Gibeon 'Railway'b S01 1 0.4 ± 2.2 0.19 ± 0.15 -0.18 ± 0.23 0.1 ± 2.2 0.15 ± 0.15 0.06 ± 0.08Gibeon 'Egg'b S02 1 2.8 ± 2.2 0.21 ± 0.15 -0.23 ± 0.23 2.6 ± 2.2 0.14 ± 0.15 0.08 ± 0.08Muonionalustab S03 1 2.7 ± 2.2 -0.02 ± 0.15 -0.05 ± 0.23 2.6 ± 2.2 -0.07 ± 0.15 0.02 ± 0.08Gibeon U03 1 0.0 ± 0.8 -0.04 ± 0.08 0.07 ± 0.05 0.1 ± 0.8 -0.02 ± 0.09 -0.02 ± 0.02Jamestown U04 1 6.8 ± 1.2 0.61 ± 0.12 -0.46 ± 0.21 6.2 ± 1.1 0.46 ± 0.10 0.22 ± 0.07
IVB ironsSanta Clarab M04 4 22.9 ± 2.2 0.96 ± 0.15 -1.36 ± 0.23 21.8 ± 2.2 0.58 ± 0.20 0.43 ± 0.07Tawallah Valleyb M05 2 9.4 ± 2.2 0.51 ± 0.15 -0.79 ± 0.23 8.6 ± 2.2 0.21 ± 0.15 0.26 ± 0.07Hobab Q01 2 14.7 ± 2.2 0.43 ± 0.21 -0.36 ± 0.23 14.4 ± 2.2 0.32 ± 0.26 0.12 ± 0.07Warburton Rangeb Q02 2 8.1 ± 2.2 0.45 ± 0.37 -0.58 ± 0.23 7.6 ± 2.2 0.25 ± 0.50 0.19 ± 0.07Warburton Rangeb R06 2 7.6 ± 2.2 0.55 ± 0.15 -0.61 ± 0.23 6.9 ± 2.2 0.30 ± 0.15 0.21 ± 0.07Iquiqueb R02 3 30.2 ± 2.2 1.73 ± 0.15 -1.27 ± 0.23 28.9 ± 2.2 1.32 ± 0.15 0.42 ± 0.07Dumontb R03 3 26.2 ± 2.2 1.37 ± 0.28 -1.13 ± 0.23 25.0 ± 2.2 1.01 ± 0.19 0.38 ± 0.07Skookumb Q04 3 7.3 ± 2.2 0.18 ± 0.15 -0.39 ± 0.30 6.9 ± 2.2 0.08 ± 0.15 0.13 ± 0.10Skookumb R05 2 7.4 ± 2.2 0.48 ± 0.15 -0.54 ± 0.23 6.9 ± 2.2 0.30 ± 0.15 0.18 ± 0.07Tlacotepecb R04 2 61.3 ± 2.2 2.76 ± 0.22 -2.63 ± 0.28 58.6 ± 2.2 1.91 ± 0.13 0.88 ± 0.10Weaver Mountainsb R08 2 7.6 ± 2.2 0.56 ± 0.44 -0.58 ± 0.23 7.0 ± 2.2 0.39 ± 0.32 0.19 ± 0.07Hoba U13 1 12.6 ± 1.7 0.61 ± 0.12 -0.48 ± 0.21 12.0 ± 1.9 0.45 ± 0.11 0.16 ± 0.07Tlacotepec U12 1 54.7 ± 1.7 2.49 ± 0.12 -2.50 ± 0.21 51.9 ± 1.9 1.65 ± 0.11 0.84 ± 0.07Warburton Range U16 1 5.4 ± 1.7 0.39 ± 0.12 -0.56 ± 0.21 4.9 ± 1.9 0.20 ± 0.11 0.19 ± 0.07
a Normalized to 198Pt/195Pt = 0.2145 (8/5) or 196Pt/195Pt = 0.7464 (6/5) using the exponential law. b Data from Kruijer et al. (2013a).
116 Resolved metal segregation ages of magmatic iron meteorites
Table 5-A4Tungsten isotope compositions of iron meteorites analysed by MC-ICPMS.
Meteorite ID N !182/183W (6/3)meas.a !182W (6/4)meas.
a !183W (6/4)meas.a !184W (6/3)meas.
a !182/183W (6/3)corr.a,c !182/184W (6/4)corr.
a,c
(±95% conf.) (±95% conf.) (±95% conf.) (±95% conf.) ±95% conf. ±95% conf.IIAB irons
Sikhote Alinb,d T01 7 -3.79 ± 0.11 -3.80 ± 0.06 -0.05 ± 0.12 0.03 ± 0.08 -3.79 ± 0.11 -3.80 ± 0.06Edmonton, Canadab,d T02 10 -3.39 ± 0.12 -3.38 ± 0.07 -0.02 ± 0.09 0.01 ± 0.06 -3.39 ± 0.12 -3.38 ± 0.07Braunaub,d T03 9 -3.32 ± 0.12 -3.40 ± 0.05 -0.08 ± 0.09 0.05 ± 0.06 -3.32 ± 0.12 -3.40 ± 0.05Ainsworthb T04 3 -4.70 ± 0.15 -4.82 ± 0.05 -0.10 ± 0.12 0.06 ± 0.08 -4.70 ± 0.15 -4.82 ± 0.05Mt. Joyb Z01 5 -3.91 ± 0.12 -3.88 ± 0.06 0.02 ± 0.09 -0.02 ± 0.06 -3.91 ± 0.12 -3.88 ± 0.06
IID ironsRodeoe R07 5 -3.16 ± 0.06 -3.01 ± 0.08 0.11 ± 0.10 -0.07 ± 0.07 -3.17 ± 0.06 -3.18 ± 0.10Rodeob,e S06 5 -3.06 ± 0.10 -2.96 ± 0.09 0.14 ± 0.09 -0.09 ± 0.06 -3.07 ± 0.10 -3.14 ± 0.10Carbob,e P01 5 -4.04 ± 0.08 -3.88 ± 0.12 0.11 ± 0.09 -0.08 ± 0.06 -4.05 ± 0.08 -4.05 ± 0.13Carbob,e P02 4 -4.14 ± 0.07 -4.02 ± 0.08 n.a. n.a. -4.15 ± 0.07 -4.19 ± 0.09Carboe P03 6 -4.27 ± 0.07 -4.07 ± 0.10 0.13 ± 0.07 -0.09 ± 0.05 -4.28 ± 0.07 -4.24 ± 0.11Carbob,e P04 5 -4.38 ± 0.05 -4.16 ± 0.13 n.a. n.a. -4.39 ± 0.05 -4.33 ± 0.14Carbob,e P05 5 -4.32 ± 0.10 -4.17 ± 0.06 n.a. n.a. -4.33 ± 0.10 -4.34 ± 0.08Carbob,e S05 5 -3.91 ± 0.08 -3.76 ± 0.06 n.a. n.a. -3.92 ± 0.08 -3.93 ± 0.08
IIIAB ironsFairview W05 6 -3.62 ± 0.07 -3.63 ± 0.04 -0.04 ± 0.04 0.03 ± 0.03 -3.62 ± 0.07 -3.63 ± 0.04Cape York CY01 T05 3 -3.43 ± 0.04 -3.41 ± 0.05 -0.01 ± 0.06 0.01 ± 0.04 -3.43 ± 0.04 -3.41 ± 0.05Cape York CY02b T06 7 -3.40 ± 0.06 -3.42 ± 0.08 -0.06 ± 0.17 0.04 ± 0.11 -3.40 ± 0.06 -3.42 ± 0.08Charcasb Z02 5 -3.90 ± 0.06 -3.97 ± 0.08 -0.06 ± 0.07 0.05 ± 0.05 -3.90 ± 0.06 -3.97 ± 0.08Boxhole Z06 6 -3.81 ± 0.08 -3.85 ± 0.05 -0.03 ± 0.07 0.03 ± 0.05 -3.81 ± 0.08 -3.85 ± 0.05Willamette W04 5 -3.29 ± 0.05 -3.29 ± 0.00 0.01 ± 0.04 0.00 ± 0.02 -3.29 ± 0.05 -3.29 ± 0.00Willamette (replicate) AG01 5 -3.34 ± 0.04 -3.29 ± 0.03 0.03 ± 0.06 -0.01 ± 0.04 -3.34 ± 0.04 -3.29 ± 0.03Willamette (Mean) W04/AG01 10 !"#"$ ± %#%" !"#&' ± %#%& %#%& ± %#%( !%#%$ ± %#%& !"#"$ ± %#%" !"#&' ± %#%&Turtle River W02 5 -3.42 ± 0.05 -3.45 ± 0.06 -0.02 ± 0.03 0.02 ± 0.03 -3.42 ± 0.05 -3.45 ± 0.06Grantb W03 5 -3.50 ± 0.07 -3.60 ± 0.06 -0.07 ± 0.08 0.05 ± 0.06 -3.50 ± 0.07 -3.60 ± 0.06Youanmi Z04 5 -3.69 ± 0.05 -3.67 ± 0.04 0.01 ± 0.03 -0.01 ± 0.02 -3.69 ± 0.05 -3.67 ± 0.04Dejbel-in-Azzeneb Z05 2 -3.48 ± 0.08 -3.50 ± 0.05 -0.02 ± 0.05 0.02 ± 0.05 -3.48 ± 0.08 -3.50 ± 0.05
IVA ironsGibeon '102'b,e N02 4 -3.22 ± 0.15 -3.31 ± 0.08 0.03 ± 0.12 -0.02 ± 0.08 -3.22 ± 0.15 -3.31 ± 0.08Gibeon 'Railway'b,e S01 5 -3.41 ± 0.11 -3.42 ± 0.08 -0.12 ± 0.10 0.08 ± 0.06 -3.41 ± 0.11 -3.42 ± 0.08Gibeon 'Egg'b,e S02 5 -3.45 ± 0.19 -3.44 ± 0.09 0.00 ± 0.13 0.00 ± 0.08 -3.45 ± 0.19 -3.44 ± 0.09Muonionalustab,e S03 5 -3.33 ± 0.07 -3.33 ± 0.07 -0.02 ± 0.09 0.01 ± 0.06 -3.33 ± 0.07 -3.33 ± 0.07
IVB ironsSanta Clarab,e M04 5 -3.66 ± 0.06 -3.48 ± 0.05 0.11 ± 0.08 -0.07 ± 0.05 -3.67 ± 0.06 -3.67 ± 0.05Tawallah Valleyb,e M05 5 -3.47 ± 0.05 -3.29 ± 0.12 0.10 ± 0.10 -0.07 ± 0.07 -3.48 ± 0.06 -3.48 ± 0.12Hobae Q01 5 -3.36 ± 0.13 -3.15 ± 0.14 0.12 ± 0.15 -0.08 ± 0.10 -3.37 ± 0.13 -3.34 ± 0.15Warburton Rangee Q02 5 -3.36 ± 0.07 -3.22 ± 0.06 0.11 ± 0.05 -0.07 ± 0.03 -3.37 ± 0.07 -3.41 ± 0.06Warburton Rangee R06 5 -3.47 ± 0.03 -3.31 ± 0.06 0.14 ± 0.06 -0.09 ± 0.04 -3.48 ± 0.04 -3.50 ± 0.07Iquiquee R02 5 -3.66 ± 0.06 -3.45 ± 0.05 0.16 ± 0.06 -0.11 ± 0.04 -3.67 ± 0.06 -3.64 ± 0.05Dumonte R03 5 -3.64 ± 0.06 -3.43 ± 0.05 0.16 ± 0.05 -0.11 ± 0.03 -3.65 ± 0.06 -3.63 ± 0.06Skookume Q04 5 -3.46 ± 0.07 -3.20 ± 0.06 0.18 ± 0.05 -0.12 ± 0.03 -3.47 ± 0.07 -3.39 ± 0.06Skookume R05 5 -3.44 ± 0.04 -3.30 ± 0.04 0.13 ± 0.04 -0.08 ± 0.03 -3.45 ± 0.04 -3.49 ± 0.05Tlacotepece R04 5 -4.22 ± 0.05 -4.06 ± 0.04 0.13 ± 0.05 -0.08 ± 0.03 -4.23 ± 0.05 -4.25 ± 0.05Weaver Mountainse R08 6 -3.40 ± 0.08 -3.26 ± 0.07 0.11 ± 0.09 -0.07 ± 0.06 -3.41 ± 0.08 -3.45 ± 0.08
a Normalized to 186W/184W = 0.92767 (6/4) or 186W/183W = 1.9859 (6/3) using the exponential law. b Tungsten isotope compositions affected by and corrected for a mass-independent effect on 183W (Section 5.5.1.3.3).c Tungsten isotope compositions corrected (if needed) for nucucleosynthetic W isotope anomalies. The nucleosynthetic W isotope anomalies of the IID and IVB irons were corrected using the following relations (Burkhardt et al., 2012a; Kruijer et al., 2013a, 2013b): !182/184W (6/4)corr = !182/184W(6/4)meas – [1.41 " !183W (6/4)Avg IVB or IID] and !182/183W(6/3)corr = !182/183W(6/3)meas – [–0.10 " !184W(6/3)Avg IVB or IID].d Data from Kruijer et al. (2012)e Data from Kruijer et al. (2013a)
Part A Chapter 5 117
Table 5-A5Tungsten isotope compositions of iron meteorites analysed by N-TIMS.
Meteorite ID N !182W (6/3)a !182W (6/4)a
(±95% conf.) (±95% conf.)IIAB irons
Filomena T10 2 -3.52 ± 0.05 -3.55 ± 0.06Negrillos T11 3 -3.38 ± 0.06 -3.38 ± 0.07Bennett County T13 3 -3.47 ± 0.05 -3.46 ± 0.07
IIIAB ironsCostilla Peak T09 1 -3.39 ± 0.07 -3.42 ± 0.09Henbury T12 1 -3.72 ± 0.05 -3.71 ± 0.06
IVA ironsGibeon U03 1 -3.29 ± 0.07 -3.21 ± 0.09Jamestown U04 1 -3.57 ± 0.09 -3.56 ± 0.12
IVB ironsHoba U13 1 -3.33 ± 0.04 -3.31 ± 0.05Tlacotopec U12 2 -4.24 ± 0.04 -4.24 ± 0.03Warburton U16 1 -3.32 ± 0.05 -3.29 ± 0.07
a Normalized to 186W/184W = 0.92767 (6/4) or 186W/183W = 1.9859 (6/3) using the exponential law.
118 Abundance and isotopic composition of Cd in iron meteorites
Chapter 6
Abundance and isotopic composition of Cd in iron meteorites
Kruijer, T.S.a,b, Sprung, P.,a,b Kleine, T.b, Leya, I.c, , Wieler, R.a
aETH Zürich, Institute of Geochemistry and Petrology, Zürich, Switzerland. bWestfälische Wilhelms-Universität Münster, Institut für Planetologie, Münster, Germany.
cUniversity of Bern, Space Research and Planetary Sciences, Bern, Switzerland.
Accepted for publication in Meteoritics and Planetary Science (2013)
Abstract Cadmium is a highly volatile element and its abundance in meteorites may help to better understand volatility-controlled processes in the solar nebula and on meteorite parent bodies. The large thermal neutron capture cross-section of 113Cd suggests that Cd isotopes might be well suited to quantify neutron fluences in extraterrestrial materials. The aims of this study are (i) to evaluate the range and magnitude of Cd concentrations in magmatic iron meteorites, and (ii) to assess the potential of Cd isotopes as a neutron dosimeter for iron meteorites. Our new Cd concentration data determined by isotope dilution demonstrate that Cd concentrations in iron meteorites are significantly lower than in some previous studies. In contrast to large systematic variations in the concentration of moderately volatile elements like Ga and Ge, there is neither systematic variation in Cd concentration amongst troilites, nor amongst metal phases of different iron meteorite groups. Instead, Cd is strongly depleted in all iron meteorite groups, implying that the parent bodies accreted well above the condensation temperature of Cd (i.e., ! 650 K) and thus incorporated only minimal amounts of highly volatile elements. No Cd isotope anomalies were found, whereas Pt and W isotope anomalies for the same iron meteorite samples indicate a significant fluence of epithermal and higher energetic neutrons. This observation demonstrates that owing to the high Fe concentrations in iron meteorites, neutron capture mainly occurs at epithermal and higher energies. The combined Cd-Pt-W isotope results from this study thus demonstrate that the relative magnitude of neutron capture-induced isotope anomalies are strongly affected by the chemical composition of the irradiated material. The resulting low fluence of thermal neutrons in iron meteorites and their very low Cd concentrations make Cd isotopes unsuitable as a neutron dosimeter for iron meteorites.
Part A Chapter 6 119
6.1 Introduction Magmatic iron meteorites are interpreted as remnants of the metal cores of planetesimals (e.g., Scott, 1972) that had accreted in the first few million years of solar system history (e.g., Blichert-Toft et al., 2010; Chen and Wasserburg, 1990; Kleine et al., 2005; Smoliar et al., 1996). Their chemical and isotopic characteristics provide insights into the high temperature history of some of the earliest protoplanets, and also provide important constraints on the solar nebula environment in which these planetary bodies accreted. Most meteorites and the terrestrial planets are depleted in volatile elements relative to the bulk solar composition as estimated from CI chondrites (e.g., Davis, 2006). Both Ga and Ge are moderately volatile elements with 50% condensation temperatures (hereafter ‘T50%’) of 885 and 971 K, respectively (Lodders, 2003). Different iron meteorite groups are distinguished by variable fractionations in Ga/Ni and Ge/Ni, indicating varying degrees of moderately volatile element depletion (e.g., Scott and Wasson, 1975). For instance, the IVA and IVB irons are most strongly depleted in moderately volatile elements with Ga/Ni and Ge/Ni ratios that are several orders of magnitude lower than values in CI chondrites. In contrast, the IIAB irons have near-chondritic Ga/Ni and Ge/Ni, and thus are relatively rich in moderately volatile elements.
Investigating the abundance and isotopic composition of Cd (Z = 48) in iron meteorites is of relevance for two main reasons. First, Cd is an even more volatile element than Ga and Ge with a 50% condensation temperature (T50%) of 650 K (Lodders, 2003). Hence, investigating the abundance of Cd in iron meteorites might provide insight into volatility-controlled processes in the solar nebula, in the respective parent bodies, or the precursor material that accreted to these planetary bodies (Rosman and Delaeter, 1974; Wombacher et al., 2003, 2008). Specifically, studying the variability of Cd concentrations in different groups of iron meteorites allows evaluating as to whether the systematic variations in moderately volatile elements like Ga and Ge also extend to more volatile elements like Cd. To mark the distinction with moderately volatile elements, Cd will henceforth be referred to as a ‘highly volatile’ element.
Second, due to the very large thermal (i.e., neutron energies < 0.5 eV) neutron capture cross-section of 113Cd (e.g., ∽27900 barns, integrated over 1"10-5 to 0.5 eV at T = 300 K, ENDFB-VI.8 300 K library) the reaction 113Cd(n,γ)114Cd can induce large deficits in 113Cd and corresponding excesses in 114Cd (e.g., Leya and Masarik, 2013). For instance, large 113Cd deficits exceeding 0.3% were found in some lunar samples (Sands et al., 2001; Wombacher et al., 2008), reflecting high fluences of thermal neutrons produced during cosmic ray-exposure on the lunar surface. Thus, Cd isotopes can potentially be used as a neutron dosimeter for extraterrestrial materials. Such neutron dosimetry is important for dating iron meteorites with some short-lived nuclides, such as the 182Hf-182W (e.g, Kleine et al., 2005; Leya et al., 2003; Markowski et al., 2006; Masarik 1997; Qin et al., 2008) and 107Pd-107Ag systems (Leya and Masarik, 2013). For instance, neutron capture effects resulting from cosmic ray exposure of the iron meteoroids caused net decreases of the 182W/184W ratios up to 1 #-unit (1 # = 0.01%). These cosmic ray-induced W isotope shifts lead to an apparent decrease in core formation model ages of up to 10 Myr, severely limiting the accuracy of Hf-W ages of iron meteorites. Obtaining accurate core formation ages of iron meteorite parent bodies using the Hf-W system requires correction of neutron capture effects, ideally using a direct neutron dosimeter (Masarik, 1997). Recently, Pt isotopes were shown to be an excellent proxy for neutron capture at epithermal neutron energies (i.e., > 0.5 eV) and higher, that is, for neutrons in the energy range that mostly affects the W isotope compositions of iron meteorites (Kruijer et al., 2013a; Wittig et al., 2013). However, given the more than two orders of magnitude higher
120 Abundance and isotopic composition of Cd in iron meteorites
thermal neutron capture cross section of 113Cd as compared to Pt isotopes, Cd isotope compositions might provide an additional neutron dose monitor for iron meteorites.
The investigation of neutron capture effects on Cd isotopes in iron meteorites is not only important for its potential as a neutron fluence dosimeter for Hf-W isotope chronometry, but also for assessing the effects of target chemistry on the resulting energy distribution of secondary neutrons. For instance, Kollár et al. (2006) studied the fluences of secondary neutrons in meteoroids due to interaction with galactic cosmic rays (GCR) with a physical model. In Fe-rich meteoroids, the ratio of the fluence of neutrons with energies above thermal to the fluence of neutrons having thermal energies appears considerably higher than in stony meteorites. This prediction is consistent with the observed neutron capture-induced shifts in Hf and Sm isotope compositions in stony and stony-iron meteorites (Sprung et al., 2010). In metal-rich meteorites (i.e., mesosiderites containing ~50 wt-% Fe metal) the relative isotopic shifts are larger on Hf (a proxy for epithermal neutrons) than on Sm (a proxy for thermal neutrons), whereas stony meteorites only show resolved isotopic shifts on Sm. Studying iron meteorites with combined Cd (mainly thermal neutrons) and Pt-W (mainly neutrons > 0.5 eV) isotope analyses will, therefore, expand the range of target materials for which the effects of sample matrix on the energy distribution of secondary neutrons can be evaluated to (almost) pure Fe-metal.
Here we report high-precision Cd concentration and combined Cd, Pt, and W isotope data for several troilite and metal samples from iron meteorites. These data are used to (i) evaluate the range of Cd concentrations in magmatic irons and, hence, to assess the magnitude of highly volatile element depletion of their parent bodies, (ii) investigate the potential of Cd isotopes as a neutron dosimeter for iron meteorites, and (iii) interpret the combined Cd-Pt-W isotope results with respect to the neutron energy distribution of the studied iron meteorites. As Cd behaves chalcophile under most conditions (e.g., Ballhaus et al., 2013; Lagos et al., 2008), its concentration is expected to be significantly higher in S-bearing minerals than in Fe-Ni metal. In addition to the pure metal phase of iron meteorites, troilite (FeS) nodules contained therein were also investigated.
6.2 Analytical methods
6.2.1 Sample preparation and chemical separation of Cd Large metal (1-20 g) and troilite samples (~0.5-4 g) from several iron meteorite groups (IAB, IC, IIAB, IIIAB, IVA, UNG) were selected for this study (Table 6.1). Metal and troilite samples were cut using a handsaw, and cleaned with ethanol and de-ionized water in an ultrasonic bath. The outermost ~10-20% of each sample were removed by leaching in dilute HCl-HNO3. Complete digestion of the metal samples was accomplished in ca. 15 ml 6 M HCl-0.06 M HF in Savillex® vials at 130°C, whereas troilite specimens were digested in concentrated HCl or HCl-HNO3 (3:1) at 130-150°C for several days. Samples for isotope dilution analyses were spiked prior to digestion and hence dissolved separately from un-spiked samples for Cd isotope analyses.
The purification of Cd for both spiked and unspiked samples involved several anion exchange chromatography steps. In the first step the samples were loaded onto a pre-cleaned anion exchange column (10 ml BioRad® AG1X8, 200-400 mesh) in 250 ml 0.5 M HCl, followed by an additional 50 ml rinse to wash off most matrix elements. Cadmium was subsequently eluted in 25 ml 2 M HNO3. The second step largely follows the procedure described by Wombacher et al. (2003). Samples were loaded in 10 ml 3 M HCl onto pre-
Part A Chapter 6 121
cleaned anion columns (2 ml BioRad® AG1X8, 100-200 mesh). Remaining matrix elements were washed off using progressively stronger HCl (0.5 M, 1 M, 2 M). After eluting Zn in 0.5 M HNO3-0.1 M HBr, Cd was eluted in 8 ml 2 M HNO3 in the case of spiked samples. However, for un-spiked samples Cd was eluted using 40 ml 4 M HF, which allowed an initial separation of Sn from Cd. After this second step the samples were dried and treated with HNO3-H2O2 several times.
As the total Cd amounts were extremely low for most of the investigated samples, even small Sn blanks from any labware or reagents lead to significant interference correction on measured Cd isotope ratios, affecting the precision and accuracy of the Cd isotope data. For spike-free samples, three additional chromatography steps were, therefore, performed to remove residual Sn. The third and fourth chromatography steps largely follow previously described procedures (Wombacher et al., 2003). For the third step, custom-made shrink-Teflon® columns (2 mm ⦰ " 5 cm) filled with 200 $L Eichrom TRU-Spec® resin were used, while the fourth step utilizes a miniaturized column (1 mm ⦰ ! 3.5 cm) using 100 $L TRU-Spec® resin. At strong HCl molarities, Cd is directly eluted from the Eichrom TRU-Spec® resin, while Sn is retained on the column. On all TRU-Spec® columns samples were loaded in 0.2 ml 8 M HCl, rinsed with 0.2 ml 8 M HCl and then with 1.4 ml (third column) or 1 ml (fourth column) 6 M HCl. The third chromatography step was repeated twice. Because the Sn blanks after the TRU-Spec® columns still appeared too high, a fifth chromatography step was introduced. On anion resin Cd is efficiently eluted in 4 M HF, while Sn is retained on the resin. The Cd samples were thus loaded onto anion exchange columns (200 $L Biorad® AG1X8, 200-400 mesh) in 0.2 ml 0.5 M HCl, followed by a 1 ml 0.5 M HCl rinse. Cadmium was then eluted using 3.5 ml 4 M HF, dried down, and re-dissolved in the measurement solution of ~0.2 ml 0.5 M HNO3. The combined yield for the various chemical separation steps was 80-100%.
The total procedural blanks for Cd isotope composition measurements were ~30 pg Cd and are negligible. For the Cd isotope concentration measurements total procedural blanks were 2-12 pg Cd, and were insignificant for most troilite samples. However, for all metal and some troilite samples the blanks are the major source of uncertainty in the calculated Cd concentration.
6.2.2 Cadmium concentration determination by isotope dilution (ID) For concentration determination, samples (0.4-0.7 g) were spiked with a 106Cd-enriched isotope tracer prior to digestion in HCl-HNO3 (3:1). The 106Cd-spike (Alfa Aesar®) was calibrated against a pure certified Cd solution standard (NIST3108). To facilitate spike-sample equilibration, the 106Cd-spike was added to the samples before the initial dissolution of the troilite or metal samples. This also ensures that possible Cd loss during digestion of the samples would not affect the accuracy of the Cd concentration data. The 106Cd/111Cd ratios of spiked samples were measured using a Nu Plasma MC-ICPMS at ETH Zürich, equipped with a Cetac® Aridus sample introduction system. Baselines were measured on-peak for 120 s using an acid blank solution at the same molarity as used for running standards and samples (0.5 M HNO3). All Cd isotope ratios were corrected for instrumental mass bias by normalizing to 116Cd/112Cd = 0.3104 using the exponential law (see also below).
To estimate the accuracy and reproducibility of the method, five aliquots of a Cd-free metal standard (NIST129c) were doped with a known amount of a certified pure Cd standard (NIST3108), consecutively spiked, digested, and processed through the full separation procedure. The external reproducibility of these analyses—each of which was measured 3-4
122 Abundance and isotopic composition of Cd in iron meteorites
times—corresponds to ±0.3% (2SD). The mean Cd concentration thus obtained (i.e., 51.70 ppb; n=5) is identical to the certified Cd concentration of the (diluted) certified NIST3108 standard (i.e., 51.64 ppb), demonstrating that our method yields accurate and reproducible results. (Note that the concentration of ~50 ppb for the NIST3108 standard was chosen based on previously reported Cd concentrations for iron meteorites (Rosman and Delaeter, 1974; Schmitt et al., 1963)). However, the major source of uncertainty in the Cd concentration data is the correction for blank, which for some (especially metal) samples was as high as 70 %. The uncertainty on the blank correction was estimated by assuming an uncertainty on the blanks of 50%. The reported uncertainties on the Cd concentrations represent the propagated uncertainties on ID measurements (0.3% 2SD) and on blank corrections (Table 6.1).
6.2.3 Cadmium isotope composition measurements Cadmium isotope compositions of four troilite samples were measured on a ThermoScientific® Neptune Plus MC-ICPMS at the University of Münster. Samples were introduced into the mass spectrometer using a Cetac® Aridus II desolvator. Total Cd ion beams varied between 6.0"10-11 and 1.2"10-10 A at a 100 $L/min uptake rate using solutions containing ~5-10 ppb Cd. Measurements consisted of 20 cycles of 4.2s each. Baselines were measured on-peak for 120 s using an acid blank solution in the same concentration as used for the samples (0.5 M HNO3). Each sample measurement was bracketed by measurements of a terrestrial solution standard (NIST3108) run at the same intensities as the samples.
Isobaric interferences from 112Sn and 114Sn on the Cd isotopes of interest were corrected by monitoring interference-free 117Sn. Interference corrected 116Cd/112Cd and values of the exponent β116/112 of the exponential mass fractionation law were calculated using an iterative approach: first, an initial β116/112 value was calculated using the raw 116Cd/112Cd ratio. This value was subsequently used to calculate an interference-corrected 116Cd/112Cd, which then allowed calculating a new β116/112 value. This procedure was repeated until the calculatedβ116/112 converged to a constant value. Possible small isobaric interferences from 113In on Cd isotope ratios were corrected by monitoring 115In and were smaller (<0.3 #) than the analytical uncertainty of the reported Cd isotope compositions.
The measured 113Cd/112Cd and 114Cd/112Cd ratios of the samples are reported as #i/112Cd (i.e., 0.01% deviations) relative to the bracketing Cd solution standard. The external reproducibility of the Cd isotope composition measurements was estimated by repeated analyses of a terrestrial metal standard (NIST129c) that was doped with Cd from a pure solution standard (Alfa Aesar), and processed through the full chemical separation. The mean values obtained for a 10 ppb solution of #113Cd = %0.05±0.45 and #114Cd = +0.06±0.28 (2 s.d., n=7) are indistinguishable from the Cd isotope composition of the bracketing standard solution, indicating that the analyses are accurate. However, for some of the meteoritic metal and troilite samples less Cd was available (e.g., as little as 0.6 ng Cd for Cape York troilite). Hence, additional uncertainty arising from lower signal intensities for these samples was estimated using the empirical relation between the internal measurement precision (2 SE) and external uncertainties (2 s.d.) obtained for Cd solution standards measured at different concentrations.
The Sn/Cd ratios of the Cd cuts obtained after the chemical separation ranged from ~0.008 for Merceditas to ~0.15 for Cape York, indicating that despite several clean up steps aimed at removing Sn from the samples, some Sn (~80-100 pg) remained in the Cd cuts. Thus, significant interference corrections of 3-110 #-units on 113Cd/112Cd were required for the iron meteorite samples (Table 6.2). However, measurements of a Cd standard solution with
Part A Chapter 6 123
variable amounts of admixed Sn demonstrate that large Sn interferences (up to Sn/Cd ! 0.1) can accurately be corrected for (Fig. 6.2). Due to the required interference corrections, the reported uncertainties for the Cd isotope data of the iron meteorite samples include the uncertainty of the Cd isotope measurements and an additional uncertainty derived from the Sn interference corrections. This additional uncertainty was estimated using an empirical correlation between the 2 s.d. vs. the Sn/Cd obtained for replicate measurements of Cd solution standards with variable amounts of admixed Sn.
6.2.4 Pt and W isotope measurements Most of the Pt and W isotope data reported here for metal samples collected adjacent to the troilite samples used for Cd isotope analyses were initially reported in Kruijer et al. (2013a; 2012; 2013b). Only Merceditas (IIIAB) was newly analyzed in this study following the methods outlined in Kruijer et al. (2013a; 2012).
Table 6.1Cadmium concentrations determined by isotope dilution.
Meteorite ID Cd [ppb] Blank correction (±2SD) [%]
IAB ironsToluca (TR) E01 4.86 ± 0.07 2.2
IC ironsArispe (M) D01 0.004 ± 0.006 72
IIAB ironsSikhote Alin (TR) G03 0.69 ± 0.01 1.6North Chile (M) J04 0.081 ± 0.001 2.7
IIIAB ironsCape York (TR) F01 1.40 ± 0.09 11Cape York (TR) G02 0.40 ± 0.01 1.8Merceditas (TR) J01 4.92 ± 0.04 0.3Djebel in-Azzene (TR) J02 0.11 ± 0.02 30
IVA ironsMuonionalusta (TR) H01 0.043 ± 0.003 13Gibeon (TR) G01 0.46 ± 0.01 2.8
Ungrouped ironsTishomingo (M) G04 0.003 ± 0.002 61
Terrestrial standardsPyrrhotine (terr.) F04 9.32 ± 0.16 2.1
TR = troilite nodule, M = metal sample.
124 Abundance and isotopic composition of Cd in iron meteorites
6.3 Results
6.3.1 Cd concentrations Cadmium isotope dilution results for the investigated samples are presented in Table 6.1 and 6.1. Cadmium concentrations in troilite samples vary between 0.1 and 4.9 ppb (Fig. 6.1). The three metal samples contain much lower amounts of Cd (3- 40 ppt) than the troilite samples. As accurate quantification of such extremely low concentrations is problematic (i.e, due to the large blank corrections), only three metal samples were analyzed in this study. The Cd concentration in troilite nodules varies significantly (up to two orders of magnitude) within a single meteorite class (e.g., IIIAB irons) and even within a single meteorite (Cape York Agpalilik). The troilite samples do not show systematic differences between different iron meteorite groups, whereas the number of metal samples is too low to discern such possible differences.
Fig. 6.1: Cadmium concentrations from this study plotted vs. CI-normalized Ge/Ni in the metal of the respective meteorite. The Ge and Ni concentrations of iron meteorites are from Kracher et al. (1980), Lovering et al. (1957), Wasson (1999), Wasson and Richardson (2001), Wasson et al. (2007), and Wasson (2011), and those for CI chondrites from Palme and Jones (2003).
10–5 10–4 10–3 10–2 10–1 100 10110–3
10–2
10–1
100
101
102
103
Cd [p
pb] Toluca, IAB
N. Chile, IIAB
Arispe, IC
Tishomingo, UNG
Gibeon, IVA
Muonionalusta, IVA
Sik. Alin, IIABCape York, IIIAB
Merceditas, IIIAB
Djebel-in-Azz., IIIAB
Troilite
Metal
(Ge/Ni)CI norm.
CI chondrites
Part A Chapter 6 125
Tabl
e 6.
2
Cd,
Pt,
and
W is
otop
e co
mpo
sitio
ns o
f mag
mat
ic ir
on m
eteo
rites
.
Met
eorit
e ID
!113/
112 C
d (6
/2)a
!114/
112 C
d (6
/2)a
!196 Pt
(6/5
)a!18
2 W (6
/4)a
"!18
2 WG
CR
b"
!11
3 Cd i-
corr.
c
±2SD
±2SD
±95%
con
f.±9
5% c
onf.
±95%
con
f.II
AB
iron
sSi
khot
e Alin
G03
0.1
±1.
70.
0±
1.6
0.27
±0.
07-3
.80
±0.
04-3
.80
±0.
0456
IIIA
B ir
ons
Cap
e Yor
k ('C
Y02
')G
02-2
.1±
2.5
-1.4
±2.
80.
02±
0.03
-3.4
2±
0.08
-3.4
2±
0.08
110
Mer
cedi
tas
K06
-0.1
±0.
50.
1±
0.7
0.06
±0.
07-3
.49
±0.
08-3
.49
±0.
083.
3IV
A ir
ons
Gib
eon
('Egg
')G
01-1
.1±
1.8
-0.9
±1.
90.
08±
0.08
-3.4
4±
0.09
-3.4
4±
0.09
48
Cad
miu
m is
otop
e co
mpo
sitio
ns w
ere
mea
sure
d in
troi
lite,
whi
le P
t and
W is
otop
e co
mpo
sitio
ns w
ere
dete
rmin
ed in
adj
acen
t met
al p
iece
s.a N
orm
aliz
ed to
116 C
d/11
2 Cd
= 0.
3104
(6/2
), 19
8 Pt/19
5 Pt =
0.2
145
(8/5
), or
186 W
/184 W
= 0
.927
67 (6
/4).
b "!18
2 WG
CR =
!18
2 Wm
eas. #
!18
2 Wpr
e-ex
posu
re. P
re-e
xpos
ure !18
2 W v
alue
s for
iron
met
eorit
e gr
oups
are
from
Kru
ijer e
t al.
(201
3a, 2
013b
).c "!11
3 Cd i-
corr.: M
agni
tude
of S
n-in
terf
eren
ce c
orre
ctio
n on
!11
3 Cd.
Unc
erta
intie
s on !i C
d ar
e ba
sed
on th
e 2S
D o
btai
ned
for r
eplic
ate
anal
yses
of t
he te
rres
trial
met
al st
anda
rd (N
IST1
29c)
,an
d ad
ditio
nal u
ncer
tain
ty d
eriv
ing
from
low
Cd
amou
nts a
nd S
n in
terf
eren
ce c
orre
ctio
ns (s
ee m
ain
text
).W
ith th
e ex
cept
ion
of M
erce
dita
s all
Pt a
nd W
dat
a ar
e fr
om K
ruije
r et a
l. (2
012,
201
3a, 2
013b
).
126 Abundance and isotopic composition of Cd in iron meteorites
6.3.2 Cd, Pt and W isotope compositions The Cd, Pt, and W isotope compositions of the studied troilite samples are presented in Table 6.2 and Fig. 6.2 and 6.3. The extremely low Cd concentrations (section 6.3.1) limited the Cd isotope composition analyses to very large troilite inclusions (>3 g). The availability of such samples is limited and hence only four troilite inclusions were analyzed for their Cd isotopic composition. All samples show #113Cd and #114Cd indistinguishable from the terrestrial value (Fig. 6.2, 6.3a). As discussed above, the slight #113Cd deficit for Cape York (Agpalilik) is very likely caused by the high Sn/Cd in the Cd cut of this sample (Fig. 6.2). The Cape York data point is therefore not shown in Fig. 6.3a. In contrast (Fig. 6.3b), some samples (in particular Sikhote Alin) show excesses in #196Pt (+0.02 to +0.30) and neutron capture effects on 182W/184W as evident from resolved Δ#182WGCR (%0.1 to %0.4). The Δ#182WGCR values represent the difference between the measured #182W of the sample and the ‘pre-exposure’ #182W (i.e., the W isotopic composition prior to cosmic ray exposure) determined for the corresponding iron meteorite group from Kruijer et al. (2013a, 2013b). The Δ#182WGCR thus represents the isotopic shift in "182W that is only due to cosmic ray interactions. The samples show a well-defined correlation in #196Pt vs.Δ#182WGCR space (Fig. 6.3b).
Fig. 6.2: Interference-corrected !113Cd plotted vs. the Sn/Cd in the respective measurement solutions. Shown are data for (i) Cd solution standards with variable amounts of admixed Sn, and (ii) the iron meteorites analyzed in this study. The thin grey bar indicates the measurement precision (~0.3 !-units, 2s.d.) based on replicate measurements of the terrestrial metal standard (NIST129c). Error bars indicate external uncertainties (2s.d.) and include additional uncertainty deriving from Sn interference corrections and the low Cd amounts available for analyses. The !113Cd data can be accurately corrected for Sn interference up to Sn/Cd " 0.1. Only Cape York has higher Sn/Cd, which likely explains the slight !113Cd deficit for this sample.
Sn/Cd
113 C
d
Solution standardsIron meteorites
Gibeon
Sikhote Alin
MerceditasCape York
10–3 10–2 10–1 100–25
–20
–15
–10
–5
0
5
Part A Chapter 6 127
6.4 Discussion
6.4.1 Cadmium abundances in the parent bodies of iron meteorites All three newly determined Cd concentrations of metal phases of iron meteorites from different groups are between four and six orders of magnitude below those of CI chondrites (Palme and Jones, 2003). The Cd concentrations are also markedly lower than values reported in two previous studies employing instrumental-neutron-activation (INAA) analyses (Rosman and Delaeter, 1974; Schmitt et al., 1963), but in agreement with another study that used thermal ionization mass spectrometry (TIMS) (Chen and Wasserburg, 1983). The cause of this discrepancy remains unclear, but may relate to large blank corrections required for the INAA data (e.g., up to several 100%; Rosman and DeLaeter, 1974).
The uniformly low Cd concentrations reported here suggest that either (i) significant fractions of highly volatile elements like Cd partitioned into the silicate portion of the iron meteorite parent bodies, (ii) iron meteorite parent bodies did not incorporate highly volatile elements like Cd in the first place, i.e., implying that Cd prevailed in the gas phase during planetesimal accretion, or (iii) Cd was lost from the parent body by reheating at some stage during its history, e.g., due to impact volatilization (e.g., Wasson et al., 2006). Below we briefly evaluate each of these scenarios.
We first consider possibility (i), i.e., that the iron meteorite parent bodies accreted from material with chondritic proportions of (volatile) elements, but that Cd primarily partitioned into the silicate mantle during metal-silicate separation. In an experimental study, Ballhaus et al. (2013) have shown that Cd metal/silicate partition coefficients are temperature-dependent and become >1 above ca. 1550 °C. Core segregation in iron meteorite parent bodies may have occurred at comparably high temperatures (Taylor, 1992), so that Cd should preferentially have partitioned into the metal phase. Furthermore, sulfide-silicate partition coefficients for Cd reported by Lagos et al. (2008) are significantly larger than 1, indicating that Cd behaves strongly chalcophile under conditions relevant for core formation. Thus, upon significant partial melting of planetesimals, Cd would predominantly partition into the metal- and sulfide-dominated phases, which would contribute significant amounts of Cd to the metal cores. The Cd concentrations of troilites from the IIIAB irons examined here vary significantly, even for different nodules contained in a single meteorite like Cape York (Fig. 6.1). This observation likely indicates that those troilite inclusions represent trapped S-rich melt liquid pockets that were isolated at distinct times during ongoing fractional crystallization of the metal core (e.g., Ulff-Møller, 1998; Wasson, 1999). This makes it difficult to estimate the Cd concentrations representative of the entire metal core. Nevertheless, given the chalcophile and siderophile character of Cd, the low Cd concentrations of the troilite and metal samples of magmatic iron meteorites reported here reflect intrinsically low Cd abundances in the iron meteorite parent bodies, indicating that these bodies are strongly depleted in volatile elements.
Concerning options (ii) and (iii), i.e., that volatility controlled the concentration of Cd in iron meteorites, a comparison of highly volatile Cd to the moderately volatile elements Ga and Ge and to the highly volatile element Tl appears useful. The concentrations of the strongly siderophile and moderately volatile elements Ga (T50% = 971 K; Lodders, 2003) and Ge (Tc = 885 K) show order-of-magnitude variations between different iron meteorite groups, a systematics that is also used for classifying iron meteorites (Scott and Wasson, 1975). The range of CI-normalized Ge/Ni and Ga/Ni ratios shown by iron meteorites are largely governed by volatility-controlled processes and are largely unaffected by parent body
128 Abundance and isotopic composition of Cd in iron meteorites
processes like fractional crystallization (Davis, 2006; e.g., Scott and Wasson, 1975). For example, the IIAB irons exhibit roughly uniform Ge/Ni and Ga/Ni ratios that are similar to those obtained for CI chondrites (Palme and Jones, 2003), whereas for the IVA irons these ratios are #2’000 times lower (see also Fig. 6.1). The strong variations in concentrations of the moderately volatile elements Ga and Ge in distinct iron meteorite parent bodies imply processing of the precursor material of these bodies at temperatures above T50% of the moderately volatile elements and, hence, well above T50% of the highly volatile elements.
Neither the Cd concentrations of the troilite nodules nor the Cd concentrations of the metal samples define clear-cut correlations with the average (Ge/Ni)CI-norm of the respective iron meteorite group, i.e., Cd concentrations seem to lack systematic variations between different iron meteorite groups. Instead and in contrast to the strongly variable depletions of moderately volatile elements, Cd shows very strong depletions relative to CI chondrites for all investigated iron meteorite groups by factors ~102-104 for the troilite and ~104-106 for metal samples (Fig. 6.1). Thallium, like Cd a highly volatile element, shows similar low concentrations for IIAB and IIIAB iron meteorites (Nielsen et al., 2006). Hence, the parent bodies of magmatic iron meteorites all appear to be strongly depleted in highly volatile elements, such as Cd, indicating that these bodies either accreted only minor amounts of volatile elements (case ii), or lost a significant fraction of their volatile elements due to a reheating event (case iii). Although the latter option cannot be excluded, it seems unlikely that such secondary losses resulted in an almost quantitative removal of Cd. Hans et al. (2013), based on Rb-Sr isotope systematics of eucrites and angrites, came to a similar conclusion for Rb in the parent bodies of these differentiated achondrites. We, therefore, conclude that Cd was never incorporated in significant amounts in the parent bodies of the magmatic irons.
The parent bodies of magmatic iron meteorites probably accreted in less than #1.5 Myr after CAI formation (Blichert-Toft et al., 2010; Kruijer et al., 2012, 2013a; Wittig et al., 2013). Possibly, ambient temperatures in the solar protoplanetary disk were higher at those early stages of planetesimal accretion and sufficiently high that volatile elements like Cd remained almost completely in the gas phase (e.g., Davis, 2006). Only minimal amounts of volatile elements would then have accreted to the parent bodies of the magmatic iron meteorites. At the same time, temperatures might have already been low enough that the moderately volatile elements could at least partly condense. This could explain why some iron meteorite parent bodies (e.g., the IIAB and IIIAB parent bodies) incorporated appreciable amounts of moderately volatile elements such as Ge or Ga. This implies that accretion of the more strongly volatile element depleted iron meteorite parent bodies (i.e., the IVA and IVB parent bodies) or their precursor material occurred at higher temperatures, at which less of the moderately volatile elements had condensed. However, the uniform and strong depletion of volatile elements like Cd in all magmatic iron meteorites indicates that their parent bodies all accreted at temperatures above the condensation temperatures of Cd (T50%! 650 K).
6.4.2 Cd isotopes as a monitor for neutron capture in iron meteorites The low Cd abundances in the metal samples restricted the Cd isotope composition measurements to large troilite inclusions that contained sufficient amounts of Cd. However, even for these samples the amount of Cd available for measurement did not exceed 1.5 ng (except for Merceditas), limiting the precision that could be achieved for Cd isotope measurements to ~1-3 #-units (2 SD). Within this precision no resolvable neutron capture-induced isotope anomalies in Cd were detected. In contrast, the Pt and W isotope compositions, measured in metal pieces sampled in immediate contact to the troilite nodules,
Part A Chapter 6 129
exhibit neutron capture-induced isotope shifts, as is most evident from the well-correlated #196Pt and Δ#182WGCR variations of these samples (Fig. 6.3b). Thus, in spite of clear evidence for a significant fluence of epithermal and higher energetic neutrons especially for Sikhote Alin, no resolvable Cd isotope anomalies could be detected in the iron meteorite samples. Consequently, Cd isotopes do not provide a useful tool to monitor and correct for neutron-capture effects on W isotope compositions. This in part is due to the low abundances of Cd in magmatic iron meteorites but as will be shown below also reflects the different neutron energies at which Cd and W are sensitive to neutron capture reactions.
Fig. 6.3: Cd, Pt, and W isotope data for iron meteorite samples from this study. (a) !113Cd vs. #!182WGCR, (b) !196Pt vs. #!182WGCR. Except for Merceditas, all W isotope data are from Kruijer et al. (2012), and the Pt isotope data from Kruijer et al. (2013a, 2013b). Error bars indicate external uncertainties (see Table 6.2). Cape York is not plotted in Fig 6.3a because of its high Sn/Cd and consequently large Sn interference correction. The model results from Leya and Masarik (2013) for neutron capture effects on Pt, Cd, and W isotopes in iron meteorites (radii 10-120 cm), and on Cd and W isotopes in ordinary chondrites for Ta/W = 0.15 (radii 10-120 cm), are shown as white circles.
6.4.3 Secondary neutron capture in iron meteorites: Importance of target chemistry Isotopic anomalies induced by secondary neutron capture have been predicted and measured for many elements in a variety of meteoritic and lunar materials (Hidaka et al., 2000; Kollár et
–0.6–0.4–0.20–0.1
0
0.1
0.2
0.3
0.4
0.5
–0.6–0.4–0.20–4
–2
0
2
4
113 C
d Merceditas
Cape York
Gibeon
196 P
t Sikhote Alin
182WGCR
Gibeon
(a)
182WGCR
(b)
Sikhote Alin
Merceditas
Ordinary chondrites(Ta/W = 0.15)
Iron meteorites
Iron meteorites
130 Abundance and isotopic composition of Cd in iron meteorites
al., 2006; Kruijer et al., 2013a; Leya et al., 2000; Leya et al., 2003; e.g., Lingenfelter et al., 1972; Masarik, 1997; Nyquist et al., 1995; Sands et al., 2001; Sprung et al., 2010; Wittig et al., 2013). The relative and absolute magnitude of these anomalies varies strongly and is not only a function of the duration of cosmic ray exposure, the geometry and size of the irradiated object, and the sampling depth therein, but is also strongly controlled by the chemical composition of the target material (e.g., Kollár et al., 2006; Leya and Masarik, 2013; Lingenfelter et al., 1972; Nyquist et al., 1995; Sprung et al., 2010). For example, because Fe strongly absorbs neutrons at epithermal energies and is an inefficient moderator, the probabilities of thermal neutron capture and the total thermal neutron flux are expected to be lower in Fe dominated than in silicate dominated target materials (Eberhardt et al., 1963; Spergel et al., 1986; Kollár et al., 2006; Leya and Masarik, 2013).
The combined Cd-Pt-W isotope results from this study confirm that the relative magnitude of neutron capture-induced isotope anomalies are strongly affected by the chemical composition of the irradiated material. Both Pt and W isotopes are most sensitive to neutron capture at epithermal and higher energies (Kruijer et al., 2013a; Wittig et al., 2013), whereas Cd has a very large neutron capture cross section at thermal energies. The correlated #196Pt and #182W variations indicate a significant fluence of epithermal and higher energetic neutrons especially for Sikhote Alin (Fig. 6.3b). In spite of the large thermal neutron capture cross section of 113Cd and the evidence for epithermal neutron capture effects in some samples, there are no resolvable Cd isotope variations (Fig 6.3a). This lack of detectable neutron capture effects in Cd in conjunction with the clear-cut evidence for neutron capture at higher neutron energies from W and Pt isotopes, therefore, provides direct testimony of a dominance of neutrons with energies higher-than-thermal in iron meteorites. In marked contrast, lunar samples show very large Cd isotope anomalies of up to 0.3% (Sands et al., 2001), which reflect a much larger fluence of thermal neutrons in silicate matrices.
Also shown in Fig. 6.3 are model calculations from Leya and Masarik (2013) for the effects of neutron capture on Cd, Pt, and W isotope compositions in iron meteoroid targets. As a comparison the modeled effects of neutron capture on Cd and W isotopes in silicate matrices, here represented by ordinary chondrites, are also displayed in Fig. 6.3a (shown here for a typical Ta/W of ~0.15 for L chondrites; e.g., Lodders and Fegley, 1998). [Note that the neutron capture effects on W isotopes in silicate matrices are also a function of the Ta/W of the target material (Leya et al., 2003), in contrast to iron meteorites which generally have Ta/W~0]. The model results were calculated using particle spectra determined using Monte Carlo techniques and evaluated cross section data files for neutron capture reactions (Leya and Masarik, 2013), and as such, are entirely independent of the (empirical) isotopic data from this study.
The model results for iron meteoroids show a general good agreement with the Pt-W-Cd isotope data presented here. In contrast, for ordinary chondrites (i.e., silicate dominated matrices) much larger neutron capture effects on Cd isotopes are predicted for the same exposure age and shielding conditions. For instance, in an ordinary chondrite (r = 65 cm), at 40 cm depth, and for an exposure age of 500 Ma, the model predicts a very large anomaly of #113Cd ! 200. In contrast, the Cd isotope anomaly predicted for an iron meteorite that was exposed under the same conditions would be ~100" smaller (#113Cd ! 2). Hence, very large neutron capture-induced anomalies on #113Cd relative to #182W are predicted for ordinary chondrites, whereas a much shallower trend is predicted for iron meteoroids (Fig. 6.3).
Thus, both the combined Cd-Pt-W isotope data and the model results confirm that in iron meteorite matrices much smaller neutron capture effects on Cd isotopes are expected
Part A Chapter 6 131
than in silicate matrices (e.g, lunar and chondrite samples, see above). Nevertheless, for a given iron meteorite sample the model still predicts larger anomalies on 113Cd than on 196Pt (e.g., ~1.5 #113Cd vs. ~0.3 #196Pt for Sikhote Alin; Fig. 6.3), such that Cd isotopes could in principle be used as a neutron dosimeter for iron meteorites. However, owing to the very low abundances of Cd in iron meteorites, Cd isotope measurements cannot be made with sufficient precision to clearly resolve #-unit level variations in 113Cd/112Cd. In contrast, the generally high Pt concentrations in iron meteorites make it possible to measure even small neutron capture-induced shifts on Pt isotope compositions with high precision, making Pt isotope the most suitable neutron dosimeter for iron meteorites.
6.5 Conclusions The Cd concentrations in all investigated iron meteorite groups are much lower as reported in some previous studies that were based on INAA. In contrast to the large systematic, variations in the concentration of moderately volatile elements like Ga and Ge, there is no clear variation in Cd concentration amongst troilites or metal phases of different iron meteorite groups. Instead, Cd is strongly depleted in all magmatic irons studied here, consistent with very low abundances of other highly volatile elements. Our results, therefore, indicate that moderately volatile element abundance systematics in iron meteorites (e.g., Ge and Ga) are decoupled from those of the more volatile elements (e.g., Cd and Tl). This decoupling can be best explained if the parent bodies of magmatic iron meteorites accreted in a temperature range in which the moderately volatile elements could partly condense, thereby accounting for variations in concentrations of moderately volatile elements such as Ge and Ga, but well above the condensation temperature of Cd. Thus, in spite of variable degrees of depletions in moderately volatile elements, all magmatic iron meteorites exhibit uniform and strong depletions in more volatile elements such as Cd.
No resolvable Cd isotope anomalies were found in the studied samples, whereas neutron capture-induced Pt and W isotope anomalies for the same samples are well resolved and well correlated. This demonstrates that neutron capture in iron meteorites, owing to their high Fe contents, mainly occurs at epithermal (and higher) energies. Our combined Cd-Pt-W isotope results, therefore, demonstrate that the relative magnitude of neutron capture-induced isotope anomalies is strongly affected by the chemical composition of the irradiated material.
The low fluence of thermal neutrons combined with the very low Cd concentrations in iron meteorites and the dominant sensitivity of W isotope compositions to neutron capture above thermal energies render Cd isotopes an inappropriate neutron dosimeter to correct cosmic ray-induced W isotope shifts in iron meteorites. The same conclusion also holds for most other isotope systems that might need corrections for neutron capture effects in iron meteorites, e.g., the 107Pd-107Ag chronometer. In contrast, the generally high concentrations of Pt in iron meteorites make Pt isotopes a powerful neutron dosimeter for these samples.
Acknowledgements Constructive reviews by Kees Welten and by the associate editor Marc Caffee are gratefully acknowledged. M. Fischer-Gödde and F. Oberli are thanked for advice and help with the MC-ICPMS analyses. M. Zbinden is acknowledged for assistance with the Cd concentration measurements. This study was supported by a Förderungsprofessor of the Swiss National Science Foundation to T. Kleine (Grant no. PP00P2_123470).
132 Abundance and isotopic composition of Cd in iron meteorites
References Ballhaus C., Laurenz V., Münker C., Fonseca R. O. C., Albarède F., Rohrbach A., Lagos M.,
Schmidt M. W., Jochum K.-P., Stoll B., Weis U. and Helmy H. M. (2013) The U/Pb ratio of the Earth's mantle—A signature of late volatile addition. Earth and Planetary Science Letters 362, 237-245.
Blichert-Toft J., Moynier F., Lee C. T. A., Telouk P. and Albarede F. (2010) The early formation of the IVA iron meteorite parent body. Earth Planet. Sci. Lett. 296, 469-480.
Chen J. H. and Wasserburg G. J., 1983. The abundances of Ag and Cd isotopes in iron meteorites14th Lunar and Planetary Science Conference #1052.
Chen J. H. and Wasserburg G. J. (1990) The Isotopic Composition of Ag in Meteorites and the Presence of Pd-107 in Protoplanets. Geochimica et Cosmochimica Acta 54, 1729-1743.
Davis A. M., 2006. Volatile evolution and loss. In: Lauretta, D. S. and McSween, H. Y. Eds.), Meteorites and the early solar system II. The University of Arizona Press, Tuscon, Arizona.
Hans U., Kleine T. and Bourdon B. (2013) Rb-Sr chronology of volatile depletion in differentiated pro-toplanets: BABI, ADOR and ALL revisited. Earth Planet. Sci. Lett. doi: 10.1016/j.epsl.2013.05.029.
Hidaka H., Ebihara M. and Yoneda S. (2000) Isotopic study of neutron capture effects on Sm and Gd in chondrites. Earth and Planetary Science Letters 180, 29-37.
Kleine T., Mezger K., Palme H., Scherer E. and Munker C. (2005) Early core formation in asteroids and late accretion of chondrite parent bodies: Evidence from Hf-182-W-182 in CAIs, metal-rich chondrites, and iron meteorites. Geochimica et Cosmochimica Acta 69, 5805-5818.
Kollár D., Michel R. and Masarik J. (2006) Monte Carlo simulation of GCR neutron capture production of cosmogenic nuclides in stony meteorites and lunar surface. Meteoritics & Planetary Science 41, 375-389.
Kracher A., Willis J. and Wasson J. T. (1980) Chemical classification of iron meteorites—IX. A new group (IIF), revision of IAB and IIICD, and data on 57 additional irons. Geochimica et Cosmochimica Acta 44, 773-787.
Kruijer T. S., Fischer-Gödde M., Kleine T., Sprung P., Leya I. and Wieler R. (2013a) Neutron capture on Pt isotopes in iron meteorites and the Hf–W chronology of core formation in planetesimals. Earth and Planetary Science Letters 361, 162-172.
Kruijer T. S., Sprung P., Kleine T., Leya I., Burkhardt C. and Wieler R. (2012) Core formation in the earliest planetesimals inferred from Hf-W chronometry on samples with minimal cosmic-ray effects. Geochimica et Cosmochimica Acta 99, 287-304.
Kruijer T. S., Touboul M., Fischer-Gödde M., Bermingham K., Kleine T. and Walker R. J., 2013b. Resolution of small differences in the time of metal segregation in iron meteorite parent bodiesLunar and Planetary Science Conference, #1920, Houston, TX.
Lagos M., Ballhaus C., Munker C., Wohlgemuth-Ueberwasser C., Berndt J. and Kuzmin D. V. (2008) The Earth's missing lead may not be in the core. Nature 456, 89-92.
Leya I., Lange H. J., Neumann S., Wieler R. and Michel R. (2000) The production of cosmogenic nuclides in stony meteoroids by galactic cosmic-ray particles. Meteoritics & Planetary Science 35, 259-286.
Leya I. and Masarik J. (2013) Thermal neutron capture effects in radioactive and stable nuclide systems. Meteoritics & Planetary Science, n/a-n/a.
Leya I., Wieler R. and Halliday A. N. (2003) The influence of cosmic-ray production on extinct nuclide systems. Geochimica et Cosmochimica Acta 67, 529-541.
Lingenfelter R. E., Canfield E. H. and Hampel V. E. (1972) The Lunar Neutron-Flux Revisited. Earth and Planetary Science Letters 16, 355-369.
Lodders K. (2003) Solar system abundances and condensation temperatures of the elements. The Astrophysical Journal 591, 1220-1247.
Lodders K. and Fegley B., 1998. The Planetary Scientist's Companion. Oxford University Press.
Part A Chapter 6 133
Lovering J. F., Nichiporuk W., Chodos A. and Brown H. (1957) The distribution of gallium, germanium, cobalt, chromium, and copper in iron and stony-iron meteorites in relation to nickel content and structure. Geochimica et Cosmochimica Acta 11, 263-278.
Masarik J. (1997) Contribution of neutron-capture reactions to observed tungsten isotopic ratios. Earth and Planetary Science Letters 152, 181-185.
Nyquist L. E., Wiesmann H., Bansal B., Shih C. Y., Keith J. E. and Harper C. L. (1995) Sm-146-Nd-142 Formation Interval for the Lunar Mantle. Geochimica et Cosmochimica Acta 59, 2817-2837.
Palme H. and Jones A., 2003. 1.03 - Solar System Abundances of the Elements. In: Editors-in-Chief: Heinrich, D. H. and Karl, K. T. Eds.), Treatise on Geochemistry. Pergamon, Oxford.
Rosman K. J. R. and Delaeter J. R. (1974) Abundance of Cadmium and Zinc in Meteorites. Geochimica Et Cosmochimica Acta 38, 1665-1677.
Sands D. G., De Laeter J. R. and Rosman K. J. R. (2001) Measurements of neutron capture effects on Cd, Sm and Gd in lunar samples with implications for the neutron energy spectrum. Earth and Planetary Science Letters 186, 335-346.
Schmitt R. A., Smith R. H. and Olehy D. A. (1963) Cadmium Abundances in Meteoritic and Terrestrial Matter. Geochimica et Cosmochimica Acta 27, 1077-1088.
Scott E. R. D. (1972) Chemical Fractionation in Iron-Meteorites and Its Interpretation. Geochimica et Cosmochimica Acta 36, 1205-1236.
Scott E. R. D. and Wasson J. T. (1975) Classification and Properties of Iron-Meteorites. Reviews of Geophysics 13, 527-546.
Smoliar M. I., Walker R. J. and Morgan J. W. (1996) Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271, 1099-1102.
Sprung P., Scherer E. E., Upadhyay D., Leya I. and Mezger K. (2010) Non-nucleosynthetic heterogeneity in non-radiogenic stable Hf isotopes: Implications for early solar system chronology. Earth and Planetary Science Letters 295, 1-11.
Taylor G. J. (1992) Core formation in asteroids. Journal of Geophysical Research: Planets 97, 14717-14726.
Ulff-Møller F. (1998) Effects of liquid immiscibility on trace element fractionation in magmatic iron meteorites: A case study of group IIIAB. Meteoritics & Planetary Science 33, 207-220.
Wasson J. T. (1999) Trapped melt in IIIAB irons; solid/liquid elemental partitioning during the fractionation of the IIIAB magma. Geochimica et Cosmochimica Acta 63, 2875-2889.
Wasson J. T. (2011) Relationship between iron-meteorite composition and size: Compositional distribution of irons from North Africa. Geochimica et Cosmochimica Acta 75, 1757-1772.
Wasson J. T., Huber H. and Malvin D. J. (2007) Formation of IIAB iron meteorites. Geochimica et Cosmochimica Acta 71, 760-781.
Wasson J. T., Matsunami Y. and Rubin A. E. (2006) Silica and pyroxene in IVA irons; possible formation of the IVA magma by impact melting and reduction of L-LL-chondrite materials followed by crystallization and cooling. Geochimica et Cosmochimica Acta 70, 3149-3172.
Wasson J. T. and Richardson J. W. (2001) Fractionation trends among IVA iron meteorites: Contrasts with IIIAB trends. Geochimica et Cosmochimica Acta 65, 951-970.
Wittig N., Humayun M., Brandon A. D., Huang S. and Leya I. (2013) Coupled W–Os–Pt isotope systematics in IVB iron meteorites: In situ neutron dosimetry for W isotope chronology. Earth and Planetary Science Letters 361, 152-161.
Wombacher F., Rehkamper M., Mezger K., Bischoff A. and Muenker C. (2008) Cadmium stable isotope cosmochemistry. Geochimica et Cosmochimica Acta 72, 646-667.
Wombacher F., Rehkamper M., Mezger K. and Munker C. (2003) Stable isotope compositions of cadmium in geological materials and meteorites determined by multiple-collector ICPMS. Geochimica et Cosmochimica Acta 67, 4639-4654.
134 Introduction to Part B
PART B
NUCLEOSYNTHETIC W ISOTOPE HETEROGENEITY AND THE
IMPLICATIONS FOR EARLY SOLAR SYSTEM CHRONOLOGY
Chapter 7 135
Chapter 7
Introduction to Part B
136 Introduction to Part B
7.1 Nucleosynthesis and nucleosynthetic isotope anomalies in meteorites Almost all isotopes of chemical elements heavier than H, He, Li, Be and B are produced by stellar nucleosynthesis. While the elements lighter than Fe are mostly produced by nuclear fusion (i.e., exothermic) in the star interiors, two main neutron capture (i.e., endothermic) processes (Burbidge et al., 1957) produce most of the nuclides beyond the Fe peak. These include (i) the slow neutron capture process (s-process) and (ii) the rapid neutron capture process (r-process). In case of the s-process, neutron densities are low compared to !-decay rates, so that, after capturing a neutron, any newly synthesized unstable nuclide will decay to a relatively nearby stable nucleus (i.e., s-process nuclides follow the ‘valley of beta-stability’ in the nuclide chart). In contrast, the r-process occurs at very high neutron densities - with neutron capture rates typically more rapid than !-decay – such that highly unstable nuclei can be ‘by-passed’ and relatively heavy nuclides be produced. Hence, the most neutron-rich isotopes of an element are often r-process nuclides. Although still debated, s-process nucleosynthesis may primarily occur in association with Asymptotic Giant Branch (AGB) stage in low (1-3 M!) to intermediate (3-8 M!) mass stars, whereas for r-process nucleosynthesis an explosive source, such as a Type II core-collapse supernova, is required. Finally, p-process nuclides represent a class of heavy, low abundance, proton-rich nuclides, whose origin is not entirely understood (see also Chapter 9).
The interstellar medium (ISM) represents a homogenized mix of isotopes synthesized in, and subsequently expelled by a multitude of stellar sources. Consequently, a variety of stellar sources and different generations of stars contributed material to the parental molecular cloud of the Sun (e.g., Burbidge et al., 1957; Nittler, 2003; Meyer and Zinner, 2006; Adams, 2010). This material partly derives from the well-mixed interstellar medium, but most likely also from one or several relatively nearby stellar sources (e.g., Adams, 2010). As a result, most chemical elements and isotopes in our solar system are very well mixed, as evidenced by relatively uniform elemental ratios and isotopic compositions of refractory elements in primitive, undifferentiated meteorites (Section 1.2.1). However, the identification of intact presolar grains (SiC, nanodiamonds and graphite) contained in chondritic meteorites with very large, often order-of-magnitude, isotopic heterogeneities (e.g., Zinner, 1998; Nittler, 2003; Zinner, 2003), provide direct and compelling evidence that (i) distinct stellar sources, and hence, nucleosynthetic sites, contributed material to the solar system and (ii) that this material was not completely homogenized and at least some intrinsic heterogeneity was preserved.
Mass-independent isotope anomalies in bulk meteorites and their components, albeit of much smaller magnitude than those in presolar grains (on the order of ~0.01%), have been identified for a large number of elements, including Mo (e.g., Dauphas et al., 2002; 2004; Burkhardt et al., 2011), Sm, Nd, and Ba (Andreasen and Sharma, 2006; Carlson et al., 2007); Ni (e.g., Regelous et al., 2008; Steele et al., 2011; 2012), Zr (Schönbächler et al., 2003; Akram et al., 2013), Ti (e.g., Niederer et al., 1981; Niemeyer and Lugmair, 1981; Niederer et al., 1985; Leya et al., 2009; Trinquier et al., 2009), Cr (Trinquier et al., 2007; Qin et al., 2010), and Ru (Chen et al., 2010). In contrast, other elements appear to have been homogeneously distributed at the bulk meteorite scale and present level of analytical resolution, including Hf (Sprung et al., 2010) and Os (Yokoyama et al., 2007; Walker, 2012). Small nucleosynthetic W isotope anomalies in bulk meteorites - consistent with heterogeneity in s- and/or r-process W isotopes - have so far been identified in IVB iron meteorites (Qin et al., 2008) and in CAI (Burkhardt et al., 2008). Why such anomalies occur for some elements but not for others is not fully known, but answering this question may provide important
Chapter 7 137
insight into the processes and time scales related to material transport and mixing in the early solar nebula.
To better understand the origin of these nucleosynthetic isotopic heterogeneities, several studies have investigated isotopic variations in the various components of chondritic meteorites, which are disclosed by several incongruent leach steps in progressively stronger acids (e.g., Schönbächler et al., 2005; Reisberg et al., 2009; Yokoyama et al., 2010; Qin et al., 2011; Burkhardt et al., 2012b; 2012a). The isotope anomalies measured for these leachates define empirical correlation lines that can be directly compared to astrophysical models for nucleosynthesis (e.g., Arlandini et al., 1999), and which might tie the observed anomalies at the bulk meteorite scale to specific nucleosynthetic environments.
The cause of nucleosynthetic isotope anomalies in bulk planetary bodies is still vividly debated. Generally, the presence of such anomalies is interpreted to reflect the incorporation of diverse presolar components in varying proportions into early-accreted planetary bodies. The degree of isotopic heterogeneity observed in meteoritic materials likely is a function of (i) the efficiency of large-scale mixing and transport of material in the protosolar nebula in space and time (e.g., Burkhardt et al., 2011; Steele et al., 2011; 2012), (ii) thermal, physical and/or chemical processing of the solar nebula that allowed sorting of distinct presolar carriers (Regelous et al., 2008; Trinquier et al., 2009; Dauphas et al., 2010; Burkhardt et al., 2012b), (iii) possibly late injected freshly synthesized material in the solar nebula (e.g., Bizzarro et al., 2007; Trinquier et al., 2007), and (iv) secondary processes like aqueous alteration on meteorite parent bodies (e.g., Yokoyama et al., 2011).
7.2 CAI: Relicts of ‘time zero’ in solar system history
7.2.1 Introduction Ca,Al-rich inclusions (CAI) are ceramic-like, refractory element-rich inclusions contained in chondritic meteorites (e.g., Christophe Michel-Lévy, 1968; Grossman, 1972; MacPherson et al., 1988; Brearley and Jones, 1998; MacPherson, 2003). The petrographic, chemical and isotopic characteristics of CAI indicate that they formed under high temperature conditions (MacPherson, 2003). The minerals contained in CAI closely resemble the first phases to condense out of a gas of solar composition after cooling from high temperatures (1300-1500 K) as predicted by thermodynamic calculations (Lord, 1965; Grossman, 1972). It is thought that CAI must have formed relatively close to the Sun (at least within <2AU) where ambient temperatures were high enough for refractory elements to have initially been in the gas phase (Boss, 1998; Ciesla, 2010). Additional evidence for early formation of CAI close to the Sun comes from evidence for the former presence in CAI of short-lived 10Be (t1/2 = 1.39 Myr) (McKeegan et al., 2000) which can only be produced by irradiation and which ties the formation of CAI to the pre-main sequence (‘T-Tauri’) phase of the Sun.
The petrographic, chemical and isotopic classification of CAI is complex (see MacPherson, 2003), but for the purpose of this thesis it is useful to distinguish: (i) (mostly) coarse-grained CAI showing evidence for a complex history that involved melting and reprocessing (e.g., type B inclusions, compact type A), and (ii) un-melted CAI (e.g., fluffy type A, fine-grained inclusions) in which pristine signatures are more readily preserved. The REE patterns of most coarse-grained CAI are unfractionated but uniformly enriched relative to CI chondrites. In contrast, fine-grained inclusions have distinctive trace element and REE patterns (“Group II”) with strong depletions in the most refractory elements and in Yb and Eu – both relatively volatile REE (Tanaka and Masuda, 1973). These patterns can likely only be explained by high-temperature fractional condensation of a solar gas (Boynton, 1975; Davis
138 Introduction to Part B
and Grossman, 1979) in which a component containing the most refractory elements was removed by condensation. The group II pattern then arises in objects that condensed from the remaining refractory element-depleted gas. Therefore, fine-grained inclusions are among the few objects directly testifying of condensation in the solar nebula.
7.2.2 Nucleosynthetic isotope anomalies in CAI Most CAI exhibit nucleosynthetic isotope anomalies for a large number of elements (including Mg, Ti, Cr, Ni, Mo, Sr, Zr, Ba, Nd, Sm, Ti, W, Hf) (e.g., Niederer et al., 1981; Niemeyer and Lugmair, 1981; Birck and Lugmair, 1988; Schönbächler et al., 2003; Trinquier et al., 2009; Sprung et al., 2010; Burkhardt et al., 2011; Wasserburg et al., 2012; Hans et al., 2013). These isotopic anomalies suggest a heterogeneous distribution of material contributed from several distinct stellar nucleosynthetic sources. However, the magnitude and origin (i.e., reflecting either s-, r-, or p-process contributions) of nucleosynthetic isotopes anomalies in CAI are often different from those observed in bulk meteorites (Burkhardt et al., 2011). So-called FUN inclusions (FUN for Fractionated and Unidentified Nuclear isotope properties) represent a very rare and distinct class of CAI with strong mass-dependent fractionations in their O isotope compositions and anomalously low initial 26Al/27Al (Wasserburg et al., 1977). As their petrographic and chemical properties closely resemble those of other CAI, FUN inclusions can only be identified based on their isotopic signatures. FUN CAI generally exhibit larger nucleosynthetic anomalies than their ‘normal’ counterparts. For instance, 'normal' coarse-grained, and some fine-grained CAI show more or less uniform excesses in the neutron-rich isotope 50Ti (Niederer et al., 1981; Niemeyer and Lugmair, 1981; Niederer et al., 1985; Schönbächler et al., 2003; Leya et al., 2009; Trinquier et al., 2009). For FUN inclusions these anomalies are significantly larger than for coarse-grained type B inclusions, and both excesses and deficits were reported. Interestingly, some fine-grained inclusions define an extensive range of 50Ti anomalies, some comparable to those of FUN inclusions (Niemeyer and Lugmair, 1981), indicating that fine-grained inclusions may bridge the gap between ‘normal’ and FUN CAI.
7.2.3 Chronology of CAI Absolute and relative radiometric ages indicate that CAI are the oldest objects that formed in the solar system. CAI have the lowest 87Sr/86Sr (Gray et al., 1973; Podosek et al., 1991), and the highest initial 26Al/27Al (e.g., Lee et al., 1976; 1977; Jacobsen et al., 2008) and initial 182Hf/180Hf (Burkhardt et al., 2008) of any meteoritic material. Absolute 207Pb-206Pb ages correspond to ~4567-4568 Ma (Amelin et al., 2002; 2010; Bouvier and Wadhwa, 2010) and are the oldest determined for any known natural material. Accordingly, CAI are commonly taken to define the start of solar system history and ‘time-zero’ in early solar system chronology. However, obtaining an accurate chronology of CAI is not without challenges. First, CAI are complex objects and individually may preserve ~1-2 Myr nebular histories of melting and evaporation (e.g., Macpherson et al., 1995; McKeegan and Davis, 2003). Hence, small age differences between CAI may exist, which may be resolved with the Al-Mg or Pb-Pb chronometers (MacPherson et al., 2010). Second, nucleosynthetic isotope heterogeneities, either due to a heterogeneous distribution of the short-lived parent isotope in the solar nebula, or alternatively, due to nucleosynthetic isotope anomalies in the daughter isotope, may hamper obtaining accurate age information (e.g., Burkhardt et al., 2012a; Wasserburg et al., 2012). In principle the different chronometers used for dating CAI and other early solar system materials should provide concordant age intervals. Hence, comparison of ages provided by different chronometers (i.e., the Al-Mg, Pb-Pb, Hf-W systems) for the samples
Chapter 7 139
can also used to identify and resolve potential issues with a particular chronometer (Nyquist et al., 2009; Kleine et al., 2012). To illustrate the aforementioned complexities, some recent issues encountered with (1) Al-Mg dating and (2) Pb-Pb dating of CAI are outlined in more detail below.
7.2.3.1 Al-Mg systematics of CAI As condensation in the solar nebula fractionated Al and Mg, the Al-Mg systematics of CAI can be used to date the timing of dust condensation and/or agglomeration. Obtaining a precise and accurate Al-Mg chronology of CAI has faced several problems and considerable debate. Two examples are highlighted below:
(1) While Al-Mg isochrons for bulk CAI may record the time of dust condensation, an internal (mineral) isochron of an igneous CAI would record the time when CAI last melted and crystallized. Thus, if CAI recorded a prolonged history of condensation and subsequent melting and crystallisation, then differences between the initial 26Al/27Al from internal and bulk Al-Mg isochrons would be expected. Indeed, several Al-Mg studies have been reported on CAI with so-called supracanonical initial 26Al/27Al (Young et al., 2005; Thrane et al., 2006), significantly higher than the canonical 26Al/27Al of ~5.2"10-5 found for most CAI (e.g., Jacobsen et al., 2008). In particular, based on the higher than canonical initial 26Al/27Al of (5.85±0.05)"10-5 obtained for a bulk CAI isochron, it has been argued that this supracanonical value records an earlier, primary episode of nebular Al/Mg fractionation of the CAI precursor material which predated crystallisation of the igneous CAI (Thrane et al., 2006). However, a later study showed that the initial 26Al/27Al of fine-grained CAI (i.e., primary condensates) is identical to that of coarse-grained CAI (and the canonical value), which renders the existence of supracanonical initial 26Al/27Al unlikely (MacPherson et al., 2010), and suggests that isochrons yielding supracanonical 26Al/27Al are disturbed. Still, the most recent Al-Mg study on CAI suggests that small age differences between different types of CAI in fact do exist, and arguing for extended processing of igneous CAI (MacPherson et al., 2012). Specifically, the most primitive, un-melted CAI, including fine-grained inclusions, all have grossly uniform canonical initial 26Al/27Al of ~5.2"10-5, whereas coarse-grained (re-)melted type B inclusions have initial 26Al/27Al ranging from ~5.2"10-5 to ~4.2"10-5, corresponding to an age span of several 100 Kyr and showing evidence for a prolonged multistage history. Thus, un-melted, refractory inclusions (fluffy type A, fine-grained, amoeboid olivine aggregates) likely represent the most pristine objects, while coarse-grained, igneous CAI (type B, compact type A) record a more prolonged history that involved melting and recrystallization (MacPherson et al., 2012).
(2) The key assumption in dating early solar system processes with extinct radionuclides is that the parent nuclide was homogeneously distributed in the early solar nebula (Section 1.2.3.1). This assumption has been vividly debated in the case of the 26Al-26Mg chronometer. For example, Larsen et al. (2011) reported evidence for widespread 26Mg heterogeneity in solar system reservoir, and reported these variations to reflect substantial heterogeneity in initial 26Al/27Al (perhaps up to 80%). However, Wasserburg et al. (2012) reported an internal Al-Mg isochron for a FUN CAI with canonical initial 26Al/27Al, but with a different initial 26Mg/24Mg than the isochrons reported by Jacobsen et al. (2008) and Larsen et al. (2011). This observation demonstrates that the isotopic heterogeneity is in Mg isotopes rather than 26Al/27Al, and does not provide evidence for heterogeneous distribution of 26Al in the solar protoplanetary disk. This is consistent with Al-Mg data for chondrules supporting a homogeneous distribution of 26Al (within 10%) in the early solar nebula (Villeneuve et al., 2009).
140 Introduction to Part B
7.2.3.2 Pb-Pb chronometry of CAI Another example of the complexities in dating CAI relates to the highly precise absolute ages provided by the Pb-Pb chronometer. Previously, Pb-Pb ages for meteorites were calculated assuming that the U isotope composition of the samples was identical to the terrestrial 238U/235U of 137.88 (e.g., Amelin et al., 2002; Jacobsen et al., 2008). However, recently discovered 238U/235U variability in CAI (Amelin et al., 2010; Brennecka et al., 2010) indicates that the assumption of a constant U isotope composition is no longer valid and requires a re-evaluation of the Pb-Pb systematics of CAI. Thus, accurate and precise Pb-Pb meteorites can only be obtained together with U isotope measurements for the same samples (Amelin et al., 2010). The Pb-Pb ages reported by some recent studies that into account U isotope variability are now consistent with an age of CAI around ~4567 Ma (Amelin et al., 2010; Connelly et al., 2012), In contrast, Pb-Pb results reported in two other recent studies indicate an age for CAI of ~4568 Ma (Bouvier and Wadhwa, 2010; Bouvier et al., 2011; Wadhwa et al., 2013), Although the earlier study did not measure U isotopes in the studied CAI (Bouvier and Wadhwa, 2010), the subsequent study did and confirmed the age of ~4568 Myr (Bouvier et al., 2011; Wadhwa et al., 2013). Thus, there is no agreement at present about the exact Pb-Pb ages of CAI, reflecting the complexity of obtaining accurate and precise absolute ages for CAI.
The problems encountered in Al-Mg and Pb-Pb chronometry highlighted above can in part also pose a problem for obtaining a precise Hf-W chronology of CAI. As the resolution of the Hf-W is not as high as that of the Al-Mg and Pb-Pb chronometers, age differences between different CAI (i.e., of only a few kyr) cannot easily be resolved. However, the discovery of nucleosynthetic W isotope anomalies in CAI introduce a serious issue for Hf-W dating and requires a more detailed assessment (Burkhardt et al., 2012a).
7.3 Aim and outline of Part B The application of Hf-W chronometry to date core formation in planetary bodies requires that the initial Hf and W isotope compositions of the solar system are accurately and precisely defined. Constraining these reference parameters is of particular relevance for obtaining a precise Hf-W chronology for early-accreted planetesimals like the parent bodies of iron meteorites. As CAI are the oldest objects that formed in the solar system (Section 1.2.3), the solar system initial 182Hf/180Hf and 182W/184W can in principle be directly derived from the Hf-W systematics of CAI (Burkhardt et al., 2008), but the presence of nucleosynthetic W isotope anomalies in CAI currently precludes accurate and precise determination of these parameters (Burkhardt et al., 2012a). Therefore, the main objective of this study is to systematically investigate the magnitude and nature of nucleosynthetic W isotope anomalies in bulk CAI in order to better constrain the initial Hf and W isotope composition of the solar system. The results of a Hf-W study on fine- and coarse-grained bulk CAI from Allende are presented in Chapter 8. Assessing the extent and origin of nucleosynthetic W isotope anomalies at the bulk meteorite scale is also essential for obtaining accurate Hf-W ages. While nucleosynthetic anomalies in the major W isotopes reflect a heterogeneous distribution of s- and r-process W isotopes (Qin et al., 2008), anomalies in the minor isotope 180W (0.13% ab.) may be indicative of p-process isotope heterogeneity in the primitive solar nebula (Schulz et al., 2013). In chapter 9, we investigate the possible existence of such p-process 180W anomalies in greater detail through an elaborate and systematic W isotope study of the IIAB and IVB iron meteorites.
Chapter 7 141
References Adams F. C. (2010) The Birth Environment of the Solar System. Annual Review of
Astronomy and Astrophysics 48, 47-85. Akram W., Schönbächler M., Sprung P. and Vogel N. (2013) Zirconium - hafnium isotope
evidence from meteorites for the decoupled synthesis of light and heavy neutron-rich nuclei The Astrophysical Journal 777, 169 (12 pp.).
Amelin Y., Kaltenbach A., Iizuka T., Stirling C. H., Ireland T. R., Petaev M. and Jacobsen S. B. (2010) U-Pb chronology of the Solar System's oldest solids with variable 238U/235U. Earth and Planetary Science Letters 300, 343-350.
Amelin Y., Krot A. N., Hutcheon I. D. and Ulyanov A. A. (2002) Lead isotopic ages of chondrules and calcium-aluminum-rich inclusions. Science 297, 1678-1683.
Andreasen R. and Sharma M. (2006) Solar Nebula Heterogeneity in p-Process Samarium and Neodymium Isotopes. Science 314, 806-809.
Arlandini C., Kappeler F., Wisshak K., Gallino R., Lugaro M., Busso M. and Straniero O. (1999) Neutron capture in low-mass asymptotic giant branch stars: Cross sections and abundance signatures. Astrophysical Journal 525, 886-900.
Birck J. L. and Lugmair G. W. (1988) Nickel and chromium isotopes in Allende inclusions. Earth and Planetary Science Letters 90, 131-143.
Bizzarro M., Ulfbeck D., Trinquier A., Thrane K., Connelly J. N. and Meyer B. S. (2007) Evidence for a Late Supernova Injection of 60Fe into the Protoplanetary Disk. Science 316, 1178-1181.
Boss A. P. (1998) Temperatures in protoplanetary disks. . Annual Review of Earth and Planetary Sciences 26, 53-80.
Bouvier A., Brennecka G. A. and Wadhwa M., 2011. Absolute chronology of the first solids in the solar systemFormation of the First Solids in the Solar System (#9054), Hawaii.
Bouvier A. and Wadhwa M. (2010) The age of the Solar System redefined by the oldest Pb-Pb age of a meteoritic inclusion. Nature Geoscience 3, 637-641.
Boynton W. V. (1975) Fractionation in the solar nebula: condensation of yttrium and the rare earth elements. Geochimica et Cosmochimica Acta 39, 569-584.
Brearley A. J. and Jones R. H., 1998. Chondritic meteorites. In: Ribbe, P. H. (Ed.), Reviews in Mineralogy. Mineralogical Society of America.
Brennecka G. A., Weyer S., Wadhwa M., Janney P. E., Zipfel J. and Anbar A. D. (2010) (238)U/(235)U variations in meteorites: Extant (247)Cm and implications for Pb-Pb dating (vol 327, pg 449, 2010). Science 327, 1200-1200.
Burbidge E. M., Burbidge G. R., Fowler W. A. and Hoyle F. (1957) Synthesis of the Elements in Stars. Reviews of Modern Physics 29, 547-650.
Burkhardt C., Kleine T., Bourdon B., Palme H., Zipfel J., Friedrich J. M. and Ebel D. S. (2008) Hf-W mineral isochron for Ca,Al-rich inclusions: Age of the solar system and the timing of core formation in planetesimals. Geochimica et Cosmochimica Acta 72, 6177-6197.
Burkhardt C., Kleine T., Dauphas N. and Wieler R. (2012a) Nucleosynthetic tungsten isotope anomalies in acid leachates of the Murchison chondrite: Implications for hafnium-tungsten chronometry. The Astrophysical Journal Letters 753, L6.
Burkhardt C., Kleine T., Dauphas N. and Wieler R. (2012b) Origin of isotopic heterogeneity in the solar nebula by thermal processing and mixing of nebular dust. Earth and Planetary Science Letters 357–358, 298-307.
Burkhardt C., Kleine T., Oberli F., Pack A., Bourdon B. and Wieler R. (2011) Molybdenum isotope anomalies in meteorites: Constraints on solar nebula evolution and origin of the Earth. Earth and Planetary Science Letters 312, 390-400.
Carlson R., Boyet M. and Horan M. (2007) Chondrite barium, neodymium and samarium isotopic heterogeneity and early Earth differentiation. Geochimica Et Cosmochimica Acta 71, A145-A145.
142 Introduction to Part B
Chen J. H., Papanastassiou D. A. and Wasserburg G. J. (2010) Ruthenium endemic isotope effects in chondrites and differentiated meteorites. Geochimica Et Cosmochimica Acta 74, 3851-3862.
Christophe Michel-Lévy M. (1968) Un chondre exceptionnel dans la météorite de Vigarano. Bull. Soc. Fr. Minéral. Cristallogr 91, 212-214.
Ciesla F. J. (2010) The distributions and ages of refractory objects in the solar nebula. Icarus 208, 455-467.
Connelly J. N., Bizzarro M., Krot A. N., Nordlund Å., Wielandt D. and Ivanova M. A. (2012) The Absolute Chronology and Thermal Processing of Solids in the Solar Protoplanetary Disk. Science 338, 651-655.
Dauphas N., Davis A. M., Marty B. and Reisberg L. (2004) The cosmic molybdenum–ruthenium isotope correlation. Earth and Planetary Science Letters 226, 465-475.
Dauphas N., Marty B. and Reisberg L. (2002) Molybdenum evidence for inherited planetary scale isotope heterogeneity of the protosolar nebula. Astrophysical Journal 565, 640-644.
Dauphas N., Remusat L., Chen J. H., Roskosz M., Papanastassiou D. A., Stodolna J., Guan Y., Ma C. and Eiler J. M. (2010) Neutron-rich Chromium Isotope Anomalies in Supernova Nanoparticles. The Astrophysical Journal 720, 1577.
Davis A. M. and Grossman L. (1979) Condensation and fractionation of rare earths in the solar nebula. Geochimica et Cosmochimica Acta 43, 1611-1632.
Gray C. M., Papanastassiou D. A. and Wasserburg G. J. (1973) Identification of Early Condensates from the Solar Nebula. Icarus 20, 213-239.
Grossman L. (1972) Condensation in Primitive Solar Nebula. Geochimica Et Cosmochimica Acta 36, 597-&.
Hans U., Kleine T. and Bourdon B. (2013) Rb-Sr chronology of volatile depletion in differentiated protoplanets: BABI, ADOR and ALL revisited. Earth and Planetary Science Letters.
Jacobsen B., Yin Q. Z., Moynier F., Amelin Y., Krot A. N., Nagashima K., Hutcheon I. D. and Palme H. (2008) Al-26-Mg-26 and Pb-207-Pb-206 systematics of Allende CAIs: Canonical solar initial Al-26/Al-27 ratio reinstated. Earth and Planetary Science Letters 272, 353-364.
Kleine T., Hans U., Irving A. J. and Bourdon B. (2012) Chronology of the angrite parent body and implications for core formation in protoplanets. Geochimica et Cosmochimica Acta 84, 186-203.
Larsen K. K., Trinquier A., Paton C., Schiller M., Wielandt D., Ivanova M. A., Connelly J. N., Nordlund Å., Krot A. N. and Bizzarro M. (2011) Evidence for Magnesium Isotope Heterogeneity in the Solar Protoplanetary Disk. The Astrophysical Journal Letters 735, L37.
Lee T., Papanastassiou D. A. and Wasserburg G. J. (1976) Demonstration of Mg-26 Excess in Allende and Evidence for Al-26. Geophys Res Lett 3, 109-112.
Lee T., Papanastassiou D. A. and Wasserburg G. J. (1977) Aluminum-26 in the early solar system: fossil or fuel? Astrophysical Journal Letters 211, L107-L110.
Leya I., Schonbachler M., Krahenbuhl U. and Halliday A. N. (2009) New Titanium Isotope Data for Allende and Efremovka Cais. Astrophysical Journal 702, 1118-1126.
Lord H. C. I. (1965) Molecular equilibria and condensation in a solar nebula and cool stellar atmospheres. Icarus 4, 279-288.
MacPherson G. J., 2003. 1.08 - Calcium–Aluminum-Rich Inclusions in Chondritic Meteorites. In: Editors-in-Chief: Heinrich, D. H. and Karl, K. T. Eds.), Treatise on Geochemistry. Pergamon, Oxford.
MacPherson G. J., Bullock E. S., Janney P. E., Kita N. T., Ushikubo T., Davis A. M., Wadhwa M. and Krot A. N. (2010) Early Solar Nebula Condensates with Canonical, Not Supracanonical, Initial Al-26/Al-27 Ratios. Astrophysical Journal Letters 711, L117-L121.
Chapter 7 143
Macpherson G. J., Davis A. M. and Zinner E. K. (1995) The Distribution of Al-26 in the Early Solar-System - a Reappraisal. Meteoritics 30, 365-386.
MacPherson G. J., Kita N. T., Ushikubo T., Bullock E. S. and Davis A. M. (2012) Well-resolved variations in the formation ages for Ca–Al-rich inclusions in the early Solar System. Earth and Planetary Science Letters 331–332, 43-54.
MacPherson G. J., Wark D. A. and Armstrong J. T., 1988. Primitive materials surviving in chondrites: refractory inclusions. In: Kerridge, J. F. and Matthews, M. S. Eds.), Meteorites and the Early Solar System. University of Arizona Press.
McKeegan K. D., Chaussidon M. and Robert F. (2000) Incorporation of Short-Lived 10Be in a Calcium-Aluminum-Rich Inclusion from the Allende Meteorite. Science 289, 1334-1337.
McKeegan K. D. and Davis A. M., 2003. 1.16 - Early Solar System Chronology. In: Editors-in-Chief: Heinrich, D. H. and Karl, K. T. Eds.), Treatise on Geochemistry. Pergamon, Oxford.
Meyer B. S. and Zinner E., 2006. Nuleosynthesis. In: Lauretta, D. C. and McSween, H. Y. Eds.), Meteorites and the early solar system II. Univ. Arizona Press.
Niederer F. R., Papanastassiou D. A. and Wasserburg G. J. (1981) The isotopic composition of titanium in the Allende and Leoville meteorites. Geochimica et Cosmochimica Acta 45, 1017-1031.
Niederer F. R., Papanastassiou D. A. and Wasserburg G. J. (1985) Absolute isotopic abundances of Ti in meteorites. Geochimica et Cosmochimica Acta 49, 835-851.
Niemeyer S. and Lugmair G. W. (1981) Ubiquitous isotopic anomalies in Ti from normal Allende inclusions. Earth and Planetary Science Letters 53, 211-225.
Nittler L. R. (2003) Presolar stardust in meteorites: recent advances and scientific frontiers. Earth and Planetary Science Letters 209, 259-273.
Nyquist L. E., Kleine T., Shih C. Y. and Reese Y. D. (2009) The distribution of short-lived radioisotopes in the early solar system and the chronology of asteroid accretion, differentiation, and secondary mineralization. Geochimica et Cosmochimica Acta 73, 5115-5136.
Podosek F. A., Zinner E. K., Macpherson G. J., Lundberg L. L., Brannon J. C. and Fahey A. J. (1991) Correlated study of initial 87Sr/86Sr and Al-Mg isotopic systematics and petrologic properties in a suite of refractory inclusions from the Allende meteorite. Geochimica et Cosmochimica Acta 55, 1083-1110.
Qin L., Alexander C. M. O. D., Carlson R. W., Horan M. F. and Yokoyama T. (2010) Contributors to chromium isotope variation of meteorites. Geochimica et Cosmochimica Acta 74, 1122-1145.
Qin L., Carlson R. W. and Alexander C. M. O. D. (2011) Correlated nucleosynthetic isotopic variability in Cr, Sr, Ba, Sm, Nd and Hf in Murchison and QUE 97008. Geochimica et Cosmochimica Acta 75, 7806-7828.
Qin L. P., Dauphas N., Wadhwa M., Markowski A., Gallino R., Janney P. E. and Bouman C. (2008) Tungsten nuclear anomalies in planetesimal cores. Astrophysical Journal 674, 1234-1241.
Regelous M., Elliott T. and Coath C. D. (2008) Nickel isotope heterogeneity in the early Solar System. Earth and Planetary Science Letters 272, 330-338.
Reisberg L., Dauphas N., Luguet A., Pearson D. G., Gallino R. and Zimmermann C. (2009) Nucleosynthetic osmium isotope anomalies in acid leachates of the Murchison meteorite. Earth and Planetary Science Letters 277, 334-344.
Schönbächler M., Lee D.-C., Rehkämper M., Halliday A. N., Fehr M. A., Hattendorf B. and Günther D. (2003) Zirconium isotope evidence for incomplete admixing of r-process components in the solar nebula. Earth and Planetary Science Letters 216, 467-481.
Schönbächler M., Rehkämper M., Fehr M. A., Halliday A. N., Hattendorf B. and Günther D. (2005) Nucleosynthetic zirconium isotope anomalies in acid leachates of carbonaceous chondrites. Geochimica et Cosmochimica Acta 69, 5113-5122.
144 Introduction to Part B
Schulz T., Münker C. and Peters S. T. M. (2013) p-Process 180W anomalies in iron meteorites: Nucleosynthetic versus non-nucleosynthetic origins. Earth and Planetary Science Letters 362, 246-257.
Sprung P., Scherer E. E., Upadhyay D., Leya I. and Mezger K. (2010) Non-nucleosynthetic heterogeneity in non-radiogenic stable Hf isotopes: Implications for early solar system chronology. Earth and Planetary Science Letters 295, 1-11.
Steele R. C. J., Coath C. D., Regelous M., Russell S. and Elliott T. (2012) Neutron-poor Nickel Isotope Anomalies in Meteorites. The Astrophysical Journal 758, 59.
Steele R. C. J., Elliott T., Coath C. and Regelous M. (2011) Confirmation of mass-independent Ni isotopic variability in iron meteorites. Geochimica Et Cosmochimica Acta 75, 7906-7925.
Tanaka T. and Masuda A. (1973) Rare-earth elements in matrix, inclusions, and chondrules of the Allende meteorite. Icarus 19, 523-530.
Thrane K., Bizzarro M. and Baker J. A. (2006) Extremely Brief Formation Interval for Refractory Inclusions and Uniform Distribution of 26Al in the Early Solar System. The Astrophysical Journal Letters 649, L159.
Trinquier A., Birck J. L. and Allegre C. J. (2007) Widespread Cr-54 heterogeneity in the inner solar system. Astrophysical Journal 655, 1179-1185.
Trinquier A., Elliott T., Ulfbeck D., Coath C., Krot A. N. and Bizzarro M. (2009) Origin of Nucleosynthetic Isotope Heterogeneity in the Solar Protoplanetary Disk. Science 324, 374-376.
Villeneuve J., Chaussidon M. and Libourel G. (2009) Homogeneous Distribution of 26Al in the Solar System from the Mg Isotopic Composition of Chondrules. Science 325, 985-988.
Wadhwa M., Bouvier A. and Janney P. E., 2013. Al-Mg systematics in a CAI from the NWA 6991 CV3 chondrite76th Annual Meteoritical Society Meeting (#5253), Edmonton, Canada.
Walker R. J. (2012) Evidence for homogeneous distribution of osmium in the protosolar nebula. Earth and Planetary Science Letters 351-352, 36-44.
Wasserburg G. J., Lee T. and Papanastassiou D. A. (1977) correlated O And Mg isotopic anomalies in Allende Inclusions: II. Magnesium. Geophysical Research Letters 4, 299-302.
Wasserburg G. J., Wimpenny J. and Yin Q.-Z. (2012) Mg isotopic heterogeneity, Al-Mg isochrons, and canonical 26Al/27Al in the early solar system. Meteoritics & Planetary Science 47, 1980-1997.
Yokoyama T., Alexander C. M. O. and Walker R. J. (2010) Osmium isotope anomalies in chondrites: Results for acid residues and related leachates. Earth and Planetary Science Letters 291, 48-59.
Yokoyama T., Alexander C. M. O. D. and Walker R. J. (2011) Assessment of nebular versus parent body processes on presolar components present in chondrites: Evidence from osmium isotopes. Earth and Planetary Science Letters 305, 115-123.
Yokoyama T., Rai V. K., Alexander M. O. D., Lewis R. S., Carlson R. W., Shirey S. B., Thiemens M. H. and Walker R. J. (2007) Osmium isotope evidence for uniform distribution of s- and r-process components in the early solar system. Earth and Planetary Science Letters 259, 567-580.
Young E. D., Simon J. I., Galy A., Russell S. S., Tonui E. and Lovera O. (2005) Supra-Canonical 26Al/27Al and the Residence Time of CAIs in the Solar Protoplanetary Disk. Science 308, 223-227.
Zinner E. (1998) Stellar nucleosynthesis and the isotopic composition of presolar grains from primitive meteorites. . Annual Review of Earth and Planetary Sciences 26, 147-188.
Zinner E., 2003. 1.02 - Presolar Grains. In: Editors-in-Chief: Heinrich, D. H. and Karl, K. T. Eds.), Treatise on Geochemistry. Pergamon, Oxford.
Part B Chapter 8 145
Chapter 8
Nucleosynthetic W isotope anomalies and Hf-W chronometry of Ca-Al-rich inclusions
Kruijer, T.S.a,b, Kleine, T.b, M. Fischer-Göddeb, Burkhardt, C.c, Wieler, R.a
aETH Zürich, Institute of Geochemistry and Petrology, Zürich, Switzerland.
bWestfälische Wilhelms-Universität Münster, Institut für Planetologie, Münster, Germany. cOrigins laboratory, Department of Geophysical Sciences, The University of Chicago.
A version of this chapter will be published in Earth and Planetary Science Letters (2013)
after moderate revisions.
Abstract The successful application of 182Hf-182W chronometry to investigate the timescales of early solar system processes requires accurate knowledge of the initial 182Hf/180Hf and 182W/184W at the beginning of solar system history. These can most directly be determined from Ca-Al-rich inclusions (CAI), the oldest dated objects that formed in the solar system. However, the interpretation of Hf-W data for CAI is complicated by the presence of nucleosynthetic W isotope anomalies, which may significantly bias the inferred initial 182W/184W of CAI. The aim of this study is to explore the extent and nature of nucleosynthetic W isotope variations in CAI to better quantify the initial Hf and W isotope compositions of the solar system. We report Hf-W data for several fine- and coarse-grained CAI from the carbonaceous CV3 chondrites Allende, NWA 6871 and NWA 6717. The five investigated fine-grained CAI exhibit large and variable anomalies in !183W (!iW equals 0.01% deviation from terrestrial values), extending to much larger anomalies than previously observed in CAI, and reflecting variable abundances of s- and r-process W isotopes. Conversely, the seven coarse-grained (mostly type B) inclusions show only small (if any) nucleosynthetic W isotope anomalies. The investigated CAI define a precise correlation between !183W and !182Wi (i.e., subscript ‘i’: corrected for radioactive decay using their measured 180Hf/184W), which provides a direct empirical means to correct the !182W of any CAI for nucleosynthetic isotope anomalies using their measured !183W. The corrected !182W vs. 180Hf/184W define a bulk CAI isochron (MWSD=1.6) with a precise initial 182Hf/180Hf of (1.03±0.05)"10-4, which, when anchored to the angrites D'Orbigny and Sahara 99555, corresponds to an age interval relative to CAI that is in good agreement with that inferred from Al-Mg chronometry. This provides evidence that 26Al was homogeneously distributed in the inner solar system, at least in the region were CAI and angrites formed. The bulk CAI isochron and !182Wi-!183W correlation yield identical initial 182W/184W isotope compositions, resulting in a mean solar system initial !182W of #3.51±0.07 (±95% conf.). Relative to this value, magmatic iron meteorites have more radiogenic ('pre-exposure') !182W values, corresponding to core formation ages of ~1-3 Ma after CAI formation and indicating a time gap of at least several 100 ka between the formation of the first solids and accretion of the oldest planetesimals.
146 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
8.1 Introduction The timespan between condensation of the first solids and melting and differentiation of the first planetary bodies probably was only a few million years (Ma). Distinguishing between different events and establishing a fine-scale chronology of early solar system evolution, therefore, requires the determination of highly precise ages for meteorites and meteorite components. The necessary time resolution is in principle provided by Pb-Pb chronometry (e.g., Amelin et al., 2010; Bouvier and Wadhwa, 2010) and by several short-lived chronometers such as the 26Al-26Mg (Jacobsen et al., 2008; Lee et al., 1976, 1977) and 182Hf-182W systems (Kleine et al., 2009). However, the precise chronological interpretation of the isotopic data is not always straightforward, and can be hampered by (i) mass-independent isotope variations in the elements of interest (e.g., Brennecka et al., 2010; Burkhardt et al., 2012; Leya et al., 2009; Wasserburg et al., 2012) (ii) a possible heterogeneous distribution of some short-lived radionuclides (especially 26Al) (e.g., Larsen et al., 2011).
Ca-Al-rich inclusions (CAI) are the oldest objects formed in the solar system (Amelin et al., 2010; Amelin et al., 2002; Bouvier and Wadhwa, 2010; Gray et al., 1973; Lee et al., 1976, 1977; Podosek et al., 1991), and are commonly used to define the start of solar system history and ‘time-zero’ in early solar system chronology as recorded in meteorites. Determining the age of CAI as well as the initial abundance of short-lived radionuclides in the CAI, therefore, is key for establishing a precise chronology of the first few Ma of solar system history. Ca-Al-rich inclusions are complex objects, however, and individually may preserve ~1-2 Ma nebular histories of melting and evaporation (e.g., Macpherson et al., 1995; McKeegan and Davis, 2003). The un-melted refractory inclusions (e.g., fluffy type A, fine-grained CAI, amoeboid olivine aggregates) likely represent the most pristine objects, while coarse-grained, igneous CAI (e.g., type B, compact type A) record a more complex history that involved (re)-melting and recrystallization (e.g., 2003; MacPherson et al., 2012). The initial isotope composition of the solar system would, therefore, most directly be determined on primitive refractory inclusions like fine-grained CAI.
The short-lived 182Hf-182W system (t1/2 = 8.9 Ma) is a powerful chronometer to determine the timescales of planetary differentiation and can also provide precise ages for individual meteorites (e.g., Kleine et al., 2009). Obtaining accurate and precise 182Hf-182W ages requires knowledge of the initial 182Hf/180Hf and 182W/184W ratios at the start of solar system history, which so far have been determined through the investigation of coarse-grained CAI (Burkhardt et al., 2008). The interpretation of the Hf-W data for CAI is complicated by the presence of nucleosynthetic W isotope anomalies in the CAI, however, which significantly bias the inferred initial 182W/184W composition of CAI (Burkhardt et al., 2012). High precision measurements of both 182W/184W and 183W/184W on bulk CAI, and knowledge of the relative nucleosynthetic contributions on 182W and 183W anomalies are, therefore, required to quantify the extent and nature of nucleosynthetic W isotope anomalies in CAI, which in turn is essential to tightly constrain the initial 182Hf/180Hf and 182W/184W of the solar system.
Fine-grained CAI have so far not been investigated for Hf-W systematics, but several features make them particularly suited to determine the initial 182Hf/180Hf and 182W/184W of the solar system. First, most fine-grained CAI are characterized by Group II REE patterns (Martin and Mason, 1974; Tanaka and Masuda, 1973), which are indicative of high-temperature fractional condensation of a gas of solar composition (Boynton, 1975; Davis and Grossman, 1979). This suggests that their Hf/W ratios may be different from those of chondrites and coarse-grained CAI, which are characterized by mostly flat REE patterns and typically have near-chondritic Hf/W ratios. Thus, based on fine-grained CAI it may be possible to determine
Part B Chapter 8 147
a bulk CAI Hf-W isochron. Second, Al-Mg data suggests that fine-grained CAI may be slightly older (~0.2 Ma) than coarse-grained, (re-)melted type B CAI, which record a prolonged multistage history (MacPherson et al., 2012). Although such time intervals cannot be resolved using Hf-W chronometry, the Al-Mg data suggest that fine-grained CAI may represent the earliest-formed, least processed, and as such, most insightful objects for investigating the pristine isotopic signatures at the start of solar system history. Finally, some fine-grained CAI, at least for some elements —such as Ti (e.g., Niemeyer and Lugmair, 1981)— appear to show more variable and larger nucleosynthetic isotope anomalies than most coarse-grained CAI. This makes fine-grained CAI ideally suited for investigating the effects of nucleosynthetic W isotope anomalies on Hf-W chronometry.
We report new Hf-W isotope data for five fine-grained CAI and seven coarse-grained CAI from the Allende, NWA 6871 and NWA 6717 CV3 chondrites. In addition, five separate digestion aliquots of a powdered slab (100 g) of Allende were also analysed for their Hf-W isotope systematics. The new Hf-W data are used to assess the relative contributions to W isotope variations in CAI of (i) 182Hf decay (i.e, radiogenic) and (ii) variable contributions of s- and r- process W isotopes (i.e, nucleosynthetic), with the ultimate goal to more precisely constrain the initial Hf and W isotope compositions of the solar system. The new Hf-W data have important implications for the timescales of planetesimal accretion and differentiation inferred from Hf-W chronometry, and for assessing the extent of possible heterogeneities in the initial abundance of short-lived radionuclides (e.g., 182Hf, 26Al) in the early solar system.
8.2 Samples and analytical methods
8.2.1 Sample preparation and chemical separation The petrographic and chemical classification (Fig. 8.12; Section 8.7.1) demonstrates that our CAI sample suite can be distinguished into (i) fine-grained CAI with Group II REE patterns and (ii) coarse-grained (mostly type B) CAI with Group I patterns. The CAI samples (~50-300 mg) were carefully sawn out of the meteorite slices using a diamond saw and then cleaned with ethanol or acetone in an ultrasonic bath to remove any adhering dust. The samples were then carefully crushed in an agate mortar and the cleanest CAI pieces were handpicked under a binocular microscope. Care was taken to remove any chondrite matrix material adhering to the CAI pieces. Finally, the CAI fractions were cleaned again in ethanol, dried, and ground to a fine powder.
The CAI samples were digested in 15 or 60 ml Savillex® vials in HF-HNO3-HClO4 (2:1:0.05) at 180-200 °C for 4-5 days on a hotplate. After digestion, samples were evaporated to dryness at 200 °C to remove HClO4 and fluoride precipitates, and then treated by repeated dry-downs in 6 M HCl-0.06 M HF. Finally, the samples were completely dissolved in 12 ml (15 ml vials) or 40 ml (60 ml vials) 6 M HCl–0.06 M HF. Approximately 1-10% aliquots (equivalent to ~2-4 ng W) of these solutions were spiked with a mixed 180Hf-183W tracer that was calibrated against pure Hf and W metal standards (Kleine et al., 2004). Additional aliquots (2-3%) were taken to determine REE concentrations (Section 8.7.1.2.1).
Tungsten was separated from the sample matrix using a two-stage anion exchange chromatography, slightly modified from previously published procedures (Kleine et al., 2004; Kleine et al., 2012; Kruijer et al., 2012). The aliquots for W isotope composition analyses were evaporated to dryness and re-dissolved in 13 ml 0.5 M HCl–0.5 M HF prior to loading onto the first anion exchange column (4 ml BioRad® AG 1"8, 200-400 mesh). The sample matrix was rinsed off the column in 10 ml 0.5 M HCl–0.5 M HF, followed by another rinse
148 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
with 6 ml 6 M HCl–1 M HF in which significant amounts of Ti, Zr and Hf were eluted. Finally, W was eluted in 8 ml 6 M HCl–1 MF. The second anion exchange chromatography step, modified after Kleine et al. (2012), provides quantitative removal of the HFSE (Ti, Zr, Hf, Ta) from the W cuts. The samples were evaporated at 200 °C with added HClO4 to destroy organic molecules, re-dissolved in 0.6 M HF–0.2% H2O2 and loaded onto pre-cleaned columns BioRad® Polyprep columns filled with 1 ml anion exchange resin (AG1"8, 200-400). Titanium, Zr, and Hf were rinsed off the column with 10 ml 1 M HCl–2% H2O2, followed by 10 ml 8 M HCl–0.01 M HF. Finally, W was eluted in 8 ml 6 M HCl–1 M HF. The W cuts were evaporated to dryness with added HClO4 (200 °C) and converted to a 0.56 M HNO3–0.24 M HF measurement solutions. The chemical separation of Hf and W for the spiked aliquots was accomplished using ion exchange chromatography techniques described in Kleine et al. (2004).
Total procedural blanks ranged from ~70 to ~120 pg for the W isotope composition analyses, and ~5 to ~30 pg W and ~2 to ~5 pg Hf for the isotope dilution analyses. The corresponding blank corrections for W isotope composition measurements were <4 ppm on !182W, and hence, negligible given the analytical resolution of our data. The blank corrections for the isotope dilution analyses were mostly !3% for both Hf and W, and were included in the uncertainty of the 180Hf/184W assuming an average uncertainty on the blank correction of 50% (Table 8.1). In a few cases, the blank corrections were slightly larger (up to 6%), primarily due to the low Hf and W amounts available for analyses.
8.2.2 Isotope measurements All isotope measurements were performed with a ThermoScientific® Neptune Plus MC-ICPMS at the University of Münster. Tungsten was introduced into the mass spectrometer using an ESI® PFA nebulizer connected to a Cetac® Aridus II desolvator. A combination of high-sensitivity Jet sampler and X skimmer cones provided total ion beam intensities of 2-2.5"10-10 A for a 30 ppb solution at a 50-60 $L/min uptake rate. Each measurement consisted of 60s baseline integrations (deflected beam) followed by 200 isotope ratio measurements of 4.2s each. Possible isobaric interferences of Os on 184W and 186W were corrected by monitoring interference-free 188Os and were negligible (<10 ppm) for all analysed samples. Instrumental mass bias was corrected by internal normalisation to either 186W/183W = 1.9859 (denoted ‘6/3’) or 186W/184W = 0.92767 (denoted ‘6/4’) using the exponential law. The W isotope measurements of the analysed samples were bracketed by measurements of a terrestrial solution standard [prepared from an Alfa Aesar® metal; (2002; Kleine et al., 2004)] and are reported as !-unit deviations (i.e, 0.01 %) relative to the mean values of the bracketing standard analyses. For samples analysed more than once, the reported !iW represent the mean of pooled solution replicates. The accuracy and reproducibility of the measurements was assessed by repeated analyses of terrestrial rock standards (BCR-2) that were digested, processed through the full chemical separation, and analysed together with each set of CAI samples (Table S1). The mean !182W (6/4) of 0.00±0.16 (2 s.d., n=12) obtained for the BCR-2 standard is identical to the terrestrial W isotope composition (Table 8-A1, Fig. 8.13), but the BCR-2 analyses show small anomalies for normalisations involving 183W, including excesses in !182W (6/3) of +0.12±0.08 and !184W (6/3) of +0.05±0.04, and a deficit in !183W (6/4) of #0.08±0.05 (±95% conf., n=12). These coupled !182W-!183W systematics have previously been observed in high-precision MC-ICPMS studies for terrestrial standards as well as silicate rock and iron meteorite samples (Kruijer et al., 2013a). They are attributed to a mass-independent W isotope fractionation only affecting 183W that is induced by W-loss during re-
Part B Chapter 8 149
dissolution of the samples in Teflon beakers. The W isotope compositions of samples can be corrected for the mass-independent effect on 183W using the different normalisation schemes for the W isotope measurements and the results obtained for the terrestrial standard (see Willbold et al., 2011; 2012; Kruijer et al., 2013a). Accordingly, the measured !182W (6/3), !183W (6/4), and !184W (6/3) of the CAI from this study were corrected using the mean !iW values obtained for the BCR-2 standard (Table 8-A1), and the associated uncertainties of the BCR-2 data (95% conf.) were propagated into the W isotope data reported in Table 8.1. Although this correction may introduce some additional uncertainty for the coarse-grained CAI, most of the nucleosynthetic W isotope anomalies of the fine-grained CAI are very large, making the mass-independent effect largely insignificant for these samples. Reported uncertainties are 95% conf. limits of the mean (if n%4) or the 2 s.d. (if n<4) estimated from replicate measurements of the BCR-2 standard as reported above (Table 8A-1). Finally, a second terrestrial basalt standard (BHVO-2) was processed and measured together with the Allende samples and yields terrestrial W isotope compositions for all normalisations (Table S1).
The uncertainty of the Hf and W isotope dilution measurements (2&) derives from the uncertainty on the isotope ratio measurement (0.2%, 2 s.d.) and from the blank correction (generally <3%). The resulting uncertainty on the 180Hf/184W ratios generally is <1.5% (2&).
150 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
Tabl
e 8.
1H
f-W
dat
a fo
r bul
k C
AI.
IDM
eteo
rite
Type
REE
pat
tern
Wei
ght
Hf
W
180 H
f/184 W
N!18
2 Wm
eas.
!183 W
mea
s.a!18
2 Wib
!182 W
nuc.
cor
r.c!18
2 Wm
eas.a
!184 W
mea
s.a!18
2 Wib
!182 W
nuc.
cor
r.c
(mg)
(ppb
) (p
pb)
(±2"
s.d.
)(±
2" s.
d.)
(±2"
s.d.
)(±
2" s.
d.)
(±2"
s.d.
)(±
2" s.
d.)
(±2"
s.d.
)(±
2" s.
d.)
(±2"
s.d.
)Fi
ne-g
rain
ed C
AI
Nor
mal
izat
ion
to 1
86W
/184 W
Nor
mal
izat
ion
to 18
6 W/18
3 WA
C01
Alle
nde
FGn.
a.52
1087
477d
2.68
9±
0.09
61
-0.3
3±
0.16
0.00
±0.
18-3
.52
±0.
25-0
.33
±0.
21-0
.29
±0.
150.
00±
0.12
-3.4
5±
0.27
-0.2
9±
0.21
AF0
1A
llend
eFG
Gro
up II
289
252
89.8
3.31
1±
0.02
91
3.35
±0.
162.
05±
0.18
-0.5
9±
0.25
0.42
±0.
210.
56±
0.15
-1.3
6±
0.12
-3.3
3±
0.27
0.39
±0.
21A
F02
Alle
nde
FGG
roup
II12
033
.0d
97.0
0.40
2±
0.00
51
-1.0
1±
0.16
1.54
±0.
25-1
.48
±0.
16-3
.21
±0.
21-3
.02
±0.
15-1
.02
±0.
16-3
.49
±0.
18-3
.15
±0.
21A
F03
Alle
nde
FGG
roup
II34
51.8
d10
80.
564
±0.
008
1-2
.33
±0.
400.
57±
0.42
-3.0
0±
0.45
-3.1
5±
0.52
-3.0
5±
0.45
-0.3
8±
0.28
-3.7
1±
0.50
-3.1
0±
0.57
AF0
4A
llend
eFG
Gro
up II
288
63.8
241
0.31
3±
0.00
22
4.47
±0.
165.
36±
0.18
4.10
±0.
16-3
.19
±0.
21-2
.68
±0.
15-3
.57
±0.
12-3
.04
±0.
18-3
.14
±0.
21C
oars
e-gr
aine
d C
AI
AC
02A
llend
e?
n.a.
104
193
158d
1.44
2±
0.04
81
-1.5
8±
0.24
0.01
±0.
18-3
.29
±0.
26-1
.60
±0.
31-1
.55
±0.
25-0
.01
±0.
12-3
.25
±0.
31-1
.55
±0.
33A
F05
Alle
nde
BG
roup
I69
1725
1224
1.66
2±
0.00
92
-1.1
9±
0.16
0.07
±0.
18-3
.16
±0.
18-1
.29
±0.
21-1
.26
±0.
15-0
.05
±0.
12-3
.21
±0.
21-1
.26
±0.
21A
I01
NW
A 6
871
Bn.
a.82
2823
847
3.93
5±
0.03
02
0.95
±0.
16-0
.08
±0.
18-3
.73
±0.
271.
06±
0.21
1.11
±0.
150.
05±
0.12
-3.5
3±
0.30
1.11
±0.
21A
I02
NW
A 6
717
Bn.
a.10
331
137
00.
991
±0.
013
1-2
.52
±0.
16-0
.16
±0.
18-3
.70
±0.
17-2
.29
±0.
21-2
.29
±0.
150.
11±
0.12
-3.4
6±
0.19
-2.2
8±
0.21
AI0
5A
llend
e?
n.a.
7723
016
21.
683
±0.
016
1-1
.15
±0.
200.
03±
0.18
-3.1
5±
0.22
-1.1
9±
0.25
-1.1
5±
0.20
-0.0
2±
0.12
-3.1
3±
0.26
-1.1
5±
0.27
AI0
6BA
llend
eB
n.a.
168
1861
1500
1.46
4±
0.00
94
-1.7
1±
0.16
0.00
±0.
18-3
.45
±0.
18-1
.71
±0.
21-1
.72
±0.
150.
00±
0.12
-3.4
4±
0.20
-1.7
2±
0.21
AI0
7A
llend
e?
n.a.
137
1081
1098
1.16
2±
0.00
64
-2.0
4±
0.16
0.04
±0.
18-3
.42
±0.
17-2
.10
±0.
21-2
.09
±0.
15-0
.03
±0.
12-3
.46
±0.
20-2
.09
±0.
21In
stru
men
tal m
ass b
ias w
as c
orre
cted
usi
ng 18
6 W/18
4 W =
0.9
2767
(den
oted
'6/3
') or
186 W
/183 W
= 1
.985
94 (d
enot
ed '6
/4')
and
the
expo
nent
ial l
aw.
a Cor
rect
ed fo
r a m
ass-
inde
pend
ent e
ffect
on
183 W
usi
ng th
e av
erag
e !i W
obt
aine
d fo
r the
BC
R-2
stan
dard
: !18
2 W (6
/3) =
+0.
12±0
.08,
!18
3 W (6
/4) =
#0.
08±0
.05,
!18
4 W (6
/3) =
+0.
05±0
.03
(±95
% c
onf.,
n=1
3).
b '!18
2 Wi':
corr
ecte
d fo
r rad
ioge
nic
cont
ribut
ions
from
182 H
f dec
ay u
sing
the
mea
sure
d 18
0 Hf/18
4 W a
nd th
e so
lar s
yste
m in
itial
182 H
f/180 H
f of (
1.03
±0.0
5)$1
0-5.
c '!18
2 Wnu
c. c
orr.':
cor
rect
ed fo
r s- a
nd/o
r r-p
roce
ss h
eter
ogen
eity
usi
ng re
gres
sion
-der
ived
slop
es sh
own
in F
ig. 8
.2.
d Con
cent
ratio
ns w
ith b
lank
cor
rect
ions
>3%
. Th
e un
certa
inty
on
the
blan
k co
rrec
tions
are
ass
umed
to b
e 50
% a
nd a
re p
ropa
gate
d in
the
unce
rtain
ties o
f the
180 H
f/184 W
ratio
s.
Part B Chapter 8 151
8.3 Results The measured W isotope compositions of the investigated CAI are presented in Table 8.1 and Fig. 8.1, those corrected for radioactive decay from 182Hf in Fig. 8.2 (Section 8.5.1), and those corrected for nucleosynthetic W isotope anomalies in Fig. 8.4 (Section 8.5.2). The !iW values are shown for different normalisations used to correct for instrumental mass bias, i.e, normalised to 186W/184W (‘6/4’) and to 186W/183W (‘6/3’). Hence, the mass bias-corrected !iW reported here represent the net effect of anomalies on all W isotopes involved for that particular normalisation. All CAI show variable deficits or excesses in !182W (6/4) (#2.5 to +4.5) (Fig. 8.1). The fine-grained CAI exhibit variable !183W (6/4) (up to +5.3), whereas the !183W of most coarse-grained CAI are indistinguishable from the terrestrial W isotope composition. Moreover, in particular for the fine-grained CAI, the !182W anomalies normalised to 186W/183W are markedly distinct from those normalized to 186W/184W (Fig. 8.1). Whereas the !182W (6/3) values display a well-defined correlation line when plotted vs. their 180Hf/184W (Fig. 8.1d), no such correlation is observed for the !182W (6/4) normalisation (Fig. 8.1c). Finally, the five replicates of a bulk Allende powder yield a mean !182W (6/4) of #1.97±0.04 (±95% conf., n=5), whereas the mean !183W of #0.02±0.05 is indistinguishable from the terrestrial value (Table 8-A2).
Fig. 8.1: Measured !182W and !183W of CAI (a,b) and Hf-W isochron diagrams with measured !182W values (c,d). The data are shown for two different normalisations: normalised to 186W/184W (a,c) and to 186W/183W (b,d). Error bars represent external uncertainties (2SD or 95% conf., see Table 8.1). Also plotted are the W isotope data for Allende CAI from Burkhardt et al. (2008). Further shown are the effects of (i) radiogenic contributions from 182Hf decay and, as a means of illustration, (ii) trends predicted for variability in s- and r-process W isotopes based on acid leachates from Murchison, Orgueil and Allende (Burkhardt and Schönbächler, 2013). .
–4 –3 –2 –1 0 1–4
–2
0
2
4
ε182
W (6
/3)
ε184W (6/3)
s- and r- process variability
182 H
f dec
ay
0 1 2 3 4–4
–2
0
2
4
180Hf/184W
ε182
W (6
/4)
NWA6717
A-ZH-5
NWA6871
(a) (b)
ε182
W (6
/4)
ε183W (6/4)
s- and r-
process variability
182 H
f dec
ay
(c)
0 1 2 3 4–4
–3
–2
–1
0
1
2
180Hf/184W
ε182
W (6
/3)
NWA6717
NWA6871
(d)
A-ZH-5
Fine-grained CAI:
Coarse-grained CAI:
Allende (this study)
Allende 'A-ZH-5'(Burkhardt et al., 2008, 2012)
Allende (this study)
NWA 6717 / 6871(this study)
Allende, Type B(Burkhardt et al., 2008, 2012)
0 2 4 6–4
–2
0
2
4
A-ZH-5
A-ZH-5
152 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
8.4 Nucleosynthetic W isotope anomalies in CAI Variations in !182W may be (i) radiogenic (i.e., from 182Hf-decay), (ii) cosmogenic (i.e., resulting from the interaction with secondary neutrons produced by interaction with cosmic rays), or (iii) nucleosynthetic (i.e., reflect a heterogeneous distribution of s- and r-process W nuclides), while variations in !183W can only be nucleosynthetic in origin. Since 184W has a larger relative s-process contribution than 182W, 183W and 186W, a deficit in s-process (or an excess in r-process) W nuclides results in elevated !183W. The elevated !183W of most of the investigated fine-grained CAI, therefore, indicates an s-deficit (or r-excess) in these samples. One fine-grained CAI (AF04) shows a very large anomaly in !183W (~+5.3), about two times the largest anomaly so far observed for CAI, namely CAI A-ZH-5 (Burkhardt et al., 2008; Burkhardt et al., 2012a). In contrast, all of the coarse-grained (mostly type B) inclusions show near-terrestrial !183W, indicating that nucleosynthetic W isotope heterogeneity is largely absent in these samples.
The extent of nucleosynthetic 182W variations is more difficult to assess, because the !182W values of the CAI result from a combination of nucleosynthetic and radiogenic effects (potential cosmogenic effects on 182W are absent in the samples, as will be discussed later). This is illustrated in a plot of !182W vs. !183W, in which the data strongly scatter and do not plot along the trend expected for correlated nucleosynthetic 182W and 183W anomalies (Fig. 1a). Determining the magnitude of nucleosynthetic 182W anomalies, therefore, first requires the quantification of the radiogenic ingrowth in each of the CAI. This will be discussed in detail further below (Section 8.5).
The observed variability in nucleosynthetic W isotope anomalies in the investigated CAI raises the question of why large and variable !183W anomalies exist in fine-grained CAI, whereas coarse-grained CAI show only small if any nucleosynthetic W isotope anomalies. Since fine-grained CAI in Allende are heavily altered (e.g., Grossman and Ganapathy, 1976; Kornacki and Wood, 1985; Krot et al., 1995), their nucleosynthetic W isotope anomalies could reflect the incongruent dissolution of pre-solar carrier phases and the mobilisation of the released W during parent body alteration (e.g., Yokoyama et al., 2011). However, significant parent body processing would also have significantly disturbed the Hf-W systematics. As we will show in Section 8.5, all investigated bulk CAI plot on a single Hf-W isochron whose 182Hf/180Hfi slope when converted to an absolute time scale corresponds to the age of CAI as given by Pb-Pb chronometry. Hence, significant secondary disturbance of the Hf-W systematics is highly unlikely (unless it occurred right at the time of CAI formation).
The nucleosynthetic W isotope anomalies in the fine-grained CAI, therefore, seem to reflect initial heterogeneities in the primitive solar nebula that has not been sampled by the coarse-grained CAI, which show only small if any 183W anomalies. Recent Al-Mg chronometry of CAI suggests that fine-grained CAI and the (precursors of) coarse-grained CAI formed contemporaneously (Jacobsen et al., 2008; Macpherson et al., 2012). In contrast, internal mineral isochrons for coarse-grained CAI show a larger range in initial 26Al/27Al that is consistent with extended processing [i.e., over several 100 ka; (MacPherson et al., 2012)]. One possibility, therefore, is that fine-grained CAI retained large and variable nucleosynthetic W isotope anomalies, because they record a slightly earlier stage in the primitive solar nebula. Coarse-grained CAI would then record a slightly later stage, i.e., after igneous reprocessing of the earliest aggregated solids, and apparent (partial) homogenization and/or remixing of nucleosynthetic W isotope heterogeneities. Yet an alternative possibility to account for the different magnitude of nucleosynthetic W isotope anomalies in the fine- and coarse-grained CAI is that they derive from two distinct nebular reservoirs, which existed at the same time
Part B Chapter 8 153
but were characterized by different relative proportions of s- and r-process W nuclides. In any case, the variable nucleosynthetic W isotope anomalies are most easily accounted for if the fine-grained CAI aggregated from variable proportions of presolar components, suggesting that the solar nebula was heterogeneous at the scale of individual CAI.
8.5 182Hf-182W systematics of bulk CAI The fine-grained CAI investigated in the present study display a sufficiently large range in Hf/W to permit determining a bulk CAI isochron (Table 8.1; Fig. 8.1). The coarse-grained CAI investigated here and in previous studies (Kleine et al., 2005; Burkhardt et al., 2008) in general exhibit a more restricted range in Hf/W (open symbols in Fig. 8.1), but for type B CAI AI01 a high 180Hf/184W of ~3.9 was determined. This CAI is exceptionally large, however, and it is unclear whether the analysed piece is representative for the bulk CAI. Given the evidence for large and variable nucleosynthetic W isotope anomalies in the investigated CAI, the !182W values of the CAI first must be corrected for nucleosynthetic anomalies before a bulk isochron can be obtained. The strong effect of nucleosynthetic W isotope anomalies on 182Hf-182W systematics becomes obvious by comparing the !182W (6/4) and !182W (6/3) values in isochron diagrams (Fig. 8.1c, 8.1d). While no isochronous relationship is observed in !182W (6/4) vs. 180Hf/184W space (Fig. 8.1c), a much better correlation is observed in !182W (6/3) vs. 180Hf/184W space (Fig. 8.1d). This is because 184W has a significantly larger s-process contribution than 182W, 183W, and 186W, and so 186W/184W-normalized W isotope ratios (e.g., !182W (6/4)) exhibit larger nucleosynthetic isotope effects than 186W/183W-normalized values (e.g., !182W (6/3)) (Fig. 8.1a,b). Below we first quantify the magnitude of nucleosynthetic W isotope heterogeneity using the measured !183W anomalies and the W isotope systematics for different normalisations (Section 8.5.1), and then evaluate the 182Hf-182W systematics of bulk CAI (Section 8.5.2).
8.5.1 Effects of nucleosynthetic W isotope anomalies on Hf-W systematics In principle the correction for nucleosynthetic W isotope anomalies can be done using (i) theoretical models for nucleosynthesis (Arlandini et al., 1999), (ii) W isotope data for SiC grains (Ávila et al., 2012) or acid leachates from primitive chondrites (Burkhardt et al., 2012a), or (iii) empirical W isotope correlations obtained from bulk CAI. As the uncertainties of theoretical models and SiC data are large or difficult to assess, using empirical W isotope data provides the most direct and precise means to correct bulk meteorite samples for nucleosynthetic W isotope variability. This is particularly true for the fine-grained CAI from this study, which exhibit large and variable nucleosynthetic isotope anomalies (Fig. 8.1), and hence, permit the determination of a precise !182W-!183W correlation line.
Investigating the co-variation of nucleosynthetic anomalies on !182W and !183W requires correction of the !182W anomalies for radiogenic contributions from 182Hf-decay. However, such corrections require accurate knowledge about the initial 182Hf/180Hf of the solar system, which is the parameter that we eventually aim to determine using a bulk CAI isochron. The slope of the correlation line for nucleosynthetic !182W and !183W anomalies and the initial 182Hf/180Hf of CAI, therefore, must be determined using an iterative approach: First, an initial estimate of the solar system initial 182Hf/180Hf is obtained from the slope of (1.01±0.06)"10-4 (±2&) of the measured !182W (6/3) vs. 180Hf/184W correlation (Fig. 8.1d). The !182W (6/3) are preferred because for this normalisation the effect of nucleosynthetic isotope anomalies is smallest and, hence, an acceptable correlation is obtained between the measured !182W (6/3) and 180Hf/184W (Fig. 8.1d). Subsequently, the measured !182W (6/3) and !182W (6/4) of the
154 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
investigated CAI are corrected for 182Hf-decay using their measured 180Hf/184W and the initial 182Hf/180Hf determined in the first stage. Then, an empirical correlation is obtained by plotting the decay-corrected “!182Wi” vs. the measured !183W (or !184W) of only the fine-grained CAI. The measured !182W of the CAI are subsequently corrected for nucleosynthetic W isotope heterogeneity using their measured !183W and the regression-derived !182Wi-!183W (6/4) (or !182Wi-!184W) slopes defined by the fine-grained CAI. The corrected !182Wi values are then plotted vs. their 180Hf/184W to obtain a bulk CAI isochron, its slope providing a new initial 182Hf/180Hf of the CAI. After repeating the above iteration once, the inferred initial 182Hf/180Hf and !182Wi-!18iW slopes converge to constant values (see Fig. 8.2 and 8.4).
After the iteration described above, the fine-grained CAI exhibit a well-defined
empirical !182Wi-!183W (6/4) correlation (Fig. 8.2a), which as noted above, reflects a heterogeneous distribution of s- and r-process isotopes relative to the Earth (Burkhardt et al., 2012a). Linear regression of the fine-grained CAI data (Fig. 8.2a) yields a positive slope of +1.43±0.07 and an intercept !182Wi of #3.58±0.22 (±95% conf.; Table 8.2). For the !182Wi vs. !184W (6/3) correlation line a slope of #0.13±0.06 and an intercept !182Wi = #3.53±0.12 (±95% conf.) are obtained (Fig. 8.2b), the latter being in good agreement with the intercept value obtained for the 186W/184W normalisation. The different regression-derived slopes in Fig.
Fig. 8.2: Tungsten isotope systematics of bulk CAI after an iterative correction for radiogenic contributions from 182Hf decay (see Section 5.1). (a,c) !182Wi (6/4) vs. !183W (6/4) and (b,d) !182Wi (6/3) vs. !184W (6/3). Shown are data and regressions for fine-grained CAI only (a,b), and for both fine- and coarse-grained CAI (c,d). Error bars on symbols indicate external uncertainties (2SD or 95% conf.; see Table 8.1). The corrections for 182Hf decay were calculated using the measured 180Hf/184W and a solar system initial of 182Hf/180HfSSI = (1.03±0.05)"10-4 (±95% conf.). Solid lines show linear regressions through the data (York, 1966). Also shown are CAI data from Burkhardt et al. (2008). Also shown are data for Allende bulk CAI from Burkhardt et al. (2008).
–4 –3 –2 –1 0 1–4.5
–4
–3.5
–3
–2.5
–2Fine-grained CAIslope = −0.13±0.06ε182Wi = −3.53±0.12 (95% conf.)MSWD = 0.82
ε182
Wi (
6/3)
ε184W (6/3)0 2 4 6
–4
–2
0
2
4Fine-grained CAIslope = 1.43±0.07ε182Wi = −3.58±0.22 (95% conf.)MSWD = 0.27
ε182
Wi (
6/4)
ε183W (6/4)
0 2 4 6–4
–2
0
2
4CAI - bulk dataslope = 1.41±0.05ε182Wi = −3.52±0.09 (95% conf.)MSWD = 1.1
ε182
Wi (
6/4)
ε183W (6/4)
(a)
–4 –3 –2 –1 0 1–4.5
–4
–3.5
–3
–2.5
–2CAI - bulk dataslope = −0.10±0.07ε182Wi = −3.44±0.07 (95% conf.)MSWD = 1.5
ε182
Wi (
6/3)
ε184W (6/3)
(b)
(c) (d)
Fine-grained CAI:
Coarse-grained CAI:
Allende (this study)
Allende 'A-ZH-5'(Burkhardt et al., 2008, 2012)
Allende (this study)
NWA 6717 / 6871(this study)
Allende, Type B(Burkhardt et al., 2008, 2012)
Part B Chapter 8 155
8.2a and 8.2b directly reflect the smaller magnitude of nucleosynthetic W isotope anomalies in the 186W/183W normalisation compared to 186W/184W-normalized values (see above).
The coarse-grained (mostly type B) CAI from this study exhibit only minor (if any) nucleosynthetic W isotope anomalies as evident from their near-terrestrial !183W (Fig. 8.1). Thus, their !182Wi
values plot close to the intercept of the !182Wi-!183W and !182Wi-!184W correlations (Fig. 8.2c,d). Regression of the combined !182Wi-!183W (6/4) data for both fine- and coarse-grained CAI (MSWD=1.1) yields a slope of +1.41±0.05 and a precise intercept !182Wi (6/4) of #3.52±0.09 (±95% conf.). Regression of the !182Wi-!184W (6/3) correlation (MSWD=1.5) yields a slope of #0.10±0.07 and an intercept !182Wi (6/3) of #3.44±0.07 (±95% conf.). All these values are indistinguishable from, albeit more precise than those obtained by regressing only the data of the fine-grained CAI (Fig. 8.2; Table 8.2). The empirical slopes obtained here from bulk CAI data are in excellent agreement with the !182Wi-!183W (6/4) and !182Wi-!184W (6/3) slopes of +1.42±0.06 and #0.10±0.13 obtained for acid leachates from the Orgueil, Murchison and Allende carbonaceous chondrites (Burkhardt and Schönbächler, 2013). This observation highlights the validity of our iterative approach to correct the W isotope compositions of CAI for nucleosynthetic heterogeneity. Moreover, after correction of the !182W (6/4) and !182W (6/3) values for nucleosynthetic W isotope heterogeneity using their !183W (6/4) and !184W (6/3) values and the empirically determined !182Wi-!183W and !182Wi-!184W slopes, the corrected !182W (6/4) and !182W (6/3) values are identical and plot on a 1:1 line (Fig. 8.3). This provides further evidence that the corrections for nucleosynthetic isotope variations are accurate.
The empirically determined !182Wi-!183W (6/4) slope of +1.41±0.05 is slightly shallower
than the slope of ~+1.686 predicted by the stellar model for s-process nucleosynthesis of Arlandini et al. (1999). The !182Wi-!184W (6/3) slope determined here of #0.10±0.07 shows even greater disparity with the value of ~#0.524 from Arlandini et al. (1999). Given that the uncertainty of the stellar model is large and difficult to assess, the agreement between the model and the data should still be considered acceptable. Nevertheless, correcting !182W data for nucleosynthetic anomalies using models of stellar nucleosynthesis would lead to
Fig. 8.3: !182W (6/4) vs. !182W (6/3) after correction for nucleosynthetic W isotope anomalies using empirically determined slopes from the !182W- !183W and !182W- !184W correlations shown in Fig. 8.2. All data points plot on the 1:1 line, indicating that the correction yields identical results for both normalisations. Error bars on symbols indicate external uncertainties (95% conf. or 2 SD). See Fig. 8.2 for legend of symbols.
–4 –2 0 2–4
–2
0
2
ε182
W (6
/4) co
rr.
ε182W (6/3)corr.
1:1 line
156 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
significantly different results than those obtained using the empirical slopes. For instance, correcting the !182Wi (6/4) of the fine-grained CAI AF04 using the slope predicted by the Arlandini et al. model would result in a value of #4.78, i.e., more than 1! lower than the initial !182W of CAI derived here. Note that with a reservoir having chondritic 180Hf/184W a difference of ~1 !182W would correspond to a calculated age difference of ~10 Ma. This underlines that the quantification of nucleosynthetic isotope anomalies using empirically determined !182Wi-!183W (6/4) and !182Wi-!184W slopes is a necessity for obtaining accurate Hf-W ages.
8.5.2 Hf-W isochron for bulk CAI After the iterative correction of nucleosynthetic W isotope anomalies, both the !182W (6/4) and !182W (6/3) values of the fine-grained CAI exhibit well-defined isochrons when plotted against their 180Hf/184W (Fig. 8.4a,b). Linear regression of the 186W/184W-normalized Hf-W data for the fine-grained CAI yields an initial 182Hf/180Hf of (1.06±0.07)"10-4 and !182Wi (6/4) = #3.65±0.15; the 186W/183W-normalized data provide an initial 182Hf/180Hf of (1.04±0.07)"10-4 and !182W (6/3) = #3.58±0.16. Both !182Wi values are indistinguishable from the intercept values obtained from the !182Wi-!183W and !182Wi-!184W correlations (Fig. 8.2a and b). Including the data for the coarse-grained CAI in the isochron regression yields an initial 182Hf/180Hf of (1.03±0.05)"10-4 and !182Wi (6/4) = #3.50±0.11 for the 186W/184W-normalized data. Similar results of 182Hf/180Hfi = (1.02±0.05)"10-4 and !182Wi (6/3) = #3.47±0.11 are obtained from the 186W/183W-normalized data. Again, the initial !182W values obtained from the isochron regression are in good agreement with the values obtained from the !182W-!183W (6/4) and !182W-!184W (6/3) correlations (Fig. 8.2; Table 8.2). It is noteworthy that two different approaches to determine initial !182W value of CAI—one based on !182Wi-!183W correlations defined by nucleosynthetic W isotope anomalies, and the other on bulk CAI isochrons—and using two different normalisations of the W isotope data (186W/184W or 186W/183W) provide results that are fully consistent. This provides powerful evidence that our approach for calculating nucleosynthetic 182W variations in the CAI is accurate.
The average initial W isotope composition obtained from (i) the !182Wi-!183W correlation (Fig. 2c), and (ii) the bulk CAI isochron (Fig. 8.4c) corresponds to !182Wi (6/4) = #3.51±0.07 (95% conf.). Similarly, a weighted average !182Wi (6/3) = #3.46±0.07 (±95% conf.) is obtained (Table 8.2). Both values agree within their uncertainties, but we use !182Wi (6/4) =
Table 8.2Linear regression results for Hf-W data of bulk CAI.
Normalization to 186W/184W Normalization to 186W/183W!182W-!iW correlations !182Wi !182Wi vs. !183W slope !182Wi !182Wi vs. !184W slope
FG CAI "3.58 ± 0.22 1.43 ± 0.07 "3.53 ± 0.12 "0.13 ± 0.06FG+CG CAI "3.52 ± 0.09 1.41 ± 0.05 "3.44 ± 0.07 "0.10 ± 0.07
Hf-W isochrons !182W (6/4)i182Hf/180Hfi !182W (6/3)i
182Hf/180Hfi
FG CAI "3.65 ± 0.15 1.06 ± 0.07 "3.58 ± 0.16 1.04 ± 0.07FG+CG CAI "3.50 ± 0.11 1.03 ± 0.05 "3.47 ± 0.11 1.02 ± 0.05
Mean FG+CG CAIa "3.51 ± 0.07 "3.46 ± 0.07FG CAI' = fine-grained CAI, 'FG+CG CAI' = fine- and coarse-grained CAI.a Mean of !182W values determined from isochron and !182Wi-!iW correlation.All uncertainties indicate 95% conf. limits of the mean.
Part B Chapter 8 157
#3.51±0.07 (±95% conf.) as the best estimate for the initial W isotope composition of CAI, because this value is least affected by any residual (unaccounted) mass-independent effect on 183W (Section 8.2.2), which would result in a slightly elevated !182Wi. The newly determined !182Wi from this study is in very good agreement with the !182Wi of #3.51±0.10 reported by Burkhardt et al. (2012a), but significantly lower than the previous initial !182Wi of #3.28±0.12, which was based on internal isochrons for type B CAI that were not corrected for nucleosynthetic W isotope anomalies. This underlines again that quantification of nucleosynthetic W isotope heterogeneity in CAI is critical to obtain accurate estimates on W isotope compositions.
The initial 182Hf/180Hf of (1.03±0.05)"10-4 obtained from the regression of the 186W/184W-normalized data for both fine- and coarse-grained bulk CAI (Fig. 8.4c) is slightly higher than but within uncertainty of the initial 182Hf/180Hf values determined by Burkhardt et al. (2008) for internal Hf-W isochrons of several type B CAI. Note that nucleosynthetic anomalies will not affect the slopes of internal isochrons but can displace them along the ordinate axis (Burkhardt et al., 2012b). Since the correction of nucleosynthetic W isotope anomalies for the mineral data for type B CAI from Burkhardt et al. (2008) is not possible, because no precise !183W data are available for these samples, the mineral data from Burkhardt et al. (2008) were not included in the regression. Clearly, obtaining new precise internal Hf-W isochrons together with precise !183W measurements to quantify nucleosynthetic isotope heterogeneity will be an important target for future work.
Fig. 8.4: Hf-W isochron diagrams for bulk CAI after iterative correction (Section 5.1) for nucleosynthetic W isotope anomalies for fine-grained CAI (a,b) and for all investigated CAI (c,d). Shown are data for two different normalisations: (a,c) !182W (6/4) and (b,d) !182W (6/3). Error bars on symbols indicate external uncertainties (2SD; see Table 1). Solid lines show linear regressions through the data (York, 1966). Corrections for nucleosynthetic W isotope anomalies for the CAI were calculated using their measured !183W or !184W and the empirically determined slopes of the !182Wi- !183W or !182Wi- !184W correlations from Fig. 8.2. Also shown are data for Allende bulk CAI from Burkhardt et al. (2008).
(a)
(c)
0 1 2 3 4 5–4
–3
–2
–1
0
1
2
180Hf/184W
Fine-grained CAI - Bulk isochron182Hf/180Hfi = (1.06±0.07) x 10-4
ε182Wi = −3.65±0.15 (95% conf.)MSWD = 0.32
ε182
W (6
/4)
0 1 2 3 4–4
–3
–2
–1
0
1
2
180Hf/184W
CAI - Bulk isochron182Hf/180Hfi = (1.03±0.05) x 10-4
ε182Wi = −3.50±0.11 (95% conf.)MSWD = 1.6
ε182
W (6
/4)
NWA6717
A-ZH-5
NWA6871
(b)
(d)
0 1 2 3 4 5–4
–3
–2
–1
0
1
2
180Hf/184W
Fine-grained CAI - Bulk isochron182Hf/180Hfi = (1.04±0.07) x 10-4
ε182Wi = −3.58±0.16 (95% conf.)MSWD = 0.35
ε182
W (6
/3)
0 1 2 3 4–4
–3
–2
–1
0
1
2
180Hf/184W
CAI - Bulk isochron182Hf/180Hfi = (1.02±0.05) x 10-4
ε182Wi = −3.47±0.11 (95% conf.)MSWD = 1.7
ε182
W (6
/3)
NWA6717
NWA6871
Fine-grained CAI:
Coarse-grained CAI:
Allende (this study)
Allende 'A-ZH-5'(Burkhardt et al., 2008, 2012)
Allende (this study)
NWA 6717 / 6871(this study)
Allende, Type B(Burkhardt et al., 2008, 2012)
A-ZH-5
A-ZH-5A-ZH-5
158 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
8.5.3 Initial 182W/184W and 182Hf/180Hf of the solar system The initial 182Hf/180Hf and 182W/184W determined from the bulk CAI isochron correspond to solar system initial values, provided that the different Hf/W of the bulk CAI were established at the time of CAI formation (i.e., time zero in cosmochemistry). Strong evidence for this comes from the initial 182Hf/180Hf of the bulk CAI Hf-W isochron itself, which yields an absolute Hf-W age for CAI of 4568.0±0.7 Ma when anchored to the angrite D’Orbigny [182Hf/180Hf = (7.15±0.17)"10-5 (Kleine et al., 2012); Pb-Pb age = 4563.37±0.25 Ma (Brennecka and Wadhwa, 2012)]. This absolute Hf-W age is in good agreement with Pb-Pb ages of CAI (Amelin et al., 2010; Bouvier and Wadhwa, 2010; Connelly et al., 2012), indicating that the Hf-W isochron for bulk CAI reflects the time of CAI formation. Nevertheless, before the initial 182Hf/180Hf and 182W/184W of the CAI isochron are interpreted as solar system initial values, it is important to consider two processes that may have had an effect on the Hf-W isochron, namely the interaction with galactic cosmic rays (Leya, 2011)and aqueous alteration on the parent body (Humayun et al., 2007)
Secondary neutron capture effects induced during cosmic-ray exposure can modify W isotope compositions of extraterrestrial materials, and thus possibly also of CAI incorporated in chondritic meteoroids (Leya, 2011). Model calculations for neutron capture effects on W isotopes in carbonaceous chondrites have shown that neutron capture of 181Ta to 182Ta and subsequent '- -decay to 182W can be relevant for CAI minerals having high Ta/W ratios (Leya, 2011). However, this effect will likely be small in bulk CAI, which generally have sufficiently low Ta/W. Although the Ta/W ratios of the CAI have not been determined directly in the present study, they can be inferred assuming that the CAI have chondritic Ta/Hf (Lodders, 2003). This results in inferred Ta/W of ~0.03-0.4 for the investigated CAI. The model from Leya (2011) predicts that for this range in Ta/W and for a relatively short exposure age of ~5.2 Ma for Allende (Scherer and Schultz, 2000), neutron capture effects in the investigated CAI will be <0.05 !182W units. Given the analytical resolution of our W isotope data, cosmic-ray effects can thus be considered negligible.
Evidence for late-stage aqueous alteration on the Allende parent body and the extensive secondary crystallisation in (fine-grained) CAI from Allende (e.g., Kornacki and Wood, 1985; McGuire and Hashimoto, 1989; 1995; Krot et al., 2004; Humayun et al., 2007) raises the question if secondary processes have remobilized Hf and W and caused Hf/W fractionation. However, the CAI investigated here have a high initial 182Hf/180Hf corresponding to an absolute age of 4568.0±0.7 Ma, i.e., the formation age of CAI. Moreover, the preservation of large nucleosynthetic W isotope anomalies for the fine-grained CAI, and the absence of such anomalies in bulk Allende (i.e., as evident from its terrestrial !183W, Table 5-A2), indicate that no significant W exchange with the matrix occurred. Finally, the lack of correlation in a diagram of !182W vs. 1/W (Fig. 8.14) demonstrates that the !182W-180Hf/184W correlation cannot be due to binary mixing. Thus, the variable Hf/W of the fine-grained CAI are not due to secondary alteration but must result from early processes in the solar nebula. Since the fine-grained CAI are characterized by group II REE patterns, which are indicative of fractional condensation in the solar nebula (Boynton, 1975; Davis and Grossman, 1979), the variable Hf/W of the fine-grained CAI most likely also result from chemical fractionation during condensation. We, therefore, interpret the initial 182Hf/180Hf and !182Wi obtained from the bulk CAI isochron as representative for the time of CAI formation, and hence, as solar system initial values.
Part B Chapter 8 159
8.6 Implications for the chronology of the early solar system
8.6.1 Iron meteorites Our new Hf-W data of bulk CAI have important implications for the Hf-W chronometry of iron meteorites. Model ages of metal segregation in iron meteorite parent bodies relative to CAI formation are obtained by comparing the measured !182W of iron meteorites to the 182W/184W isotope evolution in an unfractionated reservoir with chondritic Hf/W. A long-standing issue with respect to the Hf-W chronology of iron meteorites is that many irons show measured !182W below the initial !182W of CAI (Horan et al., 1998; Kleine et al., 2005; Markowski et al., 2006; Scherstén et al., 2006; Qin et al., 2008; Kruijer et al., 2012) resulting in inferred Hf-W model ages of metal segregation that seem older than CAI. The low !182W of iron meteorites are commonly interpreted to reflect the effects of secondary neutron capture induced during cosmic ray exposure of the iron meteoroids (Masarik, 1997; Leya and Masarik, 2013). Thus, an accurate and precise Hf-W chronology of iron meteorites relies both on the quantification of neutron capture effects in iron meteorites and accurate knowledge about the initial !182W of CAI. Recently, Pt isotopes were identified as an excellent neutron dosimeter for iron meteorites and through combined Pt and W isotope measurements it is now possible to obtain a ‘pre-exposure’ !182W of a particular iron meteorite group corrected for secondary neutron capture (Kruijer et al., 2013a; Wittig et al., 2013). Recent Hf-W chronometry demonstrates that different iron meteorite parent bodies have variable ‘pre-exposure’ !182W ranging from ~#3.40 to #3.20 (Kruijer et al., 2013a; Kruijer et al., 2013b). These values are all consistently more radiogenic than the initial !182W of #3.51±0.07 of bulk CAI derived in the present study. This observation demonstrates that CAI formation predated core formation in the parent bodies of iron meteorites by >1 Ma, a conclusion also reached by Burkhardt et al. (2008). Assuming that 26Al decay represents the heat source for planetesimal differentiation the (instantaneous) accretion of the iron meteorite parent bodies likely occurred ~0.5 Ma earlier than the time of metal segregation (Qin et al., 2008). Thus, the first planetesimals—as sampled by the parent bodies of magmatic iron meteorites—only accreted at least a few 100 ka after the formation of the first solids in the solar system.
8.6.2 Comparison to 26Al-26Mg ages and evidence for 26Al homogeneity The use of extinct short-lived chronometers like 26Al-26Mg and 182Hf-182W relies on the assumption that the initial abundances of the parent nuclides were homogeneously distributed in the early solar system. Especially for the 26Al-26Mg system the validity of this assumption is debated, however, and it has been proposed that variable 26Al/27Al in meteorites and meteorite components reflect a heterogeneous distribution of 26Al and not different formation times (e.g., Larsen et al., 2011). This can be tested through inter-calibration of different short-lived and long-lived chronometers between CAI and other well-dated samples (see e.g., Bouvier et al., 2011; Nyquist et al., 2009). Angrites are the ideal samples for such an intercalibration, because the rapid cooling of these samples ensures that differences in closure temperatures between the various chronometers do not result in resolvable age differences (Kleine et al., 2012; Lugmair and Galer, 1992; Markowski et al., 2007; Nyquist et al., 2009; Spivak-Birndorf et al., 2009). Thus, the age intervals given by the different chronometers between CAI formation and crystallisation of angrites can be directly compared with each other.
A recent Hf-W study of angrites reported indistinguishable initial 182Hf/180Hf of (7.15±0.17)"10-5 and (6.87±0.15)"10-5 for the quenched angrites D’Orbigny and Sahara 99555 (Kleine et al., 2012). Using the newly determined solar system initial 182Hf/180Hf of
160 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
(1.03±0.05)"10-4, formation intervals relative to CAI formation, (tCAI, of 4.7±0.7 Ma and 5.2±0.7 Ma are inferred for these two angrites. A combined isochron for D'Orbigny and Sahara 99555 yields an initial 182Hf/180Hf of (6.99±0.11)"10-5, corresponding to (tCAI = 5.0±0.7 Ma. The Al-Mg isochrons determined for D’Orbigny and Sahara 99555 also yield indistinguishable initial 26Al/27Al (Schiller et al., 2010; Spivak-Birndorf et al., 2009). Using the initial 26Al/27Al for CAI of (5.23±0.13)"10-5 (Jacobsen et al., 2008), these angrites define a Al-Mg formation time interval relative to CAI formation corresponding to (tCAI = 5.0±0.2 Ma (see Kleine et al., 2012), in good agreement with the time interval of (tCAI = 5.0±0.7 Ma as given by Hf-W chronometry. Note that the nominal Hf-W age for D'Orbigny and Sahara 99555 is ~0.7 Ma younger than that calculated by Kleine et al. (2012) using the previously inferred solar system initial 182Hf/180Hf of (9.72±0.44)"10-5 (Burkhardt et al., 2008), bringing the Hf-W and Al-Mg ages of these angrites in better agreement than previously recognized. Thus, the new Hf-W results are inconsistent with a grossly heterogeneous distribution of 26Al, at least in the nebular regions in which angrites and CAI formed.
On the basis of small 26Mg variations in bulk meteorites unrelated to 26Al-decay, which appear to be correlated with nucleosynthetic 54Cr anomalies, Larsen et al. (2011) argued that 26Al was heterogeneously distributed in the inner solar system. These authors inferred an initial 26Al/27Al of the angrite parent body of (1.61±0.32)"10-5 at the time of CAI formation. Relative to this initial 26Al/27Al the Al-Mg age of D'Orbigny and Sahara 99555 would be 3.8±0.3 Ma, difficult to reconcile with (tCAI = 5.0±0.7 Ma as determined by Hf-W chronometry in the present study. Thus our new Hf-W data do not support the heterogeneous distribution of 26Al as inferred by Larsen et al. (2011), but suggest that the correlation of 26Mg and 54Cr in bulk meteorites and CAI more likely reflects nucleosynthetic Mg (and Cr) isotope variations. This is consistent with evidence for nucleosynthetic Mg isotope variations in CAI, which are characterized by canonical initial 26Al/27Al (Wasserburg et al., 2012). Further work, and in particular a combined Al-Mg, Hf-W and Pb-Pb study on the very same CAI is needed to fully assess the concordance of these different dating systems.
Acknowledgements TSK thanks G. Budde, D. Cook and P. Sprung for fruitful discussions. U. Heitmann, T. Grund, and I. Ivanov-Bucher are gratefully acknowledged for their technical support. This study was supported by a Förderungsprofessor of the Swiss National Science Foundation to T. Kleine (Grant no. PP00P2_123470).
References Amelin Y., Kaltenbach A., Iizuka T., Stirling C. H., Ireland T. R., Petaev M. and Jacobsen S.
B. (2010) U-Pb chronology of the Solar System's oldest solids with variable 238U/235U. Earth and Planetary Science Letters 300, 343-350.
Amelin Y., Krot A. N., Hutcheon I. D. and Ulyanov A. A. (2002) Lead isotopic ages of chondrules and calcium-aluminum-rich inclusions. Science 297, 1678-1683.
Arlandini C., Kappeler F., Wisshak K., Gallino R., Lugaro M., Busso M. and Straniero O. (1999) Neutron capture in low-mass asymptotic giant branch stars: Cross sections and abundance signatures. Astrophysical Journal 525, 886-900.
Ávila J. N., Lugaro M., Ireland T. R., Gyngard F., Zinner E., Cristallo S., Holden P., Buntain J., Amari S. and Karakas A. (2012) Tungsten Isotopic Compositions in Stardust SiC
Part B Chapter 8 161
Grains from the Murchison Meteorite: Constraints on the s-process in the Hf-Ta-W-Re-Os Region. The Astrophysical Journal 744, 49.
Bouvier A., Spivak-Birndorf L. J., Brennecka G. A. and Wadhwa M. (2011) New constraints on early Solar System chronology from Al–Mg and U–Pb isotope systematics in the unique basaltic achondrite Northwest Africa 2976. Geochimica et Cosmochimica Acta 75, 5310-5323.
Bouvier A. and Wadhwa M. (2010) The age of the Solar System redefined by the oldest Pb-Pb age of a meteoritic inclusion. Nature Geoscience 3, 637-641.
Boynton W. V. (1975) Fractionation in the solar nebula: condensation of yttrium and the rare earth elements. Geochimica et Cosmochimica Acta 39, 569-584.
Brennecka G. A. and Wadhwa M. (2012) Uranium isotope compositions of the basaltic angrite meteorites and the chronological implications for the early Solar System. Proceedings of the National Academy of Sciences 109, 9299-9303.
Brennecka G. A., Weyer S., Wadhwa M., Janney P. E., Zipfel J. and Anbar A. D. (2010) (238)U/(235)U variations in meteorites: Extant (247)Cm and implications for Pb-Pb dating (vol 327, pg 449, 2010). Science 327, 1200-1200.
Burkhardt C., Kleine T., Bourdon B., Palme H., Zipfel J., Friedrich J. M. and Ebel D. S. (2008) Hf-W mineral isochron for Ca,Al-rich inclusions: Age of the solar system and the timing of core formation in planetesimals. Geochimica et Cosmochimica Acta 72, 6177-6197.
Burkhardt C., Kleine T., Dauphas N. and Wieler R. (2012a) Nucleosynthetic tungsten isotope anomalies in acid leachates of the Murchison chondrite: Implications for hafnium-tungsten chronometry. The Astrophysical Journal Letters 753, L6.
Burkhardt C., Kleine T., Dauphas N. and Wieler R., 2012b. Origin of nucleosynthetic isotope heterogeneity in the solar nebula inferred from Mo and W isotopes in acid leachates from Murchison. LPSC #2405, The Woodlands.
Burkhardt C. and Schönbächler M., 2013. Nucleosynthetic tungsten isotope anomalies in acid leachates of the Orgueil, Murchison, and Allende carbonaceous chondrites. Lunar and Planetary Science Conference (#1912), Houston, TX.
Connelly J. N., Bizzarro M., Krot A. N., Nordlund Å., Wielandt D. and Ivanova M. A. (2012) The Absolute Chronology and Thermal Processing of Solids in the Solar Protoplanetary Disk. Science 338, 651-655.
Davis A. M. and Grossman L. (1979) Condensation and fractionation of rare earths in the solar nebula. Geochimica et Cosmochimica Acta 43, 1611-1632.
Gray C. M., Papanastassiou D. A. and Wasserburg G. J. (1973) Identification of Early Condensates from the Solar Nebula. Icarus 20, 213-239.
Grossman L. and Ganapathy R. (1976) Trace elements in the Allende meteorite—II. Fine-grained. Ca-rich inclusions. Geochimica et Cosmochimica Acta 40, 967-977.
Horan M. F., Smoliar M. I. and Walker R. J. (1998) W-182 and Re-187-Os-187 systematics of iron meteorites: Chronology for melting, differentiation, and crystallization in asteroids. (vol 62, pg 545, 1998). Geochimica et Cosmochimica Acta 62, 1653-1653.
Huang S., Farka) J., Yu G., Petaev M. I. and Jacobsen S. B. (2012) Calcium isotopic ratios and rare earth element abundances in refractory inclusions from the Allende CV3 chondrite. Geochimica et Cosmochimica Acta 77, 252-265.
Humayun M., Simon S. B. and Grossman L. (2007) Tungsten and hafnium distribution in calcium-aluminum inclusions (CAIs) from Allende and Efremovka. Geochimica et Cosmochimica Acta 71, 4609-4627.
162 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
Jacobsen B., Yin Q. Z., Moynier F., Amelin Y., Krot A. N., Nagashima K., Hutcheon I. D. and Palme H. (2008) Al-26-Mg-26 and Pb-207-Pb-206 systematics of Allende CAIs: Canonical solar initial Al-26/Al-27 ratio reinstated. Earth and Planetary Science Letters 272, 353-364.
Kleine T., Hans U., Irving A. J. and Bourdon B. (2012) Chronology of the angrite parent body and implications for core formation in protoplanets. Geochimica et Cosmochimica Acta 84, 186-203.
Kleine T., Mezger K., Munker C., Palme H. and Bischoff A. (2004) Hf-182-W-182 isotope systematics of chondrites, eucrites, and martian meteorites: Chronology of core formation and early mantle differentiation in Vesta and Mars. Geochimica et Cosmochimica Acta 68, 2935-2946.
Kleine T., Mezger K., Palme H., Scherer E. and Munker C. (2005) Early core formation in asteroids and late accretion of chondrite parent bodies: Evidence from Hf-182-W-182 in CAIs, metal-rich chondrites, and iron meteorites. Geochimica et Cosmochimica Acta 69, 5805-5818.
Kleine T., Munker C., Mezger K. and Palme H. (2002) Rapid accretion and early core formation on asteroids and the terrestrial planets from Hf-W chronometry. Nature 418, 952-955.
Kleine T., Touboul M., Bourdon B., Nimmo F., Mezger K., Palme H., Jacobsen S. B., Yin Q. Z. and Halliday A. N. (2009) Hf-W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochimica Et Cosmochimica Acta 73, 5150-5188.
Kornacki A. S. and Wood J. A. (1985) Mineral chemistry and origin of spinel-rich inclusions in the Allende CV3 chondrite. Geochimica et Cosmochimica Acta 49, 1219-1237.
Krot A. N., MacPherson G. J., Ulyanov A. A. and Petaev M. I. (2004) Fine-grained, spinel-rich inclusions from the reduced CV chondrites Efremovka and Leoville: I. Mineralogy, petrology, and bulk chemistry. Meteoritics & Planetary Science 39, 1517-1553.
Krot A. N., Scott E. R. D. and Zolensky M. E. (1995) Mineralogical and chemical modification of components in CV3 chondrites: Nebular or asteroidal prcoessing? Meteoritics 30, 748-775.
Kruijer T. S., Fischer-Gödde M., Kleine T., Sprung P., Leya I. and Wieler R. (2013a) Neutron capture on Pt isotopes in iron meteorites and the Hf–W chronology of core formation in planetesimals. Earth and Planetary Science Letters 361, 162-172.
Kruijer T. S., Sprung P., Kleine T., Leya I., Burkhardt C. and Wieler R. (2012) Hf–W chronometry of core formation in planetesimals inferred from weakly irradiated iron meteorites. Geochimica et Cosmochimica Acta 99, 287-304.
Kruijer T. S., Touboul M., Fischer-Gödde M., Bermingham K., Kleine T. and Walker R. J., 2013b. Resolution of small differences in the time of metal segregation in iron meteorite parent bodiesLunar and Planetary Science Conference, #1920, Houston, TX.
Larsen K. K., Trinquier A., Paton C., Schiller M., Wielandt D., Ivanova M. A., Connelly J. N., Nordlund Å., Krot A. N. and Bizzarro M. (2011) Evidence for Magnesium Isotope Heterogeneity in the Solar Protoplanetary Disk. The Astrophysical Journal Letters 735, L37.
Lee T., Papanastassiou D. A. and Wasserburg G. J. (1976) Demonstration of Mg-26 Excess in Allende and Evidence for Al-26. Geophysical Research Letters 3, 109-112.
Lee T., Papanastassiou D. A. and Wasserburg G. J. (1977) Aluminum-26 in the early solar system: fossil or fuel? Astrophysical Journal Letters 211, L107-L110.
Part B Chapter 8 163
Leya I. (2011) Cosmogenic effects on 7Li/6Li, 10B/11B, and 182W/184W in CAIs from carbonaceous chondrites. Geochimica et Cosmochimica Acta 75, 1507-1518.
Leya I. and Masarik J. (2013) Thermal neutron capture effects in radioactive and stable nuclide systems. Meteoritics & Planetary Science, n/a-n/a.
Leya I., Schonbachler M., Krahenbuhl U. and Halliday A. N. (2009) New Titanium Isotope Data for Allende and Efremovka Cais. Astrophysical Journal 702, 1118-1126.
Lodders K. (2003) Solar system abundances and condensation temperatures of the elements. The Astrophysical Journal 591, 1220-1247.
Lugmair G. W. and Galer S. J. G. (1992) Age and isotopic relationships among the angrites Lewis Cliff 86010 and Angra dos Reis. Geochimica et Cosmochimica Acta 56, 1673-1694.
MacPherson G. J., 2003. 1.08 - Calcium–Aluminum-Rich Inclusions in Chondritic Meteorites. In: Editors-in-Chief: Heinrich, D. H. and Karl, K. T. Eds.), Treatise on Geochemistry. Pergamon, Oxford.
Macpherson G. J., Davis A. M. and Zinner E. K. (1995) The Distribution of Al-26 in the Early Solar-System - a Reappraisal. Meteoritics 30, 365-386.
MacPherson G. J., Kita N. T., Ushikubo T., Bullock E. S. and Davis A. M. (2012) Well-resolved variations in the formation ages for Ca–Al-rich inclusions in the early Solar System. Earth and Planetary Science Letters 331–332, 43-54.
Markowski A., Quitte G., Halliday A. N. and Kleine T. (2006) Tungsten isotopic compositions of iron meteorites: Chronological constraints vs. cosmogenic effects. Earth and Planetary Science Letters 242, 1-15.
Markowski A., Quitte G., Kleine T., Halliday A. N., Bizzarro M. and Irving A. J. (2007) Hafnium-tungsten chronometry of angrites and the earliest evolution of planetary objects. Earth and Planetary Science Letters 262, 214-229.
Martin P. M. and Mason B. (1974) Major and trace elements in the Allende meteorite. Nature 249, 333-334.
Masarik J. (1997) Contribution of neutron-capture reactions to observed tungsten isotopic ratios. Earth and Planetary Science Letters 152, 181-185.
McGuire A. V. and Hashimoto A. (1989) Origin of zoned fine-grained inclusions in the Allende meteorite. Geochimica et Cosmochimica Acta 53, 1123-1133.
McKeegan K. D. and Davis A. M., 2003. 1.16 - Early Solar System Chronology. In: Editors-in-Chief: Heinrich, D. H. and Karl, K. T. Eds.), Treatise on Geochemistry. Pergamon, Oxford.
Niemeyer S. and Lugmair G. W. (1981) Ubiquitous isotopic anomalies in Ti from normal Allende inclusions. Earth and Planetary Science Letters 53, 211-225.
Nyquist L. E., Kleine T., Shih C. Y. and Reese Y. D. (2009) The distribution of short-lived radioisotopes in the early solar system and the chronology of asteroid accretion, differentiation, and secondary mineralization. Geochimica et Cosmochimica Acta 73, 5115-5136.
Podosek F. A., Zinner E. K., Macpherson G. J., Lundberg L. L., Brannon J. C. and Fahey A. J. (1991) Correlated study of initial 87Sr/86Sr and Al-Mg isotopic systematics and petrologic properties in a suite of refractory inclusions from the Allende meteorite. Geochimica et Cosmochimica Acta 55, 1083-1110.
Pourmand A., Dauphas N. and Ireland T. J. (2012) A novel extraction chromatography and MC-ICP-MS technique for rapid analysis of REE, Sc and Y: Revising CI-chondrite and Post-Archean Australian Shale (PAAS) abundances. Chemical Geology 291, 38-54.
164 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
Qin L. P., Dauphas N., Wadhwa M., Masarik J. and Janney P. E. (2008) Rapid accretion and differentiation of iron meteorite parent bodies inferred from Hf-182-W-182 chronometry and thermal modeling. Earth and Planetary Science Letters 273, 94-104.
Scherer P. and Schultz L. (2000) Noble gas record, collisional history, and pairing of CV, CO, CK, and other carbonaceous chondrites. Meteoritics & Planetary Science 35, 145-153.
Scherstén A., Elliott T., Hawkesworth C., Russell S. and Masarik J. (2006) Hf-W evidence for rapid differentiation of iron meteorite parent bodies. Earth and Planetary Science Letters 241, 530-542.
Schiller M., Baker J. A. and Bizzarro M. (2010) Al-26-Mg-26 dating of asteroidal magmatism in the young Solar System. Geochimica et Cosmochimica Acta 74, 4844-4864.
Spivak-Birndorf L., Wadhwa M. and Janney P. (2009) 26Al–26Mg systematics in D’Orbigny and Sahara 99555 angrites: Implications for high-resolution chronology using extinct chronometers. Geochimica et Cosmochimica Acta 73, 5202-5211.
Stracke A., Palme H., Gellissen M., Münker C., Kleine T., Birbaum K., Günther D., Bourdon B. and Zipfel J. (2012) Refractory element fractionation in the Allende meteorite: Implications for solar nebula condensation and the chondritic composition of planetary bodies. Geochimica et Cosmochimica Acta 85, 114-141.
Tanaka T. and Masuda A. (1973) Rare-earth elements in matrix, inclusions, and chondrules of the Allende meteorite. Icarus 19, 523-530.
Wasserburg G. J., Wimpenny J. and Yin Q.-Z. (2012) Mg isotopic heterogeneity, Al-Mg isochrons, and canonical 26Al/27Al in the early solar system. Meteoritics & Planetary Science 47, 1980-1997.
Willbold M., Elliott T. and Moorbath S. (2011) The tungsten isotopic composition of the Earth's mantle before the terminal bombardment. Nature 477, 195-U91.
Wittig N., Humayun M., Brandon A. D., Huang S. and Leya I. (2013) Coupled W–Os–Pt isotope systematics in IVB iron meteorites: In situ neutron dosimetry for W isotope chronology. Earth and Planetary Science Letters 361, 152-161.
Yokoyama T., Alexander C. M. O. D. and Walker R. J. (2011) Assessment of nebular versus parent body processes on presolar components present in chondrites: Evidence from osmium isotopes. Earth and Planetary Science Letters 305, 115-123.
York D. (1966) Least-squares fitting of a straight line. Canadian Journal of Physics 44, 1079-1086.
8.7 APPENDIX: Supplementary text, figures, and tables
8.7.1 Petrographic and chemical classification of CAI
8.7.1.1 Petrographic classification Fragments of most of the investigated CAI were preserved and used to produce polished thin sections for petrographic and chemical classification. A JEOL 6610-LV scanning electron microscope (SEM) was used to generate back-scattered electron (BSE) images and X-ray elemental maps. A brief description of the investigated CAI is given below.
Fine-grained CAI: Allende AF01 (289 mg) is an irregular, fine-grained inclusion that occurs strongly interspersed with Allende matrix material (Fig. 8.5). The interior of the CAI is fine-grained, while the outer rim is more coarse-grained and may show evidence for secondary re-
Part B Chapter 8 165
crystallization. Allende AF02 (230 mg) is another very fine-grained but slightly larger CAI with an Al-rich interior and a thin Ca-rich outer rim (Fig. 8.6). Allende AF03 (37 mg) is a small, and very fine-grained inclusion from Allende that like AF01 seems strongly interspersed with the surrounding matrix. Allende AF04 (288 mg) is a large, irregularly shaped, fine-grained inclusion from Allende, again occurring strongly interspersed with the surrounding matrix (Fig. 8.7). The inclusion is subdivided into a Al-rich interior and a Ca-rich outer-rim. Allende AC01 (52 mg) is a fine-grained inclusion from Allende, comparable to AF01 and AF02.
Fig. 8.5: Back scatter electron (BSE) image and X-ray element map of Allende CAI AF01.
Fig. 8.6: Back scatter electron (BSE) image and X-ray element map of Allende CAI AF02.
166 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
Fig. 8.7: Back scatter electron (BSE) image and X-ray element map of Allende CAI AF04.
Coarse-grained CAI Allende AI06 (168 mg) is a rounded coarse-grained inclusion with a complex internal texture (Fig. 8.8). While the rim of the CAI mostly consists of Al-rich minerals, the interior predominantly consists of Ca-rich minerals that are interspersed with Al-rich and Mg-rich mineral grains. AI06 seems to be a typical type B inclusion. An interesting rounded inclusion with a thin rim is embedded within the CAI interior. Allende AI07 (137 mg) is a coarse-grained CAI from Allende with an irregular morphology (Fig. 8.9). The innermost portion of the CAI mostly consists of Ca-rich minerals, while the very thin outer rim contains proportionally more Al and Mg. Allende AF05 (69 mg) is a coarse-grained inclusion that compositionally seems to be similar to AI06 (Fig. 8.10). The innermost lithology consists mostly of Ca-rich minerals and both fine and coarse-grained Al-rich grains. The outer rim lithology mostly contains Al-rich minerals. Allende AI05 (77 mg) and Allende AC02 (104 mg) are coarse-grained, irregular shaped inclusions, comparable to AF05. NWA 6871 AI01 (82 mg) is a coarse-grained inclusion (Fig. 8.11), and is somewhat comparable to AI07. However, in comparison with CAI from Allende, this inclusion appears relatively pristine and unaltered. The innermost mineralogy mostly consists of Ca-rich minerals (likely melilite) with minor spinel and olivine. The outer rim of the CAI consists of Al-rich spinel and pyroxene. NWA 6737 AI02 (103 mg) is another coarse-grained type B inclusion.
Part B Chapter 8 167
Fig. 8.8: Back scatter electron (BSE) image and X-ray element map of Allende CAI AI06.
Fig. 8.9: Back scatter electron (BSE) image and X-ray element map of Allende CAI AI07.
168 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
Fig. 8.10: Back scatter electron (BSE) image and X-ray element map of Allende CAI AF05.
Fig. 8.11: Back scatter electron (BSE) image and X-ray element map of CAI AI01 from NWA 6871.
8.7.1.2 REE concentration analyses by ICPMS
8.7.1.2.1 Analytical methods Aliquots of several dissolved CAI (2-3 %) were diluted 1000-5000" with 0.28 M HNO3 (i.e., the specific dilution depending on the sample amount available) and then analysed for their REE concentrations using solution quadrupole ICPMS. The measurements were performed on a ThermoScientific® X-Series2 ICPMS at the University of Münster. Dissolved aliquots of a
Part B Chapter 8 169
terrestrial basalt standard (BHVO-2) were measured at different concentrations and were used to construct a calibration line. Aliquots of a powdered Allende 100 g slice (MS-A) and an in-house REE solution standard mixed in CI chondritic proportions were run as unknown standards. The sample solutions were doped with pure In and Tl as internal standard to monitor and correct instrumental drift in sensitivity. The uncertainties of the REE analyses are estimated to be better than 5% (2SD).
8.7.1.2.2 REE results The CI-normalised REE concentrations of Allende and a REE solution standard in CI-chondritic proportions are shown in Fig. 8.12a. The REE pattern of Allende shows perfect agreement with values reported in the literature (Pourmand et al., 2012; Stracke et al., 2012), while the solution standard shows REE concentrations identical to the expected values with a corresponding flat REE pattern. The CI-normalised REE concentrations for several of the investigated CAI are shown in Fig. 8.12b. While one coarse-grained type B inclusion (AF05) shows a typical uniformly enriched, flat ‘Group I’ pattern, all fine-grained CAI studied here show large variability in their REE concentrations, resulting in strongly fractionated REE patterns. Specifically, the fine-grained CAI exhibit uniformly enriched light REE (La, Ce, Pr, Nd, Sm) and Tm, whereas they are variably depleted in the ultra-refractory REE (Gd, Tb, Dy, Ho, Er, Lu) and in the relatively volatile REE Yb and Eu. These characteristics are typical for ‘Group II’ REE patterns (Martin and Mason, 1974) that appear to lack an ultra-refractory component (Boynton, 1975; Davis and Grossman, 1979), and are consistent with Group II patterns previously observed for fine-grained spinel-rich inclusions (e.g., Tanaka and Masuda, 1973; Huang et al., 2012). Note that, although all fine-grained CAI exhibit Group II patterns, the relative REE enrichment (or depletion) varies strongly between the different CAI (e.g, from ~3" to ~240" the CI chondrite abundance for Nd).
170 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
CI-n
orm
aliz
ed R
EE
AF05
AF04
AF03
AF02
AF01
CG CAI:
FG CAI:
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
1
2
3
4
5
REE 2013b meas.
REE 2013b calc.
Allende MS-A meas.
Avg. AllendeStracke et al., 2012, GCAAllende (B)Pourmand et al., 2012, GCA
CI-n
orm
aliz
ed R
EE
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu0.1
1
10
100
1000
(a)
(b)
Fig. 8.12: CI-normalised rare earth element (REE) concentrations of several CAI samples analyzed in this study. Shown are REE data for (a) aliquots from a powdered slice of Allende (CV3) and a pure REE standard (MS 2013b) in chondritic proportions, and (b) four fine-grained and one coarse-grained CAI from this study. All fine-grained CAI exhibit strongly fractionated ‘Group II’ patterns, while one coarse-grained type B inclusion (AF05) shows a flat, unformly enriched ‘Group I’ REE pattern.
Part B Chapter 8 171
8.7.2 Additional data tables
Table 8-A1Tungsten isotope compositions for terrestrial standards.
ID !182Wmeas.(±2SE) !183Wmeas. (±2SE) !182Wmeas. (±2SE) !184Wmeas. (±2SE) !182Wcorr. (±2SE)a
Normalization to 186W/184W = 0.92767 Normalization to 186W/183W = 1.9859
BCR-2AF06_1 0.06 ± 0.08 -0.12 ± 0.07 0.26 ± 0.07 0.08 ± 0.04 0.10 ± 0.11AF06_2 0.06 ± 0.08 -0.12 ± 0.07 0.21 ± 0.07 0.08 ± 0.04 0.05 ± 0.11AF07_1 0.14 ± 0.08 -0.11 ± 0.07 0.22 ± 0.07 0.07 ± 0.04 0.07 ± 0.11AF07_2 0.07 ± 0.08 -0.13 ± 0.07 0.22 ± 0.07 0.09 ± 0.04 0.05 ± 0.11AF07_3 -0.02 ± 0.08 -0.26 ± 0.07 0.38 ± 0.07 0.18 ± 0.04 0.03 ± 0.11AC03_1 -0.09 ± 0.08 0.01 ± 0.07 -0.12 ± 0.07 -0.01 ± 0.04 -0.10 ± 0.11AC03_2 0.09 ± 0.08 0.03 ± 0.07 0.06 ± 0.07 -0.02 ± 0.04 0.10 ± 0.11AC03_3 -0.04 ± 0.08 -0.01 ± 0.07 -0.02 ± 0.07 0.01 ± 0.04 -0.03 ± 0.11AC04_1 0.01 ± 0.08 -0.02 ± 0.07 0.14 ± 0.07 0.01 ± 0.04 0.11 ± 0.11AC04_2 -0.10 ± 0.08 -0.16 ± 0.07 0.14 ± 0.07 0.11 ± 0.04 -0.08 ± 0.11AC04_3 -0.05 ± 0.08 -0.07 ± 0.07 0.05 ± 0.07 0.04 ± 0.04 -0.04 ± 0.11AI09_1 -0.11 ± 0.08 -0.06 ± 0.07 -0.02 ± 0.07 0.04 ± 0.04 -0.11 ± 0.11AI09_2 -0.01 ± 0.08 -0.03 ± 0.07 0.03 ± 0.07 0.02 ± 0.04 -0.01 ± 0.11Average (±2SD) 0.00 ± 0.16 -0.08 ± 0.16 0.12 ± 0.27 0.05 ± 0.11 0.01 ± 0.15
BHVO-2AD06 0.10 ± 0.08 0.12 ± 0.07 -0.09 ± 0.08 -0.08 ± 0.05 0.07 ± 0.13AE04_1 -0.03 ± 0.08 0.00 ± 0.07 -0.01 ± 0.08 0.00 ± 0.05 -0.01 ± 0.13AE04_2 -0.03 ± 0.08 0.00 ± 0.07 0.00 ± 0.08 0.00 ± 0.05 0.01 ± 0.13AE04_3 -0.07 ± 0.08 -0.02 ± 0.07 -0.02 ± 0.08 0.01 ± 0.05 -0.04 ± 0.13AH07_1 -0.06 ± 0.08 0.02 ± 0.07 -0.07 ± 0.08 -0.01 ± 0.05 -0.04 ± 0.13AH07_2 -0.05 ± 0.08 -0.15 ± 0.07 0.15 ± 0.08 0.10 ± 0.05 -0.05 ± 0.13AH08_1 -0.08 ± 0.08 -0.08 ± 0.07 0.00 ± 0.08 0.05 ± 0.05 -0.10 ± 0.13AH08_2 -0.06 ± 0.08 0.02 ± 0.07 -0.07 ± 0.08 -0.01 ± 0.05 -0.04 ± 0.13Average (±2SD) -0.04 ± 0.11 -0.01 0.16 -0.01 ± 0.15 0.01 ± 0.10 -0.02 ± 0.10
a Corrected for a mass-independent effect on 183W
172 Nucleosynthetic W isotope anomalies and Hf-W chronology of CAI
Tabl
e 8-
A2
Hf-W
dat
a fo
r fiv
e di
gesti
on re
plic
ates
of a
pow
dere
d sla
b (M
S-A
) of t
he A
llend
e ca
rbon
aceo
us c
hond
rite
(CV
3).
IDW
(ppb
) H
f (pp
b)
180 H
f/184 W
(±2!
)"18
2 Wm
eas.(±
2SD
)"18
3 Wm
eas. (
±2SD
)"18
2 Wi (
±2SD
)a"18
2 Wm
eas. (±
2SD
)"18
4 Wm
eas. (
±2SD
)"18
2 Wi (
±2SD
)a
Nor
mal
izat
ion
to 1
86W
/184 W
Nor
mal
izat
ion
to 18
6 W/18
3 W
AH
0115
017
81.
394
±0.
008
-1.9
9±
0.11
-0.0
2±
0.16
-3.6
5±
0.14
-2.0
0±
0.15
0.01
±0.
10-3
.66
±0.
17A
H02
164
187
1.34
1±
0.00
8-1
.98
±0.
110.
04±
0.16
-3.5
7±
0.14
-2.0
0±
0.15
-0.0
2±
0.10
-3.6
0±
0.17
AH
0316
018
21.
347
±0.
008
-1.9
3±
0.11
0.08
±0.
16-3
.53
±0.
14-2
.08
±0.
15-0
.05
±0.
10-3
.68
±0.
17A
H04
161
184
1.35
5±
0.00
8-2
.00
±0.
110.
00±
0.16
-3.6
2±
0.14
-2.0
1±
0.15
0.00
±0.
10-3
.62
±0.
17A
H05
163
182
1.31
9±
0.00
8-1
.95
±0.
110.
02±
0.16
-3.5
2±
0.14
-1.9
9±
0.15
-0.0
1±
0.10
-3.5
6±
0.17
Aver
age
(±95
% c
onf.,
n=5
)-1
.97
±0.
040.
02±
0.05
-3.5
8±
0.07
-2.0
2±
0.05
-0.0
1±
0.03
-3.6
2±
0.06
Unc
erta
intie
s on
indi
vidu
al m
easu
rem
ents
(2SD
) wer
e es
timat
ed u
sing
repl
icat
e an
alys
es (T
able
8-A
1) o
f a te
rrestr
ial b
asal
t sta
ndar
d (B
HV
O-2
).a C
orre
cted
for r
adio
geni
c co
ntrib
utio
ns fr
om 18
2 Hf d
ecay
usin
g th
eir m
easu
red18
0 Hf/18
4 W a
nd th
e so
lar s
yste
m in
itial
182 H
f/180 H
f of (
1.03
±0.0
5)#1
0-5.
Part B Chapter 8 173
8.7.3 Additional figures
Fig. 8.13: External reproducibility of !182W (6/4) of the terrestrial basalt standards (a) BCR-2 and (b) BHVO-2 that were analysed in this study. Each symbol represents a single measurement (i.e, 1" 200 cycles) and their error bars show the associated within-run standard error of the mean (2SE). Different colours denote separate digestions of the basalt standards. Hatched areas show the long-term external reproducibility (2SD).
Fig. 8.14: !182W vs. 1/W for the CAI analysed in this study. The data do not define a correlation, indicating that the !182W – 180Hf/184W correlation (Fig. 8.4) cannot be explained by mixing between two end members.
ε182
W (6
/4)
Terrestrial basalt standard (BCR-2):ε182Wmean = 0.00±0.16 (2SD)
(a)
ε182
W (6
/4)
Terrestrial basalt standard (BHVO-2):ε182Wmean = −0.04±0.11 (2SD)
(b)
–0.4
–0.2
0
0.2
–0.4
–0.2
0
0.2
0 0.005 0.01 0.015–4
–2
0
2
1/W
ε182
W (6
/3) n
uc. c
orr.
fine-grained CAI
coarse-grained CAI
174 180W anomalies in iron meteorites: Implications for p-process heterogeneity
Chapter 9
180W Anomalies in Iron Meteorites: Implications for p-
Process Heterogeneity in the Solar Nebula
Cook, D.L.,a,b Kruijer, T.S.a,b1, Kleine, T.a
aWestfälische Wilhelms-Universität Münster, Institut für Planetologie, Münster, Germany. bETH Zürich, Institute of Geochemistry and Petrology, Zürich, Switzerland.
Submitted to Geochimica et Cosmochimica Acta (2013)
Abstract We measured tungsten (W) isotopes in 23 iron meteorites and the metal phase of the CB chondrite Gujba in order to ascertain if there is evidence for a large-scale nucleosynthetic heterogeneity in the p-process isotope 180W in the early solar nebula as recently suggested by Schulz et al. (2013). We observed large excesses in 180W (up to ! 6 ") in some irons. However, significant within-group variations in magmatic IIAB and IVB irons are not consistent with a nucleosynthetic origin. We consider the possible collateral effects on 180W from an s-deficit that is present in some iron groups, as well as the effects from 180W burnout by neutron capture reactions and the possible radiogenic in-growth of 180W from #-decay of 184Os. The presence of an s-deficit and the effects of 180W burnout generally cause only small shifts in 180W (i.e., < 1 ") and fail to explain the total observed variability. Additionally, the within-group variations are not easily reconciled with the expectations for a radiogenic origin from 184Os decay. Alternatively, the 180W nuclear reactions induced by interactions with cosmic rays may have played a role in generating the observed 180W excesses in iron meteorites, but additional work is required to test this idea.
1 As second author, I contributed to the research described in this chapter through (i) assistance with the
chemical separation of W, and (ii) involvement in the interpretation of the W isotope data, in particular with regard to the relative contributions of nucleosynthetic, radiogenic, and cosmogenic W isotope variations.
Part B Chapter 9 175
9.1 Introduction The investigation of W isotope anomalies in meteorites is of interest because of the former presence of the short-lived isotope 182Hf, which decays to 182W (t1/2 = 8.9 Ma; Vockenhuber et al., 2004). The 182Hf-182W chronometer can be used to infer timescales of accretion and core segregation on asteroids and terrestrial planets and of magmatic and metamorphic events on meteorite parent bodies (e.g., Kleine et al., 2009). However, the application of the 182Hf-182W chronometer relies on the assumption that W isotopes were homogeneously distributed in the solar nebula. This assumption appears to be valid for most meteoritic samples, although small deficits in s-process W isotopes have been observed in group IID and IVB irons, as well as some ungrouped irons (Qin et al., 2008; Kruijer et al., 2012; Kruijer et al., 2013a), CAIs (Burkhardt et al., 2008), and acid leachates of primitive chondrites (Burkhardt et al., 2012a,b).
The p-process isotopes are among the rarest isotopes in the solar system with relative isotopic abundances typically of a few percent or less (Nuclides and Isotopes, 2002). This feature could make the relative isotopic abundances of the p-process nuclides susceptible to mixing processes in the early nebula between materials produced in the different stellar sources that contributed the gas and dust from which our solar system ultimately formed. Precise measurements of p-process isotopes in meteorites may provide a means for quantifiying the magnitude and/or scale of any such initial nebular heterogeneities. Anomalies in several of the light p-process isotopes have been reported including Sr (Moynier et al., 2012), Mo (Burkhardt et al., 2011), Ru (Chen et al., 2010), and Sm (Andreasen and Sharma, 2006; Qin et al., 2011). The anomalies in Mo and Ru can be explained by small deficits in s-process isotopes of those elements rather than a true variation in the p-process isotopes in the early nebula. The results for Sr are ambiguous but probably also reflect an s- or r-process variability (Hans et al., 2013). Of these elements, only Sm isotope anomalies appear to represent acutual p-process heterogeneity (Qin et al., 2011), but the variations are rather small (i.e., ±1 "). Hence, previous studies do not suggest the presence of a general, large-scale p-process variability in the solar nebula.
Recently, excesses in 180W have been reported in several magmatic iron groups, and these data were used to argue for large-scale p-process heterogeneities in the inner nebula that were increasingly homogenized with time (Schulz et al., 2013). The relative abundance of 180W (0.12%; Völkening et al., 1991) is one of the smallest in the solar system (Arnould and Goriely, 2003). Thus, it may be a more sensitive tracer than the more abundant light p-process isotopes to isotopic variability within the nebula. We report isotopic measurements for magmatic irons and metal from the CB chondrite Gujba to examine the extent of possible nucleosynthetic W isotope anomalies in the early solar system and to assess the origin of the anomalies in the p-process isotope 180W.
9.2 Methods
9.2.1 Samples and Preparation We analysed 22 magmatic iron meteorites from the following groups: IIAB, IID, IVA, and IVB. In addition, we analysed the ungrouped iron Chinga and metal separated from the CB chondrite Gujba. Iron meteorites were cut with a slow speed saw and then mechanically cleaned with SiC paper. For Gujba, metal was handpicked after lightly crushing the sample. Metal chips of 0.5 g were used for all samples except for Muonionalusta (1.5 g). Similarly, 0.75 g of metal was separated from Gujba for analysis. For an external standard, we chose an NIST Fe-Ni steel (SRM 129c); aliquots of 0.5 g were used to match the sample size of most
176 180W anomalies in iron meteorites: Implications for p-process heterogeneity
meteorite samples. All meteoritic metal and aliquots of SRM 129c were digested in Teflon beakers using 15 ml of a 2:1 mixture of HNO3:HCl at 130 °C. After digestion, the samples were dried completely and fluxed overnight in 12 ml HCl at 130 °C to remove traces of HNO3. The samples were then dried and taken up in 75 ml of 0.5M HCl-0.5M HF for loading onto ion exchange columns. Additionally, we processed several aliquots of the NIST W solution standard (SRM 3163) as a check on our analytical techniques. These aliquots, equivalent to 1500 ng of W, were subjected to the same protocol above.
9.2.2 Separation of Tungsten The chemical separation followed that described by Kruijer et al. (2012), which is based on protocols from Horan et al. (1998) and Kleine et al. (2002; 2004). Briefly, sample solutions were loaded onto pre-cleaned anion exchange columns filled with 4 ml of resin (AG1$8; 200-400 mesh). The matrix was washed off with 10 ml of 0.5M HCl-0.5M HF, and W was eluted in 15 ml of 6M HCl-1M HF; the Muonionalusta and Gujba samples were split over 3 and 2 columns, respectively, to avoid overloading the columns, and the W cuts were recombined. All W cuts were evaporated and then dried twice in a 1:1 mixture of HNO3:H2O2. Lastly, the samples were taken up in 5 ml of 0.5M HCl-0.5M HF for loading onto a second clean-up column filled with 1 ml of anion resin (AG1$8; 200-400 mesh). After loading, the matrix was washed off with 4 ml 0.5M HCl-0.5M HF, 4.5 ml 8M HCl-0.01M HF, and 0.5 ml 6M HCl-1M HF; tungsten was eluted in 5 ml 6M HCl-1M HF. The purified W cuts were evaporated and dried twice in HNO3:H2O2. Finally, the samples were dissolved in 0.56M HNO3-0.24M HF for isotopic analyses. Total chemical yields were highly variable and typically were between ~50 to ~80%. The full procedural blank was ~230 pg, which was insignificant compared to the amount of sample W processed (~1500 ng).
9.2.3 Mass Spectrometry Isotopic measurements were made with the ThermoScientific Neptune Plus MC-ICPMS (University of Münster) in low resolution mode using standard sampler and skimmer Ni cones; several early analyses were also made using a combination of the Jet sampler and X-skimmer cones, but this routine was abandoned (see Section 9.3.1). Samples were introduced using a Cetac Aridus II desolvating system with an uptake rate of 75 %L/min. Each sample analysis consisted of a single measurement (20 cycles of 8.4 s) bracketed by measurements of the NIST W solution (SRM 3163). Measurements were made in static mode. All five W isotopes (180W, 182W, 183W, 184W, 186W) were measured simultaneously. We also monitored 178Hf, 181Ta, and 188Os which were used to correct interferences on 180W from 180Hf and 180Ta and on 184W and 186W from 184Os and 186Os. Signal intensities for both 180W and 178Hf were measured using 1012 Ohm resistors. All measurements were preceded by a 30 s measurement of the electronic background of the instrument which was subtracted from all signal intensities. Sample solutions of ! 400 to ! 500 ppb were analysed, which provided a beam of ! 45 V on 184W and ! 160 mV on 180W. Instrumental mass bias was corrected using the exponential law with either 186W/183W = 1.98594 or 186W/184W = 0.927672 (Völkening et al., 1991).
9.3 Results All data are reported in epsilon units (") with the following notation: "iW (6/3) or "iW (6/4), where i = 180, 182, or 184 and (6/3) or (6/4) indicates the ratio used for the mass bias
Part B Chapter 9 177
correction. An epsilon unit represents the relative difference of an isotope ratio between a sample and the bracketing standards in parts per 10,000.
9.3.1 Terrestrial Samples
9.3.1.1 183W Deficits The analyses of SRM 129c show small, non-mass dependent effects on 183W. Such effects have previously been reported by Willbold et al. (2011) and Kruijer et al. (2012) and were ascribed to non-mass dependent fractionation of 183W relative to the even isotopes of W related to W loss during chemical processing. This effect causes an apparent deficit in 183W and affects the "iW values for which 183W is used to correct the instrumental mass bias. However, Kruijer et al. (2012) showed that this effect can accurately be corrected using the measured "184W (6/3), and we employ similar corrections. The values for "180W (6/4) and "182W (6/4) are unaffected by the artefact on 183W and no correction is necessary.
-10 -8 -6 -4 -2 0 2 4
X/Jet-Cones SRM 3163 SRM 129c
H-Cones SRM 3163 SRM 129c
!180W (6/4)
Fig. 9.1: !180W (6/4) for replicate analyses of SRM 129c and of SRM 3163 processed through our full chemical separation protocol. Results are shown for measurements using a combination of Jet sampler and X-skimmer cones as well as those made using standard (H) cones. Analytical artifacts using Jet and X-cones are clearly resolvable. The external precision (± 1.19; 2 S.D.; n = 29) based on standard cone measurements of SRM 129c is shown by the gray vertical lines.
9.3.1.2 180W Measurements Figure 9.1 shows the results for "180W (6/4) for analyses of SRM 129c and those of SRM 3163 which were processed through our full chemical separation protocol. Initially, isotopic measurements were conducted using a combination of Jet sampler and X-skimmer cones. These cones are intended to provide increased sensitivity which could be useful for measurements of low abundance isotopes such as 180W. However, this configuration resulted in inaccurate measurements of "180W (6/4); deficits up to -9.11 " were observed in SRM 129c. The terrestrial W in SRM 129c is not expected to differ from that of SRM 3163 used as the
178 180W anomalies in iron meteorites: Implications for p-process heterogeneity
bracketing standard. Moreover, deficits up to -4.10 " were observed for the aliquots of SRM 3163 processed through chemistry. The same deficits also occur on "180W (6/3). These results demonstrate the presence of an analytical artefact using this cone configuration, which likely relates to a matrix effect introduced during the chemical purification of W. The other W isotope ratios were not adversely affected. Similar effects were reported by Schulz et al. (2013), although only in high-resolution mode. These results suggest caution when using Jet and X-cones in combination to measure low abundance isotopes. All subsequent measurements were made using standard sampler and skimmer cones (i.e., H-cones); these results are also shown in Fig. 9.1 and demonstrate that our total analytical protocol is free from artefacts when measuring with standard cones. Multiple replicates of SRM 129c were measured during each analytical session over the entire course of the study (i.e., 1 year). The long-term reproducibility (2 S.D.) based on SRM 129c for "180W (6/4) was ± 1.19 (n = 29). Internal precisions on "180W (6/4) for a single measurement were mostly better than 0.85 ". The external reproducibility for the other ratios were as follows: "182W (6/4) ± 0.10, "180W (6/3) ± 1.24, "182W (6/3) ± 0.11, and "184W (6/3) ± 0.08.
9.3.2 Meteoritic Samples All reported data are for isotopic measurements using only standard sampler and skimmer cones. Figure 9.2 shows the measured values for "180W (6/4) for all meteoritic metal samples. The ungrouped iron Chinga shows a small positive excess in "180W. Many of the IVB irons also show resolvable excesses, although two do not, and there is a high degree of variation among the IVB irons. One IIAB iron (Forsyth County) shows a clearly resolved excess, whereas one (Mt. Joy) shows an apparent deficit; most other IIAB irons have values overlapping with that of the terrestrial standard. The IID and IVA irons, as well as the CB chondrite metal, show no resolvable anomalies.
-4 -2 0 2 4 6 8
CB IVA UnG IID IIAB IVB
!180W (6/4)
Fig. 9.2: !180W (6/4) for analyses of meteoritic metal. Gray vertical lines represent the external precision of ± 1.19 based on the analyses of SRM 129c (2 S.D.).
Part B Chapter 9 179
Figure 9.3 shows the values for "184W (6/3) for all meteoritic metal samples. Most of the IVB irons are characterized by small deficits. The IID iron Carbo also shows a small, resolvable deficit of -0.09 ± 0.07 " (2 S.E.). No other samples show resolvable anomalies in "184W (6/3). The data plotted in Fig. 9.2 and 9.3 are also presented in Table 9.1 along with the values for "180W (6/3), "182W (6/3), and "182W (6/4). The latter two values can be used to assess the accuracy of the correction for the effects on 183W described in Section 9.3.1.1. In principle, both normalizations (6/3) or (6/4) should yield identical results. We can compare "182W (6/3) and "182W (6/4) in the IIAB irons for which no other correction is needed for the presence of an s-process deficit (see Section 9.4.2). The data in Table 9.1 show that after applying the correction to "182W (6/3), both normalizations give values that differ by at most 0.02 " and demonstrate the accuracy and validity of the correction. A comparison between "180W (6/3) and "180W (6/4) for IIAB irons provides additional validation.
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
CB IVA UnG IID IIAB IVB
!184W (6/3)
Fig. 9.3: !184W (6/3) for analyses of meteoritic metal. Gray vertical lines represent the external precision of ± 0.08 based on the analyses of SRM 129c (2 S.D.).
180 180W anomalies in iron meteorites: Implications for p-process heterogeneity
Tabl
e 1
Tun
gste
n is
otop
ic c
ompo
sitio
ns o
f met
eorit
ic m
etal
. Val
ues a
re e
xpre
ssed
rela
tive
to S
RM
316
3 in
par
ts p
er 1
0,00
0. T
he
nota
tion
“(6/
3)”
and
“(6/
4)”
indi
cate
the
ratio
use
d fo
r mas
s bia
s cor
rect
ion
(i.e., 18
6 W/18
3 W o
r 186 W
/184 W
). V
alue
s nor
mal
ized
usi
ng
186 W
/183 W
hav
e be
en c
orre
cted
for a
non
-mas
s dep
ende
nt e
ffec
t (se
e se
ctio
n 3.
1.1)
. SE
is th
e st
anda
rd e
rror
for t
he m
easu
rem
ent.
Sam
ple
Gro
up
!180
W (6
/3)
± 2
SE
!182
W (6
/3)
± 2
SE
!184
W (6
/3)
± 2
SE
!180
W (6
/4)
± 2
SE
!182
W (6
/4)
± 2
SE
Bra
unau
II
AB
1.
33 ±
0.6
4 -3
.39 ±
0.08
-0
.06 ±
0.05
1.
33 ±
0.6
4
-3.
39 ±
0.1
0 C
alic
o R
ock
IIA
B
0.04
± 0
.79
-3.4
4 ±
0.06
-0
.05 ±
0.05
0.
04 ±
0.7
8
-3.
44 ±
0.0
7 C
inci
nnat
i II
AB
1.
22 ±
0.7
0 -3
.42 ±
0.08
0.
00 ±
0.0
5 1.
22 ±
0.7
6
-3.
41 ±
0.0
9 C
oahu
ila
IIA
B
0.85
± 0
.56
-3.4
3 ±
0.08
0.
02 ±
0.0
4 0.
85 ±
0.5
6
-3.
43 ±
0.0
8 Fo
rsyt
h C
ount
y II
AB
2.
62 ±
0.7
6 -3
.47 ±
0.08
-0
.05 ±
0.04
2.
62 ±
0.8
0
-3.
47 ±
0.0
8 H
olla
nd’s
Sto
re
Mou
nt Jo
y N
orth
Chi
le
San
Fran
cisc
o d.
M.
IIA
B
IIA
B
IIA
B
IIA
B
1.52
± 0
.75
-2.1
3 ±
0.82
1.
29 ±
0.5
0 1.
71 ±
0.8
1
-3.4
7 ±
0.08
-3
.90 ±
0.07
-3
.50 ±
0.09
-3
.42 ±
0.05
-0.0
8 ±
0.03
-0
.03 ±
0.05
0.
00 ±
0.0
5 -0
.07 ±
0.03
1.52
± 0
.76
-2.1
3 ±
0.85
1.
29 ±
0.4
4 1.
68 ±
0.8
2
-
3.47
± 0
.09
-
3.90
± 0
.09
-
3.51
± 0
.08
-
3.40
± 0
.06
Car
bo
IID
0.
19 ±
0.5
8 -4
.19 ±
0.11
-0
.05 ±
0.05
0.
44 ±
0.5
7
-4.
02 ±
0.0
7 R
odeo
II
D
1.04
± 0
.54
-3.1
5 ±
0.11
-0
.09 ±
0.07
1.
29 ±
0.6
1
-2.
99 ±
0.1
0
Muo
nion
alus
ta
IVA
0.
36 ±
0.5
2 -3
.38 ±
0.12
0.
05 ±
0.0
5 0.
36 ±
0.5
0
-3.
38 ±
0.1
1
Dum
ont
IVB
4.
72 ±
0.7
9 -3
.65 ±
0.06
-0
.13 ±
0.04
4.9
9 ±
0.78
-3.
47 ±
0.0
8 H
oba
IVB
2.
76 ±
0.5
9 -3
.50 ±
0.10
-0
.06 ±
0.05
3.
03 ±
0.5
0
-3.
32 ±
0.0
8 Iq
uiqu
e
IVB
5.
92 ±
0.7
5 -3
.57 ±
0.07
-0
.17 ±
0.05
6.
19 ±
0.7
6
-3.
39 ±
0.0
8 Sa
nta
Cla
ra
IVB
2.
41 ±
1.0
2 -3
.56 ±
0.10
-0
.11 ±
0.06
2.
68 ±
1.0
6
-3.
38 ±
0.1
4 Sk
ooku
m
IVB
3.
19 ±
0.6
3 -3
.49 ±
0.08
-0
.04 ±
0.04
3.
46 ±
0.6
5
-3.
29 ±
0.0
9 Ta
wal
lah
Val
ley
IVB
0.
77 ±
0.6
3 -3
.56 ±
0.09
-0
.07 ±
0.06
1.
04 ±
0.6
8
-3.
38 ±
0.1
0 Ti
nnie
IV
B
3.12
± 0
.82
-3.4
3 ±
0.06
-0
.16 ±
0.04
3.
40 ±
0.8
0
-3.
25 ±
0.0
8 Tl
acot
epec
IV
B
3.68
± 0
.56
-4.2
4 ±
0.07
-0
.10 ±
0.05
3.
95 ±
0.4
8
-4.
06 ±
0.0
7 W
arbu
rton
Ran
ge
IVB
2.
13 ±
0.5
1 -3
.41 ±
0.11
-0
.09 ±
0.04
2.
40 ±
0.5
5
-3.
23 ±
0.1
0 W
eave
r Mou
ntai
ns
IVB
1.
05 ±
0.8
3 -3
.36 ±
0.11
-0
.16 ±
0.07
1.
33 ±
0.8
5
-3.
14 ±
0.0
8
Chi
nga
UnG
1.
43 ±
1.2
0 -3
.18 ±
0.10
-0
.07 ±
0.07
1.
90 ±
1.2
0
-3.
00 ±
0.0
9
Guj
ba
CB
0.
11 ±
1.3
8 -2
.86 ±
0.15
-0
.06 ±
0.07
0.
31 ±
1.3
3
-2.
86 ±
0.1
3
Part B Chapter 9 181
Tab
le
2. T
ungs
ten
isot
opic
com
posi
tions
of m
eteo
ritic
met
al a
fter c
orre
ctin
g fo
r the
eff
ects
of a
n s-
defic
it in
IID
, IV
B a
nd U
nG
sam
ples
. All
valu
es a
re e
xpre
ssed
rela
tive
to S
RM
316
3 in
par
ts p
er 1
0,00
0. T
he n
otat
ion
“(6/
3)”
and
“(6/
4)”
indi
cate
the
ratio
use
d fo
r mas
s bia
s cor
rect
ion
(i.e., 18
6 W/18
3 W o
r 186 W
/184 W
). SE
is th
e st
anda
rd e
rror
for t
he m
easu
rem
ent.
a Sam
ples
for w
hich
aliq
uots
wer
e al
so a
naly
zed
for P
t iso
tope
s (K
ruije
r et a
l., 2
013)
. b V
alue
s als
o co
rrec
ted
for t
he e
ffec
ts fr
om n
eutro
n ca
ptur
e (s
ee se
ctio
n 4.
3).
Sam
ple
Gro
up
!180 W
(6/3
) ±
2SE
!18
2 W (6
/3)
± 2
SE
!180 W
(6/4
) ±
2SE
!18
0 W (6
/4) b
± 2
SE
!182 W
(6/4
) ±
2SE
B
raun
au
IIA
B
1.33
± 0
.64
-3.3
9 ±
0.08
1.
33 ±
0.6
4 1.
33 ±
0.6
4
-3.
39 ±
0.1
0 C
alic
o R
ock
IIA
B
0.04
± 0
.79
-3.4
4 ±
0.06
0.
04 ±
0.7
8 0.
11 ±
0.7
8
-3.
44 ±
0.0
7 C
inci
nnat
i II
AB
1.
22 ±
0.7
0 -3
.42 ±
0.08
1.
22 ±
0.7
6 1.
24 ±
0.7
6
-3.
41 ±
0.0
9 C
oahu
ila
IIA
B
0.85
± 0
.56
-3.4
3 ±
0.08
0.
85 ±
0.5
6 0.
91 ±
0.5
6
-3.
43 ±
0.0
8 Fo
rsyt
h C
ount
y II
AB
2.
62 ±
0.7
6 -3
.47 ±
0.08
2.
62 ±
0.8
0 2.
74 ±
0.8
0
-3.
47 ±
0.0
8 H
olla
nd’s
Sto
re
Mou
nt Jo
y N
orth
Chi
le
San
Fran
cisc
o d.
M.
IIA
B
IIA
B
IIA
B
IIA
B
1.52
± 0
.75
-2.1
3 ±
0.82
1.
29 ±
0.5
0 1.
71 ±
0.8
1
-3.4
7 ±
0.08
-3
.90 ±
0.07
-3
.50 ±
0.09
-3
.42 ±
0.05
1.52
± 0
.76
-2.1
3 ±
0.85
1.
29 ±
0.4
4 1.
68 ±
0.8
2
1.64
± 0
.76
-1.3
6 ±
0.85
1.
47 ±
0.4
4 1.
69 ±
0.8
2
-
3.47
± 0
.09
-
3.90
± 0
.09
-
3.51
± 0
.08
-
3.40
± 0
.06
Car
bo
IID
-0
.05 ±
0.58
-4
.23 ±
0.11
-0
.04 ±
0.57
1.
49 ±
0.5
7
-4.
22 ±
0.0
7 R
odeo
II
D
0.80
± 0
.54
-3.1
9 ±
0.11
0.
81 ±
0.6
1 0.
79 ±
0.6
1
-3.
19 ±
0.1
0
Muo
nion
alus
ta
IVA
0.
36 ±
0.5
2 -3
.38 ±
0.12
0.
36 ±
0.5
0 0.
46 ±
0.5
0
-3.
38 ±
0.1
1
Dum
ont a
IVB
4.
44 ±
0.7
9 -3
.70 ±
0.06
4.4
3 ±
0.78
5.1
0 ±
0.78
-3.
71 ±
0.0
8 H
oba
a IV
B
2.48
± 0
.59
-3.5
5 ±
0.10
2.
47 ±
0.5
0 2.
92 ±
0.5
0
-3.
55 ±
0.0
8 Iq
uiqu
e a
IVB
5.
64 ±
0.7
5 -3
.62 ±
0.07
5.
63 ±
0.7
6 6.
17 ±
0.7
6
-3.
62 ±
0.0
8 Sa
nta
Cla
ra
IVB
2.
13 ±
1.0
2 -3
.61 ±
0.10
2.
12 ±
1.0
6 2.
66 ±
1.0
6
-3.
62 ±
0.1
4 Sk
ooku
m a
IVB
2.
91 ±
0.6
3 -3
.53 ±
0.08
2.
90 ±
0.6
5 3.
31 ±
0.6
5
-3.
53 ±
0.0
9 Ta
wal
lah
Val
ley
IVB
0.
49 ±
0.6
3 -3
.61 ±
0.09
0.
48 ±
0.6
8 1.
02 ±
0.6
8
-3.
62 ±
0.1
0 Ti
nnie
IV
B
2.84
± 0
.82
-3.4
8 ±
0.06
2.
84 ±
0.8
0 3.
18 ±
0.8
0
-3.
49 ±
0.0
8 Tl
acot
epec
a IV
B
3.40
± 0
.56
-4.2
9 ±
0.07
3.
39 ±
0.4
8 4.
94 ±
0.4
8
-4.
29 ±
0.0
7 W
arbu
rton
Ran
ge a
IVB
1.
85 ±
0.5
1 -3
.46 ±
0.11
1.
84 ±
0.5
5 2.
15 ±
0.5
5
-3.
46 ±
0.1
0 W
eave
r Mou
ntai
ns a
IVB
0.
77 ±
0.8
3 -3
.41 ±
0.11
0.
77 ±
0.8
5 0.
94 ±
0.8
5
-3.
38 ±
0.0
8
Chi
nga
UnG
1.
15 ±
1.2
0 -3
.23 ±
0.10
1.
17 ±
1.2
0 1.
17 ±
1.2
0
-3.
24 ±
0.0
9
Guj
ba
CB
0.
11 ±
1.3
8 -2
.86 ±
0.15
0.
00 ±
1.3
3 0.
00 ±
1.3
3
-2.
86 ±
0.1
3
182 180W anomalies in iron meteorites: Implications for p-process heterogeneity
9.4 Discussion
9.4.1 Anomalies in 180W
9.4.1.1 Accuracy of 180W Measurements Similar to the results of Schulz et al. (2013), we observe excesses in "180W (6/4) up to 6.19 " in meteoritic metal (see Fig. 9.2; Table 9.1). Before discussing the significance of these results, we consider several potential analytical biases that could affect the "180W (6/4) values. Multiple aliquots of SRM 3163 were doped with varying levels of Hf to test the interference correction on 180W from 180Hf. The results of these measurements are shown in Fig. 9.4 along with the results for all replicates of SRM 129c and all meteoritic metal samples. The Hf/W ratios for SRM 129c and the meteorite samples span the same range, and there is no correlation between "180W (6/4) and Hf/W. Hence, these data demonstrate that the interference correction is accurate for the range of Hf/W ratios measured in all samples and is not the cause of the 180W excesses. An isobaric interference on 180W can also occur from 180Ta. However, the largest Ta/W ratio measured for any sample was 6.4 $ 10-6 which corresponds to a correction on "180W (6/4) of <0.01 " and cannot be the cause of the observed excesses.
10-7 10-6 10-5 10-4-4
-2
0
2
4
6
8
SRM 3163 SRM 129c metals
!180 W
(6/4
)
Hf/W
Fig. 9.4: !180W (6/4) versus the Hf/W ratio. Measurements of aliquots of the NIST W solution SRM 3163 doped with various amounts of Hf (open triangles) are shown with all replicate measurements of SRM 129c and all measurements of meteoritic metals. Gray horizontal lines represent the external precision of ± 1.19 based on the analyses of SRM 129c (2 S.D.).
Because the yields for W are less than 100%, it is possible that mass-dependent isotopic
fractionation may occur during the chemical separation. If such fractionation occurs, the exponential law used to correct for the instrumental mass bias may be inadequate and lead to spurious results (e.g., Zhang et al., 2011). In Fig. 9.5, we plot the results for the mass-bias corrected 180W/184W (6/4) ratio against the raw 186W/184W ratio used for the correction for the analyses of SRM 129c and iron meteorites during a single analytical session for which several irons show resolvable excesses in "180W (6/4); each of the individual cycles is plotted for five
Part B Chapter 9 183
replicates of SRM 129c and the IVB irons Dumont, Iquique, Tinnie, Tlacotepec, and the ungrouped iron Chinga. The replicates of SRM 129c and the irons show a similar degree of isotopic fractionation as revealed by the measured 186W/184W ratios, and there is no correlation between this ratio and the normalized 180W/184W (6/4) ratio. Thus, the 180W excesses cannot be due to an inaccurate correction for any mass fractionation present in the samples. Figure 9.6 shows the results for 180W measurements using both mass-bias correction schemes. Within uncertainty, all samples lie along a line with a slope of 1. Because the magnitude of the 180W variations is the same for both "180W (6/3) and "180W (6/4), this shows that the anomalies as seen in Fig. 9.2 are not an artefact of the mass-bias correction.
0.9442 0.9443 0.9444 0.9445
0.003876
0.003878
0.003880
0.003882
0.003884
180 W
/184 W
(6/4
)
SRM 129c irons
186W/184W (meas)
Fig. 9.5: The mass-bias corrected 180W/184W (6/4) ratio versus the measured 186W/184W ratio. Data are for a single analytical session during which 5 replicates of SRM 129c were measured along with 5 iron meteorites which show !180W (6/4) excesses up to 6.19.
Holst et al. (2011) suggested that the presence of organic molecules (namely Cn
+ and CnHn+
species) acting as isobaric interferences could cause both apparent excesses in 180W, as well as apparent deficits in 184W (e.g., Qin et al., 2008). The metal phase of most magmatic iron meteorites contains very little C (Buchwald, 1977), but organics could be introduced by the ion exchange separations. In this case, all samples should be affected equally. The results shown in Fig. 9.1 for standard cone measurements of SRM 129c and SRM 3163 processed using our W separation chemistry do not suggest the presence of an unresolved interference; only meteoritic metals show large excesses in 180W (see Fig. 9.2). To further test our W separation chemistry, the matrix eluted during the initial purification of the IVB iron Hoba was collected, dried down, and doped with terrestrial W (i.e., SRM 3163). This sample was then re-subjected to the entire sample digestion and W separation chemistry protocol. The initial results for Hoba show a resolvable excess in "180W (6/4) (see Table 9.1). Conversely, after re-processing the sample, the "180W (6/4) value is -0.07 ± 0.40 (2 S.E.); the "184W (6/3) value also is normal within uncertainty (0.02 ± 0.05). This test corroborates the above results that no artefacts are introduced by our chemical separation. Hence, we reject the hypothesis
184 180W anomalies in iron meteorites: Implications for p-process heterogeneity
that anomalies in 180W and 184W result solely from the presence of organic molecules during isotopic measurements made in low-resolution mode.
-4 -2 0 2 4 6 8-4
-2
0
2
4
6
8 CB IVA UnG IID IIAB IVB
!180 W
(6/4
)
!180W (6/3)
Fig. 9.6: !180W (6/4) versus !180W (6/3) for all meteoritic samples. The solid grey line has a slope = 1 and is not a regression through the data. Within uncertainty, all samples lie on this line.
9.4.1.2 180W Isotope Anomalies in Iron Meteorites The results in Fig. 9.2 confirm the previous report (Schulz et al., 2013) of large, resolvable excess of 180W in magmatic iron meteorites. However, contrary to that study, we do not observe large excesses in the ungrouped iron Chinga, a group IVA iron, and most IIAB irons. Figure 9.7 compares selected data for "180W (6/4) from both studies. Although there is good agreement for the two individual IIAB irons, the weighted means for group IIAB differ beyond the uncertainties, and the agreement is poor for all other samples in the figure. In general, Schulz et al. (2013) report larger anomalies in "180W (6/4), but the reason for this discrepancy remains unclear at present.
The most significant difference between the results presented here and those of Schulz et al. (2013) is that we report within-group variations in both IIAB and IVB irons that are larger than the analytical uncertainties. These two groups have compositional trends consistent with formation via fractional crystallization of a metallic melt (e.g., Scott, 1972; Wasson et al., 2007; Walker et al., 2008) which indicates the parent melts were well-mixed prior to the onset of core crystallization. If the IIAB or IVB parent bodies were characterized by "180W anomalies inherited from the solar nebula, these should be homogenous throughout the metallic cores, and large variations among group members are not expected. In contrast, the IVB irons have "180W (6/4) values ranging from 1.04 ± 0.68 to 6.19 ± 0.76. Similarly, only one IIAB iron shows a well-resolved "180W (6/4) excess, and some scatter is present within this group. These within-group variations argue strongly against a nucleosynthetic origin for the "180W anomalies and suggest a process acting on the parent bodies is likely responsible. Moreover, the lack of resolvable "180W excesses in many samples in our dataset is
Part B Chapter 9 185
inconsistent with a large scale p-process heterogeneity in W isotopes in the early nebula as proposed by Schulz et al. (2013).
0 1 2 3 4 5 6
0
1
2
3
4
5
IIAB mean IIAB UnG IVA IVB
!180 W
(thi
s st
udy)
!180W (Schulz et al.)
Fig. 9.7. Values of measured !180W (6/4) for selected samples compared to the results given in Schulz et al. (2013). Samples: IVB (Weaver Mountains), UnG (Chinga), IVA (La Grange (Schulz et al.); Muonionalusta (this study)), IIAB (Forsyth County, Holland’s Store). The weighted mean for IIAB irons excludes Mount Joy (this study) due to large neutron capture effects; the IIAB mean reported in Schulz et al. excludes Ainsworth also for this reason. The solid black line has a slope = 1.
Schulz et al. (2013) also used a correlation between "180W and "182W to argue for a
temporal decrease in the magnitude of the 180W anomalies that reflected an increasing homogenization of the early nebula. However, one problem with this interpretation is that variable 182W deficits in iron meteorites not only reflect different times of accretion and metal segregation, but also variable effects of secondary neutron capture during cosmic ray exposure of the iron meteoroids (Kleine et al., 2005; Leya et al., 2003; Masarik, 1997). We compare our results for "180W (6/4) with the pre-exposure values for "182W (6/4) obtained by Kruijer et al. (2013a,b); the pre-exposure "182W (6/4) values are corrected for cosmic ray-induced shifts in W isotopic compositions and reflect the time of core segregation on the iron meteorite parent bodies, assuming these bodies formed from a common reservoir with a chondritic Hf/W ratio. Figure 9.8 shows the mean "180W (6/4) versus the pre-exposure "182W (6/4) for group IIAB, IID, IVA, and IVB irons. The pre-exposure "182W for the IVB irons is ! 20 ppm higher than of the IIAB irons, indicating that core formation on the IIAB parent body pre-dates that on the IVB parent body (Kruijer et al., 2013b), yet the largest "180W (6/4) excesses occur in the IVB irons, whereas those of the IIAB irons are significantly smaller. Furthermore, there is no correlation (r2 = 0.051) between the mean "180W (6/4) values for groups IIAB, IID, IVA, and IVB irons and their pre-exposure "182W (6/4) values. Thus, we see no evidence for an increasing homogenization of 180W with time in the early nebula as proposed by Schulz et al. (2013).
186 180W anomalies in iron meteorites: Implications for p-process heterogeneity
-3.5 -3.4 -3.3 -3.2 -3.1 -3.0
0
1
2
3
4
IVA IID IIAB IVB
!180 W
(6/4
)
!182W (6/4)
Fig. 9.8: !180W (6/4) versus the pre-exposure !182W (6/4) of four iron meteorite groups. For !180W, the weighted means and errors are shown for groups IIAB, IID, and IVB. The pre-exposure !182W values are from Kruijer et al. (2013a,b).
9.4.2 Effects from an s-Deficit The IVB irons show small deficits in "184W (6/3); the weighted mean and error for the group is -0.11 ± 0.01 ". Qin et al. (2008), Kruijer et al. (2013a), and Wittig et al. (2013) also reported similar deficits in IVB irons (i.e., -0.08 ± 0.01 ", -0.09 ± 0.01 ", -0.14 ± 0.08 ", respectively). These deficits are consistent with the presence of a small deficit in the s-process isotopes of W in the IVB irons relative to the terrestrial isotopic composition (Qin et al., 2008). The IID irons also have a similar deficit (Kruijer et al., 2013a), and the ungrouped iron Chinga also appears to contain an s-deficit (Kruijer et al., 2012). The s-process deficit of W isotopes is corroborated by deficits of s-process isotopes of Ru (Chen et al., 2010; Fischer-Gödde et al., 2012) and Mo (Dauphas et al., 2002; Burkhardt et al., 2011) in many iron meteorite groups. The s-deficit in W isotopes also causes collateral effects in the ratios besides "184W (6/3). For example, the "182W (6/3) values will appear to be less radiogenic (i.e., more negative) compared to "182W (6/4) values (e.g., Kruijer et al., 2013a). This can be an additional sign that an s-process deficit is present and results from the fact that the magnitude of the effect is unequal between the two different mass bias correction ratios. Our "182W (6/3) and "182W (6/4) results for IVB and IID irons and the ungrouped iron Chinga are entirely consistent with this expectation (see Table 9.1). The results for 180W should also be affected by the s-deficit, and in Fig. 9.9 we show the "180W (6/4) versus "184W (6/3) for the IVB, IID, and Chinga irons. Burkhardt et al. (2012a,b) showed that the s-deficit in W isotopes follows the nucleosynthesis model of Arlandini et al. (1999). The Arlandini et al. model predicts that deficits in "184W (6/3) are accompanied by excesses in "180W (6/4). This collateral effect is nearly sufficient to explain the small shifts in "180W (6/4) measured in Tawallah Valley and Weaver Mountains (IVB), the IID irons, and Chinga. The results in Fig. 9.9 demonstrate that the s-deficit can influence "180W (6/4) but cannot explain the total excesses in most IVB irons. Furthermore, the lack of an s-deficit in the IIAB irons also negates this as a possible source of
Part B Chapter 9 187
any excesses measured in that group. After correction for the s-deficit, both normalizations (6/3 and 6/4) yield the same "180W and "182W (see Table 9.2).
-0.3 -0.2 -0.1 0.0 0.1-1
0
1
2
3
4
5
6
7 UnG IID IVB
!180 W
(6/4
)
!184W (6/3)
Fig. 9.9: !180W (6/4) versus !180W (6/3) for irons that show evidence for an s-process deficit in W isotopes. The black line shows the expected effects on W isotopes based on the nucleosynthesis model of Arlandini et al. (1999).
9.4.3 Neutron Capture Effects In outer space, the interaction between meteorites and high-energy cosmic rays induce nuclear reactions that lead to a production of secondary neutrons within the outer depths of the meteorite. As these neutrons travel through the meteorite matrix, they are slowed to (epi)thermal energies and may then be recaptured by other nuclei with large neutron capture cross-sections. Such secondary neutron capture reactions may lead to measureable changes in the isotopic composition of a particular element. Theoretical models predicted that such effects may be important for W isotopes in samples with long cosmic ray exposure histories (Masarik, 1997; Leya et al., 2003), and these effects have been verified experimentally in iron meteorites (e.g., Markowski et al., 2006; Kruijer et al., 2013a; Wittig et al., 2013) as burnout of 182W, which leads to apparently less radiogenic compositions. Such cosmic ray effects are predicted to also cause burnout of 180W with a magnitude that is ~1.5 times the effect on 182W (Leya and Masarik, 2013). We can compare our measured "182W (6/4) values to the pre-exposure values determined for IIAB, IID, IVA, and IVB irons (Kruijer et al., 2013a,b) to calculate the expected effect on 180W for each sample in these groups; Table 9.2 lists the 180W (6/4) values after correction for neutron capture effects. For most irons, the effect is relatively minor (&0.55 ") and cannot account for the total "180W (6/4) variation observed in group IVB (see Fig. 9.10). However, the effect on Carbo (IID) is 1.53 " and explains why the measured value is lower than in Rodeo (IID), which shows no evidence for irradiation effects. Similarly, 180W burnout of 0.77 " in Mount Joy (IIAB) largely explains why this sample appears to contain a deficit in "180W (6/4). Schulz et al. (2013) also noted that significant burnout effects on 180W are likely present in their sample of the IIAB iron Ainsworth, which has a long
188 180W anomalies in iron meteorites: Implications for p-process heterogeneity
cosmic ray exposure age. Although we show that neutron capture reactions can lead to measureable changes in "180W, this effect fails to fully explain the within-group variability in IIAB and IVB irons.
-4.4 -4.2 -4.0 -3.8 -3.6 -3.4 -3.2 -3.0
0
1
2
3
4
5
6
7
IVB
!18
0 W (6
/4)
!182W (6/4)
Fig. 9.10: !180W (6/4) versus !182W (6/4) for IVB irons. All values have been corrected for the s-deficit. A vector shows the maximum expected burnout of !180W (6/4) caused by capture of secondary neutrons during exposure to cosmic rays. The pre-exposure !182W (6/4) for group IVB irons is -3.26 (Kruijer et al., 2013a); the pre-exposure !180W (6/4) is not known, so we have chosen the maximum observed value as the anchor point for the vector.
9.4.4 Radiogenic Effects Theoretical calculations imply that 184Os may be unstable (t1/2 >1 Ga) and could undergo #-decay to 180W (Sperlein and Wolke, 1976). Schulz et al. (2013) considered a possible in-growth of 180W from the #-decay of 184Os and concluded that this effect would be insignificant relative to the large excesses measured in "180W. More recently, these same authors argued that the decay of 184Os is in fact responsible for the majority of the observed excesses based on a correlation between "180W (6/4) and Os/W ratios (Peters et al., 2013). Although we did not determine Os and W concentrations in our samples, we can use our dataset to test this idea. Tungsten concentration data are readily available in the literature for all of our samples, but Os concentration data are scarce. However, Ir concentration data are also available for all our samples. Both Os and Ir behave as compatible elements during core crystallization, and their concentrations are highly correlated (r2 ! 0.95) in magmatic irons (McCoy et al., 2011; Ryan et al., 1990; Walker et al., 2008). Therefore, we will use Ir as a proxy for Os. Figure 9.11a shows the "180W (6/4) versus the Ir/W ratio for all irons we analysed; the "180W (6/4) values have been corrected for both an s-deficit and for burnout of 180W by neutron capture. Although there is a positive trend in Fig. 9.11a, a large amount of scatter exists; a regression through the data has an MSWD of 8.1, and samples with very similar Ir/W ratios, and presumably similar Os/W ratios, have significantly different "180W (6/4) values. The converse is also true; samples with significantly different Ir/W ratios have similar "180W (6/4) values;
Part B Chapter 9 189
for example, the IIAB irons analysed represent a large portion of the crystallization sequence (i.e., Ir concentrations change by a factor of ~100; Wasson et al., 2007), yet seven out of nine of these samples have "180W values that are indistinguishable within the uncertainties, which is inconsistent with radiogenic in-growth of 180W as the source of the variations. Fig. 9.11b shows the data only for groups IIAB and IVB; the results of linear regressions through both groups are also shown. Each group lies along a different trend. However, the Re-Os crystallization ages of group IIAB and IVB irons are indistinguishable (Smoliar et al., 1996), such that 184Os decay would not be expected to produce separate trends in "180W due to its very long half-life (i.e., ' 1 Ga).
0 2 4 6 8 10 12-3-2-101234567
(a)
IVA UnG IID IIAB IVB
!180 W
(6/4
)
Ir/W
0 2 4 6 8 10 12-3-2-101234567
(b)
IIAB IVB
!180 W
(6/4
)
Ir/W
190 180W anomalies in iron meteorites: Implications for p-process heterogeneity
Fig. 9.11 (previous page): (a) !180W (6/4) versus Ir/W ratio for all irons. The !180W (6/4) values have been corrected for both an s-deficit and for burnout from neutron capture. (b) !180W (6/4) versus Ir/W ratio for IIAB and IVB irons. Solid lines are linear regressions for each group. The Ir/W ratios were calculated from data in Campbell et al. (2002), Campbell and Humayun (2012), Walker et al. (2008), Wasson and Huber (2006), Wasson and Richardson (2001), Wasson et al. (2007), and Scott (1978). Errors on Ir/W ratios are taken as ± 11 % and are smaller than the symbol sizes. Gray horizontal lines represent the external precision for !180W (6/4) of ± 1.19 based on the analyses of SRM 129c (2 S.D.).
As an alternative approach, we can use the results from Sperlein and Wolke (1976) to
calculate a maximum possible effect on 180W from 184Os decay. Although Sperlein and Wolke did not observe any experimental evidence for the #-decay of 184Os, they calculated a possible lower limit on the half-life of 5.6 $ 1013 years. In our sample set, Iquique has the largest Os/W ratio of 16.1 (Walker et al., 2008). Assuming an initial 180W/184W ratio equal to the terrestrial value (Völkening et al., 1991) and allowing for 4567 Ma of decay, the in-growth from 184Os decay would shift the "180W (6/4) value by + 1.45. This value is too low by a factor of ! 4.3 compared to the value in Iquique after correcting for the s-deficit and burnout by neutron capture. For several other IVB irons (e.g., Tinnie, Skookum, Warburton Range), decay of 184Os would account for an even smaller proportion of the observed excesses. If the 184Os half-life were shorter, as suggested by Peters et al. (2013), it could be possible to explain the observed excess in Iquique, yet the internal inconsistencies between the predicted and observed "180W (6/4) values for various samples would remain. Thus, our dataset does not seem to support a radiogenic ingrowth by 184Os decay as the origin for the variations in 180W, but this possibility cannot be fully excluded since we did not measure Os/W ratios in our samples.
We showed in Fig. 9.4 that several samples analysed by us and by Schulz et al. (2013) give different values for "180W (6/4). The IIAB irons lack an s-deficit, and those shown in Fig. 9.4 also lack large neutron capture effects, and these factors cannot account for the discrepancy between studies. In the case of the IVB iron Weaver Mountains, the s-deficit will affect both samples equally, and the neutron capture effects are too small to explain the disagreement. Peters et al. (2013) also noted that they obtained results that do not agree with those of Schulz et al. (2013), including the IIAB iron Holland’s Store and the IVB iron Weaver Mountains. To explain the discrepancies, Peters et al. (2013) suggested sample heterogeneity in the metal phase leading to variable radiogenic in-growth of 180W. This potential explanation requires that taenite and kamacite have unequal Os/W ratios and different studies sample the irons at a scale that reflects this heterogeneity. However, neither Os nor W show a marked preference for taenite relative to kamacite (e.g., Campbell and Humayun, 1999; Campbell et al., 2003; Ash et al., 2007) and little Os-W fractionation between these phases is expected. The metal phase of IVB irons does consist of a mixture of taenite and kamacite, but this occurs as a very fine continuous intergrowth with taenite lamellae on the order of several µm in diameter (Buchwald, 1975). This microscopic heterogeneity would not influence the results for whole rock samples with masses of 0.25 to 1 g, as used in this study and by Schulz et al. (2013). In the case of the IIA irons (e.g., Forsyth County, Holland’s Store), the metal phase consists only of kamacite; no taenite is present (Buchwald, 1975). Therefore, modal variability in the metal phase of IIA and IVB irons is not a viable explanation for the sample to sample variation in "180W between the different studies.
Part B Chapter 9 191
9.4.5 Possible Spallation Effects Irradiation by cosmic rays is known to cause isotopic variations within a single meteorite. This isotopic effect occurs because the magnitudes of spallation reactions and secondary neutron capture effects depend on the depth below the meteorite surface (e.g., Herzog, 2004). For example, secondary neutron capture reactions on 182W explain why different pieces of the same iron meteorite can have distinct "182W (6/4) values (e.g., Kruijer et al., 2013b). Fig 9.12 shows "192Pt (8/5) versus "180W (6/4) for seven IVB irons; the "192Pt values were determined on aliquots from the same sample pieces used for 180W analyses and were measured in a companion study by Kruijer et al. (2013b). The "180W (6/4) values in Fig. 9.12 have been corrected for an s-deficit and burnout of 180W by neutron capture. Kruijer et al. (2013b) and Wittig et al. (2013) demonstrated that Pt isotopes in iron meteorites are affected by capture of secondary neutrons and that variations in "192Pt indicate a history of exposure to cosmic rays. The positive trend in Fig. 9.12 suggests that cosmic ray irradiation exerts some control on the 180W abundance. One possibility could be spallation reactions on isotopes of siderophile elements with masses > 180 (e.g., Re, Os, Ir, Pt). Because the abundance of 180W is so small, it may be prone to spallation effects that fail to produce detectable isotopic shifts in other neighbouring nuclides. The IVB irons are enriched in the heavy siderophile elements relative to other magmatic groups (e.g., Kelly and Larimer, 1977), which could explain why this group shows both the largest "180W excesses and significant within-group variation. The trend in Fig. 9.11a also suggests a possible link between the abundance of heavy siderophile elements and the anomalies in "180W. An alternative process could be secondary neutron capture reactions on an as yet unrecognized target nuclide that leads directly, or indirectly via subsequent radioactive decay, to a net gain of 180W atoms. These ideas are speculative, and additional theoretical models of spallation processes are needed to evaluate if reactions induced by cosmic rays provide a plausible source for the observed "180W anomalies.
0 1 2 3 4 5 6 70
10
20
30
40
50
60
IVB
!192Pt
(8/5
)
!180W (6/4)
Fig. 9.12: !192Pt (8/5) versus !180W (6/4) for seven IVB irons. The !180W (6/4) values have been corrected for an s-deficit and for neutron capture burnout of 180W. Platinum data are from Kruijer et al. (2013b) obtained on pieces of the same samples used for the 180W measurements. Errors on !192Pt are smaller than the symbol size.
192 180W anomalies in iron meteorites: Implications for p-process heterogeneity
9.5 Conclusions We confirm that large 180W excesses (up to ! 6 ") occur in magmatic iron meteorites, but significant within-group variations among IIAB and IVB irons are not consistent with the inheritance of an initial widespread p-process heterogeneity of W isotopes in the solar nebula. Furthermore, we find no evidence for an increasing homogenization of 180W with time in the early solar system. Thus, in contrast to the conclusions of Schulz et al. (2013), the solar nebula appears to have been well mixed with regard to p-process isotopes of W. The presence of an s-deficit in W isotopes (e.g., IVB irons) causes apparent positive shifts in "180W, but these are insufficient to account for most of the observed excesses. Neutron capture effects (i.e., burnout of 180W) can lead to measureable effects in 180W; however, this process fails to explain the total variations observed. In addition to the within-group variations, a comparison of our data to those of Schulz et al. (2013) suggest that sample to sample variations within a single iron meteorite are present; Peters et al. (2013) noted a similar observation. We suggest that a process acting on meteorite parent bodies is responsible for the majority of the 180W anomalies. Nuclear reactions induced by exposure to cosmic rays and/or decay of 184Os are possibilities; however, most IIAB irons have indistinguishable "180W values and do not seem to support a radiogenic origin for 180W excesses. Additional work is required to test these two ideas.
Acknowledgements We thank Tim McCoy and Linda Weizenbach (Smithsonian), Denton Ebel and Shawn Wallace (Natural History Museum, NYC), and Addi Bishoff (U. Münster) for providing samples. We also thank Ulla Heitmann for sample preparation and Mario Fischer-Gödde for laboratory assistance.
References Andreasen R. and Sharma M. (2006) Solar nebula heterogeneity in p-process samarium and
neodymium isotopes. Science 314, 806-809. Arlandini C., Käppeler F., Wisshak, Gallino R., Lugaro M., Busso M., and Straniero O.
(1999) Neutron capture in low-mass asymptotic giant branch stars: Cross sections and abundance signatures. Astrophy. J. 525, 886-900.
Arnould M. and Goriely S. (2003) The p-process of stellar nucleosynthesis: astrophysics and nuclear physics status. Phys. Reports 384, 1-84.
Ash R. D., Luong M. V., Walker R. J., McDonough W. F., and McCoy T. J. (2007) Trace element fractionation in kamacite and taenite in IVA irons. . Lunar Planet. Sci. XXXVIII, abstract #2383, Lunar and Planetary Institute, Houston (CD-ROM).
Buchwald V. F. (1975) Handbook of Iron Meteorites, Vol 1. Berkeley, CA: University of California Press, 243 p.
Buchwald (1977) Buchwald, V. F. (1977) The mineralogy of iron meteorites. Philos. Trans. R. Soc. London Ser. A 286, 453-491.
Burkhardt C., Kleine T., Bourdon B., Palme H., Zipfel J., Friedrich J. M., and Ebel D. S. (2008) Hf-W mineral isochron for Ca,Al-rich inclusions: Age of the solar system and the timing of core formation in planetesimals. Geochim. Cosmochim. Acta 72, 6177-6197.
Burkhardt C., Kleine T., Dauphas N., and Wieler R. (2012a) Nucleosynthetic tungsten isotope anomalies in acid leachates of the Murchison chondrite: Implications for hafnium-tungsten chronometry. Astrophy. J. 753:L6, 1-6.
Part B Chapter 9 193
Burkhardt C., Kleine T., Dauphas N., and Wieler R. (2012b) Origin of isotopic heterogeneity in the solar nebula by thermal processing and mixing of nebular dust. Earth Planet. Sci. Lett. 357-358, 298-307.
Burkhardt C., Kleine T., Oberli F., Pack A., Bourdon B., and Wieler R. (2011) Molybdenum isotope anomalies in meteorites: Constraints on solar nebula evolution and origin of the Earth. Earth Planet. Sci. Lett. 312, 390-400.
Campbell A. J. and Humayun M. (1999) Microanalysis of platinum group elements in iron meteorites using laser ablation ICP-MS. Lunar Planet. Sci. XXX, abstract #1974, Lunar and Planetary Institute, Houston (CD-ROM).
Campbell T. J. and Humayun M. (2012) Siderophile element abundances in the Ni-rich ataxites Gebel Kamil, Dumont, and Tinnie. Lunar Planet. Sci. XLIII, abstract #2833, Lunar and Planetary Institute, Houston (CD-ROM).
Campbell A. J., Simon S. B., Humayun M, and Grossman L (2003) Chemical evolution of metal in refractory inclusions in CV3 chondrites. Geochim. Cosmochim. Acta 67, 3119-3134.
Campbell A. J., Humayun M., and Weisberg M. K. (2002) Siderophile element constraints on the formation of metal in the metal-rich chondrites Bencubbin, Weatherford, and Gujba. Geochim. Cosmochim. Acta 66, 647-660.
Chen J. H., Papanastassiou D. A., and Wasserburg (2010) Ruthenium endemic isotope effects in chondrites and differentiated meteorites. Geochim. Cosmochim. Acta 74, 3851-3862.
Dauphas N., Marty B., and Reisberg L. (2002) Molybdenum evidence for inherited planetary scale isotope heterogeneity of the protosolar nebula. Astrophy. J. 565, 640-644.
Fischer-Gödde M., Burkhardt C., and Kleine T. (2012) Ruthenium isotope anomalies in meteorites and the cosmic Mo-Ru correlation. Lunar Planet. Sci. XLIII, abstract #2492, Lunar and Planetary Institute, Houston (CD-ROM).
Hans U., Kleine T., and Bourdon B. (2013) Rb-Sr chronology of volatile depletion in differentiated protoplanets: BABI, ADOR and ALL revisited. Earth Planet Sci. Lett. in press.
Herzog G. F. (2004) Cosmic-ray exposure ages of meteorites. In Treatise on Geochemistry, Volume 1 (ed. Davis A. M.) p. 347-380. Elsevier Pergamon: San Diego, CA.
Holst J. C., Paton C., and Bizzarro M. (2011) Are non-radiogenic W isotope anomalies in iron meteorites analytical artifacts? Workshop on Formation of the First Solids in the Solar System, abstract #9065.
Horan M. F., Smoliar M. I., Walker R. J. (1998) 182W and 187Re-187Os systematic of iron meteorites: Chronology for melting, differentiation, and crystallization in asteroids. Geochim. Cosmochim. Acta 62, 545-554.
Kelly W. R. and Larimer J. W. (1977) Chemical fractionations in meteorites—VIII. Iron meteorites and the cosmochemical history of the metal phase. Geochim. Cosmochim. Acta 41, 93-111.
Kleine T., Mezger K., Münker C., Palme H., and Bischoff A. (2004) 182Hf-182W isotope systematics of chondrites, eucrites, and Martian meteorites: Chronology of core formation and mantle differentiation in Vesta and Mars. Geochim. Cosmochim. Acta 68, 2935-2946.
Kleine T., Mezger K., Palme H., Scherer E., and Münker C. (2005) Early core formation in asteroids and late accretion of chondrite parent bodies: Evidence from 182Hf-182W in CAIs, metal-rich chondrites and iron meteorites. Geochim. Cosmochim. Acta 69, 5805-5818.
194 180W anomalies in iron meteorites: Implications for p-process heterogeneity
Kleine T., Münker C., Mezger K., and Palme H. (2002) Rapid accretion and early core formation on asteroids and the terrestrial planets from Hf-W chronometry. Nature 418, 952-955.
Kleine T., Touboul M., Bourdon B., Nimmo F., Mezger Kl, Palme H., Jacobsen S. B., Yin Q-Z., and Halliday A. N. (2009) Hf-W chronology of the accretion and early evolution of asteroids and terrestrial planets. Geochim. Cosmochim. Acta 73, 5150-5188.
Kruijer T. S., Sprung P., Kleine T., Leya I., Burkhardt C., and Wieler R. (2012) Hf-W chronometry of core formation in planetesimals inferred from weakly irradiated iron meteorites. Geochim. Cosmochim. Acta 99, 287-304.
Kruijer T. S., Fischer-Gödde M., Kleine T., Sprung P., Leya I., and Wieler R. (2013a) Neutron capture on Pt isotopes in iron meteorites and the Hf-W chronology of core formation in planetesimals. Earth Planet. Sci. Lett. 361, 162-172.
Kruijer T. S., Touboul M., Fischer-Gödde M., Bermingham K., Kleine T., and Walker R. J. (2013b) Resolution of small differences in the time of metal segregation in iron meteorite parent bodies. Lunar Planet. Sci. XLIV, abstract #1920, Lunar and Planetary Institute, Houston (CD-ROM).
Leya I. and Masarik J. (2013) Thermal neutron capture effects in radioactive and stable nuclide systems. Meteor. Planet. Sci. 48, 665-685.
Leya I., Wieler R., and Halliday A. N. (2003) The influence of cosmic-ray production on extinct nuclide systems. Geochim. Cosmochim. Acta 67, 529-541.
Markowski A., Quitté G., Halliday A. N., and Kleine T. (2006) Tungsten isotopic compositions of iron meteorites: Chronological constraints vs. cosmogenic effects. Earth Planet. Sci. Lett. 242, 1-15.
Masarik J. (1997) Contribution of neutron-capture reactions to observed tungsten isotopic ratios. Earth Planet. Sci. Lett. 152, 181-185.
McCoy T. J., Walker R. J., Goldstein J. I., Yang J., McDonough W. F., Rumble D., Chabot N. L., Ash R. D., Corrigan C. M., Michael J. R., and Kotula P. G. (2011) Group IVA irons: New constraints on the crystallization and cooling history of an asteroidal core with a complex history. Geochim. Cosmochim. Acta 75, 6821-6843.
Moynier F. Day J. M. D., Okui W., Yokoyama T., Bouvier A., Walker R. J., and Podosek F. A. (2012) Planetary-scale strontium isotopic heterogeneity and the age of volatile depletion of early solar system materials. Astrophy. J. 758:45, 1-7.
Nuclides and Isotopes – Chart of the Nuclides (2002) Sixteenth Edition, Berkeley, CA: Lockheed-Martin, 88 p.
Qin L., Carlson R. W., and Alexander C. M. O’D. (2011) Correlated nucleosynthetic isotopic variability in Cr, Sr, Ba, Sm, Nd and Hf in Murchison and QUE 97008. Geochim. Cosmochim. Acta 75, 7806-7828.
Qin L., Dauphas N., Wadhwa M., Markowski A., Gallino R., Janney P. E., and Bouman C. (2008) Tungsten nuclear anomalies in planetesimal cores. Astrophy. J. 674, 1234-1341.
Peters S. T. M., Münker C., Becker H., and Schulz T. (2013) Tungsten-180 anomalies in iron meteorites reflect alpha decay of osmium-184. Lunar Planet. Sci. XLIV, abstract #2073, Lunar and Planetary Institute, Houston (CD-ROM).
Ryan D. E., Holzbecher J., and Brooks R. R. (1990) Rhodium and osmium in iron meteorites. Chem. Geol. 85, 295-303.
Schulz T., Münker C., and Peters S. T. M. (2013) p-Process 180W anomalies in iron meteorites: Nucleosynthetic versus non-nucleosynthetic origins. Earth Planet. Sci. Lett. 362, 246-257.
Part B Chapter 9 195
Scott E. R. D. (1972) Chemical fractionation in iron meteorites and its interpretation. Geochim. Cosmochim. Acta 36, 1205-1236.
Scott E. R. D. (1978) Iron meteorites with low Ga and Ge concentrations-composition, structure and genetic relationships. Geochim. Cosmochim. Acta 42, 1243-1251.
Smoliar M. I., Walker R. J., and Morgan J. W. (1996) Re-Os ages of group IIA, IIIA, IVA, and IVB iron meteorites. Science 271, 1099-1102.
Sperlein R. F. and Wolke R. L. (1976) A search for alpha instability in 184Os. J. Inorg. Nucl. Chem. 38, 27-29.
Vockenhuber C., Oberli F., Bichler M., Ahmad I., Quitté G., Meier M., Halliday A. N., Lee D-C., Kutschera W., Steier P., Gehrke R. J., and Helmer R. G. (2004) New half-life measurement of 182Hf: Improved chronometer for the early solar system. Phys. Rev. Lett. 93:17, 1-4.
Völkening J., Köppe M., and Heumann K. G. (1991) Tungsten isotope ratio determinations by negative thermal ionization mass spectrometry. Internat. J. Mass Spec. Ion Proc. 107, 361-368.
Walker R. J., McDonough W. F., Honesto J., Chabot N. L., McCoy T. J., Ash R. D., Bellucci J. J. (2008) Modeling fractional crystallization of group IVB iron meteorites. Geochim. Cosmochim. Acta 72, 2198-2216.
Wasson J. T. and Huber H. (2006) Compositional trends among IID irons; their possible formation from the P-rich lower magma in a two-layer core. Geochim. Cosmochim. Acta 70, 6153-6167.
Wasson J. T., Huber H., and Malvin D. J. (2007) Formation of IIAB iron meteorites. Geochim. Cosmochim. Acta 71, 760-781.
Wasson J. T. and Richardson J. W. (2001) Fractionation trends among IVA iron meteorites: Contrasts with IIIAB trends. Geochim. Cosmochim. Acta 65, 951-970.
Willbold M., Elliott T., and Moorbath S. (2011) The tungsten isotopic composition of the Earth’s mantle before the terminal bombardment. Nature 477, 195-198.
Wittig N., Humayun M., Brandon A. D., Huang S., and Leya I. (2013) Coupled W-Os-Pt isotope systematics in IVB iron meteorites: In situ neutron dosimetry for W isotope chronology. Earth Planet. Sci. Lett. 361, 152-161.
Zhang J., Dauphas N., Davis A. M., and Pourmand A. (2001) A new method for MC-ICPMS measurement of titanium isotopic composition: Identification of correlated isotope anomalies in meteorites. J. Anal. At. Spectrom. 26, 2197-2205.
196 Chapter 10
Chapter 10
Synthesis and outlook The endeavour of this thesis was to explore the processes and timescales involved in the accretion and differentiation of planetesimals and planetary embryos in the solar protoplanetary disk. In particular, the main ambition was to better constrain the chronology of core formation (metal segregation) in the parent bodies of iron meteorites using Hf-W chronometry. Prior to this thesis, the effects of secondary neutron capture on W isotope compositions represented the main obstacle to obtain accurate Hf-W ages of iron meteorites. The primary objectives of this thesis were (i) to obtain W isotope compositions of iron meteorites that are unaffected by neutron capture (part A), and (ii) to quantify nucleosynthetic W isotope anomalies in Ca-Al-rich inclusions (CAI) in order to accurately constrain the initial Hf and W isotope compositions of the solar system (Part B).
To this purpose we utilized two different approaches to obtain W isotope compositions of iron meteorites that are unaffected by neutron capture. In the first approach we used combined noble gas and W isotope analyses, together with published cosmic ray exposure ages to identify samples whose 182W signatures are unaffected by cosmic rays. Through this means we identified several samples from magmatic iron meteorites with negligible cosmic ray induced W isotope shifts (Chapter 3). As these samples do not require any correction on their measured W isotope composition, they represent key samples to firmly establish the time scales of metal segregation in the parent bodies of iron meteorites. However, the best approach to quantify neutron capture-induced W isotope shifts in iron meteorites is by measuring the neutron fluence directly using an independent neutron dose monitor. In a first attempt we investigated the suitability of Cd isotopes as neutron dose monitor. However, due to the low thermal neutron capture fluence and low Cd concentrations in iron meteorites, Cd isotopes appeared unsuitable to quantify neutron capture effects on W isotopes in iron meteorites (Chapter 6).
Nevertheless, the discovery of Pt isotopes as an excellent neutron-capture dosimeter presents an adequate solution to the problem of neutron capture-induced W isotope shifts in iron meteorites (Chapter 4). Combined Pt-W isotope analyses on iron meteorites provide a well-defined Pt-W isotope correlation line for any iron meteorite group. The intercept ‘pre-exposure’ value represents the 182W/184W composition that is unaffected by neutron capture and that can be reliably interpreted in terms of Hf-W chronology. Note here that the iron meteorite samples identified in Chapter 3 that are completely unaffected by neutron capture proved invaluable to exactly nail down the pre-exposure values of each magmatic iron meteorite group to a very high precision (~3-7 ppm, 95% conf.). Collectively, the quantification of neutron capture effects in iron meteorites using combined Pt-W isotope analyses, together with the improved techniques for high-precision W isotope measurements, now allow the investigation of the accretion and metal segregation history of iron meteorites parent bodies with unprecedented temporal resolution.
Dating early solar system processes at high temporal resolution using Hf-W chronometry also requires accurate and precise knowledge of the initial Hf and W isotope composition of the solar system. Quantification of nucleosynthetic W isotope anomalies in CAI is fundamentally important to accurately constrain these important reference parameters. An important discovery from this thesis is that, in particular, fine-grained CAI exhibit large
Synthesis and outlook 197
and variable nucleosynthetic W isotope anomalies, which reflect a heterogeneous distribution of s- and/or r- process W isotopes in the primitive solar nebula (Chapter 7). The empirical slope of the !182Wi- !183W correlation defined by fine-grained CAI can be used to correct any CAI or meteorite sample for nucleosynthetic W isotope heterogeneity. In addition, after correction for nucleosynthetic W isotope anomalies, the investigated CAI define the first well-defined Hf-W isochron for bulk CAI, which provides reliable estimates of the initial Hf and W isotope compositions of the solar system. The precise solar system initial values thus obtained are of fundamental importance to obtain a precise chronology for the accretion and differentiation of meteorite parent bodies.
Fig. 10.1: Evolution of planet formation in the early solar system in dimensions of space (m) and time (years). The coloured ellipses accentuate the two main results from this thesis based on Hf-W chronometry: (i) The solar system initial Hf and W isotope compositions as defined by bulk CAI and (ii) the timescales involved in the accretion of and metal segregation in the parent bodies of iron meteorites. The grey field defines the spatial and temporal scales at which heating by 26Al decay will internally melt planetary bodies. Also shown are the conventional, hierarchical processes of planet formation (curved arrows). An alternative scenario in which large planetary embryos accreted early and rapidly (thick green arrow) may be required to overcome the radial drift and fragmentation barrier. Figure modified after Dauphas and Chaussidon (2011, An. Rev.).
One curious observation that in part motivated us to pursue the research presented in this
thesis is that many iron meteorites have measured !182W that are lower than the solar system initial W isotope composition defined by CAI. Prior to this thesis this observation led to the exciting but highly speculative idea that some iron meteorites, even after correcting for cosmic ray effects, might perhaps have lower !182W than CAI and thus derive from planetesimals that accreted earlier than the oldest preserved refractory inclusions. If true, such a conclusion would have caused a fundamental overhaul of the current theories of disk evolution and planet formation. However, the Hf-W results for iron meteorites and CAI from
10–3
100
103
106
10–6
104 105 106 107 108 1090
1 Myr 10 Myr 100 Myr
1,000 km
100 km
1 km
10,000 km
10 km
Spat
ial s
cale
Size
(m)
Time after of solar system formation (yr)
Primitive CAI (e.g., !ne-grained)
Extended processing of igneous CAI (e.g., type B)
Radial drift and fragmentation (meter size) barrier
Chondrule formation
Giant Moon forming impact
Condensation of refractory dust
Accretion and core formationon Earth completed
ic and
Oligarchihicay growth
runawaay
Chaotic growthwD
issipation of gas in protoplanetary disk
Dust
coag
ulat
ion
(1 cm)
(10 "m)
Accretion of iron meteorite parent bodies
Extended period ofmetal segregation
Rapid accretion of largeplanetary embryos?
198 Chapter 10
this thesis demonstrate that this hypothesis is incorrect. The precisely defined solar system initial !182W obtained from bulk CAI (Chapter 7) and the pre-exposure !182W of the iron meteorite parent bodies (Chapters 4-5) together unequivocally show that the iron meteorite parent bodies accreted at least several 100 kyr after the formation of the first solids in the solar system (Fig. 10.1).
A major outcome of this thesis is the discovery of small differences (~5-20 ppm) in pre-exposure !182W between the different magmatic iron meteorite groups (Chapter 5). Note that the accurate and precise identification of such small 182W heterogeneities was only possible thanks to the advances in high precision W isotope measurements by MC-ICPMS (Chapters 3-5) and the quantification of neutron capture effects on W isotope compositions (Chapters 3-4). The small 182W heterogeneities provide more detailed insight and at higher temporal resolution into the accretion and differentiation history of early-accreted planetesimals. In particular, this is the first study reporting resolved Hf-W ages for magmatic iron meteorites. Specifically, the pre-exposure !182W variations exhibit well-defined correlations with the abundances of moderately volatile elements (i.e., Ga, Ge and S). We interpret the !182W vs. S correlation as reflecting the different melting temperatures of the iron meteorite parent bodies, which thus segregated their cores at distinct times after they accreted about concurrently. Collectively, our results demonstrate that the iron meteorite parent bodies most likely accreted around only few hundred kyr after CAI formation and subsequently segregated their cores at distinct instances during a time interval of ~1-2 Myr after formation of the first solids (Fig. 10.1). The 182W heterogeneities also place constraints on the nebular environment in which the iron meteorite parent bodies accreted. For example, the higher pre-exposure !182W of the IID and IVB irons may in part reflect that their precursor planetary bodies accreted and differentiated in chemically distinct nebular reservoirs with different Hf/W. An alternative for the IID core is that it formed by two metal segregation steps at distinct times.
The main focus of this thesis was on the five major iron meteorite groups, which essentially sample only five planetesimals that accreted in the solar protoplanetary disk. Clearly, these represent only a minority of the differentiated planetesimals that once existed or that still reside in the asteroid belt. Nonetheless, the number of studied planetesimals may be expanded through future work on additional groups of magmatic iron meteorites (e.g., IC, IIC, IIG, IIIF). The ungrouped iron meteorites have hardly been studied but together may represent as many as 50 different parent bodies. Combined Pt and W isotope analyses on additional groups of iron meteorites would provide additional pre-exposure 182W/184W and eventually more detailed constraints on the time scales of planetesimal accretion and differentiation of the parent bodies of iron meteorites. Future work will also allow assessing as to whether the newly discovered correlations of 182W/184W with moderately volatile elements, and the implications thereof (chapter 5), will also hold for other iron meteorite parent bodies.
Another intriguing observation that will have to be investigated in more detail is the possible internal variability in pre-exposure !182W between different members particular iron meteorite group. Quantifying such heterogeneity in pre-exposure !182W will reveal more details about the internal structure of planetesimal cores and the processes of accretion and metal segregation in planetesimals. For example, the presence or absence of 182W heterogeneities within a parent body core could help to assess as to whether core formation in planetesimals occurred instantaneously or over an extended period of time that involved several episodes of (accretion and) metal segregation. The IIIAB irons represent the largest iron meteorite group and combined Pt and W isotope analysis on additional IIIAB irons will
Synthesis and outlook 199
thus almost certainly help to better quantify the degree of 182W heterogeneity in the IIIAB core, and thus of the processes and time scales of core formation.
The application of short-lived nuclides in early solar system chronology relies on the assumption that the parent nuclides were homogeneously distributed in the solar system. Investigating the concordance of different short-lived (Al-Mg, Hf-W) and long-lived (Pb-Pb) chronometers by obtaining formation intervals between different early solar system events – e.g., by dating angrites and CAI - not only allow to test this assumption, but also to convert relative ages of short-lived chronometers to an absolute time scale. The Hf-W results for bulk CAI (Chapter 7) indicate that the Al-Mg and Hf-W chronometers currently show concordant formation intervals between angrites and CAI, suggesting that 182Hf and 26Al were homogeneously distributed in the primitive solar nebula, at least within the temporal resolution provided by Hf-W chronometry. However, the Pb-Pb ages of CAI accounting for U isotope variability show considerable variation and do not in all cases agree with the Hf-W and Al-Mg age intervals. Obtaining precise internal Hf-W isochrons (that take into account nucleosynthetic variations), Al-Mg isochrons and Pb-Pb (that take into account U isotope variability) may help to resolve this issue.
Understanding the cause of nucleosynthetic isotope anomalies in meteorites and their components can provide essential insight into the stellar sources that contributed nuclides to the solar system, and the processes acting on matter in the solar protoplanetary disk. We demonstrated that p-process 180W heterogeneities are absent in iron meteorites (Chapter 9) and that only the IVB and IID iron meteorite groups show small nucleosynthetic W isotope anomalies that are slight heterogeneity in the abundance of s- or r-process W isotopes in the primitive solar nebula (Chapters 3-5). The observation that fine-grained CAI show large and variable W nucleosynthetic isotope anomalies, whereas coarse-grained CAI do not (Chapter 7) may be explained by different degree and duration of nebular processing between igneous (e.g., coarse-grained type B inclusions) and primitive CAI (e.g., fine grained inclusions) (Fig. 10.1). However, the exact origin and cause of the varying nucleosynthetic W isotope anomalies in CAI is not yet completely understood, as is the genetic relationship between coarse- and fine-grained CAI. A systematic, multi-element study of nucleosynthetic isotope anomalies on the exact same fine- and coarse-grained CAI may help to establish the full extent and nature (e.g., variability in s-, r- or p-process isotopes) of nucleosynthetic isotope anomalies in these objects, and also help to assess whether there is any genetic relationship between primitive CAI (e.g., fine-grained inclusions) and igneous CAI (e.g., coarse-grained type B inclusions). The results from such a study may provide fundamental insight into the processes and time scales related to material transport and mixing in the primitive solar nebula, and eventually may also tie the origin of matter in the solar system to a specific nucleosynthetic environment.
Reconstructing the planet forming processes in the first few million years of solar system history is and will be of fundamental significance to ultimately understand how planetary systems and in particular, terrestrial planets like the Earth are built. Collaborations between different scientific disciplines like astronomy, astrophysics, cosmochemistry will continue to be of great value in this context. Slowly but progressively such scientific knowledge will contribute to a better understanding of the origin of the solar system, its planetary system, the Earth, and perhaps, the origin of life. Ultimately we may then be able to guess about our own origin and eminence within the vast expanse of the universe.