UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Metal- and alteration-zoning, and hydrothermal flow paths at the moderately-tilted, silica-saturated… Jago, Christopher Paul 2008

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

Item Metadata

Download

Media
24-ubc_2008_spring_jago_christopher.pdf [ 121.81MB ]
Metadata
JSON: 24-1.0052588.json
JSON-LD: 24-1.0052588-ld.json
RDF/XML (Pretty): 24-1.0052588-rdf.xml
RDF/JSON: 24-1.0052588-rdf.json
Turtle: 24-1.0052588-turtle.txt
N-Triples: 24-1.0052588-rdf-ntriples.txt
Original Record: 24-1.0052588-source.json
Full Text
24-1.0052588-fulltext.txt
Citation
24-1.0052588.ris

Full Text

Metal- and alteration-zoning, and hydrothermal flow paths at the moderately-tilted, silica-saturated Mt. Milligan Cu-Au alkalic porphyry deposit by CHRISTOPHER PAUL JAGO B.Sc., University of Toronto, 2005  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Geological Sciences) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) March, 2008  © Christopher Paul Jago, 2008  ABSTRACT The Mt. Milligan deposit is a tilted (~45°) Cu-Au alkalic porphyry located 155 km northwest of Prince George, B.C., Canada. It is the youngest of the BC alkalic porphyry deposits, all of which formed between 210 to 180 Ma in an extensive belt of K-enriched rocks related to the accretion of the Quesnellia-Stikinia superterrane to ancestral North America. Mt. Milligan has a measured and indicated resource of 205.9 million tonnes at 0.60 g/t Au and 0.25% Cu containing 3.7 million oz. gold, and 1.12 billion lb. copper. Shoshonitic volcanic and volcaniclastic andesites host mineralization. These have been intruded by a composite monzonitic stock (MBX stock), and associated sill (Rainbow Dike). Early disseminated chalcopyrite-magnetite and accessory quartz veins are associated with K-feldspar alteration in the MBX stock. A halo of biotite alteration with less extensive magnetite replaces host rocks within a ~150 m zone surrounding the stock, while K-feldpsar alteration extends along the Rainbow Dike and permeable epiclastic horizons. Peripheral albite-actinolite-epidote assemblages surround the Ksilicate zone. Albite-actinolite occurs at depth, and epidote dominates laterally. Copper and Au grade are maximal where the albite-actinolite assemblage overprints biotite alteration. Gold grade is moderate in association with epidote, whereas Cu is depleted. The post-mineral Rainbow Fault separates the core Cu-rich zone from a downthrown Au-rich zone. A similar zonation of metals occurs in the hanging-wall (66 zone), where a Cu-bearing, potassically-altered trachytic horizon transitions to a funnel-shaped zone of pyrite-dolomite-sericite-chlorite alteration with elevated gold. Sulfide S-isotope compositions range from -4.79 δ34S in the central Cu-Au orebody to near-zero values at the system periphery, typical of alkalic porphyries. Sulfur isotope contours reflect the magmatic-hydrothermal fluid evolution, and indicate late-stage ingress of peripheral fluids into the Cu-Au zone. Carbonate C- and O-isotope compositions corroborate the magmatic fluid path from the Cu-Au rich zone to Au-rich zone with decreasing depth. Strontium isotopic compositions of peripheral alteration minerals indicate a laterally increasing meteoric fluid component. Changes in majorii  and trace element composition of epidote and pyrite across the deposit are also systematic. These provide additional vectors to ore, and confirm the kinematics of the Rainbow Fault.  iii  TABLE OF CONTENTS ABSTRACT.....................................................................................................................ii TABLE OF CONTENTS .................................................................................................iv LIST OF TABLES...........................................................................................................ix LIST OF FIGURES ..........................................................................................................x ACKNOWLEDGMENTS...............................................................................................xvi DEDICATION ..............................................................................................................xvii CHAPTER 1: DIAGNOSTIC ALTERATION-MINERALIZATION ZONING AT THE MT. MILLIGAN SILICA-SATURATED CU-AU ALKALIC PORPHYRY DEPOSIT ................1 1.1 RATIONALE FOR STUDY........................................................................................1 1.2 STUDY METHODOLOGY ........................................................................................3 CHAPTER 2: CU-PORPHYRY SYSTEMS – CALC-ALKALINE VS. ALKALINE...........9 2.1 CU-PORPHYRY SYSTEMS .....................................................................................9 2.2 ALKALINE VS. CALC-ALKALINE ...........................................................................11 2.3 ALKALINE VS. CALC-ALKALINE – IGNEOUS ROCKS.........................................11 2.4 ALKALINE VS. CALC-ALKALINE – ALTERATION.................................................12 2.5 ALKALINE VS. CALC-ALKALINE – METAL RATIOS.............................................13 CHAPTER 3: GEOLOGICAL SETTING........................................................................16 3.1 REGIONAL SETTING.............................................................................................16 3.2 REGIONAL STRATIGRAPHY ................................................................................16 3.3 REGIONAL STRUCTURE ......................................................................................19 3.4 GEOLOGY OF THE MT. MILLIGAN PORPHYRY AREA .......................................20 3.5 DEPOSIT STRATIGRAPHY ...................................................................................21 3.6 PROPERTY GEOLOGY – MBX MAIN DEPOSIT...................................................22 3.6.1 Witch Lake succession ......................................................................................22 iv  3.6.2 Apparent breccia (volcanic-conglomerate) vs. pseudo-breccia..........................26 3.6.3 Lower Trachyte ..................................................................................................28 3.6.4 Upper Trachyte ..................................................................................................29 3.6.5 Pyroclastic flow ..................................................................................................31 3.6.6 MBX stock..........................................................................................................31 3.6.7 Rainbow Dike.....................................................................................................34 3.6.8 Faulting/brecciation associated with the MBX stock ..........................................36 3.6.9 Late-mineral dikes .............................................................................................38 3.6.10 Structurally-defined ore zones .........................................................................40 CHAPTER 4: DEPOSIT HISTORY AND PREVIOUS WORK .......................................43 4.1 HISTORY OF THE MT. MILLIGAN DEPOSIT ........................................................43 4.2 PREVIOUS WORK .................................................................................................45 CHAPTER 5: ALTERATION AND SULFIDE MINERALIZATION.................................49 5.1 ANALYTICAL TECHNIQUES .................................................................................49 5.1.1 Core logging ......................................................................................................49 5.1.2 Petrography .......................................................................................................49 5.1.3 SWIR (short wave infrared) spectra...................................................................49 5.1.4 SEM (scanning electron microscope) ................................................................50 5.1.5 XRD (X-ray diffractometry).................................................................................50 5.2 OVERVIEW ............................................................................................................50 5.2.1 MBX zone ..........................................................................................................50 5.2.2 66 zone ..............................................................................................................55 5.2.3 DWBX zone .......................................................................................................55 5.3 K-FELDSPAR ± QUARTZ (potassic shell).............................................................55 5.4 QUARTZ VEINS .....................................................................................................57 5.4.1 Quartz-K-feldspar ± sulfide veins (E2-stage) .....................................................57 5.4.2 Accessory quartz veinlets ..................................................................................58 5.5 K-FELDSPAR-BIOTITE-ACTINOLITE (inner calc-potassic shell)...........................59 5.6 MAGNETITE (potassic- and inner calc-potassic shell) ..........................................60 5.7 EARLY STAGE SULFIDE VEINS (potassic- and inner calc-potassic shell) ..........61 5.7.1 Early chalcopyrite veins (E3-stage) .................................................................. 62 5.7.2 Chalcopyrite-pyrite veins (E4-stage)................................................................. 62 5.7.3 Magnetite ± pyrite, quartz, chalcopyrite, calcite veins (E5-stage) ......................62 5.7.4 Replacement mineralization...............................................................................62  v  5.8 CU-AU GRADE (potassic- and inner calc-potassic shell) ......................................64 5.9 66 ZONE (potassic- and inner calc-potassic shell) ................................................65 5.10 CU-AU GRADE – 66 ZONE (potassic- and inner calc-potassic shell) .................68 5.11 INTERMEDIATE STAGE SULFIDE VEINS (potassic- and inner calc-potassic shell) ......................................................................................................................................69 5.11.1 Pyrite-magnetite veins (T1-stage)....................................................................69 5.12 NA-FELDSPAR-ACTINOLITE-EPIDOTE (sodic-calcic / outer calc-potassic shell)... .......................................................................................................................................70 5.13 SULFIDE MINERALIZATION (sodic-calcic / outer calc-potassic shell)................73 5.13.1 T3-stage veins .................................................................................................73 5.14 CU-AU GRADE (sodic-calcic / outer calc-potassic shell).....................................75 5.15 EPIDOTE-NA-FELDSPAR-ACTINOLITE (inner-propylitic shell)..........................76 5.16 SULFIDE MINERALIZATION (inner-propylitic shell)............................................77 5.17 CU-AU GRADE (inner-propylitic shell).................................................................77 5.18 DOLOMITE-ANKERITE-SERICITE-CHLORITE (carbonate-phyllic zone)............78 5.19 CHLORITE-HEMATITE-CARBONATE (chloritic zone)........................................81 5.20 SULFIDE MINERALIZATION (carbonate-phyllic zone) .......................................82 5.21 CU-AU GRADE (carbonate-phyllic zone) ............................................................85 5.21.1 CARBONATE-PHYLLIC VEIN (L1-stage).......................................................85 5.22 EPIDOTE-CHLORITE (outer-propylitic shell).......................................................86 5.23 SULFIDE MINERALIZATION (outer-propylitic shell) ...........................................88 5.24 CU-AU GRADE (outer-propylitic shell) ................................................................88 5.25 OTHER INTERMEDIATE TO LATE STAGE VEINS.............................................88 5.25.1 Pyrite-carbonate veins (L3-stage)....................................................................88 5.25.2 Pyrite-calcite veins (L4-stage)..........................................................................89 5.25.3 Calcite-chlorite ± tourmaline-quartz veins (L5 stage) .......................................89 5.25.4 Hematite-dolomite veins (L6 stage) .................................................................89 5.26 ALTERATION AND MINERALIZATION PARAGENESIS – summary ..................91 vi  5.27 DISCUSSION: DIAGNOSTIC ALTERATION ZONING AT THE MT. MILLIGAN MAIN DEPOSIT .............................................................................................................93 5.27.1 Lateral zonation .................................................................................................94 5.27.2 Vertical zonation ..............................................................................................100 5.28 PART ONE CONCLUSIONS ..............................................................................105 5.28.1 Host rock control..............................................................................................105 5.28.2 Alteration and mineralization............................................................................105 CHAPTER 6: GEOCHEMICAL DISPERSION AT THE SILICA-SATURATED MT. MILLIGAN CU-AU ALKALIC PORPHYRY DEPOSIT ................................................108 6.1 INTRODUCTION ..................................................................................................108 6.1.1 Geochemical dispersion – fluid evolution...........................................................108 6.1.2 Epidote-pyrite ore vector and structural indicator...............................................108 6.2 ANALYTICAL TECHNIQUES ...............................................................................109 6.2.1 Strontium isotopes .............................................................................................109 6.2.2 Sulfur isotopes ...................................................................................................109 6.2.3 Carbon and oxygen isotopes .............................................................................110 6.2.4 Epidote major elements – EMPA .......................................................................110 6.2.5 Epidote trace elements – LA-ICP-MS ................................................................110 6.2.6 Pyrite trace elements – LA-ICP-MS ...................................................................111 6.2.7 Fluid inclusions ..................................................................................................111 6.3 STRONTIUM ISOTOPES .....................................................................................112 6.3.1 Theory ...............................................................................................................112 6.3.2 Sr composition of alteration minerals.................................................................113 6.3.3 Sr-isotope results...............................................................................................114 6.4 SULFUR ISOTOPES ............................................................................................116 6.4.1 Theory ...............................................................................................................116 6.4.2 Examples ...........................................................................................................118 6.4.3 S-isotope results ................................................................................................120 6.5 CARBONATE C- AND O-ISOTOPES...................................................................128 6.5.1 C-isotope systematics........................................................................................128 6.5.2 O-isotope systematics .......................................................................................129 6.5.3 C- and O-isotope results....................................................................................130 6.6 TRACE ELEMENTS IN EPIDOTE AND PYRITE...................................................132 6.6.1 Epidote chemistry ..............................................................................................133 6.6.2 Previous work with epidote in porphyry systems ...............................................134 6.6.3 Epidote results – major elements – EMPA.........................................................136 6.6.4 Epidote results – trace elements – LA-ICP-MS..................................................138 6.6.5 Pyrite results – trace elements – LA-ICP-MS.....................................................152 vii  6.7 GEOTHERMOMETRY..........................................................................................164 6.7.1 Chalcopyrite-pyrite S-isotope geothermometry..................................................164 6.7.2 Fluid inclusion analysis and geothermometry ....................................................167 6.8 DISCUSSION .......................................................................................................171 6.8.1 Sr-isotopes – magmatic vs. meteoric fluids........................................................172 6.8.2 S-isotopes..........................................................................................................173 6.8.3 C- and O-isotopes..............................................................................................174 6.8.4 Geothermometry and isotope synthesis ............................................................176 6.8.5 Interpretation of Cu-Au and Au-only ore zones..................................................178 6.8.6 Epidote trace elements – ore vectors ................................................................179 6.8.7 Pyrite trace elements – ore vectors ...................................................................180 6.8.8 Trace elements in combined epidote-pyrite alteration .......................................184 6.9 CONCLUSION......................................................................................................184 6.9.1 Recommendations for future work (and other conjectures) ...............................186 REFERENCES ............................................................................................................189 APPENDIX 1: MT. MILLIGAN MAIN DEPOSIT VOLCANIC STRATIGRAPHY AND ALTERATION ..............................................................................................................201 APPENDIX 2: SULFUR ISOTOPIC COMPOSITIONS EXPRESSED AS δ34S VALUES FOR SULFIDE FROM THE MT. MILLIGAN CU-AU ALKALIC PORPHYRY DEPOSIT..... .....................................................................................................................................205 APPENDIX 3: EMPA DATA FOR EPIDOTE AT THE MT. MILLIGAN CU-AU ALKALIC PORPHYRY DEPOSIT................................................................................................208  viii  LIST OF TABLES Table 1: Styles, and spatial and temporal distribution of alteration assemblages at the Mt. Milligan Cu-Au alkalic porphyry deposit ...................................................................53 Table 2: Vein stages of the Mt. Milligan Cu-Au alkalic porphyry deposit ......................54 Table 3: Average, median, and standard deviation of Cu and Au grade within alteration zones at relative depth...................................................................................................66 Table 4: Rb and Sr tracer concentration (ppm) in peripheral alteration minerals from the Mt. Milligan Cu-Au alkalic porphyry deposit ...........................................................115 Table 5: Carbon and oxygen isotope analysis results, and conversion to fluid values for the Mt. Milligan Main deposit .......................................................................................130 Table 6: Median composition (wt.% oxide) of epidote at the Mt. Milligan Main deposit (EMPA) ........................................................................................................................137 Table 7: Median concentration (ppm) of trace elements in epidote at the Mt. Milligan Main deposit with synopsis of interpreted lateral and vertical trends, and ore vector potential .......................................................................................................................141 Table 8: Median concentration (ppm) of trace elements in pyrite for the Mt. Milligan Main deposit with synopsis of interpreted lateral and vertical trends, and ore vector potential .......................................................................................................................160 Table 9: Characteristics of fluid inclusions .................................................................169  ix  LIST OF FIGURES Figure 1: Map of British Columbia showing the location of the accreted Quesnel and Stikine ocean arc terranes and major alkalic porphyry deposits. The Mt. Milligan deposit is located at the end of an east-southeast shift in the southern limb of the Hogem Batholith ..............................................................................................................3 Figure 2: Late Permian - Early Jurassic continental configuration illustrating the breakup of Pangea and the accretion of ocean arc superterrane to the western margin of ancestral North America, and the timing of alkalic porphyry deposits in British Columbia . .........................................................................................................................................4 Figure 3: Rb vs. (Y+Nb) plot for rocks from the Mt. Milligan Cu-Au alkalic porphyry showing granitoid fields for various tectonomagmatic environments ...............................4 Figure 4: Plan view showing interpreted geology (including the MBX Main deposit, Southern Star deposit, and Goldmark stock), major faults, four ore zone divisions within the MBX Main deposit (DWBX, WBX, MBX, and 66), orientation of the hinged crosssection, and the location of drill holes investigated ..........................................................5 Figure 5: Profile view along the hinged cross-section showing interpreted geology, major structures, four ore zone divisions of the MBX Main deposit (DWBX, WBX, MBX, and 66), and the location of drill holes investigated .........................................................6 Figure 6: Regional geology and tectonic setting of the southern Hogem Batholith (in north-central Quesnellia) and its relationship to the Mt. Milligan Cu-Au porphyry presented with simplified aeromagnetic topography ......................................................18 Figure 7: DDH core graveyard located 2 km west of the MBX stock in a valley between the N-S trending chain of low-lying mountains...............................................................20 Figure 8: View looking east across Witch Lake from helicopter flying southward above the Pinchi Fault valley....................................................................................................21 Figure 9: Geochemistry for Mt. Milligan host rocks and intrusives ...............................23 Figure 10: Examples of host rocks of the Witch Creek Succession ..............................25 Figure 11: Schematic illustration of a subaqueous, in situ, and resedimented hyaloclastite...................................................................................................................26 Figure 12: Volcanic-conglomerate, apparent- and/or pseudo-breccia ..........................27 Figure 13: Lower Trachyte (Tephriphonolite) ................................................................29  x  Figure 14: Upper Trachyte (Tephriphonolite) ................................................................30 Figure 15: Volcanic-conglomerate with juvenile clasts (hyaloclastite?) in upper DWBX zone...............................................................................................................................31 Figure 16: MBX stock. Drillcore sections showing the variety of composition, texture, alteration, and sulfide mineralization of monzonitic to monzodioritic rocks ....................33 Figure 17: Datamine image showing the MBX Main deposit in 3-D. Orientation is similar to that of the cross-section in Figure 2 ...............................................................34 Figure 18: Rainbow Dike (monzodiorite)......................................................................35 Figure 19: Brecciation associated with the MBX stock .................................................37 Figure 20: Late-mineral dikes .......................................................................................40 Figure 21: Copper grade, Au grade, and Cu-Au ratio plotted on hinged cross-section..... .......................................................................................................................................41 Figure 22: Mineral-specific sequence of alteration and sulfide mineralization compiled from detailed core logging, petrography, SWIR and XRD spectral data ........................51 Figure 23: Hinged cross-section showing lateral zoning of alteration assemblages .....52 Figure 24: Potassic alteration. Individual frames for compiled K-feldspar, biotite, and magnetite alteration shells, and their distribution across the hinged cross-section........56 Figure 25: Quartz veins ................................................................................................58 Figure 26: Calc-potassic alteration of basaltic-trachyandesite......................................60 Figure 27: Early to transitional-stage veins and replacement sulfide minerals .............63 Figure 28: Intermediate stage potassic alteration in the 66 zone..................................67 Figure 29: Sodic-calcic (outer calc-potassic) and propylitic alteration stages. Individual frames for compiled Na-feldspar, actinolite, and epidote alteration shells, and their distribution across the hinged cross-section ..................................................................71 Figure 30: Sodic-calcic (outer calc-potassic) alteration.................................................72 Figure 31: Mineralization and replacement textures in sodic-calcic and inner-propylitic alteration shells..............................................................................................................74 Figure 32: Inner-propylitic alteration .............................................................................76 xi  Figure 33: Carbonate-phyllic and chlorite alteration. Individual frames for compiled dolomite/ankerite, sericite, and chlorite alteration shells, and their distribution across the hinged cross-section......................................................................................................78 Figure 34: Carbonate-phyllic alteration .........................................................................80 Figure 35: Chlorite alteration ........................................................................................82 Figure 36: Chalcopyrite and pyrite modes estimated in the field, shown on the hinged cross-section .................................................................................................................83 Figure 37: Mineralization in carbonate-phyllic zone ......................................................84 Figure 38: Outer-propylitic alteration.............................................................................87 Figure 39: Other intermediate to late-stage veins .........................................................90 Figure 40: Copper-Au grade juxtaposed with alteration shells along the hinged crosssection ...........................................................................................................................92 Figure 41: Transition from E4- veins to P2-stage veins through the intermediate T3 stage as an analogue for the transition from the calc-potassic to outer-propylitic assemblage through the intermediate sodic-calcic and inner-propylitic assemblages ...96 Figure 42: 66 zone branch of the hinged cross section showing Au grade. Inferred faults extending from the upper and lower margins of the Upper Trachyte towards the Great Eastern Fault are shown. Gold grade is elevated along the lower fault at 98-107 m depth. Locations and results of XRD analyses of carbonate-phyllic altered rocks are shown ..........................................................................................................................104 Figure 43: Schematic cross-section of the Mt. Milligan Main deposit with a scenario for the development of alteration zonation and related ore-body in the Early Jurassic. Moderate tilting and normal faulting in the Tertiary results in the present day structural geometry......................................................................................................................106 Figure 44: Systematic variation of the 87Sr/86Sr ratio of the oceans during the Phanerozoic as indicated by chemostratigraphy. The Sr-ratio for 182.5 Ma (0.70747), the measured age of the Rainbow Dike, is indicated ...................................................114 Figure 45: Profile view of MBX zone showing Sr isotope sample locations for Nafeldspar, actinolite, and epidote ...................................................................................115  xii  Figure 46: Initial 87Sr/86Sr ratios at 182.5 Ma for samples indicated in Fig. 45 showing dominantly magmatic signatures. The best-fit lines for measured 87Rb/86Sr and 87Sr/86Sr values of individual alteration minerals show deviations from 182.5 Ma isochrons suggesting inter-mineral 87Sr mobility. The regression line for all analyses is within error of 182.5 Ma model age for the Rainbow Dike..............................................................116 Figure 47: Ohmoto (1972) pH versus log ƒO2 diagram at 250ºC modified to show the path of a magmatic-hydrothermal fluid in an alkalic porphyry system..........................118 Figure 48: Plot of δ34Ssulfide versus lateral distance from the center of the vertically reoriented MBX stock. The plotted δ34Ssulfide values form an upward-widening distribution suggesting reduction of oxidized magmatic fluids down the thermal gradient as fluids moved away from the MBX stock ..................................................................121 Figure 49: S-isotope δ34Ssulfide values vs. lateral distance from center of MBX stock, grouped according to associated alteration. Transition from negative to positive values represents cooling and reduction of the mineralizing fluid to background values, providing a general analogue for paragenetic sequence .............................................122 Figure 50: S-isotope sample locations within the hinged cross-section and associated δ34S values. The dominant alteration for individual sample points is indicated...........125 Figure 51: Contoured δ34Ssulfide values in profile across hinged cross-section showing early-intermediate and late stage fluid domains and pathways. The integrated ‘total’ profile illustrates the likelihood of convective recycling of fluids in the MBX zone, and overprinting of the early-intermediate by late stage assemblage in the lower DWBX and MBX zones ..................................................................................................................126 Figure 52: A) δ13C - δ18O plot that compares the fluid values calculated from Mt. Milligan carbonates with isotopic composition of fluids from a variety of source reservoirs as compiled by Rollinson (1993). B) Hinged cross-section showing location of carbonate isotope samples, and predicted path of magmatic fluid ..........................131 Figure 53: Location of epidote-pyrite samples in relation to the compiled Na-feldspar and epidote alteration shells ........................................................................................133 Figure 54: A) Coordination polyhedra of epidote showing arrangement of A- and Msites. B) Notation of the M positions in the octahedral chains.....................................135 Figure 55: EMPA data showing median pistachite ratio [Fe3+/(Fe3+/Al)] versus lateral distance from the center of the MBX stock for sampled epidote, with 1-sigma error bars representing distribution of values per sample. Trends for median values are shown...... .....................................................................................................................................137 Figure 56: Examples of LA-ICP-MS time-resolved spectra for Mt. Milligan epidote showing trace element counts-per-second ..................................................................139 xiii  Figure 57: Trace elements in epidote measured by LA-ICP-MS. Ten analyses per sample are represented by median values with 1-sigma error bars to represent the distribution of values per sample. Trends for median values are shown.....................142 Figure 58: Representative REE (and Y) in epidote measured by LA-ICP-MS. Ten analyses per sample are represented by median values with 1-sigma error bars to represent the distribution of values per sample. Trends for median values are shown...... .....................................................................................................................................149 Figure 59: REE spidergrams for secondary epidote across the Mt. Milligan Cu-Au alkalic porphyry deposit. Fractionation of LREE increases toward the MBX stock resulting in steeper profiles, particularly in the DWBX zone ........................................150 Figure 60: LREE ratios showing stockward fractionation of LREE towards higher La concentration relative to other LREE, and graph of Eu anomaly magnitude versus lateral distance ............................................................................................................151 Figure 61: Median values for LA-ICP-MS analyses of As, Cu, Th and 3La/(Ce+Pr+Nd) in epidote plotted according to sample location on the hinged cross-section to further illustrate trends defined in lateral distance versus compositional space. Amount of throw in meters between chemically similar epidote on either side of the Rainbow Fault is shown ..........................................................................................................................153 Figure 62: Trace elements in pyrite that increase in concentration with distance laterally from the MBX stock. .......................................................................................156 Figure 63: Trace elements in pyrite that show behaviour other than an increase in median concentration with lateral distance from the MBX stock..................................163 Figure 64: Median values for LA-ICP-MS analyses of Mn, Co, and As in pyrite plotted according to sample location on the hinged cross-section to further illustrate trends defined in lateral distance versus compositional space ...............................................165 Figure 65: Equilibrium temperature (pyrite-chalcopyrite geothermometer) vs. lateral distance from the center of the vertically re-oriented MBX stock, with associated alteration......................................................................................................................166 Figure 66: Hinged cross-section of the Mt. Milligan Main deposit showing fluid inclusion data (bottom) including homogenization temperatures not pressure corrected (minimum trapping temperature) and pressure corrected temperatures at 1200 bars (~3.5 km depth). Sulfur isotope pyrite-chalcopyrite geothermometer results are included (top) for comparison ..................................................................................................................170  xiv  Figure 67: Synthesis of results from Sr-, S-, C-, and O- isotope analysis, with pressurecorrected fluid inclusion geothermometry, and pyrite-chalcopyrite S-isotope geothermometry. Results are plotted in cross-section against areas of K-feldspar and Na-feldspar alteration to better constrain the hydrothermal plumbing, and are contrasted with the resultant Cu and Au ore shells .......................................................................177 Figure 68: Graph showing the location of highest concentration of the specified trace element in epidote for the Mt. Milligan Main deposit, as interpreted from median values of LA-ICP-MS analyses. The best intra-system ore vectors are highlighted. Depth is indicated with associated ore-zone (DWBX, WBX, MBX, and 66) ...............................180 Figure 69: Examples of LA-ICP-MS time-resolved spectra for pyrite showing trace element counts-per-second. A) Internal variations in pyrite of LILE-REE-HFSE incompatible elements. B) Ag-Au telluride with Ba spike at pyrite grain edge .............183  xv  ACKNOWLEDGEMENTS This work forms part of a larger MDRU-CODES research project investigating shallowand deep-level alkalic mineral deposits, in collaboration with the GSC. Funding and field support is provided by Amarc Resources Ltd., Anglogold-Ashanti, Barrick Gold Corp., Imperial Metals Corp., Lysander Minerals Corp., Newcrest MIning Ltd., Newmont Mining Corp., Novagold Resources Inc., Teck Cominco Ltd., NSERC-CRD and Geoscience BC.  Logistical and field support from Terrane Metals Corp., Eastfield  Resources Ltd. and Lihir Gold Mine is also acknowledged. The former Placer Dome Inc provided funding and field support for the Mt. Milligan project in 2005. Additional funding for the Mt. Milligan project was supplied through SEG and GSA student research grants, and the Colin D. Spence Memorial Scholarship in Geology through UBC. I would like to thank Dr. Richard Tosdal for guidance and supervision throughout this project. Additional thanks goes to my committee members Dr. Greg Dipple, Dr. James Mortensen, and Dr. Claire Chamberlain for additional guidance. Additional thanks to Dr. Ken Hickey, for graphing assistance, Dr. Mati Raudsepp for direction in the use of SEM and microprobe, and Dr. Dominique Weis for assistance with Sr isotope analyses. Special thanks to Anna Fonseca, Darren O’Brien, and Dr. Kirstie Simpson for support in field work; Charles Tarnocai for instruction on ASD spectra interpretation and software use; Janet Gabides (Department of Earth and Ocean Sciences, UBC) for C- and Oisotope analyses; the Pacific Center for Isotope and Geochemical Research (PCIGR) for Sr analyses; Arne Toma (MDRU resource centre coordinator) for office assistance, Anthony Harris (Centre of Excellence in Ore Deposits - CODES) for fluid inclusion experiments and helpful discussion, Sarah Gilbert (CODES) for instruction on use of LA-ICP-MS, and Dr. David Cooke (CODES) for additional guidance and support.  xvi  DEDICATION This work is dedicated in loving memory of Harold and Patrick McDonagh, my grandfather and uncle. It is also dedicated to my wife, Aidyl, for her invaluable love and support. Now, here we are standing at the ocean side…  xvii  CHAPTER 1: DIAGNOSTIC ALTERATION-MINERALIZATION ZONING AT THE MT. MILLIGAN SILICA-SATURATED CU-AU ALKALIC PORPHYRY DEPOSIT 1.1 RATIONALE FOR STUDY In magmatic-hydrothermal systems, metasomatic alteration occurs via interaction of existing mineral assemblages with circulating fluids of magmatic and/or meteoric origin. Shifting physio-chemical conditions within the host rock cause reconfiguration of components into more stable phases, bringing about changes in mineralogy, texture, and composition analogous to metamorphic facies (McMillan and Panteleyev, 1988). Since the resultant alteration assemblages depend on the intensive parameters of the altering fluid, including temperature (ranging from ~600-200ºC in Cu-porphyry systems), pressure, bulk composition, ion speciation, redox state, and pH, the secondary minerals produced can be viewed as diagnostic of the evolving fluid with respect to the host protolith. Moreover, conditions and processes of ore formation can be deduced from the relationship of metals to alteration gangue, and their geochemical signature. For calc-alkaline Cu-porphyry deposits, the world’s largest source of copper, the genetic association of sulfide mineralization and metasomatic alteration has been well documented (Titley and Beane, 1981). This was significant because it implies that alteration halos, typically extending into a much larger volume of rock (>10km3) than ore zones (<4km3), can be used as thermodynamic- and geochemical vectors to ore at the center of these hydrothermal systems (Seedorff, 2005). It is therefore paramount in the porphyry environment that the paragenetic sequence of alteration from high- to lowtemperature mineral assemblages be fully documented.  Because alkaline systems  deviate subtly but significantly from calc-alkaline systems, the sequencing of alteration in this increasingly economical deposit type warrants further attention than has been previously given. The Mt. Milligan Main deposit provides an important case study in sulfide- and alteration mineral zonation for alkalic porphyry systems in B.C. and the greater Circum-Pacific. 1  The site is of particular interest because it is interpreted to be moderately tilted ~30-50º (based on the geometry of the intrusive stock, and the dip of layering to the horizontal axis of drillcore in fine-grained trachytic units). This geometry makes the magmatichydrothermal system amenable to the study of vertical and lateral changes over a range of paleodepths. Moreover, an important fault (Rainbow Fault) separates a Cu-Au-rich core zone from an Au-rich, Cu-poor zone inferred to be the shallower Au-enriched segment of the deposit, thereby increasing the total vertical exposure provided through drilling. The Mt. Milligan site is located 155 km northwest of Prince George, B.C., and is the youngest dated (183-182 Ma) of the seven major mineralized alkalic porphyry systems in British Columbia (Afton/Ajax, Copper Mt./Ingerbelle, Galore Creek, Lorraine, Mt. Polley, and possibly Red Chris).  The B.C. alkalic deposits formed in two pulses  between 210 and 180 million years ago within the conjoined Quesnellia-Stikinia intraoceanic island arc superterrane (a ~1300 km north-westerly trending belt of rocks comprising central B.C.). At that time, Atlantic rifting and spreading was underway causing ancestral North America to be displaced westward toward the offshore terrane complex (Mathews and Monger, 2005). Convergence and collision of the island arc with the western continental margin formed central B.C. and the Cordillera.  The  tectonic complexities therein resulted in increased (and petrologically unique) plutonic and volcanic activity of shoshonitic affinity (Fig. 1-3). The Mt. Milligan Cu-Au alkalic porphyry deposit is located in central Quesnellia, at the terminus of a ~45 km east-southeast structural trend extending from the southern edge of the Hogem Batholith along Chuchi Lake.  The deposit (including the Main and  Southern Star monzonitic stocks, and hydrothermally affected host rock) is hosted in mildly shoshonitic volcanic rocks of the Takla Group (Fig. 4). It has a measured and indicated resource of 417.1 million tonnes at 0.41 g/t Au and 0.21% Cu (as reported by Terrane Metals Corp., Aug. 21st, 2007) containing 5.5 million Oz gold, and 1.9 billion Lb copper. Compared to other Cu-porphyry deposits in B.C., Mt. Milligan is Au-enriched.  2  Figure 1: Map of British Columbia showing the location of the accreted Quesnel and Stikine ocean arc terranes and major alkalic Cu-Au porphyry deposits. The Mt. Milligan deposit is located at the end of an east-southeast shift in the southern limb of the Hogem Batholith. Image after M. Jackson (2008).  1.2 STUDY METHODOLOGY The present M.Sc. study incorporates two field seasons of detailed core logging focused on defining alteration- and metal-zonation (and its relationship to the host volcanic stratigraphy) in a silica-saturated alkalic Cu-Au porphyry system. Drillcore from 15 diamond-drill holes distributed within 50 m of a hinged cross-section were logged in detail and sampled (over 550 rock specimens taken; Fig. 5). Fourteen of the 15 drill  3  holes were drilled vertically but represent combined vertical and lateral transitions with increasing depth, assuming a moderate eastward tilt of the deposit. The hinged crosssection spans a central monzonitic stock (MBX stock) and andesitic rocks west and southeast of the stock, where ore is best developed.  Figure 2: Late Permian - Early Jurassic continental configuration illustrating the break-up of Pangea and the accretion of ocean arc superterrane to the western margin of ancestral North America, and the timing of Cu-Au alkalic porphyry deposits in British Columbia (image after Matthews and Monger, 2005).  Figure 3: Rb vs. (Y+Nb) plot for rocks from the Mt. Milligan Cu-Au alkalic porphyry showing granitoid fields for various tectonomagmatic environments after Pearce et al. (1984).  4  Part 1 (Ch. 1-5) and Part 2 (Ch. 6) of the current study have been written as stand alone documents for publication. In the first part, alteration and metal zonation is described with respect to lithology and volcanic stratigraphy. Data is compiled from core logging,  Figure 4: Plan view showing interpreted geology (including the MBX Main deposit, Southern Star deposit, and Goldmark stock), major faults, four ore zone divisions within the MBX Main deposit (DWBX, WBX, MBX, and 66), orientation of the hinged cross-section, and the location of drill holes investigated. Units not listed are listed in Fig. 5. Original lithology provided by Placer Dome Inc.  5  6  Figure 5: Previous page.  Profile view along the hinged cross-section showing interpreted  geology, major structures, four ore zone divisions of the MBX Main deposit (DWBX, WBX, MBX, and 66), and the location of drill holes investigated. Original lithology provided by Placer Dome Inc.  petrography, spectroscopy, XRD, and SEM analysis (plus additional support from previous years’ logging reports provided by Placer Dome Inc.) with the principal focus on alteration-mineralization changes and effect on ore grade. Individual alteration zones are described in macro- and microscopic detail. Textural changes are highlighted and replacement reactions postulated. The pattern of alteration and metal zonation is used to interpret changes in the hydrothermal fluid, as well as define deposit-scale architecture with regard to post-mineral tilting and faulting. Based on this structural geometry, crosscutting vein stages and alteration assemblages, a paragenetic sequence is developed (with associated sulfide minerals and ore grade).  This is  described in terms of a combined lateral and vertical evolution. Finally, the overall development - combined alteration and structure - of the Mt. Milligan Cu-Au alkalic porphyry is reviewed. Part two (Ch. 6) of the study defines lateral and vertical geochemical dispersion halos within the Mt Milligan Main deposit. Systematic geochemical dispersion in porphyry systems is particularly important for brownfields exploration (exploration of areas proximal to mines and known mineral deposits), as it can provide vectors to additional nearby resources. Sulfide minerals from the compiled Mt. Milligan rock specimens were analyzed for sulfur isotope ratios, which provide a proxy for oxidation and temperature evolution of the magmatic-hydrothermal fluid (Wilson et al., 2006). The results were correlated with alteration and used to support the paragenetic sequence of Part one. When plotted spatially, the S-isotopes suggest deposit-scale outflux (thermally retrograding) and influx  (thermally  prograding)  fluid  pathways.  Additionally,  assuming  isotopic  equilibration, the pyrite-chalcopyrite isotope geothermometer of Kajiwara and Krause (1971) was used to estimate temperature of ore formation. This is contrasted with 7  temperatures derived from fluid inclusions in intrusion-related quartz veins and peripheral secondary epidote to develop a general thermal evolution for the system. Carbonates associated with late-stage phyllic overprinting at the shallowest depths and along intrusive contacts central to the system were analyzed for C and O isotopes to verify the source from which these elements were derived (magmatic versus meteoric). Likewise, Sr isotope analysis of secondary minerals Na-feldspar, actinolite, and epidote points to the nature of source fluids, and the relative proportions of fluid mixing in the peripheral alteration assemblages. Primitive fluid signatures occur in both cases, although mixing with external fluids is also indicated. Additionally, the deposit-wide mineral association epidote-pyrite was analyzed for traceelement composition to determine if spatial changes in concentration of any particular element, set or relation of elements provide a systematic ore vector. Epidote is used for its diagnostic solid solution behaviour between pistachite and clinozoisite, and for its structural ability to incorporate fluid-borne incompatible elements, thereby representing a proxy for detailed compositional changes in the magmatic-hydrothermal fluid from the system core to periphery. Additionally, spatial trends in trace element composition of epidote-pyrite are used to verify post-mineral normal faulting with respect to the Rainbow Fault, and to estimate the amount of throw between the MBX zone and the downdropped 66 zone.  8  CHAPTER 2: CU-PORPHYRY SYSTEMS – CALC-ALKALINE VS. ALKALINE 2.1 CU-PORPHYRY SYSTEMS As intermediate-to-felsic magmas crystallize inward from the margins of upper crustal magma chambers, the remaining differentiate is an H2O-saturated felsic or alkalic melt/fluid which collects by thermogravitational- and Soret diffusion at the apex of cupolas in the root zone of porphyritic stocks, pipes or dikes (Wilson, 1995). Porphyritic intrusions are generated from these wet magmas (~4.0 wt% H2O), which exsolve H2O at a sufficiently low pressure causing the magma to become unfluxed and rapidly quenched to a sugary groundmass (Seedorff et al., 2005).  Magmatic fluids rise  buoyantly along the margins of a porphyritic intrusion until becoming trapped by an impermeable cap of brittle-ductile contact-altered volcanic host rock (Fournier, 1999). Eventually,  this  fluid-saturated  carapace  deforms  to  accommodate  internal  overpressures that may be caused by fluid accumulation, resurgent boiling, and/or further intrusive activity. The surrounding host rock fails under extreme tensile stress and the initial pulse of supercritical fluid-vapour either creates or exploits permeable structures (such as breccia bodies) thus preparing the ground for subsequent fluid pulses (Burnham, 1979). Metasomatic alteration spreads by ionic exchange, volatileaddition, and hydrolysis reactions linked to propagating hydraulic fractures in the host rock that evolve to high density from early-stage microfractures at depth to later openspace fissures higher in the system (Lowell and Guilbert, 1970; Engvik et al., 2005). Common fluid pathways include pre-existing weaknesses in the host rock such as volcaniclastic matrix, shear banding, bedding, perlitic cracks, and grain boundaries of phenocrysts. Additionally, lateral brecciation and/or fluid release can occur within less competent volcanostratigraphic horizons, particularly along the deeper flanks of the carapace where the energy of resurgent boiling is less. A breach can result in normal dike intrusion along a plane of weakness, driven by internal overpressure of the magma (Burnham, 1979).  9  In the classic Lowell and Guilbert static model (1970) for calc-alkaline Cu-porphyry systems, the intrusive body (typically a composite intrusion) is surrounded by a sequence of hypogene alteration shells reflecting changes in the K+/H+ ratio and decreasing temperature of the altering fluid. An inner ‘potassic’ shell where all protolith constituents are chemically unstable is dominated by the composition of the magmatic fluid (an oxidized hypersaline brine) that has concentrated incompatible and volatile elements of the intrusive magma including alkalis, halogens, sulfur, silica, and metals such as Cu, Au, Mo, Ag, Pb, and Zn (depending on the degree of magmatic fractionation, chemistry of the intrusion, and intensive parameters). The composition of the potassic assemblage essentially reflects the last differentiate of the fractionating magma (Gustafson and Hunt, 1975). In more mafic host rocks, such as andesite, there is higher Mg2+ and Fe2+ activity, and the potassic fluid will alter the surrounding protolith to hydrothermal biotite, whereas in more felsic rocks, such as monzonite or granite, the dominant alteration mineral will be K-feldspar. Potassic alteration occurs at lithostatic pressure, but may be punctuated by brief episodes of elevated fluid pressure resulting in brecciation and explosive escape of volatiles (McMillan and Panteleyev, 1988). In calc-alkaline systems, the potassic zone develops upward into ‘phyllic’ (quartzsericite-pyrite), and ‘argillic’ (quartz, kaolinite, chlorite) zones as hydrolysis increases with decreasing K+/H+ ratio and temperature of the magmatic-hydrothermal fluid (Seedorff et al., 2005). Mixing of inwardly convecting, meteoric-derived fluids and nonacidic magmatic fluids on the periphery of the hydrothermal system generates a zone of low  temperature  Propylitic  ‘propylitic’  alteration  weakens  alteration laterally  (epidote-Na-feldspar-calcite-chlorite-pyrite). outward  to  unaltered,  or  regionally  metamorphosed, country rock. The related ore-shell surrounds the central intrusion as a vertical pipe-like body extending from potassic to phyllic alteration zones. Sulfide occurs in a sequence of microveinlets to veins, and as disseminations (Guilbert and Park, 1986).  10  2.2 ALKALINE VS. CALC-ALKALINE The above model is idealized, and thereupon does not capture the vagaries of individual calc-alkaline Cu-porphyry deposits.  However, it becomes even more  abstracted with regard to alkalic Cu-Au porphyry deposits which differ from the calcalkaline type in composition of intrusives, associated alteration assemblages, and metal ratios (Stanley, 1992). Barr et al. (1976) and Ney and Hollister (1976) distinguished the alkaline suite of Cu-porphyry deposits in the Canadian Cordillera from the calc-alkaline variety previously described in the work of Lowell and Guilbert (1970). In British Columbia, the alkalic subgroup is restricted to the Early Mesozoic Quesnel and Stikine terranes within the Intermontane morphogeologic belt (Lang et al., 1995; Fig. 1). The spatial-temporal trend of alkalic porphyritic intrusives continues into the Tertiary period along the eastern Rocky Mountain front from Canada to Mexico. This trend includes Au-enriched deposits such as Cripple Creek, Colorado and Zortman-Landusky, Montana (Douglass and Campbell, 1994). The alkaline subgroup in British Columbia has been subdivided into silica-saturated, and silica-undersaturated types (Lang et al., 1998). Equigranular-to-porphyritic alkalic igneous rocks characterize the silica-saturated type (including the Mt. Milligan Cu-Au alkalic porphyry), ranging from diorite to monzonite compositions that lack modal quartz, but  have  weak  quartz-  or  feldspathoid-normative  compositions.  Alternatively,  megaporphyritic alkalic igneous rocks characterize the silica-undersaturated type, dominated by syenite and pyroxenite that have modal and normative feldspathoid compositions (Lang et al., 1998). The above classifications are based on the geochemical distinction of alkaline versus calc-alkaline suites of volcanic rocks, as presented by Barager and Irvine (1971) and MacDonald (1968). 2.3 ALKALINE VS. CALC-ALKALINE – IGNEOUS ROCKS The primary distinguishing feature between alkaline and calc-alkaline porphyry deposits is the composition of the source intrusion, and the implications this has on 11  mineralization style. Calc-alkaline systems are characterized by intrusions containing modal quartz (quartz diorite, granodiorite, and quartz monzonite) whereas alkaline systems range from diorite-to-syenite silica-deficient compositions. In calc-alkaline systems, hydraulic fracturing by silicic metalliferous fluids (and concomitant pressure quench) results in the trapping of precipitated Cu-sulfide ± Mo in quartz cement. Accordingly, ore is associated with a multi-generation quartz vein stockwork in both intrusive and proximal volcanic rocks. Alternately, the British Columbia alkaline systems exsolved silica-poor fluids that failed to generate quartz cement, and therefore lack the mineralized quartz stockwork so characteristic of calc-alkaline and Australian alkaline systems such as Cadia-Ridgway (Wilson et al., 2003).  Instead, sulfide veins and  veinlets with accessory quartz occur within and around the intrusions accompanied by replacement mineralization of Fe2+ bearing minerals. However, a low density of quartz veins may occur in silica-saturated alkaline systems such as the Mt. Milligan Cu-Au porphyry. 2.4 ALKALINE VS. CALC-ALKALINE – ALTERATION Another distinguishing characteristic of alkaline systems is the absence of strong hydrolytic alteration. Typically, a magnetite-rich potassic zone passes into propylitic alteration without any intermediary phyllic or argillic stages, suggesting elevated pH of magmatic-hydrothermal fluids inhibiting hydrolytic assemblages (Stanley, 1992), or acid buffering by abundant primary and secondary K-feldspar.  Nevertheless, an alkalic  lithocap of limited ‘intermediate argillic’ (sericite-chlorite-kaolinite-pyrite-carbonate) or phyllic alteration may develop (Cooke et al., 1998; Holliday, and Cooke, 2007). Other alteration styles associated with alkaline systems include ‘sodic’, ‘sodic-calcic’, and ‘calc-potassic’ assemblages, plus the near absence of quartz as an alteration mineral but the presence of abundant carbonate (Lang et al., 1998).  Sodic-calcic  alteration is best developed in silica-saturated alkalic porphyries, whereas the calcpotassic assemblage is common to the silica-undersaturated type. Both show variable spatial and temporal relationships with potassic alteration.  12  Distinctive sodic-calcic alteration occurs at the deep center of systems in an inverted cup-shaped volume beneath and peripheral to potassic alteration (Seedorff et al., 2005). Oligoclase replacement of K-feldspar, actinolite replacement of mafics, and redistribution of Cu, are characteristic (Dilles and Einaudi, 1992). Throughout the Great Basin of the Western U.S., Na-rich hydrothermal alteration of Permian- to Jurassic-aged igneous and volcanic rocks is thought to indicate a combination of igneous processes and paleogeogaphy (low-lying arc or marginal basin settings), involving convection of isotopically-heavy saline fluids of marine-, formation- and/or meteoric origin (Battles and Barton, 1995; Dilles and Einaudi, 1992). Alternatively, for the British Columbia alkalic porphyry province, Lang et al. (1998) suggest that sodic alteration is related to intrusionspecific magmatic fluids, and possibly to early fluid saturation of the source intrusion resulting in an elevated fluid Na/K ratio. Calc-potassic alteration in silica-undersaturated alkalic porphyry deposits such as Galore Creek and Mt. Polley (Lang et al, 1998), and Lorraine B.C. (Nixon and Peatfield, 2003) is comprised of high temperature minerals (~500º) including garnet, diopside, Kfeldspar, anhydrite, and chalcopyrite-bornite. Fluid inclusion data from Galore Creek suggest a magmatic-hydrothermal origin for the calc-potassic fluid. Based on experimental data of Holland (1972), Lang et al. (1998) hypothesize a highly saline aqueous fluid with strong partitioning of Ca2+ at a late stage of syenite crystallization. Alkalic Cu-Au porphyry deposits also differ from calc-alkalic varieties of western USA and South America in having a spatially smaller alteration footprint, typically several hundred meters in radius versus kilometer scale, which makes exploration of this deposit type additionally challenging. The alkalic variety also tends to occur in clusters of diminutive intrusions (pipes, stocks and dikes) commonly with variable character of mineralization within an individual cluster (Holliday & Cooke, 2007). 2.5 ALKALINE VS. CALC-ALKALINE – METAL RATIOS The calc-alkaline variety of Cu-porphyry deposits is associated with the mineral assemblage Cu-Mo-Ag.  Orebodies typically contain between 0.4 and 1% Cu with 13  lesser amounts of Mo, and Ag (Seedorff, 2005). Alternately, the alkaline variety is associated with Cu-Au-Ag, where orebodies range from 0.2 and 1% Cu, and as high as 2.5 g/t Au as indicated for the Ridgeway Au-Cu deposit, New South Wales (Wilson et. al, 2003). Apart from the differences outlined above, the broad similarities between alkalic and calc-alkalic porphyry deposits suggest the alkalic type represents the Auenriched endmember of a continuum of porphyry deposits, and is consistent with gradations in alkalinity from calc-alkalic Cu-Mo deposits to high-K calc-alkalic Cu-Mo-Au deposits, to silica-saturated alkalic Cu-Au-Ag deposits, to silica-undersaturated, highly alkaline systems (Lang et al., 1998). However, a thermal divide in the nepheline-quartz binary system separates silica-saturated and silica-undersaturated phases to 1118ºC at Na-feldspar, implying a break in the continuum between these alkalic types at lower temperatures. Gammons and Williams-Jones (1997) report that Au solubility as chlorocomplexes at Cu-porphyry temperatures (~500ºC) is highest for fluids that are oxidized, acidic, highly saline, and K-rich. The differences in metals between the deposit types may reflect compositional variation in the fluids derived from magmas of different composition, degree of magmatic differentiation, and/or timing of fluid saturation. It may be that the higher concentration of alkali and alkali-earth elements, which are fluxing agents that lower magma viscosity (Chang et al., 1987), stabilizes the magma to lower temperatures, allowing for prolonged differentiation and concentration of incompatible elements. Furthermore, Au concentration in magnetite is 10 to 100 times greater than silicates. In magmas where magnetite remains above the solidus until a late stage, loss of Au will be minimized, and the potential to form an Au ore body via interaction with hydrothermal fluids increased (Stanley, 1992). The abundance of hyrothermal magnetite associated with potassic alteration and breccia cement in alkaline systems supports late stage magnetite precipitation. Additionally, high K concentrations resulting in abundant K-feldspar crystallization may favour the precipitation of biotite over magnetite as the Fe-bearing phase, according to the equilibrium equation provided by Stanley (1992). KAlSi3O8 + H2O + Fe3O4 ↔ KFe3AlSi3O10(OH)2 + 0.5O2 14  Not only would biotite crystallization suppress Au sequestration in cumulate magnetite, but also, according to Burnham (1979), would buffer the fluid composition to a lower KCl/NaCl ratio, resulting in magmatic fluids capable of sodic alteration.  15  CHAPTER 3: GEOLOGICAL SETTING 3.1 REGIONAL SETTING The alkalic intrusions of the Paleozoic-Mesozoic Stikinia and Quesnellia terranes comprising the intermontane region of British Columbia are tectonic and geochemical anomalies. These intrusions are associated with regionally extensive successions of calc-alkaline to mildly alkaline rocks of shoshonitic affinity  (K-enriched mafic to  intermediate composition) that were produced by complex subduction processes during amalgamation of the late oceanic arc superterrane to ancestral North America (Nelson and Bellefontaine, 1996; Muller and Groves, 1993; Fig. 1-3). Shoshonites are rare in island arcs and are usually associated with melt derived from an upper mantle that has been metasomatically enriched in incompatible and volatile elements (Foley et al., 1987). They are thought to involve deviations from steady-state subduction processes, such as reversal in subduction polarity followed by extensional tectonics (Kennedy et al., 1990; Richards, 1995). 3.2 REGIONAL STRATIGRAPHY The middle to late Paleozoic basements of Quesnellia and Stikinia are less well understood than overlying Mesozoic sequences (Monger et al., 1991; Mortensen et al., 1995). North-central Quesnellia in the late Paleozoic consisted of the Mississippian volcano-sedimentary Lay Range arc and the Pennsylvanian-Permian Slide Mountain marginal basin (a likely back-arc basin between Quesnellia and ancestral North America). Cessation of calc-alkaline volcanism with accompanying uplift and mild deformation has been attributed to the late Permian, while blueschist melange of the Cache Creek terrane indicates subduction beneath Quesnellia (Mihalynuk et al.,1994). Lower Triassic sedimentary rocks of the Takla Group overly this basement in a regional decollement (Mihalynuk et al., 1994). There is conspicuous stratigraphic similarity between the volcanic and sedimentary units of Quesnellia and Stikinia during the Mesozoic (Miller, 1987). Quesnellia at this time had a two-phase development. The first phase is composed of Upper Triassic 16  Nicola and Takla Groups (Carnian to Norian, ~227-210 Ma) in the southwest and northeast, respectively. These groups consist of basal sedimentary rocks overlain by volcanic and volcaniclastic successions dominated by augite-phyric basalts and andesites of calc-alkaline to shoshonitic affinity. Coeval intrusions are also present. These high-K and mildly shoshonitic rocks range over a 1000 km strike length in North Quesnellia (Mortimer, 1987; Nelson et al., 1992). The Witch Lake volcanic succession of the Takla Group hosts the Mt. Milligan Cu-Au alkalic porphyry. Late Triassic blueschists along the Pinchi fault support eastward subduction of Cache Creek (Mihalynuk, 1994) until the Late Triassic-Early Jurassic hiatus. The hiatus represents a brief period of uplift combined with cessation of volcanism and increase in sedimentation and plutonism. The Triassic-Jurassic transition (210-200 Ma) temporally coincides with the main phase of alkalic porphyry deposits and related batholithic reservoirs in both Quesnellia and Stikinia (Lang et al., 1995). The Hogem batholith intrusive suite (206±8 to 171±6 Ma) of central Quesnellia is interpreted as the intrusive equivalent of the Takla group (Garnett, 1978). The second stage of development in the Mesozoic consists of Early Jurassic carbonate and clastic sedimentary sequences unconformably overlying the Triassic volcanics. However, in the Mt. Milligan area, volcanism continued after the Late Triassic-Early Jurassic hiatus resulting in the paraconformably overlying Chuchi Lake and Twin Creek successions (Pleisbachian-Toarcian, ~196-180 Ma) which exhibit greater compositional heterogeneity than the Upper Triassic sequence and are dominantly comprised of plagioclase-  augite-phyric  subalkaline  to  shoshonitic  lithologies  (Nelson  and  Bellefontaine, 1996). The Mt. Milligan Cu-Au porphyry and related plutons are coeval with this second stage. U-Pb zircon ages from the Mt. Milligan deposit include 189 ± 3.3 Ma from the Heidi Lake stock, 183 ± 3 Ma from the North Slope stock, and 182.5 ± 4.3 Ma from the Southern Star stock (Mortensen et al., 1995). Additionally, a U-Pb rutile age of 182.5 ± 4 Ma has been obtained from the Rainbow Dike apophysis of the MBX stock (Nelson and Bellefontaine, 1996). Most recently, titanite from the MBX stock (DDH 90-616) has been dated at 186.9 ± 0.5 Ma (Richard Friedman, 2008, written 17  18  Figure 6: Previous Page. Regional geology and tectonic setting of the southern Hogem Batholith (in north-central Quesnellia) and its relationship to the Mt. Milligan deposit presented with simplified aeromagnetic topography.  The ESE deflection of the southern Hogem batholith  continues toward the deposit as a magnetic anomaly suggesting a buried extension of the batholith as the root of the Mt. Milligan Cu-Au porphyry system.  Image after Nelson and  Bellefontaine (1998), and Fonseca (2005).  communication). Subduction had ceased by ~186-181 Ma with the accretion of Quesnellia to ancestral North America (Murphy et al., 1995; Nixon, 1993), so the Mt. Milligan porphyritic stocks represent the final plutonic activity of the local magmatic system. 3.3 REGIONAL STRUCTURE The linear northwesterly trend of the Hogem batholith parallel to the Pinchi fault system separating the Quesnellia and Cache Creek terranes attests to fundamental structural control over its emplacement (Nelson and Bellefontaine, 1996). Additionally, a sudden structural break in regional trend beneath Chuchi Lake, indicating a pre-Triassic fault, extends east-southeast from the southern edge of the batholith, transverse to the arc. The implied fault lies along trend with an east-southeast shift in the Hogem regional magnetic high (Fig. 6), indicating deflection by this basement structure. The magnetic anomaly continues ~25 km eastward to the Mount Milligan intrusive suite, a separate monzonite-diorite-granite pluton located ~7 km north of the porphyry deposit (Nelson and Bellefontaine, 1996). Compositions and textures suggest the pluton is an extension of the Hogem batholith, which implies the monzonitic porphyry stocks of the Mt. Milligan deposit to the south also stem from a buried extension of the Hogem. Fault strands of the post-Jurassic Manson-Macleod Lake Fault system dissect the area southeast of the Hogem Batholith (including the Mt. Milligan deposit) into sets of horst and grabens (Nelson et al., 1992).  19  3.4 GEOLOGY OF THE MT. MILLIGAN PORPHYRY AREA The Mt. Milligan Cu-Au porphyry is located within the Nechako Plateau near the southern limits of the Swannell Range of the Omineca Mountains of north-central British Columbia. The deposit is situated on the eastern slopes of a north-south trending chain of peaks that includes Mount Milligan (1,508 m ASL) 8.5 km northwest of the deposit (Fig. 7). Until ~11,000 years ago, the region was covered by the Cordilleran Ice Sheet, which had migrated eastward and northward from the Coast Ranges, altering the preglacial landscape. Consequently, the area has been blanketed by glacial drift resulting in a hummocky landscape of drumlins, flutings, eskers, glacial lakes, and bog-like fens surrounding Rainbow Creek (Fig. 8).  The area is situated ~40 km north of the  continental divide, and drainage is to the northeast via the Nation River into Williston Lake, forming part of the north-flowing Peace-Mackenzie River watershed.  Figure 7: DDH core yard located 2 km west of the MBX stock in a valley between the N-S trending chain of low-lying mountains. Photo by Michael Leznoff (field assistant).  20  Figure 8: View looking east across Witch Lake from helicopter flying southward along the Pinchi Fault valley. The Mt. Milligan Cu-Au porphyry deposit is located ~45 km east over the N-S trending ridge. Note the hummocky glacial landscape. Photo by author.  3.5 DEPOSIT STRATIGRAPHY Because the strata of the Upper Triassic Takla Group are complexly variable and interfingering, Nelson and Bellefontaine (1996) subdivided the stratigraphy into informal ‘successions’ rather than distinct units. These include the Rainbow Creek, Inzana Lake, Witch Lake, and Chuchi Lake successions. The basal Rainbow Creek succession is comprised of grey and limy slate, argillite, and lesser siltstone units. The overlying Inzana Lake succession is a submarine sequence of rocks that grade upward from sedimentary to non-porphyritic volcanic/volcaniclastic rocks that interfinger with the overlying Witch Lake succession. This succession is comprised of submarine augite-phyric volcanic and derivative volcaniclastic rocks of 21  basaltic-andesitic composition. Within 10 km north of the porphyry deposit, a ~3 km thick package of the Inzana Lake succession is conformably overlain by the ~5 km thick Witch Lake succession (Nelson and Bellefontaine, 1996). The Lower Jurassic Chuchi Lake succession is the uppermost unit in the region (Nelson and Bellefontaine, 1996). Heterolithic volcanic-conglomerates, lapilli tuffs, polymictic lahars, and partly subaereal flows compose the Chuchi Lake succession. Feldsparphyric andesitic to trachyandesitic flows are common, with lesser augite- and olivinephyric basalts and trachytes. The main-stage monzonitic intrusions of the Mt. Milligan porphyry deposit are geochemically similar to rocks of the Chuchi Lake succession, and may be coeval. Accordingly, the mineralized porphyritic stocks and host Witch Lake succession rocks may be separated in age by ~20 Ma (after Barrie, 1993; Lang, 1992). 3.6 PROPERTY GEOLOGY – MBX MAIN DEPOSIT The host sequence for the Mt. Milligan deposit consists of coherent volcanic and volcanic-conglomerate flows (or hypabyssal sills) of the Witch Lake succession (Takla Group). These mafic volcanic/volcaniclastic rocks are interbedded with finely laminated and variably brecciated stratiform epiclastic bodies known as the Upper and Lower Trachyte (Fonseca, 2006). At the MBX Main deposit herein studied, this host rock package has been intruded by the monzonitic MBX stock and apophysial Rainbow Dike. The deposit has been divided into four zones (DWBX, WBX, MBX, and 66 zones) based on location of ore and interpreted structure (Sketchley et al, 1995). 3.6.1 Witch Lake succession The Witch Lake succession is comprised of a moderately northeast dipping/facing panel of alternating coherent- and clastic porphyritic basaltic-trachyandesites (after Barrie, 1993; Lang, 1992; Fig. 9-10). Clinopyroxene (augite) phenocrysts, typically ~2.5 mm in diameter, predominate, constituting ~30% of the rock on average. Clinopyroxenes are pale  green  pleochroic,  and  selectively  altered  to  sheaf-textured  actinolite.  Glomeroporphyritic clinopyroxene commonly occur and are indicative of coherent lavas and syn-volcanic intrusions (McPhie, et al. 1993). In the footwall of the Harris Fault in 22  23  Figure 9: Previous page. Geochemistry for Mt. Milligan host rocks and intrusives. Data is after Barrie (1993) and Lang (1992) and is not corrected for potassic alteration. Minimum alkali values are the best indicators of primary lithology. A) Chemical classification and nomenclature of volcanic rocks using the total alkali versus silica (TAS) diagram after Le Bas et al. (1986). Alkalienrichment of host rocks to phono-tephrite (MBX zone) and tephri-phonolite (Upper and Lower Trachyte) reflects potassic alteration. B) CIPW normalized data of Barrie (1993) and Lang (1992) for Mt. Milligan intrusive rocks plotted within the IUGS classification scheme after LeMaitre (1989). Alkali-enrichment of the MBX stock from quartz-monzonite/monzodiorite, and the Rainbow Dike from monzodiorite reflects potassic alteration. C) Alkalinity Classification Diagram for the Mt. Milligan Cu-Au alkalic porphyry plotted with alkaline versus subalkaline fields after Barager and Irvine (1971), and silica-saturated versus silica unsaturated fields after Lang et al. (1994). Alkalienriched (silica-undersaturated) data points reflect potassic alteration. D) Alkalinity Classification Diagram (K2O vs. SiO2 wt%) illustrating the shoshonitic nature of the property. Alkalinity classification after Lang (1992).  the DWBX zone (DDH 90-675), clinopyroxene phenocrysts reach up to 15 mm in diameter and may constitute 40-50% of basaltic units. In the MBX zone, feathery Kfeldspar needles in basaltic-trachyandesite groundmass have ‘bow-tie’ spherulite morphologies indicating devitrification of glassy igneous rocks at intermediate temperatures (650-400ºC; McPhie et al., 1993). Toward the 66 zone, beyond ~250 m from the MBX stock, the host rock lightens in hue to a dull green reflecting a lack of secondary biotite. It contains ~25% clinopyroxene (lesser hornblende), and 3-5% subhedral plagioclase phenocrysts (~0.5 - 1 mm), which are  commonly  fractured  and  partially  resorbed  into  the  trachytic  matrix.  Compositionally, the rock is trachyandesite, reflecting higher silica and alkali content than the basaltic-trachyandesite of the MBX zone (Fig. 9). Plagioclase constitutes 2550% of the trachytic-textured feldspar groundmass, and cryptocrystalline K-feldspar forms the remainder. Plagioclase composition is oligoclase-andesine which is slightly more calcic than oligoclase in the basaltic-trachyandesite and Rainbow dike of the MBX zone, but comparable to plagioclase in the MBX stock and Lower Monzonite dike in the 66 zone.  24  Figure 10: Examples of host rocks of the Witch Creek Succession. A) K-feldspar-Na-feldsparactinolite altered hornblende- augite-phyric trachyandesite (DDH 90-598 at 123.3 m). B) Coarse augite phenocrysts (4-15 mm) in chlorite-altered basaltic-trachyandesite with ribboned pyritecarbonate L3 vein and Na-feldspar-epidote halo (90-675 at 174.2 m).  C) Weakly chloritized  trachyandesite with oligoclase-andesine fragments (1 mm) and augite phenocrysts (90-648 at 176.5). D) Basaltic-trachyandesite with chlorite-altered devitrified groundmass and glomeroporphyritic augite (90-641 at 118).  The host rocks have previously been genetically classified as crystal-, lithic-, and lapillituffs, with minor augite-porphyritic flows and heterolithic debris flows (Rebagliati, et al., 1990; Nelson and Bellefontaine, 1996; Sketchley et al., 1995; Delong, 1996). The genetic term ‘tuff’ implies a pyroclastic origin, which may not be appropriate if the andesites are submarine flows or hypabyssal intrusions. Accordingly, for the present study, the host rocks are simply referred to compositionally as basaltic-trachyandesites 25  or trachyandesites. Volcaniclastic or apparent-breccia units with gravel to cobble sized clasts of porphyritic andesite could well be described as lapilli tuffs (clasts < 64 mm) if these represent subaqueous pyroclastic agglomerates (Nelson and Bellefontaine, 1996), but may also be interpreted as resedemented hyaloclastite (Fig. 11), and are best described simply as volcanic-conglomerates where volcaniclastic origin is unambiguous.  Figure 11: Schematic illustration of a subaqueous, in situ, and resedimented hyaloclastite. The growing hyaloclastite pile is intruded by its feeder dike / stock. Unstable in situ hyaloclastite is resedimented downslope. Abundant conglomeritic trachybasalt to trachyandesitic host rocks at Mt. Milligan suggest a hyaloclastite pile scenario. Image is not to scale (modified from Yamagishi, H., 1987; and McPhie et al., 1993).  3.6.2 Apparent breccia (volcanic-conglomerate) vs. pseudo-breccia Apparent- and/or pseudo-breccia textures comprised of rounded gravel- to cobble-sized clasts of basaltic-trachyandesite in a compositionally similar matrix are common throughout the host rocks (Fig. 12). Rock texture is described as ‘pseudo-breccia’ where phenocrysts continue across the edge of an apparent clast into the surrounding andesite. This is an alteration texture in coherent volcanic rock that occurs by fluid flux through large diameter perlitic fractures during submarine quenching (McPhee et al., 26  1993). Pseudo-clasts represent the isolated remnants of an earlier alteration stage. Discernment of clastic versus coherent protolith is often complicated by reaction fronts along the edges of apparent clasts. Where the rock texture appears more volcaniclastic than coherent, it is described as ‘apparent breccia’. Petrographic analysis of apparent breccia in the DWBX zone indicates a volcaniclastic protolith since clinopyroxene and plagioclase crystals are truncated at the clast edge (Fig. 12C). The hanging-wall of the Lower Trachyte in the Cu-Au mineralized MBX zone consists of ~30 m thickness of apparent breccia with cobble-sized (3-4 cm) clasts. Historically, the host rock sequence has been described as being dominated by monolithologic fragmental varieties of augite-porphyritic rocks (Sketchley et al., 1995).  Figure 12: Volcanic-conglomerate, apparent- and/or pseudo-breccia.  A) Na-plagioclase rich  andesite clast in chloritized, K-feldspar rich andesitic matrix (DDH 90-598 at 215 m). B) Similar volcanic-conglomerate (DDH 90-600 at 166 m). C) Microphotograph of contact (with broken  27  clinopyroxene and plagioclase phenocrysts) between Na-plagioclase rich clast (left) and Kfeldspar-actinolite-chlorite altered andesitic matrix (right; DDH 90-598 at 215 m). D) Apparent- or pseudo-breccia? in sodic-calcic altered MBX zone. Biotite-altered ‘clasts’ are surrounded by Nafeldspar-epidote-pyrite. There is no apparent change in the underlying porphyritic texture (DDH 90-639 at 158 m).  3.6.3 Lower Trachyte The Lower Trachyte is a ~70 m thick stratiform unit of intensely altered rock situated ~180 m below the Rainbow Dike (Fig. 5). At depth in DDH 90-652, the trachyte is pale pink in colour and has thin mafic laminations (S0) separated by ~1.5 mm that resemble shear bands, but could also represent bedding planes in fine epiclastic material (Nelson and Bellefontaine, 1996; Fig. 13A). These laminations, oriented at ~30º to the horizontal axis of drillcore, provide the basis for moderate tilting of the deposit in conjunction with the interpreted orientations of the MBX stock and Rainbow Dike (Sketchley et al., 1995). Compositionally, the Lower Trachyte is more silica-deficient than is implied in the term trachyte, and would be more accurately described as tephriphonolite (after Barrie, 1993; Lang, 1992; Fig. 9). The use of trachyte in the naming of this unit is in reference to trachytic microtexture, rather than composition. In DDH 90-652, combined SWIR (short-wavelength infrared) and XRD (X-ray diffractometer) spectral analysis indicates a Lower Trachyte composition of microclineadularia-albite-muscovite-phengite-illite-biotite-chlorite-pyrite. In thin-section, it has a very fine-grained trachytic texture with abundant ~200 µm acicular feldspar microlites that wrap around pyrite grains. Ghosts of former augite phenocrysts are composed of biotite-chlorite ± actinolite, and may be partially disintegrated into acicular fragments that have been remobilized into the partings (or shear fractures) between layers. Pyrite has replaced some phenocrysts, and has precipitated in the partings (~500 µm wide; Fig 13D) where it is typically chlorite mantled. Closer to the MBX stock (DDH 90-628), the Lower Trachyte has a waxy pink-orange tone, and lacks mineralized laminations. It is crackle-brecciated and cemented with K-  28  feldpsar-quartz-ankerite. In thin section, it alternates between a microlitic acicular texture, and a fine recrystallized granular texture composed of Na- and K-feldspar.  Figure 13: Lower Trachyte (Tephriphonolite). A) Unfoliated K-feldspar-phengite-muscoviteankerite-illite altered trachyte (DDH 90-628 at 199 m). B) Microphotograph of recrystallized granular texture comprised of K-feldspar-Na-feldspar-quartz with interstitial ankerite-sericitechlorite (DDH 90-628 at 199 m). C) Pyrite laminations at ~30º to the horizontal axis of drillcore in K-feldspar-phengite-Na-feldspar-chlorite-illite altered trachyte (DDH 90-652 at 193.7 m). D) Microphotograph of very fine-grained trachytic texture and close-up of mineralized partings (DDH 90-652 at 193.7 m).  3.6.4 Upper Trachyte The 66 zone Upper Trachyte unit is a ~20 m thick sequence of weakly vesicular trachytic-textured rock, recrystallized fine granular lenses, and faults (3-5 m wide) with gravel-sized infill (Fig. 5). Compositionally, the unit plots as tephriphonolite (after Barrie,  29  1993; Lang, 1992, Fig. 9), but may be potassically-altered trachyandesite, based on the presence of relict mafic phenocrysts in thin section. Similar to the Lower Trachyte, sulfide-mineralized hairline fractures resemble shear bands at 35º to the horizontal axis of drillcore (Fig. 14A). Similarly oriented layering is present in granular lenses composed of K-feldspar (~0.1 mm grains), quartz (5-10 modal %), and few plagioclase fragments. Sulfide-bearing sheeted magnetite veins mark the lower contact with andesitic rocks (Fig 14C-D). The veins are ~2 mm wide, and sub-parallel to layering. The unit terminates in magnetite-cemented breccia (DDH 91-815) with rotated and displaced pebble- to granule-sized trachytic clasts altered by K-feldspar and dolomite.  Figure 14: Upper Trachyte (Tephriphonolite): A) K-feldspar altered trachyandesite with Cumineralized hairline fractures at 35º to the horizontal axis of drillcore (DDH 90-643 at 58 m). B) Microphotograph of trachytic texture of K-feldspar matrix with relict phenocrysts replaced by chalcopyrite-biotite (DDH 90-643 at 58 m).  C) Copper-mineralized sheeted magnetite veins at  footwall contact of Upper Trachyte. Dotted lines outline magnetite veins (DDH 90-643 at 80 m). D) Microphotograph of magnetite veins in trachyte with disseminated dolomite/ankerite (DDH 90-643 at 80 m).  30  The similar attitude and depth of the Upper Trachyte and Rainbow Dike suggests continuity between these units. However, the return to high-temperature alteration and mineralization in the hanging-wall of the Rainbow Fault indicates either syn-mineral or post-mineral faulting and separate stratigraphic levels for the two units. 3.6.5 Pyroclastic flow In contrast to the rest of the deposit, the hanging-wall segment of the DWBX zone is comprised of intermittently faulted, polymict fragmental units (possibly resedimented hyaloclastite). Juvenile clasts with non-deformed cuspate and tabular shards, and vesicles filled by hydrous micro-minerals, comprise the western fringe of the system in DDH 90-600 (Fig. 15).  Figure 15: Volcanic-conglomerate with juvenile clasts (hyaloclastite?) in upper DWBX zone. A) Drillcore section showing chloritized volcanic-conglomerate or hyaloclastite (DDH 90-600 at 24 m). B) Microphotograph of same showing devitrified pyroclastic shards with cuspate and platey shapes (McPhie et al., 1993).  Augite and hornblende phenocrysts and fragments altered to  actinolite-chlorite. Matrix is chlorite-sericite altered. Vesicles are filled with fibrous, radiating crystals (zeolites?), quartz, chlorite, calcite and pyrite (DDH 90-600 at 24 m).  3.6.6 MBX stock The MBX stock is a northwest striking ellipsoid intrusive body, with a principal axis ratio of ~2:1, and a diameter of ~400 m. It has a circular near-surface contour due to tilting. The stock is composite with at least three phases, commonly plagioclase-phyric. 31  Compositions range from quartz monzonite to monzodiorite (CIPW normative), and plot within alkaline (Na2O+K2O vs. SiO2 wt%) and shoshonitic (K2O vs. SiO2 wt %) fields (after Barrie, 1993; Lang, 1992; Fig. 9). A variety of textures are present including: 1) plagioclase-phyric, 2) crowded plagioclase-phyric, 3) flow-aligned plagioclase, 4) medium-grained equigranular, 5) xenolithic monzonite (containing both monzonitic and biotite-magnetite hornfels xenoliths), 6) magmatic-hydrothermal breccia (with pink K-feldspar cement), and 7) apparent-breccia (resulting from pervasive K-feldspar alteration) (Fig. 16). Separate intrusive phases were identified using textural changes and the ratio of modal plagioclase to mafics. Contacts between the phases are typically gradational, and/or obscured by pervasive K-feldspar alteration. The compiled data from mapping of the stock is insufficient for defining the intrusive architecture. Nonetheless, interpolation between drill-holes 90-667 and 90-597 in the WBX zone suggests a similar attitude of intrusive phases as the Rainbow Dike in the MBX zone. However, this could be an artefact of sampling, which followed this trend. In the WBX zone (DDH 90-667), the intrusive sequence includes an early monzonite phase with crowded sericitized plagioclase phenocrysts in a K-feldspar rich groundmass. This phase forms the outer ~40 m rim of the stock. Farther inside the stock, the early phase is cut by crowded plagioclase-phyric monzodiorite with a more biotitic groundmass. Both phases can be mineralized. Weakly sericitized plagioclasephyric diorite is the last phase to occur. In general, the rocks are comprised of >60% crowded plagioclase (oligoclaseandesine). Laths are typically 1-4 mm (but can be as coarse as 7 mm), with thin albite rims (<100 µm). Primary K-feldspar phenocrysts (~2.5 mm) also occur. Plagioclase is commonly zoned and replaced by a fine dusting of sericite. Apatite (<1.2 mm) has a mode of 1-2%.  The monzonite groundmass is comprised of ~400 µm grains of  K-feldspar (~80%), Na-plagioclase (10%), hydrothermal biotite (5%) after primary biotite, and magnetite (1-5%). 32  Figure 16: MBX stock. Drillcore sections showing the variety of composition, texture, alteration, and sulfide mineralization of monzonitic to monzodioritic rocks. A) Sericitized plagioclase-phyric monzonite with K-feldpsar rich matrix, and minor disseminated magnetite (DDH 90-616 at 126.0 m depth). B) Weakly sericitized crowded plagioclase-phyric monzonite with K-feldspar altered rims, and interstitial biotite.  Disseminated sulfide replaces biotite surrounding chalcopyrite veinlet  (DDH 90-597 at 223 m). C) Plagioclase-phyric monzodiorite with ~10% magnetite-bearing matrix (DDH 90-667 at 133.0 m). D) Intense K-feldspar-quartz alteration with clotted biotite replaced by  33  sulfide (DDH 90-597 at 223 m). E) Microphotograph of weakly altered monzonite. Sericitized phenocrysts have Na-plagioclase rims. Apatite (opague) at lower right (DDH 90-597 at 123 m). F) Microphotograph of strong potassic alteration converting monzonite to a granular K-feldsparquartz assemblage. K-feldspar replaces sericitized plagioclase laths (90-597 at 223 m).  Figure 17: Datamine image showing the MBX Main deposit in 3-D. Orientation is similar to that of the cross-section in Figure 2. Note the change in the Rainbow Dike east of the MBX stock, where the sill-like attitude suddenly dips inward to form a bowl-shape. Image produced by G. Lusteg (2006), courtesy of Placer Dome Inc.  3.6.7 Rainbow Dike The monzodiorite Rainbow Dike protrudes outward from the southeast-to-eastern margin of the MBX stock (Fig. 4-5). Southeast of the MBX stock, it is a ~50 m thick, east-dipping stratiform body for ~250 m, and would be better described as a sill (Fig. 17). Due east of the stock it has a bowl-shape comprised of a steeply dipping sill-like  34  body that changes to a vertical curvilinear dike-like body ~200 m east from the stock. The peculiar shape might indicate intersection of strata with a pre-mineral, moderately west-dipping, north-south trending fault; or structural control by an inward dipping cone sheet related to emplacement of the MBX stock. Geochemical data (Barrie, 1992) suggests the Rainbow Dike is more silica-undersaturated than the MBX stock.  Figure 18:  Rainbow Dike. A) Potassically altered monzodiorite with carbonate filled micro-  fractures at 30º to the horizontal axis of drillcore (DDH 90-639 at 74.2 m).  B) Biotite-altered  crowded plagioclase-phyric medium-grained monzodiorite (DDH 90-641 at 97.6 m). C) Microphotograph of Rainbow Dike with sub-aligned, crowded oligoclase laths in a fine granular Kfeldspar groundmass. Ankerite and pyrite occur along grain edges and within microfractures. (DDH 90-641 at 105 m).  Where least altered, the Rainbow Dike is a crowded plagioclase-phyric monzodiorite (CIPW-normalized after Barrie, 1993; Lang, 1992; Fig. 9) with smaller diameter 35  phenocrysts (~2 mm) than those common to the MBX stock (Fig. 18). Plagioclase composition is albite-oligoclase (Michel-Levy method). The groundmass is comprised of pale grey K-feldspar grains (~100 µm) with <20% disseminated biotite, trace carbonate and sericite. Within 25 m of the stock (DDH 90-628), the dike contains gravel-sized monzonitic xenoliths that are probably derived from the MBX stock. The dike is typically in fault contact with host rock, but contacts can also be gradational, particularly within ~50 m of the stock, resulting in a ‘monzonite-latite hybrid’ unit used by previous workers. Secondary biotite increases towards the sides of the dike, and is commonly replaced by sulfide at 1-2 modal %. Carbonate-filled micro-fractures are also common. In DDH 90-639 they occur at ~30º to the horizontal axis of drillcore with similar appearance and attitude as sulfide-mineralized partings in the Lower and Upper Trachyte. A magnetite halo (<50 m) surrounds the dike and is closely associated with extensive biotite alteration in the host rock. At 200 m from the MBX stock (DDH 90652), the dike is in contact above and below with 7-10 m wide, strongly magnetic latemineral porphyritic diorite dikes. 3.6.8 Faulting/brecciation associated with the MBX stock 3.6.8.1 Fault zone and brecciation along stock margin At the footwall contact of the MBX stock, near the Lower Trachyte, a <35 m wide fault and breccia zone occurs within monzonite and surrounding biotite-hornfels (Fig. 19A). The Lower Trachyte grades outward from this fault zone, and develops into the Cu-Au mineralized massive-to-laminated trachytic unit ~50 m southeast of the stock. The fault intervals are composed of carbonate-rich gouge and gravel in monzonite, becoming chloritic in biotite hornfels. Monzonite clasts are typical of the MBX stock. They are moderately K-feldspar altered with weakly sericitized sub-aligned plagioclase, 2-3% disseminated magnetite, ~1% Cu-sulfide after biotite, and ankeritic fracture fill. Alternately, the wall rock contacts are intensely K-feldspar altered (<70 modal %), with <40% magnetite, ~20% carbonate, 3-5% disseminated and vein chalcopyrite, sheeted pyrite and magnetite veins, and grades up to 1.1 wt% Cu and 1.1 g/t Au (DDH 90-684).  36  Figure 19:  Brecciation associated with the MBX stock.  A) Magmatic-hydrothermal jigsaw-fit  breccia at the MBX stock margin, with K-feldspar cement (E1-type veins) and magnetite-rich monzodioritic clasts (DDH 90-628 at 24 m). B) Clast-supported magmatic-hydrothermal breccia with rounded pebble-to-granule sized monzonite fragments in K-feldspar-magnetite cement, cut by chalcopyrite veinlet (DDH 90-597 at 152.2 m). C) Clast-rotated dolomite cemented breccia located ~50 m from the stock margin in the MBX zone (DDH 90-628 at 133.1 m).  D)  Microphotograph of dolomite breccia with ankerite-dolomite-Na-feldspar-K-feldspar-biotite altered clasts.  3.6.8.2 MBX breccia body Historic drill logs indicate a variably jigsaw-brecciated to clast-rotated breccia body extending the length of the MBX stock. It ranges in thickness from 2 m in the WBX zone, to 50 m beneath the poorly mineralized center of the stock, to 5 m near the MBX zone (Fig. 5). When plotted on the hinged cross-section, it appears to have a similar attitude as the Rainbow Dike, but this is poorly constrained.  Where observed in 37  drillcore, it varies from a hematitic pink K-feldspar cemented crackle-breccia to a milled breccia with rounded monzonite pebbles in a magnetite-rich matrix. Chalcopyrite veinlets cut both clasts and matrix (Fig. 19B). Previous workers describe the breccia cement as K-feldspar-carbonate-sericite ±chlorite ±clay, with 1-3% magnetite and 1-2% fine disseminated sulfide. They also document stockwork pyrite veins (5 mm) with K-feldspar-quartz selvages, and late sericite stringers in brecciated zones. At shallow levels of the stock, the breccia is described as mylonitic, referring to milling and/or pulverization of the breccia clasts, possibly with flow texture. 3.6.8.3 Dolomite-cemented breccia A 6 m horizon of clast-rotated breccia in vuggy dolomite cement extends outward from the margin of the MBX stock for ~50 m within the footwall of the Rainbow Dike (Fig. 19C-D). Brecciation post-dates potassic alteration, and may be coeval with Na-feldsparcarbonate-chlorite alteration. Fire-assay data indicates strong Cu-Au mineralization near the stock margin (DDH 90-630) with ~0.95 wt% Cu and 8 g/t Au in at the upper and lower contact of the breccia, and ~0.52 Cu and ~3.5 g/t Au within the breccia itself. In the 66 zone, a similar dolomite breccia follows the footwall of the Lower Monzonite Dike (DDH 90-648). Gold grade is elevated to ~0.4 g/t for a thickness of 30 m below the dike and breccia. Based on C and O isotopes in Part 2 of the present thesis (Ch. 6.8.3.2), dolomite cementation probably represents post-mineral incursion of seawater along permeable and/or faulted contacts between intrusions and host rock. 3.6.9 Late-mineral dikes Late-mineral dikes in the MBX Main deposit include northeast-trending, moderately northwest-dipping trachyte and monzonite dikes. Northwest-trending, steeply northeastdipping porphyritic diorite dikes are the youngest intrusive rock to occur (Sketchley et al., 1995).  38  Figure 20: Late-mineral dikes. A) Trachyte dike (DDH 90-643 at 51.7 m). B) Monzodiorite dike (90675 at 142.0 m). C) Monzonite dike (90-600 at 14.2 m). D) Porphyritic diorite dike cut by epidote veinlet. The dike has an Au-rich upper contact and footwall (91-815 at 141.0).  3.6.9.1 Trachyte dike Trachyte dikes (<15 m wide) are fine-grained, grey pink dikes with ~3% disseminated biotite, and are typically magnetite-bearing (Delong, 1996). They occur in the southwestern portion of the Main deposit, in the northern portion of the Southern Star deposit (outside of the scope of the present study), and at the hanging-wall contact of the Upper Trachyte in the 66 zone (DDH 90-643; Fig. 20A). 3.6.9.2 Monzonite and monzodiorite dikes Late-mineral monzonite and monzodiorite dikes (<10 m wide) are encountered within the MBX stock, and DWBX zone, but otherwise do not intersect the cross-section of the present study. Where seen, they are comprised of flow-aligned plagioclase laths (< 5  39  mm) in a fine-grained mafic-poor, K-feldspar groundmass (Fig. 20B-C). Mafic phases may be replaced by pyrite. 3.6.9.3 Hornblende- plagioclase-phyric diorite dikes Northwest-trending, steeply northeast-dipping hornblende- plagioclase-phyric diorite dikes are the youngest intrusive rock to occur. They are common in the northeastern portion of the MBX stock, Main and 66 zones (Delong, 1996).  The dikes can be  pervasively magnetic, and appear relatively unaltered in hand sample.  Zoned  plagioclase phenocrysts (<3 mm) comprise ~20% of the diorite and are randomly oriented, whereas hornblende phenocrysts represent ~5%. In thin section, plagioclase is partially altered to sericite, and hornblende is altered to chlorite-epidote-carbonate. The groundmass is chloritized, contains fine disseminated magnetite (3-5 modal %), and trace K-feldspar and epidote. A late epidote veinlet cuts a diorite dike in the 66 zone indicating continued hydrothermal activity after dike emplacement (Fig 20D). The ~11 m thick dike has an Au mineralized halo averaging 2.45 g/t over 10 m in the footwall and 22.5 g/t at the upper contact (Appendix 1). 3.6.10 Structurally-defined ore zones The deposit has been divided into four zones based on location of ore and interpreted structure (Rebagliati, 1990; Sketchley et al., 1995; Fig. 21). 3.6.10.1 DWBX zone West of the stock is the north-northwest striking, steeply east-dipping Harris Fault, which separates the DWBX (down-dropped WBX) on the west from the WBX zone to the east. An additional fault (the DWBX Fault), identified in this study, post-dates the Harris Fault, dips shallowly to the west and separates upper and lower segments of the DWBX zone. Chlorite-altered pyroclastic and volcaniclastic trachyandesites (and abundant minor faults) dominate the hanging-wall segment, whereas propyliticallyaltered, pyrite-bearing volcanic-conglomerate is characteristic of the footwall segment. A ~60 m envelope of biotite-hornfels surrounds the downdropped monzonite with a ~30 m inner chalcopyrite-pyrite zone. 40  Figure 21: Copper grade, Au grade, and Cu-Au ratio plotted on hinged cross-section. Data is binned into 5 ranges using the Jenks Natural Breaks classing method, which is based on identifying groupings that naturally exist in the data (Jenks and Caspall, 1971). Fire assay data provided by Placer Dome Inc.  3.6.10.2 WBX zone The WBX zone includes the western portion of the MBX stock, the deepest continuous portion, plus a ~40 m wide biotite-altered envelope of MBX monzonite and host rock that is cut-off by the DWBX fault. Copper grade is typically greater than 0.3 wt.%. The WBX zone has the highest Cu/Au ratio of the deposit.  41  3.6.10.3 MBX and 66 zones The MBX (magnetite breccia) zone represents the main Cu-Au ore body. It is located immediately southeast of the MBX stock along strike of the Rainbow Dike and Lower Trachyte horizons. The Rainbow Fault, an east-northeast trending, moderately southeast dipping cross-fault, truncates the Rainbow Dike and MBX zone to the south, separating it from the downthrown 66 Zone. The north-striking, shallowly east-dipping, regional Great Eastern Fault (a strand of the Manson-Macleod lake fault system) cuts off the MBX zone and hydrothermal system immediately east of the bowl-like portion of the Rainbow Dike. The fault separates the Mt. Milligan system from early Tertiary volcanic and sedimentary rocks. A second cross-fault, the east-northeast trending subvertical Oliver Fault, lies immediately north of the MBX stock. Historically, the Rainbow Fault was considered to be a pre-mineral structure trending northward and dipping shallowly to the east, concordant to bedding (Sketchley et al., 1995). It was thought to have guided emplacement of the Rainbow Dike southeast of the MBX stock. In 2004, Placer Dome reinterpreted the geology and the Rainbow Fault to strike northeast and offset the hydrothermal system as a post-mineral normal fault. The fault, with attitude 050, 50º SE separates the Cu-Au rich zone (MBX zone) and the Cu-poor, Au-rich zone (66 Zone).  42  CHAPTER 4: DEPOSIT HISTORY AND PREVIOUS WORK 4.1 HISTORY OF THE MT. MILLIGAN DEPOSIT The following deposit history is derived from Nelson and Bellefontaine (1996), Lusteg (2006), and Terrane Metals publications (2006, 2007). In 1937, while en route to his placer claims, prospector George Snell collected Aubearing pyritic andesite float on the western flank of Mount Milligan. Snell collected five more samples west of Mitzi Lake with assay values of trace to 148.8 g/t Au. At that time, the Consolidated Mining and Smelting Company examined area with negative results. With passing years, Snell’s anomalous gold discovery was relegated to local exploration folklore. More than 30 years later, in 1972, Pechiney Development Ltd drilled five DDH’s near Heidi Lake, ~2 km south of the Snell discovery, based on the location of IP and soil geochemical anomalies, but again generated negative results. In 1983, Selco Inc., exploring for Cu-Au porphyries, staked a larger twelve claims block immediately east of the deposit. During preliminary surveys they recovered abandoned Pechiney core that was propylitically altered with anomalous Au-As values. In 1984, Selco amalgamated with BP Resources Canada Ltd, who were also investigating the BP-Chuchi property in the Nation Lakes district, ~35 km northwest of Mt. Milligan. In 1983, prospector Richard Haslinger discovered Cu-Au mineralized bedrock in a creek bank only 100’s of meters east of the BP Resources claims. He staked the claims in 1984, which were optioned by BP later that year. In 1984 and 1985, BP conducted soil geochemical, magnetic and IP surveys, plus a trenching program near Heidi Lake. By 1986, BP was downsizing their mineral exploration program and abandoned the project before the system center was located.  43  Ex-BP geologist Mark Rebagliati marketed the BP claims to Lincoln Resources Ltd. who resumed drilling in 1987 through an option agreement made with BP in April, 1986. Lincoln entered into an option agreement with Haslinger in July 1986. In September 1987, the Lincoln drill program shifted east under Rebagliati’s direction to explore a combined magnetic, geochemical, and IP anomaly in glaciofluvial drift within a Haslinger claim, and the Cu-Au mineralized MBX zone of the Mt. Milligan Main deposit was discovered. In 1988, Lincoln reorganized as United Lincoln Resources and expanded their claims surrounding the Mt. Milligan deposit. Later that year United Lincoln amalgamated with Continental Gold Corp. In 1989, the new Continental Gold Corp. expanded the Cu-Au mineralized zone through additional drilling, and in 1990, more claims were acquired in the region. Between September 1990 and January 1991, Placer Dome Inc. purchased the BP mineral claims and acquired the Continental Gold shares through a $266 million takeover bid (Northern Miner, daily news, 4/24/2006) making Placer Dome the sole proprietor of the Mt. Milligan project. In late 1990 and 1991 Placer Dome continued exploration drilling, but by 1992, due to falling metal prices, had written down the lowgrade/large tonnage property as subeconomic, not worth the capital investment to develop. Since its discovery, a total of 194,467 m of diamond drilling in 911 holes had defined the Mt. Milligan deposit. In 1996, Placer Dome re-evaluated the project with an updated geological model for the Main zone designed to improve grade estimates, and in 1998, an economic reevaluation was completed for the combined Main and Southern Star deposits. In 2004, historical data was integrated in a GIS, and geostatistically processed to enhance visualization of combined physical and chemical properties. An updated 3-D model was constructed for a new resource estimation that included the re-oriented Rainbow Fault. A 14 DDH drill program under the direction of Anna Fonseca was conducted to provide fresh material for metallurgical testing. In 2005, a BLEG stream sediment sampling 44  program was undertaken to determine downstream dispersion of metals, and the present M.Sc. study (through MDRU/UBC) to investigate alteration geometry and construct a 3-D alteration model was in its first field season. In early 2006, Barrick Gold acquired Placer Dome through a $10.4 billion takeover bid (Northern Miner, daily news, 4/24/2006) as Placer Dome was completing the updated pre-feasibility study for Mt. Milligan. As part of the deal, Goldcorp purchased all of Placer Dome’s Canadian assets. In April 2006, Atlas Cromwell purchased five of the former Placer Dome projects from Goldcorp, including Mt. Milligan, in exchange for preferred shares, and changed its name to Terrane Metals. The president and CEO of Terrance Metals is the former head of Canadian Exploration for Placer Dome, Rob Pease. In 2006, Terrane Metals outsourced an engineering and technical team to prepare a feasibility study on Mt. Milligan involving a two-phase drill program.  The Phase I  program returned Cu-Au grade consistent with historic results for the property with 8,200 m of large diameter drilling for metallurgical test work. The Phase II program was designed to upgrade ‘inferred’ resources into ‘measured’ and ‘indicated’ categories through 9,300 m of drilling. Phase II expanded Cu-Au mineralization into the DWBX zone (which had not been included in historic resource estimates), and recovered locally high Ag grade (~1000 g/t) in the 66 zone. The results were incorporated into an updated grade block model, open pit resource estimate, and mine plan. Terrane also supported the second field season of this M.Sc. project in 2006. 4.2 PREVIOUS WORK Previous work done on the Mt. Milligan deposits includes an M.Sc. thesis by Delong (1996).  Delong (1996) used 114 sample points distributed across the MBX, WBX,  DWBX zones, and Southern Star deposit at the 1000 m elevation level for a 2dimensional analysis of metal concentration and alteration mineral modes. The thesis focused on variations in biotite composition between igneous biotite related to the magmatic intrusion and hydrothermal biotite related to potassic alteration in the MBX 45  and 66 zones. Changes in biotite composition were modelled statistically by converting the cation compositional data to Thompson exchange components (Thompson, 1986). Delong (1996) concluded that subtly lower Mg/(Mg+Fe) values of hydrothermal biotite in the 66 zone were not the result of direct substitution of Fe for Mg, but rather resulted from the incipient breakdown of biotite to chlorite. This breakdown is proposed to have occurred as retrograde fluids overprinted earlier biotite alteration during thermal decay of the hydrothermal system, as isotherms collapsed inward against the intrusion. In addition, Delong researched chemical models of metal zoning and Au distribution in porphyry (Jones, 1992) and volcanogenic massive sulfide deposits (Huston and Large, 1989). These studies show that Au can be transported as a chloro-complex (AuCl2-) in early higher temperature hydrothermal fluids, and later as a bisulfide complex [Au(HS)2-] in more evolved, acid fluids. Accordingly, the continuation of elevated Au grade from the MBX to the 66 zone suggests the 66 zone may have encountered two phases of Au deposition.  The first phase would have involved a higher temperature fluid (350-  275°C), potassic alteration, magnetite precipitation, and the precipitation of Cu and Au from chloro-complexes. The second phase would involve a lower temperature fluid (250-200°C), propylitic alteration, precipitation of pyrite, and precipitation of Au from bisulfide complexes. Delong concludes that the two-phase model of Au deposition is consistent with results from the biotite analysis which suggest the Au- and pyrite-rich 66 zone saw at least two phases of alteration: potassic, and an overprinting propylitic phase in which hydrothermal biotite was altered to chlorite. Stanley (1993) also addressed the two-phase thermodynamic model for Mt. Milligan. He showed that the Cu/Au abundance ratios of the mineralized rock in the potassic alteration zone were more strongly correlated and had a higher magnitude than areas of propylitic overprinting. The Cu/Au ratio in the potassic zone is expected to be relatively constant because the saturation ratio of Cu and Au chloro-complexes is temperature insensitive. In a bubble-plot graph of Cu versus Au concentration in the potassicallyaltered MBX West Zone, Stanley (1993) estimated a mode ratio of 15,000 wt.% Cu/Au(ppm) (represented as a straight line through the origin) – a value consistent with 46  the composition expected for the hydrothermal fluid at ~350°C. In a second plot for the MBX East Zone, the area of focus for the present study, Stanley observed a trend toward lower ratios, or an increase in Au abundance (although the mode value of 15,000 remained the same). He accounted for this by proposing Au precipitation from a bisulfide complex at lower temperatures. In a geophysical study of the Mt. Milligan site, Oldenburg et al. (1997) corroborated the geochemical/geological work done by Stanley and Delong. Four geophysical data sets [magnetic  susceptibility,  DC  resistivity,  induced  polarity  (IP),  and  airborne  electromagnetic (EM)] were inverted to develop 3-D distributions of physical properties. Magnetic susceptibility is primarily a measure of magnetite content. IP is a measure of the chargeability of electrically conductive materials, and will be stronger where there is sulfide. DC resistivity is a measure of sub-surface electrical conductivity and will be strongest where there is significant sulfide mineralization, fluid filled fractures, or pore water. These conductivity, chargeability, and susceptibility models were compared to a geologic cross-section at 9600N compiled from data for 600 drill-holes, and to a 3-D model of gold concentration (provided by Placer Dome Inc.). The results showed that low conductivity and high susceptibility correlate with the monzonitic stock and fault regions, whereas high chargeability correlates with the volcanic host rocks bordering the MBX stock and Rainbow Dike. Interestingly, an offset was observed between the area of highest Au concentration (nested against the MBX stock within the hanging-wall of Rainbow Dike) and the region of highest chargeability (centered farther east along the Rainbow Dike). The highest Au concentration is expected to be located within strong potassic alteration where metals precipitated from chloro-complexes. The fact that the chargeability high resides east of this illustrates continued sulfide mineralization (dominated by pyrite) beyond the ore-zone. However, there is a smaller Au-rich region on the inner tip of the chargeability high that may represent a separate Au population precipitated from a bisulfide complex. Also of note, the larger Au-rich zone near the stock is situated within moderate magnetic susceptibility likely describing the mineral assemblage: magnetite-chalcopyrite-pyrite. The smaller gold-rich zone occurs in low magnetic susceptibility supporting Stanley’s argument that cooler hydrothermal fluids, 47  containing bisulfide gold, reacted with secondary magnetite (precipitated from an earlier potassic alteration stage) to form Au-enriched pyrite, minor hematite, and water. Au(HS)-2 + Fe3O4 = Au0 + FeS2 + [Fe2O3] + H2O + eIn their study, Oldenburg et al. (1997) also propose that high chargeability data on both the west and east sides of the MBX stock at 9600N supports the argument (based on trachytic marker beds) that the deposit has been tilted and rotated clockwise subsequent to formation. It suggests an annulus of sulfide mineralization, or pyrite halo, around a vertically oriented stock. Other important contributions include the fieldwork of Rebagliati (1988) and Sketchley et al. (1995) in defining the deposit geology and ore body; geochemical analysis of Mt. Milligan rocks by Lang (1992) and Barrie (1993); U-Pb dating of zircons recovered from separate stocks of the monzonite cluster by Mortensen et al. (1995); regional mapping of the Nation Lakes area and U-Pb dating of rutile from the Rainbow Dike by Nelson and Bellefontaine (1996); and alteration modelling and detailed petrography by Fonseca (2004, 2005).  48  CHAPTER 5: ALTERATION AND SULFIDE MINERALIZATION 5.1 ANALYTICAL TECHNIQUES 5.1.1 Core logging The bulk of geological information presented in this thesis was obtained through detailed re-logging of over 3600 m of drillcore from 15 holes produced between 19891991 by Continental Gold and Placer Dome Inc. The holes were chosen from within 50 m of a hinged cross-section spanning the 4 structurally defined ore zones on the west and southeast of the central MBX stock, and within the stock itself. 5.1.2 Petrography Detailed petrographic analysis of over 100 thin sections selected from ~550 rock specimens was conducted at the University of British Columbia. 5.1.3 SWIR (short wave infrared) spectra As a reconnaissance method for identifying alteration minerals, 160 drillcore samples across the MBX and 66 zones were spectrally analyzed by the author, and corroborated by A. Fonseca (associated with Analytical Spectral Devices, Inc.).  Spectra were  compiled with results from petrography, SEM and XRD in the identification of alteration mineral assemblages. Reflectance spectra were collected with an ASD TerraSpec® device provided by Placer Dome Inc. with a full spectral range (350-2500 nm), a 6 to 7nm spectral resolution, and rapid data collection (1/10th of a second per spectrum). Spectra were identified with SPECMIN-PRO version 3.1 software using spectra libraries collected by ASD (over 1500 spectra) and USGS (400 spectra). These libraries are focussed on minerals that have detectable absorption features in the visible (0.4-0.7µm), near-infrared (0.71.3µm), and short-wave infrared (1.3-2.5 µm) ranges of the EM spectrum. The ASD device identified hydrous and hydroxyl-bearing minerals of mafic to felsic composition (including tourmaline and Na-amphiboles), carbonates, and some Cu49  sulfates. K- and Na-feldspars were not identified, except for their hydrolyzed alteration products. 5.1.4 SEM (scanning electron microscope) Scanning electron microscopy was performed on a Philips XL-30 with a Princton Gamma-Tech (PGT) EDS/IAS system at the University of British Columbia. A 15 kV beam was used with a spot size of 5 um, and a BSE detector at a working distance of 10.0 mm. SEM was used following petrography for more accurate mineral identification, such as for multiphase pseudomorphs of clinopyroxene phenocrysts. It was also used to observe and document mineral replacement textures in sodic-calcic and innerpropylitic assemblages, to generate qualitative carbonate compositions in the carbonate-phyllic assemblage, and to determine the location of Au in sulfide mineralization. 5.1.5 XRD (X-ray diffractometry) Eight XRD analyses were performed on a Siemens D5000 X-ray Powder Diffractometer at the University of British Columbia using DIFFRACplus software. Intensely altered fine-grained specimens with bulk mineralogy unidentified by petrography and SEM were selected for XRD analysis. These include the intensely and pervasively altered Lower Trachyte, inner-propylitic alteration veins in the MBX zone, and carbonate-phyllic alteration along the upper margin of the Rainbow Dike and in the 66 zone. 5.2 OVERVIEW The following review of alteration-mineralization sequencing at the Mt. Milligan Cu-Au alkalic porphyry deposit is summarized in Fig. 22-23,and Table 1 and 2. 5.2.1 MBX zone In the MBX zone, Cu-Au bearing minerals are associated with potassic alteration and magnetite along the brecciated margin of the MBX stock, and the stratiform Lower Trachyte and Rainbow Dike within the biotite alteration shell. Sulfide-bearing quartz veins are also concentrated at the MBX stock margins within monzonite and biotite 50  Figure 22: Paragenetic sequence for alteration and sulfide minerals determined from detailed core logging, petrography, ASD and XRD spectral data, and SEM analyses.  hornfels. A zone of sodic-calcic alteration (Na-feldspar-actinolite-epidote) has overprinted the outer margin of the potassic shell and develops outward to inner- and outer-propylitic alteration (epidote-Na-feldspar-calcite-actinolite-chlorite), and regional chloritic alteration. Copper and Au grade are maximized where albitization of the potassic zone is strongest. Moderate Au grade continues outward within the pyrite halo associated with the peripheral assemblages. A carbonate-phyllic (dolomite-ankerite51  52  Figure 23: Previous page.  Hinged cross-section showing lateral zoning of alteration  assemblages: potassic and calc-potassic, sodic-calcic, and inner- and outer-propylitic. Carbonate-phyllic alteration is superposed on potassic alteration in the downdropped 66 zone. Location of alteration shells was complied from core logging and historic drill logs.  53  54  sericite-pyrite) vein within the distal Rainbow Dike has elevated Au-Cu grade. Latestage epidote-chlorite-pyrite has exploited permeable stratigraphic horizons within the biotite shell. 5.2.2 66 zone In the 66 zone, above the Rainbow Fault, the potassic assemblage reappears, and is marked by pervasive K-feldspar alteration and Cu mineralization of the Upper Trachyte, and biotite alteration of surrounding trachyandesite. Sheeted magnetite veins affect the lower contact of the Upper Trachyte. The unit terminates in a magnetite breccia, which transitions into a zone of intense carbonate-phyllic alteration. Elevated Au grade is present within minor faults and along late-mineral dike contacts within the carbonatephyllic assemblage, but decreases with distance from the Upper Trachyte. Gold grade sharply decreases in the outer-propylitic and chloritic alteration surrounding the carbonate-phyllic shell. 5.2.3 DWBX zone In the lower DWBX zone, a ~30 m envelope of Cu-Au mineralization, associated with potassic alteration and magnetite, is nested along the upper contact of the monzonite body (potentially the downdropped MBX stock) where it is cut off by the Harris Fault. Pervasive outer-propylitic alteration of volcanic-conglomerate borders this biotite hornfels envelope, and becomes more chloritic towards the Harris Fault. Late pyrite veins (L3) that cut the potassic zone may be genetically linked with the outer-propylitic assemblage. 5.3 K-FELDSPAR ± QUARTZ (potassic shell) Within the stock, pervasive hematitic (pink) K-feldspar alteration converts the porphyritic monzonite to a more equigranular K-feldspar-quartz assemblage (~0.1 mm grains; Fig. 16F). Sericitized plagioclase phenocrysts are replaced by K-feldspar, and K-feldspar forms rims on relict laths. Disseminated quartz (<5 modal %) is present in halos to Kfeldspar-quartz veinlets (E1-stage). It also occurs interstitially to K-feldspar in potassically-altered groundmass. 55  Secondary K-feldspar, magnetite, and Cu-sulfide minerals are positively correlated. Mineralized monzonite typically has >10% secondary K-feldspar in groundmass. Sulfide replaces biotite, and commonly has a mode of 1-2% (chalcopyrite > pyrite > bornite). Where alteration is strongest, disseminated biotite has been remobilized into clots (<10 mm; Fig. 16D). Trace rutile occurs with biotite as a replacement phase after primary biotite. Magnetite mode increases to 15-25% in strongly altered crackle-breccia and minor clast-rotated breccia.  Figure 24: Potassic alteration. Each frame shows the compiled alteration shell for the key mineral and its distribution across the hinged cross-section of the present study. Note the re-  56  intensification of alteration in the hanging-wall of the Rainbow Dike, concentrated along the Upper Trachyte horizon.  In the deep stock, the ratio of average modal ‘plagioclase: mafic mineral: K-feldspar’ (ignoring minor and accessory minerals) is 38.5%: 24.0%: 37.5% [DDH's 90-667, 90597]. Higher in the stock, towards the Rainbow Dike, the ratio changes to 28.0%: 11.0%: 61.0% [90-616], reflecting increased infiltration of potassic fluids at shallower levels. Pervasive K-feldspar alteration, K-feldspar cemented crackle-breccia, and E1 veins are prominent at the southwest margin of the stock (Fig. 19A). Pervasive pink K-feldspar alteration extends along the Rainbow Dike for approximately 20 m outboard of the MBX stock. Farther from the contact, cloudy grey K-feldspar and biotite represent potassic alteration along the dike. Alternately, pink K-feldspar replaces the upper margin of the Lower Trachyte for up to 260 m from the stock, reflecting either a more felsic and permeable protolith, or a higher intensity of alteration (Fig. 24). Because of the limited drillcore on site for the Lower Trachyte, it is largely without data in Fig. 24, and it likely enveloped in a more extensive K-feldspar shell. 5.4 QUARTZ VEINS 5.4.1 Quartz-K-feldspar ± sulfide veins (E2-stage) Early-intermediate E2-stage quartz veins (A-type veins of Gustafson and Hunt, 1975) are composed of quartz (70%), K-feldspar (~10%) and calcite ± anhydrite at <15 modal % (Fig. 25). Feldspar (grey perthite in thin section), calcite and Cu-sulfide (chalcopyrite and trace bornite) are interstitial to quartz grains, which range in size from 0.5 to 3 mm. Thin pink K-feldspar selvages (~500 µm) are present in monzonitic rocks, whereas biotite-magnetite halos (~10 mm) occur in biotite hornfels. Vein widths range from 4 to 40 mm. They increase in size and density in a <30 m wide zone of faults (gougebearing) and K-feldspar cemented crackle-breccia at the margin of the MBX stock (especially within the transitional contact with the Lower Trachyte), and in the surrounding biotite hornfels. These veins have irregular, undulating walls indicating emplacement in a ductile host at elevated temperatures. Farther within in the stock, E2  57  veins have more parallel walls indicating greater host rock control, and may show internal symmetry such as a sulfide centre-line. Such veins may be segmented and displaced along hairline fractures. In the WBX zone, the deepest portion of the MBX stock (with the highest Cu/Au ratio), rare bornite-bearing quartz veins (~8 mm) are mottled by Fe-carbonate that may be a younger overprint. 5.4.2 Accessory quartz veinlets Quartz veinlets are abundant within the Rainbow Dike for ~170 m beyond the MBX stock, and generally do not extend more than 10 m into the host rock. They are intercut with pyrite veinlets indicating multiple quartz vein episodes during the pyrite-dominant sulfide stage. Irregular, milky quartz veins are present in the intensely altered Lower Trachyte for ~90 m from the MBX stock.  Figure 25: Quartz veins. A) Pyrite-mineralized A-type quartz vein with 8 mm biotite halo (DDH 90616 at 256.7 m). B) Pyrite-chalcopyrite-magnetite bearing A-type quartz vein with magnetite halo  58  and pinch-and-swell morphology suggesting emplacement in ductile biotite hornfels (DDH 90-675 at 236.5 m). C) Coarsely ribboned chalcopyrite-pyrite-magnetite bearing A-type quartz-K-feldspar vein in fault zone between the MBX stock and Lower Trachyte (DDH 90-616 at 158.0 m). D) Microfaulted quartz vein with sulfide centerline and 0.5 mm K-feldspar selvage (DDH 90-616 at 148 m).  5.5 K-FELDSPAR-BIOTITE-ACTINOLITE (inner calc-potassic shell) Pervasive biotitization of andesitic host rock is the most extensive early alteration event, reaching ~260 m from the MBX stock on the southeast and ~130 m on the west (Fig. 24). In the 66 zone, biotite alteration re-intensifies and extends southeast for another 180 m above the Rainbow Fault. Biotite alteration affects the margins of the Rainbow Dike up to 70 m from the MBX stock, and becomes pervasive throughout the dike by 150 m. The monzodiorite groundmass hosts biotite-filled microfractures and disseminated clusters (<1 mm) that can become the dominant groundmass phase. In basalt-trachyandesite, pervasive biotite alteration consists of green-brown pleochroic biotite flakes (15 µm) intergrown with feathery K-feldspar microlites (Fig. 26C). The Kfeldspar-biotite association comprises ~60% or more of the mineral mode. Microcrystalline apatite (<30 µm) is disseminated throughout, and is observable with SEM. Calcite is also weakly disseminated, and may be a byproduct of calc-potassic alteration (see Ch. 5.27.1.1). In hand specimen, biotite-altered host rocks are dark grey to black in colour. Porphyritic textures are obscured, but not destroyed, except in recrystallized biotite-magnetite hornfels within ~60 m of the MBX stock. Where biotitization is strongest, both biotite and K-feldspar replace hornblende and partially replace clinopyroxene. Clinopyroxene is more commonly altered to sieve-textured actinolite and/or uralite. The presence of actinolitized clinopyroxene with pervasive biotite and K-feldspar alteration is diagnostic of the ‘inner calc-potassic’ assemblage of Holliday and Cooke (2007).  59  Figure 26: Inner calc-potassic alteration of basaltic-trachyandesite. A) Pervasive biotite alteration of groundmass with clinopyroxene phenocrysts replaced by actinolite-pyrite-Na-feldspar-biotiteK-feldspar (DDH 90-639 at 35 m). B) Microphotograph (plane polars) of actinolitized phenocrysts with subsequent pyrite-calcite replacement (DDH 90-639 at 35 m). C) Green-brown pleochroic biotite alteration in thin section and K-feldspar-Na-feldspar-calcite-pyrite-chalcopyrite-enargite after clinopyroxene (DDH 90-639 at 51 m). D) SEM microphotograph showing concentration of Nafeldspar as a clinopyroxene replacement phase with biotite and K-feldspar. Ab = Na-feldspar, Ksp = K-feldspar, bt = biotite, py = pyrite (DDH 90-639 at 13 m).  5.6 MAGNETITE (potassic- and inner calc-potassic shell) Finely disseminated magnetite (up to ~10% mode) belongs to both the potassic and calc-potassic assemblages, with which it is thermodynamically stable (Guilbert and Park, 1986; Holliday and Cooke, 2007). However, it is less extensive laterally (Fig. 24) in part due to magnetite-destruction by peripheral and/or retrograde fluids. A magnetite  60  halo (20-80 m) surrounds the MBX stock and Rainbow Dike, as well as the Upper Trachyte in the 66 zone. Magnetite is focussed within faulted contacts between the Rainbow Dike and host rock (Appendix A). In basaltic-trachyandesite of the MBX zone, magnetite surrounds quartz-K-feldspar (E2) veins (Fig. 25B). Additionally, magnetite-chalcopyrite veins (E5) occur at the upper contact of the Rainbow Dike within ~70 m of the MBX stock. Magnetite stringers and fine disseminations are present within the Rainbow Dike up to 180 m from the stock and are observed cutting older quartz veinlets.  At 230 m from the stock, only trace  magnetite remains, and hematite becomes the prevalent Fe-oxide. In the transition between calc-potassic and sodic-calcic zones (DDH 90-639), pyritemagnetite occurs as disseminated clots and clusters with chlorite halos. Disseminated pyrite can have magnetite ± chalcopyrite rims, and the same is present in halos to chalcopyrite-pyrite (E4) veins. Magnetite ± sulfide-quartz-carbonate (E5) veinlets are less common in the sodic-calcic assemblage. Magnetite and chalcopyrite abundances correlate positively. Within the envelope of biotite hornfels in the DWBX zone, magnetite occurs with clotty, disseminated chalcopyrite-pyrite. Disseminated magnetite and magnetite ± sulfidecarbonate (E5) veinlets are <5 modal %. 5.7 EARLY STAGE SULFIDE VEINS (potassic- and inner calc-potassic shell) Early chalcopyrite- and magnetite-chalcopyrite ± pyrite veins are concentrated within 15 m of the MBX stock in the DWBX zone, and within 100 m in the MBX zone. Veins in the MBX zone are generally concentrated within ~15-20 m of the Rainbow Dike. Early vein types are low in abundance and density. 5.7.1 Early chalcopyrite veins (E3-stage) Early chalcopyrite veins are 4 to 7 mm wide with trace carbonate between chalcopyrite grains and diffuse carbonate-magnetite halos (>10 mm) that overprint biotite alteration 61  (Fig. 27A). Vein walls are typically sharp, but undulating.  In the MBX stock,  chalcopyrite veinlets are associated with K-feldspar-magnetite alteration. 5.7.2 Chalcopyrite-pyrite veins (E4-stage) Chalcopyrite-pyrite veins are most abundant in the hanging-wall of the Rainbow Dike (within 15 meters of the MBX stock), and in the brecciated MBX stock margin above the Rainbow Dike where they may reopen earlier E2 quartz veins. They are also preserved where sodic-calcic alteration overprints the potassic zone between the Rainbow Dike and Lower Trachyte. Veins are on the order of 2-3 mm wide, commonly with magnetite halos. In thin section, subhedral pyrite grains (0.5 mm) are entrained in anhedral chalcopyrite (Fig. 27B). The sulfide minerals appear to be in textural equilibrium, and precipitated from the same fluid. 5.7.3 Magnetite ± pyrite, quartz, chalcopyrite, calcite veins (E5-stage) Magnetite veins are typically 3 mm wide with carbonate-chamosite halos (>2 mm) that overprint secondary biotite and replace actinolitized phenocrysts. Vein walls can be irregular. Anastomosing E5 veins proximal to a fault in the footwall of the Rainbow Dike are suggestive of in situ brecciation of the host rock, or stockwork (Fig. 27C). Sheeted magnetite veins at the footwall contact of the Upper Trachyte are similar in appearance. 5.7.4 Replacement mineralization Within biotite-altered host rock, clinopyroxene phenocrysts are pseudomorphed by combinations of chalcopyrite-pyrite-magnetite-biotite-K-feldspar-Na-feldspar-carbonatequartz-apatite (Fig. 27G-H). Rims of Na-feldspar are concentrated around sulfidized phenocrysts, suggesting a sodic component to the clinopyroxene (Fig. 26D). Chalcopyrite forms inclusions in pyrite, and rims to pyrite grains. Trace monazite, observed on SEM, can also adhere to pyrite. In the surrounding groundmass, disseminated chalcopyrite-magnetite forms weak halos to pyrite mineralization.  62  63  Figure 27: Previous page. Early to transitional-stage veins and replacement sulfide minerals. A) Chalcopyrite (E3) vein with biotite halo (DDH 90-675 at 254 m). B) Chalcopyrite-pyrite (E4) vein with pyrite grains entrained in chalcopyrite (DDH 90-628 at 38 m). C) Magnetite-chalcopyrite ± pyrite (E5) veins in dendritic array with carbonate-chamosite halos (DDH 90-639 at 116.2 m). D) Pyrite-magnetite ± chalcopyrite (T1) veins (DDH 90-639 at 50.6 m). E) T1 vein in thin section showing magnetite surrounding coarse pyrite and interstitial carbonate replacing trace chalcopyrite (DDH 90-639 at 50.6 m). F) T1 type vein cutting monzonite in fault zone between the MBX stock and biotite hornfels (DDH 90-616 at 194.5 m). G-H) Chalcopyrite-pyrite-magnetite ± enargite replacement of clinopyroxene phenocrysts (DDH 90-639 at 34.5 and 50.6 m). Py =pyrite, cpy = chalcopyrite, en = enargite, mt = magnetite, bt = biotite, chl = chlorite, cbt = carbonate, cc = calcite.  5.8 CU-AU GRADE (potassic- and inner calc-potassic shell) Average Cu grade in the potassic shell increases from ~0.16 wt% at depth in the explored portion of the MBX zone (DDH 90-616) to ~0.32 wt% surrounding the Rainbow Dike (DDH 90-639). Similarly, average Au grade increases from 0.20 to 0.71 g/t. The average Cu/Au ratio decreases from ~0.87 to 0.60 wt%/g·t-1 (weight percent divided by grams per tonne) indicating a lateral and vertical increase in Au relative to Cu (Fig. 21, Table 3). In DDH 90-628, Cu and Au grade is elevated within the Lower Trachyte to peak average values of ~0.50 wt% Cu and ~0.88 g/t Au, representing the bestcombined ore grade of the deposit. The Cu/Au ratio is subequal at 0.83 wt%/g·t-1, similar to mineralization at depth below the Lower Trachyte. In the potassically-altered DWBX zone, average Cu and Au grade increases toward the Harris Fault, along the hanging-wall of the downdropped monzonite, from ~0.22 wt% Cu and ~0.32 g/t Au to ~0.58 wt% Cu and ~0.39 g/t Au. Additionally, the average Cu/Au ratio increases from ~1.21 to 1.56 wt%/g·t-1. This is significant because it indicates either: 1) that the center of Cu mineralization (in the direction of increasing Cu/Au ratio) was located somewhere in the hanging-wall of the pre-faulted DWBX zone; 2) that the monzonite body in the DWBX zone is unlikely to be a deeper segment of the MBX stock along which ore fluids were channelled from depth.  64  In the K-feldspar altered WBX zone, the deepest portion of the MBX stock, the average grade is ~0.33 wt% Cu and ~0.17 g/t Au. The average Cu/Au ratio is the highest in the deposit at ~2.37 wt%/g·t-1. Accordingly, the WBX zone is associated with the most primitive ore fluid. 5.9 66 ZONE (potassic- and inner calc-potassic shell) Potassic alteration continues into the 66 zone, where cloudy K-feldspar alteration is coincident  with  (and  overprints)  pervasive  biotite  ±  magnetite  alteration  of  trachyandesite (Fig. 28A). Within the intermittently faulted Upper Trachyte (~30 meters southeast of Rainbow Fault), up to 10 modal % chalcopyrite accompanies strong K-feldspar alteration. Chalcopyrite fills vesicular cavities, and occurs with biotite ± pyrite-magnetite-dolomite in microfractures at 30º to the horizontal axis of drillcore. It also occurs in similarly oriented layering in fine-grained recrystallized lenses. The same assemblage has replaced mafic phenocrysts. Magnetite alteration intensifies around the Upper Trachyte, particularly along the lower contact where sheeted magnetite veins are concentrated (Fig. 14C-D). Farther to the southeast, in DDH 91-815, the unit becomes a matrix-supported, magnetite-cemented breccia with milled trachyte pebbles and granules (clasts rotated and displaced) ranging in size from 5 mm to 20 µm, that have been dolomite-altered (Fig. 28C). The matrix contains 3-10% disseminated chalcopyrite and pyrite. The hanging-wall contact of the Rainbow Fault is comprised of a monomict breccia of cobble-sized trachyandesite(?) clasts in a sand-sized matrix (Fig. 26D). Cloudy grey Kfeldspar constitutes >70 modal % of textural-destructive potassically-altered clasts. Disseminated biotite and shreddy biotite clots constitute less than 3% of the rocks. According to the lithology provided by Placer Dome Inc., this unit (identified as a trachytic bedded tuff) follows the hanging-wall contact of the Rainbow Fault until reaching the Lower Monzonite dike in the deepest extent of the 66 zone (Fig. 4-5). 65  66  Figure 28: Intermediate stage potassic alteration in the 66 zone. A) Auriferous K-feldspar-pyrite overprinting biotite-altered trachyandesite. Footwall of the Upper Trachyte (DDH 91-815 at 103 m). B) Microphotograph of the Upper Trachyte with biotite-filled microfracture network overprinted by pyrite-epidote-dolomite clots (DDH 91-815 at 86). C) Magnetite-cemented milled breccia with Kfeldspar-dolomite altered trachytic clasts. Southeast terminus of Upper Trachyte (DDH 91-815 at 81.1 m). D) Intensely K-feldspar altered clasts in breccia at the upper contact of Rainbow Fault (bedded trachyte unit - Placer Dome Inc.) with clotted biotite replaced by sulfide. Epidote-albitechlorite alteration is present between clasts (DDH 90-643 at 102.5). E) Xenolithic monzonite with  67  trachyte xenolith (DDH 91-815 at 183 m).  F) Thin section image of xenolithic monzonite.  Oligoclase-andesine phenocrysts in K-feldspar groundmass. Mafics minerals are replaced by pyrite and minor carbonate. Ksp = K-feldspar, bt = biotite, chl = chlorite, dol = dolomite, ep = epidote, ab = albite, cpy = chalcopyrite, py = pyrite, trach = trachyte, monz = monzonite.  Because of the limited intersection of the “trachytic bedded tuff” unit with the hingedcross section of the present study, the rock appears to have a fault-parallel orientation along the hanging-wall of the Rainbow Fault suggestive of a tectonic breccia.  If  potassic alteration were consistent along the length of the unit (as implied by historic drill-logs, and considering the return of biotite-magnetite-chalcopyrite alterationmineralization in the hanging-wall) it would imply that the Rainbow Fault was synmineral, and potentially reactivated post-mineral. The fault zone itself is comprised of chlorite-calcite gauge, as well as calcareous clays with pyrite-mineralized residual clasts (formerly clinopyroxene-phyric) that have been potassically altered and hydrolyzed. However, the compiled Placer Dome drill maps indicate the unit is a stratigraphic horizon, similar to the Lower Trachyte of the MBX zone, in which case K-feldspar alteration is stratigraphically controlled, not tectonic. Farther to the southeast (in DDH 91-815), a ~30 m thick horizon of brecciated monzonite in monzonitic cement occurs 30 m above the Lower Monzonite Dike (Appendix A). The unit also contains pebble-sized trachyte xenoliths and is moderately-to-strongly altered to K-feldspar (Fig. 28E-F). The rock is comprised of over 60% K-feldspar, with 5-7 modal % pyrite-chalcopyrite, and variable carbonate (3-10%). In DDH 90-650, this horizon continues as an “altered latite debris flow” according to historic drill-logs, but was not available on-site for confirmation. Interestingly, the monzonitic unit has magnetite-rich porphyritic diorite dikes situated above and below, similar to the Rainbow Dike in DDH 90-652, and is of equivalent thickness as the Rainbow Dike. 5.10 CU-AU GRADE – 66 ZONE (potassic- and inner calc-potassic shell) Average grade in the potassically altered Upper Trachyte is ~0.24 wt% Cu, and ~0.40 g/t Au (DDH 90-643), which is slightly lower than the grade surrounding the Rainbow 68  Dike in the MBX zone. The average Cu/Au ratio is sub-equal at 0.81 wt%/g·t-1, which is similar to mineralization associated with the Lower Trachyte in the MBX zone (Fig. 21, Table 3). Average Cu grade decreases along strike of the Upper Trachyte body to 0.13 wt% in DDH 91-815. Alternately, Au grade increases to an average value of ~0.85 g/t. The average Cu/Au ratio sharply decreases to 0.37 wt%/g·t-1, reflecting the antithetic behaviour of the metals. 5.11 INTERMEDIATE STAGE SULFIDE VEINS (potassic- and inner calc-potassic shell) As early stage veins evolved, they become increasingly pyritic, magnetite-poor, and increased in size, abundance and density (Fig. 27D-F).  This transition marks the  beginning of the intermediate vein stage. 5.11.1 Pyrite-magnetite veins (T1-stage) Coarse-grained pyrite-magnetite veins are <20 mm wide with undulating, veins walls that can have minor apophysial offshoots.  Pyrite grains (<5 mm) are subhedral to  euhedral and cemented by interstitial magnetite and calcite. In thin section, trace chalcopyrite is present with interstitial calcite, but appears to be in disequilibrium. Chalcopyrite is also present along pyrite grain edges. Diffuse carbonate halos (~1 cm) overprint biotite alteration and actinolitized phenocrysts in the host rock. In the MBX zone, T1-stage veins are present in the hanging-wall of the Rainbow Dike, and within sodic-calcic alteration between the Rainbow Dike and Lower Trachyte. In the deepest sampled section of the MBX zone (DDH 90-616), a T1 vein, together with sheeted magnetite (E5) veins, is present in the faulted contact between the MBX stock and biotite hornfels.  69  5.12 NA-FELDSPAR-ACTINOLITE-EPIDOTE (sodic-calcic / outer calc-potassic shell) The sodic-calcic shell at the Mt. Milligan Main deposit is similar in mineralogy and zonal relationship as the “outer calc-potassic” shell of Holliday and Cooke (2007). Sodiccalcic alteration in the MBX zone increases inward and downward towards the MBX stock and marks the perimeter of elevated Cu grade (Fig. 21, 23, 29). The sodic-calcic assemblage is characterized by Na-feldspar replacement of K-feldspar in basaltictrachyandesite groundmass, and actinolite/uralite replacement of clinopyroxene phenocrysts (Fig. 30). Accessory alteration minerals include pyrite, epidote, chlorite, calcite, apatite, titanite, riebeckite, and quartz. Groundmass K-feldspar and Na-feldspar are commonly in gradational contact. Only the subsequent replacement of Na-feldspar by epidote indicates preceding Na-feldspar replacement of K-feldspar, and not the opposite. Actinolitized (uralitized) clinopyroxenes with thin Na-feldspar halos are common to both the inner calc-potassic, and sodic-calcic zones. In basaltic-trachyandesite overlying the Lower Trachyte (DDH 90-639), there is extensive actinolite/uralite replacement of clinopyroxene, resulting in pale green pleochroic, sheaf-textured phenocrysts, with low to second order birefringence. These commonly disintegrate into randomly oriented actinolite needles in the altered groundmass. Actinolite ± chlorite-carbonate filled fractures (T2-stage veinlets) are observed branching away from clots of pyrite-epidote-Na-feldspar. Such veins are thought to control the distribution of sodic-calcic alteration at temperatures of ~400450ºC (Seedorff et al., 2005). The feathery grey K-feldspar groundmass is pervasively pseudomorphed by Na-feldspar. Where albitization is strongest, the primary “bow-tie” spherulitic texture (McPhie et al., 1993) of the groundmass is enhanced. Accessory epidote appears to have nucleated on the edges of calcic phases such as clinopyroxene and Ca-plagioclase, but commonly replaces Na-plagioclase in the groundmass. Partial magnetite destruction is associated with the sodic-calcic zone, which is generally less magnetic than the inner calc-potassic zone.  70  Figure 29: Sodic-calcic (outer calc-potassic) and propylitic alteration stages. Each frame shows the compiled alteration shell for the indicated mineral and its distribution across the hinged cross-section of the present study.  At the upper margin of the intensely altered Lower Trachyte (DDH 90-628), XRD analysis indicates the assemblage: albite-quartz-ankerite-dolomite-biotite-clinochloremuscovite. Na-feldspar has replaced the K-feldspar comprising fine granular lenses. A late carbonate-phyllic overprint is also evident, and it is unclear if Na-feldspar is related  71  to an earlier chalcopyrite-bearing sodic-calcic assemblage, or the later chalcopyritedestructive carbonate-phyllic stage.  Figure 30: Sodic-calcic (outer calc-potassic) alteration. A) Selective-pervasive albitization of basaltic-trachyandesite and weak epidote after Na-plagioclase (DDH 90-639 at 135 m). B) Bow-tie  72  texture of Na-plagioclase groundmass with scattered actinolite needles (DDH 90-639 at 166 m). C) Disintegration of actinolitized phenocrysts within albitized groundmass (DDH 90-639 at 127.6 m). D) Reaction front between Na-plagioclase and K-feldspar alteration (DDH 90-639 at 193.8 m). E) Pyrite-epidote-carbonate replacing Na-plagioclase groundmass (DDH 90-639 at 193.8 m). F) Pyritechalcopyrite T3 vein with epidote selvage and Na-plagioclase halo overprinting biotite along coarse layering (flow-banding?) at 45º to the horizontal axis of drillcore (DDH 90-639 at 182 m). Ksp = K-feldspar, bt = biotite, ab = Na-plagioclase, act = actinolite, ep = epidote, cbt = carbonate, py = pyrite.  5.13 SULFIDE MINERALIZATION (sodic-calcic / outer calc-potassic shell) Pyrite veins and veinlets in the groundmass of basaltic-trachyandesite dominate the sodic-calcic assemblage. Pyrite also selectively replaces actinolitized clinopyroxene phenocrysts. Epidote replacement of Na-plagioclase is closely associated with pyrite in clots, clusters and veins. Concentration of Na-feldspar and pyrite is highest in the matrix of volcanicconglomerate, in the partings of coarse layering or flow banding, within in situ breccia fractures, in groundmass-hosted microfractures, and along phenocryst boundaries. The development of the epidote-pyrite within the same permeabilities indicates the epidotepyrite association is retrograde to Na-feldspar-pyrite. 5.13.1 T3-stage veins Pyrite veins (T3-stage) have a selvage of highly irregular chalcopyrite ± epidote (~1 mm), an inner halo of albitization (10-20 mm) with spotty epidote, and an outer Kfeldspar halo. Spatially, the zoned halo sequence is typically: pyrite ± chalcopyrite (vein) → chalcopyrite ± epidote (selvage) → Na-feldspar-actinolite ± epidote-chlorite-calcite → K-feldspar → biotite-K-feldspar (wall rock) In the T3 veins, epidote grows along the vein walls and traps chalcopyrite between grains, resulting in the serrated texture of chalcopyrite commonly seen in vein selvages (Fig. 31A). This texture is similar to the “cusp and carie” replacement texture described 73  Figure 31: Mineralization and replacement textures in sodic-calcic and inner-propylitic alteration shells. A) T3-stage pyrite vein with chalcopyrite-epidote selvage, inner Na-feldspar halo, and outer K-feldspar halo (DDH 90-639 at 129.3 m). B) T3-stage vein with Au-bearing chalcopyrite interstitial to pyrite (DDH 90-639 at 182.1 m).  C) Au-bearing chalcopyrite inclusion in pyrite.  74  D) Gold accreted to edge of pyrite grain within epidote selvage. E) Atoll texture. Pyrite-epidote cluster with pyrite replacing chalcopyrite (DDH 90-639 at 150.7 m). F) Cusp and carie texture. Chalcopyrite trapped within, or being replaced by epidote alteration in selvage of P1 vein (DDH 90641 at 112.2 m). Ab = albite, ep = epidote, act = actinolite, cpy = chalcopyrite, py = pyrite, Au = gold.  by Guilbert and Park (1986), suggesting epidote replacement of chalcopyrite. Accompanying “atoll” replacement textures in pyrite indicate replacement of chalcopyrite by pyrite. Trace chalcopyrite resides in albitized enclaves within selvages, and adheres to actinolitized phenocrysts in the vein halo. It appears that Au and chalcopyrite inclusions in pyrite result from pyritization of Aubearing chalcopyrite-magnetite from the underlying calc-potassic assemblage. This is based on petrographic observations including replacement textures, the presence of Au-bearing chalcopyrite interstitial to pyrite in T3 veins (~40 µm Au grain observed on SEM), and Au-bearing chalcopyrite and magnetite inclusions in pyrite grains (Fig. 31). The partial replacement of chalcopyrite by pyrite within the sodic-calcic zone represents the inner front of pyritization more common to the inner- and outer-propylitic assemblages, where chalcopyrite does not occur as veins, but is preserved in trace amounts between pyrite grains and as inclusions in pyrite. 5.14 CU-AU GRADE (sodic-calcic / outer calc-potassic shell) Average Cu grade remains at ~0.38 wt% upward and outward in the sodic-calcic shell, whereas average Au grade increases from ~0.58 to 0.84 g/t, and the average Cu/Au ratio decreases from ~1.04 to 0.77 wt%/g·t-1 (Fig. 21, Table 3). Ore grade within the upper sodic-calcic shell is similar to that surrounding the Rainbow Dike in the calcpotassic zone, except slightly elevated in both Cu and Au (and Cu/Au ratio). Accordingly, the fringe of sodic-calcic alteration in the MBX zone has close to the highest grade of the deposit. It is comparable to Cu-Au grade in the Lower Trachyte (DDH 90-628), which may be a deeper continuation of the same assemblage, but drillcore was unavailable for confirmation. 75  5.15 EPIDOTE-NA-FELDSPAR-ACTINOLITE (inner-propylitic shell) In the MBX zone, the innermost propylitic alteration is situated in the footwall of the Rainbow Dike ~150 m from the margin of the MBX stock. Distinct pale green veins (P1stage peripheral veins) and alteration bands composed of a fine-grained mixture of epidote-Na-feldspar characterize the inner-propylitic assemblage (Fig. 32). Accessory minerals are K-feldspar-biotite-ferroactinolite-chamosite/clinochlore and calcite. The epidote-Na-feldspar mixture replaces biotite-altered groundmass resulting in coarse pseudo-breccia textures. The assemblage is typically associated with pyrite (or a central pyrite veinlet) and does not co-exist with magnetite. Calcite mantles epidote- and actinolite-altered phenocrysts. Disseminated titanite and apatite microcrysts are observable with SEM. Titanite is a byproduct of chloritization of biotite (Eggleton and Banfield, 1985).  Figure 32: Inner-propylitic alteration. A) Microcrystalline epidote-Na-plagioclase-actinolite-calcite vein (P1-stage) with associated pyrite, and cloudy K-feldspar halo (DDH 90-641 at 112.2). B) In thin section the P1 vein is composed of actinolite needles, fine-grained epidote, and Naplagioclase after K-feldspar. SEM analysis indicates the prismatic blue mineral in the upper right  76  corner is epidote, although it resembles zoisite (DDH 90-641 at 112.2).  C) Very fine grained  epidote-Na-feldspar alteration with coarse pyrite and chlorite halo after biotite (DDH 90-641 at 132.1 m). D) Pyrite-rich P1-stage alteration band overprinting chloritized trachyandesite (DDH 90815 at 127 m).  In thin-section, the epidote-Na-feldspar-ferroactinolite alteration is brown-orange pleochroic and very fine-grained (Fig. 32B). Clinopyroxenes are partially to completely replaced by actinolite ± epidote-pyrite-calcite-chlorite. Pyrite appears to have nucleated on phenocrysts and extended into the groundmass within microfractures, forming epidote-rimmed pyrite aggregates and clots, which can be coarse grained (~10 mm). As in sodic-calcic alteration, the zoning observed around coarse-grained T3-stage pyrite veins and clots is: pyrite (vein or clot) → epidote ± Na-feldspar-actinolite-chlorite → Na-feldspar-calcite ± epidote → K-feldspar → biotite-K-feldspar (wall rock). 5.16 SULFIDE MINERALIZATION (inner-propylitic shell) The inner-propylitic assemblage is dominated by pyrite (~98% of total sulfide) occurring as veins, clots, and pseudomorphs. Trace chalcopyrite is present as inclusions in pyrite, irregular rims to pyrite grains, and as fine disseminations surrounding pyrite. Gold is observed within chalcopyrite trapped between pyrite grains, in chalcopyrite inclusions within pyrite, and adhering to epidote-rimmed pyrite grain edges (Fig. 31C-D). Pyrite disseminations and pseudomorphs are common to both the overprinted calc-potassic and superposed inner-propylitic assemblages with a modal abundance ranging from 2% to 5%. 5.17 CU-AU GRADE (inner-propylitic shell) Average Cu grade sharply plummets to ~0.09 wt% outside the sodic-calcic zone, while the average Au grade decreases modestly to ~0.67 g/t (Fig. 21, Table 3). The average Cu/Au ratio decreases to 0.19 wt%/g·t-1 reflecting the disproportionate loss of Cu with respect to Au.  77  Figure 33: Carbonate-phyllic and chlorite alteration. Each frame shows the compiled alteration shell for the indicated mineral and its distribution across the hinged cross-section of the present study. Note the concentration of late stage alteration along permeable pathways, faults, and lithological contacts.  5.18 DOLOMITE-ANKERITE-SERICITE-CHLORITE (carbonate-phyllic zone) In the 66 zone, a 60 m wide halo of intense carbonate-phyllic alteration occurs in the footwall of the Upper Trachyte magnetite-cemented breccia (~130 meters southeast of the Rainbow Fault), and widens to a ~90 m zone toward the Great Eastern Fault (Fig. 33). Trachyandesite is altered to a salmon pink-orange assemblage of K-feldspar-Na78  feldspar-muscovite-phengite-dolomite-ankerite-pyrite-hematite-biotite-chlorite, and clays illite-montmorillonite-halloysite (Fig. 34A-B). Trace Na-Fe amphibole (riebeckitearfvedsonite), Cu-sulphate, and Cu-sulfosalts are indicated by spectroscopy. Previous workers have documented a weak supergene influence (Sketchley et al., 1995). In thin-section, shreddy textured clinopyroxene and hornblende phenocrysts are replaced by dolomite-ankerite ± sericite-quartz (Fig. 34C). The groundmass is composed of selectively albitized needles of plagioclase and K-feldspar, and pervasive fine-grained sericite and hematite. Sericite forms halos to hairline fractures, and halos to pyrite pseudomorphs, which can also be rimmed by chlorite. Pyrite mineralization is linked to a microfracture network in the groundmass commonly filled by carbonate. Dolomite ± quartz occurs in veinlets (L2-stage) and stockwork. Late dendritic hematite stringers cut across pyrite veinlets and reopen carbonate veins. The fact that carbonate-phyllic alteration has such sharp boundaries with chloritized trachyandesite, and ranges from 60 to 135 m depth in DDH 90-650 and 45 to 140 m depth in DDH 90-648 implies structural and/or stratigraphic control. Intensity of alteration hinders protolith identification, but the altered zone appears to include both coherent and volcaniclastic units. Two fault zones at 45-56 m and 98-107 m depth in DDH 90-648 appear to have been the main focus for carbonate-phyllic fluids. These contain chlorite-sericite-clay gouge and carbonate-clay gouge, respectively. The shallower fault is situated at the upper contact of the carbonate-phyllic assemblage, where it contacts chlorite-altered trachyandesite. The deeper fault may be associated with the Upper Trachyte magnetite breccia in DDH 91-815, which is situated at the same elevation of 980-990 m ASL. In the MBX zone, similar carbonate-phyllic alteration is present at the upper margin of the Lower Trachyte, within ~90 m of the MBX stock in DDH 90-628, where it overprints the potassic ± sodic-calcic assemblage (Fig. 33, 13A). It also occurs along the upper margin of the Rainbow Dike, where it can be tracked from the MBX stock to the L1stage vein that cuts the dike in the 66 zone footwall (DDH 90-652; Fig. 34D). 79  Figure 34: Carbonate-phyllic alteration.  A) Salmon pink phengite-dolomite-illite ± brucite-  arfvedsonite alteration with pyrite pseudomorphs and dolomite veinlet stockwork (DDH 90-650 at 60.5 m). B) Stockwork of pyrite veinlets in carbonate-phyllic alteration (DDH 90-648 at 123 m). C) Dolomite pseudomorphs in trachytic-textured groundmass. Alteration is muscovite-illite-chloritebiotite. Pyrite also replaces phenocrysts and is finely disseminated in the groundmass (DDH 90648 at 59.9 m). D) Carbonate-phyllic L1-stage vein in distal Rainbow Dike with Au-bearing sulfide  80  (5.11 g/t Au). Alteration is dolomite-ankerite-sericite-illite (90-652 at 81.1 m). E) Carbonate-phyllic alteration at the upper margin of the Rainbow Dike, ~15 m from the MBX stock. Alteration is quartz-ankerite-adularia-muscovite-biotite-Na-feldspar-pyrite (90-628 at 31 m). F) Ankerite veinlet cutting pyrite veinlet in Lower Trachyte. Remobilized chalcopyrite is replaced by Cu-sulfosalt. Alteration is muscovite-ankerite-brucite-illite (90-628 at 201.1 m). Ank = ankerite, cp = chalcopyrite, py = pyrite, en = enargite (Cu-sulfosalt)  5.19 CHLORITE-HEMATITE-CARBONATE (chloritic zone) Trachyandesite surrounding carbonate-phyllic alteration in the 66 zone, has a cloudy green appearance due to ~20-50 modal % chlorite (Fig. 33, 35). In thin section, the groundmass consists of trachytic-textured K-feldspar needles with interstitial chlorite, carbonate, 3-5 modal % disseminated pyrite, and fine-grained hematite. Clinopyroxene and hornblende phenocrysts are partially altered to dolomite-chlorite-actinolite-pyrite and trace rutile. Spotty epidote occurs within Na-feldspar, particularly in rims to pyrite clots. Carbonate veinlets form localized stockwork. Dendritic hematite-carbonate veinlets (L6 vein-stage) with muscovite-illite halos are also present, particularly in the hanging-wall of the carbonate-phyllic zone.  Specularite and hematite stringers are  common. Trace cuprite is observed in thin section indicating a weak supergene influence. Chlorite alteration of trachyandesite in the 66 zone appears to be coincident with carbonate-phyllic alteration, and is overprinted by the inner-propylitic assemblage at depth (Fig. 35D). These relationships suggest: 1) retrograde chloritization of secondary biotite (with accompanying oxidation of magnetite to hematite) peripheral to carbonatephyllic alteration, or 2) a background regional chloritization of trachyandesite overprinted by carbonate-phyllic, inner- and outer-propylitic alteration. The former appears to be the case above the Upper Trachyte, where the chlorite-hematite assemblage is spatially consistent with biotite-magnetite alteration closer to the Rainbow Fault, whereas the latter describes the deeper 66 zone where clinopyroxene and andesine phenocrysts are relatively unaltered.  81  Figure 35: Chlorite alteration.  A) Strongly chloritized trachyandesite with dendritic L6-stage  hematite veinlets and dolomitized phenocrysts (DDH 91-815 at 31 m). B) Chloritized trachyandesite with L2-stage carbonate veinlets (DDH 90-650 at 198.9 m). C) Plane polar image showing chlorite-filled microfractures surrounding K-feldspar-sericite domains in trachyandesite groundmass (DDH 90-648 at 35.4 m). D) Na-feldspar-epidote alteration overprinting chloritized groundmass (no preceding potassic alteration stage) with pyrite replacement of clinopyroxene and hornblende phenocrysts (DDH 90-648 at 167 m).  5.20 SULFIDE MINERALIZATION (carbonate-phyllic zone) In the carbonate-phyllic altered 66 zone, pyrite ranges from 5 to 20 modal %, representing the highest abundance within the deposit-scale pyrite halo (Fig. 36). Trace Au-bearing chalcopyrite occurs between pyrite grains, and as inclusions in pyrite (Fig. 37). Hematite is similar, and forms inclusions with carbonate in pyrite. Tarnished inclusions of pyrite within pyrite indicate multiple stages of Fe-sulfide deposition. 82  83  Figure 36: Previous page. Chalcopyrite and pyrite modes estimated in the field. The chalcopyrite zone is associated with potassic and inner calc-potassic alteration within and adjacent the MBX stock, and in the Upper Trachyte. In the MBX zone, the chalcopyrite shell diminishes outboard of DDH 90-639 where pyrite intensifies in the sodic-calcic and inner-propylitic (90-641) assemblages. The deposit-scale pyrite halo is associated with sodic-calcic and inner-propylitic alteration in the MBX zone. In the 66 zone, it is associated with carbonate-phyllic and inner-propylitic alteration, and less so the calc-potassic, outer-propylitic, and chloritic assemblages.  Figure 37: Mineralization in carbonate-phyllic zone. A) Pyrite pseudomorph/clot illustrating the proportion of pyrite to chalcopyrite in the carbonate-phyllic assemblage. Chalcopyrite is trapped between pyrite gains (DDH 90-650 at 60.5 m). B-C) Gold and carbonate in chalcopyrite-mineralized interstices to pyrite grains (DDH 90-652 at 81.1 m).  D) Gold inclusion in pyrite adjacent to  chalcopyrite-carbonate inclusion (DDH 90-652 at 81.1 m). Cpy = chalcopyrite, py = pyrite, Au = gold, cbt = carbonate.  84  In the MBX zone, ankerite veins associated with the carbonate-phyllic overprint of the Lower Trachyte contain sulfosalt (enargite-luzonite or tetrahedrite-tennantite). The sulfosalt has replaced chalcopyrite remobilized from the underlying potassic assemblage (Fig. 34F). 5.21 CU-AU GRADE (carbonate-phyllic zone) Average grade in the carbonate-phyllic altered 66 zone is ~0.01 wt% Cu and ~0.92 g/t Au closest to the Upper Trachyte body in DDH 90-650 (Fig. 21, Table 3). Assay data indicates elevated Au in permeable horizons (conglomerate or breccia) extending from potassic (2.35 g/t Au avg.) to carbonate-phyllic alteration (2.67 g/t Au avg.) in the footwall of the Upper Trachyte (Appendix 1). The highest measured Au grade (22 g/t Au) occurs at the upper contact of a late-mineral porphyritic diorite dike (DDH 91-815 at 138 m), and elevated Au grade continues in the footwall of the same dike (2.45 g/t Au avg.). Farther to the southeast in DDH 90-648, average Au grade decreases to ~0.46 g/t, and the average Cu grade remains low at ~0.02 wt%. The high average Au grade in the carbonate-phyllic assemblage is comparable to that in the sodic-calcic shell and the potassically altered Lower Trachyte of the MBX zone, and is maximal for the deposit. Alternately, the decreasing grade away from the Upper Trachyte is similar to that of inner- and outer-propylitic alteration stages, except that Au is locally concentrated within non-coherent strata and faults, and along dike structures. 5.21.1 CARBONATE-PHYLLIC VEIN (L1-stage) Approaching the Rainbow Fault footwall contact in DDH 90-652, a 1.7 m wide vein of dolomite-ankerite-sericite-illite-pyrite cuts the lower margin of the Rainbow Dike (L1stage vein). Coarse euhedral pyrite clusters (<80 mm grains) constitute ~30 modal %, with ~2% accessory chalcopyrite (Fig. 34D). A 3 mm wide chalcopyrite veinlet cuts across pyrite. Fire assay data indicate a sudden jump in grade to 0.29 wt% Cu and 5.11 g/t Au, similar to elevated Au grade occurring along the footwall of the Upper Trachyte in the 66 zone. Coincidentally, this is also the section of the Rainbow Dike  85  where magnetite-rich late-mineral porphyritic diorite dikes intrude the upper and lower faulted contacts (Appendix 1). 5.22 EPIDOTE-CHLORITE (outer-propylitic shell) The outer-propylitic stage is distinguished by pervasive epidote alteration of the host rock groundmass or matrix, with accessory chlorite (Fig. 38). Epidote is present as: 1) very fine-grained veins and halos to microfractures; 2) coarser, pervasive interlocking grains (<0.4 mm) replacing trachyandesite groundmass; 3) halos to P2-stage pyrite veins (Fig. 38D); and 4) a mafic phenocryst replacement phase, intergrown with pyrite. The P2-stage veins are <50 mm wide straight-walled veins composed of coarse irregular pyrite interspersed with epidote ± chlorite, hematite, and trace montmorillonite and/or illite. Vein halos are epidote-chlorite-calcite (<40 mm) in the DWBX and 66 zones, and Na-feldspar-chlorite-calcite in the MBX zone. Other minerals of the sodic-calcic and inner-propylitic suite (Na-feldspar-calciteactinolite) are present, but are subordinate to epidote. Sodic-plagioclase occurs as rims and halos to epidote patches. Biotite alteration is retrograded to chlorite, such as the ~20 m wide chloritized halo surrounding the potassically-altered Lower Trachyte in DDH 90-652. Pyrite (<5 modal %) is disseminated throughout the rock with fine-grained (~20 µm) hematite. It partially or completely replaces mafic phenocrysts and forms in groundmass-hosted micro-fractures, typically intergrown with epidote. In the MBX zone, outer-propylitic alteration occurs along permeable horizons (apparentbreccia and/or stratiform fault zones) above and below the Rainbow Dike. In the dike footwall (DDH 90-639), brecciated host rock surrounding a stratiform fault at 120 m depth (Appendix 1) contains P2-stage veins (Fig. 38E). Chlorite alteration, as illustrated in the compiled chlorite alteration shell (Fig. 33), is also focussed along this horizon.  86  Figure 38: Outer-propylitic alteration. A) Pervasive epidote-pyrite replacing albitized groundmass in trachyandesite (DWBX zone; DDH 90-598 at 95.0 m). B) Thin section image of epidote-pyrite overprinting actinolitized phenocrysts and Na-plagioclase altered groundmass (DWBX zone; DDH 90-598 at 166.5). C) Epidote-calcite-chlorite-quartz-pyrite alteration of trachyandesite with clotting of chloritized mafics (66 Zone; DDH 90-652 at 139.5 m). D) P2-stage vein with hematite centerline and halo of pervasive epidote alteration rimmed by Na-feldspar (66 zone; DDH 91-815 at 263 m).  87  E) Pyrite-epidote P2-stage vein with inner Na-feldspar halo and diffuse chlorite-carbonate halo. Late stage vein in fault zone cutting the Cu-Au ore shell (MBX zone; DDH 90-639 at 117.2 m). Py = pyrite, hem = hematite, ep = epidote, ab = albite, act = actinolite, chl = chlorite, qz = quartz, cc = calcite, cbt = carbonate.  The outer-propylitic assemblage is most prevalent in the DWBX zone where it abuts biotite hornfels within ~60 m of the monzonite stock. It extends outward to the DWBX fault separating the upper and lower DWBX zone. Epidote-chlorite alteration exploited the permeable matrix of conglomeritic host rocks that dominate the DWBX zone stratigraphy. It subsequently replaced secondary Na-plagioclase in pebble- to cobblesized fragments, and diffused into biotite-chlorite altered coherent units. 5.23 SULFIDE MINERALIZATION (outer-propylitic shell) Pyrite is the dominant sulfide mineral in the outer-propylitic shell. Similar to the innerpropylitic assemblage, trace chalcopyrite occurs as inclusions in pyrite, irregular rims on pyrite grains, and is disseminated in epidote-calcite halos to pyrite. Trace Au-Ag telluride, identified with LA-ICP-MS time-resolved spectra (see Ch. 6.8.6), also occurs on pyrite grain edges. 5.24 CU-AU GRADE (outer-propylitic shell) Average grade is ~0.03 wt% Cu and ~0.33 g/t Au, which is ~50% lower than Au grade associated with the inner-propylitic assemblage (Fig. 21, Table 3). The average Cu/Au ratio is ~0.09 wt%/g·t-1 which is also ~50% lower. 5.25 OTHER INTERMEDIATE TO LATE STAGE VEINS 5.25.1 Pyrite-carbonate veins (L3-stage) Pyrite-carbonate veins (L3-stage) occur throughout the DWBX zone but increase in density within biotite hornfels (DDH 90-598) and chlorite-altered host rock (DDH 90-675) near the Harris Fault. The L3 veins are similar to B-type veins of Gustafson and Hunt (1975) in being continuous planar structures with parallel walls and internal banding. They can also be mildly undulating. Pyrite is granular, but can be ribboned or banded, suggesting sequential veinlets along a single fissure. 88  In the DWBX zone, L3 veins are chalcopyrite bearing within 15 m of the stock. Beyond the biotite hornfels, L3 veins have weak epidote selvages and halos. Similar veins occur in the shallow MBX stock, and in the 66 zone associated with weak epidote alteration of chloritized trachyandesite (Fig. 39A-B, Table 2). B-veins are typically associated with calcic alteration (Seedorff, 2005). 5.25.2 Pyrite-calcite veins (L4-stage) Discrete, coarse granular pyrite veins with interstitial calcite (L4-stage) continue to the latest stages of mineralization and are observed cutting volcanic-conglomerate in the propylitized DWBX zone (Fig. 39C). Some veins are epidote-mantled and may be genetically related to the L3 veins. 5.25.3 Calcite-chlorite ± tourmaline-quartz veins (L5 stage) Calcite-chlorite ± tourmaline-quartz veins (L5-stage) occur in the MBX stock and 66 zone. They are parallel walled, continuous, and typically ~10 mm in width. Calcite is commonly vuggy, and can be bladed, signifying boiling of fluids (Simmons and Cristenson, 1994). Chloritic segments of the vein occur separately from calcite (Fig. 39D-E). Quartz crystals align the inner wall of the vein in both calcite and chlorite segments. Tourmaline forms clots of acicular crystals intergrown with quartz, and may be weakly disseminated in calcite. In the carbonate-phyllic altered 66 zone (DDH 90648), a 20 mm wide complex vein, comprising L4- and L5-stage veins is suggestive of incipient epithermal processes (Fig. 39E). 5.25.4 Hematite-dolomite veins (L6-stage) Hematite-dolomite  veins  (L6-stage)  with  muscovite-illite  halos  cut  chloritized  trachyandesite in the hanging-wall of carbonate-phyllic alteration in the 66 zone. (Fig. 39F).  89  Figure 39: Other intermediate to late-stage veins.  A) Ribboned pyrite-calcite L3 vein cutting  biotite hornfels (DWBX zone; DDH 90-598 at 272 m). B) Ribboned pyrite-calcite L3 vein with 20 mm weakly disseminated epidote halo (66 Zone; DDH 90-648 at 173.5 m). C) Coarse-grained pyrite L4 vein with interstitial carbonate and K-feldspar-sericite halo (DDH 90-598 at 175.7 m). D) Chlorite-calcite-quartz L5 vein cutting K-feldspar altered monzonite (MBX stock; DDH 90-667 at 65 m). E) L3- and L5-stage veins exploiting a common fracture in the carbonate-phyllic zone (DDH 90-648 at 130.0). F) Hematite-carbonate L6 vein with muscovite-illite halo in strongly chloritized trachyandesite. Hanging-wall of the carbonate-phyllic zone (DDH 90-648 at 58.5).  90  5.26 ALTERATION AND MINERALIZATION PARAGENESIS – summary The intimate association of metal- and alteration zonation with ore grade at the Mt. Milligan Main deposit is discernable when comparing the compiled alteration shells to fire-assay data provided by Placer Dome Inc. (Fig. 40). The calc-potassic shell is most extensive in the MBX zone, reaching ~260 m from the MBX stock, but is largely overprinted by later alteration stages beyond ~130 m. Copper and Au have a greater than 1:1 relationship [wt%/(g/t)] in the deepest levels (lower DWBX zone, WBX zone), a sub-equal relationship at intermediate levels (deep MBX zone, Lower Trachyte), and are less than unity surrounding the Rainbow Dike. The Cu/Au ratio generally decreases upward and outward with increasing Cu and Au grade, although a ~75 m wide interval of highest-grade mineralization occurs within the Lower Trachyte in DDH 90-628. Potassic alteration is also present in the downdropped 66 zone, where it is centered on the fault-bound Upper Trachyte. Copper and Au grade is slightly lower than the upper MBX zone. However, Au grade increases by 100% where the potassic assemblage terminates in a magnetite-cemented milled breccia (DDH 91-815), and carbonate-phyllic alteration intensifies. The sodic-calcic shell represents an intermediate zone between calc-potassic and propylitic-stage assemblages. It is strongest along the upper margin of the Lower Trachyte, but extends below the Lower Trachyte as close as ~50 m to the MBX stock. From this deepest extent upward to the Rainbow Dike footwall (DDH 90-639), the sodiccalcic shell becomes ~70% richer in Au grade. This represents the best combined CuAu grade of the deposit apart from that within the Lower Trachyte (DDH 90-628), which may be a deeper portion of the same assemblage. Inner-propylitic alteration (Na-feldspar-epidote-pyrite) is located outboard of the sodiccalcic shell in the MBX zone (at ~150 m from the MBX stock), and also overprints the calc-potassic assemblage. Chalcopyrite-magnetite is incompatible with epidote 91  alteration, which is reflected in the low Cu grade, whereas Au grade remains moderately high.  Figure 40: Copper-Au grade juxtaposed with alteration shells along the hinged cross-section. Alteration shells are as indicated in Figure 36. Copper mineralization drops off sharply beyond the sodic-calcic shell, but moderate Au grade continues into the outer-propylitic zone. Copper mineralization and (calc)-potassic alteration recurs in the 66 zone surrounding the Upper Trachyte. Gold grade is elevated at the southeastern tip of the Upper Trachyte (magnetitecemented breccia) and into the carbonate-phyllic zone. Copper-gold ratio is >1 in most of the MBX stock, and within the Lower Trachyte <50 m from the MBX stock margin. It is near unity in the calc-potassic and sodic-calcic shells, but decreases upward and outward. It is <1 in the inner-  92  and outer-propylitic, and carbonate-phyllic shells. Fire assay data was provided by Placer Dome Inc.  Late-stage carbonate-phyllic alteration in the 66 zone develops outward from potassic alteration centered on the Upper Trachyte, and also occurs in a 1.7 m wide vein at the lower margin of the Rainbow Dike (DDH 90-652). Gold grade reaches peak values where the carbonate-phyllic assemblage commences (~4-5 g/t). It decreases outward to modest levels (0.1-0.6 g/t), except along minor faults and dike contacts that remain at elevated grade (~1-3 g/t). Chalcopyrite occurs in trace amounts within the dominant pyrite phase, but is significant in that it hosts gold. In the MBX zone, carbonate-phyllic alteration overprints the upper margin of the Rainbow Dike outward of ~230 m from the stock. It follows the Lower Trachyte for at least 90 m, where Cu-sulfosalt replacement of chalcopyrite is observed in ankerite veins. Outer-propylitic alteration (epidote-chlorite-pyrite) is peripheral to all other alteration stages in the MBX and 66 zones, but also cuts across the earlier assemblages along permeable horizons. Much of the lower DWBX zone is overprinted by the outerpropylitic stage reflecting an abundance of permeable volcanic-conglomerate as host rock. Gold grade is moderate to weak, and Cu is insignificant. The deposit-wide pyrite halo (as described geophysically by Oldenberg et al., 1997) is associated with peripheral assemblages (sodic-calcic, inner- and outer-propylitic) and the carbonate-phyllic assemblage where pyrite abundance is typically 1-5 modal %. 5.27 DISCUSSION: DIAGNOSTIC ALTERATION ZONING AT THE MT. MILLIGAN MAIN DEPOSIT The alteration and metal zoning at the Mt. Milligan Main deposit can be divided into vertical and lateral components. Laterally, alteration progresses from (calc)-potassic to sodic-calcic to inner- and outer-propylitic assemblages, spanning ~350 m in the MBX zone. Vertically, alteration progresses from potassic to carbonate-phyllic assemblages, spanning ~300 m in the MBX and 66 zones.  93  5.27.1 Lateral zonation The lateral progression is best expressed in sodic-calcic and inner-propylitic alteration of basaltic-trachyandesite between the Lower Trachyte and Rainbow Dike in the MBX zone. These alteration stages are most diagnostic for determining paragenetic sequence because they represent an intermediate region between (calc)-potassic alteration derived from magmatic fluids, and outer-propylitic alteration signifying entrainment of thermally prograding external fluids of meteoric or formation origin (McMillan and Panteleyev, 1988; Seedorff, 2005). Alteration and sulfide mineralization in the sodic-calcic zone shares properties with both the potassic and outer-propylitic end members. The evolution of Cu-bearing sulfide veins, and the zoned halos of T3-stage veins, provides a macroscopic example of lateral alteration paragenesis on the deposit-scale. In hydrothermal systems characterized by Fe-S-O, veins become more magnetite-poor and pyritic with decreasing temperature, as fluids evolve to lower pH (Barnes and Czamanske, 1967) and higher ƒS2 (Meyer and Hemley, 1967). 5.27.1.1 E3- or E4-stage veins In the MBX zone, veins began as fractures in calc-potassic altered host rock that were mineralized by oxidized metalliferous fluids to form chalcopyrite ± pyrite veins (E3- or E4-stage) with magnetite halos, typical of the calc-potassic assemblage at temperatures of ~600-400ºC (Fig. 41A). The associated calc-potassic alteration leached Ca2+ from the host rock in reactions involving biotite replacement clinopyroxene and hornblende, such as:  8(Ca,Mg,Fe)2(Si,Al)2O6 + 4KCl + 4HCl + 2H2O augite  94  = 4KMgFe2AlSi3O10(OH)2 + 4CaCl2 + O2  (1)  biotite  4KCa2(Mg,Fe)5Si6Al2O22(OH)2 + 4FeOrock + 4KCl + 12HCl hornblende = 8KMgFe2AlSi3O10(OH)2 + 8CaCl2 + 2H2O + O2  (2)  biotite  These reactions would have increased the Ca2+ content of circulating fluids, and may have produced byproduct calcite if aqueous H2CO3 were present: CaCl2 + H2CO3 = 2HCl + CaCO3  (3)  5.27.1.2 Incipient T3-stage veins At the onset of sodic-calcic alteration (450-400ºC; Seedorff et al, 2005), sulfide veins became increasingly pyritic, commonly with a thin selvage of chalcopyrite (incipient T3stage; Fig. 41B), The veins also developed Na-feldspar halos reflecting Na+ saturation of the fluid, and outward diffusion from the vein. Sodic-feldspar replaced groundmass Kfeldspar, which can be present as larger halos to Na-feldspar. The relationship suggests a few scenarios: 1) a preceding K-feldspar halo with Na-feldspar superposed on it; 2) removal of K+ from potassically-altered groundmass and re-precipitation outboard of albitization, and 3) inward diffusion of FeO, MgO and H2O from the biotite-altered andesitic host: KMgFe2AlSi3O10(OH)2 = KAlSi3O8 + 2FeO + MgO + H2O biotite  (4)  K-feldspar  95  Figure 41: Transition from E4- to P2-stage veins through the intermediate T3 stage as an analogue for the transition from the calc-potassic to outer-propylitic assemblage through the intermediate sodic-calcic and inner-propylitic assemblages.  96  In the latter case, this diffusion may be reflected in the magnetite halos to E3 and E4 veins, and the increased Fe2+ content of T3 veins. Traces of chalcopyrite-magnetite are disseminated outside the vein within the selvage and albitized halo. Leaching of Ca2+ from host rock continued in the sodic-calcic zone, with actinolitization of clinopyroxene, and minor chloritization of actinolite: 4(Ca,Fe,Mg)2Si2O6 + 2HCl = Ca2(Mg,Fe)5Si8O22(OH)2 + CaCl2 augite  (5)  actinolite  Ca2(Mg,Fe)5Si8O22(OH)2 + FeOrock + 4HCl + H2O actinolite = (Fe,Mg)6Si4O10(OH)8 + 4SiO2 + 2CaCl2  (6)  chlorite  5.27.1.3 Developed T3-stage veins Around 300ºC, Ca2+ became insoluble and began to diffuse outward from the vein, reacting with Na-feldspar and Fe2+ to form a magnetite-destructive inner epidote halo: CaNaAl2Si6O16 + FeOrock + CaCl2 + H2O + O2 andesine = Ca2Fe3+Al2Si3O13(OH) + 3SiO2 + NaCl + HCl  (7)  epidote The veins grew wider and become more pyritic (T3-stage). Pyrite replaced chalcopyrite and magnetite within the epidote halo, resulting in “atoll” textures of relict chalcopyrite included in pyrite (Gustafson and Hunt, 1975). Epidote also appears to have replaced chalcopyrite, or displaced it to more stable enclaves in the groundmass. As the veins 97  matured to the P2 stage, the inner epidote halo replaced the outer Na-feldspar halo, and converted Fe-bearing minerals to pyrite, leaving only trace chalcopyrite trapped between or included within pyrite grains (Fig. 41C). 5.27.1.4 Applying the vein model to deposit-scale alteration zoning Considering the above interpretation of zoned halos to L3-stage veins, the lateral transition from calc-potassic to sodic-calcic, inner- and outer-propylitic shells in the MBX zone can be viewed as a spatial-temporal sequence of K+, Na+, and Ca+ saturation in the hydrothermal fluid as it moved away from the MBX stock in Fe-rich andesitic host rock, with decreasing temperature. Accordingly, the truncation of the Cu ore shell at ~130 m from the MBX stock (at the transition between sodic-calcic and inner-propylitic shells) may not simply reflect a sudden change in the intensive parameters of the hydrothermal fluid that caused insolubility of chloro-complexes below ~450ºC (Gammons and Jones, 1997). It could also indicate the front of inwardly migrating peripheral assemblages during thermal collapse of the system below 300ºC, analogous to replacement of the outer Na-feldspar halo by the inner epidote halo in L3-stage veins. In this scenario, the Cu-Au ore shell may have been more expansive at higher temperatures, within the pre-overprinted inner- and outer calc-potassic (sodic calcic) zones. Furthermore, the occurrence of moderate Au grade in the inner-propylitic shell (~0.5 g/t), and less so the outer-propylitic shell (~0.3 g/t) may not simply reflect Au precipitated from Au(HS)2- at lower temperatures (~350-300ºC) in retrograde chloritizing fluids (Gammons and Jones, 1997; Delong, 1995). It might also be related to leaching of Cu by deep peripheral fluids, as documented at the Ann Mason deposit (Dilles and Einaudi, 1992), and pyritization of Au-bearing chalcopyrite-magnetite from the underlying calc-potassic assemblage, preserving relict Au- and Au-bearing chalcopyrite inclusions. Subsequently, Au blebs located on the rims of epidote-mantled pyrite grains would have precipitated from bisulfide-complexes at lower temperature.  98  5.27.1.5 Alternate interpretation Alternatively, the inferred precipitation of chalcopyrite after pyrite in E2-stage veins (pyrite entrained in chalcopyrite) may be carried forward into later stage veins but with a greatly reduced Cu abundance. In this view, trace chalcopyrite rims and halos to pyrite, and pyrite grain interstices filled with Au-bearing chalcopyrite (as observed in the sodiccalcic and inner-propylitic zones) could represent the final dregs of sulfide mineralization after formation of pyrite. Accordingly, there would be no change in the chloro-complexing behaviour of Cu and Au from early to late alteration stages, only a change in the abundance of Cu, which is significantly reduced outside the magnetitebearing calc-potassic shell, probably due to temperature decrease through fluid mixing. In Part 2 of the thesis, Sr isotope analysis of actinolite, Na-feldspar, and epidote from the MBX zone explores the proportions of magmatic- to meteoric-derived fluids in the lateral alteration sequence to help determine the flow paths and mixing zone of these fluid end-members. 5.27.1.6 DWBX zone In the DWBX zone, the absence of a Cu-mineralized sodic-calcic shell and innerpropylitic shell suggests the outer-propylitic assemblage advanced farther toward the system center during thermal collapse than in the MBX zone. A ~60 m wide zone of volcanic-conglomerate with Na-feldspar rich andesite clasts in epidote-chlorite-pyrite altered andesitic matrix contacts biotite hornfels ~60 m outboard of the downdropped monzonite. Pyrite-calcite L3 veins span both zones suggesting pyrite-mineralizing fluids encroached upon disseminated chalcopyrite-magnetite within ~35 m of the stock. Alternatively, if the volcanic-conglomerate is in fact pseudo-breccia (see Ch. 3.6.2), it would indicate a deep, early sodic-calcic alteration stage preceeding calc-potassic and propylitic stages, with possible leaching of Cu from potassic protore and redeposition at shallower level ore zones (MBX zone), such has been interpreted for the Ann Mason porphyry Cu deposit, Nevada (Dilles and Einaudi, 1992).  99  5.27.2 Vertical zonation Potassic alteration within the Upper Trachyte and surrounding trachyandesite in the 66 zone is key to understanding alteration- and metal zoning in a vertical sense at the Mt. Milligan Main deposit. The potassic assemblage represents a sudden, incongruous jump up the thermal gradient to temperatures in the range of 650-400ºC where Kfeldspar, biotite and magnetite are thermodynamically stable (Burnham & Ohmoto, 1980) from temperatures below 300ºC in the footwall of the Rainbow Fault, where epidote is stable (Seedorff et. al., 2005: Fig. 36, 40). This alteration geometry implies the 66 zone is a downdropped segment from shallower depths of the paleohydrothermal system, and from closer proximity to the MBX stock. The amount of throw (vertical displacement) on the Rainbow fault is unknown, but is probably on the order of ~95 m based on differences in epidote chemistry in the footwall and hanging-wall (see Ch. 6.6.4.5, Fig. 61). Coincidentally, this is also the vertical difference between Rainbow Dike in DDH 90-652 and the 66 zone xenolithic brecciated monzonite body (DDH 91-815), which share a common relationship to porphyritic diorite dikes. Assuming a 95 m throw, the paleo-vertical distance from the Lower Trachyte in the MBX zone to chlorite-altered trachyandesite in the uppermost 66 zone is ~300 m. This represents the vertical extent of drillcore examined on the southeast side of the MBX stock, the heart of the Cu-Au resource at the Mt. Milligan deposit. 5.27.2.1 Potassic to carbonate-phyllic transition The vertical component of alteration zoning at the Mt. Milligan deposit is represented by the transition of potassic to carbonate-phyllic alteration immediately southeast of the Upper Trachyte in the hanging-wall block of the 66 zone. Within the potassically-altered Upper Trachyte (spanning over 100 m from the Rainbow Fault), dolomite has partially replaced mafic phenocrysts, and is present in the biotitesulfide filled micro-fracture network.  At the footwall contact of the Upper Trachyte,  carbonate cuts across and alongside magnetite veins, and reaches a modal abundance as high as 30%. Additionally, where the Upper Trachyte terminates in a matrix100  supported magnetite-cemented breccia in DDH 91-815, the milled trachyte clasts contain dolomite veinlets. These relationships indicate that precipitation of dolomite ± ankerite overlapped with potassic alteration, and is not simply a younger overprint. Texture-destructive carbonate-phyllic alteration of trachyandesite in the 66 zone forms a sideways funnel-shape that stems from the magnetite-cemented breccia at ~81 m depth (DDH 90-815), and expands laterally outward for at least 100 m between 60-135 m depth (DDH 90-650), and 45-140 m depth (DDH 90-648, Appendix 1). Faults at 45-56 m and 106-107 m depth (DDH 90-648) appear to be the main structural control. Pyrite δ34S values form a zoned halo around the deeper fault, providing geochemical evidence for fault control of oxidized H2S-bearing fluids (see Ch. 6.4.3.2). Carbonate replacement of clinopyroxene and hornblende phenocrysts in the 66 zone implies a CO2-rich fluid or vapour. The following are examples of replacement processes that may have occurred. Byproducts are muscovite and hematite: 12Ca(Fe,Mg)SiAlO6 + 4KCl + 24H2CO3 augite = 12Ca(Fe,Mg)(CO3)2 + 4KAl2(AlSi3O10)(OH)2 + 4HCl + 18H2O + 3O2 dolomite/ankerite  (8)  muscovite  12KCa2(Mg,Fe)5Si6Al2O22(OH)2 + 48H2CO3 + 4HCl + 6O2 hornblende = 24Ca(Fe,Mg)(CO3)2 + 8KAl2(AlSi3O10)(OH)2 + 18Fe2O3 + 48SiO2 + 4KCl + 54H2O (9) dolomite/ankerite  muscovite  hematite  101  In the MBX zone, the carbonate-phyllic altered Lower Trachyte contains ankerite veinlets with chalcopyrite replaced by Cu-sulfosalt. This implies the carbonate-phyllic fluid had elevated activity of As and Sb. dominant As  3+  The As(OH)3 complex is considered the  carrier in moderate- to high-temperature hydrothermal fluids with  densities higher than 5 g.cm-3 (Pokrovski et al., 2000). Sulfosalt replacement indicates a sulfidation reaction, and production of accessory Fe oxide:  2CuFeS2 + 2(As,Sb)(OH)3 + 4H2S + 2O2 = 2Cu(As,Sb)S4 + Fe2O3 + 7H2O chalcopyrite  sulfosalt  (10)  hematite  5.27.2.2 Phyllic alteration in porphyry-related systems For most of the past 30 years, phyllic alteration in Cu-porphyry systems has been thought to reflect ingress of convecting meteoric water, mixing with magmatic fluids, and a retrograde temperature regime at hydrostatic pressure (Gustafson and Hunt, 1975; McMillan and Panteleyev, 1988). This was based on stable isotope studies of North American calc-alkaline systems (Taylor, 1974), and on the presence of fluid inclusions with lower salinities than magmatic-derived hypersaline brines (Henley and McNabb, 1978). More recently, it has been shown that phyllic alteration in Cu-porphyry systems can be generated solely from magmatic fluids with declining K+/H+ and temperature between 600-300ºC (Seedorff et al., 2005; Harris and Golding, 2002), such that sericite becomes the most stable K-silicate. Additionally, it has been proposed that low salinity fluids associated with phyllic alteration may be exsolved from magma chambers at a late stage, after the main pulse of hypersaline brine and vapour has already escaped to the porphyry environment (Shinohara and Hedenquist, 1997). In Part 2 of the present thesis, the source of the apparent Au-mineralizing CO2-bearing fluids at Mt. Milligan is explored through C and O-isotope analysis of ankerite and dolomite from the carbonatephyllic assemblage. The results support derivation from a magmatic-hydrothermal fluid.  102  5.27.2.3 Phase separation in the 66 zone The spatial relationship between magnetite-cemented breccia and carbonate-phyllic alteration in the 66 zone suggests the possibility of phase separation in the magmatichyrothermal fluid. If phase separation occurred, the acidic components (H2S, HCl, SO2) would have partitioned into the vapour with CO2 and H2O, whereas the brine would have collected KCl, NaCl, HS- and most of the dissolved Au (Gammons and WilliamsJones, 1997). Partitioning of CO2 into vapour occurs at >350ºC at <5 km depth (Baker, T., 2002). The host trachyandesite would have immediately buffered the acidic, CO2bearing vapour, accounting for the intense texture-destructive alteration, dolomitization of mafic phenocrysts, and sericite-illite overprint in the ~100 m wide carbonate-phyllic corridor. Loss of acid components would have caused pH increase in the residual brine, and a potential shift from muscovite to adularia stability field (at pH > 6 under 400ºC; Craw, 1997). This pH increase may have triggered precipitation of Au above the boiling horizon (Hedenquist and Henley, 1985). Accordingly, the focus of boiling should be reflected in the location of elevated Au-grade, with the potential presence of adularia. In the 66 zone, this relationship occurs within the fault zone at 98-107 m depth (980-990 m ASL) extending southeast from the Upper Trachyte magnetite-breccia (Fig. 42). Elevated Au grade is indicated by fire assay data (>1g/t Au between 102 and 107 m in DDH 90-648), and XRD analysis indicates the presence of adularia at 99 m. Adularia is not present in deeper and shallower rocks of the carbonate-phyllic assemblage that were analyzed by XRD. Although the jump in Au grade, secondary mineral assemblage, character of alteration, and spatial relationships in the 66 zone suggest the possibility of phase separation, the only additional evidence is from an L5-stage carbonate-chlorite ± tourmaline vein with bladed calcite. Unfortunately, the carbonate-phyllic zone is not represented in the fluid inclusion study of the present thesis (see Ch. 6.7.2), and the presence of fluid inclusions consisting largely of vapour, indicative of boiling (Kesler, 2005), is unknown at this time. Hydrogen and oxygen isotope analysis of alteration minerals within the transition zone between potassic- and carbonate-phyllic assemblages, such as biotite and muscovite,  103  could also support or refute the possibility of phase separation (Harris and Golding, 2002).  Figure 42: 66 zone branch of the hinged cross section showing Au grade. Alteration shells are as indicated in Figure 36. Inferred faults extending from the upper and lower margins of the Upper Trachyte towards the Great Eastern Fault are shown. Gold grade is elevated along the lower fault at 98-107 m depth. Locations and results of XRD analyses of carbonate-phyllic altered rocks are shown (A, B, and C). Sample B nearest the 98-107 m fault has elevated Au grade and contains adularia.  See text for discussion. Orth = orthoclase, micro = microcline, ab = albite, adul =  adularia, musc = muscovite, dol = dolomite, ank = ankerite, bt = biotite, qz = quartz, py = pyrite.  5.27.2.4 Carbonate-phyllic alteration analogue The Early Jurassic Red Chris porphyry Cu-Au deposit of northwestern British Columbia (Stikine terrane; Fig. 1), which skirts the boundary between alkalic and calc-alkalic geochemistry, also has abundant carbonate alteration occurring throughout the  104  paragenetic sequence (Baker et al., 1997). The main ore zone (with a Cu/Au ratio of ~2) is associated with quartz-stockwork in pervasive quartz-sericite-ankerite (QSC) alteration. Baker et al. (1997) report that QSC alteration overprints the potassic assemblage, but they do not resolve the nature and source of the hydrothermal fluid. However, preliminary fluid inclusion work at Red Chris shows co-existing brine and vapour-rich fluids within the potassic assemblage implying phase separation of a magmatic and/or meteoric-derived fluid. 5.28 PART ONE CONCLUSIONS 5.28.1 Host rock control The ore fluid appears to have channelled outward from the MBX stock margin along permeable stratigraphic horizons (Lower and Upper Trachyte) and lithological contacts (Rainbow Dike). It diffused pervasively into the surrounding host rock causing reconfiguration of components to more stable mineralogical assemblages. Post-mineral tilting and extensional faulting along strands of the post-Jurassic Manson-Macleod Lake transcurrent fault system) resulted in the present-day geometry (Nelson et al., 1992, Fig. 43). 5.28.2 Alteration and mineralization There is an intimate relationship between sulfide- and alteration zonation at the Mt. Milligan Main Deposit. Lateral zonation is best developed in the MBX zone, whereas vertical zonation is best represented in the downthrown 66 zone. 5.28.2.1 Lateral zoning – MBX zone The Cu-Au ore zone coincides with chalcopyrite-pyrite mineralization and magnetitebearing potassic alteration in the MBX stock and basaltic-trachyandesite host rock. The magnetite-associated inner calc-potassic shell extends ~130 m outward from the stock margin, and abruptly terminates within the sodic-calcic overprint. In contrast, biotite alteration continues another ~150 m into the host rock, but is overprinted by the sodiccalcic (~100-150 m from the stock margin), inner-propylitic (~150-220 m), and outerpropylitic assemblages (>220 m). Generally, the outer-propylitic shell represents 105  106  Figure 43:  Previous page. Schematic cross-section of the Mt. Milligan Main deposit with a  scenario for the development of alteration zonation, and related ore-body, in the Early Jurassic. Moderate tilting and normal faulting in the Tertiary results in the present day structural geometry.  epidote-chlorite alteration of trachyandesite beyond the limits of calc-potassic alteration. Pyrite is the predominant sulfide outboard of the sodic-calcic assemblage, and contains modest Au grade. 5.28.2.2 Vertical zoning – MBX zone In the 66 zone, the Upper Trachyte hosts a magnetite-bearing Cu-mineralized calcpotassic assemblage, which is surrounded by sodic-calcic, inner- and outer-propylitic assemblages with increasing depth. Dolomite-ankerite replacement of mafic phases is present in the Upper Trachyte but increases in intensity to the southeast forming a funnel-shaped body of carbonate-phyllic alteration in the host rock.  The similar  elevation of magnetite breccia in DDH 91-815 to an Au-enriched, adularia-bearing fault in DDH 90-648 suggests the possibility of phase separation as the mechanism for localized gold deposition in the 66 zone. This fault, and a shallower fault at the contact between carbonate-phyllic and chloritic assemblages appear to be the main focus of hydrothermal fluids. The geometry of alteration in the 66 zone indicates the Rainbow Fault hanging-wall block may be rotated more than 45º.  107  CHAPTER 6: GEOCHEMICAL DISPERSION AT THE SILICASATURATED MT. MILLIGAN CU-AU ALKALIC PORPHYRY DEPOSIT 6.1 INTRODUCTION 6.1.1 Geochemical dispersion – fluid evolution Geochemical dispersion halos related to sulfide- and alteration zonation are investigated to determine ore fluid pathways, paragenetic sequence, and provide information on fluid sources. Strontium, sulfur, carbon, and oxygen isotope analyses from minerals sampled from drillcore across the deposit, presented herein, provide a lateral and vertical view of the combined evolution of the hydrothermal system. Results are compiled with thermal constraints from pressure-corrected fluid inclusion homogenization temperatures and pyrite-chalcopyrite sulfur isotope geothermometry. The integration of all data, geochemical and geothermometric, suggests Cu-Au mineralization at the Mt. Milligan Main deposit was derived from a CO2-bearing, low-tomoderate salinity, magmatic fluid. Copper-gold precipitation from this fluid formed a ~130 m halo surrounding the stock, and developed outward to an Au-only halo in the mixing zone with meteoric-derived fluids.  This is reflected in the lateral alteration  sequence of (calc)-potassic to sodic-calcic alteration, and from sodic-calcic to inner- and outer-propylitic assemblages with increasing proportion of meteoric fluids. Vertically, there is less evidence for meteoric fluid influx. A CO2-bearing magmatic fluid that may have undergone pH increase through interaction with limestone beds in volcanic stratigraphy (and/or phase separation) dominates the Au-only shell at shallow levels. 6.1.2 Epidote-pyrite ore vector and structural indicator Outboard of biotite hornfels surrounding the central stock, epidote and pyrite are common minerals in the host rock, and their trace element composition may provide a systematic ore-vector. In as much, the results of trace element analysis for this mineral association is significant for exploration programs throughout the Circum-Pacific involving alkalic porphyry systems. Epidote is used for its diagnostic solid solution behaviour between pistachite and clinozoisite, and for the ability to incorporate fluidborne incompatible elements (Franz and Liebscher, 2004; Frei et al., 2004), thereby 108  providing a proxy for detailed compositional changes in the magmatic-hydrothermal fluid across the system. Results indicate V, Mn, Ga, Sb, Zr, As (and possibly Bi) are the best ore-vector elements in epidote. In pyrite, these are Mn, As, Zr, Pb, and Bi. Fractionation of LREE in epidote toward the system core also provides a robust orevector. The lateral increase of trace element abundance in pyrite suggests incompatible elements (LILE-REE-HFSE) remained in solution until reduction enabled substitution into epidote and pyrite at a late stage. The spatial trends in trace element composition are also used to verify post-mineral normal faulting of the Rainbow Fault, as proposed by Placer Dome Inc. in 2004. 6.2 ANALYTICAL TECHNIQUES 6.2.1 Strontium isotopes High precision Sr-isotope analysis was performed at the Pacific Center for Isotopic and Geochemical Research (PCIGR) at the University of British Columbia.  A Thermo  Finnigan Triton (TIMS) was used to analyze hand-picked mineral separates for Nafeldspar, actinolite, and epidote.  Full trace element recovery was achieved with  standard hot plate dissolution in a Sallivex beaker® on 100-250 mg of monomineralic powder. A standard cation exchange column of Bio-rad AG50W-X8 resin (100-200 mesh) was used to separate Sr from Hf and REE. Strontium isotope compositions were measured in static mode with relay matrix rotation on a single Ta, and double Re-Ta filaments of the TIMS. The Sr reference material used was the SRM 987 standard. The data were corrected for mass fractionation by normalizing to  87  Sr/86Sr = 0.1194 using  the exponential law. In run errors were generally better than 0.000010 (2σ). 6.2.2 Sulfur isotopes A total of 116 sulfide samples were analyzed for δ34Spyrite (90 samples) and δ34Schalcopyrite (26 samples). Sulfide was collected as monomineralic powders from drillcore by handheld drill. Analyses of the sulfur isotope compositions were performed at the University of Tasmania (Hobart, Tasmania), using a VG Sira Series II mass spectrometer in accord to the conventional sulfur isotope techniques of Robinson & Kusakabe (1975). Standards used are the same as in Wilson (2003). Sulfur isotope compositions are 109  expressed in standard notation relative to Canon Diablo Troilite (δ34S value of 0‰, Shima et al., 1963), with a precision of ±0.1. 6.2.3 Carbon and oxygen isotopes A total of 12 carbonate analyses were carried out on 8 monomineralic powders collected from drillcore. Analyses were performed at the Pacific Center for Isotope and Geochemical Research (University of British Columbia) using the gas bench and a Delta PlusXL mass spectrometer in continuous flow mode. Samples were acidified with 99% phosphoric acid in helium-flushed sealed vials, and the headspace gas was measured in a helium flow. Corrections for fractionation were done through repeat analyses of UBC internal carbonate standards BN 13, BN 83-2, H6M. These have been calibrated against NBS international standards NBS 18 & 19. The δ13C (VPDB) and δ18O (VSMOW) values were corrected to VPBD and VSMOW based on an average of multiple analyses of NBS standards 18 and 19. 6.2.4 Epidote major elements – EMPA Electron-probe micro-analyses of epidote were done on a fully automated CAMECA SX-50 instrument, operating in the wavelength-dispersion mode with the following operating conditions: excitation voltage, 15 kV; beam current, 20 nA; peak count time, 20 s; background count-time, 10 s; spot diameter, 5 µm. Data reduction was done using the 'PAP' φ(ρZ) method (Pouchou & Pichoir 1985). For the elements considered (Na, Mg, Al, Si, K, Ca, Ti, Cr, Mn, Fe), the following standards, X-ray lines and crystals were used: albite, NaKα, TAP; kyanite, AlKα, TAP; diopside, MgKα, TAP; diopside, SiKα, TAP; orthoclase, KKα, PET; diopside, CaKα, PET; rutile, TiKα, PET; synthetic magnesiochromite, CrKα, LIF; synthetic rhodonite, MnKα, LIF; synthetic fayalite, FeKα, LIF. Epidote structural formulas were calculated by normalizing to 8 cations (excluding the H in the hydroxyl). 6.2.5 Epidote trace elements – LA-ICP-MS LA-ICP-MS analysis of 18 epidote samples (in 1-inch epoxy mounts) collected from drillcore was performed on an New Wave UP-213 Nd:YAG Q-switched laser ablation 110  system at the University of Tasmania. Beam diameter ranged between 40-110 µm, depending on the size of epidote analyzed. Analyses were collected from a suite of 26 elements including the alkali-earth elements (88Sr, 65  Cu,  172  Yb,  66  Zn,  175  89  Y,  90  Zr,  95  Mo,  197  Lu); actinides (232Th,  and the non-metal  75  Au); REE (139La, 238  140  137  Ba); transition metals (51V,  55  141  163  Ce,  U); other metals (69Ga,  Pr,  118  146  Sn,  Nd,  121  147  Sb,  As. Concentrations were calculated relative to  Sm,  205 43  Tl,  153  Eu,  208  Pb,  Mn,  209  Dy,  Bi);  Ca (assumed to  be stoichiometric). The international standard NIST612 was used as the primary standard, and the basaltic glass BCR-2 was used as a secondary standard. Standards were taken every hour to correct for instrument drift. After removal of analyses below detection limits, the average analytical error is 4.7±3.4%(1 σ). 6.2.6 Pyrite trace elements – LA-ICP-MS LA-ICP-MS analyses of 10 pyrite samples were collected from the same sample set. Beam diameter ranged between 55-110 µm, depending on the size of pyrite analyzed. Analyses were taken for a suite of 26 elements including alkali-earth element (137Ba) transition metals (49Ti, 182  W,  (75As, to  57  197 82  53  Cr,  55  Au); other metals (  Mn,  118  Sn,  57  Fe,  121  59  Co,  Sb,  Se, 125Te); and actinides (232Th,  205  238  60  Tl,  Ni, 208  65  Cu,  Pb,  66  209  Zn,  90  Zr,  95  Bi); REE (  Mo,  139  107  Ag,  111  Cd,  La); non-metals  U). Concentrations were calculated relative  Fe (assumed to be stoichiometric). The internal standard STDGL2b-2 was used.  This standard was quantified using a combination of laser ablation analyses based on previous in house standards, and NIST 612. correct for instrument drift.  Standards were taken every hour to  After removal of analyses below detection limits, the  average analytical error is 5.6 ± 4.5% (1σ). 6.2.7 Fluid inclusions Microthermometry was performed by Anthony Harris at the Center for Ore Deposit Research, at the University of Tasmania, Hobart, Tasmania.  Measurements were  conducted on a Linkam THMSG 600 heating-freezing stage (University of Tasmania) at 1-atm, calibrated with synthetic fluid inclusions produced by Syn Flinc and a variety of compounds of known melting compositions. Measurements below 10ºC are accurate to within ±0.5ºC, and measurements above this temperature to ±2ºC.  Calculations of 111  salinity from first-melting temperatures were made assuming a NaCl-H2O system as modelled by Bodnar (1994). 6.3 STRONTIUM ISOTOPES 6.3.1 Theory The isotopic composition of strontium (87Sr/86Sr) in whole rocks and individual minerals can be used as a tracer to indicate the source magma/fluid from which the rock or mineral crystallized. The convention is to use the absolute  87  Sr/86Sr ratio versus the  delta (δ) notation. Strontium has four naturally occurring isotopes (88Sr, 87Sr, 86Sr, 84Sr), all of which are stable. The  87  Sr isotope is the radiogenic product of negative beta  decay of 87Rb: 87  Rb → 87Sr + β- +  where β- is the negative beta particle,  +Q  is an anti-neutrino, and Q is the energy  produced by decay. As a Group IA alkali metal, Rb has an ionic radius (1.48 Å) that is sufficiently similar to K (1.33 Å) to substitute for potassium in K-bearing rocks and minerals (Faure, 1977). Furthermore, as an LILE, rubidium increases in concentration (along with K) with degree of magmatic fractionation, whereas Sr substitutes for Ca and gets sequestered in early plagioclase. Accordingly, more evolved igneous and volcanic rocks have higher concentrations of 87Rb (and consequently higher 87Sr/86Sr ratios) than less evolved rock-types, allowing for differentiation between different geological environments based on the abundance of  87  Sr. The average  87  Sr/86Sr increases (and  becomes more positively skewed) from tholeiitic oceanic basalts (~0.7030) to oceanic islands (~0.7040) to island arcs (~0.7045) to continental arcs (~0.7060; Faure, 1977). 6.3.1.1 Sr chemostratigraphy The composition of strontium in the oceans (typically between 0.7070 and 0.7090) is controlled by the mixing of reservoirs including the continental crust (via weathering), upper mantle (via seafloor hydrothermal activity), and recycling of marine carbonate (Corfield, 2003; Faure, 1977). It has been found that Sr in the oceans is isotopically homogenous with slow change rates and long residence times (Peterman et al., 1970). 112  Changes in Sr composition of the ocean reflect large-scale shifts (~10 millions of years) in the kinds of rocks exposed to chemical weathering, to periods of worldwide volcanic activity, and to changing proportions of crustal versus upper mantle input. Because of the isotopic homogeneity and slow change-rate, oceanic Sr chemostratigraphy provides a robust record of oceanic Sr composition throughout the Phanerozoic eon. 6.3.2 Sr composition of alteration minerals The Witch Lake succession, the host rocks to the Mt. Milligan deposit, are interpreted as subaqueous flows and/or hypabyssal sills (Nelson and Bellefontaine, 1996). The model age for the deposit (182.5 ± 4 Ma; Mortensen et al, 1995; Nelson and Bellefontaine, 1996) coincides with the proposed age for cessation of subduction and accretion of Quesnellia to ancestral North America at ~186-181 Ma (Nixon, 1993; Murphy et al., 1995). It also coincides with the partly subaereal Chuchi Lake succession (~196-180 Ma; Nelson and Bellefontaine, 1996), which may be represented by pyroclastic conglomerate in the upper DWBX zone. Consequently, there is some uncertainty as to the paleohydrologic setting of the Mt. Milligan deposit and related volcanic ediface, whether subaqueous or subaereal. To explore the role of external fluids at the Mt. Milligan Main deposit, Sr-isotope analysis was conducted on separates of secondary minerals Na-feldspar, actinolite, and epidote from the sodic-calcic, inner- and outer-propylitic alteration shells of the MBX zone. Epidote-bearing alteration assemblages in Cu-porphyry systems are thought to reflect convective influxing of non-magmatic fluids (meteoric or formation) toward the upflow path of hypogene fluids in the potassic zone (McMillan and Panteleyev, 1988). The Sr isotope ratios of the peripheral alteration minerals should reflect this process since magmatic and meteoric fluid reservoirs are isotopically distinct. The reference curve for oceanic Sr chemostratigraphy indicates a value of 0.70747 at 182.5 Ma (Fig. 44), the rutile U-Pb modal age for the Rainbow Dike (Nelson and Bellefontaine, 1996). Strontium geochemistry from the MBX stock indicates a primitive value of 0.70330 (Lang, 1998), which is on the low side of the expected range for 113  oceanic island arc rocks of intermediate composition, and implies magma genesis from a depleted (<0.7045) mantle source (Kesler et al., 1975). The Sr composition of the Mt. Milligan alteration minerals should fit within this endmember range (0.70330–0.70747) representing the span between magmatic and marine fluids at 182.5 Ma.  Figure 44: Systematic variation of the  87  Sr/86Sr ratio of the oceans during the Phanerozoic as  indicated by chemostratigraphy (modified from Corefield, 2003).  The Sr-ratio for 182.5 Ma  (0.70747), the measured age of the Rainbow Dike, is indicated.  6.3.3 Sr-isotope results Results from Sr-isotope analysis, including the calculated initial Sr ratios [(87Sr/86Sr)o] are given in Table 4. Alteration mineral sample locations in the MBX zone are shown in Fig. 45. The results are also plotted in a Rb-Sr isochron diagram (Fig. 46) with the best fit lines of individual minerals, best fit line of all minerals, and isochrons for 182.5 Ma against which the regression lines can be contrasted (see Ch. 6.9.1.1). The clustering of (87Sr/86Sr)o values immediately above the measured value for the MBX stock (Lang, 1998) illustrates the dominance of magmatic Sr in the alteration mineral genesis, and of  114  the sodic-calcic and propylitic assemblages they represent. However, displacement to values higher than 0.70330 reflects an ambient fluid component. The relatively primitive Sr-isotope composition of secondary epidote is relevant for zonal S-isotope fractionation across the Mt. Milligan Main deposit. It implies an absence of biogenic or marine-derived sulfur in the magmatic-hydrothermal system.  Figure 45:  Profile view of MBX zone showing Sr-isotope sample locations for Na-feldspar,  actinolite, and epidote. The compiled Na-feldspar and epidote alteration shells are also shown.  115  Figure 46: Initial  87  Sr/86Sr ratios (y-axis) at 182.5 Ma for samples indicated in Fig. 45 showing  dominantly magmatic signatures. The best-fit lines for measured  87  Rb/86Sr and  87  Sr/86Sr values of  individual alteration minerals show deviations from 182.5 Ma isochrons suggesting inter-mineral 87  Sr mobility (see Discussion for interpretation). The regression line for all analyses is within  error of 182.5 Ma model age for the Rainbow Dike.  6.4 SULFUR ISOTOPES 6.4.1 Theory Sulfur has four stable isotopes (32S,  33  S,  34  S, and  36  S). The ratio of  34  S (95.02%) to  32  S  (4.21%) in the sample as it deviates from the sulfur isotope standard (Canyon Diablo Troilite, CDT) is used to measure sulfur isotopic composition, expressed as δ34S: δ34S = (34S/32Ssample - 34S/32Sstandard) / (34S/32Sstandard) x 103 ‰  116  Ohmoto (1972) and Rye and Ohmoto (1974) explored the systematics of the sulfur isotope system, and concluded that δ34S is a function of many variables including pH, ƒO2, ƒS2, the mole fraction of each sulfur species, the initial isotope ratios, and temperature. Temperature controls the equilibrium constants (K) for reactions in the fluid involving sulfur, activity co-efficients of aqueous species, the ionic strength of the fluids (I) approximated by K+ and Na+ concentration, and the relative isotopic enrichment among sulfur species (δa- δb). Common aqueous sulfur species include H2S, HS-, S2-, SO42-, HSO4-, KSO4-, and NaSO4. The prevalent species largely depends on ƒO2 and pH of the fluid. In the log ƒO2 vs. pH space of Rye and Ohmoto (1974; Fig. 47A), at temperatures relevant to Cuporphyry mineralization, high ƒO2 conditions are represented by the sulfate anion field (XΣSO4 > 99%); high pH conditions are represented by the S- anion field (XS- > 99%); and low pH-low ƒO2 conditions are represented by the H2S field (XH2S > 99%). As long as fluid conditions constrain sulfur within a field, then changes in pH and ƒO2 will not affect δ34S.  However, between fields, where several different sulfur species are present,  small changes in pH or ƒO2 can have a significant effect on isotopic fractionation, as much as ±20 δ34S ‰ for one unit pH or ƒO2 (Fig. 47B). Copper porphyry deposits are considered to initiate between the H2S and SO4 fields, within the tight cluster of δ34S contours, and then evolve toward the H2S field with increasing acidity and reduction. The δ34S contours are clustered near pyrite-magnetite, pyrite-hematite, and barite soluble/insoluble boundaries, so that sulfides in equilibrium with magnetite, hematite, or a sulfate will tend to show variability in δ34S with changing pH and/or ƒO2. Furthermore, since oxygen forms stronger bonds with  34  S than  32  S (under equilibrium  conditions), the heavy sulfur isotope is sequestered in sulfate anions leaving a higher proportion of  32  S available for precipitation of early sulfide, resulting in depleted δ34S  signatures (negative). As the fluid evolves toward the H2S field, it moves up the δ34S contours resulting in less negative values of precipitated sulfide.  117  Figure 47: pH versus log ƒO2 at 250ºC.  A) Domains of H2S, SO4, and S2 speciation in a  hydrothermal fluid. Contours between zones represent mixing proportions between species. B) δ34S contours for pyrite [square brackets] and barite (curved brackets) precipitated from solution at the prevailing pH-ƒO2 conditions. The red arrow suggests the path of a magmatichydrothermal fluid in an alkalic porphyry system. At higher temperature, changes in pH and/or ƒO2 cause less δ34Ssulfide variation. Alternately, increasing ƒS2 expands the pyrite stability field, steepens the contours, and increases the potential for δ34Spyrite variation, such as in the pyrite halo within outer-propylitic and carbonate-phyllic alteration shells (modified from Ohmoto, 1972).  6.4.2 Examples Research into the alkalic porphyry deposits of British Columbia (Deyell and Tosdal, 2005) and New South Wales (Lickfold, 2001; Wilson et al., 2006) has described vertical and lateral S-isotope zonation characterized by low δ34S values in the deposit cores, and a gradual increase in δ34S values outward to the system periphery where δ34S values merge with near-zero background values. This is an important geochemical feature of alkalic porphyry systems due to the applicability as an ore vector. The diminutive size of hypogene alteration halos in alkalic systems (compared to the calc118  alkaline variety) limits the effectiveness of alteration gangue zonation as a vectoring method, and warrants alternative geochemical methods, such as S-isotope zoning (Wilson et al., 2006; Holliday and Cooke, 2007). Within British Columbia, both the Mt. Polley and Afton alkalic porphyry deposits exhibit S-isotope zonation with more negative δ34Ssulfide values proximal to the mineralized core. However, a degree of mixing with meteoric-derived fluid is implied by several anomalously positive values (Deyell and Tosdal, 2005). In the Cadia District, NSW, well-developed sulfur isotope zonation has been identified at Ridgeway and Cadia Hill, with δ34S values ranging from -3.7‰ and -9.4‰ respectfully in the deposit core to near-zero values in propylitic margins. Wilson et al. (2006) attribute the temporal-spatial δ34Ssulfide zonation to a combination of shifting redox conditions (H2S/SO4) and decreasing temperature of magmatic fluids at a constant composition of bulk sulfur (δ34SΣS). Temperature-controlled S-isotope fractionation between 600-300ºC alone cannot account for the range of δ34Ssulfide values observed.  An important  assumption is no external input of depleted δ34Ssulfide derived from biological sources, based on the absence of sedimentary sulfides in the volcano-sedimentary wall rocks, and primitive Sr-isotope compositions of secondary epidote. They also assume conditions of isotopic equilibrium as outlined above in the work of Ohmoto (1972). Wilson et al. (2006) propose a physio-chemical process wherein  34  S-enriched aqueous  sulfate species are reduced through interaction with primary Fe2+ bearing minerals in the host rock. Abundant Fe3+ bearing alteration minerals (magnetite, hematite, and epidote) in potassic and propylitic zones are diagnostic of this inorganic sulfate reduction. The authors provide examples of sulfate reduction reactions involving the production of hematite, magnetite, and epidote:  8FeO(rock) + SO42-(aq) + 2H+(aq) = 4Fe2O3(s) + H2S(aq) hematite 119  = 12FeO(rock) + SO42-(aq) + 2H+(aq) = 4Fe3O4(s) + H2S(aq) magnetite  CaFe2+Si2O6(s) + SO2-4(aq) + 2Al(OH)3(aq) + H4SiO4(aq) +2H+(aq) hedenbergite = CaFe3+Al2Si3O12(OH)(s) + H2S(aq) + 4.5H2O(l) + 1.25O2(aq) epidote The oxidation of ferrous to ferric iron in the host rock reduces the hydrothermal fluid by increasing the aqueous H2S/SO4 ratio, and gradually liberates  34  S to be utilized in  sulfide generation. Similar reduction of mineralizing fluids has been documented for the calc-alkaline Cu-porphyry system at Bingham, Utah where Bowman et al. (1987) used thermochemical calculations to show that log ƒO2 decreases from -17.5 in the potassic core to -29 in the propylitic fringe. 6.4.3 S-isotope results A total of 116 samples distributed laterally and vertically across the hinged crosssection of the present study were analyzed for δ34Spyrite (90 samples) and δ34Schalcopyrite (26 samples). The results, given in Appendix B and plotted in Fig. 48, show a welldeveloped zonation of δ34S values for an oxidized magmatic-hydrothermal system free of sedimentary sulfide. The deposit core contains the full range of δ34Ssulfide values from -4.79‰ in chalcopyrite from a quartz vein at the upper contact of the Rainbow Dike less than 20 m from the MBX stock, to +0.18‰ in chalcopyrite filled fractures at the center of the stock. The highest value of +0.35‰ occurs at the western margin of the hydrothermal system, ~250 m west of the MBX stock, hosted in chlorite-altered volcanic conglomerate in the DWBX fault footwall.  120  6.4.3.1 δ34Ssulfide vs. lateral distance The plot of δ34Ssulfide versus lateral distance from the center of the vertically reoriented MBX stock has an upward-widening distribution of δ34Ssulfide values resembling a whale tail in contour (Fig. 48-49), such that the peripheral wings represent distal pyrite samples with background (near-zero) values. δ34Ssulfide values between -5‰ and -3‰ (Fig. 49A) Values between -5‰ and -3‰ δ34Schalcopyrite occur in jigsaw-fit breccia with clasts of monzonite and biotite hornfels in pink K-feldpsar cement. These form the base of the distribution along the margin of the MBX stock (±200 m from the stock center).  Figure 48: Plot of δ34Ssulfide versus lateral distance from the center of the vertically reoriented MBX stock. The plotted δ34Ssulfide values form an upward-widening distribution suggesting reduction of oxidized magmatic fluids down the thermal gradient as fluids moved away from the MBX stock. The data supports a lack of contamination by isotopically enriched (positive) sedimentary sulfide, oceanic fluids or evaporites.  121  Figure 49: S-isotope δ34S sulfide values vs. lateral distance from center of MBX stock, grouped according to associated alteration. Transition from negative to positive values represents cooling and reduction of the mineralizing fluid to background values, providing a general analogue for paragenetic sequence, assuming the ore fluid evolves to cooler temperatures and lower oxidation state. The deposit is deepest to shallowest from the DWBX zone through the MBX zone, to the 66 zone. The earliest sulfide mineralization occurs at the stock margins and along the Rainbow Dike. See text for detailed description.  122  δ34Ssulfide values between -3‰ and -2.5‰ (Fig. 49B) Between -3‰ and -2.5‰, δ34Ssulfide values spread outward to 400 m in the MBX zone (following the potassically-altered Rainbow Dike), but remain at -180 m in the DWBX zone. A halo of biotite alteration surrounds the Rainbow Dike. Within the Lower Trachyte, a δ34Schalcopyrite value of -3.07‰ occurs within an ankerite bearing carbonatephyllic overprint of sodic-calcic and/or potassic alteration. δ34Ssulfide values between -2.5‰ and -1.5‰ (Fig. 49C) Between -2.5‰ and -1.5‰, K-feldspar hosted chalcopyrite retracts inward to the center of the MBX stock. Magnetite-sulfide (E4, E5) veins and veinlets occur with biotite alteration in the MBX zone. Sulfide-bearing quartz veins (E2) continue from the earliest stage, and are concentrated at the stock margin, within biotite hornfels (WBX and MBX zones), and in the Lower Trachyte (less than 240 m from the stock center). The sodiccalcic and inner-propylitic overprints (220-340 m, and 340-420 m from the stock center, respectively) also occur in the -2.5‰ and -1.5‰ range, as does sulfide associated with potassic alteration and magnetite in the 66 zone. In the DWBX zone, chalcopyritemagnetite clots and chalcopyrite veins (E3) occur in biotite hornfels (between -150 and -190 m). δ34Ssulfide values between -1.5‰ and +0.5‰ (Fig. 49D-E) Between -1.5‰ and +0.5‰, the δ34S values are spatially most widespread. In the MBX stock, pyrite and chalcopyrite veinlets continue with pink K-feldspar alteration. Pyritecarbonate veins (L3) also occur. In the biotite-altered MBX zone, pyrite pseudomorphs occur below the Lower Trachyte near the stock margin. In the 66 zone, pyrite veinlets (T1) and pseudomorphs occur with biotite-magnetite alteration above and within the Upper Trachyte. Also within this δ34S range, carbonatephyllic alteration occurs in a 1.7 m wide vein (L1) in the Rainbow Dike (at 424 m from center stock), and with pyrite-dolomite pseudomorphs in trachyandesite (between 700 and 760 m from center stock).  123  Cutting across all zones within the -1.5‰ to +0.5‰ range, is pyrite associated with the outer-propylitic assemblage. In the 66 and DWBX zones, it is characterized by epidotechlorite-pyrite groundmass replacement, and ribboned pyrite-calcite veins (L3). In the MBX zone, late pyrite-epidote veins (P2) cut the ore shell. In the DWBX zone, pyrite pseudomorphs in chloritized host rock in the footwall of the Harris Fault have δ34S values starting at -1.3‰. Outer-propylitic alteration farther west of the stock begins at -0.5‰. 6.4.3.2 δ34Ssulfide in profile – another view When plotted on the hinged cross-section (Fig. 50-51), the δ34Ssulfide values are lowest in the Rainbow Dike and Lower Trachyte where they contact the MBX stock. For example, a depleted chalcopyrite sample (-4.08‰) occurs in K-feldspar-cemented breccia at the edge of the MBX stock within the strongly altered and faulted contact with the Lower Trachyte. The δ34Ssulfide contours then move upward and outward across the MBX zone, giving the impression of a magmatic-hydrothermal plume rising from the Lower Trachyte (at ~180 m depth and 25 m from the MBX stock, DDH 90-616) to the farthest reaches of the Rainbow Dike (at ~50 m depth and ~300 m from the MBX stock, DDH 90-652), although this may be an artefact of sampling resulting from poor representation of the Lower Trachyte (core not available on site for sampling). δ34Ssulfide values between -5‰ and -3‰ The most negative values (-5‰ to -3‰) occur within ~60 m of the MBX stock in the Rainbow Dike, ~90 m in the Lower Trachyte, and in K-feldspar cemented breccia deep in the MBX stock (WBX zone). δ34Ssulfide values between -3‰ and -2‰ In the MBX zone, δ34Ssulfide values between -3‰ and -2‰ form an area that follows the Rainbow Dike outward for ~180 m from the stock margin, covers the extent of (calc)potassic, sodic-calcic, and inner-propylitic alteration of host rock, and retracts at least 50 m into the MBX stock. Similar δ34Schalcopyrite values occur in K-feldspar cemented 124  125  Figure 50:  Previous page.  S-isotope sample locations within the hinged cross-section and  34  associated δ S values. Sample points are coloured according to the dominant alteration. Red numbers represent δ34Schalcopyrite. Black numbers represent δ34Spyrite.  Figure 51: Contoured δ34Ssulfide values in profile across hinged cross-section showing earlyintermediate (red-orange-yellow) and late stage (blue) fluid domains and pathways (assuming δ34Ssulfide provides a proxy for oxidation state and temperature decrease of the evolving fluid). The integrated ‘total’ profile illustrates the likelihood of convective recycling of fluids in the MBX zone, and overprinting of the early-intermediate by late stage assemblage in the lower DWBX and MBX  126  zones. The red dashed line indicates the vertical extent of the system where the MBX stock is corrected for post-mineral tilting (post-mineral faulting is ignored). The early-intermediate contours encompass the potassic, calc-potassic, sodic-calcic, and inner-propylitic alteration shells, whereas the late stage contours correlate with the outer-propylitic, carbonate-phyllic, and chloritic shells.  breccia in the deep WBX zone (the lowest reaches of the MBX breccia body that cuts the stock). In the carbonate-phyllic altered Upper Trachyte (66 zone), a single δ34Spyrite value of -2.04 occurs in a specularite-cemented, clast-rotated milled breccia, where the hematite has likely replaced magnetite. δ34Ssulfide values between -2‰ and -1‰ In the MBX zone, values of δ34Spyrite between -2‰ and -1‰ are present: 1) beyond the biotite shell into the outer-propylitic assemblage; 2) below the Lower Trachyte in biotite hornfels ~60 m from the MBX stock; 3) outward to ~250 m along the Rainbow Dike (approximately 100 m before truncation by the Rainbow Fault) in a carbonate-phyllic vein (L1-stage). The downdropped 66 zone is dominated by δ34Spyrite values between -2‰ and -1‰ occurring with: 1) potassic and magnetite alteration above and within the Upper Trachyte; 2) inner-propylitic alteration below the Upper Trachyte; 3) carbonate-phyllic alteration in a fault continuing southeast from the lower faulted contact of the Upper Trachyte. In the WBX zone, the upper portion of the MBX stock has δ34Schalcopyrite values between -2‰ and -1‰ associated with pervasive K-feldspar alteration of monzonite. Similar values for pyrite occur in a wedge of biotite hornfels at the upper contact of the stock. δ34Ssulfide values between -1‰ and +0.5‰ In the MBX zone, the least oxidized δ34Spyrite values between -1‰ and +0.5‰ occur in the outer-propylitic shell, more than 220 m from the MBX stock margin. Anomalous pyrite-epidote veins (P2) with near-zero δ34S values occur within ~100 m of the MBX 127  stock in apparent-brecciated or conglomeritic host rock above the Rainbow Dike, and next to a ~3 m wide chloritized fault in the dike footwall. In the WBX zone, near-zero δ34Spyrite values occur in pink K-feldspar cemented jigsaw-fit breccia at the center of the stock (~100 m depth, DDH 90-597). These are shallower than more negative values associated with similar brecciation at 190-230 m depth. In the carbonate-phyllic altered 66 zone, a halo of near-zero δ34Spyrite values is situated above and below the trend of more negative values associated with the fault extending beyond the lower contact of the Upper Trachyte (at ~980 m ASL). The near-zero halo grades outward to chlorite alteration of similar value. Pyrite in epidote-chlorite alteration at the footwall of the Lower Monzonite dike is also near-zero. In the upper DWBX zone, pyrite veinlets and pseudomorphs in chlorite-altered pyroclastic and volcaniclastic host rocks have near-zero δ34Spyrite values. In the lower DWBX zone, near-zero δ34Spyrite values are associated with the outer-propylitic overprint. 6.5 CARBONATE C- AND O-ISOTOPES To further explore the observed association between elevated Au grade and carbonatephyllic alteration at the Mt. Milligan Main deposit, six dolomite and two ankerite representative samples were collected from drillcore across the 66 and MBX zones to be analyzed for carbon and oxygen isotopic ratios. The purpose of this reconnaissance study was to determine the nature of carbon species and associated CO2-bearing fluids, whether derived from a meteoric or magmatic source. 6.5.1 C-isotope systematics Carbon has two stable isotopes: abundance) to  13  12  C and  13  C.  The ratio of  12  C (98.89% natural  C (1.11%) in the sample as it deviates from the carbon isotope  standard (PDB belemnite) is used to measure carbon isotopic composition, expressed as δ13C: 128  δ13C = (13C/12C sample - 13C/12C PDB) / (13C/12C PDB) x 103 ‰ The systematics of C isotopes are similar to those of S isotopes. Fractionation of δ13C in hydrothermal systems is primarily a function of temperature (greater fractionation at lower T), but also of fluid and wall rock composition, dissolved carbon species, pH, and ƒO2 (Rye and Ohmoto, 1974). Pressure effects are negligible < 10 kbar. The heavy 13  C isotope is enriched in oxidized carbon species versus C0 (organic matter, graphite)  and CH4 (most reduced). Carbonate minerals precipitate from fluids containing oxidized carbon species including CO2, H2CO3, HCO3-, and CO3-2, which may be derived from a magmatic source, oxidation of reduced carbon reservoirs (sedimentary biological materials, or graphite), or by leaching of sedimentary carbonates. Each of these sources has a distinctive δ13C range that aids in determination of the CO2 source, although fractionation processes are complex and prevent a unique solution (Rollinson, 1993). For example, mixing of two reservoirs, sedimentary organic carbon (-26‰ to -38‰) and seawater (~0‰) could generate δ13C values in the mantle range (-2‰ to -8‰). Accordingly, δ13C values need to be interpreted with careful regard to the known geology. In addition, spatial (and temporal) variation in δ13C of carbonate minerals can help constrain the geometry of the plumbing system (Ohmoto and Goldhaber, 1997). 6.5.2 O-isotope systematics Oxygen has three stable isotopes: abundance) to  16  18  O,  17  O, and  16  O. The ratio of  18  O (0.205% natural  O (99.757%) in the sample as it deviates from the oxygen isotope  standard (V-SMOW – Vienna standard mean ocean water) is used to measure oxygen isotopic composition, expressed as δ18O: δ18O = (18O /16O sample - 18O /16OV-SMOW) / (18O /16OV-SMOW) x 103 ‰ Fractionation of δ18O in meteoric water is dependent on Raleigh fractionation processes in the Earth’s atmosphere. Fractionation to higher δ18O values in rocks than meteoric 129  water reflects isotopic exchange in silicates and carbonates. These minerals incorporate  18  O to varying amounts depending on temperature and water/rock ratio  (Ohmoto, 1986). The sensitive temperature dependency of  18  O fractionation allows for  determination of H2O source from diverse reservoirs (magmatic versus meteoric) using hydrothermally derived minerals and fluid inclusions.  6.5.3 C- and O-isotope results Results from carbon and oxygen isotope analyses are shown in Table 5 and plotted spatially in Fig 52. Isotopic ratios were corrected for carbonate-fluid isotopic fractionation to generate ratios of the responsible fluid. It was assumed that the fluid was buffered such that the dominant carbonate species was H2CO3, based on the 130  stability of sericite in the carbonate-phyllic assemblage, temperature of formation (>100ºC), and lack of CH4 in the oxidized ore fluid (Rowins et al., 1997).  Figure 52:  A) δ13C - δ18O plot that compares the fluid values calculated from Mt. Milligan  carbonates with isotopic composition of fluids from a variety of source reservoirs as compiled by Rollinson (1993). Numbers beside low T values correspond to red numbers in the profile below. B) Hinged cross-section showing location of carbonate isotope samples, predicted path of magmatic fluid (blue arrow), and Na-feldspar and epidote alteration shells representing sodiccalcic and propylitic alteration zones respectively. See Discussion for interpretation.  131  Carbonate-fluid 13C fractionation calculations Corrections for carbonate-fluid  13  C isotope fractionation followed the approximation by  Ohmoto (1972) that: δ13CH2CO3 ≈ δ13CCO2 ≈ δ13Cfluid.  All analyses were treated as  dolomite-CO2 fractionations after Ohmoto and Rye (1979), assuming negligible differences between ∆13Cdolomite-CO2, and ∆13Cankerite-CO2 at high temperature (310-235ºC). Carbonate-fluid 18O fractionation calculations Fractionation of  18  O in dolomite and ankerite from CO2 was calculated using the  empirical factors of Rosenbaum and Sheppard (1986) following other recent studies that investigate carbonate-fluid  18  O fractionation (Bierlein et al, 2004; Rosenberg and  Larsen, 2000). Calculations for dolomite and ankerite fractionation at Mt. Milligan (ranging from 400-175ºC) assumed the linear regression formulae of Rosenbaum and Sheppard apply above 150ºC (R2 = 0.9996 and 0.9931 for ankerite and dolomite, respectively). 6.6 TRACE ELEMENTS IN EPIDOTE AND PYRITE The thorough sampling of epidote-pyrite in drillcore across the Mt. Milligan Main deposit provided an opportunity to assess whether individual mineral deposits have characteristic trace element signatures that vary in space and time, reflecting fluid evolution, or whether these signatures are buffered by host rock (Cooke, 2006). Accordingly, LA-ICP-MS analyses of 18 epidote-pyrite samples, extending across sodiccalcic, inner- and outer-propylitic shells, were collected and examined for trace element variation (Fig. 53). Since epidote-pyrite is common throughout the system, with increasing abundance from the stock to the periphery, systematic changes in epidote and pyrite composition have the potential to be used as intra-system vectors to ore. This is particularly important within regional lower-greenschist metamorphic facies characterized by the same mineral suite (chlorite-actinolite-epidote) as propylitic alteration (Cooke et al., 1999). Samples were selected from each of the DWBX, MBX, and 66 zones. Since the deposit is moderately tilted to the east, samples from the lower DWBX zone represent a deeper 132  portion of the deposit, although the single upper DWBX sample represents an intermediate level or shallower. On the southeastern side of the stock, the MBX zone represents an intermediate level, whereas the more distal 66 zone is shallowest. Although the relative depth of the sections is uncertain due to postmineral faulting, Cu/Au ratios support the above interpretation (see Ch. 3.6.10 Fig. 21, and Ch. 5 for detailed review). To simplify interpretation, the 66 zone is here defined as the Rainbow Fault hanging-wall block, and is not applied to the area beyond the Cu-mineralized shell in the footwall (as defined by Placer Dome Inc.). The post-faulting depth spans ~670 m from the lower DWBX to 66 zones as is shown with reference to the estimated paleovertical line bisecting the MBX stock in Fig. 53. Additionally, the distribution of samples crosses the Rainbow Fault to the 66 zone, allowing comparison of trace element composition on either side.  Figure 53: Location of epidote-pyrite samples in relation to the compiled Na-feldspar and epidote alteration shells along the hinged cross-section. Also shown is the datum (center MBX stock) from which easting values are measured for the vertically re-oriented stock, with example measurements (positive and negative).  6.6.1 Epidote chemistry The epidote group minerals comprise a binary solid solution between monoclinic clinozoisite [Ca2Al3Si3O12(OH)] and pistachite [Ca2Fe3+Al2Si3O12(OH)] (Spear, 1993). Compositions are typically less ferric than Fe3+/(Fe3+ + Al3+) = 0.33, or Ps33 (Deer et al.,  133  1966), but values as high as 0.5 have been reported (Miyashiro and Seki, 1958). A value above 0.33 occurs in the present study, but likely represents addition of Fe2+. Epidote solid solutions can be described as: [(A1)(A2)][(M1)(M2)(M3)]Si3O12(OH) representing a structure with chains of edge-sharing octahedra parallel b that are crosslinked by SiO4 tetrahedra and double-tetrahedral Si2O7 groups (Frei et al., 2004; Fig. 54). The ‘A’ denominator represents nine- or ten-fold coordinated sites, whereas M1, M2, and M3 are in six-fold (octahedral) coordination (Deer et al, 1992). The symmetric M1 and M2 sites are normally occupied by Al3+, whereas both Al3+ and Fe3+ substitute into the distorted M3 site. The ‘A’ sites represent large irregular cavities between crosslinked octahedral chains, with the A2 site being larger and more distorted (Frei et al., 2004). Larger cations, usually Ca2+, occupy the A sites. The tendency for substitution of trace elements, such as LILE, REE, actinides (Th, U), and transition metals, into the A and M lattice sites of the epidote-clinozoisite structure [Ca2Al2(Al3+, Fe3+)Si3O12(OH)] gives it the potential to be used as an ore vector and proxy for fluid evolution in hydrothermal systems. 6.6.2 Previous work with epidote in porphyry systems Previous work exploring the use of epidote as an indicator mineral in calc-alkaline Cuporphyry environments was undertaken by Bowman et al. (1987) at Bingham, Utah, and by Norman et al. (1991) for the smaller system at Tintic, Utah. These studies used EMPA analysis of major elements to determine systematic compositional changes. At Bingham, it was found that compositions clustered around end-member epidote (Ps = 0.33) ranging in composition from Ps26.6  to Ps30.1 with no substantial zonation.  However, MnO content of vein epidote was observed to increase from 0.08 to 0.73 wt.% with distance from the deposit center. Alternatively, at Tintic, Mn and Fe3+ content decreased outward from the biotite zone resulting in a five-time increase in clinozoisite 134  Figure 54: A) Coordination polyhedra of epidote showing arrangement of A and M sites. B) Notation of the M positions in the octahedral chains. Image modified from Franz and Liebscher, 2004; and Dollase, 1968.  activity of epidote, suggesting a shift from more oxidized to more reduced conditions with increasing distance from the system core. A previous study by Ballantyne (1981) of three Cu-porphyry systems in Arizona (North Silver Bell, Safford, and the Christmas  135  Mine) showed no systematic variation in major element composition of epidote within the propylitic halo, and compositional variation between Ps24 and Ps34. 6.6.2.1 Compositional heterogeneity The high variability of epidote compositions observed in Cu-porphyry systems reflects the sensitivity of Fe3+ and Al3+ substitution to a complex suite of variables including shifting ƒO2, ƒCO2, temperature, bulk rock and fluid compositions, and pH induced changes in Al and Fe speciation within the fluid (Arnason et al., 1993).  However,  according to Frei et al. (2004), epidote crystal chemistry and structure are the first-order control of trace element composition since crystal structures are more rigid than fluid phases, and accept only ions of similar radius and charge as the major cations that normally occupy the lattice sites. Because of this micro-scale variability in trace element composition of epidote, multiple analyses of individual epidote grains must be taken to calculate a single representative composition within the context of a paragenetically constrained sampling traverse (Cooke, 2006). 6.6.3 Epidote results – major elements – EMPA Electron microprobe analyses (EMPA) of major elements (K, Na, Ca, Mg, Al, Cr, Ti, Mn, Fe, Si) were collected from 12 of the 18 epidote samples used for trace element analysis as a reconnaissance study of compositional variation across the deposit. Results are given in Appendix 3, and median values are compiled in Table 6. The establishment of pistachite ratios [Fe3+/(Fe3+ + Al3+)] was of particular interest. Fifteen to 20 spots were taken for each sample (10 mm2 epoxy-mounted wafers) in order to statistically compensate for compositional variability.  A structural formula was  calculated by normalizing to eight cations (excluding H in the hydroxyl). All iron was assumed to be ferric, except in seven instances where the A-site could not be filled by a combination of Ca, Mn, and Mg, and the trace remainder was subtracted from Fetot as Fe2+. As with trace element results described below, median values are used for each  136  sample in order to minimize the outlier effect, and the distribution of values is represented with one-sigma error bars (Fig. 55).  Figure 55: EMPA data showing median pistachite ratio [Fe3+/(Fe3+/Al)] versus lateral distance from the center of the MBX stock for sampled epidote, with 1-sigma error bars representing distribution of values per sample. Red dashed line shows the trend of median values.  137  6.6.3.1 Pistachite ratios of epidote Median pistachite ratios for the Mt. Milligan Main deposit range from Ps36 to Ps25. When plotted against lateral distance from the center of the MBX stock, epidote shows a systematic decrease in Fe3+ content. Furthermore, repetition of the decreasing trend across the Rainbow Fault in the 66 zone is consistent with post-mineral normal faulting, but is weakly constrained. In the MBX zone, median pistachite ratios decrease from Ps30 in the potassic zone to Ps25 in the footwall of the Rainbow Fault in the outer-propylitic shell. Likewise, in the 66 zone, the ratios decrease from Ps30 at the margin of the potassically-altered Upper Trachyte to Ps26.5 in the outer-propylitic assemblage. In the DWBX zone, ratios decrease sharply from Ps36 to Ps26 between potassic and outer-propylitic alteration. 6.6.4 Epidote results – trace elements – LA-ICP-MS Ten LA-ICP-MS analyses were collected from each of the 18 epidote samples to statistically compensate for microscopic variability. Analyses were taken for a suite of 26 elements including the alkali-earth elements (88Sr,  137  55  Ce,  Mn,  65  Cu,  172  66  Zn,  Y,  90  Zr,  163  Dy,  209  Bi); and the non-metal  Yb,  175  89  95  Mo,  197  Lu); actinides (232Th, 75  Au); REE (139La, 238  140  Ba); transition metals (51V, 141  U); other metals (69Ga,  Pr,  118  146  Sn,  Nd,  121  147  Sb,  Sm,  205  Tl,  As. Concentrations were calculated relative to  153  Eu,  208  Pb,  43  Ca  (assuming stoichiometric abundance) after correcting for machine drift. Integration intervals were selected from time-resolved spectra (counts per second) to maximize sampling of clean epidote and omit obvious inclusions (Fig. 56). Internal variations were retained. Once the data was converted to ppm, additional attention was required to eliminate values that had integrated outliers (such as Zr-REE-U-Th enriched titanite inclusions), or were below detection limit. After data correction, the average analytical error was 4.7% with a standard deviation (1-sigma) of 3.4%. The data were plotted against lateral distance from the vertically corrected MBX stock, using median values to minimize outlier effects. One-sigma error bars represent the spread in values. In this way, trends could be identified with greater confidence, and data clutter minimized. Vertical changes are inferred between zones: DWBX (deep), MBX (intermediate), 66 138  (shallow). The results are summarized in Table 7 and Fig. 57. Concentration profiles are described below in order of alkali-earth elements, transition metals, other metals, non-metals, actinides, and REE.  Figure 56: Examples of LA-ICP-MS time-resolved spectra for Mt. Milligan epidote showing trace element counts-per-second. A) Clean, unzoned epidote analysis with stoichiometric (flat-lying) 57  Fe and  43  Ca spectra. Other abundant elements with flat-lying spectra include Sr, Mn, V, Ga, Y,  139  and LREE (DDH 90-641 at 139 m). B) Inferred titanite (or zircon) micro-inclusion in epidote with elevated Zr, HREE, U and Th spectra highlighted in bold (DDH 90-643 at 122 m).  Alkali-earth elements Alkali-earth elements with valences of +2 are accommodated into the distorted A-site cavities of the epidote structure because they have a larger ionic radius than the Ca2+ they substitute for (Frye et al., 2004). The high degree of compositional variation of Sr in epidote depends on Sr bulk rock abundance, fluid-rock interaction, and the presence of other Sr-bearing minerals including Ca-plagioclase, titanite, apatite, and calcite. Transition metals and other metals Many transition metals have several oxidation states resulting in a variety of substitution mechanisms in the epidote structure reflecting the prevailing redox conditions. Each of the elements V, Mn, Cu, Zn, Zr, Mo and Au can have a valence of +2. However, elements V and Mn can be oxidized to +3 and +4, whereas Au and Mo exist in a variety of oxidation states ranging from -1 to +7 and +2 to +6, respectively. Divalent ions are accommodated into the A-sites by direct substitution for Ca2+, and will be incorporated into the undistorted, smaller A1 site if the ion has a smaller radius than Ca2+. Vanadium, Mn, Cu, Zn and Pb can all substitute in epidote as divalent ions. Trivalent ions with small ionic radius, such as many transition metals, will preferentially substitute into the octahedral M-sites by replacing Al3+ or Fe3+. Most trivalent elements show a strong preference for the more distorted M3 site, usually occupied by Fe3+. Gallium, Sb, Tl, and Bi are each trivalent and preferentially substitute for Fe3+ in the distorted M3 site, but will substitute for Al3+ as well. Tetravalent ions with small ionic radius, or High Field Strength Elements (HFSE), are commonly incorporated into epidote M-sites through coupled substitutions with a divalent ion, replacing two Al3+ and/or Fe3+ ions in order to maintain charge (Frye et al., 2004). Zirconium and Sn are examples.  140  141  142  143  Figure 57:  Previous two pages.  Trace elements in epidote measured by LA-ICP-MS.  Ten  analyses per sample are represented by median values (to reduce positive skew due to outlier analyses) with 1-sigma error bars to represent the distribution of values per sample. Trends for median values are shown in red dashed line. See text for detailed discussion.  Non-metals Arsenic is the only non-metal analyzed for trace element composition in epidote. Arsenic has two oxidation states at +3 and +5, and is expected to substitute preferentially for Fe3+ in the distorted M3-site of the epidote structure, or via coupled substitutions with a univalent metal for two M-site Al3+ and/or Fe3+ ions if oxidized to the +5 valence. Actinides The actinides (Th4+ and U4+) behave as LILE (large-ion lithophile elements) and can substitute for Ca2+ in epidote A-sites with concomitant replacement of two M-site ions (Al3+ and/or Fe3+) with two divalent ions to balance the charge.  Alternately, the  actinides can replace A-site REE3+ through a coupled substitution with a divalent ion replacing M-site Al3+ or Fe3+ to balance charge (Frye et al., 2004). REE and Y Due to their large ionic radii, trivalent REE and Y preferentially substitute for Ca2+ in the more distorted A2-site of the epidote structure with concomitant replacement of a trivalent M-site Fe3+ or Al3+ ion with a divalent ion, usually Mg2+ or Fe2+. Igneous rocks of intermediate composition typically show an enrichment of LREE relative to MREE and HREE (Frye et. Al, 2004). 6.6.4.1 Epidote results – alkali-earth elements (Ba, Sr) Barium and strontium in epidote have similar trends in the MBX and DWBX zones, but are antithetic in the 66 zone. Despite the high compositional variability of Sr in epidote, trends are more apparent than in Ba. Strontium is the most abundant trace element in epidote averaging ~1530 ppm. Median Sr concentration decreases in the DWBX and MBX zones by 100’s of ppm from the deposit center toward the outer-propylitic 144  assemblage, and jumps in value across the DWBX fault. Median Ba likewise decreases in the DWBX zone (~10 ppm), and shows no change in the MBX zone. In the 66 zone, median Sr sharply increases (~500 ppm) from the footwall of the potassically-altered Upper Trachyte to the outer-propylitic halo, whereas Ba decreases. The results suggest that Sr and Ba variation in epidote is more vertical than lateral, with a steep lateral increase in Sr concentration at shallow depths (66 zone), possibly indicating a change in host rock stratigraphy to trachyandesite interlayered with calcareous beds, as suggested by carbon isotope analysis. 6.6.4.2 Epidote results – transition metals and other metals (V, Mn, Cu, Pb, Zn, Ga, Sb, Th, Bi, Zr, Sn, Mo, Au) Vanadium and manganese are the 3rd and 2nd most abundant trace elements in epidote averaging ~430 and ~1200 ppm, respectively. The trends in median V and Mn concentration show a systematic lateral increase away from the MBX stock. In the MBX zone, median V increases less dramatically than Mn and has a higher degree of scatter about the trend. Alternately, the trend in Mn represents the third best fit to the spread in values (R2 = 0.5664 in the MBX zone) reported in the present study (other than Bi and As) and suggests increasing piemontite [Ca2Al2(Mn3+,Fe3+)Si3O12(OH)] exsolution towards the system periphery. Both elements provide robust intra-system ore-vectors. Results suggest no vertical change in V or Mn concentration.  The return to lower  values across the Rainbow Fault is consistent with post-mineral normal faulting. The trends in median copper and lead concentration in epidote decrease laterally outward by ~5-10 ppm from the MBX stock (although there is no change in median Pb within the 66 zone). Copper has highest compositional variability in the DWBX zone, particularly within outer-propylitic alteration, where the spread in values per sample reflects abundance of chalcopyrite micro-inclusions. Lead has increasing variability toward the center of the system in the MBX zone (in sodic-calcic and inner-propylitic alteration) reflecting a separate population of Pb-enriched epidote, rather than inclusions of galena.  145  Median Pb concentration jumps to a higher value across the DWBX fault, and does not change across the Rainbow Fault, whereas the repeated decreasing trend in median Cu is consistent with post-mineral normal faulting. There are no systematic changes in median Cu with elevation. Median Pb appears to increase vertically at deeper levels (DWBX zone), but shows no change at shallower levels (66 zone), such that maximal Pb concentration in epidote occurs nearest to the stock at intermediate levels. Zinc abundance in epidote shows no clear variation within the MBX and 66 zones, but median Zn increases outward in the DWBX zone by ~15 ppm from potassic to outerpropylitic alteration, reflecting increased abundance of sphalerite and chalcopyritesphalerite micro-inclusions in the outer-propylitic zone. Median Zn decreases vertically such that highest concentration in epidote occurs at the periphery of the deepest levels. The trends in median gallium and antimony concentration in epidote increase laterally outward from the system center. In the MBX zone, median Ga increases weakly (~5 ppm) toward the Rainbow Fault, whereas median Sb has a steeper lateral increase (~80 ppm). Scatter about the Ga trend reflects a separate population of Ga-enriched epidote forming halos to pyrite. In the 66 and DWBX zones, the trends increase more sharply for median Ga (~20 ppm). Gallium jumps to a higher value across the DWBX fault, whereas Sb shows no change. Both Ga and Sb drop in concentration across the Rainbow Fault, consistent with post-mineral normal faulting. Results show no vertical change, and highest concentrations are located at the periphery of intermediate levels. The systematic behaviour in median Ga and Sb, and the abundance of Ga and Sb in epidote (10’s of ppm) provide a potential ore-vector. Thallium concentration in epidote is close to detection limits of the mass spectrometer and the analyses are therefore less reliable. Nonetheless, median Tl and bismuth behave similarly in the MBX (increasing laterally) and 66 zones (decreasing laterally). There is no variation in median Tl in the DWBX zone, but median Bi increases laterally outward (~1 ppm).  146  Results suggest concentration of Bi in epidote is greatest toward the periphery of intermediate levels, whereas the highest Tl occurs proximal to the stock at shallow levels. The change in trend across the Rainbow fault, particularly for Bi, confirms a structural discontinuity. The good linear fit to the Bi data (R2 = 0.8049 in the MBX zone) might make XBi a reliable ore-vector, but concentrations are low and the change in behavioural trend at shallow levels introduces complexity. The trend in median zirconium concentration in epidote increases laterally outward (10-20 ppm) from the core of the system, and reaches a maximum of 130 ppm in the upper DWBX zone, coupled with a maximum spread in values. Increasing compositional variability reflects internal fluctuations of REE-HFSE (growth zones), and titanite micro-inclusions. The return to lower median Zr across the Rainbow Fault is again consistent with post-mineral normal faulting. The systematic lateral increase at all levels, and abundance of Zr in epidote, provides a good ore-vector. However, care must be taken to minimize Zr-enriched inclusions where integrating time-resolved spectra and correcting data. The concentration of tin in epidote shows no change in the MBX and DWBX zones, however there is increased variability in Sn toward the stock in MBX and 66 zones reflecting REE-HFSE zoning in epidote (sodic-calcic alteration), and Sn micro-inclusions in the 66 zone (potassic alteration). Across the Rainbow Fault, median Sn jumps in value and decreases sharply toward the outer-propylitic fringe, coupled with decreasing compositional variability. Results suggest Sn concentration in epidote is highest near the center of the system at shallow levels. Molybdenum and gold values are at the detection limits of the mass spectrometer, resulting in fewer analyses from which to calculate a median value and spread. Accordingly, the results are less reliable, particularly for the DWBX and 66 zones. However, there appears to be a trend in the MBX zone where median Mo and Au concentration decreases laterally from sodic-calcic to outer-propylitic assemblages.  147  Gold occurs as micro-inclusions in epidote, whereas Mo fluctuates with incompatible elements Ba-REE-HFSE. 6.6.4.3 Epidote results – non-metals The median concentration of arsenic in epidote increases systematically outward from the system core with sufficient range (~40 ppm) to provide a robust ore-vector. Median values conform well to a linear trendline within MBX (R2 = 0.8393) and DWBX (R2 = 0.6279) zones. Furthermore, a distinct break in the trend and return to low values across the Rainbow Fault clearly indicates post-mineral normal faulting. No vertical change in concentration is evident. 6.6.4.4 Epidote results – actinides The profiles for thorium and uranium in epidote are similar to Sn, indicating incompatible behaviour similar to HFSE’s. Both lack any change in the trend of median values within the MBX and DWBX zones, but have a sharp lateral decrease in the 66 zone, such that maximum Th and U concentration is centrally located at shallow levels. Compositional variability is greatest near the stock in the MBX zone and toward the Rainbow Fault in the 66 zone, particularly for thorium. The jump in median values of Th and U across the Rainbow Fault, coupled with a change in behavioural trend, again indicates a structural discontinuity. 6.6.4.5 Epidote results – La (LREE), Sm (MREE), Lu (HREE) and Y Generally, REE in epidote show parallel behaviour across the deposit. Concentration is highest for LREE (and Y), and decreases through MREE and HREE, with increasing amplification of the data profile (Fig. 58). In the MBX zone, the trend in median REE values increases laterally to the outer-propylitic shell (although this may be due to a separate population of REE-enriched epidote in the outer-propylitic shell overprinting a laterally decreasing trend). Scatter about the trend reflects compositional heterogeneity and REE-HFSE zoning. In the DWBX and 66 zones, median REE concentration decreases laterally outward, and shows less scatter than in the MBX zone. In the upper DWBX zone, median REE values jump to higher concentration and compositional 148  variability, reflecting increased REE-HFSE zoning and titanite (or zircon) microinclusions. The results show a vertical increase at deep levels (DWBX zone), and no vertical increase at intermediate to shallow levels, such that the maximum REE concentration in epidote occurs near the stock at deep and shallow levels, but at the periphery of intermediate levels. The change in trend across the Rainbow Fault again highlights the structural discontinuity.  Figure 58: Representative REE (and Y) in epidote measured by LA-ICP-MS. Ten analyses per sample are represented by median values (to reduce positive skew due to outlier analyses) with 1-sigma error bars to represent the distribution of values per sample. Median trends are shown in red dashed line. See text for detailed discussion.  BSE normalized REE in epidote When median compositional values of the analyzed suite of REE are normalized to Bulk Silicate Earth (McDonough and Sun, 1995) and plotted as spidergrams for each sample, the sequence of graphs shows a subtle steepening of LREE profiles toward the 149  150  Figure 59: Previous page. REE spidergrams for secondary epidote across the Mt. Milligan Main deposit. Sample locations for individual spidergrams are indicated on the hinged cross-section. The compiled epidote alteration shells are also shown.  The paleo-vertical datum is shown  (dashed red line on cross-section) with example lateral distance measurements. REE data is normalized to Bulk Silicate Earth of McDonough and Sun (1995). The Y-axes of spidergrams are not at a common scale, due to spatial variation in trace element abundance. Fractionation of LREE increases toward the MBX stock resulting in steeper profiles, particularly in the DWBX zone. This is illustrated with dashed red lines on spidergrams B to F (DWBX zone) representing tangents to the LREE curves. Numerical values beside tangent lines are steepness values of the tangent from horizontal. These values generally increase toward the stock.  Figure 60: LREE ratios showing stockward fractionation of LREE to higher La concentration relative to other LREE, and graph of Eu anomaly magnitude versus lateral distance from the vertically  corrected  MBX  stock.  Data  points  represent  chondrite  normalized  median  concentrations (10 analyses per sample). Interpreted trends are shown in red dashed line.  stock (particularly in the DWBX zone), indicating increasing LREE fractionation (La>Ce>Pr>Nd) toward the system core (Fig. 59). Plots of La/Ce, La/Pr, and La/Nd show a predictable decrease outward from the center of the system, providing a robust  151  intra-system ore-vector (Fig. 60A-C). Moreover, each plot has a return to higher values and a repetition of the decreasing trend across the Rainbow Fault, consistent with postmineral normal faulting. In addition, positive Eu anomalies generally increase in magnitude toward the outerpropylitic assemblage, particularly in the DWBX zone, reflecting a lateral increase in the substitution of Eu2+ (1.09 Å) for Ca2+ (1.14 Å), of similar ionic radius. A graph of median Eu/0.5(Sm+Dy) in epidote, representing the relative height of the Eu anomalies, shows the expected outward increase in the DWBX and MBX zones. The slope of the trend shallows from the DWBX to MBX zones, and decreases outward in the 66 zone (Fig. 60D). This suggests the Eu anomaly is greatest on the periphery of deep levels, and moves inward toward the stock with decreasing depth, until the trend has reversed at shallow levels. 6.6.4.6 Epidote trace elements in profile To support the spatial trends derived from concentration versus lateral distance profiles, the median values for As, Cu, Th, and 3La/(Ce+Pr+Nd) in epidote have been plotted on the hinged cross-section to provide a combined lateral-vertical view of the compositional trends (Fig. 61). Arsenic represents profiles that increase laterally to the system periphery with maximum abundance at intermediate levels. Copper represents laterally decreasing trends with greatest abundance near the stock at intermediate levels. Thorium represents vertically increasing trends with maximum abundance towards the stock at shallow levels. The relationship 3La/(Ce+Pr+Nd) represents the fractionation of LREE to higher La concentration (relative to Ce, Pr, and Nd) at the center of the system. Increasing the size of the data set and sampling density would reduce the proportion of anomalous values. Trends are clearly discernable, nonetheless. 6.6.5 Pyrite results – trace elements – LA-ICP-MS Epidote alteration at the Mt. Milligan Main deposit is commonly intergrown with pyrite, such that the pyrite halo surrounding the Cu ore zone is co-spatial with sodic-calcic,  152  Figure 61: Median values for LA-ICP-MS analyses of As, Cu, Th and 3La/(Ce+Pr+Nd) in epidote plotted according to sample location on the hinged cross-section to further illustrate trends defined in lateral distance versus compositional space. The relationship 3La/(Ce+Pr+Nd) represents the stockward fractionation of LREE towards relatively higher La concentration, and is  153  derived from normalized values of LREE (McDonough and Sun, 1995).  Red dashed line and  numerical values indicate amount of throw in meters between chemically similar epidote on either side of the Rainbow Fault. See text for further description.  inner- and outer-propylitic alteration assemblages. Accordingly, LA-ICP-MS analyses for pyrite were collected from the same 18 samples used for epidote trace element analyses. Thereupon, pyrite-epidote alteration could be examined as a system, rather than epidote as an isolated mineral. Furthermore, because epidote-pyrite dominates the peripheral and late paragenesis, it is widespread throughout the deposit, and has the potential to be tracked within the system footprint, and traced to ore. As with epidote, ten analyses were taken for each of the 18 samples to generate a median concentration to be plotted against lateral distance from the vertically corrected stock, with one-sigma error bars to represent the spread in values per sample. Concentrations  were  calculated  relative  to  57  Fe  (assuming  stoichiometric  concentrations) after correcting for machine drift. Analyses below detection limit were eliminated from the data set. Careful attention was paid to time-resolved spectra to eliminate spikes representing inclusions such as epidote, titanite or chalcopyrite, however micro-inclusions within an otherwise undisturbed pyrite signature were included in the integration. Differentiation between epidote and titanite inclusions from spectra is not possible since both minerals concentrate Ti, Zr, REE, and HFSE above background levels of pyrite (Deer et al., 1992). However the abundance of epidote in the outer-propylitic assemblage suggests these inclusions are epidote. The average analytical error for all elements after eliminating analyses below the detection limit, and those with anomalous inclusions, is 5.6 ± 4.5% (1-sigma). LA-ICP-MS analyses were taken for a suite of 26 elements including alkali-earth element (137Ba) transition metals (49Ti, 53Cr, 55Mn, 57Fe, 59Co, 60Ni,  65  metals (118Sn,  209  121  Sb, 205Tl, 208Pb,  Cu, 66Zn, 90Zr, 95Mo, 107Ag, 111Cd, 182W, 197Au); other Bi); REE (139La); non-metals (75As,  82  Se, 125Te); and  actinides (232Th, 238U).  154  Substitution for Fe2+ in pyrite (FeS2) is homovalent for elements that are in a divalent state. Elements that occur at higher valency will involve coupled substitutions such as multiple Fe2+ ions replaced by tetravalent ions, or combinations of trivalent or other oddnumbered valency ions (such as Au3+ and As3+) with univalent ions (such as Cu1+ and Ag1+) to balance charge.  Anions (Se-2, Te-2) can substitute for S2- by direct ion  exchange (Chouinard et al., 2005). 6.6.5.1 Pyrite results – elements that increase laterally outward Most trace elements in pyrite exhibit a systematic increase in concentration outward from the MBX stock (Fig. 62, Table 8). These elements include: Ba, Ti, Cr, Mn, Co, Zn, Ag, Zr, Cd, W, Au, Sn, Sb, Tl, Pb, Bi, La, As, and Te. Accompanying the outward increase in median trace element abundances, the value spread per sample for many elements also increases such that the most distal samples within individual zones (DWBX, MBX, and 66) have the maximum compositional variation. This reflects cryptic zoning of trace elements in pyrite, and increased abundance of micro-inclusions. It suggests multiple stages of pyrite generation in the outer-propylitic assemblage, with micro-inclusions incorporated through pyritization of altered mafic phenocrysts and host rock groundmass. Furthermore, each of the laterally increasing pyrite trace element profiles shows a sudden drop in median values across the Rainbow Fault and DWBX Fault (except for La and Sn in the DWBX zone) consistent with post-mineral normal faulting, and a more centralized pre-fault location for the upper DWBX and 66 zone hanging-wall blocks. Pyrite trace elements – highest potential ore-vectors Median values for manganese and arsenic in pyrite increase in abundance by 10’s of ppm from sodic-calcic to outer-propylitic alteration. Median As concentration in the DWBX zone increases exponentially toward the DWBX fault (y = 0.0345e0.0175x, R2 = 0.8926). The large spread of values in the most distal samples reflects compositional heterogeneity of pyrite in the outer-propylitic assemblage. In individual analyses, As concentration shows broad variations in time-resolved spectra, but 155  156  157  Figure 62: Previous two pages. Trace elements in pyrite (measured by LA-ICP-MS) that increase in concentration with lateral distance from the MBX stock. Ten analyses per sample are represented by median values (to reduce positive skew due to outlier analyses) with 1-sigma error bars to represent the distribution of values per sample. Interpreted trends are shown in red dashed line. See text for detailed discussion.  approaches homogeneity. Manganese concentration increases toward grain edges and epidote/titanite inclusions. Cobalt is the third most abundant trace element in pyrite (along with Ni and Ti) averaging ~55 ppm, and has high variability per sample. Despite this variability, the trend in median values increases laterally outward by ~50 ppm in the DWBX zone, ~100 ppm in the MBX zone, and ~150 ppm in the 66 zone suggesting increasing lateral concentration to shallower levels. Spectra from the most Co-rich analyses (1000-2000 ppm) show uniformly high stoichiometric Co and Ni. Median zirconium and lead concentration in pyrite increases by 3-5 ppm across the DWBX, MBX, and 66 zones, whereas bismuth increases by ~1 ppm.  Outwardly  increasing variability in Zr concentration reflects Zr-U-Th zoning and inclusions of epidote/titanite. Variations in Bi spectra commonly mirror Pb. Lead and Bi increase in abundance toward epidote/titanite and chalcopyrite inclusions, forming halos or co-  158  precipitating with Cu. Spectra indicate galena-chalcopyrite-epidote/titanite multiinclusions in P2 veins (66 zone), and galena inclusions near the Rainbow Fault (MBX zone). Each of Zr, Pb and Bi are consistently above detection limits despite low concentrations. There is no discernable change in concentration vertically. Pyrite trace elements – moderate potential ore-vectors Median zinc concentration in pyrite shows a well-defined outward increase across the DWBX zone of ~2 ppm, and more uniform values across the MBX and 66 zones. The maximum compositional variability occurs near the DWBX fault where Zn spectra mirror Mn and increase toward pyrite grain edges. In the sodic-calcic shell of the MBX zone, Zn micro-inclusions (sphalerite?) occur where there is elevated Cu. Zinc concentration in pyrite is highest on the periphery of deep levels, and decreases in abundance to shallower levels. The trend in median antimony concentration increases on the order of 1 ppm toward the system periphery in the MBX and DWBX zones, with an increasing value spread reflecting zoning in pyrite and higher concentration at pyrite grain edges. Antimony generally follows the Mn spectra. There is no variation in Sb composition in the 66 zone. Increase in the MBX zone is exponential near the Rainbow Fault (y = 0.0011e0.0114x, R2 = 0.7711). Antimony concentration in pyrite is highest at periphery of deep to intermediate levels, and decreases to shallow levels. Concentrations of cadmium, gold, thallium, thorium and uranium in pyrite approach detection limits. Median values increase by ~0.1 ppm towards the system periphery. The spread in values for Au and Tl increases outward in the DWBX zone and is maximized near the DWBX fault in the outer-propylitic assemblage. Gold occurs as AuAg-Te micro-inclusions (DWBX zone) and Au micro-inclusions (MBX zone) near the edge of pyrite grains. It is commonly associated with elevated Cu values (~300-400 ppm). Thallium micro-inclusions occur rarely with elevated Cu in sodic-calcic and innerpropylitic shells of the MBX and 66 zones. Variations in U and Th spectra generally follow zoning of Zr, and epidote/titanite inclusions. There is no vertical change in 159  160  median concentration of Cd, Au Th, and U in pyrite, but Tl is greatest on the periphery of deep levels. The trend in median tin concentration increases weakly by ~0.2 ppm toward the system periphery, and does not change in the 66 zone. Variability of Sn concentration in pyrite reflects cassiterite micro-inclusions (DWBX zone), increasing values toward pyrite grain edges, and elevated Sn in epidote/titanite inclusions. There is no clear change in Sn abundance with height in the system. The trend in median lanthanum (LREE) increases laterally on the order of ~0.1 ppm across the MBX and DWBX zones, but individual samples exhibit a large spread in values due to compositional zoning and epidote/titanite micro-inclusions where the La spectrum broadly follows Mn-Ti-Zr-U-Th ± Cr. The increasing trend in the 66 zone is less well defined. Concentrations approach detection limits in La poor analyses. The highest concentration of La in pyrite occurs on the periphery of deep levels, and decreases with height vertically. Titanium is the second most abundant trace element in pyrite averaging ~70 ppm. The trend in median Ti increases on the order of 200-300 ppm across the DWBX and MBX zones, but only by ~20 ppm across the 66 zone, indicating decreasing concentration to shallow levels. The increased value spread per sample in the outer-propylitic zone reflects increased compositional heterogeneity, Ti zoning, and epidote/titanite microinclusions. Variations in Ti spectra are commonly reflected in Cr-Zn-La-W-Th-U ± Mn and Cu. Maximum Ti concentration occurs on the periphery of intermediate levels. The trend in median tellurium composition increases laterally outward by ~1 ppm across the DWBX, MBX, and 66 zones, but is poorly constrained in the DWBX zone. The increasing spread in values per sample in the outer-propylitic assemblage is due to internal variations with Pb and Bi, and in part to Ag-Au-Te micro-inclusions.  161  Pyrite trace elements – lowest potential ore-vectors Median barium concentration increases laterally by ~5 ppm across the DWBX and MBX zones, and by ~0.3 ppm in the 66 zone indicating decreasing abundance to shallower levels. The value spread per sample is maximal in the footwall of the DWBX fault. Compositional variability is due to Ba-Mn-Zn enriched pyrite grain edges, and Ba inclusions (sulfate?) in the MBX zone (sodic-calcic shell). A pyrite rim in the outerpropylitic DWBX zone hosts an Au-Ag-Te inclusion containing Ba. Median chromium concentration increases by ~5 ppm across the DWBX and MBX zones, and by ~2 ppm in the 66 zone. Variability of Cr within individual samples is due to LILE-HFSE zoning in pyrite where the Cr spectrum generally follows Mn-Zr-La-Th-U ± Zn, W, and Sn spectra. The majority of silver analyses are above detection limits. The trend in median Ag concentration has the steepest lateral increase (~1 ppm) in the DWBX zone, flattens in the MBX zone, and increases again in the 66 zone.  Silver concentration (and  variability) in pyrite is highest on the periphery of deep levels where it fluctuates with Te ± Au. A second population at shallow levels is associated with Pb-Bi ± Sb, Cu. For many analyses, tungsten values approach detection limits. The trend in median W concentration increases by only ~0.2 ppm across the DWBX, MBX, and 66 zones. The highest spread in values per sample occurs in the outer-propylitic assemblage. Variations in W concentration follow Ti. 6.6.5.2 Pyrite results – elements that do not increase laterally outward Of the 25 analyzed trace elements in pyrite, only five show behaviour other than a lateral increase in concentration away from the center of the system (Fig. 63). Copper is the only trace element that systematically decreases in abundance towards the system periphery, whereas Mo, Ni, and Se show asymmetric behaviour with different trends at different levels.  162  Median copper concentration and the spread of values per sample is highest proximal to the MBX stock (in sodic-calcic alteration) and decreases laterally outward to the periphery of the system. Compositional variability reflects Cu zoning in pyrite, and abundance of chalcopyrite micro-inclusions. Concentrations are highest at intermediate levels.  Figure 63:  Trace elements in pyrite that show behaviour other than an increase in median  concentration with distance from the MBX stock. See text for discussion.  Only ~15% of representative analyses for molybdenum concentration in pyrite are above detection limits of the mass spectrometer. The trend in median Mo increases laterally outward to the system periphery by ~0.05 ppm in the DWBX zone, and decreases by the same amount in the 66 zone. There is no change in the MBX zone, although the spread in values increases toward the stock.  Molybdenum occurs as  micro-inclusions with no apparent affinity with other elements, except perhaps Cu. Results suggest an inward shift of Mo concentration with height such that the highest Mo values occur at the periphery of deep levels, but close to the stock at shallow levels.  163  Along with Ti and Co, nickel is the most abundant trace element in pyrite, averaging ~75 ppm. Median Ni concentration increases toward the system periphery in the DWBX zone by ~100 ppm, and decreases in the MBX and 66 zones by a similar amount (50100 ppm). There is considerable spread in values per sample, and scatter about the trends. The spread in values reflects compositional heterogeneity of pyrite since Ni spectra are generally stoichiometric and lack internal variations. Nickel spectra are closely associated with Co, and secondarily to Pb, Bi, and Se. Similar to Mo, results suggest a shifting inward of maximum Ni concentration towards the stock with decreasing depth. The trend in median selenium concentration increases toward the system periphery in the DWBX zone by ~20 ppm, but decreases in the MBX (~50 ppm) and 66 zones (~10 ppm). The decrease in the MBX zone is the best constrained of the trendlines, echoing a similar trend in Cu. Selenium spectra are evenly stoichiometric, comparable to Ni, Bi, and As. Selenium abundance in pyrite is greatest near the stock at intermediate levels, and decreases vertically to shallower levels. 6.6.5.3 Pyrite trace elements in profile To support interpretations derived from graphs of trace element concentration versus lateral distance from the MBX stock, the median values of Mn, Co, and As have been plotted on the hinged cross-section to provide a combined lateral-vertical view of the compositional trends (Fig. 64). Each represents a profile that increases laterally toward the system periphery. Again, increasing the sample size and density would improve the visible trends, and decrease the proportion of anomalous values. 6.7 GEOTHERMOMETRY 6.7.1 Chalcopyrite-pyrite S-isotope geothermometry 6.7.1.1 Theory Fractionation of sulfur isotopes occurs during precipitation of sulfide minerals from sulfide melts or hydrothermal fluids, and is controlled by variable bond-strengths between sulfides such that certain minerals are more enriched in 34S than others (Sakai, 164  Figure 64: Median values for LA-ICP-MS analyses of Mn, Co, and As in pyrite plotted according to sample location on the hinged cross-section to further illustrate trends defined in lateral distance versus compositional space.  Each element shows a laterally increasing trend.  See text for  further description.  1968; Bachinski, 1969). The degree of S-isotope fractionation between cogenic sulfides (∆A-B = δ34SA-δ34SB = ~103lnαA-B) is a linear function of the temperature of isotopic equilibration (expressed as equations in the form: δ34SA -δ34SB = (C * 106)/T2, where C is an empirically-derived constant), and approximates the temperature of sulfide precipitation as long as equilibrium conditions are met. Temperatures calculated from isotope fractionation include a degree of uncertainty, but generally conform to more 165  robust geothermometric techniques (such as pressure-corrected fluid inclusion homogenization temperatures) to within ∆A-B ±0.2, such that error increases with decreasing  δ34SA-δ34SB  and  increasing  temperature  (S.  Rowins,  personal  communication). Consequently, cogenetic minerals with the most dissimilar isotope fractionation behaviour provide the best sulfide pairs with which to determine equilibrium temperature, such as pyrite-galena. Enrichment in  34  S decreases on the  order of: sulfate > pyrite > sphalerite > chalcopyrite > galena. The  pyrite-chalcopyrite  geothermometer  is  insufficient  for  precisely  defining  mineralization temperature, but provides an approximate estimate for temperatures in the 600-250ºC range. Sulfide geothermometry should be used in conjunction with fluid inclusion homogenization temperatures (TH), to better constrain temperature estimates, and to determine isotopic equilibrium (sulfide coprecipitation) or disequilibrium depending on whether the two methods are in agreement or not.  Figure 65:  Equilibrium temperature vs. lateral distance from the center of the vertically re-  oriented MBX stock showing associated alteration. The average equilibrium temperature (356ºC) is based on 10 of the 12 samples ranging 317-496ºC. The remaining elevated temperatures likely signify isotopic disequilibrium, but could represent an early Cu-mineralizing event involving magmatic fluids at ~600ºC (within error).  166  6.7.1.2 Chalcopyrite-pyrite geothermometer results Twelve pyrite-chalcopyrite pairs (24 samples) of the 116 sulfur-isotope samples collected across the Mt. Milligan Main deposit appeared to be in textural equilibrium, and were used for geothermometry according to the constants determined by Kajiwara and Krause (1971). Of the 12 analyses, 10 ranged between 496-317ºC with an average of 356 +67/-50ºC (Fig. 65), which is within the expected range for sulfide precipitation in Cu-porphyry systems (~600-200ºC; Seedorff et al., 2005). The hottest of these (496 +127/-85ºC was taken from the 1.7 m wide auriferous carbonate-phyllic L1 vein in the Rainbow Dike (DDH 90-652). The remaining two analyses from the WBX and DWBX zone have magmatic values of 585ºC and 726ºC, respectively, which probably reflect isotopic disequilibrium and separate sulfide precipitation events. The 726 (+342/-168)ºC result was taken from an E3 chalcopyrite vein in biotite hornfels (suggesting either modification of the Cu-mineralized zone by later pyrite-generating fluids (isotopic disequilibrium), or early precipitation from cupiferous fluids at magmatic temperatures ~650-560ºC (within error). 6.7.2 Fluid inclusion analysis and geothermometry Microthermometric observations were made on 6 samples of hydrothermal minerals (quartz, epidote, and calcite) collected from potassic and outer-propylitic alteration shells of the Mt. Milligan Main deposit. The intent of this reconnaissance study was to better constrain the physicochemical nature of the responsible fluids. Temporal variations in fluid inclusions were identified with detailed petrographic observations using doubly polished thick sections.  Primary inclusion populations were the main  focus of heating and freezing measurements. These were observed along welldeveloped growth zones in quartz, and isolated in negative crystal form within quartz and calcite, whereas irregular shaped primary inclusions were measured in epidote. Fluid inclusions are classified into five types according to the associated alteration (P = potassic, OP = outer-propylitic), character (primary, secondary), thermodynamic behaviour, and composition. Results are compiled in Table 9 and presented spatially in Fig. 66. Pressure corrected homogenization temperatures were calculated using the  167  salinity-dependent liquid-vapour curve and isochores (T vs. P curve at constant volume) of Shepherd et al. (1985) at an estimated pressure of 1200 bars (~3.5 km depth). Type P-I Type P-I fluid inclusions occur in the potassically altered WBX zone, within the deepest sampled portion of the MBX stock. They are two-phase (liquid-vapour) with 30-50 vol.% vapour. Homogenization occurs by vapour bubble disappearance and meniscus fading at temperatures up to 320ºC. Meniscus fading at TH (temperature of homogenization) is diagnostic of a supercritical fluid (Shepherd et al., 1985), and the observed return of the vapour bubble within 2-10ºC of cooling from TH is also diagnostic (A. Harris, personal communication). First ice melting temperatures Te (-42ºC) from freezing experiments suggest a NaCl-H2O ± CaCl2 ± MgCl2 fluid complex. These inclusions trapped a lowsalinity (9.5 wt.% NaCl equiv.), near-critical, aqueous, probably magmatically derived fluid trapped in a one-phase field. Type P-II Type P-II fluid inclusions are observed in the same sample as P-I inclusions from the WBX zone, and are also two-phase, but have lower TH (270-173ºC). They occur in crosscutting secondary inclusion trails. Type P-III Type P-III fluid inclusions occur in quartz veins cutting biotite hornfels within ~30 m of the MBX stock in the potassically altered DWBX and MBX zones. These are liquidvapour inclusions (20-30 vol.% vapour) that also homogenize by vapour disappearance at 245-210ºC. Inclusions occur as isolated negative crystal shapes (MBX zone), or as crosscutting inclusion trails (DWBX zone). They may contain one or more daughter crystals (± sylvite ± calcite ± halite ± hematite ± unidentified opague).  Estimated  salinities are 16 wt.% NaCl equivalent, which is higher than the other two-phase inclusion types. During freezing, observed gas hydrates melted at 3.8-7.6ºC, and Te was measured at -47ºC, indicating a CO2-bearing NaCl-H2O ± CaCl2 ± MgCl2 -rich fluid.  168  169  170  Figure 66: Previous page. Hinged cross-section of the Mt. Milligan Main deposit showing fluid inclusion data (bottom) including homogenization temperatures not pressure corrected (minimum trapping temperature) and pressure corrected temperatures at 1200 bars (~3.5 km depth). Sulfur isotope pyrite-chalcopyrite geothermometer results are included (top) for comparison. The two data sets indicate the fluids remained at elevated temperature along the main potassic fluid conduits (MBX stock, Rainbow Dike, Lower Trachyte), but cooled outward into the surrounding host rock and outer-propylitic assemblage. Salinity increases from deep to shallow levels (DWBX to 66 zone) suggesting a transition to brinier fluids through degassing and/or mixing with equilibrated external fluids. The compiled albite and epidote alteration shells are also shown.  Type OP-I Type OP-I fluid inclusions occur in epidote, quartz, and calcite from the outer-propylitic assemblage of the DWBX, MBX, and 66 zones. These are simple two-phase liquidvapour primary inclusions (30-50 vol.% vapour) that also homogenize to liquid. From deep to shallow levels of the fossil hydrothermal system, the OP-I fluid inclusions indicate a cooling trend accompanied by increasing salinity. In the DWBX zone, they exhibit the highest TH (~248ºC) and lowest salinities (4.5 wt.% NaCl equiv.). In the MBX zone, they have lower TH (220-175ºC) and higher salinities (10 wt.% NaCl equiv.). In the 66 zone, TH is lowest (~200ºC), and salinities are highest (13 wt.% NaCl equiv.). Type OP-II Type OP-II fluid inclusions are observed in the same sample as OP-I inclusions in the DWBX zone. They are two-phase secondary inclusions with lower TH (~215ºC), low salinities (~3.5 wt.% NaCl equiv.), and can be CO2-bearing (forming gas hydrates melting at 6.0ºC). 6.8 DISCUSSION The mineral chemistry and isotope systems (Sr, S, C, and O) examined in the present study can be combined to develop a spatial-temporal picture of fluid evolution for the Mt. Milligan Main deposit including fluid pathways, shifting ƒO2, changes in fluid chemistry, and changing proportions of source fluids. Complimenting this data are  171  geothermometric results that bracket mineralization temperatures and define the thermal evolution of the system. 6.8.1 Sr-isotopes – magmatic vs. meteoric fluids In Fig. 46, Na-feldspar has the highest magmatic signature ranging from 94.7 to 93.6% magmatic (assuming a 0.70330  88  Sr/86Sro value is 100% magmatic fluid and a 0.70747  value is 100% seawater). Epidote has the largest range from 86.7 to 76.7% magmatic, and actinolite ranges from 79.1 to 78.3%.  Accordingly, the sequence from most  primitive to least is: Na-feldspar, epidote (1), actinolite, epidote (2). Despite the strong magmatic Sr signature, a subordinate amount of mixing with external fluids is also implied. The range of epidote values suggests increased mixing with external fluids over time, or locally variable proportions of mixing. Complimentary analyses of δ18O and δD would help clarify the nature of the external fluid, whether isotopically-depleted meteoric water, or isotopically-enriched formation waters (Bowman et al., 1987; Norman et al., 1991). However, for the purpose of the present study, it is inferred that the sodic-calcic assemblage (and less so the inner- and outer-propylitic assemblages) is dominated by a magmatic fluid component, with a secondary meteoric fluid component.  Formation waters (andesite equilibrated) would increase the  proportion of external fluids since the  88  Sr/86Sro signature would be closer to magmatic  values than seawater. 6.8.1.1 Isochron analysis Assuming alteration was coeval with the emplacement of the Rainbow Dike (182.5 Ma), the measured values for individual alteration minerals (as well as all minerals) should form parallel isochrons with a slope of y=0.0021x (Fig. 46) unless closed-system behaviour was not approached and 87Rb and/or 87Sr was added or lost (Faure, 1977). The regression line for all seven analyses is concordant with the 182.5 Ma isochron within analytical error (±46 Ma) suggesting an approach to closed-system behaviour for the lateral alteration sequence as a whole.  However, the individual minerals have 172  divergent slopes indicating diffusion-related gains and losses of  87  Rb or  87  Sr between  minerals (and between the alteration shells they represent), assuming closed system behaviour. The slope for Na-feldspar (Na altered K-feldspar) is less than the 182.3 Ma isochron suggesting  87  Rb or  87  Sr loss from the mineral structure, or a thermal resetting of the  isotope clock by a subsequent heating event at ~84.9 Ma under closed system conditions. This scenario is not supported by similar resetting of the other minerals, and is rejected. Actinolite (177 Ma) is the most concordant with the 182.5 Ma isochron suggesting only slight losses of  87  Rb or 87Sr. Epidote is the most divergent mineral with  an isochron slope of ~400.6 Ma, which is incompatable with the early Jurassic geology. The higher slope implies diffusion-related gains of  87  Rb or  87  Sr, likely reflecting the  propensity of epidite to incorporate LILE and other incompatible elements into its structure. Microprobe analyses of epidote from the Mt. Milligan Main deposit indicate K concentrations up to 1967 ppm (avg. = 280 ppm, median = 81 ppm; Appendix 3), and substitution of Rb in trace amounts is expected. However, the more likely scenario is Sr (1.12 Å) substitution for Ca (1.14 Å), of similar ionic radius. It appears that  87  Sr and  87  Rb isotopes were disproportionately leached with Ca and K  from the sodic-calcic shell and redistributed to the inner- and outer-propylitic shells, resulting in the  87  Sr depletion and enrichment observed for Na-feldspar and epidote  respectively, assuming closed system conditions. Moreover, the concordance of Nafeldspar, actinolite and epidote collectively with the model age for the Rainbow Dike supports alteration processes dominated by magmatic fluids. 6.8.2 S-isotopes The path of the mineralizing fluid from highest to lowest oxidation state and temperature, as reflected in the δ34Ssulfide proxy, originated at the stock margin in the MBX zone and channelled along the Lower Trachyte and Rainbow Dike in an apparent magmatic-hydrothermal plume associated with (calc)-potassic and marginal sodic-calcic alteration of the host rock (Fig. 51). The oxidation state then evolved laterally outward 173  through inner- and outer-propylitic shells outboard of the sodic-calcic shell, and vertically to carbonate-phyllic alteration surrounding the main potassic fluid conduit (Upper Trachyte) at shallow levels. The fluid evolved to less negative δ34Ssulfide toward the center of the potassically-altered MBX stock suggesting recession of geotherms from the stock margin. However, negative values (< -2) are present in the deep WBX zone (~190-230 m) in an early monzonite phase that has been crackle-brecciated and pervasively altered by pink Kfeldspar. These depleted values are probably linked to those in the MBX zone, but data coverage is lacking. In the MBX zone, minor faults and brecciated/conglomeritic horizons with reduced sulfide suggest inward lateral convection of late fluids during thermal collapse of the system (Fig. 51). Ingress of peripheral fluids towards the stock appears to be more developed and widespread in the lower DWBX zone. In the 66 zone, a halo of near-zero δ34Spyrite values surrounds more negative values associated with a fault continuing southeast from the lower contact of the Upper Trachyte. The near-zero values indicate outward diffusion of auriferous carbonatephyllic fluids into host rock surrounding the main fluid conduit. 6.8.3 C- and O-isotopes Magmatic source fluids have been proposed for some orogenic gold vein-type deposits (Burrows et al., 1986), and moderately reduced intrusion-related gold deposits (Baker, 2002). Baker argues that intrusion-related systems involve high-temperature (>350ºC) immiscible brines that co-exist with low salinity CO2-bearing vapour that was exsolved from the parent magma at an earlier stage. The CO2-rich emission has the property of elevating the critical temperature of fluids, potentially causing primary boiling of saline volatiles at greater depth and higher temperature. This would allow for prolonged generation of Au-complexing volatiles within the intrusion over a wider depth range. Furthermore, in alkaline solutions, the solubility of calcite increases with temperature 174  and the addition of either NaCl or CaCl2 (Holland and Malinin, 1979), such that brines of alkaline chemistry are capable of assimilating carbon from deep-seated crustal sources, or of maintaining magmatic CO2 in solution. Alternately, if the carbonate-phyllic assemblage at the Mt. Milligan Main deposit is related to meteoric fluid dilution of magmatic fluids, as has been suggested for phyllic alteration shells in Cu-porphyry systems (McMillan and Panteleyev, 1988), the isotopic signature of the carbonates should exhibit this. 6.8.3.1 Carbonate source fluids The δ13C-δ18O plot of Fig. 52A compares fluid values calculated from Mt. Milligan carbonates with isotopic composition of fluids from a variety of source reservoirs as compiled by Rollinson (1993). The cluster of values at the edge of the magmatic H2O box indicates these carbonates were derived from magmatic fluids. Mixing trends are observed between magmatic and meteoric fluids (suggesting the Mt. Milligan system could have been partly subaereal at the time of mineralization), and from meteoric fluids to seawater. A single value in the limestone box suggests the possibility of limestone beds in the 66 zone. 6.8.3.2 Carbonate source fluids in profile The profile view (Fig. 52B) shows the location of carbonate samples, and illustrates that analyses with magmatic or near-magmatic values (1-4) are located at the upper contact of the Lower Trachyte, within the Rainbow Dike, and within meters of shallow faults extending beyond the Upper Trachyte. The blue arrow shows the predicted path for magmatic fluids. Values with dominantly meteoric values (5-7) suggest convection of meteoric fluids outboard of the magmatic fluid path.  Fluid mixing is reflected in  alteration assemblages where sodic-calcic alteration (5) occurs at the magmaticmeteoric interface close to the stock, whereas outer-propylitic and chloritic alteration occurs with meteoric fluids (6-7). Values on the meteoric-seawater (7-8) mixing trend occur in dolomite-cemented breccias at the margin of the MBX stock, and the Lower Monzonite dike (66 zone), suggesting incursion of seawater at a late stage. 175  6.8.4 Geothermometry and isotope synthesis The combined geothermometric data sets (pressure-corrected fluid inclusion TH, and pyrite-chalcopyrite S-isotope fractionation) indicate the fluids were hottest (~455-370 ºC) and supercritical in the WBX zone (Fig. 66). Low δ34S values (< -2) from deeper in the WBX zone similarly reflect an early-stage potassic fluid that may be linked to mineralization higher up in the MBX zone. In the MBX zone, fluid temperatures remained elevated (~355ºC) within the jigsawbrecciated, K-feldspar cemented MBX stock margin above the Rainbow Dike, and the faulted contact between the stock and Lower Trachyte. These mineralizing fluids were CO2-bearing, and more saline (<16 wt.% NaCl equiv.) than the critical fluids in the stock (<9.5 wt.% NaCl equiv.) suggesting phase separation at the stock margin. Additionally, δ34S values are at a minimum (< -4) indicating fractionation of sulfur isotopes in an oxidized fluid between aqueous sulfate anion and the precipitated sulfide. The potassic fluids (~344ºC) diffused into the surrounding trachyandesitic host rock forming a ~130 m shell of (calc)-potassic alteration and Cu-Au mineralization. At the outer margin of this shell, mixing with meteoric-derived (formation?) water, as indicated by the (87Sr/86Sr)o ratio of Na-feldspar and actinolite (94% and 78% magmatic), formed a ~60 m wide shell of sodic-calcic alteration, marking the outer limit of the Cu ore zone. Outboard of the Cu mineralized shell, potassic alteration and high temperature fluids (<496ºC) were channelled along the Rainbow Dike, and less so within the Lower Trachyte and trachyandesite host rock. At ~230 m from the stock, a dolomite-ankeriteK-feldspar-muscovite vein (L1-stage) with elevated Au and Cu grade (5.11 g/t, 0.29 wt.%) cuts the Rainbow Dike and represents the superposed carbonate-phyllic assemblage. The C- and O-isotope ratio for the vein dolomite is indicative of a CO2bearing magmatic fluid. As such, the inferred fluid was potentially the same as that in quartz-hosted, CO2-bearing fluid inclusions (P-III) from the potassically-altered stock margin. The P-III inclusions may be CaCl2 - and MgCl2-bearing, suggesting a fluid capable of producing dolomite.  176  Host rocks ~150 m laterally outward from the sodic-calcic shell in the MBX zone are overprinted by inner- and outer-propylitic alteration stages, and show increased proportion of meteoric or formation water, as indicated by (87Sr/86Sr)o in epidote (77% magmatic), and meteoric values for C- and O-isotope ratios in dolomite. Salinity of epidote-hosted primary fluid inclusions (OP-I) increases from 10 wt.% (NaCl equiv.) to 13% from the MBX to 66 zone.  Figure 67: Synthesis of results from Sr-, S-, C-, and O- isotope analysis, with pressure-corrected fluid inclusion geothermometry, and pyrite-chalcopyrite S-isotope geothermometry. Results are plotted in cross-section against K-feldspar and Na-feldspar alteration to better constrain the hydrothermal plumbing, and are contrasted with the resultant Cu and Au ore shells. See text for discussion.  In the 66 zone, magmatic C- and O-isotope ratios in dolomite occur within the Upper Trachyte (particularly where it becomes a magnetite-cemented milled breccia) and  177  continue to the southeast toward the Great Eastern Fault along a fault horizon stemming from the lower contact of the Upper Trachyte.  A halo of more positive  δ34Spyrite values surrounds more negative values adjacent the fault and indicates diffusion of auriferous carbonate-phyllic fluids into the host rock. There are no fluid inclusion or S-isotope geothermometer results for the Upper Trachyte, but pressurecorrected TH in epidote from the outer-propylitic shell surrounding the carbonate-phyllic zone indicates temperatures of ~310ºC. It is therefore estimated that temperatures within the Upper Trachyte were in the range of 400-310ºC, based on the occurrence of potassic alteration and magnetite precipitation.  Finally, incursion of seawater (as indicated by C- and O-isotopes) resulted in clastrotated dolomite-cemented breccias at the previously mineralized margins of the MBX stock and Lower Monzonite dike in the MBX- and 66 zones, respectively. 6.8.5 Interpretation of Cu-Au and Au-only ore zones Lateral development The path of magmatic and meteoric fluids proposed in Fig. 67 suggests the transition from the inner Cu-Au ore shell to the peripheral Au-only shell coincides with the transition from magmatic- to mixed magmatic-meteoric fluids. Copper precipitation at the interface between fluids suggests cooling below ~350ºC and dilution were the principal mechanisms. Outboard of the sodic-calcic zone, Au remained in solution with minor amounts of Cu into the increasingly meteoric mixed-fluid zone, where pyrite is the predominant sulfide. Here, Au grains (~30 um) are commonly entrained in chalcopyrite that occurs in pyrite grain interstices, and trapped as inclusions in amalgamated pyrite. Accordingly, it seems that H2S levels were buffered by the trachyandesitic host rock so that AuCl2- remained the dominant aqueous complex (along with CuCl2) at a steadily decreasing solubility to 300ºC (Gammons and Jones, 1997). There is less petrographic evidence for Au transport as Au(HS)2- in an H2S-rich fluid devoid of Cu, as previously argued (Stanley, 1992; Delong, 1996), although rare Au blebs on pyrite grain edges devoid of chalcopyrite support the latter.  178  Vertical development In the 66 zone, the Au-only shell develops outboard of the potassically-altered Cu-Au mineralized Upper Trachyte. The transition to Au-only mineralization appears to be linked to a magnetite-cemented breccia that marks the shift from potassic to carbonatephyllic alteration within the Upper Trachyte.  A halo of auriferous carbonate-phyllic  alteration surrounds the breccia. Gold grade decreases steadily with distance (over ~100 m) from the Upper Trachyte except along minor faults and late-mineral dikes. Isotope analysis of carbonates indicates the fluids were magmatic in origin. A dolomite sample from the outer margin of the 66 zone (90-648 at 74 m depth) plots in the limestone box of the δ13C-δ18O diagram (Fig. 52A), suggesting pH increase as a mechanism for Au precipitation from the magmatic fluid. Phase separation may also have been a factor, based on the geometry of carbonate-phyllic alteration surrounding magnetite-cemented breccia, plus the presence of adularia and elevated Au grade in a minor fault where δ34Spyrite values are most negative (see Ch. 5.27.2.3). 6.8.6 Epidote trace elements – ore vectors The investigation of epidote trace element chemistry across the Mt. Milligan Main deposit has suggested intra-system lateral and vertical trends despite frequently large compositional ranges in individual samples (Table 7, Fig. 57). Elements that show systematic lateral increases in concentration are: V, Mn, Ga, Sb, Zr and As (all have highest concentration at the periphery of intermediate levels along with Bi, Y, MREE, and HREE). Elements that show a systematic lateral decrease in concentration are: Pb, Cu, Au, and Mo. Elements that increase in concentration vertically are Ba, Sr, Tl, Sn, Th, U, and LREE (all are more concentrated in the system core, except Sr, which increases laterally outward). Zinc is the only element that increases laterally at deep levels, and decreases vertically (Fig. 68). The median compositional values of all elements (except for Pb and Zn) show either a sudden change or repeat in trend across the Rainbow Fault indicating a structural break, and post-mineral normal faulting. Epidote chemistry in the cross-sectional view 179  of the Main deposit supports a throw of ~95 m between the MBX zone and downdropped 66 zone (Fig. 61). The trace elements in epidote that provide the best intra-system vector to ore are: V, Mn, Ga, Sb, Zr, As, and possibly Bi. However, the best use of epidote as an ore-vector may be through chondrite-normalized LREE ratios: La/Ce, La/Py, La/Nd, or a multiple such as 3La/(Ce+Pr+Nd), using median (or average) values derived from multiple analyses of the epidote crystal(s) comprising each sample, as done in the present study.  Furthermore, the height of positive Eu anomalies appears to correlate with  intensity of epidote alteration at deep to intermediate levels, and a relationship such as Eu/0.5(Sm+Dy) can be used to vector towards the system center.  Figure 68: Graph showing the location of highest concentration of the specified trace element in epidote for the Mt. Milligan Main deposit, as interpreted from median values of LA-ICP-MS analyses. Elements in red provide the best intra-system ore vectors. Depth is indicated with associated ore-zone (DWBX, WBX, MBX, and 66).  6.8.7 Pyrite trace elements – ore vectors Similar to epidote, the trace element composition of pyrite across the Mt. Milligan Main deposit also shows intra-system lateral and vertical trends despite frequently large compositional ranges in individual samples. 180  Elements with median concentrations that increase laterally outward and are maximal on the periphery of deep to intermediate levels include: Mn, As, Zn, Sb, La, Ti, Ba, and Ag. Elements that increase laterally, but show no vertical change include: Zr, Pb, Bi, Cd, Au, Tl, Th, U, Sn, Te, Cr, and W. Only the median trend for Co increases both laterally and vertically (Fig. 62). Elements that show an inward shift of maximum values from the periphery at depth towards the stock at shallower levels include Mo, Ni, and Se. The only element that systematically decreases away from the stock is Cu. The trace elements in pyrite that provide the best ore-vectors (based on a minimal spread in values about a median, abundance above detection limits, and the most systematic change in concentration spatially) include: Mn, As, Zr, Pb, and Bi. Each of these may be oxidized to a high valence. The fact that most trace elements in pyrite increase laterally toward the outer-propylitic assemblage, particularly elements with the potential to be at an oxidation state greater than +4 (Ti, Cr, Mn, As, Zr, W, Sn, Sb, Au, Pb, Bi, Th, U), suggests these elements remained in solution in an oxidized fluid, and due to their high valency were incompatible with the prevailing mineralogy. The increased substitution of these elements in pyrite on the system periphery suggests fluid reduction through widespread epidote alteration, with a resultant decrease in oxidation state. Furthermore, the general increase in value spread per sample toward the periphery reflects increased compositional heterogeneity, micro-inclusions, cryptic zoning, and pyrite rims enriched in incompatible elements (Fig. 69). A prolonged period of pyrite mineralization would generate the increased variability through shifting conditions of fluid-rock interaction (including changing fluid-rock compositions, ƒO2, ƒCO2, pH, and temperature). Moreover, if mixing of magmatic and meteoric fluids occurred, as indicated by Sr, C, and O isotopes, shifting parameters would be expected.  181  Of the suite of trace elements that increase laterally in concentration, each shows a return to lower median values across the Rainbow Fault, followed by a repetition of the increasing trend in the 66 zone. This indicates a more central pre-fault location of the 66 zone, and post-mineral normal faulting. Furthermore, for many elements (Tl, Cr, Sn, W, Mn, Sb, Te, Ti, As, Zr, Th, U) the laterally increasing spread in values per sample sharply decreases across the Rainbow Fault, and then resumes again. Alternately, elements that have a laterally decreasing trend toward the system periphery (Cu, Se) suddenly increase in concentration across the Rainbow Fault, again consistent with post-mineral normal faulting. Copper variability (or abundance of micro-inclusions), which is maximal closest to the MBX stock, also increases again in the hanging-wall of the 66 zone, consistent with the return to (calc)-potassic alteration as mapped in the present study. These combined characteristics of the pyrite trace element profiles provide additional proof that the 66 zone is a downdropped segment from a higher level of the paleo-hydrothermal system. An interesting prospect raised by the pyrite trace element analysis is the role of Te in the precipitation of Au and Ag in the DWBX zone, as the Au-Ag-Te spectra are commonly in parallel. Past workers have drawn attention to the association of Au-Ag tellurides with high Au/Ag ratios in alkalic epithermal deposits (Bonham and Giles, 1983; Mutschler et al., 1985) and have suggested that these grade downwards into alkalic porphyry-type Cu-Au systems, such as the Mt. Milligan Main deposit herein studied. The coincidence of a spike in Ag-Au-Te and Ba on the edge of a Cu-enriched pyrite grain rimmed by epidote (Fig. 69B) implies that Au-Ag telluride precipitated at a late stage from a fluid enriched in incompatible trace elements (including the large ion lithophile, Ba). Pressure-corrected fluid inclusion data from epidote in the same sample indicates temperatures of 355ºC (primary) and 300ºC (secondary) from CO2-bearing fluids. These temperatures, and the location of the telluride on the grain edge suggests Au solubility dominated by Au(HS)2- in an H2S enriched fluid (Gammons and Jones, 1997). The δ34Spyrite value of the sample is -0.19‰, indicating an approach to background (or reduced) conditions of pyritization in the outer-propylitic assemblage. 182  Figure 69: Examples of LA-ICP-MS time-resolved spectra for Mt. Milligan pyrite showing trace element counts-per-second. A) Internal variations (cryptic zoning) in pyrite of LILE-REE-HFSE incompatible elements (DDH 90-641 at 139 m). B) Red dashed ellipse outlines Ag-Au telluride with Ba spike (BaSO4 or BaCO3?) at pyrite grain edge rimmed by epidote (DDH 90-598 at 127 m). Fluid inclusions from the same sample indicate a secondary CO2-bearing fluid at 300ºC.  183  6.8.8 Trace elements in combined epidote-pyrite alteration Looking at trace elements common to the suite of analyzed elements for both pyrite and epidote (Mn, Cu, Zn, Zr, Mo, La, Au, Sn, Tl, Pb, Th, and U), partitioning of elements between the two minerals can be ascertained. Each of the trace elements has higher concentration in epidote, except for Au and As (approximately equal in both minerals), and Cu (more abundant in pyrite), reflecting the higher capacity of epidote to accommodate incompatible elements. Only five elements exhibit the same trends in both epidote and pyrite. Copper decreases in both minerals toward the system periphery, whereas Mn, Zr, Sb, and As increase. Zinc concentration increases laterally in both minerals in the DWBX zone, as does Tl in the MBX zone. The remainder of the trace elements behave antithetically in epidote and pyrite.  Each exhibits laterally decreasing (or uniform) concentration in  epidote paired with laterally increasing concentration in pyrite. This may reflect the preference in epidote for elements V, Mn, Zr, Ga, Sb, and As on the system periphery, precluding other incompatible elements, which were then partially incorporated in pyrite as oxidation state decreased through propylitization. 6.9 CONCLUSION Investigation of Sr-, S-, C-, and O-isotope systematics across the Mt. Milligan Main deposit has revealed geochemical dispersion halos that mirror the zonation of sulfide and alteration gangue assemblages mapped in Part 1 of the present study.  The  combined isotopic results suggest mineralization at the Mt. Milligan Cu-Au alkalic porphyry system was derived from a CO2-bearing magmatic fluid that either rose up through the MBX stock (forming K-feldspar cemented crackle breccia), or along it’s margins from a deeper source. Once in the MBX zone, the ore fluid formed an upwardwidening plume resulting in the (calc)-potassically altered Cu-Au ore shell, and then channelled along the Rainbow Dike until reaching Ca2+, Mg2+, and Fe2+ saturation (through loss of H+) with respect to carbonate (Zheng and Hoefs, 1992). This marks the vertical transition from potassic to carbonate-phyllic alteration, and from Cu-Au to Auonly mineralization. 184  Laterally outward from the Cu-Au mineralized (calc)-potassic shell, the magmatic fluid mixed with fluids of meteoric derivation resulting in the sodic-calcic, inner- and outerpropylitic shells. Gold abundance deceases gradually outward, and there is evidence for both chloro- and bisulfide complexing of Au within the pyrite-dominant sulfide assemblage, probably reflecting superposed Au-mineralization stages with decreasing temperature between 450-300ºC (Gammons and Jones, 1997). Shifts in the trace element abundances of epidote and pyrite across the Main deposit reflect lateral changes in hydrothermal fluid composition. Despite a high degree of variability in trace element concentration per sample, some elements show systematic behaviour that could be used to vector towards the system center. In epidote, these elements include: V, Mn, Ga, Sb, Zr, As (and possibly Bi). In pyrite, they include: Mn, As, Zr, Pb, and Bi. Additionally, fractionation of LREE in epidote toward the system core provides a robust ore-vector, and the height of positive Eu anomalies also appears to have a spatial trend. In pyrite, the general increase in trace element concentration toward the system periphery suggests incompatible elements (LILE-REE-HFSE) remained in solution to lower temperatures (<350ºC) until reduction through epidote alteration enabled substitution into epidote and pyrite mineral structures at a late stage. Gold precipitation as Ag-Au telluride is associated with this late stage in the outerpropylitic shell of the DWBX zone. The trends in median concentration of trace elements for both epidote and pyrite almost always show an offset across the Rainbow Fault that verifies post-mineral normal faulting, as proposed by Placer Dome Inc. 6.9.1 Recommendations for future work (and other conjectures) Much of the work contained in the present thesis was conducted as reconnaissance studies of various aspects of the Mt. Milligan Main deposit. It would be worthwhile to follow up these investigations to generate more complete picture of the system. •  The internal structure of the MBX stock should be investigated in greater detail to follow up observations made through core logging in this project. Interpolation between drill-holes 90-667 and 90-597 in the WBX zone has suggested a similar 185  attitude of composite monzonitic phases as the Rainbow Dike outside the stock. The inferred MBX breccia body appears to be similarly oriented. Furthermore, Cu/Au ratios associated with the downdropped monzonite body in the lower DWBX zone suggest the center of Cu mineralization was located somewhere in the hanging-wall of the pre-faulted DWBX zone, and that the monzonite body may not be a deeper segment of the MBX stock. Alternatively, this may be due to Cu leaching in the DWBX zone related to an early deep sodic-calcic alteration stage that has been largely overprinted. Relogging the entire stock, focused on identifying intrusive contacts, with the collection of magnetic susceptibility readings as an additional aid, could clarify: 1) the intrusive architecture and phase sequence; 2) the relationship of brecciation to intrusion within the stock; and 3) the pathway of mineralization. •  A recurring question regarding the Mt. Milligan system is why intrusion of the Rainbow Dike, potassic alteration, and related brecciation occurs paleohorizontally instead of vertically. Historically, these have been interpreted as having been controlled by stratigraphy (Sketchley et al, 1995). However, hydrothermal fluids are buoyant, and will tend to move vertically in a pressure gradient. Accordingly, it may be worthwhile to reconsider stratigraphic indicators. Perhaps the interpreted bedding of epiclastic horizons, or shear banding, does not indicate horizontality. This relates back to the architecture of the MBX stock. Is the system actually tilted and/or rotated closer to 90º? Alternately, the Rainbow Dike and related mineralization could represent a deeper flank of the MBX stock where the energy of resurgent boiling was less, such that dike intrusion and fluid release occurred along stratigraphic planes of weakness, such as a cone sheet, driven by internal overpressure of the magma (Burnham, 1979).  •  The role of hornblende- plagioclase-phyric diorite dikes needs to be further investigated. Their occurrence in DDH 90-652 and DDH 91-815 in the 66 zone is proximal to carbonate-phyllic alteration and elevated Au grade. It is tempting to interpret these late-mineral dikes as lamprophyres. Fractionation of 186  lamprophyric melts has been shown to produce alkalic magmas (McDonald et al., 1986; Leat et al., 1988), and these dikes could represent mantle-derived volatile-rich (H2O + CO2) source melts from the lower crust (McDonald et al., 1986). Furthermore, their conduits may coincide with those for deep-sourced Aubearing fluids (Muller and Groves, 1995). Their presence in the 66 zone raises speculation in regard to the Au source, which may be a separate event from CuAu mineralization in the MBX zone. Their proximity to the L1-stage vein (DDH 90-652) could explain the elevated S-isotope geothermometer result (496ºC). Alternatively, these may simply be late diorite dikes that provided permeable contacts for Au-bearing fluids diffusing outward from the Upper Trachyte. Geochemical analysis of the dikes could be undertaken to determine if they are of lamprophyric composition. Thermochemical modelling using MELTS (Ghiorso et al.) could show whether their fractionation product matches geochemistry of the least altered MBX stock and/or Rainbow dike. •  Fluid inclusion analyses from the Upper Trachyte and carbonate-phyllic altered 66 zone need to be obtained to complete the thermal-, salinity-, and fluid composition profile for the Main deposit.  These should confirm or deny the  occurrence of phase separation. Accompanying O- and D-isotope analysis of alteration minerals in the transition between potassic- and carbonate-phyllic assemblages, such as biotite and muscovite, would also be diagnostic (Harris and Golding, 2002). •  Additional analyses of C- and O-isotopes of carbonate across the entire deposit would aid in contouring magmatic versus meteoric fluid pathways. This could be accompanied by SWIR, XRD, and/or microprobe analyses of carbonate minerals to determine if the Mt. Milligan Main deposit is similar to the Red Chris Cu-Au porphyry where carbonate composition varies spatially and temporally (Baker et al., 1999). At Red Chris, the potassic zone is associated with siderite, the QSC zone with ankerite, and the chlorite zone with calcite.  187  •  Similarly, additional Sr-isotope analyses of Na-feldspar, actinolite, and epidote from the lateral alteration sequence would better define the magmatic to meteoric fluid contours. Accompanying O- and D-isotope analyses could help determine whether marginal fluids were isotopically-depleted meteoric water, or isotopically-enriched formation waters (after Bowman et al., 1987; Norman et al., 1991). It would be interesting to test whether the best fit line for all analyses in the isochron diagram remains within error of the Rainbow Dike isochron at 182.5 Ma, and to confirm the inferred loss and gain of  87  Sr and/or  87  Rb between Na-  feldspar and epidote. •  Increasing the sample size and sample density of epidote major- and trace element analyses would improve visualization of deposit-scale trends. LA-ICPMS sampling traverses across individual epidote and pyrite grains would provide quantitative evidence of trace-element zoning.  •  Additional cross-sections covering the east and north parts of the Rainbow Dike and MBX stock would complete the 3-dimensional picture of the MBX Main deposit. Related C- and S-isotope analyses would help constrain the source direction of primitive magmatic fluids, which in the current profile appears to be from the west.  •  A similar study could focus on the Cu-Au mineralized Southern Star stock.  188  REFERENCES Arnason, J.G., Bird, D.K., and Liou, J.G., 1993, Variables controlling epidote composition in hydrothermal and low-pressure regional metamorphic rocks: In Hock, V. and Koller, F. (eds.) 125 years Knappenwand – proceedings of a symposium held in Salzburg, Austria, p. 17-25. Bachinski, D.J., 1969, Bond strength and sulfur isotopic fractionation in coexisting sulfides: Economic Geology, v. 64, p. 56-65. Baker, T., Ash, C.H., Thompson, J.F.H., 1997, Geological setting and characteristics of Red Chris porphyry copper-gold deposits, northwestern British Columbia: Exploration and Mining Geology, v. 6, p. 297-316. Ballantyne, G.H., 1981, Chemical and mineralogical variations in propylitic zones surrounding porphyry copper deposits. PhD thesis (unpublished), University of Utah, p. 208. Baker, T., 2002, Emplacement depth and CO2-rich fluid inclusions in intrusion related gold deposits: Economic Geology, v. 97, p. 1109-1115. Barager, W.R.A and Irvine T.N., 1971, A guide to the chemical classification of the common volcanic rocks: Canadian Journal of Earth Sciences, v. 8, pp. 523-548. Barnes, H.L., Czamanske, G.K., 1967, Solubilities and transport of ore minerals: In Geochemistry of Hydrothermal Ore Deposits, ed. by H.L. Barnes, Holt (Rinehart and Winston), New York, p. 334-381. Barr, D.A., Fox, P.E., Northcote, K.E., and Preto, V.A., 1976, The alkaline suite of porphyry copper deposits – a summary: Canadian Institute of Mining and Metallurgy, Special Vol. 25, p. 359-367. Barrie, C.T., 1993, Petrochemistry of shoshonitic rocks associated with porphyry copper-gold deposits of central Quesnellia, British Columbia, Canada: Journal of Geochemical Exploration, v. 48, p. 225-258. Battles, D.A., and Barton, M.D., 1995, Arc-related sodic hydrothermal alteration in the western United States: Geology, v. 23, p. 913-916.  189  Bierlein, F.P., Arne, D.C., and Cartwright, I., 2004, Stable isotope (C, O, S) systematics in alteration haloes associated with orogenic gold mineralization in the Victorian gold province, SE Australia: Geochemistry: Exploration, Environment, Analysis, v. 4, p. 191-211. Bodnar, R.J., 1994, Synthetic fluid inclusions: XII. The system H2O–NaCl. Experimental determination of the halite liquidus and isochores for a 40 wt% NaCl solution: Geochimica et Cosmochimica Acta, v. 58, p. 1053–1063. Bonham Jr., H.F., and Giles, D.L., 1983, Epithermal gold/silver deposits – the geothermal connection: Geothermal Resources Council, p. 384. Bowman, J.R., Parry, W.T., Kropp, W.P., Kruer, S.A., 1987, Chemical and isotopic evolution of hydrothermal solutions at Bingham, Utah: Economic Geology, v. 82, p. 395-428. Burnham, C.W., 1979, Shallow emplacement and solidification of hydrous magmas: In Geochemistry of Hydrothermal Ore Deposits, 2nd ed., John Wiley & Sons, New York, p. 104-119. Burnham, C.W., 1979, Magmas and hydrothermal fluids. In Geochemistry of Hydrothermal Ore Deposits, 2nd ed., John Wiley & Sons, New York, p. 404-460. Burnham and Ohmoto, 1980, Late stage processes of felsic magmatism, In Granite Magmatism and Related Mineralization: Society of Mineral Geologists, Special Issue 8, p. 247. Burrows, D.R., Wood, P.C., Spooner, E.T.C., 1986, Carbon isotope evidence for a magmatic origin for Archaean gold-quartz vein ore deposits: Nature, v. 321, p. 851-854. Chang, H.Y., Lee, T.F., and Ejima, T., 1987, Effect of alkali-metal oxide on mold flux viscosity: Transactions of the Iron and Steel institute of Japan, v. 27, p. 797-804. Chouinard, A, Paquette, J., and Willianms-Jones, A.E., 2005, Crystallographic controls on trace element incorporation in auriferous pyrite from the Pascua epithermal high-sulfidation deposit, Chile-Argentina: Canadian Mineralogist, v. 43, p. 951-963. Cooke, D.R., 2006, Epidote trace element geochemistry and geochronology: Centre for Ore Deposit Research, University of Tasmania, p. 1-12. 190  Cooke, D.R., Heithersay, P.S., Wolfe, R., Calderon, A.L., 1998, Australian and western Pacific porphyry Cu-Au deposits: Journal of Australian Geology and Geophysics, v. 17, p. 97-104. Cooke, D.R., Crawford, T., and Cartwright, I., 1999, Narromine intrusive complex – petrology, whole rock geochemistry, alteration mineralogy and oxygen isotope systematics: In Origin and metallogenesis of Ordovician volcanic belts of central western NSW, p. 1-8. Corfield, R., 2003, Signatures of time: Chemistry World, No. 11, November. Craw, D., 1997, Fluid inclusion evidence for geothermal structure beneath the Southern Alps, New Zealand: New Zealand Journal of Geology and Geophysics, v. 40, p. 43-52. Delong, R.C., Godwin, C.I., Harris, M.W., Caira, N.M., and Rebagliati, C.M., 1991, Geology and alteration at the Mount Milligan gold-copper porphyry deposit, central British Columbia, In Geological Fieldwork 1990: British Columbia Ministry of Energy, Mines and Petroleum Resources Paper 1991-1, p. 199-206. Deer, W.A., Howie, R.A., and Zussman, J, 1992, An Introduction to the Rock Forming Minerals: Longman, London, p. 528. Delong, R.C., 1996, Geology, alteration, mineralization and metal zonation of the Mt. Milligan porphyry copper-gold deposits: M.Sc. thesis, University of British Columbia. Deyell, C.L., and Tosdal, R.M., 2005, Alkalic Cu-Au deposits of British Columbia: sulfur isotope zonation as a guide to mineral exploration: British Columbia Geological Survey, p. 191-208. Dilles, J.H., and Einaudi, M.T., 1992, Wall-rock alteration and hydrothermal flow paths about the Ann-Mason porphyry copper deposit, Nevada – a 6-km vertical reconstruction: Economic Geology, v. 87, p. 1963-2001. Dollase, W.A., 1968, Refinement and comparison of the structures of zoisite and clinozoisite: American Mineralogist, v. 53, p. 1882-1898. Douglass, S.E., and Campbell, A.R., 1994, Characterization of alkaline rock-related mineralization in the Nogal Mining District, Lincoln County, New Mexico, Economic Geology, v. 89, p. 1306-1321. 191  Eggleton, R.A., and Banfield, J.F., 1985, The alteration of granitic biotite to chlorite: American Mineralogist, v. 70, p. 902-910. Engvik, A.K., Bertram, A., Kalthoff, J.F., Stockhert, B., Austrheim, H., Elvevold, E., 2005, Magma-driven hydraulic fracturing and infiltration of fluids into the damaged host rock, an example from Dronning Maud Land, Antarctica: Journal of Structural Geology, v. 27, p. 839-854 Faure, G., 1977, Principles of Isotope Geology: John Wiley and Sons Inc., New York, p. 464. Franz, G., Liebscher, A., 2004, Physical and chemical properties of epidote minerals – An introduction: Reviews in Mineralogy and Geochemistry, v. 56, p. 1-82. Friedman, I., and O’Neil, J.R., 1977, Compilation of stable isotope fractionation factors of geochemical interest: In Data of Geochemistry, 6th ed., Fleischer, M. (ed.), United States Government Printing Office, Washington. Foley, S.F., Venturelli, G., Green, D.H., Toscani, L., 1987, The ultrapotassic rocks: characteristics, classification, and constraints for petrogenetic models: Earth Science Reviews, v. 24, p. 81-134. Fonseca, A., 2004, Report on diamond drilling on Mount Milligan property, NorthCentral British Columbia (NTS 93N/01, 93O/04) 124º04’ W longitude / 55º08’ N latitude: Placer Dome Inc. internal report. Fonseca, A., 2005, Mt. Milligan Deposit, background paper: Placer Dome Inc. internal report. Fonseca, A., 2005, Mt. Milligan alteration modelling project, petrographic report: Placer Dome Inc. internal report. Fonseca, A., 2006, Mt. Milligan Cu-Au porphyry: Placer Dome Inc. poster at the Association for Mineral Exploration B.C. Roundup Conference, Vancouver. Fournier, R.O., 1999, Hydrothermal processes related to movement of fluid from plastic into brittle rock in the magmatic-epithermal environment: Economic Geology, v. 94, p. 1193–1211. Frei, D., Liebscher, A., Franz, G., Dulski, P., 2004, Trace element geochemistry of epidote minerals: Reviews in Mineralogy and Geochemistry, v. 56, p. 553-605.  192  Gammons, C.H., and Williams-Jones, A.E., 1997, Chemical mobility of gold in the porphyry-epithermal environment: Economic Geology, v. 92, p. 45-59. Garnett, J.A., 1978, Geology and mineral occurrences of the southern Hogem Batholith: B.C. Ministry of Energy, Mines, and Petroleum Resources, Bulletin 70, p. 75. Guilbert, J.M., and Park, C.F., Jr., 1986, The Geology of Ore Deposits: New York, Freeman, p. 985. Gustafson, L.B., Hunt, J.P., 1975, The porphyry deposit at El Salvador, Chile: Economic Geology, v. 70, p. 857-912. Harris, A.C., and Golding, S.D., 2002, New evidence of magmatic-fluid-related phyllic alteration: Implications for the genesis of porphyry Cu deposits: Geological Society of America, v. 30, p. 335-338. Hedenquist, J.W., and Henley, R.W., 1985, The importance of CO2 on freezing point measurements of fluid inclusions: evidence from active geothermal systems and implications for epithermal ore deposition: Economic Geology, v. 80, p. 1379-1406. Henley, R.W., and McNabb, A., 1978, Magmatic vapour plumes and ground-water interactions in porphyry copper emplacement: Economic Geology, v. 73, p. 1-20. Holland, H.D., 1972, Granites, solutions, and base metal deposits: Economic Geology, v. 67, p.281-301. Holland, H.D., and Malinin, S.D., 1979, The solubility and occurrence of non-ore minerals: In Barnes, H.L. (ed.) Geochemistry of Hydrothermal Ore Deposits, 2nd ed., Wiley, New York, p. 461-508. Holliday, J.R., and Cooke, D.R., 2007, Advances in geological models and exploration methods for copper ± gold porphyry deposits: In Proceedings of Exploration 07: Fifth Decennial International Conference on Mineral Exploration, p. 791-809. Huston, D.L., and Large, R.R., 1989, A chemical model for distribution of gold in volcanogenic massive sulphide deposits: Ore Geology Reviews, v. 4, p. 171-200. Jackson, M., 2008, The relationship between brecciation, intrusion, and mineralization, Northeast Zone, Mount Polley Mine, British Columbia: M.Sc. thesis, University of British Columbia.  193  Jenks, G.F., and Caspall, F.C., 1971, Error on Choroplethic Maps. Definition, Measurement, Reduction: In Annals of the Association of American Geographers. v. 61. p. 217-244. Jones, B.K., 1992, Application of metal zoning to gold exploration in porphyry copper systems: Journal of Geochemical Exploration, v. 43, p. 127-155. Kajiwara, Y. and Krause, J., 1971, Sulphur isotope partitioning in metallic sulphide systems: Canadian Journal of Earth Sciences, v. 8, p. 1397-1408. Kennedy, A.K., Grove, T.L., Johanson, R.W., 1990, Experimental and major element constraints on the evolution of lavas from Lihir Island, Papua New Guinea: Journal of Geophysical Research, v. 95, p. 6929-6942. Kesler, S.E., 2005, Ore-forming fluids: Elements, v. 1, p. 13-18. Kesler, S.E., Jones, L.M. and Walker, R.L., 1975, Intrusive rocks associated with porphyry copper mineralization in island-arc areas: Economic Geology, v. 70, p. 515-526. Lang, J.R., 1992, Mount Milligan Geochemistry: MDRU Porphyry Cu-Au Project, Year 1, Section 7, p. 2-49. Lang, J.R., Lueck, B., Mortensen, J.K., Russell, J.K., Stanley, C.R., Thompson, J.F.H., 1995, Triassic-Jurassic silica-undersaturated and silica-saturated alkalic intrusions in the Cordillera of British Columbia: Implications for arc magmatism: Geology, v. 23, p. 451-454. Lang, J.R., Stanley, C.R., Thompson, J.F.H., Dunne, K.P.E., 1998, Na-K-Ca magmatichydrothermal alteration in alkalic porphyry Cu-Au deposits, British Columbia: B.C. Ministry of Energy and Mines Open File 1998-8, p. G3-G30. Lang, J.R., Thompson, J.F.H. and Stanley, C.R. (1998): Alkalic Cu-Au-Ag Porphyry Deposits in the Canadian Cordillera: Tectonic Setting, Magmatic Affiliations and Hydrothermal Characteristics: In Metallogeny of Volcanic Arcs, B.C. Geological Survey, Short Course Notes, Open File 1998-8, Section G. Le Bas, M., Le Maitre, R.W., Strekeisen, A., Zanettin, B., 1986, A chemical classification of volcanic rocks based on the total alkali-silica diagram: Journal of Petrology, v. 27, p. 745-750.  194  Leat, P.T., Thompson, R.N., Morrison, M.A., Hendry, G.L., Dicken, A., 1988, Silicic magmas derived by fractional crystallization from Miocene minette, Elkhead Mountains, Colorado: Mineralogical Magazine, v. 52, p. 577-585. LeMaitre, R.W, ed., 1989, A Classification of Igneous Rocks and Glossery of Terms: Blackwell Scientific Publications, London, U.K. Lickfold, V., 1992, Intrusive history and volatile evolution of the Endeavour porphyry CuAu deposits , Goonumbla district, NSW, Australia. Ph.D. thesis, University of Tasmania, Hobart, p. 243. Loucks, R.R., and Mavrogenes, J.A., 1999, Gold solubility in supercritical hydrothermal brines measured in synthetic fluid inclusions: Science, v. 284, p. 2159-2163. Lowell, D.J, and Guilbert, J.M., 1970, Lateral and vertical alteration- mineralization in porphyry ore deposits: Economic Geology; July 1970; v. 65, p. 373-408. Lusteg, G., 2006, Technical Report on the Mt. Milligan Project: Omineca mining district, British Columbia: G.N. Lusteg Consulting Ltd., p. 62. Macdonald, G.A., 1968, Composition of Hawaiian lavas: Mem. Geological Society of America, v. 116, pp. 477–522. Macdonald, R., Rock, N. M. S., Rundle, C. C., Russell, O. J., 1986, Relationships between late Caledonian lamprophyric, syenitic, and granitic magmas in a differentiated dike, southern Scotland: Mineralogical Magazine, v. 50, p. 547−557. Mathews, W.H., and Monger, J.W.H, 2005, Roadside geology of Southern British Columbia: Mountain Press, Missoula, Montana, p. 403. McDonough, W.F. and Sun S., 1995, The composition of the Earth: Chemical Geology, v. 120, pp. 223–253. McMillan, W.J, and Panteleyev, A., 1988, Porphyry Copper Deposits: In Ore Deposits Models, Geoscience Canada, Reprint Series 3, p. 45-59. McPhie, J., Doyle, M., Allen, R., 1993, Volcanic Textures: A guide to interpretation of textures in volcanic rocks: CODES, Hobart, Tasmania, p. 198. Meyer, C. and Hemley, J.J., 1967, Wallrock alteration. In Geochemistry of Hydrothermal Ore Deposits, ed. by H.L. Barnes, Holt (Rinehart and Winston), New York, p. 166–235. 195  Mihalynuk, M.G., Nelson, J., Diakow, L.J., 1994, Cache Creek terrane entrapment: Oroclinal paradox within the Canadian Cordillera: Tectonics, v. 13, p. 575-595. Miller, M.M., 1987, Displaced remnants of a northeast Pacific fringing arc, Upper Paleozoic terranes of Permian McCloud faunal affinity, western US: Tectonics, v. 6, 807-830. Miyashiro, A., and Seki, Y., 1958, Enlargement of the composition field of epidote and piemontite with rising temperature: American Journal of Science, v. 256, p. 423-430. Monger, J.W.H., 1991, Wheeler, J.O., Tipper, H.W., Gabrielse, H., Harms, T., Struik, L.C., Campbell, R.B., Dodds, C.J., Gehrels, G.E., O’Brien, J., 1991, Cordilleran terranes. In Geology of the Cordilleran Orogen in Canada, Geology of Canada, v. 4. Edited by H. Gabrielse and C.J. Yorath, p. 281-327. Mortensen, J.K., Ghosh, D.K., Ferri, F., 1995, U-Pb geochronology of intrusive rocks associated with copper-gold porphyry deposits in the Canadian Cordillera: Canadian Institute of Mining and Metallurgy, Special Vol. 46, p. 142-158. Mortimer, N., 1987, The Nicola Group: Late Triassic and Early Jurassic subductionrelated volcanism in British Columbia: Canadian Journal of Earth Sciences, v. 24, p. 2521-2536. Muller, D., and Groves, D.I., 1995, Indirect associations between lamprophyres and gold-copper deposits: In Potassic Igneous Rocks and Associated Gold-Copper Mineralization, Springer, Berlin, v. 56. p. 135-158. Murphy, D.C., Gerasimoff, M., van der Heyden, P., Parrish, R.R., Klepacki, D.W., McMillan, W., Struik, L.C., Gabites, J., 1995, New geochronological constraints on Jurassic deformation of the Western Edge of North America, Southeastern Canadian Cordillera, In Jurassic Magmatism and Tectonics of the North American Cordillera: Geological Society of America, Special Paper 299, p. 159-172. Mutschler, F.E., Griffin, M.E., Stevens, D.S., and Shannon Jr., S.S., 1985, Precious metal deposits related to alkaline rocks in the North American Cordillera – an interpretive review: Geological Society of South Africa, Transactions, v. 88, p. 355-377. 196  Nelson, J.L., and Bellefontaine, K.A., 1996, The geology and mineral deposits of NorthCentral Quesnellia; Tezzeron Lake to Discovery Creek, Central British Columbia: British Columbia Ministry of Employment and Investment, Geological Survey Branch, Bulletin 99, p. 100. Nelson, J.L., Bellefontaine, K., Rees, C., MacLean, M., 1992, Regional geological mapping in the Nation Lakes area: B.C. Ministry of Energy, Mines, and Petroleum Resources, Geological Fieldwork 1991, p. 103-118. Ney, C.S., Hollister, V.F., 1976, Geological setting of porphyry deposits of the Canadian Cordillera: Canadian Institute of Mining and Metallurgy, Special Vol. 15, p. 21-29. Nixon, G.T., 1993, 40Ar-39Ar and U-Pb geochronometry of the Polaris Alaskan-type complex, British Columbia: precise timing of Quesnellia-North America interaction. Geological Association of Canada / Mineralogical Association of Canada, Program and Abstracts Joint Annual Meeting, Edmonton, P. A76. Nixon, G.T, and Peatfield, G.R, 2003, Geological setting of the Lorraine Cu-Au porphyry deposit, Duckling Creek Syenite Complex, North-central British Columbia: B.C. Ministry of Mines and Energy, Open File Report 2003-4, p. 24. Norman, D.K., Parry, W.T., Bowman, J.R., 1991, Petrology and geochemistry of propylitic alteration at Southwest Tintic, Utah: Economic Geology, v. 86, p. 13-28. Ohmoto, H., 1972, Systematics of sulfur and carbon isotopes in hydrothermal ore deposits: Economic Geology, v. 67, p. 551-578. Ohmoto, H., 1986, Stable isotope geochemistry of ore deposits: Reviews in Mineralogy and Geochemistry, v. 16, p. 491-559. Ohmoto, H., and Goldhaber, M.B., 1997, Sulfur and carbon isotopes, In Geochemistry of Hydrothermal Ore Deposits, 3rd edition: John Wiley & Sons, New York, p. 517-612. Ohmoto, H., and Rye, R. O., 1979, Isotopes of sulfur and carbon: In Barnes, H. L. Ed., Geochemistry of Hydrothermal deposits, John Wiley & Sons, p. 509-567 Oldenberg, D.W., Li, Y., Ellis, R.G., 1997, Inversion of geophysical data over a copper gold porphyry deposit: A case history for Mt. Milligan: Geophysics, v. 62, p. 1419-1431. 197  Pearce, J.A., Harris, N.B.W., Tindle, A.G., 1984, Trace element discrimination diagrams for the tectonic interpretation of granitic rocks: Journal of Petrology, v. 25, p. 956-983. Peterman, Z.E., Hedge, C.E., Tourtelot, H.A., 1970, Isotopic composition of strontium in sea water throughout Phanerozoic time: Geochemica et Cosmochemica Acta, v. 34, p. 105-120. Pokrovski, G., Zakirov, I., Roux, J, 2000, Arsenic speciation and transport in hydrothermal vapour and low-density crustal fluids: Goldschmidt 2000, Journal of Conference Abstracts, v. 5, p. 807. Rebagliati, C.M., 1988, Mt. Milligan property assessment report, Omineca mining division NTS 93 /1, latitude 55º08’ N, longitude 124º04’ W: for United Lincoln Resources, Inc. Richards, J.P., 1995, Alkalic-type epithermal gold deposits – a review: Mineralogical Association of Canada Short Course Series, v. 23, p. 367–400. Robinson, B.W., and Kusakabe, M., 1975, Quantitative preparation of sulphur dioxide, for 34S/32S analyses, from sulphides by combustion with cuprous oxide: Analytical Chemistry, v. 47, p. 1179-1181. Rollinson, H.R., 1993, Using carbon isotopes: In Using Geochemical Data: Evaluation, Presentation, Interpretation: Longman Press. Rosenbaum, J. and Sheppard, S.M.F., 1986, An isotopic study in siderites, dolomites and ankerites at high temperatures: Geochemica et Cosmochemica Acta, v. 50, p. 1147-1150. Rosenberg, P.E., and Larsen, P.B., 2000, Isotope geochemistry of ankerite bearing veins associated with Coeur D’Alene mining district, Idaho: Economic Geology, v. 95, p. 1689-1699. Rowins, S. M. Groves, D. I., McNaughton, N. J., Palmer, M. R., and Eldridge, C. S., 1997, A reinterpretation of the role of granitoids in the genesis of Neoproterozoic gold mineralization in the Telfer Dome, Western Australia: Economic Geology, v. 92, p. 133-160. Rye, R.O, and Ohmoto, H., 1974, Sulfur and carbon isotopes and ore genesis: a review: Economic Geology, v. 69, p. 826-842. 198  Sakai, H., 1968, Isotopic properties of sulfur compounds in hydrothermal processes: Journal of Geochemistry, v. 2, p. 29-49. Seedorff, E., Dilles, J.H., Proffett, J.M., Einaudi, M.T., Zurcher, L., Stavast, W.J.A., Johnson, D.A., Barton, M.D., 2005, Porphyry Deposits: Characteristics and Origin of Hypogene Features: Economic Geology: 100th Anniversary Volume, p. 251-298. Shepherd, T.J., Rankin, A.H., and Alderton, D.H.M., 1985, A practical guide to fluid inclusion studies: Blackie, London, p. 191-211. Shima, M., Gross, W. H., and Thode, H. G, 1963, Sulphur isotope abundances in basic sills differentiated granites and meteorites: Journal of Geophysical Research, v. 68, p. 2838-2847. Shinohara, H., and Hedenquist, J.W., 1997, Constraints on magma degassing beneath the Far Southeast porphyry Cu-Au deposit, Philippines: Journal of Petrology, v. 38, p. 1741-1752. Simmons, S.F., Christenson, B.W., 1994, Origins of calcite in a boiling geothermal system: American Journal of Science, v. 294, p. 361-400. Sketchley, D.A., Rebagliati, C.M., Delong, C., 1995, Geology, alteration and zoning patterns of the Mt. Milligan copper-gold deposit: Canadian Institute of Mining and Metallurgy, Special Vol. 46, p. 650-665. Spear, F.S., 1993, Metamorphic phase equilibria and pressure-temperature-time paths: Mineralogical Society of America, Washington, D.C., p. 799. Stanley, C.R., 1992, Porphyry copper-gold systems of British Columbia: MDRU Annual Technical Report – Year 1, p. 13.2-13.30. Taylor, B.E., 1974, The application of oxygen and hydrothermal isotope studies to problems of hydrothermal alteration and ore deposition: Economic Geology, v. 69, p. 843-883. Thompson, J.B. Jr., 1986, Compositional space; an algebraic and geometric approach: Mineralogical Society of America Reviews in Mineralogy, v. 10, p. 1-31. Titley, S.R. and Beane, R.E., 1981, Porphyry copper deposits: Economic Geology 75th Anniversary Volume, p. 214-269.  199  White, W.M., 2006, Radiogenic isotope geochemistry: In Geochemistry, an on-line textbook, John-Hopkins University Press. Wilson, A.J, 2003, The geology, genesis, and exploration context of the Cadia goldcopper porphyry deposits, New South Wales, Australia: PhD thesis, University of Tasmania. Wilson, A.J., Cooke, D.R., Harper, B.J., 2003, The Ridgeway gold-copper deposit: a high-grade alkalic porphyry deposit in the Lachlan Fold Belt, New South Wales, Australia: Economic Geology, v. 98, p. 1637-1666. Wilson, A.J., Cooke, D.R., Harper, B.J., Deyell, C.L., 2006, Sulfur isotope zonation in the Cadia district, southeastern Australia: exploration significance and implications for the genesis of alkalic porphyry gold-copper deposits: Mineralium Deposita, v. 42, p. 465-487. Wilson, M., 1995, Magmatic differentiation: Geological Society of London Memoirs, v. 16, p. 205-218. Yamagishi, H., 1987, Studies on the Neogene subaqueous lavas and hyaloclastites in southwest Hokkaido: Report of the Geological Survey of Hokkaido v. 59, p. 117. Zheng, Y.F., and Hoefs, J., 1992, Carbon and oxygen isotopic covariations in hydrothermal calcites: Mineralium Deposita, v. 28, p. 79-89.  200  APPENDIX 1: MT. MILLIGAN MAIN DEPOSIT VOLCANIC STRATIGRAPHY AND ALTERATION  201  202  203  204  APPENDIX 2: SULFUR ISOTOPIC COMPOSITIONS EXPRESSED AS δ34S VALUES FOR SULFIDE FROM THE MT. MILLIGAN CU-AU ALKALIC PORPHYRY DEPOSIT  205  206  207  208  209  210  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0052588/manifest

Comment

Related Items