Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Alunite and high sulfidation gold-silver-copper mineralization in the El Indio-Pascua Belt, Chile-Argentina 2002

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

Item Metadata

Download

Media
ubc_2002-731510.pdf [ 16.06MB ]
Metadata
JSON: 1.0052344.json
JSON-LD: 1.0052344+ld.json
RDF/XML (Pretty): 1.0052344.xml
RDF/JSON: 1.0052344+rdf.json
Turtle: 1.0052344+rdf-turtle.txt
N-Triples: 1.0052344+rdf-ntriples.txt
Citation
1.0052344.ris

Full Text

ALUNITE AND HIGH SULFIDATION GOLD-SILVER-COPPER MINERALIZATION IN THE EL INDIO-PASCUA BELT, CHILE-ARGENTINA by CARI L. D E Y E L L B.Sc.H., Queen's University, 1995 A thesis submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Department of Earth and Ocean Sciences) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December, 2001 ©Cari L. Deyell, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of rxir4s\*~ CiC&fJ^ £ r x e W , \ The University of British Columbia Vancouver, Canada DE-6 (2/88) A B S T R A C T High sulfidation precious-metal deposits, one of the principal styles of epithermal mineralization, are characterized by minerals diagnostic of a high sulfidation state (pyrite, enargite) and acidic hydrothermal conditions, particularly alunite (KAl3(SC>4)2(OH)6) and alunite-group minerals. One of the world's foremost high sulfidation districts, the El Indio- Pascua Belt, straddles the Chile-Argentina border in the Cordillera Principal of the Andes. The region contains the world-class El Indio mine, the extensive Pascua-Lama and Veladero exploration projects, and several smaller prospects and mineralized alteration zones. This belt has been the subject of numerous studies at both a regional- and deposit-scale, and provides an excellent site to constrain genetic and exploration models for high sulfidation deposits. The genesis and timing of high sulfidation deposits in the El Indio-Pascua Belt has been constrained by detailed field studies, supplemented by geochemical, stable isotopic, and 4 0 Ar- 39AT analyses of alunite, a critical mineral throughout the paragenesis of most systems in this region. Widespread barren, pre-mineral alteration containing abundant alunite may have provided permeability controls and metals for subsequent mineralization. Economic mineralization in the district formed between ca. 6 to 9.5 Ma, and is associated commonly with syn-mineral alunite. This alunite has textural, chemical and isotopic characteristics consistent with a magmatic-hydrothermal origin and requires the presence of relatively oxidizing, acidic, and sulfur-rich mineralizing fluids. Data suggest that these fluids were derived from the condensation of a magmatic vapor plume, further supported by the presence of syn-ore magmatic steam alunite at the Tambo deposit. This conclusion is consistent with emerging evidence from other systems indicating the importance of vapor transport for metals in high sulfidation systems. Alteration and mineralization in the El Indio-Pascua Belt were dominated by magmatic fluids, even in the near-surface environment. The negligible contribution of meteoric fluids is attributed to prolonged magmatic activity, an arid climate, and depressed water tables resulting from tectonic uplift and repeated incision. These features provide new constraints for the genetic model for high sulfidation deposits, although the specific setting of the El Indio-Pascua Belt may have provided a unique combination of processes explaining the formation and preservation of these prolific mineralizing systems. ii T A B L E OF CONTENTS Abstract ii Table of Contents ii i List of Tables vii List of Figures ix Acknowledgements xiii Chapter 1: Introduction High Sulfidation Deposits 1 The E l Indio-Pascua Belt 3 The Role of Alunite 6 Scientific Contributions of this Study 9 Presentation 10 References 12 Chapter 2: The Nature of Hydrothermal Alteration at the Pascua-Lama High Sulfidation Deposit; Chile-Argentina Introduction 17 Geological Framework 20 Mineralization 21 Alteration: Characteristics and Distribution 24 Methods of identification 24 Alteration Types 25 Distribution of Alteration 32 Alunite Geochemistry. 37 Stable Isotope Study 47 Sulfur Isotope Relations 47 Oxygen Isotope Relations 51 Geochronology 56 Discussion: Controls on Alteration 58 Summary: Alteration Paragenesis in an Evolving Hydrothermal System 60 References 64 in Chapter 3: Alunite in an Evolving Magmatic-Hydrothermal System; The Tambo High Sulfidation Deposit, El Indio District, Chile Introduction 70 Geological Framework 72 Regional geology 72 Local geology 73 Mineralization 73 Methods of Analysis 76 Alunite Paragenesis in the Tambo system 81 Geochronology 85 Stable Isotope Study 87 Geochemistry of Alunite 97 Fluid Inclusion Gas Chemistry 106 Discussion 111 Role of Magmatic Fluids: In Alteration 112 Role of Magmatic Fluids: In Mineralization 113 Origins of Magmatic Steam 114 Summary 118 References 119 Chapter 4: The Role of Alunite in Exploration: Evidence from High Sulfidation Deposits in the El Indio-Pascua Belt, Chile Introduction 129 Geology of the El Indio-Pascua Belt 131 Methods of Analysis 134 Environments of Acid Sulfate Alteration 138 1. Magmatic-Hydrothermal Environment 138 2. Magmatic Steam Environment 153 3. Steam-Heated Environment 159 4. Supergene Environment 163 Discussion: Implications for Exploration 168 References 173 iv Chapter 5: Alunite-Natroalunite Stability Relations Introduction 179 Stability Relations in the K^O/NazO-AbOs-SiCV^O-SOs System. 180 Effects of Fluid Composition on the Alunite-Natroalunite Solid Solution 188 Controls on Alunite-Natroalunite Compositions in the El Indio-Pascua Belt 196 Depositional Temperature 196 Fluid Composition 198 Summary 202 References 202 Chapter 6: Formation of High Sulfidation Precious-Metal Deposits: New Evidence from the El Indio-Pascua Belt, Chile. Introduction 206 The High Sulfidation model 207 The El Indio-Pascua Belt 208 High Sulfidation Systems in the E l Indio-Pascua Belt 211 Pre-Mineral Alteration 211 Mineralization 217 Post-mineralization and peripheral alteration 217 Discussion: Comparison to Other High Sulfidation Systems 219 The origin and importance of early alteration 219 Mineralizing processes 219 Magmatic-hydrothermal to supergene transition 223 The near-surface environment 224 Constraints on the High Sulfidation Model 224 References 226 Appendix A: Geochemical Analysis A . l Analytical techniques: Electron Microprobe ......232 Analytical parameters 232 Analytical standard 237 Methods of recalculation 239 Detection limits 240 V Error estimation 240 Statistical analysis 241 A 2 . A N A L Y T I C A L M E T H O D S : Geochemistry of Alunite Separates 243 Sample preparation 243 Analysis 243 Error estimates 243 References 253 Appendix B: Stable Isotope Analysis Sample Preparation 254 Analytical Methods 254 Sample Descriptions and Results 256 References 264 Appendix C: 4 0 Ar- 3 9 Ar Geochronology Analytical Methods 265 Analytical Data 265 References 271 Appendix D: Thermodynamic Data and Calculations D. 1. pH-log fD 2 Diagram 272 Au solubility as Au(HS)i 272 Au solubility as Au(HS)° 275 Solubility of Au as AuCli 276 D.2. Alunite-natroalunite stability relationships 276 Activity diagrams 276 Solid solution models 278 References 280 Appendix E: PIMA Analysis and Interpretation Methods of Analysis 282 Pascua-Lama Alteration Study 282 vi L I S T O F T A B L E S Table 1.1. Alteration assemblages; mineralogy and occurrence in high sulfidation 2 systems. Table 1.2. Minerals of the alunite supergroup (with formulas). 6 Table 2.1. Selected minerals of the alunite-jarosite family (with formulas). 38 Table 2.2. Summary of E P M A and trace element ICP-MS data for Pascua alunite 39 samples. Table 2.3. Whole rock geochemical data for Colorado Unit. 43 Table 2.4. Sample descriptions and stable isotope results for all Pascua samples. 48 Table 2.5. 4 0 Ar / 3 9 Ar data for alunite and jarosite from Pascua. 57 Table 3.1. Paragenetic sequence and characteristics of 8 stages of Tambo alunite. 77 Table 3.2. Average whole rock geochemical data for unaltered rocks of the Tilito 79 Formation. Table 3.3. Minerals of the alunite-jarosite family identified in the Tambo area. 84 Table 3.4. 4 0 Ar / 3 9 Ar data for Tambo alunite. 86 Table 3.5. Sample descriptions and stable isotope results for all Tambo samples. 88 Table 3.6. Summary of E P M A and trace element ICP-MS data for Tambo alunite 99 samples. Table 3.7. Fluid inclusion gas chemistry data for Tambo alunite (Stages 2, 3, and 108 banded veins). Table 4.1. Minerals of the alunite supergroup. 136 Table 4.2. Characteristics of the 4 environments of acid sulfate alteration in the El 139 Indio-Pascua Belt. Table 4.3. Physical, geochemical, and isotopic characteristics of alunite in the El 140 Indio-Pascua Belt. Table 4.4. Summary of E P M A data for E l Indio-Pascua Belt alunite according to 144 alteration type (magmatic-hydrothermal, magmatic steam, steam-heated, late stage). Table 4.5. Variation in sulfur isotope data for magmatic steam alunite. 158 Table 5.1. Thermodynamic properties of natroalunite at 25°C and 1 bar. 188 vii Table 5.2. Estimates of In K and Margules parameters for alunite-natroalunite mixing 191 reaction. Table 5.3. Statistical comparison of Del Carmen alunite E P M A data versus rest of El 200 Indio-Pascua Belt. Table 5.4. Average major element geochemistry for host lithologies from Del 201 Carmen, Tambo, and Pascua areas. Table 6.1. Geological characteristics of high sulfidation deposits and prospects in the 212 E l Indio-Pascua Belt. Table 6.2. Characteristics of ore and hypogene alteration assemblages in the E l Indio- 213 Pascua Belt. Table 6.3. Characteristics of late stage and near-surface processes in the El Indio- 214 Pascua Belt. viii L I S T O F F I G U R E S Figure 1.1 Location map of E l Indio-Pascua Belt showing regional tectonic features 4 and major deposits and prospects. Figure 1.2 Schematic diagram showing environments of acid sulfate alteration. 8 Figure 2.1 Regional geology of the Pascua-Lama district. 18 Figure 2.2 Surficial alteration: Pascua-Lama region. 19 Figure 2.3 Outcrop and SEM photos of APE-type mineralization. 23 Figure 2.4 Photomicrographs of each paragenetic stage of alunite at Pascua. 26 Figure 2.5 Section CA-EW. Distribution of alteration. 33 Figure 2.6 Section CA-00. Distribution of alteration, major lithological units, Au 34 mineralization, and PIMA sample points. Figure 2.7 Section CA-EW: Esperanza area. Distribution of alteration, major 35 lithological units, Au mineralization, and PIMA sample points. Figure 2.8 Section CA-EW: Frontera zone. Distribution of alteration, major 36 lithological units, Au mineralization, and PIMA sample points. Figure 2.9 Summary of EPMA data: K vs Na for all stages of alunite. 41 Figure 2.10 REE data for Pascua alunite normalized to host granite. 42 Figure 2.11 Range of K/Na ratios for AS I and AS II alunite vs elevation. 46 Figure 2.12 8 3 4S data for Pascua alunite, jarosite, barite, and sulfides. 49 Figure 2.13 8 3 4 S, 8 1 8 0 , and 8D relations for AS I, AS II, and Esperanza alunite. 52 Figure 2.14 8 3 4 S, 8 1 8 0 , and 8D relations for steam-heated, late vein, AS III alunite 53 and jarosite. Figure 2.15 Schematic sections showing evolution of Pascua system from pre-, to 61 syn-, to post-mineral processes. Figure 2.16 Thermochemical model of the Pascua APE mineralizing event in terms 63 of pH and the fugacity of O2. Figure 3.1 Geology of the El Indio district. 71 Figure 3.2 Map of Tambo property. 74 Figure 3.3 Schematic section of Tambo deposit showing distribution of alteration. 75 Figure 3.4 Photos and photomicrographs for all paragenetic stages of Tambo 82 alunite. ix Figure 3.5 Range in 8 3 4S data for Tambo alunite, barite, and sulfides. 89 Figure 3.6 8 3 4 S, 8 1 8 0 , and 6D relations for Stage 1, Stage 2, and Bx Sylvestre 91 alunite. Figure 3.7 8 3 4 S, 8 1 8 0 , and 8D relations for Stage 3 and Banded vein alunite. 93 Figure 3.8 8 3 4 S, 8 1 8 0 , and 8D relations for steam-heated, late vein, and Ca-vein 95 alunite. Figure 3.9 REE data for all stages of Tambo alunite normalized to host rock 102 compositions. Figure 3.10 K-Na-Ba ternary plot for Stage 2 alunite - based on E P M A data. 104 Figure 3.11 K+Na versus Ca for huangite-bearing vein alunite - based on E P M A 104 data. Figure 3.12 (a) Ternary H2S-SO2-HCI data from fluid inclusion analyses of Tambo 110 alunite. (b) Ternary HCI-HF-H2 data from fluid inclusion analyses of Tambo alunite. Figure 3.13 Schematic section of Cerro Elefante depicting the proposed model for 116 (A) Stage 2 alteration and Au ore formation and (B) magmatic steam processes. Figure 3.14 Phase relations in the system NaCl-EbO - used to illustrate origins of 117 magmatic steam alunite. Figure 4.1 Schematic diagram of a high sulfidation system showing environments 130 of acid sulfate alteration. Figure 4.2 Simplified geology of E l Indio-Pascua Belt. 132 Figure 4.3 Photos and photomicrographs of typical magmatic-hydrothermal 142 alteration. Figure 4.4 Photo of Del Carmen alunite with visible zoning. 142 Figure 4.5 Histogram showing range of alunite Na E P M A data for different origins 145 of acid sulfate alteration. Figure 4.6 Ternary diagrams showing range of compositions (K-Na-Ba and K-Na- 145 Ca) for magmatic-hydrothermal alunite. Figure 4.7 Photomicrographs showing compositional zoning in magmatic- 146 hydrothermal alunite. Figure 4.8 SEM-EDS element maps: distribution of elements in Wendy alunite. 147 x J Figure 4.9 Examples of PIMA spectra for magmatic-hydrothermal alunite. 149 Figure 4.10 Range of REE data for magmatic-hydrothermal alunite, normalized to 149 host rock compositions. Figure 4.11 8 3 4 S, 8 1 8 0 , and 8D relations for magmatic-hydrothermal alunite and 150 associated sulfides. Figure 4.12 Photo of fluid inclusion in Pascua alunite. 152 Figure 4.13 Ternary H2S-SO2-HCI data from fluid inclusion analyses of Tambo 152 alunite. Figure 4.14 Photo of magmatic steam alunite (Tambo deposit). 154 Figure 4.15 Photomicrograph showing compositional zoning in magmatic steam 156 alunite. Figure 4.16 Range of REE data for magmatic steam alunite. 156 Figure 4.17 8 3 4 S, 8 1 8 0, and 8D relations for magmatic steam alunite. 157 Figure 4.18 Photo of steam-heated alteration, Tambo deposit. 160 Figure 4.19 SEM photomicrograph of typical steam-heated assemblage. 160 Figure 4.20 REE data for steam-heated alunite. 162 Figure 4.21 S 1 8 0 , and 8D relations for steam-heated and late stage alteration. 162 Figure 4.22 Photo showing late stage alunite-jarosite alteration, Tambo deposit. 165 Figure 4.23 SEM photomicrographs of late stage alteration. 165 Figure 4.24 REE data for late stage alteration, normalized to host rock compositions. 167 Figure 4.25 8 3 4S and 8 1 8 0 relations for late stage alunite and supergene jarosite. 167 Figure 4.26 The use of alunite in mineral exploration. 172 Figure 5.1 Calculated stability diagrams for the system K20-Al20 3-Si02-H 20-S03 182 at (a) 200°, (b) 300 0 and (c) 380 °C, and 1000 bars. Figure 5.2 Effect of pressure on the stability of alunite relative to kaolinite and 183 muscovite in the system K20-Ai203-Si02-H20-SC>3. Figure 5.3 Calculated stability diagrams for the system Na20-Al203-Si02-H20-S03 184 at (a) 200°, (b) 300 0 and (c) 380 °C, and 1000 bars Figure 5.4 Effect of variable alunite Na substitution on the K20-Al203-Si02-H 20- 186 S 0 3 system at 350°C, 500 bars. Figure 5.5 Model fluid:composition curves for an ideal alunite-natroalunite solid 190 solution. xi Figure 5.6 Effect of temperature and fluid K + / N a + fluid composition on a non-ideal 192 alunite-natroalunite solid solution, 500 bars. Figure 5.7 Dominance fields for KTNa species, based on equilibrium calculations. 192 Figure 5.8 Effect of temperature and fluid KCl /NaCl fluid composition on a non- 194 ideal alunite-natroalunite solid solution, 500 bars. Figure 5.9 Histogram showing the range of Na substitution in alunite from different 197 environments of acid sulfate alteration. Figure 5.10 Range of K/Na ratios for Pascua AS I and AS II alunite with elevation 197 (based on E P M A results). Figure 5.11 Histogram showing alunite Na contents for Pascua alteration (AS I) 199 based on dominant clay mineralogy. Figure 5.12 SEM photomicrograph showing oscillatory zoning in Del Carmen 199 alunite. Figure 5.13 Histogram of alunite Na contents for Del Carmen magmatic 201 hydrothermal alunite versus magmatic hydrothermal alunite from all other E l Indio-Pascua Belt properties. Figure 6.1 Simplified geology map of the El Indio-Pascua Belt. 209 Figure 6.2 Barren Oligocene to Upper Miocene hydrothermal alteration in the El 215 Indio-Pascua Belt. Figure 6.3 Summary of stable isotope data for alunite and associated alteration 216 minerals from the El Indio-Pascua Belt. Figure 6.4 Photos of syn-ore alunite and associated mineralization. 218 Figure 6.5 The Pascua alunite-pyrite-enargite (APE) mineralizing event in terms of 222 pH and the fugacity of O2 at 275°C and 0.5 molal total dissolved sulfur. xii A C K N O W L E D G E M E N T S I came to U.B.C. nearly six years ago with the intention of doing a M.Sc. in acid rock drainage, and somehow got diverted into this economic geology project along the way. This was largely the fault of John Thompson, and I found myself working on a small high sulfidation property in north-central B.C. within a few months of my arrival in Vancouver. After a year or so, the option arose to expand the project into a Ph.D. thesis. At that time, all I knew was that it would involve travelling to South America, which seemed like a pretty good deal. I diligently signed up for an introductory Spanish course, but quickly discovered upon my arrival in La Serena that my abilities to count to 10 just weren't going to help that much. Chileans are generally incomprehensible, even to fellow Latinos, and I was thoroughly confused. My plans to study the supergene remobilization of precious-metals were erased within hours of my arrival in Chile, after my first meeting with Dave Heberlein. Luckily Alan Clark, one of my previous professors from Queen's University, was present at the time, and all of a sudden I found myself about to embark on a comprehensive study of 'alunita' in the E l Indio-Pascua Belt. Needless to say, my first field season was considerably overwhelming. I eventually managed to pick up enough Spanish to figure out at what time the miners would be blasting in the pit I was working in that day - although I didn't always get that one quite right. Obviously I survived (minus one truck), and immediately signed myself up for another season. After all this time, there are so many people that I would like to thank for their contributions to this thesis. First and foremost, I would like to thank the Thompson family for putting up with me for all these years. I am indebted to John for the scientific support and constant reminders to not panic at every turn. I am sure Teck will be glad to have his full attention once again. Just one more to go, John! I am also sure that Anne will be very glad to see an end to my impositions on countless evenings and weekends. I thank her for her enthusiasm for the research topic. Her knowledge of the subject, including field and PIMA expertise, was invaluable. I also wish to thank Lee Groat for accepting me to U.B.C. in the first place, and allowing this project to evolve in the direction that it did. I appreciate all the resources that Lee put at my disposal throughout this process. This project was funded by Barrick Gold Corporation, with additional support from NSERC scholarships and grants. I would like to thank Jay Hodgson at Barrick for setting up the financial support for this project, and Dave Heberlein for the initial research ideas and providing all of the administrative and logistical support required, both in the field and in La Serena. This thesis would not have been possible without the collaboration of Barrick geologists and consultants, who provided the geological framework that my project depends on. Kevin Heather was a great help my first season, and I thank him for his interest in the project. Others xiii on this list include: Raul Guerra and all the geologists at Km. 34.; Nivaldo Rojas at El Indio; Dean Williams; Luis Perez and the other Del Carmen geologists; Bob Leonardson, Jean-Francois Metail, and the Pascua boys - Javier Vega, Pedro Vera, Fernando Rojas; Jose Noriega and the Lama exploration staff; Jack Hamilton, Graham Nixon, and Raymond Jannas. Logistical support and practical assistance was provided by numerous people, particularly Jessica Cuellar, Gabriel Sanchez, and John Kieley. A special thanks to Tono Traslavina and the Pascua crew for the salsa instruction. My fellow researchers were of immense help, and made life in Chile very enjoyable. Thomas Bissig provided thoughtful discussion and exchange of samples, maps, data, and ideas. I also thank him for making sure my samples made it through the Queen's Geochron Lab. Annick (a.k.a. Miss Pascua) Chouinard was a pleasure to work with, and her knowledge of Pascua was an invaluable resource. I am sure 'Club Resort Pascua' will never be the same - and I have the photos to prove it! A large component of this thesis is based on a stable isotope study that was completed at the U.S.G.S. Denver facility. I am indebted to Bob Rye for his support and his knowledge of acid sulfate systematics. It is a pleasure working with him, and his scientific contributions are very much appreciated. Cyndi Kester provided infinite assistance in the lab and this project would not have been possible without her. I thank her for her quick response to my many queries over the past few years. Gary Landis is also thanked for his persistence with the fluid inclusion gas analyses and his contributions to this project. Many other individuals at U.B.C. have made important contributions to this thesis. Shane Ebert is gratefully acknowledged for his prompt reviews of many versions of these chapters and numerous discussions on all things epithermal. I thank Greg Dipple for his patience in helping me get through the frightening world of thermodynamics and Margules parameters. Dick Tosdal was a great source of contacts at the U.S.G.S. and is thanked for his comments on several of these chapters. Steve Rowins is acknowledged for his assistance with the fluid inclusion stage. My committee members, K Fletcher and John Jambor, are thanked for their input. Many of their suggestions have been incorporated into this version of the thesis. I would also like to thank my friends and fellow graduate students at U.B.C. over the years. Shannon Shaw and Anita Lam provided wise advice upon my arrival in Vancouver. I thank Craig Nichol, Mike Buchanan, Vanessa Gale, Mike St. Pierre, Meghan Lewis, and Emily Chastain for the countless caffeine and sugar breaks; Tina Roth for the gallons of tea, commiseration, and helpful advice; and Scott Heffernan for providing welcome distraction and understanding over the last year. And finally, I want to thank my family for their support through this long process. Their constant questions o f ' Are you done yet?', and the inevitable follow-up, 'Well, what the .... are you doing out there anyway?', were a continual source of motivation. And, most importantly, I actually managed to beat my brother to the 'doctor' title, with a few months to spare! xiv Chapter 1 INTRODUCTION H I G H S U L F I D A T I O N D E P O S I T S Epithermal deposits were originally recognized by Lindgren (1933) as a distinct class of precious-metal deposit that formed at relatively low temperatures and shallow depths. High sulfidation deposits are one of the two principal types of epithermal systems, and are characterized by the presence of high sulfidation state minerals (e.g., enargite and luzonite) and acid sulfate alteration assemblages (quartz, alunite, kaolinite, pyrophyllite: Hedenquist, 1987; Arribas, 1995). Deposits of this type were previously termed enargite-gold (Ashley, 1982), high sulfur (Bonham, 1984; 1986), quartz-alunite gold (Berger, 1986), acid-sulfate (Heald et al., 1987), and alunite-kaolinite (Hayba et al., 1985), in reference to either the alteration mineralogy or conditions of formation. Intensive exploration and the discovery of several world-class epithermal deposits in the last 10 to 20 years have greatly advanced our understanding of the high sulfidation environment. Research has focused on the mineralizing environment and ore-forming processes at specific deposits (e.g., Stoffregen, 1987; Vennemann et al., 1993; Sillitoe and Lorsen, 1994; Arribas et al., 1995; Hedenquist et al., 1998; Jannas et al., 1999). Studies in many areas are hindered by the pervasive nature of alteration, the depth of oxidation, and the lack of good quality fluid inclusions. Regardless, the major elements of the genetic model are broadly accepted as follows: • Deposits form in subduction-related arc settings and are typically associated with volcanic centers. Mineralization is coeval with local volcanic rocks and related intrusions. • In many deposits, pervasive pre-ore alteration is widespread. Alteration results from the condensation of magmatic vapors that are generated by phase separation of exsolved magmatic fluids. Disproportionation or dissociation of S 0 2 and HC1, respectively, create extremely low pH, reactive fluids. Vuggy residual quartz is the product of extreme base leaching and marks the principal fluid upflow channel. Alteration typically grades outwards to an acid sulfate alteration zone containing quartz, alunite ± clays, and surrounding advanced argillic, argillic and propylitic assemblages (Table 1.1). This sequence reflects the progressive neutralization and cooling of acidic fluids outwards from the main fluid conduit. 1 a> o c V 3 O o O E cu U3 > . CO c o TO T3 X : o cu o o __. ro g Q . > , CO o cu o "cO Q . O •o c CO I ro 3 CT C O CO o cu > CO CO ro E CO ro CJ o cu CD 7 3 C CO co cu CO (0 | - C a o (0 I <* <u I u o K o a> c in o c O) CO ro E cu to CD Q - CO X I O CO O CO 3 CT ai _ I A E <u CO co < CO 3 o O ) CD 3 ro 3 •D CO CU rr •2 .E cu > ro CO X I as" 3 v_ 5^ c o TD CD O CO CO 3 C T O CJ c o TD 2 ro cu x: E CO CU cu c a> O) o o. x: cu x: c a> o c cu O ) cu a. 3 CO •o c ro CO cu o CJ g o CO "D c 3 O w. 3 co CO CJ C L cu cu z r >. CO >, c CJ x: 3 11= Q . N CO O „ cu >. CU c C L o « i SD  C L CO o CJ -a cu T3 o CU co >. a. c CU o ro C L CO 3 C T C 3 0) ro "5 CO "D < CL < CJ cu E CO >> x: C L o >> C L O — . CU o ro CO cu c o N c o « .2 ro 2 « "5 CO CJ CO cu c cu O J o C L o C L « •a « 3 C T C n CO " O cu o c CO > •a < T3 CU CJ c CO > •o ca v-o 2 « 12 o ro cu c cu o> o C L > , Ix: o ro ix c o CO cu ro xi E CD CO CO ro D3 i CO - 03 o E c o E 2 E o ro C L *— co 3 CT o cu E CO a5 < c g to i_ 03 ro ro L_ cu x: a. \_ cu Q . •o" CO CU L_ Q . CO CU T J i ro o a. C L 2 o TD C L CU ro cu E CO — IS a. o 103 3 ro 03 C L E 03 !c i CD a. cu cu I-a ro o C L C g ro 2 ro g 5> L_ ro "D 03 O C CO > T3 « C _g ro > ' 3 C T 03 C L CO 3 CT 2 '5 a> co g 03 CO 2 • Mineralization postdates regional alteration and is focused in veins, breccia zones, and permeable or reactive lithologies. The composition of ore-fluids is poorly constrained. These fluids are typically inferred to be less acidic and less oxidized than the early alteration fluids, and possibly of mixed magmatic-meteoric origin. This two-stage model is invoked to explain localized mineralization and conversely, the lack of mineralization in many regional alteration zones (lithocaps). • Formation of these deposits at shallow levels, close to the paleosurface, influence their geometry, the vertical zonation of alteration with respect to temperature and pressure (depth), and the potential interaction of deep ascending and shallow descending fluids. The nature of landforms, elevation, relief, and access to meteoric water will affect both deposit formation and preservation. T H E E L I N D I O - P A S C U A B E L T The E l Indio-Pascua Belt is situated in the Central Andean Cordillera Principal and straddles the Chile-Argentina border between latitudes 29°20' and 30°30' S (Fig. 1.1). This region has been the subject of intensive exploration for approximately twenty years, resulting in the development of several mines (e.g., E l Indio, Tambo), delineation of extensive resources (e.g., Pascua-Lama, Veladero), and discovery of numerous mineralized alteration zones. It is one of the foremost districts of high sulfidation mineralization in the world. Total production from the district to date has been over 310 t gold, 31,000 t silver, and 1 Mt of copper, mostly from the El Indio iriine. Mining activities at E l Indio are diminishing however, and the adjacent Tambo mine was recently closed (in 2000). Exploration is currently focused in the Pascua-Lama and Veladero districts, at the northern limit of the belt. The El Indio-Pascua Belt is located in the "flat slab" segment of the Andes, between two other well-known mineralized areas: the Maricunga porphyry Au and high sulfidation Au-Ag belt to the north (26-28° S), and the Los Pelambres - E l Teniente porphyry Cu-Mo belt to the south (32-34° S). The Maricunga Belt occurs at the northern margin or transition zone of the "flat slab" region, but the mineralized systems in the belt formed before or during the very early stages of slab flattening, 24-20 and 14-13 Ma respectively (Vila and Sillitoe, 1991; Muntean and Einaudi, 2001). In contrast, porphyry Cu-Mo deposits in the Los Pelambres - E l Teniente district formed in the late Miocene (10-5 Ma: Skewes and Stern, 1994; Kay et al., 1999), and are coeval with high sulfidation systems in the E l Indio-Pascua Belt. Several metallogenetic models 3 o CN L O C M O CO CO CO 5 9U0Z OIUBO|OA |EJ}uaQ +- ir auoz 0!}BUj6euje 'uojpnpqns ley auoz 0!UBO|OA ujamnos 11 • °0 oypej 5 / 3 A c • ° 0 S u c t! .2 o 0«g u c J 3 — 1 en ° <u u o o « . CO 3 15" 8- C3 3 o u c 03 O i 03 _ Q S T3 W .2 CD - 1 3 a- 60 u E *3 tn C .a u o s ̂ S - .5 •sS § ° 8 a • • 3 ™ I ! ! ° £ § D c « O "2 £ *V g "a 5 o o U -S 4 have been proposed for the flat-slab region and its borders (Skewes and Stern, 1994; Sasso and Clark, 1998; Kay et al., 1999; Kay and Mpodozis, 2001), including the most recent contribution from Bissig (2001). Hydrothermal alteration of late Eocene to late Miocene age is recognized throughout the E l Indio-Pascua Belt, but economic Au (Cu, Ag) mineralization is restricted to a narrow interval between 6 to 9.5 Ma (Bissig, 2001). Most deposits and prospects are of the high sulfidation- type. Low sulfidation-style mineralization is recognized only in the Rio del Medio vein, located near E l Indio, although E l Indio itself has characteristics of both high and low sulfidation environments (Jannas et al., 1999; Bissig, 2001). Ore deposition throughout the region is inferred to have occurred beneath a relatively flat landscape, at a time when magmatism was minimal (Bissig, 2001). There is no evidence for major volcanic edifices related to mineralizing centres. The current topography of the belt is attributed to post-Miocene valley incision and glaciation (Viet, 1996; Bissig, 2001). The El Indio-Pascua Belt provides an excellent site for studying epithermal processes and high sulfidation deposits for several reasons. • The systems in this region are typically well preserved, due to their young age and favorable tectonic and climatic conditions. Near-surface features of these deposits are preserved as steam-heated alteration zones and regional planar landforms (Bissig, 2001). • Access is possible via the extensive mining and exploration activities of Barrick Gold Corporation, the predominant landholder in the region. This company has been operating in the belt for over 20 years and all drill core, underground workings, and exploration data were made available for the purposes of this study. • A relatively complete geological framework has been determined for this region, based on numerous studies at both a deposit- (e.g., Siddeley and Araneda, 1986; Jannas et al., 1999; Chouinard and Williams-Jones, 1999) and regional-scale (e.g., Thiele, 1964; Maksaev et al., 1984; Martin et a l , 1995; Kay and Mpodozis, 2001; Bissig, 2001). • Alunite is spatially and genetically associated with precious metal mineralization in many deposits and prospects, and hence provides direct information on mineralizing processes (see below). 5 T H E R O L E O F A L U N I T E Alunite (KA1 3(S0 4)2(0H) 6) and alunite-group minerals (Table 1.2) are the key constituents in acid sulfate alteration. Alteration of this type is a characteristic feature of high sulfidation epithermal gold deposits (Cooke and Simmons, 2000; and references therein) and forms under conditions of low pH and highly oxidized fluid chemistry (Holland, 1965; Henley and McNabb, 1978; Stoffregen, 1987). These conditions can be generated by several mechanisms in the epithermal environment, related to either hypogene magmatic condensates, steam-heated processes, or supergene oxidation. The abundance and variability of alunite in high sulfidation systems has been recognized by several authors (e.g., Bethke, 1984; Rye et al., 1992; Sillitoe, 1993; Thompson, 1992; Arribas, 1995; Hedenquist et al., 2000). Recent studies indicate that alunite can be used to determine the nature of alteration fluids and the physio-chemical environment of mineral deposition. The presence of syn-mineral alunite in the E l Indio-Pascua Belt also provides direct information on ore-forming processes. Research to date has focused on specific characteristics of alunite-group minerals, including: • Field relations: The distribution, textures, mineral associations, and paragenesis of acid sulfate alteration are commonly used as an exploration tool to differentiate between ore- proximal, hypogene alteration and near-surface assemblages (e.g., Thompson, 1992; Sillitoe, 1993; Hedenquist et al., 2000). Table 1.2 Minerals of the alunite supergroup (current usage: Jambor, 1999). rtV.V.V.V.V.V.V.V.V.V.V.V.V.V.V.V.V.V.V Al > F e T A T " " "" Alunite-jarosite alunite KAI 3 (S0 4 )2 (OH) 6 jarosite K F e 3 ( S 0 4 ) 2 ( O H ) 6 group natroalunite NaAI 3 (S0 4 ) 2 (OH) s natrojarosite N a F e 3 ( S 0 4 ) 2 ( O H ) 6 minamiite (Na,K,Ca) 2 AI 6 (S0 4 ) 4 (OH) 1 2 hydronium jarosite ( H 3 0 ) F e 3 ( S 0 4 ) 2 ( O H ) s huangite CaAI 6 (S0 4 ) 4 (OH) 1 2 argentojarosite A g F e 3 ( S 0 4 ) 2 ( O H ) 6 walthierite BaAI 6 (S0 4 ) 4 (OH ) i 2 beaverite Pb (Fe ,Cu) 3 Fe 3 (S0 4 ) 2 (OH, H 2 0 ) 6 ammonioalunite NH 4 AI 3 (S0 4 ) 2 (OH) 6 ammoniojarosite (NH 4 )Fe 3 (S0 4 ) 2 (OH) 6 schlossmacherite (H 3 0,Ca)AI 3 (S0 4 ) 2 (OH) 6 plumbojarosite P b F e 6 ( S 0 4 ) 4 ( O H ) , 2 Beudanti te g roup svanbergite SrAI 3 [ (P,S)0 4 ) ] 2 (OH,H 2 0) 6 beaudantite PbFe 3 [ (As ,S)0 4 ) ] 2 (OH,H 2 0) 6 woodhouseite CaAI 3 [ (P,S)0 4 ) ] 2 (OH,H 2 0) 6 hinsdalite P b A I 3 [ ( P , S ) 0 „ ) ] 2 ( O H , H 2 0 ) 6 Crandall i te g r o u p crandallite C a A I 3 [ ( P 0 3 ( 0 „ 2 ( O H ) 1 / 2 ] 2 ( O H ) 6 goyazite S r A I 3 [ ( P 0 3 ( 0 „ 2 ( O H ) 1 / 2 ] 2 ( O H ) 6 gorceixite B a A I 3 ( P O „ ) ( P 0 3 « O H ) ( O H ) 6 plumbogummite PbAI 3 (P0 4 ) 2 (OH,H 2 0 ) 6 florencite-(Ce) CeAI 3 (P0 4 ) 2 (OH) 6 f l o r e n c i ^ 6 • Geochemistry: Several experimental (Hemley et al., 1969; Stoffregen and Cygan, 1990) and empirical studies (e.g., Aoki, 1991; Aoki et al., 1993; Thompson, 1992; Hedenquist et al., 1994; Arribas et al., 1995) have correlated variations in alunite-group chemistry to specific environments or temperatures of deposition. The temperature dependence of K:Na substitution is of particular interest in mineral exploration, since this feature can be readily identified in the field by the use of portable short-wave infrared (SWIR) spectrometers (e.g., PIMA, FieldSpec Pro). • Stable-isotope systematics: Alunite-group minerals contain four stable-isotope sites; D ( 0 H ) , 34S(so4)> 1 8OOH> and 1 8 0 S 0 4 , more than any other common mineral. Isotopic variations are related to the source and type of fluids, rates of processes, and the physical-chemical environment of deposition (Rye et al., 1992). Isotopic studies are particularly useful in conjunction with analyses of coexisting minerals such as clays (kaolinite, dickite, pyrophylfite), sericite, and sulfides. • Geochronology: Alunite is useful for age determinations by K-Ar and 4 0 Ar / 3 9 Ar methods due to the large concentration of potassium in end-member alunite (and jarosite). Alunite age data have been applied to supergene and weathering events (see Vasconcelos, 1999 for details) and paleoclimate studies (e.g., Bird et al., 1990; Sillitoe et al., 1991; Arehart and O'Neil, 1993; Vasconcelos et a l , 1994). Hypogene alunite can also be used to date hydrothermal events and associated precious- or base-metal mineralization (e.g., Alpers and Brimhall, 1988; Sillitoe et al., 1991; Perello, 1994; Arribas et a l , 1995; Bissig, 2001; this study), provided alteration is coeval with mineralization. Using a combination of these characteristics and previous stable-isotope studies, Rye et al. (1992) defined four specific environments of acid sulfate alteration (Fig. 1.2). These are: • The magmatic-hydrothermal environment, where the disproportionation of magmatic S 0 2 in condensed magmatic vapor forms H 2S and H 2 S 0 4 below ca. 350°C. The resulting acidic fluids react with wall rock to form extensive zones of acid sulfate alteration (this assemblage is equivalent to hypogene chloride-sulfate alteration as summarized in Hedenquist et al., 2000). • The magmatic steam environment, where alunite is believed to form from the expansion of rapidly ascending S0 2-rich magmatic vapor following sudden depressurization of the hydrothermal system (Rye, 1993). 7 • The steam-heated environment, where alunite forms from the condensation of vapour derived from an underlying hydrothermal system and the oxidation of H 2S gas above the water table. • The supergene environment, where the supergene oxidation of sulfides can produce an assemblage of kaolinite ± quartz, alunite, jarosite, with iron oxide and oxyhydroxide minerals. The terminology and relations defined by Rye et al. (1992) are used throughout this dissertation to specify different origins of alunite and acid sulfate alteration assemblages. Figure 1.2. Schematic diagram of a high sulfidation epithermal system showing environments magmatic-hydrothermal, magmatic steam, and steam-heated altera- tion. Supergene processes not shown. Modified from Rye et al. (1992). 8 S C I E N T I F I C C O N T R I B U T I O N S O F THIS S T U D Y The principal goal of this thesis is to better constrain genetic and exploration models for high sulfidation systems in the premier district for this deposit-type, the E l Indio-Pascua Belt, and to improve the general understanding of this deposit-type worldwide. Prior to this study, only limited data were available on alteration and ore-forming events in the E l Indio-Pascua Belt. Investigations were carried out in association with exploration activities and concurrent research in this region to clarify the relation between alteration and high sulfidation mineralization. In particular, research presented in this dissertation is focused on: • The origin and importance of early alteration. Economic mineralization in the E l Indio- Pascua Belt was preceded by the development of numerous and widespread barren alteration zones. Pre-mineral, barren alteration is typical of most high sulfidation systems and may be an important factor in the development of some economic deposits. The distribution and characteristics of pre-mineral alteration, and the timing with respect to ore formation in the Tambo and Pascua-Lama districts, are addressed. • Nature of alunite-natroalunite stability relationships. Very little information is available regarding occurrences of natroalunite and the nature of the alunite-natroalunite solid solution. This relationship is of particular interest in mineral exploration, given the potential to distinguish ore-proximal, hypogene alteration from lower temperature assemblages. The present study examines the distribution of alunite-natroalunite throughout the E l Indio Pascua Belt, using stable-isotope and geochronological data to constrain the origin of alteration. Experimental data from Stoffregen and Cygan (1990) is used to predict the effect of fluid compositions on the alunite-natroalunite solid solution. • Mineralizing processes. While there is general consensus that source of fluids and metals in high sulfidation deposits is magmatic (e.g., Hedenquist and Lowenstern, 1994; Arribas, 1995; Cooke and Simmons, 2000), there is still considerable uncertainty regarding the physical nature of the ore fluids. In particular, the relative contributions of vapor versus hypersaline liquid to the transport and deposition of gold, silver, and copper are the cause of much debate. Data from this study are used to generate a model for high sulfidation mineralization in the E l Indio-Pascua Belt, with implications for epithermal environments in general. • The transition from magmatic-hydrothermal to supergene processes. Late-stage processes in the E l Indio-Pascua Belt include the deposition of magmatic steam alunite and late-stage acid 9 sulfate alteration. Magmatic steam processes are currently poorly understood, but new data from this study are used to clarify the nature of these events and their relation to magmatic- hydrothermal alteration. An environment transitional between magmatic-hydrothermal and supergene processes is defined based on stable-isotope and 4 0 Ar - 3 9 Ar data for late-stage acid sulfate alteration. • The near-surface environment. Despite the importance of near-surface alteration to the recognition of high sulfidation environments, very few data are available regarding the characteristics of these assemblages. 4 0 Ar - 3 9 Ar alunite ages are used to constrain the timing of steam-heated alteration relative to underlying mineralization in the E l Indio-Pascua Belt. Stable-isotope data, in conjunction with paleoclimate and physiographic data, record the relative contributions of meteoric and magmatic fluids in near-surface processes. P R E S E N T A T I O N This dissertation is organized in manuscript format in accordance with the requirements of the University of British Columbia. Each chapter represents either a complete manuscript or a major contribution to future publications. A brief description of each chapter, listing co-authors and their respective contributions, is given below. Chapter 2: The nature of hydrothermal alteration at the Pascua-Lama high sulfidation deposit, Chile-Argentina. Deyell, C.L., Leonardson, R., Chouinard, A., Rye, R.O., and Bissig, T. This chapter presents detailed work by the author on the characteristics and distribution of hydrothermal alteration in the Pascua-Lama district, with new 4 0 Ar - 3 9 Ar ages to constrain the timing of alteration events. The data presented herein is based on field studies, combined with detailed short-wave infrared (SWIR), petrographic, and geochemical analyses by the author. A l l stable-isotope data were collected by the author at Bob Rye's U.S.G.S. facility. The regional geological framework is based on the recent study by Bissig (2001), with additional information from R. Leonardson and the exploration staff at Barrick Chile Ltda. and Barrick Exploraciones Argentina. The styles and distribution of mineralization in the Brecha Central area (the central part of the Pascua system) are the subject of A. Chouinard's Ph.D. thesis (currently in progress at McGi l l University). Work by the author is focused on the broader system. This paper represents the major contribution to a manuscript in preparation for submission to Economic Geology. 10 Detailed results from A. Chouinard's study will be submitted as a companion paper. Chapter 3: Alunite in an evolving magmatic-hydrothermal system: The Tambo high sulfidation deposit, E l Indio district, Chile. Deyell, C.L., Rye, R.O., Landis, G.P., andBissig, T. This paper presents a model for the evolution of the Tambo high sulfidation system, based on new field, paragenetic, geochronologic, and geochemical constraints determined by the author. The paper emphasizes differences of scale and controls between Tambo and Pascua- Lama, but documents a similar hydrothermal evolution. The manuscript has been submitted to Chemical Geology for publication in a special volume on 'Sulfate Minerals in Hydrothermal Environments'. The paper has also been informally reviewed by two U.S.G.S. scientists and J. Hedenquist. Their comments are greatly appreciated and have been incorporated into this version of the paper. The regional geological framework for the deposit is based on recent work by Bissig (2001) and previous studies in this area (Jannas et al., 1999; Siddeley and Araneda, 1990). A l l stable-isotope analyses were performed by the author at R. Rye's U.S.G.S. laboratory. G. Landis and R. Rye are responsible for fluid inclusion gas analyses and the interpretation of this data summarized herein. Chapter 4. The role of alunite in exploration. Evidence from high sulfidation deposits in the E l Indio-Pascua Belt, Chile-Argentina. Deyell, CL. and Thompson, A.J.B. The chapter provides an evaluation of the use of alunite and acid sulfate alteration for mineral exploration in the E l Indio-Pascua Belt. Field relations, paragenetic constraints, and physical, geochemical, and stable-isotopic characteristics for different alteration assemblages are summarized, based on research by the author in the E l Indio-Pascua Belt. Particular emphasis is placed on the characteristics of magmatic-hydrothermal alteration and its relation to precious- metal mineralization. The paper is intended for submission to the Journal of Geochemical Exploration. Several ideas on the use of alunite in exploration have evolved from previous work by A. Thompson (and E.U. Petersen) in hydrothermal systems in Utah and northern Chile (unpub. data). Chapter 5. Alunite-natroalunite stability relationships. Deyell, CL. and Dippie, CM. The focus of this paper is the nature of the alunite-natroalunite solid-solution and the occurrence of these minerals in the E l Indio-Pascua Belt. Experimental data of Stoffregen and 11 Cygan (1990) are used to predict the effect of variable temperature and fluid K:Na contents on the composition of alunite-natroalunite solid-solution. Chapter 6. Formation of high sulfidation precious metal deposits: New evidence from the E l Indio-Pascua Belt, Chile-Argentina. Deyell, CL. This paper provides new constraints on the genetic model for high sulfidation systems based on data from the El Indio-Pascua Belt. This chapter is the major contribution to a collaborative paper (in preparation) with J. Thompson, T. Bissig, and A. Chouinard and is intended to summarize results of recent research in this belt. J. Thompson contributed his ideas to the regional framework of high sulfidation rnineralization and relations to other systems in the Andean Cordillera. R E F E R E N C E S Alpers, C.N., and Brimhall, G.H., 1988. Middle Miocene climatic change in the Atacama Desert, northern Chile: Evidence from supergene mineralization at La Escondida. Geological Society of America Bulletin, 100, 1640-1656. Aoki, M., 1991. Mineralogical features and genesis of alunite solid solution in high temperature magmatic-hydrothermal systems. Journal of the Geological Survey of Japan, 277, 31-32. Aoki, M., Comsti, E.C., Lazo, F.B., and Matsuhisa, Y. , 1993. Advanced argillic alteration and geochemistry of alunite in an evolving hydrothermal system at Baguio, northern Luzon, Phillipines. Resource Geology, 43, 155-164. Arehart, G.B. and O'Neil, J.R., 1993. D/H ratios of supergene alunite as an indicator of paleoclimate in continental settings: Climate change in continental isotope records. Geophysics Monographs, 78, 277-284. Arribas, A., Jr., 1995. Characteristics of high sulfidation epithermal deposits, and their relation to magmatic fluids. In Thompson, J.F.H., ed., Magmas, Fluids and Ore Deposits. Mineralogical Association of Canada Short Course Notes, 23, 419-454. Arribas, A., Jr., Cunningham, C.G., Rytuba, J.J., Rye, R.O., Kelly, W.C., McKee, E.H., Podwysocky, M.H., and Tosdal, R.M., 1995. Geology, geochronology, fluid inclusions, and 12 stable isotope geochemistry of the Rodalquilar Au alunite deposit, Spain. Economic Geology, 90, 795-822. Ashley, R.P., 1982. Occurrence model for enargite-gold deposits. U.S. Geological Survey Open File Report 82-795, 144-147. Bethke, P.M., 1984. Controls on base- and precious-metal mineralization in deeper epithermal environments. U.S. Geological Survey Open File Report 84-890, 40 p. Berger, B.R, 1986. Descriptive model of epithermal quartz-alunite Au. In Cox, D.P., and Singer, D.A., eds., Mineral Deposit Models. U.S. Geological Survey Bulletin 1693, p. 158. Bird, M.I., Chivas, A.R., and McDougall, I., 1990. An isotopic study of surficial alunite in Australia: 2. Potassium-argon geochronomogy. Chemical Geology, 80, 133-145. Bissig, T., 2001. Metallogenesis of the Miocene El Indio-Pascua gold-silver-copper belt, Chile/ Argentina: Geodynamic, geomorphological and petrochemical controls on epithermal mineralization. Unpublished Ph.D. thesis, Queen's University. Bonham, H.F., Jr, 1984. Three major types of epithermal precious metal deposits. Geological Society of America Abstracts with Program, 16, 449. Bonham, H.F., Jr., 1986. Models for volcanic-hosted epithermal precious metal deposits: A review. In Proceedings of the International Volcanological Congress, Symposium 5, Hamilton, New Zealand 1986. Aukland, New Zealand, 13-17. Chouinard, A. and Williams-Jones, A.E. 1999. Mineralogy and petrography of ore and alteration assemblages at the Pascua Au-Ag-Cu deposit, Chile. Internal for Barrick Gold and Barrick Chile Ltda. Cooke, D.R. and Simmons, S.F., 2000. Characterisitcs and genesis of epithermal gold deposits. In Hagemann. S.G., and Brown, P.E., eds., Reviews in Economic Geology, v. 13, 221-244. Hayba, D.O., Bethke, P.M., Heald, P, and Foley, N.K., 1985. Geologic, mineralogic, and geochemical characteristics of volcanic-hosted epithermal precious-metal deposits. Reviews in Economic Geology v. 2, 129-167. Heald, P., Foley, N.K., and Hayba, D.O., 1987. Comparative anatomy of volcanic-hosted epithermal deposits: acid-sulfate and adularia-sericite types. Economic Geology, 82, 1-26. Hedenquist, J.W., 1987. Mineralization asscoiated with volcanic-related hydrothermal systems in the Circum-Pacific basin. In Horn, M.K., ed., Transactions of the Fourth Circum-Pacific Energy and Mineral Resources Conference, August, 1986. Singapore, American Association of Petroleum Geologists, Tulsa, OK, 513-524. Hedenquist, J.W, Arribas, A., Jr., and Gonzalez-Urien, E., 2000. Eploration for epithemal gold 13 deposits. In Hagemann, S.G. and Brown, P.E., eds., Gold 2000. Reviews in Economic Geology, vol. 13., 245-277. Hedenquist, J.W., Arribas, Jr., A., and Reynolds, T.J., 1998. Evolution of an intrusion-centered hydrothermal system: Far Southeast-Lepanto porphyry and epithermal Cu-Au deposits, Phillipines. Economic Geology, 93, 373-404. Hedenquist, J.W. and Lowenstern, J.B., 1994. The role of magmas in the formation of hydrothermal ore deposits. Nature, 370 (6490), 519-527. Hemley, J.J., Hostetler, P.B., Gude, A.J. , and Mountjoy, W.T., 1969. Some stability relations of alunite. Economic Geology, 64, 599-612. Henley, R.W., and McNabb, 1978. Magmatic vapor plumes and ground-water interaction in porphyry copper emplacement. Economic Geology, 73, 1-20. Holland, H.D., 1965. Some applications of thermodynamic dta to problems of ore deposits, II. Mineral assemblages and the composition of ore-forrning fluids. Economic Geology, 60, 1101-1166. Jannas, R.R., Bowers, T.S., Petersen, U., Beane, R.E., 1999. High-sulfidation deposit types in the E l Indio District, Chile. In B.J. Skinner, ed., Geology and ore deposits of the Central Andes. SEG Special Publication #7, 27-59. Kay, S.M., Mpodozis, C , and Coira, B., 1999. Neogene magmatism, tectonism, and mineral deposits of the Central Andes (22° to 33° S latitude). In Skinner, B.J., ed., Geology and Ore Deposits of the Central Andes, SEG Special Publication #7, 27-59. Kay, S.M., and Mpdozis, C , 2001. Central Andean ore deposits linked to evolving shallow subduction systems and thickening crust. GSA Today, 11(3), 4-9. Lindgren, W., 1933. Mineral deposits, 4 t h Edition. McGraw-Hill Book Company, New York, 930 p. Maksaev, V., Moscoso, R., Mpodozis, C , and Nasi, C , 1984. Las unidades volcanicas y plutonicas del cenozoico superior en la Alta Cordillera del Norte Chico (29°-31° S): Geologia, alteracion hidrothermal y mineralisacion. Revista Geologica de Chile, 21, 11-51. Martin, M, Clavero, J., and Mpodozis, C , 1995. Estudio geologico regional del Franja E l Indio, Cordillera de Coquimbo. Servicio National de Geologia y Minera, Santiago, Chile. Registered report IR-95-06, 232p. Martin, M.W., Clavero, J.R., Mpodozis, C , 1999. Late Paleozoic to Early Jurassic tectonic development of the high Andean Principal Cordillera, E l Indio Region, Chile (29-30° S). Journal of South American Earth Sciences, 12, 33-49. 14 Muntean, J.L., and Einaudi, M.T., 2001. Porphyry-epithermal transition: Maricunga Belt, Northern Chile. Economic Geology, 96 (4), 1445-1472. Perello, J.A., 1994. Geology, porphyry Cu-Au, and epithermal Cu-Au-Ag mineralization of the Tombulilato district, North Sulawesi, Indonesia. Journal of Geochemical Exploration, 50, 221-256. Rye, R.O., 1993. The evolution of magmatic fluids in the epithermal environment: The stable isotope perspective. Economic Geology, 88, 733-753. Rye, R.O., Bethke, P.M., and Wasserman, M.D., 1992. The stable isotope geochemistry of acid sulfate alteration. Economic Geology, 87 (2), 225-262. Sasso, A .M. , and Clark, A.H. , 1998. The Farahon Negro Group, Northwest Argentina: Magmatic hydrothermal and tectonic evolution and implications for Cu-Au metallogeny in the Andean back-arc. Society of Economic Geologists Newsletter, 34, 8-18. Siddeley, G, and Areneda, R., 1985. Gold-silver occurences of the El Indio belt, Chile. Geology of the Andes and its relation to mineral and energy resources. Symposium, Chile, 18 p. Sillitoe, R.H., 1993. Epithermal models: Genetic types, geometric controls, and shallow features. In R.V. Kirkham, W.D. Sinclair, R.I. Thorpe and J.M. Duke, eds., Mineral Deposit Modelling. Geological Association of Canada Special Paper 40, 403-417. Sillitoe, R.H., 1999. Styles of high sulfidation gold, silver and copper mineralization in the porphyry and epithermal environments. PacRim '99. Bali, Indonesia, 10-13 October, Proceedings, 29-44. Sillitoe, R.H., and Lorson, R,C, 1994. Epithermal gold-silver-mercury deposits at Paradise Peak, Nevada: Ore controls, porphyry gold association, detachment faulting, and supergene oxidation. Economic Geology, 89, 1228-1248. Sillitoe, R.H., McKee, E.H., and Vila T., 1991. Reconnaissance K-Ar geochronology of the Maricunga gold-silver belt, northern Chile. Economic Geology, 86, 1261-1270. Skewes, M.A., and Stern, C.R., 1994. Tectonic trigger for the formation of late Miocene Cu- rich breccia pipes in the Andes of central Chile. Geology, 22 (6), 551-554. Stroffregen, R., 1987. Genesis of acid-sulfate alteration and Au-Cu-Ag mineralization at Summitville, Colorado. Economic Geology, 82, 1575-1591. Stoffregen, R.E., and Cygan, G., 1990. An experimental study of Na-K exchange between alunite and aqueous sulfate solutions. American Mineralogist. 75, 209-220. Thompson, A.J.B, 1992. Alunite compositions and textures: Relationships to precious metal 15 mineralization. In New Developments in Lithogeochemistry, Mineral Deposit Research Unit Short Course #8, 20-21 February. Thompson, A.J.B., and Thompson, J.F.H., 1996. Atlas of alteration: A field and petrographic guide to hydrothermal alteration minerals. Geological Association of Canada - Mineral Deposits Division. 119 p. Vasconcelos, P.M., 1999. 4 0 Ar / 3 9 Ar geochronology of supergene processes in ore deposits. In Lambert, D.D., Ruiz, J., eds., Application of radiogenic Isotopes to Ore Deposits Research and Exploration. Reviews in Economic Geology vol 12, 73-113. Vasconcelos, P.M., Brirnhall, G.H., Becker, T.M., and Renne, P.R., 1994. 4 0 Ar / 3 9 Ar analysis of supergene jarosite and alunite: Implications to the paleoweathering history of the western USA and West Africa. Geochimica et Cosmochimica Acta, 58, 401-420. Venneman, T.W., Muntean, J.L., Kesler, S.E., O'Neil, J.R., Valley, J.W., and Russell, N., 1993. Stable isotope evidence for magmatic fluids in the Pueblo Viejo epithermal acid sulfate Au- Ag deposit, Dominican Republic. Economic Geology, 88, 55-71. Viet, H., 1996. Southern westerlies during the Holocene deduced from geomorphological and pedological studies in the Norte Chico, northern Chile (27-33 degrees S). Paleogeography, Paleoclimatology, Paleoecology, 123, 107-119. Vila, T., and Sillitoe, R.H., 1991. Gold-rich porphyry systems in the Maricunga Belt, northern Chile. Economic Geology, 86 (6), 1238-1260. 16 Chapter 2 T H E NATURE OF HY D R O T H E R M A L ALTERATION AT T H E P A S C U A - L A M A H I G H SULFIDATION DEPOSIT; C H I L E - A R G E N T I N A I N T R O D U C T I O N The Pascua-Lama Au-Cu-Ag deposit is located at the north end of the El Indio-Pascua Belt, in the Cordillera Principal of Chile (Region III) and Argentina (San Juan province; Fig. 2.1). The deposit occurs at approximately 4500 to 5100 metres above sea level (a.s.l.) in an area hosting widespread zones of hydrothermal alteration that are visible as distinct colour anomalies and topographic features. The recognition of these zones led to the discovery of high sulfidation gold mineralization in the Cerro Nevado area (Chile) in 1977 by Compania Minera San Jose S.A. The project was explored by a series of operators, including Lac Minerals Ltd., until the acquisition of this company by Barrick Gold Corporation in 1994. Continued exploration has uncovered several zones of both sulfide-rich and oxidized (sulfide-poor) mineralization to the east of the original Cerro Nevada discovery. Estimated proven- and probable-resources at Pascua-Lama total 269 Mt at 1.95 g/t Au and 66 g/t Ag (1.1 g/t Au cut-off: data from Mining Journal, London, April, 2001). Pascua-Lama is classified as a high sulfidation deposit on the basis of the abundance of acid-sulfate alteration and minerals of high sulfidation state, such as enargite. Several stages of alteration and alunite deposition are recognized from the near paleosurface to several hundred metres depth. The Au-Cu-Ag ore occurs in many forms. The largest deposit is Brecha Central (Fig. 2.2) which hosts high-grade gold ore in alunite-pyrite-enargite breccia matrix fill and banded veins, and in extensive alteration zones of silica-pyrite ± Fe-sulfate. The Frontera area, east of Brecha Central, straddles the Chile-Argentina border and hosts similar styles of mineralization. Smaller, high grade oxidized veins and breccias occur in the Esperanza and Penelope deposits, located to the SW and SE of Brecha Central, respectively (Fig. 2.2). Several other mineralized zones have been discovered recently in the Lama district, including El Morro Oueste and Filo Frederico Norte. New 4 0 Ar / 3 9 Ar data suggest that Au-Ag mineralization (and 17 Fi gure 2.1. Regional geology of the Pascua-Lama area. Adapted from Martin et al. (1995) using unpublished data from D. Heberlein. See text for descriptions of lithological units. 18 CO 111 Q . Z o < = u >» = a. o 0. < CU C m ° £ ss 5 S 2 o o E < a> co CU +-> ra cu I E ro CU ro c o «3 U 0) CO « - o cu c cu _ > o~ cu o — J O x o ^ » » S r i I- l - A \ \ \ N o > _> - C P " C o < u J U E 1 1 S « o .2 -a 4-i CO T3 a s M cn * I D . l i l § O CD 1 o l 3 > SS cn -> 3 ca u p cn S O N O N O N cn C - 3 o rt tl cn CL) U £ o c j—i a p ' f i cn _ CD I _s _< C I * .5 u co £ < 6 cn C _ 2 1 w co o S tH S CO O H l - J s m ^ U X 3 £ 8 OH CO y -a 3 a, u 0 0 o g E a o 19 related alteration) at Lama may have narrowly pre-dated the main Au-Cu-Ag event at Pascua (Bissig, 2001) and will not be considered here. In this paper, we examine the nature and distribution of alteration in the Pascua district to constrain the origin and evolution of the deposit. Focus is placed on the relation between alunite and ore-stage events, and its implications for magmatic-hydrothermal processes in the near-surface environment. G E O L O G I C A L F R A M E W O R K The El Indio-Pascua Belt (Fig. 2.1) straddles the Chile-Argentina border in the Cordillera Principal of the Andes, between latitudes 29°20' and 30°30' S. In addition to Pascua-Lama, the district hosts widespread zones of hydrothermal alteration and several world- class epithermal deposits and prospects, including E l Indio, which has produced over 10 Moz of gold, and Veladero, with resources of 15 Moz Au and 230 Moz Ag. These hydrothermal systems were emplaced within a NNE-SSW striking tectonic depression bound by the steeply W-dipping high-angle reverse Banos del Toro fault to the west and opposing structures in the Valle del Cura region in the east. Upper Paleozoic -to- lower Jurassic basement in this block consists predominantly of calc-alkaline felsic intrusive suites and volcanic rocks (Martin et al., 1999) and is overlain by up to 1500 m of Tertiary subaerial volcanic strata. The latter are extensively preserved in the southern part of the belt, but are less widespread in the vicinity of Pascua-Lama and Veladero (Martin et al., 1995; Maksaev et al., 1984). The Tertiary volcanic sequence (summarized in Bissig, 2001 and Martin et al., 1995) consists of a thick succession (up to 1200 m) of dacitic and rhyodacitic tuffs and subordinate basaltic and andesitic flows of the 23-27 Ma Tilito Formation. The overlying 17.5-21 Ma Escabroso Formation is composed predominantly of andesitic lavas and hypabyssal intrusive bodies as well as volcaniclastic sediments. Following a minor deformation event, continued igneous activity is represented by the 14-17 Ma andesitic Cerro de las Tortolas Formation and the contemporaneous Infiernillo intrusive unit. Magmatism decreased markedly thereafter and is recorded only as isolated occurrences of dacitic tuffs of the Vacas Heladas Formation (11 to 12.7 Ma). A single 7.8 Ma dacite dike, the only igneous rock roughly coeval with the mineralization in the district, is reported from Pascua (Bissig, 2001). Rhyolitic tuffs of the Vallecito Formation (5.5-6.2 Ma) are restricted to the Valle del Cura (Ramos et al., 1989) and some occurrences in the wider E l Indio-Tambo area. Volcanism finally ceased in the Upper 20 Pliocene after the eruption of the 2 Ma Cerro de Vidrio rhyolite dome in the NE of the belt (Bissig, 2001). In the immediate Pascua area, the geology is dominated by intrusive rocks of the Pastos Blancos group (Upper Paleozoic to Lower Jurassic) that outcrop to the north of Brecha Central and the Frontera zone. The intrusions are largely fine- to coarse-grained porphyritic granites and small stocks of dacite porphyry and granodiorite. Crystal and pumice tuffs of mostly dacitic composition and of probable Paleozoic age occur on the western edge of the Esperanza zone. Dacitic tuffs of Vacas Heladas age (11 to 12.7 Ma) are also recognized in the eastern part of the property (Sector Morro). Both the extrusive and intrusive units are intruded by small stocks of Bocatoma-age (36-30 Ma) hornblende-biotite diorite and various small dikes of andesitic to rhyolitic composition. The Pascua-Lama deposit occurs roughly at the intersection of two structural corridors; the E-SE oriented Pascua corridor and the N-NE oriented Pedro corridor (Hamilton, 1998). Numerous pre- or syn-mineral breccias are sited on, and are elongate parallel to, ESE fractures of the Pascua system. The largest of these is Brecha Central, which is a polymictic, matrix- supported breccia. Breccia fragments are typically altered to a quartz-alunite assemblage and the matrix is composed exclusively of alunite and pyrite-enargite. Other matrix- and clast- supported breccias occur east of Brecha Central in the Penelope area in Argentina, and also further SW in the Esperanza area. The character and geometry of the breccias suggest that they are phreatomagmatic in origin, although a few post-mineral breccias of magmatic origin are also present. Structural indicators suggest that breccias were emplaced during a period of extension that is related to local intrusive activity and the development of small-scale horst- graben structures and block faults in the Pascua area (Hamilton, 1998). Down-dropped blocks in between the Quebrada de Pedro and Quebrada de Pascua systems contain late volcaniclastic and lacustrine sequences of probable Vacas Heladas age that are not preserved elsewhere (Nixon, 1998). There is also a suggestion that normal movement along the Pedro fault system has down-dropped the area west of this zone. M I N E R A L I Z A T I O N There are two principal styles of Au (and Cu ± Ag) mineralization in the Brecha Central and Frontera areas. Alunite-pyrite-enargite (APE) mineralization is concentrated in the matrix of the phreatic breccias and as banded alunite-sulfide veins and minor disseminations in a 21 surrounding stockwork zone (Fig 2.3). Pyrite-szomolnokite (PZ) mineralization is characterized by pyrite and szomolnokite (FeS0 4 H 2 0) in strongly silicified, partly leached alteration zones. The APE mineralizing event is the largest in terms of volume. It consists of alunite, enargite, and pyrite in various amounts. Barite is common and occurs as individual grains and massive aggregates. Native sulfur occurs with alunite and enargite-pyrite in the matrix of Brecha Central but is rarely observed in the banded veins. Accessory minerals include diaspore, anglesite, pyrophyllite, stibnite, cassiterite, goldfieldite, covellite, galena, and trace chalcopyrite. Gold occurs primarily as native gold and calaverite (AuTe2) as inclusions in enargite, rarely in pyrite or alunite, and occasionally in fractures cross-cutting sulfide grains. Minor 'electrum' also occurs as inclusions in both pyrite and enargite. APE-type mineralization occurs primarily in the deeper parts of the deposit, typically between about 4450-4700 m (all elevations are reported as metres above sea level), and below PZ mineralization. Average Au grades are about 1.2-2 g/t Au (values vary slightly between Brecha Central and stockwork samples). APE ore also contains some of the highest Au and Cu grades - up to 20 g/t and 2% respectively. The PZ style of mineralization is distributed above and lateral to the APE assemblage. Mean Au grades in the PZ zone (2.1 g/t) are slightly higher than APE grades, and Cu contents average 0.2%. Copper is attributed to minor enargite that is commonly partly replaced by arsenolite or szomolnokite. Other accessory minerals include anglesite, barite, and rare covellite, gypsum, voltaite, and paracoquimbite. Gold and silver are thought to occur as sub- microscopic inclusions in pyrite and enargite. A third style of gold mineralization (which is equivalent to silica-gold mineralizing facies of Chouinard and Williams-Jones, 1999) is concentrated in the Esperanza area. Occurrences are also present locally in the Brecha Central and Frontera zones, usually above PZ-type mineralization. Gold occurs almost exclusively in native form associated with selective to massive silicic alteration. Jarosite, alunite, barite, and scorodite are common but sulfides are rare to absent (Caceres et a l , 1997). Trace electrum and Ag-Cu sulfosalts occur locally. Gold grades average about 1.2 g/t (in Brecha Central) and Cu mineralization is insignificant. The timing of mineralization in the Esperanza zone is unclear but is thought to be roughly contemporaneous with the main Au events in the district. The occurrence of visible gold as fine coatings on oxidized fracture surfaces at Esperanza also suggests supergene remobilization of the primary mineralizing assemblage. 22 Figure 2.3. Examples of APE mineralization. (A) Banded alunite-pyrite-enargite veins in silicified and AS I altered-wall rock. Alunite = white bands, sulfides = dark bands. Maria Tunnel. (B) SEM backscatter image of the AS II assemblage. Enargite has inclusions of calaverite (AuTe2) and one inclusion of stibnite (Sb2S3), and is intergrown with alunite, barite, and quartz. Alex Tunnel. Scale bar = 100 u.m. Photo courtesy of A. Chouinard. 23 Silver is concentrated in the upper parts of the Brecha Central and Frontera zones, as black Ag-rich veinlets and disseminations in silicified alteration zones. This assemblage consists of fine-grained native silver, electrum, native selenium, iodargyrite (Agl), chlorargyrite (AgCl), and barite. Accessory phases include jarosite, galena, aguilarite (Ag4SeS), native gold, and tiemannite (HgSe). Silver-enriched zones are typically found above and/or slightly offset from the zones of highest Au and Cu concentrations and overprints both APE and PZ-type mineralization. A L T E R A T I O N : C H A R A C T E R I S T I C S A N D D I S T R I B U T I O N Alteration is widespread in the Pascua district, occurring in an area of approximately 15 km 2. Alteration types identified on the property include intense acid sulfate and silicic assemblages with peripheral argillic and propylitic alteration. Deep quartz-muscovite alteration has been described in early company reports (R. Guerra, pers. comm.) but was not observed over the course of this study. Alteration is typically pervasive and texturally destructive, making it difficult to identify original rock types and controls on the distribution of alteration. The relations between assemblages are complex and subtle changes in mineralogy, particularly when clay and sericitic minerals are present, are not easily identified in the field. Methods of identification An extensive study was initiated in 1998 to refine pre-existing alteration maps and to provide detailed information regarding the distribution of specific alteration minerals. Alunite is of particular interest because it is directly associated with the A P E and silica-gold styles of mineralization. Samples were selected from surface and underground exposures, drill core, and reverse-circulation drillholes (RDH). The character and mineralogy of the assemblages were determined by systematic short-wave infrared spectroscopy (SWIR) techniques, with supplemental petrography and X-ray diffraction. Samples for SWIR analysis were taken every 10-20 m from the Alex Tunnel and from selected drill core along cross-sections CA-EW, and CA-00. Several drill holes located approximately within the 4680 m level were also sampled to supplement the Alex Tunnel data. A l l analyses were made using a Portable Infrared Mineral Analyzer (PIMA) - a field portable spectrometer that measures reflectance spectra in the SWIR band. Spectra were analyzed using PIMA View 3.0 software. Mineral identification is made by comparison of wavelength positions, intensity and shape of absorption troughs, and the overall 24 shape of the spectrum in comparison to standards in the SPECMIN database (e.g., Thompson et al., 1999). PIMA samples were carefully selected to minimize intergrown sulfides and surficial sulfate or alunogen material that can obscure the SWIR spectra (Thompson et al., 1999). Abundant jarosite was also avoided during sampling because its strong spectral response can conceal the presence of other, less abundant, minerals. Alteration Types 1. Acid Sulfate Alteration a) Early Acid Sulfate (AS I) The earliest recognized alunite-bearing alteration in the Pascua area is characterized by fine-grained intergrowths of alunite ± quartz, clays (kaolinite-dickite-pyrophyllite) and pyrite. This stage of alteration formed prior to brecciation and is superimposed by the main mineralizing event (Chouinard and Williams-Jones, 1999). The assemblage is widespread and extends from surface to the maximum depth investigated by drilling (ca. 4100 m). AS I alteration occurs as aggregates that replaced feldspar phenocrysts, as disseminations in the wallrock matrix, and rarely as irregular alunite veinlets and stringers (Fig. 2.4a). Only primary quartz phenocrysts and rare rutile remain from the original wallrock. Alunite is the main constituent of the assemblage and comprises 10 to 40 volume percent of the altered rocks. Alunite is white, cream, yellow, or pink and forms lath-like, elongate tabular or anhedral crystals 25 to 200 pim in length. Fine to medium-grained bladed alunite crystals also infill vugs within a silicified matrix that appear to have formed from the complete dissolution of feldspars or tuff fragments. Rare inclusions of alumino-phospho-sulfate (APS) minerals are present. AS I accessory minerals include pyrite, diaspore, and zunyite. Pyrite occurs as a minor to trace constituent interstitial to, or as inclusions in, alunite. Locally, small vugs in alunite appear to have resulted from the leaching of pyrite grains. Zunyite occurs exclusively at depths greater than 4500 m and only where pyrophyllite is present. Diaspore is slightly more abundant than zunyite and occurs as clusters of subequant grains, locally as inclusions in alunite. Diaspore also occurs in the deeper parts of the deposit, below about 4550 m. AS I alteration is overprinted by at least one silicification event ("silicification selectiva"; Chouinard and Williams-Jones, 1999). This event is characterized by fine-grained, 25 Figure 2.4. (A) SEM backscatter image showing irregular APS grains (bright zones) in cores of AS I alunite. DDH-116, 289m. Scale bar = 25 um. (B) SEM backscatter image of AS II alteration with K-rich (light) and Na-rich (dark) alunite associated with diaspore, pyrite, and enargite. Lama DDH-05, 337m. Scale bar = 50 um. (C) SEM backscatter image of REE-bearing APS inclusions in AS II alunite. DDH-111, 189.85m. Scale bar = 20 um (D) SEM backscatter image of steam heated alteration from near-surface above the Frontera Zone. DDH-119,47m. Scale bar = 200 um. 26 Figure 2.4. (Continued) (E) SEM backscatter image showing oscillatory PO4 ± Sr zoning (light bands) in coarse-grained AS III alunite (sample PS-26c). Scale bar = 100 um. (F) SEM backscatter image of fine-grained illite (ill) cross-cutting and replacing AS I alunite (alun). DDH-111, 120m. Scale bar = 50 urn. (G) Alunite ± jarosite late stage veinlets cross-cutting AS I altered wall rock. DDH-057, 154m. Scale bar = 1 cm. (H) SEM backscatter image showing supergene alunite, jarosite and intermediate alunite-jarosite solid solution Sample PM-33, Alex Tunnel. Scale bar = 20 urn. 27 granular quartz that locally dominates over the acid sulfate assemblage. Locally, alunite is also replaced by pyrophyllite, dickite, or fine-grained illite (see discussion below). Trace to significant amounts of jarosite also overprint the AS I assemblage. A l l jarosite is late and occurs exclusively in open spaces in the rock matrix, overprinting alunite, and as small cross- cutting veinlets. b) Acid Sulfate II (AS II) A second stage of alunite is coeval with APE-type mineralization. As noted above, this event is characterized by alunite-pyrite-enargite deposition as disseminations and open-space fill in the Brecha Central area, and as banded alunite-sulfide veins in the surrounding stockwork zone. Locally, alunite and sulfides are intimately intergrown, particularly in underground exposures in the Maria Tunnel, with smooth grain boundaries. Narrow silicified alteration halos surround the banded veins and indicate that AS II fluids interacted with the surrounding wallrock. This interaction was of limited extent (Chouinard and Williams-Jones, 1999) and it has not been possible to determine the effect of the AS II event on previously AS I-altered rock by observation and petrography alone. For this reason, the following descriptions are limited to matrix and vein alunite clearly associated with APE sulfides. AS II alunite occurs as coarse-grained clusters and rosettes in open vugs of the breccia matrix and as densely intergrown euhedral crystals in the banded veins. Grains are typically tabular to bladed, ranging from 30 |im to >2 cm in length. Colours range from white to grey to pink. Barite inclusions are common and occur along growth planes in coarse alunite crystals. Trace inclusions of pyrite, enargite, galena, anglesite, APS (Fig. 2.4c), and rare diaspore are also present (Fig. 2.4b). Native sulfur occurs intergrown with alunite and sulfides in the matrix of Brecha Central. Farther east in the Frontera zone, massive alunite-enargite-pyrite veins occur with accessory pyrophyllite and diaspore (Fig. 2.4b). In this zone, two phases of alunite are identifiable, with K-dominant tabular alunite grains overgrown and cemented by a more Na- rich alunite. Both varieties are intergrown with the AS II sulfides and are considered to be part of the same event. In the Esperanza area, the absence of sulfides makes it difficult to distinguish between acid sulfate alteration events. However, larger vugs and open space in breccia matrices are filled with coarse, white to cream, bladed alunite crystals up to 1 cm long that texturally and chemically resemble the AS II alunite identified in the Brecha Central and Frontera zones. Vertical zonation in the AS II assemblage is difficult to determine because of limited 28 drill-core data above the 4680 level. A near-surface vein assemblage of alunite-gypsum ± native sulphur cross-cuts AS I alteration above Brecha Central. This assemblage is thought to represent the shallow (i.e., low temperature) equivalent of the banded alunite-sulfide veins, but paragenetic relationships are unclear. c) Surficial Alteration Blanket A third acid sulfate assemblage occurs at or near the present-day surface and consists of silica minerals (cristobalite) ± kaolinite, alunite, and native sulfur. Alteration forms a blanket- like zone at elevations greater than approximately 4900 m and locally penetrates to greater depths along structures. Alteration is typically pervasive and commonly occurs as powdery, friable masses. Although original textures are rarely preserved, locally the rocks have been affected only by strong leaching, leaving a residual vuggy quartz rock with discernable porphyritic textures. Minor opaline silica intermixed with fine-grained quartz also occurs in irregular patches near the base of the alteration zone. The mineralogy, texture, and distribution of this assemblage are consistent with the steam-heated environment (e.g., Schoen et al., 1974). Alunite is a minor constituent in the surficial blanket assemblage. Alteration is locally zoned with an upper layer of residual or vuggy quartz (cristobalite) that grades downwards into a kaolinite and/or alunite-bearing zone. Where present, alunite occurs in irregular patches and disseminations in veinlets and open spaces. It is extremely fine-grained (< 5 \im) and forms pseudo-cubic, tabular, and anhedral crystals that appear as grey, irregular masses in thin section (Fig. 2.4d). Fine-grained jarosite occurs locally as overgrowths on alunite and linings of open vugs. d) Near-surface, Open-space Fill Alunite (AS III) A third stage of alunite deposition occurs in isolated surface exposures of coarse vug- and matrix-fill alunite above the Alex Tunnel. The alunite is medium- to coarse-grained, clear to golden yellow, and occurs as a lining in vugs in the matrix of the surficial breccia (Fig. 2.4e). Rare inclusions of barite and jarosite are present. Similar occurrences are noted as coarse vein and vug-fill alunite that cross-cuts AS I alteration in several near-surface drillholes, but the relationship of this event to AS II and steam-heated alteration is unclear. 29 e) Late-stage alunite (± jarosite) veins A late-stage of alunite occurs as veins, veinlets and local disseminations throughout the Pascua district. Jarosite is commonly associated with alunite but is always late. Accessory quartz is present locally. Alteration may occur in a fine-grained, granular to powdery assemblage, forming cryptocrystalline, hard lenses and veins, or as medium- to coarse-grained, euhedral veins (Fig. 2.4g). Local euhedral tabular crystals line veins and cavities. Veins are white to yellow to orange-brown, depending on the amount of jarosite present. Alunite is commonly fine- to medium-grained with tabular to anhedral crystals locally intermixed with silica. A pseudo-cubic crystal habit is observed in some finer-grained (<5 u,m) samples. Cross- cutting relations indicate that this stage of alteration post-dates both the AS I and AS II events. f) Supergene jarosite (± alunite) As noted, jarosite is present as a late mineral in association with acid sulfate, silicic, and argillic alteration assemblages. It occurs interstitial to alunite, quartz, and illite as fine-grained disseminations, granular crystals in vugs and open spaces, and locally as veinlets that dissect alunite grains or cross-cut AS I and argillic alteration. The distribution of jarosite is strongly fault and fracture controlled and extends to the maximum depths investigated by drilling (about 4100 m). Accessory hematite is present locally. Alunite rarely occurs intergrown with supergene jarosite, in thin veinlets and fine-grained aggregates in the rock matrix. The two cannot be distinguished in thin section but SEM-EDS analysis shows pseudo-cubic alunite grains (<5 t̂m) overgrown by successive bands of jarosite and compositional zones intermediate to alunite-jarosite end-members (Fig. 2.4h). 2. Silicic Alteration and Silicification At least four different silicification events can be distinguished on the basis of mineral associations and paragenetic relations. The terminology provided in parentheses is from Chouinard and Williams-Jones (1999). Minor vuggy silica (Vuggy Silica I) and locally massive, grey silicification (Silicification I) pre-date the emplacement of Brecha Central, on the basis of the presence of grey, densely silicified and locally strongly leached, vuggy silica fragments in the breccia. Most silicification occurred post-Brecha Central, however, and is contemporaneous with silica-gold-, pyrite-, and pyrite-szomolnokite -type mineralization. This stage of alteration includes the patchy development of vuggy silica (Vuggy Silica II) in which 30 primary feldspars and mafic minerals are partly to completely leached. Primary quartz phenocrysts are typically preserved, and the groundmass is completely replaced by fine-grained granular quartz. Rutile occurs as tiny clusters along grain margins of the original quartz grains, but there is also evidence for the local redistribution and remobilization of Ti during this event (Chouinard and Williams-Jones, 1999). Silicic II alteration also formed post-Brecha Central and comprises massively silicified rocks, selectively silicified, partly leached rocks, and localized silica stockwork and veins. The white to grey-coloured silica alteration consists of massive, fine-grained quartz commonly with later syn-ore minerals (e.g. pyrite, enargite, szomolnokite) that infill leached vugs and cavities. 3. Argillic Alteration Argillic alteration is characterized by kaolinite, quartz ± illite, montmorillonite, interlayered illite-smectite, and rare gypsum and dickite. Feldspars are partly to completely replaced by fine-grained, platy masses of clay minerals that can only be differentiated by XRD or SWIR analysis. Accessory pyrite and late jarosite occur interstitially to kaolinite-illite. Alteration is typically pervasive, although original granitic textures are locally visible. Argillic alteration is typically gradational to acid sulfate and propylitic assemblages at the lateral margins of the Brecha Central and Frontera zones. Illite (± kaolinite and quartz) also locally cross-cuts and overprints AS I alteration in close proximity to Brecha Central (e.g., "illite overprint" in Chouinard and Williams-Jones, 1999). Some fragments in Brecha Central also contain illite. This alteration is rarely recognizable in hand specimen, but petrographic and SEM-EDS analyses show that illite corroded and replaced AS I alunite grains (Fig. 2.4f). Observations suggest that the alteration occurred following the AS I event but before brecciation and mineralization. The distribution of this assemblage is not well-constrained and it is unclear how, or if, it relates to the widespread argillic alteration assemblage. 4. Propylitic Alteration A weak alteration assemblage of chlorite ± kaolinite, gypsum, pyrite, and illite with rare epidote and smectite occurs peripheral to the zones of intense alteration and the major breccia and vein systems in the Pascua-Lama area. Typically, feldspars, biotite, or hornblende are partly to completely replaced by fine-grained masses of sericite (kaolinite-illite-smectite), chlorite, and minor Ti-oxide. Minor patches and disseminations of epidote are common. Rare 31 fine-grained pyrite occurs as clusters and disseminations in the rock matrix. Seams of kaolinite- illite, gypsum, and patchy limonite-hematite locally cross-cut the propylitic assemblage. Distribution of Alteration The distribution of alteration assemblages is illustrated at surface (Fig. 2.2) and in two cross-sections; CA-EW (Fig. 2.5) and CA-00 (Fig. 2.6). Section CA-EW is oriented nearly E- W and extends from Esperanza to the Frontera zone. Section CA-00 is oriented roughly orthogonal to CA-EW and transects both high-grade Au and Cu mineralization in Brecha Central and relatively weakly altered rocks to the north and south. Alteration maps were completed by systematic SWIR analyses from selected drill core and RC chips (individual data points shown; Figs. 2.6 to 2.8) and by extrapolating drill core logs from other holes on section (not shown). Only certain alteration assemblages are mapable at the scale of these diagrams. Due to the difficulties in separating the AS I and AS II assemblages, the two are combined as 'Acid Sulfate Alteration' for the purposes of this compilation. Most of the alteration is presumed to have formed during the earlier event, due to the consistency in style and mineralogy of the assemblage. Likewise, individual silicification events are also not differentiated and are grouped as one unit. Results from this study illustrate differences in the styles of alteration between the Esperanza and Brecha Central/Frontera zones. At Esperanza, silicic alteration is prevalent at and near surface and is underlain by extensive acid sulfate alteration (Fig. 2.7). Kaolinite is the most common clay mineral associated with the acid sulfate assemblage. Dickite and pyrophyllite occur locally. A small siliceous knob composed mostly of opaline silica outcrops at the top of the Esperanza zone. This may represent silica precipitation at the water table coincident with the base of an eroded steam-heated zone (e.g., Schoen et al., 1974). Isolated occurrences of argillic alteration, intermixed with patchy chlorite-bearing propylitic alteration (not shown) occur at the base of the silicic zone. This assemblage is largely hosted in a small stock of diorite and, on the basis of sparse drill core data, does not appear to extend into the surrounding granitic rocks. Jarosite is prevalent beneath the silicic alteration zone. Abundant veins and disseminations of jarosite ± alunite also cross-cut the silicic alteration zone but are too irregular to be shown at the scale of this diagram. East of the Esperanza area, silicic alteration is present primarily as irregular pods or lens-like bodies at intermediate elevations (approximately 4700-4900 m) and is surrounded by 32 33 Figure 2.6. Section CA-00. (a) Distribution of alteration zones. Also shown are locations of major faults, (b) Major lithological units and distribution of Au mineralization, (c) Lo- cations of PIMA sample points and distribution of dominant clay mineralogy. 34 Figure 2.7. Detailed section of Esperanza area from section CA-EW. (a) Distribution of alteration zones. Also shown are locations of major faults, (b) Major lithological units and distribution of Au mineralization, (c) Locations of PPMA sample points and distribution of dominant clay mineralogy. 35 36 acid sulfate assemblages (Fig. 2.5, 2.6). Surficial blanket-type alteration occurs in an irregular, tabular zone that extends from surface to about 4900 m. Locally, this zone is underlain by kaolinite-rich argillic alteration and silicic assemblages (Fig. 2.8) that do not appear to be associated with the overlying steam-heated zone. Between the Central and Frontera faults, alteration is less intense, with irregular occurrences of kaolinite ± illite-bearing argillic alteration and the absence of significant silicic alteration. This area coincides with very low Au grades, and is hosted primarily in granodiorite with cross-cutting diorite dikes (not shown). Alteration in and at the margins of Brecha Central (Fig. 2.6) is dominated by silicic and acid sulfate assemblages. To the north of the breccia, alteration grades outward to more widespread propylitic and illite/smectite-bearing argillic assemblages. Jarosite is generally restricted to granitic rocks at the margins of Brecha Central and is concentrated at intermediate to low elevations, peripheral to areas with the highest Au and Cu grades. A strong vertical zonation in clay mineralogy associated with acid sulfate alteration is evident in both the Brecha Central and Frontera zones (Figs. 2.6, 2.8). Pyrophyllite occurs almost exclusively below about 4550 m. Dickite is dominant above this zone to about 4700 m. Kaolinite is most common at elevations above 4700 to 4750 m. A L U N I T E G E O C H E M I S T R Y To evaluate the variability of alunite in each alteration assemblage, the chemical compositions of different stages of alunite were determined by analysis of individual alunite grains and sample composites. End-member alunite is represented by the formula KA1 3 (S0 4 )2 (OH) 6 but naturally occurring minerals of the alunite supergroup can show a wide range of chemical substitutions (e.g., Jambor, 1999). The most common substitution of Na for Kdefines the alunite-natroalunite solid solution. Other substitutions for K + may include Ca, Ba, Sr, REE (Ce, La), Pb, Ag, H 3 0 , and N H 4 . Substitution of Fe 3 + for A l 3 + defines the alunite-jarosite solid solution and substitution of (P0 4) 2" for (S0 4) 2" forms APS minerals of the crandallite or beudantite groups (Table 2.1; Jambor, 1999; Stoffregen et al., 2000). Analytical methods Samples from each stage of alteration were classified on the basis of distribution, mineral associations, and paragenetic relations. These observations were supplemented by detailed petrographic study, and only samples clearly belonging to each alteration type were 37 used for further geochemical and isotope analyses. Alunite grain mounts and polished sections were analyzed by SEM-EDS and electron probe microanalysis (EPMA). E P M A data were collected on a fully-automated C A M E C A SX-50 microprobe, operating in the wavelength- dispersion mode with the following operating conditions: excitation voltage, 15 kV; beam current, 10 nA; beam diameter, 15 |im; and total count time under 65 seconds. Analytical parameters were chosen to minimize beam damage caused by the volatilization of alkali elements (e.g., Petersen and Thompson, 1992). However, the large beam size results in poor representation of finer-grained samples (supergene and steam-heated alteration in particular) and thin oscillatory zones and/or growth bands. Details of analytical methods are given in Appendix A. Based on petrographic and E P M A results, representative samples were selected for ICP- MS trace-metal and REE analysis (details Appendix A). Supergene alunite could not be adequately separated from intermixed jarosite ± quartz, but all other alunite types were analyzed. The purity of separates was verified by powder X-ray diffraction but trace amounts (<10 %) of quartz or clays may be present. Table 2.1. Selected minerals of the alunite-jarosite family. A l u n i t e - j a r o s i t e g r o u p alunite K A I 3 ( S 0 4 ) 2 ( O H ) 6 natroalunite N a A I 3 ( S 0 4 ) 2 ( O H ) 6 m inam iite ( N a , K , C a ) 2 A I 6 ( S 0 4 ) 4 ( O H ) , 2 huangi te C a A I 6 ( S 0 4 ) 4 ( O H ) i 2 walthierite B a A I 6 ( S 0 4 ) 4 ( O H ) , 2 Jarosite K F e 3 ( S 0 4 ) 2 ( O H ) 6 B e u d a n t i t e g r o u p svanberg i te S r A I 3 ( P 0 4 ) ( S 0 4 ) ( O H ) 6 w o o d h o u s e i t e C a A I 3 ( P 0 4 ) ( S 0 4 ) ( O H ) 6 C r a n d a l l i t e g r o u p crandal l i te C a A I 3 [ ( P 0 3 ( O w 2 ( O H ) i / 2 ] 2 ( O H ) 6 f lorenc i te - (Ce) C e A I 3 ( P 0 4 ) 2 ( O H ) 6 f lorenci te- (La) L a A I 3 ( P 0 4 ) 2 ( O H ) 6 Analytical results E P M A and trace-element geochemical data for all stages of alunite are summarized in Table 2.2. Alunite K and Na data for all E P M A analyses collected during this study are plotted in Figure 2.9. REE data (Fig. 2.10) are normalized to a sample of unaltered basement granite from the Pascua area (sample 99thb 139a, Colorado Unit; Table 2.3). Also shown in Figure 2.10 is the normalized REE composition of a dacite dike that is coeval with mineralization in the district (sample 99thbl30b, Table 2.3). 38 <*> ON Q. s to CO II c O C\l II c a E to co II c co II Q. S CO — Oi 'Si 0 0 < N — •o co n c CO SI 9 I ci CO o O "Cf q o i- ci T - O CO CM O CO o o o O CO O 7 - f - CO o o O CT) O i- o 0) CO T— o O O i- O) CO CO T - CO CD CO CM ci O T - O O -r- I s O CO o i- s O CM O T - q ci 39 Table 2.2b. Selected ICP-MS trace element and REE concentrations for bulk (~1 gram) alunite samples. Trace contaminants may be present. Data below detection limits (0.1 ppm for REE; 2 ppm for Zn) not shown. Estimates of precision, based on duplicate analyses, given in Appendix B. Steam Late Stage AS I* AS II AS III Heated Veins Sample #: P09 P11 P06 P05 P32 P-SH P15 P17 P (ppm) 110 135 470 580 840 740 790 2260 Sr 41.4 187.3 878 137 615 1095 685 2400 Ba 220 230 6170 6120 1720 1140 1240 2180 Pb 340 554 696 2530 1765 688 128 3170 Sb 12.2 22.7 1 5 31 4 2 1.5 Ag 3.5 2.7 1.2 0.4 2.8 1.0 0.4 1.4 Cu 31 12 22 122 21 54 28 27 Bi 13.1 10.3 0.2 0.4 4.5 0.9 0.6 0.2 Zn 16 2 - 18 - 2 - 48 La 15 14.8 17.5 20 39 37.5 39 32 Ce 29.5 26.8 19.5 20 33 54.5 51.5 65 Pr 3.2 2.7 1.5 1.7 1.4 5.5 4.5 8.7 Nd 8.5 5.8 3.5 4 2 16.5 14 37 Sm 1 0.4 0.3 0.4 0.1 2 2.3 11.3 Eu 0.1 - 0.1 - - 0.4 0.4 2.8 Gd 0.5 0.2 0.2 0.3 0.1 1.4 1.7 11.1 Tb - - - - - 0.1 0.2 1.3 Dy 0.3 0.4 - 0.1 - 0.8 1 3.6 Ho - 0.1 - - - 0.1 0.2 0.3 Er 0.1 0.7 - - - 0.6 0.6 0.3 Tm - 0.1 - - - 0.1 - - Yb 0.2 1.6 - - - 0.9 0.8 0.3 Lu - 0.4 - - - 0.1 0.1 - * may contain up to 10% quartz. 40 ^ AS 1 alunite j >• 0.0 0.2 0.4 0.6 0.8 Na (a.p.f.u.) 0.0 0.2 0.4 0.6 0.8 Na (a.p.f.u.) =» 0.8 0.2 Na (a.p.f.u.) •/V— •v, • Esperanza "v. • • mm m • * SH alunite • • AS II alunite 0.0 0.2 0.4 0.6 Na (a.p.f.u.) 1 0 * 0.2 0.4 Na (a.p.f.u.) => 0.8 CL 0.2 Na (a.p.f.u.) "tj. • • AS III alunite • • • LV alunite • • • Figure 2.9. K versus Na contents for each stage of Pascua alunite based on EPMA results. Each point represents one analysis from a total of m samples for each stage: AS (m=18); AS II (m=10); Esperanza (m=3); AS III (m=2); steam heated (SH; m=2); and late vein (m=4). Data points below detection limits for K (0.01 a.p.f.u.) and Na (0.2 a.p.f.u.) not shown. 41 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Figure 2.10. REE data for Pascua alunite normalized to a sample of host granite (Colorado Unit). Also shown is normalized REE data for one sample of dacite (Pascua dacite) that is contemporaneous with alteration and mineralization in the area. (A) AS I, AS II samples (sample number given in parentheses). (B) AS III, steam heated (SH), and late vein (LV) alunite. Data points below detection not plotted. 42 Table 2.3. Geochemical data for unaltered host rock (Colorado Unit) used for normalization of alunite REE. Data for Pascua dacite given for comparison. Unpub. data, T. Bissig. Pascua Dacite Colorado Unit Sample # 99thb130b 99thb139a Na 2 0 (wt %) 2.94 3.32 K 2 0 1.84 4.39 CaO 5.68 0.36 MgO 2.07 0.11 MnO 0.27 0.05 F e 2 0 3 (Fe total) 4.45 0.82 A l 2 0 3 16.99 12.28 P 2 O 5 0.19 0.01 S i 0 2 63.06 76.35 TiO z 0.66 0.08 Total 98.23% 97.79% REE ICP-MS (ppm) La 24.7 19.7 Ce 51.2 43.3 Pr 6.2 4.9 Nd 23.7 17.3 Sm 4.1 3.8 Eu 1.1 0.2 Gd 3.4 3.1 Tb 0.4 0.6 Dy 2.3 3.4 Ho 0.4 0.7 Er 1.1 2.2 Tm 0.2 0.3 Yb 1.0 2.5 Lu 0.1 0.4 Acid Sulfate I: AS I alunite is dominantly potassic. Substitution of Na ranges from below detection to 0.6 atoms per formula unit (a.p.f.u.), based on E P M A analyses (Fig. 2.9). Irregular compositional zoning is common with trace P, Sr, and Ba substitution. Rare oscillatory zoning is present in deeper samples, in which compositions alternate in bands of nearly pure alunite, alunite with minor Na, and thin Ba or Ca-bearing alumino-phospho-sulfate (APS) zones. APS grains also occur locally as inclusions and as cores of AS I alunite. Where present, APS minerals contain minor Sr, REE, and trace Ba. Many of these APS inclusions and zones are too narrow (<5-10 |im) to obtain accurate E P M A analyses. Trace-element contents are better represented by the ICP-MS data that indicate elevated Sb and Bi concentrations, 43 particularly in comparison to AS II alunite. REE contents are variable. One sample (P09) is relatively depleted in REE and exhibits strong fractionation of heavy-REE (HREE) compared to host-rock compositions. The second sample (P l l ) is depleted in mid-REE (MREE), but high HREE contents may be attributed to the presence of REE-bearing APS zones that are visible by SEM backscatter imaging. Acid Sulfate II: The chemistry of AS II alunite is more variable than AS I alteration. The range of Na substitution is similar for the two stages, but APS zoning with Ca, Ba, Sr ± REE is more common in the AS II event. Walthierite is present as rare irregular zones or bands, particularly in alunite sampled from the Maria Tunnel. Substitution of Ca locally approaches end-member huangite in irregular zones from a vein at the margins of Brecha Central. Again, geochemical variability is not adequately represented by E P M A data due to the small size of APS inclusions and chemical zonation. ICP-MS data show significantly elevated Sr, Ba, and Pb concentrations in AS II alunite, in comparison to the AS I samples. A l l samples show strong fractionation in HREE compared to the host lithology. Acid Sulfate III: SEM-EDS analysis indicates crystals are nearly stoichiometric (> 0.8 a.p.f.u. K) with minor Na or P substitution in oscillatory growth zones. Sb and Pb contents are higher than average and data reflect extreme fractionation in HREE relative to host-rock compositions. Blanket zone: Alunite in this assemblage is typically homogenous and K-dominant, with up to 0.3 a.p.f.u Na substitution. Trace amounts of P and Sr ± Fe occur in irregular zones. Sr and Sb are elevated relative to most other alunite types, but Ba and Pb contents are low. Alunite exhibits limited REE fractionation with respect to the granite host rock. Late-stage alunite veins: Alunite of this stage is nearly stoichiometric, with Na substitution rarely exceeding 0.2 a.p.f.u. ICP-MS data indicate significantly elevated P, Sr, and Pb, relative to all other alunite types. This sample also exhibits a very different REE fractionation pattern, with strong enrichment of M R E E and depletion in HREE. The strong Eu anomaly is similar to increased Eu contents in the coeval dacite. Interpretation of results With few exceptions, AS I and AS II alunite are similar in their major-element compositions. Average Na contents are comparable for both stages but the distribution of Na- versus K-rich alunite differs between the two events. AS II alunite on average shows increasing 44 Na contents with increasing depth, which is not observed in AS I samples (Fig. 2.11). As well, Na contents are higher in AS II samples from the eastern (Lama) side of the property, in comparison to AS II alunite in the Brecha Central and Frontera zones (significant difference at 95% confidence level, based on t-test results; Appendix A). Alunite veins at Lama were sampled to greater depths than those in Pascua however, and hence the increase in average Na/ K may reflect the depth (and therefore temperature) of formation rather than a spatial distribution. High concentrations of Na have been correlated with elevated temperatures of alunite formation in other deposits (Stroffregen and Alpers, 1987; Stoffregen and Cygan, 1990; Aoki, 1991; Hedenquist et al., 1994; Aoki et al., 1993; A.J.B. Thompson, unpub. data). Results from this study also indicate that AS I and AS II alunite are generally enriched in Ba, Ca ± P 0 4 , Sr, REE in comparison to AS III and near-surface, blanket zone alunite occurrences. Similar observations are noted in other high sulfidation environments (e.g., Aoki, 1991; Stoffregen et al., 2000; Thompson and Petersen, unpub. data). Late-stage alunite, however, exhibits variable compositions that are consistent with the range of data reported for supergene alunite in other studies (e.g., Stoffregen and Alpers, 1987; Dil l , 2001). REE fractionation patterns are extremely variable. AS I alteration in particular has very different REE signatures for the two samples analyzed. Results for sample P09 are similar to those reported in recent studies of advanced argillic alteration in other high sulfidation environments (e.g. Fulignati et al., 1999; Arribas et al., 1995). The observed loss of HREE may be the result of preferential uptake of LREE into the alunite structure (Schwab et al., 1990). However, grains with elevated zones of Sr ± P 0 4 in the second sample (PI 1) are correlated with much higher HREE contents. This enrichment may be attributed to mineralogical controls on REE substitution and the stabilization of the alunite crystal structure by larger bivalent ions such as Sr and Ba (Schwab et al., 1990). REE contents may also be controlled by the replacement of apatite grains in the host rock (Stoffregen and Alpers, 1987) or local variations in temperature, redox conditions, or fluid/rock ratios (e.g., Michard, 1989; Lottermoser, 1992). REE contents of AS III alunite, and to a lesser extent AS II, exhibit extreme fractionation of HREE compared to host-rock compositions. We propose that the data reflect differential partitioning of REE into the source fluids (Flynn and Burnham, 1978; Reed, 1995) and minimal wallrock interaction. Results for AS III alunite are similar to those obtained for magmatic steam alunite in the Tambo deposit (Chap. 3). 45 c o s AS II alunite AS 1 alunite 0 1 2 3 N a / K (m.p.f.u.) Figure 2.11. Range of Na/K ratios for AS I and AS II alunite with elevation. Based EPMA results for a total of n data points (AS I n=351; AS II n=316). 46 S T A B L E - I S O T O P E S T U D Y Complete stable-isotope systematics (8 S, 8 D , 5 Oso4, and 8 O O H ) were determined for alunite and associated alteration minerals in order to characterize the source of these components. Only samples clearly belonging to each paragenetic stage were analyzed. Alunite from each stage of alteration and two samples of jarosite were analyzed for 8 3 4 S, and a sub-set of samples was selected for 8 D and 8 1 8 0 analysis. Sulfur data were also obtained for AS II- stage pyrite, enargite, and barite. Clay minerals are not included in this study because material could not be adequately separated for analysis. A l l data were collected at the U.S. Geological Survey Isotope Laboratory in Denver. The 8 3 4 S analyses were determined by an on-line method using an elemental analyzer coupled to a Micromass Optima mass spectrometer following Giesemann et al. (1994). Analyses of 8 D were performed by a step heated technique modified from Godfrey (1962). Oxygen-isotope data were collected for alunite (both sulfate and hydroxyl groups) and barite. Analyses were done using the conventional BrF 5 method described in Wasserman et al. (1992). Several samples were also analyzed by pyrolysis with a Finnigan TC/EA coupled to a Finnigan Delta Plus X L mass spectrometer using continuous-flow methods modified from Kornexl et al. (1999). Details of sample preparation and analytical methods are reported in Appendix B. Brief sample descriptions and all isotope data are listed in Table 2.4. Sulfur-Isotope Relations A l l sulfur-isotope results are summarized in Fig. 2.12. The 8 3 4 S value of bulk sulfur in the magma (5 3 4 S S S ) cannot be directly determined, since no large bodies of syn-ore volcanic rocks have been identified in the Pascua district. Estimation of 5 3 4 S S s (2 to 4%o; Fig. 2.5) is therefore based on data for magmatic steam alunite (AS III; see discussion below). The main results from this study are: • The range of 8 3 4 S values for AS I and AS II alunite are similar, from 13 to 25%o. Data for AS II barite (22.1%c) and Esperanza alunite (18.3%o) also fall within this range. • AS II enargite and pyrite data are narrowly clustered around -3 to -6%o. AS I pyrite could not be separated in sufficient quantity for analysis, but Beane (1988) reported a 8 3 4 S value of -3.5%o for disseminated pyrite intergrown with alunite in quartz monzonite from the Brecha Central area. This value is plotted for comparison on Figure 2.12. 47 Table 2.4a. Sample descriptions and stable isotope results for Pascua alunite. All data is given in per mil, elevation in metres above sea level. Stages are: Acid Sulfate I (AS I); Acid Sulfate II (AS II); Acid Sulfate III (AS III); Steam heated (SH); Late stage veins (LV); Esperanza (Esp); and Supergene (SP). Abbreviated mineral names; alunite (alun), quartz (qtz), kaolinite (kao), dickite (dick), pyrophyllite (pyl), jarosite (jar), enargite (enarg), sil (silica), bar (barite), native sulfur (S). Sample Lab# Stage Elev. (m) Description 5D 5 M S 6 1 80S 04 818CW DDH-137A, 23.1m P09a AS I 4665 Qtz-alun alteration of monzonite -37 24.7 19.9 9.2 DDH-184, 441.3m P10a AS I 4616 Qtz-alun alteration of diorite -52 22.0 13.8 12.4 DDH-195A, 242.0m P11a AS I 4612 Qtz-alun alteration of monzonite -33 24.5 16.9 12.2 DDH-154, 185.9m P12a AS I 4570 Qtz-alun-dick alteration; Frontera -52 15.7 LM-20, 94.8m P13a AS I 4990 Qtz-alun-py alteration of monzonite -29 21.0 20.7 14.7 DDH-116, 289.5m P21a AS I 4435 Qtz-alun alteration; Lama 15.3 DDH-152, 107.0m P22a AS I 4630 Qtz-alun-dick alteration; Frontera 12.6 PSD-98-108, 374m P23a AS I 4435 Qtz-alun-pyl-zuny alteration -43 18.5 15.6 9.4 DDH-017, 122.5 P24a AS I 4369 Qtz-alun-pyl alteration; near Esperanza 17.1 DDH-017, 108.0m P28a AS I 4369 Qtz-alun-pyl alteration; near Esperanza 16.7 DDH-168, 260m P33a AS I 4647 Qtz-alun-pyl alteration -27 20.4 DDH-116, 157.0m P01a AS II 4551 Alun-py-enarg banded vein -46 17.6 15.8 10.9 DDH-137A, 113.9m P02a AS II 4580 Brecha Central Sur matrix; alun-enarg-py -34 19.8 16.5 13.0 DDH-137A, 152.1m P03a AS II 4544 Alun-py-enarg banded vein -51 14.7 17.4 15.6 DDH-154, 224.3m P04a AS II 4545 Matrix alun-enarg-py; Frontera -44 22.2 15.7 10.7 PS-99-03 P05a AS II 4680 Brecha Central matrix; alun-enarg-py-S -41 14.1 18.0 13.6 PS-98-16d P06a AS II 4674 Alun-enarg vein; Tunel Maria -40 24.7 19.6 10.4 LM-03, 301.0m P07a AS II 4519 Alun-py-enarg vein; Lama 17.9 DDH-172, 435.1m P25a AS II 4393 Alun-jar vein -47 18.9 15.6 14.4* Annick 440 P27a AS II 4680 Brecha Central matrix; alun-enarg-py -36 12.6 18.2 15.5* PS-26c P32a AS III 4995 Near-surface, coarse matrix alun -41 2.8 20.4 13.5 DDH-129, 111.6m P14a LV 4582 Alun-jar vein -40 7.0 20.7 13.5 DR-368NE, PM-33 P16a LV 4680 Alun-jar +/- silica vein; Tunel Alex -72 4.3 9.7 9.0 DDH-154, 310.5m P17a LV 4489 Alun +/- jar vein -79 6.2 12.7 5.5 LM-20,160.1m P15a SH 4960 Alun-jar matrix; Lama, near surface -48 1.6 19.1 13.1* DDH-119, 47.7m P18a SH 4994 Alun-sil +/- kao, S; near surface -42 -0.1 16.9 13.2 DDH-122, 4.3m P19a SH 5089 Alun-sil +/- kao, S; near surface -68 6.1 DDH-119. 34.0m P20a SH 5006 Alun-sil +/- kao, S; near surface -58 2.4 18.7 13.3 DDH-057, 76.5m P26a Esp 4873 Alun in bx matrix; Esperanza -35 18.3 17.4 12.4 ' calculated using BrF5 total oxygen data, except continuous flow samples marked with See discussion in text. Table 2.4b. Sample descriptions and stable isotope results for jarosite, barite, and sulfides. Abbreviations as in Table 2.4a. Sample Lab# Stage Elev. (m) Description SD 8a4S 5 1 80 S 0 4 8 1 80O H Jarosite: DDH-135, 162.7m DDH-198, 168m PJ30 PJ31 SP SP 4528 4718 Jar +/- alun vein cutting QA alteration Matrix of opaline qtz bx -129 -178 -2.5 3.3 4.5 2.3 Barite: Annick 440 Pb29 AS II 4680 Barite breccia matrix in Dr-440-SE 22.1 14.4 Pyrite: DDH-137A, 113.9m DDH-137A, 152.1m LM-03, 301.0m P02p P03p P07p AS II AS II AS II 4580 4544 4519 Brecha Central Sur matrix; alun-enarg-py Alun-py-enarg banded vein Alun-py-enarg vein; Lama -3.8 -5.3 -3.4 Enargite: DDH-137A, 113.9m DDH-154, 224.3m PS-98-16d LM-03a, 433m P02e P04e P06e P08e AS II AS II AS II AS II 4580 4545 4674 4398 Brecha Central Sur matrix; alun-enarg-py Matrix alun-enarg-py; Frontera Alun-enarg vein; Tunel Maria Massive vein enarg -6.0 -3.6 -4.5 -5.6 4 8 AS III alunite was sampled from coarse grained, euhedral crystals obtained near-surface in the matrix of Brecha Central. The alunite has a low 534S value of 2.8%o. 834S values for alunite in the near-surface, blanket zone (0 to 6%o) are slightly enriched relative to sulfides in the deposit. Samples of late-stage alunite were selected from drill core and underground exposures in the Alex Tunel. The 834S values range from 4 to 7%o and are also enriched relative to AS II sulfides. Jarosite (sample Pj31) was separated from a series of fine-grained veinlets that cross-cut a strongly silicified alteration zone approximately 300 m below the present-day surface. This sample has the lightest 834S value (-2.5%o) of all sulfates in the deposit. Sample Pj31, from veinlets that cross-cut AS I altered wallrock, has 834S data (3.3%o) similar to those of late- stage alunite. Late Stage Veins Steam Heated AS I AS I A S I -10 es e Enargite • Alunite o Pyrite A Jarosite V Barite • Esperanza alunite _ 245_-305_£_ H 2S/S0 4 - 1 to 3 085fc_ HoS/SO,, - 2 "«3"« O • 10 8 3 4 S (per mil) 20 30 Figure 2.12. Range in 8 S data (in %o) for all stages of Pascua alunite. Also shown is data for associated sulfides, jarosite, and barite. Estimated 8 3 4S £ S ~ 2 to 4 per mil. Range of A 3 4 S a i u n . p y temperatures for AS I and AS II stages are given, as well as estimated H2S/SO2 ratios. See text for discussion. 49 The high 8 S values for AS I, AS II, and Esperanza alunite are significantly different from all other stages of alunite and jarosite in the deposit. The 14 to 30%o difference between alunite and sulfides is consistent with sulfate derived from the disproportionation of S 0 2 (Rye et al., 1992; Rye, 1993) and is interpreted to reflect a magmatic-hydrothermal origin. The large variation in alunite 8 3 4 S values is attributed to deposition over a range of temperatures. These temperatures can be estimated by the isotopic fractionation equation of Ohmoto and Lasaga (1982) for coexisting alunite and pyrite. Data for the one AS I sample reported in Beane (1988) give an equilibrium A 3 4 S a i u n - P y temperature of 380° C. This represents a maximum depositional temperature for the AS I event. Using an average 5 3 4S pyrite value of -4%o, depositional temperatures for other AS I alunite samples are estimated at 190° to 350°C. Likewise, equilibrium temperatures calculated for three AS II alunite-pyrite pairs (samples P02, P03, and P07) range from 245° to 305° C. Calculated sulfide-sulfate (H 2S/S0 4) ratios for AS I and AS II fluids are approximately 1 to 3. These values are in the range of magmatic-hydrothermal processes (Rye et al., 1992; Rye, 1993) and suggest mildly reducing fluid conditions during alunite deposition in both pre- and syn-mineral events. Low 8 3 4 S values for all AS III, blanket zone, and late-stage alunite are interpreted to reflect different origins. Sulfur-isotope data for AS III alunite, in conjunction with oxygen- isotope results, are consistent with a magmatic steam origin. Alunite of this type is considered to have 8 3 4 S values close to that of the bulk sulfur in the magma, based on relationships determined by Cunningham et al. (1984) and Rye et al. (1992) for alunite at Marysvale, Utah. The 8 3 4 S values for blanket-type alunite, combined with high 818Oso4 values, are consistent with a steam-heated origin, although data indicate limited sulfur isotope exchange between aqueous S 0 4 and H 2 S. The 8 3 4 S data for jarosite sample Pj31 are consistent with sulfate derived from the oxidation of pre-cursor sulfides. The slightly heavier sulfur values for late-stage alunite are interpreted to represent either sulfate derived from the oxidation of sulfides or H 2S degassed during the collapse of the hydrothermal system. In either case, limited sulfur-isotope exchange between sulfate and sulfide species is inferred, on the basis of the range of 8 3 4 S values. Oxygen-isotope systematics provide further clues as to their origin (see below). 50 , Oxygen-Isotope Relations Oxygen-isotope systematics for each stage of alteration, including both sulfur-oxygen and oxygen-hydrogen relationships, are given in Figures 2.13 and 2.14. Also shown are the range of calculated fluid compositions in equilibrium with alunite of each stage. The composition of paleo- meteoric waters is estimated at 8D = -100 ± 10%o (B. Taylor, pers. comm.). However, data from this study and from Jannas et al. (1999) indicate that this value may have been slightly lighter (ca. -125%o). Variations of this magnitude are expected over the duration of the hydrothermal system, and a range of meteoric 5D values is shown on all isotope plots accompanying the following discussions. AS I and AS II alunite: Depositional temperatures calculated from A 1 8 0 S O 4 - O H values for A S I and A S II alunite are unreasonable, ranging from -50° to >800°C. Oxygen isotopic temperatures for individual samples are also consistently higher than those determined by A 3 4 S a i u n . p y fractionations. These results are attributed to disequilibrium effects due to retrograde OH exchange (Rye et al., 1992). Fluid compositions for the two alteration events are nearly identical (Fig. 2.13) and are shown as one field with 8DH2o between -25 and -47%o. These data overlap with values for felsic magmatic fluids (Taylor, 1992) and the volcanic vapor field (Giggenbach, 1992), and indicate a dominant magmatic component in the source fluids for both paragenetic stages. Lower 8DH2o values are generally accompanied by a downward shift in 8 1 80 H 2o and are interpreted to represent limited mixing between magmatic fluids and meteoric waters. This trend is not attributed to 8D depletion by open-system degassing (Taylor, 1988) since there is no systematic depletion over time, from the early A S I to the later A S II assemblage. AS III: The equilibrium fluid composition for A S III alunite (Fig. 2.14b) overlaps with the range of A S I and A S II fluids and indicates a dominant magmatic fluid signature. However, the depositional temperature calculated from A 1 8 O S O 4 - O H is lower than expected (110 °C), and may indicate equilibration with a low 5DH2o fluid. Overall the isotope data, combined with textural and geochemical observations, are consistent with a magmatic steam origin for A S III alunite (Rye et al., 1992). Blanket zone: Calculated A 1 8 OSO4-OH temperatures of deposition for alunite of this stage are 130°, 150° and 270 °C. The lower temperature data are reasonable for environments in which aqueous sulfate forms at shallow levels from the oxidation of H2S by atmospheric oxygen, and indicate that oxygen-isotope equilibrium was obtained between the aqueous sulfate 51 20 15 1 u 3 O 10 -10 • A S I S " ^ • ASI8 1 8 0 O H • AS I I S * ^ A AS II8 1 80O H • O ESP8 ' 8 0 O H Pyrite A / 10 834S (per mil) 15 20 25 30 Figure 2.13a. 5 3 4 S and 5 1 8 Oda ta for alunite from stages A S I, A S II, and Esperanza (ESP). 8 1 8 O S 0 4 , and 8 I 8 0 O H values for each sample are joined by lines. Range of 8 3 4 S data for A S II pyrite is given for reference. 81 O (per mil) Figure 2.13b. 8D, 8l8Oso4, and 8 1 8OOH values for alunite from stages AS I, AS II, and Esperanza (ESP). Range of calculated 8DH2o and 8180H20 for fluids in equilibrium with AS I and AS II are shown as one shaded fields. Fluid compositions calculated from equations of Stoffregen et al.. (1994) and Rye and Stoffregen (1995). Temperatures used for calculations (AS I; 200-380°C: AS II; 240-350°C) are derived from textural, A^Saiun.py, and A I 8OSO4-OH isotope data. Lines and fields are MWL = meteoric water line of Craig (1961); FMW = composition of fluids dissolved in felsic magmas (Taylor, 1988); kaolinite line of Savin and Epstein (1970); V.V (volcanic vapor) = range of water compositions discharged from high temperature fumaroles (Giggenbach, 1992); SASF (light grey) = supergene alunite S0 4 field; and SAOZ = supergene alunite OH zone as described in Rye et al. (1992). 52 20 15 H O 1°1 oo 1o Pyrite O • • S H B 1 ^ o S H 8 , 8 0 O H • L V S 1 ^ o L V 5 l 8 0 O H T AS III S 1 ^ V AS III5 1 80O H • Jar6 1 8 0 S 0 4 • Jar8 1 8 0 O H 10 15 8 3 4S (per mil) Figure 2.14a. 8 3 4S and 8 1 80data for steam heated (SH), late vein (LV) and AS III alunite, and one sample of jarosite (jar). S 1 8 0 S 0 4 , and 8 1 8 0 0H values for each sample are joined by lines. Range of 8 3 4S data for AS II pyrite is given for reference. Figure 2.14b. 8D, 8 1 8 0 S 0 4 , and 8 1 8 0 O H values steam heated (SH) and AS III alunite. Range of calculated 8D H 2 o and 8 1 8 0 H 2 o for fluids in equilibrium with alunite are given - calculated from equations of Stofffegen et al. (1994) and Rye and Stoffregen (1995). Temperatures used for calculations are derived from textural and A 1 8 O s 0 4 - O H isotope data. Lines and fields as described in Fig. 2.13b. 53 Figure 2.14c. 8D, 8 1 8 0 S 0 4 , and 8 1 8 0 O H values late vein alunite (LV) and jarosite (jar). Range of calculated 8D H 2 o and 8 ' 8 0 H 2o for fluids in equilibrium with each sample are given - calculated from equations of Stoffregen et al.. (1994) and Rye and Stoffregen (1995). Temperatures used for calculations are derived from textural and A 1 8 0 S 0 4 - O H isotope data. Lines and fields as per Fig. 2.13b. Also shown is SJOZ = supergene jarosite O H zone and SJSF = supergene jarosite sulfate field as described in Rye and Alpers (1997); and the estimated composition of paleo- meteoric water. 54 and water in the fluid. The third sample gives an unreasonably high A 1 8 O S O 4 - O H temperature, given its shallow depth of formation, which suggests that A 1 8 O S O 4 - O H may have been affected by post depositional exchange. Calculated fluid compositions in equilibrium with alunite (Fig. 2.14b) have 8DH2o values of -40 to -55%o and 5 1 80 H 2o data between -2 and +3%o. The values are significantly heavier than those of paleometeoric water and are not typical of steam-heated environments (e. g., Rye et al., 1992). Some of this enrichment may be attributed to evaporation effects (Henley and Stewart, 1983), but the degree of enrichment (> 45%o) is much greater than that in any system reported to date (J. Hedenquist, pers. comm.). Based on these results, we propose that steam-heated alteration involved a significant magmatic fluid component. Late-stage alunite: Fig. 2.14c highlights reference fields for supergene alunite and jarosite as given in Rye et al. (1992) and Rye and Alpers (1997), respectively, for comparison to late-stage alunite and jarosite from this study. Alunite 8 1 8 0 0 H data plot within the Supergene Alunite OH Zone (SAOZ), but two of the three 51 80So4 values fall outside of the Supergene Alunite Sulfate Field (SASF). The latter samples have large, positive A 1 8 0 S O 4 - O H values that are not consistent with a supergene origin (Rye et al., 1992). Depositional A 1 8 0 S O 4 - O H temperatures average 85°C, higher than expected for the supergene environment. Temperature calculations for these two alunite samples suggest sufficient residence time of aqueous sulfate following oxidation that allowed for oxygen isotopic equilibrium with water in the fluid. The third sample exhibits obvious disequilibrium between oxygen-isotope species. Calculated 8D H 2 o values for late-stage alunite (Fig. 2.14c) range between -35 and -75%o with 5 1 80H20 less than 0%o. The lowest 5 1 80H20 values are correlated with progressively lighter 5 1 80H20 contents, suggesting a substantial magmatic contribution to the source fluids mixed with varying amounts of meteoric water. Overall, data for late-stage alunite are not consistent with a supergene origin. Results can be interpreted to reflect either the oxidation of precursor sulfides or of H 2S degassed from the hydrothermal system. Based on the abundance of sulfide in the Pascua deposit and visible signs of oxidation, we suggest the former is more likely. Oxidation could result from atmospheric processes above the water table. Alternatively, recent evidence for alteration in the Tambo deposit (Chap. 3) indicates that oxidation by S0 2-rich magmatic vapors could be significant in the late-stages of a shallow hydrothermal system. Similar processes may be applicable at Pascua, given the evidence for near-surface magmatic steam alunite. 55 Jarosite: Oxygen-isotope data are available only for one sample of jarosite (Pj31). The 8 1 8 O O H data is slightly heavier than the Supergene Jarosite OH Zone (SJOZ) defined by Rye and Alpers (1997), but 81 8Oso4 data fall within the Supergene Jarosite Sulfate Field (SASF; Rye and Alpers, 1997). These values may indicate limited OH exchange with low pH waters (Rye and Alpers, 1997) or formation from partly exchanged meteoric waters. The composition of fluid calculated in equilibrium with jarosite is close to the predicted composition of paleometeoric waters and is consistent with formation in the supergene environment. G E O C H R O N O L O G Y Alunite from four paragenetic stages and one sample of jarosite were selected for 4 0 Ar / 3 9 Ar dating. AS III alunite (sample P32) was collected and dated previously (Bissig, 2001) and is included here for comparison. The objective of this study was to determine the duration of magmatic-hydrothermal activity in the Pascua area and to confirm the timing of different alteration events. A l l analyses were carried out at the Queen's University (Kingston, Canada) 4 0 Ar- 3 9 Ar laboratory, equipped with a Mass Analyzer Products M A P 216 mass spectrometer. About 10 mg of each sample was irradiated for 7.5 h at the McMaster nuclear reactor in Hamilton, Canada using biotite standard Mac-83 as a radiation-flux monitor (24.36 ± 0.17 Ma; Sandeman et al., 1999). Samples were step-heated using a defocused L E X E L 3500 Ar laser-beam. Ages were calculated using the decay constants suggested by Steiger and Jager (1977), and all errors are given at 2a. Analytical details are provided in Appendix C. Results and Interpretation: Of the four alunite samples dated, only one (sample P05) yielded a reasonable plateau age (Table 2.5). This sample consists of coarse-grained AS II alunite from the matrix of Brecha Central. The alunite is intergrown with pyrite, enargite and native sulfur and is therefore considered to be coeval with the APE-ore event. Its age (8.78 ± 0.63 Ma) is consistent with the range of ages reported in Bissig (2001) for APE mineralization (8.7 ± 0.2 to 8.1 ± 0.1 Ma). The relatively large error for this sample is attributed to poor absorption of the laser-beam and insufficient release of argon. 56 Q co CNJ CJ) OO LO CO CD CD CNJ O q iri in co r-- co cn CM CJ) CM LO r-- od CT) CNJ 00 oq co CNJ cn CNJ Q co o co o CD CT) o 00 CD C8 O O O O CD CO r-- o o co LO CO o o CO O) CNJ o o CJ) co co 3 CO 2 CL q r-- CNI LO LO CJ) CO CJ) < 3 CO CD « Q_ co T3 "O CD CO CD D) CO 0) Q. E CO co CO r- oo CJ) co co CJ) LO d d d oo oo 00 CO CT) CJ) od od Cji i — u? ? 5" CJ) CJ) co CO CNJ d — ' — ' — oo oo oo CNI co T - o i 03 o i od co CO LO CJ) CNI o o CNI t ~ => m CO I- r- H h- CD CD CD CD 0) c c c c CO 3 3 O < < < < Ja  - E X > 0 . c o CO _ l c o < < CO CNI 00 o CO T— co. 0- 0- CL 0_ CL 57 AS III alunite, sampled at surface from coarse vug-fill crystals at the margin of Brecha Central, yielded an age of 8.38 ± 0.17 Ma (Bissig, 2001). Based on these results, this alunite is coeval with AS II alteration and ore deposition. A sample of late jarosite (sample Pj31) also yielded a reasonable age (7.98 ± 0.43 Ma) on the second dating attempt. This sample is characterized by a supergene isotopic signature and is the youngest alteration event recorded in the Pascua-Lama district. Samples of late vein and steam-heated alunite did not release sufficient argon and their reported ages represent only rough estimates (7.97 ± 1.59 and 9.14 ± 1.98 Ma, respectively). Repeated attempts to date AS I alunite were unsuccessful. D I S C U S S I O N : C O N T R O L S O N A L T E R A T I O N The distribution of alteration and mineralization in high sulfidation systems is strongly controlled by the focus of fluid flow and the related permeability of the host rocks (Sillitoe, 1999). Permeability can be generated by structural, hydrothermal, or lithological processes, and the relative importance of each will vary with the local environment (Sillitoe, 1993). Lithological and hydrothermal brecciation are typically important controls in shallow epithermal systems (< 500 m depth; Sillitoe, 1999). Preferential alteration and precious-metal deposition in poorly lithified, porous, clastic, and volcaniclastic sedimentary rocks are common in several deposits (e.g., Pierina, Volkert et al., 1998; Pueblo Viejo, Kesler et al., 1981). Hydrothermal-induced permeability is attributed to rock fragmentation, brecciation, and rock leaching (Sillitoe, 1999), and several deposits host alteration and mineralization in hydrothermal breccias (e.g. Paradise Peak, Nevada, John et al., 1991; Tambo, Chile, Jannas et al., 1999). Structural controls dominate at the intermediate levels of high sulfidation systems (Sillitoe,.1999). These controls include either regional-scale features or structures related to the emplacement of shallow intrusions (Sillitoe, 1999; Hedenquist et al., 2000). Topographical effects are also important in the shallow epithermal environment. These effects are particularly evident in deposits where steam-heated alteration (generated above paleo-water tables) is preserved. Progressive or intermittent lowering of the water table during hydrothermal activity can cause overprinting of deeper high sulfidation mineralization and magmatic-hydrothermal alteration assemblages by the steam-heated environment. The nature of alteration at Pascua has many similarities with other high sulfidation 58 systems. On a broad scale, intense silicic and acid sulfate alteration zones grade outward to argillic and propylitic assemblages. Detailed mapping and SWIR analyses have shown that the distribution of alteration on a smaller scale is often much more complex however, and is influenced by several factors including elevation and lithological, structural, and topographical controls. The effects of each of these factors are described below. Lithology: Lithological controls are subordinate at Pascua. The deposit is largely hosted in intrusive rocks of the Pastos Blancos group that are relatively homogenous with respect to composition and texture. Permeability controls are apparent only in the Esperanza area where tuffaceous units with a high degree of primary porosity are altered to massive, silica-rich assemblages (section CA-EW). Compositional variations in host lithology appear to influence alteration only on a local scale. For example, less intense argillic and propylitic alteration in the Esperanza area is associated with a small stock of presumed Bocatoma-age diorite that intrudes the Pastos Blancos sequence. Similarly, intrusions of granodioritic and dioritic composition occur peripheral to Brecha Central (section CA-00) and are altered to argillic and propylitic assemblages, whereas the adjacent granitic rocks are intensely altered. In this case, lithological effects were likely combined with the progressive reduction and cooling of acid sulfate fluids away from the main breccia conduit. Structure and brecciation: The distribution of alteration is strongly influenced by structural controls, especially in the Brecha Central and Frontera areas. The AS II assemblage in particular is preferentially hosted in intra-mineral breccias and surrounding stockwork zones. Breccias formed along pre-existing ESE- and N- to NNE-trending structures of the Quebrada de Pascua and Quebrada de Pedro fault systems (Hamilton, 1998). North-trending faults and sub- horizontal fracture zones also control the AS II assemblage and stockwork mineralization in the Maria Tunnel and Pedro Este area. Likewise, steam-heated and late-stage acid sulfate alteration used existing structures to propagate downwards from surface and from the margins of the deposit. Elevation: Elevation has had a strong effect on the mineralogy and distribution of several alteration assemblages, due to both topographical and temperature-depth relationships. Topography has had the largest influence on the distribution of steam-heated and supergene assemblages, since the acid sulfate fluids responsible for alteration in each case were generated above the paleo-water table. The blanket-type distribution of steam-heated alteration at Pascua is typical of this environment (e.g. Schoen et al., 1974) and its preservation indicates that limited erosion has taken place in this region over the past 8 m.yr. The predominance of 59 jarosite and scorodite also appear to have been controlled by present-day topography (Chouinard and Williams-Jones, 1999). Jarosite is most abundant at the margins of Brecha Central, on the flanks of Quebrada de Pedro, roughly enveloping the sulfide-rich ore. A consistent vertical zonation in the mineralogy of the AS I assemblage indicates a gradual cooling of magmatic-hydrothermal fluids with increasing elevation. The co-existence of alunite, pyrophyllite, zunyite, and diaspore at depths greater than about 4500 m is consistent with temperatures of 200-380° C (Hemley et al., 1980; Reyes, 1990). The kaolinite-dominant assemblage at higher elevations suggests that temperatures had cooled to less than about 150- 200° C. Similarly, silicic alteration typically occurs above AS II alteration in the Brecha Central and Frontera zones, at elevations between 4600 to 4900 m. This suggests a decrease in silica solubility with gradual cooling or neutralization of the upwelling fluids. Equivalent lateral zonation is present in both the AS I and silicic assemblages at the margins of the hydrothermal system as alteration grades outwards to argillic and propylitic assemblages. S U M M A R Y : A L T E R A T I O N P A R A G E N E S I S I N A N E V O L V I N G H Y D R O T H E R M A L S Y S T E M Data from this study, combined with 4 0 Ar / 3 9 Ar ages presented in Bissig (2001), indicate that episodic magmatic-hydrothermal activity occurred in the Pascua-Lama region over a period of at least 1 to 2 m.yr. The deposit is complex, with multiple stages of alteration and several forms of Au, Cu, and Ag mineralization. In this study, we have focussed on the nature of alteration in the Esperanza, Brecha Central, and Frontera areas but several newly discovered zones in the Lama district have not been considered. The combined areal extent of this district is over 15 km 2 and recent resource estimates indicate that this region, including Veladero, is one of the largest high sulfidation systems discovered to date. Throughout this study, we have examined the relation between alunite-bearing alteration assemblages and ore-stage events, using petrographic, SWIR, geochemical, and isotopic methods. Based on these results, we can group alteration in the Pascua district into 3 intervals, according to the relation between alteration and precious-metal deposition. These are: (1) pre- mineral magmatic-hydrothermal alteration that occurred prior to ca. 8.8 Ma; (2) syn-mineral magmatic-hydrothermal alteration and related Au and Cu mineralization between about 8.8-8.1 Ma; and (3) limited post-mineral, late-stage oxidation and supergene alteration that record the collapse of the hydrothermal system. The features of each of these stages are described in the following section and are illustrated schematically in Figure 2.15. 60 Figure 2.15. Schematic cross sections showing the evolution of the Pascua deposit from pre-mineral (A), syn- mineral (B), and post-mineral (C) stages. A. Widespread, early hydrothermal alteration in the greater Pascua re- gion (prior to - 8.8 Ma). B . The initiation of brecciation and Au-Ag-Cu mineralization in the Esperanza, Brecha Central, and Frontera areas (age - 8.8-8.1 Ma). C. Late stage processes with local incursion of meteoric fluids and late pulses of magmatic steam. Post-ore faulting is inferred. 61 Pre-mineral alteration: The consistency of style and distribution of most wallrock alteration in the Pascua district suggests the alteration was generated by a single, large event. These features, combined with stable-isotope data of alunite and associated alteration minerals, indicate a magmatic- hydrothermal origin for this stage of alteration. The widespread acid sulfate (AS I) assemblage is consistent with the condensation of high-temperature magmatic vapors and subsequent disproportionation of magmatic S 0 2 at temperatures <400°C (Holland, 1965). Isotope data indicate that vapors were absorbed into fluids with magmatic H- and O-isotope signatures and there is no evidence for significant mixing with meteoric waters. These fluids would have become increasingly acid and oxidizing following the dissociation of H 2 S 0 4 and HC1 with decreasing temperature. Variations in the clay mineralogy of the AS I assemblage record a progressive cooling of these hydrothermal fluids with increasing elevation. Similarly, the zonation of acid sulfate and silicic assemblages to surrounding argillic and propylitic alteration is typical of high sulfidation systems and indicates progressive neutralization of hypogene hydrothermal fluids by wallrock reaction outward from the main fluid conduits (e.g. Stoffregen, 1987; Arribas, 1995). Vuggy silica alteration, which is generated from leaching by hydrothermal fluids of pH <2 and temperatures less than about 250° C (Stoffregen, 1987), is relatively scarce in the Pascua area. This scarcity suggests that either the early magmatic- hydrothermal fluids were not sufficiently acidic to leach A l from the host rock, or that fluids rose along multiple conduits and were not focused in any one area. Given the large areal extent of alteration and the abundance of pre-existing structural conduits, we consider the latter explanation the more likely. Syn-minerai alteration: Deposition of AS II alunite and associated Cu and Au mineralization occurred in the Brecha Central and Frontera zones at approximately 8.8-8.1 Ma. Wallrock alteration appears to be limited to structurally controlled acid sulfate (AS II) and silicic assemblages. Abundant alunite precipitated with the sulfide-rich APE mineralizing facies along open fractures and in the matrix of hydrothermal breccias. Isotopic data for this stage are nearly identical to the early acid sulfate alteration event and again indicate a dominantly magmatic fluid source and depositional temperatures between 200-350° C. Data are consistent with the condensation of high temperature magmatic vapors into fluids of dominantly magmatic origin (e.g., Rye et al., 1992). These fluids ascended to elevations of ca. 4500-4700 m, where they precipitated an 62 assemblage of alunite with enargite, pyrite, and native sulfur. The f02 and pH conditions of mineral deposition can be modeled on the basis of this assemblage, at temperatures calculated from alunite-pyrite isotope pairs (Fig. 2.16). These calculations indicate that ore precipitated from relatively oxidized, acidic, and sulfur-rich fluids. These findings are not consistent with a near-neutral, reduced ore-fluid that is inferred for many high sulfidation deposits (e.g., Hedenquist et al., 1998: details see Chap. 6). The occurrence of coarse-grained magmatic steam alunite above Brecha Central suggests that episodic pulses of high-pressure, high temperature magmatic vapors were released to the near-surface environment (e.g., Rye, 1993: details see Chap. 3). The timing of steam- heated alteration above the Brecha Central and Frontera areas is unconstrained. The 4 0 Ar / 3 9 Ar ages are inconclusive, and alteration may be related to either pre- or syn-mineral magmatic- hydrothermal activity, or likely both. Steam-heated alteration forms only under sub-aerial conditions and the relatively large thickness of this alteration zone suggests a progressive descent of the water table. However, isotopic data for alunite indicate that there was little, if any, meteoric water involved in the alteration process. Figure 2.16. The Pascua APE mineralizing event in terms of pH and the fugacity of 0 2 at 275°C and 0.5 molal total dissolved sulfur. Stability relations among minerals are calculated using reactions given in Appendix D. Pressure equals vapour saturation at temperature. The grey area represents the chemical characteristics of APE mineralizing fluids and is contained within the stability fields of alunite, pyrite, enargite, and native sulfur. Mineral abbreviations: alunite (alun), enargite (en), hematite (hem), magnet- ite (mag), muscovite (muse), pyrrhotite (po), pyrite (py), native sulfur (S°), tennantite (tenn). 63 Post-mineral alteration: Late-stage acid sulfate alteration events record the gradual collapse of the magmatic- hydrothermal system. Isotopic data indicate a progressive mixing of magmatic and meteoric fluids as temperatures cooled to <100°C. Alunite ± quartz and jarosite precipitated as thin veins and disseminations from sulfate-rich fluids that we propose were generated from the oxidation of precursor sulfides. Due to the overlap in 4 0 Ar / 3 9 Ar age data, these processes may have been active at surface or at the margins of the system even as magmatic-dominated fluids were still ascending at depth in the core of the deposit. The wide range of trace-element (eg., Pb, Ba, Zn) contents in late-stage alunite was likely derived from the remobilization of the precursor AS II sulfide assemblage. These late-stage, acidic, moderate-temperature fluids may have been responsible for the enrichment of Ag observed at higher elevations in the Brecha Central area (discussed in Chouinard and Williams-Jones, 1999). Supergene jarosite has been identified based on paragenetic observations and stable- isotope data. The occurrence of supergene jarosite suggests that most magmatic activity in this region had stopped by 7.5 Ma at the latest. This is the youngest alteration event recognized in the district. Evidence for the supergene remobilization of Au is observed in the Esperanza area, where visible Au occurs as coatings on high-level, oxidized fracture surfaces. Similar processes are apparent at the Penelope deposit on the Lama side of the property and are important factors in determining the economic feasibility of the region as a whole. R E F E R E N C E S Aoki, M., 1991. Mineralogical features and genesis of alunite solid solution in high temperature magmatic-hydrothermal systems. Journal of the Geological Survey of Japan, 277, 31-32. Aoki, M., Comsti, E.C., Lazo, F.B., and Matsuhisa, Y., 1993. Advanced argillic alteration and geochemistry of alunite in an evolving hydrothermal system at Baguio, northern Luzon, Phillipines. Resource Geology, 43, 155-164. Arribas, A., Jr., 1995. Characteristics of high sulfidation epithermal deposits, and their relation to magmatic fluids. Mineralogical Association of Canada Short Course Notes, v. 23, 419- 454. Arribas, A., Jr., Cunningham, C.G., Rytuba, J.J., Rye, R.O., Kelly, W.C., McKee, E.H., 64 Podwysocky, M.H., and Tosdal, R.M., 1995. Geology, geochronology, fluid inclusions, and stable isotope geochemistry of the Rodalquilar Au alunite deposit, Spain. Economic Geology, 90, 795-822. Beane, R.E., 1988. Hydrothermal alteration and mineralization at the Nevada prospect, Chile. Unpublished company report, Barrick Corp. Bissig, T., 2001. Metallogenesis of the Miocene El Indio-Pascua gold-silver-copper belt, Chile/ Argentina: Geodynamic, geomorphological and petrochemical controls on epithermal mineralization. Unpublished Ph.D. thesis, Queen's University. Caceres, C , Cordova, M., Rojas, F., and Vega, J., 1997. Projecto Mina Pascua - Informe de exploraciones: Temporada 1996-1997. Internal company report. Chouinard, A. and Williams-Jones, A.E. 1999. Mineralogy and petrography of ore and alteration assemblages at the Pascua Au-Ag-Cu deposit, Chile. Internal for Barrick Gold and Barrick Chile Ltda. Cooke, D.R. and Simmons, S.F., 2000. Characterisitcs and genesis of epithermal gold deposits. In Hagemann, S.G. and Brown, P.E., eds., Gold 2000. Reviews in Economic Geology, 13, 221-244. Craig, H., 1961. Isotopic variations in meteoric waters. Science, 133, 1702-1703. Cunningham, C.G., Rye, R.O., Steven, T.A., and Mehnert, H.H., 1984. Origins and exploration signficance of replacement and vein-type alunite deposits in the Marysvale volcanic field, west central Utah. Economic Geology, 79: 50-71. Di l l , H.G., 2001. The geology of aluminum phosphates and sulfates of the alunite group minerals; a review. Earth-Science Reviews, 53, 35-93. Flynn, F.R., and Burnham, C.W., 1978. An experimental determination of rare earth partition coefficients between a chloride-containing vapor phase and silicate melts. Geochimica et Cosmochimica Acta, 42, 685-701. Fulignati, P., Gioncada, A., and Sbrana, A., 1999. Rare-earth element behaviour in the alteration facies of the active magmatic-hydrothermal systems of Vulcano (Aeolian Islands, Italy). Journal of Volcanology and Geothermal Research, 88, 325-342. Gieseman, A., Jager, H.J., Norman, A.L., Krouse, H.R., and Brand, W.A., 1994. On-line sulfur- isotope determination using and elemental analyzer coupled to a mass spectrometer. Analytical Chemistry, 65(18). Giggenbach, W.F., 1992. Magma degassing and mineral deposition in hydrothermal systems along convergent plate boundaries. Economic Geology, 87, 1927-1944. 65 Godfrey, J.D., 1962. The deuterium content of natural waters and other substances. Geochimica et Cosmochimica Acta, 27: 43-52. Hamilton, J.V. 1998. Aspects of the structural setting, alteration, and Au-Ag-Cu mineralization at the Pascua deposit, east-central Chile. Unpublished company report, Barrick Corp. Hedenquist, J.W., Arribas, A.R., and Gonzalez-Urien, E., 2000. Exploration for epithermal gold deposits. In Hagemann, S.G. and Brown, P.E., eds., Gold 2000. Reviews in Economic Geology, 13, 245-278. Hedenquist, J.W., Matsuhisa, Y. , Izawa, E., White, N.C., Giggenbach, W.F., and Aoki, M., 1994. Geology, geochemistry, and origin of high sulfidation Cu-Au mineralization in the Nansatsu District, Japan. Economic Geology, 89, 1-30. Hemley, J.J., Montoya, J.W., Marinenko, J.W., and Luce, R.W., 1980. Equilibria in the system A l 2 03-S i0 2 -H 2 0 and some general implications for alteration/mineralization processes. Economic Geology, 75, 210-228. Henley, R.W., and Stewart, M.K., 1983. Chemical and isotopic changes in the hydrololoy of the Tauhara geothermal field due to exploitation at Wairakei. Journal of Volcanology and Geothermal Research, 15, 285-314. Holland, H.D., 1965. Some applications of thermochemical data to problems of ore deposits, II. Mineral assemblages and the composition of ore-forming fluids. Economic Geology, 60, 1101-1166. Jannas, R.R., Bowers, T.S., Petersen, U., Beane, R.E., 1999. High-sulfidation deposit types in the E l Indio District, Chile. In Skinner, B.J. ed., Geology and ore deposits of the Central Andes, SEG special pub. #7, 27-59. Jambor, 1999. Nomenclature of the alunite supergroup. Canadian Mineralogist, 37. 1323- 1341. John, D.A., Nash, J.T., Clark, C.W., and Wulftange, W.H., 1991. Geology, hydrothermal alteration, and mineralization at the Paradise Peak gold-silver-mercury deposit, Nye County, Nevada. In Raines, G.L., Lisle, R.E., Schafer, R.W., and Wilkinson, W.H., eds., Geology and Ore deposits of the Great Basin, Symposium Proceedings. Geological Society of Nevada, Reno, 1020-1050. Kesler, S.E., Russel, N., Seaward, M., Rivera, J., McCurdy, K., Cumming, G.L., and Sutter, J. F., 1981. Geology and geochemistry of sulfide mineralization underlying the Pueblo Viejo gold-silver-oxide deposit, Dominican Republic. Economic Geology, 76, 1096-1117. Kornexl, B. E., Gehre, M., Hofling, R., and Werner, R.A., 1999. On-line 8 1 8 0 measurement of 66 organic and inorganic substances. Rapid Communications in Mass Spectrometry, 13, 1685- 1693. Lottermoser, B.J., 1992. Rare earth elements and hydrothermal ore formation processes. Ore Geology Reviews, 7, 25-41. Maksaev, V., Moscoso, R., Mpodozis, C , and Nasi, C , 1984. Las unidades volcanicas y plutonicas del cenozoico superior en la Alta Cordillera del Norte Chico (29°-31° S): Geologia, alteration hidrothermal y mineralisacion: Revista geologica de Chile, 21, 11-51. Martin, M , Clavero, J., and Mpodozis, C , 1995. Estudio geologico regional del Franja El Indio, Cordillera de Coquimbo. Servicio National de Geologia y Minera, Santiago, Chile. Registered report IR-95-06, 232p. Martin, M.W., Clavero, J.R., Mpodozis, C , 1999. Late Paleozoic to Early Jurassic tectonic development of the high Andean Principal Cordillera, E l Indio Region, Chile (29-30° S). Journal of South American Earth Sciences, 12, 33-49. Michard, A., 1989. Rare earth element systematics in hydrothermal fluids. Geochimica et Cosmochimica Acta, 53, 745-750. Nixon, G.T., 1998. The volcanic geology of Pascua: Implications for mineral exploration. Unpublished company report, Barrick Corp. Ohmoto, H., and Lasaga, A.C., 1982. Kinetics of reactions between aqueous sulfates and sulfides in hydrothermal systems. Geochemica et Cosmochimica Acta, 46, 1727-1746. Petersen, E.U., and Thompson, A.J.B., 1992. Analysis of alunite and other thermally sensitive minerals. Microprobe Analyst (Internal report for the Dept. of Geology and Geophysics, University of Utah), April, 2-6. Ramos, V. A., Mahlburg Kay, S, Page, R. N. and Munizaga, F., 1989. La ignimbrita Vacas Heladas y el cese del volcanismo en el Valle del Cura, Provincia de San Juan. Revista Asociacion Geologica Argentina, 44, 336-352. Reed, M.J., 1995, Distribution of rare earth elements between aqueous fluids and granitic melt. Ph.D. thesis, Univ of Maryland at College Park. Reyes, A.G. , 1990. Petrology of Phillipine geothermal systems and the application of alteration mineralogy to their assessment. Journal of Volcanology and Geothermal Research, 43, 279- 309. Rye, R.O., 1993. The evolution of magmatic fluids in the epithermal environment: The stable isotope perspective. Economic Geology, 88, 733-753. Rye, R.O., and Alpers, C.N., 1997. The stable isotope geochemistry of jarosite. USGS Open 67 File Report 97-88, 28 p. Rye, R.O., Bethke, P.M., and Wasserman, M.D., 1992. The stable isotope geochemistry of acid sulfate alteration. Economic Geology, 87, 225-262. Rye, R.O, and Stoffregen, R.E., 1995. Jarosite-water oxygen and hydrogen isotope fractionations: preliminary experimental data. Economic Geology, 90, 2336-2342. Sandeman, H.A., Archibald, D.A.; Grant, J.W.; Villneuve, M.E. and Ford, F.D. 1999. Characterisation of the chemical composition and 4 0 Ar - 3 9 Ar systematics of intralaboratory standard MAC-83 biotite. In Radiogenic Age and Isotopic Studies: Report 12; Geological Survey of Canada, Current Research 1999-F, p.13-26. Savin, S.M., and Epstein, S., 1970. The oxygen and hydrogen isotope geochemistry of clay minerals. Geochimica et Cosmochimica Acta, 34, 24-42. Schwab, R.G., Herold, H, Gotz, C. and de Oliveira, N.P., 1990. Compounds of the crandallite type: Synthesis and properties of pure rare earth element phosphates. Neues Jahrbuch fuer Mineralogie, 6, 241-254. Schoen, R., White, D.E., and Hemley. J.J., 1974. Argillization by descending acid at Steamboat Springs, Nevada. Clays and Clay Minerals, 22, 1-22. Sillitoe, R.H., 1993. Epithermal models: Genetic types, geometric controls, and shallow features. In Kirkham, R.V., Sinclair, W.D., Thorpe, R.I., and Duke, J.M., eds. Mineral Deposit Modelling; Geological Association of Canada, Special Paper 40, p.403-417. Sillitoe, R.H., 1999. Styles of high sulfidation gold, silver and copper mineralization in the porphyry and epithermal environments. PacRim '99. Bali, Indonesia, 10-13 October, Proceedings, p. 29-44. Stoffregen, R., 1987. Genesis of acid sulfate alteration and Au-Cu-Ag mineralization at Summitville, Colorado. Economic Geology, 82, 1575-1591. Stoffregen, R.E. and Alpers, C.N., 1987. Woodhouseite and svanbergite in hydrothermal ore deposits: Products of apatite destruction during advanced argillic alteration. Canadian Mineralogist, 25, 201-211. Stoffregen, R.E., Alpers, C.N, and Jambor, J.L., 2000. Alunite-jarosite crystallography, thermodynamics, and geochronology. In Alpers, C.N., Jambor, J.L., and Nordstrom, D.K., eds., Sulfate Minerals; Crystallography, geochemistry, and environmental significance. Reviews in Mineralogy and Geochemistry, 40, p. 454-480. Stoffregen, R.E., and Cygan, G., 1990. An experimental study of Na-K exchange between alunite and aqueous sulfate solutions. American Mineralogist. 75, 209-220. 68 Stoffregen, R.E., Rye, R.O., and Wasserman, M.D., 1994. Experimental studies of alunite: I. 1 8 0 - 1 6 0 and D-H fractionation factors between alunite and water at 250-450°C. Geochimica et Cosmochimica Acta, 58, 903-916. Steiger, R. H. and Jager, E., 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359-362. Taylor, B E . , 1988. Degassing of rhyolitic magmas: Hydrogen isotope evidence and implications for magmatic-hydrothermal ore deposits. Canadian Institute of Mining and Mineralogy Special Volume, 39, 33-49. Thompson, A.J.B., Hauff, P.L., and Robitaille, A.J. 1999. Alteration mapping in exploration: Application of short-wave infrared (SWIR) spectroscopy. SEG newsletter, October 1999. Thompson, A.J.B., and Petersen, E.U., 1995. Characteristics of alunite in relation to ore- forming environment. Unpublished data. Volkert, D.F., McEwan, C.J.A., Garay, E., 1998. Pierina Au-Ag deposit, Cordillera Negra, North-Central Peru. In Pathways '98: Extended Abstracts Volume. 69 Chapter 3 A L U N I T E I N A N E V O L V I N G M A G M A T I C - H Y D R O T H E R M A L S Y S T E M ; T H E T A M B O H I G H S U L F I D A T I O N D E P O S I T , E L INDIO D I S T R I C T , C H I L E I N T R O D U C T I O N The Tambo high sulfidation gold deposit is located in the El Indio-Pascua Belt, within the Main Cordillera of Chile (Fig. 3.1). This belt hosts widespread zones of hydrothermal alteration and several world-class epithermal deposits and prospects, including E l Indio (>10 Moz Au produced), Pascua-Lama (proven and probable resources of 17 Moz Au and 560 Moz Ag) and Veladero (15 Moz Au, 230 Moz Ag). The Tambo deposit is relatively small in comparison, with a total production of just over 0.8 Moz Au from several tectonic-hydrothermal breccias and high- grade veins. Previous studies of the El Indio district (Siddeley and Areneda, 1985; Bennet, 1995; Jannas, 1995; Jannas et al., 1999) have constrained the geological setting of the Tambo deposit. This region has many features in common with other high sulfidation systems, although the deposit is unique in several aspects. 1. Limited erosion has preserved many features of the hydrothermal system, including the near- paleosurface. As such, multiple stages of alunite (KAl3(S04) 2(OH) 6) and acid sulfate alteration can be recognized over several hundred metres of vertical extent. These stages include each of the four major types of acid sulfate alteration described by Rye et al. (1992): magmatic-hydrothermal, magmatic steam, steam-heated, and supergene. 2. Mineralization is sulfide- and Cu- poor in comparison to most other deposits of this type (e. g., Cooke and Simmons, 2000; Sillitoe, 1999). Ore is hosted primarily in barite and alunite ± quartz. 3. A second stage of Au mineralization is hosted in, and associated with, alunite of magmatic steam origin (this study; Jannas et al., 1999). In this paper, we examine shallow-level magmatic-hydrothermal processes within a well- constrained geological framework. The goal of this investigation was to clarify the evolution of the Tambo deposit and, in particular, to determine the role of magmatic fluids, the mechanisms 70 El Indio Tambo o T UJ TERTIARY [TTTn Vallecito Fm: rhyolitic pyroclastics H Vacas Heladas Fm: dacitic pyroclastics r~' Cerro de las Tortolas Fm: predominantly andesitic flows Infiernillo Unit: granodioritic to dioritic intrusives Escabroso Fm: andesitic flows and volcaniclastic sediments I I Tilito Fm: dacitic pyroclasitic rocks PALEOZOIC TO JURASSIC: predominantly felsic intrusive and volcanic rocks {H§ Cuartitos Sequence m Pastos Blancos Group (undifferentiated) m Colorado and Chollay Units Fault Approx. area shown on map to left 20 km Figure 3.1. Geology of the El Indio district adapted from Jannas et al. (1999). Locations of the Tambo and El Indio properties are shown. See inset map for area location. 71 of ore-metal transport, and the nature of magmatic steam processes. Emphasis is placed on the temporal and spatial relationship between alunite and Au mineralization, because the isotopic, geochronologic, and geochemical characteristics of alunite can be used to resolve the nature of source fluids (e.g., Bethke, 1984; Rye et al., 1992). Eight different stages of alunite have been recognized at the Tambo deposit. The relative timing of each event was determined from the paragenetic sequence defined by Jannas (1995), with additional constraints from field relations and new 4 0 Ar / 3 9 Ar ages for alunite. G E O L O G I C A L F R A M E W O R K Regional geology The E l Indio-Pascua Belt is located in the center of a presently amagmatic segment of the Andean Cordillera. The lack of recent volcanism is attributed to subhorizontal subduction of the Nazca plate beneath the South American continent between 27° 30' and 33° S (Barazangi and Isacks, 1976). The district is approximately 100 km long and 30-40 km wide and straddles the Chile-Argentina border (Fig. 3.1). Hydrothermal systems formed within a NNE-SSW- striking tectonic depression bounded by the steeply west-dipping high-angle reverse Baiios del Toro Fault to the west and opposing structures in the Valle del Cura region to the east. Within this block, an Upper Paleozoic to lower Jurassic basement, predominantly composed of calc- alkaline felsic intrusive suites and volcanic rocks (Martin et al., 1999), is overlain by up to 1500 m of Tertiary subaerial volcanic strata. The latter are extensively preserved in the southern part of the belt, where the Tambo and E l Indio mines are located (Fig. 3.1), but are less widespread in the northern extremity of the district near Pascua-Lama and Veladero (Martin et al., 1995; Maksaev et al., 1984). The Tertiary volcanic sequence is briefly summarized here after Bissig (2001) and Martin et al. (1995). A thick succession (up to 1200 m) of dacitic and rhyodacitic tuffs with volcaniclastic sediments and subordinate basaltic and andesitic flows comprises the 23-27 Ma Tilito Formation. It is separated from the overlying 17.5-21 Ma Escabroso Formation by an angular unconformity marked by a persistent regolith horizon. The latter consists predominantly of andesitic lavas and hypabyssal intrusive bodies as well as volcaniclastic sedimentary rocks. After a brief deformation event, igneous activity continued and is represented by the 14-17 Ma andesitic Cerro de las Tortolas Formation and the contemporaneous Infiernillo Intrusive Unit. Magmatism decreased markedly thereafter. Dacitic tuffs of the Vacas Heladas Formation erupted from isolated centers between 11 and 12.7 72 Ma. A single 7.8 Ma dacite dike, the only igneous rock coeval with mineralization in the district, is reported from Pascua (Bissig, 2001). The 5.5-6.2 Ma Vallecito Formation rhyolitic tuffs are restricted to an ignimbrite sheet in the Valle del Cura (Ramos et al., 1989) and some occurrences in the E l Indio-Tambo area. Volcanism ceased in the Upper Pliocene after eruption of the 2 Ma Cerro de Vidrio rhyolite dome in the northeast corner of the belt (Bissig, 2001). Local geology The geology of the El Indio-Tambo district is dominated by an upper Oligocene to Upper Miocene sequence of intensely fractured volcanic and volcaniclastic rocks. The epithermal mineralization is hosted by rhyolitic ash-flow tuffs of the Tilito Formation ("Amiga tuff'; Jannas et al., 1999) and, at E l Indio, is overlain by Escabroso Formation andesitic lavas. Intrusive units and hydrothermal alteration of Escabroso and Infiernillo age are also present in the area. However, 4 0 Ar / 3 9 Ar ages for both El Indio and Tambo mineralization are significantly younger than these intrusions. The El Indio vein system formed between 7.8 and 6.2 Ma (Bissig, 2001), whereas the Tambo ore was deposited between 7 and 9 Ma (this study; Bissig, 2001). Tuffs of the Vacas Heladas Formation were deposited to the southwest of E l Indio in the Azufreras region, and rhyolites of the Vallecito Formation occur ca. 4 km south of Tambo. The structural grain of the E l Indio-Tambo district is dominated by NNE- and NW- trending normal and high-angle reverse faults and conjugate shear faults (Martin et al. 1995; Caddey, 1993). Mineralization is thought to have occurred during an early compressional event that was followed by regional relaxation and the development of shear fractures and normal oblique-slip faults (Caddey, 1993). Mineralization Gold (and Ag) mineralization at Tambo is hosted in three major hydrothermal breccias and several small high-grade veins. Two of the major breccia deposits, Kimberly and Wendy, are on the north-west and south flanks of Cerro Elefante, respectively (Fig. 3.2 and 3.3). These deposits have lateral and vertical dimensions of up to 300 m and 400 m, respectively, and consist of matrix-supported, monolithic breccias that contain altered fragments of Amiga tuff (Jannas et al., 1999). The main ore zone occurs between 4200 m and 4000 m a.s.l., with grades up to 4-10 g/t Au and 10 g/t Ag (Jannas et al., 1999). The Canto Sur deposit is located at a higher elevation on Cerro Canto, which separates the El Indio and Tambo mines. A l l three breccias are located at, or close to, the intersection of north-northwest and east-northeast fault 73 Figure 3.2. Map of the Tambo property modified from Jannas et al. (1999). Shaded areas indicate of mineralization. 74 75 systems in areas of intense shears and extensional joints (Jannas et al., 1999). Barite is common in both the Canto Sur and Kimberly breccias, and the original matrix has been completely replaced by barite-alunite ± quartz cement. No barite is present at Wendy, however, and the breccia matrix consists entirely of alunite ± quartz. Vein mineralization at Tambo is primarily hosted by the Reina, Deidad, and Indigena structures (Fig. 3.2). These veins are typically narrow (1-4 m wide), of limited vertical extent (< 200 m), and consist of primarily barite and alunite ± quartz. Ore grades average 15 g/t Au and 30 g/t Ag, but locally exceed 1000 g/t Au (Jannas et al., 1999). Two stages of mineralization were recognized by Jannas et al. (1999). Early Au mineralization (Stage 2; Table 3.1) is closely associated with barite in both the veins and the breccia deposits. Electrum, minor sulfides (including enargite, galena, and pyrite), native tellurium, other Te minerals, and native gold are concentrated along growth planes in the barite. Trace sulfides and gold occur interstitial to alunite overgrowths on barite grains. Gold also occurs locally on the rims of the breccia fragments. The second stage of mineralization (Stage 3; Table 3.1) is characterized by native gold that occurs in fractures and vugs, and interstitial to grains of alunite, hematite, jarosite, and barite. Inclusions of native gold also occur in alunite. Sulfides are rare, although minor enargite occurs at depth in the Wendy breccia, intergrown with coarse alunite grains. Early tellurides are typically replaced by minerals such as poughite, rodalquilarite, and tellurite (Jannas et al., 1999). M E T H O D S O F A N A L Y S I S Alteration assemblages were identified by field and petrographic observations, using the paragenetic relations described by Jannas et al. (1999). Field work was carried out between February, 1998 and April, 1999. Access at the time was limited to surficial outcrops, open pits (Kimberly, Wendy, and Canto Sur), and drill core. The mineralogy and character of each paragenetic stage were determined by examination of thin and polished sections using standard optical methods and X-ray diffraction (XRD). In addition, short-wave infrared (SWIR) analyses were made in situ using a Portable Infrared Mineral Analyzer (PIMA), a field-portable spectrometer that measures reflectance spectra in the SWIR band. Spectra were analyzed using PIMA View 3.0 software, and mineral identification was made by comparison to standards in the SPECMTN database (Thompson et al., 1999). Representative samples from each 76 CD CD "O £3 w u CD bo a, B S 2 d U r 1 CD •5 •<=> % Vi CD 6 0 CD d 'ss "5 C ! J 5 -—> a o CD •£= CO C g G 43 ^ O s ^ co d , Si O CD CD co -tl to & 'tl M tu 43 1) o o CD O cd C o o CO 4 * T3 e CO ^ CD N 3 N Er t! CD CO to ;3 CD W S f 1> 43 Co ^—^ >1 CO CD OH .13 a . . j = _c3 cO CD 1-1 CO 2 "o -o .c CO CD 6 0 cd < CD t; o O H CD 6 0 53 =S 4> '5b S3 cd 3> 43 ? CD 43 CD 43 CD CD a. c a, C co 6 0 . _ ^ CD § .SP CD CO 43 CO 6 0 03 CO to cu 2 •s X 3o to . .E a> cu • - > c oil ca c I co cu .Q to *i a s CO I 13 0) to x i = £ «> to > CQ CN CU O) CO CO CO co CO >- r- •*- ' o S £ > ne d m in e ja r o gr ai  m in e +1 gr ai  J Z IO ne - 3 CN ne - c m LL _3 o •H to co co M o -H to CO 00 to CO CD o D) +1 CO CM 0 0 CN -H co CN o +1 CO CN o +1 CN 0> CO CO o -*rf CO "S i ® £ . £ ro c a, ro ^ ^ o ro.E > E c c ro £ U- -S -c '5 £j +1 ro'? ^ § CO co CO O I U O o CO CJ c " D o ne  'co c se , ve  Fi ne  co ar  an d i c < cu CL) Q> > O ro ro _H X o ro <u E .55, _- .22 ro ro — CO ^ 7 3 XI to - g m  ro ro co cu ro ~ cr > 5 . CD i_ - CD > ° -s C O ) _ CO E C . CD ^ i - 5-g ro g uj io >; 3 < = , < 3s 15^ c: it +1 ^" .2 = § .9 ro co ^ ro «I Q. O 43 •| iS > ro x ro ' CD CD c c 3 CD ^ =3 o o O c .o .N CD .C 77 paragenetic stage were selected for further analysis. Six samples were dated by 4 0 Ar / 3 9 Ar laser step-heating methods. Al l analyses were carried out at the Queen's University 4 0 Ar- 3 9 Ar laboratory equipped with a Mass Analyzer Products M A P 216 mass spectrometer. About 10 mg of each sample were irradiated for 7.5 h at McMaster nuclear reactor in Hamilton, Canada, using biotite standard Mac-83 as a radiation flux monitor (24.36 ± 0.17 Ma; Sandeman et al., 1999). J-values for individual samples were determined by second-order polynomial interpolation and, for the samples discussed in this article, were about 2.4 x 10"3 and varied by less than 0.02 x 10"3. Samples were step-heated using a defocused L E X E L 3500 Ar laser-beam. Ages were calculated using the decay constants suggested by Steiger and Jager (1977), and all errors are given at 2a. Geochemical characteristics of alunite were determined by analysis of individual grains and sample composites. Grain mounts and polished sections were analyzed by scanning- electron microscopy with energy-dispersion spectroscopy (SEM-EDS) and electron-probe microanalysis (EPMA). E P M A data were collected on a fully-automated C A M E C A SX-50 microprobe, operating in the wavelength-dispersion mode with the following operating conditions: 15 keV excitation voltage, 10 nA beam current, 15 \xm beam diameter, and total count time under 65 seconds. Analytical parameters were chosen to minimize beam damage caused by the volatilization of alkali elements (Petersen and Thompson, 1992). Data reduction was done with the 'PAP' method of Pouchou and Pichoir (1985). Alunite separates (approximately 1 g samples) were carefully hand-picked and treated using a 1:1 HF/H 2 0 solution to remove silicate contaminants (Wasserman et al., 1992). The purity of separates was verified by XRD, but trace to minor amounts (i.e., < 10 volume %) of quartz, clays, or other impurities may be present. Samples were analyzed by inductively-coupled plasma mass spectrometry (ICP-MS) for trace metal and rare-earth-element (REE) contents by ALS Chemex Laboratories, Canada. For trace metal analysis, samples were dissolved in a mixture of perchloric, nitric and hydrofluoric acids to ensure total dissolution. For REE determination, samples were fused with lithium metaborate prior to analysis. Alunite R E E data are normalized to average REE content of volcanic tuffs from the Tilito Formation (T. Bissig, unpub. data; average values given in Table 3.2) that host ore in the El Indio district. 78 Table 3.2: Average major and rare earth element geochemical data for unaltered rocks of the Tilito Fm (n=3). Data from Bissig (2001). Tilito Fm Na 2 0 (wt %) 2.85 K 2 0 3.85 CaO 4.48 MgO 1.25 MnO 0.07 F e 2 0 3 ( F e total) 4.33 A l 2 0 3 16.03 P 2 O 5 0.15 S i 0 2 63.39 T i 0 2 0.57 Total 97.35% REE ICP-MS (ppm) La 31.6 Ce 62.6 Pr 7.2 Nd 25.2 Sm 4.8 Eu 1.0 Gd 3.9 Tb 0.6 Dy 3.5 Ho 0.7 Er 2.1 Tm 0.3 Yb 2.1 Lu 0.3 Complete stable isotopic analyses (5 3 4S, 8D, 8 1 8 0 S o 4 , and 8 1 8 0 0 H ) were determined for alunite from each paragenetic stage. Sulfur-isotope data were also obtained for selected barite and enargite samples. A l l data were collected at the U.S. Geological Survey Isotope Laboratory in Denver, USA. Samples were separated and prepared according to the procedure of Wasserman et al. ( 1 9 9 2 ) . Sulfur isotopic analyses were determined by an on-line method using an elemental analyzer coupled to a Micro mass Optima mass spectrometer (Giesemann et al., 1 9 9 4 ) . Analytical precision is better than ± 0.2%o. D/H analyses were performed by a step- heated technique modified from Godfrey ( 1 9 6 2 ) , with analytical precision better than ± 3%o. Alunite oxygen-isotope data were collected for both sulfate and hydroxyl oxygen. For S 1 8 0 S 0 4 79 analysis, alunite was dissolved in a hot NaOH solution, and sulfate was precipitated as BaS0 4 (Wasserman et al., 1992). The BaS0 4 precipitate was then analyzed by fluorination with BrF 5 at 580°C in accordance with the standard analytical procedure of Clayton and Mayeda (1963), and as used by Pickthorn and O'Neil (1985). Analytical precision is estimated at ± 0.15%o. 8 1 8 O O H was determined by material balance using the total-0 isotopic composition of alunite (measured by fluorination with BrF 5 , described above) and 8 I 8 0 S o 4 results. Several oxygen- isotope analyses were also determined by pyrolysis with a Finnigan TC/EA coupled to a Finnigan Delta Plus X L mass spectrometer using continuous-flow methods modified from Kornexl et al. (1999; details in Appendix B). Analytical precision is better than ± 0.3%o. Fluid-inclusion gas analyses were carried out at the U.S. Geological Survey fluid- inclusion and noble-gas laboratories in Denver, USA. A l l samples were initially crushed in a sealed stainless steel tube to mechanically release gas from inclusions. After analysis of this gas, the crushed alunite was heated at 200°C for 1 h in the same tube to generate thermally released gas. The temperature chosen was sufficient to decrepitate fluid inclusions (especially vapor-rich) that still remained after crushing, yet avoided thermal decomposition of alunite. The released gases from both extractions were split for gas-composition analysis on a calibrated quadrupole mass spectrometer (Pfeiffer Vacuum Prisma) and the remaining gas was analyzed for He and Ne isotopes on a high-resolution M A P 215-50 noble-gas instrument. Gas species were analyzed at approximately 10"8 torr, with each analysis consisting of an average of 40 to 60 matrix calculations that correct for ion sensitivity and gas ionization fragmentation, and which calibrate the intensity response to standards. Results were determined as mole % summed to 100% when normalized to nitrogen. The remaining gas was then processed to extract He and Ne by cryogenically removing other gases on charcoal, gettering the gas (Saes ST707), and concentrating the purified He and Ne in a small-volume charcoal trap. Helium, and then Ne, were separately analyzed by selective thermal desorption from the charcoal and expansion into the mass spectrometer. Helium and Ne calibration was achieved by processing laboratory air corrected for barometric pressure, temperature, and relative humidity using identical procedures (4He=6.414 e"12 mol/volt ± 0.016%, 3He=9.951 e"20 mol/cps ± 0.144%, 3He/ 4He detector discrimination = 1.03074 ± 1.68%; Ne isotopes are treated as normalized percentages and ratios only, but corrected for 4 0Ar* 4" and CC^"1-1" isobaric interferences to 2 0 Ne and 2 2Ne). 80 A L U N I T E P A R A G E N E S I S I N T H E T A M B O S Y S T E M Eight stages of alunite are recognized in the Tambo district. The paragenetic sequence (Table 3.1) is based on field and petrographic observations, using relations and terminology (e. g., Stages 1, 2, and 3) presented in Jannas et al. (1999). The characteristics of each stage are described below. a) Brecha Sylvestre: Brecha Sylvestre is a relatively small, barren breccia body at the southwestern end of Cerro Elephante (Fig. 3.2). The breccia occurs between 4430 and 4480 m (all elevations are reported as elevation above sea level), approximately 200 m above the zones of highest gold production in the Wendy and Kimberly orebodies. The breccia is mainly clast- supported, locally silicified, and shows evidence of strong acid-leaching which resulted in removal of most of the matrix material (Bennet, 1995). The matrix consists of open vugs lined with alunite and minor quartz (Fig. 3.4a). Alunite typically occurs as pale white to yellow, elongate bladed to acicular crystals from 20 to 200 |nm in length. Jarosite is also common and overgrows silicified breccia fragments and matrix crystals. b) Stage 1 - Pre-ore acid sulfate alteration: Widespread acid sulfate alteration of Amiga Tuff wallrock is recognized throughout the Tambo area (Stage 1). This assemblage is surrounded by argillic- and propylitic- altered rocks away from the main orebodies (this study; Jannas, 1995). The occurrence of altered clasts with silicified rims in the hydrothermal breccias indicates that alteration occurred prior to brecciation and subsequent mineralization. Alteration occurs as fine intergrowths of alunite-quartz that selectively replaced feldspar crystals and locally forms massive clusters in the matrix of the Amiga tuff (Fig. 3.4b). Pumice clasts within dacite tuff fragments in the breccia pipes are also replaced (Jannas, 1995). Original rock textures are generally preserved. Associated minerals include clays (kaolinite-dickite), pyrophyllite, rare pyrite, and diaspore. Alunite commonly occurs as clusters of tabular to bladed crystals between 50 \im to 2 mm in length. Crystals locally have cores of alumino- phospho-sulfate (APS) minerals, including woodhouseite, svanbergite, and florencite (Table 3.3). Alunite grains are also locally pitted with corroded margins, and are rarely overprinted by fine-grained (10-20 jLim) alunite ± quartz. c) Stage 2 - Early Gold Mineralization: Stage 2 alunite post-dates the emplacement of the three major breccias and is closely associated early Au deposition. Alunite typically occurs as euhedral, golden-yellow bladed crystals up to 1 cm long that have overgrown barite and breccia fragments. Finer-grained tabular crystals less than 50 u,m in length also occur. Alunite 81 Figure 3.4. A) Medium-grained, tabular alunite crystals in the matrix of Brecha Sylvestre (PTS, crossed polars). Scale bar = 50 um. B) S E M backscatter image of Stage 1 quartz (qtz)-alunite (alun) alteration. Bright zones within alunite grains are inclusions of woodhouseite-svanbergite. Scale bar = 25 um. C) Walthierite (light grey)-alunite (dark grey) zoning in Stage 2 alunite from the Canto Sur deposit (SEM back- scatter image). Alunite is hosted in barite (white). Scale bar = 50 um (SEM backscatter image). D) Native A u (white) associated with alunite (grey) from Wendy (DDH-92a, 206.7m; S E M backscatter image). Bright circle in centre of image due to electron beam damage. Scale bar = 25 um. E) Fine-grained (white) Stage 3 alunite (alun) overgrowing coarser-grained Stage 3 alunite (grey) in the matrix of the Wendy breccia. Minor enargite (en) is associated with coarse-grained alunite. Scale bar = 1 cm. F) Banded alunite-hematite vein from the Kimberly deposit. Scale bar = 1 cm. 82 Figure 3.4. (Continued) G) Zoned alunite (light grey)- huangite (dark grey)- woodhouseite/svanbergite (white) grains from the Kimberly pit (SEM backscatter image). Scale bar = 10 urn. H) Steam heated alteration in outcrop near the top of Cerro Elefante. Hammer for scale. I) Fine-grained, pseudocubic alunite grains (grey) with minor jarosite (white). Scale bar = 10 um (SEM backscatter image). of this stage is recognized by complex compositional zoning and the presence of walthierite (Ba alunite end-member; Table 3.3). Walthierite was first recognized in the Reina vein (Beane, 1991; L i et al., 1992), where it forms irregular to oscillatory compositional zones (Fig. 3.4c) and aggregates in coarse-grained alunite crystals. Stage 2 alunite also occurs in the Kimberly and Canto Sur breccias in close association with barite and trace quartz. d) Stage 3 - Late Gold Mineralization: Deposition of Stage 3 alunite overlapped with the later stage of Au deposition and continued after mineralization ceased. Alunite is characterized by its nearly stoichiometric composition and variable crystal size and shape. Fine-grained alunite occurs as lath-like to thin bladed crystals, rarely greater than 30 pin long, that form overgrowths on breccia fragments and Stage 2 barite crystals in the Kimberly and Canto Sur breccia pipes. Alunite commonly exhibits the characteristic 'earthy' texture described by Jannas et al. (1999) and locally contains inclusions of native gold (Fig. 3.4d). Coarser grained varieties (up to 2 cm) of white to pinkish-brown, bladed alunite occur in veins, as cement to breccia fragments, as overgrowths on Stage 2 alunite crystals, and as local 83 intergrowths with enargite (Fig. 3.4e). Fine-grained alunite crystals also postdate the coarser bladed veins and breccia matrix. Minor amounts of quartz and kaolinite are associated with this stage. Late jarosite and scorodite that overprint Stage 3 alunite are common. e) Coarse banded alunite (± hematite) veins: Coarsely crystalline veins up to 20 cm wide of alunite ± hematite (Fig. 3.4f) are common in the upper and marginal parts of the Kimberly open pit, and occur locally in the Wendy deposit. The veins cross-cut both Stage 1 alteration and hydrothermal breccias, but the timing of the veins relative to Stage 3 alunite is unknown. The veins are characterized by coarse alunite crystals (1 to 5 mm long) that grew in successive bands outward from the wallrock. Minor hematite and rare quartz ± jarosite occur interstitial to the alunite crystals. The presence of hematite gives the veins a distinct reddish colour that allows them to be easily distinguished from the surrounding wallrock. Where hematite is absent, the veins have a cream-white colour. f) Huangite-bearing veins: Another variety of late veins occurs in the Kimberly and Kimberly West open pits. These veinlets are thin, pink to white, cryptocrystalline, and are composed of alunite, huangite (Ca alunite end-member; Table 3.3) ± kaolinite, quartz, and trace pyrite (Fig. 3.4g). Alunite occurs as clusters of tabular to anhedral grains averaging 50 to lOOiJm in length, locally overprinted by late jarosite and scorodite. The veinlets cut Stage 1 alteration, but timing with respect to Stage 2 or later events is unknown. Huangite had not been recognized previously in the Tambo area, but was first described in the El Indio district in association with kaolinite and pyrite (± alunite, pyrophyllite, sericite, minamiite, woodhouseite) adjacent to the El Indio Campana B vein (Li et al., 1992). Alunite in this vein has been dated at 6.21 ± 0.26 Ma and represents the youngest alteration event in the district (Bissig, 2001). Table 3.3. Minerals of the alunite supergroup identified in the Tambo area. Alunite-jarosite group alunite natroalunite minamiite huangite walthierite jarosite KAI 3(S0 4)2(OH)6 NaAI3(S04)2(OH)6 (Na,K,Ca)2AI6(S04)4(OH)i2 CaAI 6(S0 4)4(OH)i 2 BaAI 6(S0 4)4(OH)i 2 KFe 3(S0 4)2(OH) 6 Beudantite group svanbergite woodhouseite SrAI3(P04)(S04)(OH)6 CaAI3(P04)(S04)(OH)6 Crandallite group crandallite florencite-(Ce) florencite-(La) CaAI3[(P03(Oi/2(OH)i/2]2(OH)6 CeAI3(P04)2(OH)6 LaAI3(P04)2(OH)6 84 g) Near-surface alteration: A powdery, friable assemblage of silica- kaolinite ± alunite and native sulfur is present at higher elevations (4300 - 4500 m) in the Tambo district (Fig. 3.4h) and occurs as alteration blankets on top of Cerro Elephante, Cerro Canto, and the adjacent Azufreras and Sol Poniente areas. Alteration is commonly pervasive, with the matrix of the host rocks and feldspars in tuff fragments completely replaced. Alteration also extends downward along fractures, particularly beneath the surficial alteration zone on Cerro Elefante, and locally overprints older acid sulfate assemblages. Alunite in the near-surface zone is restricted to irregular pods and lenses and occurs as fine-grained (<15-20 Lim), tabular, and less commonly pseudo-cubic crystals intergrown with fine-grained silica. The mineralogy and distribution of this assemblage is typical of steam-heated alteration (e.g., Schoen et al., 1974). h) Late alunite veins ±jarosite: Late, fine-grained to cryptocrystalline alunite ± jarosite veinlets are common in the Tambo area. These veins occur at all depths investigated by drilling (down to 3800 m elevation) and cross-cut all other alteration types. Alunite is typically fine- to medium-grained (10-50 \im), tabular to bladed in habit, and is locally intermixed with quartz. Alunite is commonly overgrown and locally cross-cut by fine-grained jarosite. In rare occurrences, extremely fine-grained (2-10 |im), pseudo-cubic alunite is intergrown with kaolinite ± jarosite (Fig. 3.4i) in veinlets that cross-cut wallrock altered by illite-smectite ± chlorite. Sufficient alunite of this last style could not be separated for further analysis. G E O C H R O N O L O G Y Six alunite samples were dated by the 4 0 Ar / 3 9 Ar laser step-heating method to supplement existing age data for the Tambo area (Jannas et al., 1999; Bissig, 2001) and to confirm the relative timing of the alteration events. Results (Table 3.4) indicate that Stage 1 wallrock alteration (11.0 ± 0.3 Ma) occurred 2 to 3 m.yr. prior to either Stage 2 or Stage 3 ore deposition. Alteration is contemporaneous with Vacas Heladas volcanism, and extensive hydrothermal alteration of this age can be traced throughout the E l Indio-Pascua Belt (Bissig, 2001). A sample of coarse, vug-fill alunite from Brecha Sylvestre, dated at 10.4 ±0 .3 Ma (Bissig, 2001), is also pre-ore and may overlap with Stage 1 alteration. The timing of Stage 2 mineralization is constrained by 4 0 Ar / 3 9 Ar data (8.7 ± 0.2 Ma) for a sample of coarse-grained alunite from the Reina vein that contains inclusions of gold. Alunite in this sample is also intergrown with barite that contains precious metals. Stage 3 85 b CN CN CD O) CM CD d CM CO LO CO CO* CM LO CO CO ro ro CM CN a CD CN CN CO pi CD CD h- rd CD CM CU to 9> I cS 0) 3 CO CO 10- CD to "O CD CO (1) CM + co to "D. CM -4. + CO CM -4. + O 5 cu "a. E co co cu CO CO T— ro •sr O s CN in s o o cd d d d o o o o o o d o o o d o d o d o d o CO CO oc? CM Lr? CO T— 12 .7 8)  T— d CD CO d 12 .7 8)  11 .1 0 8. 72  I 6. 97 1 8. 97  ( 7. 27  I 8. 15  ( CN i> ro • So CD 00 0 0 d CD (0 .1 4)  I ( 0. 27  (0 .4 1]  CO- CO d (0 .1 4)  (2 .8 0]  O CO CO CO CO in CM o CO CO CM LO CD in m CO co d d d d d CM h- 00 CD 00 CD 00 00 ro o T— od cd od i< TB -1 06  TB -1 10  TB -1 03  TB -1 01  TB -1 04  TB -1 11  T1 0a  T0 1 in  T3 5 T1 7 T 02 a T 33  CM C a- ve  SH  LV  BV  86 mineralization is slightly younger, based on data for medium-grained alunite (8.2 ± 0.2 Ma) sampled from the matrix of the Kimberly breccia. The near-surface, blanket style of alteration (8.9 ± 0.4 Ma) was sampled from Cerro Elephante, above the Kimberly deposit. Alteration overlaps in time with Stage 2 and may represent the surficial expression of early gold mineralization. The timing of huangite-bearing veins (8.6 ± 0.4 Ma) relative to mineralization is uncertain. The relatively large error for this analysis is attributed to the small grain size of alunite and a smaller quantity of Ar gas released during heating. These veins may overlap with either Stage 2 or Stage 3, or may be intermediate between the two. Cross-cutting relations could not be determined in the field. It is clear, however, that alunite in these veins is unrelated to the much younger huangite-bearing Campana B vein at E l Indio (6.2 ± 0.3 Ma; Bissig, 2001). Late stage alunite (± jarosite) was sampled from thin veinlets that cross-cut Stage 1 alteration in the Canto Sur deposit. The plateau age for this sample (7.3 ± 0.1 Ma) is approximately 1 m.yr. younger than the main mineralizing events in the Kimberly and Wendy areas. It should be noted, however, that Bissig (2001) reported an age of 7.1 ± 0.2 Ma for alunite thought to be related to Au deposition in the Canto Sur breccia. The paragenesis of this sample is unclear. The style and mineralogy of both alteration and ore assemblages at Canto Sur are similar to those in other parts of the Tambo deposit. These features suggest a common origin. If mineralization in this area is younger, however, then late-stage alteration is coeval with, and not younger than, Canto Sur ore-stage alunite. S T A B L E - I S O T O P E S T U D Y Stable-isotope analyses of alunite and associated alteration minerals were determined for each paragenetic stage. Alunite contains four stable-isotope sites (8 3 4S, 8D, 81 80So4, and 8 1 8 O O H ) and data can be used to characterize the fluid source and environment of deposition for each stage of alteration (Rye et al., 1992). Sample descriptions and isotope results are presented in Table 3.5. The data show considerable variation in alunite 8 3 4S, with values ranging from 0 to +27%o (Fig. 3.5). Limited data for Stage 3 enargite range from about -2 to - 4%o, consistent with values reported in Jannas et al. (1999). Estimation of 8 3 4S for bulk sulfur in the magma (l-3%o; Fig. 3.5) is based on data for magmatic steam alunite (cf. Rye et al., 1992) and agrees with estimates presented by Jannas et 87 Table 3.5. Sample descriptions and stable isotope results for all Tambo alunite, barite, and enargite samples. Elevation given in metres above sea level. Abbreviated mineral names; alunite (alun), quartz (qtz), kaolinite (kao), dickite (dick), pyrophyllite (pyl), walthierite (wal), jarosite (jar), enargite (enarg), scorodite (scord), hematite (hem), bar (barite), native sulfur (S). S a m p l e L a b # S t a g e D e p o s i t E l e v . D e s c r i p t i o n S D 8 3 4 S 8 1 8 O s 0 4 8 l 8 0 O H 1 (m) (per mil) Alunite: WND-92A, 190.0m T09a 1 Wendy 4113 Qtz-alun alteration -98 6.2 9.4 3.7 WND-92A, 290.8m T10a 1 Wendy 4135 Qtz-alun alteration -53 22.1 15.8 10.1 KBD-85, 16.0m T16a 1 Kimberly 4113 Qtz-alun-pyl alt 7.2 KBD-85, 306.8m T29a 1 Kimberly 3887 Qtz-alun alteration -32 23.1 17.2 14.3 KBD-85, 94.1m T30a 1 Kimberly 4052 Qtz-alun alteration -30 1.1 20.3 14.3 WND-92a, 82.7m T31a 1 Wendy 4090 Qtz-alun-dick alteration 12.4 RA-01 T01a 2 Reina 4010 Vein alun with bar, walth, Au -28 27.3 21.0 15.5 RA-4, 41.6m T24a 2 Reina 4025 Coarse alun with walth -27 27.4 20.3 16.3 CS-29 T26a 2 Canto Sur 4522 Fine-grained alun, bar with Au -55 27.3 16.5 10.8 CS-01b T27a 3 Canto Sur 4474 Alun with bar in breccia matrix -38 2.1 19.6 15.3 CS-08 T02b 3 Canto Sur 4474 Matrix alun over bar -78 1.8 10.1 3.6 LN-24 T03a 3 Kimberly West 4130 Matrix alun, overgrowing T03b -47 5.5 17.7 13.2 LN-24 T03b 3 Kimberly West 4130 Matrix alun, overgrowing qtz -46 1.3 19.2 12.8 WND-86, 155.0m T04a 3 Wendy 4017 Matrix alun, with enarg and scord -63 1.7 15.0 8.1 WND-86, 237 m T06a 3 Wendy 3960 Matrix alun hosting enarg -85 1.7 11.9 5.7 CS-04 T20a 3 Canto Sur 4475 Matrix alun -41 -0.2 WND-92a, 37.5m T21a 3 Wendy 4080 Matrix alun hosting enarg -80 1.7 15.9 4.9 WND-92a, 231.8m T22a 3 Wendy 4122 Matrix alun -54 2.8 19.7 14.7 KB-43 T23a 3 Kimberly 4075 Matrix alun -79 0.7 LN-09 T25a 2 Kimberly West 4240 Matrix alun with bar, qtz -42 25.7 19.6 16.4 WND-92a, 13.7m T28a 3 Wendy 4075 Fine-grained matrix alun -98 2.8 11.1 3.6 KB-08 T36a 3 Kimberly 4050 Matrix alun with scord -51 3.5 18.0 12.0 KB-47 T07a MS Kimberly Coarse banded vein with minor hem -40 0.5 20.5 14.0 KB-33B T08a MS Kimberly 4100 Coarse banded vein with hem -48 1.1 16.4 8.8 WN-07 T32a MS Wendy 4180 Banded vein with hem -69 1.4 15.9 9.7 KB-01 T33a MS Kimberly 4050 Coarse banded vein -60 0.4 18.8 15.6* KB-44 T34a MS Kimberly 4075 Coarse banded vein with hem -74 0.3 11.3 KB-09 T14a Ca Kimberly 4060 Thin veinlet cross-cutting altered tuff -118 1.8 11.7 3.2 LN-02 T35a Ca Kimberly West 4260 Thin veinlet cross-cutting altered tuff -56 9.2 20.6 14.2 KB-37B T13a SH Kimberly 4480 Near-surface qtz-kao-alun-S alteration -53 5.9 18.8 12.3 KB-40 T17a SH Kimberly 4460 Near-surface qtz-kao-alun-S alteration -13 4.9 23.2 18.8 WN-14 T18a SH Cerro Elephante 4450 Powdery qtz-alun veinlets -62 16.6 15.4 11.9* CS-08 T02a LV Canto Sur 4474 Fine-grained alun-jar over Stage 3 alun -111 -0.2 10.0 2.5 KBD-85, 306.8m V T11a LV Kimberly 3887 Alun-jar veinlet in Stage 1 alteration -110 5.4 7.0 1.3 KBD-85, 125.7m T12a LV Kimberly 4027 Alun-qtz Qar) veinlet -57 2.0 15.7 9.8 WND-86, 226.5m T15a LV Wendy 3969 Brown jar-alun vein -68 4.3 16.1 9.3 Barite: RA-01 Tb01 2 Reina 4050 Coarse vein fill, associated with T01a 26.8 13.8 CS-08 Tb02 2 Canto Sur 4474 Coarse matrix crystals, with alun 24.5 17.3 CS-30 Tb03 2 Canto Sur 4396 Coarse matrix crystals, mineralized 23.8 17.1 Enargite: WND-86, 155.0m T04e 3 Wendy 4017 Enarg intergrown with alun -2.4 WND-86, 237m T06e 3 Wendy 3960 Enarg intergrown with alun -3.7 ' calculated using BrF5 total oxygen data, except samples marked by * 88 Late stage S t e a m heated Ca-aluni te ve ins 1 B a n d e d alunite ve ins S tage 3 S t a g e 2 S t a g e 1 -10 8 3 4 S V O O • ! 1 1 mm mm mm 10 15 5 3 4S (per mil) o enargite A barite • alunite mm 20 25 30 Figure 3.5. Range in 8 S data (in %0) for all stages of Tambo alunite. Also shown is data for associated enargite (this study; Jannas et al, 1999) and barite from Stages 2 and 3. Es- timated 834Sj;s ~ 1 to 3 per mil. See text for discussion. 89 al. (1999). The composition of paleometeoric waters is estimated at 8D = -100 ± 10%o (B. Taylor, pers. comm.). However, isotope data for jarosite of supergene origin from the Pascua deposit (Chap. 2), located to the north of E l Indio-Tambo, suggest that this value has been slightly lighter {ca. -125%o). Variations of this magnitude are expected during the 3-4 m.yr. duration of the Tambo hydrothermal system, and a range of meteoric SD values is shown on all isotope plots accompanying the following discussions. Stable-isotope systematics for each paragenetic stage are presented below. For ease of data interpretation and discussion, some samples are grouped with other, related, stages. In each section, 5 3 4S data are discussed first, followed by 8 1 8 0 and 8D relations. Brecha Sylvestre, Stage 1, and Stage 2: Magmatic-hydrothermal alunite Figure 3.6 summarizes the 8 3 4 S, SD, and 5 1 8 0 data for Brecha Sylvestre, Stage 1 and Stage 2 alunite. Stage 1834S data range from 1 to 23%o (Fig. 3.6a). Stage 2 alunite is heavier, with 8 3 4S values of 26 to 27%o. Sulfur data for the two Brecha Sylvestre samples (14 and 15%o) are lighter than Stage 2 alunite, but are intermediate to Stage 1 values. Alunite 8D and 8 1 8 0 values are shown in Figure 3.6b, along with the range of calculated fluid compositions in equilibrium with samples from each alteration event (calculated from equations in Stoffregen et al., 1994). Both Stage 2 and Brecha Sylvester alunite fluids overlap with the compositions of fluids in felsic magmas (Taylor, 1992). Most Stage 1 alunite fluids also plot within this range, indicating a dominant magmatic fluid component. One lighter 8DH2o value (-94%o) suggests local mixing with meteoric fluids. In both Stages 1 and 2, there is a linear trend to lighter isotopic values. Discussion: The large range in Stage 1 8 3 4S data is unusual. The heaviest 8 3 4S values (22 to 23%o) are consistent with derivation from sulfate that equilibrated with H2S during disproportionation of SO2 (Rye et al., 1992; Rye, 1993), but other data are not. We propose two alternative explanations for this variability. 1) Data may represent magmatic-hydrothermal alunite formed from aqueous sulfate sulfur species that did not always reach isotopic equilibrium with H 2 S during disproportionation of SO2. This could result from shorter residence times for aqueous sulfate during early vapor- plume development, condensation, and alunite precipitation in Stage 1. 2) Alternatively, data can be interpreted as magmatic-hydrothermal alunite that is partially to completely replaced by either magmatic steam or steam-heated alunite. 90 o 00 "oo 25 20 15 10 Enarg A -5 • Stage 1 S 1 8 O s o 4 A Stage 1 6 1 8 0 Q H • Stage25 1 8 O S C M O Stage28 1 8 0 O H • B x S 6 1 8 0 S 0 4 • BxS 8 1 8 0 O H A 10 15 S 3 4S l A 6 I 9 0 20 25 30 Figure 3.6a. 5 3 4S and 8 1 80data (in %>) for alunite from Stage 1, Stage 2, and Brecha Sylvestre (BxS). 5 1 8 0 S 0 4, and 8 I 8 0 0 H values for each sample are joined by lines. Range of 8 3 4S data for associated sulfides is also given. -15 -5 5 15 25 8 1 8 0 (per mil) Figure 3.6b. 8D, 8 1 8 0 So4, and 8 1 8 0 0 H values for alunite from Stage 1, Stage 2, and Brecha Sylvestre (BxS). Range of calculated 8DH2o and 8 i 8 OH2O for fluids in equilibrium with each stage are shown as shaded fields (grey = Stage 1; 200-280°C: white = Stage 2; 180-250°C). Fluid compositions calculated from equations of Stoffregen et al (1994). Temperatures used for calculations are derived from textural, A 3 4 S a : u n . p y , and A 1 8 0 S O4-OH isotope data. Lines and fields are M W L = meteoric water line of Craig (1961); F M W = composition of waters dissolved in felsic magmas (Taylor, 1988); kaolinite line of Savin and Epstein (1970); V . V (volcanic vapor) = range of water compo- sitions discharged from high temperature fumaroles (Giggenbach, 1992); SASF (light grey) = supergene alunite S 0 4 field; and SAOZ = supergene alunite OH zone as described in Rye et al. (1992). Also shown (white circle) is the estimated composition of paleo-meteoric waters (B. Taylor, pers. comm.). 91 Based on textural and petrographic observations for Stage 1 alunite, the second model is preferred. Alunite with the lowest 8 3 4S values has macroscopic characteristics of Stage 1 alteration but the alunite is not associated with clays or APS inclusions typical of the Stage 1 alunite. These grains appear to have precipitated in open vugs created by the leaching of feldspar phenocrysts. Partial replacement of APS-bearing alunite by a more homogeneous, K- rich variety is also observed in one sample. Similar processes are inferred at E l Salvador (J. Hedenquist, pers. comm.), where partial replacement of hypogene by supergene alunite resulted in intermediate 8 3 4S values. Magmatic fluid signatures for most samples of Stage 1 alunite are also consistent with a magmatic-hydrothermal origin. High 83 4S values for Stage 2 alunite (and barite) and fluid compositions in the magmatic range are consistent with a magmatic-hydrothermal origin for this stage of alteration. Homogenization temperatures determined by Jannas (1995) for fluid inclusions in Stage 2 barite range from 140° to 280°C. Calculated A 1 8 O S O 4 - O H temperature values (e.g., Rye et al., 1992; Stoffregen et al., 1994) for Stage 2 alunite are nearly identical (140°to 240°C) and suggest that equilibrium between oxygen species was maintained. Stage 3 and Banded Alunite Veins; Transitional and Magmatic Steam alunite Figure 3.7 summarizes the 8 3 4S, 8D, and 8 1 80 data for Stage 3 and banded vein alunite. 83 4S values for both events are slightly greater than the values of Stage 3 enargite (about -2 to - 4%o). Stage 3 alunite exhibit almost no 83 4S variation (± 0.2%o) between coarse breccia-fill alunite and fine-grained alunite overgrowths (Appendix B). Only minor 83 4S zoning is noted within large (1 to 2 cm) Stage 3 crystals with values ranging between 1.0 to 2.0%o. Banded vein alunite has consistently lower 8 3 4S values (about 0 to 1.5%o) than Stage 3 samples. Only minor 8 3 4S variation is reported between different banded vein samples or between growth bands in a single vein (Appendix B). The narrow range of values for both banded vein and Stage 3 alunite indicates that alunite sulfur isotope equilibrium was not attained between aqueous sulfate and H2S. Calculated fluid compositions for alunite are shown in Figure 3.7b. Stage 3 fluids have 8DH2o values between about -30 and -95%o, and 8 , 8 0 H 2o between -1 and 13%o. Banded vein alunite fluids occur in a more restricted range, with 8DH2o values between -35 and -65%o. Data for both events overlap with magmatic waters (Taylor, 1992), but Stage 3 and banded vein fluids are typically depleted in deuterium relative to Stage 2 magmatic-hydrothermal fluids. A 92 25 O "~"c 20 15 10 Hnargi t LA c 4 4 -OTf A A f I I I A • S t a g e 3 S ' 8 O s o < A Stage 3 8 , 8 0 O H • B V 5 , 8 0 S O 4 O B V 8 , 8 0 O H 10 534S Figure 3.7a. 8 3 4S and 8 I 8 0 data (in %o) for Stage 3 and banded vein (BV) alunite. 5 1 8 0 S 0 4 , and 8 1 8 0 O H values for each sample are joined by lines. Range of S 3 4S data for Tambo sulfides is also given. Figure 3.7b. 8D, 8 1 8 0 S 0 4 , and 8 1 8 0 O H values for Stage 3 and banded vein ( B V ) alunite. Range of calculated 5DH2O and 8 1 8 0 H 2o for fluids in equilibrium with each stage are shown as shaded fields (grey = Stage 3; 200-300°C: white = B V ; 200-300°C). Fluid compositions calculated from equations of Stoffregen et al (1994). Temperatures used for calculations are derived from textural, A ^ S ^ p y , and A 1 8 O s 0 4 - O H isotope data. Lines and fields are given in Figure 3.6. 93 correlation between decreasing 8DH20 and 8 1 8OH2O values for both alunite types suggests that this depletion can be attributed mainly to mixing between magmatic and meteoric fluids. Deuterium depletion of late-stage magmatic fluids (Taylor, 1988) may also contribute to this trend. Discussion: Data for banded vein alunite indicate a major magmatic component in the source fluids. Results are consistent with a magmatic steam origin. Stage 3 data are more variable and show greater degrees of mixing between magmatic and meteoric fluids. This variation suggests episodic pulses of vapor released from the magma rather than continued, open-system degassing. The larger range of 8 3 4S values for alunite of this stage suggests limited exchange between aqueous sulfur species. Deposition is thought to have occurred in an environment that is transitional between magmatic-hydrothermal and magmatic steam. Depositional temperatures calculated from A 1 8 0 S O 4 - O H for banded vein alunite average about 110°C, with the exception of one higher value at 320°C (sample T33a). Independent temperature estimates are not available for this stage of alteration, but temperatures are expected to be at least as high as those encountered in the magmatic-hydrothermal environment (i.e., 180 to 300°C). Retrograde isotopic exchange is not expected given the coarse grain size of the alunite; hence, the origin of the low 8 1 8 O O H values is unclear. Similar behaviour is reported for magmatic steam alunite at Alunite Ridge, Utah (Rye et al., 1992). The range in A 1 8 OSO4-OH temperatures for Stage 3 alunite is greater than for the banded veins (10° to 330°C), but most values are below 200°C. Again, several temperatures are unreasonably low but others are within the expected range for shallow environments of deposition. Huangite-bearing vein alunite 8 3 4 S, 8D, and 8 1 8 0 data for two samples of huangite-bearing vein alunite are plotted in Figure 3.8. Results for the two samples are significantly different. One sample has low 8 3 4S (1.8%o) and 81 80So4 (11.7%o) values. The other sample has a 8 3 4S value of about 9%o, and has 8 1 8 0 data in the range of magmatic-hydrothermal alunite. Calculated 8DH2o compositions in equilibrium with the two samples are -115 and -52%o, respectively. The first sample (T35) has a fluid composition which overlaps that of Tambo magmatic-hydrothermal and magmatic steam occurrences. Data for the latter (sample T14a) indicate a dominant meteoric fluid component. Discussion: Depositional temperatures for the two huangite-bearing samples are 94 o 1o 25 20 15 10 5 4 O H Enarg <!> • I I 6 o o 0 • i i 0 o • S H 8 1 8 O s 0 4 o S H 8 1 8 0 O H • L V 5 , 8 0 S 0 4 o L V 5 1 8 0 O H • Ca-vein 5 1 8 0 S O 4 • Ca-vein 5 1 8 0 O H 10 15 S , 4 S 20 Figure 3.8a. 5 3 4 S and 8 1 8 0 data (in %o) for steam heated (SH), late vein ( L V ) and huangite-bearing vein (Ca-vein) alunite. 8 l 80So4, and 8 1 8 OOH values for each sample are joined by lines. Range of 8 3 4 S data for Tambo sulfides is given for reference. -15 -5 5 15 25 8 i 6 0 (per mil) Figure 3.8b. 8D, 8 l 8 0 S o4 , and 8 ' 8 0 0 H values for steam heated (SH), late vein (LV) and huangite-bearing vein (Ca- vein) alunite. Range of calculated SDH2o and 8' 8 0 H 2o for fluids in equilibrium with each stage are shown as symbols (SH = steam heated fluids; L V = late vein fluids). Fluid compositions calculated from equations of Stoffregen et al (1994) . Temperatures used for calculations are derived from textural, A 3 4 S a | u n . p y , and A I 8 O s 0 4 - O H isotope data. Lines and fields are given in Figure 3.6. 95 similar and average 105°C. No independent temperature estimates are available for these veins however and it is unclear whether these values are reasonable. Temperatures of formation for the mineralogically similar E l Indio Campana vein were estimated to be 225-275°C (Jannas, 1995), but lower temperature occurrences of Ca-rich alunite have been reported elsewhere (e.g., Stoffregen and Alpers, 1987; Arribas et al., 1995). The origin of these veins and their relation to other alteration or mineralizing events is uncertain. Near-surface alteration Figure 3.8 summarizes the 8 3 4S, 8D, and 8 1 8 0 data for alunite sampled from the near- surface alteration blanket on Cerro Elefante. Two samples (T13, T17) have 8 3 4S values slightly enriched relative to bulk sulfur in the deposit, and have high 8 1 8 0 S 0 4 values. The third sample (T18), selected from white, powdery veins that cross-cut Stage 1 alteration just below the blanket zone, has a much larger 8 3 4S value (16.6%o) and lower 8 1 8 0 S 0 4 value (15.4%o) compared to the first two samples. Calculated fluid compositions in equilibrium with alunite are plotted in Figure 3.8b. 8 D H 2 o data for the first two samples vary between about -10 and -55%o, with 8 1 8 OH2O data between 2 and 7%o. These values are significantly enriched relative to those for meteoric water, and indicate a dominant magmatic fluid component. Deuterium enrichment by evaporation (Henley and Stewart, 1983) is inferred for the highest 8D value. Discussion: Sulfur and oxygen isotopic data for this stage of alteration are consistent with a steam-heated origin. Limited S-isotope exchange between aqueous S 0 4 and H 2S is inferred for the first two samples. Calculated A 1 8 0 S O 4 - O H temperatures of deposition are 110° and 210 °C. Results are reasonable, although perhaps slightly high in the latter, for environments in which aqueous sulfate forms at shallow levels from the oxidation of H 2S by atmospheric oxygen (Rye et al., 1992; Ebert and Rye, 1997). Oxygen isotopic temperatures also suggest that oxygen-isotope equilibrium was obtained between the aqueous sulfate and water in the fluid. The third sample, given its shallow depth, has an unreasonably high A 1 8 0 S o 4 - O H temperature of 290 °C, although 8 1 8 0 0 H data are in the range of the previous two samples. Significant sulfur isotope exchange occurred between aqueous S 0 4 and H 2 S, but A 1 8 0 S O 4 - O H may have been affected by post-depositional exchange." 96 Late vein alunite Sulfur data for late vein alunite are slightly greater than values for sulfides in the deposit (Fig. 3.8a). A l l 5 1 8 0 0 H results plot within the Supergene Alunite OH Zone (SAOZ) defined by Rye et al. (1992); however, all but one alunite 8 1 8 0 S o 4 data plot outside of the Supergene Alunite Sulfate Field (SASF; Rye et al., 1992). These results, combined with large positive A 1 8 0 S O 4 - O H values, are not consistent with a supergene origin. Similarly, calculated A 1 8 0 S O 4 - O H temperatures average about 100° C and are higher than expected for deposition in the supergene environment. Calculated fluid compositions in equilibrium with alunite samples are shown in Figure 3.8b. Two samples with the lowest 8D H 2o and 8 1 8 0 H 2 o values have fluids very close to the predicted composition of meteoric waters. The other two samples have much larger 8D H 2o and 8 1 8 0 H 2 o values, indicating a substantial magmatic contribution. These fluid compositions overlap those for magmatic-hydrothermal and magmatic steam occurrences. Discussion: Results indicate that the late alunite veins were not generated under supergene conditions. Alunite formed from mixed magmatic-meteoric fluids at moderate temperatures {ca. 100°C). The absence of significant sulfides in the orebody and surrounding host rocks requires that the majority of sulfur must have been derived from condensed magmatic vapors (either H 2S or S0 2). Data are consistent with aqueous sulfate that formed from the oxidation of degassed H 2 S as the hydrothermal system cooled. Alternatively, pulses of S0 2-rich magmatic steam rapidly condensed into mixed magmatic-meteoric fluids in the final stages of the system. G E O C H E M I S T R Y O F A L U N I T E Major and trace-element compositions were determined to evaluate the variability of alunite and alunite-group minerals in each alteration assemblage. End-member alunite is represented by the formula KA1 3(S0 4) 2(0H) 6 but naturally occurring minerals of the alunite supergroup can have a wide range of chemical substitutions (Jambor, 1999). The most common substitution of Na for K defines the alunite-natroalunite solid solution. Other substitutions for K + may include Ca, Ba, Sr, REE (Ce, La), Pb, Ag, H 3 0 , and NH4. Substitution of Fe 3 + for A l 3 + defines the alunite-jarosite solid solution, and substitution of (P0 4) 2" for (S0 4) 2" forms APS minerals of the crandallite and beudantite groups (Table 3.3: as summarized in Jambor, 1999). 97 Several experimental (Hemley et al., 1969; Stoffregen and Cygan, 1990) and empirical studies (e.g., Aoki, 1991; Aoki et al, 1993; Thompson, 1992; Hedenquist et al., 1994; Arribas et al., 1995) have correlated variations in alunite-group chemistry to specific environments or temperatures of deposition. Similarly, published data on REE contents of acid sulfate-altered rocks suggest that REEs may be influenced by a number of factors including; (a) crystallographic controls (Schwab et al., 1990); (b) the availability and type of complexing agents (Wood, 1990a; Wood, 1990b; Lottermoser, 1992); and (c) environmental controls such as pH, Eh, temperature (e.g., Michard, 1989). It is also likely that vapor-melt partition coefficients (Flynn and Burnham, 1978; Candela and Piccoli, 1995; Reed, 1995) affect the REE content of magmatic vapors, and subsequent alteration, in the epithermal environment. Summary of Results Chemical data for all stages of Tambo alunite are given in Table 3.6. E P M A data (Table 3.6a) are summarized for each paragenetic stage. The large beam size required for analysis resulted in poor representation of finer-grained samples (supergene and steam-heated alteration in particular) and thin oscillatory zones and growth bands. Selected ICP-MS trace metal concentrations are listed in Table 3.6b. REE results presented in this study (Fig. 3.9) are normalized to average REE contents of unaltered host rock (Tilito Formation: Table 3.2). Data for each paragenetic stage are summarized in the following discussion. Magmatic-hydrothermal alunite, represented by Brecha Sylvestre, Stage 1, and Stage 2 alteration, exhibits a wide range of compositions. Brecha Sylvestre alunite samples are compositionally simple, with local enrichment in Pb and Sb, minor Na, and trace P, Sr, and Ba. Alunite is depleted in REE relative to host-rock values, and moderate HREE fractionation is apparent. In contrast, Stage 1 alunite is chemically heterogeneous, with irregular trace-element (Na, Ca, Ba, P, Sr) substitution. Average Na contents are the highest among the analyzed alunite types (statistically significant at 95% confidence limit, based on t-test results: Appendix A). Calcium and Ba substitutions locally approach huangite and walthierite end-member compositions. REE results show Stage 1 alunite samples are depleted in mid-REE (MREE) relative to the host Tilito Formation, and exhibit minor light-REE (LREE) fractionation. Stage 2 alunite has a unique geochemical signature characterized by high Ba and the presence of walthierite. Compositions intermediate between walthierite and alunite are common and a nearly complete solid-solution exists between the two end-members (Fig. 3.10). Stage 2 alunite in the Kimberly and Canto Sur deposits are commonly oscillatory zoned (Fig. 98 <u I > "5. E o 10 O O CO o q q d CN 9 9 d d CN CO I in CN q I o d CD cn o> I W 01 o I o d E i- o o o co o N- cn O CM O CO X CO Q. E q d q d o o CN Si q d Q. E CQ W ^ _ O) >o •<J- O CN - cn o m T - o o o 8 o LO O CO CD q q T- O CN CN CO I CM IT) LO O LO LO cn 0) q co co cn co o co d 5 "5. E <0 10 CN CO U3 0 CM CO CO d O O CM O CO I CO CO d d cu 01 i- o o o 0) LO O CO LO O CO O O CN | 6 r N 6 CO o d CM CO Si CD W 0- W i l E LU CO (0 W rt z o m w °- w 99 Table 3.6b. ICP-MS trace element and REE concentrations for bulk (~1 gram) alunite samples. Trace contaminants may be present. Data below detection limits (0.1 ppm for REE) not shown. Estimates of precision, based on duplicate analyses, given in Appendix B. S t a g e 1* S t a g e 2 S t a g e 3 Sample #: T09 T10 T01 T27 T03b T04 T22 T28 P (ppm) 970 380 1570 120 1180 1210 1700 1270 Sr 668 208 511 461 411 696 728 1580 Ba 340 630 4080 3850 3600 600 280 1470 Pb 116 856 1515 1745 251 608 428 338 Sb - 11 317 10 10 29.5 31 15 Ag 2.7 0.9 79.4 4.0 5.8 21.4 0.2 3.0 Cu 1218 10 32 20 17 53 18 69 Bi 22 1.5 6.5 0.8 13.4 102.0 2.3 32.0 Zn 4 2 - 2 6 48 - 46 La 45.3 25.5 10.5 13.3 66.5 90.3 Ce 86.5 45 9.5 9.2 79.5 81.3 Pr 9.2 4.3 0.6 0.5 4.1 5.1 Nd 24.8 11.5 1.5 0.8 6 6.8 Sm 3.3 1.6 0.2 - 0.2 0.1 Eu 0.3 0.3 0.2 0.2 0.2 0.1 Gd 1.8 1.3 0.1 - 0.3 0.1 Tb 0.2 0.3 - - - - Dy 1 1.1 - - 0.2 - Ho 0.2 0.3 - - - - Er 1 1.1 - - - - Tm 0.1 0.3 - - - - Yb 1.4 1.6 - - - - Lu 0.2 0.3 - - - - * May contain up to 10% quartz. 100 Table 3.6b. (Continued) Banded Vein Ca-vein Steam-Heated Late Stage Bx Sylvestre Sample #: T33 T32 T14 T17 T11 T02A B03 P Sr Ba Pb Sb Ag Cu Bi Zn 2300 1535 1120 580 28.5 18 0.3 1410 582 450 628 22 0.6 19 46.2 2 630 246 830 59 0.5 18 0.6 8 940 353 1030 916 10 1.4 44 1.0 18 1490 2760 2140 486 14.5 3.0 44 32.3 70 1580 677 2430 2240 14.3 7.9 1 6.0 680 511 650 1700 78.5 0.8 25 29.2 2 La Ce Pr Nd Sm Eu Gd Tb D y Ho Er Tm Yb Lu 156 198 13.4 24 1.1 0.3 1 0.1 0.3 0.1 0.1 81.5 116.5 7.9 13 0.6 0.1 0.5 0.1 48 109.5 15.5 65.5 16.8 2.5 9.3 0.6 0.7 0.1 47.5 85 8.7 27.5 3.7 0.5 2.3 0.2 0.7 0.1 0.5 0.8 0.1 29 50 5.9 22 5 0.9 3.2 0.4 1.1 0.1 0.3 0.3 77 47 2.1 3 0.2 0.4 0.1 33 3 8 0.6 0.1 0.5 0.1 0.1 0.1 0.1 101 Figure 3.9. REE data for all stages of Tambo alunite normalized to average REE composition of the Tilito Fm. Individual sample numbers are given in the legend accompanying each plot. Data points be- low detection are not plotted. (A) Bx Sylvestre, Stage 1 and Stage 2 alunite. (B) Stage 3 and Banded vein alunite. (C) Huangite-bearing vein alunite (Ca-vein), steam heated, and late vein samples. 102 3.4c), typically with a core of Na-bearing alunite overgrown by walthierite and rimmed by nearly stoichiometric alunite. Walthierite has not been identified in the Wendy deposit, and Stage 2 alunite in this area occurs as coarse bladed crystals that are zoned in Ca, Na, P, and minor Ba. Stage 2 alunite is typically Na-poor (less than 0.2 a.p.f.u.), although the mineral locally approaches natroalunite end-member compositions. Barium, Pb, Sb, and Ag contents on average are the highest in Stage 2 compared to the contents in other alunite types. Stage 2 alunite exhibits strong fractionation in LREE, no detectable heavy-REE (HREE), and a strong, positive Eu anomaly. Stage 3 and banded vein alunite are chemically similar. Both are nearly stoichiometric, with minor Na (< 0.4 a.p.f.u) and trace-metal substitution. Oscillatory zoning with P 0 4 ± Sr substitution is common, and banded vein alunite in particular has the highest P contents compared to other stages of alunite. In contrast, Ca, Ba, and Pb contents are low. E P M A analyses indicate almost no chemical variation across successive growth bands in the alunite veins. Stage 3 and banded vein alunite are both enriched in L R E E , with values two to five times higher than those of the host rock, but are significantly depleted in HREE. These trends are more pronounced in the banded vein samples with (La/Yb) n ratios >100. As their name suggests, huangite-bearing veins are characterized by the presence of huangite. It typically occurs as thick overgrowths on alunite grains that have woodhouseite- svanbergite cores. E P M A data indicate a full range of substitution between alunite and huangite end-member compositions (Fig. 3.11). Oscillatory zoning between huangite and minamiite compositions occurs locally. ICP-MS results confirm very high Ca contents with elevated Na concentrations and few trace elements. Alunite of this stage is significantly enriched in L R E E and M R E E relative to all other alunite types. HREE are strongly fractionated. Chemical data for steam-heated alunite samples are limited, due to the small grain size and relatively rare occurrence of alunite in the steam-heated alteration zone. This alunite is nearly stoichiometric, with minor Na, Sr, and P substitution. Fractionation of HREE is moderate, with minor enrichment in LREE compared to the host Tilito Formation. Geochemical data for the late veins are limited because of the fine-grained nature of the alunite. Results indicate that the mineral is chemically variable, with high Sr contents and locally elevated P 0 4 , Ba, Pb, Cu, and Zn. Sodium contents are low, and REE contents are extremely variable. One sample (T i l ) shows no L R E E fractionation relative to host-rock values but has a significant depletion in HREE. Another sample (T02a) shows similar behavior 103 N a Figure 3.10. Normalized K-Na-Ba compositions of Stage 2 alunite. Based on recalculated EPMA data (58 analyses total from 4 samples). 0.8 0.2 0.0 < >•».• ; ; • • i • • • : • • : • • • > • 1 1 • 0.0 0.1 0.2 0.3 0.4 0.5 C a a.p.f.u. Figure 3.11. Measured K+Na versus Ca content of huangite-bearing vein alunite. Axes are given as atoms per formula unit (a.p.f.u.). Data taken from EPMA analyses (n=47) of multiple grains in two samples, KB-09 and LN-02. 104 to Stage 2 alunite, with extreme HREE fractionation and a large, positive Eu anomaly. E P M A analyses of extremely fine-grained, pseudo-cubic alunite from irregular veinlets indicate nearly stoichiometric compositions, minor Na substitution (<0.2 a.p.f.u.), and trace Fe. Slightly larger (10-15 jam) pseudo-cubic grains of Ca- and Sr-bearing APS minerals (likely woodhouseite- svanbergite) occur intergrown with alunite and kaolinite in these veins. Interpretation Chemical data for Tambo magmatic-hydrothermal alunite are consistent with other studies that report heterogeneous alunite compositions and elevated concentrations of Na, Ba, Ca, Sr, P 0 4 , and REE in high temperature environments (Stroffregen and Alpers, 1987; Stoffregen and Cygan, 1990; Aoki, 1991; Aoki et al., 1993; Hedenquist et al., 1994; A.J.B. Thompson, unpub. data). These elements typically occur as APS minerals overgrown by alunite or alunite-natroalunite (e.g., Stoffregen and Alpers, 1987; Aoki et al., 1993), as observed in Stage 1 samples from this study. The abundance of walthierite in Stage 2 is unusual, although similar compositions are noted for ore-stage alteration at the Pascua deposit, located at the northern end of the El Indio-Pascua Belt (Chap. 2). Little geochemical information is available for near-surface alunite occurrences. Limited data for steam-heated alteration show significant variations (Ebert and Rye, 1997; A.J. B. Thompson, unpub. data), with compositions extending to the alunite and natroalunite end- member. The former is consistent with results reported in this study. Supergene alunite is generally K-rich in comparison to that in higher temperature occurrences, although trace- element concentrations are extremely variable, with Ca, Sr, R E E and P 0 4 substitutions reported (Stoffregen and Alpers, 1987; Arribas et al., 1995). The elevated trace-metal concentrations for samples of late vein alunite examined in this study are attributed to the metal content of source fluids, since stable-isotope results are not consistent with the oxidation of precursor sulfides. Few chemical analyses have been published for magmatic steam alteration. Alunite in the Tambo banded veins is chemically similar to that from Alunite Ridge, Utah, and both are characterized by fine-scale, sawtooth bands of P 0 4 and Sr substitution (Cunningham et al., 1984). Barium substitution is more common at Alunite Ridge however, and is rarely detected in the Tambo banded veins. Recent studies on the REE contents of hydrothermal fluids (Michard, 1989; Lewis et al., 1997, 1998) and of hydrothermally altered rocks (Hopf, 1993; Arribas et al., 1995; Fulignati et al., 1999), have shown that REE can be mobilized under the conditions of acid sulfate 105 alteration. Magmatic-hydrothermal alteration has been characterized by strong fractionation of HREE with respect to LREE, when compared to fresh-rock equivalents (Arribas et al., 1995; Fulignati et al., 1999). However, compositional data for REE in Stage 1 alunite at Tambo and the host-rocks are not consistent with these published trends. We propose that the presence of APS inclusions and growth zones with elevated P 0 4 and Sr contents contribute to the 'capturing effect' described by Schwab et al. (1990) and allow for the incorporation of HREE into the alunite crystal structure. Similar behavior is noted for Tambo steam-heated alunite which shows relatively minor HREE fractionation compared to host-rock values. Results are analogous to those described by Lewis et al. (1998) for altered rocks at Yellowstone National Park. Alunite sampled from Stages 2 and 3 and the banded veins all exhibit strong fractionation of HREE compared to the host-rock composition. This trend is most pronounced in the Stage 3 and banded vein samples, whereas Stage 2 samples have comparatively lower REE concentrations. In all three stages, alunite precipitated in open spaces (either in veins or breccia matrices) with little evidence for wallrock interaction. The chemistry of these alunites should therefore reflect their source fluid composition. Experimental studies of REE partitioning between a Cl-bearing supercritical fluid (i.e., 'vapors') and a residual melt have shown a relative enrichment in L R E E compared to HREE in the vapor phase (Flynn and Burnham, 1978; Reed, 1995). No information is available for these phases at lower temperatures, but the chemical signature of the Stage 2, Stage 3, and banded vein alunite may reflect the composition of a HREE-depleted fluid derived from the supercritical vapor-phase. In contrast, other forms of alunite derived their R E E contents from a combination of magmatic fluids and REEs leached from the host rocks. C H E M I S T R Y O F F L U I D - I N C L U S I O N G A S E S Bulk sample fluid-inclusion techniques are widely used for the analysis of inclusion volatiles, mainly H 2 0 , C 0 2 , C H 4 , N 2 , H 2 , and the noble gases (Shepperd and Rankin, 1998). Quadrupole mass spectrometric (QMS) analysis in particular has been applied to the identification of a large number of gases at low concentration in a variety of ore deposits and geothermal systems (e.g., Guha et al., 1990; Landis and Hofstra, 1991; Graney and Kesler, 1995b; Lindaas et al., 1998). A thorough review of this technique and its limitations are given in Graney and Kesler (1995a). 106 Compositions of fluid-inclusion gas in Tambo alunite were determined by QMS analysis (Landis and Rye, in prep) in order to characterize Stage 2 and 3 ore fluids and banded vein alunite (i.e., magmatic steam) fluids. Repeated attempts to analyze fine-grained Stage 1, steam- heated, and late vein alunite failed to release sufficient quantities of gas. We present herein only a summary of results and the reader is referred to Landis and Rye (in prep.) for details. Alunite samples were selected from coarse-grained assemblages and were examined petrographically prior to analysis. Stage 2 alunite, selected from the Reina vein, contains mostly small (<10 jam), apparently primary, vapor-rich inclusions, but the results of necking are common. Other, slightly larger inclusions (10-15 jam) have irregular shapes and may be secondary or pseudo-secondary in origin. These inclusions are vapor-dominant, but with more variable liquid-to-vapor ratios. Stage 3 alunite was selected from the Canto Sur deposit and contains many small (5-10 jam), vapor-dominant inclusions. Banded vein alunite, sampled from the Kimberly deposit, typically contains small, irregular, vapbr-rich inclusions tightly clustered along growth zones. Results and Interpretation Data for crush- and thermally-released gas from alunite samples are reported in Table 3.7. The quantity of gas released by crushing is significantly smaller than that by thermal release, and is likely the result of opening very few and mostly larger inclusions along secondary planes in the alunite. The genetic significance of these mechanically released gases is not known, but the gases probably represent mixtures of multiple generations and secondary trapping events. Thermally-released gases are derived from decrepitation of the much more uniform occurrence of small primary inclusions (after removal of anomalous and secondary inclusions by crushing). These gases are generally considered to be representative of magmatic volatiles at the time of deposition (Graney and Kesler, 1995a). However, due to analytical effects, measured H 2 S concentrations are considered a minimum value (Graney and Kesler, 1995b). The presence of sulfide daughter minerals would indicate significant post-depositional changes in sulfur gas species, but none were detected at Tambo. Similarly, significant reaction of S 0 2 gas with sulfate bound in the structure of alunite is unlikely (Landis and Rye, in prep.). Ratios of major gas species, H 2S-S0 2-HC1 (Fig. 3.12a), for the three alunite depositional events illustrate a distinct transition in volatile chemistry. Stage 2 fluids are H2S-dominant (H 2 S/S0 2 ~ 6), whereas magmatic steam alunite (banded vein sample) is characterized by a 107 Table 3.7. Fluid inclusion gas chemistry data for Stage 2, Stage 3, and banded vein alunite. Abbreviations: MH = magmatic hydrothermal, MS = magmatic steam, a = atmosphere, cps = counts per second. Blank fields are not detected. Accuracy is 3-5% based on prepared standard gas mixtures and analysis of synthetic fluid inclusions. Sample T01a T02b T08a Description Stage 2 Stage 3 Banded Vein Method Crush Thermo Crush Thermo Crush Thermo Weight (g) 8 8 6.6 6.6 7.4 7.4 IV/o 13.6131 0.1961 19.5394 0.5174 48.6233 20.2117 0 2 ppm 48 44 19 59 127 1045 Ar ppm 4070 200 4431 654 14228 54992 H 2 % 65.09 1.44 65.65 6.68 13.59 45.38 HP0/* 11.3008 97.551 4.0415 91.1306 21.7247 29.9172 H 2S ppm 20 17 12 12 22 1 S 0 2 ppm 38 3 24 30 43 48 HCI ppm 47 8 9 41 10 3 HF ppm 313 2672 90 2234 0 673 C H 4 % 0.212 0.014 0.43 0.083 0.143 0.5 C 0 2 % 1.637 0.458 2.173 0.764 1.141 3.264 C O % 7.69 0.045 7.712 0.52 20.128 — He ppm 44 4He volts 0.085718 1.479663 0.029593 0.945581 0.425582 0.107442 4He moles-gm"1 6.872E-14 1.186E-12 2.876E-14 9.189E-13 3.689E-13 9.313E-14 3Hecps 2.771531 8.834217 1.057146 9.002538 17.721322 2.350376 3He moles-gm"1 3.447E-20 1.099E-19 1.594E-20 1.357E-19 2.383E-19 3.161E-20 R=3He/4He 5.016E-07 9.263E-08 5.542E-07 1.477E-07 6.460E-07 3.394E-07 R/R a 0.3625 0.0669 0.4004 0.1067 0.4668 0.2452 ^Ne . 91.772 90.009 91.35 91.564 85.109 90.098 2 1 Ne 0.29924 0.28346 0.2571 0.25246 4.23851 0.39388 ^Ne 7.92907 9.70787 8.39272 8.1838 10.65245 9.50805 ^Ne/^Ne 0.0033 0.0031 0.0028 0.0028 0.0498 0.0044 ^Ne/^Ne 0.0377 0.0292 0.0306 0.0308 0.3979 0.0414 '^Ne/^Ne 0.0864 0.1079 0.0919 0.0894 0.1252 0.1055 108 disequilibrium gas assemblage with high H 2 , abundant reduced carbon species, and high S0 2 . This chemical shift is also illustrated in Figure 3.12b, with compositions recast to modeled equilibrium mole fractions at approximated temperatures and pressures for each alunite stage (using the N A S A - C E A program and thermochemical data set; Gordon and McBride, 1994; McBride and Gordon, 1996: details see Landis and Rye, in prep). Helium is commonly used as a tracer in studies of fluid flow in hydrothermal systems, because isotopic compositions of He from atmospheric, upper-mantle, and crustal or radiogenic sources differ substantially. The atmospheric 3He/ 4He ratio (Ra) is 1.39 + 0.01 (e.g., Mamyrin et al., 1969) and is typically used as a reference for comparison to 3He/ 4He (R) gas data in the form R/Ra. Mantle-derived He has R/R a values greater than 8 (Craig and Lupton, 1981; Lupton, 1983) whereas crustal 4He is enriched from both radiogenic U-Th decay and nuclear spallation reactions, creating low R/R a values of < 0.01 in upper crustal elevated U-Th rocks (Gerling et al., 1971; Morrison and Pine, 1955). Thermal release 3He/ 4He and R/R a in this study range from 9.3e-8 to 3.4e-7 and 0.06 to 0.25, respectively. There is no indication of elevated 3He of mantle origin in any of the samples in this study, and data suggest a crustal magmatic component. In particular, banded vein samples have the highest R/Ra values. Stable isotopic analyses indicate minor meteoric fluid was involved during any of these stages; hence, the volatiles and magmatic steam in particular must have been derived directly from magma, with He incorporated in the melt during partial melting in the lower crust. Similarly, the isotopic and elemental composition of Ne of the earth is significantly different from that of the present atmosphere, and can provide further indications of fluid source and the amount of crustal versus magmatic interaction. Neon isotopes are typically compared against end-member compositions of the atmosphere and of mid-ocean-ridge basalts (MORB), O A O1 O O which show enrichment in Ne and Ne relative to Ne (e.g., Poreda et al, 1984; Marty, 1989). In contrast, crustal fluids typically are enriched in 2 1 Ne and 2 2 Ne relative to 2 0 Ne from crustal nucleogenic-radiogenic processes (Kennedy et al., 1990). Neon isotopic data from this study are consistent with He results and suggest a crustal enrichment in 2 1 Ne and 2 2 Ne relative to 2 0 Ne (Table 3.7). 109 1 0.8 0.6 0.4 0.2 0 H2S Figure 3.12a. Ternary H2S-SO2 -HCI data from fluid inclusion analyses of Tambo alunite. Hydrogen Chloride Figure 3.12b. Ternary HC1-HF-H2 data from fluid inclusion analyses of Tambo alunite. Methods of mole fraction recalculation detailed in Landis and Rye (in prep.) 110 DISCUSSION Fluid-inclusion and isotopic studies of several high sulfidation deposits have indicated that magmatic fluids contribute to both alteration and mineralization within these systems (e.g., Hedenquist and Lowenstern, 1994; Arribas, 1995; Cooke and Simmons, 2000). This magmatic component originates from the exsolution of a fluid from the melt during retrograde (or resurgent) boiling (Burhnam, 1979; Burnham and Ohmoto, 1980). When the exsolved fluid reaches its solvus, the inception of aqueous immiscibility leads to the formation of low-density vapor and dense hypersaline liquid. Evidence for phase separation is seen in fluid inclusions from many porphyry copper deposits, which contain both hypersaline liquid-rich inclusions and associated vapor-rich inclusions (Burnham, 1979; Roedder, 1984; Bodnar, 1985). This process is critical for the partitioning of ore metals from the melt to the hydrothermal system (Candela and Piccoli, 1995; Williams et al., 1995), and the strong effect on the physical behavior of the system is due to the large density contrast between the vapor and liquid phases. The aqueous vapor will rise to shallow depths and either discharge at surface as volcanic fumaroles, or become absorbed into the groundwater system. The dense, hypersaline liquid remains at depth and is recorded as saline fluid inclusions in intrusive rocks and porphyry ore deposits (Hedenquist and Lowenstern, 1994 and references therein). Ore-metal partitioning between these two phases will determine where (potential) mineralization wil l occur (see discussion below). As fluids ascend from their underlying magma source, they are further influenced by the transition from brittle to plastic behavior in the lithosphere (Fournier, 1999). This transition typically occurs at about 370° to 400°C in continental hydrothermal systems (Fournier, 1999). Abrupt changes in the pressure regime across this transition zone are thought to be responsible for porphyry Cu mineralization in several systems (e.g., Fournier, 1967; Gustafson and Hunt, 1975; Cunningham, 1978), and more recently for porphyry Au mineralization (Muntean and Einuadi, 2000). Similarly, several authors have related epithermal mineralization to the escape of magmatic fluids into hydrostatically pressured hydrothermal systems (e.g. Fournier, 1987; Rye, 1993; Deen et al., 1994; Hedenquist et al., 1998). In the following discussion, we examine the role of magmatic fluids in the Tambo high sulfidation system and the nature of magmatic steam processes. i l l Role of Magmatic Fluids in Alteration It is generally accepted that hypogene acid sulfate alteration results from oxidized and acidic fluids that are generated by the condensation of magmatic volatiles enriched in S0 2 , HC1, and HF. The disproportionation of these acid species at temperatures below ca. 300-350°C (Hemley et al., 1969) creates acidic fluids with a pH of about 1. These fluids are sufficiently acidic to leach most components from the host rock, leaving a vuggy quartz residue. The progressive neutralization of these fluids by reaction with wallrock forms successive alteration envelopes of acid sulfate, advanced argillic, argillic, and propylitic assemblages outward from the main fluid conduit (e.g., Steven and Ratte, 1960). Stable-isotope evidence indicates that the degree of meteoric water interaction during hypogene alteration is variable. At the Julcani deposit for example, the condensed liquids responsible for acid sulfate alteration were of magmatic origin (Deen et al., 1994). Rodalquilar also has evidence for a strong magmatic component during hypogene alteration (Arribas et al., 1995). In contrast, many other high sulfidation deposits have evidence for mixed magmatic and meteoric alteration-fluids (e.g., Summitville, Rye et al., 1990; Nansatsu, Hedenquist et al., 1994; Pueblo Viejo, Venneman et al., 1993). Magmatic contributions to the near-surface environment are less well defined. Steam- heated alteration forms from the oxidation of H 2S gas at or above the water table (Schoen et al., 1974). The fluids responsible for alteration typically have a meteoric isotopic signature (e.g., Rye et al., 1992; Ebert and Rye, 1997), however deuterium enrichment (10 to 20%o) is common in steam-heated fluids from active geothermal systems. This enrichment is attributed to the effects of either subsurface boiling (Truesdell et al., 1977) or evaporation (Henley and Stewart, 1983). At Tambo, the large difference between isotopically light meteoric water and significantly heavier magmatic fluids emphasizes processes involving these two components. Stable-isotope evidence indicates that there is a dominant magmatic component in all stages of hypogene alteration and alunite deposition. Minor meteoric water is involved in Stage 1 and Stage 3 alteration, but overall, magmatic-hydrothermal and magmatic steam processes from initial Stage 1 alteration through to the banded alunite veins involve a significant magmatic fluid contribution. Similarly, fluids responsible for steam-heated alteration are significantly enriched in deuterium (by up to 100%o) compared to local meteoric water. This enrichment is much greater than that attributed to either boiling or evaporation mechanisms to date (maximum of ca. 40%0; J . Hedenquist, pers. comm.). Thus, whereas evaporation may have played a role in 112 steam-heated processes, we conclude that much of the water involved in alteration was of magmatic origin. Late stage alteration at Tambo also involved a large magmatic fluid component. On the basis of stable isotopic data, we suggest that alteration resulted from the mixing of magmatic fluids (either magmatic steam or condensed magmatic vapors) with meteoric waters. This magmatic fluid signature is recorded in alunite deuterium values that show a maximum 50%o enrichment above that of meteoric water. Overall, there is a strong magmatic fluid component to alteration at the Tambo deposit, even in near-surface and late-stage processes. These fluids likely consisted of condensed magmatic vapors that accumulated from episodic magmatic activity over a period of 4 m.y. These fluids effectively displaced, or overwhelmed, meteoric groundwaters in the immediate vicinity. We propose that this process is attributed to local climatic and physiographic effects, related to an semi-arid climate and regional erosional events that are thought to have significantly depressed water tables throughout the El Indio-Pascua Belt (Bissig, 2001; details see Chap. 6). Role of Magmatic Fluids in Mineralization Although the consensus is that high sulfidation ore metals are of magmatic origin (e.g., Hedenquist and Lowenstern, 1994; Arribas, 1995; Cooke and Simmons, 2000), there is considerable debate regarding the nature of these fluids, and particularly the relative contributions of vapor versus hypersaline liquid to the formation of Au, Ag, and Cu mineralization. Based on fluid-inclusion and isotopic studies on several high sulfidation deposits, the nature of mineralizing fluids were summarized by Cooke and Simmons (2000). Three possible ore-fluids were given as: a) Low-salinity, acid waters with high concentrations of reduced sulfur. Under these conditions, Au (and possibly Cu) would be transported as bisulfide complexes. Fluids could result from either the mixing of magmatic volatiles with meteoric waters (Sillitoe, 1983; Arribas, 1995), particularly under high pressure conditions (Hedenquist, 1995), or at the base of a rising vapor plume (Rye, 1993). b) Metal-bearing, acidic brines derived from a magma carrying chloride-complexed metals (White, 1991; Hedenquist et al., 1994). c) Reduced magmatic vapors transporting Au and Cu in the vapor phase as sulfide or chloride species (Arribas, 1995; Mountain, 1999). 113 Given recent studies of metal fractionation between magmatic brines and vapors and evidence from fluid and melt inclusions, there is growing evidence that supports the proposal that magmatic vapors have the capacity to transport metals such as Cu and Au (Lowenstern et al., 1991; Heinrich et al., 1992; Shinohara, 1994; Audetat et al., 1998; Heinrich et al., 1999; Ulrich et al., 1999). In some cases, Au, Ag, and Cu have been shown to partition preferentially into a high-pressure vapor phase (Heinrich et al., 1999). These results are particularly significant to epithermal mineralization, given the much larger mobility and mass flux of a vapor phase relative to brine. At Tambo, Stage 2 ore is characterized by vapor-rich, H2S-dominant, low-salinity fluids (this study; Jannas et al., 1999). There is no evidence for a hypersaline fluid component; nearly all fluid-inclusions measured by Jannas (1995) contain less than 4 wt. % NaCl equivalent, and measured concentrations of CI in inclusion gases are negligible. We conclude that Au was transported as sulfide-complexed metals in a magmatic vapor phase, either as volatile species (e.g., AuS(g)) or as bisulfide complexes, as inferred by Heinrich et al. (1999) for Au transport in their experiments. The S-rich, Fe-poor nature of the Stage 2 assemblage is consistent with vapor separation from a brine (Heinrich et al., 1999). Similarly, the high Te content of the mineralizing assemblage is also consistent with vapor-phase transport (Cooke and McPhail, 2000). The mechanisms of Stage 3 mineralization are less clear, but again there is no evidence for a hypersaline brine component. Stage 3 fluids are vapor-rich, although more oxidized (S0 2 - rich) than those responsible for Stage 2 Au deposition. There are two viable scenarios for mineralization. Gold may have precipitated directly from the magmatic vapor phase without prior condensation, as postulated for magmatic steam alunite (see discussion below). Alternatively, Stage 3 ore fluids may have represented mixed magmatic vapor condensates (analogous to Stage 2 fluids) and S0 2-rich magmatic steam. Gold was either transported in the ascending mixture of vapor + steam as sulfide-complexes or volatiles species, or was present in previously condensed magmatic fluids. In the latter case, Au deposition would result from oxidation caused by periodic pulses of magmatic steam that mix with residual vapor condensates. Origins of Magmatic Steam The magmatic steam environment was initially defined by Rye et al. (1992), based on data for coarsely crystalline alunite veins from Marysvale, Utah (Cunningham et al., 1984). 114 Alunite in these veins is characterized by near-equilibrium oxygen-isotope values but disequilibrium sulfur-isotope values and 8D in the magmatic range. Documented occurrences of this type are rare, and the mechanism of magmatic steam deposition is poorly understood. Magmatic steam processes are a significant part of the Tambo deposit however, due to their abundance and association with Stage 3 ore. Stable-isotope and fluid-inclusion data from this study have helped to characterize the magmatic steam environment. Important considerations are summarized below. • Magmatic steam fluids are S0 2-rich. This was inferred by Cunningham et al. (1984) on the basis of the absence of sulfide in the Marysvale veins. Fluid-inclusion data from this study confirm a transition from H 2S-rich magmatic-hydrothermal fluids to S0 2-rich magmatic steam fluids. • Alunite is characterized by nearly constant 8 3 4S values close to the value of bulk sulfur in the magma (Rye et al., 1992). Given that sulfur-isotope equilibrium between aqueous sulfate and sulfide is obtained quickly at high temperature (Ohmoto and Lasaga, 1982), a rapid rate of vapor ascent and mineral deposition is required once fluids are exsolved from the magma (Rye et a l , 1992; Rye, 1993). • Inclusion fluids are vapor-rich and of low salinity. These data indicate that magmatic steam fluids consisted mostly of vapor that had already undergone phase separation. However, under low pressure conditions, some Cl may be incorporated into the vapor- phase as HC1° (Fournier, 1999). • Magmatic steam alunite forms late in the development of the hydrothermal system. A similar transition from early magmatic-hydrothermal to late magmatic steam processes has also been documented at the Red Mountain deposit, Colorado (Bove et al., 1990; Rye et al., 1992). On the basis of these constraints and the ideas presented in Fournier (1999), we propose a conceptual model for the origin of magmatic steam alunite at Tambo. This model is illustrated schematically in Fig. 3.13, on the basis of physical parameters given in Fig. 3.14. Magmatic steam processes were initiated by the emplacement of a magma body within 1 to 2 km of surface (Fig. 3.14). Given the shallow depth of emplacement, and perhaps coupled with the thermal collapse of the previous magmatic-hydrothermal system, a high thermal gradient would have existed across the brittle-plastic transition zone. Fluids exsolved from the magma 115 Kimberly Bx \ « C Elefante Bx Sylvestre Wendy Bx Steam heated alteration Magmatic-hydrothermal alteration Hypersaline fluids 'Equilibrated' magmatic fluids Adapted from Fournier (1999) Figure 3.13. Schematic section of Cerro Elefante depicting the proposed model for (A) Stage 2 al- teration and Au ore formation and (B) magmatic steam processes. Stable isotope data indicate Stage 2 alunite and barite formed from condensed magmatic vapor with H 2 S / S O 2 ~ 6 and equilibrated sul- fur species with no evidence for meteoric fluid involvement. Magmatic fluids likely evolved from deep crystalline rocks between the brittle-plastic transition and carapace of the magma (Rye, 1993) during episodic rupture of the brittle-plastic transition (Fournier, 1999). Steam heated alteration was active near-surface during the magmatic-hydrothermal event. Magmatic steam alunite formed from S02-dominant magmatic vapors that exsolved from a new pulse of magma emplaced to shallow depth. Decompression and rapid cooling would have resulted due to high temperature gradients and progressive lowering of the water table (Bissig et al., submitted, b). A vapor + brine mixture was released into the hydrostatic regime. The vapor quickly ascended along pre-existing structures and alunite (+ trace quartz, hematite) precipitated along fracture surfaces. 116 Temperature (°C) 0 200 400 600 800 Figure 3.14. Diagram showing phase relations in the system NaG-F^O (modified from Fournier (1999)). Depths assuming lg/cm3 hydrostatic load are given. G = gas, L = liquid, S = solid salt. Curve A shows the three-phase boundary, G+L+S for the system NaCl-f^O. Curve B shows the three phase boundary, G+L+S, for the system NaG-KG-F^O, with Na/K in solution fixed by equilibration with albite and K feldspar at the indicated temperatures. Approximate position of the brittle-plastic transition is given. Reference lines: dotted line = condensation point curves for magmatic vapor containing 10 wt % dissolved NaCl: dashed lines = contours of constant wt % NaCl dissolved in brine. This diagram is used to illustrate the origin of magmatic steam alunite veins from fluids exsolved from a magma emplaced at shallow depth (1-2 km). 117 and released across this boundary would therefore experience a sudden decompression (i.e., 'flashing') and rapid cooling (Fournier, 1999). These fluids were likely a mixture of vapor + brine, but the large volume of 'vapor' generated under these conditions would have ascended very quickly, away from the parent brine. These vapors would have been S02-dominant, due to the shallow depth of vapor release (Gerlach and Casadevall, 1986; Carroll and Webster, 1994) and sudden decompression buffered H 2 S/S0 2 gases in the magmatic fluids (Gerlach, 1993). According to Fournier (1999), the vapors would also have been capable of carrying minor Si and moderate HC1°. Under these low-pressure conditions, however, the solubility of quartz in vapor is too low to permit major silica deposition (Fournier and Potter, 1982), in contrast to the occurrence of banded quartz veins in some porphyry gold deposits (e.g., Muntean and Einaudi, 2000). Flashing of the vapor-rich fluids would result in gas + salt (Fig. 3.14) and apparent supersaturation with respect to alunite (with trace quartz and hematite). The mechanisms of alunite precipitation are unclear, but it is likely that S 0 2 in the vapors would react with H 2 upon cooling (Gerlach, 1993) to form sulfate. Any HC1° in the vapor would remain unreactive in the absence of liquid water (Fournier, 1999). Given the semi-arid conditions and depressed water table in the Tambo area at this time (Bissig, 2001), little excess water would have been present at the depths of alunite deposition (within about 500-800 m of the surface), and any HC1° would have remained in the vapor phase. S U M M A R Y The geological framework determined by Jannas et al. (1999) for the Tambo deposit has allowed a detailed study of shallow-level magmatic-hydrothermal processes. Tambo is characterized by several alteration and mineralizing events, related to episodic magmatic activity over a period of at least 4 m.y. We have focussed on the characteristics of alunite and acid sulfate alteration assemblages. Textural and paragenetic observations, combined with geochronologic, stable-isotope, and geochemical data, have helped to define the evolution of alunite source fluids and depositional processes in the epithermal environment. Results from this study have shown: 1. There is a strong magmatic signature to all stages of alteration and ore-related events in the Tambo area. Meteoric fluid contribution is minimal, even in the near-surface steam-heated environment and during the final stages of the hydrothermal system. 118 2. Vapor-phase Au transport is inferred for both Stage 2 and Stage 3 mineralization. There is no evidence for a magmatic brine contribution to ore-stage fluids. 3. The fluids responsible for both alteration and mineralization evolve from a high-temperature magmatic vapor plume. The transition from magmatic-hydrothermal alteration to magmatic steam-dominated processes reflects changes in pressure conditions and the hydrodynamic regime into which magmatic volatiles were released. R E F E R E N C E S Aoki, M , 1991. Mineralogical features and genesis of alunite solid solution in high temperature magmatic-hydrothermal systems. Journal of the Geological Survey of Japan, 277: 31-32. Aoki, M., Comsti, E.C., Lazo, F.B., and Matsuhisa, Y., 1993. Advanced argillic alteration and geochemistry of alunite in an evolving hydrothermal system at Baguio, northern Luzon, Phillipines. Resource Geology, 43: 155-164. Arribas, A., Jr., 1995. Characteristics of high sulfidation epithermal deposits, and their relation to magmatic fluids. In Thompson, J.F.H. ed., Magmas, Fluids and Ore Deposits. Mineralogical Association of Canada Short Course Notes, 23: 419-454. Arribas, A., Jr., Cunningham, C.G., Rytuba, J.J., Rye, R.O., Kelly, W C , McKee, E.H., Podwysocky, M.H., and Tosdal, R.M., 1995. Geology, geochronology, fluid inclusions, and stable isotope geochemistry of the Rodalquilar Au alunite deposit, Spain. Economic Geology, 90: 795-822. Audetat, A., Giinther, D., and Heinrich, C.A., 1998. Formation of magmatic-hydrothermal ore deposit: Insights with LA-ICP-MS analysis of fluid inclusions. Science, 279: 2091-2094. Barazangi, M. and Isacks, B. L., 1976. Spatial distribution of earthquakes and subduction of the Nazca Plate beneath South America. Geology, 4: 686-692. Beane, R.E., 1991. Deposit correlations in the E l Indio Belt, Chile: Mineral paragenesis, fluid inclusions, and sulfur isotopes. Internal report for Barrick Gold Corporation. Bennet, R.E., 1995. A combined hydrothermal alteration study and exploration target evaluation of the Tambo mining district, E l Indio mineral belt, Region IV, Chile. Internal report for Compania Minera San Jose Ltda. 119 Bethke, P.M., 1984. Controls on base- and precious-metal mineralization in deeper epithermal environments. U.S. Geological Survey Open File Report 84-890, 40 p. Bissig, T., 2001. Metallogenesis of the Miocene El Indio-Pascua gold-silver-copper belt, Chile/ Argentina: Geodynamic, geomorphological and petrochemical controls on epithermal mineralization. Unpublished Ph.D. thesis, Queen's University, Kingston, Canada. Bodnar, R.J., 1995. Fluid-inclusion evidence for a magmatic source of metals in porphyry copper deposits. Mineralogical Association of Canada Short Course Series, 23: 139-152. Bove, D.J., Rye, R.O., and Hon, K , 1990. Evolution of the Red Mountain alunite deposits, Lake City, Colorado. US Geological Survey Open File Report, 90-0235. Burham, C.W., 1979. Magmas and hydrothermal fluids. In Barnes, H.L., ed., Geochemistry of Hydtrothermal Ore Deposits. New York, Wiley: 71-36. Burham, W.C., and Ohomoto, H., 1980. Late-stage processes of felsic magmatism. Mining Geology Special Issue 8: 1-11. Caddey, S.W., 1993. Structural analysis of the E l Indio-Tambo mining district, Chile: exploration guides, ore controls, and target areas. Internal report for Barrick Gold Corporation. Candela, P.A. and Piccoli, P.M., 1995. Model ore-metal partitioning from melts into vapor and vapor-brine mixtures. In Thompson, J.F.H., ed., Magmas, Fluids, and Ore Deposits. Mineralogical Association of Canada Short Course Series, 23: 101-127. Carroll, M.R., and Webster, J.D., 1994. Solubilities of sulfur, noble gases, nitrogen, chlorine, and fluorine in magmas. Reviews in Mineralogy, 30: 231-279. Cooke, D.R., and McPhail, D.C., 2000. Epithermal Au-Ag-Te mineralization, Acupan, Baguio District, Phillipines: numerical simulations of mineral deposition. Economic Geology, 96: 109-131. Cooke, D.R. and Simmons, S.F., 2000. Characterisites and genesis of epithermal gold deposits. In Hagemann, S.G. and Brown, P E . ed., Gold 2000. Reviews in Economic Geology, 13: 221-244. Craig, H., 1961. Isotopic variations in meteoric waters. Science, 133: 1702-1703. Craig, H., and Lupton, J.E., 1981.3He and mantle volatiles in the ocean and the ocean crust. In Emiliani, C, ed., The Sea: 391-428. Cunningham, C.G., 1978. Pressure gradients and boiling as mechanisms for localizing ore in porphyry systems. US Geological Survey Journal of Research, 6: 745-754. Cunningham, C.G., Rye, R.O., Steven, T.A., and Mehnert, H.H., 1984. Origins and exploration 120 signficance of replacement and vein-type alunite deposits in the Marysvale volcanic field, west central Utah. Economic Geology, 79: 50-71. Deen, J.A., Rye, R.O., Munoz, J.L., and Drexler, J.W., 1994. The magmatic-hydrothermal system at Julcani, Peru: Evidence from fluid inclusions and hydrogen and oxygen isotopes. Economic Geology, 89: 1924-1938. Ebert, S.W. and Rye, R.O., 1997. Secondary precious metal enrichment by steam-heated fluids in the Crowfoot-Lewis hotspring gold-silver deposit and relation to paleoclimate. Economic Geology, 92: 578-600. Fournier, R.O., 1967. The porphyry copper deposit exposed in the Liberty open-pit mine near Ely, Nevada. Part I. Syngenetic formation. Economic Geology, 62: 57-81. Fournier, R.O., 1987. Conceptual models of brine evolution in magmatic-hydrothermal systems. US Geological Survey Professional Paper, 1350, v.2: 1487-1506. Fournier, R.O., and Potter, R.W., 1982. An equation correlating the solubility of quartz in water from 25 to 900°C at pressures up to 10,000 bars. Geochimica et Cosmochimica Acta, 46: 1969-1973. Fournier, 1999. Hydrothermal processes related to movement of fluid from plastic into brittle rock in the magmatic-hydrothermal environment. Economic Geology, 94: 1193-1212. Flynn, F.R., and Burnham, C.W., 1978. An experimental determination of rare earth partition coefficients between a chloride-containing vapor phase and silicate melts. Geochimica et Cosmochimica Acta, 42: 685-701. Fulignati, P., Gioncada, A., and Sbrana, A., 1999. Rare-earth element behaviour in the alteration facies of the active magmatic-hydrothermal systems of Vulcano (Aeolian Islands, Italy). Journal of Volcanology and Geothermal Research, 88: 325-342. Fulignati, P. and Sbrana, A., 1998. Presence of native gold and tellurium in the activee high- sulfidation hydrothermal system of the La Fossa volcano (Vulcano, Italy). Journal of Volcanology and Geothermal Research, 86: 187-198. Gammons, C.H., and William-Jones, A.E., 1997. Chemical mobility of gold in the porphyry- epithermal environment. Economic Geology, 92: 43-59. Gerlach, T.M., 1993. Oxygen buffering of Kilauea volcanic gases and the oxygen fugacity of Kilauea basalt. Geochimica et Cosmochimica Acta, 57: 795-814. Gerlach, T.M., and Casadevall, T.J., 1986. Evaluation of gas data from high-temperature fumaroles at Mount St. Helens, 1980-1982. Journal of Volcanology and Geothermal Research, 28 (1-2): 107-140. 121 Gerling E., K., Mamyrin, B.A.,a nd Tolstikhin, I.N., 1971. Helium isotope composition in some rocks. Geochemical International, 8: 755-761. Gieseman, A., Jager, H.J., Norman, A.L., Krouse, H.R., and Brand, W.A., 1994. On-line sulfur- isotope determination using and elemental analyzer coupled to a mass spectrometer. Analytical Chemistry, 65 (18). Giggenbach, W.F., 1992. Magma degassing and mineral deposition in hydrothermal systems along convergent plate boundaries. Economic Geology, 87: 1927-1944. Godfrey, J.D., 1962. The deuterium content of natural waters and other substances. Geochimica et Cosmochimica Acta, 27: 43-52. Gordon, S., and McBride, B.J., 1994. Computer Program for Calculation of Complex Chemical Equilibrium Compositions and Applications. I. Analysis. N A S A Reference Publication 1311. Graney, J.R., and Kesler, S.E, 1995a. Gas composition of inclusion fluid in ore deposits: Is there a relation to magmas?. In Thompson, J.F.H., ed., Magmas, Fluids, and Ore Deposits. Mineralogical Association of Canada Short Course vol. 23: 221-246. Graney, J.R., and Kesler, S.E., 1995b. Factors affecting gas analysis of inclusion fluid by quadrupole mass spectrometry. Geochimica et Cosmochimica Acta, 59: 3977-3986. Guha, J., Lu, L.Z., and Gagnon, M., 1990. Gas composition of fluid inclusions using solid probe mass spectrometry and its application to study of mineralizing processes. Geochimica et Cosmochimica Acta, 54: 553-558. Gustafson, L.B., and Hunt, J.P., 1975. The porphyry copper deposit at E l Salvador, Chile. Economic Geology, 70: 857-912. Hedenquist, J.W., 1995. The ascent of magmatic fluid: Discharge versus mineralization. Mineralogical Association of Canada Short Course Series, v. 23: 263-289. Hedenquist, J.W. and Arribas, A., Jr., 1999. Epithermal gold deposits: I. Hydrothermal processes in intrusion-related systems, and II. Characteristics, examples, and origin of epithermal gold deposits. In F. Molnar, J. Lexa, and J.W. Hedenquist, eds., Epithermal Mineralization of the Western Carpathians. Society of Economic Geologists, Guidebook Series, 31: 13-63. Hedenquist, J.W. and Lowenstern, J.B., 1994. The role of magmas in the formation of hydrothermal ore deposits. Nature, 370: 519-527. Hedenquist, J.W., Simmons, S.F., Giggenbach, W.F., and Eldridge, C.S., 1993. White Island, New Zealnad, volcanic-hydrothermal system represent the geochemical environment of 122 high sulfidation Cu and Au ore deposition. Geology, 21: 731-734. Hedenquist, J.W., Matsuhisa, Y., Izawa, E., White, N.C., Giggenbach, W.F., and Aoki, M., 1994. Geology, Geochemistry, and origin of high sulfidation Cu-Au mineralization in the Nansatsu District, Japan. Economic Geology, 89: 1-30. Heinrich, C.A., Gunther, D., Audetat, A., Ulrich, T., and Frischknecht, R., 1999. Metal fractionation between magmatic brine and vapor, determined by microanalysis of fluid inclusions. Geology, 27: 755-758. Heinrich, C.A., Ryan, C.G., Mernagh, T.P., and Eadington, P.J., 1992. Segregation of ore metals between magmatic brine and vapor: A fluid inclusion study using PIXE micranalysis. Economic Geology, 87: 1566-1583. Hemley, J.J., Hostetler, P.B., Gude, A.J. , and Mountjoy, W.T., 1969. Some stability relations of alunite. Economic Geology, 64: 599-612. Henley, R.W., and Stewart, M.K., 1983. Chemical and isotopic changes in the hydrololoy of the Tauhara geothermal field due to exploitation at Wairakei. Journal of Volcanology and Geothermal Research, 15: 285-314. Hopf, S., 1993. Behavior of rare earth elements in goethermal systems of New Zealand. Journal of Geochemical Exploration, 47: 333-357. Jambor, J.L., 1999. Nomenclature of the alunite supergroup. Canadian Mineralogist, 37: 1323- 1341. Jannas, R.R., 1995. Reduced and oxidized high sulfidation deposits of the El Indio deposit, Chile. Ph.D. thesis, Cambridge, Massachusetts, Harvard University, 421 p. Jannas, R.R., Bowers, T.S., Petersen, U., Beane, R.E., 1999. High-sulfidation deposit types in the El Indio District, Chile. In Skinner, B.J., ed., Geology and ore deposits of the Central Andes. SEG special publication #7: 27-59. Kavaleris, I., 1994. High Au, Ag, Mo, Pb V, and W content of fumarolic deposits at Merapi volcano, Central Java, Indonesia. Journal of Geochemical Exploration, 50: 480-492. Kennedy, B.M., Hiyagon, H., and Reynolds, J.H., 1990. Crustal neon: a striking uniformity. Earth Science Planetary Letters, 98: 277-286. Kornexl, Barbara E., Matthias Gehre, Reiner Hofling, and Roland A. Werner, 1999. On-line 8 1 8 0 measurement of organic and inorganic substances. Rapid Communications in Mass Spectrometry 13: 1685-1693. Landis, G.P., and Hofstra, A.H. , 1991. Fluid inclusion gas chemistry as a potential mineral exploration tool: Case studies from Crede, CO, Jerrit Canyon, NV , Coeur d'Alene district, 123 ID and MT, southern Alaska epithermal veins, and mid-continent MVT's . Journal of Geochemical Exploration, 42: 25-60. Lewis, A.J., Palmer, M.R., Sturchio, N.C., and Kemp, A.J. , 1997. The rare earth element geochemistry of acid-sulphate and acid-sulphate-chloride geothermal systems from Yellowstone National Park, Wyoming, USA. Geochimica et Cosmochimica Acta, 61: 695- 706. Lewis, A.J., Komminou, A., Yardley, B.W.D., and Palmer, M R . , 1998. Rare earth element speciation in geothermal fluids from Yellowstone National Park, Wyoming, USA. Geochimica et Cosmochimica Acta, 62: 657-663. L i , G , Peacor, D R . , Essene, E.J., Brosnahan, D.R., and Beane, R.E., 1992. Walthierite and huangite; two new minerals of the alunite group from the Coquimbo region, Chile. American Mineralogist, 77: 1275-1284. Lindaas, S.E., Chomiak, B.A., Campbell, A.R., and Norman, D.I., 1998. Enargite-hosted fluid inclusion gases from the Lepanto, Philippines, high-sulfidation Cu-Au deposit. Abstracts with Programs - Geological Society of America. 30 (7): 20. Lottermoser, B.J., 1992. Rare earth elements and hydrothermal ore formation processes. Ore Geology Reviews, 7: 25-41. Lowenstern, J.B., Mahood, G.A., Rivers, M.L., and Sutton, S.R., 1991. Evidence for extreme partitioning of copper into a magmatic vapor phase. Science, 257: 1405-1409. Lupton, J.E., 1983. Terrestrial inert gases: istope tracer studies and clues to primordial components in the mantle. Annual Reviews in Earth Science, 11: 371-414. Maksaev, V., Moscoso, R., Mpodozis, C , and Nasi, C , 1984. Las unidades volcanicas y plutonicas del cenozoico superior en la Alta Cordillera del Norte Chico (29°-31° S): Geologia, alteration hidrothermal y mineralisation. Revista geologica de Chile, 21: 11-51. Mamyrin, B.A., Anufriev, G.S., Kamenski, I.L., and Tolstikhin, I.N., 1969. Determination of the isotopic composition of atmospheric helium. Geochemistry International, 7: 498-505. Martin, M , Clavero, J., and Mpodozis, C , 1995. Estudio geologico regional del Franja El Indio, Cordillera de Coquimbo. Servicio National de Geologia y Mineria, Santiago, Chile. Registered report IR-95-06, 232p. Martin, M.W., Clavero, J.R., Mpodozis, C , 1999. Late Paleozoic to Early Jurassic tectonic development of the high Andean Principal Cordillera, E l Indio Region, Chile (29-30°S). Journal of South American Earth Sciences, 12: 33-49. Marty, B. 1989. Neon and xenon isotopes in MORB: implications for the earth-atmosphere 124 evolution. Earth Science Planetary letters, 94: 45-56. McBride, B.J., and Gordon, S., 1996. Computer Program for Calculation of Complex Chemical Equilibrium Compositions and Applications. II. Users Manual and Program Description. N A S A Reference Publication 1311: 177p. Michard, A., 1989. Rare earth element systematics in hydrothermal fluids. Geochimica et Cosmochimica Acta, 53: 745-750. Morrison, P., and Pine, J., 1955. Radiogenic origin of the helium isotopes in rock. New York Academy Science Annals, 62: 71-92. Mountain, B.W., 1999. Trace metals in geothermal fluids: I. Metal behavior at the deep source. In Simmons, S.F., Morgan, O.E., and Dunstall, M, eds., New Zealand Geothermal Workshop: 21 s r , Aukland, November, 1999. Proceedings: 61-66. Muntean, J.L., and Einaudi, M.T., 2000. Porphyry gold deposits of the Refugio District, Maricunga Belt, Northern Chile. Economic Geology, 95: 1445-1472. Ohmoto, H., and Lasaga, A.C., 1982. Kinetics of reactions between aqueous sulfates and sulfdies in hydrothermal systems. Geochimica et Cosmochimica Acta, 46: 1727-1746. Petersen, E.U., and Thompson, A.J.B., 1992. Analysis of alunite and other thermally sensitive minerals. Microprobe Analyst (Internal report for the Dept. of Geology and Geophysics, University of Utah), April: 2-6. Pickthorn, W.J., and O'Neil, J.R., 1985. 1 8 0 relations in alunite mineral: potential single mineral thermometer. Geological Society of America Abstracts with Programs, 17(7): 686. Poreda, R., Radicati di Brozolo, F., 1984. Neon in mid-Atlantic ridge basalts. Earth Science Planetary Letters, 69: 277-289. Pouchou, J.L. & Pichoir, F., 1985. PAP 0(pZ) procedure for improved quantitative microanalysis. Microbeam Analysis, 1985: 104-106. Ramos, V. A., Mahlburg Kay, S, Page, R. N. and Munizaga, F., 1989. La ignimbrita Vacas Heladas y el cese del volcanismo en el Valle del Cura, Provincia de San Juan. Revista Asociacion Geologica Argentina, 44: 336-352. Reed, M.H., 1992. Origin of diverse hydrothermal fluids by reaction of magmatic volatiles with wallrock. Japan Geological Survey Report, 279: 135-140. Reed, M.J., 1995. Distribution of rare earth elements between aqueous fluids and granitic melt. Ph.D. thesis, University of Maryland at College Park. Roedder, E., 1984. Fluid inclusions. Reviews in Mineralogy, 12, 644 p. Rye, R.O., 1993. The evolution of magmatic fluids in the epithermal environment: The stable 125 isotope perspective. Economic Geology, 88: 733-753. Rye, R.O., Bethke, P.M., and Wasserman, M.D., 1992. The stable isotope geochemistry of acid sulfate alteration. Economic Geology, 87: 225-262. Sandeman, H.A., Archibald, D.A.; Grant, J.W.; Villneuve, M E . and Ford, F.D. 1999. Characterisation of the chemical composition and 4 0 Ar - 3 9 Ar systematics of intralaboratory standard MAC-83 biotite. Radiogenic Age and Isotopic Studies: Report 12; Geological Survey of Canada, Current Research 1999-F: 13-26. Savin, S.M., and Epstein, S., 1970. The oxygen and hydrogen isotope geochemistry of clay minerals. Geochimica et Cosmochimica Acta. 34: 24-42. Schoen, R., White, D., and Hemley, J.J., 1974. Argillization by descending acid at Steamboat Springs, Nevada. Clays and Clay Minerals, 22: 1-22. Schwab, R.G., Herold, H, Gotz, C. and de Oliveira, N.P., 1990. Compounds of the crandallite type: Synthesis and properties of pure rare earth element phosphates. Neues Jahrbuch fuer Mineralogie, 6: 241-254. Shepperd, T.J., and Rankin, A.H., 1998. Fluid Inclusin Techniques of Analysis. In Richards, J. P., and Larson, P.B., eds., Reviews in Economic Geology, vol. 10: 125-150. Shinohara, H., 1994. Exsolution of immiscible apor and liquid phases from a crystallizing silicate melt: Implications for chlorine and metal transport. Geochimica et Cosmochimica Acta, 58: 5215-5221. Siddeley, G, and Areneda, R., 1985. Gold-silver occurences of the El Indio belt, Chile. Geology of the Andes and its relation to mineral and energy resources. Symposium, Chile, 18 p. Sillitoe, R.H., 1983. Enargite-bearing massive sulfide deposits high in porphyry copper systems. Economic Geology, 78: 348-352. Sillitoe, R.H., 1993. Epithermal models: Genetic types, geometric controls, and shallow features. In R. V. Kirkham, W.D. Sinclair, R.I. Thorpe and J.M. Duke, eds., Mineral Deposit Modelling; Geological Association of Canada, Special paper 40: 403-417. Sillitoe, R.H., 1999. Styles of high sulfidation gold, silver and copper mineralization in the porphyry and epithermal environments. PacRim '99. Bali, Indonesia, 10-13 October, Proceedings: 29-44. Steiger, R. H. and Jager, E., 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36: 359-362. Steven, T.A., and Ratte, J.C., 1960. Geology of ore deposits of the Summitville district, San 126 Juan Mountains, Colorado. US Geological Survey Professional Paper 343: 70 p. Stoffregen, R.E. and Alpers, C.N., 1987. Woodhouseite and svanbergite in hydrothermal ore deposits: Products of apatite destruction during advanced argillic alteration. Canadian Mineralogist, 25: 201-211. Stoffregen, R.E., Rye, R.O., and Wasserman, M.D., 1994. Experimental studies of alunite: I. 1 8 0 - 1 6 0 and D-H fractionation factors between alunite and water at 250-450°C. Geochimica et Cosmochimica Acta, 58: 903-916. Stoffregen, R.E., and Cygan, G., 1990. An experimental study of Na-K exchange between alunite and aqueous sulfate solutions. American Mineralogist. 75: 209-220. Taran, Y.A. , Bernard, A., Gavilanes, J - C , and Africano, F., 2000. Native gold in mineral precipitates from high-temperature volcanic gases of Colima volcano, Mexico. Applied Geochemistry, 15: 337-346. Taylor, B.E., 1988. Degassing of rhyolitic magmas: Hydrogen isotope evidence and implications for magmatic-hydrothermal ore deposits. Canadian Institute of Mining and Mineralogy Special Volume, 39: 33-49. Taylor, H.P., Jr., 1979. Oxygen and hydrogen isotope relationships in hydrothermal mineral deposits. In Barnes, H.L. ed., Geochemistry of Hydrothermal Ore Deposits. Wiley Interscience, New York: 236-277. Thompson, A.J.B, 1992. Alunite compositions and textures: Relationships to precious metal mineralization. Mineral Deposit Research Unit Short Course #8: New Developments in Lithogeochemistry. University of B.C., Canada. Thompson, A.J.B., Hauff, P.L., and Robitaille, A.J. 1999. Alteration mapping in exploration: Application of short-wave infrared (SWIR) spectroscopy. Society of Economic Geologists Newsletter, Oct 1999: 39. Truesdell, A.H. , Nathenson, M., and Rye, R.O., 1977. The effects of boiling and dilution on the isotopic compositions of Yellowstone thermal waters. Journal of Geophysical Research, 82: 3694-3704. Ulrich, T., Gunther, D. and Heinrich, C.A., 1999. Gold concentrations of magmatic brines and the metal budget of porphyry copper deposits. Nature, 399: 676-679. Venneman, T.W., Muntean, J.L., Kesler, S.E., O'Neil, JR. , Valley, J.W., and Russell, N., 1993. Stable isotope evidence for magmatic fluids in the Pueblo Viejo epithermal acid sulfate Au- Ag deposit, Dominican Republic. Economic Geology, 88, 55-71. Wasserman, M.D., Rye, R.O., Bethke, P.M., and Arribas, Jr., A., 1992. Methods for separation 127 and total stable isotope analysis of alunite. Open file report 92-9, US Department of the Interior, Geological Survey. White, N.C., 1991. High sulfidation epithermal gold deposits: Characteristics and a a model for their origin. Geological Survey of Japan Report 277: 9-20. Williams, T.J., Candela, P.A., and Piccoli, P.M., 1995. The partitioning of copper betweem silicate melts and two-phase aqueous fluids: An experimental investigation at 1 kbar, 800°C and 0.5 kbar, 850°C. Contributions to Mineralogy and Petrology, 121: 388-399. Wood, S.A., 1990a. The aqueous geochemistry of the rare-earth elements and yttrium 1. Review of available low-temperature data for inorganic complexes and the inorganic REE speciation of natural waters. Chemical Geology, 82: 159-186. Wood, S.A., 1990b. The aqueous geochemistry of the rare-earth elements and yttrium 2. Theorectical predictions of speciation in hydrothermal solutions to 350°C at saturation water vapor pressure. Chemical Geology, 88: 99-125. 128 Chapter 4 T H E R O L E O F A L U N I T E I N E X P L O R A T I O N : E V I D E N C E F R O M H I G H S U L F I D A T I O N D E P O S I T S I N T H E E L I N D I O - P A S C U A B E L T , C H I L E I N T R O D U C T I O N Acid sulfate alteration is a characteristic feature of high sulfidation epithermal gold deposits (Cooke and Simmons, 2000; and references therein). Alteration is defined by alunite ± kaolinite (or dickite, pyrophyllite), quartz, and pyrite (Hemley and Jones, 1964; Meyer and Hemley, 1967) and forms under conditions of low pH and highly oxidized fluid chemistry (Holland, 1965; Henley and McNabb, 1978; Stoffregen, 1987). These conditions can be generated by several mechanisms in both the epithermal environment and in lithocaps overlying porphyry deposits. Four specific environments of acid sulfate alteration were defined by Rye et al. (1992; Fig. 4.1) on the basis of field characteristics, paragenetic associations, geochemical signatures, and stable-isotope systematics; each of which may be present within high sulfidation systems. These include: • The magmatic-hydrothermal environment, where the disproportionation of magmatic S 0 2 in condensed magmatic vapor forms H 2S and H 2 S 0 4 below ca. 350°C. The resulting acidic fluids react with wallrock to form extensive zones of acid (chloride-) sulfate alteration. • The magmatic steam environment, where alunite is believed to form from the expansion of rapidly ascending S0 2-rich magmatic vapor following sudden depressurization of the hydrothermal system (Rye, 1993). • The steam-heated environment, where alunite forms from the condensation of vapour derived from an underlying hydrothermal system and the oxidation of H 2S gas above the water table. • The supergene environment, where the supergene oxidation of sulfides can produce an assemblage of kaolinite ± quartz, alunite, jarosite, with iron oxide and oxyhydroxide minerals. 129 Figure 4.1. Schematic diagram of a high sulfidation epithermal system showing environments magmatic-hydrothermal, magmatic steam, and steam heated alteration. Supergene processes not shown. Effective exploration for high sulfidation deposits requires an understanding of the origin of acid sulfate alteration, since each assemblage has different genetic and spatial relation to potential mineralization (Bethke, 1984; Thompson, 1992; Sillitoe, 1993). Alunite and associated alteration minerals are now easily and routinely identified in the field by short-wave infrared (SWIR) field spectrometers and hyperspectral techniques (including airborne surveys), but the identification of critical ore-related assemblages is often difficult due to multiple or superimposed alteration events. To date, a number of techniques have been used to differentiate between the environments of acid sulfate alteration. These include field characteristics (e.g., distribution, mineral assemblages, textures, paragenetic relations), chemical composition, age dating, and stable isotopic signatures of alunite and associated minerals (e.g., Bethke, 1984; Rye et al., 1992; Sillitoe, 1993; Thompson, 1992; Arribas et al., 1995; Jannas et al., 1999; Thompson et al., 1999; Hedenquist et al., 2000). Few studies however include detailed data for more than one type of alteration from a single deposit or district. These observations are critical for 130 exploration at either a regional- or deposit-scale if complex alteration patterns are apparent. In this paper, we present results from an extensive study of alteration associated with high sulfidation Au-Cu-Ag mineralization in the E l Indio-Pascua Belt. This belt straddles the Chile-Argentina border between latitudes 29°20' and 30° S in the Main Andean Cordillera (Fig. 4.2). The region hosts widespread zones of hydrothermal alteration and several world-class epithermal deposits, including E l Indio-Tambo, which has produced over 11 Moz of gold, Pascua-Lama (proven-and-probable resources 17 Moz Au and 560 Moz Ag), and Veladero (15 Moz Au, 230 Moz Ag). Many smaller occurrences such as the Del Carmen and Salitrales prospects are also recognized. The comprehensive nature of this study is possible because: • The deposits in this region are young (ca. 6 to 9 Ma; Bissig, 2001) and typically well preserved, due to favorable tectonic and climatic conditions. • Several deposits have been previously studied in detail (e.g., E l Indio-Tambo, Jannas et al., 1999; Pascua-Lama, Chouinard and Williams-Jones, 1999), and metallogenetic and geochronologic constraints have been determined (Bissig, 2001). • Access is available via the extensive mining and exploration operations of Barrick Gold Corporation, the principal operator in the region. • Multiple stages of acid sulfate alteration are recognized in each deposit, including one or several magmatic-hydrothermal, magmatic steam, steam-heated, and/or supergene assemblages. We present herein a summary of the physical, chemical, stable-isotope, and age characteristics of alunite-group minerals, and their role in differentiating between origins of acid sulfate alteration. Our discussion is focused on high sulfidation systems in the El Indio- Pascua Belt, but applications to exploration in other epithermal districts are addressed. G E O L O G Y OF T H E E L INDIO-PASCUA B E L T The El Indio-Pascua Belt (Fig. 4.2) host several high sulfidation deposits and prospects in Paleozoic intrusions and volcanic rocks (e.g., Pascua-Lama, Veladero) and Oligocene to Middle Miocene volcanic and volcaniclastic rocks (e.g., E l Indio-Tambo, Sancarron). Mineralization in the belt occurred ca. 9 to 6 Ma, between Vacas Heladas and Vallecito age volcanism (Bissig 2001; Jannas et al., 1999). The geology and styles of alteration at the El 131 S i m p l i f i e d g e o l o g y : • Upper Pliocene | Cerro de Vidrio: rhyolite dome V////X Upper Miocene Vallecito Formation: ^ rhyolitic pyroclastis Middle to Upper Miocene Vacas Heladas Formation: dacitic pyroclastics 1 Oligocene to Middle Miocene Cerro de las Tortolas Formation: predominantly andesitic flows; Infiernillo intrusive unit; Granodioritic to dioritic intrusives Escabroso Formation: Andesitic flows and volcaniclastic sediments; hypabyssal diorites and granodiorites Tilito Formation: predominantly dacitic pyroclastic rocks Bocatoma intrusive unit: diorites and granodiorites Paleozoic to Jurassic Predominantly felsic intrusive and volcanic rocks Faults Reverse faults ~| widely reactivated Mined deposits Major exploration projects Minor exploration projects International border V Chile-Argentina 20 km •M Figure 4.2. Map of the El Indio-Pascua Belt showing locations of epithermal deposits and prospects. Simplified geology is given according to the stratigraphy defined in Bissig (2001). BdTF = Banos del Toro fault. Adapted from Bissig (2001). 132 Indio and Tambo deposits have been described previously (see Chap. 3: Jannas et al., 1999; Jannas et al., 1990; Siddeley and Areneda, 1986). Both deposits are hosted in intensely altered Tertiary rhyodacitic volcanic rocks. E l Indio is characterized by sulfide-rich mineralization in complex enargite-pyrite (copper dominant) and later gold-quartz vein systems (Jannas et al., 1999). Silicic, argillic, and sericitic alteration assemblages are dominant. Alunite is rare. It occurs in patchy kaolinite- and sericite-bearing alteration assemblages in the proximity of the Campana, Viento, and Mula Muerta copper veins and as banded alunite-enargite-pyrite veins in the Campana and Brechita-Huantina areas (Jannas et al., 1999). In contrast, minerahzation at the Tambo deposit is hosted in sulfide-poor alunite-barite-gold breccias and several high grade alunite-barite veins. The deposit contains at least six stages of acid sulfate alteration, including magmatic-hydrothermal (Stages 1 and 2), magmatic steam (includes Stage 3 and banded alunite veins), steam-heated, and apparent supergene (details see Chap. 3). The Pascua property is located at the northern end of the El Indio-Pascua Belt, straddling the Chile-Argentina border. Gold, copper, and silver mineralization occurs in several forms (Chouinard and Williams-Jones, 1999). These include abundant alunite-pyrite-enargite breccia matrix fill and banded veins, and extensive silica-pyrite ± Fe-sulfate alteration zones. Several smaller, high grade oxidized veins and breccias occur in the Esperanza and Penelope deposits. Mineralization and alteration is hosted primarily in intrusive rocks of the Carboniferous-Lower Triassic Pastos Blancos group. Several stages of acid sulfate alteration are recognized (Chap. 2). Pre-ore (AS I) and syn-ore (AS II) alteration are both magmatic- hydrothermal in origin. Magmatic steam alunite (AS III) occurs locally near-surface. Steam- heated and supergene assemblages are also recognized. Del Carmen is a smaller prospect located south of the E l Indio mine in the San Juan province of Argentina. Gold (± copper, silver) mineralization occurs as veins, infilling breccia matrices, or disseminations in association with sulfides; principally pyrite and enargite (Noriega and Perez, 1997). Rare native Au is observed. Stratigraphy in the project area consists of a sequence of pyroclastic flows and lapilli tuff flows that correspond to the upper part of the Dona Ana Formation. Andesitic to dacitic tuffs and lavas of the Cerro de Tortalas Formation overlie the Doha Ana rocks and are interbedded with pyroclastic flows and volcanic breccias (Gallardo, 1995). Hydrothermal breccias outcrop as isolated bodies intruding the tuff sequences. Alteration is associated primarily with the Cerro de las Tortalas Formation and assemblages grade laterally from vuggy silica to quartz-alunite to quartz-alunite-kaolinite-illite 133 to smectite-pyrite (Noriega and Perez, 1997). The Salitrales prospect is located a few kilometers south-east of E l Indio. The area has not been explored extensively, but contains several anomalous Au showings (Paleczeck et al., 1996). The property consists of a small porphyritic dacite stock that intrudes a sequence of Miocene volcanics. Anomalous Au, Ag, As, Sb, Pb, Hg, and B i concentrations occur in soils related to siliceous (vuggy quartz) and acid sulfate alteration assemblages. Alteration is typically structurally controlled and is concentrated in and around several strongly silicified breccias of probable hydrothermal origin (Paleczeck et al., 1996). Minor pyrite and enargite are found at surface. Alunite is typically overprinted with late jarosite ± scorodite. M E T H O D S OF ANALYSIS Methods used for the characterization of alunite and acid sulfate alteration in the E l Indio-Pascua Belt are summarized below. A detailed account of sample preparation and analytical techniques is provided in Appendices A, B, and C. Field Techniques: Detailed alteration mapping and sampling at E l Indio-Pascua Belt properties was completed between 1998 and 1999. Access was made possible through the extensive mining and exploration activities of Barrick Gold Corporation. This allowed an examination of both surface and underground exposures, and detailed sampling of drill core, wherever available. Considerable effort was made to study complete vertical and lateral sections across alteration zones. Petrography: Detailed petrographic studies are required to accurately detenriine the paragenesis of alteration minerals and/or assemblages. Both transmitted and reflected light techniques were used throughout this study. Scanning electron microscopy was often necessary to identify fine-grained assemblages and intergrowths of alunite and clays that could not be distinguished in thin section. SWIR Spectroscopy: Alunite and associated minerals are easily detected with field portable SWIR spectrometers (e.g., PIMA, FieldSpec Pro). The spectrometers detect the energy generated by vibrations within molecular bonds, in the 1300 to 2500 nm region of the electromagnetic spectrum (Thompson et al., 1999). SWIR is particularly sensitive to certain molecules and radicals, including OH", H 2 0 , N H / , C0 3 2 " , and cation-OH" bonds such as A l - OH, Mg-OH, and Fe-OH. Minerals are distinguished on the basis of distinctive features and wavelength positions, and by the character of the SWIR profile. SWIR spectroscopy is suitable 134 for many alteration minerals such as phylosilicates, clays, carbonates, and selected sulfates - including alunite (Thompson et al., 1999). In this study, a PIMA instrument was used in the field for regional reconnaissance work and more detailed alteration mapping at Pascua (details see Chap. 2). A database of over 5000 SWIR spectra was collected from samples across the E l Indio-Pascua Belt. This study complements initial SWIR analyses compiled by Bennet (1995) for the Tambo region. Geochemistry: Alunite belongs to a large group of minerals, the alunite supergroup, that consists of over 40 mineral species. Minerals have the general formula DG 3(T04)2(OH,H 20) 6 and can be sub-divided into 3 groups based on the T-site occupancy (Jambor, 1999). The alunite group is characterized by (S0 4) 2" dominant minerals; the beudantite group contains both S 0 4 and either P 0 4 or A s 0 4 ; and T 0 4 in the crandallite group represents one or both of P 0 4 and As0 4 . End-member compositions are rarely attained in natural alunite occurrences and extensive solid solution is typical for one or more of the D, G, and T sites (Stoffregen et al., 2000). The most common and naturally abundant minerals belong to the alunite-natroalunite and jarosite-natrojarosite solid solution series, but many other minerals are known to occur in high sulfidation epithermal environments (Table 4.1). In this study, geochemical characteristics of alunite-group minerals (hereafter referred to as alunite, except where specified) were determined by analysis of individual grains and sample composites. Grain mounts and polished sections were analysed by scanning electron microscope with energy-dispersion spectroscopy (SEM-EDS) and electron probe microanalysis (EPMA). E P M A analytical parameters were chosen in order to minimize beam damage caused by the volatilization of alkali elements (Petersen and Thompson, 1992). This requires a large beam size (15-20 u.m), that results in poor representation of finer-grained samples (i.e. particularly supergene and steam-heated alunite) and thin compositional growth bands. Alunite separates were also analyzed by inductively-coupled plasma mass spectrometry (ICP-MS) for trace metal and rare earth element (REE) contents. Approximately 1 gram of material from each sample was separated for analysis and treated using a 1:1 HF/H 2 0 solution to remove silicate contaminants, if present (Wasserman et al., 1992). The purity of separates was verified by powder X-ray diffraction (XRD) but trace amounts of quartz, clays, or other impurities may be present. Samples were digested with perchloric, nitric and hydrofluoric acids for trace-metal analysis. REE were determined by lithium metaborate fusion and ICP-MS techniques. Stable-isotope Systematics: Alunite-group minerals contain four stable-isotope'sites; D 135 o I ~ I ~ x t o — I ~ , r 0 3 ,r ~ to ' O - i O => ~s o O co o co o co CO CO " •• - ' - - 1 t - T u ; J-I 2 J3 ± - < CL 6. 0- CO X o co 0 « IO 1 E | CO Q} O CO o ~ "c o C <D O X> o > E E S* 2 £ = CO -O CO CL x O X 2 o ; o ~ ~ x - i x o o f/i -—- - ^ v _S? CJ < ra ™ « B I " ^ z £ o D z i o qo I™ x - x - x" 1 I o a a So ^ ~ X c o x x n O 2 s r J O J O O - j c o q ~ 5 O O O o < J O CO CO Q." f - " < 3 < So i x q i q O I x" I I I 2 aa° O O O O o O CO < < < < « n co a CQ Q . O - J .•e o 0) ~ .2 E E o> -2 o » « c £ f O E £ 5 S 2 E °> sa o E co co I o 01 o 136 (OH); 3 4S(so4)> 1 8OOH> and 1 8 0 S 0 4 . Isotopic variations are related to the source and type of fluids, rates of processes, and the physical-chemical environment of deposition (Rye et al., 1992). Isotopic studies are particularly useful in conjunction with analyses on coexisting minerals such as clays (kaolinite, dickite, pyrophyllite), muscovite/sericite, and sulfides. The stable-isotope systematics of the four environments of acid sulfate alteration are described in detail in Rye et al. (1992) and Rye (1993). A l l data for this study (Appendix B) was collected at the U.S. Geological Survey Isotope Laboratory in Denver. Geochronology: Alunite is useful for age determinations by K-Ar and 4 0 Ar / 3 9 Ar methods due to the large concentration of potassium in end-member alunite (and jarosite). Alunite age data have been applied to supergene and weathering events (see Vasconcelos, 1999 for details) and paleoclimate studies (e.g., Vasconcelos et a l , 1994; Arehart and O'Neil, 1993; Sillitoe et al., 1991; Bird et a l , 1990). Hypogene alunite can also be used to date hydrothermal events and associated precious- or base-metal mineralization (e.g., Bissig, 2001; Arribas et al., 1995; Perello, 1994; Sillitoe et al., 1991; Alpers and Brimhall, 1988; etc.), provided alteration is coeval with mineralization. Data for this study were collected at the Queen's University 4 0 Ar - 3 9 Ar laboratory, equipped with a Mass Analyzer Products MAP 216 mass spectrometer. About 10 mg of each sample was irradiated for 7.5 hours at McMaster nuclear reactor in Hamilton, Canada using biotite standard Mac-83 as a radiation flux monitor (24.36 ± 0.17 Ma; Sandeman et a l , 1999). Samples were step-heated using a defocused L E X E L 3500 Ar laser-beam. Ages are calculated using the decay-constants suggested by Steiger and Jager (1977) and all errors are given at 2a. Fluid inclusions: Fluid inclusions data for alunite are rare. These studies are difficult because inclusions are typically small, necking and decrepitation are common (this study; Jannas et al., 1999; Beane, 1988), and it is often difficult to determine the paragenesis of inclusion populations. Associated phases such as quartz, barite, and/or sulfides may be suitable for fluid inclusion analysis, provided they are coeval with alteration. In this study, only relatively coarse-grained alunite (e.g., magmatic-hydrothermal and magmatic steam) contained sufficiently large inclusions to study petrographically. Enargite samples from both Pascua and Tambo were unsuitable for analysis (A. Chouinard, pers comm.). Analyses of gas species contained in fluid inclusions were determined for several coarse-grained alunite samples from the Tambo area. Thermally released gas from ca. 1 gram alunite separates was analyzed by Pfeiffer quadrupole mass spectrometer (Prisma) at the U.S.G. S. facilities in Denver. Analytical details provided in Chapter 3. 137 E N V I R O N M E N T S O F A C I D S U L F A T E A L T E R A T I O N Characteristics of the four environments of acid sulfate alteration in the E l Indio-Pascua Belt are summarized in Table 4.2. The nature of each of these environments is described in the following section, with particular reference to their relation to potential high sulfidation ore. Detailed physical, chemical, and isotopic data specifically for alunite-group minerals within each assemblage are also given (Table 4.3). 1. Magmatic-Hydrothermal Environment Magmatic-hydrothermal alteration is most closely related to potential high sulfidation (hypogene) mineralization (e.g., Sillitoe, 1999; Hedenquist et al., 2000). Alteration results from high temperature acidic fluids produced from the condensation of magmatic volatiles containing S 0 2 , H 2 S, HCF, and HF. The dissociation of dominant acidic species, HC1 and H 2 S0 4 , below about 300 to 350°C (Hemley et al., 1969; Knight, 1977) creates acidic fluids with a pH of about 1. These fluids are sufficiently acid to leach most components from the host rock, often leaving a vuggy quartz residue. The progressive neutralization of these fluids by reaction with wallrock forms successive alteration envelopes of alunite ± dickite/Tcaolinite, to kaohnite/dickite ± pyrophyllite, to illite-smectite, and smectite-chlorite outward from the main fluid conduit. Strong structural or hthological controls can influence the shape and distribution of alteration as fluids flow down the hydraulic gradient, along the most permeable channel. The permeability provided by early stage residual quartz commonly controls subsequent fluid flow and precious- metal deposition (e.g., Summitville; Stoffregen, 1987). In some deposits however, the silicic zone is absent and gold ore is hosted in acid sulfate (e.g., la Mejicana, Losada-Calderon and McPhail, 1996) or advanced argillic assemblages (e.g., Pueblo Viejo, Kesler et al., 1981). Characteristics of magmatic-hydrothermal alteration in the E l Indio-Pascua Belt are extremely variable. Alteration occurs as widespread acid sulfate and silicic assemblages that grade outwards to argillic and propylitic alteration on a broad scale in both the E l Indio-Tambo and Pascua-Lama districts. This style of alteration ('wallrock' alteration) typically pre-dates precious metal mineralization. Acid sulfate alteration is also directly associated with precious- metal mineralization in the Tambo, Pascua, and Del Carmen areas. Alteration of this type ('gangue' alunite) is strongly structurally controlled and occurs as disseminations and open- space filling in sulfate-rich (e.g., Tambo) and sulfide-rich (e.g., Pascua, Del Carmen) breccias 138 0) c cu O) a> Q. 3 w 9? •a tS 5 ro co ^ CO . c o > s I I CO CO ro co ro >> 3 ro 8 • " • § . 3 • ro xj c o 0 C L = ° § o Q) co ro o •ti ro CO i - 0 cu > Q. c o co 2> o . . g CD !Z T5 — > > ro -c" - J -*—> ro £ +1 3 CO «P 0 CD o af 0 '-5 LL O 1— u.- O ro o -3 CO <o^ 3 v 8 CO CO 2 CD t« ro iE a: 3 CO o ^ 3 ro ^ o "D CO 0 ro N D_ l - i C l i ? CD CO ^ < £ CD o J5 CD ro S- CO 0 ° § ro I ro ° 5-= ™ Q .^ < y_ ° S o Q- c c <3 e 73 a> co a> X E co CD CO CD ro^ O ) — o o O - ±= 2. = r- O O = " O CO o o ro £ _ o ro O CO CO N " O C CD C O >>2 ro W CD c 3 "d" ro fro ro o 9- ro -2- = +i CO o > o < 2 o • c >» co o 5 ro „ 0 5 !s O CO CD E £ ro E ro 0 (0 o *3 co E co S +1 5fi = CO 8 ^ . £ •= 0 I I I 0 co T3 CO 8 I CO CD CÔ i i E 0 x: +1 c 3 TD 0 ro 0 2 £ ° E i2 ro co h- to •- O - o 3 x: co a. >» >» 0 0 = = O ) X J § 8 2 mow . o 0 X3 o E o ~ 0 a. ro a) 0 JS ro o W 3 CO D c >* CO CO E u 0 0 3 CO c CO O co E c?1 S o o ro 5 CO ro .g ro "S E ^ • i_ > w CD i - 2 c o ±3 »- i- o co o 0 "5 T3 ~ 2 «5 CD *+- 2 CO Q.-0 s i § s o o It: Q. CO O c 0 H o OL ro " O 2 X5 c 0 E CO -*-» 0 o ro CO as  ar g i a l - >, !H > ° -ro o ^ CO o TD > , . C O 3 ro N +i & CT^ C C L 3 CO 0 J£i " ? o .9 0 E o w w co O CD o CO 0 C L o 0 > c LU ro 0 O) O r- 'o. ro c o 2 o £ a. o 2 ro ^ cu ro co 0 c C L 3 c ra l 0 o r al  * ^ 0 ro rib u  M in  m bl  -•—« CO se  b as  C L ^ > Z ro S ro o °- ~ CO a: o .2 0 JO O 0 o 139 t» c V 2> 0) a 3 CO •a to CU E CO cu (0 E CO CU to o CO E C O CO s CO I CO c 3 CD D) C CO O o o CD C C CD •a co P ro ?> o to CD Q - " " o . CO CO CD - 3 ( 0 CO O CO § g | — — o Ii CD > ro 3 D ) CD • E = : CD ° CO Q . .2 O TO — CO g . 8 " i d E JD 5 a> ro co £ > t » a c o >;= £ ~ CT > co l i s 3 .„ CO a. CD co . CD CO CO CO E y v- to o „ a) « .E t: C CO O P <D E co .E .or 2 "S • CO 5 ^ CD CO CD «= CD > TO JO co ro § Si jo . £ 1 CT CD ~ o - <*= "D CO B CD £ - — — CO CD ro co ^ c: • | 2 £• • 2 ra § > CO "to ro . CO o CO CD J O CO CO I t ro co 45 CC < •a CD T l ro J O _ ; >> E CD "D is » ro J Z or c ro ro CO CO QT _ro ro 3 3 JD O ro 'o •*~* ro TO CD TO ro CO T O 1 E CO =J- co o D> L O T O L O E f 2 2 . a. CD ,>- T3 CD c CO • _ •§j <o E G « i CO - 1 o i to c V O i l ) b 7 6 E "O LO O T O CO CD 'o o | : T O a ln  CD C CD CD T O CO J O E CO _ CD J O 52 O CO X co o CD 'p. 3 o 3 ro J O o t3 E CO s T O t— M e E o to >. ro o T O •II CO ™ o <?>LO CD CM ( 0 o CO CM 8 ^ 2 8 o CO c o 3 Lo  CO co d 3 o Q . 3 O CO CO CD T O Z "5 CO CD J O .2, .ro c CO 3 > CO J O " 2 CO iun w ro ed  ro 3 > Q. CD >> CD "EZ >* CO CD ro Z o o _ i ( J * i_ 0 ) E o 1c 6 'c o 3 0 ) ro >. ro 0 z c .9 uj •g w .« 0- or ^ ro ̂  * J T CQ -= c 3 § I co ro . c o )̂ E CO J O CD o>^ .2 •co ro o. > < .9 15 o .2, • 5 c CD = 3 c ° m o CO = ° ro >, CO o = Z ° CO i —I 3 ^ . CO CO 3 .2? c 9? 'E ™ 9 cc N CL CD" ro - —I TO j«: ro t o ™ <u 0 = « . 1 -ii S " to o °- = o Q ^ T J o •= S"g. z 2 -° ro i _ LU - C ro yj co II CO LU LU c C£ O to « S E .2 CD o o ro S «> CO X J o c »- ro CO JO +l o CL - • B i -ro — CTJ = S o a « +i s 1 S § o ? | 2 2 '5 co £ ro co > o . > J E CD Q s i § = CO ai co JO i_" .2 co to ro" > O c ca •2 co CD 2 CD TO X J CD LU 9- ro LU o o c c ts ro-B is . to O co J O o co CO E = 9? o £ O i= ro o 5 ro i f f o t . o ~ ra J= o .E <->-». •= c e o ' o N CO cb 2 z ^ ° s CO o CL < 9s K l to 2 ra /n  s/ . m ic  at u  ra /n  s/ . he  c O) CD G e o c  tfl 9 o f - 0 s < ra jo co = a 3 CO CD E 6 CD m CO a § 2 to co > : in I o ! CM ' O ll CO 3 CD CO 3 to _ £ . Si d 5 >.co E Lo 8 E 3 „ 2 S o - 9 0 ^ 5 ro I ro CO S CD ro ro CO S O T3 3 C O "O O ro to Q 1 E CO ™ J O RO ro—^ aS ro' E § ro " c 3 o .E j= O c c ™ | 8 - CD N CO CD J 2-o LO CO A CD § ^ J O CM 15 g >> .2 o to CL CO 3 tZ. CD 3 S D o to ° -a X ) c 3 CO CO 3 ro I CD CO > 0 ) ro ro i to ra CD CO RO S E ° — to c co | I o Q E co m J : J -v o A V A CD J = c o §> g co ro j= A " ; co CD L_ ro or J O .2 ro > CD CD 2 = O TJ U C o « J 3 CO J = "to O 18 . o i= co o o fO JC UJ 2 m £ CB m a. o o = . 2 tu 0) o S 2 to c o w 3 O c T3 3 140 and veins. Paragenetic and textural observations indicate that alunite is coeval with mineralization in these deposits. This close temporal and spatial relationship between alunite and hypogene ore is rare. Similar observations are documented only at la Mejicana deposit, Argentina (Losada-Calderon and McPhail, 1996) and, to a much lesser extent, at Lepanto in the Philippines (Hedenquist et al., 1998). Differences between gangue and wallrock alteration will be highlighted in the following discussion, to emphasize ore-proximal characteristics. Physical characteristics Mineral associations: Wallrock alteration occurs as aggregates replacing feldspar phenocrysts, disseminated in the wallrock matrix, infilling vugs left from the leaching of fragments or feldspar phenocrysts, and as irregular alunite veinlets and stringers. Original lithic textures are commonly but not always preserved. Alunite is typically intergrown with quartz, pyrite ± kaolinite ± dickite ± pyrophyllite and common accessory rninerals; diaspore, zunyite, APS (Fig. 4.3a). Alteration can be vertically and horizontally zoned with quartz-alunite- kaolinite near surface and at external margins, grading to pyrophyllite-zunyite and/or diaspore at depth. Vertical zonation is particularly apparent at Pascua, where alunite-pyrophyllite (+ zunyite, diaspore) assemblages occur at depths below about 4550 metres above sea level (m.a.s. 1.; all elevations reported hereafter are a.s.l.). Alunite-dickite is most common between 4550- 4750 m, and kaohnite-alunite alteration is dominant above this zone. Gangue alunite occurs in association with barite, sulfides (pyrite, enargite-luzonite, covellite, tetrahedrite-tennantite, chalcopyrite) ± quartz, native sulfur (Fig. 4.3b). At Tambo, alunite overgrows barite and cements breccia fragments or occurs in coarse-grained veins intergrown with barite and lesser quartz. At Pascua, alunite occurs in the matrix of several hydrothermal breccias, intergrown with precious metal-bearing sulfides. Banded veins in surrounding stockwork zones contain coarse-grained alunite, pyrite, and enargite in successive bands parallel to the vein wall. Accessory diaspore and pyrophyllite occur locally. Crystal size and habit: Magmatic-hydrothermal alunite is typically medium to coarse- grained (ca. 20-250 um), but grain size can range between 5 fim to 2 cm. Degree of crystallinity and crystal morphology are variable. Bladed and tabular habits are common. Very thin, elongate to acicular grains are found locally in open spaces of vugs and breccia matrices. Pseudo-cubic forms are rare. Alunite colour ranges from white, pink, yellow, cream, to grey, and single crystals are locally visibly zoned (Fig. 4.4). Colour is often related to inclusions of 141 sulfides, jarosite, or hematite contained in alunite. No consistent relation between alunite composition and colour has been determined. Figure 4.3. (A) on left. Typical disseminated magmatic-hydrothermal acid sulfate alteration with alunite, pyrophyllite, and pyrite. Small APS grains are found in cores to some alunite grains. (DDH - LM03, 300.1m; Pascua-Lama). (B) on right. Banded alunite (white-cream)-pyrite-enargite (dark) vein from Pascua deposit (Maria tunnel). Figure 4.4. Coarse-grained magmatic-hydrothermal alunite from Del Carmen. Individual crystals are visibly zoned. 142 Geochemical characteristics Compositional variability is common in magmatic-hydrothermal alunite from the El Indio-Pascua Belt (Table 4.4). Most minerals belong to the alunite-natroalunite solid-solution series but walthierite, minamiite, huangite, beudantite-group (e.g., woodhouseite, svanbergite) and crandallite-group (e.g., florencite) compositions occur locally. Data are consistent with other studies that indicate elevated P 0 4 with variable Na, Ca, Sr, Ba, Pb, and/or REE in alunite of magmatic-hydrothermal origin (e.g., Aoki, 1991; Aoki et al, 1993; Thompson, 1992). Results from this study are also consistent with experimental (Stoffregen and Cygan, 1990) and empirical evidence (as summarized in Stoffregen et al., 2000) that indicate greater Na substitution is favored at higher depositional temperatures (Fig. 4.5). Elevated Ba and, to a lesser extent Ca, are also characteristic of magmatic-hydrothermal alunite at several localities (Fig. 4.6). Zoning: Compositional variations in alunite occur as intracrystalline growth bands or irregular zones, rather than as discrete grains of homogeneous composition. Disseminated alunite are commonly cored with phosphate ± Ca, Sr, REE-rich compositions surrounded by alunite-natroalunite solid-solution (this study; Aoki, 1991). At Salitrales, these cores are typically Na-rich and are rimmed with P 0 4 ± Ca ± Sr and finally K-rich alunite. Oscillatory zoning between alunite-natroalunite and beudantite or crandallite compositions is also common, both in wallrock and gangue alunite. This zoning is particularly evident in Stage 2 alunite at Tambo, where alunite (or alunite-natroalunite) cores are overgrown by successive bands of walthierite and K-dominant alunite (Fig. 4.7a). Other well developed zoning is noted between huangite-alunite compositions (see Fig. 3.4g). P 0 4 enrichment is often coupled with Sr or Ca substitution (Fig. 4.8), likely by the reaction S 0 4 2 " + D + P 0 4 3 " + D 2 + , where D represents the cation site in 12-fold coordination (Stoffregen et al., 2000). Irregular geochemical zoning is also observed at several E l Indio-Pascua belt localities, particularly between alunite and natroalunite compositions (Fig 4.7b). Similar features are reported in the Baguio district, Philippines (Aoki et al., 1993) where they are interpreted to represent post-depositional Na depletion. Compositional zoning is rarely evident in thin section, due to similarities in optical properties of the alunite-group minerals (Stoffregen et al., 2000). Zoning that is visible petrographically rarely corresponds to significant compositional differences and may represent inclusions or impurities concentrated along growth surfaces. Backscatter imaging is most 143 o Ol S <o in a II Z c £ CJ X to o co 5 U K >. o> X II u c 0 a CM E Ol a 2 C M CD 5 Tf 1- CO o CO b CM IO OJ CO CO O y- CO CO t- CNJ CO r - O O i o i n b b tf> CM C\J CO OJ CO q co ^ b q q b b q q b b o o 0) i- q b to o q oi b co i- r- m co m t- co 00 CM CM CM t- T - 2 O CQ CO O CO d T- r\ OJ CO CO T - O CO I q b CM 0 ) q co O ^ T - | CM O O CO O T- O »- CM CM CO CO CO TI- CO d d d d 144 80 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Atoms p.f.u. Na Figure 4.5. Histogram showing the range of Na substitution in alunite from 4 environments of acid sulfate alteration. Based on EPMA data from El Indio-Pascua Belt properties, including Pascua-Lama, El Indio, Tambo, Del Carmen, and Salitrales. Details of analytical procedures given in Appendix A. Total number analyses (n): magmatic-hydrothermal n = 976; magmatic steam n = 138; steam heated n = 47; late stage and supergene n = 43. Na K Figure 4.6. Ternary diagrams showing range of compositions (K-Na-Ba and K-Na-Ca) for all El Indio-Pascua Belt magmatic-hydrothermal alunite. Based on EPMA results. 145 Figure 4.7a. Oscillatory zoning between alunite (dark) and ierite (light) compositions (Canto Sur area, Tambo). walth- Figure 4.8. SEM photomicrograph (top) showing compositional zoning in magmatic-hydrothermal alunite (Tambo, Stage 2. Wendy deposit. DDH-92a, 290m). Qualitative EDS element maps (bottom) show distribution of K, S, P, Ca, Sr, and Ba. Lightest coloured zones correlate with highest concentration of each element. 147 effective for identifying (qualitatively) geochemical variations in alunite. In some cases, these variations can also be detected by SWIR analyses (Fig. 4.9) or by X-ray diffraction (XRD), if phases are present in sufficient quantity. In general, compositional zoning is best developed in alunite that occurs in small vugs or open spaces, either within wallrock or large barite and sulfide crystals. Alunite sampled from larger veins and breccias is typically more homogeneous. These observations suggest that geochemical zonation is not strongly affected by large-scale variations in external factors such as pressure, temperature, or composition of the source fluid, as suggested for strong growth zoning in hydrothermal garnets (e.g., Jamtveit, 1999; Crowe et al., 2001). Instead, variations in alunite-group mineral chemistry may be attributed to feedback mechanisms associated with supersaturation and growth kinetics, in a method analogous to the porous media experiments of Putnis et al. (1992) and Prieto et al. (1997). REE signature: REE data for magmatic-hydrothermal alunite are summarized in Fig. 4.10. Results indicate that while magmatic-hydrothermal alunite cannot be distinguished from other alteration types on the basis of REE fractionation trends, REE patterns can be used to separate wallrock and gangue alunite. Gangue alunite is characterized by a strong fractionation of heavy REE (HREE) when normalized to host rock composition. In contrast, wallrock alunite typically has a U-shaped REE distribution with variable enrichment of light REE (LREE) and HREE, and depletion in middle REE (MREE). Similar patterns were observed by Terakado and Fujitani (1998) for magmatic-hydrothermal alteration in southwestern Japan. These differences are likely related to fluid-rock ratios and the amount of fluid-rock interaction (e.g., Lottermoser, 1992), and are discussed in detail in Chapters 2 and 3. Stable-isotope characteristics Isotopic equilibrium between all sulfur and oxygen species is expected in the magmatic- hydrothermal environment, due to the highly acidic fluids and high temperatures of formation. However, the alunite OH-site is prone to retrograde isotopic exchange during cooling (Rye et al., 1992). Alunite in the E l Indio-Pascua Belt are characterized by high S 3 4 S values (Fig. 4.11a) and overlap with data for barite, where available. One exception is alunite from the Brechita-Huantina vein (El Indio deposit) that has much lighter 8 3 4S values of 6 to 7%o. Sulfur- isotope data for pyrite from this vein is similarly depleted. In general, alunite samples provide reliable A 3 4 S a i u n j t e-pyrite temperature estimates, ca. 200-350°C, and indicate equilibrium 148 ~ i 1 i 1 1 1 1 1 r 1500 1700 1900 2100 2300 Wavelength in nm Figure 4.9. PUMA spectra for three samples of magmatic-hydrothermal alteration from the El Indio-Tambo district. Compositional variation is evident from features in the ca. 1480 nm range (enclosed in box). A) alunite-waltherite zoning (Canto Sur deposit, Tambo); B) alunite-huangite zoning (Kimberly deposit, Tambo); and C) typical example of pure alunite (Pascua deposit). 0.01 4 1 , , , , , , , , , , , , r- La Ce Pr Nd Sn- Eu Gd Tb Dy Ho Er Trr Yb Lu Figure 4.10. Range of REE data for El Indio-Pascua Belt magmatic-hydrothermal alunite, normalized to average host rock compositions for each area. Shown are range of values for disseminated, wallrock alteration (black; n=4) and gangue alteration (grey; n=9). 149 Brechita-Huantina |- • » Del Carmen Tambo: Stage 2 Stage 1 Pascua: AS AS I 8 3 4 S Figure 4.1 la. Range of 8 3 4S data for E l Indio-Pascua Belt magmatic-hydrothermal alunite and associated sulfides, where available. Data for each paragenetic stage of at Tambo and Pascua separated (details see Chap. 2 and 3). Also shown are range of depositional temperatures calculated from A 3 4 S a i u n . P y (Ohmoto and Lasaga, 1982), where possible. Figure 4.11b. Range of 8 1 O and 8D for fluids calculated in equilibrium with magmatic-hydrothermal (M-H) alunite (dark grey) from all study areas. Also shown is range of alunite fluids for Tambo Stage 1 alunite (stippled area), which has been overprinted by later steam heated or magmatic steam fluids (see discussion in text). Lines and reference fields: meteoric water line (Craig, 1961); P M W = primary magmatic water field of Taylor (1979); volcanic vapor = range of water compositions discharged from high temperature fumaroles (Giggenbach, 1992); SASF = supergene alunite S 0 4 field and SAOZ = supergene alunite O H zone as described in Rye et al. (1992). 150 fractionation between aqueous H 2S and S0 4 . Fluid compositions in equilibrium with alunite can be calculated based on fractionation data from Stoffregen et al. (1994). S 1 80H2o and 8DH2o results indicate that most alunite fluids have a dominant magmatic signature (Fig. 4.12b) and little meteoric water was involved during alteration. However, a large range in alunite 8 1 8 0 H 2 o and 8 D H 2 0 are calculated for an early stage of alteration at the Tambo deposit (Stage 1). This variation is attributed to the overprinting of previously altered wallrock by later magmatic steam or descending steam- heated fluids (details see Chap. 3). Fluid inclusions Fluid inclusions in wallrock alunite are typically too small to study (< 2u.m). Coarser- grained vein and open space filling alunite generally contain larger, 2-phase inclusions (5- lO/Um). Inclusions are typically vapor-rich, although results of necking are common. Daughter crystals are observed only in samples of gangue alunite from Pascua (AS II alteration). These include a translucent phase, probably halite or sylvite, and locally an unknown opaque phase, possibly sulfide or hematite (Fig. 4.12). Most attempts to measure inclusion homogenization or freezing temperatures were unsuccessful due to decrepitation. Fluid inclusion gas chemistry data for alunite are hmited at this time, but results from the Tambo deposit indicate that magmatic-hydrothermal fluids are characterized by high H 2 S/ S 0 2 ratios (Fig. 4.13). Measured C l contents are low and are consistent with petrographic observations and hmited fluid inclusion data from Jannas et al. (1999) that show no evidence for highly saline mineralizing fluids in this deposit. Age relations The relation of magmatic-hydrothermal alteration to gold mineralization in the E l Indio- Pascua Belt is complex. Evidence from this study and Bissig (2001) indicate that the El Indio- Tambo and Pascua-Lama districts are characterized by multiple alteration and mineralizing events. In each region however, at least one stage of magmatic-hydrothermal alteration occurred prior to the main stage of precious-metal mineralization, which is typical of most high sulfidation systems (e.g., Cooke and Simmons, 2000; Hedenquist et al., 2000). At Tambo, barren wallrock alteration occurs up to 3 m. yr. prior to mineralizing events. At Pascua, the time gap between early wallrock alteration and mineralization is at most 0.5 m. yr. 151 Figure 4.12. Fluid inclusion in magmatic-hydrothermal alunite (Pascua deposit; DDH-137a, 152.1m) with small vapor bubble and two daughter crystals. S02 1 ,o 1 0.8 0.6 0.4 0.2 0 Figure 4.13. Ternary H2S-S02-HC1 data from fluid inclusion analyses of Tambo alunite (details see Chap. 3). Range of data for magmatic-hydrothermal and magmatic steam are shown. Also shown is data for Stage 3 alunite from Tambo that is transitional between magmatic-hydrothermal and mag- matic steam (circle labeled 'transitional). 152 A second stage of alunite deposition is coeval with mineralization at both Tambo and Pascua, as well as locally at E l Indio (Brechita-Huantina and Campana veins; Jannas et al., 1999). In each of these cases alunite is intergrown with, or host to, precious-metal bearing sulfides or native gold. A similar relation is inferred for Del Carmen, where alunite is locally intergrown with enargite and pyrite, but age constraints are not available for this region. The presence of abundant syn-mineral alunite is unique to the E l Indio-Pascua Belt, and its significance is discussed in a later chapter (Chap. 6). 2. Magmatic Steam Environment The nature of magmatic steam alunite is poorly understood at this time, and documented occurrences are rare. Alunite of this type occurs in abundance at Alunite Ridge in Marys vale, Utah in coarse-grained mono-mineralic veins up to 20 metres wide (Cunningham et al., 1984), and at the Red Mountain deposit in Colorado as coarse-grained crystals cementing breccia fragments and infilling veins (Bove et al., 1990; Rye et al., 1992). Magmatic steam alunite is of particular interest in the E l Indio-Pascua Belt because of its abundance and association with late gold mineralization at the Tambo deposit (Chap. 3). Localized occurrences of magmatic steam alunite are also documented near-surface in at Pascua. Alunite of this type is thought to form from the expansion of rapidly ascending S0 2-rich magmatic vapor (Rye et al., 1992), resulting either from the sudden depressurization of the hydrothermal system (Rye, 1993), or from magma emplaced to shallow depths (Fournier, 1999; Gerlach and Casadevall, 1986). A detailed discussion on this topic is included in Chapter 3, and only a summary of physical and geochemical data from the E l Indio-Pascua Belt will be presented here. Physical characteristics Magmatic steam alunite is most abundant in the Tambo deposit, where it occurs in several forms. • Coarse-grained crystals (1 to 5 mm) occur in mono-mineralic, sub-vertical, banded veins (Fig. 4.14). Hematite is common and occurs as inclusions in, or small grains interstitial to, alunite and trace quartz. • Coarse-grained (1-3 mm) alunite also occurs in stockwork veins and cementing breccia fragments. Alunite is locally overgrown or banded with much finer-grained (<50u.m), lath- shaped alunite crystals. 153 • A third variety of alunite (Stage 3) occurs as fine-grained (<30 fxm), lath-like to thin bladed crystals in sugary, earthy masses. Native gold occurs interstitial to, or as inclusions in, alunite of this stage. At Pascua, magmatic steam alunite occurs near-surface as coarse-grained (1-3 mm) crystals lining open vugs and veins. Minor jarosite is found as overgrowths on magmatic steam alunite at both deposits, but its origins are unclear. Jarosite may form from a magmatic steam phase (Rye and Alpers, 1997), or represent an overprinting supergene event. Figure 4.14. Banded magmatic-steam alunite vein (Kimberly deposit, Tambo) in contact with wallrock (at left). 154 Geochemical characteristics Magmatic steam alunite is K-dominant with less than 0.2 atoms per formula unit (a.p.f. u.) Na substitution, on average (Fig. 4.5). Elevated Sr and P 0 4 concentrations are common and typically occur in oscillatory growth zones detectable by EDS backscatter imaging (Fig. 4.15). Results are similar to those reported for Alunite Ridge, Utah (Cunningham et al., 1986; Stoffregen and Alpers, 1987). However, Cunningham et al. (1986) also report locally elevated Ba concentrations, inversely correlated with Sr and P 0 4 zoning, in magmatic steam alunite from Alunite Ridge. Similar trends were not seen in this study and nearly all Ba is below detection (Table 4.4). REE signature: Magmatic steam alunite from both Tambo and Pascua exhibit extreme HREE fractionation and hmited enrichment of La and Ce, compared to host rock values (Fig. 4.16). Results are similar to data for magmatic-hydrothermal gangue alunite reported above. Both forms of alunite are precipitated in open spaces, likely without significant wallrock interaction. REE signatures should therefore reflect the composition of the source fluids/ vapors. Several experimental studies have demonstrated the differential partitioning of LREE into an.exsolved magmatic vapor phase (e.g., Flynn and Burnham, 1978; Reed, 1995), although crystallographic controls on REE substitution described by Schwab et al. (1990) may also account for the significant depletion of HREE in magmatic steam alunite. Stable-isotope characteristics Magmatic steam alunite in the E l Indio-Pascua Belt is characterized by near zero 8 3 4S values (Fig. 4.17a). At Tambo, only minor 8 3 4S variation is reported between different samples of banded vein alunite or between growth bands in a single vein (Table 4.5). These nearly constant 8 3 4S values should be close to the 8 3 4S of bulk sulfur in the magma (Rye et al., 1992) and likely reflect the quantitative, disequilibrium oxidation of magmatic S 0 2 (and possibly H 2 S: Cunningham et al., 1997). Alunite 8D and 8 1 8 0 S 0 4 values are near magmatic values and suggest that little to no meteoric water was present in the source fluid (Fig. 4.17b). Depositional temperatures calculated from A 1 8 0 S O 4 - O H are typically in the range of 70-200°C and are considered too low for precipitation from a magmatic vapor phase. Similar observations are noted by Rye et al. (1992) for Marys vale alunite. Isotopic results from the Tambo deposit also indicate that there is a transitional phase between magmatic-hydrothermal and magmatic steam environments (Chap. 3). Sulfur data for 155 Figure 4.15. Backscatter image showing P0 4 ± Sr zoning (light grey/white) in magmatic steam alunite (Pascua deposit). La Ce Pr Nd Sir Eu Gd Tb Dy Ho Er Trr Yb Lu Figure 4.16. Range of REE data for El Indio-Pascua Belt magmatic steam alunite (n=5; including sam- ples of Stage 3 alteration from Tambo). Data are normalized to average host rock compositions for each region. 156 Supergene/ A # • • • • • Late Stage • Alunite o Alunite: Tambo Stage 3 A Jarosite Steam Heated Magmatic Steam C « « D » 0 O Sulfides ' Magmatic-Hydrothermal - Alunite -IK 10 15 8 3 4 S (per mil) Figure 4.17a. Sulfur isotope data for magmatic steam, steam heated and supergene/late stage alunite and jarosite for the El Indio-Pascua Belt. Range of data for magmatic-hydrothermal alunite and sulfides is given for reference. -40 CD S -80 • to -120 El Indio- Pascua Belt paleo meteoric waters Magmatic. Steam FMW Tambo Stage 3 SASF- -20 -10 0 10 20 5 1 8 0 (per mil) Figure 4.17b. Range of 8 1 80 and 8D for fluids calculated in equilibrium with magmatic steam alunite (dark grey) from all study areas. Also shown is range of alunite fluids for Tambo Stage 3 alunite (white), transitional between magmatic-hydrothermal and magmatic steam environments (see discussion in text). Descriptions of lines and fields as per Fig. 4.1 lb. 157 Stage 3 alunite (8 S up to 5.5%o) are slightly heavier than values from the banded veins. This suggests incomplete sulfur isotope exchange between S 0 4 and H 2S species in some samples, possibly due to slightly longer fluid residence times. Also, in contrast to banded vein samples, calculated fluid compositions in equilibrium with Stage 3 alunite show limited mixing between magmatic and meteoric fluids. Fluid inclusions Fluid inclusions in magmatic steam alunite are typically very small (<5 u.m) and vapor- rich (this study; Jannas et al., 1999). Gas analyses indicate that they are SOydominant (Fig. 4.11), particularly in comparison to magmatic-hydrothermal alunite. H 2S and HC1 concentrations are extremely low, but abundant reduced carbon species are detected. Overall data suggest that volatiles were derived directly from the magma (further details provided in Chap. 3). Table 4.5. Variation in sulfur isotope data (in per mil) for magmatic steam alunite from a single vein. Sample# Analysis 8 S Sample* Analysis S^S KB-02a 1 -0.1 2 0.8 3 1.1 4 1.6 5 1.8 6 1.0 7 0.7 8 0.6 9 -1.1 10 -0.2 11 0.0 12 -0.4 13 -0.4 14 -0.5 15 -0.1 16 0.8 17 -0.8 18 -0.2 KB-02b 1 0.4 2 -0.2 3 0.1 4 -0.2 5 0.0 6 -0.7 7 0.2 8 0.1 9 -0.7 10 0.2 19 0.4 158 Age relations Paragenetic and geochronologic evidence indicate that magmatic steam alunite formed late in the development of magmatic-hydrothermal systems in the E l Indio-Pascua Belt. 4 0 Ar / 3 9 Ar dating of magmatic steam alunite at Pascua indicates that it is slightly younger than, or contemporaneous with, the main gold mineralizing event (Chap. 2). At Tambo, Stage 3 alunite (8.2 ± 0.2) is younger than the gold-bearing magmatic-hydrothermal Stage 2 event (8.7 ± 0.2). However, the coarse-grained banded magmatic steam alunite veins did not yield accurate 4 0 Ar / 3 9 Ar ages, despite several dating attempts. 3. Steam-heated Environment Steam-heated alteration forms from the condensation of vapour derived from an underlying hydrothermal system and the oxidation of H 2 S. Sulfuric acid is produced by the reaction: H 2S + 2 0 2 <-» H 2 S 0 4 (Schoen et al., 1974). This reaction takes place at or above the water table, and therefore at temperatures below 100°C, but fluids may descend along fractures to depths where higher temperatures are encountered, usually in the range of 90-160°C (Rye et a l , 1992). Steam-heated alteration typically occurs as tabular, blanket-like zones that are defined by the paleo-water table (Fig. 4.18). These zones will be thicker where paleo-water tables are deep, such as in arid regions or beneath topographic highs (like stratavolcanoes) in mountainous terrain (Sillitoe, 1993). There is no direct genetic relation between steam-heated alteration and precious metal mineralization, but alteration occurs above or lateral to the mineralizing system, if present. In cases of groundwater table collapse (e.g., Simmons and Browne, 1990; Ebert and Rye, 1997), alteration may overprint epithermal mineralization. Most our of knowledge of steam-heated processes is taken from low sulfidation systems and active geothermal areas. Much less information is available for steam-heated alteration above high sulfidation deposits, likely because the paleosurface is preserved in few systems (e. g., La Coipa, Paradise Peak, Yanacocha, Pierina; Sillitoe, 1999). In the El Indio-Pascua belt, steam-heated alteration occurs at upper elevations as visibly distinct, discontinuous alteration zones. In the E l Indio-Tambo region, steam-heated assemblages occur above the Kimberly, Wendy, and Canto Sur deposits and as extensive alteration zones in the nearby, barren, Azufreras and Sol Poniente areas. At Pascua, alteration forms a nearly continuous blanket-type zone above the Brecha Central area at elevations above 4950 m. 159 Figure 4.18. Steam heated alteration at Cerro Elephante, Tambo deposit. Figure 4.19. SEM micrograph of typical steam heated alteration, Pascua deposit (DDH-119, 47.7m). Scale bar = 10 um. Physical characteristics Steam-heated alteration in the El Indio-Pascua Belt consists of kaolinite, cristobalite (± trydimite), native sulfur ± trace opal. Alunite is relatively rare. Where present, it is restricted to irregular pods and lenses of fine-grained (ca. 5-50 |im), tabular, lath-shaped, pseudo-cubic and less commonly acicular crystals that appear as grey, irregular masses in thin section (Fig. 4.19). Alteration is typically distributed in massive to extremely porous, friable zones, although cross- cutting veins and disseminations of this assemblage typically extend to greater depth along fractures. Alteration in several steam-heated systems is vertically zoned (Sillitoe, 1993; Hedenquist et al., 2000) and at Pascua, an upper layer of residual, vuggy quartz grades downwards into a kaolinite and/or alunite-bearing zone. Geochemical characteristics Geochemical data for steam-heated alunite in the El Indio-Pascua Belt are limited, due to the relatively scarcity of alunite and fine-grained nature of alteration. Overall, our results indicate that alunite is nearly stoichiometric with minor Na and locally elevated Sr, Sb, Fe, and P 0 4 contents (Table 4.4). No zoning is visible. Geochemical data from other steam-heated alteration zones are also limited but generally show both alunite and natroalunite end-member compositions (Thompson and Petersen, 1995; Ebert and Rye, 1997). Other substitutions are rare. 160 REE signatures: Steam-heated alunite typically show slight HREE fractionation compared to host rock values (Fig. 4.20). Some samples exhibit a U-shaped M R E E depletion trend, similar to magmatic-hydrothermal wallrock alteration. REE fractionation patterns of E l Indio-Pascua Belt alunite are similar to those of acid-sulfate waters in Yellowstone National Park and suggest REE are complexed by sulphate and fluoride in acidic, low temperature, low salinity fluids (Lewis et al., 1998). Stable-isotope characteristics Sulfur-isotope data for steam-heated alunite are variable (Fig. 4.17a). 5 3 4S values average 4%o and are slightly to significantly heavier than data for underlying sulfides. Data reflect partial sulfur isotope exchange with H 2 S, possibly as a result of longer fluid residence times with aqueous sulfate, or the interaction of steam-heated fluids with higher temperature magmatic steam. These values are not typical of other steam-heated environments, where alunite 8 3 4S are nearly equal to H 2S derived from the hydrothermal fluid and equivalent to underlying sulfides, if present (Rye et a l , 1992). Depositional temperatures based on A 1 8 0 S O 4 - O H for steam-heated alunite average between 90° and 150°C and are consistent with data from other studies (e.g., Rye et al., 1992). Results indicate oxygen isotopic equilibrium between alunite and the fluid. Few samples give unreasonably high temperatures however, suggesting post-depositional oxygen-isotope exchange. Calculated § 1 80 H 2o §DH20 values for E l Indio-Pascua Belt alunite are shown in Figure 21. These values are much heavier than the paleo-meteoric water composition, estimated at 8D = -100 ± 10%o (B. Taylor, pers. comm.). Data are inconsistent with the predictions of Rye et al. (1992) and results from other studies (Ebert and Rye, 1997), which indicate steam-heated fluids are typically derived from meteoric waters. Our results suggest that much of the water present in the steam-heated zone in the E l Indio-Pascua Belt formed from the condensation of vapors derived from magmatic fluids. This is attributed to prolonged magmatic-hydrothermal activity in the area and a progressive lowering of the water table induced by regional erosional events (Bissig, 2001). Age relations Steam-heated acid sulfate alteration is often considered to be coincident with underlying 161 •vT\ V ^ 1 0 • Pascua (P15) - V - Pascua (P-SH) H i Tambo (T17) La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Figure 4.20. REE data for steam heated alteration, Pascua and Tambo deposits. Al l data normalized to average host rock compositions. Data below detection not plotted. Figure 4.21. Range of 5 8 0 and 8D for fluids calculated in equilibrium with steam- heated (dark grey) and late stage oxidation (stippled) alunite from all study areas. Also shown is range of supergene jarosite fluids (black) from the Pascua deposit. Descriptions of lines and fields as per Fig. 4.11b. 162 mineralization, if present (Rye et a l , 1992). Age relations in the E l Indio-Pascua Belt are more complex however. Deposits occur in regions with evidence for multiple stages of alteration and mineralization than span several million years. Steam-heated processes could have been associated with any, or all, of these events. The only reliable age data for steam-heated alteration overlying mineralization in this region is from the Tambo deposit (8.9 ± 0.4 Ma), and shows that alteration is coeval with the earliest stage of ore deposition (Stage 2). 4. Supergene Environment Supergene acid sulfate alteration results from the weathering of any sulfide-rich zone under atmospheric conditions. In the strictest sense, supergene oxidation therefore only occurs above (or locally immediately below) the water table and has no genetic relation to epithermal precious-metal mineralization. Alteration typically occurs in topographically controlled zones but acid fluids may drain downward locally along faults and open fractures, resulting in cross- cutting veins and disseminations in the weathering profile. The development of supergene alunite is also strongly controlled by climate and local environmental effects (e.g., Sillitoe, 1999). Well-developed supergene profiles are expected only in semi-arid to arid environments where relatively low water/rock ratios can result in acidic, sulfate-rich waters during weathering. Temperatures of formation typically average 20-40°C, depending on the local environment, but oxidation reactions are strongly exothermic and temperatures may reach up to 80°C (Rye et al., 1992). Supergene processes are apparent throughout the E l Indio-Pascua Belt. The effects of oxidation include ubiquitous iron oxide and hematite staining on surficial and near surface exposures of both mineralized and barren host rock. Secondary gypsum is abundant locally, particularly in surface outcrops in the Pascua district and in the Del Carmen area. Hydrous iron sulfates such as szomolnokite, voltaite, and coquimbite are found in-situ in underground exposures of sulfide-rich ore in the Pascua area and also develop rapidly on the surface of drill core once exposed to surficial atmospheric conditions, and particularly to added water. In contrast, definitive occurrences of supergene acid sulfate alteration are rare. Supergene jarosite is recognized at the Pascua district (Chap. 2) and localized occurrences of alunite-jarosite are also considered supergene in origin, based on paragenetic characteristics (described below). Thin veinlets and disseminations of porcellaneous alunite ± late jarosite are 163 common, but textural and isotopic evidence has shown that alteration is not of typical supergene origin (Chap. 2, 3). Instead, alteration ('late stage alteration') is thought to have occurred during the final stages of the hydrothermal systems, from mixing of magmatic and meteoric fluids at temperatures ca. 80° to 150°C. These assemblages can rarely be distinguished from true supergene alteration on the basis of field or textural characteristics, unless cross-cutting relations are evident. The distinction may be of importance in exploration however, since the higher temperatures of late stage alteration and involvement of magmatic fluids have a greater potential for remobilization or enrichment of hypogene precious-metal mineralization. At Pascua for example, late stage acidic, moderate temperature fluids may have been responsible for Ag enrichment at high elevations in the Brecha Central area (A. Chouinard, pers. comm.). Characteristics of both late stage and supergene alteration are described in the following section, and any differences between the two assemblages are emphasized. Physical characteristics As mentioned, late stage and supergene acid sulfate alteration are difficult to distinguish on a textural basis. Late stage alteration throughout the E l Indio-Pascua Belt occurs as chalky, porcellaneous masses, fine powdery aggregates, fracture coatings, and thin, cryptocrystalline veinlets that cross-cut mineralization and all other alteration types (Fig 22). Alunite is common, and occurs as fine- to medium-grained (10-50 >um), tabular to bladed crystals, often intermixed with quartz. Locally, alunite is overgrown or cross-cut by fine-grained jarosite, with ubiquitous iron oxides and common scorodite. This later assemblage is considered supergene in origin. Rare alunite is present and typically occurs as fine-grained (0.5-5 urn) pseudo-cubic crystals (Fig. 4.23a), although grains up to 30|Lim are found locally. Anhedral, lath-like to rhombohedral forms are rare. Pseudocubic to subhedral grains of woodhouseite-svanbergite composition are also observed in thin veinlets cross-cutting argillically altered rock in the Tambo deposit (Kimberly Oueste), mtermixed with kaolinite and pseudo-cubic alunite grains (Fig. 4.23b). Based on these relations, the APS grains are considered supergene in origin. Similar observations have been reported in other supergene environments (e.g., Stoffregen and Alpers, 1987, Dil l et al., 1997). Geochemical characteristics Only qualitative geochemical data is available for supergene alunite in the E l Indio- 164 Figure 4.23. (a) top left. SEM micrograph of pseudocubic alunite crystals (grey) showing oscillatory compositional zoning between alunite and more Na-rich compositions. Jarosite (white) occurs interstitial to alunite grains. Scale bar = 5um. (b) top right. SEM micrograph of supergene APS (light grey) with alunite (grey) and kaolinite (dark grey). Scale bar = lOurn. (c) bottom. SEM micrograph of veinlet with supergene alunite overgrown by jarosite and alunite-jarosite solid solution (Pascua deposit, 4680 level). Scale bar = 20um. 165 Pascua Belt, due to the fine-grained nature of alteration and relatively rare occurrence of alunite in the supergene assemblage. Alunite is typically K-dominant (Table 4.4), and rarely exhibits very fine oscillatory zoning between K-dominant and more Na-rich compositions. Solid solution between alunite and jarosite end-member compositions also occurs in thin veinlets at the Pascua deposit (Fig. 4.23c). Geochemical data for late stage alteration (Fig. 4.5) indicate that average alunite compositions are K-rich in comparison to higher temperature occurrences (at 95% confidence level, based on t-test statistics: Appendix A). Trace element concentrations are extremely variable with locally elevated Ca, Sr, P 0 4 ± REE and trace metal concentrations (Pb, Cu, Zn ± Ba) that are likely derived from hypogene sulfides. Data are consistent with other studies that report variable supergene alunite compositions, largely depending on the contents of precursor sulfides and the availability of ions in the weathering profile (Arribas et al., 1995; Stoffregen and Alpers, 1987; ibid, 1992, Thompson and Petersen 1995, Dil l et al., 1997). REE signatures: REE data are available only for late vein alunite since sufficient material of supergene alunite could not be separated. Results are given in Figure 24 and show variable patterns of REE fractionation. One sample exhibits strong depletion in HREE. The other two samples have concave-shaped patterns of M R E E enrichment and also show significant depletion in HREE. The pattern of M R E E enrichment is opposite to the U-shaped pattern observed in magmatic-hydrothermal wallrock alteration and steam-heated samples (Figs. 10 and 16), although to a lesser degree. We suggest that REE fractionation is affected by temperature variations and differences in the availability of complexing agents with increasing meteoric water contribution (Wood, 1990; Lottermoser, 1992), but more data is required to fully understand these processes. Stable-isotope characteristics Stable-isotope data are available only for samples of late stage alunite (Fig. 4.25) and one sample of supergene jarosite from Pascua. The alunite is characterized by low S^S (typically less than 6%o) and 5 1 8 0 0 H (less than 10%o) values. Data generally do not fall within the reference fields defined by Rye et al. (1992) for supergene alunite. E l Indio-Pascua Belt alunite A 1 8 O S O 4 - O H values are all positive and therefore also not consistent with a supergene origin (e.g., Rye et al., 1992). Several samples give A1 80SO4-OHtemperatures between 80° to 140 °C. These values are considered reasonable for late stage oxidation in the presence of 166 c o o LU Ol 0.01 -\7 \ - • • • Pascua - Frontera zone (P17) — T a m b o - Canto Sur (T02a) —a Tambo - Kimberly (T11) La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Figure 4.24. REE data for late stage acid sulfate alteration; Pascua and Tambo deposits, normalized to host rock compositions. Data below detection limits not shown. All data 20 15 £ i i • i •-• - T # alunite 8 1 8 O S 0 4 1 © alunite 8 1 8 0 O H 1 • jarosite 8 1 8 0 S 0 4 f T j - • jarosite 8 1 8 0 O H i j i - i 1 T • 6 i j _ Sulfides C 0 -5 0 5 10 8 3 4 S (per mil) Figure 4.25. Sulfur and oxygen isotope data for El Indio-Pascua Belt late stage alunite and supergene jarosite. 818Oso4 and 5 I 8 O O H data for each sample joined by lines. Range of 834S data for hypogene sulfides given for reference. 167 magmatic fluids. Other A I 8 OSC -4-OH values are grossly out of equilibrium and suggest non- equilibrium or kinetic controls on alunite 8 1 8 0 S 0 4 values. Calculated alunite 5 1 80H2o and 8D H 2o data indicate variable mixing between magmatic and meteoric fluids (see Fig. 4.21). Jarosite 8 3 4S data from the Pascua deposit are slightly enriched relative to sulfides in the deposit. Jarosite 8 1 8 0 data generally fall within the reference fields defined by Rye and Alpers (1997) for supergene environments. However, 8 1 8 0 0 H data are slightly heavier than the Supergene Jarosite OH Zone (SJOZ) and may indicate formation from partially exchanged meteoric waters. Jarosite calculated fluid compositions are nearly identical to the estimated composition of paleo-meteoric waters (see Fig. 4.21). Age relations Results from the E l Indio-Pascua Belt illustrate that the timing of supergene or oxidation processes within magmatic-hydrothermal systems can be complicated. Both supergene and late stage acid sulfate assemblages post-date alteration and mineralization related to the magmatic- hydrothermal system. 4 0 Ar / 3 9 Ar ages reported for the Pascua deposit (Chap. 2; Bissig, 2001) indicate supergene jarosite (7.98 ± 0.43 Ma) formed within 1 Ma of the main mineralizing event. Similarly, late stage alunite (± jarosite) veins from the Tambo deposit (Canto Sur pit: 7.25 ± 0.14 Ma) formed ca. 1 Ma after the main period of mineralization in the Kimberly and Wendy areas, although these veins may be contemporaneous with Au mineralization in this area (Bissig, 2001). DISCUSSION: IMPLICATIONS FOR E X P L O R A T I O N Recognition of the different types of acid sulfate alteration is critical for effective exploration in high sulfidation systems, particularly in poorly exposed terrains. In the field, the identification of these assemblages will rely mostly on the spatial distribution of alteration, macroscopic textures, and mineral assemblages. Rigorous field work and careful mapping are required. Particular attention must be paid to local features such as climate (both current and paleo- conditions), topography, lithology, and structure that can affect both the distribution and nature of alteration zones. For example, both alteration and mineralization in high sulfidation systems are strongly controlled by permeability, which can be generated by either structural, 168 hychothermal, or lithological processes. The relative importance of each will vary with the local environment (Sillitoe, 1993). The distribution of alteration can be further complicated by superimposed alteration events, particularly in areas of groundwater table collapse (Simmons and Browne, 1990; Ebert and Rye, 1997). Field mapping in epithermal systems can be aided by in-situ SWIR analyses, which are particularly helpful for identifying fine-grained mixtures and subtle changes in mineralogy (e. g., kaolinite-dickite-pyrophyllite) that can be difficult to distinguish in hand specimen. In some cases, SWIR techniques can also be used to detenriine the chemistry of alunite-group minerals (Thompson et al., 1999) and may aid in the identification of higher temperature alteration assemblages. In areas with multiple stages of alteration, field work should be combined with careful petrographic observations and simple X R D or SEM-EDS analysis, in order to better define paragenetic relations and mineral chemistry. Detailed isotopic or fluid-inclusion studies can provide additional information regarding the nature of the source fluids and physio- chemical environment of deposition. Successful use of such complementary information can distinguish higher temperature, magmatic-hydrothermal alteration zones that have a direct spatial (and possibly temporal) relation to potential precious-metal mineralization. Exploration in the El Indio-Pascua Belt The E l Indio-Pascua Belt is a unique region which hosts several world-class Au (Cu-Ag) deposits and extensive areas of hydrothermal alteration. High sulfidation systems within this belt have many features in common with other deposits, although several differences can be identified. For example; • There is only limited evidence for syn-ore magmatic activity (Bissig, 2001). • Deposits and prospects are not associated with major volcanic edifices. • The largest (most economic) deposits in this region have evidence for multiple, superimposed alteration and mineralizing events. • A shallow erosional level has preserved many features of the near paleo-surface. • Abundant alunite is coeval with, and host to, precious-metal mineralization. Results from this study have highlighted the complexity of relations between acid sulfate assemblages of different origin. Effective exploration in this region (as in other regions) must therefore focus on identifying ore-related, hypogene alteration assemblages. Several key considerations are outlined below. 169 1. Magmatic-hydrothermal alteration can be differentiated from lower temperature steam- heated and supergene occurrences by a combination of factors which include: • Mineral associations, and particularly a direct spatial and genetic relation of alunite with sulfides or high-temperature alteration minerals (e.g., dickite, pyrophyllite, diaspore, zunyite). • Alunite crystal habit: coarse-grained crystals with a bladed to tabular habit. • Variable alunite geochemistry, specifically elevated Ba contents and complex or oscillatory geochemical zoning. 2. Differentiating between barren wallrock alteration and potentially ore-bearing (gangue) alunite is much more difficult. In the El Indio-Pascua Belt, mineralization and associated gangue is typically strongly structurally controlled and exploration should therefore focus on determining favorable structural fluid conduits - after magmatic-hydrothermal alteration zones have been delimited. 3. Magmatic steam alunite may also host gold ore, but cannot reasonably be targeted by regional-scale exploration programs. Alunite of this type is extremely variable in texture and physical characteristics, and it may be difficult to distinguish visually from magmatic- hydrothermal and even steam-heated assemblages. Detailed paragenetic, petrographic, and isotopic studies would be required to confirm alunite of magmatic steam origin. 4. Steam-heated alteration overlies precious metal mineralization in all of the areas included in this study, with the exception of Salitrales. Alteration is coeval with gold deposition in the Tambo deposit, and possibly in the Pascua area as well. However, large areas of barren steam-heated alteration also occur to the south of E l Indio-Tambo. Without an in-depth examination of these barren alteration zones, we have no means to distinguish potential 'ore-related' steam-heated alteration. 5. The economic feasibility of deposits in the El Indio-Pascua Belt is not dependent on supergene or late-stage oxidation processes. However, enrichment or remobilization of precious-metal assemblages can occur on a local scale. Applications to Exploration World-wide While the above features apply specifically to the El Indio-Pascua Belt, general exploration criteria can be applied to high sulfidation systems world-wide. A decision-tree 170 (Fig. 4.26) highlights the use of alunite and the critical observations that must be made to explore effectively for deposits of this type. Emphasis is placed on distinguishing between types of acid sulfate alteration. Reference is made to Tables 4.2 and 4.3, which summarize characteristics of alunite and associated alteration assemblages in the El Indio-Pascua Belt. Many of these features may apply to high sulfidation systems in general, but specific alunite characteristics, and particularly alunite chemical compositions, will vary in different geological settings. Alunite may be of particular use in two common exploration scenarios; 1) regions with poor outcrop, or 2) regions with widespread alteration but no obvious mineralization. In both cases, a combination of paragenetic, textural, or geochemical features of alunite could help by first confirming that alteration is magmatic-hydrothermal in origin, and secondly, by providing focus to exploration activities by identifying higher-temperature, and therefore potentially ore- proximal, assemblages. Specifically, coarse-grained alunite, associated with kaolinite or dickite and sulfides, are characteristic of a magmatic-hydrothermal origin. Complex chemical zoning and elevated Na, Ba, P, and Ca contents are also indicative of magmatic-hydrothermal alunite. These determinations could be made from simple petrographic or geochemical (SWIR, XRD) analyses. In some cases, more detailed SEM-EDS and E P M A analyses may also be useful to identify zoning and trace-element variations in alunite minerals. Figure 4.26 emphasizes the importance of the combination of techniques needed to explore successfully for high sulfidation systems. Careful field mapping, with emphasis on alteration assemblages and structural features, can be supplemented by more detailed petrographic and alunite chemical analyses to generate viable drill targets. 171 Alteration mapping distribution of alteration mineral assemblages (SWIR) alunite? 2 no. C h e c k other ev idence for H S sys tem: (e.g., strongly silicified z o n e s , breccias) . C o n s i d e r other types of systems. yes ..Type of Alteration? (see Table A.T) Steam Heated Supergene check for p resence of M H alteration below (if possible) check for remobi l ized A u (fracture coat ings, d isseminat ions) (assay) Ind ica tespa leodepth below typical bulk H S sys tems , •consider deeper types of sys tems (structurally control led ve ins / rep lacement ; porphyry deposits) •look for lower temperature a s s e m b l a g e Magmatic-Hydrothermal Unknown? poor exposure , amb iguous textures ineral assemblage? Muscovite ± pyrophyllite Dickite, kaolinite Look for assoc ia ted sulfides i v u g g y quartz zones Detailed mapping, surficial sampling (assay) Drill targets F o c u s on favorable structural sett ings (brecc ias , f rac ture networks) ± v u g g y quartz zones ± favo rab le (permeable) l i thologies Extensive alteration but no metals or obvious field relationships Poor outcrop Alunite mineralogy (Table'4,.3*) Need to identify higher temperature, potentially ore-prox imal M H alteration: • mineral zoning; e.g. changes in alunite chemistry indicating focus • ev idence for mineral ization - mineral assoc ia t ions . Figure 4.26. Decision-tree illustrating the use of alunite in mineral exploration. *Reference is made to Tables 4.2 and 4.3 that apply specifically to the E l Indio-Pascua Belt, but many features may apply to high sulfidation systems in general. 172 REFERENCES Alpers, C.N., and Brimful, G.H., 1988. Middle Miocene climatic change in the Atacama Desert, northern Chile: Evidence from supergene mineralization at La Escondida. Geological Society of America Bulletin, 100, 1640-1656. Aoki, M. , 1991. Mineralogical features and genesis of alunite solid solution in high temperature magmatic-hydrothermal systems. Journal of the Geological Survey of Japan, 277, 31-32. Aoki, M. , Comsti, E.C., Lazo, F.B., and Matsuhisa, Y., 1993. Advanced argillic alteration and geochemistry of alunite in an evolving hydrothermal system at Baguio, northern Luzon, Phillipines. Resource Geology, 43, 155-164. Arehart, G.B. and O'Neil, J.R., 1993. D/H ratios of supergene alunite as an indicator of paleoclimate in continental settings: Climate change in continental isotope records. Geophysics Monographs, 78, 277-284. Arribas, A., Jr., Cunningham, C.G., Rytuba, J.J., Rye, R.O., Kelly, W.C., McKee, E.H., Podwysocky, M.H., and Tosdal, R.M., 1995. Geology, geochronology, fluid inclusions, and stable-isotope geochemistry of the Rodalquilar Au alunite deposit, Spain. Economic Geology, 90, 795-822. Beane, R.E., 1988. Hydrothermal alteration and mineralization at the Nevada prospect, Chile. Unpublished company report, Barrick Corp. Bennet, R.E., 1995. A combined hydrothermal alteration study and exploration target evaluation of the Tambo mining district, El Indio mineral belt, Region IV, Chile. Internal report for Compania Minera San Jose Ltda. Bethke, P.M., 1984. Controls on base- and precious-metal mineralization in deeper epithermal environments. U.S. Geological Survey Open File Report 84-890, 40 p. Bird, M.I., Chivas, A.R., and McDougall, I., 1990. An isotopic study of surficial alunite in Australia: 2. Potassium-argon geochronomogy. Chemical Geology, 80, 133-145. Bissig, T., 2001. Metallogenesis of the Miocene El Indio-Pascua gold-silver-copper belt, Chile/ Argentina: Geodynamic, geomorphological and petrochemical controls on epithermal mineralization. Unpublished Ph.D. thesis, Queen's University, Kingston, Canada. Bove, D.J., Rye, R.O., and Hon, K., 1990. Evolution of the Red Mountain alunite deposits, Lake City, Colorado. US Geological Survey Open File Report, 90-0235, 29 p. Cooke, D.R. and Simmons, S.F., 2000. Characterisitcs and genesis of epithermal gold deposits. 173 In Hagemann, S.G. and Brown, P.E., eds., Gold 2000. Reviews in Economic Geology, vol. 13., 221-244. Chouinard, A. and Williams-Jones, A.E., 1999. Mineralogy and petrography of ore and alteration assemblages at the Pascua Au-Ag-Cu deposit, Chile. Internal Report for Barrick Gold Corp. Crowe, D.E., Riciputi, L.R., Bezenek, S., Ignatiev, A., 2001. Oxygen isotope and trace element zoning in hydrothermal garnets: Windows into large-scale fluid-flow behavior. Geology, 29, 479-482. Cunningham, C.G., Rye, R.O., Bethke, P.M., Logan, M.A.V., 1997. Vein alunite formed by degassing of an epizonal stock. EOS, Transactions 78, F746. Cunningham, C.G., Rye, R.O., Steven, T.A., and Mehnert, H.H., 1984. Origins and exploration significance of replacement and vein-type alunite deposits in the Marysvale volcanic field, west central Utah. Economic Geology, 79, 50-71. Di l l , H.G., Bosse, H.R., Henning, K.H., Fricke, A., Ahrendt, H., 1997. Mineralogical and chemical variations in hypogene and supergene kaolin deposits in a mobile fold belt; the Central Andes of northwestern Peru. Mineralium Deposita, 32, 149-163. Ebert, S.W. and Rye, R.O., 1997. Secondary precious metal enrichment by steam-heated fluids in the Crowfoot-Lewis hotspring gold-silver deposit and relation to paleoclimate. Economic Geology, 92, 578-600. Fournier, 1999. Hydrothermal processes related to movement of fluid from plastic into brittle rock in the magmatic-hydrothermal environment. Economic Geology, 94, 1193-1212. Flynn, F.R., and Burnham, C.W., 1978. An experimental determination of rare earth partition coefficients between a chloride-containing vapor phase and silicate melts. Geochimica et Cosmochimica Acta, 42, 685-701. Gallardo, V., 1995. Proyecto Del Carmen; Provincia de San Juan, Argentina. Internal Report for Barrick Gold Corporation. Gerlach, T.M., and Casadevall, T.M., 1986. Fumarole emissions at Mount St. Helens volcano, June 1980 to October, 1981: Degassing of a magma-hydrothermal system. Journal of Volcanology and Geothermal Research, 28, 141-160. Hedenquist, J.W, Arribas, A., Jr., and Gonzalez-Urien, E., 2000. Eploration for epithemal gold deposits. In Hagemann, S.G. and Brown, P.E., eds., Gold 2000. Reviews in Economic Geology, vol. 13., 245-277. Hedenquist, J.W., Arribas, Jr., A., and Reynolds, T.J., 1998. Evolution of an intrusion-centered 174 hydrothermal system: Far Southeast-Lepanto porphyry and epithermal Cu-Au deposits, Phillipines. Economic Geology, 93, 373-404. Henley, R.W., and McNabb, 1978. Magmatic vapor plumes and ground-water interaction in porphyry copper emplacement. Economic Geology, 73, 1-20. Hemley, J.J., and Jones, W.R., 1964. Chemical aspects of hydrothermal alteration with emphasis of hydrogen metasomatism. Economic Geology, 59, 538-569. Hemley, J.J., Hostetler, P.B., Gude, A.J. , and Mountjoy, W.T., 1969. Some stability relations of alunite. Economic Geology, 64, 599-612. Holland, H.D., 1965. Some applications of thermodynamic dta to problems of ore deposits, II. Mineral assemblages and the composition of ore-forming fluids. Economic Geology, 60, 1101-1166. Jambor, J.L., 1999. Nomenclature of the alunite supergroup. Canadian Mineralogist, 37, 1323- 1341. Jamveit, B., 1999. Crystal growth and intracrystalline zonation pattern in hydrothermal environments. In Jamtveit, B. and Meakin, P., eds., Growth, Dissolution and Pattern Formation in Geosystems, 65-84. Jannas, R.R., Beane, R.E., Ahler, B.A., and Brosnahan, D.R., 1990. Gold and copper mineralization at the El Indio deposit, Chile. Journal of Geochemical Exploration, 36, 233- 266. Jannas, R.R., Bowers, T.S., Petersen, U., Beane, R.E., 1999. High-sulfidation deposit types in the El Indio District, Chile. In Skinner, B.J. ed., Geology and ore deposits of the Central Andes, SEG special pub. #7, p.27-59. Kesler, S.E., Russell, N., Seaward, M., Rivera, J., McCurdy, K, Cumming, G.L., and Sutter, J. F., 1981. Geology and geochemistry of sulfide mineralization underlying the Pueblo Viejo gold-silver oxide depoist, Dominican Republic. Economic Geology, 76, 1096-1117. Knight, J.E., 1977. A thermochemical study of alunite, enargite, luzonite, and tennantite deposits. Economic Geology, 72, 1321-1336. Losada-Calderon, A.J. , and McPhail, D.C., 1996. Porphyry and high-sulfidation epithermal mineralization iin the Nevados del Famatina Mining District, Argentina. In Camus, F., Sillitoe, R.H., and Petersen, R., eds., Andean copper deposits: New discoveries, mineralization, styles, and metallogeny. Society of Economic Geologists Special Publication No. 5, 91-118. Lottermoser, B.J., 1992. Rare earth elements and hydrothermal ore formation processes. Ore 175 Geology Reviews, 7, 25-41. Meyer, C.A. and Hemley, J.J., 1967. Wallrock alteration. In Baines, H.L., ed., Geochemistry of Hydrothermal Ore Deposits. New York, Holt, Rinehart, and Winston, 166-235. Noriega, J. and Perez, L., 1997. Proyecto Del Carmen - Informe temporada 1997. Internal report for Barrick Gold Corporation. Perello, J.A., 1994. Geology, porphyry Cu-Au, and epithermal Cu-Au-Ag mineralization of the Tombulilato district, North Sulawesi, Indonesia. Journal of Geochemical Exploration, 50, 221-256. Petersen, E.U., and Thompson, A.J.B., 1992. Analysis of alunite and other thermally sensitive minerals. Microprobe Analyst (Newsletter for Laboratory for Electron Microprobe Analysis, University of Utah), April 1992. 2-5. Prieto, M., Feernandez-Gonzalez, A., Putnis, A., and Fernandez-Diaz, L., 1997. Nucleation, growth, and zoning phenomena in cystallizing (Ba,Sr)C0 3, Ba(S0 4 ,Cr0 4 ) , (Ba,Sr)S04, and (Cd,Ca)C03 solid solutions from aqueous solutions. Geochimica et Cosmochimica Acta, 61, 3383-3397. Putnis, A., Prieto, M., and Fernandez-Diaz, L., 1995. Fluid supersaturation and crystallization in porous media. Geological Magazine, 132, 1-13. Reed, M.J., 1995. Distribution of rare earth elements between aqueous fluids and granitic melt. Ph.D. thesis, University of Maryland at College Park. Ripmeester, J.A., Ratcliffe, C L , Dutrizac, J.E., and Jambor, J.L., 1986. Hydronium ion in the alunite-jarosite group. Canadian Mineralogist, 24,435-447. Rye, R.O., 1993. The evolution of magmatic fluids in the epithermal environment: The stable- isotope perspective. Economic Geology, 88, 733-753. Rye, R.O., and Alpers, C.N., 1997. The stable-isotope geochemistry of jarosite. USGS Open File Report 97-88., 28 p. Rye, R.O., Bethke, P.M., and Wasserman, M.D., 1992. The stable-isotope geochemistry of acid sulfate alteration. Economic Geology, 87, 225-262. Rye, R.O, and Stoffregen, R.E., 1995. Jarosite-water oxygen and hydrogen isotope fractionations: preliminary experimental data. Economic Geology, 90, 2336-2342. Sandeman, H.A., Archibald, D.A.; Grant, J.W.; Villneuve, M.E. and Ford, F.D., 1999. Characterisation of the chemical composition and 4 0 Ar - 3 9 Ar systematics of intralaboratory standard MAC-83 biotite. In Radiogenic Age and Isotopic Studies: Report 12; Geological Survey of Canada, Current Research 1999-F, 13-26. 176 Schwab, R.G., Herold, H, Gotz, C. and de Oliveira, N.P., 1990. Compounds of the crandallite type: Synthesis and properties of pure rare earth element phosphates. Neues Jahrbuch fuer Mineralogie, 6, 241-254. Schoen, R., White, D.E., and Hemley, J.J., 1974. Argillization by descending acid at Steamboat Springs, Nevada. Clays and Clay Minerals, 22, 1-22. Siddeley, G. and Areneda, R., 1985. Gold-silver occurences of the El Indio belt, Chile. Geology of the Andes and its relation to mineral and energy resources, symposium, Santiago, Chile, 18 p. Sillitoe, R.H., 1993. Epithermal models: Genetic types, geometric controls, and shallow features. In Kirkham, R.V., Sinclair, W.D., Thorpe, R.I., and Duke, J.M., eds., Mineral Deposit Modelling. Geological Association of Canada, Special paper 40, 403-417. Sillitoe, R.H., 1999. Styles of high sulfidation gold, silver and copper mineralization in the porphyry and epithermal environments. PacRim '99. Bali, Indonesia, 10-13 October, Proceedings, p. 29-44. Sillitoe, R.H., McKee, E.H., and Vila T., 1991. Reconnaissance K-Ar geochronology of the Maricunga gold-silver belt, northern Chile. Economic Geology, 86, 1261-1270. Simmons, S.F., and Brown, P.R.L., 1990. Mineralogie, alteration, and fluid inclusion studies of epithermal gold-bearing veins at the Mt. Muro prospect, central Kalimantan (Borneo), Indonesia: Geology, geochemistry, origin, and exploration. Journal of Geochemical Exploration, 35, 63-104. Steiger, R. H. and Jager, E., 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359-362. Stroffregen, R., 1987. Genesis of acid-sulfate alteration and Au-Cu-Ag mineralization at Summitville, Colorado. Economic Geology, 82, 1575-1591. Stoffregen, R.E. and Alpers, C.N., 1987. Woodhouseite and svanbergite in hydrothermal ore deposits: Products of apatite destruction during advanced argillic alteration. Canadian Mineralogist, 25, 201 -211. Stoffregen, R.E. and Alpers, C.N., 1992. Observations on the unit-cell dimensions, H 2 0 contents, and 5D values of natural and synthetic alunite. American Mineralogist, 77, 1092- 1098. Stoffregen, R.E., Alpers, C.N, and Jambor, J.L., 2000. Alunite-jarosite crystallography, thermodynamics, and geochronology. In Alpers, C.N., Jambor, J.L., and Nordstrom, D.K., 177 eds., Sulfate Minerals; Crystallography, Geochemistry, and Environmental Significance. Reviews in Mineralogy and Geochemistry, vol. 40, 454-480. Stoffregen, R.E., and Cygan, G., 1990. An experimental study of Na-K exchange between alunite and aqueous sulfate solutions. American Mineralogist. 75, 209-220. Stoffregen, R.E., Rye, R.O., and Wasserman, M.D., 1994. Experimental studies of alunite: I. 1 8 0 - 1 6 0 and D-H fractionation factors between alunite and water at 250-450°C. Geochimica et Cosmochimica Acta, 58, 903-916. Terakado, Y. , and Fujitani, T., 1998. Behavior of rare earth elements and other trace elements during interactions between acidic hydrothermal solutions and silicic volcanic rokcs, southwestern Japan. Geochimica et Cosmochimica Acta, 62, 1903-1917. Thompson, A.J.B, 1992. Alunite compositions and textures: Relationships to precious metal mineralization. In New Developments in Lithogeochemistry, Mineral Deposit Research Unit Short Course #8, 20-21 February. Thompson, A.J.B., Hauff, P.L., and Robitaille, A.J. 1999. Alteration mapping in exploration: Application of short-wave infrared (SWIR) spectroscopy. Society of Economic Geologists Newsletter, 39,, October 1999. Vasconcelos, P.M., 1999. 4 0 Ar / 3 9 Ar geochronology of supergene processes in ore deposits. In Lambert, D.D., Ruiz, J., eds., Application of radiogenic Isotopes to Ore Deposits Research and Exploration. Reviews in Economic Geology vol 12, 73-113. Vasconcelos, P.M., Brimhall, G.H., Becker, T.M., and Renne, P.R., 1994. 4 0 Ar / 3 9 Ar analysis of supergene jarosite and alunite: Implications to the paleoweathering history of the western USA and West Africa. Geochimica et Cosmochimica Acta, 58, 401-420. Wasserman, M.D., Rye, R.O., Bethke, P.M., and Arribas, Jr., A., 1992. Methods for separation and total stable-isotope analysis of alunite. Open file report 92-9, US Department of the Interior, Geological Survey. Wood, S.A., 1990. The aqueous geochemistry of the rare-earth elements and yttrium 1. Review of available low-temperature data for inorganic complexes and the inorganic REE speciation of natural waters. Chemical Geology, 82,159-186. 178 Chapter 5 ALUNITE-NATROALUNITE STABILITY RELATIONS I N T R O D U C T I O N Acid-sulfate alteration is defined as the assemblage alunite, quartz ± clays (kaolinite, dickite, pyrophyllite) and pyrite (Hemley and Jones, 1964; Meyer and Hemley, 1967). Empirical evidence and thermodynamic models indicate that acid-sulfate alteration forms under conditions of low pH and highly oxidized fluid chemistry (Holland, 1965; Henley and McNabb, 1978; Stoffregen, 1987), but experimental data on the relation of alunite and alunite-group minerals to other alteration minerals is limited. The only data available on stability relations for minerals in the K20-Ai203-Si02-H20-S03 system are provided by Hemley et al. (1969), who examined reactions between alunite, kaolinite/pyrophyllite, muscovite, and K-feldspar. Likewise, little information is available regarding occurrences of natroalunite and the nature of the alunite-natroalunite solid-solution. The substitution of N a + for K + is the most common substitution in natural alunite (Stoffregen et al., 2000) and the temperature dependence of this reaction is of particular interest in mineral exploration. Recent studies (e.g., Aoki, 1991; Aoki et al., 1993; Thompson, 1992; Arribas et al., 1995) suggest that higher Na contents are characteristic of higher temperature environments of formation, and thus may indicate ore- proximal alteration zones. Alunite from lower temperature environments has been reported to have more K-rich compositions (e.g., Arribas et al., 1995; A.J.B. Thompson, unpub. data), although few studies suggest that a range of K- and Na-rich compositions can occur (e.g., Stoffregen and Alpers, 1992; Zotov, 1971; Reyes, 1990). Experimental data for the alunite- natroalunite solid-solution is only available from Stoffregen and Cygan (1990), who developed a mixing model for the alunite-natroalunite alkali exchange reaction at 250° to 450°C. Their data are consistent with observations that higher Na contents are favored at higher temperatures. However, many other factors are likely to affect K/Na substitution in naturally occurring alunite, including host-rock composition and nature of source fluids, the latter of which has not been addressed to date. 179 Based on the scarcity of information available regarding equilibrium relations of alunite, natroalunite, and associated alteration minerals, the goals of this chapter are: 1. To compare the stability relations of alunite and natroalunite with associated alteration minerals in the K20/Na20-Al 2 03-Si02-S03-H20 systems. This comparison also provides a means to evaluate the experimental results and thermodynamic calculations of Hemley et al. (1969) in light of more recent thermodynamic data. 2. To predict the effect of fluid composition, and K/Na fluid ratios in particular, on the composition of the alunite-natroalunite solid-solution. The range of alunite-natroalunite solid-solution in the El Indio-Pascua Belt, and controls on Na substitution, are also discussed. STABILITY RELATIONS IN THE K J O / N A J O - A L J O J - S I O ^ - I ^ O - S O T , S Y S T E M The stability relations of alunite and associated alteration minerals (kaolinite, pyrophyllite, muscovite, and K-feldspar) were first examined experimentally by Hemley et al. (1969). These experiments were conducted only for the K20 -A l 2 03 -S i02 -H 2 0 -S03 system at 200°, 300 °, and 380 °C and 15,000 psi total pressure. Experimental results were expressed in terms of molar concentrations of H2SO4 and K 2 S O 4 and were transformed into activity-activity plots on the basis of dissociation calculations (Hemley et al., 1969). Invariant points for the reactions were also evaluated independently of the experimental results using thermodynamic data for all phases. Thermodynamic properties (see Table 3 in Hemley et al., 1969) for most minerals were evaluated by Hemley et al. (1969) using high-temperature thermodynamic data from Robie and Waldbaum (1968) or by extrapolating low-temperature data according to the heat-capacity method of Criss and Cobble (1964). Standard free-energy data for alunite were calculated from the experiments of Shomate et al. (1946). The calculated invariant points of Hemley et al. (1969) did not agree with their experimental results. However, more recent thermodynamic data are now available for several minerals in the K ^ O - A ^ C V S i C V ^ O - S O s system (Helgeson et al., 1978). These data, and data for all Na-equivalent minerals, are used to construct updated activity-activity diagrams and to compare the equilibrium conditions of formation for alunite- and natroalunite-bearing acid- sulfate assemblages. 180 Methods and Results: Balanced reactions among minerals in the K 20-Al203-Si02-H 20-S03 system (i.e. alunite, kaolinite, pyrophyllite, muscovite, and K-feldspar) are given in Appendix D. Equations involving natroalunite and all of the Na analogues of the K mineals were also determined (Appendix D). Standard molal Gibbs free energies for each reaction were calculated using SUPCRT92, a software package which permits the calculation of thermodynamic properties for minerals, gases, aqueous species, and reactions at specified P and T conditions (Johnson et al., 1992). The SUPCRT92 thermodynamic database was modified to include data for natroalunite from Stoffregen and Cygan (1990). A l l reactions were determined at 200°, 300 0 and 380 °C, and 1000 bars in the presence of quartz. These P-T conditions were chosen to allow comparison with the experiments of Hemley et al. (1969). Mineral stability fields at various P-T conditions were determined by plotting log(aX+)2(c7SC>4) versus log(aH+)2(aSC>4), where X = either K or Na (Hemley et al., 1969; Stoffregen, 1987), and aS04 is the thermodynamic activity of SO4 in fluid (and likewise for all other species). Activity diagrams calculated for the system K 2 0-Al 2 03 -Si0 2 -H 2 0-S03 are shown in Figure 5.1. These plots indicate that: • Alunite is stable at increasingly higher pH values with increasing temperature, at fixed <Zso4- « The activity of K + required at equilibrium among alunite, muscovite, and kaolinite decreases with increasing temperature (at fixed flso4)- • With increasing temperature, the stability field of muscovite shrinks relative to those of kaolinite and K-feldspar. • Equilibrium relations are relatively insensitive to pressure (Fig. 5.2), within the range of epithermal conditions. Changes in pressure (at this scale) affect only the absolute position, but not the relations of the stability fields. Activity diagrams calculated for the system Na 2 0-Al 2 03-Si02-H20 -S03 are shown in Figure 5.3. These plots indicate that: • The effects of temperature and pressure in the Na20 and K2O systems are similar. Natroalunite is also stable at increasingly higher pH values and lower Na activities with increasing temperature, at fixed aso4- 181 ( . ^OSe^UxeJBoi (. 3"ose) 3( +»e)6o| C O C o o CO c o T 3 ( 0 c3 *2 o C O 2* 1) > o •8 « I S © 3 CN O O N JT N O O O N 6 ^ O >, O « <C > N O — 23 •c '3 <u ll > CS •a > cS -*-» o c cn ca >-. u "> « D c - 1 0 u On | B 2 2 Sb '3 1 & B D co -*-» co II r3 a * J . o o O H o in O N ,0 ' " H C3 C 1-1 CU CL) hS l - l . M H s a 6 0 £2 CS o CD 1 1 UU o •a .2 - n o II a, T3 (. 3 t 'OSB) z( +>lB)6o| 182 -2 -12 I i i i i i I 8 9 10 11 12 13 14 -log(aH+)2(aS042") Figure 5.2. Effect of pressure on the stability of alunite relative to kaolinite and muscovite in the system K20-Al203-Si02-H20-S03.. 183 in CO JO o o o o oo CO E a> "&> >. IS) ro £ < £ , i / = / / / o / ' c ? / , Q _ / : ^ I JZ Q. O k_ >> O co to I CO o o O CO JO o o o o o 00 £ 0) (.3t7OSe)^(+e/ve)6o| o oo CO I co o 8 E CN " <*> a C/3 73 q § o s 2 a «2 g a -a T3 " Q (.2"OSe)jj(+BA/B)Bo| CO JO o o o o o CM E 3 >. CO z cd (U X) O O O o (.^OSe^+e/veJBoi 184 • Significantly higher aqueous Na activity is required for equilibrium between natroalunite and kaolinite/pyrophyllite than for K + activity in the equivalent K 2 0 system (at fixed flso4)- Slightly higher pH values are also required at equilibrium. Paragonite is uncommon in natural systems and to our knowledge, has not been reported to occur in high sulfidation environments. A much more accurate representation of Na-bearing acid sulfate alteration is the assemblage natroalunite + kaolinite/pyrophyllite + muscovite/illite. We can approximate this assemblage in the K 2 0-A l 2 03-S i02-H 2 0-S03 system by using the mixing model of Stoffregen and Cygan (1990) for the alunite-natroalunite solid-solution. The activity of various alunite compositions (ranging from XKain,ain = 0.2 to 1.0, where X is the mole fraction of alunite in alunite-natroalunite solid-solution: Appendix D) can be calculated to illustrate the effects of partial Na-substitution in alunite relative to other alteration phases (Fig. 5.4). As shown, increasing the Na content of alunite expands its stability field relative to kaolinite/pyrophyllite. Equilibrium among these minerals occurs at lower aqueous K + concentrations, as expected, and slightly higher pH values at all temperatures studied (assuming constant flso4) • The magnitude of this change is typically small (about 1 to 1.5 log units K 2S04) and appears to be greatest at intermediate temperatures (i.e., 350°C). Interpretation of Results: • In this study, the thermodynamic equilibria calculated on the basis of data in Helgeson et al. (1978) and Stoffregen and Cygan (1990) are much closer to the dissociation-constant triple points than the thermodynamic calculations of Hemley et al. (1969). Our calculations of the alunite-kaolinite-muscovite invariant point are within 0.5 log units of experimental H 2S04 and K 2 S 0 4 results at 200° and 300°C. The difference is slightly greater at 380°C (up to 1 log unit). Recent thermodynamic data cannot be compared directly to the experimental results of Hemley et al. (1969) without further determination of dissociation effects for K + and H + aqueous species. • The pressure conditions (1000 bars) used in these activity models are higher than expected for alteration in most areas of the El Indio Belt, and for many high sulfidation systems in general. However, we have shown that phase relations are relatively insensitive to pressure, which is consistent with the experimental results of Hemley et al. (1969). General relations 185 Figure 5.4. Effect of variable alunite Na substitution on the K 2 O - A L 2 O 3 - S 1 O 2 - H 2 O - S O 3 system. Dashed lines represent X K a l n . a i „ = 1.0, 0.8, 0.6, 0.4, and 0.2. 186 illustrated here can therefore be applied to high sulfidation environments, regardless of P of formation. Activity-activity diagrams indicate that both end-member alunite and natroalunite (and various intermediate compositions) are stable relative to kaolinite/pyrophyllite and micas at temperatures typical of acid-sulfate alteration (about 200-350°C). At constant sulfate concentrations, the natroalunite-kaolinite (or pyrophyllite) assemblage is stable to higher pH values than potassium equivalents. The assemblages alunite/natroalunite + kaolinite (or pyrophyllite) and alunite/natroalunite + kaolinite + muscovite (or illite) are common in naturally occurring systems. Muscovite (or illite) are rarely observed in the El Indio Belt but are present elsewhere, particularly in the roots of some magmatic-hydrothermal systems (e.g., Hedenquist et al., 2000; Sillitoe, 1999). The shrinkage of the mica field at high temperatures (Figs 5.1 and 5.3) suggests that small changes in H + and #K+ would be sufficient to move from an assemblage containing pyrophyllite to one containing K-feldspar. This may explain why sericite is relatively rare in high sulfidation systems, and the El Indio Belt in particular. Detailed short-wave infrared (SWIR) analyses at Pascua, for example, have shown that advanced argillic and acid-sulfate assemblages can occur directly adjacent to relatively fresh, K-feldspar bearing granites, without an intermediate sericitic or illitic alteration zone. To our knowledge, an equilibrium assemblage of alunite + K-feldspar or alunite + K-feldspar + muscovite has not been reported in naturally occurring systems. This apparent absence suggests that the elevated K2SO4 concentrations required to stabilize such assemblages are rarely, if ever, obtained in nature (Hemley et al., 1969). The metastable formation of alunite must also be considered in natural systems. As noted by Hemley et al. (1969), the rate of crystallization of silicate minerals at low temperatures is much slower relative to sulfate minerals. It is therefore possible that alunite-natroalunite could form from the breakdown of feldspathic rocks under conditions that would otherwise produce K-mica or kaolinite, if equilibrium were established. Also, disequilibrium acid- sulfate assemblages in high sulfidation deposits are likely, due to the overprinting or telescoping of different alteration assemblages. In the El Indio Belt in particular, late-stage (lower temperature) alteration commonly overprints higher temperature assemblages. This overprint can lead to complex phase relations that do not represent equilibrium conditions. 187 E F F E C T S O F F L U I D C O M P O S I T I O N O N T H E A L U N I T E - N A T R O A L U N I T E S O L I D - S O L U T I O N Thermodynamic data for natroalunite (Table 5.1) were determined from the alkali- exchange experiments of Stoffregen and Cygan (1990), who examined the reaction: KA1 3(S0 4)2(0H) 6 + N a + = NaAl3(S04)2(OH)6 + K + (5.1) The distribution coefficient (Kd) for reaction (5.1) is defined as: Kd = mNaJ™K+ where X N 3 indicates the mole fraction of Na in the solid, and m is the molality in aqueous solution. Experimentally determined values of lnlCj and X N A were fit by Stoffregen and Cygan (1990) using a subregular Margules model, in order to describe the nonlinear variations in their data. Results from their study indicate that natroalunite is favoured at higher temperatures, because the equilibrium constant for reaction 5.1 decreases with decreasing temperature. Table 5.1. Estimated thermodynamic properties of natroalunite at 25°C and 1 bar (Stoffregen and Cygan, 1990). A G f -4622.40 ± 1.91 kJmol" 1 AHf -5131.97 kJmol" 1 S 321.08 JKmol " 1 V 141.15 cm 3 mol"1 C P(a) 641.5 JK- 'mof 1 C P (b) -7.87 x 10"3 J K ^ m o l 1 C P (c) -234.12 x 105 J-K-mol"1 The mixing model developed by Stoffregen and Cygan (1990) can also be used to examine the effect of fluid composition, particularly K/Na contents of a source fluid, on the composition of alunite-natroalunite solid-solution. Using reaction 5.1, the activity of both alunite and natroalunite can be modeled against variable fluid compositions assuming either: (a) Ideal solid-solution - This is the simplest model and nearly fits data from Stoffregen and Alpers (1992) on unit-cell dimensions for samples on the alunite-natroalunite binary. (b) Non-ideal solid-solution - This model is most consistent with the findings of Stoffregen and Cygan (1990). 188 Both models require calculating log K for reaction 5.1 knowing that: „ [ a N a l n , a l n ] [ a A : + , / / ] A 5 1 = [ a Kaln,aln][ < 3 A'a+, /7] Where aNain,ain = activity of natroalunite in the solid phase, aKain,ain = activity of alunite in the solid phase, «K+ A = activity of K + in the fluid phase, «Na+ A = activity of Na + in the fluid phase. Re-arranging this expression gives: aK+ _ f l K.aln . aln (5.3) The value of K is fixed and calculable at various P and T conditions. A l l reactions in this study were calculated for 500 and 1000 bars pressure, and for temperatures ranging from 25° to 450°C, using SUPCRT92 (Johnson et al., 1992). a) Ideal Solid-solution In an ideal solution: ajtaiMn = XKain,ain and aKain,ain = 1- «Nain,ain (e.g., Nordstrom and Munoz, 1985). Rearranging equation 5.3 gives: aK+ \aNa+ J fl "Naln.aln Equation 5.4 can be determined over a range of Xicain.ain = 0.1 to 0.9 and plotted against temperature (Fig. 5.5a). Pressure has a minimal effect on the behavior of the solid-solution, as illustrated in Figure 5.5b. K i A - K 5 A (5.4) 0* b) Non-ideal Solid-solution The activity of a phase in asymmetric (sub-regular) binary solid-solutions can be calculated from measured Margules parameters (e.g., Anderson and Crerar, 1993): RT\nyx =(2WG2 -WGl)X22 +2(WGl -WG2)X23 (5.5) and, aiP'T =X,P'T*riPJ (5.6) 189 o o CL E CD 400 300 200 100 0.001 fluid Figure 5.5a. Model fluid/composition curves for an ideal alunite-natroalunite solid solution at 500 bars pressure. Lines represent molar fraction of alunite (i.e., XKain-ain)- 200 h O o CL E CD 100 50 bars 500 bars 0.001 0.01 0.1 K + /Na + fluid Figure 5.5b. Model fluid/composition curves for an ideal alunite-natroalunite solid solution at variable pressure. Lines represent molar fraction of alunite (i.e., X^m-airO- 190 where Wo = Margules term for phases 1 or 2 Yi = activity coefficient for phase 1 For the purposes of this study, phases 1 and 2 are alunite and natroalunite, respectively. Mixing parameters (WG ) were determined by Stoffregen and Cygan (1990) for alunite and natroalunite at 250°C, 500 bars; 350°C, 500 bars; and 450°C, 1000 bars (Table 5.2). Table 5.2. Estimates of In K ( 5 A) and Margules parameters for alunite-natroalunite mixing reaction from Stoffregen and Cygan (1990). l a error estimates given in parentheses. In K(5.i) W G , N A (J-mol-1) W G , K (Jmof1) 450°C, 1000 bars -0.99 (0.05) 1837(427) 3159(435) 350°C, 500 bars -1.73 (0.26) 2867 (1050) 4785 (1229) 250°C, 500 bars -2.56 (0.42) 4668 (2091) 6443 (4836) Equations for the variation of Wo,Na and WG,K with temperature (e.g., Essene, 1982) were generated based on a least-squares regression (Appendix D). Using these estimates for Wo,Na and WG,K in equations 5.5 and 5.6, activity values for both alunite and natroalunite were calculated for a range of solid-solution compositions (XicaiMn = 0.1 to 0.9 at 500 and 1000 bars, and 100° to 450 °C : Appendix D). Calculated activity values for alunite and natroalunite are input into equation 5.3, which can be re-arranged as: = *5A ^Kaln.aln \aNa+ J fl _ a Naln.aln _ Plotting equation 5.7 at variable temperatures shows the combined effects of temperature and K + / N a + fluid activity on the composition of alunite-natroalunite solid-solution (Fig. 5.6). These ionic species (K + and Na +) will dominate at lower temperatures in the magmatic- hydrothermal environment (region I, Fig. 5.7), but complexes of Cl may be significant at higher temperatures (region III, Fig. 5.7). We can represent these higher temperature conditions using KC lTNaCr species as follows: 191 400 Figure 5.6. Effect of temperature and fluid K /Na fluid composition on a non-ideal alunite- natroalunite solid solution, 500 bars. Lines represent molar fraction of alunite (i.e. XKain,ain)- 100 - 0 I , , : , , , , , , 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 T o t a l M C i * Figure 5.7. Dominance fields for K/Na species, based on equilibrium calculations (see Appendix D for details. Region I = K + and Na +; Region II = K + and NaCl 0; Region III = KC1° and NaCl 0. 192 K + + NaCl° = N a + + K C l 0 Knowing logA^ 5 8=log (5.8) V QK+ J + log Re-arranging this equation gives: log ' K C l V ° N a C l J = l o g £ 5 8 - log r \ aNa+ \ a K + J "KCI V ° N a C l J (5.9) Log K for reaction 5.8 can be calculated at P,T using K + / N a + activities calculated previously (reaction 5.7). Reaction 5.9 can be plotted in the same manner as the previous reactions (Fig. 5.8). At intermediate temperatures (region II; Fig. 5.7), neither paired ionic species nor complexes of CI are predominant. Instead, NaCl predominates over Na + , but K + predominates over KC1. In this temperature range, therefore there is need to consider the reaction: KA13(S04)2(0H)6 K + NaCl = NaAl 3(S0 4)2(OH) 6 + K + + CI' (5.10) To plot this reaction, assumptions must be made concerning the activity of CI" in solution. However, under conditions typical of high sulfidation systems (i.e., total salinity less than about 0.5 M), this region is relatively small and will not be considered further. Discussion: • The calculations indicate that the alunite-natroalunite solid-solution is relatively insensitive to change in pressure. Varying the pressure from 500 to 1000 bars shifts equilibrium temperature determinations by less than ±10°C and fluid ratios by less than 0.1 log units K/Na, in both the ideal and non-ideal solid-solution models. These errors are likely smaller than the uncertainty in the thermodynamic calculations. • The alunite-natroalunite solid-solution approaches calculated ideal behavior at high temperatures (e.g., above about 400°C). At low temperatures, large differences are observed between the ideal and non-ideal models (Figs 5.5 and 5.6). The latter consistently predicts higher Na contents in alunite formed at constant temperature and fluid K + / N a + activities. 193 Figure 5.8. Effect of temperature and fluid KCl/NaCl fluid composition on a non- ideal alunite-natroalunite solid solution, 500 bars. Lines represent molar fraction of alunite (i.e. XKain,ain). 194 Overall, higher temperatures favor greater substitution of Na in alunite under both ideal and non-ideal conditions, at constant fluid K/Na ratios. This increase is consistent with the experimental results of Stoffregen and Cygan (1990) and is consistent with empirical observations of several magmatic-hydrothermal systems (e.g., Aoki, 1991; Thompson, 1992; Stoffregen et al., 2000 and references therein). At low to intermediate temperatures (i.e. 150-225°C), the non-ideal solid-solution model (Fig. 5.6) predicts that a large range of equilibrium alunite compositions (Xicain-ain = about 0.2 to 0.6) can form from a very narrow range of fluid compositions. However, at higher temperatures, our model predicts that more than an order of magnitude change in K/Na fluid content is required to precipitate both alunite and natroalunite end-member compositions, at equilibrium. This suggests that small-scale zoning between end-member alunite-natroalunite compositions at high temperature is unlikely the result of changes in compositions of the bulk fluid (see further discussion below). The calculations indicate that low K/Na fluid ratios are required for significant Na- substitution in alunite at all temperatures (at equilibrium). For example, at temperatures typical of magmatic-hydrothermal alteration (e.g., 200-350°C), fluid K/Na activities <0.01 are required to form nearly pure natroalunite (Figs. 5.6 and 5.8). Geochemical data for K/Na in active volcanic systems (e.g., fumaroles, volcanic gas condensates) are scarce, but a few studies suggest that molar K/Na ratios range between 0.1 to 0.8 (Reed, 1997; Giggenbach, 1997; Symonds et al., 1990). This range is well within a field for nearly pure alunite, according to our models. These conditions are consistent with the formation of nearly pure magmatic steam alunite but magmatic-hydrothermal alunite typically show a much greater degree of compositional variability (this study; Rye, 1993; Stoffregen et al, 2000). This suggests that the chemistry of alteration minerals is affected by factors other than simply equilibrium temperature and fluid composition. This is not surprising, given the dynamic nature of the magmatic-hydrothermal environment and the complexity of magma-fluid-rock interactions. 195 CONTROLS ON ALUNITE-NATROALUNITE COMPOSITIONS IN THE E L INDIO-PASCUA B E L T The compositional variability of alunite and alunite-group minerals in the El Indio- Pascua Belt has been described in previous chapters. Alunite K/Na ratios are of interest in this region because of their potential use as an indicator for higher temperature (e.g., magmatic- hydrothermal) alteration that is directly associated with Au ± Cu, Ag mineralization at several deposits and prospects, including Pascua-Lama, Tambo, El Indio, and Del Carmen. The effects of depositional temperature on the composition of the alunite-natroalunite solid-solution in the El Indio-Pascua Belt are discussed below. The influence of fluid composition on alunite mineral chemistry is also addressed. Depositional Temperature Geochemical data for alunite-group minerals from this study indicate that higher depositional temperatures favor a greater range of Na-substitution. Electron microprobe (EPMA) data (Fig. 5.9) show that magmatic-hydrothermal alunite has a much larger range in Na substitution (Fig. 5.9), and nearly twice the average Na content (0.17 m.p.f.u.), than steam-heated (averaging 0.07 m.p.f.u. Na) or late-stage and supergene alteration (0.09 m.p.f.u. Na). Estimated temperatures of deposition for magmatic-hydrothermal alunite range from about 140° to 380°C, compared with lower-temperature steam-heated (ca. 90-180°C), and late stage alteration (ca. 60- 150°C), on the basis of stable isotope, fluid inclusion, and paragenetic evidence (Chapters 2 and 3). Differences in K:Na contents are likely exaggerated by the larger data-set for higher temperature occurrences; quantitative data for low-temperature alunite are limited by the small grain size of most samples. However, these differences are statistically significant at a 95% confidence level, based on t-test results (Appendix A). E P M A data are also supplemented by qualitative SEM-EDS analyses of more than 125 polished thin sections and individual grain mounts that have shown similar trends. Geochemical data provide further evidence for two stages of magmatic-hydrothermal alteration (AS I and AS II; details see Chapter 2) at the Pascua deposit. A correlation between alunite Na content and increasing depth (and by inference, temperature) is evident for AS II alteration (Fig. 5.10) that occurs in banded alunite-sulfide veins and in the matrix of mineralized breccias. An earlier stage of acid sulfate alteration minerals (AS I) occurs as aggregates and disseminations in wallrock over a large areal extent. The distribution of AS I alteration is irregular and strongly structurally controlled, and there is no consistent relation between alunite 196 80 60 o 40 20 • Magmatic hydrothermal I Steam heated • Late stage/supergene 0.0 0.1 Atoms p.f.u. Na Figure 5.9. Histogram showing the range of Na substitution in alunite from 3 environments of acid sulfate alteration (Rye et al., 1992). Based on EPMA data from El Indio-Pascua Belt properties, including Pascua-Lama, El Indio, Tambo, Del Carmen, and Salitrales. Details of analytical procedures given in Appendix B. Total number analyses («): magmatic- hydrothermal n = 976; steam heated n = 47; late stage and apparent supergene n = 43. T 4800 • A S II alunite A S I alunite ro > LU 4600 H 4400 1 2 Na/K (a.p.f.u.) Figure 5.10. Range of K/Na ratios for Pascua AS I and AS II alunite with elevation. Based on EPMA results for a total of n data points (AS I n = 335; AS II n = 331). 197 Na content and depth. Instead, the influence of temperature on this earlier assemblage is evident from correlations between alunite geochemistry and the mineralogy of associated clays (Fig. 5.11). Alunite associated with pyrophyllite has lower average K contents than alunite associated with dickite (statistically significant at 95% confidence level, based on t-test results: Appendix A). In hydrothermal systems, pyrophyllite typically forms between ca. 200° to 350°C, whereas dickite forms at temperatures less than ca. 200° to 250°C (Hemley et al., 1980; Reyes, 1990). Fluid Composition The effects of source-fluid composition on alunite-natroalunite solid-solution are difficult to quantify, given the complexity of natural systems. Both local and distal scale fluid controls on alunite geochemistry can be inferred in the E l Indio-Pascua Belt, however, given constraints from the solid-solution models presented above. Local controls Complex geochemical zoning and banded alunite-natroalunite crystals are common at several localities in the El Indio Belt (e.g., Del Carmen: Fig. 5.12). There is significant debate in the literature as to whether oscillatory zoning in hydrothermal minerals reflects local or large- scale variations in fluid composition (see discussions in Shore and Fowler, 1996, and Crowe et al., 2001). Our solid-solution models (Fig. 5.6 and 5.8) predict that more than an order of magnitude change in K/Na fluid content is required to precipitate both alunite and natroalunite end-members (at equilibrium). Such large shifts in bulk fluid composition are unlikely in the natural environment, particularly in the short time periods required to form zoned crystals. Zoning between end-member alunite and natroalunite is therefore unlikely the result of changes in bulk fluid compositions and may be attributed to more local effects such as supersaturation or crystallographic controls, analogous to the mechanisms inferred by Putnis et al. (1995) and Prieto et al. (1997) from their porous-media solid-solution experiments. Distal controls On a larger scale, extrinsic variations in fluid composition (e.g., due to lithological variations, fluid:rock interaction) can be shown to affect bulk alunite-natroalunite compositions. Data from the Del Carmen project in particular suggest that host-rock chemistry can affect the composition of alunite-natroalunite solid-solution. Magmatic-hydrothermal alunite in this area is less K-rich on average (Table 5.3) than that from the nearby E l Indio-Tambo district (this study; 198 100 >N o c CD C T CD 80 60 40 20 • B Dickite I I Pyrnphyl l i fp 0.0 0.2 0.4 0.6 Atoms p.f.u. Na 0.8 1.0 Figure 5.11. Histogram showing alunite Na contents for Pascua alteration (stage AS I) based on dominant clay mineralogy. Alunite data based on EPMA results for a total of n analyses (dickite series, n = 99; pyrophyllite series, n = 137). Clay identification by SWIR analysis (PIMA). Figure 5.12. SEM backscatter image showing oscillatory zoning between Na-rich (dark) and K-rich (light) compositions in Del Carmen alunite. Bright circle in centre due to beam damage. 199 Bennet, 1995) and the Pascua-Lama area. K-depletion is attributed to increased Na substitution in Del Carmen alunite, based on EPMA (Fig. 5.13) and SWIR results. The style, timing, and characteristics of alteration are relatively consistent for all three regions (this study; Bissig, 2001), but their host lithologies differ. Alteration at Del Carmen is hosted primarily in the andesitic Cerro de las Tortolas Formation and the contemporaneous Infiernillo Intrusive Unit (Noriega and Perez, 1997). This unit has higher average Na20/K20 than either the Tilito Formation, which hosts the El Indio-Tambo deposits, or Pastos Blancos Group intrusions in the Pascua area (Table 5.4: Bissig, 2001). Intense, acid sulfate alteration of the andesitic host rock likely resulted in higher Na/K in the alteration fluids, which is reflected in the composition of the Del Carmen alunite. Variations in host-rock composition are most likely to influence alunite that forms directly from wallrock reaction with low fluid:rock ratios, rather than alunite deposited as open-space fill in veins and breccia matrices. Table 5.3. Summary of EPMA Na data for magmatic-hydrothermal alunite from Del Carmen compared to the rest of the El Indio-Pascua Belt. Statistical analysis is based on recalculated atoms per formula unit Na for n analyses in each group. Standard deviation (CT), t-statistic, degrees of freedom (DF), and probability at 95% confidence level (p) are given: based on SYSTAT t-test (details Appendix A). Group n Mean CJ t value DF P El Indio-Pascua Belt 909 0.158 0.134 -6.529 70.6 0.000 Del Carmen 66 0.301 0.175 In contrast, the solid-solution models presented earlier also predict that very small changes in fluid composition could precipitate a range of alunite-natroalunite solid-solution at low temperatures. Fluid variation of this magnitude may be present in the near-surface environment because of variable mixing between magmatic fluids and meteoric waters and variable fluid:rock ratios. Significant variation in alunite composition has been recorded for other low-temperature (supergene or sedimentary) acid-sulfate environments (e.g., Stoffregen and Alpers, 1992; Hemley et al., 1969; Stoffregen et al., 2000). However, data from the El Indio-Pascua belt do not exhibit similar variation. Late-stage and supergene alunite from this study is K-dominant and rarely contains greater than 0.3 m.p.f.u. Na substitution (Fig. 5.9), even 200 60 40 c D CT CO 20 El Indio-Pascua Belt (except Del Carmen) 3 Del Carmen I n 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Atoms p.f.u. Na Figure 5.13. Histrogram showing alunite Na contents fro Del Carmen magmatic-hydrothermal alunite versus magmatic-hydrothermal alunite from all other El Indio-Pascua Belt properties. Alunite data based on EPMA results for a total of n analyses (Del Carmen series, n = 66; El Indio-Pascua Belt, n = 909). Table 5.4. Average major element geochemistry for host lithologies from the Del Carmen (Infiernillo; n=5), Tambo (Tilito; n=3), and Pascua (Colorado Unit, n=l) areas. Data from Bissig (2001). Unit Cerro de las T6rtolas Tilito Pastos Blancos (Infiernillo) (Chollay) Deposit Del Carmen El Indio-Tambo Pascua S i0 2 62.22 62.12 76.35 Al 2 0 3 16.39 16.39 12.28 Ti0 2 0.74 0.66 0.08 F e 2 0 3 (Fe total) 5.32 5.05 0.82 MnO 0.09 0.09 0.05 MgO 2.43 1.59 0.11 CaO 4.22 4.35 0.36 Na 2 0 3.74 2.89 3.32 K 2 0 2.57 4.24 4.39 P 2 O 5 0.19 0.16 0.01 Total 98.00 97.63 97.79 Na 2 0 /K 2 0 1.5 0.7 0.8 201 though significant mixing between meteoric and magmatic fluids is indicated by stable isotopic results (details see Chapters 2 and 3). These findings indicate that large-scale fluctuations in source-fluid composition did not result in variable alunite-natroalunite compositions in low- temperature assemblages. Data suggest that equilibrium conditions were rarely achieved in the near-surface environment, or that temperature exerted the greatest influence on the composition of alunite-natroalunite solid-solution. Summary • Geochemical results from the El Indio-Pascua Belt are consistent with our solid-solution model calculations, as well as with experimental (Stoffregen and Cygan, 1990) and empirical evidence (e.g., Stoffregen et al., 2000). Data suggest that higher temperatures of formation favor increased incorporation of solid-solution Na in alunite. • In the magmatic-hydrothermal environment, both depositional temperature and large-scale fluid chemistry may affect average alunite-natroalunite compositions. Intracrystalline zonation or chemical banding between alunite and natroalunite compositions is likely due to local supersaturation effects or crystallographic controls. • Large-scale fluid controls on alunite chemistry are not evident in low-temperature alteration. Alunite-natroalunite chemistry is most likely influenced by depositional temperatures or disequilibrium conditions. R E F E R E N C E S Andersen, G.M. and Crerar, D.A, 1993. Thermodynamics in Geochemistry; The Equilibrium Model. Oxford University Press, New York. Aoki, M., 1991. Mineralogical features and genesis of alunite solid-solution in high temperature magmatic-hydrothermal systems. Journal of the Geological Survey of Japan, 277, 31-32. Aoki, M., Comsti, E.C., Lazo, F.B., and Matsuhisa, Y., 1993. Advanced argillic alteration and ge