w\ u»V C v ♦ ■• W V £. u ** •jfltef- * / •- * v °^- * • • " A v <»4fc* ^/ .-iaKaSr. V,* *V vv •° .»' '.' >°^K V 6 ^V \f«^. ; ^^ v V** » ^ v »bv* .h •»°- «P •>•-.- '*. 9: A°* --B >J jP-^, 'fgtMBU £"•>■ CI q,. '• .Ti» aO' V •♦TT. • •" ^ ~o. -. .t. •" aP °o /-: 4 V«* i*^ v^V'— '* *« A v» <+ ♦'TVS* ^0 tor. ^. «- : ; G°\ »*q* -i !>* .*• ^- *> >0 ^ v •^^ l* o°' # . ^ 'bV 1 v >°^K v •o v .♦! ^q*. ., > ^ : .^ »!••- *V c v ♦ 6°+ - o JO *^ A>^<» »\t r * A^ ^ • > AV «*v . v ^v •: ■q, "•.. q#. ••-•° aP^ ^ *. »^^y/^x'y^v^'>/^x>VAv^^^^ IC 8897 Bureau of Mines Information Circular/1982 Platinum Availability- Market Economy Countries A Minerals Availability System Appraisal By T. F. Anstett, D. I. Bleiwas, and C. Sheng-Fogg UNITED STATES DEPARTMENT OF THE INTERIOR Information Circular 8897 Platinum Availability- Market Economy Countries A Minerals Availability System Appraisal By T. F. Anstett, D. I. Bleiwas, and C. Sheng-Fogg UNITED STATES DEPARTMENT OF THE INTERIOR James G. Watt, Secretary BUREAU OF MINES Robert C. Horton, Director As the Nation's principal conservation agency, the Department of the Interior has responsibility for most of our nationally owned public lands and natural resources. This includes fostering the wisest use of our land and water resources, protecting our fish and wildlife, preserving the environmental and cultural values of our national parks and historical places, and providing for the enjoyment of life through outdoor recreation. The Department assesses our energy and mineral resources and works to assure that their development is in the best interests of all our people. The Department also has a major responsibility for American Indian reserva- tion communities and for people who live in island territories under U.S. administration. no. ml This publication has been cataloged as follows: Library of Congress Cataloging in Publication Data Anstett, T. F. (Terrance F.) Platinum availability [in] market economy countries. (Information circular/Bureau of Mines) 8897) Bibliography: p. 13 Supt of Docs no ; I 28 27: 1 . Platinum ores. 2. Platinum I. Bleiwas, Donald I. II. Sheng-Fogg.C. (Catherine) III. Title. IV Series: Information circular(United States. Bureau of Mines): 8897 TN295U4 [TN490P7] 622s [553.4'22] 82-600204 iii PREFACE The Bureau of Mines Minerals Availability Program is assessing the worldwide availability of nonfuel critical minerals. The program collects, compiles, and evaluates information on active, developed, and explored deposits and on mineral processing plants worldwide. Objectives are to classify domestic and foreign resources, to identify by cost evaluation resources that are reserves, and to prepare analyses of mineral availability. This report on the availability of platinum is part of a continuing series of Minerals Availability System (MAS) reports to analyze the availability of minerals from domestic and foreign sources. Analyses of other minerals are currently in progress. Questions about the MAS program should be addressed to Director, Division of Minerals Availability, Bureau of Mines, 2401 E Street, N.W., Washington, D.C. 20241. v CONTENTS Page Preface iii Abstract 1 Introduction 2 Acknowledgments 2 MAS evaluation and analysis procedure 2 Deposits analyzed in this study 3 Republic of South Africa 4 Rustenburg 6 Impala 6 Western Platinum 6 Der Brochen 6 Zimbabwe 6 Colombia 6 Canada 7 United States 7 Stillwater Complex 7 Salmon River 7 Ely Spruce and Minnamax 7 Other platinum sources 8 Republic of South Africa 8 Brazil 8 Canada 8 _ Page Other platinum sources — Continued United States 8 Soviet Union 3 Zimbabwe 3 Engineering and cost evaluation 9 Assumptions 9 Operating and capital costs 9 Analysis of platinum availability 10 Total availability 10 Annual availability 10 Availability from domestic sources 11 Analysis of factors affecting platinum availability 11 Impact of byproduct credits 11 Impact of operating and capital costs 1 -| Secondary supply from U.S. sources i 2 Technological enhancements -jo Summary and conclusions 1 2 Bibliography 13 Appendix A. — Central economy countries 14 Appendix B— Mining and processing operations for a typical South African property 14 ILLUSTRATIONS 1 . Flowsheet, MAS evaluation procedure 3 2. Classification of mineral resources 3 3. Location map, South African deposits 5 4. Location map, Zimbabwean deposits 6 5. Total platinum potentially available from primary platinum deposits 10 6. Annual platinum potentially available at selected incentive prices 11 7. Potential platinum availability, without byproduct credits 11 8. Platinum potentially available from "low cost" deposits, with a 25-percent operating cost increase 12 B-1 . Flowsheet, South African PGM ore processing 15 B-2. Flowsheet, conventional refining process 15 B-3. Flowsheet, OPNIM refining process 16 TABLES 1 . Estimated 1 980 world platinum and palladium production 2 2. Platinum deposit information 4 3. Platinum resource information, demonstrated level, by country 4 4. Byproduct commodity prices 9 5. Estimated typical operating costs 1 PLATNIUM AVAILABILITY— MARKET ECONOMY COUNTRIES A Minerals Availability System Appraisal By T. F. Anstett, 1 D. I. Bleiwas, and C. Sheng-Fogg 2 ABSTRACT The Bureau of Mines investigated the availability of platinum from known major deposits in market economy countries. Fifteen of these deposits contain demonstrated resources totaling 302 million troy ounces of platinum. Detailed geologic, engineering, and cost evaluations were performed on each deposit to determine its potential for platinum production. There are large amounts of demonstrated platinum potentially available from market economy deposits that were not analyzed owing to lack of adequate data or to technological limitations. These include 297 million troy ounces contained in the UG2 and Platreef of the Bushveld Complex, South Africa. This analysis indicates that about 1 10 million troy ounces (2.3 million ounces annually) of platinum are potentially available at the January 1980 producer price of $420 per troy ounce, nearly all from South Africa. An additional 85 million troy ounces occurring in Zimbabwean deposits would become available only through more than a quadrupling of that price. In order to meet projected world demand of 3.5 million troy ounces in 1 990, platinum price may have to increase. Only one domestic deposit, the Salmon River in Alaska, is capable of producing platinum at $420 per troy ounce. Regardless of price, at assumed production capacities, domestic deposits could supply less than 10 percent of annual U.S. requirements. 1 Geologist, Minerals Availability Field Office, Bureau of Mines, Denver, Colo. 2 Physical scientist, Minerals Availability Field Office, Bureau of Mines, Denver, Colo. INTRODUCTION The National Material and Minerals Policy. Research and Development Act of 1980 established, among other objec- tives, a goal of providing for the security of foreign mineral imports. In order to establish and implement a coherent, effective policy, it is necessary to develop and analyze data on mineral resources for the United States and the rest of the world. The purpose of this report is to present the results of the Bureau of Mines analysis of the availability of platinum from domestic and other market economy sources. 3 The price of platinum, like that of other precious metals, is often subject fo speculative pressures quite unrelated to industrial utility; however, the metal possesses a wide variety of physical and chemical properties that make it essential in many industrial applications, including auto emissions control, petroleum refining, and glass fiber manufacturing. The metal is also used as a catalyst in the chemical and pharmaceutical industries, and in a wide variety of electrical and electronic devices. Platinum is recovered as the primary metal in ores from South Africa and placer deposits in various countries throughout the world. Significant quantities of the metal are recovered as a byproduct from the nickel-copper Noril'sk and Sudbury deposits of the Soviet Union and Canada, respectively. Table 1 shows 1980 world production of platinum by country and geologic source. A more detailed discussion of platinum production and consumption is available from other Bureau of Mines sources {13, 16).' The United States obtains approximately 90 percent of its platinum from foreign sources; the remaining 10 percent is 3 Only market economy countries were evaluated in this study. "Market economy country " is a term used by the Bureau ot Mines to denote a country that does not have a centrally planned economy (appendix A). 4 Italicized numbers in parentheses refer to items in the bibliography preceding the aooendixes. derived as a byproduct of gold and copper production and from recycling. A disruption of the supply of this strategic metal could have a significant impact on the United States. The Republic of South Africa, the world's leading producer of platinum, has historically been a reliable source of supply to the United States. However, the political uncertainty of many African countries has served to raise concern in the United States regarding the continued availability of this important metal. Although previous reports (e.g., 17) have addressed, in general terms, the subject of platinum resources in known deposits, a search of the literature revealed that no comprehensive studies or analyses have been published regarding potential availability on an individual deposit 5 basis. The procedure for conducting this analysis was as follows: 1. The quantity and grade of platinum resources were evaluated in relation to physical, technological, and other conditions that affect production from each deposit. 2. The capital investments and operating costs associated with mining, concentrat'ng, smelting, and refining operations were estimated. 3. A cost analysis for each deposit was performed. The results of the analysis indicate the unit cost and associated tonnage of platinum that could be recovered at specific production levels for each deposit. 4. Cost-production relationships were combined and analyzed in the form of a platinum availability curve to show the platinum production potential at various commodity prices. 5. Sensitivity analyses were performed in order to determine the effects of various parameters (e.g., operating costs, byproduct prices) on platinum availability. 5 For purposes of this study, "deposit" is hereafter defined as that portion of an exploitable or potentially exploitable mineralized body owned and/or operated by a specific entity. TABLE 1. — Estimated 1980 world platinum and palladium production Production, troy ounces Country Source Platinum Palladium Total PGM 1 2,500 7,200 9,850 162,000 186,000 404,585 14,345 14,345 120 120 NA NA 590 12,900 29,700 NA 1,891,000 806,000 3,100,000 812,500 2,177,500 3,250,000 1,000 2,000 3,348 700 5,300 NA 3,000 > 6,800 9,800 Australia Canada Colombia Ethiopia Finland Japan Republic of South Africa USSR United States Yugoslavia Zimbabwe Total Byproduct of nickel ores Byproduct of nickel-copper ores Placer . . do Byproduct of copper ores Byproduct of nickel and copper ores 2 PGM 1 ores Placer, and byproduct of nickel-copper ores . . . Placer, and byproduct of gold and copper ores NA Byproduct of nickel ore 2,900,065 3,220,500 6,792,638 NA Not available. 1 Platinum-group metals (platinum, palladium, iridium, osmium, rhodium, ruthenium). 2 Imports from several countries including Australia, Canada, Indonesia, Papua New Guinea, and the Philippines, Sources: 13, table 4; 16, table 9. ACKNOWLEDGMENTS The authors wish to thank J. Roger Loebenstein, William C. Butterman, and Christine M. Moore of the Bureau of Mines, Division of Nonferrous Metals, for their assistance in identifying the properties and associated resource tonnages included in this report. Personnel at Bureau of Mines Field Operations Centers in Juneau, Alaska, Denver, Colo., and Spokane, Wash., supplied much of the data for the domestic deposits included in this study. MAS EVALUATION AND ANALYSIS PROCEDURE A flow chart of the Minerals Availability System (MAS) evaluation and analysis procedure is shown in figure 1 . Data collection for foreign deposits was performed under contract, while data for domestic deposits were collected by Bureau of Mines personnel at Field Operations Centers in Spokane, Wash Denver, Colo., and Juneau, Alaska. This section of the report presents the criteria used to select deposits for the analysis, the sources of data used for grade and tonnage Identification and selection of deposits To n n age and grade determination Enginee ring and cost eva I u ation | Mineral Indust r ies I J Location « ! System I I (MILS) | j data j MAS computer data base Taxes, royalties, cost indexes, prices, ere... Deposit report preparation MAS per manent deposit files Data sel ection and va I idation Variable and parameter adjustments Economic analysis Dafa I Sensitivity ana lysis Availabilityi curves Analytical reports j± W Data Availability curves Analytical reports FIGURE 1. — Flowsheet, MAS evaluation procedure. determination, and the various assumptions used to evaiuate the amount of platinum potentially available from each deposit. Engineering and cost evaluation procedures are outlined in a later section. Selection was limited to deposits in market economy countries containing demonstrated platinum resources according to the definitions outlined in the reserve-resource classification system developed by the Bureau of Mines and U.S. Geological Survey (fig. 2). The producing deposits analyzed accounted for 91 percent of 1980 primary platinum production from market economy countries. The only deposits containing byproduct platinum that were included in the analysis were the Ely Spruce and Minnamax deposits in Minnesota. They were included because of their potential to contribute significantly to U.S. production. Deposits of the Sudbury district of Canada and the Empress Mine fn Zimbabwe, which produce platinum as a byproduct of nickel, were not analyzed owing to a lack of required data. Information on grades, tonnages, and other physical criteria affecting platinum availability was obtained from numerous sources, including Bureau of Mines and Geologic- al Survey publications, professional journals, industry pub- lications, annual reports, company 10K reports and prospec- tuses filed with the U.S. Securities and Exchange Commis- sion, and data provided by private companies and contrac- tors. The personal knowledge and judgments of Bureau of Mines engineers and geologists were frequently utilized. Since many of the deposits analyzed are nonproducers, certain assumptions relative to prospective operations on those deposits were made. Where appropriate, the assump- tions pertinent to each deposit are stated. Cumulative Production IDENTIFIED RESOURCES UNDISCOVERED RESOURCES Demonstrated Inferred Probability Range fnr\ Measured Indicated \0[ j Hypothetical } Speculative l ECONOMIC Res< jrve se Inferred Reserve Base I 1 ' 4- - i MARGINALLY ECONOMIC Bo SUB- EC0N0MIC 1 Other Occurrences Includes nonconventional and low-grade materials FIGURE 2.— Classification off mineral resources. DEPOSITS ANALYZED IN THIS STUDY Table 2 contains a list of deposits investigated by the Bureau of Mines. Owing to a lack of adequate resource, engineering, and/or cost information, detailed analyses could not be performed for several of the deposits. Those that were not analyzed are marked with an asterisk; a more detailed discussion of these deposits may be found in the section "Other Platinum Sources." Table 3 contains resource information for the deposits that were analyzed. The Bureau of Mines has established a reserve base for platinum, which contains the in-place demonstrated (mea- sured plus indicated) resource from which reserves are established, and includes those resources that are currently economic or marginally economic and some of those that are currently subeconomic (fig. 2). The platinum reserve base contains about 513 million troy ounces. The deposits analyzed in this study contain a total of approximately 302 million troy ounces of platinum. Included in the reserve base, but not in this analysis, are 350 million troy ounces in the Norilsk and Sudbury deposits of the Soviet Union and Canada, respectively, and the UG2 and Platreef intervals of the Bushveld Complex, South Africa. These deposits were excluded primarily owing to a lack of sufficient data or to technological limitations. However, included in this analysis are 139 million troy ounces in the Great Dyke of Zimbabwe, for which data were available. Because of high production costs, the Zimbabwean resources have been placed in the subeconomic category shown on figure 2. The deposits analyzed as part of this study were evaluated at the demonstrated level owing to the lack of credibility in resource data at lower levels of geologic confidence and the fact that there are adequate demonstrated resources to fulfill demand for the foreseeable future. However, there are estimated to be an additional 1 .0 to 1 .3 billion troy ounces of identified platinum in known market economy deposits (13). Following is a discussion, by country, of each of the deposits analyzed for this report. REPUBLIC OF SOUTH AFRICA The Republic of South Africa is the world's largest supplier of platinum, accounting for 65 percent of world production in 1980 (table 1). Nearly all of South Africa's platinum is derived from operations on the Merensky Reef of the Bushveld Igneous Complex, the largest known layered magmatic sequence in the world, occupying an area of 26,000 square miles. The Merensky Reef is generally continuous through- out the entire complex and contains a total of 159 million troy ounces of platinum at the demonstrated level. Buchanan (2) has estimated the total platinum reserves 6 in the Merensky Reef to be 333 million troy ounces, calculated to a depth of 3,900 feet. Platinum-group metal (PGM) grade averages about 0.25 troy ounces per metric ton (oz/t), with platinum constituting 61 percent and palladium 26 percent of total PGM content (2, p. 56). 6 Although Buchanan refers to his figures as "reserves," the term is not defined in his paper, and does not necessarily correspond with the term as defined in U.S. Geol. Survey Circ. 831 (25). TABLE 2. — Platinum deposit information State and or Country and deposit Type of deposit Owner and/or operator Status 1 Type 2 Alaska: Salmon River •Salt Chuck Minnesota: Ely Spruce Minnamax Montana: Stillwater Anaconda Brazil: "Santa Catarina "Ouro Branco Canada: Lac des lies "Sudbury Colombia: Choco Pacifico South Africa: 3 Der Brochen Impala Rustenburg Western Zimbabwe: 'Empress Hartley Musengezi Selukwe Wedza Placer Ultramafic intrusive .do .do .do .do NA Sedimentary Layered volcanic . . Ultramafic intrusive Placer Ultramafic intrusive do do do do. do. do. .do. do. Hanson Properties, Inc. . Fox Geological Consultants. Ltd. Inco Ltd AMAX Exploration, Inc. Stillwater PGM Resources Anaconda Minerals Co. . . NA NA Boston Bay Mines Ltd. Numerous Mineros del Choco . . . Platinum Proprietary Ltd Impala Platinum Holdings Ltd. . Rustenburg Platinum Mines Ltd. Western Platinum Ltd Empress Nickel Mining Co., Ltd. n 8 ( 4 ) Dev PI Exp UG Exp OP Exp UG Dev UG Dev UG NA NA Exp Unknown Exp OP Prd UG Prd PI Exp UG Prd UG Prd UG Prd UG Prd UG Exp UG Exp UG Exp UG Exp UG * Deposit not analyzed for availability study. NA Information not available. ' Dev — Developing mine: Exp — Explored deposit; Prd — Producing mine. 2 OP — Open pit; PI — Placer; UG— Underground. 3 Includes Merensky Reef deposits only. * Ownership not accurately defined, but Union Carbide Management Services Ltd., (Zimbabwe) Group are known to be actively participating in potential development. Anglo-American Management Services Ltd., and Rio Tinto Mining TABLE 3 Platinum resource information, demonstrated level, by country 1 pla.irlumgrade, materfeftillion Platinum, million troy ounces Country oz/t 2 metric tons Contained Recoverable Canada, Colombia 3 (*) (7) i ~ South Africa 0.153 1,041 159 112 United States C) O 3 2 Zimbabwe .064 2,175 139 85 Total NAp NAp 302 200 1 Does not include Sudbury Complex of Canada, or UG2 and Platreef deposits of South Africa. 2 Troy ounces platinum per metric ton of ore. 3 Canada and Colombia have been combined to avoid disclosing individual company data. 4 Aggregate figures not calculated owing to a mixture of several different deposit types. Percent of MAS total 56 1 43 100 Four separate Merensky Reef deposits have been included in this study, three of which are currently being mined and account for all of South African ore production from the Reef (fig. 3). The producers are Rustenburg, Impala, and Western Platinum. Der Brochen is an undeveloped deposit. Production is dominated by Rustenburg, which accounted for 56 percent of total South African platinum output in 1979. Impala accounted for 41 percent, and Western Platinum supplied the remaining 3 percent (2, p. 58). All of the deposits are mined primarily for their PGM content, with nickel and copper as major byproducts. \ , IO°S f~ ZIMBAB^ a/eI / 20°S /Rus- r enbura/ L-^ / VTS REPUBLIC OF /? 30°S Ur fyf ■"^ SOUTH / ^ >» AFRICA/ -> Ill Hi 111 f N IO % Because of the highly regular and predictable nature of the dip, strike, thickness, and grade of the ore-bearing interval, the Merensky Reef is mined using simple, conventional stoping methods. This allows for large areas of relatively shallow deposits to be brought into production in less than a year; consequently, production and development rates can respond quickly to, and keep pace with, changes in the price of and demand for platinum. LEGEND O A D X Approximate boundary of Bushveld Complex Rustenburg sections Western Platinum Impala Der Brochen N CT3 Rustenbur Pretoria Lydenburg 50 1 ■ ' ■ ' ■'! ' I00 Miles 50 I00 I50 Kilometers FIGURE 3. — Location map, South African deposits. Rustenburg The Rustenburg deposit, leased and operated by Rusten- burg Platinum Mines Ltd., is located in the western Transvaal Province. It consists of four sections, all of which are currently being mined. The sections, with the approximate percentage each contributes to total Rustenburg annual production, are Rustenburg (67), Amandalbult (15), Union (15). and Atok (3). Platinum production in 1979 from the four sections totaled 1 .2 million troy ounces (2, p. 58). Each of the sections has its own concentration plant and smelter. Copper and nickel refining occurs at Rustenburg, and 75 percent of the PGM refining is done in Wadeville, Germiston, South Africa, located 25 miles east of Johannesburg. The other 25 percent is refined in Royston, England. Impala The Impala lease holdings, which consist of several mining operations, produced 870,000 troy ounces of platinum in 1979, or 41 percent of total Republic of South Africa production in that year (2. p. 58). The deposit is leased and operated by Impala Platinum Holdings Ltd. It is located mainly on Bafokeng tribal ground in the western part of the Transvaal, approximately 10 miles north of the town of Rustenburg. The Impala concentrator and smelter are located onsite, and refining occurs in the village of Springs, near Johannesburg, about 115 miles from the mine. Western Platinum The Western Platinum operation, owned by Western Platinum Ltd., is located in the western Transvaal about 20 miles east of the town of Rustenburg. The mine produced 74,000 troy ounces of platinum in 1979, or 3 percent of total Republic of South Africa production in that year (2, p. 58). Ore from the Western Platinum mine is milled and smelted onsite. The copper-nickel matte is shipped to Falconbridge's refinery in Kristiansand, Norway, and the resulting PGM-. bearing sludge is returned to the Lonrho refinery at Brakpan, South Africa, for final processing. Der Brochen Der Brochen is an explored, nonproducing deposit located in the eastern part of the Transvaal, south of the town of Lydenburg. Mineral rights to the deposit are owned by Platinum Proprietary Ltd. of Lydenburg. For purposes of this analysis, a development plan similar to that of the Western Platinum operation was assumed, with production modeled to begin in 1988. It was assumed that processing and smelting would be done onsite, with copper-nickel refining performed in a Norwegian refinery, and PGM concentrates returned to a refinery in Lydenburg. ZIMBABWE The Great Dyke of Zimbabwe includes four separate igneous complexes containing large quantities of platinum in association with palladium, nickel, and copper (fig. 4). The names of the complexes are Hartley, Selukwe, Wedza, and Musengezi. The PGM of the Great Dyke occur in a 30-inch-thick interval, mineralogically similar to the Merensky Reef of the Bushveld Complex; this interval is continuous in all four complexes. The PGM-bearing interval averages 0.25 percent nickel, 0.25 percent copper, and 0.10 to 0.16 oz/t of combined platinum and palladium ( 12). The four complexes together contain a total of about 139 million troy ounces of platinum at the demonstrated level. Currently, there are no major platinum mining operations on the Great Dyke. An experimental mine, the Mimosa, was — i Musengezi =T Complex _ Hortley Complex Selukwe Complex Wedzo Complex 50 ~~ I n 100 150 Kilometers FIGURE 4.— Location map, Zimbabwean deposits. established in the Wedza Complex and produced during 1970, 1971, and 1976 to 1978. A second test operation started in 1980 at Rio Tinto's Zinca Prospect in the Hartley Complex. Development drilling is proceeding there, with a test mine shaft planned by 1982. For purposes of this analysis, all of the deposits were modeled to begin production in 1 984, preceded by 4 years of development. Mine production rates are equivalent to existing chromite operations on the Great Dyke, while processing procedures were modeled after platinum opera- tions in South Africa. It has been assumed that the concentrate would be smelted in Zimbabwe, with the resulting PGM-bearing nickel-copper matte refined in Nor- way. Final PGM refining would be performed in Zimbabwe. COLOMBIA According to official production records, Colombia accounts for less than 1 percent of world platinum supply. However, it is estimated that only about half of the total Colombia platinum production is traded on the free market, and the remainder is unreported. Production is from placer deposits located along the eastern margin of the Choco coastal plain, adjacent to the western slope of the Andes Mountains, in the Choco Department. The Colombian placers led world production as late as 1923, but were later surpassed by the rich South African Bushveld Complex. Official production in 1980 was 14,345 troy ounces. This study includes the Choco Pacifico platinum operation, which accounts for about 50 percent of reported Colombian production. Independent small producers in the Choco Department account for most of the remaining reported production. Concessions on the Choco deposit are presently held by Mineros del Choco, a workers' cooperative that acquired the land and equipment from the Choco Pacifico Co. in 1975. The deposit consists of platinum, gold, and silver in Pleistocene and Recent alluvial gravels along the San Juan River. Thickness of the gravel beds ranges from 30 to 50 feet. The operation uses standard bucket line dredges with a trommel-screen-jig circuit for gold and silver recovery. The three dredges currently in operation have a total capacity of about 6.7 million metric tons per year, but production has averaged less than 4.5 million metric tons annually since 1974. Most of the dredges and power generation equipment have been in use for 30 to 50 years and are in need of complete replacement. At present, Mineros del Choco reportedly is bankrupt, and the capital investment necessary to replace aged equipment and prove additional reserves is not likely to be forthcoming. However, several other companies have recently shown an interest in further developing the deposit. CANADA Canadian production of platinum totaled about 162,000 troy ounces in 1980. Although this figure is about 6 percent of world production in that year, Canada's resources account for far less than 1 percent of total known platinum resources in market economy countries. Platinum-group metals in Canada are largely derived as byproducts of nickel-copper production from the Sudbury Complex. Owing to a lack of adequate information at the time of this study, the Sudbury deposits were not analyzed in detail. However, they are discussed in a later section of this report. Lac des lies is an undeveloped deposit located about 55 miles northwest of Thunder Bay, Ontario. It is believed to be a volcanogenic sulfide deposit containing disseminated grains of PGM, along with copper, nickel, and gold. Altogether, nine mineralized zones occurring in an area 1 .7 miles by 0.2 mile have been identified. The deposit contains approximately 22 million metric tons averaging 0.125 oz/t PGM {19, p. 15). For purposes of this analysis it was assumed that the deposit would be open pit mined, with milling done onsite, and that the mill concentrate would be sold to Inco Ltd. at Copper Cliff, located in the Sudbury District. Production was modeled to begin in 1983. UNITED STATES In the United States, platinum is presently recovered mainly as a byproduct of copper refining, but the amount produced is far less than 1 percent of the world total. Properties included in this study are Stillwater PGM Resources' and Anaconda's deposits in the Stillwater Complex in Montana, and Inco Ltd.'s Ely Spruce and AMAX Exploration, Inc.'s Minnamax deposits in the Duluth Gabbro, Minn. All are nonproducers. Also included is the Salmon River placer deposit, in Alaska, which is currently under development. Stillwater Complex The Stillwater Complex is an elongate, tabular assemblage of Precambrian differentiated mafic and ultramafic igneous rock that crops out along the northern front of the Beartooth Mountains in south-central Montana. The Complex is exposed over a length of 25 miles and a width of 1 to 5 miles. The ore-bearing interval is generally continuous throughout the complex and averages approximately 6 feet in thickness. Stillwater PGM Resources, a joint venture partnership between Manville Products Corp. and Chevron Resources Corp., and Anaconda Minerals Co. are the two concerns that control virtaully the entire known Stillwater resource. Buchanan (2, p. 57) has estimated the platinum reserves of the entire complex to be 7 million troy ounces. The two deposits evaluated at the demonstrated level for this analysis contain less than one-third of this total; however, there are believed to be substantial additional tonnages present at a lower level of geologic confidence. The combined platinum and palladium grade of the mineralized zone, common to both deposits, is 0.6 to 0.83 oz/t (9). Both operations will probably use a modified shrinkage stoping method, with ore milled and floated onsite. Several plans dealing with the post-milling treatment of ore have been considered, ranging from construction of a nearby smelter to shipping the nickel-copper concentrates to a nickel-copper smelter and refiner. For this evaluation, the concentrates were assumed to be sent to Inco's Copper Cliff facilities in Ontario and the resulting PGM concentrates to Royston, England. In August 1981, Anaconda submitted an operating permit application to the Montana Department of Lands, proposing to begin development of the Stillwater deposit. Pending approval by the State, full production, planned to be in the range of 30,000 to 35,000 troy ounces of platinum per year (about 3 percent of U.S. requirements), could begin by late 1984 or early 1985 (9). Stillwater PGM Resources reportedly plans to submit a similar plan sometime in early 1982. For purposes of this analysis, both properties have been modeled to begin production in 1984. Salmon River The Goodnews Bay Mine, located along the Salmon River near the Bering Sea in western Alaska, has been the largest producer of primary platinum in the United States. Production from 1934 to 1975 totaled 641 ,000 troy ounces. It has been estimated that the deposit could yield an additional 500,000 troy ounces at a rate of 1 0,000 ounces per year, equivalent to about 1 percent of domestic platinum needs (10). The platinum is located within the lower 6 to 8 feet of an alluvial gravel section that measures up to 250 feet in thickness, and within the upper 3 feet of altered sedimentary and igneous bedrock. Although production ceased after the 1975 season, development resumed on a limited basis in 1980 after the deposit was purchased by Hanson Properties, Inc., of Spokane, Wash. For this analysis, it was assumed that future operations will continue to use dredging as the primary mining method, and the gravels will be milled using gravity methods. The gravels will be sized using a trommel and the resulting fines sent to riffle tables, followed by jigs and sluices. The resulting concentrate was modeled to be sent to Engelhard Industries in New Jersey for refining. There is a possibility of including an underground mining operation on the deep reserve areas if technical difficulties can be overcome. Potential problems include poor roof stability and excessive water. For purposes of this report, only surface operations were considered. Ely Spruce and Minnamax AMAX Exploration's Minnamax and Inco's Ely Spruce deposits in northeastern Minnesota could supply a significant amount of platinum as a byproduct of copper-nickel production. The Minnamax deposit consists of a mineralized zone within the Duluth Gabbro, a large basic intrusion that occurs over more than 100 square miles in the Lake Superior region. The Duluth Gabbro itself has been intruded by a number of other units, one of which is the South Kawishiwi Intrusive. The Ely Spruce deposit consists of a mineralized zone located within this secondary intrusive. Together, the two deposits contain less than 800,000 troy ounces of platinum at the demonstrated level. Ely Spruce would most likely be mined by open pit methods, and Minnamax would be mined by various underground methods, as dictated by the nature of the ore and host rock. For this analysis, concentrates from both deposits have been modeled to be shipped to a smelter- refinery complex to be built in Duluth, Minn. In August 1981. AMAX announced indefinite postpone- ment of its project owing to depressed metal prices (8). Work on Inco's Ely Spruce project was suspended in November 1975, following sinking of a development shaft that was completed in 1968 (11, p. 302). However, in order to be included in this analysis, both deposits were modeled to begin production in 1983. OTHER PLATINUM SOURCES There are several world sources of platinum that were excluded from analysis because of a lack of adequate cost and or engineering data. As information becomes available, these sources will be included in future studies. A few of the deposits are significant or potential producers and therefore merit a brief discussion. REPUBLIC OF SOUTH AFRICA In addition to the Merensky Reef, from which practically all South African platinum is produced, the Bushveld Complex contains the UG2 and Platreef platinum-bearing intervals. Together, the UG2 and Platreef contain 297 million troy ounces of platinum at the demonstrated level. Total platinum reserves in the two units have been estimated by Buchanan (2. p. 56) to be on the order of 597 million troy ounces, calculated to a depth of 3,900 feet. The UG2 is approximately 2 feet thick and located between 330 and 1,155 feet below the Merensky Reef. Its existence has long been known, but exploitation has never been viable, owing to smelting problems associated with the relatively high chromite content. It contains a higher grade of PGM than the Merensky (0.28 oz/t versus 0.25 oz/t), including nearly three times as much rhodium, the most valuable of the PGM (2, p. 56). The high precious-metal grades, substantial tonnage, and accessibility of the UG2 have encouraged several companies to attempt to develop processes to remove chromite prior to smelting or, in the case of Texasgulf Inc., to acquire a process using a high-temperature plasma furnace (3, p. 75). Western Platinum Ltd. has reportedly announced plans to exploit the UG2 using a process developed by South Africa's National Institute of Metallurgy ( 14). However, because of the lack of cost data concerning the project and the fact that the process has not yet been proven on a commercial scale, the UG2 was not included in this analysis. The Platreef interval, which ranges between 0.22 and 0.84 oz t PGM, is located at the contact between the Bushveld Complex and underlying rocks (2, pp. 56, 58). Owing to the disseminated and erratic distribution of mineralization throughout the Platreef, it would most likely be mined by low-cost open pit methods. BRAZIL Brazil's national reserves of platinum, all located in the southern part of the country in the State of Santa Catarina, amount to 766,000 metric tons of measured ore with a grade of 0.09 oz/t. Platinum also occurs in quartzitic and conglomeratic lenses within a metasedimentary Precambrian sequence exposed in the Ouro Branco Mountains, in the State of Minas Gerais. Other occurrences of platinum and PGM are reported in the States of Goias and Rondonia. Irruptive is a large, layered complex of igneous origin. The nickel-copper ores are sulfide-rich portions of a discon- tinuous sublayer at the base of the complex, containing an average of 0.03 oz/t PGM. Platinum constitutes 44 percent of the total PGM content and palladium, 46 percent. Buchanan (2, p. 57) has estimated the reserves of the entire complex to be on the order of 3.4 million troy ounces of platinum and 3.6 million troy ounces of palladium. Inco Ltd. accounts for 80 to 90 percent of Canadian platinum output, primarily from its 1 1 underground mines at Sudbury. Falconbridge Nickel Mines, Ltd., currently operates six underground mines at Sudbury and accounts for the remaining 10 to 20 percent of Canadian platinum production. Inco operates its own smelter-refinery complex at Copper Cliff, Ontario, with final recovery of PGM at Acton, England. Falconbridge smelts concentrates at Sudbury, refines nickel-copper at its plant in Kristiansand, Norway, and either recovers the PGM at its precious metals refinery in Norway, or ships the PGM-rich residue to the Engelhard refinery in Newark, N.J. UNITED STATES In addition to Stillwater, Salmon River, Ely Spruce, and Minnamax, there are numerous other U.S. deposits with reported occurrences of platinum. One of the most notable reported occurrences, the Salt Chuck Mine, was investigated as part of this study but was excluded from the availability analysis because resources at the demonstrated level were considered to be too small to be of major significance at this time. Salt Chuck is a past producer of PGM located on the Kasaan Peninsula of southeastern Alaska. Mineralization occurs in a concentrically zoned ultramafic complex within the Coast Range Intrusive. The mine produced less than 300,000 metric tons ot ore intermittently between 1 907 and 1941. Demonstrated resources are less than 1,000 troy ounces of PGM, while total mineralized material at the inferred level is estimated to be only about 600,000 metric tons at 0.014 oz/t PGM; however, there are believed to be more deposits in the area. SOVIET UNION The Soviet Union produced about 812,500 troy ounces of platinum in 1980, or 28 percent of the world total (table 1). The copper-nickel Noril'sk deposits are believed to account for 90 percent of total Soviet PGM production, and Buchanan (2, p. 57) has reported reserves of platinum to be 50 million troy ounces. Other central economy countries (e.g., Poland) also contain potentially significant platinum resources. ZIMBABWE CANADA Nearly all Canadian platinum production is a byproduct of nickel-copper mining in the Sudbury District. The Sudbury Zimbabwe produced 3,000 troy ounces of platinum in 1980, mainly as a byproduct of nickel production at the Empress Mine. Mineralization occurs within an amphibolite intrusive in Precambrian greenstones of the Bulawayan System. ENGINEERING AND COST EVALUATION After a deposit was selected for inclusion in the analysis, an engineering evaluation was performed. For producing deposits, production rates, capacities, and other pertinent production information were adapted for use in the study. For deposits not yet in production, appropriate mining and concentration methods, production rates, capacities, and other important factors were estimated using standard engineering procedures. As an example, a complete description of mine, mill, and postmill processing for a typical South African operation is given in appendix B. The processes described were also used as a general model for nonproducing deposits. Capital expenditures were calculated for exploration, acquisition, development, mine plant and equipment, and constructing and equipping the mill plant. Capital expendi- tures for the different mining and processing facilities include the costs of mobile and stationary equipment, construction, engineering, facilities and utilities, and working capital. Facilities and utilities (infrastructure) is a broad category that includes, among other things, the costs of the water system, fire protection, roads, fences, and fuel and power distribution. Working capital is a revolving cash fund required for operating expenses such as labor, supplies, taxes, and insurance. Mine, mill, smelting, and refining operating costs, and transportation charges were also calculated for each deposit. The total operating cost is a combination of direct and indirect costs. Direct operating costs include materials, utilities, direct and maintenance labor, and payroll overhead. Indirect operating costs include technical and clerical labor, adminis- trative costs, facilities maintenance and supplies, and research. Other costs in the analysis are fixed charges including taxes, insurance, depreciation, deferred expenses, and interest payments. When possible, actual company cost data were used. If these were not available, the required capital and operating costs were estimated by standardized costing techniques or from a cost-estimating manual prepared for the Bureau of Mines (4). The Bureau's costing system is designed for preparing capital and operating cost estimates through the use of equations, curves, and factors. The system, based on an average of the costs for existing mining operations in the United States and Canada, covers operations of different sizes. Correct use of the costing system will produce a reliable estimate, which historically has fallen within 25 percent of actual costs. These data are then used to perform a financial analysis for each deposit. All capital and operating costs have been adjusted to January 1980 dollars. Foreign deposit costs have been adjusted to U.S. dollars using foreign exchange rates, productivity factors, and inflation rates. These costs reflect the cost of doing business within the foreign country in U.S. dollar equivalents. The Bureau of Mines developed the Supply Analysis Model (SAM) to perform economic analyses that present the results as the commodity price (referred to as "incentive price") needed to provide a stipulated rate of return (7). The rate of return used in this study is the discounted cash flow rate of return (DCFROR), most commonly defined as the rate of return that makes the present worth of cash flows from an investment equal to the present worth of all after-tax investments (23). For this study, a 15-percent DCFROR was considered as a necessary rate of return to cover the opportunity cost of capital plus risk. Prices used for byproducts are shown in table 4. Analyses were conducted in terms of constant January 1980 dollars. No escalation of either costs or prices was included, since it is assumed that any increase in cost would be offset by an increase in price. TABLE 4 Byproduct commodity prices 1 Commodity Price, January 1980 Cobalt .'. $ 25.00 per pound Copper' 1 .00 per pound Gold 1 475.00 per troy ounce Iridium 350.00 per troy ounce Nickel ..'. 3.20 per pound Osmium 152.50 per troy ounce Palladium 1 54.57 per troy ounce Rhodium 800.00 per troy ounce Ruthenium 45.00 per troy ounce Silver 1 12 00 per troy ounce 1 Based on January 1980 commodity prices, except for copper, gold, and silver, which were lowered from temporary highs to more historically realistic longrun prices. ASSUMPTIONS The objective of this analysis was to determine the amount of platinum potentially available from each deposit, regard- less of demand or other market conditions. In order to accomplish this, certain assumptions were used in the evaluation. They are • Preproduction activities for all nonproducing deposits began in January 1980. • Every operation produces at full capacity throughout its life. • Every operation is able to sell all of its output at the determined incentive price. • For operations for which complete processing facilities do not presently exist and are not modeled to be constructed, existing facilities elsewhere have the capacity to handle all products. In this study, the potential availability of platinum is based upon a minimum lead time for nonproducing deposits to achieve full production capacity. The preproduction period allows for the engineering and construction time necessary to initiate production under the proposed development plan. Any additional time lags and potential costs associated with filing environmental impact statements, acquiring permits, financing, etc., have not been included. Tonnage-price data from deposits producing platinum as the primary commodity were aggregated to construct the total and annual availability curves presented in this report. The Ely Spruce and Minnamax deposits, where platinum would be a byproduct, are not included in the availability curves, because platinum recovery from these deposits would be primarily a function of primary product (copper- nickel) prices and production, and, therefore, an economic evaluation that would attempt to calculate a platinum price necessary to exploit these deposits would not be representa- tive of the true availability of platinum from these sources. OPERATING AND CAPITAL COSTS Table 5 shows typical weighted-average operating costs per troy ounce refined platinum for underground mines in South Africa and Zimbabwe, and for placer deposits. Mining costs in South Africa are comparatively cheap, primarily because of the well-defined, predictable nature of the Merensky Reef, which allows for simple, conventional mining methods. In terms of dollars per metric ton of ore mined, mining costs for Zimbabwe would be nearly twice those for South Africa ($21 versus $1 1 ) owing to engineering problems associated with mining the Great Dyke. Total Zimbabwean operating costs would be nearly five times those for South Africa, largely because of the grade difference between the Great Dyke and Merensky Reef. 10 TABLE 5.— Estimated typical operating costs Mine type Ore treated. million metric- tons per year Weighted-average dollars per troy ounce retined platinum Mine Mill Misc.' Refining Total Underground: South Africa . Zimbabwe . . . Placer: U.S. and Colombia 1.3-15 8 .36 12.0-13.0 $151 796 468 $33 107 $28 175 $16 16 27 $228 1.094 501 ' Includes costs related to townsite facilities. Capital costs are an important factor in any decision to develop a mining property. The figures below are approxi- mate capital costs for an underground operation in Zim- babwe treating 360,000 metric tons of ore per year. Costs are in terms of January 1980 dollars per annual troy ounce of refined platinum. Plant and equipment: Mine v > $ 950 Mill 1.100 Smelter-PGM refinery complex 450 Total 2,500 The figures represent capital costs of begining an operation, exclusive of exploration, acquisition, and develop- ment costs. The smelter would produce a nickel-copper matte containing PGM, which would then be refined. For purposes of this analysis, the matte was assumed to be sent to Norway for refining of the copper and nickel, with the resulting PGM-bearing sludge returned to Zimbabwe for final refining into the separate PGM components. The figures above include costs of building a smelter and PGM refinery in Zimbabwe, but the costs associated with building a nickel-copper refinery in Norway are not included. ANALYSIS OF PLATINUM AVAILABILITY TOTAL AVAILABILITY The total amount of recoverable platinum potentially available from the deposits included in the analysis is 200 million troy ounces, more than 99 percent of which is contained in primary platinum deposits. The four producing deposits (three from South Africa, and Choco Pacifico in Colombia) account for 106 million troy ounces, or 53 percent of the total. Figure 5 shows the total amount of platinum potentially available from primary deposits at various incentive price levels. The curve has been smoothed to avoid disclosing proprietary deposit data. There are about 110 million troy ounces of platinum available at the January 1980 producer price of $420 per troy ounce. About 99 percent of the total amount potentially available at $420 is from producing deposits; all platinum mines currently producing could operate profitably at that price. Highly significant in terms of future potential availability is the fact that there would be only a very small amount of additional platinum available between $420 and $1 ,800 per troy ounce. Nonproducing Zimbabwean deposits contain 42 percent of the total platinum recoverable from deposits included on the curve and account for nearly all of the platinum available at relatively high cost. The discovery of 50 I0O I50 200 TOTAL RECOVERABLE PLATINUM, milliontroy ounces FIGURE 5. — Total platinum potentially available from primary platinum deposits. Costs are in January 1980 dollars. Incentive price includes a 1 5-percent rate of return on investments. new economically viable deposits or proven development of the vast resources contained in the UG2 and/or Platreef could have a large impact on the availability of platinum at economically viable prices. The weighted-average incentive price for producing South African properties is $140 per troy ounce, compared with $1,830 for the nonproducing Zimbabwean deposits. The large disparity in price is due principally to the differences in platinum grade between the Merensky Reef and the Great Dyke. The Merensky Reef platinum grade is about 2.5 times higher than that of the Great Dyke. ANNUAL AVAILABILITY The total availability curve has been disaggregated in order to illustrate the annual amount of platinum potentially available at selected incentive prices through the year 2000. For the annual availability curves, it was assumed that all deposits are producing at full capacity, preproduction activities for nonproducing deposits began in 1980, and there are no additions to demonstrated resources through time. Deposits containing byproduct platinum (Ely Spruce, Minna- max) are not included on the curves. The estimated world demand for primary platinum in 1978 was about 2.7 million troy ounces; 40 percent of the demand was from the United States. Annual world demand is expected to increase about 30 percent by 1990 to nearly 3.5 million troy ounces (13). As shown on figure 6, at the January 1 980 producer price of $420 per ounce, an estimated 2.3 million troy ounces of platinum could potentially be produced annually from market economy countries through the year 1992. Over 99 percent of this total would be supplied from mines that are currently producing. At the 1980 level of production, an additional 0.8 million troy ounces could be potentially available annually from deposits in the Soviet Union. In order to meet the 1 990 projected world demand of 3.5 million troy ounces, analyses indicate that platinum price would have to increase. However, owing to the highly regular and predictable nature of the Merensky Reef, South African producers are capable of rapidly expanding production from existing mines in response to increases in demand. This situation could result in resisting further price increases. Furthermore, it is expected that nonproducing deposits, particularly the UG2 and Platreef, will be brought into production within the near future. The annual availability curve indicates that the amount of metal potentially available at $420 per troy ounce could decrease by about 44 percent (from 2.3 million to 1 .3 million troy ounces) by 1993. This decrease is due largely to the 11 4 1 1 1 I u _^~ ■ . 3 O f / / \ o £ 3 \ _ c o \ / ..... $2,000 h ? 3 2 2 Y - h- < y 0. V $1,500 liJ \ _J OD $ 420 < . rr I Ul > o o UJ 19 i i i 80 1984 1988 1992 1996 20C YEAR FIGURE 6. — Annual platinum potentially avail- able at selected incentive prices. Costs are in January 1980 dollars. Incentive price includes a 1 5-percent rate off return on investments. cessation of production from the Impala deposit, which currently accounts for 40 percent of South African produc- tion. However, Impala, like other South African operators, has the potential to expand operations into other resources (e.g., UG2) of the Bushveld Complex. AVAILABILITY FROM DOMESTIC SOURCES The total amount of recoverable platinum potentially available in the U.S. deposits analyzed is about 2.3 million troy ounces, about one-third of which is contained in the Ely Spruce and Minnamax deposits, which would produce byproduct platinum. At assumed production capacities, U.S. deposits are capable of producing 113,000 troy ounces annually. However, in 1990, none can be produced from primary platinum deposits at the January 1 980 price of $420, and large quantities of imports will be required to meet projected domestic needs. Of course, the total amount of platinum potentially available could be substantially higher if currently unknown platinum-bearing deposits come into production or if the deposits analyzed prove to contain additional demonstrated resources. The amount of platinum available on an annual basis (regardless of price) could be different if production levels are higher or lower than those assumed for this analysis. ANALYSIS OF FACTORS AFFECTING PLATINUM AVAILABILITY IMPACT OF BYPRODUCT CREDITS The incentive price for a deposit is partially a function of the marketability of the byproduct commodities and the prices received for them. The following values represent weighted- average percentages of revenues derived from the various products from South African and Zimbabwean deposits, based on the byproduct prices shown in table 4 and $420 per troy ounce for platinum: South Africa Zimbabwe Platinum 58 41 Palladium 9 15 Copper-nickel 21 44 Other 12 South African operators are less dependent on byproduct revenues than Zimbabwean operators would be if they were able to operate economically at these prices, or if commodity «J 3,600 c ° 3,200 g ^2,800 = 2,400 o UJ 2,000 o rr °- 1,600 UJ > H 1,200 z UJ Z 800 5 2 400 No byproduct credits *-i Bose case a 50 100 150 200 TOTAL RECOVERABLE PLATINUM, million troy ounces FIGURE 7 — Potential platinum availability, without byproduct credits. Costs are in January 1980 dollars. Incentive price includes a 15- percent rate of return on investments. prices were to remain constant relative to each other but increase to a level at which the deposits could be operated economically. One of the assumptions inherent in the availability curves is that each operation is able to sell all of its output at the determined incentive price for the primary commodity. A necessary element of this assumption is the condition that each operation is able to sell all of its byproducts at the assigned prices. The inability of potential Zimbabwean producers to market byproduct commodities, especially copper and nickel, could have a severe impact on platinum production. In order to quantify the effects of byproduct revenues on total potential platinum availability, all credits from these commodities were eliminated. The results are shown in figure 7. Only one deposit, Salmon River, could produce at $420; total availability at that price decreased to virtually nothing. The weighted-average incentive price for South African producers increased to $475. Were byproduct prices to increase by 25 percent, the total amount of platinum available at $420 would increase by only 6 percent. IMPACT OF OPERATING AND CAPITAL COSTS In order to assess the impact on potential platinum availability of increases in costs of production such as energy, materials, and labor, operating costs for all deposits were increased by 25 percent. The weighted-average incentive price for producers increased by $60. Although this change represents a 43-percent increase, producing deposits remained substantially below the January 1980 producer price of $420, and total platinum available at that price was not affected. The weighted-average incentive price for Zimbabwean deposits increased by 20 percent, to nearly $2,200. Figure 8 shows the resulting availability curves for deposits in the "low cost" portion of the total availability curve. Capital costs, including initial investments and periodic reinvestments, account for only a small percentage of total production costs. Therefore, even large increases in capital costs would have an insignificant effect on availability. This is 12 c 1 I 400 - 1 1 1 — // // / / 25-perce 1 ost >ncreose^ / / / / ~Bose cose - ■ 1 1 — a 25 50 75 100 125 TOTAL RECOVERABLE PLATINUM, milbon troy ounces FIGURE 8. — Platinum potentially available from "low cost" deposits, with a 25-percent operating cost increase. Costs are in January 1980 dollars. Incentive price includes a 15-percent rate of return on investments. true even for nonproducing deposits, for which complete mine and processing facilities would be constructed. In fact, a 25-percent capital cost increase for Zimbabwean deposits resulted in only 3-percent increase in incentive price. SECONDARY SUPPLY FROM U.S. SOURCES The quantity of platinum recycled in the United States was approximately 687,000 troy ounces in 1980 (16, p. 3). This figure could rise dramatically if recycling of automobile catalytic converters becomes commercial on a large scale. The U.S. automobile industry accounted for 46 percent of U.S. platinum consumption in 1980 (16, p. 6), and it has been estimated that, by the late 1980s, about 300,000 troy ounces of platinum will become available annually in the United States in the form of spent converters (21). Were the domestic automobile industry to become, in effect, nearly self-sustaining in terms of platinum production and consump- tion through recycling of catalytic converters, U.S. demand for platinum from primary sources could decrease dramati- cally. TECHNOLOGICAL ENHANCEMENTS In view of Western Platinum Ltd.'s announced plans to exploit the UG2 of the Bushveld Complex using a process developed by South Africa's Institute of Metallurgy, the problem of high chromite content of UG2 ore may have been solved. Other South African producers are known to be developing their own processes to treat UG2 ore. The commercial feasibility of any of these processes could have a dramatic impact on the availability of platinum. The Bureau of Mines research program is oriented toward alleviating dependence on imports of critical and strategic minerals. The Bureau's Reno (Nev.) Research Center has devised a method for preparing a platinum-palladiumi concentrate from Stillwater ore. Flotation using mercap- tobenzthiazole (MBT) collector in sulfuric acid resulted in a final concentrate grade as high as 28.5 oz/t (1). A pilot flotation mill, processing ore at approximately 100 pounds per hour, will be operated to provide data for an evaluation of the technical and economic feasibility of the flotation process. Future research will focus on improvement on the two-stage leaching technique for extraction of PGM. The Twin Cities (Minn.) Research Center has conducted differential flotation studies to separate the copper and nickel content of the Duluth Gabbro sulfides into their respective fractions. About 50 percent of the PGM are being recovered in the nickel fraction. Platinum-group metals occur in minute quantities and are difficult to analyze. Future research on recovery of PGM from the nickel-rich matte by leaching process is planned for 1 982. The Albany (Oreg.) Research Center has conducted detailed beneficiation of potential platinum resources from four Alaskan deposits: Salt Chuck, Snettisham, Union Bay, and Yakobi Island (6). Gravity concentration, magnetic separation, and froth flotation procedures for each deposit were developed. The Avondale (Md.) Research Center and the Defense Property Disposal Service (U.S. Department of Defense) initiated programs to recover precious metals from electronic scrap. The Avondale Research Center plans to produce precious metal-bearing concentrates from electronic relays, switches, and automotive components. Development of hydrometallurgical procedures for selectively removing base metais from the concentrates is also part of this research program. SUMMARY AND CONCLUSIONS Major known platinum deposits located in five market economy countries were chosen for detailed analysis. The total amount of recoverable platinum potentially available at the demonstrated resource level from those deposits that do or would produce platinum as the primary product is about » 200 million troy ounces, with producing mines accounting for slightly greater than half of the total. The amount of platinum potentially available at or below the January 1980 producer price of $420 per troy ounce is about 1 10 million troy ounces. South African producing mines, which contain nearly 53 percent of the recoverable platinum in the deposits analyzed, have a weighted-average incentive price of only $140 per troy ounce. Zimbabwean deposits, all of which are nonpro- ducing, contain 42 percent of the total and have a much higher incentive price, averaging $1 ,830. In terms of dollars per troy ounce of refined platinum, Zimbabwean operating costs are about five times those for producing South African deposits. The large difference in cost is due principally to grade differences between the Merensky Reef and the Great Dyke. At a price of $420, an estimated 2.3 million troy ounces of platinum could be produced annually from market economy countries. This quantity appears adequate to meet current demand; however, an increase in platinum price may be required in order to satisfy the projected 1990 world demand of 3.5 million ounces. In 1990, there will be no known U.S. primary platinum deposits capable of providing platinum at $420 per ounce. Regardless of price, at assumed production capacities, domestic deposits are capable of providing less than 10 percent of annual U.S. requirements, and large quantities of imports will be necessary to satisfy projected demand. There are several situations that could offset a significant increase in price at projected platinum demand levels: • Currently, South African producers appear capable of rapidly expanding production from existing mines in re- sponse to increases in demand. • Although production costs are not yet known, commer- cial development of the UG2 and Platreef of the Bushveld Complex would more than double the amount of platinum available from South African deposits. • Improvements in processing technology could increase the amount of platinum available in producing or currently unexploitable deposits. • Discovery of new deposits could also contribute to potential supplies. • Expanded recycling efforts (e.g., of automobile catalytic converters) could significantly reduce demand for platinum from primary sources. 13 BIBLIOGRAPHY 1. Bennetts, J., E. Morrice, and M. M. Wong. Preparation of Platinum-Palladium Flotation Concentrate From Stillwater Complex Ore. BuMines Rl 8500, 1981, 18 pp. 2. Buchanan, D. L. Platinum — Great Importance of Bushveld Complex. World Min., v. 33, No. 9, August 1980, pp. 56-59. 3. Chemical Engineering (New York). Plasma Process Is Ready for Metals Recovery. V. 86, No. 5, February 1979, pp. 75-79. 4. Clement, G. K., Jr., R. L. Miller, P. A. Seibert, L. Avery, and H. Bennett. Capital and Operating Cost Estimating System Manual for Mining and Beneficiation of Metallic and Nonmetallic Minerals Except Fossil Fuels in the United States and Canada. BuMines Special Pub., 1980, 149 pp. Also available as: STRAAM Engineers, Inc. Capital and Operating Cost Estimat- ing System Handbook — Mining and Bneficiation of Metallic and Nonmetallic Minerals Except Fossil Fuels in the United States and Canada. Submitted to the Bureau of Mines under contract JO255026, 1977, 374 pp.; available from the Minerals Availability Field Office, Bureau of Mines, Denver, Colo. 5. Compania Minera Choco Pacifico S.A. 1973 Annual Report. 23 pp. 6. Dahlin, D. C, A. R. Rule, and L. L. Brown. Beneficiation of Potential Platinum Resources From Southeastern Alaska. BuMines Rl 8553, 1981, 14 pp. 7. Davidoff, R. L. Supply Analysis Model (SAM): A Minerals Availability System Methodology. BuMines IC 8820, 1980, 45 pp. 8. Duluth News Tribune. AMAX Cuts Babbitt Staff. Aug. 1 1 , 1 981 , sec. A, p. 2. 9. Engineering and Mining Journal. Anaconda Begins Permit Process for Stillwater Complex. V. 182, No. 9, September 1981, p. 11. 10. Spokane Industralist Re-Opens Only U.S. Plantinum Mine. V. 182, No. 7, July 1981, p. 41. 11. Financial Times (London). Mining International Yearbook. 1981, 714 pp. 12. Harrison, M. W. Mineral Resources of Zimbabwe — An Overview. Economic Conference, Zimbabwe Promotion Council, September 1980, 17 pp.; available upon request from T. F. Anstett, Bureau of Mines, Denver, Colo. 13. Jolly, J. H. Platinum-Group Metals. Ch. in Mineral Facts and Problems, 1980 Edition. BuMines Bull. 671, 1981, pp. 683-706. 14. Mining Journal (London). Western Platinum to Exploit UG2 and Expand Production by 50,000 Oz/Yr. Mining Annual Review 1981, p. 473. 15. Mohide, T. P. Platinum Group Metals— Ontario and the World. Ontario Ministry of Natural Resources, Ontario Mineral Policy Background Paper No. 7, 1979, 162 pp. 16. Moore, C. M. Platinum-Group Metals in the First Quarter of 1981. BuMines Mineral Industry Surveys, June 22, 1981, 11 pp. 17. National Materials Advisory Board (National Research Coun- cil). Supply and Use Patterns for the Platinum-Group Metals. 1980, 198 pp; available from NTIS, PB 80-179088. 18. Newman, S. C. Platinum. Inst. Min. and Met., Trans., Sec. A, v. 32, No. 797, pp. A52-68. 19. Northern Miner. Platinum Find Has Tonnage Potential. V. 60, No. 32, Oct. 24, 1974, pp. 1, 15. 20. Page, N. J., and J. C. Dohrenwend. Mineral Resource Potential of the Stillwater Complex and Adjacent Rocks in the Northern Part of the Mount Wood and Mount Douglas Quadrangles, Southwestern Montana. U.S. Geol. Survey Circ. 684, 1973, 9 pp. 21. Rosso, J. P. (Gemini Industries, Inc.). Private communication, 1982; available upon request from T. F. Anstett, Bureau of Mines, Denver, Colo. 22. Stanton, R. L. Ore Petrology. McGraw-Hill Book Co., New York, 1972, 713 pp. 23. Stermole, F. J. Economic Evaluation and Investment Decision Methods. Investment Evaluations Corp., Golden, Colo , 2d ed 1974,449 pp. 24. U.S. Bureau of Mines. Platinum-Group Metals. Sec. in Mineral Commodity Summaries 1982. P. 114. 25. U.S. Geological Survey and U.S. Bureau of Mines. Principles of a Resource/Reserve Classification for Minerals. U.S. Geol. Survey Circ. 831, 1980, 5 pp. 14 APPENDIX A.— CENTRAL ECONOMY COUNTRIES Albania Bulgaria China Cuba Czechoslovakia German Democratic Republic Hungary Kampuchea Laos Mongolia North Korea Poland Romania Soviet Union Vietnam APPENDIX B.— MINING AND PROCESSING OPERATIONS FOR A TYPICAL SOUTH AFRICAN PROPERTY Owing to the highly regular and predictable nature of the dip. strike, thickness, and grade of the ore zone, the Merensky Reef of the Bushveld Complex is mined using simple, straightforward methods. Access to ore can be gained from inclined and vertical shafts. The mechanized longwall mining method is being introduced in most of the mines, taking the place of labor-intensive hand loading of blasted ore, and hand tramming. Face scrapers move ore to gullies where the ore is loaded into cars pulled by battery locomotives, which discharge their load into passes. From the passes, the ore is hoisted to the surface. The processing of ore to matte is shown in figure B-1 . The mined ore is hoisted to the surface, crushed, ground, and then floated to recover the metallic minerals. In addition to flotation, some mining companies include gravity separation of a metallic mineral fraction, which is sent directly to the precious metals refinery. The flotation concentrate, which in some cases is pelletized, is sent to an electric furnace where it is partially oxidized to a "green matte." The "green matte" is transferred to converters where, under an oxygen-rich atmosphere, it is melted and oxidized, resulting in a "white matte." The "white matte" is sent to a nickel-copper refinery, and the resulting sludge from nickel-copper refining is sent to a precious metals refinery for further separation. The precious-metal concentrates from the nickel and copper refinery stages have a PGM content which normally ranges between 25 percent and 75 percent. The first stage of the refining process is the removal of the base metal content by roasting, followed by leaching processes. The individual PGM are then separated and refined by chemical methods. Until the early 1970's, the refinery methods most commonly used were variations of the so-called "convention- al process," based on selective precipitation methods. In recent years, however, new techniques based on solvent extraction and ion-exchange methods have been developed. These are replacing the older methods on a commercial scale. CONVENTIONAL PROCESS A simplified flowsheet illustrating the sequence of steps involved in the conventional process is given in figure B-2. In the initial step, aqua regia (3 volumes concentrated HCI combined with 1 volume concentrated HN0 3 ) is used to dissolve Pt, Pd, and Au, as chlorides, leaving an insoluble residue containing the other PGM. Gold is removed from this solution by the addition of a reducing agent, while platinum is precipitated as an impure platinum sponge, which is then redissolved and purified by bromate hydrolysis. Palladium is then removed from the filtrate by a process similar to that for platinum. The treatment comprises the formation of dichlorodiammine palladium (II) and its precipita- tion as a high-purity salt, which is then ignited to form palladium sponge. The insoluble residue remaining after initial aqua regia extraction of the platinum and palladium is blended with available concentrates high in Rh, Ir, and Ru. This mix is then ■ smelted with lead oxide ore, lead carbonate, fluxes, and carbon, and the molten charge is poured into a conical mold. After solidification, the slag is removed and the lead (which contains the PGM) is melted and granulated. A parting process with HN0 3 dissolves most of the Ag, Pb, and any residual Pt and Pd. The filtrate is then treated for the recovery of each of these metals. The insoluble residue from the HN0 3 treatment consists of Rh, Ir, Ru, and Os in concentrated form. The subsequent separation of these metals involves a much more elaborate series of steps that commonly includes fusion, leaching, numerous chemical purification stages, precipitation, roast- ing, oxidation or reduction, and frequently includes (for Os, Ru, and Ir) ignition and reduction by hydrogen to yield metal. All of the filtrates and insoluble residues generated in the separations for the individual metals are recycled to the appropriate stages earlier in the process so that metal losses at this stage are very small. However, some osmium is usually lost as the volatile tetraoxide during smelting or lead granulation operations. Although the processes vary in detail from plant to plant, those using the conventional procedures all employ methods based on selective precipitation techniques from a mixed PGM solution. In terms of the degree of separation achieved, however, these are generally rather inefficient, and even where conditions are favorable for complete removal of the desired elements from solution, the precipitate is contamin- ated by entrained mother liquor and requires thorough washing. This tends to lead to multiple purifications involving redissolution and reprecipitation, which, in turn, result in the need for increased recycling and retreatment. Recoveries achieved during the refining stage are typically about 98 percent. However, the percentage of metal recovered in refined form is only 75 percent of the assayed PGM content of ore, while nickel and copper recoveries are approximately 79 percent. The reason for these discrepan- cies is that substantial portions of the total metal content are lost during the milling and earlier nickel-copper concentration stages. SOUTH AFRICA NATIONAL INSTITUTE OF METALLURGY (OPNIM) PROCESS A new process for the separation of PGM and gold based on solvent extraction was developed by the South African National Institute of Metallurgy (NIM) in the early 1970's. A generalized flowsheet for the OPNIM process is given in figure B-3. A commercial plant has now been installed at the Lonrho platinum refinery at Brakpan, South Africa, for extraction of Rh, Ru, Ir, and Os. Impala Platinum is also understood to be using the process, and Matthey Rustenburg Refiners is carrying out research independently on solvent extraction techniques. The solvent extraction process requires preleaching treatments that result in the easy separation of the base metals by selective leaching and allows complete dissolution 15 Mining I Ore size reduction crushing and grinding Froth flotation »■ Gravity concentration PGM refining of heavies, 30-40% PGM Thickening, filtering, drying Pelletizing Smelting I Converting ~~t — Cooling T Mat t e size reduction crushing and grinding I Magnetic separation Pressure leaching to give PGM residue T Pressure leaching nonmagnetics Electrowinning PGM residue refining 60% PGM FIGURE B-1 Flowsheet, South African PGM ore processing. I Preleaching treatment I Aqua regia leaching ■* > Insolubles i Leach liquor f Gold sponge Gold precipitation i Platinum precipitation * | Platinum purification ♦ Palladium precipitation Impure "salt" 1 i ' Zinc reduction Palladium purification 1 Effluent Lead smelting 1 Slags Lead and Parting in HN0 3 and aqua PGM Pt, Pd, Au 1 regia 1 eacmng Peroxide fusion : RuO«, Os0 4 Acidification I Ruthenium-osmium separation , t. , ; Iridium purification | Ruthenium and osmium distillation I Iridium precipitation Rhodium purification and precipitation FIGURE B-2. — Flowsheet, conventional refining process. 16 Preleaching treatments f Base-metal leaching Base-metal effluent J Leaching with CI and HCI Low-PGM residue Gold extraction Stripping Metal precipitation ~l 1 Palladium extraction Stripping "Salt" precipitation 1 Platinum extraction Stripping Salt precipitation - Osmium distillation Os0 4 Salt precipitation - J 1 Ruthenium extraction • Stripping - Salt precipitation "1 _ J Iridium extraction Stripping Salt precipitation H 1 Rhodium extraction Stripping "Salt" precipitation — 1 Fin al barren effl uent FIGURE B-3.— Flowsheet, OPNIM refining process. of the PGM by a single CI-HCI leach. The process dissolves at least 99 percent of the primary PGM (Pt and Pd) and Au, and 95 percent of the remaining PGM. The solvent extraction steps are carried out in continuous countercurrent operations. Metal recoveries from stripping solutions are accomplished by the precipitation of insoluble salts or complexes, except in the case of gold. Palladium is recovered prior to platinum, and osmium is recovered by a distillation process rather than by solvent extraction. The main advantages of the OPNIM process are the savings in terms of time and cost. The processing time for recovery of secondary PGM is reduced from between 4 and 6 months for the conventional process to 20 days for the OPNIM process, and 99.95 percent purity of products is achieved. In addition, the process uses 20 percent of the labor of the conventional process because the techniques are simpler; they can be more highly automated and involve less reprocessing. Finally, it is reported that the capital costs of equipment required for the extraction of the secondary PGM with OPNIM may be reduced by 50 percent in comparison with those for the conventional process. fcU.S. Government Printing Office : 1982 - 383-747/8693 *°o V V^V V 1 ** A* • fife*' ^ c* * o, *o . , * A <. ♦7777 • .0*" >». '• • » * J\ <. *' 7VV* ^0 V V ..J^Lv > »* »1"»* *> ^ *-• .v • ^K2« «? v ^, oVJ&W* av ?v •«■»• «? "£, oV^aAy* av ***, 'Kara* «? "3> •v.J&Ay* 4 9 *'\ -zmt; «►♦'-«. °*. *• T* 4> J? .*;v^. ** ^ . % ^fci-- ^ .** ^wa\ ^ Water- ^ .** .v&ta-. ** .*♦ /■ *^* v • V4* • W • "fev* ■r.-y* V^'/ v^'V ' %"^ f v K ^°- ^9 ^ ^ A* ^ • a v *^ - ' 4 'fr ■# •"