UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Origin of rutile-bearing ilmenite Fe-Ti deposits in Proterozoic anorthosite massifs of the Grenville.. 2008

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

Item Metadata

Download

Media
ubc_2008_fall_morisset_caroline-emmanuelle.pdf [ 8.32MB ]
Metadata
JSON: 1.0052388.json
JSON-LD: 1.0052388+ld.json
RDF/XML (Pretty): 1.0052388.xml
RDF/JSON: 1.0052388+rdf.json
Turtle: 1.0052388+rdf-turtle.txt
N-Triples: 1.0052388+rdf-ntriples.txt
Citation
1.0052388.ris

Full Text

ORIGiN OF RUTILE-BEARING ILMENITE FE-TI DEPOSITS IN PROTEROZOIC ANORTHOSITE MASSIFS OF THE GRENVILLE PROVINCE by Caroline-Emmanuelle Morisset A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Geological Sciences) UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July2008 © Caroline-Emmanuelle Morisset, 2008 Abstract The Saint-Urbain and Big Island rutile-bearing ilmenite Fe-Ti oxide deposits are located in the composite 450 km2 Saint-Urbain anorthosite (1055-1046 Ma, U-Pb zircon) and in the Lac Allard intrusion (1057-1062 Ma, U-Pb zircon) of the 11,000 km2 Havre-Saint Pierre anorthosite suite, respectively, in the Grenville Province of eastern Canada. Slow cooling rates of 3-4°C/m.y. are estimated for both anorthosites, based on combined U-Pb zircon/rutile/apatite and40Ar/39rbiotite/plagioclase geochronology, and resulted from emplacement during the active Ottawan Orogeny. Slow cooling facilitated (1) diffusion of Zr from ilmenite and rutile, producing thin (10-100 microns) zircon rims on these minerals, and (2) formation of sapphirine via sub-so lidus reactions of the type: spinel + orthopyroxene + rutile ± corundum —* sapphirine + ilmenite. New chemical and analytical methods were developed to determine the trace element concentrations and Hf isotopic compositions of Ti-based oxides. Rutile is a magmatic phase in the deposits with minimum crystallization temperatures of 781°C to 1016°C, calculated by Zr-in rutile thermometry. Ilmenite present in rutile-free samples has higher Xhem (hematite proportion in ilmenite), higher high field strength element concentrations (Xhem = 30-17; Nb = 16.1-30.5 ppm; Ta 1.28-1.70 ppm), and crystallized at higher temperatures than ilmenite with more fractionated compositions (Xhem = 21-11; Nb = 1.36-3.11 ppm; Ta = <0.18 ppm) from rutile-bearing rocks. The oxide deposits formed by density segregation and accumulation at the bottom of magma reservoirs, in conditions closed to oxygen, from magmas enriched in Fe and Ti. The initial‘76Hf/’7fof rutile and ilmenite (Saint Urbain [SU] = 0.28219-0.28227, Big Island [BI] = 0.28218-0.28222), and the initial Pb isotopic ratios (e.g. 206Pb/4: SU = 17.134-17.164, BI = 17.012-17.036) and 87Sr/6r (SU = 0.70399-0.70532, BI = 0.70412-0.70427) of plagioclase from the deposits overlap with the initial isotopic ratios of ilmenite and plagioclase from each host anorthosite, which indicates that they have common parent magmas and sources. The parent magmas were derived from a relatively depleted mantle reservoir that appears to be the primary source of all Grenvillian anorthosite massifs and existed for --600 m.y. along the margin of Laurentia during the Proterozoic. 11 Table of Contents Abstract.ii Table of Contents iii List of Tables vii List of Figures x Acknowledgements xv Co-authorship Statement xvii Chapter I Geologic framework of Proterozoic anorthosite massifs and their associated Fe-Ti oxide deposits, and overview of the dissertation 1 1.1- Introduction 2 1.1.1- Objectives of the dissertation 2 1.1.2- History of the studied deposits 4 1.1.3- Proterozoic anorthosite-mangerite-charnockite-granite intrusive suites 4 1.1.4- Proposed parental magmas to anorthosite 7 1.1.5- Crystallization and emplacement of anorthosite massifs 8 1.1.6- The Grenville Province and the Grenville Orogeny 9 1.1.7- Fe-Ti oxide deposits 9 1.2 Outline of the dissertation and contributions to the research 11 1.3 References 15 Chapter II Crystallization ages, cooling histories, and tectonic setting of two Proterozoic anorthosite massifs, Grenville Province, Québec: Saint-Urbain and Lac Allard (Havre-Saint-Pierre) 20 2.1- Introduction 21 2.2- Overview of the Grenville Province 23 2.3- Geology of the Saint-Urbain region 24 2.4- Geology of the Havre-Saint-Pierre region 29 2.5-Method 31 2.5.1- Sampling and separation techniques 31 2.5.2- Zircon treatment 32 2.5.3- Zircon dissolution 32 2.5.4- Rutile and apatite treatment/dissolution 33 2.5.5- Isotopic ratio and U-Pb concentration analysis 34 2.5.6- 40Ar/39r biotite and plagioclase 34 2.6- Results 36 2.6.1- U-Pb zircon 36 2.6.2- U-Pb rutile 45 2.6.3- U-Pb apatite 50 2.6.4- 40Ar/39r biotite and plagioclase 53 2.7- Discussion 53 2.7.1 - Crystallization age of the Saint-Urbain anorthosite and Saint-Anne du Nord orthopyroxene granodiorite 53 2.7.2- Crystallization age of the Big Island deposit and Havre-Saint-Pierre anorthosite 58 111 2.7.3- Cooling history of the Saint-Urbain and Lac Allard anorthosites 59 2.7.4 AMCG magmatism and relationship to tectonics of the Grenville Province 63 2.8- Conclusions 65 2.9- Acknowledgments 66 2.10- References 67 Chapter III Origin of zircon rims around ilmenite in mafic plutonic rocks of Proterozoic anorthosite suites 74 3.1- Introduction 75 3.2- Locality descriptions 76 3.3- Analytical methods 82 3.4- Morphology and Composition of the Zircon in the Rims 85 3.5- Discussion 88 3.5.1 - Zircon precipitation from late hydrothermal fluids 88 3.5.2- Crystallization of the zircon rim from an evolved interstitial liquid 90 3.5.3- Zircon formation following oxidation-exsolution of baddeleyite from ilmenite 91 3.5.4- Formation of a zircon rim by diffusion of Zr from ilmenite and reaction along grain boundaries 92 3.5.5- Implications of the zircon rims for U-Pb geochronology in plutonic rocks....99 3.6- Conclusions 100 3.7- Acknowledgements 101 3.8- References 102 Chapter IV Geochemistry and Hf isotopic systematics of rutile and ilmenite from Fe-Ti oxide deposits associated with Grenvillian Proterozoic anorthosite massifs 106 4.1- Introduction 107 4.2- Locality description and sampling 108 4.2.1- Saint-Urbain anorthosite and associated Fe-Ti oxide deposits 108 4.2.2- Havre Saint-Pierre anorthosite (Lac Allard lobe) and Big Island Fe-Ti oxide deposit 112 4.3-Method 114 4.3.1- Mineral separation 114 4.3.2- XRF analyses 114 4.3.3- Acids 114 4.3.4- Rutile and ilmenite digestion 115 4.3.5- Trace element concentrations by ICP-MS 116 4.3.6- Hf column chemistry 123 4.3.7- Hf isotopic analyses by MC-ICP-MS 133 4.4-Results 135 4.4.1 Rutile chemistry 135 4.4.2- Ilmenite chemistry 141 4.4.3- Hf isotopic compositions 143 4.5- Discussion 148 4.5.1- Crystallization sequence based on the high field strength and major element variations 148 iv 4.5.2 — V and Cr concentrations in ilmenite: the influence of crystallization assemblages and f02 152 4.5.3- Tetrad effect in chondrite-normalized REE patterns of rutile and ilmenite ..153 4.5.4- Hf isotopic constraints on deposit-anorthosite relationships 156 4.5.5- Magma source constraints 157 4.6- Conclusions 161 4.7- Acknowledgments 162 4.8- References 164 Chapter V Pb-Sr isotope geochemistry of the Saint-Urbain and Lac Allard (Havre Saint-Pierre) anorthosites and associated Fe-Ti oxide deposits: Implications for the isotopic systematics of Proterozoic anorthosites 169 5.1- Introduction 170 5.2- Locality description 172 5.2.1- Saint-Urbain anorthosite and associated Fe-Ti oxide deposits 172 5.2.2- Havre-Saint-Pierre anorthosite (Lac Allard lobe) and Big Island Fe-Ti oxide deposit 174 5.3-Method 177 5.3.1- Mineral separation 177 5.3.2- XRF analyses 177 5.3.3- Microprobe analyses 178 5.3.4- ICP-MS analyses 178 5.3.5- Pb and Sr isotopes 178 5.4-Results 181 5.4.1- Major and trace elements in plagioclase 181 5.4.2- Major and trace elements in apatite 187 5.4.3- Pb isotopic compositions 191 5.4.4- Sr isotopic compositions 191 5.5- Discussion 199 5.5.1- REE contents of the parent magmas to the Fe-Ti deposits 199 5.5.2- Pb and Sr isotopic constraints on deposit-anorthosite relationships 202 5.5.3- Contamination and emplacement of the Saint-Urbain deposits 203 5.5.4- Magma source constraints for the Saint-Urbain and Lac Allard anorthosites 205 5.5.5- Isotopic systematics of Proterozoic anorthosites 212 5.6- Conclusion 219 5.7- Acknowledgments 220 5.8- References 222 Chapter VI Rutile-bearing ilmenite deposits of associated with Proterozoic anorthosite massifs of the Grenville Province (Québec) 228 6.1- Introduction 229 6.2- Locality description 231 6.2.1 - Saint-Urbain anorthosite and associated Fe-Ti oxide deposits 231 6.2.2- Havre-Saint-Pierre anorthosite (Lac Allard lobe) and Big Island Fe-Ti oxide deposit 236 6.3-Method 237 v 6.3.1- XRF analyses .237 6.3.2- Microprobe analyses 239 6.4- Results 239 6.4.1- Whole rock geochemistry 239 6.4.2- Oxide mineral chemistry 242 6.4.3- Silicate mineral chemistry 248 6.5- Discussion 256 6.5.1- Cumulate controls on the geochemistry of the Fe-Ti oxide deposits 256 6.5.2- Significance of orthopyroxene compositions 256 6.5.3- Origin of sapphirine 258 6.5.4- Origin of rutile lenses within ilmenite 261 6.5.5- Controls on ilmenite compositions from Saint-Urbain and Big Island and from other ilmenite-bearing intrusions 263 6.5.6- Experimental constraints on oxide stability, oxygen fugacity and magma compositions 266 6.5.7- Rutile saturation at Saint-Urbain and Big Island 269 6.5.8- Saint-Urbain and Big Island Fe-Ti oxide ore deposits: accumulation by density9 273 6.6- Conclusions 274 6.7- Acknowledgements 275 6.8- References 276 Chapter VII Conclusions 280 7.1- Conclusions 281 7.2- References 286 Appendices 287 Appendix a Table 2.A 1 40Ar/39r step-heating results of biotite from Fe-Ti oxide deposits in the Saint-Urbain and Lac Allard anorthosites 288 Appendix b Table 2.A2 40Ar/39r step-heating results of plagioclase from the Fe-Ti oxide deposits in the Saint-Urbain and Lac Allard anorthosites 294 Appendix c Table 4.A1 Sample location and description 296 Appendix d Table 4.A2 Rutile duplicate and replicate analyses 297 Appendix e Table 4.A3 Ilmenite duplicate and replicate analyses 298 Appendix f 4.A4 Purity of the mineral separates 300 Appendix g Table 6.A1 Sample locations and descriptions 301 Supplementary electronic file on CD-ROM Microprobe analyses (.xls file) vi List of Tables Chapter II Table 2.1 Geochronology sample locations and descriptions 28 Table 2.2 Zircon U-Pb TIMS analytical data 37 Table 2.3 Rutile and apatite U-Pb TIMS analytical data 46 Table 2.4 Ages and closure temperatures for samples of Saint-Urbain and Lac Allard 60 Chapter III Table 3.1 Rutile compositions determined by electron microprobe 77 Table 3.2 Bulk hemo-ilmenite compositions determined by XRF, Saint-Urbain, Québec 80 Table 3.3 Selected whole rock compositions determined by XRF, Saint-Urbain, Québec 81 Table 3.4 Representative ilmenite compositions determined by electron microprobe, Laramie anorthosite complex, Wyoming 83 Table 3.5 Representative zircon compositions determined by electron microprobe 84 Table 3.6 Zircon U, Pb and Th contents determined by isotope dilution mass spectrometry, Saint-Urbain, Québec 89 Chapter IV Table 4.1 Major element oxide (XRF) and trace element compositions (HR-ICP-MS) of rutile separates 117 Table 4.2 Major element compositions (XRF) of ilmenite separates 119 Table 4.3 Trace element compositions (HR-ICP-MS) of ilmenite separates 124 Table 4.4 Hf isotopic compositions of samples from the Saint-Urbain area 136 Table 4.5 Hf isotopic composition of samples from the Havre Saint-Pierre area ..137 vii Chapter V Table 5.1. Sample locations and descriptions 175 Table 5.2 Major element concentrations (XRF) of plagioclase separates 182 Table 5.3 Trace element compositions (HR-ICP-MS) of plagioclase separates 184 Table 5.4 Major element concentrations (EPMA) of apatite 189 Table 5.5 Trace element concentrations (ICP-MS) of apatite separates 190 Table 5.6 Pb isotopic compositions (MC-ICP-MS) of leached plagioclase separates 192 Table 5.7 Pb isotopic compositions (MC-ICP-MS) of whole rock samples 194 Table 5.8 Sr isotopic compositions of plagioclase and apatite separates from Saint-Urbain and Big Island deposits 195 Table 5.9 Sr isotopic compositions of whole rock samples 197 Table 5.10 La and Lu content in the liquid inverted using different partition coefficients and mineral phases 200 Table 5.11 Different source models for Proterozoic anorthosite massifs based on Hf, Sr, Pb and Nd isotopic geochemistry 209 Chapter VI Table 6.1 Whole rock analyses by XRF 240 Table 6.2 Representative major element compositions (XRF) of ilmenite separates 243 Table 6.3 Representative spinel compositions determined by electron microprobe 246 Table 6.4 Representative rutile compositions determined by electron microprobe 247 Table 6.5 Corundum compositions determined by electron microprobe 249 Table 6.6 Representative plagioclase compositions determined by electron microprobe 250 Table 6.7 Representative orthopyroxene compositions determined by electron microprobe 253 viii Table 6.8 Representative sapphirine compositions determined by electron microprobe 254 Table 6.9 Representative biotite compositions determined by electron microprobe 255 Table 6.10 Misfit of the closest mass balance for the reactions forming sapphirine 262 Table 6.11 Calculated relative f02 from ilmenite compositions using QUILF 267 ix List of Figures Chapter I Figure 1.1 Simplified geological map of the Grenville Province adapted from Davidson (1998) 3 Figure 1.2 Compilation of U-Pb zirconlbaddeleyite crystallization ages for Proterozoic AMCG magmatism worldwide (updated from Scoates & Chamberlain 1995) 5 Chapter II Figure 2.1 Simplified geological map of the Grenville Province adapted from Davidson (1998) and Corriveau et al. (2007) 22 Figure 2.2 Simplified geological map of the Lac Saint-Jean and Saint-Urbain anorthosite areas (modified from Hébert et al. 2005) 25 Figure 2.3 Photographs of field relationships for geochronological samples (a-e: Saint-Urbain; f: Big Island) 26 Figure 2.4 Simplified map of the Havre-Saint-Pierre anorthositic suite (after Gobeil et al. 2003) 30 Figure 2.5 Corcordia diagrams for U-Pb data from analyzed zircon from Saint-Urbain (a-f) and Big Island (g-j) 39 Figure 2.6 Concordia diagram for U-Pb data from analyzed rutile fractions from the Saint-Urbain deposits (a-d) and Big Island dyke (e-f) 48 Figure 2.7 Textures of minerals (e.g. rutile, apatite, plagioclase and biotite) Analyzed for U-Pb and Ar-Ar geochronology from the Saint-Urbain and Big Island deposits 49 Figure 2.8 Concordia diagram for U-Pb data from analyzed apatite fractions from the Saint-Urbain deposits 51 Figure 2.9 40Ar/39r incremental-heating age spectra for biotite from the Saint-Urbain and Big Island Fe-Ti oxide deposits and their respective host anorthosites 54 Figure 2.10 40Ar/39r incremental-heating age spectra for plagioclase from the Saint-Urbain and Big Island Fe-Ti oxide deposits and their respective host anorthosites 55 Figure 2.11 Summary diagram of U-Pb zircon ages determined in this study 56 x Figure 2.12 Cooling histories of the Saint-Urbain (a) and the Lac Allard lobe of the Havre-Saint-Pierre anorthosite suite (b) 61 Figure 2.13 Compilation of crystallization ages for Proterozoic AMCG magmatism worldwide (updated from Scoates & Chamberlain 1995) 64 Chapter III Figure 3.1 Simplified map of the Grenville Province modified from Davidson (1998) showing Proterozoic anorthosite massifs and associated mangerites/granites and the localities of samples under study (star symbols) 78 Figure 3.2 Back-scattered electron images showing the textural associations of zircon rims from Mirepoix, Saint-Urbain and Laramie 86 Figure 3.3 Back-scattered electron images showing the textural associations of zircon rims from Big Island and Methuen 87 Figure 3.4 Relationships between whole rock Zr (ppm) and Ti02 (wt%) and hemo-ilmenite Zr (ppm) 94 Figure 3.5. Zircon radius vs. ilmenite radius calculated from Fraser et al. (1997) (a) and volume of ilmenite/volume of zircon vs. Zr (ppm) in ilmenite diagram (b) 95 Chapter IV Figure 4.1 Simplified geological map of the Grenville Province adapted from Davidson (1998) 109 Figure 4.2 Simplified geological maps of the Saint-Urbain anorthosite area and related Fe-Ti oxide ore deposits 110 Figure 4.3 Structural and textural characteristics of the Saint-Urbain and the Big Island Fe-Ti oxide ores 111 Figure 4.4 Simplified geological map of the Lac Allard lobe, part of the Havre Saint-Pierre anorthosite suite (after Gobeil et a!. 2003) 113 Figure 4.5 Primitive mantle-normalized trace element diagram showing analyses of the synthetic titanite 1500 and 150 by different analytical methods. ..122 Figure 4.6 La concentrations in ilmenite measured by HR-ICP-MS on different acid solutions 127 Figure 4.7 Schematic representation of the developed methodology for the separation of Hf from high Ti-bearing minerals (>40 wt% Ti02) 129 xi Figure 4.8 Elution diagrams of Hf and Ti for the second column of the Hf separation procedure 130 Figure 4.9 Trace element binary diagrams of rutile separates 138 Figure 4.10 CI chondrite-normalized REE diagrams of rutile and ilmenite from the Saint-Urbain and Big Island deposits 139 Figure 4.11 Primitive mantle-normalized trace element diagram of rutile from Saint-Urbain and Big Island deposits (primitive mantle-normalizing values from McDonough & Sun, 1995) 140 Figure 4.12 Major and trace element binary diagrams of ilmenite separates 142 Figure 4.13 Primitive mantle-normalized trace element diagram for ilmenite separates (primitive mantle-normalizing values from MeDonough & Sun, 1995) 144 Figure 4.14 Trace element binary diagrams of ilmenite separates 145 Figure 4.15 Initial Hf isotopic ratios for Saint-Urbain (a) and Big Island (b) ilmenite and rutile separates 147 Figure 4.16 Temperature calculated from the Zr (ppm) content of rutile at Saint-Urbain and Big Island 149 Figure 4.17 Diagram of Ta vs. Nb showing possible scenarios to explain the Nb-Ta correlation in ilmenite 150 Figure 4.18 Diagrams showing the variation of Y/Ho with Zr/Hf (a) and Dy (ppm) with Y/Ho (b) for rutile and ilmenite from Saint-Urbain and Big Island 155 Figure 4.19 Initial‘76Hf/177fversus Nb (ppm) for rutile from Saint-Urbain and Big Island 158 Figure 4.20 eHf versus time (Ga) showing the depleted mantle Hf model ages for the highest EHf from rutile and ilmenite in the Saint-Urbain and Big Island deposits 159 Chapter V Figure 5.1 Simplified geological map of the Grenville Province adapted from Davidson (1998) and Corriveau et al. (2007) 171 Figure 5.2 Simplified geological maps of the Saint-Urbain anorthosite area and related Fe-Ti deposits 173 xii Figure 5.3 Simplified geological map of the Lac Allard lobe, part of the Havre-Saint-Pierre anorthosite suite (after Gobeil et al. 2003) 176 Figure 5.4 CI chondrite-normalized REE diagram of the plagioclase from the Saint-Urbain and Big Island deposits and their respective host anorthosites and apatite separates from the Saint-Urbain deposits 188 Figure 5.5 Pb isotopic compositions of leached plagioclase separates from the Saint-Urbain (light gray fields) and Big Island (dark gray field) deposits and respective anorthosite massifs (Saint-Urbain and Lac Allard) as well as country rocks 193 Figure 5.6 Sr isotopic compositions of plagioclase and apatite from the Saint-Urbain and the Big Island deposits compared to the composition of plagioclase from their respective anorthosite host rocks 198 Figure 5.7 CI chondrite-normalized REE diagrams showing modeled melt compositions inverted from plagioclase and apatite (see discussion for the choice of partition coefficients) for (a) Saint-Urbain and from plagioclase for (b) Big Island 201 Figure 5.8 Initial EHf vs. initial 87Sr/6r showing gHf from ilmenite and rutile (Chapter 4) and the initial Sr isotopic compositions of plagioclase and apatite from the same rocks 204 Figure 5.9 Sr-Nd-Hf isotopic geochemistry of plagioclase separates and whole rocks from Proterozoic anorthosite massifs 207 Figure 5.10 Pb isotopic geochemistry of Proterozoic anorthosite massifs 208 Figure 5.11 206Pb/41versus crystallization age for Proterozoic anorthosite massifs 214 Figure 5.12 Available ENd(t) versus crystallization age for Proterozoic anorthosite massifs 217 Figure 5.13 Available 87Sr/6r() versus crystallization age for Proterozoic anorthosite massifs 218 Chapter VI Figure 6.1 Simplified geological maps of (a) the Grenville Province adapted from Davidson (1998), (b) the Saint-Urbain anorthosite area after Rondot (1989), and (c) the Lac Allard lobe, part of the Havre-Saint-Pierre anorthosite suite (after Gobeil et al. 2003) 230 Figure 6.2 Geologic relations exposed in the pit walls and adjacent outcrops around the Saint-Urbain deposits 233 xiii Figure 6.3 Photographs of field relationships observed in the Saint-Urbain and Big Island Fe-Ti oxide ore deposits 234 Figure 6.4 Photomicrographs and hand-sample photographs of samples showing the different rock types and textures found in the Saint-Urbain and Big Island deposits 235 Figure 6.5 Geology of the Big Island massive Fe-Ti oxide dyke 238 Figure 6.6 Triangular RO-T02-R03diagram showing the magnetite-ulvospinel, hematite-ilmenite and pseudobrookite-ferropseudobrookite solid solutions 244 Figure 6.7 MgO (wt%) versus Xhem for ilmenite from the Saint-Urbain (a) and Big Island (b) deposits 245 Figure 6.8 Histogram showing the An content of plagioclase and XMg of orthopyroxene for the Saint-Urbain and Big Island deposits 252 Figure 6.9 Major element chemistry of whole rock samples from the Saint-Urbain and Big Island areas 257 Figure 6.10 Photomicrographs showing textural relationships between spinel, orthopyroxene and sapphirine 260 Figure 6.11 Comparison of ilmenite XFe vs. MgO (wt%) for available whole grain compositions in magmatic systems 265 Figure 6.12 Diagrams of iSiog f02 vs. XFe and XTi showing the experimental results for oxide stability fields at 1100°C (a) and 1000°C (b) from Lattard et al. (2005) with the natural ilmenite and rutile compositions from Saint-Urbain and Big Island superimposed 270 xiv Acknowledgements First, I would like to thank James Scoates, my supervisor, for his never-ending enthusiasm and interest in the project that has permitted me to complete my research. During the past five years at UBC, his constant and unfailing support has been essential to the finalization of my doctoral dissertation. Not only has he edited my writing in a language that is not mine, but he has constantly helped me to clarify and formulate the ideas presented in this thesis. My gratitude is also immediately extended to Dominique Weis, my co-supervisor, for her indispensable insight and input into my thesis, particularly regarding the analytical chemistry and isotope work that was carried out at the Pacific Centre for Isotopic and Geochemical Research. I also want to highlight her <<fast eyes>> at seeing correlations between isotopic data as well as catching and deciphering any calculation mistakes that I made. I would also like to add that her trips to Belgium were essential as she always brought back my favourite chocolate, without which the long days of work would have been much more difficult. The caring help of Jacqueline Vander Auwera, especially at the beginning of my Ph.D. which I started in Belgium in 2002, is greatly appreciated. Without her support, I would never have started studying these <<rocks>>, which led to my obtaining my first grants. In my first year at the Université of Liege, Jean-Clair Duchesne taught me mineral separation techniques and, along with Guy Bologne, XRF and ICP-MS analyses; I would like to thoroughly thank both of them. Furthermore, the very keen interest with which Kelly Russell and Dick Tosdal, who were on my supervisory committee at UBC, received my work and followed its progression during my five years in the department has been a constant encouragement — much more than they realize. * It is a pleasure to mention here everyone that helped me through the long process of analytical work at the PCIGR. Bruno Kieffer, for his training in the clean lab as well as for analyzing the Sr isotopic ratios on the Triton. Rich Friedman, for all his U-Pb analyses on zircon, rutile and apatite; he carried out all the chemistry and the analyses on the TIMS himself with the help of Rachel Lishansky and Hai Lin, and I warmly thank them all. Bert Mueller, for the time spent with me discussing how to analyze the REE in ilmenite and rutile and also for the all mornings invested in tuning the Element2 ICP-MS. Jane Barling, for her training on the Nu Plasma MC-ICP-MS. Vivian Lay, for many good discussions on analytical chemistry and her help running some samples when I was desperately running out of time. Thomas Ulirich, for his Ar-Ar analyses of biotite and plagioclase. Wilma Pretorious, who provided advice on how to prepare samples and standards to run on the Element. And last but not least, to all students that are working in the labs — for we are all such good friends and what else but shared surroundings and friendship could help us keep going through those long days of column chemistry! * xv A special word of gratitude needs to be paid to André Rahier. Dominique met him after six months of unsuccessful attempts trying to adapt the Hf chemistry for ilmenite. The various e-mail exchanges with him lead to the breakthrough that permitted us to understand the chemical reactions involved in each step of the long Hf chemistry process and finally lead to a successful separation of Hf from high-Ti minerals, which is one of the important contributions of this work. Thanks as well to Heinz-Juergen Bernhardt for his help with the microprobe analyses at Bochum in Germany and also to Mati Raudsepp for his training on the SEM and while analyzing samples on the microprobe at EOS. Terry Gordon’s assistance in balancing chemical equations is greatly appreciated, not to mention his positive attitude towards my work. I would like him to know that it really uplifted my spirits during the last months of work. * I gratefully acknowledge the support of Rio Tinto Fer et Titane Inc. for the financial and logistical aid provided throughout this project. In particular, the ceaseless efforts of Martin Sauvé, who was essential in the organization of my field seasons, including all trips to Big Island to which there is no road access. He always kept a lively interest in the development of my research. I shall always be very grateful to Kerry Stanaway, now retired, who believed in the sound importance of undertaking detailed research to understand geological problems. Finally, I would like to mention Jacques Dumouchel, who appreciated my work and decided to continue supporting it, and Yves Bourque who offered me a summer job at the Lac Tio mine (Havre-Saint-Pierre) in 1998, from which the initial interest I took in understanding the formation of Fe-Ti deposits developed. Bernard Charlier, of the Université de Liege, who also studies Fe-Ti oxide deposits, spent five weeks in the field with me in the summer of 2002, and I am very appreciative of his help with mapping and sampling. * All my friends at EOS, that are still here or that are gone, supported and helped me, making these years in Vancouver so enjoyable. I must underline my faithful accomplices, my officemate throughout the past five years, Elspeth Barnes, and my almost-neighbour, Inês Garcia Nobre Silva. Without you two in Vancouver, it would simply not have been the same. Finalement, from deep in my heart, I want to tell Marie-Christine, ma seur... my sister, that I could not have succeeded without her help, advice and constant solidarité! And to my father and Errol, my brother in law, for their tireless support. I would never have been able to conclude this work without all those hours of conversation, with the only purpose of encouraging me. Je sais que vous êtes là et ça fait toute la difference! Merci a tous! xvi Co-authorship Statement The five manuscripts in this dissertation are all co-authored by my supervisor James Scoates. Chapters II, IV, V and VI are also co-authored by my co-supervisor Dominique Weis. They participated in each step of the research from the development of each project to the editing of the manuscripts and provided financial support during the course of this study. Chapter II Crystallization ages and cooling histories of the Saint-Urbain and Lac Allard (Havre Saint-Pierre) anorthosite massifs, Grenville Province, Québec Authors: Caroline-E. Morisset, James S. Scoates, Dominique Weis and Richard Friedman Richard Friedman provided all U-Pb geochronological results (9 zircon, 6 rutile and 4 apatite), including chemistry, analysis, data reduction and editing. Chapter III Origin of zircon rims around ilmenite in mafic plutonic rocks of Proterozoic anorthosite suites Authors: Caroline-E. Morisset and James S. Scoates Chapter IV Geochemistry and Hf isotopic systematics of rutile and ilmenite from Fe-Ti oxide deposits associated with Grenvillian Proterozoic anorthosite massifs Authors: Caroline-E. Morisset, James S. Scoates, Dominique Weis and André Rahier André Rahier helped develop the Hf separation column chemistry for high-Ti minerals. Chapter V Pb-Sr isotope geochemistry of the Saint-Urbain and Lac Allard (Havre-Saint-Pierre) anorthosites and associated Fe-Ti oxide deposits: Implications for the isotopic systematics of Proterozoic anorthosites Authors: Caroline-E. Morisset, James S. Scoates and Dominique Weis Chapter VI Origin of rutile-bearing Fe-Ti oxide ore deposits of Proterozoic anorthosite massifs of the Grenville Province (Québec) Authors: Caroline-E. Morisset, James S. Scoates, Dominique Weis, Jacqueline Vander Auwera and Martin Sauvé Jacqueline Vander Auwera contributed to field and analytical aspects of the study and will provide revisions of the manuscript before publication. Martin Sauvé, from Rio Tinto Fer et Titane mc, the company that provided significant financial support for the project, provided logistical support for field work and will revise the manuscript before publication. xvii Chapter I Geologic framework of Proterozoic anorthosite massifs and their associated Fe-Ti oxide deposits, and overview of the dissertation 1 1.1- Introduction 1.1.1- Objectives of the dissertation In this dissertation, a comprehensive geological study of two rutile-bearing ilmenite ore deposits associated with Proterozoic anorthosite massifs is presented. The two deposits are found within the Grenville Province of Québec, Canada (Figure 1.1): (1) the Saint-Urbain deposits (eight discrete bodies) occur within the Samt-Urbain anorthosite, and (2) the Big Island dyke crosscuts the Lac Allard lobe of the Havre-Saint-Pierre anorthositic suite. Proterozoic anorthosite massifs of Québec and Norway host the largest economically significant magmatic iron-titanium (Fe-Ti) oxide deposits (i.e. Lac Allard and Tellnes, respectively). These deposits contain hemo-ilmenite (ilmenite with exsolution lamell of hematite) as their sole oxide and can contain up to 35 wt% titanium-oxide (Ti02)and 65 wt% iron-oxide(0l)(FeOT); TiO2 is used primarily as white pigment in paints and plastics. Very few rutile-bearing ilmenite deposits have been documented, however, the presence of rutile with ilmenite can significantly increase the Ti02 content of the ores (up to 53 wt% based on the results of this study). To understand the magmatic processes that enrich a basaltic (or mafic) magma in Fe and Ti and that result in the saturation and concentration of ilmenite + rutile, and whether these processes are related to crystallization of the associated anorthosite massifs, an integrated petrologic, geochronologic, geochemical and isotopic study was undertaken. The principal objectives of this research are to investigate: (1) the timing of the emplacement of the Saint-Urbain and Lac Allard intrusions relative to the Grenville orogeny; (2) the cooling history of these two intrusions to evaluate the post crystallization evolution of the massifs and their mineralization; (3) the genetic link between the anorthosites and the Fe-Ti oxide deposits; (4) the source of the magma parental to the anorthosites and Fe-Ti oxide ores; (5) the character of the mineral phases present in the deposits (i.e. magmatic, metamorphic or sub-solidus); and (6) the conditions that allow ilmenite + rutile saturation in mafic magmas. 2 Figure 1.1. Simplified geological map ofthe Grenville Province adapted from Davidson (1998). Inset map in the lower right part shows the relative location of the map area in North America. Anorthosite massifs and related mangerite and granitic rocks (AMCG suites) are identified as well as associated Fe-Ti±P mineral deposits as identified in Corriveau et al. (2007): (a) Irvy and Desgrosbois; (b) Saint-Hypolyte; (c) Saint-Urbain; (d) Mine Canada Iron; (e) Saint-Charles; (f) La Hache-Est; (g) Buttercup; (h) Lac Brulé; (i) Lac Dissimieu; (j) Lac La Blache; (k) Rivière Pentecôte; (1) Canton Amaud; (m) Lac Raudot; (n) Magpie; (o) Big Island; (p) Tio Mine; (q) Everett. 3 1.1.2- History of the studied deposits Jacques Cartier recognized the presence of iron ore in the anorthosite of Saint-Urbain in the 1 500s (Rose 1969), and the first geological descriptions of the rutile-ilmenite Fe-Ti oxide deposits were made by Logan (1850), Hunt (1853), and Warren (1912). Exploitation of the Saint-Urbain deposits has been intermittent. The Furnace deposit was mined for two years in 1872. During the First World War, the General Electric Company mined ore at Saint-Urbam from the deposits that now carry this name. In 1928, the Dupont Chemical Company extracted ore for four years, stopped, and than restarted during the Second World War, operating from 1940 to 1946. Over the past 60 years, many different companies have owned the mining rights of the area and periodically exploited the Fe-Ti resources (e.g. America Titanic Iron Company, Continental Iron & Titanium Mining Limited) (Rose 1969), but production of massive ilmenite ceased in the 1970s (Corriveau et al. 2007). During the 2002 and 2004 field seasons of the study, the claims to the area, which are now owned by Gestion Ora-Mirage Ltée, were in the possession of Bertrand Brassard and Ressources d’Arianne Inc. The Big Island deposits were first described by Retty (1942) during summer mapping for the “Ministère des Mines du Québec”. Except for one study carried out by Cloutier (1982) and a section in the work of Bergeron (1986) on the Big Island mineralization, most of the geological work in the area has been concentrated on the Lac Allard (Lac Tio) mine, the world’s largest magmatic ihnenite deposit, which is located only 20 km northeast of Big Island and has been in operation since 1950. The mining rights for the Big Island area currently are owned by Rio Tinto Fer et Titane Inc. 1.1.3- Proterozoic anorthosite-mangerite-charnockite-granite intrusive suites Proterozoic rocks of the Grenville Province of Canada contain the world’s largest concentration of anorthosite massifs, which are commonly referred to as anorthosite mangerite-charnockite-granite (AMCG) suites. These intrusive suites are a characteristic feature of the Middle Proterozoic (crystallization ages range from 2.1-0.9 Ga) and are located in the North American, European, Asian (China and India) and African continents (Figure 1.2). With the exception of a few older massifs, AMCG magmatism 4 700 900 1100 1300 1500 1700 1900 2100 2300 I I I I I I I I I .I-J(1) Arnanunat . : a) a) Lofoten (Norway) cay Korosten (Ukraine) 0 <> Horse Creek (USA) * Lanying + Damiao (China) . Manicouagan Imbricate Zone cr3 Wigborg (Finland/Russia) i. Mealy Mtns QJ Sairni (Russia) )I< Bengal (India) Mazury (Poland) , Wolf River (USA) )1( Kunene (Namibia/Angola) 0 Jonhoping (Sweden) X Harp Lake+Michikamau O Laramie RMère Pentecôte Nain t De Ia Blache A.ih’V Lac Saint-Jean + Mattawa Adirondacks Morin Oaxacan (Mexico) tZ Atikonak Grenville Li i Havre-Saint-Pierre X NainSaint-Urbain Montpelier(USA) ‘G Horse Creek/Laramie/ Labrieville • Wolf River)4( )4( ChflkaLaka (India) 0 Scandinavia + Europe Q Rogaland (Norway) )K Asia + Africa + Mexico ê( Uluguru (Tanzania) I 700 900 1100 1300 1500 1700 1900 2100 2300 Age (Ma) Figure 1.2 Compilation of U-Pb zircon!baddeleyite crystallization ages for Proterozoic AMCG magmatism worldwide (updated from Scoates & Chamberlain 1995). The Grenville massifs are identified as triangles. The AMCG suites are found mostly in eastern Canada, except when noted. The grey bands indicate timing of orogenic events; Labradorian and Pinawarian orogenies from Gower & Krogh (2002), Shawinigan, Ottawan and Rigolet orogenies from Rivers (1997) and Rivers et al. (2002). For a complete list of the references for the ages, see Figure 2.14 in Chapter 2. 5 located in the Grenville Province was contemporaneous with the different episodes of metamorphism associated with the Grenville orogeny (i.e. from Ca. 1190 Ma to Ca. 1000 Ma) (Figure 1.2). Anorthosite massifs themselves are composed of different rocks types: pure anorthosite (>90% plagioclase), leucotroctolite, leuconorite, leucogabbronorite, leuconorite, and more mafic lithologies such as troctolite, norite and gabbronorite (Emslie et al. 1994). Other members of the AMCG suites are not always all present in every suite, and include mangerite (orthopyroxene monzonite), charnockite (orthopyroxene granite), biotite-hornblende granite and rapakivi granite. Finally, large layered intrusions can also be associated with AMCG suites (e.g. Kiglapait in Nain, Labrador, Morse 1978; Bjerkreim-Sokndal in Rogaland, Norway, Wilson et al. 1996). Different views on the source of the magmas that produced AMCG magmatism have been developed. For examples, Emslie et al. (1994) suggest that anorthosite crystallized from a mantle-derived magma, whereas mangerite-charnockite-granite magmas originate from melting of the lower crust with a variable mantle contribution. Based on petrologic experiments, Vander Auwera et al. (1998) proposed that polybaric fractional crystallization of crustally-derived jøtunitic magma (hypersthene monzodiorite) (Longhi et al. 1999) could produce mangerite, quartz mangerite and chamockite compositions and also produce anorthosite by crystal accumulation along the differentiation trend. Scoates & Chamberlain (2003) proposed that mantle-derived high-Al gabbros, with some amount of crustal contamination, could be parental magmas to anorthosite and the residual magmas from crystallization of anorthosite (ferrodiorite) would differentiate to monzonitic compositions; chamockite (if present) would be mostly derived from the melting of the crust. Various proposals thus exist for the formation of AMCG suites and the most currently debated topic is whether the source of the magmas parental to Proterozoic anorthosites primarily have a mantle (Emslie 1978; Morse 1982; Wiebe 1992; Emslie et al. 1994; Ashwal 1993) or a lower crustal (e.g. Taylor et al. 1984; Longhi et al. 1999) origin. 6 1.1.4- Proposed parental magmas to anorthosite The parental magmas to Proterozoic anorthosites must account for all types of rocks and compositions that are found within the massifs (e.g. plagioclase [An7030], olivine, orthopyroxene, clinopyroxene, Fe-Ti oxides). Two compositions are proposed to be the parental magma: high-Al basalt and jøtunite. High-Al basalts found as dikes or as chilled margins of mafic intrusions in anorthosite suites are considered to have a mantle origin (Emslie 1980; Nolan & Morse 1986; Emslie et al. 1994; Mitchell et al. 1995; Scoates & Mitchell 2000). Longhi et al. (1999) proposed that there is a continuum in compositions from high-Al gabbros to primitive jøtunite (hypersthene monzodiorite found as chilled margins in the Rogaland Anorthositic Province, Norway) and that these magmas have a crustal origin. An experimental study (Vander Auwera & Longhi, 1994) ofjotunite reproduced the compositions observed in the Bjerkreim-Sokndal layered intrusion from Rogaland. The major objection to a mantle-derived origin for these magmas is the presence of a thermal divide in the system olivine-plagioclase-wollastonite-ilmenite orthoclase-quartz. The compositions used in the experiments of Longhi et al. (1999) from Harp Lake and Rogaland (high-Al basalt and primitive jøtunite, respectively) lie directly on the thermal divide at 1275°C (10-13 kbar, pressure of the first crystallization stage of anorthosite as explained below). Fractional crystallization of a basaltic magma cannot lead to a composition that lies on the thermal divide, thus it has been argued that melting of the lower crust would generate melts of these compositions (Longhi et al. 1999; Duchesne et al. 1999; Longhi 2005). However, such melts would favor crystallization of orthopyroxene rather than olivine and therefore cannot produce the olivine-bearing anorthosites and related troctolites that are observed in many AMCG suites, especially in the largest massifs (Scoates & Mitchell, 2000; Scoates, 2003). To further evaluate the parental magmas to anorthosites, Longhi (2005) calculated the fractionation and assimilation fractionation paths for different mafic magma compositions (e.g. Hawaiian tholeiitic basalt, continental flood basalt, komatiite). These calculations demonstrate that, in general, when plagioclase saturation is reached, the XMg (Mg/(Mg+Fe)) of the residual magma is too low to account for the compositions of the ferromagnesian mineral in the anorthosite. He thus proposed that melting of the most 7 mafic part of a deep-crustal layered intrusion, perhaps delaminated, could produce the appropriate melt compositions to be parental to anorthosite massifs. Radiogenic isotopic studies (Rb-Sr, Sm-Nd, Pb-Pb) are an extremely powerful tool for assessing the source of mafic magmatism, but have been considered to be inconclusive in identifying the source of the magmas for individual massifs. This is because young (<200 m.yr.) crust can have a similar isotopic signature to the mantle from which it is extracted from due to the long half-lives of each isotopic system. New Hf, Pb, and Sr isotopic results in this dissertation, combined with available Pb, Sr and Nd isotopic compositions for anorthosite massifs, bring new insights to this issue and help to resolve the source of parent magmas to Proterozoic anorthosites. 1.1.5- Crystallization and emplacement of anorthosite massifs Despite the arguments on the source of the parent magmas, there is a general consensus that anorthosite massifs are produced by polybaric crystallization (e.g. Emslie 1978, 1985; Morse 1982; Longhi & Ashwal 1985). Crystallization starts in a deep-seated magma chamber located in the lower crust or at the mantle-crustal boundary where plagioclase crystallizes and separates by flotation from the magma (e.g. Fram & Longhi, 1992). This crystal mush, which includes suspended plagioclase, orthopyroxene megacrysts and fractionated interstitial liquid, becomes unstable and begins to ascend to higher crustal levels (10 to 25 km depth, Emslie 1985) as a diapir (Longhi & Ashwal 1985; Bamichon et a!. 1999) or along conduits through the crust (Royse & Park 2000; Scoates & Chamberlain 2003). During transport, the crystallized portions of the massifs recrystallize and are deformed (e.g. flattened and elongated orthopyroxene, recrystallization to granular plagioclase with 120° triple junctions — Lafrance et al. 1996). A metamorphic overprint in some massifs of the Grenville Province (e.g. 1.16-1.15 Ga Lac-Saint-Jean — Higgins & van Breemen 1996 and Hébert 2001; 1.16-1.15 Ga Adirondack — Hamilton et a!. 2004) is characterized by the formation of metamorphic garnet. Crystallization and cooling ages of AMCG suites are essential for relating the timing of emplacement of the massifs to Grenvillian metamorphic events as described below. 8 1.1.6- The Grenville Province and the Grenville Orogeny The Grenville Province formed on the margin of Laurentia. With the exception of Archean and Paleoproterozoic sedimentary sequences on the margin of Laurentia (Rivers 1997, Davidson 1998) that are mostly found near the Front of the Grenville Province, most of the Grenville Province crust found in southern Québec is composed of accreted magmatic arcs. The 1.96-1.84 Ga Labradoria arc (Nd model ages calculated from Dickin 2000) was incorporated into Laurentia during short-lived, south-dipping subduction leading to the 1.71-1.66 Ga Labradorian orogeny (Gower & Krogh 2002). The 1.75-1.65 Ga Quebecia arc (Nd model ages calculated from Dickin 2000) was probably accreted to Laurentia within 100 m.yr. of its formation (Dickin & Higgins 1992). A continental margin magmatic arc, which alternated between extensional and compression systems, prevailed during the period from 1.50 to 1.23 Ga (Rivers & Corrigan 2000). The Grenville orogeny was produced during Himalayan-type continent-continent collision from 1.19-0.98 Ga and occurred in three phases: (1) the Shawinigan orogeny from 1.19- 1.14 Ga, (2) the Ottawa orogeny from 1.08-1.02 Ga; and (3) the Rigolet orogeny from 1.01-0.99 Ga (Rivers 1997; River eta!. 2002). Gower & Krogh (2002) consider that continent-continent collision occurred only in the interval from 1.08-1.02 Ga. Recent paleogeographic reconstructions suggest that continent-continent collision occurred at ca. 1.10 Ga at the level of present-day Texas and that the active subduction zone beneath northern Laurentia resulted in Amazonia colliding with Laurentia only around ca. 1 Ga (Li et a!. 2008). 1.1.7- Fe-Ti oxide deposits More than 20% of the surface of the Grenville Province is covered by AMCG suites and within these suites, Fe-Ti oxides deposits are commonly associated with the anorthosites. Four different types of Fe-Ti oxide deposits are found within the anorthosites of the Grenville Province: (1) magnetite-ilmenite-apatite (e.g. Buttercup); (2) Ti-magnetite-ilmenite (e.g. Magpie); (3) ilmenite-rutile (e.g. Saint-Urbain and Big Island); and (4) ilmenite (e.g. Tio mine) (Figure 1.1) (Hébert eta!. 2005, Corriveau et al. 2007). Because some of these deposits are massive oxide dykes crosscutting the 9 anorthosites, it been proposed that they crystallized from a Fe-Ti-O melt produced by immiscibility from a silicate magma (Force 1991). Fe-Ti oxide + apatite rocks (nelsonites) have long been considered to represent a Fe-Ti-P-rich immiscible melt from a silicate magma (Philpotts 1967, Kolker 1981). However, no experimental results support this hypothesis for known magmatic compositions and crustal conditions (Lindsley 2003, Tollari et al. 2006). The most commonly accepted explanation for the formation of massive Fe-Ti oxide deposits and nelsonites is by mineral accumulation (Emslie 1975; Frost & Simons 1991; Duchesne 1996 and 1999; Dymek & Owens 2001; Charlier et al. 2006). Because these deposits are associated with anorthosite massifs and because crystallization of large quantities of plagioclase could substantially enrich the residual magma in Fe and Ti, a direct link between anorthosites and Fe-Ti oxide deposits has been proposed (Emslie 1975, Ashwal 1993, Duchesne 1996). The genetic link between the Saint-Urbain and Big Island deposits and their host anorthosites is assessed in this study based on Hf-Pb-Sr isotopic geochemistry. The focus of this dissertation is the rutile-bearing massive ilmenite deposits of the Saint-Urbain and Big Island deposits located within the Saint-Urbain and Lac Allard anorthosite, respectively. The presence of widespread sapphirine (Mg, Fe, Al silicates) (up to 5 modal%) is a unique feature of these two deposits, thus one important question of this study was to evaluate whether the observed mineral assemblage was primary or secondary, related to sub-solidus reactions or metamorphism of the Fe-Ti oxide deposits. Combined, the results presented in this dissertation significantly advance our understanding of the crystallization and sub-solidus evolution of rutile-bearing Fe-Ti oxide deposits associated with Proterozoic anorthosites of the Grenville Province of eastern Canada. The crystallization ages and slow-cooling histories documented for the Saint-Urbain and Lac Allard anorthosites by U-Pb and40Ar/39r geochronology indicate emplacement during the Ottawan Grenvillian events, and give support to the role of sub solidus reactions in the deposits (e.g. zircon rims around ilmenite, sapphirine formation). The maj or element and trace element concentrations of minerals (ilmenite, rutile, spinel, corundum, plagioclase, orthopyroxene, sapphirine, biotite) indicate a magmatic origin for the rutile and allow for a characterization of the compositions and f02 evolution of the magmas parental to the deposits. The mineral compositions are also used to balance 10 reactions for the sub-solidus formation of sapphirine. Radiogenic isotopic compositions (Hf, Pb, Sr) illustrate that the deposits and their respective host anorthosites share the same source. Based on these new isotopic results and previously published results from anorthosite worldwide, a mantle source for the generation of magmas parental to Proterozoic anorthosite massifs is defined. Finally, by comparing the compositions of ilmemte from the Saint-Urbain and Big Island deposits to naturally-occurring ilmenite from other Fe-Ti deposits and to experimental results, a model for the formation of rutile bearing ilmenite deposits is proposed. 1.2 Outline of the dissertation and contributions to the research This dissertation was prepared in manuscript format and contains five main chapters, each designed as a manuscript to be published in an international scientific journal. Some repetition of the locations of the studies and the geological context is inevitable between the different chapters, although each was adapted to the specific topic of the chapter. Research results related to this dissertation were presented at one national and five international conferences (Morisset et al. 2003, Morisset et al. 2005a and b, Morisset et al. 2006a and b, Morisset et al. 2007). Field work and sample collection was carried in three separate field seasons: five weeks in summer 2002, two weeks in summer 2004 and one week in summer 2005. All chemical and analytical work in this study was carried at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia with the exception of: (1) mineral separation, XRF analyses and plagioclase and apatite trace elements analyses by ICP-MS that were done at the Université de Liege (Belgium); and (2) microprobe analyses that were undertaken at the University of Bochum (Germany) and at the Electron Microbeam / X-Ray Diffraction Facility of the University of British Columbia. In Chapter II, the first precise crystallization ages for the Saint-Urbain and Lac Allard anorthosites are presented as well as the cooling histories for both intrusions based on U Pb geochronology (9 zircon, 6 rutile and 4 apatite) and40Ar/39rgeochronology (7 biotite and 4 plagioclase). All mineral separates for U-Pb and40Ar/39rwere prepared by C.-E. Morisset, except for one sample that was prepared by Hai Lin. The chemistry 11 and mass spectrometric analyses were carried by Rich Friedman with the assistance of Rachel Lishansky in the clean lab and Hai Lin with the mass spectrometer. All40Ar/39r analyses and data reduction were carried by Thomas Ulrich. In Chapter III, a study that characterizes the formation of zircon rims on ilmenite and rutile from the Saint-Urbain and Big Island Fe-Ti oxide deposits is presented. The rims form by sub-solidus diffusion of Zr from Ti-based oxides and reaction to form zircon along the grain margins. Microprobe compositions of ilmenite, rutile and zircon and back-scattered imaging were obtained by C.-E. Morisset with training from Mati Raudsepp. These results were first acquired in the context of a graduate course at UBC (EOSC 521) with Mati Raudsepp. In addition to samples from the Saint-Urbain and Big Island deposits, samples from other Proterozoic intrusions were analyzed, including the Mirepoix layered intrusion (Lac Saint-Jean anorthosite massif, Morisset 2002), the unrecrystallized Twin Lakes intrusive complex (Ontario), and the Laramie anorthosite (Wyoming, western U.S.). In Chapter IV, the geochemistry of rutile and ilmenite is used to establish the magmatic nature of rutile, the compositional evolution of ilmenite during crystallization, and the link between the anorthosite and the deposits at each location. The analytical results include major and trace elements compositions (XRF, HR-ICP-MS) and Hf isotopic compositions (MC-ICP-MS) of rutile and ilmenite separates (11 rutile separates, 36 ilmenite separates for major elements, and 20 ilmenite separates for complete trace elements and Hf isotope ratios). The Hf isotopic compositions of four whole rocks were analyzed to characterize the signature of the country rocks. Mineral separation and major element analyses were carried out at the Université de Liege, Belgium, with guidance from Jean-Clair Duchesne, Guy Bologne and Jacqueline Vander Auwera. The methods for the digestion of ilmenite and rutile and for the HR-ICP-MS analyses were formulated with the assistance of Bert Mueller and Wilma Pretorious. The chemical purification of Hf from ilmenite was developed with the help of André Rahier and is detailed in the first part of this chapter. The Hf isotopic compositions were measured on a MC-ICP-MS with the training of Jane Barling. In Chapter V, the nature of the link between the deposits and the anorthosites is confirmed and the source of the parental magmas to the anorthosites is evaluated based 12 on combined major and trace element geochemistry and Pb-Sr isotopic compositions. The major element contents (XRF and microprobe) and trace element concentrations (ICP-MS) of plagioclase separates (17 separates analyzed for major elements, 21 separates analyzed for trace elements) and apatite separates (5 separates analyzed for trace elements and Sr isotopic compositions) were mostly determined at the University of Liege with the help of Guy Bologne under the supervision of Jacqueline Vander Auwera. The Pb and Sr isotopic compositions of leached plagioclase (14 samples) and four whole rock samples from the surrounding area at each location were also determined as well as the Sr isotopic composition of apatite (5 samples) from the Saint-Urbain deposits. All sample digestion and chemical purifications for the isotopic results were prepared in clean laboratories (Class 1000) with training from Bruno Kieffer. The Pb isotopic compositions were measured on the MC-ICP-MS under the supervision of Jane Barling, and the Sr isotopic compositions were measured on a TIMS by Bruno Kieffer. The results are used to constrain the REE concentrations of the parental magmas. The initial isotopic ratios indicate that the Saint-Urbain deposits are only slightly contaminated (<5%) by the gneissic country rock. More importantly, combination of the new results with all available Pb, Sr and Nd isotopic compositions are used to propose a mantle origin for the Proterozoic anorthosites of the Grenville Province. Finally, Chapter VI is an evaluation of the origin of the rutile-bearing ilmenite deposits based on representative electron microprobe analyses of mineral phases present in the Saint-Urbain and Big Island deposits (rutile, spinel, corundum, plagioclase, orthopyroxene, sapphirine and biotite). The microprobe analyses were performed at the Bochum University (Germany) with the help of H.-J. Bernhardt. Major and trace elements analyses (XRF) of whole rocks from each deposit, host anorthosite and surrounding rocks were carried out at the Université de Liege with the guidance of Jean Clair Duchesne, Guy Bologne and Jacqueline Vander Auwera. Natural ilmenite compositions of other Fe-Ti deposits and layered intrusions and experimental ilmenite compositions from three-or multi-component systems were compiled and used to define the conditions of crystallization and the stability of Fe-Ti oxide minerals and rutile saturation. With the help of Terry Gordon, it was possible to balance reactions for the formation of sub-solidus sapphirine in the studied deposits. Based on the results 13 presented in this chapter, and integrating results from previous chapters, a model for the formation of rutile-bearing Fe-Ti deposits in Proterozoic anorthosites is presented. The concluding Chapter VII provides a summary of the knowledge gained from the research carried out for this dissertation. This study marks an important contribution to the understanding of the formation of rutile-bearing Fe-Ti oxide deposits and of their host rock anorthosites in terms of their source and their magmatic and post-magmatic evolution. Ultimately, these results can be integrated to provide better insight into Proterozoic magmatism, especially that related to anorthosite-mangerite-charnockite granite suites. Chapters 2, 4 and 6 each contain appendices where additional material used in the separate studies is located (e.g. complete sample locations and descriptions, analytical tables). The CD-ROM associate with the thesis contains the complete microprobe dataset used in Chapter 6. 14 1.3 References Ashwal, L.D. (1993) Anorthosite. Minerals and Rocks (ed. P.J. Wyllie, A. El Goresy, W. von Engelhardt and T. Hahn) Spinger-Verlag, Berlin. pp.422 Barnichon, J.D., Havenith, H., Hoffer, B., Charlier, R., Jongmans, D., Duchesne, J.C. (1999): The deformation of Egersund-Ogna anorthosite massif, south Norway: finite- element modeling of diapirism. Tectonophysics 303, 109-130. Bergeron, M (1986): Mineralogie et géochimie de Ia suite anorthositique de la region du Lac Allard, Québec: evolution des membres mafiques et origine des gItes massifs d’ilménite. Unpublished Ph.D. dissertation, Université de Montréal, 485 p. Charlier, B., Duchesne, J.-C., & Vander Auwera, J. (2006): Magma chamber processes in Telines ilmenite deposit (Rogaland Anorthosite Province, SW Norway) and the formation of Fe-Ti ores in massif-type anorthosites. Chemical Geology 234, 264-290. Cloutier, M.A. (1982): La mineralization de rutile-hemo-ilmenite-sapphirine du lac Big Island, region du lac Allard, Quebec. B.Sc. thesis. École Polytechnique, Université de Montréal. 42 p. Cordani, U.G., D’Agrella-Filho, M.S., Brito-Neves, B.B., & Trindade, R.I.F. (2003): Tearing up Rodinia: the Neoproterozoic paleogeography of South American cratonic fragments. Terra Nova 15, 350-359. Corriveau, L., Perrault, S. and Davidson, A. (2007): Prospective metallogenic settings of the Grenville Province. In: Mineral Deposits of Canada: A Synthesis of Major Deposit-Types, District Metallogeny, the Evolution of Geological Provinces, and Exploration Methods (ed. W.D. Goodfellow). Geological Survey ofCanada, Mineral Deposits Division, Special Publication 5, 819-847. Davidson, A. (1998): An overview of the Grenville Province geology, Canadian Shield. In: Geology of the Precambrian Superior and Grenville Provinces and Precambrian Fossils in North America (eds S. B. Lucas & M. R. St-Onge). Geological Survey of Canada, Geology of Canada 7, 205-270. Dickin A.P. (2000): Crustal formation in the Grenville Province: Nd-isotope evidence. Canadian Journal ofEarth Sciences 37, 165-18 1. Dickin, A.P., and Higgins, M. (1992): Sm!Nd evidence for a major 1.5 Ga crust-forming event in the central Grenville province. Geology 20, 137—140 Duchesne, J.C. (1996): Liquid ilmenite or liquidus ilmenite: A comment on the nature of ilmenite vein deposits. In: Petrology and Geochemistry of Magmatic Suites of Rocks in the Continental and Oceanic Crusts. A Volume Dedicated to Professor Jean Michot (ed. D. Demaiffe). Université Libre de Bruxelles. Royal Museum for Central Africa. Tervuren, Belgium. 73-82. Duchesne, J.C. (1999): Fe-Ti deposits in Rogaland anorthosites (South Norway): geochemical characteristics and problems of interpretation. Mineralium Deposita 34, 182-198. 15 Duchesne, J.-C., Liégeois, J.-P., Vander Auwera, J. & Longhi, J. (1999): The crustal tongue melting model and the origin of massive anorthosites. Terra Nova 11, 100- 105. Dymek, R.F. & Owens, B.E. (2001): Petrogenesis of apatite-rich rocks (nelsonites and oxide-apatite gabbronorites) associated with massif anorthosites. Economic Geology 96, 797-815. Emslie, R.F. (1975): Major rock units of the Morin complex, south-western Quebec. Canadian Geological Survey paper 74-48, 37 p. Emslie, R.F. (1978): Anorthosite massifs, rapakivi granites, and late Proterozoic rifting of North America. Precambrian Research 7, 6 1-98. Emslie, R.F. (1980): Geology and petrology of the Harp Lake complex, central Labrador: an example of Elsonian magmatism. Geological Survey of Canada, Bulletin 293, pp136. Emslie, R.F. (1985): Proterozoic anorthosite massifs. In: The Deep Proterozoic Crust in the North Atlantic Provinces (eds A.C. Tobi & J.L.R. Touret), Reidel, Dordrecht, 39- 60. Emslie, R.F., Hamilton, M.A. & Thériault, Ri. (1994): Petrogenesis of a Mid Proterozoic anorthosite-mangerite-charnockite-granite (AMCG) complex: isotopic and chemical evidence from the Nain Plutonic Suite. Journal of Geology 102, 539- 558. Force, E.R. (1991): Geology of titanium-mineral deposits. Geological Society of America, Special Paper 259, pp 112. Frost, B.R. & Simons, J.P. (1991): Fe-Ti oxide deposits of the Laramie anorthosite complex: their geology and proposed economic utilization. Wyoming Geological Association Guidebook, 42nd Field Conference, 41-48. Fram, M. S. & Longhi, J. (1992): Phase equilibria of dikes associated with Proterozoic anorthosite complexes. American Mineralogist 77, 605-6 16. Gower, C.F. & Krogh, T.E. (2002): A U-Pb geochronological review of the Proterozoic history of the eastern Grenville Province. Canadian Journal of Earth Sciences 39, 795-829. Hamilton, M.A., McLelland, J. & Selleck, B. (2004): SHRIMP U-Pb zircon geochronology of anorthosite-mangerite-charnockite-granite suite, Adirondack Mountains, New York: Ages of emplacement and metamorphism. In: Proterozoic Tectonic Evolution of the Grenville Orogen in North America (eds R.P. Tollo, L. Corriveau, J.M. McLelland, M.J. Bartholomew). Geological Society of America Memoir 197, 337-3 55. Hébert, C. (2001): Géologie de la region de lac Mélonèze (SNRC 22E/10). Ministère des ressources naturelles, Quebec, RG2001-02. 25 p. Hébert, C, Cadieux, A.-M. & van Breemen, 0. (2005): Temporal evolution and nature of Ti-Fe-P mineralization in the anorthosite-mangerite-charnockite-gramte (AMCG) suites of the south-central Grenville Province, Saguenay — Lac St. Jean area, Québec, Canada. Canadian Journal ofEarth Sciences 42, 1865-1880. 16 Higgins, M.D. & van Breemen, 0. (1996): Three generations of anorthosite-mangerite charnockite-granite (AMCG) magmatism, contact metamorphism and tectonism in the Saguenay-Lac-St.-Jean region of the Grenville Province, Canada. Precambrian Research 79, 327-346. Hunt, T.S. (1853): Reports of Progress. Geological Survey of Canada. Kolker, A. (1982): Mineralogy and geochemistry of Fe-Ti oxide and apatite (nelsonite) deposits and evaluation of the liquid immiscibility hypothesis. Economic Geology 77, 1146-1158. Lafrance, B., John, B.A. & Scoates, J.S. (1996): Syn-emplacement recrystallization and deformation microstructures in the Poe Mountain anorthosite, Wyoming. Contributions to Mineralogy and Petrology 122, 43 1-440. Li, X.L., Bogdanova, S.V., Collins, A.S., Davidson, A., De Waele, B., Ernst, R.E., Fitzsimons, I.C.W., Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karistrom, K.E., Lu, S., Natapov, L.M., Pease, V., Pisarevsky, S.A., Thrane, K., Vernikovsky, V. (2008): Assembly, configuration, and break-up history of Rodinia: A synthesis. Precambrian Research 160, 179-210. Lindsley, D.H. (2003): Do Fe-Ti oxide magmas exist? Geology: Yes; Experiments: No! In: Ilmenite Deposits and Their Geochemical Environment, Volume Special Publication No. 9. (eds J.C. Duchesne & A. Korneliussen), Norges Geologiske Undersokelse, 34-35. Logan, W.E. (1850): Geology of the vicinity of St. Paul and Murray Bay. Geological Survey of Canada, Progress Report, 1849-1850. Longhi, J. (2005): A mantle or mafic crustal source for Proterozoic anorthosites? Lithos 83, 183-198. Longhi, J. & Ashwal, D. (1985): Two-stage models for lunar and terrestrial anorthosites: petrogenesis without a magma ocean. Journal of Geophysical Research 90 sup., C571-C584. Longhi, J., Vander Auwera, J., Fram, M.S. & Duchesne, J.C. (1999): Some phase equilibrium constraints on the origin of Proterozoic (massif) anorthosite and related rocks. Journal ofPetrology 40-2, 339-362. Mitchell, J.N., Scoates, J.S. & Frost, C.D. (1995): High-Al gabbros in the Laramie Anorthosite Complex, Wyoming: implications for the composition of melts parental to Proterozoic anorthosite. Contributions to Mineralogy and Petrology 119, 166-180. Morisset, C.E. (2002): Étude pétrologique et geochimique de la minéralisation en oxydes de fer et titane et en apatite du complexe du lac Mirepoix (Anorthosite du lac Saint- Jean, Quebec). Mémoire de maItrise, Université du Québec a Montréal, Montréal, 116 p. Morisset, C.E., Charlier, B., Cloots, R., Brassard, B. & Vander Auwera, J. (2003): The rutile and hemo-ilmenite mineralizations of Québec. The Biennal SGA Meeting — Abstract, Athens. Morisset, C.E., Scoates, J.S. & Weis, D. (2005a): Exsolution origin for zircon rims around hemo-ilmenite in magmatic Fe-Ti oxide deposits. Geochimica et Cosmochimica Acta 69-10, supplement 1, A16. 17 Morisset, C.E., Scoates, J.S. & Weis, D. (2005b): Pb isotopic compositions of plagioclase from Fe-Ti ore deposits related to the Saint-Urbain and Havre Saint-Pierre Anorthosites (Grenville Province, Canada). AGU Fall Meeting — Abstract, San Francisco, V41G-1543. Morisset, C.E., Scoates, J.S. & Weis, D. (2006a): Constraining the origin of Grenvillian Proterozoic anorthosite massifs using the Pb isotopic compositions of plagioclase. GAC-MAC Meeting — Abstract, Montréal. Morisset, C.E., Scoates, J.S. & Weis, D. (2006b): Trace element and Hf isotopic compositions of magmatic rutile from Fe-Ti oxide ore deposits related to the Proterozoic anorthosite massifs. AGU Fall Meeting — Abstract, San Francisco, V33A- 0644 Morisset, C.E., Scoates, J.S. & Weis, D. (2007): Rutile-bearing hemo-ilmenite deposits in Proterozoic anorthosite massifs of Quebec. Cordilleran RoundUp — Abstract, Vancouver. Morse, S.A (1978): Kiglapait Geochemistry I: systematics, sampling, and density. Journal ofPetrology 20, 555-590. Morse, S.A. (1982): A partisan review of Proterozoic anorthosites. American Mineralogist 67, 1087-1100. Morse, S.A (2006): Labrador massif anorthosites: Chasing the liquids and their sources. Lithos 89, 2020-221. Nolan, K.M. & Morse, S.A. (1986): Marginal rocks resembling the estimated bulk composition of the Kiglapait intrusion. Geochimica et Cosmochimica Acta 50, 2381- 2386. Philpotts, A.R. (1967): Origin of certain iron-titanium oxide and apatite rocks. Economic Geology 62, 303-3 15. Retty, J.A. (1942): Lower Romaine river area, Saguenay County. Département des Mines, Québec, Rapport Geologique 19, 12 p. Rivers, T. (1997): Lithotectonic elements of the Grenville Province: review and tectonic implications. Precambrian Research 86, 117-154. Rivers, T. & Corrigan, D. (2000): Convergent margin on southeastern Laurentia during the Mesoproterozoic: tectonic implications. Canadian Journal of Earth Sciences 37, 359-383. Rivers, T. & Ketchum, J., Indares, A. & Hynes, A. (2002): The high pressure belt in the Grenville Province: architecture, timing, and exhumation. Canadian Journal ofEarth Sciences 39, 867-893. Rose, E.R. (1969): Geology of titanium and titaniferous deposits of Canada. Geological Survey of Canada, Economic Geology Report No 25, 177 p. Royse, K.R. & Park, R.G. (2000): Emplacement of the Nain anorthosite: diapiric versus conduit ascent. Canadian Journal ofEarth Sciences 37, 1195-1207. Scoates, J.S (2003): Constraining the source of Proterozoic anorthosites. Geological Society of America, Abstracts with Programs 35, 394. 18 Scoates, J.S. & Mitchell, J.N. (2000): The evolution of troctolitic and high Al basaltic magmas in Proterozoic anorthosite plutonic suites and implications for the Voisey’s Bay massive Ni-Cu sulfide deposit. Economic Geology 95, 677-701. Scoates, J.S. & Chamberlain, K.R. (2003): Geochronologic, geochemical and isotopic constraints on the origin of monzomtic and related rocks in the Laramie anorthosite complex, Wyoming, USA. Precambrian Research 124, 269-304. Taylor, S.R., Campbell, I.H., McCulloch, M.T., McLennan, S.M. (1984): A lower crustal origin for massif-type anorthosites. Nature 311, 372-374. Tollari, N., Toplis, M.J. & Barnes, S.-J. (2006): Predicting phosphate saturation in silicate magmas: An experimental study of the effects of melt composition and temperature. Geochimica et Cosmochimica Acta 70, 15 18-1536. Vander Auwera, J. & Longhi, J. (1994): Experimental study of a jotunite: constraints on the parent magma composition and crystallization conditions, P, T, f02 of the Bjerkreim-Sokndal layered intrusion, Norway. Contributions to Mineralogy and Petrology 118, 60-78. Vander Auwera, J., Longhi, J. & Duchesne, J.-C. (1998): A liquid line of descent of the jøtunite (hypersthene monzodiorite) suite. Journal ofPetrology 39, 439-468. Warren, C.H., (1912): The ilmenite rocks near St-Urbain, Quebec; a new occurrence of rutile and sapphirine. American Journal ofScience 33, 263-277. Wiebe, R.A. (1992): Proterozoic anorthosite complexes. In: Proterozoic Crustal Evolution (ed. K.C. Condie). Elsevier, Amsterdam, pp 215-261. Weil, A., Van der Voo, R., Mac Niocaill, C. & Meert, J.G. (1998): The Proterozoic supercontinent Rodinia: paleomagnetically derived reconstructions for 1100 to 800 Ma. Earth and Planetaiy Science Letters 154, 13-24. Wilson, J.R., Robins, B., Nielsen, F.M., Duchesne, J.C., Vander Auwera, J. (1996): The Bjerkriem-Sokndal layered intrusion, Southwest Norway. In: Layered Intrusions (ed R.G. Cawthron). Elsevier, Amsterdam, pp 23 1-255. 19 Chapter II Crystallization ages, cooling histories, and tectonic setting of two Proterozoic anorthosite massifs, Grenville Province, Québec: Saint-Urbain and Lac Allard (Havre-Saint-Pierre)’ A version of this chapter was submitted for publication to an international scientific journal. Morisset, C.-E., Scoates J.S., Weis, D. & Friedman, R. 20 2.1- Introduction Anorthosite-mangerite-oharnockite-granite (AMCG) intrusive suites are a major component of the Proterozoic Grenville Province of eastern Canada and the northeastern U.S. (Figure 2.1) and of the Sveconorvigian Province of southern Norway and Sweden. The oldest anorthosite massifs, including the 1640 Ma Mealy Mountains (Emslie & Hunt, 1990) and 1648-1628 Ma intrusions of the Manicouagan Imbricate Zone (Indares et al., 1998), are coeval with the pre-Grenvillian Labradorian deformation event (1710- 1600 Ma; Gower & Krogh, 2002). Only a few anorthosites have ages in the interval 1 370 to 1325 Ma and they do not appear to be associated with any regional-scale deformation event (e.g. 1365 Ma Rivière Pentecôte, Emslie & Hunt, 1990; 1327 Ma De la Blache, Gobeil et al., 2002). In contrast, the majority of the youngest Grenvillian massifs, including the very large Lao Saint-Jean and Havre-Saint-Pierre anorthosite suites, were emplaced during the main deformation events from 1200 to 980 Ma related to the accretionary tectonics that formed the Grenville Province (Rivers, 1997; Rivers et al., 2002). Determining precise crystallization ages and cooling histories of AMCG suites and comparing them to regional deformation and metamorphic events is critical for evaluating the tectonic setting required to generate Proterozoic anorthosites. Over the past 15-20 years, significant progress has been made in documenting the precise ages of Grenvillian AMCG suites, however a number of intrusions still remain undated. In this study, we present U-Pb zircon ages for the Saint-Urbain anorthosite, which hosts numerous small Fe-Ti oxide deposits, and for the Lac Allard lobe of the giant Havre Saint-Pierre anorthosite suite, which hosts the Lao Tio mine, the world’s largest magmatic ilmenite deposit, and the Big Island massive Fe-Ti oxide dyke. To constrain the thermal evolution of the Saint-Urbain and Lac Allard anorthosites, U-Pb rutile/apatite and 40Ar/39rbiotite/plagioclase ages were determined from the associated Fe-Ti oxides ores. Combined, these new results allow for the construction of the cooling history of these two AIVICG massifs and assessment of their relationship to Grenvillian tectonic events. 21 Figure 2.1 Simplified geological map of the Grenville Province adapted from Davidson (1998) and Corriveau et al. (2007). Ages for Labradoria and Quebecia are based on Nd model ages from Dickin (2000). Inset map in the lower right part of the figure shows the relative location of the map area in North America (from Davidson, 2008). Anorthosite massifs: (a) Adirondack; (b) Morin; (c) Saint-Urbain; (d) Lac Saint-Jean; (e) Mattawa; (f) Labrieville; (g) De La Blache; (h) Rivière Pentecôte; (i) Havre-Saint-Pierre; (j) Lac Allard; (k) Lac Fournier; (1) Atikonak; (m) Mealy Mountains; (n) Harp Lake; (o) Vieux-Fort; (p) Nain plutonic suite; (q) Roseland; (r) Laramie. Boxes indicate areas outlined in more detail in Figures 2.2 and 2.4. Anorthosite Mangerite Granite Gabbro Late Paleoproterozoic orogens 1.84-1.96 Ga Labradoria 1.65-1.75 Ga Quebecia Central metasedimentary belt 22 2.2- Overview of the Grenville Province The Proterozoic Grenville tectonic province stretches from Mexico to Labrador in North America (Davidson, 2008). The Grenville Front, which is marked by ductile and brittle faults, is the northern limit of deformed or transported rock with the Archean Superior Province (Davidson, 1998). Rivers et al. (1989) defined three regions within the Grenville Province: (1) the parautochthonous belt, which is adjacent to the Grenville Front and contains rocks of the Superior Province that have been deformed and transported; (2) the allochthonous polycyclic belt, which is located in the interior of the orogen and contains pre-deformed rocks unrelated to the Superior Province (no Archean rocks have been identified in this region) (Davidson, 1998); and (3) the allochthonous monocyclic belt in the south part of the orogen where supracrustal rocks have been deformed, metamorphosed and intruded by plutonic rocks only during Grenvillian time (Figure 2.1). AMCG plutonic suites represent 20% of the exposed Grenville Province crust and are distributed throughout the polycyclic allochthonous belt (Figure 2.1). The Grenville Province formed on the margin of Laurentia, which was built by multiple island arc and continent-continent collisions from 1.98 to 1.79 Ga and that resulted in the formation of Paleoproterozoic orogenic belts around the Superior Province (e.g. Trans-Hudson, Figure 2.1 inset) (Hoffman, 1989). Three separate orogenic events are believed to have occurred during the formation and deformation of the margin of Laurentia prior to the Grenvillian Orogeny (Rivers, 1997): the Labradorian (1960-1660 Ma), Pinwarian (1500-1450 Ma) and Elzevirian (1250-1190 Ma) events. During these periods, the south margin of the Laurentian continent was the site of Andean-type subduction. Labradorian activity is concentrated in the northeast section of the Grenville Province where an oceanic back arc, with a significant proportion ofjuvenile crust, was accreted to the southeastern part of Laurentia (Gower, 1996, Rivers, 1997). A continental margin magmatic arc that alternated between extensional and compression systems prevailed during the period from 1500 to 1230 Ma (Rivers & Corrigan, 2000). During relatively brief extensional periods the formation of continental back-arc basins (e.g. Wakeham Supergroup) and marine back-arc basins (e.g. Central Metasedimentry Belt) occurred (Rivers & Corrigan, 2000). These basins were closed during the pre-Grenvillan 23 accretionary orogenies of the Pinawarian and Elzivirian. Finally, the Grenville orogeny was produced during continent-continent collision from 1190-980 Ma. The Grenville orogeny occurred in three phases: (1) the Shawinigan orogeny 1190-1140 Ma, (2) the Ottawa orogeny 1080-1020 Ma; and (3) the Rigolet orogeny 1010-990 Ma (Rivers, 1997; River et al., 2002). Gower & Krogh (2002) consider that continent-continent collision occurred only in the interval from 1080-1020 Ma, however recent palaeogeographic reconstructions suggest that continent-continent collision was diachronous or progressive; it occurred at ca. 1100 Ma in the south (present Texas) and that the active subduction zone beneath northern Laurentia resulted in Amazonia colliding with Laurentia only around ca. 1000 Ma (Li et al., 2008). 2.3- Geology of the Saint-Urbain region The Saint-Urbain anorthosite is a small pluton (—450 km2) located north of Baie Saint-Paul within the allochthonous polycyclic belt of the Grenville Province (Figures 2.1 and 2.2). The massif is predominantly andesine anorthosite (Figure 2.3a) and locally contains xenoliths of labradoritic composition (Dymek, 2001). Fe-Ti oxide mineralization occurs in eight discrete deposits found in the southwestern part of the anorthosite (Bignell; Coulomb West; Coulomb East; General Electric; Séminaire; Furnace; Bouchard; and Glen) (Figure 2.2b) (Chapter 6). Contacts with the host anorthosite range from sharp (Figure 2.3b) to gradational. The mineralogy of the deposits is defined by variable proportions of hemo-ilmenite (subsequently referred to as ilmenite for simplicity), rutile, plagioclase, sapphirine, orthopyroxene, apatite, biotite, pleonaste spinel with trace amounts of corundum, sulphide (pyrite, pyrrhotite, and chalcopyrite), and zircon (Figure 2.3c) (Chapter 6). Jøtunite (ferrodiorite) outcrops at the southeast contact of the massif and layered oxide-apatite gabbronorite (OAGN) occurs along the northwestern limit of the intrusion (Figures 2.2 and 2.3d). The jøtunites are interpreted by Icenhower et al. (1998) to be residual products of anorthosite crystallization and the presence of layering suggests that the OAGN may be cumulates formed from jøtunitic magma. 24 LEGEND (a) Paleozoic Cover Rocks Proterozoic other intrusions Anorthosite 1020-1010 Ma Valin anorthosite suite Labrieville anorthosite Mattawa anorthosite Charnockite, and monzonite 1080-1045 Ma period Saint-Urbain anorthosite jJ Saint-Anne du Nord opx-granodiorite EJ Mangerite, charnockite, and granite 1160-1140 Ma Lac St-Jean anorthosite suite Anorthosite, gabbronorite, charnockite, and mangerite “1327 Ma De Ia Blache plutonic suite 1 Anorthosite, gabbronorite Gneiss complex Fault LEGEND (b) Anorthosite Des Martres Mangeritic rocks Group Jotunite E] Saint-Tite-des Gneiss Complex Caps Group .—“ Road * Samples Charlevoix impact structure inner limit — — — outer limit Figure 2.2 Simplified geological map ofthe Lac Saint-Jean and Saint-Urbain anorthosite areas (modified from Hébert et al., 2005). (a) The ages of the 1160-1140 Ma Lac Saint-Jean anorthosite suite area are from Higgins & van Breemen (1992). Numbers in the legend on the right represent different intrusions. References for the crystallization age of the massifs are numbered in brackets after the age on the map and are as follows: (1) Gobeil et al. (2002); (2) Hébert et al. (2005); (3) Owens et al. (1994); (4) Hébert et al. (1998); (5) Higgins & van Breemen (1996); (6) Hervet et al. (1994); and (7) this study. (b) Simplified geological map of the Saint-Urbain anorthosite after Rondot (1989). Sample locations for geochronology are indicated with a star and sample number. (SANO) Saint-Anne du Nord orthopyroxene granodiorite; (RMO) Rivière Malbaie orthopyroxene granodiorite; (SUA) Saint-Urbain anorthosite; Des Martres Group and Saint Tite-des-Caps Group metasediments; (BSP) town ofBaie-Saint-Paul. 25 Figure 2.3 Photographs of field relationships for geochronological samples (a-e: Saint-Urbain; f: Big Island). a) anorthosite from the Lac des Cygnes area; dark areas are unrecrystallized plagioclase and white areas are strongly recrystallized plagioclase (10cm knife for scale); b) sharp contact between the Bignell Fe-Ti oxide deposit with host anorthosite (40 cm long hammer for scale); c) banded plagioclase with ilmenite-rutile horizon from the Séminaire deposit 2033-D (2.5 cm diameter coin for scale); d) layered oxide-apatite gabbronorite (OAGN) with slightly folded plagioclase-rich and oxide-pyroxene-rich layers from the northwest border ofthe Saint-Urbain massif (outcrop is 1.5 m high); e) strongly foliated orthopyroxene granodiorite from the Saint-Anne-du-Nord massif (2 cm diameter coin for scale); f) sharp contact between the Big Island dyke and anorthosite of the Lay Allard anorthosite (40 cm long hammer for scale). 26 The Saint-Urbain anorthosite intruded charnockitic migmatites of unknown age, which are part of the Quebecia crust as defined by Dickin (2000). The anorthosite is bordered along its western margin by the large (1800 km2) Saint-Anne du Nord orthopyroxene granodiorite (Figure 2.3e) (also referred as the Lac des Martres mangerite, e.g. Dymek, 2001). The temporal relationship between the Saint-Urbain anorthosite and the granodiorite cannot be established by field observations as the contact is not exposed (Rondot, 1989; Icenhower et al., 1998). Both the Saint-Urbain and Saint-Anne du Nord intrusions are associated with an extensive area (‘—20,000 km2)of mangerite, charnockite and granite located to the west and north of the Saint-Urbain anorthosite (Figure 2.2a). These massifs were grouped into the 1080-1061 Ma Vanel anorthositic suite (Hébert & van Breemen, 2004a), but are now part of an unnamed 1080-1045 Ma magmatic suite (Hébert et al., 2005). The association of the Saint-Urbain anorthosite with these massifs is based on a relatively imprecise 1079 ± 22 Ma mineral-whole rock Sm-Nd isochron age for the Saint-Urbain massif (Ashwal & Wooden, 1983). The Saint-Urbain anorthosite and Saint-Anne du Nord orthopyroxene granodiorite are separated from the other massifs by the Saint-Fulgence shear zone, which likely represents a major thrust fault (Hébert & Lacoste, 1998) (Figure 2.2a). Metamorphic grade in the region varies from greenschist to granulite facies (Rondot, 1989). The presence of charnockitic migmatite in the Pare des Laurentides Complex, which outcrops near the Saint-Urbain anorthosite, is indicative of a high-grade metamorphic environment (Rondot, 1989). The temperature and pressure conditions of the country rock north and south of the Saint-Urbain intrusions have been determined by Rondot (1989) by geothermobarometry of metasedimentary rocks from the Des Martres (temperatures from 540 to 800°C and pressures from 5.1 to 6.3 Kbar) and the Saint-Tite des-Caps Groups (temperature of 700°C and pressure of 5.5 Kbar) (Figure 2.2b). The epicentre of the Charlevoix meteorite impact is located east of Baie-Saint-Paul (Figure 2.2). Impactites close to the epicentre give K-Ar ages of 372 and 342 Ma (Rondot, 1971). The inner circle (long-dashed line) on Figure 2.2b shows the limit of impact metamorphism and the subsequent post-crater collapse radius is shown by the outer circle (short-dashed line). The samples from this study (Table 2.1) were collected 27 Table 2-1 Geochronology sample locations and descriptions Deposit or intrusion Sample - Coordinate - Rock type Geochronology analyses E N 2006-B 1 2006-Gi 2006-C2 3 82 064 3 81 968 3 82 125 Saint-Urbain region 52 61112 Ilmenitite 52 66 105 Ilmenitite with rutile 52 66 382 Anorthosite 2006-C4 3 82 050 52 66 384 Anorthosite Bignell Bignell Saint-Urbain Anorthosite Saint-Urbain Anorthosite Bignell Bignell Coulomb East Coulomb East Saint-Urbain Anorthosite Lac des Martres General Electric Séminaire Séminaire Furnace Furnace Furnace Saint-Urbain Saint-Urbain Big Island dyke Big Island dyke Big Island dyke Big Island dyke Big Island dyke Lac Allard Anorthosite Lao Allard Anorthosite Lao Allard Anorthosite 2006-Gi 2006-F 1 2015-A4 2015-B4 2020 2023 2030-B2 2033-A2 2033-D 2036-BiB 2036-B3 2036-D2 2042-A 2043 2102 2103-B2 2l04-D 21 06-D 2109-A 2114-B 3 81 968 5266 105 Ilmenitite with rutile 3 82 125 52 66 382 Nelsonite 3 82 943 52 65 611 Ilmenite anorthosite 3 82 900 52 65 611 Ilmenite-rutile anorthosite 3 74 742 52 85 079 OAGN 368016 5282933 Mangerite 3 82 918 52 65 806 Ilmenitite with rutile 3 82 465 52 65 531 Biotite contact deposit anorthosite 3 82 455 52 65 524 Ilmenite-rutile anorthosite 3 83 090 52 66 680 Megacrystic leuconorite 3 83 090 52 66 680 Nelsonite 3 83 070 52 66 670 Nelsonite 3 79 295 52 77 212 Megacrystic leuconorite 379516 5280813 Anorthosite Lac Allard region (Havre-Saint-Pierre Suite) 4 50 494 55 88 945 Ilmenitite 4 50 494 55 88 945 Ilmenitite 4 50 514 55 88 945 Ilmenitite with rutile 4 50 530 55 88 945 Ilmenitite with plagioclase 4 50 615 55 88 960 Ilmenitite with rutile 450405 5588760 Anorthosite Ar-Ar: bt U-Pb: rt U-Pb: zrc Ar-Ar: bt U-Pb: rt U-Pb: ap Ar-Ar: bt, p1 U-Pb: rt U-Pb: zrc U-Pb: zrc U-Pb: rt Ar-Ar: bt Ar-Ar: bt, p1; U-Pb: rt Ar-Ar: bt; U-Pb: ap U-Pb: ap U-Pb: ap Ar-Ar: bt, p1; U-Pb: zrc U-Pb: zrc U-Pb: zrc Ar-Ar: bt, p1 U-Pb: rt Ar-Ar: bt U-Pb: rt Ar-Ar: p1; U-Pb: zrc U-Pb: zrc U-Pb: zrc 2123-B 450235 5588866 Anorthosite 2132 472000 5600000 Anorthosite UTM are in NAD 27. UTM zone 19 for Saint-Urbain area and zone 20 for Havre-Samt-Pierre area. Bt (biotite); rt (rutile); zrc (zircon); ap (apatite); OAGN (oxide-apatite gabbronorite). 28 from areas well outside the limit of thermal metamorphism (Trepman et al., 2005) and mostly beyond the limit of brittle deformation related to the impact event. 2.4- Geology of the Lac Allard region (Havre-Saint-Pierre) The 11,000 km2 Havre-Saint-Pierre anorthosite suite is located in the allochthonous polycyclic belt of the Grenville Province at the limit between the Quebecia and Laurentia crust as defined by Dickin (2000) (Figure 2.1). In this region of the Grenville Province, three tectonostratigraphic domains have been identified (Gobeil et al., 2003) (Figure 2.4). The Lac-à-l’Aigle Domain is present to the northwest of the Havre-Saint-Pierre massif and contains granulite facies metamorphic rocks. The Saint-Jean Domain contains the Manitou Gneiss Complex with deformed and recrystallized mafic and felsic intrusions, the Matamec Intrusive Complex and the Havre-Saint-Pierre anorthosite suite, which have all been thrust over the Lac-à-l’Aigle Domain (Figure 2.4). The Natashquan Domain contains the Buit Complex (1535 ± 4 Ma: U-Pb detrital zircon, Wodicka et al., 2003), which was metamorphosed to amphibolite facies, and the greenschist facies Wakeham Supergroup. Madore et al. (1999) established that the peak metamorphic event reached 550°C and 3.5 kbar in the Buit Complex. A metamorphic cooling age of 1052 +61-4 Ma (U-Pb rutile: TCB --400°C, Wodika et al., 2003) for the Buit Complex indicates that peak metamorphism (i.e. 550°C) occurred slightly earlier. The Havre-Saint-Pierre anorthosite is believed to have intruded the Buit Complex (pers. comm. Serge Perreault, 2004), which also crops out in the Saint-Jean Domain along the St. Lawrence River (Figure 2.4). The Havre-Saint-Pierre anorthosite suite has been divided into four lobes by van Breemen & Higgins (1993) and into seven different units by Gobeil et al. (2003) (Figure 2.4). The giant Lac Allard ilmenite deposit and the Big Island dyke are contained within the Lac Allard lobe for which we present the first published geochronological results. A single sample from the North-West lobe yields three different U-Pb ages (Wodika et al., 2003): (1) an older age of ca. 1139 Ma (U-Pb zircon, one fraction) is interpreted to represent inherited zircon; (2) an intermediate age of 1129 ± 3 Ma (U-Pb interstitial zircon) is interpreted as the crystallization age; and (3) a younger 1082 ± 16 Ma (U-Pb metamorphic zircon) is consistent with other metamorphic dates in the region. Wodika et 29 Pa le oz oi c ro ck s M at am ec co m pl ex La c- à-l ’A igl e D om ain i M ag pi e co m pl ex Po iss et co m pl ex M an ito u co m pl ex Ba un e ga bb ro To rtu e an o rt ho si te Th om an o rt ho si te H av re -S ai nt -P ie rre an o rt ho si tic su ite 4 A no rth os ite M an ge rit e Fi gu re 2. 4 Si m pl ifi ed m ap o ft he H av re -S ai nt -P ie rre an o rt ho sit ic su ite (af ter G ob ei le ta l. 20 03 ). a) M ap sh ow in g th e di ffe re nt do m ai ns o ft he ar ea an d th e di ffe re nt lo be s o ft he Fl av re -S ai nt -P ie rre an o rt ho sit e. N um be rs in th e le ge nd in di ca te di ffe re nt in tru sio ns . (3a )M ag pi e- W es t; (3b )N or th -W es t; (3c )S he ld ra ke ;( 3d )L ac Be ze l; (3e )R iv iè re -a u- To nn er e ;( 3f) La c A lla rd ;( 3g ) R iv iè re Ro m ai ne . R ef er en ce s fo rt he cr ys ta lli za tio n ag es o ft he m as sif s ar e n u m be re d in br ac ke ts af te rt he ag es o n th e m ap an d ar e as fo llo w s: (1) W od ic ka et al. 20 03 ;( 2) v an B re em en & H ig gi ns 19 90 ; ( 3) Em sli e & H un t 19 90 ;( 4) th is st ud y. b) Lo ca tio n o fB ig Is la nd m as siv e Fe -T io x id e dy ke . St ar s in di ca te po sit io ns o fs am pl es co lle ct ed fo rg eo ch ro no lo gy (+ sa m pl e n u m be rs ). // // /• / La cA lla rd Lc .,,. . H 10 61 M a (4) // / / g Is la nd dy ke . 9 ± 6. 6 M a (4’ M an ge rit e e n v e lo p 1 0 11 29 ±3 M a( 3) 0 20 km c ii La te -to po st- G re nv ill e in tru sio ns Sa in t-J ea n D om ain [Z S] Ca na tic he co m pl ex Fo um ier an o rt ho sit ic su ite K . A no rth os ite G ra ni te ,m an ge rit e N at as qu an D om ain W ak eh am G ro up Bu it Co m pl ex a!. (2003) and Gobeil et al. (2003) integrated more recent mapping (Perreault, 2003) and geochronological results indicating that the Rivière-au-Tonnerre massif (1062 ± 4 Ma: U-Pb zircon; van Breemen & Higgins, 1993) is distinct from the South-West lobe (Lao Brézel). Finally, an age of 1126 +71-6 Ma (U-Pb zircon; Emslie & Hunt, 1990) was determined for the mangeritic envelope of the Lac Allard lobe, and has been interpreted as the crystallization age of this part of the anorthosite suite (Emslie & Hunt, 1990). The magmatic age of the North-West lobe overlaps with those presented by Emslie & Hunt (1990) for the mangeritic envelope, but is 60 to 70 m.y. older than the Rivière-au Tonnerre massif. The mangerite was emplaced before regional metamorphism, however no metamorphic zircons were found in the samples analyzed by Emslie & Hunt (1990). According to Wodika et al. (2003), the Rivière-au-Tonnerre massif is too young to have been affected by regional metamorphism. The Lao Allard lobe has been mapped by Hocq (1982) and the eastern part, which hosts the giant Lao Tio deposit, was further described by Madore et al. (1999). The massif is an andesine anorthosite and blocks of labradorite anorthosite are observed locally in the Big Island deposit. Hocq (1982) estimated high- pressure emplacement conditions (850°C and 7 kbar) for the Lao Allard lobe based on spinel reaction coronae around olivine in contact with plagioclase. 2.5- Method 2.5.1 - Sampling and separation techniques Rutile and apatite grains were separated at the Department of Geology of the Université de Liege (Belgium) by crushing the samples to 60-160 j.tm to liberate the grains and using heavy liquids (bromoform, heated Thalium Clerici solution and Methyl Iodine) and a Frantz Isodynamic Separator following the method outlined in Duchesne (1966). All other mineral treatment protocols and all analyses were carried out at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia, Vancouver, Canada, as described below. Zircon grains were separated from samples using conventional crushing, grinding, and Wilfley table techniques, followed by final concentration using heavy liquids and magnetic separation. Mineral fractions for analysis were selected based on grain morphology, quality, size and low 31 magnetic susceptibility. All reagents used for zircon and accessory mineral processing were sub-boiled or in the case of water, of>18M(cm purity. 2.5.2- Zircon treatment The clearest, crack- and inclusion-free zircon grains available from each sample were hand-picked under magnification in ethanol or methanol. Prior to dissolution, zircons were separated into three groups, including (1) untreated grains (i.e. no physical or chemical abrasion), (2) grains that were physically abraded to minimize the effects of post-crystallization Pb-loss, using the technique of Krogh (1982), and (3) grains that were annealed and chemically abraded (CA) employing procedures slightly modified from those outlined in Mundil et al. (2004) and Mattinson (2005) as reported in detail in Scoates & Friedman (2008). In the case of CA-TIMS pre-treatment, several to approximately 30 grains per sample were selected and treated together in a single beaker. The grains were annealed in quartz glass crucibles in a muffle furnace at 900°C for 48 hours. They were transferred into 10 mE pyrex beakers, ultrasonicated in 3N HNO3 for 15 minutes, warmed to 60 ± 10°C for 30 minutes, rinsed with water followed by acetone and then air-dried. Zircon grains were transferred into 3.5 mL PFA screwtop beakers, HF (50%, 500 jiL) and HNO3 (14 N, 50 giL) were added, and caps were closed finger-tight. The beakers were placed in 125 mL PTFE Teflon® liners (up to four per liner) along with approximately 2 mL HF and 0.2 mL concentrated HNO3.The liners were then slid into stainless steel ParrTM high-pressure dissolution vessels, which were sealed and brought up to 200°C for 16 hours. After cooling, the beakers were removed from the liners and the leached zircon grains were removed from the leachate; they were pipetted into clean 3.5 mL PFA Teflon® beakers, rinsed with water, ultrasonicated and warmed to 60 ± 10°C for 15 and 30 minutes in 6 N HC1. After a final water and acetone rinse and air-dry, the CA pre-treated zircon grains were ready for final dissolution. The grains were transferred to disposable petrie dishes in ethanol for final selection. 2.5.3. Zircon dissolution Untreated, physically and chemically abraded zircons selected for analysis were all dissolved as described below. Grains were weighed and transferred into 300 giL PFA 32 microcapsules into which 50 ILL of 50% HF, 5 ILL of 14 N HNO3 and a weighed small drop of 233235U-05Pbisotopic tracer were added. The microcapsules were placed in 125 mL PTFE liners (8-13 microcapsules per liner, with —2 mL HF/0.2 mL HNO3), inserted and sealed in Parr high-pressure steel jackets and digested for 40 hours at 240°C. After cooling, the microcapsules were removed and the resulting solutions were dried on a hotplate at 130°C. The fluorides were dissolved in 3.1 N HC1 (for ion exchange chemistry) or 6.2 N HC1 (no chemistry) in the microcapsules/125 mL liners/high pressure jackets for 12 hours at 210°C. Separation and purification of Pb and U employed ion exchange column techniques slightly modified from those described by Parrish et al. (1987), in which Pb and U are sequentially eluted into a single beaker. Purified and no chemistry fractions were dried to chlorides in 7 mL PFA beakers after adding 2 jiL of 0.5 NH3P04. 2.5.4- Rutile and apatite treatment/dissolution Hand-picked multi-grain rutile and apatite fractions were transferred into 10 mL pyrex beakers. About 2 mL of 1 N HNO3 (rutile) and water (apatite) was added and grains were ultrasonicated for 5 minutes and warmed to 60 ± 10°C for 10 minutes, rinsed with water followed by acetone and then air-dried. Fractions were weighed and transferred to 3.5 mL screwtop PFA Teflon® beakers. Approximately 1 mL of a 10:1 mixture of 50% HF and 14N HNO3 (rutile) and 6.2 N HCI (apatite) was added followed by a weighed drop of 233235U-0Pbisotopic tracer. The beakers were capped tightly and dissolved on a hotplate at 130°C for a minimum of 48 hours. For rutile, the beakers were uncapped and taken to dryness at 130°C. Approximately 1 mL of 6.2 N HC1 was added; beakers were capped and placed on a hotplate at 130°C for 24 hours. For rutile and apatite, the beakers were again uncapped, dried and 1 mL of 3.1 N HC1 was added and again they were capped and left on a hotplate at 130°C for 24 hours. Anion exchange column procedures were slightly modified from that of zircon, as described below. Twice the volume of the anion exchange resin was used compared to zircon (0.5 vs. 0.25 mL, respectively). U was eluted into a 7 mL PFA beaker with 0.5 N HBr followed by elution of Pb into a separate beaker with 6.2 N HC1. The U was taken nearly to dryness at 110°C, and 1 ml. of 6.2 N HCI was added. Resin in the columns was washed with water and 33 conditioned with 6.2 N HC1, and the U solution was reintroduced into each column and washed with 8 N HNO3 to remove iron. U was eluted with water into the same beaker into which Pb was previously eluted. Samples were dried after addition of 2 tL of 0.5 N H3P04.Samples were then loaded and analyzed in the same manner as described below. 2.5.5- Isotopic ratio and U-Pb concentration analysis All samples, including leachates, were loaded onto single zone-refined Re filaments; 5 .tL of silicic acid activator (prepared with slight modifications from the formulation described in Gerstenberger & Haase, 1997) was pipetted onto small sample droplet in 7 mL PFA Teflon® beaker and this mixture was pipetted directly onto the filament and heated to a faint glow. Isotopic ratios were measured using a modified single collector VG-54R thermal ionization mass spectrometer equipped with a Daly photomultiplier. All isotopes were measured in peak-switching mode on the Daly detector. Procedural blanks for U were in the range of 0.1-1 pg (rutile and zircon) and for Pb were 5-30 pg (rutile) and 1-3 pg (zircon) during the course of this study (2004-2007). U fractionation was determined directly on individual runs using a 233235Utracer, and Pb isotopic ratios were corrected for a fractionation of 0.23% to 0.37%/amu for Daly runs, based on replicate analyses of the NBS-98 1 and NBS-982 Pb reference materials and the values recommended by Thirwall et al. (2000). All analytical errors were numerically propagated through the entire age calculation using the technique of Roddick (1987). Standard concordia diagrams were constructed with Isoplot 3.00 (Ludwig, 2003). Regression intercepts, concordia ages and weighted means calculated using also Isoplot 3.00 (Ludwig, 2003). Unless otherwise noted, all errors are quoted at the 2a level of uncertainty. 2.5.6- 40Ar/39rbiotite and plagioclase Whole rock samples were crushed in an agate conic crusher and were sieved to keep the 60-160 jim and >160 jim cuts. Biotite grains were collected in the >160 jim cut and were handpicked under a binocular microscope. Plagioclase grains were separated using heavy liquids (bromoform) and a Frantz Isodynamic Separator, and handpicked under a binocular microscope. Mineral separates were washed in acetone, dried, wrapped in 34 aluminum foil and stacked in an irradiation capsule with similar-aged samples, neutron flux monitors (Fish Canyon Tuff sanidine, 28.02 Ma, Renne eta!., 1998), optical grade CaF2 and potassium glass. The samples were irradiated in two different groups. The first group was irradiated on October 29 through November 4, 2005 and the second group was irradiated on February 15 through 17, 2006 at the McMaster Nuclear Reactor in Hamilton, Ontario, for 225 MWH, with a neutron flux of approximately 3x10’6 neutrons/cm2.Analyses (n=36) of 12 neutron flux monitor positions produced uncertainty of <0.5% in the J value for the first group and analyses (n=57) of 19 neutron flux monitor positions produced uncertainty of <0.5% in the J value for the second group. The samples from the first group were analyzed on December 31, 2005 though January 2, 2006, and samples from the second group were analyzed on March 17 through 20, 2006. The mineral separates were step-heated at incrementally higher powers in the defocused beam of a lOW CO2 laser (New Wave Research MIR1 0) until fused. The gas evolved from each step was analyzed by a VG5400 mass spectrometer equipped with an ion-counting electron multiplier at PCIGR. All measurements were corrected for total system blank, mass spectrometer sensitivity, mass discrimination, radioactive decay during and subsequent to irradiation, as well as interfering Ar from atmospheric contamination and the irradiation of Ca, Cl and K. Isotope production ratios were: (40Ar/39r)K= .0302 ± 0.00006,(37Ar/9r)ca1416.4 ± 0.5,(36Ar/9r)caO.3952 ± 0.0004, CaJK1.83 ± 0.01(37ArCa/91K). The plateau and correlation ages were calculated using Isoplot 3.09 (Ludwig, 2003). Uncertainty is quoted at the 2a (95% confidence) level and is propagated from all sources, but not for mass spectrometer sensitivity and age of the flux monitor. The best statistically-justified plateau and plateau ages for both samples were picked based on the following criteria: (1) three or more contiguous steps comprising more than 50% of the 39Ar; (2) a probability of fit of the weighted mean age greater than 5%; (3) a slope of the error-weighted line through the plateau ages equals zero at 5% confidence; (4) the ages of the two outermost steps on a plateau are not significantly different from the weighted mean plateau age (at 1 .8a six or more steps only); and (5) the outermost two steps on 35 either side of a plateau must not have non-zero slopes with the same sign (at 1 .8G nine or more steps only). 2.6- Results 2.6.1- U-Pb zircon All U-Pb results for zircon are presented in Table 2.2 and concordia diagrams for each sample are shown in Figure 2.5. In Figure 2.5, all ages are presented with 2a uncertainty calculated without and with decay-constant errors, however, within the text, the 2a uncertainty stated does not include the decay-constant errors. Anorthosite, wall of Bignell deposit, Saint-Urbain (sample 2006-C2) Sample 2006-C2 is an anorthosite collected 2 metres from the contact with the Bignell Fe-Ti oxide deposit (Figure 2.2). The picked zircon grains are separated into three fractions. Grains in the non-abraded fraction NA1 were elongate, clear, transparent and relatively large (60 x 100 jim to 170 x 300 jim), and some grains contained small fluid inclusions. Zircon grains in the abraded fractions Al and A2 were rounded, some were anhedral, clear, transparent, and of variable size (140 x 210 to 71 x 128 jim). The U-Pb data from all three zircon fractions are concordant and overlapping (0.0-0.5% discordance) with 207Pb/6 ages ranging from 1052.8 to 1058.0 Ma. The calculated concordia age, based on the uranogenic Pb*/U and Pb*IPb* isotopic ratios (Ludwig, 1998), on all three fractions is 1053.6 ± 2.6 Ma, which is interpreted as the crystallization age of the Saint-Urbain anorthosite near the Bignell deposit. Leuconorite, west-central area, Saint-Urbain (sample 2042-A) Sample 2042-A is a megacrystic leuconorite with cm-size plagioclase and orthopyroxene. Zircon crystals in this sample were anhedral, clear, transparent and varied in size from 60 x 100 jim to 100 x 250 jim. Five fractions were analyzed, of which two were non-abraded (NA1 and NA2), two were mechanically abraded (Al and A2), and one was chemically abraded (CA2). The data from all fractions are concordant (-0.3 to 0.5% discordance), except for NA1 which is slightly discordant (1.8%), and span a range 36 Ta bl e 2. 2 Zi rc on U -P b TI M S an al yt ic al da ta Fr ac tio ns W t U” P b 2°6P b ’ Pb ° Tl iJ U 25 4 (m g) (pp m) (pp m) Pb (pg ) A no rth os ite ,w all o fB ig ne ll de po sit s, Sa in t-U rb ain (sa mp le 20 06 -C 2) N A I, 1 4 15 8.3 29 .3 36 29 1.9 0. 45 0. 17 74 6 ± 0. 18 1. 82 09 ± 0. 34 0. 07 44 2 ± 0. 27 A l, 1 4 73 .1 13 .8 71 1 4. 6 0.5 3 0. 17 79 4 ± 0. 34 1. 82 97 ± 0. 60 0. 07 45 8 ± 0. 45 A2 ,1 4 69 .0 13 .3 14 59 2.1 0. 62 0. 17 74 4 ± 0. 29 1. 82 53 ± 0. 54 0.0 74 61 ± 0.4 8 Le uc on or ite ,w es t-c en tra la re a, Sa in t-U rb ain (sa mp le 20 42 -A ) N A I, 2 6 32 .4 6.4 12 51 1.7 0. 88 N A 2, 2 6 52 .9 10 .2 12 90 2. 7 0. 67 A l, 1 16 72 .1 13 .9 56 50 2. 2 0.6 5 A2 ,1 5 67 .8 12 .8 14 82 2.5 0. 60 CA 2, 1 1.7 48 .7 9.2 74 8 1.2 0. 60 A no rth os ite ,L ac de sC yg ne s, Sa in t-U rb ain (sa mp le 20 43 ) N A I, 2 4 58 .8 10 .4 14 14 1.8 A l, 1 9 99 .5 18 .0 36 14 2. 7 A2 ,1 5 86 .1 15 .4 22 35 2.1 A3 ,1 6 89 .5 16 .6 94 6 6.4 A 4, 2 6 96 .7 18 .2 12 25 5.2 O xi de -a pa tit e ga bb ro no rit e, n o rth w es tc o n ta ct , Sa in t-U rb ain (sa mp le 20 20 ) N A I, 1 7 57 .5 10 .9 16 12 2. 8 0.5 3 N A 2, 1 11 57 .9 12 .4 32 52 2. 2 1.0 9 CA 2, I 17 .3 31 .4 5.9 75 63 0.8 0. 50 CA 4 19 .1 35 .2 6.8 76 40 1.0 0. 58 CA 5 11 .0 54 .7 10 .6 84 99 0.8 0. 62 Sé m in ai re de po sit , S ai nt -U rb ain (sa mp le 20 33 -D ) N A I, - ‘ 30 71 1.1 0.9 83 12 .4 15 .3 0. 16 73 4 ± 0.7 3 1. 64 30 ± 2. 49 0.0 71 21 ± 2.2 1 A l, — 40 11 3 1.0 1.1 62 27 .2 19 .7 0. 16 59 9 ± 0. 48 1. 64 47 ± 1.7 5 0. 07 18 6 ± 1.5 5 A2 ,- ‘ 40 11 8 1.1 1.0 12 7 11 .8 17 .5 0. 16 01 4 ± 0. 72 1. 55 64 ±3 .2 6 0. 07 04 9 ± 2. 97 Sa in t-A nn e du N or d o rt ho py ro xe ne -g ab br on or ite (sa mp le 20 23 ) N A I, 1 7 13 7.2 25 .8 44 10 2. 4 0. 54 N A 2, 1 5 13 1.8 23 .6 29 89 2. 4 0.3 5 A l, I 6 13 2.3 24 .6 42 28 2.1 0. 48 C A l, 1 23 .9 16 3.5 30 .4 31 99 0 1.3 0. 46 CA 2, 1 7.5 17 3.4 31 .4 20 11 0 0.7 0.3 5 Bi g Is la nd de po sit , L ac A lla rd ,H av re -S ai nt -P ie rre (sa mp le 21 02 ) N A I, — 80 46 4 10 .3 2. 0 12 23 0 4. 2 0.8 1 0. 17 37 3 ± 0.2 1 1. 76 89 ± 0.2 3 0. 07 38 4 ± 0. 14 N A 2, — 70 34 5 4. 8 1.0 68 58 2. 6 0. 88 0. 17 37 8 ± 0.1 3 1. 76 77 ± 0. 20 0. 07 37 8 ± 0. 11 N A 3, - ‘ 50 29 2 8.0 1.6 27 49 9.2 0. 97 0. 17 37 2 ± 0. 12 1. 76 67 ± 0. 20 0. 07 37 6 ± 0.1 1 A l, — 50 22 3 10 .1 2. 0 10 81 22 .6 0. 76 0. 17 17 3 ± 0.1 1 1. 73 89 ± 0. 21 0. 07 34 4± 0. 13 98 6.9 ± 31 .4 96 3. 4 ± 87 .8 /9 3 0. 50 93 8 98 7.5 ± 22 .2 98 1. 9 ± 62 .0 /6 4. 5 0. 53 62 9 95 3.1 ± 40 .3 94 2. 6 ± 11 7/ 12 6 0. 48 88 3 10 34 .1 ± 2. 9 10 37 .2 ± 5.8 /5 .8 0. 79 00 0 10 33 .7 ± 2. 6 10 35 .3 ± 4. 4/ 4. 4 0.8 61 12 10 33 .3 ± 2.5 10 34 .8 ± 4. 3/ 4. 3 0. 88 85 4 10 23 .1 ± 2. 7 10 26 .1 ± 5.4 /5 .4 0. 84 43 9 0.6 0.9 0.3 - 0.2 - 0.3 0.8 - 0.1 0. 4 1.0 0.5 Is ot op ic ra tio s ± ls ,% 5 A pp ar en t a ge s± 2s , M a g rT’ % 2 5 6 P b / 3 5 U 2 5 7 P b / 3 5 u 2 0 7 P b /6b 2 0 6 P b / 3 5 U 2 5 7 P b / 3 5 U 2 5 7 P b /4b di sc or da nc e 0.1 70 51 ± 0. 32 1. 73 19 ± 0. 58 0. 07 36 7 ± 0. 44 0. 17 55 6 ± 0. 26 1. 79 44 ±0 .5 1 0. 07 41 3 ± 0. 42 0. 17 65 5 ± 0. 12 1. 80 47 ±0 .2 1 0. 07 41 4 ± 0. 14 01 74 23 ± 0. 24 1. 77 64 ±0 .9 2 0. 07 39 5 ± 0.8 3 0. 17 55 2 ± 0. 27 1. 79 06 ± 0.8 3 0. 07 39 9 ± 0. 74 0. 40 0. 17 19 4 ± 0. 28 1. 74 60 ± 0.5 3 0. 07 36 5 ± 0.4 1 0. 37 0. 17 76 4 ± 0. 16 1. 81 98 ± 0. 30 0. 07 43 0 ± 0. 24 0. 36 0. 17 58 5 ± 0. 17 1. 79 84 ± 0.2 7 0. 07 41 7 ± 0. 18 0. 44 0. 17 81 6 ± 0.2 1 1. 83 35 ± 0.3 5 0. 07 46 4 ± 0. 24 0.5 3 0. 17 67 5 ± 0. 17 1. 81 75 ± 0. 32 0. 07 45 8 ± 0. 24 0. 17 74 8 ± 0. 23 1. 82 65 ± 0. 40 0. 07 46 4 ± 0. 34 0. 17 69 0 ± 0. 17 1. 82 06 ± 0. 29 0. 07 46 4 ± 0. 21 0. 17 80 9 ± 0.1 3 1. 83 37 ± 0.1 8 0. 07 46 8 ± 0. 13 0. 17 83 5 ± 0.1 5 1. 83 29 ± 0. 19 0. 07 45 4 ± 0. 15 0. 17 84 0 ± 0.1 8 1. 83 24 ±0 .4 4 0. 07 45 0 ± 0. 37 10 53 .1 ± 3. 5 10 53 .0 ± 4. 5 10 52 .8 ± 10 .8 /1 0. 9 0. 61 82 3 0. 0 10 55 .7 ± 6. 6 10 56 .2 ± 7. 8 10 57 .1 ± 18 .1/ 18 .3 0. 65 69 8 0.1 10 53 .0 ± 5.7 10 54 .6 ± 7.1 10 58 .0 ± 19 .0/ 19 .3 0. 48 00 6 0.5 10 14 .9 ± 6. 0 10 20 .5 ± 7. 4 10 32 .4 ± 17 .8/ 18 0. 64 82 5 1.8 10 42 .7 * 5. 1 10 43 .4 ± 6.7 10 45 .0 ± 16 .6/ 16 .8 0. 59 40 2 0. 2 10 48 .1 ± 2. 4 10 47 .2 ± 2. 8 10 45 .2 ± 5.6 /5 .6 0. 79 05 8 - 0.3 10 35 .4 ± 4. 6 10 36 .9 ± 12 .0 10 40 .1 ± 33 .0 /3 3. 8 0. 50 72 2 0.5 10 42 .4 ± 5. 1 10 42 .0 ± 10 .9 10 41 .2 ± 29 .7 /3 0. 3 0. 48 09 4 - 0.1 10 22 .8 ± 5. 3 10 25 .7 ± 6.8 10 31 .8 ± 16 .4/ 16 .6 0. 63 60 7 0.9 10 54 .1 ± 3. 1 10 52 .6 ± 4. 0 10 49 .5 ± 9. 6/ 9. 7 0. 62 10 7 - 0.5 10 44 .3 ± 3. 2 10 44 .9 ± 3.5 10 46 .1 ± 7. 0/ 7.1 0. 78 09 5 0.2 10 56 .9 ± 4. 1 10 57 .5 ± 4. 6 10 58 .7 ± 9. 5/ 9. 6 0. 74 57 7 0. 2 10 49 .2 ± 3. 3 10 51 .8 ± 4. 2 10 57 .1 ± 9. 5/ 9. 5 0. 70 43 7 0.8 10 53 .2 ± 4. 4 10 55 .0 ± 5.3 10 58 .9 ± 13 .4 /1 3. 6 0. 54 51 0 10 50 .0 ± 3.3 10 52 .9 ± 3.8 10 58 .8 ± 8. 4/ 8. 4 0. 70 09 9 10 56 .5 ± 2. 5 10 57 .6 ± 2. 3 10 59 .8 ± 5. 2/ 5. 2 0. 68 32 5 10 58 .0 ± 2. 9 10 57 .3 ± 2.5 10 56 .0 ± 6. 0/ 6. 0 0. 62 66 5 10 58 .2 ± 3. 6 10 57 .1 ± 5.8 10 54 .9 * 15 .0/ 15 .1 0. 53 45 9 99 7.5 ± 13 .4 99 0. 0 ± 8.7 95 7.6 ± 12 .9 10 46 .1 ± 3. 4 10 48 .7 ± 3.4 10 54 .0 ± 6.5 /6 .5 0. 79 83 7 10 44 .8 ± 2. 8 10 44 .4 ± 3.5 10 43 .7 ± 7. 7/ 7. 7 0.7 31 91 10 49 .4 ± 2. 7 10 50 .7 ± 2. 9 10 53 .2 ± 5.9 /5 .9 0. 76 31 8 10 55 .5 * 2. 3 10 58 .6 ± 3. 0 10 65 .0 ± 6. 9/ 6. 9 0. 67 48 2 10 55 .6 ± 3, 7 10 57 .3 ± 2. 8 10 60 .8 ± 2. 8/ 2. 8 0. 94 24 6 10 32 .7 ± 4. 1 10 32 .9 ± 2. 4 10 32 .6 * 2. 2 10 21 .6 ± 2. 0 0. 17 61 9 ± 0. 18 1. 80 89 ± 0. 26 0. 07 44 6 ± 0. 16 0. 17 59 5 ± 0. 15 1. 79 72 * 0. 27 0. 07 40 8 ± 0. 19 0. 17 68 0 * 0. 14 1. 81 44 ±0 .2 2 0. 07 44 3 ± 0.1 5 0. 17 78 9 ± 0. 12 1. 83 64 ± 0.2 3 0. 07 48 7 ± 0. 17 0.1 77 91 ± 0. 19 1. 83 28 ± 0.2 1 0. 07 47 2 ± 0. 07 - 3.8 - 0. 9 - 1. 7 0.5 0.3 0.2 0.5 Ta bl e 2. 2 (co nti nu ed ) Fr ac tio n’ W I Ub Pb ” 2 0 6 P b ’ Pb ’ l’ h/ lf Js ot op jcr ati os ±l s,% 5 A pp ar en ta ge s± 2s ,M a 5 r’ 0 /, (m g) (pp m) (pp m) ° P b (pg ) 2 0 6 P b / 3 8 U 2 5 7 P b / 3 5 U 2 0 7 P b /6b 2 0 6 P b ? 3 8 U 2 0 7 P b / 3 5 U 2 0 7 P b /6b di sc or da nc e C A l, 1 11 .9 2. 2 0. 9 31 3 1.0 5.0 6 0. 17 77 5 ± 0. 51 1. 84 58 ± 1.6 9 0. 07 53 2 ± 1.5 4 10 54 .7 ± 10 .0 10 61 .9 ± 22 .3 10 76 .9 ± 60 .5 /6 2. 9 0. 43 88 7 2. 2 CA 2, 1 9.6 4. 9 1.1 67 0 0.8 1.2 6 0. 17 71 3 ± 0. 46 1. 80 75 ± 2. 35 0. 07 40 1 ± 2. 13 10 51 .3 ± 9. 0 10 48 .2 ± 30 .8 10 41 .7 ± 83 .7 /8 8. 5 0. 55 20 1 - 1. 0 A no rth os ite ,L ac A lla rd ,H av re -S ai nt -P ie rre (sa m pl e 21 14 -B ) N A Il 10 32 .6 6.0 18 77 1.9 0. 49 0. 17 54 0 ± 0. 22 1. 80 22 ± 0. 39 0. 07 45 2 ± 0. 3 10 41 .8 ± 4. 2 10 46 .3 ± 5.1 10 55 .5 ± 12 .1 /1 2.1 0. 63 43 7 1.4 N A 2, I 11 18 .6 3.4 94 3 2. 4 0.5 5 0. 17 26 8 * 0. 33 1. 75 09 ± 0. 78 0. 07 35 4 ± 0. 66 10 26 .8 ± 6. 4 10 27 .5 ± 10 .1 10 28 .8 ± 26 .4 /2 6. 9 0. 55 74 4 0.2 A l, 1 12 10 .8 3.9 69 2 2.1 4. 07 0. 17 97 5 ± 0. 43 1. 85 59 ± 0. 98 0. 07 48 8 ± 0. 83 10 65 .6 ± 8. 5 10 65 .5 ± 13 .0 10 65 .4 ± 33 .1 /3 3. 9 0. 54 34 3 0.0 A2 ,1 12 12 .4 4. 6 66 4 2. 6 4. 28 0. 18 00 6 ± 0. 48 1. 85 77 ± 1.2 8 0. 07 48 3 ± 1.1 3 10 67 .3 ± 9. 4 10 66 .2 ± 16 .9 10 63 .9 ± 44 .7 /4 6. 0 0. 49 18 7 - 0.3 CA l, 1 7.5 95 .4 16 .5 37 77 2. 0 0.3 4 0. 17 08 2 ± 0. 14 1. 75 95 ± 0.2 8 0. 07 47 0 * 0.2 3 10 16 .6 ± 2. 6 10 30 .7 ± 3.6 10 60 .5 ± 9. 0/ 9. 1 0. 60 49 5 4. 5 CA 2, 1 9.7 22 .9 4. 5 23 84 1.0 0.7 2 0. 17 87 6 ± 0. 18 1. 84 13 ± 0. 36 0. 07 47 1 ± 0. 29 10 60 .2 ± 3. 5 10 60 .3 ± 4. 8 10 60 .7 ± 11 .8 /1 1. 9 0. 59 32 9 0. 0 A no rth os ite ,L ac A lla rd ,H av re -S ai nl -P ie rre (sa mp le 21 23 -B ) N A I,— 35 26 17 8.5 36 .0 12 14 0 4.1 0. 89 0. 17 43 5 ± 0. 11 1. 78 70 ± 0. 16 0. 07 43 4 ± 0. 08 10 36 .0 ± 2. 1 10 40 .7 ± 2.1 10 50 .6 ± 3. 2/ 3. 2 0. 89 40 2 1.5 N A2 , — 35 95 12 3.9 25 .8 22 39 0 5.7 0.9 9 0. 17 56 6 ± 0. 12 1. 80 30 ± 0. 16 0. 07 44 4 ± 0. 08 10 43 .2 ± 2. 3 10 46 .5 ± 2.1 10 53 .4 ± 3. 1/ 3. 1 0.8 91 51 1.0 A l, 10 67 24 2. 3 50 .6 32 24 0 5.5 1.0 0 0. 17 60 8 ± 0. 10 1. 80 64 ± 0. 15 0. 07 44 1 ± 0. 08 10 45 .5 ± 1.9 10 47 .8 ± 1.9 10 52 .5 ± 3. 0/ 3. 0 0. 88 26 0 0.7 A2 ,1 5 33 25 8. 2 55 .0 18 22 0 5.1 1.1 0 0. 17 55 5 ± 0. 10 1. 80 12 ± 0. 15 0. 07 44 2 ± 0. 07 10 42 .6 ± 2. 0 10 45 .9 ± 2. 0 10 52 .7 ± 2. 9/ 2. 9 0. 90 46 5 1.0 L 10 34 5.1 71 .9 65 0 58 .8 1.0 3 0. 17 45 7 ± 0. 15 1. 78 74 ± 0. 25 0. 07 42 6 ± 0. 17 10 37 .2 ± 2. 8 10 40 .9 ± 3.3 10 48 .5 ± 6. 7/ 6. 8 0. 77 02 3 1.2 CA 3- l, 1 8 69 .4 14 .6 28 81 2.1 1.0 3 0. 17 59 9 ± 0. 20 1. 80 05 ± 0.4 1 0. 07 42 0 ± 0.3 5 10 45 .0 ± 3. 8 10 45 .6 ± 5.4 10 46 .9 ± 14 .1 /1 4. 2 0.5 29 81 0. 2 CA 3- 2, 5 8 38 6.5 81 .3 16 33 2. 6 1.0 1 0. 17 67 6 ± 0. 19 1. 82 37 ± 0. 5 0. 07 48 3 ± 0.4 3 10 49 .3 ± 3. 7 10 54 .0 ± 6.6 10 63 .9 ± 17 .2 /1 7. 4 0. 53 46 2 1.5 CA 4, 1 2.1 12 6.4 28 .2 22 92 1.3 1.2 6 0. 17 77 5 ± 0. 16 1. 83 30 ± 0. 53 0. 07 48 0 ± 0. 46 10 54 .7 ± 3. 1 10 57 .4 ± 6.9 10 63 .0 ± 18 .5/ 18 .7 0. 53 59 2 0.8 A no rth so ite ,L ac A lla rd ,H av re -S ai nt -P ie rre (sa mp le 21 32 ) N i, 1 5 80 .1 19 .6 26 99 1.7 1.7 0 0. 17 85 0 ± 0. 18 1. 84 10 ± 0. 27 0. 07 48 0 ± 0. 19 10 58 .8 ± 3. 6 10 60 .2 ± 3.5 10 63 .2 ± 7. 6/ 7. 6 0. 71 07 7 0.4 N A2 , 1 4 10 1.5 20 .1 44 3 10 .5 0.7 7 0. 17 62 4 ± 0. 19 1. 82 92 ± 1.4 9 0. 07 52 8 ± 1.3 9 10 46 .4 ± 3. 6 10 56 .0 ± 19 .6 10 75 .8 ± 54 .9 /5 6. 9 0. 57 64 5 3.0 A l, 1 5 21 7. 5 50 .2 67 03 1.8 1.3 8 0. 17 90 8 ± 0. 08 1. 84 45 ± 0. 15 0. 07 47 0 ± 0. 09 10 62 .0 ± 1.5 10 61 .5 ± 2. 0 10 60 .4 ± 3. 4/ 3. 4 0. 90 68 9 - 0.2 A2 ,1 20 52 .7 9.3 47 65 2.5 0.2 6 0. 17 86 0 ± 0. 08 1. 83 90 ± 0. 15 0. 07 46 8 ± 0. 09 10 59 .3 ± 1.5 10 59 .5 ± 2. 0 10 59 .9 ± 3. 5/ 3.5 0. 90 70 9 0.1 Fr ac tio n ID fo llo w ed by th e n u m be ro fg ra in s; A - ai ra br ad ed ;N A - u n a br ad ed ;C A - c he m ic al ly a br ad ed b U c o rr e c te d fo r b la nk o f 1 p g ± 20 % ,a n d fr ac tio na tio n, m e a su re d fo re a c h ru n w ith a d o u b le 2 3 3 2 3 5 U sp ik e R ad io ge ni c Pb dM es au re d ra tio c o rr e c te d fo rs pi ke an d Pb fr ac tio na tio n o f0 .2 3- 0. 37 % /a m u ± 20 % (D aly c o lle ct or )w hi ch w as de te rm in ed by re pl ic at e an al ys is o fN B S Pb SR M 98 1 o r SR M 98 2 st an da rd m at er ia l ‘ To ta lc o m m o n Pb in a n a ly si s ba se d o n bl an k is ot op ic c o m po si tio n o f Z t P b / 2t4P b l 8 . 5 ± 3% ,2°7P b / 0 b = 1 5 .5 ± 3% a n d 2 0 8 P b / 4b ± 0. 5% M od el Th /IJ c a lc ul at ed fr om ra di og en ic 2 0 8 P b an d th e 2 0 7 P b / 6b ag e o ff ra ct io n g C or re ct ed fo rf ra ct io na tio n, 1 pg U (w he re a pp lic ab le ), 1- 2 pg o fb la nk Pb an d c o m m o n Pb c o m po si tio ns m e a su re d fr om a ss o c ia te d pl ag io cl as e o r c a lc ul at ed fr om a ss o c ia te d w ho le ro c k c o m po si tio ns (se e t ab le 5. 7 an d 5. 9 in C ha pt er 5 fo rc o m m o n Pb c o m po si tio ns ) c o rr e la tio n c o e ff ic ie nt % di sc or da nc e 00 DC 0 Bignell anorthosite 2006-C2 ‘ 0.179 10651053.6 ± 2.6/2.8 Ma 0.177MSWD 0 048 0.175 Ne” 0.169 0.173 0.171 1045 0.167 — 1.79 1.81 1.83 1.85 1.87 1.68 suuconjo:2çL 1.72 1.76 1.80 1.84 0. iou 0.179 0.178 0.177 0.176 0.181 0.179 0.177 0.175 0.173 0.171 0.16 0.174 0.170 ? 0.166 p0.162 0.158 0.154 S-U OAGN 2020 1057.4 ± 1.5/1.7 Ma MSWD = 0.03 LDC anorthosite 2043-A 10 Concordia 1055.0 ± 2.4/2.6 Ma MSWD = 0.07 Al - A3 A4 1040- A2 1030 NA1 Upper intercept 1057 ± 21 Ma MSWD = 1.06 0.180 0.179 0.178 0.177 0.176 1065 N 1.70 1.74 1.78 1.82 1.86 Séminaire 2033-D ±7fl.3Ma 1.80 1.81 1.82 1.83 1.84 1.85 1.86 0.181 fSAN opx granodiorite 2023 1070 1060.8 ± 2.8 Ma - 0.179 1060 o5o2 0.173 207Pb/35U Figure 2.5... 1.4 1.5 1.6 1.7 1.8 1.75 1.77 1.79 1.81 1.83 1.85 1.87 207Pb/35U 39 0.177 0.175 0.173 0.171 0.169 1.68 1.76 106 207Pb/35U 207Pb/35U Figure 2,5 Concordia diagrams for U-Pb data from analyzed zircon from Saint-Urbain (a-f) and Big Island (g-j). Each ellipse represents the result ofthe analysis of a single fraction, as identified in Table 2.2 (e.g. Al, NAI), and corresponds to the associated 2 uncertainties. The white ellipses indicate concordia ages. For each sample, the 2 uncertainty is indicated first without including the decay-constant errors followed by the 2 uncertainty including the decay-constant errors. (a) Sample 2006-C2, Saint-Urbain anorthosite (SU) at one metre from the contact with the Bignell deposit; (b) sample 2042-A, megacrystic leuconorite from the Saint-Urbain anorthosite; (c) sample 2043-A, recrystallized anorthosite from Monts du Lac des Cygnes (LCD) of the Saint-Urbain anorthosite; (d) sample 2020, oxide-apatite gabbronorite (OAGN) from the northwest border of the Saint-Urbain anorthosite; (e) sample 2033-D, Séminaire deposit; (f) sample 2023, orthopyroxene granodiorite from the Saint-Anne du Nord intrusion (SAN); (g) sample 2102, Big Island dyke; (h) sample 2114-B, Lac Allard anorthosite, Havre-Saint-Pierre anorthosite suite (HSP), 500 metres south of the dyke; (i) sample 2123-B, Lac Allard anorthosite, Havre-Saint-Pierre anorthosite suite, 10 metres south of the dyke; and (j) sample 2132, Lac Allard anorthosite, Havre-Saint-Pierre anorthosite suite, at the site of the Lac Allard mine. See text for complete descriptions ofthe samples. The black band is the concordia curve including decay constant errors. 0.181 0.179 1070 i’ 0.183 0.181 0.179 Big Island dyke 2102 1052.9 ± 6.5/6.6 Ma MSWD = 0.24 101” HSP anorthosite 1km from dyke 2114-B 1061.6 ± 3.0/3.2 MSWD = 0.001 L 0.177 0.175 0.173 0.171 1.84 1.92 1.68 1.76 1.84 1.92 HSP anorthosite lÔm from dyke 2123-B Upper intercept 1057.4 ± 5.7/8.4 fr MSWD = 0.62 1r” CA3-1 ‘0 0 0.179 0.177 0.175 0.173 Al 1.76 1.78 1.80 1.82 1.84 1.86 1.74 1.78 1.82 1.86 1.90 40 of207Pb/6 ages from 1032.4 to 1058.0 Ma. A concordia age of 1046.2 ± 3.1 Ma, based on the U-Pb results calculated with fractions Al, NA2 and CA2, is interpreted as the crystallization age of the leuconorite. The younger207Pb/6 and Pb/U ages and slightly discordant behaviour of fractions A2 and NA1 are consistent with Pb-loss from these zircon grains. Anorthosite, Lac des Cygnes, Saint-Urbain (sample 2043-A) Sample 2043-A is a fine-grained pink recrystallized anorthosite with relic patches of unrecrystallized blue anorthosite from the Lac des Cygnes area (Figure 2.3c). Zircon grains extracted from this sample were rounded fragments of transparent, slightly yellowish crystals that varied in size from 50 x 100 jim to 140 x 250 jim. The zircon grains were divided in one non-abraded fraction (NA 1) and four abraded fractions (Al, A2, A3 and A4). The U-Pb results for all fractions are concordant (-0.5 to 0.9% discordance), but the 207Pb/6 ages vary from 1031.8 to 1058.7 Ma (Figure 2.5c). The concordia age, based on data from fractions Al and A3, is 1055.0 ± 2.4 Ma, which is interpreted as the crystallization age. The younger 207Pb/6 and Pb/U ages for fractions A2, A4 and NA1 are attributed to Pb-loss in these zircon grains. Oxide-apatite gabbronorite, northwest contact, Saint-Urbain (sample 2020) Sample 2020 is an oxide-apatite gabbronorite (OAGN) that outcrops at the northeast border of the anorthosite massif (Figure 2.2b). This lithology is composed primarily of Fe-Ti oxide minerals and pyroxene with plagioclase-rich horizons (Figure 2.3d). Two non-abraded fractions (NA1 and NA2) and three chemically abraded fractions (CA2, CA4 and CA5) of zircon were analyzed. The Al fraction contained large anhedral grains of zircon (40 x 340 jim to 300 x 300 jim), which were clear and transparent. The NA2, CA2, CA4 and CA5 fractions contained anhedral zircon grains that were clear and transparent and varied in size from 100 x 250 to 100 x 500 jim. The U-Pb data of all fractions are concordant (-0.3 to 0.9% discordance) with 207Pb/6 ages ranging from 1054.9 to 1059.8 Ma. The 207Pb/6 and Pb/U ages from the two non-abraded fractions are younger than the others suggesting that zircon in these fractions lost Pb subsequent to crystallization. The concordia age of 1057.4 ± 1.5 Ma, calculated from the data of the 41 three overlapping chemically abraded fractions, is interpreted as the crystallization age of the OAGN. Séminaire deposit, Saint-Urbain (sample 2033-D) Sample 2033-D is an ilmenite-rutile-rich anorthosite (Figure 2.3c) from the Séminaire deposit in the Saint-Urbain anorthosite. Zircon grains from this sample were anhedral, clear, transparent and free of inclusions, and the grains were separated into three fractions. Fraction Al contained the coarser grain fragments measuring up to 225 x 280 jim and fraction A2 contained smaller fragments (75 x 150 jim) before they were mechanically abraded. The non-abraded fraction NA1 contained fragments up to 110 x 190 jim. The U concentrations in the three fractions of zircon from this sample are extremely low (l ppm), which resulted in low 206Pb/4 (<130), as well as very high Th/U ratios (>15) (Table 2.2). The chemistry of these zircon grains is representative of zircon found as rims on hemo-ilmenite and likely formed by diffusion of Zr from hemo-ilmenite and reaction along grain boundaries (Chapter 3). The U-Pb data from all fractions are concordant (-3.8 to -0.8%) with 207Pb/6 ages ranging from 942.6 to 981.9 Ma, distinctly younger that ages determined from the other Saint-Urbain samples. The concordia age, calculated from data of fractions Al and NA1, is 992 ± 7.2 Ma. This is interpreted to be the formation age of the zircon rims, thus this is a cooling age not an igneous crystallization age. The younger 207Pb/6 and Pb/U age given by the results from fraction A2 is considered to represent Pb-loss from the zircon grains. Saint-Anne dii Nord orthopyroxene granodiorite (sample 2023) Sample 2023 is a foliated orthopyroxene granodiorite from the large intrusion that borders the Saint-Urbain massif to the west (Figure 2.2), and it is composed of coarse K feldspar and plagioclase megacrysts, orthopyroxene, clinopyroxene, green homblende, and minor magnetite, ilmenite and quartz (Figure 2.3e). The zircon grains from this sample were elongated, clear, and transparent, and varied in size from 150 x 400 jim to 60 x 200 jim. Five fractions were analyzed, two non-abraded (NA1 and NA2), one physically abraded (Al) and two chemically abraded (CAl and CA2). On Figure 2.5f, 42 the U-Pb data from all fractions are concordant to slightly discordant (-0.3 to 0.9%) and give 207Pb/6 ages from 1054.9 to 1059.8 Ma. The 207Pb/6 and Pb/U ages become progressively younger from the chemically abraded fractions to the abraded fractions to the non-abraded fractions, which is consistent with Pb-loss from these grains. Data from fraction CA2 is concordant and yields a 207Pb/6 age of 1060.8 ± 2.8 Ma, which we interpret as a minimum age for the crystallization of the orthopyroxene granodiorite. Big Island deposit, Lac Allard lobe, Havre-Saint-Pierre (sample 2102) Sample 2102 is a massive ilmenitite from the Big Island dyke in the Lac Allard anorthosite (Figures 2.4 and 2.30 and contained interstitial, clear and transparent zircon grains that varied in size from 100 x 200 gm to 400 x 400 gm. In Figure 2.5g, the U-Pb data from six fractions are concordant (-1 to 0.5% discordance, except for fraction CAl, which is 2.2% discordant). Curiously, the results from the physically abraded fraction (Al) yield a younger 207Pb/6 age of 1023.1 Ma compared to the non-abraded fractions (NA1, NA2 and NA3) (1034.8-1037.2 Ma), which might imply that fraction Al represents zircon from subsolidus rim material (see Chapter 3) or a mixed population of rim and magmatic zircon grains. U-Pb results for the two chemically abraded fractions (CAl and CA2) yield older 207Pb/6 ages, from 1041.7 to 1076.9 Ma, with relatively large associated uncertainty due to the small sample size. The concordia age of 1052.9 ± 6.5 Ma, calculated from concordant and overlapping results for CAl and CA2 fractions, is interpreted as the crystallization age of the dyke. Zircons grains from fractions NA 1, NA2 and NA3 have undergone Pb-loss. Anorthosite, Lac Allard lobe, Havre-Saint-Pierre (sample 2114-B) Sample 2114-B is located 500 metres south of the Big Island dyke in the Lac Allard anorthosite (Figure 2.4). The sample contains anti-perthitic and recrystallized plagioclase and orthopyroxene megacrysts with minor ilmenite. Zircon grains extracted from this sample were clear and transparent, anhedral, and varied in size from 145 x 215 jim to 285 x 430 jim, and the picked grains were split into six fractions. The U-Pb data from one non-abraded fraction (NA2), the two mechanically abraded fractions (Al and A2) and the chemically abraded fraction (CA2) are concordant, whereas the non-abraded 43 fraction NA1 and chemically abraded fraction CAl are discordant (1.4-4.5% discordance). The results from all analyzed fractions span a range of207Pb/6 ages from 1028.8 to 1065.4 Ma. The concordia age, calculated from concordant and overlapping U-Pb results for fractions Al, A2 and CA2, is 1061.3 ± 3.0 (2a) Ma and is interpreted as the crystallization age of this anorthosite. The zircon grains of fractions NA1, NA2 and CAl experienced some post-crystallization Pb-loss. Anorthosite, Lac Allard lobe, Havre-Saint-Pierre (sample 2123-B) Sample 2123-B is a leucogabbronorite located about 20 metres south of the Big Island dyke (Figure 2.4). Plagioclase in the sample shows 120° triple junctions, which are interpreted as a high-temperature recrystallization texture. Anhedral zircon fragments from the sample were clear and transparent, without inclusions, and varied in size from 25 x 25 .tm to 50 x 200 j.m. The zircon grains were separated into seven fractions, including two non-abraded (NA1 and NA2), two physically abraded (Al and A2), and three chemically abraded (CA3- 1, CA3-2, and CA4) fractions, as well as the leachate from the chemical abrasion process. The U-Pb data are concordant to slightly discordant (1.5 to 0.2% discordance) and yield 207Pb/6 ages from 1046.9 to 1063.9 Ma. An upper intercept age of 1057.4 ± 5.7 Ma is considered to be the minimum crystallization age for this anorthosite. Anorthosite, Lac Allard lobe, Havre-Saint-Pierre (sample 2132) Sample 2132 is an anorthosite sample from the wall rock of the Lac Tio magmatic ilmenite deposit. The anorthosite is pinkish-blue in colour with 10 cm-sized plagioclase crystals in a matrix of fine-grained plagioclase. Some cm-long orthopyroxene crystals are also present. The zircon grains extracted form this sample were clear and transparent anhedral fragments, varying in size from 90 x 100 jim to 111 x 240 jim. U-Pb data from the mechanically abraded (Al, A2) fractions and one non-abraded (NA 1) fraction are concordant and the results from the non-abraded (NA2) fraction are discordant (3.0%). The 207Pb/6 age results from these fractions span a range from 1059.9 to 1075.8 Ma. A concordia age calculated with the U-Pb data of fractions NA1, Al and A2 is 1060.5 ± 1.9 Ma. The upper intercept of a regression using the data from all fractions yields a 44 slightly more precise age of 1060.5 ± 1.8 Ma, which is interpreted as the crystallization age of the anorthosite. 2.6.2- U-Pb rutile All U-Pb analytical results for rutile are presented in Table 2.3 and Figure 2.6. The importance of adequate common Pb corrections for samples containing large proportions of Pb (e.g. rutile and apatite) has been addressed in Verts et al. (1996), Chamberlain & Bowring (2000), and Schoene & Bowring (2006). Two methods are generally used to determine the common Pb isotopic compositions: (1) use of a Pb growth model (e.g. Stacey & Kramers, 1975); and (2) use of Pb isotopic compositions from a co crystallizing low-U mineral (e.g. feldspar, sulphide). In this study, we have used the Pb isotopic composition of coexisting plagioclase (Chapter 5) from the same sample, or from a nearby sample, to correct for common Pb. A U-Pb age from rutile records the time when the temperature of the cooling rock reached the closure temperature of Pb diffusion in rutile, which is about 650-700°C (Cherniak, 2000), but is dependant on grain size and cooling rate. All U-Pb ages indicated in the section below for rutile are interpreted as cooling ages (see section 7.3). Saint-Urbain deposits (samples 2006-Gi, 2015-B4, 2030-B2 and 2033-D) Rutile grains were separated from samples from four different Fe-Ti oxide deposits in the Saint-Urbain anorthosite: (1) sample 2006-Gi (Bignell) is a massive oxide rock composed of ilmenite and rutile (Figure 2.7a); (2) sample 2015-B4 (Coulomb East) is an oxide-bearing leuconorite that contains plagioclase, orthopyroxene, sapphirine, ilmenite and rutile (Figure 2.7b); (3) sample 2030-B2 (General Electric) is a massive oxide rock that contains ilmenite and rutile; and (4) sample 2033-D (Séminaire) is an oxide-bearing anorthosite, with plagioclase, ilmenite and rutile. Rutile grain size in the samples varied from 0.3 xO.9mmto3.3x5.5 mm. For sample 2006-Gi (Bignell), six multi-grain fractions were analyzed. U-Pb results from fractions R3, R5 and R6 are concordant to slightly discordant (-0.5 to 0.9%) and yield a restricted range of207Pb/6 ages from 908.1 to 908.9 Ma. The data from the other three fractions are discordant (1 .2-4.1%) with a range of207Pb/6 ages from 900 45 Bi gn el ln iti le (R u2 00 6-G 1) R 1, 4 60 R 2, 7 71 R 3, 9 59 R4 ,1 0 88 R5 ,1 0 96 R6 ,1 0 86 Co ul om b Ea st ru til e (R u2 01 5-B 4) R i, 3 R2 ,5 R3 , 7 10 32 ± i3 6 12 62 ± 34 1/ 43 9 90 0 ± 15 8 90 9 ± 44 4/ 62 2 87 5 ± 19 3 83 1 ± 54 4/ 83 8 Sé m in ai re ru tii e (R u2 03 3-D ) R i, 4 R2 ,4 R3 ,4 91 4.1 ± 6. 1 91 6. 9 ± 4. 7 89 2 * 25 10 17 * 70 94 9 * 12 3 91 5 ± 32 94 3 ± 41 92 9 ± 29 87 5 ± 19 3 0. 59 23 10 .2 0. 67 47 4. 7 0. 61 02 - 7. 9 0. 64 58 - 19 .3 0. 95 58 0.1 0. 59 23 - 36 .7 91 5 ± 38 7 79 1 ± 93 0/ 24 59 85 9 ± 38 5 67 6 ± 97 2/ 27 74 11 26 ± 35 4 15 19 ± 71 1/ 13 79 0. 63 60 - 24 0. 60 23 - 40 .5 0. 61 97 41 .3 Bi g Is la nd dy ke ru til e (R u2 10 9- A) R i,4 0 21 2 0. 24 0. 04 25 .6 0. 14 61 .2 0. 28 7 0. 16 60 ± 3. 17 1.6 1 15 ± 38 .9 0. 07 04 0 ± 36 .6 R 2, 40 21 6 0. 25 0. 04 25 .1 0. 12 65 .5 n o T h 0. 15 39 ± 3. 31 1.2 31 1 ± 49 .7 0. 05 80 3 ± 47 .6 99 0 ± 58 97 5 ± 48 7 94 0 ± 10 46 /3 85 7 92 3 ± 57 81 5 ± 55 7 53 1 ± 13 22 /9 99 9 0. 72 83 - 5. 7 0. 65 70 - 79 .3 Bi gn el ia pa tit e (A p2 00 6-F i) A 1, 20 88 A 2, 20 10 9 A 3, 25 10 8 A4 ,2 0 94 A5 ,1 5 89 Ta bl e 2. 3 R ut ile an d ap at ite U -P b TI M S an al yt ic al da ta Fr ac tio n W t U1’ Pb ° ° P b t Pb */ Pb 0 P b t T h/ U 5 Is ot op ic ra tio s± is ,% ” A pp ar en ta ge s± 2s ,M a 5 r’ % (m g) (pp m) (pp m) 1 ’ P b (pg ) 2 0 5 P b / 2 3 8 U 2 0 7 P b / 2 3 5 U 2 0 7 P b / 2 0 6 P b 2 0 6 P b / 2 3 8 U 2 0 7 P b / 2 3 5 U 2 0 7 P b / 2 0 6 P b di sc or da nc e 14 .6 2. 0 19 7.5 2. 69 46 .1 0. 00 5 0. 14 93 ± 0. 08 1. 44 46 ± 0. 49 0. 07 01 6 ± 0. 44 12 .7 1.8 21 7. 4 2. 98 43 .5 0. 00 2 0. 15 20 ± 0. 14 1. 48 05 ± 0. 50 0. 07 06 3 ± 0. 45 13 .5 1.9 49 6. 5 7. 06 16 .0 n o T h 0. 15 13 ± 0. 09 1. 45 15 ± 0. 48 0. 06 95 9 ± 0. 44 11 .2 1.6 29 9. 9 4. 18 32 .8 n o T h 0. 14 82 ± 0. 17 1. 41 05 ± 0. 56 0. 06 90 5 ± 0. 51 14 .2 2. 0 28 2. 8 3. 95 49 .2 n o T h 0. 15 14 ± 0. 19 14 49 1 ± 0. 47 0. 06 94 1 ± 0. 38 13 .6 1.9 24 9.1 3. 44 48 .3 n o T h 0. 15 14 ± 0. 14 1. 44 42 ± 0. 49 0. 06 92 0 ± 0. 43 10 8 0. 27 0. 04 34 .9 0. 28 16 .4 0. 25 9 0. 15 47 ± 1.2 8 1.7 64 3 ± 10 .5 0. 08 27 1 ± 9. 77 93 0. 29 0. 04 40 .1 0. 36 11 .5 0. 23 3 0. 14 92 ± 1.2 3 1. 42 66 ± 13 .2 0. 06 93 6 ± 12 .5 76 0. 30 0. 04 35 .7 0. 28 11 .9 0.1 33 0. 14 84 ± 1.4 7 1. 36 64 ± 16 .5 0. 06 67 7 ± 15 .6 90 7. 6 ± 5.9 92 2. 4 ± 6. 0 91 0. 5 ± 5. 8 89 3. 4 ± 6. 7 90 9. 5 ± 5.6 90 7. 5 ± 5.8 89 7. 2 ± 1.4 91 2. 3 ± 2. 3 90 8.1 ± 1. 5 89 0. 6 ± 2. 8 90 8. 9 ± 3. 3 90 8. 6 ± 2. 3 92 7.3 ± 22 .2 89 6. 4 * 20 .6 89 2.1 ± 24 .6 93 3 ± 18 /1 8 94 7 ± 18 /1 8 91 6 ± 18 /1 8 90 0 ± 21 /2 1 91 1 ± 16 /1 6 90 5 ± 18 /1 8 0. 65 69 4.1 0. 47 17 3.9 0. 53 34 0. 9 0. 47 49 1.2 0. 62 94 0.2 0. 53 11 - 0.5 0. 59 59 28 .5 0. 60 88 1.5 0. 61 02 - 7. 9 G en er al El ec tri c ru til e (R u2 03 0-B 2) R i, 5 31 3.3 0. 47 75 .0 0. 88 17 .3 0. 04 7 0. 15 23 ± 0. 36 1. 53 05 ± 3. 36 0. 07 28 6 ± 3. 16 R 2, 5 42 4. 2 0. 6 58 .5 0. 64 41 0. 10 9 0. 15 28 ± 0. 27 1. 49 72 ± 2. 41 0. 07 10 5 ± 2. 24 R 3, 5 29 0. 30 0. 57 35 .7 0. 28 11 .9 0.1 33 0. 14 84 ± 1. 47 1. 36 64 ± 16 .5 0. 06 67 7 ± 15 .6 41 0.2 3 0. 04 31 .2 0. 19 7. 9 n o T h 0. 17 09 ± 3. 70 1. 59 80 ± 36 .5 0. 06 78 2 ± 34 .2 38 0.2 5 0. 04 35 .0 0. 26 5.5 6 0. 95 6 0. 15 86 ± 6. 94 1. 54 63 ± 85 .6 0. 07 07 0 ± 78 .9 28 0. 62 0. 08 37 .6 0. 28 8. 6 n o Th 0. 15 26 ± 1.8 9 1. 30 94 ± 22 .4 0. 06 22 4 ± 21 .3 10 10 ± 12 3/ 13 4 95 9 ± 89 /9 4 83 1 ± 54 4/ 83 8 Bi g Is la nd dy ke ru tii e (R u2 10 4-D ) R i,5 37 0. 32 0. 05 25 .3 0.1 3 15 .2 0.2 71 0. 16 18 ± 2. 81 1. 46 14 ± 32 .1 0. 06 55 1 ± 30 .4 R 2, 5 36 0. 34 0.0 5 32 .1 0. 22 8.2 8 0. 16 3 0. 15 54 ± 2. 84 1. 32 97 ± 33 .3 0. 06 20 7 ± 31 .6 R 3, 8 35 0. 28 0.0 5 26 .3 0. 15 11 .1 0.4 71 0. 15 59 ± 3. 17 2. 03 14 ± 26 .0 0. 09 45 2 ± 24 .2 96 9 ± 45 5 86 3 ± 10 04 /3 20 0 94 9 ± 10 55 94 9 ± 17 25 /9 99 9 85 0 ± 25 8 68 2 ± 71 7/ 13 37 96 7 ± 51 93 1 ± 49 93 4 ± 55 1.7 0. 44 28 .3 0. 26 15 0 1. 92 6 0. 17 80 ± 0. 36 1. 94 48 ± 3. 46 0. 07 92 5 ± 3. 20 10 55 .9 ± 7. 0 10 97 ± 46 11 79 ± 12 1/ 13 2 0. 75 35 11 .3 1.8 0.4 3 27 .1 0. 22 21 8 1. 62 2 0. 18 14 ± 0. 61 1. 98 18 ± 6. 92 0. 07 92 3 ± 6. 33 10 75 ± 12 11 09 ± 93 11 78 ± 23 2/ 27 3 0. 96 23 9.5 1.7 0.4 1 28 .4 0. 26 17 2 1. 84 2 0. 17 23 ± 0. 61 1. 94 50 ± 4. 79 0. 08 18 5 ± 4. 34 10 25 ± 12 10 97 ± 64 12 42 ± 16 1/ 18 0 0. 77 36 18 .9 1.9 0.4 5 28 .5 0. 25 16 9 1.7 28 0. 17 59 ± 0. 33 1. 84 78 ± 3. 56 0. 07 61 9 ± 3. 30 10 44 .5 ± 6. 4 10 63 ± 47 11 00 ± 12 7/ 13 8 0. 80 26 5.5 1.7 0.4 5 26 .5 0.2 1 19 4 1.6 61 0.1 90 1 ± 0. 25 2. 09 21 ± 3. 09 0. 07 98 0 ± 2. 86 11 22 .1 ± 5. 2 11 46 ± 43 11 92 ± 10 9/ 11 7 0. 93 66 6. 4 Ta bl e 2. 3 (co nti nu ed ) Fr ac tio ns W t U b P b z Pb d Pb */ Pb 0 Pb ’ Th JU 5 Is ot op ic ra tio s± ls ,% h A pp ar en t a ge s± 2s ,M ah r % (m g) (pp m) (pp m) 2°4P b (pg ) 2 0 6 P b / 2 3 8 U 2 0 7 P b / 2 3 5 U 2 0 7 P b / 2 0 6 P b 2 0 6 P b / 2 3 8 U 2o7P b /n 5U 2 0 7 P b / 2 0 6 P b di sc or da nc e Fu rn ac e ap at ite (A p2 03 6-B IB ) A2 ,5 0 19 7 0. 63 0. 14 22 .0 0. 10 28 2 1. 42 9 0. 16 77 ± 0. 62 1. 68 02 ± 8.1 3 0. 07 26 5 ± 7. 59 10 00 ± 11 10 01 ± 10 4 10 04 ± 28 1/ 34 3 0. 89 68 0.5 A3 ,1 00 25 9 0. 64 0.1 3 23 .0 0. 12 28 4 1.4 21 0. 15 39 ± 1.1 7 1. 28 75 ± 16 .0 0. 06 06 7 ± 14 .9 92 3 ± 20 84 0 ± 18 3 62 8 ± 53 8/ 81 9 0. 92 13 - 50 .5 A5 ,1 5 98 0. 57 0.1 1 22 .4 0. 10 10 8 1. 25 8 0. 15 54 ± 1.4 4 1. 78 12 ± 13 .4 0. 08 31 6 ± 12 .2 93 1 ± 25 10 39 ± 17 4 12 73 ± 41 4/ 56 9 0. 84 80 28 .8 A6 ,1 5 78 0. 50 0.1 1 21 .5 0. 09 94 .7 1. 47 6 0. 16 00 ± 1.8 7 1.6 97 1 ± 21 .2 0. 07 69 5 ± 19 .6 95 7 ± 33 10 07 ± 27 2 11 20 ± 63 0/ 10 75 0. 90 30 15 .7 Fu ra nc e ap at ite (A p2 03 6-B 3) A 1, 30 73 1.9 0. 39 33 .9 0. 36 80 .1 1. 57 4 0. 15 78 ± 0. 41 1. 52 65 ± 4. 00 0. 07 01 5 ± 3. 71 94 4. 7 ± 7. 2 94 1 ± 49 93 3 ± 14 5/ 16 0 0. 72 40 - 1. 4 A 2, 30 10 4 1.8 0. 37 33 .7 0. 36 11 1 1.5 31 0. 15 78 ± 0. 59 1. 40 46 ± 6. 08 0. 06 45 6 ± 5. 62 94 4 ± 10 89 1 ± 72 76 0 ± 22 1/ 25 7 0. 80 16 - 26 .1 A3 , 5 0 13 9 2. 0 0. 40 37 .0 0. 42 13 4 1. 50 6 0. 15 50 ± 0. 36 1. 52 34 ± 2. 58 0. 07 13 0 ± 2. 38 92 8. 7 ± 6. 1 94 0 ± 32 96 6 ± 94 /1 00 0.6 13 1 4.1 Fu rn ac e ap at ite (A p2 03 6-D 2) A l, 25 15 6 5.7 1.6 55 .8 1.0 7 24 4 2. 90 4 0. 16 97 ± 0. 17 1.7 41 1 ± 1. 30 0. 07 44 3 ± 1.1 6 10 10 .2 ± 3. 2 10 24 ± 17 10 53 ± 46 /4 7 0. 83 66 4. 4 A 2, 35 19 3 6.2 1.7 53 .6 1.0 1 33 9 2. 91 9 0. 16 75 ± 0. 11 1. 68 13 ± 1. 00 0. 07 28 1 ± 0. 91 99 8.1 ± 2. 0 10 01 ± 13 10 09 ± 36 /3 7 0. 91 39 1.1 A 3, 40 21 6 5.3 1.5 54 .3 1.0 2 31 6 2. 85 0 0. 16 67 ± 0. 11 1.6 14 1 ± 0. 99 0. 07 02 3 ± 0. 89 99 3. 8 ± 2. 1 97 6 ± 12 93 5 ± 36 /3 7 0. 86 20 - 6.8 aF ra ct io n ID (ru til e: R i, R 2, et c. ; ap at ite : A l, A 2, et c. ), fo llo w ed by th e n u m be ro fg ra in s bU bl an k c o rr e c tio n o f0 .1 -1 .0 pg ± 20 % ;U fr ac tio na tio n c o rr e c tio ns w e re m e a su re d fo re a c h ru n w ith a do ub le 23 32 35 U sp ik e. ° R ad io ge ni c Pb ;a ll ra w Pb da ta c o rr e c te d fo rf ra ct io na tio n o f0 .2 3- 0. 37 % /a m u ± 20 % de te rm in ed by re pe at ed a n a ly si s o fN B S- 98 2 re fe re nc e m a te ria l. dM ea su ie d ra tio c o rr e c te d fo rs pi ke an d Pb fr ac tio na tio n. eR at io o fr a di og en ic to c o m m o n Pb T ot al c o n m io n Pb in a n a ly si s ba se d o n bl an k is ot op ic c o m po si tio n: 2 t t P b / 2 0 4 P b = 18 .5 ± 3% ,2 0 7 P b / 2 0 4 P b = 15 .5 -1 5. 8 ± 3% , 2 0 8 P b / 2°4P b = 36 .4 -3 8. 3 ± 0. 5% . tM od el Tb /U de riv ed fr om ra di og en ic 2 tt P b an d th e 2 0 7 P b / 2 0 6 P b a ge o ff ra ct io n. hF ra ct io na tio n bl an k an d c o m m o n Pb -c or re ct ed fo rp ro ce du ra lb la nk s, w hi ch w e re — 5- 30 pg du rin g th e c o u rs e o ft hi s st ud y (2 00 4- 20 07 ). C om m on Pb is ba se d o n m e a su re d is ot op ic c o m po si tio ns o fa ss o c ia te d pl ag io cl as e. ‘ C or re la tio n c o e ff ic ie nt . D is co rd an ce in % to o rig in . D0 0.154 D C’, 0 920 ‘ L Coulomb East I Ri0.160 2015-B4 940908.3 ± 1.5 Ma 903 ± 11 Ma0.152 MSWD = 0.14 Bignell 2006-G1 0.156 MSWD 0 79 R6 0.1520.150 900 0.148 0.148 0.144 380j’ 0.146 0.140 1.37 1.41 1.45 1.49 0.6 1.0 1.4 1.8 2.2 0.155 0.153 7/7 0.151 f I / General Electric \ / 2030-B20.149 \ 914.1 ± 4.9 Ma _7890 MSWD = 0.22 0.147 1.3 1:4 1.5 1.6 1.7 Ld Séminaire 2033-D 941 ±40 Ma MSWD = 0.67 U 0.20 0.18 0.16 0.14 0.12 0.185 0.175 0.165 0.155 0.145 0.135 1 2 3 4 5 D 0 ‘0 0 C” 0.18 0.17 1040 Ri 0.16 BigIsl R3 and 2104-D 943±25Ma R2 880.15 ‘7-MSWD =0.0075 0.14 0 12 3 4 0 • !• __ L 105 R2 Ri Big Island 2109-A 850/ MSWD = 0.23 1 2 3 207Pb/35U 207Pb/35U Figure 2.6 Concordia diagrams for U-Pb data from analyzed rutile fractions from the Saint-Urbain deposits (a-d) and Big Island dyke (e-f). Each ellipse represents the results ofthe analysis ofa single fraction, as indicated in Table 2.3 (e.g. Ri, R2), and corresponds to the associated 2 uncertainties unless otherwise noted. The white ellipse with thick outline indicates the concordia age; 2 uncertainties are reported first without including the decay-constant errors and then including the decay-constant errors. (a) sample 2006-B 1, Bignell deposit; (b) sample 201 5-B4, Coulomb West deposit; (c) sample 2030-B2, General Electric deposit; (d) sample 2033-D, Séminaire deposit, the uncertainty in this age is calculated as t MSWD”2(t is a Student’s t-factor that “takes into account the fact that the true scatter of the data-point population is only estimated from a finite number ofactual points”; Ludwig, 2003); (e) sample 21 04-D, west side of the Big Island dyke; (f) sample 2109-A, east side of the Big Island dyke. The black band is the concordia curve including decay constant errors. 48 Figure 2.7 Textures of minerals (e.g. rutile, apatite, plagioclase and biotite) analyzed for U-Pb and Ar-Ar geochronology from the Saint-Urbain and Big Island deposits. (a) Sample 2006-Gi; photomicrograph (plane polarized light) of ilmenite-rutile assemblages in the Bignell deposit; note the large size of the rutile grains (>2 mm); (b) sample 2015-A4; photomicrograph (cross polarized light) of recrystallized plagioclase with 1200 triple junctions from the Coulomb East deposit; (c) sample 201 5-B4; photomicrograph (plane polarized light) showing ilmenite rutile-sapphirine-plagioclase-orthopyroxene-apatite assemblage from the Coulomb East deposit; (d) sample 203 6- B 1B; photomicrograph (cross polarized light) showing abundant and large (>1 mm) biotite crystals at the interface between ilmenite and plagioclase from the Furnace deposit; (e) sample 2036-D2; photomicrograph (plane polarized light) showing an apatite-ilmenite (nelsonite) rock from the Furnace deposit; (f) sample 2104-D; photomicrograph (plane polarized light) showing an ilmenite-rutile-sapphirine assemblage from the Big Island deposit. Scale bars are noted on each photomicrograph. Abbreviations (following Kretz, 1983): (rt) rutile; (ilm) ilmenite; (plag) plagioclase; (ap) apatite; (spr) sapphirine; (opx) orthopyroxene; (bt) biotite. 49 to 947 Ma. A concordia age, calculated from the results of fractions R3, R5 and R6, is 908.3 ± 1.1 Ma. For the three other samples, the U-Pb data are all associated with relatively larger uncertainties due to the very low proportion of radiogenic Pb (Pb*/Pb = 0.19-0.88, Table 2.3) and yield a wide range of207Pb/6 ages from 831 to 1262 Ma. Concordia ages for each sample were calculated with results from all analyzed fractions and the ages are 903 ± 11 Ma for sample 2015-B4 (Coulomb East), 914.1 ± 3.0 Ma for sample 2030-B2 (General Electric), and 941 ± 39 Ma for sample 2033-D (Séminaire). Big Island deposit (samples 2104-D and 2109-A) Sample 2104-D is a massive oxide rock from the west part of the Big Island dyke and contains ilmenite, rutile and sapphirine (Figure 2.7c), whereas sample 2109-A is a massive oxide rock sampled from the eastern part of the dyke that is composed of ilmenite and rutile. Rutile grain size varies from 0.9 x 1.25 mm to 2 x 2.6 mm. As for the Saint-Urbain samples, the proportion of radiogenic Pb is very low (Pb*/Pb = 0.12-0.22, Table 2.3) leading to large uncertainties associated with the U-Pb results. The U-Pb data from all fractions of sample 21 04-D and sample 2109-A are concordant and yield ages from 531 to 1519 Ma. The concordia age for sample 21 04-D, calculated using all fractions, is 943 ± 25 Ma and the concordia age for sample is 2109-A, also using all fractions, is 962 ± 32 Ma. 2.6.3- U-Pb apatite U-Pb apatite results are presented in Table 2.3 and Figure 2.8. The common Pb correction was done using the Pb isotopic compositions of co-existing plagioclase (Chapter 5) unless otherwise indicated. A U-Pb age from apatite dates the time when the temperature of the cooling rock reached the closure temperature of Pb diffusion in apatite (TCB = 400-450°C; Cherniak et aL, 1991) (see section 7.3). Bignell deposit, Saint-Urbain (sample 2006-Fl) Sample 2006-Fl is a rock containing mostly ilmenite and apatite (Figure 2.7d). The apatite grains were elongate, varying in size from 0.2 x 1 mm to 1 x 2 mm. Five fractions 50 0.174 Furnace 2036-BiB ‘10200.194 Bignell 2006-Fl A5 0.170 932.6 ± 14/14 Ma 1—7 A6Upper intercept0.190 1192 ±203 Ma 0.166MSWD = 0.66 0.162 . 0.182 D 0.186 104 108 _ ,1 0.1580.178 0.174 MSWD = 0 074 -_--—_‘ N 0.154 A5 0.1500.170 0.166 0.146 1.5 1.7 1.9 2.1 2.3 0.4 0:8 1.2 1.6 2.0 2:4 2.8 0.1715960 LC 0.160 0.1705 Unresolvable 0.156 Furnace 206-b2 Ld 0.1695D 0.158 0.1685 C 0.1675 Furnace 2036-B/0.154 947.3 ± 8.7/9 Ma 0.1655 0.1665 MSWD = 1.6 __________________ 0.152 _____________________________ 1.15 1.25 1.35 1.45 1.55 1.65 1.75 1.54 1.62 1.70 1.78 2o7Pb/nsU 2o7Pb/nsU Figure 2.8 Concordia diagram for U-Pb data from analyzed apatite fractions from the Saint-Urbain deposits. Each ellipse represents the results ofthe analysis of a single fraction, as indicated in Table 2.3 (e.g. Al, A2), and corresponds to the associated 2 uncertainties. The white ellipse with thick outline indicates the concordia age; 2 uncertainties are reported first without including the decay-constant errors and then including the decay-constant error. (a) sample 2006- Fl, Bignell deposit; (b) sample 2036-B 1B, Furnace deposit, the uncertainty in this age is calculated as t MSWD’; (c) sample 2036-B3, Furnace deposit, the uncertainty in this age is calculated as t MSWD; (d) sample 2036-D2, Furnace deposit. The black band is the concordia curve including decay constant errors. 51 of apatite grains were analyzed and the U-Pb results are discordant. This sample does not contain plagioclase and using the Pb isotopic compositions of plagioclase from a nearby sample of the Bignell deposit yields to an upper intercept age of 1192 Ma with a very large associated uncertainty (203 m.y.) (Figure 2.8a). The Pb isotopic composition of co existing ilmenite was determined for use in the common Pb correction (C.-E. Morisset, unpublished data), however the analytical uncertainty on the isotopic ratios of the ilmenite was so large that they were unusable. Alternatively, an age of 1075 ± 30 Ma was calculated using the Pb growth model of Stacey & Kramers (1975) for the common Pb correction, which barely overlaps with the U-Pb zircon age of 1053.6 ± 2.8 Ma for the Bignell anorthosite. Field relationships clearly indicate that the deposits intrude anorthosite, thus the older apatite age, which has a closure temperature much lower than for zircon, is geologically meaningless. The age given by the 3-D isochron diagram calculated with Isoplot 3.00 is 798 ± 45 Ma. This age is about 100 m.y. younger than the 4O39 biotite ages (section 6.4), which correspond to closure temperatures around 335°C (Harrison et al., 1985) and which are lower than the closure temperature for Pb diffusion in apatite. Thus, no reliable age information was possible based on the U-Pb systematics of apatite from the Bignell deposit. Furnace deposit, Saint-Urbain (samples 2036-BiB, 2036-B3 and 2036.-D2) Sample 2036-B 1 B is a megacrystic leuconorite with plagioclase and orthopyroxene, 10 to 25 cm in length, and a matrix mostly composed of ilmenite and apatite. Sample 2036-B3 is a nelsonite (ilmenite + apatite) that occurs near the location of sample 2036- BiB and sample 2036-D2 is also a nelsonite located 20 metres west of the Furnace deposit. The U-Pb results from most fractions are concordant and yield a large range of 207Pb/6 ages from 628 to 1273 Ma. The concordia age of sample 2036-BiB is 932.6 ± 14 Ma, based on three fractions, and the concordia age of sample 203 6-B3 is 947.3 ± 8.7 Ma, based on two fractions. No reliable age could be determined from the U-Pb data of sample 203 6-D2. 52 2.6.4- 40Ar/39rbiotite and plagioclase Biotite in the Saint-Urbain and Big Island samples occurs mostly as small crystals (0.9 x 1.3 mm to 1 x 1.44 mm) located between ilmenite and plagioclase (Figure 2.7e). Plagioclase is typically equigranular (0.5 x 0.6 mm to 1 x 1.2 mm) with 1200 triple junction grain intersections (Figure 2.7f). At Saint-Urbain, both biotite and plagioclase from the deposits and the host anorthosite were selected. Sample 2033-A2 from the Séminaire deposit was collected from a cm-thick horizon of biotite present at the contact between the deposit and the anorthosite. At Lac Allard, biotite and plagioclase selected for analysis were from the Big Island dyke. All40Ar/39r results are presented in Appendices a (Table 2.A1) and b (Table 2.A2), and40Ar/39r incremental-heating age spectra are shown for biotite in Figure 2.9 and for plagioclase Figure 2.10. Many of the biotite samples and several of the plagioclase samples show low apparent ages for the low-temperature increments indicating Ar loss (Figures 2.9 and 2.10). Two of the plagioclase samples (Séminaire 2033-D and Big Island 2103-B2) have saddle-shape spectra with older apparent ages for the low-temperature increments. All biotite and plagioclase results yield integrated plateau ages, which are interpreted as cooling ages that correspond to the temperature of closure to diffusion of Ar. The integrated plateau ages given by the biotite results from Saint-Urbain range from 885.8 ± 4.6 Ma to 931.3 ± 4.8 Ma, and biotite from the Big Island dyke yields a plateau age of 952.4 ± 5.1 Ma. The plateau ages for plagioclase from Saint-Urbain span a range from 860.3 ± 8.3 Ma and 884.4 ± 5.4 Ma, and plagioclase from Big Island yields a plateau age of 913 ± 12 Ma. 2.7- Discussion 2.7.1 - Crystallization age of the Saint-Urbain anorthosite and Saint-Anne du Nord orthopyroxene granodiorite The crystallization age of the Saint-Anne du Nord orthopyroxene granodiorite (1060.8 ± 2.8 Ma) indicates that it is the oldest intrusion of the studied suite from the Saint-Urbain area (Figure 2.11). This age is consistent with U-Pb zircon ages of 1060 ± 10 Ma for three quartz mangerite samples from this intrusion as reported in Icenhower et 53 -.- 1ILJ Coulomb East 2015-A4-2 Lc 950 850 750 650 Plateau age = 892.2 ± 4.7 MaIncludes 79.7% of the 39Ar 550 MSWD = 1.7 450o io 4o 60 80 10 Seminaire 20033-D e 800 600 Plateau age = 885.8 ± 4.6 Ma 400 Includes 70% of the 39Ar MSWD = 0.95 0 20 40 60 80 100 910 890 870 850 830 810 790 1200 Bignell 2006-B4 L 1000 __________________________ 800 600 400 Plateau age = 931.3 ± 4.8 Ma Includes 62% of the 39Ar 200 MSWD = 1.5 U 770- 0 20 40 60 80 100 0 •1 frr Bignell anorthosite 2006-C4 l t .i Plateau age = 886.7 ± 5.4 Ma Includes 86% of the 39Ar MSWD = 0.64 20 40 60 80 100 (U (0 ci) (0 ci) Séminaire contact 2033-A2 W Plateau age = 900.4 ± 4.8 Ma Includes 87.4% of the 39Ar MSWD = 1.3 0 ) 1400 1200 1000 800 600 400 200 0 1000 800 600 400 200 20 40 60 80 100 Furnace 2036-BiB Plateau age = 875.7 ± 4.5 Ma Includes 68.5% of the 39Ar MSWD = 1.7 1000 Big Island dyke 2103-B2 900 800 700 Plateau age = 952.4 ± 5.1 Ma Includes 6 1.2% of the 39Ar 600 MSWD = 0.77 500 0 20 40 60 80 100 Cumulative 40Ar/39r percent Figure 2.9 40ArI39r incremental-heating age spectra for biotite from the Saint-Urbain and Big Island Fe-Ti oxide deposits and their respective host anorthosites. (a) Biotite 2006-B4, Bignell deposit; (b) biotite 2006-C4, anorthosite host of Bignell deposit; (c) biotite 2015-A4, Coulomb East deposit; (d) biotite 2033-A2, contact between the Séminaire deposit and the host anorthosite; (e) biotite 2033-D, Séminaire deposit; (f) biotite 2036-B1B, Furnace deposit; (g) biotite 2103-B2, Big Island deposit.0 20 40 60 80 Cumulative 40Ar/39r percent 100 54 Figure 2.1040Ar/39rincremental-heating age spectra for plagioclase from the Saint-Urbain and Big Island Fe-Ti oxide deposits and their respective host anorthosites. (a) Plagioclase 201 5-A4, Coulomb East deposit; (b) plagioclase 2033-D, Séminaire deposit; (c) plagioclase 2042-A, Saint-Urbain anorthosite; (d) plagioclase 2103- B2, Big Island deposit. 1400 1200 1000 800 600 400 200 00 ‘V ci) (V z ci) Séminaire 2033-D Plateau age = 860.3 ± 8.3 Ma Includes 50.1% of the 39Ar MSWD = 0.83 = I 0 20 40 60 80 100 Coulomb East 2015-A4 Plateau age = 884.4 ± 5.4 Ma Includes 57.8% of the 39Ar MSWD = 1.7 20 40 60 80 100 Leuconorite 2042-A Plateau age = 873.4 ± 5.4 Ma Includes 61.2% of the 39Ar MSWD = 0.75 _ —t 0 20 40 60 80 100 Cumulative 40Ar/39r percent 1500 1300 1100 900 700 500 1900 1700 1500 1300 1100 900 700 500 1700 1500 1300 1100 900 700 500 300 Big Island dyke 2103-B2 Plateau age = 913 ± 12 Ma Includes 89.3% of the 39Ar MSWD = 0.64 0 20 40 60 80 100 Cumulative 40Ar/39r percent 55 1040 1045 1050 1055 1060 1065 1070 I I La Saint-U rba in Area orthopyroxene-granodiorite I I oxyde-apatite-gabbronorite Lac des Cygnes anorthosite Bignell anorthosite I I Leuconorite Anorthosite close to mangerite contact Lac Allard Area Tio mine I I Anorthosite close to Big Island dyke Big Island dyke I I I I 1040 1045 1050 1055 1060 1065 1070 Age (Ma) Figure 2.11 Summary diagram of U-Pb zircon ages determined in this study. (a) Saint-Urbain area; (b) Lac Allard anorthosite and Big Island deposit, Havre-Saint-Pierre anorthosite suite. Error bars indicate 2 uncertainty associated with each age not including the decay constant errors. 56 a!. (1998). The 1057.4 ± 1.5 Ma oxide-apatite gabbronorite (OAGN) from the northwest margin of the Saint-Urbain anorthosite overlaps in age (within error) with the Bignell anorthosite (1053.6 ± 2.6 Ma) and the Lac des Cygnes anorthosite (1055.0 ± 2.4 Ma), indicating crystallization of the southern and northern parts of the Saint-Urbain massif at ca. 1053-1056 Ma (Figures 2.2b and 2.11). Based on deformation observed in the OAGN (e.g. folded layering; Figure 2.3d), we suggest that the OAGN was intruded prior to emplacement of the northern part of the massif (Lac des Cygnes). The 1046.2 ± 3.1 Ma megacrystic Saint-Urbain leuconorite, located between the two anorthosite samples (Figure 2.2b), is the youngest phase of the Saint-Urbain massif indicating that it is a composite intrusion. Based on the results of this study, magmatism in the Saint-Urbain region occurred over a period of about 15 m.y. from 1061 to 1046 Ma. Higgins & van Breemen (1996) and Hébert et al. (2005) defined a 1080-1045 Ma magmatic episode in the Saguenay-Lac Saint-Jean region that contains the Chicoutimi and Poulin de Courval mangérites (Figure 2.2). This AMCG magmatic event is sandwiched between an older 1160-1140 Ma event defined by the Lac Saint-Jean anorthositic suite (Higgins & van Breemen, 1996) and a younger 1020-10 10 Ma event defined by the Valin anorthosite suite containing the Labrieville and Mattawa anorthosite massifs (Figure 2.2). There is no evidence indicating that regional metamorphism younger than the Saint Urbain anorthosite has affected the area. The last magmatic activity of the Lac Saint-Jean anorthosite suite (1142 ± 2 Ma) in the southwest part of the intrusion is coeval with the formation of zircon coronas (1142 ± 3 Ma) on baddeleyite (1157 ± 2 Ma) in the southeastern part of the suite (Higgins & van Breemen, 1992). These coronas have been interpreted as forming in response to contact metamorphism by unidentified intrusions in the area. The crystallization ages imply that emplacement of the Lac Saint-Jean anorthositic suite was contemporaneous with a regional metamorphic event related to the Shawinigan orogeny (1190-1140 Ma) of Rivers (1997); there is no evidence that the suite has been affected by a later metamorphic event (Higgins & van Breemen, 1992; Hébert, 2001). Metamorphic zircon from the country rock amphibolite of the Labrieville anorthosite yields a metamorphic age of 1015 ± 18 Ma (Owens et al., 1994), which has been interpreted as the age of the Grenvillian metamorphism or the effect of a heating 57 episode due to the numerous intrusions in the region. Finally, the U-Pb zircon systematics from a syenite sample from the older 1327 Ma De la Blache plutonic suite (Gobeil et al., 2002) provide a lower intercept age with concordia of 1084 ± 27 Ma, which is the metamorphic age in this area. Combining the results from published metamorphic ages in the area and knowledge of the three orogenies for the Grenville Province (Rivers, 1997), emplacement of the Lac Saint-Jean massif was coeval with the Shawinigan orogeny (1190-1140 Ma) and the Saint-Urbain anorthosite was coeval with the Ottawa orogeny (1080-1020 Ma) recorded by the De la Blache plutonic suite and by the 1080-1045 Ma magmatic event defined by Hébert et al. (2005). The Rigolet orogeny (10 10-990 Ma) has not been observed in this area, which is consistent with the observation that deformation associated with this event is typically concentrated close to the Grenville Front (Rivers et al., 2002). 2.7.2- Crystallization age of the Big Island deposit and Havre-Saint-Pierre anorthosite Some of the intrusive lobes of the large Havre-Saint-Pierre anorthositic suite have previously been dated. However, prior to this study, the Lac Allard lobe, which hosts the Lac Tio and Big Island deposits was of unknown age. The crystallization age of anorthosite from the site of the Lac Tio deposit (sample 2132) is 1060.5 ± 1.9 Ma. The anorthosite sample located 25 km southwest of the Lac Allard mine, close to the mangerite contact, is dated at 1061.6 ± 3.0 Ma (sample 2114-B), which overlaps the age of the Lac Tio anorthosite within error (Figure 2.11). The age of the anorthosite sample (2123-B) located less than 5 metres from the Big Island dyke is 1057.4 ± 5.7 Ma, and overlaps within error with the other two anorthosites and with the 1052.9 ± 6.5 Ma Big Island dyke (Figure 2.11). Field relations indicate that the dyke crosscuts the anorthosite (Figure 2.30. The crystallization age of ca. 1061 Ma for the Lac Allard lobe is the same as that for the Rivière-au-Tonnerre lobe (1062 ± 4 Ma, van Breemen & Higgins, 1993) and is much younger than the Rivière Sheldrake massif (1139-1129 Ma, Wodicka et al., 2003), which overlaps with mangerite envelope (1126 +71-6 Ma, Emslie & Hunt, 1990). Emplacement of the Havre-Saint-Pierre anorthosite suite thus occurred over a 60 million year period, with the first stage following the Shawinigan orogeny (1190-1140 Ma). The 58 U-Pb metamorphic zircon age of the Rivière Sheldrake massif of Ca. 1080 Ma, and the U-Pb zircon age of 1079 ± 5 Ma (Loveridge, 1986) and U-Pb rutile age of 1052 +61-4 Ma both from the Buit Complex, imply that the second stage of Havre-Saint-Pierre anorthosite magmatism occurred just after the peak of the Ottawan orogeny (1080-1020 Ma) in the area. As observed for the Saint-Urbain massif, the Rigolet orogeny does not appear to be recorded in the Havre-Saint-Pierre area. 2.7.3- Cooling histories of the Saint-Urbain and Lac Allard anorthosites For each of the anorthosite massifs and their associated Fe-Ti oxide deposits, bulk closure temperatures (TCB) for the diffusion of Pb in zircon, rutile and apatite as well as for the diffusion of Ar in biotite and plagioclase were calculated with the formula TCB = EI(Rln(eARD1Tc2IEa(dTldt))), presented in Hodges (2003) (adapted from Dodson, 1973), where R is the gas constant and A is a constant depending on the geometry of diffusion. For each system, the diffusivity (Dj) and activation energy (E) constants were derived from the following studies: Pb-zircon (Cherniak & Watson, 2000), Pb-rutile (Chemiak, 2000), Pb-apatite (Cherniak et al., 1991), Ar-biotite (Harrison et al., 1985), and Ar plagioclase (Kelley et al., 2002). For each sample, we calculated a maximum bulk closure temperature (TCB max) using the largest mineral radius (a) observed in the sample and a cooling rate dT/dt = 5°C/m.y., and a minimum closure temperature (TCB mm), using the smallest mineral radius (a) observed in the sample and a slower cooling rate (dT/dt 1 °CIm.y.). The results of these calculations are shown in Table 2.4 and in Figure 2.12. At Saint-Urbain, several different cooling rates can be estimated. Using only results from zircon, rutile and apatite, the best fit line gives a relatively slow cooling rate of 2.7 ± 0.2°C/m.y. (Figure 2.12). With the rutile, biotite and plagioclase results (excluding the older biotite40Ar/39r age of ca. 930 Ma), the calculated cooling rate is relatively fast at 7.3 ± 1.2°C/m.y. This would imply that the rate of cooling increased at around 940 Ma, which would require a period of more rapid uplift. Using the combined apatite, rutile, biotite and plagioclase results, the calculated cooling rate is 2.7 ± O.6°C/m.y. Finally, use of all the geochronological results yields an cooling rate of 3.3 ± 0.2 °CIm.y. (Figure 2. 12a). Because there is no evidence for tectonic activity after 980 Ma, we favour the 59 Table 2.4 Aces and closure temperatures for samnies of Saint-Urbain and Lac Allard 2033-D U-Pb zircon 2006-C2 U-Pb zircon 2042-A U-Pb zircon 2043-A U-Pb zircon 2020 U-Pb zircon 2023 U-Pb zircon average zircon (exc. 2033-D) 2006-0 1 U-Pb rutile 2015-B4 U-Pb rutile 2030-B2 U-Pb rutile 2033-D U-Pb rutile average rutile 2036-B lB U-Pb apatite 2036-B3 U-Pb apatite ave apatite 2006-B4 Ar-Ar biotite 2006-C4 Ar-Ar biotite 201 5-A4 Ar-Ar biotite 2033-A2 Ar-Ar biotite 2033-D Ar-Ar biotite 2036-BiB Ar-Ar biotite ave biotite 2015-A4 Ar-Ar plagioclase 2033-D Ar-Ar plagioclase 2042-A Ar-Ar plagioclase ave plagioclase Saint-Urbain 992.4 7.2 38-42 1053.6 2.6 35-85 1046.2 3.1 30-125 1055.0 2.4 25-70 1057.4 1.5 20-160 1060.8 2.8 30-75 1055 ± 11 908.3 1.1 903 11 914.1 3.0 941 39 917 ± 34 932.6 14 100-800 947.3 8.7 90-180 940 ± 21 931.3 4.8 1000-2500 886.7 5.4 3000 892.2 4.7 500-1000 900.4 4.8 3000 885.8 4.6 700-850 875.7 4.5 450-700 895 ± 39 884.4 5.4 100-500 860.3 8.3 250-500 873.4 5.4 380-800 873 ± 24 1058 ± 8 <1020 136 <800 129 971 128 899 117 989 131 893 116 962 127 886 115 1000 132 877 114 966 127 893 116 932 ± 93 585 60 518 54 619 64 566 59 585 60 518 54 585 60 518 54 561± 78 565 25 445 21 499 23 441 22 488 ± 116 399 28 356 26 406 28 393 28 368 27 335 25 406 28 393 28 362 26 345 26 356 26 332 25 370 ± 55 313 10 233 8 313 10 268 9 345 10 285 9 293 ± 79 946 ± 85 585 60 585 60 552 ± 77 b TCB max (largest radius and fastest cooling rate) T mm (smallest radius and slower cooling rate) Sample - Method - Age ± 2s effective TCB ± TCB minC ± radius (urn)° max” 50-100 150-200 50-100 50-100 2102 2114-B 2123-B 2132 ave zircon 21 04-D 2109-A ave rutile U-Pb zircon U-Pb zircon U-Pb zircon U-Pb zircon U-Pb rutile U-Pb rutile Lac Allard (Havre-Saint-Pierre suite) 1052.9 6.5 50-200 1061.6 3 70-145 1057.4 5.7 30-75 1060.5 1.9 45-55 943 25 50-100 926 32 50-100 934.5 ± 24 1011 134 996 131 966 127 952 125 914 928 893 909 119 122 116 118 518 54 518 54 2103-B3 Ar-Ar biotite 952.4 8.3 500-850 362 26 335 25 2l03-B3 Ar-Arplagioclase 913 12 600-1250 360 11 306 9 a half the size of the smallest size of the mineral based on grain measuments and thin section observations 60 1100 1050 1200 — 1000 .1.bt-plag 0.3 ± 0.7 rt-bt-plag 7.3 ± 1.2 Big Island & Lac Allard rt-plag 6.9 ± 2.7 I I 1100 1050 1000 950 900 850 800 Age (Ma) Figure 2.12 Cooling histories of the Saint-Urbain (a) and the Lac Allard lobe of the Havre-Saint-Pierre anorthosite suite (b). Maximum (TCB max — gray symbols) and minimum (TCB mm — white symbols) bulk closure temperatures were calculated using the largest defined radius and a faster cooling rate (dT/dt = 5°C/m.y.) and using the smallest defined radius and a slower cooling rate (dT/dt = 1°C/m.y.), respectively, for each result (see Table 2.4). Errors bars indicate 2a uncertainty on the ages. Closure temperature uncertainty is estimated based on the errors given for the diffusion parameters for each isotopic system (see text for references on diffusion). The larger symbols with thick black lines show the average result for each mineral system (±2a). The cooling rates are the slope ofthe best fit lines to the data points using least square regression and the estimated uncertainty for each cooling rate is the error on the slope. Abbreviations: (zrc) zircon; (rt) rutile; (bt) biotite; (plag) plagioclase. 1000 950 900 850 800 La Saint-Urbain all 3.3 ± 0.2 800 600 400 200 0 1000 800 • zircon • rutile A apatite • biotite plagioclase 0 I C,) 0 I— I I Lb all 3.7 ± 0.3 rt-plag 3.2 ± 0.2 zrc-bt 5.5 ± 0.5 400 200 0 4*— numbers indicate cooling rates in °C/m.y. 61 interpretation that the cooling rate was relatively constant at about 3°CIm.y. Similar cooling rates have been observed elsewhere in the Grenville Province. Hanes et al. (1988), based on 40ArI39r ages on muscovite, biotite, microcline and plagioclase as well as40Ar/39r amphiboles ages from Lopez-Martinez (1982), calculated a cooling rate of 3°C/m.y. for the trondjemite batholiths (Eastern Elzevir terrane) in the Central Metasedimentary Belt. Martignole & Reynolds (1997) determined a cooling rate of 1.5°C/m.y. for a transect from the Morin anorthosite to the Grenville Front (Figure 2.1) based on U-Pb rutile (95 5-945 Ma), 40Ar/39rbiotite (90 1-889 Ma) and40ArI39rK- feldspar (859-808 Ma) ages, which are similar to the ages found at Saint-Urbain for the same minerals (Table 2.4). For the Lac Allard lobe of the Havre-Saint-Pierre anorthosite suite, estimated cooling rates are 3.2 ± 0.2°C/m.y. using the zircon and rutile results and 5.5 ± 0.5°CIm.y. using the zircon and biotite results. Based on the slower cooling rate, there would have to be an increase to 6.9 ± 2.7°C/m.y. (rutile-plagioclase) that is not supported by any known geological event in the area. Using the faster cooling rate, there would be a decrease in the cooling rate to <0.5°C/my. with the biotite and plagioclase results; a cooling rate of <0.5°C/m.y. is much slower than the average orogenic cooling rate (10°C/m.y., Reiners & Brandon, 2006) and is thus unlikely. Our preferred cooling rate is 3.9 ± 0.4°C/m.y. based on the combined results, which is similar to that determined for the Saint-Urbain anorthosite above. Cooling rates from the Saint-Urbain and Lac Allard anorthosites can be compared to those established for other AIvICG suites. Scoates & Chamberlain (2003) determined that the 1.43 Ga Laramie anorthosite complex (Wyoming) cooled from 1000°C to 350°C (temperature of the country rock) in about 30 m.y., which yields a relatively fast cooling rate of 20-25°C/m.y. The cooling rate of the Kiglapait layered intrusion in the 1.3 Ga Nain Plutonic Suite of Ladrador (Yu & Morse, 1992) was similar to that established for Laramie. Dörr et al. (2002) showed that a granitic intrusion in the Mazury AMCG complex (Poland) took approximately 100 m.y. to cool from 900°C to 300°C with a change in the cooling rate at 600°C (27°CIm.y. to 3°C/m.y.). The late Grenvillian 1.01 Ga Labrieville anorthosite massif (Owens et al., 1994; Owens & Tomascak, 2002), emplaced after the peak of the metamorphism in the area (ca. 1045), has a cooling rate of 62 7. 1°C/m.y., slightly faster than for Saint-Urbain and Lac Allard (e.g. 3-4°C/m.y.). Thus, The Saint-Urbain and Lac Allard anorthosites cooled much more slowly than most other AMCG plutonic suites. Because the emplacement of these two massifs is broadly coeval with the peak of metamorphism in their respective location, they experienced very slow cooling at rates similar to other regions of the Grenville Province as described above. 2.7.4- AMCG magmatism and relationship to tectonics of the Grenville Province Anorthosite-mangerite-charnockite-granite (AMCG) magmatism extends for over one billion years in the Proterozoic from the 2020-2010 Ma Arnanuat massif in Labrador (Hamilton et al., 1998) to the 950-930 Ma Rogaland massif in Norway (Schärer et a!. 1996; Andersen & Griffin 2004) (Figure 2.13). Within the Grenville Province of eastern Canada, AIVICG magmatism occurred episodically from the large Mealy Mountains massif (1640 Ma; Emslie & Hunt, 1990) and small intrusions in the Manicouagan Imbricate Zone (1648-1628 Ma; Indares et a!., 1998), which are coeval with deformation and metamorphism associated with accretion of magmatic arcs during the Labradorian deformation event (1710-1600 Ma; Gower & Krogh, 2002), to the 975 Ma Vieux Fort massif(Heaman et al., 2004). Several major AMCG Complexes (e.g. Harp Lake, Michikamau) were emplaced at ca. 1.45 Ga north of the Grenville Front, following the Pinwarian orogeny (1520-1460 Ma; Gower & Krogh, 2002). Major suites, including the Lac Saint-Jean, Marcy (Adirondacks), Morin, and parts of the Havre-Saint-Pierre anorthosite suites (ca. 1167-1130 Ma), were emplaced during and just following the Shawinigan orogeny (1190-1140 Ma), which is one phase of the larger Grenvillian orogenic event (Rivers 1997; Rivers et al. 2002). Subsequent AMCG magmatism (ca. 1080-1055 Ma) in the Lac Saint-Jean area and parts of the Havre-Saint-Pierre anorthosite suite, including the Lac Allard lobe, and smaller intrusions like the Saint-Urbain anorthosite, is coeval with the 1080-1020 Ma Ottawan orogeny (Figure 2.13). Finally, the youngest intrusions of the Labrieville and Vieux Fort were emplaced during and following the Rigolet orogeny (10 10-980 Ma). The AMCG suites emplaced during the Grenville orogeny may be related to convective thinning of the lithosphere following major continent-continent collisions during the final phases of assembly of the Grenville tectonic province (Corrigan & 63 900 1100 1300 1500 1700 1900 Age (Ma) Figure 2.13 Compilation of crystallization ages for Proterozoic AMCG magmatism worldwide (updated from Scoates & Chamberlain, 1995). Massifs are found in eastern Canada, except when noted. Arnanunat — Hamilton et al. (1998); Lofoten (Norway) — Corfu (2004); Korosten (Ukraine) — Amelin et al. (1994); Horse Creek — Scoates & Chamberlain (1997); Lanying & Damiao (China) — Zhang et al. (2007); intrusions ofthe Manicougan Imbricate Zone — Indares et al. (1998); Wigborg (Finland/Russia) — Vaasjoki et al. (1991), Aviola et al. (1999); Mealy Mountains — Emslie & Hunt (1990); Salmi (Russia) —Amelin et al. (1997); Bengal (India) — Chatterjee et al. (2008); Mazury (Poland) — Dörr et al. (2002); Wolf River (USA) — Van Schmus et al. (1975); Kunene (Namibia/Angola) — DrUppel et al. (2007), Mayer et al. (2004); Jonhoping (Sweden) — Brander & Söderlund (2007); Laramie (USA) — Scoates & Chamberlain (1995), Frost et al. (1990), Verts et al. (1996), Scoates & Chamberlain (2003); Harp Lake, Michikamau — Krogh & Davis (1973); Rivière Pentecôte — Emslie & Hunt (1990); Nain— Simmons et al. (1986), Simmons & Simmons (1987), Emslie & Loveridge (1992), Hamilton et al. (1994), Amelin et al. (1994), Berg et al. (1994); De la Blache — Gobeil et al. (2002); Lac Saint-Jean — Higgins & van Breemen (1992), Hervet et al. (1994), Higgins et al. (2002), Hébert & van Breemen (2004b), Hèbert et al. (2005); Mattawa— Hébert et al. (2005); Adirondacks — McLelland et al. (2004); Morin — Emslie & Hunt (1990), Doig et al. (1991); Oaxacan (Mexico) — Keppie et al. (2003); Atikonak — Emslie & Hunt (1990); Havre-Saint Pierre — Emslie & Hunt (1990), van Breemen & Higgins (1993), Wodika et al. (2003), this study; Saint-Urbain — this study; Montpellier (USA) —Aleinikoffet al. (1996); Labrieville — Owens et al. (1994); Vieux Fort— Heaman et al. (2004); Chilka Laka and Bolangir (India) — Krause et al. (2001); Rogaland — Schärer et al. (1996), Andersen & Griffin (2004); Uluguru (Tanzania) — Tenczer et al. (2006). The grey band indicate timing of orogenic events; Labradorian and Pinawarian orogenies from Gower & Krogh (2002), Shawinigan, Ottawan and Rigolet orogenies from Rivers (1997) and Rivers et al. (2002). 700 .1-i C 0 2100 2300 Amanunat X. . : Q C]) Lofoten (Norway) .9 Korosten (Ukraine) G Horse Creek (USA) Lanying + Damiao (China) Manlcouagan Imbricate Zone IDD Wigborg (Finland/Russia) / Mealy Mtns Saimi (Russia) )l< Bengal (India) ( Mazury (Poland) Wolf River (USA) )I( Kunene (Namibia/Angola) 0 Jonhoping (Sweden) X Harp Lake+Michikamau O Lararnle Rivière Pentecôte XXX Nain De Ia Blache Lac Saint-Jean + Mattawa Adirondacks Morin Oaxacari (Mexico))K iT Atikonak 11 Havre-Saint-Pierre Saint-Urbain Montpeiier (USA) Labilevilie *( )I( Chilka Laka (India) 1eux Fort © Rogaland (Norway) ( Uiuguru (Tanzania) Grenville X Nain D Horse Creek/Laramie/ Wolf River 0 Scandinavia + Europe X Asia + Africa + Mexico C C 0 I— C .! 700 900 1100 1300 1500 1700 1900 2100 2300 64 Hanmer, 1997). Convective thinning, or delamination of the lithosphere, was followed by uplift and extension, and would have lead to partial melting of upwelling asthenospheric mantle thus producing the parent magmas to AMCG suites. This mechanism for formation of the AMCG parent magmas is consistent with Pb-Sr-Nd-Hf isotopic evidence from the Saint-Urbain and Lac Allard massifs, and from Proterozoic Grenvillian anorthosites in general, for an upper mantle source that evolved with time during the Proterozoic (Chapter 4 and 5). The new ages from the Saint-Urbain anorthosite (ca. 1055 Ma) and the Lac Allard lobe (ca. 1060 Ma) indicate emplacement of these two intrusions during the Ottawan orogeny, which supports the proposal of Corrigan & Hanmer (1997) that AMCG magmatism can be generated in a convergent margin tectonic setting. The documented slow cooling rates (3-4°C/m.y.) found for both intrusions is consistent with emplacement into an active orogenic terrane. 2.8- Conclusions The age and cooling path of two Proterozoic anorthosite-mangerite-charnockite granite (AMCG) suites (Saint-Urbain and Lac Allard) in the polycyclic allochthonous belt of the Grenville Province were determined by U-Pb zircon/rutile/apatite and 4O,39 biotite/plagioclase geochronology from the anorthosites and their respective Fe-Ti oxide ore deposits. U-Pb zircon crystallization ages (Lac des Cygnes anorthosite: 1055.0 ± 2.4 Ma, Bignell anorthosite: 1053.6 ± 2.6 Ma, and megacrystic leuconorite: 1046.2 ± 3.1 Ma) reveal that the small (—-450 2) Saint-Urbain anorthosite massif is a composite intrusion. The Saint-Urbain anorthosite post-dates crystallization of a 1057.4 ± 1.5 Ma layered oxide-apatite gabbronorite on the margin of the intrusion and the 1060.8 ± 2.8 Ma Saint-Anne du Nord orthopyroxene granodiorite. These intrusions belong to the 1080-1045 Ma magmatic period in this part of the Grenville Province (Hébert et al., 2005) and are coeval with Grenvillian metamorphism of the Ottawan orogeny. Crystallization ages for anorthosites from the Lac Allard lobe of the large (11,000 2) Havre-Saint-Pierre anorthosite suite are 1060.5 ± 1.9 Ma, 1061.6 ± 3.0 Ma and 1057.4 ± 5.7 Ma, which are similar to the 1062 ± 4 Ma Rivière-au-Tonnere massif (van Breemen & Higgins, 1993). These results indicate that emplacement of the Havre-Saint 65 Pierre suite occurred over 60 million years. The Big Island massive Fe-Ti oxide dyke is slightly younger (1052.9 ± 6.5 Ma), which is supported by a crosscutting relationship with the host anorthosite. Emplacement of the Lac Allard lobe is coeval with regional metamorphism of the Ottawan orogeny. The results of this study allow for estimation of the cooling rates for the Saint-Urbain and Lac Allard anorthosites of 3-4°C/m.y. These slow cooling rates are interpreted to have resulted from emplacement of the massifs into active orogenic terranes. Similar cooling rates have been documented in other parts of the Grenville Province suggesting that the massifs cooled during slow unroofing of the terrane. These two massifs are part of an extended 1160 to 975 Ma AMCG magmatic event, which is coeval with continent- continent collision of the Grenville orogeny and which produced some of the largest anorthosite massifs (e.g. Lao Saint-Jean, Havre-Saint-Pierre) in the Grenville Province. 2.9- Acknowledgements Rio Tinto Iron and Titanium Inc provided all logistical support in the field as well as major financial support for the analytical component of this study. H. Lin carried out much of the mineral separation and provided assistance with grain abrasion and mass spectrometry. Assistance in the clean lab for U-Pb geochronology was provided by R. Lishansky. T. Ullrich is gratefully acknowledged for the careful Ar-Ar analyses and data reduction. C.-E. Morisset was supported by a NSERC PGS-B scholarship. This research was supported by a NSERC CRD to J.S. Scoates and D. Weis and by NSERC Discovery Grants to J.S. Scoates and D. Weis. 66 2.10- References Aleinikoff, J.N., Wright Horton Jr., J. & Walter, M. (1996): Middle Proterozoic age for the Montpelier anorthosite, Goochland terrane, eastern Piedmont, Virginia. GSA Bulletin 108, 1481-1491. Amelin, Y.V., Heaman, L.M., Verchogliad, V.M. & Skobelev, V.M. (1994): Geochronological constraints on the emplacement history of an anorthosite — rapakivi granite suite: U-Pb zircon and baddeleyite study of the Korosten complex, Ukraine. Contributions to Mineralogy and Petrology 116, 411-419. Amelin, Y.V., Larin, A.M. & Tucker, R.D. (1997): Chronology of multiphase emplacement of the Salmi rapakivi granite-anorthosite complex, Baltic Shield: implications for magmatic evolution. Contributions to Mineralogy and Petrology 127, 3 53-368. Andersen, T. & Griffin, W. L. (2004): Lu-Hf and U-Pb isotope systematics of zircons from the Storgangen intrusion, Rogaland Intrusive Complex, SW Norway: implications for the composition and evolution of Precambrian lower crust in the Baltic Shield. Lithos 73, 27 1-288. Ashwal, L.D. & Wooden, J.L. (1983): Isotopic evidence from the eastern Canadian shield for geochemical discontinuity in the Proterozoic mantle. Nature 306, 679-680. Aviola, R., Johanson, B.S., Rämö, O.T, Vaasjoki, M. (1999): The proterozoic Ahvenisto rapakivi granite—massif-type anorthosite complex, southeastern Finland; petrography and U-Pb chronology. Precambrian Research 95, 89-107. Berg, J.H., Emslie, R.F., Hamilton, M.A., Morse, S.A., Ryan, A.B. & Wiebe, R.A. (1994): Anorthositic, granitoid and related rocks of the Nain Plutonic Suite. IGCP #290, #315 Excursion — Nain area (ed. R.F. Emslie). pp 69. Brander, L. & SOderlund, U. (2008): Mesoproterozoic (1.47-1.44 Ga) orogenic magmatism in Fennoscandia; Baddeleyite U-Pb dating of a suite of massif-type anorthosite in S. Sweden. International Journal of Earth Sciences. DO! 10.1007/s0053 1-007-0281-0. Chatterjee, N., Crowley, J.L., & Ghose, N.C. (2008): Geochronology of the 1.55 Ga Bengal anorthosite and Grenvillian metamorphism in the Chotanagpur gneissic complex, eastern India. Precambrian Research 161, 303-316. Chamberlain, K.R. & Bowring, S.A. (2000): Apatite-feldspar U-Pb thermochronometer: a reliable, mid-range (450 °C), diffusion-controlled system. Chemical Geology 172, 173-200. Cherniak, D.J. (2000): Pb diffusion in rutile. Contributions to Mineralogy and Petrology 139, 198-207. Cherniak, D.J. & Watson, E.B. (2000): Pb diffusion in zircon. Chemical Geology 172, 5- 24. Cherniak, D.J., Landford, W.A, & Ryerson, F.J. (1991): Lead diffusion in apatite and zircon using ion implantation and Rutherford Backscattering techniques. Geochimica et Cosmochimica Acta 55, 1663-1673. Corfu, F. (2004): U-Pb age, setting and tectonic significance of the anorthosite mangerite-charnockite-granite suite, Lofoten-Vesterâlen, Norway. Journal of Petrology 45, 1799-1819. 67 Corrigan, D & Hanmer, S. (1997): Anorthosites and related granitoids in the Grenville orogen: A product of convecting thinning of the lithosphere. Geology 25, 61-64. Corriveau, L., Perrault, S. and Davidson, A. (2007): Prospective metallogenic settings of the Grenville Province. In: Mineral Deposits of Canada: A Synthesis of Major Deposit-types, District Metallogeny, the Evolution of Geological Provinces, and Exploration Methods (ed. W.D. Goodfellow). Geological Survey of Canada, Mineral Deposits Division, Special Publication 5, 819-847. Davidson, A. (1998): An overview of the Grenville Province geology, Canadian shield. In: Geology of the Precambrian Superior and Grenville Provinces and Precambrian Fossils in North America (eds S.B. Lucas & M.R. St-Onge). Geological Survey of Canada, Geology ofCanada 7, 205-270. Davidson, A. (2008): Late Paleoproterozoic to mid-Neoproterozoic history of northern Laurentia: An overview of central Rodinia. Precambrian Research 160, 5-22. Dickin A.P. (2000): Crustal formation in the Grenville Province: Nd-isotope evidence. Canadian Journal ofEarth Sciences 37, 165-18 1. Dodson, M.H., 1973. Closure temperature in cooling geochronological and petrological systems. Contributions to Mineralogy and Petrology 40, 259-274. Doig, R. (1991): U-Pb zircon ages of Morin anorthosite suite rocks, Grenville Province, Québec. Journal of Geology 99, 729-738. Dörr, W., Belka, Z., Marheine, D., Schastok, J., Valverde-Vaquero, P. & Wiszniewska, J. (2002): U-Pb and Ar-Ar geochronology of anorogenic granite magmatism of the Mazury complex, NE Poland. Precambrian Research 119, 101-120. DrUppel, K., Littmann, S., Romer, R.L. & Okrusch, M. (2007): Petrology and isotope geochemistry of the Mesoproterozoic anorthosite and related rocks of the Kunene Intrusive Complex, NW Namibia. Precambrian Research 156, 1-31. Duchesne, J.C. (1966): Separation rapide des minéraux des roches. Annales de la Société géologique de Belgique 87, 347-3 56. Dymek R. F. (2001): An Introduction to the CRUML Belt ofMassifAnorthosites, Grenville Province, Québec. Excursion guide, Friends of Grenville (FOG) Meeting. pp 50. Emslie R. F. & Hunt P. A. (1990): Ages and petrogenetic significance of igneous mangerite-charnockite suites associated with massif anorthosites, Grenville Province. Journal ofGeology 102, 539-558. Emslie, R.F. & Loveridge, W.D. (1992): Fluorite-bearing early and middle Proterozoic granites, Okak Bay area, Labrador: geochronology, geochemistry, and petrogenesis. Lithos 28, 87-109. Frost, C.D., Meier, M. & Oberli, F. (1990): Single-crystal U-Pb zircon age determination of the Red Mountain pluton, Laramie Anorthosite Complex, Wyoming. American Mineralogist 75, 2 1-26. Gerstenberger, H. and Haase, G. (1997): A highly effective emitter substance for mass spectrometric Pb isotope ratio determinations. Chemical Geology 136, 309-312. Gobeil, A., Hébert, C., Clark, T., Beaumier, M. & Perreault, 5. (2002) : Géologie de Ia region du lac De La Blache (22K103 et 22K104). Ministère des Ressources naturelles, Québec, RG 2002-0 1, 53p. Gobeil, A., Brisebois, D., Clark, T., Verpaelst, P., Madore, L., Wodicka, N. & Chevé, S. (2003): Géologie de Ia moyenne Côte-Nord. In: Géologie et ressources minérales da 68 la partie est de la Province de Grenville (eds D. Brisebois & T. Clark). Ministère des Ressources naturelles, de la Faune et des Parcs, Québec, DV 2002-03, 9-58. Gower, C.F. (1996): The evolution of the Grenville Province in eastern Labrador, Canada. In: Precambrian Crustal Evolution in the North Atlantic Region (ed. T.S. Brewer). Geological Society Special Publication 112, 197-218. Gower, C.F. & Krogh, T.E. (2002): A U-Pb geochronological review of the Proterozoic history of the eastern Grenville Province. Canadian Journal ofEarth Sciences 39, 795-829. Hamilton, M.A., Ryan, A.B., Emslie, R.F. & Ermanovics, I.F. (1998): Identification of Paleoproterozoic anorthositic and monzonitic rocks in the vicinity of the Mesoproterozoic Nain Plutonic Suite, Labrador. Geological Survey ofCanada, Current Research 1998-F, 23-40. Hanes, J.A., Clark, S.J. & Archiblad, D. A. (1988): An 40Ar/39rgeochronological study of the Elzevir batholith and its bearing on the tectonothermal history of the southwestern Grenville Province, Canada. Canadian Journal ofEarth Sciences 25, 1834-1845. Harrison, T.M., Ducan, I. & McDougall, I. (1985) Diffusion of40Ar in biotite, temperature, pressure and compositional effects. Geochimica et Cosmochimica Acta 49, 2461-2468. Heaman, L.M., Gower, C.F. & Perreault, S. (2004): The timing of Proterozoic magmatism in the Pinware terrane of southeast Labrador, easternmost Québec and northwest Newfoundland. Canadian Journal ofEarth Sciences 41, 127-150. Hébert, C. (2001): Géologie de la region de Lac Mélonèze. Ministère des Ressources naturelles Québec, RG 2001-02. pp.25 Hébert, C, & Lacoste, P. (1998): Geologie de la region de Jonquiere-Chicoutimi (SNRC 22D/06). Ministère des Ressources naturelles Québec, RG 96-08. pp 31. Hébert, C. & van Breemen, 0. (2004a): Redefinition of the Lac-St.-Jean Anorthosite, Central Grenville Province, Québec, based on compositional, structural, geochronological, and mineral deposit features. EOS Transactions AGU, 85-17, Joint Assembly Supplementary, Abstract V53A-03. Hébert, C. & van Breemen, 0. (2004b): Mesoproterozoic basement of the Lac St. Jean Anorthosite Suite and younger Grenvillian intrusions in the Saguenay region, Québec: Structural relationships and U-Pb geochronology. In: Proterozoic Tectonic Evolution of the Grenville Orogen in North America (eds R.P. Tollo, L. Corriveau, J.M. McLeland & M.J. Bartholomew). Geological Society ofAmerica, Memoir 197, 65-80. Hervet, M., van Breemen, 0. & Higgins, M.D. (1994): U-Pb igneous crystallization ages of intrusive rocks near the southeastern margin of the Lac-St-Jean anorthosite complex, Grenville Province, Québec. Current Research - Geological Survey of Canada, Report: 1994-F, 115-124. Hébert, C, van Breemen, 0. & Lacoste, P. (1998): Tectonic setting and U-Pb zircon age of the Poulin-de-Courval mangerite, Saguenay-Lac Saint-Jean area, Grenville Province, Québec. Current Research - Geological Survey of Canada, Report: 1998-F, 69-76. Hébert, C, Cadieux, A.-M. & van Breemen, 0. (2005): Temporal evolution and nature of Ti-Fe-P mineralization in the anorthosite-mangerite-charnockite-granite (AMCG) 69 suites of the south-central Grenville Province, Saguenay — Lac St. Jean area, Québec, Canada. Canadian Journal ofEarth Sciences 42, 1865-1880. Higgins, M.D. & van Breemen, 0. (1992): The age of the Lac-Saint-Jean Anorthosite Complex and associated mafic rocks, Grenville Province, Canada. Canadian Journal ofEarth Sciences 29, 1412-1423. Higgins, M.D. & van Breemen, 0. (1996): Three generations of anorthosite-mangerite charnockite-granite (AMCG) magmatism, contact metamorphism and tectonism in the Saguenay-Lac-St.-Jean region of the Grenville Province, Canada. Precambrian Research 79, 327-346. Higgins, M.D., Ider, M. & van Breemen, 0. (2002): U-Pb ages of plutonism, wollastonite formation, and deformation in the central part of the Lac-Saint-Jean anorthosite suite. Canadian Journal ofEarth Sciences 39, 1093-1105. Hocq M. (1982): Region du lac Allard. Ministère de l’énergie et des ressources, Québec. pp 99. Hodges, K.V. (2003): Geochronology and thermochronology in orogenic systems. In: Treatise on Geochemistry 3, 263-292. Hoffman, P.F. (1989): Precambrian geology and tectonic history of North America. In: The Geology of North America; an overview (eds A. W. Bally & A. R. Palmer) The Geology ofNorth America, Geological Society ofAmerica, 447-512. Icenhower, J.P., Dymek, R.F. & Weaver, B.L. (1998): Evidence for an enriched mantle source for jøtunite (orthopyroxene monzodiorite) associated with the St. Urbain anorthosite, Quebec. Lithos 42, 191-212. Indares, A., Dunning, G., Cox, R., Gale, D. & Connelly, J. (1998): High-pressure, high- temperature rocks from the base of thick continental crust: Geology and age constraints from the Manicouagan Imbricate Zone, eastern Grenville Province. Tectonics 17, 426-440. Kelley, S.P., Wartho, J.A. & Elphick, S.C. (2002): Experimantal determination of Ar diffusion and solubility in plagioclase and leucite. Geochimica et Cosmochimica Acta 66-15A, 390. Keppie, J.D., Dostal, J., Cameron, K.L., Solari, L.A., Ortega-Gutiérrez, F. & Lopez, R. (2003): Geochronology and geochemistry of Grenvillian igneous suites in the northern Oaxacan Complex, southern Mexico: tectonic implications. Precambrian Research 120, 365-389 Krause, 0., Dobmeier, C., Raith, M.M. & Mezger, K. (2001): Age of emplacement of massif-type anorthosites in the Eastern Ghats Belt, India: constraints from U-Pb zircon dating and structural studies. Precambrian Research 109, 25-38. Kretz, R. (1983): Symbols for rock-forming minerals. American Mineralogist 68, 277- 279. Krogh, T.E. (1982): Improved accuracy of U-Pb zircon ages by the creation of more concordant systems using an air abrasion technique. Geochimica et Cosmochimica Acta 46, 637-649. Krogh, T.E. & Davis, G.L. (1973): The significance of inherited zircons on the age and origin of igneous rocks — an investigation of the Labrador adamellites. Carnegie Institute Washington Year Book 72, 610-613. Li, X.L., Bogdanova, S.V., Collins, A.S., Davidson, A., De Waele, B., Ernst, R.E., Fitzsimons, I.C.W., Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karlstrom, K.E., Lu, 70 S., Natapov, L.M., Pease, V., Pisarevsky, S.A., Thrane, K., Vernikovsky, V. (2008): Assembly, configuration, and break-up history of Rodinia: A synthesis. Precambrian Research 160, 179-2 10. Loveridge, W.D. (1986): U-Pb ages on zircon from rocks of the Lar Morhiban map area, Quebec. In: Current Research, part A. Geological Survey of Canada, paper 86-lA, 523-530. Ludwig, K.R. (1998): On the treatment of concordant uranium-lead ages. Geochimica et Cosmochimica Acta 62, 665-676. Ludwig, K.R. (2003): Isoplot 3.09, A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center, Special Publication No. 4. Madore, L., Verpaelst, P., Brisebois, D, Hocq, M. & Dion, D.J. (1999): Géologie de la region du lac Allard (SNRC 12L/1 1). Ministère des Ressources naturelles, Québec. RG 98-0 1, pp37. Mattinson, J.M. (2005): Zircon U-Pb chemical abrasion (“CA-TIMS”) method: combined annealing and multi-step partial dissolution analysis for improved precision and accuracy of zircon ages. Chemical Geology 220, 47-66. Martignole, J. & Reynolds, 0. (1997): 40Ar/39r thermochronology along a western Quebec transect of the Grenville Province, Canada. Journal ofMetamorphic Geology 15, 283-296. Mayer, A., HofiTlann, A.W., Sinigoi, S., & Morais, E. (2004): Mesoproterozoic Sm-Nd and U-Pb ages for the Kunene Anorthosite Complex of SW Angola. Precambrian Research 133, 187-206. McLelland, J.M., Bickford, M.E., Hill, B.M., Clechenko, C.C., Valley, J.W. & Hamilton, M.A. (2004): Direct dating of Adirondack massif anorthosite by U-Pb SHRIMP analysis of igneous zircon: implications for AMCG complexes. Geological Society of America Bulletin 116-11/12, 1299-1317. Mundil, R., Ludwig, K.R., Metcalfe, I., and Renne, P.R. (2004): Age and timing of the Permian mass extinctions: U/Pb dating of closed-system zircons. Science 305, 1760- 1763. Owens, B.E. & Tomascak, P.B. (2002) Mesoproterozoic lamprophyres in the Labrieville Massif, Québec: clues to the origin of alkalic anorthosites? Canadian Journal of Earth Sciences 39, 983-997. Owens B.E., Dymek R.F., Tucker R.D., Brannon J.C. & Podosek F.A. (1994): Age and radiogenic isotopic composition of a late- to post-tectonic anorthosite in the Grenville Province: the Labrieville massif, Québec. Lithos 31, 189-206. Parrish, R., Roddick, J.C., Loveridge, W.D., and Sullivan, R.W. (1987): Uranium-lead analytical techniques at the geochronology laboratory, Geological Survey of Canada. In Radiogenic Age and Isotopic Studies, Report 1, Geological Survey of Canada, Paper 87-2, 3-7. Perreault, S. (2003): Contrasting styles of Fe-Ti mineralization in the Havre-Saint-Pierre anorthosite suite, Quebec’s North Shore, Canada. In: Ilmenite Deposits and Their Geochemical Environment, Volume Special Publication No. 9. (eds J.C. Duchesne & A. Korneliussen), Norges Geologiske Undersøkelse. 87-90. Reiners, P.W. & Brandon, M.T. (2006): Using thermochronology to understand orogenic erosion. Annual Review ofEarth and Planetary Science 34, 419-466. 71 Renne, P.R., Swisher, C.C., Demo, A.L., Kamer, D.B., Owens, T.L. and DePaolo, D.J. (1998): Intercalibration of standards, absolute ages and uncertainties in40Ar/39r dating. Chemical Geology 145, 117-152. Rivers, T. (1997): Lithotectonic elements of the Grenville Province: review and tectonic implications. Precambrian Research 86, 117-154. Rivers, T. & Corrigan, D. (2000): Convergent margin on southeastern Laurentia during the Mesoproterozoic: tectonic implications. Canadian Journal ofEarth Sciences 37, 359-383. Rivers, T., Martignole, J., Gower, C.F. & Davidson, A. (1989): New tectonic divisions of the Grenville Province, southeast Canadian Shield. Tectonics 8, 63-84. Rivers, T. & Ketchum, J., Indares, A. & Hynes, A. (2002): The high pressure belt in the Grenville Province: architecture, timing, and exhumation. Canadian Journal ofEarth Sciences 39, 867-893. Roddick, J.C. (1987): Generalized numerical error analysis with application to geochronology and thermodynamics. Geochimica et Cosmochimica Acta 51, 2129-2135. Rondot, J. (1971): Impactite of the Charlevoix Structure, Québec, Canada. Journal of Geophysical Research 76, 54 14-5423. Rondot, J. (1989): Géologie de Charlevoix. Ministère de l’énergie et des ressources, Québec. MB 89-21. 606 p. Schärer, U., Wilmart, E. & Duchesne, J.-C. (1996): The short duration and anorogenic character of anorthosite magmatism: U-Pb dating of the Rogaland complex, Norway. Earth and Planetary Science Letters 139, 335-350. Schoene, B. & Bowring, S.A. (2006): U-Pb systematics of the McClure Mountain syenite: thennochronological constraints on the age of the40Ar/39r standard MMhb. Contributions to Mineralogy and Petrology 151, 615-630. Scoates J.S. & Chamberlain K.R. (1995): Baddeleyite (ZrO2)and Zircon (ZrSiO2)from anorthositic rocks of the Laramie anorthosite complex, Wyoming: Petrologic consequences and U-Pb ages. American Mineralogist 80, 1317-1327. Scoates, J.S. & Chamberlain, K.R. (1997): Orogenic to post-orogenic origin for the 1.76 Ga Horse Creek anorthosite Complex, Wyoming, USA. Journal of Geology 105, 33 1-343. Scoates, J.S. & Chamberlain, K.R. (2003): Geochronologic, geochemical and isotopic constraints on the origin of monzonitic and related rocks in the Laramie anorthosite complex, Wyoming, USA. Precambrian Research 124, 269-3 04. Scoates, J.S. & Friedman, R.F. (in press): Precise age of the platiniferous Merensky Reef, Bushveld Complex, South Africa, by the U-Pb zircon chemical abrasion ID-TIMS technique. Economic Geology. Simmons, K.R. & Simmons, E.C. (1987): Petrogenetic implications of Pb- and Sr isotopic compositions for rocks from the Nain anorthosite complex, Labrador. Abstracts with programs - Geological Society of America 19, 845. Simmons, K.R., Wiebe, R.A., Snyder, G.A. & Simmons, E.C. (1986): U-Pb zircon age for the Newark Island layered intrusion, Nain anorthosite complex, Labrador. Abstracts with programs - Geological Society of America 18, 751. Stacey, J.S. & Kramers, J.D. (1975): Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters 26, 207-221. 72 Tenczer, V., Hauzenberger, C.A., Fritz, H., Whitehouse, M.J., Mogessie, A., Walibrecher, E., Muhongo, S. & Hoinkes, G. (2006): Anorthosites in the Eastern Granulites of Tanzania — New SIMS zircon U-Pb age data, petrography and geochemistry. Precambrian Research 148, 85-114. Trepman, C.A., Götte, T. & Spray, J.G. (2005): Impact-related Ca-metasomatism in crystalline target-rocks from the Charlevoix structure, Quebec, Canada. Canadian Mineralogist 43, 553-567. Vaasjoki, M., Rämö, O.T. & Sakko, M. (1991): New U-Pb ages from the Wiborg rapakivi area: constraints on the temporal evolution of the rapakivi granite anorthosite-diabase dyke association of southeastern Finland. Precambrian Research 51, 227-243. van Breemen 0. & Higgins M. D. (1993): U-Pb zircon age of the southwest lobe of the Havre-Saint-Pierre Anorthosite Complex, Grenville Province, Canada. Canadian Journal ofEarth Sciences 30, 1453-1457. Van Schmus, W.R., Medaris, JR., L.G. & Banks, P.O. (1975): Geology and age of the Wolf River batholith, Wisconsin. Geological Society ofAmerica Bulletin 86, 907- 914. Verts, L.A., Chamberlain, K.R. & frost, C.D. (1996): U-Pb sphene dating of metamorphism: the importance of sphene growth in the contact aureole of the Red Mountain pluton, Laramie mountains, Wyoming. Contributions to Mineralogy and Petrology 125, 186-199. Wodicka, N., David, J., Parent, M., Gobeil, A. & Verpaelst, P. (2003): Geochronologie U-Pb et Pb-Pb de la region de Sept-Iles—Natshquan, Province de Grenville, moyenne Côte-Nord. In: Géologie et ressources minérales da la partie est de la Province de Grenville (eds D. Brisebois & T. Clark). Ministére des Ressources naturelles, de la Faune et des Parcs, Québec, DV 2002-03. 59-118. Yu, Y. & Morse, S.A. (1992): Age and cooling history of the Kiglapait Intrusion from an 4O39 study. Geochemica et Cosmochimica Acta 56, 2471-2485. Zhang, S.-H., Liu, Shu-Wen, Zhao, Y., Yang, J-H., Song, B. & Liu, X.-M. (2007): The 1.75-1.68 Ga anorthosite-mangerite-alkali granitoid-rapakivi granite suite from the northern North China Craton: Magmatism related to a Paleoproterozoic orogen. Precambrian Research 155, 287-3 12. 73 Chapter III Origin of zircon rims around ilmenite in mafic plutonic rocks of Proterozoic anorthosite suites1 ‘A version of this chapter has been published. Morisset, C.E. & Scoates, J.5. (2008): Origin of zircon rims around ilmenite in mafic plutonic rocks of Proterozoic anorthosite suites. Canadian Mineralogist 46, 289- 304. 74 3.1- Introduction Ilmenite and zirconium-bearing minerals occur in close association in a diverse range of igneous to metamorphic rocks. Baddeleyite (Zr02) is found either as lamellae in ilmenite or as blebs around ilmenite grains within the Basistoppen sill, east Greenland (Naslund 1987), in the Stillwater complex, Montana (Loferski & Arculus 1993), in the Rum layered intrusion, Scotland (Kersting et a!. 1989), and in mafic granulite and amphibolite rocks of the Proterozoic Lindâs Nappe, western Norway (Bingen et a!. 2001). Srilankite (Ti2ZrO6)and ilmenite have been observed together in oceanic gabbros (Morishita et al. 2004) and Norwegian granulitic rocks (Bingen eta!. 2001). Zircon forms a rim around ilmenite and hemo-ilmenite (FeTiO3- e20;ilmenite with exsolution lamell of hematite commonly referred to as hemo-ilmenite) in the metamorphic rocks of the Lindâs Nappe (Bingen eta!. 2001) and in the Tellnes Fe-Ti oxide deposit, Southwest Norway (Charlier et a!., 2007). Thompson & Peck (2003) have also described the presence of a thin rim of zircon around ilmenite in the Fe-Ti oxide deposits of the Morin anorthosite, Grenville Province, Canada. In this study, we examined samples from five different igneous intrusions that each contain rock types featuring ilmenite, hemo-ilmenite or rutile (Ti02)having a discrete rim of zircon. The samples are mostly Fe-Ti oxide ores from Proterozoic anorthosite massifs and layered intrusions of the Grenville Province (Québec, Canada), ranging from massive ores (>90 vol% oxides) through semi-massive (25-90 vol%) to disseminated (<25 vol%) oxide-rich rocks. The emplacement and crystallization of each intrusion postdates regional metamorphism. We examine various hypotheses for the origin of a zircon rim, including precipitation from late-hydrothermal fluids, crystallization from evolved high-temperature interstitial liquid, oxidation-exsolution of ilmenite to produce baddeleyite lamellae, and diffusion of trace amounts of zirconium (Zr) through the structure of ilmenite and other Ti-based oxides to the grain boundaries. 75 3.2- Locality descriptions We studied 10 different samples from five Fe-Ti oxide deposits related to Proterozoic anorthosite massifs and gabbroic intrusions (four in the Grenville Province of eastern Canada and one in the western United States) (Figure 3.1). Each sample contains ilmenite rimmed by zircon; rutile is present in 4 of the 10 samples. Zircon rims around ilmenite, with or without rutile, occur in. Below, we provide a brief description of the geology, age and mineralogy of each location to characterize the lithologic association and relative thermal histories of the samples studied. At the easternmost locality (Figure 3.1), the Big Island deposit is a dyke of massive Fe-Ti oxide that intrudes the Lac Allard lobe (southeast part) of the 11,000 km2 Havre Saint-Pierre anorthositic suite. This anorthositic suite is divided into lobes that range in age from 1129 ± 3 Ma (U-Pb zircon; Wodicka et at. 2003) for the northwestern part and 1062 ± 4 Ma (U-Pb zircon; van Breemen & Higgins 1993) for the southwestern part. A preliminary U-Pb zircon age for the Lac Allard lobe is ca. 1060 Ma (Morisset, unpublished data), about 20 m.yr. after peak metamorphism in the region, which is dated at 1082 ± 16 Ma from metamorphic zircon in the older part of the anorthosite massif (Wodicka et at. 2003). The mineralogy of the Big Island dyke consists of hemo-ilmenite (19> Xhem < 28), rutile (Ti02>97.9 wt%; Table 3.1), plagioclase (An3950), orthopyroxene (71 > Mg# < 78; Mg# = Mg/(Mg+Fe2)),hercynite, sapphirine and biotite, as well as trace amounts of apatite, corundum, sulphide, and grains of zircon (mineral compositions from Morisset, unpublished data). Rutile has two textural habits: (1) primary cumulus rutile grains (Morisset et a!. 2006), and (2) secondary rutile, present as irregular lenses with hematite within ilmenite grains, which is formed by oxidation (e.g., Haggerty 1976). All samples are characterized by an equigranular texture suggesting that the rocks have undergone high-temperature recrystallization during slow cooling. Samples from the second locality are from the Lac Mirepoix layered complex within the 200 km2 Mattawa anorthosite (1016 ± 2 Ma: U-Pb zircon; Hébert eta!. 2005), which is located east of the giant Lac Saint-Jean anorthositic suite (Owens & Dymek 2005) (Figure 3.1). The crystallization sequence of the Mirepoix layered intrusion is defined by plagioclase (An4153)and hemo-ilmenite (5 <Xhem> 24), followed by the appearance of 76 Table 3.1 Rutile compositions determined by electron microprobe Saint-Urbain Big Island Sample 2006-DI 2006-D2 2006-D3 2006-D8 2006-D9 2102-1 2102-2 2102-3 2102-4 rim core rim core rim grain grain L grain SiO2wt% 0.01 bd bd 0.01 0.01 0.05 0.04 0.01 0.01 Ti02 99.78 99.15 99.70 99.48 99.77 99.11 98.38 99.58 97.94 A1203 bd bd bd bd bd bd bd bd bd v203 bd bd bd bd bd bd bd bd bd Cr203 0.32 0.33 0.32 0.31 0.33 0.10 0.07 0.09 0.09 Fe203 0.15 0.17 0.27 0.24 0.20 0.22 0.42 0.21 1.05 MgO bd bd bd bd bd bd bd bd bd CaO bd bd bd bd 0.01 0.18 0.24 0.16 0.16 MnO bd bd bd bd bd bd bd bd bd ZnO bd bd bd bd bd 0.01 bd bd 0.01 Na20 bd 0.04 0.03 bd 0.05 0.04 bd 0.04 0.01 Zr02 0.06 0.13 0.04 0.18 0.09 0.06 0.13 0.06 0.06 Hf02 bd bd bd bd bd bd bd bd bd Nb205 0.03 0.03 bd 0.02 bd 0.01 0.40 0.02 0.03 Total 100.43 99.93 100.39 100.30 100.46 99.78 99.69 100.16 99.41 Siapfu 0.000 - - 0.000 0.000 0.001 0.001 0.000 0.000 Ti 0.996 0.995 0.995 0.995 0.995 0.995 0.992 0.996 0.990 Al - - - - - - V - - - - - - - - - Cr 0.003 0.003 0.003 0.003 0.003 0.001 0.001 0.001 0.001 Fe3+ 0.001 0.002 0.003 0.002 0.002 0.002 0.004 0.002 0.011 Mg - - - - - - - - Ca - - - 0.000 0.00.3 0.004 0.002 0.002 Mn - - - - - Zn - - - - - 0.000 - - 0.000 Na - 0.001 0.000 - 0.001 0.001 - 0.001 0.000 Zr 0.000 0.001 0.000 0.001 0.001 0.000 0.001 0.000 0.000 Hf - - - - - - - - - Nb 0.000 0.000 - 0.000 - 0.000 0.002 0.000 0.000 Cation suni 1.002 1.002 1.002 1.002 1.002 1.003 1.005 1.002 1.005 Zr (ppm) 455 979 318 1322 634 452 925 440 419 T(°C) 678 750 647 782 708 677 744 675 670 Cations are calculated on the basis of 2 oxygens. L (1amall within ferrian-ilmenite). Detection limit for Zr02 is 0.04 wt% (296 ppm Zr) and for Hf02 is 0.14 wt% (1696 ppm Hf), bd (below detection). T (°C) is calculated with the formula of Watson et al. (2006) where T (°C) = (44701(7.36-log Zr (ppm)))- 273. 77 Figure 3.1 Simplified map ofthe Grenville Province modified from Davidson (1998) showing Proterozoic anorthosite massifs and associated mangerites/granites and the localities of samples under study (star symbols). Inset in lower right part ofthe figure shows the location ofthis map in North America and locality 5. The localities are: (1) Big Island Fe-Ti oxide deposit in the Lac Allard anorthosite (Havre-Saint-Pierre anorthosite suite), Québec; (2) Mirepoix layered intrusion in the Mattawa anorthosite, Québec; (3) Saint-Urbain Fe-Ti oxide deposits in the Saint-Urbain anorthosite, Québec; (4) Methuen massive Fe-Ti oxide deposit in the Twin Lakes intrusion, Ontario; (5) Laramie anorthosite, southeast Wyoming (U.S.A.). 78 orthopyroxene (60 <Mg#> 73), clinopyroxene (69 <Mg#> 75) and magnetite with apatite (mineral compositions from Morisset 2002). Layers of massive hemo-ilmenite and nelsonite (a rock rich in Fe-Ti oxides and apatite) occur throughout the sequence. Samples examined for this study are composed mainly of equigranular ilmenite with trace amounts of plagioclase and biotite. The third locality is represented by the massive ilmenite and rutile deposits contained within the 450 km2 Saint-Urbain anorthosite (1079 ± 22 Ma: Sm-Nd mineral- whole rock; Ashwal & Wooden 1983; preliminary U-Pb zircon age of ca. 1054 Ma; Morisset, unpublished data) located about 45 km northeast of Québec City (Figure 3.1). The Saint Urbain deposits consist of eight discrete irregularly shaped bodies up to 50 metres across and 120 metres in length, some of which contain nelsonite (Dymek & Owens 2001). The mineralogy of the deposits is very similar to that of the Big Island dyke, with hemo ilmenite (14 <Xhem> 27; Table 3.2), rutile (TiO2>99.1 wt%; Table 3.1), plagioclase (An4450), orthopyroxene (71 <Mg#> 75), hercynite, sapphirine, biotite, apatite and sulphide (mineral chemistry from Morisset, unpublished data). Two different textures of rutile are also observed at this locality. Selected whole rock compositions for samples from the Saint-Urbain Fe-Ti oxide deposits are presented in Table 3.3. The fourth locality is the Methuen massive ilmenite deposit, which is contained within the 8 km2 Twin Lakes intrusive complex, southeastern Ontario (Ketchum et a!. 1988) (Figure 3.1). The central part of this complex is composed of gabbro, gabbronorite and oxide-rich cumulates, and the margins vary from quartz diorite to monzodiorite. Unlike the other intrusions in this study, this intrusion has no direct link to anorthosites. Based on relative age constraints, the Twin Lakes intrusive complex appears to be related to other —1 080 Ma diorite to monzodiorite intrusions of the Elzevir terrane in the Central Metasedimentary Belt of this part of the Grenville Province (e.g., Lumbers et a!. 1991). The Methuen deposit contains massive to disseminated hemo-ilmenite with minor magnetite (<5 vol. %). Along with the Fe-Ti oxides, the deposit rocks contain tabular plagioclase crystals (—An40), ortho- and clinopyroxene, quartz, biotite and sulphide. The well-preserved igneous textures of the Methuen rocks indicate that they have not been affected by subsequent metamorphism or deformation. 79 Table 3.2 Bulk hemo-ilmenite compositions determined by XRF, Saint-Urbain, Québec Sample 2006-D1 2009-B1 2015-B4 2033-E SiO2wt% 0.00 0.04 0.14 0.11 Ti02 45.27 44.62 47.39 44.36 A1203 bd 0.03 0.03 0.04 V203 0.27 0.29 0.30 0.28 Cr203 0.47 0.15 0.18 0.10 Fe203 53.56 55.32 52.02 55.84 MgO 2.99 2.73 3.22 2.88 CaO 0.04 0.01 0.02 0.08 MnO 0.21 0.16 0.16 0.21 ZnO 0.01 0.01 0.00 0.01 LOl -3.10 -2.82 -3.17 -3.10 Total 99.71 100.53 100.28 100.80 Fe203 14.46 16.30 11.22 17.44 FeO 35.17 35.11 36.71 34.56 Total 98.89 99.44 99.37 100.06 Si apfu 0.000 0.001 0.003 0.003 Ti 0.855 0.841 0.887 0.830 Al - 0.001 0.001 0.001 V 0.005 0.006 0.006 0.005 Cr 0.009 0.003 0.003 0.002 Fe3 0.273 0.307 0.210 0.326 Mg 0.112 0.102 0.120 0.107 Ca 0.00 1 0.000 0.00 1 0.002 Mn 0.004 0.003 0.003 0.004 Fe2 0.739 0.736 0.764 0.719 Zn 0.000 0.000 0.000 0.000 Cation sum 2.000 2.000 1.999 2.000 Xilm 84 83 88 82 Xhem 16 17 12 18 Ni ppm 234 286 229 165 Cr 3202 1058 1209 697 Co 52 84 108 65 Nb 0 1 19 20 Zr 11 24 53 6 Cations are calculated on the basis of 2 cations and Fe3 asFe2+Mg+Mn-Ti followed by a calculation based on 3 oxygens. Xilm (ilmlilm+hem); Xhem (hemlilm+hem); ilm (Ti-Mg-Mn+Al/2); hem (0.5*(Fe+Fe3Mg+Mn Ti) (Lindsley & Frost 1992). 80 Table 3.3 Selected whole rock compostitions determided bu XRF, Saint-Urbain, Québec 2006-Di 2009-B1 2015-B2 2033-D SiO2wt% 0.67 26.84 28.37 28.91 Ti02 53.79 25.29 23.05 23.45 A1203 1.24 13.75 14.10 14.85 Fe203 41.97 21.98 20.16 18.21 MnO 0.20 0.07 0.06 0.07 MgO 2.66 2.39 3.25 2.03 CaO 0.05 4.19 4,25 5.44 Na20 0.01 4.12 4.24 4.28 K20 0.01 0.35 0.38 0.31 0.03 0.17 0.38 0.05 LOl 0.87 0.16 0.36 1.00 Total 101.50 99.32 98,61 98.62 Rbppm 12.5 11,7 11.6 9,2 Sr 11 748 647 749 Y 10 9 10 6 Zr 693 245 272 282 Nb 40 20 20 21 Th 26 13 14 11 U 19 2 5 4 Pb 36 26 23 24 Co 43 40 49 39 Cu 43 19 37 19 Ga 91 75 66 63 Ni 386 155 182 92 Zn 375 117 73 87 81 The fifth locality is represented by a sample of olivine leucogabbro from the unmetamorphosed 1.43 Ga Laramie anorthosite complex (Wyoming, USA) (Scoates & Chamberlain 1995) (Figure 3.1, inset). The sample SR287 was collected from the leucogabbroic layered zone of the Poe Mountain massif, the northernmost of the anorthosite massifs in the Laramie complex. The sample consists predominantly of equigranular plagioclase (80 vol%, An51), olivine, clinopyroxene, and ilmenite (<5 vol%, Xhem = 1-4; Table 3.4), biotite, apatite and interstitial zircon. 3.3- Analytical methods The morphology, size and thickness of zircon rims from the five different intrusions were studied using back-scattered electron (BSE) images from a Philips XL-30 Scanning Electron Microscope (SEM) in the Department of Earth and Ocean Sciences at the University of British Columbia (accelerating voltage of 15 kV and beam current of 70 pA). Zircon, ilmenite and rutile compositions were determined using a Cameca SX 50 electron microprobe with four WDS detectors. Data reduction was carried out using the “PAP” I(pZ) method of Pouchou & Pichoir (1985). The accelerating voltage was set at 15 kV, and the beam current at 60 nA, and the beam diameter was 1 jim. For rutile and ilmenite, counting times were 20 seconds on the peak and 10 seconds on the background for Ca, Ti, Cr, Fe and Zn; 40 seconds on the peak and 20 seconds on the background for Na, Mg, Al, Si, V, and Mn; and 60 seconds on the peak and 30 seconds on the background for Zr, Nb and Hf. For zircon, counting times were set at 20 seconds on the peak and 10 seconds on the background for Si and Zr, and 60 seconds on the peak and 30 seconds on the background for Mg, Al, P, Ca, Ti, Fe, Hf, Th and U. The Th02,U02 and P205 contents in zircon are below detection limits (Table 3.5). Whole-rock and mineral separates were prepared and analyzed by XRF at the Department of Geology of the Université de Liege, Belgium. Ilmenite was separated by crushing the samples down to 60-160 jim to liberate the grains and then using heavy liquids (bromoform and heated Clerici solution) and a Frantz Isodynamic Separator following the method outlined in Duchesne (1966). After the purity of the separated material was verified under a binocular microscope, ilmenite was crushed to a powder in an agate mortar. Whole-rock 82 Table 3.4 Representative ilmenite compositions determined by electron microprobe, Laramie Anorthosite Complex, Wyoming. SR287-1 SR287-2 SR287-4 SR287-5 SR287-6 SR287-7 SR287-8 SR287-9 rim core rim core rim core rim core SiO2wt.% 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.02 Ti02 51.75 51.92 52.12 51.93 51.90 51.83 51.60 51.47 Al203 bd bd bd bd bd bd bd bd v203 bd bd bd bd bd bd bd bd Cr203 bd bd bd bd bd bd bd bd FeO 46.36 46.28 47.04 45.69 46.46 45.96 46.98 46.60 MgO 0.73 0.92 0.74 0.75 0.64 0.74 0.68 0.80 CaO 0.02 0.00 0.02 0.04 0.01 0.00 0.02 0.01 MnO 0.51 0.50 0,46 0.48 0.50 0.44 0.48 0.48 ZnO bd bd 0.06 bd 0.02 0.04 0.06 0.01 Zr02 bd bd bd bd bd bd bd bd Hf02 bd bd bd bd bd bd bd bd 205 0.04 0.06 0.05 0.05 0.06 0.06 0.07 0.08 Total 99,42 99.70 100.50 98.95 99.61 99.08 99.91 99.45 Fe203 1.72 2.02 2.16 0.90 1.60 1.24 2.53 2.46 FeO 44.74 44.55 45.09 44.88 45.02 44.85 44.71 44.38 Total 99.60 100.01 100.78 99.09 99.83 99.27 100.19 99.71 Si apfu 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Ti 0.983 0.980 0.978 0.990 0.984 0.987 0.975 0.976 Al - - - - - v3 - - - - - - Cr - - - - - - - - Fe3 0.033 0.038 0.041 0.017 0.030 0.024 0.048 0.047 Fe2 0.945 0.935 0,941 0.952 0,949 0.950 0.939 0.936 Mg 0.027 0.034 0.028 0.028 0.024 0.028 0.025 0.030 Ca 0.000 0.000 0.000 0.001 0.000 0.000 0.000 0.000 Mn 0,011 0.011 0.010 0.010 0,011 0.009 0.010 0.010 Zn - - 0.00 1 - 0.000 0.00 1 0.00 1 0.000 Zr - - - - Hf - - - - - - - - Nb 0.000 0.001 0.001 0.001 0.001 0.001 0.001 0.001 Cation sum 2.000 2.000 2.00 1 2.001 2.00 1 2.00 1 2.00 1 2.000 Xilm 98 98 98 99 98 99 98 98 Xhem 2 2 2 1 2 1 2 2 Calculation of mineral formulae and end-members as in Table 3.2. Detection limits as in Table 3.1. (bd) below detection. 83 Ta bl e 3.5 R ep re se nt at iv e zi rc on co m po sti on sd et er m in ed by el ec tro n m ic ro pr ob e Bi g Is la nd Sa in t-U rb ai n La ra m ie Sa m pl e 21 02 -2 21 02 4* 21 02 -5 21 02 11 * 21 09 -A -4 21 09 -A -5 21 09 -A -6 21 09 -A -8 20 06 -D i- 20 06 -D 1- 20 06 -D - 20 06 -D - SR 28 7- 3 SR 28 7- 6 1 5 12 13 S i O 2w t% 32 .4 1 32 .4 7 32 .5 6 32 .7 9 32 .0 0 32 .2 6 32 .5 0 32 .2 7 32 .5 6 32 .3 2 32 .4 5 32 .7 6 32 .2 1 32 .4 5 T i0 2 0.2 1 0. 03 0. 02 0. 06 0. 05 bd 0. 05 bd 0. 04 bd 0. 39 0. 31 0. 43 0. 04 A 1 203 0.0 1 bd 0.0 1 bd 0.0 1 bd 0.0 1 bd bd bd 0.0 1 bd bd 0.0 1 M gO 0.0 1 bd 0.0 1 bd bd 0.0 1 bd bd bd bd 0. 01 0. 19 0. 03 0. 02 Ca O 0. 06 0. 04 0. 03 0. 04 0. 01 0. 01 0.0 1 bd bd bd bd 0. 01 bd bd Fe O 0. 21 0. 06 0. 13 0. 10 0. 48 0. 18 0. 24 0. 04 0. 26 0. 15 0. 43 1. 19 1. 18 0. 20 Z r0 2 67 .1 3 67 .0 0 65 .3 9 66 .5 1 67 .7 9 67 .5 3 66 .7 1 67 .2 9 66 .7 4 67 .3 9 66 .2 3 66 .6 4 66 .4 5 66 .7 9 H f0 2 1. 06 0. 96 1.6 3 0. 97 0. 72 0, 52 0. 56 0. 63 0. 69 0. 68 0. 75 0. 95 1. 07 0. 64 T h0 2 bd bd bd bd bd bd bd bd bd bd bd bd bd bd U 0 2 bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd bd To ta l 10 1. 10 10 0. 57 99 .7 9 10 0. 49 10 1. 08 10 0. 53 10 0. 09 10 0. 26 10 0. 34 10 0. 56 10 0. 29 10 2. 06 10 1. 43 10 0. 16 Si ap fu 0. 98 6 0. 99 2 1. 00 2 0. 99 9 0. 97 7 0. 98 6 0. 99 5 0. 98 8 0. 99 9 0. 98 8 0. 99 2 0. 98 7 0. 97 9 0. 99 3 Ti 0. 00 5 0. 00 1 0. 00 1 0. 00 1 0. 00 1 - 0. 00 1 - 0. 00 1 - 0. 00 9 0. 00 7 0. 01 0 0. 00 1 A l 0. 00 0 - 0. 00 0 - 0. 00 0 - 0. 00 0 - - - 0. 00 1 - - 0, 00 0 M g 0. 00 0 - 0. 00 0 - - 0. 00 0 - - - - 0. 00 0 0. 00 8 0, 00 2 0. 00 1 Ca 0. 00 2 0. 00 1 0. 00 1 0, 00 1 0. 00 0 0. 00 0 0. 00 0 - - - - 0. 00 0 - - F e 2 0. 00 5 0. 00 1 0. 00 3 0. 00 2 0. 01 2 0. 00 5 0. 00 6 0. 00 1 0. 00 2 0, 00 4 0. 01 1 0. 03 0 0. 03 0 0. 00 5 Zr 0. 99 6 0. 99 8 0. 98 1 0. 98 9 1. 00 9 1. 00 6 0. 99 6 1. 00 5 0. 98 9 1. 00 4 0. 98 7 0. 97 9 0. 98 5 0. 99 7 H f 0. 00 9 0. 00 8 0. 01 4 0. 00 8 0. 00 6 0. 00 5 0. 00 5 0. 00 6 0. 00 8 0. 00 6 0. 00 7 0. 00 8 0. 00 9 0. 00 6 T h - - - - - - - - U4+ - - - - - - - - - P - - - - - - - - - - - - - - Ca tio n su m 2. 00 4 2. 00 2 2. 00 2 2. 00 2 2. 00 6 2. 00 3 2, 00 3 2. 00 1 2. 00 2 2. 00 2 2, 00 6 2. 01 9 2. 01 6 2. 00 3 Ti pp m 12 49 15 0 13 9 34 9 29 9 - 27 2 - 26 3 - 23 57 18 71 25 59 25 4 T (°C ) 14 71 10 52 10 41 11 92 11 64 - 11 48 - 11 42 - 16 53 15 82 16 79 11 36 A na ly se s ar e al lf ro m zi rc on rim s o f 15 pm th ic kn es s o r m o re . * zi rc on gr ai n an al ys es (no n-r im )f ro m sa m pl e 21 02 . D et ec tio n lim its ar e: T i0 2 = 0, 02 w t% (12 0 pp m Ti ), T h0 2 = 0. 09 w t% (15 82 pp m Th ), U 0 2 0. 13 w t% (22 92 pp m U )a n d P205 = 0. 08 w t% (44 7p pm P), bd (be low de te ct io n). Ti pp m ar e ca lc ul at ed u sin g 4 de ci m al so fw t% T i0 2, T( °C )i sc al cu la te d w ith th e fo rm ul a o fW at so n e ta !. 20 06 (lo g( Ti, pp m )= 6. 01 ±0 .3 - (50 80 ±3 0/T (K )). powders or mineral powders were heated at 1000°C for a minimum of 4 hours to ensure 3+that all the iron was oxidized to Fe . Synthetic standards were prepared for whole-grain hemo-ilmenite analysis. The samples were analyzed for major elements on fused glass disks (Li-borate), and for trace elements, on pressed powder pellets with an ARE 9800 XP automatic spectrometer. Standard curves were produced with a mix of natural and in-house synthetic standards (Bologne and Duchesne, 1991). 3.4- Morphology and Composition of the Zircon in the Rims Zircon rims were observed around ilmenite grains (ilmenite, hemo-ilmenite) in each of the examined samples, independent of the amount of Fe-Ti oxide present or the overall texture (e.g. equigranular or tabular). A rim is typically found at the contact of ilmenite with plagioclase or biotite (Figures 3.2a, b, e, f, 3.3c, f). Less commonly, a rim is observed along the contacts between ilmenite grains, and rutile where present (Figures 3.2c, d, 3.3a, c, e). In a few samples, there are examples of a zircon rim adjacent to secondary minerals (white mica ± chlorite ± carbonate) formed from the alteration of plagioclase (Figures 3.2c and 3.3c). A typical rim measures from a few .im to 100 1.Lm in thickness and commonly is continuous along grain boundaries (Figures 3.2a, b). In some examples (Figures 3.2a, b and 3.3c), the rim is thicker where it occupies embayments defined by intersecting grain-boundaries. In samples from the Saint-Urbain deposits (Figure 3 .2d), the zircon rims consist of individual grains measuring about 10 im by 50 tim. Several samples from this study (2102, Big Island; SR287, Laramie) contain two distinct morphologies of zircon: (1) large (>250 tm) grains that fill interstitial space either between oxides (2102, Big Island; Figures 3.3a,b) or plagioclase (SR287, Laramie), and (2) thin (< 50 J.Lm down to several jim) rims around oxide grains. The minor-element compositions of the zircon rims analyzed in this study are not uniform (Table 3.5). All in situ analyses were made in the thickest parts of individual rims, typically 20-75 jim across. The Ti contents vary from below detection limits by electron microprobe (120 ppm Ti) to 0.39 wt% Ti02 (139 to 2559 ppm Ti) (Table 3.5). The latter values are significantly higher than previously reported for zircon from mafic rocks (e.g. < 17 ppm; Belousova et at. 2002). One possibility is that the high Ti contents 85 Figure 3.2 Back-scattered electron images showing the textural associations of zircon rims from Mirepoix, Saint-Urbain and Laramie. (a) 1072-D (Mirepoix); hemo-ilmenite rimmed by zircon along contact with biotite and plagioclase. (b) 1 072-D (Mirepoix), hemo-ilmenite rimmed by zircon along contact with biotite and plagioclase. Note the thicker zircon rim in an embayment bordered by three grains of ilmenite. (c) 2006-D1 (Saint-Urbain); zircon rim along the interface between hemo-ilmenite and rutile grains. Note the much larger scale of this image compared to the other images. (d) 2006-D 1 (Saint-Urbain); zircon rim composed of numerous individual zircon grains. Note also the spinel grains contained within the hemo-ihnenite. (e) SR287 (Laramie); zircon rim around single ilmenite grain in contact with biotite. (f) SR287 (Laramie); zircon rim around ilmenite grain in contact with biotite. Note the enlarged rim at the tip ofthe ilmenite grain in the top left quadrant. Scale bars are indicated on each image. Abbreviations: (bt) biotite; (ilm) ilmenite; (plag) plagioclase; (rut) rutile and (zrc) zircon. 86 Figure 3.3. Back-scattered electron images showing the textural associations of zircon rims from Big Island and Methuen. (a) 2102 (Big Island); zircon as distinct grains and as thin rims between hemo-ilmenite grains. (b) 2102 (Big Island); detail of the bottom-left grain of zircon from (a) showing secondary baddeleyite along the margin of the zircon grain and in distinct fractures cutting zircon. (c) 2109-A (Big Island); continuous zircon rims of unequal thickness along the contact between hemo-ilmenite and plagioclase (altered). A thin zircon rim follows the limit of the rutile lens contained within hemo-ilmenite. (d) 2109-A (Big Island); cathodoluminescence image ofthe view shown in (c) (outlined by white box). (e) 2109-A (Big Island); zircon rim at the contact of an hemo-ilmenite grain and a rutile lens contained within ilmenite. Hematite exsolution lamellae on either side ofthe rutile lens can be observed. (f) Tl 10.9 (Methuen); discontinuous zircon rim at the contact of hemo-ilmenite with biotite and plagioclase. Scale bars are indicated on each image. Mineral abbreviations as defined in the caption ofFigure 3.2. 87 reported in Table 3.5 may result from the effects of secondary fluorescence. John Foumelle (personal communication, 2007) demonstrated that, at 40 tm from the contact with ilmenite, up to 100 ppm Ti in zircon may be due to secondary fluorescence of Ti in the adjoining ilmenite. Apparent Ti contents in zircon can rise up to 1000 ppm at 10 tm from the contact. Thus, the high Ti02 contents, and abnormally high temperatures calculated using the Ti-in-zircon thermometer of Watson et al. (2006) (Table 3.5), are likely due to secondary fluorescence of Ti in the ilmenite. The analyzed rims also contain between 0.52 and 1.63 wt% Hf02. Large variations in Hf02 are observed within individual samples and even within a single rim. The U-Pb-Th concentrations of four zircon fractions (0.04 to 0.12 mg) from one Saint-Urbain sample were determined by isotope dilution on a VG 5400 single collector thermal ionization mass spectrometer at the Pacific Centre for Isotopic and Geochemical Research, University of British Columbia. The measured U concentrations are extremely low (0.92-1.11 ppm) and the ThIU value varies between 15 and 20 (Table 3.6). 3.5- Discussion Below we address four possible mechanisms by which a zircon rim could form around ilmenite (and rutile when present) in plutonic rocks that have not undergone post- crystallization metamorphism: (1) zircon precipitation from late hydrothermal fluids, (2) zircon crystallization from evolved, high-temperature interstitial liquid, (3) zircon rim formation by oxidation-exsolution of ilmenite to form baddeleyite and subsequent reaction, and (4) zircon formation by diffusion of zirconium from ilmenite (and rutile) at subsolidus conditions in slowly cooled plutonic rocks. 3.5.1 - Zircon precipitation from late hydrothermal fluids In a few samples from Saint-Urbain and Big Island, a zircon rim is found along oxide grain boundaries that are adjacent to plagioclase that has been strongly to completely altered to fine-grained white mica (Figures 3.2c, 3.3c). The presence of secondary alteration is evidence that hydrothermal fluids circulated through the rocks and reacted 88 Table 3.6 Zircon U, Pb and Th contents determined by isotope dillution mass spectrometry, Saint-Urbain, Québec Sample 2033-D 2033-D 2033-D 2033-D Al A2 NA1 NA2 Uppm 1.04 1.11 1.10 0.92 Pb 1.28 1.07 0.91 0.78 Th 20.32 19.27 16.69 na U/Th 19.5 17.4 15.1 na (not analysed) 89 with the pre-existing igneous minerals, thus a potential hydrothermal origin for the zircon rims needs to be considered. The occurrence of zircon has been documented in some hydrothermal systems, such as replacement deposits and in quartz-fluorite veins (e.g., Rubin eta!. 1993; Hoskin 2005), and the solubility of Zr is enhanced by the presence of F in alkaline hydrothermal systems (Rubin et a!. 1993; Aja et a!. 1995). Grains of hydrothermal zircon described by Rubin et a!. (1993) from a F-rich rhyolite of the Sierra Blanca intrusions (Texas, USA) occur as veinlets connecting overgrowths on magmatic zircon. In a study of the Boggy Plain zoned pluton located in eastern Australia, Hoskin (2005) observed that zircon crystals precipitated from fluids are murky-brown in colour and contain no internal texture, in contrast to the oscillatory zoning commonly observed in grains of igneous zircon. Hoskin (2005) also compared the composition (analyzed by secondary ion mass spectrometry and laser ablation inductively coupled plasma mass spectrometry) of hydrothermal zircon to igneous zircon found in the same rock. Hydrothermal zircon is distinguished by much higher Hf02 (3.4 to 4.9 wt%) and LREE relative to HREE (i.e., flat chondrite-normalized REE patterns) compared to igneous zircon, which typically contains less than 1 wt% Hf02 and is characterized by strong HREE enrichment relative to the LREE (Belousova et a!. 2002). The P (‘-‘800 to 4000 ppm), U (550 to 13,000 ppm) and Th (—450 to 6000 ppm) contents of the Boggy plain hydrothermal zircon are substantially higher than concentrations determined for igneous zircon (Belousova et a!. 2002). The compositions of the analyzed rims from this study are not consistent with a formation from hydrothermal fluids; in situ analyses by electron microprobe (Table 3.5) and bulk separate analyses by isotope dilution (Table 3.6) reveal very low contents of P (<447 ppm), U (0.9-1.1 ppm) and Th (15-19 ppm), much lower than the concentrations typically of in hydrothermal zircon. In addition, there is a lack of F-bearing phases even in the relatively rare altered areas adjacent to Ti-based oxide minerals with zircon rims. 3.5.2- Crystallization of the zircon rim from an evolved interstitial liquid Although present in very low abundances (<<1 vol %), zircon is a relatively common accessory mineral in anorthositic and related rocks from Proterozoic anorthosite massifs and has been used to precisely date the age of crystallization of anorthosites worldwide 90 (see compilation of U-Pb zircon!baddeleyite crystallization ages in Scoates & Mitchell 2000). On the basis of literature descriptions and observations, the morphology of zircon in anorthosites varies from euhedral to interstitial and poikilitic, with grain sizes typically up to 1-1.5 mm in diameter. The U concentrations in zircon from Proterozoic anorthosites range from 9 to 463 ppm, and nearly 50% of the published values occur in the 25-75 ppm range (Scoates & Chamberlain 2003). These textural and compositional characteristics indicate late crystallization of zircon from an evolved, albeit low-U, interstitial liquid at temperatures just above the solidus for anorthosites (—900- 1000°C; e.g. Frost & Lindsley 1992). The measured concentrations of U (1 ppm) within zircon rims found in sample 2033-D (Saint-Urbain; Table 3.6) are significantly lower than the range observed for interstitial zircon in Proterozoic anorthosites. The zircon rims thus do not simply represent a thin film crystallized along oxide grain boundaries from late zircon-saturated interstitial liquid. As described above, some samples from Big Island (2102; Figures 3.3a,b) and Laramie (SR287) do contain two morphological types of zircon: large (>250 jim) zircon grains and thin (<50 jim down to several jim) zircon rims around oxide grains. Uranium concentrations determined by isotope dilution for the relatively coarse, interstitial zircon grains from sample 2102 are —1 0 ppm (Morisset, unpublished data). These concentrations are an order of magnitude higher than concentrations from rim fragments in sample 2033-D, at the low end of the range observed for interstitial zircon from Proterozoic anorthosites, which is consistent with crystallization of the coarser grained zircon from evolved interstitial liquid. However, the reproducible textural setting of the zircon rims exclusively along ilmenite (and rutile) grain boundaries and their extremely low U concentrations (--1 ppm) do not support an origin by crystallization from a late-stage magmatic liquid. 3.5.3- Zircon formation following oxidation-exsolution of baddeleyite from ilmenite Small (1-20 jim) blebs of baddeleyite in ilmenite from the Basistoppen sill, East Greenland, have been interpreted as exsolution lamellae that formed in the host ilmenite at Zr concentrations well below the solubility limit of Zr in ilmenite (Naslund 1987). The formation of rutile by the oxidation of ilmenite follows the reaction given by Haggerty (1976): (9FeTiO3+ Fe2O3)+ 02 (5FeTiO3+ 3Fe2O)+ 4TiO2,and if Zr 91 replaces Ti in the equation, then baddeleyite is produced rather than rutile. Naslund (1987) proposed that simultaneous exsolution of Fe-Cr-spinel (reduction) and baddeleyite (oxidation) from ilmenite occurred during cooling of the high-level Basistoppen sill. Following this, Bingen et al. (2001) interpreted the presence of zircon rims around hemo-ilmenite grains in rocks of the Lindás Nappe in the Caledonides of western Norway to represent former baddeleyite that had exsolved from ilmenite and that migrated to grain boundaries prior to granulite-facies metamorphism. During granulite facies metamorphism, the baddeleyite reacted with silica at grain boundaries to form metamorphic zircon (Bingen et al. 2001). Despite extensive examination of ilmenite grains by SEM from the five different intrusions in this study, we did not find a single occurrence of baddeleyite blebs or lamellae in ilmenite. The only occurrence of baddeleyite is in sample 2102 (Big Island), in secondary, fracture-filling structures cutting zircon and extending from the fractures along ilmenite or rutile-zircon grain boundaries (Figure 3.3b). This texture indicates that the pre-existing grain of zircon reacted to produce baddeleyite. In addition, oxidation of ilmenite or hemo-ilmenite should produce lenses of rutile and hematite within ilmenite (e.g., Haggerty 1976), such as those observed in the Big Island and the Saint-Urbain deposits in this study (Figure 3.3c); no baddeleyite was observed in association with secondary rutile and hematite produced by oxidation. Oxidation products, or oxidation-reduction products, also were not observed in the samples studied from the other three localities. Thus, we find no evidence that the zircon rims around ilmenite documented in this study formed by reaction of baddeleyite or that baddeleyite originally exsolved from ilmenite. 3.5.4- Formation of a zircon rim by diffusion of Zr from ilmenite and reaction along grain boundaries The textural restriction of zircon rims to the margins of ilmenite and hemo-ilmenite (and rutile when present) in the Proterozoic anorthosites and gabbroic intrusions studied, and the extremely low measured U contents, are consistent with derivation of zirconium from the adjacent Ti-based oxides by diffusion and subsequent reaction along the grain boundaries to form thin, jim-scale, films of zircon. To assess the viability of this mechanism, below we evaluate (1) the partitioning behaviour of Zr into ilmenite (and 92 rutile) during crystallization from silicate magmas, (2) the amount of zircon present that could be produced from Zr-bearing ilmenite, and (3) the mechanism for diffusion of Zr out of ilmenite. With respect to Zr partitioning in ilmenite (FeTiO3),the ionic radii of Ti4 and Zr4in 6-fold coordination are 0.61 and 0.72 A (Shannon 1976), a difference of about 18%; thus substitution by direct exchange of Zr4 for Ti4 in ilmenite is permissible. Compatible behaviour of Zr in ilmenite was originally suggested by Taylor & McCallister (1972), who reported Zr contents up to 0.42 wt% in ilmenite from Apollo 15 lunar rocks. Later experimental studies on lunar evolution demonstrated that the partition coefficient for Zr between ilmenite and silicate liquid (D) ranges from 0.28 to 0.38 in synthetic high-Ti Mare basalt (McCallum & Charette 1978; McKay et al. 1986; Nakamura et al. 1986). Fujimaki et al. (1984) found that Zr is compatible (D = 3) in ilmenite megacrysts from kimberlite. More recently, compatible behaviour has also been proposed for Zr in ilmenite from the Skaergaard intrusion, with D 2.3 based on measured trace element contents of ilmenite separates and the estimated composition of the Skaergaard magma (Jang & Naslund 2003). Thus, Zr is compatible in ilmenite in ferrobasaltic systems, which would include the Fe-Ti-enriched basaltic magmas that produce significant accumulations of ilmenite in Proterozoic anorthosites (e.g. Duchesne 1999; Dymek & Owens 2001; Charlier et a!. 2006). Finally, Zr is compatible in rutile (Ti02) over a wide range of compositions, with D in the range of 2.7-13.1 in basaltic melts (Xiong et a!. 2005) and 1.1-8.8 for andesitic and rhyolitic melts (Kiemme et a!. 2005). A positive correlation between whole-rock Ti02 content (a proxy for ilmenite ± rutile abundance) and whole-rock Zr concentration for samples from massive, disseminated and unmineralized samples from the Saint-Urbain and Big Island deposits provides strong evidence that Zr was compatible in ilmenite during crystallization and formation of these deposits (Figure 3.4a). Massive ores of ilmenite contain 350-500 ppm Zr, and samples with abundant rutile deviate from the reference line on Figure 3 .4a to higher TiO2 and Zr contents. The Zr concentrations in ilmenite are highly variable and may be extremely low (6-57 ppm) when compared to those of the respective whole rocks (Figure 3.4b). This finding is consistent with diffusive loss of Zr from ilmenite at high 93 900 ..—.. 800 I: 500 400 300 200 100 900 ..—., 800 E D. 700 600 500 2 400 300 200 100 Whole rock TiO, (wt%) 0 10 20 30 40 50 60 Ferrian-ilmenite Zr (ppm) Figure 3.4. Relationships between whole rock Zr (ppm) and Ti02(wt%) and hemo-ilmenite Zr (ppm). (a) Diagram of whole rock Zr (ppm) vs. whole rock Ti02 (wt%) for samples from Saint-Urbain and Big Island showing a positive correlation indicating that Zr is compatible in hemo ilmenite. The two samples with higher Zr contents contain abundant rutile. (b) Diagram of whole rock Zr concentration (ppm) vs. hemo ilmenite Zr concentration (ppm) for samples from the Saint-Urbain and Big Island deposits. 0 10 20 30 40 50 60 b . . . . . . . . 94 temperatures to form the rims. However, the viability of this mechanism requires a demonstration that liberation of reasonable amounts of Zr from ilmenite can produce the observed quantities of zircon present in the rims. To accomplish this, we have applied the formulation of Fraser eta!. (1997) (equation in Figure 3.5a), used by them to calculate the volume of zircon produced from metamorphic net-transfer equations during breakdown of a Zr-bearing phase (e.g., garnet with 20-55 ppm Zr). First, we calculated an array of theoretical lines (Figure 3.5a) to predict the radius (in jim) of zircon (density = 4.65 glcm3)that could be produced from ilmenite (density = 4.72 g/cm3)of variable grain size (up to 2 mm radius) that contains 20 to 900 ppm Zr, on the basis of the equation of Fraser et at. (1997). For example, a spherical grain of ilmenite with a 1 mm radius containing 100 ppm Zr could produce a spherical zircon grain with a radius of 60 j.tm (Figure 3.5a), assuming that all Zr is expelled from the ilmenite grain and that silica is available to form zircon. Next, we re-arranged the formulation of Fraser et a!. (1997), shown as an inset in Figure 3 .5b, to express the initial concentration of Zr in ilmenite as a function of the volume of ilmenite divided by the volume of zircon formed (Viimenite/Vzircon) (solid curve on Figure 3 .5b). By evaluating the volume of ilmenite and zircon in the samples studied, our goal was to assess whether the calculated concentration of Zr in ilmenite (i.e., the initial Zr content) was realistic. However, accurately defining the volume of ilmenite and zircon from the distribution of these phases in thin section is difficult because the mineral grains themselves are not present as idealized geometrical forms (i.e., spheres): a thin section represents only a single 2D-slice through a rock with complex 3D grain boundaries. In addition, the exact shape of zircon in the rims is difficult to quantify, as irregularities in grain boundaries may have facilitated concentration of zircon at specific locations, and some material may have migrated into embayments in ilmenite (Figure 3.2b). A more tractable approach to estimate the relative volumes of ilmenite and zircon involves using mineral and whole-rock compositions (i.e. mass quantities) as well as mineral density. Major-element and select trace-element data on separated hemo ilmenite grains (Table 3.2) and respective (or nearby) whole rocks (Table 3.3) are available for four samples from the Saint-Urbain deposits. In the simplest case, for rocks 95 EI,) - C 0 N C 0 25000 20000 15000 10000 ci) 2 — 5000 . 0 100 200 300 400 500 600 700 0 > Zr ppm in ilmenite Figure 3.5. Zircon radius vs. ilmenite radius calculated from Fraser et al. (1997) (a) and volume of ilmenite/volume of zircon vs. Zr (ppm) in ilmenite diagram (b). (a) Relationship between Zr concentration in ilmenite, ilmenite radius and zircon radius based on the formulation of Fraser et al. (1997), adapted for ilmenite (formula is shown in the inset) where V is the volume of reacting mineral; V is the volume of zircon produced; pz is the density ofzircon (4.65 g/cm3); M, is the molar mass of zirconium; [ZrJ, is the concentration of zirconium in ppm in the reacting mineral; p is the density of the reacting mineral (ilmenite = 4.72 g/cm3);and M is the molar mass ofzircon (ZrSiO4). (b) Calculated ilmenite volume/zircon volume vs. Zr content of ilmenite. The black line shows the results from figure (a): the triangles indicate results for selected samples from the Saint-Urbain deposits (see Discussion). The reorganized formula ofFraser et al. (1997), as well as a representation to scale ofmineral volumes for sample 2033-D, are also shown. 0 0.5 1 1.5 2 Ilmenite radius (mm) VIlm 490268 2033 D b L0 [z rijim ilmenite0VoIumezircon 2009-Bi D 2006-Di15B2 2033- 96 that do not contain rutile and where all of the Zr is derived from ilmenite, the initial Zr content of ilmenite can be calculated by dividing the concentration of Zr in the whole- rock by the fraction of ilmenite present. The proportion of ilmenite in each rock can be calculated using the Ti02 contents of the whole rocks and the corresponding Ti02 contents of the ilmenite. When rutile is present, the proportion of rutile can be calculated with the Ti02 not included in the ilmenite. For example, the proportion of ilmenite in sample 2006-Di is 0.78 and for rutile it is 0.18, and the calculated ratio of the volume of ilmenite to zircon is 1096. Results for the four whole-rocks and associated ilmenite are plotted in Figure 3 .5b, and the ratio Vjimenjte/Vzjrcon varies from 860 to 1415. Using the Fraser et a!. (1997) formula, 345-568 ppm Zr from the reacting mineral would be needed to produce the calculated volume of zircon. The range of Zr concentrations obtained for hemo-ilmenite is within the one observed in the ilmenite of the Skaergaard intrusion (213 to 1400 ppm; Jang & Naslund 2003). Considering a mineral/melt partition coefficient of 2.3 (Jang & Naslund 2003), the magmas in equilibrium with ilmenite would have had Zr concentrations between 150 to 247 ppm. This estimated range is consistent with the Zr content of a basaltic liquid that would have fractionated non-Zr-bearing minerals such as olivine, pyroxene and plagioclase (average Zr contents for basalts from the Archean to Cenozoic are between 82-138 ppm, Condie 1993). Rutile, which has higher partition coefficients for Zr than ilmenite, is present in many of the samples from Saint-Urbain and Big Island (see Figures 3.2c, d and 3.3e) and could potentially have contributed Zr to the formation of some of the zircon rims. Electron microprobe analyses of rutile reveal a zonation in Zr content for all analyzed grains (Table 3.1). For the Saint-Urbain rutile, the central parts of the grains contain 698-1322 ppm Zr, and the margins adjacent to zircon rims contain 318-634 ppm Zr. The lower Zr contents of rutile adjacent to the zircon rims are consistent with transfer of Zr from rutile across a phase boundary (e.g., Watson & Baxter 2007). Calculated temperatures, based on the Zr-in-rutile geothermometer of Watson et a!. (2006), vary between 717-782°C for the cores and 647-708°C for the rims. The apparent lower temperatures of the rims may correspond to the closure temperatures for the transfer of Zr from rutile to the zircon rims. 97 The trace-element compositions of the zircon in the rims also directly reflect a genetic link to the adjacent of grains of ilmenite. As noted previously, the U contents of fragments of zircon rims from sample 2030-D (Saint-Urbain) are extremely low (- 1 ppm), much lower than in typical igneous zircon. Uranium is highly incompatible in ilmenite (DjlmU = 0.0082 for alkali basalt; Zack & Brumm 1998), thus the formation of low-U zircon as a consequence of transfer of Zr from ilmenite is expected. The structure of ilmenite is favourable to the chemical diffusion of Zr toward the grain margins. Ilmenite (Fe2Ti4O3)is an ordered rhombohedral oxide (space-group symmetry R3) formed of stacked layers of oxygen octahedra that are 2/3-filled by cations (Lindsley 1976; Waychunas 1991). The structure of the exsolved ilmenite is defmed as discrete layers of Ti4 octahedra that alternate with layers of Fe2 forming the ilmenite lamellae and layers ofFe3 forming the hematite lamellae (e.g., McEnroe et al. 2002; Robinson et a?. 2004). Cations of Zr4 incorporated into ilmenite during crystallization from Fe-Ti-enriched ferrobasaltic magmas will reside within the Ti layers. Chemical potential gradients along the layers of Ti octahedra may arise through establishment of concentration gradients from regions of relatively high Zr (200-1400 ppm in ilmenite; Jang & Naslund 2003; this study) to Zr-poor regions external to individual grains of ilmenite. Cations ofZr4 could migrate along the Ti octahedral layers by a series of small “jumps” to equivalent sites within each layer. Because diffusion is a temperature- dependent phenomenon (e.g. Watson & Baxter 2007), the diffusion of Zr in ilmenite would be enhanced at the high temperatures corresponding to subliquidus to subsolidus conditions following crystallization (e.g. 1000°C down to -750°C). In this study, zircon rims were observed around ilmenite in all examined samples from slowly cooled plutonicrocks. Zircon rims around ilmenite are not expected in ilmenite-bearing basaltic lavas owing to rapid cooling associated with emplacement at the Earth’s surface. High- level intrusions may also cool too rapidly to allow for significant diffusion of Zr in ilmenite (e.g., high-Zr ilmenite from the Skaergaard intrusion; Jang & Naslund 2003). The formation of a rim around ilmenite requires reaction between the diffused Zr with silica to form zircon (ZrSiO4). Zircon rims do occur directly adjacent to silicates (e.g. plagioclase, biotite), which may have contributed the required silica. However, some of the samples studied are massive oxide ores with only minor silicate minerals 98 present and1.tm-thick rims may be observed sandwiched between ilmenite grains or along ilmenite-rutile contacts (i.e. silica-deficient environments). An additional source for the Si is ilmenite itself. In 6-fold coordination, the radius of Si4 is 0.40 A (Shannon 1976), thus minor substitution of Si4 for Ti4 should be possible. Electron-microprobe analyses of ilmenite generally reveal detectable 5i02 (0.01-0.02 wt%), and bulk separates of ilmenite can show a range of Si02 contents (up to 0.4 wt%), although some quantity of microscopic silicate minerals adhering to separated ilmenite is probably inevitable. Where necessary, the diffusion of Si and Zr from ilmenite could thus provide the cations for zircon formation along grain boundaries between Ti-based oxide minerals. Finally, it is possible that the presence of a high-temperature aqueous fluid may have aided in the mobilization of Zr from ilmenite (and rutile) by a dissolution-and reprecipitation mechanism. Dymek & Schiffries (1987) observed vermicular intergrowths of quartz + plagioclase (calcic myrmekite) that represent about 1 vol.% of the andesine anorthosites from the Saint-Urbain massif. They proposed that these intergrowths result from the interaction of cumulus plagioclase and high-temperature aqueous fluid derived from fluid-saturated interstitial melt localized at grain boundaries. Locally-derived aqueous fluid along oxide mineral grain boundaries may have allowed for partial dissolution of ilmenite (and rutile) and liberation of Zr, which could then react with silica in the fluid to form zircon. A dissolution-and-reprecipitation origin of zircon from ilmenite in response to reaction of pre-existing ilmenite grains with high- temperature aqueous fluid should produce either a porous rim around ilmenite grains or chemical zoning between the core of the unreacted ilmenite and the reacted rim (e.g. Putnis, 2002). In the samples examined in this study, neither of these features is observed. And, importantly, only the Zr present in the rim portions of individual ilmenite grains would have been available to form zircon, which is inconsistent with the mass balance calculations presented above. 3.5.5- Implications of the zircon rims for U-Pb geochronology in plutonic rocks Application of U-Pb geochronology to zircon rims around ilmenite in slowly cooled plutonic rocks could potentially provide information on the timing of zircon formation. The closure temperature for diffusion of Pb out of zircon is likely in excess of 950- 99 1000°C (Cherniak & Watson 2000), temperatures higher than the estimated temperatures for formation of the zircon rim under subsolidus conditions. Thus, the U-Pb age of a zircon rim where ilmenite is the source of Zr should directly date the age of rim formation. Potential issues in precise U-Pb dating include (1) the low U concentrations (—1 ppm U) in the zircon rims and correspondingly small amounts of radiogenic Pb produced, (2) the effective identification and separation of rim material from a rock where coarser interstitial zircon also occurs, for analysis by thermal ionization mass spectrometry, or (3) the small width of the rims, commonly less than 10 Jim, for analysis by microbeam techniques (LA-ICP-MS or SIMS). Nevertheless, even the determination of relatively imprecise dates from zircon rims on ilmenite could be useful in quantifying the thermal evolution history of a given plutonic rock. 3.6- Conclusions 1- Thin (1-100 jim) zircon rims have been observed around ilmenite in samples from the five different mafic intrusions, mostly related to Proterozoic anorthosite suites in the Grenville Province of Québec, Canada. The low Hf02,P, Th and U contents of the zircon rims preclude their formation by precipitation from hydrothermal fluids and by crystallization from evolved, high-temperature, zircon-saturated interstitial liquids or aqueous fluids. 2- The zircon rims are restricted to the margins of Ti-based oxide minerals (ilmenite, hemo-ilmenite, and rutile when present) indicating that the Zr was provided by the adjacent Ti-based oxide. 3- A positive correlation between whole rock Ti02 and Zr for samples from the Saint-Urbain and Big Island Fe-Ti oxide ore deposits in this study (up to 500 ppm Zr in massive hemo-ilmenite ores) demonstrates that Zr is compatible in ilmenite that crystallizes from parental ferrobasaltic magmas. 4- Mass balance calculations of the initial Zr concentration in ilmenite for samples from the Saint-Urbain deposits yield values from 345-568 ppm, which are below the solubility limit of Zr in ilmenite. Exsolution of baddeleyite from ilmenite, which would then migrate to grain boundaries and react to form 100 zircon, requires oxidation (Naslund, 1987). No evidence for baddeleyite exsolution lamellae or conversion of pre-existing baddeleyite to zircon has been found in this study. 5- We propose that the origin of the jtm-scale zircon rims around ilmenite occurs by diffusion of Zr4 along the octahedral Ti layers of the ilmenite structure at sub-solidus temperatures from regions of relatively high Zr (1 OOs of ppm) to Zr-poor regions external to the ilmenite grains. Silica to form zircon is provided by adjacent silicate minerals or by trace silica in ilmenite itself. Based on the results of this study, we predict that j.tm-scale zircon rims around ilmenite may be a common feature in slowly-cooled plutonic rocks formed from the crystallization of ferrobasaltic magmas. 3.7- Acknowledgements Rio Tinto Iron & Titanium provided critical logistic support in the field and analytical costs. We thank Mati Raudsepp for his help with the electron-microprobe analyses and SEM imaging, Richard Friedman for the isotope dilution analyses of zircon, and Bernard Charlier for his assistance in the field. We are grateful to Jacqueline Vander Auwera and Jean-Clair Duchesne (University of Liege, Belgium) for careful training of C.-E. Morisset in Fe-Ti oxide petrology, mineral separation techniques and XRF analysis during a 2-year period of work at Liege. Comments and advice from Dominique Weis significantly improved our interpretation and presentation. We thank J.-C. Duchesne, H.R. Naslund and R.F. Martin for very useful reviews and editorial comments. C.-E. Morisset was supported by an NSERC PGS-B scholarship. Research support is gratefully acknowledged from a NSERC Discovery Grant to J.S. Scoates and a NSERC CRD to J.S. Scoates and D. Weis. 101 3.8- References Aja, S.U., Wood, S.A. & Williams-Jones, A.E. (1995): The aqueous geochemistry of Zr and the solubility of some Zr-bearing minerals. Appl. Geochem. 10, 603-620. Ashwal, L.D. & Wooden, J. L. (1983): Isotopic evidence from the eastern Canadian shield for geochemical discontinuity in the Proterozoic mantle. Nature 306, 679- 680. Belousova, E. A., Griffin, W. L., OReilly, S. Y. & Fisher, N. I. (2002): Igneous zircon: trace element composition as an indicator of source rock type. Contrib. Mineral. Petrol. 143, 602-622. Bingen B., Austrheim, H. & Whitehouse, M. (2001): Ilmenite as a source for zirconium during high-grade metamorphism? Textural evidence from the Caledonides of Western Norway and implications for zircon geochronology. J. Petrol. 42, 355-375. Bologne, G. & Duchesne, J.C. (1991): Analyse des roches silicatées par spectrométrie de fluorescence X: precision et exactitude. Professional Papers, Geological Survey of Belgium 249, 1-11. Charlier, B., SkAr, 0., Korneliussen, A., Duchesne, J.-C. & Vander Auwera, J. (2007): Ilmenite composition in the Tellnes Fe-Ti deposit, SW Norway: fractional crystallization, postcumulus evolution and ilmenite-zircon relation. Contrib. Mineral. Petrol. 154, 119-134. Charlier, B., Duchesne, J.-C. & Vander Auwera, J. (2006): Magma chamber processes in the Telines ilmenite deposit (Rogaland Anorthosite Province, SW Norway) and the formation of the Fe-Ti ores in massif-type anorthosites. Chem. Geol. 234, 264-290. Cherniak, D.J. & Watson, E.B. (2000): Pb diffusion in zircon. Chem. Geol. 172, 5-24. Condie, K.C. (1993): Chemical composition and evolution of the upper continental crust: contrasting results from surface samples and shales. Chem. Geol. 104, 1-37. Davidson, A. (1998): An overview of the Grenville Province geology, Canadian shield. In: Geology of the Precambrian Superior and Grenville Provinces and Precambrian Fossils in North America (S. B. Lucas & M. R. St-Onge, eds). Geol. Surv. Can., Geology ofCanada 7, 205-270. Duchesne, J.C. (1999): Fe-Ti deposits in Rogaland anorthosites (South Norway): geochemical characteristics and problems of interpretation. Miner. Deposita 34, 182-198. Duchesne, J.C. (1966): Separation rapide des minéraux des roches. Ann. Soc. Géol. Belg. 87, 347-356. Dymek, R.F. & Owens, B .E. (2001): Petrogenesis of apatite-rich rocks (nelsonites and oxide-apatite gabbronorites) associated with massif anorthosites. Econ. Geol. 96, 797-815. Dymek, R.F. & Schiffries, C.M. (1987): Calcic myrmekite: possible evidence for the involvement of water during the evolution of andesine anorthosite from St-Urbain, Quebec. Can. Mineral. 25, 291-3 19. Fraser, G., Ellis, D. & Eggins, S. (1997): Zirconium abundance in granulite-facies minerals, with implications for zircon geochronology in high-grade rocks. Geology 25, 607-6 10. 102 Frost, B.R. & Lindsley, D.H. (1992): Equilibria among Fe-Ti oxides, pyroxenes, olivine, and quartz: Part II. Application. Am. Mineral. 77, 1004-1020. Fujimaki, H., Tatsumoto, M. & Aoki, K. (1984): Partition coefficients of Hf, Zr, and REE between phenocrysts and groundmass. Proceedings of the Fourteenth Lunar and Planetary Science Conference. I Geophys. Res. B662-B672. Haggerty, S.E. (1976): Oxidation of opaque mineral oxides in basalts. In: Oxide minerals (ed. D. Rumble). Mineral. Soc. Amer., Rev. Mineral 3, Hgl-HglOO. Hébert, C, Cadieux, A.-M. & van Breemen, 0. (2005): Temporal evolution and nature of Ti-Fe-P mineralization in the anorthosite-mangerite-charnockite-granite (AMCG) suites of the south central Grenville Province, Saguenay — Lac St. Jean area, Quebec, Canada. Can. I Earth Sci. 42, 188 1-1906. Hoskin, P.W.O. (2005): Trace-element composition of hydrothermal zircon and the alteration of Hadean zircon from the Jack Hills, Australia. Geochim. Cosmochim. Acta 69, 637-648. Jang, Y.D. & Naslund, H.R. (2003): Major and trace element variation in ilmenite in the Skaergaard Intrusion: petrologic implications. Chem. Geol. 193, 109-125. Kersting, A.B., Arculus, R.J., Delano, J.W. & Loureiro, D. (1989): Electrochemical measurements bearing on the oxidation state of the Skaergaard layered intrusion. Contrib. MineraL Petrol. 102, 376-388. Ketchum, J.W.F, Scoates, J.S. & Duke, J.M. (1988): The Methuen ilmenite deposits, southeastern Ontario. Geol. Assoc. Can. — Mineral. Assoc. Can., Program Abstr. 13, A65. Kiemme, S., Prowatke, S., Hametner, K & GUnther, D. (2005): Partitioning of trace elements between rutile and silicate melts: Implications for subduction zones. Geochim. Cosmochim. Acta 69, 2361-237 1. Lindsley, D.H. (1976): The crystal chemistry and structure of oxide minerals as exemplified by the Fe-Ti oxides. In Oxide Minerals (D. Rumble, ed.). Rev. Mineral. 3, L1-L60. Lindsley, D.H. & Frost, B.R. (1992): Equilibria among Fe-Ti oxides, pyroxenes, olivine, and quartz: Part I. Theory. Am. Mineral. 77, 987-1003. Loferski, P.J. & Arculus, R.J. (1993): Multiphase inclusions in plagioclase from anorthosites in the Stiliwater Complex, Montana: implications for the origin of the anorthosites. Contrib. Mineral. Petrol. 114, 63-78. Lumbers, S.B., Wu, T.W., Heaman, L.M., Vertolli, V.M. & MacRae, N.D. (1991): Petrology and age of the A-type Mulok granite batholith, northern Grenville Province, Ontario. Precambrian Res. 53, 199-231. McCallum, I.S. & Charette, M.P. (1978): Zr and Nb partition coefficients: Implications for the genesis of mare basalts, KREEP, and sea floor basalts. Geochim. Cosmochim. Acta 42, 859-869. McEnroe, S.A., Harrison, R.J., Robinson, P. & Langenhorst, F. (2002): Nanoscale haematite-ilmenite lamellae in massive ilmenite rock: an example of ‘lamellar magnetism’ with implications for planetary magnetic anomalies. Geophys. I mt. 151, 890-912. McKay, G., Wagstaff, J. & Yang, S.-R. (1986): Zirconium, hafnium, and rare earth element partition coefficients for ilmenite and other minerals in high-Ti lunar mare basalts: an experimental study. Proceedings of the Sixteenth Lunar and Planetary Science Conference, Part 2. J. Geophys. Res. 91, D229-D237. 103 Morishita, T., Maeda, J., Miyashita, S., Matsumoto, T. & Dick, H.J.B. (2004): Magmatic srilankite (Ti2ZrO6)in gabbroic vein cutting oceanic peridotites: an unusual product of peridotite-melt interactions beneath slow-spreading ridges. Am. Mineral. 89, 759-766. Morisset, C.E. (2002): Étude pétrologique et géochimique de la minéralisation en oxydes defer et titane et en apatite du complexe du lac Mirepoix (Anorthosite du lac Saint- Jean, Quebec). Mémoire de maItrise, Université du Québec a Montréal, Montréal. Morisset, C.E., Scoates, J.S. & Weis, D. (2006): Trace element and Hf isotopic compositions of magmatic rutile from Fe-Ti oxide ore deposits related to Proterozoic anorthosite massifs. Eos Trans. AGU, Fall Meet. Suppi. 87(52), Abstract V33A-0644. Nakamura, Y., Fujimaki, H., Nakamura, N., Tatsumoto, M., McKay G.A. & Wagstaff, J. (1986): Hf, Zr, and REE partition coefficients between ilmenite and liquid: implications for lunar petrogenesis. I Geophys. Res. B91-4, D239-D250. Naslund, H.R. (1987): Lamellae of baddeleyite and Fe-Cr spinel in ilmenite from the Basistoppen Sill, East Greenland. Can. Mineral. 25, 9 1-96. Owens, B.E. & Dymek, R.F. (2005): Rediscovery of the Mattawa anorthosite massif, Grenville Province, Quebec. Can. I Earth Sci. 42, 1699-1718. Pouchou, J.L. & Pichoir, F. (1985): PAP q5(pZ)procedure for improved quantitative microanalysis. Microbeam Analysis, 104-106. Putnis, A. (2002): Mineral replacement reactions: from macroscopic observation to microscopic mechanisms. Mineral. Mag. 66, 689-708. Robinson, P., Harrison, R.J., McEnroe, S.A. & Hargraves, R.B. (2004): Nature and origin of lamellar magnetism in the hematite-ilmenite series. Am. Mineral. 89, 725- 747. Rubin, J.N., Henry, C.D. & Price, J.G. (1993): The mobility of zirconium and other “immobile” elements during hydrothermal alteration. Chem. Geol. 110, 29-47. Scoates, J.S. & Chamberlain, K.R. (1995): Baddeleyite (ZrO2)and zircon (ZrSiO4)from anorthositic rocks of the Laramie anorthosite complex, Wyoming: Petrologic consequences and U-Pb ages. Am. Mineral. 80, 1317-1327. Scoates, J.S. & Chamberlain, K.R. (2003): Geochronologic, geochemical and isotopic constraints on the origin of monzonitic and related rocks in the Laramie anorthosite complex, Wyoming, USA. Precambrian Res. 124, 269-304. Scoates, J.S. & Mitchell, J.N. (2000): The evolution of troctolitic and high Al basaltic magmas in Proterozoic anorthosite plutonic suites and implications for the Voisey’ s Bay massive Ni-Cu sulfide deposit. Econ. Geol. 95, 677-701. Shannon, R.D. (1976): Systematics of interatomic distances in oxides (R.G.J. Strens, ed.). John Wiley & Sons, London, United Kingdom. Taylor, L.A & McCallister, R.H. (1972): An experimental investigation of the significance of zirconium partitioning in lunar ilmenite and ulvospinel. Earth Planet. Sd. Lett. 17, 105-109. Thompson, J. & Peck, W.H. (2003): Metamorphic zircon in Fe-Ti ores, Morin Anorthosite Complex (Grenville Province, Québec). Geol. Soc. Am., Abstr. Prog. 35, 85. 104 van Breemen, 0. & Higgins, M.D. (1993): U-Pb zircon age of the southwest lobe of the Havre-Saint-Pierre Anorthosite Complex, Grenville Province, Canada. Can. I Earth Sd. 30, 1453-1457. Watson, E.B. & Baxter, E.F. (2007): Diffusion in solid-Earth systems. Earth Planet. Sci. Lett. 253, 307-327. Watson, E.B, Wark, D.A. and Thomas, J.B. (2006): Crystallization thermometers for zircon and rutile. Contrib. Mineral. Petrol. 151, 4 13-433. Waychunas, G.A. (1991): Crystal chemistry of oxides and oxyhydroxides. In Oxide Minerals: Petrologic and Magnetic Significance (D.H. Lindsley, ed.). Rev. Mineral. 25, 11-68. Wodicka, N., David, J., Parent, M., Gobeil, A. & Verpaelst, P. (2003): Géochronologie U-Pb et Pb-Pb de la region de Sept-Iles—Natshquan, Province de Grenville, moyenne Côte-Nord. In Géologie et ressources minérales da Ia partie est de la Province de Grenville (D. Brisebois and T. Clark, eds). Géologie Québec, Québec (59-118). Xiong, X.L., Adam, J. & Green, T.H. (2005): Rutile stability and rutile/melt HFSE partitioning during partial melting of hydrous basalt: Implications for TTG genesis. Chem. Geol. 218, 339-359. Zack, T. & Brumm, R. (1998): Ilmenite/liquid partition coefficients of 26 trace elements determined through ilmenite/clinopyroxene partitioning in garnet pyroxenites. In Extended Abstracts - 7th International Kimberlite Conference, Cape Town, South Africa (986-988). 105 Chapter IV Geochemistry and Hf isotopic systematics of rutile and ilmenite from Fe-Ti oxide deposits associated with Grenvillian Proterozoic anorthosite massifs’ A version of this chapter has been submitted for publication to an international scientific journal. Morisset, C.-E., Scoates J.S., Weis, D. & Rahier, A. 106 4.1- Introduction Although widespread in metamorphic and granitic rocks, rutile (Ti02)is not commonly present in basaltic rocks due to the high solubility of Ti02 in Mg-Fe-rich melts (Ryerson & Watson, 1987). The chemistry of rutile has been investigated mainly for its role in high field strength element depletion, which is characteristic of subduction zone magmatism (e.g. Foley et a!. 2000; Kiemnie et al. 2005; Xiong et a!. 2005). In contrast, ilmenite (FeTiO3)is present in a wide range of both metamorphic and magmatic rocks (mafic to felsic). The major and minor element chemistry of ilmenite has been extensively studied because: (1) ilmenite stability is related to the oxygen fugacity of the magmas from which it crystallizes, (2) extensive sub-solidus reactions occur during cooling between ilmenite and magnetite (and coexisting ferromagnesian silicates), (3) and ilmenite plays an important role in the Fe-Ti-enrichment path of basaltic magma (e.g. Buddington & Lindsley 1964; Duchesne 1972; Lindsley & Frost 1992; Toplis & Carrol 1995; Jang & Naslund 2003; Charlier et al. 2007). Fe-Ti deposits, typically composed of ilmenite ± Ti-magnetite, are common in Proterozic anorthosite massifs. Both rutile and ilmenite are present in the Saint-Urbain and Big Island Fe-Ti oxide ore deposits that occur within Proterozoic anorthosite massifs of the Grenville Province, Québec (Saint-Urbain anorthosite and Lac Allard lobe of the large Havre-Saint-Pierre anorthositic suite, respectively). With the exception of rutile bearing nelsonite in the Roseland anorthosite (Virginia, USA) (Kolker 1982), these are the only reported Fe-Ti oxide ore deposits that contain rutile. The presence of magmatic rutile in these deposits indicates that the Fe-Ti-enriched parental magmas achieved a level of Ti-enrichment rarely observed in igneous systems. In this study, we present major element (XRF), trace element (HR-ICP-MS), and Hf isotopic compositions (MC ICP-MS) of rutile and ilmenite separates from the Saint-Urbain and Big Island deposits and ilmenite separates from their respective host anorthosites. A major component of the study involved developing the chemical protocols for trace elements analysis in solution, without fusion, of rutile and ilmenite and analysis by HR-ICP-MS, and Hf ion exchange chemistry for ultra-rich Ti-bearing minerals (40-95 wt% Ti02) in large samples (>100 mg). Combined, the results are used to constrain the origin of these unusual rutile 107 bearing Fe-Ti oxide deposits and their genetic link with their host anorthosite, as well as the source of the anorthosite parent magmas based on Hf isotopic systematic. 4.2- Locality description and sampling Proterozoic anorthosite massifs represent —10% of the exposed surface area of the Grenville Province (Figure 4.1). These massifs are composed mostly of anorthosite (>90% plagioclase; An50 ± io) and subordinate leucotroctolite, leuconorite and leucogabbro. Mangerite, charnockite and granite are almost always associated with the anorthosite massifs, and form magmatic suites commonly referred to as AMCG (anorthosite-mangerite-chamockite-granite) suites. Oxide deposits associated with Grenvillian anorthosite massifs have been classified into three major types by Hébert et al. (2005): (1) Fe-Ti; (2) Ti-Fe-P; and (3) P deposits. The Saint-Urbain and Big Island deposits belong to the first type of deposit. 4.2.1 - Saint-Urbain anorthosite and associated Fe-Ti oxide deposits The Saint-Urbain anorthosite (age = 1053 Ma, Chapter 2) is a small (—450 2) predominantly andesine anorthosite pluton (Dymek, 2001), located north of Baie-Saint Paul, and contained within the allochthonous polycyclic belt of the Grenville Province (Figure 4.1). The anorthosite intrudes undated charnockitic migmatite (Rondot, 1989) and is bounded in the west by the ca. 1060 Ma Saint-Anne du nord orthopyroxene gabbrodiorite (SANG, Figure 4.2a) (Chapter 2). Fe-Ti oxide mineralization occurs in eight discrete deposits found in the southwestern part of the anorthosite (Bignell; Coulomb West; Coulomb East; General Electric; Séminaire; Furnace; Bouchard, and Glen) (Figure 4.2). These irregular-shaped bodies measure between 70 m x 160 m and 3 m x 24 m in dimension and have sharp to gradational contacts with the host anorthosite (Figure 4.3a). The mineralogy of the deposits (Figure 4.3b) includes variable proportions of hemo-ilmenite (referred to as ilmenite for simplicity; Figure 4.3c), rutile, plagioclase, sapphirine, orthopyroxene, apatite, biotite, pleonaste spinel and with trace corundum, suiphide (pyrite, pyrrhotite, and chalcopyrite) and zircon. In this study, major element 108 Figure 4.1 Simplified geological map ofthe Grenville Province adapted from Davidson (1998). Inset map in the lower right part shows the relative location of the map area in North America. Anorthosite massifs and related mangerite and granitic rocks (AMCG suites) are identified as well as associated Fe-Ti±P mineral deposits as shown in Corriveau et al. (2007): (a) Irvy and Desgrosbois; (b) Saint-Hypolyte; (c) Saint-Urbain; (d) Mine Canada Iron; (e) Saint-Charles; (f) La Hache-Est; (g) Buttercup; (h) Lac Brulé; (i) Lac Dissimieu; (j) Lac La Blache; (k) Rivière Pentecôte; (1) Canton Arnaud; (m) Lac Raudot; (n) Magpie; (o) Big Island; (p) Tio Mine; (q) Everett. Boxes 2 and 3 indicate the areas covered by Figures 4.2 and 4.4. 109 Anorthosite Des Martres Mangerltic rocls Group Jotunite Salnt-Tite-des Gneiss complex Caps Group ....— Road ..- Trail * Samples Figure 4.2 Simplified geological maps of the Saint-Urbain anorthosite area and related Fe-Ti oxide ore deposits. (a) Geological map after Rondot (1989): (SANG) Saint-Anne du Nord orthopyroxene-granodiorite; (RMO) Rivière Malbaie orthopyroxene granodiorite; (SUA) Saint-Urbain anorthosite. Stars indicate sample locations of anorthosite (2006-C2 and 2043), SANG (2023) and gneissic country rock (2034). The box indicates the area covered in b. (b) Map showing the location of the Fe-Ti oxide ore samples identified by stars (after Rose, 1969): (2006) Bignell; (2009) Coulomb West; (2015) Coulomb East; (2030) General Electric; (2033) Seminaire; and (2036) Furnace. Bouchard and Glen (2031) deposits are not identified. 110 :--- 1• Figure 4.3 Structural and textural characteristics ofthe Saint-Urbain and the Big Island Fe-Ti oxide ores. a) Photograph showing a sharp contact between the Bignell deposit and host anorthosite. b) Photomicrograph (transmitted light) of ilmenite-rutile leuconorite from the Coulomb East deposit showing coexisting hemo-ilmenite with rutile that contains hematite lamellae, plagioclase, sapphirine, orthopyroxene, plagioclase and biotite (sample 2015-B4). c) Photomicrograph (polarized reflected light) ofhemo-ilmenite with the gray-white lamellae ofhematite exsolved from the anisotropic dark gray to brownish gray ilmenite (sample 2006-B4). d) Photomicrograph (reflected and transmitted light) of the “urbainite” from the Big Island deposit showing the mineral assemblage hemo-ilmenite — rutile — sapphirine (sample 2104-D). Abbreviations: (plag) plagioclase, (sa) sapphirine, (Fe-ilm) hemo-ilmenite, (ap) apatite, (bt) biotite, (opx) orthopyroxene, (ru) rutile. 111 compositions are available for 24 ilmenite separates from the deposits (see section 4.4.1 Mineral separation below). A sub-set of 15 ilmenite separates and 9 rutile separates were analyzed for trace element concentrations by HR-ICP-MS and Hf isotopic compositions by MC-ICP-MS. Ilmenite from the two host rock anorthosites, one two meters from the contact with the Bignell deposit (2006-C2) and the other from the northern part of the anorthosite massif (2043), were analyzed for their trace element and Hf isotopic compositions (Figure 4.2a). The Hf isotopic compositions of the orthopyroxene gabbronorite SANG (2023) and country rock gneiss (2034) (Figure 4.2) were also determined. 4.2.2- Havre Saint-Pierre anorthosite (Lac Allard lobe) and Big Island Fe-Ti oxide deposit The large (11,000 km2)Havre Saint-Pierre anorthositic suite is also located in the allochthonous polycyclic belt of the Grenville Province (Figure 4.1). The mostly andesine anorthosite (Hocq 1982, Madore et al. 1999) was divided by van Breemen & Higgins (1993) into four lobes (1- Lac Allard lobe; 2-North-East lobe; 3-North-West lobe; 4- South-West lobe). The Big Island deposit (age = 1053 Ma, Chapter 2), as well as Lac Tio, which is the world’s largest magmatic ilmenite deposit, occur within the Lac Allard lobe (age = 1060 Ma, Chapter 2), north of Havre Saint-Pierre on the St-Lawrence River. The Lac Allard lobe is completely surrounded by an older mangeritic envelope (age = 1148 Ma, Emslie & Hunt 1990) (Figure 4.4a). The Big Island deposit is a 30 m wide by 250 rn-long dyke (Figure 4.4b) and has similar mineralogy to the Saint-Urbain deposits described above (Figure 4.3c). Major element analyses (XRF) were determined for 11 ilmenite separates from the dyke (2100 to 2110) and two separates from nearby ilmenite pods (2111 and 2127). Trace element concentrations (HR-ICP-MS) and Hf isotopic compositions were analyzed on five ilmenite and two rutile separates from the dyke, as well as ilmenite separated from the anorthosite host rock (2114-C) (Figure 4.4b). Finally, samples from the mangerite (2130) and the country rock Buit Complex (2131) were analyzed for their Hf isotopic compositions. A complete list of the sample locations and descriptions is provided in Appendix 4.A1. 112 Figure 4.4 Simplified geological map ofthe LacAllard lobe, part ofthe Havre Saint-Pierre anorthosite suite (after Gobeil eta!. 2003). (a) Map ofthe Lac Allard lobe showing sample locations of the mangeritic envelope (2130) and the country rock gneiss (2131) represented as stars. The box indicates the area covered in b. (b) Map of the Big Island dyke area with sample locations from the massive Fe-Ti oxide dyke (2100 to 2110), massive oxide concentrations within the anorthosite (2111 and 2127-B), and anorthosite (2114-C). Paleozoic cover rocks Massive oxides rocks Anorthosite Mangerite * Bult Gnelss Complex 113 4.3- Method 4.3.1 - Mineral separation Mineral separates were prepared at the Department of Geology, Université de Liege, Belgium. Whole rock samples (1-2 kg) samples were crushed to 60-160 m grain size to liberate the grains. Rutile and ilmenite were separated using heavy liquids (bromoform and heated Thalium Clerici solution) and a Frantz Isodynamic Separator following the method outlined in Duchesne (1966). After the purity of the separated material was verified under a binocular microscope, the rutile and hemo-ilmenite grains were washed in 1 8M2 cm water and sub-boiled acetone before being pulverized with an agate mortar- and-pestle. 4.3.2- XRF analyses Mineral powders were heated at 1000°C for a minimum of 4 hours to ensure that all the iron was oxidized to Fe3. Major elements were analyzed at the Department of Geology, Université de Liege, on fused Li-borate glass disks (20 times dilution) and trace elements on pressed powder pellets with an ARL 9800 XP automatic spectrometer. Standard curves were produced with a mix of natural and in-house synthetic standards (Bologne & Duchesne, 1991). 4.3.3- Acids Acids used for mineral digestion and ICP-MS analyses at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia were all sub-boiling distilled in Teflon® bottles. The HF acid used for Hf column chemistry was sub-boiled, HC1 was quartz-distilled and HC1O4and H20were Baseline® from Seastar Chemicals Inc. 114 4.3.4- Rutile and ilmenite digestion For the rutile samples, 100 mg of powder was digested in a steel-jacketed acid- washed high-pressure PTFE bomb for 5 days at 190°C in a mixture of 5.0 mL of -28 N HF, 0.7 mL of-14 N HNO3,and 0.7 mL of’70% HC1O4. There was no residue observed in the sample solutions following digestion and 10% (j— 10 mg) aliquots were taken and placed in 7 mL Savillex® for ICP-MS analysis. The aliquots were dried down on a hotplate at 180°C before being converted to chlorides with 3 mL of -6 N HCI for 24 hours at 120°C. Once dried, aliquots were re-dissolved in 2% HC1 with a trace of HF to be analyzed by ICP-MS. The REE were analyzed on aliquots converted to nitrates and analyzed in a solution of 2% HNO3with a trace of HF. The remaining sample solutions were dried down on a hotplate at 180°C before being re-bombed in 6 mL of -6 N HC1 for 24 hours at 190°C. Rutile samples in HC1 were clear and yellow. Samples were transferred to 15 mL Savillex®, dried on a hotplate at 130°C and re-dissolved in 2 mL of 1.5 N HC1 with 16 l of ---28 N HF in preparation for ion exchange column 1 as discussed below. For the ilmenite samples, 100 mg of ilmenite powder was digested in 15 mL Savillex® on a hotplate at 120°C in a mixture of 8 mE of—28 N HF and 1 mL of—14 N HNO3 for 3 days. A greenish residue was observed. The samples were dried on a hotplate at 120°C and converted to chlorides in 6 mL of--6 N HC1 for 24 hours at 120°C. The solutions were clear and red. Aliquots (10%) for trace element determinations were taken and both the aliquots and the remaining of the sample solutions were dried down. The 10% aliquots were then re-dissolved in 2% HC1 solution with a trace amount of HF and were analyzed on the ICP-MS. The remaining samples were re-dissolved in 1.5 N HC1 with a trace amount of HF and were ready for ion exchange column 1 as discussed below. The rutile and ilmenite minerals in solution in HF can be dried down once and transformed to chlorides (and nitrates in the case of rutile), however after a second drying, the sample will not dissolve unless a minimum amount of I-IF is added. This is because a minimum quantity of fluorine must be present to maintain large quantities of titanium in solution. Thus, for rutile and ilmenite, the amount of HF added to each 115 sample was calculated so that 6 moles of fluorine were available for every 1 mole of Ti present based on the XRF analyses (Tables 4.1 and 4.2). 4.3.5- Trace element concentrations by ICP-MS Samples were analyzed on a double-focusing sector field Element2 ICP-MS (Thermo Finnigan, Bremen, Germany) at the PCIGR. The internal standard chosen was indium (see Pretorius et al. 2006). To ensure that matrix effects were limited, the rutile solutions were diluted 1 0,000x and the ilmenite samples were diluted 6,000x. To determine these optimum dilutions, at least six dilutions ranging from 22,000x to 3,000x of the same sample were run. The 1 0,000x dilution for rutile and 6,000x dilution for ilmenite were the lower dilutions that give the same concentration as the highest dilution (e.g. 22,000x). We could observe matrix effect in lower dilutions. Rutile samples were run in a matrix of 2% HNO3 + 0.01% HF and ilmenite samples were run in 2% HCI + 1% HF. The presence of trace amounts of HF in the analyzed solution can potentially reduce the accuracy of the rare earth elements (REE) analyses because of their partitioning into insoluble Ca- and Mg-fluoride phases (Yokohama et al. 1999). In this study, the CaO contents (0.01-0.06 wt% in rutile; 0.00-0.19 in ilmenite) and MgO contents (0.33-0.37 wt% in rutile; 1.19-3.69 wt% in ilmenite) are low compared to the mafic and ultramafic rocks studied by Yokohama et al. (1999) and thus the potential precipitation of these fluoride phases was less problematic. However, to test whether the addition of very small quantities of HF would affect the recovery of the REE, we analyzed synthetic titanite spiked with 150 ppm and 1500 ppm of different REE provided by Stefan Prowatke from the Mineralogisches Institut at Ruprecht-Karls University, Germany. The titanite samples were analyzed by ICP-MS following the same procedure as the rutile samples (i.e. 2% HNO3 + 0.0 1% HF). A complete trace element spectra of titanite is shown in Figure 4.5. All the REE concentrations differ by <10% relative to the Prowatke values, except for La and Gd which differ by 13%. This shows that the minor presence of HF does not significantly affect the measured concentrations of the REE. Pb, Th, U, Nb, Hf, Ta, Sr, Y and Zr were also added to the synthetic titanite. Our results show <7% relative difference between the amount of added material (e.g. 150 ppm and 1500 ppm), except for Pb and Rb. For these two elements, we obtained a relative difference between 116 Table 4.1 Major element oxide (XRF) and trace element compositions (HR-ICP-MS) of rutile separates Location Saint-Urbain Deposit Bignell Bignell Coulomb Coulomb Coulomb General General General West West East Electric Electric Electric Sample 2006-DI 2006-G1 2009-Bi 2009-D2 2015-B4 2030-B2 2030-B6 2030-C4 XRF (wt%) ULG Si02 0.22 0.07 0.17 0.19 0.17 0.39 0.35 0.22 Ti02 97.41 97.62 97.80 97.24 98.06 96.57 95.32 98.29 A1203 bd bd bd bd bd bd bd bd Fe203 0.18 0.17 0.00 0.21 0.05 0.50 1.46 0.00 MnO bd bd bd bd bd bd bd bd MgO 0.36 0.35 0.36 0.37 0.37 0.39 0.37 0.33 CaO 0.06 0.03 0.01 0.02 0.02 0.01 0.02 0.00 Total 98.23 98.06 98.19 97.90 98.57 97.84 97.49 98.64 ICP-MS (ppm) UBC Sc 4.81 4.92 6.32 5.15 5.82 4.86 5.38 6.71 V 1073 704 1169 1065 1181 984 1098 1149 Cr 1755 1095 621 658 621 415 548 605 Mn 15.4 30.3 4.6 32.3 12.9 20.5 10.6 4.6 Co 31.7 32.7 1.1 23.2 4.1 79.1 2355 27.8 Ni 100 52 6 31 14 136 2742 24 Cu 15 13 55 28 13 41 642 18 Zn 247 306 302 336 165 296 278 346 Rb 2.8 2.6 2.6 2.5 2.6 2.4 2.5 2.6 Sr 1.7 1.2 2.2 2.5 1.9 1.1 1.5 1.2 Y 1.83 1.80 0.28 1.01 0.21 2.13 1.87 1.28 Zr 2674 2710 1318 2894 1554 5958 4306 3327 Nb 135 198 105 194 122 228 550 161 Sn 26.41 21.67 10.18 18.79 10.71 25.83 22.53 13.60 La 0.700 0.451 0.356 0.188 0.205 0.183 0.682 0.604 Ce 1.122 0.611 0.799 0.353 0.515 0.339 0.922 0.938 Pr 0.136 0.059 0.110 0.047 0.073 0.039 0.109 0.084 Nd 0.566 0.214 0.548 0.230 0.323 0.192 0.443 0,340 Sm 0.096 0.05 1 0.090 0.036 0.066 0.025 0.062 0.056 Eu 0.023 0.019 0.041 0.016 0.016 0.013 0.029 0.020 Gd 0.103 0.061 0.098 0.045 0.053 0.059 0.093 0.070 Th 0.025 0.017 0.013 0.014 0.008 0.022 0.022 0.015 Dy 0.261 0.190 0.070 0.116 0.041 0.280 0.225 0.146 Ho 0.097 0.084 0.012 0.043 0.008 0.120 0.087 0.063 Er 0.397 0.385 0.039 0.189 0.022 0.665 0.45 1 0.360 Tm 0.076 0.074 0.006 0.036 0,004 0,144 0.107 0.088 Yb 0.622 0.617 0.043 0.306 0.033 1.367 1.076 0.900 Lu 0,104 0.120 0.008 0.054 0.006 0.248 0.218 0.165 Hf 60.8 64.0 38.2 69.6 45.1 111 100 74.8 Ta 9.6 11.0 7.4 12.4 8.2 13.8 36.4 10.6 Pb 1.4 7.0 4.9 1.9 1.3 7.4 7.4 1.3 Th 0.170 0.085 0.023 0.036 0.072 0.142 0.130 0.064 U 9.05 15.30 0.23 0.22 0.33 4,18 0.62 0.74 (RE) replicate analysis of the same sample solution; (DU) total procedural duplicate of the sample; bd: below detection limit 117 Table 4.1 (continued) Location Saint-Urbain Big Island Deposit General Séminaire Séminaire Dyke Dyke Electric Sample 2030-C4 2033-D 2033-D 2104-D 2109-A RE DU XRF (wt%) ULG Si02 0.21 0.17 0.08 0.38 Ti02 98.32 96.87 97.56 96.63 A1203 bd bd bd bd Fe203 0.00 0.54 0.39 0.04 MnO bd bd bd bd MgO 0.35 0.46 0.33 0.35 CaO 0.01 0.03 0.01 0.01 Total 98.70 97.91 98.26 97.36 ICP-MS (ppm) UBC Sc 6.61 4.59 4.65 5.44 6.26 V 1129 1005 997 1099 1040 Cr 598 612 611 574 447 Mn 4.8 36.4 36.1 9.7 8.1 Co 27.3 66.0 66.2 3.2 3.2 Ni 23 85 90 8 5 Cu 17 259 272 9 33 Zn 298 127 130 256 228 Rb 2.9 2.4 2.6 2.5 2.5 Sr 1.1 2.6 2.3 1.2 1.0 Y 1.22 0.52 0.49 1.45 1.78 Zr 3283 1643 1577 3931 7796 Nb 157 155 157 234 249 Sn 13.64 14.38 14.35 23.41 27.19 La 0.560 0.331 0.314 0.078 0.067 Ce 0.871 0,814 0.727 0.176 0.150 Pr 0.033 0.102 0.105 0.022 0.022 Nd 0.411 0.477 0.460 0.126 0.106 Sm 0.030 0.066 0.066 0.021 0.017 Eu 0.020 0.017 0.018 0.013 0.013 Gd 0.000 0.055 0.064 0.032 0.049 Tb 0.015 0.010 0.010 0.016 0.027 Dy 0.143 0.065 0.061 0.176 0.410 Ho 0.061 0.022 0.020 0.069 0.195 Er 0.354 0.072 0.064 0.336 0.964 Tm 0.083 0.013 0.011 0.066 0.190 Yb 0.867 0.093 0.081 0.558 1.610 Lu 0.152 0.018 0.014 0.110 0.290 Hf 74.2 51.9 52.0 86.5 144 Ta 10.6 10.1 10.2 16.0 16.6 Pb 1.4 23.2 24.2 5.3 8.4 Th 0.061 0.181 0.179 0.030 0.023 U 0.69 0.72 0.74 0.40 0.25 118 Ta bl e 4. 2 M ajo re le m en t c o m po sit io ns (X RF )o fi lm en ite se pa ra te s Sa in t-U rb ai n S i0 2 (w t% ) T i0 2 A 1 203 v203 C r 203 F e 2O3T M nO Zn O M gO Ca O LO l To ta l F e 203 Fe O Si ap fu Ti A l V Cr F e 3 F e 2 M n Zn M g Ca To ta l X ilm X he m RO R203 T O 2 0. 07 bd 43 .0 5 44 .7 8 0. 06 0. 03 0. 27 0. 29 0. 08 0. 18 57 .4 8 54 .8 2 0. 17 0. 19 bd 0. 01 2. 61 2. 86 0.0 1 0. 01 - 3. 17 - 3. 11 10 0. 64 10 0. 06 19 .81 15 .9 6 33 .8 9 34 .9 7 0. 00 2 0. 00 0 0. 80 8 0. 84 4 0. 00 2 0. 00 1 0. 00 5 0. 00 6 0. 00 2 0. 00 4 0. 37 2 0. 30 1 0. 70 8 0. 73 3 0. 00 4 0. 00 4 0. 00 0 0. 00 0 0. 09 7 0. 10 7 0. 00 0 0. 00 0 2. 00 0 2. 00 0 0. 06 bd 0. 13 bd 42 .7 1 45 .2 7 44 .3 0 46 .3 6 0. 04 0. 00 0. 10 0. 02 0. 28 0. 27 0. 30 0. 28 0. 08 0. 47 0. 18 0. 17 58 .1 4 53 .5 6 55 .3 2 53 .2 5 0. 17 0.2 1 0. 25 0. 23 0.0 1 0.0 1 0.0 1 0. 01 2. 44 2. 99 2. 78 3. 02 0.0 1 0. 04 0. 08 0. 01 - 2. 88 - 3. 10 - 2. 86 - 3. 37 10 1. 06 99 .7 1 10 0. 58 99 .9 7 20 .4 7 14 .4 6 16 .83 13 .1 4 33 .8 9 35 .1 7 34 .6 4 36 .0 9 0. 00 1 0. 00 0 0. 00 3 0. 00 0 0. 80 2 0. 85 5 0. 83 3 0. 87 2 0. 00 1 0. 00 0 0. 00 3 0. 00 1 0. 00 6 0. 00 5 0. 00 6 0. 00 6 0. 00 2 0. 00 9 0. 00 4 0. 00 3 0. 38 5 0. 27 3 0. 31 6 0. 24 7 0. 70 8 0. 73 9 0. 72 4 0. 75 4 0. 00 4 0. 00 4 0. 00 5 0. 00 5 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 09 1 0. 11 2 0. 10 4 0. 11 2 0. 00 0 0. 00 1 0. 00 2 0. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 0. 04 0. 26 0. 13 44 .6 2 43 .0 4 42 .8 6 0. 03 0. 09 0. 16 0. 29 0. 30 0. 28 0. 15 0. 18 0. 09 55 .3 2 56 .5 9 56 .8 5 0. 16 0. 15 0. 15 0.0 1 bd 0.0 1 2. 73 2. 82 2. 67 0.0 1 0. 19 0. 02 - 2. 82 - 2. 66 - 2. 86 10 0. 53 10 0. 97 10 0. 36 16 .3 0 19 .3 4 19 .47 35 .1 1 33 .5 2 33 .6 3 0. 00 1 0. 00 6 0. 00 3 0. 84 1 0. 80 7 0. 80 8 0. 00 1 0. 00 3 0. 00 5 0. 00 6 0. 00 6 0. 00 6 0. 00 3 0. 00 4 0. 00 2 0. 30 7 0. 36 3 0. 36 7 0. 73 6 0. 69 8 0. 70 5 0. 00 3 0. 00 3 0. 00 3 0. 00 0 0. 00 0 0. 00 0 0. 10 2 0. 10 5 0. 10 0 0. 00 0 0. 00 5 0. 00 1 2. 00 0 2. 00 0 1. 99 9 Lo ca tio n D ep os it Bi gn el l Bi gn el l Bi gn el l Bi gn el l Bi gn el l Bi gn el l Co ul om b Co ul om b Co ul om b Co ul om b Co ul om b Co ul om b Co ul om b G en er al G en er al W es t W es t W es t W es t Ea st Ea st Ea st El ec tri c El ec tri c Sa m pl e 20 06 -B 4 20 06 -B 7 20 06 -C l 20 06 -D l 20 06 -E S 20 06 -G i 20 09 -B l 20 09 -B 3 20 09 -B 5 20 09 -D 2 20 15 -A 4 20 15 -B 4 20 15 -C 2 20 30 -A 2 20 30 -B 2 0. 01 0. 16 0. 14 bd 0. 12 0. 05 47 .0 9 47 .9 5 47 .3 9 42 .8 6 44 .9 5 44 .3 1 0. 01 0. 00 0. 03 0. 00 0.1 1 0. 07 0. 29 0. 27 0. 30 0. 27 0. 25 0. 26 0. 13 0. 12 0. 18 0. 06 0. 16 0. 13 51 .4 9 50 .9 3 52 .0 2 57 .4 6 55 .8 9 55 .31 0. 20 0. 17 0. 16 0. 16 0. 17 0. 17 0. 01 0.0 1 0. 00 bd 0.0 1 0. 01 3. 69 3. 34 3. 22 2. 93 3. 09 3. 14 0. 01 0. 05 0. 02 bd 0. 07 0. 02 - 3. 08 - 3. 36 - 3. 17 - 3. 18 - 2. 78 - 2. 79 99 .8 4 99 .6 2 10 0. 28 10 0. 56 10 2. 02 10 0. 69 11 .95 9. 80 11 .2 2 20 .6 1 17 .2 8 17 .4 4 35 .5 8 37 .0 1 36 .7 1 33 .1 6 34 .7 5 34 .0 7 0. 00 0 0. 00 4 0. 00 3 0. 00 0 0. 00 3 0. 00 1 0. 88 3 0. 90 1 0. 88 7 0. 80 3 0. 83 2 0. 83 0 0. 00 0 0. 00 0 0. 00 1 0. 00 0 0. 00 3 0. 00 2 0. 00 6 0. 00 5 0. 00 6 0. 00 5 0. 00 5 0. 00 5 0. 00 3 0. 00 2 0. 00 3 0. 00 1 0. 00 3 0. 00 3 0. 22 4 0. 18 4 0. 21 0 0. 38 6 0. 32 0 0. 32 7 0. 74 2 0. 77 3 0. 76 4 0. 69 1 0. 71 5 0. 71 0 0. 00 4 0. 00 3 0. 00 3 0. 00 3 0. 00 4 0. 00 4 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 13 7 0. 12 4 0. 12 0 0. 10 9 0. 11 3 0. 11 7 0. 00 0 0. 00 1 0. 00 1 0. 00 0 0. 00 2 0. 00 1 2. 00 0 1. 99 9 1. 99 9 2. 00 0 2. 00 0 2. 00 0 79 83 79 84 82 86 83 79 79 87 89 88 78 82 81 21 17 21 16 18 14 17 21 21 13 11 12 22 18 19 45 46 45 46 45 47 46 45 45 47 47 47 45 45 45 11 8 11 8 9 7 9 10 10 6 5 6 11 9 9 45 46 44 46 45 47 46 45 45 47 47 47 45 45 45 Ca tio ns ar e ca lc ul at ed o n th e ba sis o f2 ca tio ns an d Fe 3+ as Fe 2+ +M g+ M n- Ti fo llo w ed by a ca lc ul at io n ba se d o n 3 o x yg en s. X ilm (il m/ ilm +h em ); X he m (he ml ilm +h em ); u rn (T i-M g-M n+ A1 12 ); he m (0 .5* (F e2 ++ Fe 3+ +M glM nT i) (L ind sle y& Fr os t 19 92 ). RO (M gO +C aO +M nO +Z nO +F eO ;a ll in m o la r % ); R203 ( A 1 203+ V 203+ C r 203+ F e 203; al li n m o la r % ); T O 2 (T i0 2 in m o la r% ). Ta bl e 4. 2 (co nti nu ed ) Lo ca tio n S i0 2 (w t% ) T i0 2 A 1 203 v203 C r 203 F e 2O3T M nO Zn O M gO Ca O LO l To ta l F e 203 Fe O Si ap fu Ti A l V Cr F e 3 F e 2 M n Zn M g Ca To ta l X ilm X he m RO R203 T O 2 bd bd bd bd bd bd 39 .7 3 43 .5 0 43 .1 4 39 .7 3 39 .9 3 43 .5 0 0. 02 bd 0. 00 0. 04 bd 0.0 1 0. 30 0. 30 0. 25 0. 30 0. 30 0. 26 0. 16 0. 17 0. 15 0. 17 0. 14 0. 16 61 .1 9 56 .4 0 57 .2 3 61 .6 2 61 .1 5 56 .8 2 0. 10 0.1 1 0. 12 0. 10 0. 11 0.1 1 0.0 1 bd bd 0. 01 bd 0.0 1 2. 55 2.9 1 2. 79 2. 57 2. 67 3. 02 bd bd bd bd bd 0. 03 - 3. 00 - 3. 46 - 3. 47 - 3. 10 - 3. 15 - 3. 05 10 1. 07 99 .9 5 10 0. 22 10 1. 43 10 1. 16 10 0. 86 26 .6 5 18 .81 19 .7 7 27 .1 2 26 .6 5 19 .4 4 31 .0 8 33 .8 3 33 .7 0 31 .0 5 31 .0 4 33 .6 3 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 74 5 0. 81 8 0. 81 0 0. 74 2 0. 74 6 0. 81 4 0. 00 1 0. 00 0 0. 00 0 0. 00 1 0. 00 0 0. 00 0 0. 00 6 0. 00 6 0. 00 5 0. 00 6 0. 00 6 0. 00 5 0. 00 3 0. 00 3 0. 00 3 0. 00 3 0. 00 3 0. 00 3 0. 50 0 0. 35 4 0. 37 1 0. 50 6 0. 49 8 0. 36 4 0. 64 8 0. 70 7 0. 70 4 0. 64 4 0. 64 5 0. 69 9 0. 00 2 0. 00 2 0. 00 3 0. 00 2 0. 00 2 0. 00 2 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 09 5 0. 10 8 0. 10 4 0. 09 5 0. 09 9 0. 11 2 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 1 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 Sa in t-U rb ai n D ep os it G en er al G en er al G le n Sé m in ai re Sé m in ai re Sé m in ai re Fu rn ac e Fu rn ac e Fu rn ac e D yk e — — D yk e D yk e D yk e D yk e D yk e El ec tri c El ec tri c Sa m pl e 20 30 -B 6 20 30 -C 4 20 31 -C 20 33 -A l 20 33 -B 20 33 -E 20 36 -A 20 36 -B 3 20 36 -D 2 21 00 21 01 -D 21 02 21 03 -A l 21 04 21 04 -D Bi g Is la nd 0. 15 0.1 1 0. 09 bd 0. 05 0. 11 bd 0.0 1 bd 43 .1 4 44 .2 9 43 .3 4 43 .0 5 44 .4 6 44 .3 6 40 .5 2 40 .9 9 39 .9 8 0. 13 0. 09 0. 09 0. 04 0.0 1 0. 04 0. 02 0. 03 0. 00 0. 30 0. 29 0. 25 0. 27 0. 25 0. 28 0. 30 0. 33 0. 36 0. 18 0. 14 0. 13 0. 07 0. 07 0. 10 0. 18 0.1 1 0. 08 55 .1 7 55 .5 6 56 .3 8 57 .0 4 53 .8 7 55 .8 4 60 .8 6 59 .9 4 61 .5 9 0. 18 0.2 1 0. 15 0. 18 0. 16 0. 21 0. 16 0. 17 0. 14 bd 0.0 1 bd 0. 01 0.0 1 0. 01 bd 0.0 1 bd 3. 06 2. 93 3. 09 3. 05 3. 14 2. 88 2. 39 2. 29 2. 22 0. 06 0. 01 0. 05 0. 05 0. 05 0. 08 0. 01 0. 01 0.0 1 - 2. 95 - 2. 81 - 3. 17 - 3. 09 - 3. 10 - 3. 10 - 3. 18 - 3. 08 - 3. 14 99 .4 3 10 0. 83 10 0. 40 10 0. 67 98 .9 8 10 0. 80 10 1. 27 10 0.8 1 10 1. 24 18 .3 0 17 .3 2 19 .3 4 20 .2 6 15 .83 17 .4 4 25 .2 7 23 .6 9 26 .1 8 33 .1 7 34 .4 0 33 .3 2 33 .0 9 34 .2 2 34 .5 6 32 .0 3 32 .6 2 31 .8 6 0. 00 4 0. 00 3 0. 00 2 0. 00 0 0. 00 1 0. 00 3 0. 00 0 0. 00 0 0. 00 0 0. 81 7 0. 83 0 0. 81 2 0. 80 6 0. 84 5 0. 83 0 0. 75 8 0. 77 2 0. 75 0 0. 00 4 0. 00 3 0. 00 3 0. 00 1 0. 00 0 0. 00 1 0. 00 1 0. 00 1 0. 00 0 0. 00 6 0. 00 6 0. 00 5 0. 00 5 0. 00 5 0. 00 5 0. 00 6 0. 00 7 0. 00 7 0. 00 4 0. 00 3 0. 00 3 0. 00 1 0. 00 1 0. 00 2 0. 00 3 0. 00 2 0. 00 1 0. 34 7 0. 32 5 0. 36 2 0. 38 0 0. 30 1 0. 32 6 0. 47 3 0. 44 6 0. 49 1 0. 69 8 0. 71 7 0. 69 4 0. 68 9 0. 72 3 0. 71 9 0. 66 6 0. 68 3 0. 66 5 0. 00 4 0. 00 4 0. 00 3 0. 00 4 0. 00 3 0. 00 4 0. 00 3 0. 00 4 0. 00 3 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 11 5 0. 10 9 0. 11 5 0. 11 3 0. 11 8 0. 10 7 0. 08 9 0. 08 5 0. 08 2 0. 00 2 0. 00 0 0. 00 1 0. 00 1 0. 00 1 0. 00 2 0. 00 0 0. 00 0 0. 00 0 1. 99 9 1. 99 9 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 80 82 79 78 83 82 74 75 73 72 80 79 72 72 79 20 18 21 22 17 18 26 25 27 28 20 21 28 28 21 45 45 45 45 46 45 43 44 43 43 45 45 43 43 45 10 9 10 11 8 9 14 13 14 15 10 10 15 15 10 45 45 45 45 46 45 43 44 43 43 45 45 43 43 45 Ta bl e 4. 2 (co nti nu ed ) Lo ca tio n Bi g Is la nd D ep os it D yk e D yk e D yk e D yk e D yk e Is la nd N -D yk e Sa m pl e 21 05 21 07 -C 21 08 21 09 -A 21 10 -A 21 11 21 27 -B S i 0 2(w t% ) 0. 06 bd bd bd bd bd bd T i0 2 38 .6 6 39 .6 4 38 .8 8 44 .0 5 38 .8 3 40 .5 6 38 .3 9 A 1 203 0. 38 0. 22 0. 05 bd 0. 03 0. 05 0. 03 V203 0.3 1 0. 30 0. 27 0. 26 0. 27 0. 29 0. 28 C r 203 0. 16 0. 17 0. 17 0. 13 0. 12 0. 06 0. 10 F e 203r 61 .8 5 61 .3 9 62 .4 6 56 .2 3 62 .1 9 60 .8 7 62 .0 6 M nO 0. 10 0. 10 0. 10 0. 11 0. 10 0. 12 0. 16 Zn O 0.0 1 0. 01 bd 0. 01 bd bd 0.0 1 M gO 2. 66 2. 72 2. 56 2. 92 2. 54 2. 54 1. 96 Ca O 0. 02 bd bd 0. 01 0.0 1 0.0 1 bd LO T - 2. 63 - 2. 73 - 3. 04 - 3. 03 - 2. 69 - 2. 86 - 2. 53 To ta l 10 1. 57 10 1.8 1 10 1. 45 10 0. 68 10 1. 40 10 1. 65 10 0. 47 F e 203 28 .5 8 27 .2 7 28 .8 0 18 .1 0 28 .5 3 25 .5 1 27 .7 6 Fe O 29 .9 3 30 .7 0 30 .2 9 34 .3 1 30 .2 9 31 .8 2 30 .8 6 Si ap fu 0. 00 1 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 Ti 0. 72 2 0. 73 8 0. 72 6 0. 82 6 0. 72 8 0. 75 7 0. 73 1 A l 0. 01 1 0. 00 6 0. 00 1 0. 00 0 0. 00 1 0. 00 2 0. 00 1 V 0. 00 6 0. 00 6 0. 00 5 0. 00 5 0. 00 5 0. 00 6 0. 00 6 Cr 0. 00 3 0. 00 3 0. 00 3 0. 00 3 0. 00 2 0. 00 1 0. 00 2 F e 3 0. 53 4 0. 50 8 0. 53 8 0. 34 0 0. 53 5 0. 47 7 0. 52 9 F e 2 0. 62 1 0. 63 6 0. 62 9 0. 71 5 0. 63 1 0. 66 1 0. 65 3 M n 0. 00 2 0. 00 2 0. 00 2 0. 00 2 0. 00 2 0. 00 3 0. 00 4 Zn 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 M g 0. 09 8 0. 10 0 0. 09 5 0. 10 8 0. 09 4 0. 09 4 0. 07 4 Ca 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 0. 00 0 To ta l 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 2. 00 0 X ilm 70 72 70 81 70 74 71 X he m 30 28 30 19 30 26 29 RO 42 42 42 45 42 43 42 R203 16 15 16 10 16 14 16 T O 2 42 42 42 45 42 43 42 100000 10000 a) 2 1000 100 2(0(1) 0.1 Figure 4.5 Primitive mantle-normalized trace element diagram showing analyses ofthe synthetic titanite 1500 and 150 by different analytical methods. Primitive mantle-normalizing values from McDonough and Sun (1995). The SIMS and EMS results were provided by Stefan Prowatke from the Mineralogisches Institut at Ruprecht-Karls University, Germany, and the HR-ICP-MS results were determined in this study. Rb Th U Nb Ta La Ce Pb Pr Sr Nd Zr Hf Sm Eu Gd Tb Dy Ho Er Yb Y Lu 122 32 and 42%. Values for Pb and Rb are indicated in the data Tables 4.1 and 4.3, but no interpretations are based on these concentrations due to poor accuracy for which we have no explanation. Ilmenite solutions were run twice, once in 2% HC1 + 1% HF and once in 2% HC1 + 0.0 1% HF. The results shown in Table 4.3 are the concentrations determined with the higher content of HF, which were chosen because the reproducibility was better. The higher amount of HF in solution does not appear to have affected the REE contents as demonstrated in Figure 4.6. For example, the La contents in 2% HC1 + 1% HF compared to the La contents in 2% HC1 + 0.01% HF yield a slope of 1.02, which implies that the La content in each type of matrix is the same. 4.3.6- Hf column chemistry Column chemistry for the separation of Hf from rutile was initially performed as described in Weis et a!. (2007), a modified procedure from Patchett & Tatsumoto (1980) and Blichert-Toft et al. (1997). This chemistry was established for basaltic to granitic rocks with no more than 4 wt% Ti02. Because rutile in this study contains between 38 and 144 ppm Hf, enough Hf was recovered, even with 2 to 10% yield, to run precise isotopic ratio analyses. The good reproducibility of the duplicates (i.e. 50 ppm) indicates that despite this low yield, no fractionation of Hf occurred during column separation. We attribute these low yields to the fact that rutile is composed of 95-98 wt% Ti02 and that the chemistry used was calibrated for much lower Ti concentrations. The ilmenite (>35 wt% Ti02)samples of this study have very low concentrations of Hf (2 ppm average) and if the ilmenite samples had been processed in the same way as the rutile samples, there would not have been enough Hf to analyze for isotopic ratios by MC-ICP-MS. Modifications to the ion exchange column chemistry protocols were made to obtain higher yields (described below). Following theses modifications, yields for Hf were between 75 and 100% for samples that passed through the second column once (total of 3 columns) and 50 to 70% for samples that passed through the second column twice (total of 4 columns). 123 HR-ICP-MS (ppm) UBC Sc 39.95 40.10 15.00 V 2075 1936 838 Cr 599 574 842 Mn 1437 1394 1479 Co 34.6 34.3 45.9 Ni 77 73 78 Cu 16 16 19 Zn 121 68 76 Rb 1.3 1.4 2.0 Sr 2.3 2.5 32.1 Y 0.50 0.52 6.91 Zr 31,45 32.06 116 Nb 20.79 20.61 28.35 Sn 7.20 7.32 0.65 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U 0.341 0.351 17.31 0.802 0.846 44.90 0.104 0.106 4.102 0.427 0,433 14.51 0.087 0,100 2,031 0.051 0.053 1.410 0.095 0.111 2.111 0.015 0.015 0.401 0,091 0.094 1.326 0.0 19 0.02 1 0,347 0.077 0.080 0.555 0.018 0,017 0.074 0.177 0.185 0.469 0.037 0.037 0.077 1.40 1.61 3.90 1.28 1,33 1,69 7.1 7.6 1.5 0.023 0.029 0.167 0.044 0,052 0.4 15 42.89 66.57 46.22 2010 2071 1976 3566 1116 955 1875 1385 1512 79.3 117.8 88.4 235 291 127 99 5 2 79 45 127 1.4 1.4 1.5 1.6 3.9 5.6 0.28 0.47 0.58 19.94 22.77 28.76 1.52 1.36 2.83 2.81 0.91 1.14 0,187 0.309 0.504 0.374 0.913 1.386 0,044 0,120 0,179 0,164 0,543 0,804 0,032 0,127 0.174 0.010 0.050 0,041 0.003 0.093 0.134 0.005 0.018 0.022 0.034 0.086 0.113 0.007 0.016 0.021 0,044 0.039 0.055 0.0 12 0.006 0.008 0.163 0.066 0.069 0.035 0.013 0,013 0.74 0.64 1.05 0.01 0.00 0.20 15.5 6.8 4.8 0.015 0.034 0.149 0.078 0,002 0.008 62.13 1962 1118 1357 100.1 181 4 97 1.4 2.5 0.49 22.43 2,38 0.70 0.436 1.372 0.169 0.751 0.169 0.044 0,114 0.019 0.087 0.0 15 0.042 0.006 0,061 0,011 0.68 0.18 12.9 0,064 0.003 Table 4,3 Trace element compositions (HR-ICP-MS) of ilmenite separates Location Saint-Urbain Deposit Bignell Bignell anorth. Bignell Coulomb Coulomb Coulomb Coulomb West East East East Sample 2006-Cl 2006-Cl 2006-C2 2006-Dl 2009-Bl 2015-A4 2015-A4 2015-B4 RE DU 44,67 1921 851 1465 88.4 125 2 106 1.5 5,7 0.56 26.85 2.79 1.09 0,483 1,324 0.175 0.770 0.171 0.037 0.135 0.021 0.114 0.019 0.054 0.008 0,064 0.012 1.03 0.18 4.8 0.150 0.009 RE: replicate analysis of the same solution; DU: duplicate analyses are multiple digestions of a powdered mineral separate. Anorth. (anorthosite). 124 Table 4.3 (continued) Location Saint-Urbain Deposit Coulomb General General General General Séminaire Séminaire Furnace East Electric Electric Electric Electric Sample 2015-C2 2030-A2 2030-B2 2030-B6 2030-C4 2033-Al 2033-E 2036-B3 ICP-MS (ppm) UBC Sc 35.06 34.97 35.41 34.17 49.07 33.49 35.10 39.72 V 1918 1930 1954 1817 2075 1847 1875 2374 Cr 422 1016 1029 1001 872 518 637 778 Mn 1318 1448 1466 1409 1759 1485 1752 1437 Co 59.0 77.7 78.7 83.5 95.1 50.5 94.3 92.9 Ni 177 159 161 248 222 79 139 98 Cu 16 8 8 7 23 47 36 25 Zn 54 43 43 21 24 27 94 20 Rb 1.3 1.4 1.4 1.3 1.4 1.4 1.5 1.5 Sr 0.5 2.2 2.2 2.2 2.1 1.4 6.4 2.2 Y 0.25 0.26 0.26 0.27 0.23 0.23 0.69 0.24 Zr 40.45 41.15 41.67 18.05 17.35 78.46 23.49 39.36 Nb 17.33 16.13 3,94 5.92 1.56 16.98 23.73 22.80 Sn 4.16 4.01 4.06 3.64 2.10 4.24 6.12 4.79 La 0.048 0.148 0.149 0.220 0.188 0.252 1.035 0.139 Ce 0.113 0.299 0.303 0.447 0.328 0.219 2.092 0.290 Pr 0.0 15 0.040 0.040 0.054 0.034 0.023 0.224 0.037 Nd 0.062 0.154 0.156 0.186 0.135 0.106 0.826 0.148 Sm 0.016 0.034 0.034 0.036 0.030 0.013 0.164 0.030 Eu 0.005 0.014 0.014 0.019 0.013 0.005 0.054 0.010 Gd 0.012 0.034 0.035 0.039 0.025 0.010 0.195 0.038 Th 0.003 0.006 0.006 0.007 0.005 0.003 0.029 0.005 Dy 0.021 0.034 0.035 0.038 0.021 0.019 0.155 0.022 Ho 0.005 0.007 0.007 0,008 0.004 0.004 0.027 0.005 Er 0.023 0.019 0.019 0,021 0.013 0.018 0.076 0.018 Tm 0.006 0.004 0,004 0.003 0.003 0.004 0.010 0.003 Yb 0.082 0.041 0.041 0.020 0.025 0.048 0.081 0.031 Lu 0.014 0.009 0.009 0.004 0.006 0.009 0.018 0.007 Hf 2.28 2.59 0.94 0.97 0.68 4.13 1.95 2.48 Ta 1.18 1.14 0.19 0.39 0.05 1.15 1.47 1.41 Pb 10.7 7.5 7.6 12.1 11.2 14,4 9.9 7.8 Th 0.017 0.016 0.016 0.025 0.019 0.010 0.042 0.017 U 0.023 0.007 0.007 0.004 0.004 0.007 0.011 0.025 125 Table 4.3 (continued) JCP-MS (ppm) UBC Sc V Cr Mn Co Ni Cu Zn Rb Sr Y Zr Nb Sn La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U 44.41 34.28 2573 2267 511 210 1185 2197 51.0 106.5 15 78 6 8 75 75 1.3 2.0 0.6 3.9 0.26 0.58 66.68 127.26 30.48 26.47 5.20 0.41 0.085 0.429 0.216 0.846 0.03 1 0.082 0.143 0.333 0.030 0.058 0.007 0.035 0.032 0.059 0.004 0.008 0.018 0.045 0.004 0.009 0.010 0.040 0.001 0.007 0.006 0.118 0.001 0.019 3.38 5.20 1.70 1.80 9.7 1.3 0.018 0.026 0.007 0.014 Location Saint-Urbam Big Island Deposit Furnace anortho- Dyke Dyke Dyke Dyke Dyke Dyke anorth. site Sample 2036-D2 2043 2100 2101-D 2104-D 2104-D 2108 2109-A 2114-C RE 28,46 32.75 38.60 39.79 26.91 35,51 22.62 2169 1836 1877 1780 2246 1837 2067 1094 800 1054 1014 1244 931 654 917 976 947 912 892 987 1643 128.0 150,7 138.3 141.2 129.4 146.5 120.9 176 300 282 288 235 299 160 7 7 97 100 13 50 16 65 81 28 65 55 47 95 1.4 1.6 1.5 1.6 1,4 1.5 1,4 0.6 0.6 0.8 0.7 0.5 0.8 7.9 0.28 0.23 0.21 0,21 0.24 0.24 0.94 55.88 19.28 21.98 21.94 61,96 19.58 48.23 22.80 3.08 3.11 3.20 23,41 3.05 22.65 6.01 1.87 2.87 2.95 5.55 1.94 0.91 0.122 0.054 0.048 0.050 0,107 0.049 1.149 0.345 0,146 0.110 0.112 0.226 0.095 2.843 0.042 0.018 0.013 0.015 0.028 0.014 0.263 0.197 0.080 0.056 0.061 0.117 0.047 1.104 0.047 0.020 0.012 0.010 0.022 0.013 0.159 0.015 0.008 0.008 0.010 0.004 0.007 0.091 0.043 0.019 0.014 0.014 0.017 0.010 0.186 0.006 0.003 0.002 0.002 0.002 0.003 0.037 0.024 0.016 0.012 0.012 0.015 0.014 0.138 0.005 0.003 0.003 0.003 0.003 0.003 0.047 0.019 0.012 0.012 0.013 0.012 0.011 0.098 0.004 0.003 0.003 0.003 0.003 0.002 0.0 13 0.039 0.028 0.034 0.032 0.032 0.017 0.135 0.008 0.005 0.006 0.006 0.007 0.004 0.026 3.09 1.01 1.16 1.13 4.04 1.07 2.43 1.46 0.24 0.18 0.19 1.53 0.18 0.96 7.2 8.7 16.6 17.0 7.3 10,8 0.3 0.007 0.006 0.005 0.008 0.004 0.005 0.030 0.011 0.005 0.002 0,002 0.004 0.002 0.028 126 0.6U- = 0 .1 0.4 = -° 03 (1 0.2 2 . 0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 La (ppm) in 2% HCI + <0.01% HF Figure 4.6 La concentrations in ilmenite measured by HR-ICP-MS on different acid solutions. Error bars are 2 sigma. y= 1.03x-O.02 R2=O.98 127 1st column: Fe and REE elimination The ilmenite samples were dissolved in 2 mL of 1.5 N HC1 + 7 j.tL of-28 N I-IF (for complete digestion technique, see section 3.4.4) and centrifuged for 7 minutes before being loaded on a Teflon® column filled with Bio-Rad® AG5OW-X8 100-200 mesh resin (20 cm high x 1 cm diameter) as shown in Figure 4.7. One mL of 1.5 N HC1 was added to the column followed by 9.5 mL of 2.5 N HC1. This combined total of 12.5 mL of acid contained 95% of the Ti, 90-100% of the Hf, 50% of the Zr, 10% of the Cr and —1% of the Fe and was collected in 15 mL Savillex® and dried down on a hotplate at 120°C. Iron is the next element to be eluted in a strong yellow complex. To ensure the best Fe separation possible at this step, the 15 mL Savillex® was carefully removed as soon as the first drop of yellow acid was observed even if the 12.5 mL of acid had not fmished passing through the column. The bulk of the Fe and the REE were washed with one full reservoir of 6 N HC1. The resin was then washed with another full reservoir of 6 N HC1 and backwashed in 1.5 N HC1 followed by cleaning with one reservoir of 1.5 N HC1, after which the column was ready for other samples. 2d column: Ti elimination (—90%) We changed the purpose of the second column, which in whole rock chemistry is typically used to separate phosphorus contained in the sample. The presence of insoluble Hf- and Zr-phosphates, in acids other than HF, can prevent a successful/pure separation of Hf (Patchett & Tatsumoto 1980). Given that there is no detectable P in ilmenite, this was not of importance for the samples in this study. Also, the second column has been used to separate Cr as Cr3 will be oxidized to Cr6 during the evaporation step (see below) and can disturb the separation on the last column as discussed in detail by Blichert-Toft (2001). Multiple tests to understand the behavior of Ti and Hf on this column were made to achieve a maximum separation of Ti at this stage and to improve the yield. This was critical due to the large amount of Ti in the samples and greatly facilitates the later steps of the Hf purification process. In Figure 4.8, we show elution curves for ilmenite sample 2030-B2 processed under different conditions. The resin used in all cases is the Bio-Rad® AG1-X8 100-200 mesh anionic resin. In Figure 4.8a, all the Ti and Hf is in the form of TiF62 and HfF62 in 0.1 N 128 Steps: 1column 2’ column Evaporation 3rd column Resin: AG5OW-X8 Scale: Load sample in 2 mLof 1.5 N HCI AG1-X8 AG5OW-X8 1:2 Load sample in HCIO4 + 300 jiLof 2.5 N HCI + 30 pLof H20 Wash Ti with 5 mL of 2.5N HCI Collect: first 10 mL in 2.5 N HCI 5 mL in 0.5 N HCI/0.3 N HF Yield: 10% of Cr 50% of Zr 90-100% of Hf 95% of Ti 1% of Fe 5-7% of Cr 50% of Zr 73-100% of Hf 10% of Ti 1% of Fe 0.5% of Cr 50% of Zr 70-100% of Hf no detectable Ti or Fe Figure 4.7 Schematic representation of the developed methodology for the separation of Hf from high Ti-bearing minerals (>40 wt% Ti02). See text for details. 1:6 1:2 1:1.5 Load sample in 1 mL of 0.1 N HF/0.5 N HCI + 7 pL of conc HF Add 0.5 mL of to sample and until volume is Remove trace of HF first 7 mL in 0.1 N HF/0.5 N HCI 129 0.04 14000 0.03 10000 0.02 6000 2000 0.01 2 2 2.5 N HCI Figure 4.8 Elution diagrams ofHfand Ti for the second column ofthe Hfseparation procedure. a) Result for a smaller sample size (—0.05 g) and lower normality ofHF acid (0.1 N HF): Hfand Ti stick to the resin in 0.1 N HF/0.5 N HC1 and are eluted together in 2.5 N HC1. b) Results for a larger sample size (—0.1 g) and lower normality ofHF acid (0.1 N HF): >73% ofthe Hfand only 10% ofthe Ti are eluted in 0.1 N HF!0.5 N HC1, the remaining metals are eluted in 2.5 N HC1. c) Results for a larger sample size (—0.1 g) and higher normality ofHF acid (1 N HF): 20 to 30% ofthe Hfand Ti are eluted in 1 N HF/0.5 N HC1, the remaining metals are eluted in 2.5 N HC1. d) Results for a larger sample size (—0.1 g) and highest normality of HF acid (4 N HF): 50-55% of the Hf and Ti are eluded in 4 N HF/0.5 N HG, the remaining metals are eluted in 2.5 N HC1. Sample 2030-62 Sample 2030-62 • Hf: 0.058 pg [_a 0.06 • Hf: 0 076 pg Lb5000ci Ti: 11,540 pg 0.04 3000 ppp 1000 AZ2ZI0.02 3 2 2 2 21 1 1 1 1 1 0 1 1 1 2 2 2 2 4- r 4- 0 C) a 9- r C) a 0.06 0.04 0.02 0 0.16 0.12 0.08 0.04 0 0.1 N HF/0.5 N HCI Sample 2006-Di 0.1 N HF/0.5 N HCI 2.5 N HCI Sample 2033-Al •Hf: ó.257g a Ti: 20,593 pg 7000 cnnn t .Jvvu 0 -h 3000 zi 1000 9000 7000 5000 3000 zi 1000 • Hf: 0.093 pg J27P\ 3 3 3 22.5 2.5 2.5 0 1.5 N HF/0.5 N HCI 2.5 N HCI mL eluted 3 3 3 22.S 2.5 2.5 4 N HF/0.5 N HCI 2.5 N HCI mLeluted 130 HF/O.5 N HC1, which will stick to the anionic resin until no more F (in the form of HF) is provided and only 2.5 N HCI is added to the column. The relative selectivity of the resin is 22 for CF and 16 for F (Bio-Rad® instruction manual), which means that CF has a higher affinity with the resin than F. When large amounts of CF are introduced, TiF62 and HIP6 are eluted. In Figure 4.8b, the size of the sample is increased, but the normality of the acid remains the same. The majority of the Hf is eluted in the first 3 mL, as well as some Ti, but most of the Ti is eluted when the 2.5 N HC1 is added to the column. Because most of the sites will be occupied by cr and some F, to ensure that all the TiF62 and HfF62 are able to attach to the resin, the correct sample/resin ratio is needed. From Figures 4.8a and 4.8b, we could have exceeded the correct sample/resin ratio or we could have a sample too large for the quantity of HF present. If the quantity of HF present is used first to complex the Ti as TiF62,the Hf might not all be complexed as HfF62 if the quantity of HF is not sufficient and therefore it would be eluted in the first few mL. Cases presented in Figures 4.8c and 4.8d show results for samples that are as large as in Figure 4.8b (e.g. —O.1 g), however the elutions are different (note change in vertical scale). We can thus rule out the possibility that the sample/resin ratio was exceeded. In this case, it is the quantity of sample versus the normality of HF that controls the elution of Hf and Ti. For the cases in Figures 4.8c and 4.8d, the samples are dissolved in 1.5 N HF/0.5 N HC1 and 4 N HF/O.5 N HC1, respectively. Because the acidity is higher, part of the Ti and Hf will be present as TiHF6 and HfHF6 and, because of the lower charge, some Hf and Ti will be lost in the first mL of elution. Some of the Ti and Hf will be found as TiF62 and HfF62 and will be eluted only when 2.5 N HC1 is put on the column. In light of these results, we chose to process the samples following the example in Figure 4.8b as detailed below. The Bio-Rad® AG1-X8 100-200 mesh anionic resin was first pre-cleaned in batch three times with 18 M2 cm water (fine particles are decanted), followed by three steps of 6 N HC1 and rinsing in 18 M cm. A polypropylene column (Poly-Prep® Bio-Rad) was filled with a bed of fresh resin (3.7 cm high x 1 cm diameter) for each sample (Figure 4.7). The resin was then cleaned again in the column with three cycles of 10 mL of 6 N HC1 followed by 10 mL of 18 M cm water, 10 mL of 24N HF, three steps of 10 mL of 18 M2 cm water, and fmally with 10 mL of 0.1 N HF/0.5 N HC1. The samples were 131 dissolved in 1 mL of 0.1 N HF/0.5 N HC1 + 7 tL of28 N HF, which is equivalent to 1 N HF/0.5 N HC1 for the first mL. The samples were loaded on the column after being centrifuged for 7 minutes. Six mL of 0.1 N HF/0.5 N HC1 were added to the column and the total 7 mL of acid were collected in a 10 mL Teflon® beaker. Only 10-15% of the initial Ti content of the sample was collected here along with 73-100% of the Hf, 50% of the Zr, 5-7% of the Cr, and <1% of the Fe. If the Cr content of the sample is high, a second pass on this column can be done. This second pass would be carried out as described in Weis et al. (2007). This sample is loaded in 1 mL of 0.1 N HF/0.5 N HC1, followed by cleaning of the Cr and the rest of the matrix in 10 mL of 0.1 N HF/0.5 N HC1, and finally collection of the sample in 5 mL of 2.5 N HC1. After this second pass, only 6-10% of the initial Ti content of the sample is collected along with 67-73% of the Hf, 4 1-45% of the Zr, 0.02% of the Cr, and no detectable Fe. Evaporation in HC1O4:HF elimination This step is necessary to successfully separate the remaining Ti from Hf and Zr on the 3r column. The last column is a cationic exchange column and thus if some Hf is complexed as HfF62,it will not stick to the resin. Four evaporation steps are done in HC1O4to evaporate all the HF from the sample (Figure 4.7). First, 500 gL of HC1O4 are added to the 7 mL collected after the second column. The sample is evaporated at 180°C in a perchioric hood on a hotplate until the volume reaches about 250 pi. Another 500 jiL of HC1O4 are added to the remaining 250 j.tL and the volume is taken down to —250 jiL again. This is repeated two more times. The volume obtained at the end of each evaporation step can vary from one sample to another depending on the quantity of Ti present. In a too small volume, a white precipitate of Ti02 incorporating the Hf and Zr will be formed. To put the precipitate back in solution, a drop of concentrated HF is added and the evaporation procedure must be started all over again. 3rd column: Ti elimination The addition of —30% H2O in the system is important to form compounds like Hf(H2O)4instead of Hf(OH)3,which have a lower charge. The cations with lower 132 charge will not attach as well to the Bio-Rad® AG5OW-X8 200-400 mesh cation exchange resin used in this column. Ti will more likely form a complex such as TiO(H20)and, because of the lower charge, will be eluted before the collection of Hf and Zr. Adding greater amounts of 11202 to the sample does not improve separation and creates large amounts of oxygen bubbles in the column thus making elution difficult. The relatively acidic environment created by the 2.5 N HC1 added at the beginning of this step of column chemistry is necessary to ensure that the reaction Hf’1 + H20 E--) Hf02 + .4+. 4++ 2H moves to the left and so that all Hr will be complexed as Hf(H20) and will attach well to the resin. Once some HF is introduced in the system, Hf and Zr will form anionic complexes with F, which are not retained by the cationic resin. The resin was first cleaned in batch three times with 18 M cm water, then three times with 6 N HC1 and once with 4 N HF before being rinsed again in 18 M cm water. Teflon® columns were filled to a height of 12 cm x 0.5 cm of diameter (Figure 4.7). Once in the column, the resin was rinsed with 12 mL of 2.5 N HC1. To prepare the sample, 300 j.tL of 2.5 N HC1 + 30 1iL ofH20was added to the 250 ji.L drop of HC1O4 containing the sample from the previous evaporation step. After the sample was loaded on the column, the Ti was washed in 0.4 mE followed by 5 mL of 2.5 N HC1 + a trace of H20 (e.g. 100 ul ofH20 in 100 mL of 2.5 N HC1). The Hf and Zr were then collected in 7 mE Savillex® with 5 mL of 0.5 N HC1/0.3 N HF. For samples passed once on the second column, the yield was 75-100% for Hf, 50% for Zr, and 0.5% Cr. For samples passed twice on the second column, the yield was 63-71% for Hf, 44-47% for Zr, and 0% for Cr. The contents of Ti and Fe in these 5 mL collected were below the ICP-MS detection limit. 4.3.7- Hf isotopic analyses by MC-ICP-MS Hf isotopic compositions were analyzed by static multi-collection using a Nu Plasma MC-ICP-MS with a desolvating nebulizer (DSN) at the Pacific Centre for Isotopic and Geochemical Research (PCIGR), University of British Columbia. Masses‘72Yb ‘74Hf, ‘75Lu,‘76Hf,‘77Hf, ‘78Hf ‘79Hfand‘80Hfwere measured simultaneously in collectors L3 to 114. Hf isotopic measurements were normalized toi79Hf/l7f 0.7325 using an exponential correction. 133 The corrections of isobaric interferences on‘74Hf and on‘76Hfwere done using natural abundances of isotopes (‘72Yb: 0.2183;‘74Yb: 0.3 183; 176yb: 0.1276;‘75Lu: 0.97416; ‘76Lu: 0.02584) and by measuring 172y10 and‘75Lu, which are corrected for mass bias as recorded by‘79Hf/’7f. Rutile samples were analyzed during one session and ilmenite samples were analyzed during three sessions. When the rutile samples were analyzed in August 2006 (n = 13),‘72Yb/’7Hfwas 0.000 1 ± 0.0003 and‘75Lu/’7Hfwas 0.00006 ± 0.000 1, which corresponds to 280 ± 350 ppm for Yb and 5 ± 6 ppm for Lu on ‘76Hf, respectively. For the ilmenite samples measured in March 2007 that were passed twice on the second column,‘72Yb/’7Hfwas 0.0006 ± 0.002 and‘75Lu/’7Hfwas 0.00003 ± 0.000 1, which corresponds to 127 ± 510 ppm of Yb and 3 ± 10 ppm of Lu on ‘76Hf, respectively. For the ilmenite samples also analyzed in March 2007 that were passed once on the second column,‘72Yb/177Hfwas 0.0007 ± 0.000 1 and‘75Lu/177Hfwas 0.00003 ± 0.00006, which corresponds to 147 ± 281 ppm of Yb and 3 ± 6 ppm of Lu on ‘76Hf, respectively. Earlier ilmenite samples analyzed in August 2006 before the chemistry was modified (yields between 3 and 5%) had‘72Yb/’7Hfof 0.0007 ± 0.0014 and‘75Lu/’7Hfof 0.0003 ± 0.0006, corresponding to 1583 ± 2956 ppm of Yb and 32 ± 61 ppm of Lu on‘76Hf, respectively. The modification of the chemistry improved the purity of the samples. It is also important to note that the ilmenite samples had approximately the same proportion of Yb and Lu to Hf whether they went once or twice on the second column. However, the samples that were passed only once on the second column have a better reproducibility and higher yields as previously stated. Interference- corrected analyses of duplicate ilmenite samples processed before and after the chemistry was modified have the same measured values within 2 standard deviations. No corrections were necessary for interferences of‘80Ta and‘80W on‘80Hf‘80Hf/’77f 1.886940 ± 0.000 130 (2 SD) for 54 samples analyzed, which is comparable to the values of180Hf/’77= 1.886887 ± 0.000129(2 SD) for the JMC-475 Hf standard analyzed (n = 66) during data collection. The JMC-475 Hf standard solution was analyzed in-between every two samples to record any drift during the analytical run. The‘76Hf/’7fratios were corrected to the standard average during the run normalized to the JMC-475 Hf value of 0.282160 (Stevenson & Patchett, 1990; Blichert-Toft et al., 1997; Vervoort & Blichert-Toft, 1999). The average‘76Hf/177fratio for the 66 standards run during the 134 collection of the data presented in Table 4.4 and 4.5 is 0.282176 ± 0.000018(2 SD). Total procedural blanks have 40 pg of Hf, which represents <0.00 1% for the rutile samples and <0.05% for the ilmenite samples. 4.4- Results 4.4.1 Rutile chemistry The major element composition of rutile from the Saint-Urbain and Big Island Fe-Ti oxide deposits shows a limited amount of variation (Ti02 = 95.3 to 98.3 wt%; Fe203= below detection to 1.46 wt%; MgO = 0.33 and 0.46 wt%) (Table 4.1). In contrast, the strongly compatible 4+ and 5+ charged high field strength (HFS) elements vary over a large concentration range: Nb = 105 to 550 ppm, Ta = 7.4 to 36.4 ppm, Zr = 1318 to 7796 ppm, and Hf= 38.2 to 144 ppm (Table 4.1). The HFS elements are also strongly positively correlated with each other (Figure 4.9) and MgO is inversely correlated with the HFS elements. Chondrite-normalized REE patterns display two distinct shapes for rutile (Figure 4.lOa). The majority of the samples from Saint-Urbain and the two Big Island samples are characterized by a U-shaped profile with relatively enriched LREE and HREE and depleted MREE, which results in a overall LaCN/LUCN ratio lower than one (0.02-0.7). At Saint-Urbain, an anti-clockwise rotation in the REE patterns, illustrated by a change in the LaCN/LUCN from sample 2006-D1 (0.7) to sample 2030-B2 (0.08), is correlated with Nb content. Three samples (2009-B1, 2015-B4, and 2033-D) are enriched only in the LREE resulting in a LaCN/LUCN above one (1.91-4.62). The ratio of LaJLu is inversely exponentially related to the Nb concentration (Figure 4.9). The range of Eu/Eu* is from 0.86 to 1.53 (small negative to positive Eu anomaly) and two samples have a ratio of’--l (2006-Gi and 2030-B2: Eu/Eu* 1.03) (Figure 4.lOa). The compatible HFS elements in rutile all show prominent positive anomalies in an extended primitive mantle-normalized trace element diagram (Figure 4.11). Pb concentrations determined by isotope dilution (0.04-2 ppm, Chapter 2) for some samples 135 Ta bl e 4. 4 H fi so to ni c co m n o si tio ns o fs am n le s fro m th e Sa in t-U rb ai n ar ea C2 SE o n H fc o n ce n tr at io ns ar e 0. 05 fo r ilm en ite an d 0. 8 fo rr u til e an d w ho le ro ck d l7 6 u se d in all ca lc ul at io ns is l.8 67 x1 0- 1 1( SO de rlu nd et al ., 20 04 ); sa m pl es ar e ag e- co rr ec te d to 10 53 M a bu ts am pl e 20 23 to 10 61 M a (se eC ha pt er 2) pr es en t-d ay ch on dr ite iso to pi c co m po si to ns u se d ar e ‘7 6 H g ” 7 7 H f = 0. 28 27 72 an d ‘7 6 L u I ’ 7 7 H j0 .0 3 3 2 (B lit ch ert -T of t& A lb ar êd e 19 97 ) de pl et ed m an tle iso to pi c co m po si tio ns u se d ar e 1 7 6 H f Y 1 7 7 H f’ = 0 .2 8 3 2 2 4 an d ‘7 6 L u / ’ 7 7 H f” 0 .0 3 8 1 3 (V erv oo rt & B lic he rt- To ft 19 99 )( 1) w ith ‘7 6 L u / ’ 7 7 H f o ft he m in er al , ( g) 1 7 6 L u / ’ 7 7 H f= 0 .0 0 7 2 , (h) ‘7 6 L u / ’ 7 7 H 0 .0 1 6 5 D ep os it Sa m pl ea M at er ia l 1 7 6 H fY 2S E Lu ” Hf c ‘7 6 L u / (* l0 ) c H f 1’ ‘7 6 H 1 7 T T g T h ‘7 7 H fm (* 10 6) (p pm ) (p pm ) ‘7 7 H f ‘7 7 H f (G a) 0. 00 72 0. 01 65 B ig ne ll 20 06 -C l Ilm en ite 0. 28 22 71 5 0. 03 7 1. 40 0. 00 37 3 0. 28 21 97 ± 10 3. 00 * 0. 36 1. 47 1. 49 1. 68 B ig ne ll 20 06 -D i Ilm en ite 0. 28 23 34 * 5 0. 03 5 0. 74 0. 00 66 5 0. 28 22 02 * 18 3. 16 ± 0. 63 1. 50 1. 49 1. 67 Co ul om b W es t 20 09 -B l Ilm en ite 0. 28 22 96 * 13 0. 01 3 0. 64 0. 00 27 9 0. 28 22 40 ± 21 4. 52 ± 0. 75 1. 39 1. 42 1. 58 C ou lo m bE as t 20 15 -A 4 Ilm en ite 0. 28 22 77 * 7 0, 01 3 1. 05 0. 00 17 5 0. 28 22 42 ± 11 4, 59 + 0. 40 1. 38 1. 42 1. 57 Co ul om b Ea st 20 15 -A 4D u Ilm en ite 0. 28 22 76 * 6 0. 01 2 1. 03 0. 00 16 9 0. 28 22 42 ± 10 4. 59 ± 0. 36 1. 38 1. 42 1. 57 C ou lo m bE as t 20 15 -B 4 Ilm en ite 0. 28 22 78 ± 8 0. 01 1 0. 68 0. 00 23 7 0. 28 22 31 + 15 4. 19 ± 0. 52 1. 40 1. 44 1. 60 Co ul om b Ea st 20 15 -C 2 Ilm en ite 0. 28 22 53 ± 5 0. 01 4 2. 28 0. 00 08 8 0. 28 22 35 ± 6 4. 34 ± 0. 22 1. 38 1. 43 1. 59 G en er al El ec tri c 20 30 -A 2 Ilm en ite 0. 28 22 33 ± 5 0. 00 9 2. 59 0. 00 04 9 0. 28 22 23 ± 7 3. 92 + 0. 23 1. 38 1. 45 1. 62 G en er al El ec tri c 20 30 -B 2 Ilm en ite 0. 28 22 48 ± 7 0. 00 9 0. 94 0. 00 13 5 0. 28 22 34 ± 12 4. 28 + 0. 41 1. 40 1. 43 1. 59 G en er al El ec tri c 20 30 -B 2D u Ilm en ite 0. 28 22 60 ± 19 0. 00 9 0. 94 0. 00 13 5 0. 28 22 22 * 23 3. 85 ± 0. 83 1. 39 1. 45 1. 62 G en er al El ec tri c 20 30 -B 2D u2 Ilm en ite 0. 28 22 47 ± 5 0. 00 9 0. 94 0. 00 13 5 0. 28 22 21 ± 10 3. 82 ± 0. 34 1.4 1 1. 45 1. 62 G en er al El ec tri c 20 30 -B 6 Ilm en ite 0. 28 22 41 ± 6 0. 00 4 0. 97 0. 00 06 2 0. 28 22 29 ± 9 4. 12 ± 0. 33 1.4 1 1. 44 1. 60 G en er al El ec tri c 20 30 -C 4 Ilm en ite 0. 28 22 39 ± 10 0. 00 6 0. 68 0. 00 13 0 0. 28 22 13 ± 15 3. 54 ± 0. 54 1. 39 1. 47 1. 64 Sé m in ai re 20 33 -A l Ilm en ite 0. 28 22 51 ± 25 0. 00 9 4. 13 0. 00 03 1 0. 28 22 45 ± 26 4. 67 ± 0. 93 1. 42 1.4 1 1. 57 Sé m in ai re 20 33 -A ID u Ilm en ite 0. 28 22 25 ± 6 0. 00 9 4. 13 0. 00 03 1 0. 28 22 19 ± 7 3. 76 ± 0. 24 1. 37 1. 46 1. 63 Sé m in ai re 20 33 -E Ilm en ite 0. 28 22 44 ± 5 0. 01 8 1. 95 0. 00 13 1 0. 28 22 18 ± 7 3. 73 ± 0. 26 1. 40 1. 46 1. 63 Fu rn ac e 20 36 -B 3 Ilm en ite 0. 28 22 24 + 17 0. 00 7 2. 48 0. 00 04 0 0. 28 22 16 ± 18 3. 66 * 0. 64 1.4 1 1. 46 1. 64 Fu rn ac e 20 36 -D 2 Ilm en ite 0. 28 21 96 ± 7 0. 00 1 3. 38 0. 00 00 5 0. 28 21 90 ± 8 2. 73 * 0. 28 1.4 1 1.5 1 1. 70 Fu rn ac e 20 36 -D 2D u Ilm en ite 0. 28 21 91 + 23 0. 00 1 3. 38 0. 00 00 5 0. 28 21 95 ± 24 2. 92 ± 0. 86 1. 44 1. 50 1. 69 B ig ne ll 20 06 -D l R ut ile 0. 28 22 31 ± 6 0. 10 60 .8 0. 00 02 4 0. 28 22 26 ± 7 4. 02 * 0. 26 1. 43 1. 44 1.6 1 B ig ne ll 20 06 -G i R ut ile 0. 28 22 43 ± 9 0. 12 64 .0 0. 00 02 7 0. 28 22 37 ± 9 4. 41 * 0. 33 1. 39 1. 43 1. 59 Co ul om b W es t 20 09 -B i R ut ile 0. 28 22 54 ± 7 0. 01 38 .2 0. 00 00 3 0. 28 22 53 ± 8 4. 98 ± 0. 30 1. 38 1. 40 1. 55 Co ul om b W es t 20 09 -D 2 R ut ile 0. 28 22 51 ± 7 0. 05 69 .6 0. 00 01 1 0. 28 22 48 ± 8 4. 80 ± 0. 28 1. 35 1.4 1 1. 56 Co ul om b Ea st 20 15 -B 4 R ut ile 0. 28 22 66 * 14 0. 01 45 .1 0. 00 00 2 0. 28 22 66 ± 15 5. 42 ± 0. 54 1. 36 1. 38 1. 52 G en er al El ec tri c 20 30 -B 2 R ut ile 0. 28 22 37 * 5 0. 25 11 1. 0 0. 00 03 2 0. 28 22 31 * 6 4. 20 ± 0. 20 1. 33 1. 44 1. 60 G en er al El ec tri c 20 30 -B 6 R ut ile 0. 28 22 27 ± 6 0. 22 10 0. 5 0. 00 03 1 0. 28 22 21 ± 7 3. 84 ± 0. 25 1. 38 1. 45 1. 62 G en er al El ec tri c 20 30 -C 4 R ut ile 0. 28 22 44 * 6 0. 17 74 .8 0. 00 03 1 0. 28 22 38 * 7 4, 44 ± 0. 25 1. 40 1. 43 1. 58 Sé m in ai re 20 33 -D R ut ile 0. 28 22 48 * 8 0. 02 51 .9 0. 00 00 5 0. 28 22 47 ± 9 4, 75 ± 0. 33 1. 38 1.4 1 1. 56 Sé m in ai re 20 33 -D D u R ut ile 0. 28 22 38 * 10 0. 01 52 .0 0. 00 00 4 0. 28 22 37 ± 12 4. 41 ± 0. 41 1. 36 1. 43 1. 59 A no rth os ite 20 06 -C 2 Ilm en ite 0. 28 22 68 * 6 0. 07 7 3. 90 0. 00 28 2 0. 28 22 13 ± 8 3. 54 ± 0. 27 1. 37 1. 47 1. 64 A no rth os ite 20 43 Ilm en ite 0. 28 22 54 ± 5 0. 01 9 5. 20 0. 00 05 2 0. 28 22 44 ± 6 4. 65 ± 0. 20 1. 37 1. 42 1. 57 M an ge rit e 20 23 W ho le ro ck 0. 28 23 14 ± 4 0. 59 7 14 .1 6 0. 00 59 9 0. 28 21 94 ± 15 3. 06 ± 0. 52 1. 50 M an ge rit ic gn ei ss 20 34 W ho le ro ck 0. 28 23 67 ± 4 2. 33 5 25 .3 9 0. 01 30 7 0. 28 21 08 ± 15 - 0. 17 ± 0. 52 1.8 1 aD u ar e du pl ic at e v al ue s sh ow n to as se s da ta qu al ity bu t a re n o t u se d in ca lc ul at io ns an d Fi gu re s b2 SE o n Lu co n ce n tr at io ns ar e 0. 00 1 fo ri lm en ite an d 0. 02 fo rr u til e an d w ho le ro ck Ta bl e 4. 5 H fi so to pi c co m po sit io n o fs am pl es fro m th e La c A lla rd ar ea (H av re Sa in t-P ie rre su ite ) D ep os it Sa m pl ea M at er ia l ‘7 6 H 17 2S E Lu b H f 1 7 6 L u / 1 7 6 H f/ (*1 0.6 ) EH f T D M TD M h ‘7 7 H fm (* 1O i (pp m) (pp m) ‘7 7 H f 1 7 7 H f (G a) 0. 00 72 0. 01 65 B ig ls la nd 21 00 Ilm en ite 0. 28 22 11 ± 6 0. 00 8 3. 09 0. 00 03 8 0. 28 22 03 ± 7 3.2 1 ± 0. 23 1. 42 1. 48 1. 67 Bi g Is la nd 2l OO Du Ilm en ite 0. 28 22 14 ± 7 0. 00 8 3. 09 0. 00 03 8 0. 28 22 06 ± 8 3.3 1 ± 0. 27 1. 42 1.4 8 1. 66 Bi g Is la nd 21 01 -D Ilm en ite 0. 28 22 17 ± 7 0. 00 5 1.0 1 0. 00 07 7 0. 28 22 02 ± 10 3. 15 ± 0. 37 1.4 3 1. 49 1. 67 Bi g Is la nd 21 08 Ilm en ite 0. 28 22 08 ± 5 0. 00 7 4. 04 0. 00 02 6 0. 28 21 82 ± 6 2. 44 ± 0. 22 1.4 5 1. 52 1. 72 Bi g Is la nd 21 08 D u Ilm en ite 0. 28 21 87 ± 19 0. 00 7 4. 04 0. 00 02 6 0. 28 22 03 ± 20 3. 20 ± 0. 70 1. 42 1. 48 1. 67 Bi g Is la nd 21 09 -A Ilm en ite 0. 28 22 29 ± 8 0. 00 4 1. 07 0. 00 05 3 0. 28 22 18 + 11 3. 73 ± 0. 38 1. 40 1. 46 1.6 3 Bi g Is la nd 21 04 -D Ru til e 0. 28 22 18 ± 5 0.1 1 86 .5 0. 00 01 8 0. 28 22 15 ± 5 3.6 1 ± 0. 19 1.4 1 1. 49 1. 67 Bi g Is la nd 21 09 -A Ru til e 0. 28 22 21 ± 6 0. 29 14 3. 6 0. 00 02 9 0. 28 22 15 ± 6 3. 63 ± 0. 21 1.4 1 1. 46 1. 64 A no rth os ite 21 14 -C Ilm en ite 0. 28 22 27 + 5 0. 02 6 2. 43 0. 00 15 0 0. 28 21 97 ± 7 3. 18 ± 0. 23 1. 44 1. 46 1. 64 M an ge rit e 21 30 w ho le ro ck 0. 28 22 58 ± 5 0. 48 17 .4 0. 00 39 5 0. 28 21 73 ± 11 2. 37 ± 0. 41 1. 50 G ne iss 21 31 w ho le ro ck 0. 28 24 91 ± 5 2. 88 31 .7 0. 01 29 0 0. 28 22 36 ± 14 4. 36 ± 0. 48 1. 54 aD U ar e du pl ic at e v al ue s sh ow n to as se s da ta qu al ity bu ta re n o tu se d in ca lc ul at io ns an d Fi gu re s b 2S E o n Lu co n ce n tr at io ns ar e 0. 00 1f or ilm en ite an d 0. 02 fo r r u til e an d w ho le ro ck ° 2S E o n H fc o n ce n tr at io ns ar e 0. 05 fo ri lm en ite an d 0. 8 fo rr u til e an d w ho le ro ck d 1 7 6 u se d in all ca lc ul at io ns is 1. 86 7x 10 1 1 (S öd erl un de ta l., 20 04 ); sa m pl es ar e ag e- co rre ct ed to 10 53 M a an d sa m pl e 21 14 -C to 10 61 M a (se eC ha pt er 2) ep re se nt ..d ay ch on dr ite iso to pi c co m po sit on su se d ar e 1 7 6 H f Y 1 7 7 H f = 0. 28 27 72 an d 1 7 6 L u ! ’ 7 7 H f0 .0 3 3 2 (B litc he rt- To ft & A lb ar èd e 19 97 ) f, g an d h pr es en t-d ay de pl et ed m an tle iso to pi c co m po sit io ns u se d ar e ‘7 6 H f 7 ’ 7 7 H f = 0. 28 32 24 an d ‘7 6 L u / ’ 7 7 H f 0 .0 3 8 13 (V erv oo rt & B lic he rt- To ft 19 99 )( t) w ith ‘7 6 L u J ’ 7 7 H f o ft he m in er al , ( g) ‘7 6 L u / 1 7 7 H f0 .0 0 7 2 , (h) ‘7 6 L & 1 7 7 H f0 .0 1 6 5 200 150 2 100 50 0 1000 Figure 4.9 Trace element binary diagrams of rutile separates. Correlation coefficient values (R2) are indicated to quantitatively demonstrate the good correlation between the compatible elements in rutile. The lower R2 value for Nb- Zr relationship is probably due to the diffusion ofZr to form zircon rims (Morisset & Scoates, 2008). Zr (ppm) 3000 5000 7000 100 200 300 600 I I I A 500 400 -300 R2=0.69 1000 3000 5000 7000 Zr (ppm) 400 500 600 Nb (ppm) 50 I I 40 A 30 -20 A 0 AA, A 100 200 300 400 500 600 Nb (ppm) 138 100 La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Figure 4.10 CI chondrite-normalized REE diagrams of rutile and ilmenite from the Saint-Urbain and Big Island deposits. CI chondrite-normalizing values from McDonough & Sun (1995). a) Rutile from the Saint-Urbain and Big Island deposits. b) Ilmenite separates from the Saint-Urbain deposits and from two samples ofthe host anorthosite. c) Ilmenite separates from the Big Island deposit and from the host anorthosite. Grey fields showing the range of compositions in ilmenite from the Skaergaard intrusion (Jang & Naslund 2003) for comparison. Saint-Urbain’s rutile •Z! 2006-Di •—— 2006-Gi 2015-B4 •-— 2009-Bi 2030-B2 2009-D2 —4— 2104-D 2030-B6 2030-C4 •—*— 2033-D Saint-Urbain’s ilmenite L) I C-) I I C-) 0 C.) I • 2015-B4 . rutiIe-bearng rocks —— 2015-C2 —•-- 2006-C2 —0—2043 10 1 10 1 0.01 1 0.1 0.01 . 2030-A2 — 2033-Al 2030-B2 2033-E 2030-B6 ——2036-B3 2030-C4 —0—- 2036-D2 a-. Big Island Hmenite V.. • 2101 2104-D 2101-D • 2108 .—. I • . host rock ilmenite 2109-A rutüe-bearing rocks - 2114-C 139 10 00 — a- - 2 00 6- D i — a— 20 09 -B 1 — a— 20 06 -G i — a— 20 09 -D 2 Fi gu re 4.1 1 Pr im iti ve m an tle -n or m al iz ed tra ce el em en td ia gr am o fr u til e fro m Sa in t-U rb ai n an d Bi g Is la nd de po sit s (pr im itiv em an tle - n o rm al iz in g v al ue s f ro m M cD on ou gh & Su n, 19 95 ).a )R ut ile fro m Sa in t-U rb ai na n dB ig Is la nd . b )N on -n or m al iz ed tr ac e el em en td ia gr am o fa v ai la bl ep ar tit io n co ef fic ie nt sf or ru til e. D at a so ur ce s: (1) Fo le y et al. 20 00 ,(2 )J en ne re ta !. 19 93 ,(3 )K le m m e et al. 20 05 ,( 4) Ba sa lt: M cC al lu m & Ch ar re tte 19 78 ,W en dl an dt 19 90 ,S ch m id te t a l. 20 04 an dX io ng et al 20 05 .A nd es in e: Sc hm id t e ta !. 20 04 an dG re en & Pe ar so n 19 87 .I th yo lit eo r gr an ite :S ch m id te ta l. 20 04 . ci 4-, C (U 1 o E 0 .11 Rb Th U Nb Ta La Ce Pb Pr Sr 20 15 -B 4 20 30 -B 2 20 30 -B 6 Nd Zr Hf Sm Eu Gd Tb Dy Ho Er Yb Y Lu 20 30 -C 4 — a— 20 33 -D I 21 04 -D — a- 21 09 -A show a positive anomaly (Figure 4.11), which suggests that Pb is more compatible in rutile than the light REE. Small negative Sr anomalies are typical of all analyzed rutile and may indicate prior extensive crystallization of plagioclase of the parent magma to the deposits. The samples with a strong enrichment in the HREE (Lu 0.008-0.290 ppm) also have a larger negative Y anomalies (Figure 4.11). Although the transition metals are compatible in rutile (e.g. Foley et al. 2000; Klemme et al. 2005; Xiong et al. 2005), they are not necessarily correlated with each other or with the compatible HFS elements in rutile from Saint-Urbain and Big Island. For example, V concentrations (704-1169 ppm) show no correlation with Cr (415-1755 ppm), and are negatively correlated with the HFS elements. Scandium (4.81-6.71 ppm) and Mn (4.6-36.4 ppm) concentrations vary positively with V, but are not correlated with other elements. Cobalt and Ni are generally low (Co = 1.1-66 ppm; Ni = 5-100 ppm), except for sample 2030-B6 where both elements are