Metals, Energy and Sustainability the Story of Doctor Copper and King Coal Metals, Energy and Sustainability Barry Golding • Suzanne D

Barry Golding Suzanne D. Golding

Metals, Energy and Sustainability The Story of Doctor Copper and King Coal Metals, Energy and Sustainability Barry Golding • Suzanne D. Golding

Metals, Energy and Sustainability The Story of Doctor Copper and King Coal

123 Barry Golding Suzanne D. Golding Sherwood, QLD School of Earth and Environmental Sciences Australia University of Queensland Brisbane Australia

ISBN 978-3-319-51173-3 ISBN 978-3-319-51175-7 (eBook) DOI 10.1007/978-3-319-51175-7

Library of Congress Control Number: 2016962033

© Springer International Publishing AG 2017 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Cover illustrations: NOAA Okeanos Explorer Program, Galapagos Rift Expedition 2011; Interior of one of the copper mines on the Paris Mountain: 1792 by John Warwick Smith. By kind permission of the National Library of Wales; Moche smiths smelting, at the National Museum of Archaeology and Anthropology in Lima, Peru, photograph by kind permission of Nathan Benn. See also Figures 2.6, 3.19 and 3.52 from this book.

Printed on acid-free paper

This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Acknowledgements

Metals, Energy and Sustainability—The Story of Doctor Copper and King Coal evolved as we visited copper and coal mining sites throughout the world. Our initial training was gained at Mount Morgan where the then General Manager Loy Hennessy encouraged a scientific approach to the mining and extraction of copper. We are deeply indebted to the people who enabled our visits to Timna in Israel, Parys Mountain and the staithes of Newcastle on Tyne in Great Britain, the Rio Tinto mine and museum in Spain, Bingham Canyon and North Antelope Rochelle Mine mines in the United States and finally Chuquicamata in Chile. We are also grateful to the many organisations acknowledged throughout the book that kindly agreed to our using their material. Information has been sourced widely; however, special thanks are due to Daniel Edelstein from the U.S. Geological Survey and Prof. Rod Eggert from the Colorado School of Mines. In Great Britain, Prof. Paul Younger kindly provided ideas and relevant places to visit. Professor Harry Campbell not only read and edited the final chapter but was also the driving force behind the paper and Ph.D. on which Chap. 4 is based. John Reid and Brian Warner who worked with us at Mount Morgan gave valuable technical input into the metallurgy sections of the book. Nevertheless, any mistakes are our own and we would be grateful to receive advice on any errors spotted or necessary corrections. Thank you Petra van Steenbergen from Springer who supported our proposed book and the continuing support of Springer when deadlines came and went. Finally, the book was only possible with the encouragement and support from our son Laurence who kept our computers operational over the years it has taken to research and write this book.

v Contents

1 Doctor Copper and King Coal ...... 1 1.1 Why Doctor Copper and King Coal ...... 1 1.2 Copper: A Most Useful Metal ...... 3 1.3 Properties that Make Copper Desirable...... 8 1.4 Coal: The Energy for the Industrial Revolution ...... 13 1.5 Our Current Use of Coal ...... 14 1.6 The Essential Properties of Coal...... 15 References...... 19 2 Copper and Coal Resources ...... 21 2.1 Copper Mineralogy and Formation ...... 22 2.2 Coal Composition and Formation ...... 30 References...... 35 3 Copper and Coal Through the Ages ...... 37 3.1 Earliest Metalsmiths ...... 38 3.2 Out of the Dark Ages ...... 66 3.3 The First Modern Mines...... 92 3.4 The Age of Electricity ...... 110 3.5 The Mega Mines ...... 136 References...... 146 4 The Future for Copper and Coal ...... 157 4.1 Consumption of Copper and Coal Since 1940 ...... 157 4.2 Deﬁning Sustainability ...... 158 4.3 Sustainability Predictions Based on Reserves ...... 161 4.4 Evaluating the Sustainability of Copper and Coal...... 170 4.5 Sustainability of Coal Production ...... 173 4.6 Sustainability of Copper Production ...... 176 4.7 Conclusions ...... 182 References...... 183 Glossary...... 187 Index ...... 191

vii About the Authors

Barry Golding commenced his mining career in the Dawson Valley Colliery in 1965. The mine supplied coal for the Mount Morgan Limited copper reverberatory furnace. He accepted a cadetship with Mount Morgan Limited in 1966 and gained his degree in mining engineering in 1968. Barry worked in metalliferous and coal mining in Australia and gold and chrome mining in South Africa and has worked continuously in the mining industry from 1968 onward apart from two years in the army that included a year in Vietnam. Barry completed a Postgraduate Diploma of Applied Economics followed by a Master of Economics by Research in 2002. The focus for both degrees was applying cost–beneﬁt and cost-effectiveness techniques to resource management incorporating risk analysis and sustainable development. In 2011 he was awarded his Ph.D. in Economics at the University of Queensland for his thesis entitled Metals Energy and Sustainability.

Suzanne D. Golding commenced her geology career at the Mount Morgan gold– copper mine in 1967. She graduated from the University of Queensland in 1970 with a ﬁrst class honours degree in geology. Suzanne worked in mineral exploration and production in Australia and gold and coal mining in South Africa through the 1970s. In 1982 she was awarded a Ph.D. in Geochemistry from the University of Queensland for work on gold mineralisation in the Kalgoorlie-Norseman region, Western Australia. Suzanne has worked continuously at the University of Queensland since 1982 and is currently Professor in the School of Earth and Environmental Sciences. She has published more than 160 journal articles and book chapters and edited a pioneering text on coal seam gas entitled Coalbed Methane: Scientiﬁc, Environmental and Economic Evaluation.

ix List of Figures

Figure 1.1 Old King Coal’s crown in danger—Library of Congress ...... 5 Figure 1.2 The Iceman’s copper axe: Photo credit: South Tyrol Museum of Archaeology—www.iceman.it ...... 6 Figure 1.3 Known world copper production—After (Schmitz 1979) ...... 7 Figure 1.4 Various copper alloy colours—courtesy of the International Copper Association, Australia ...... 12 Figure 1.5 Consumption of copper by the major consuming countries...... 12 Figure 1.6 World coal consumption by sector...... 14 Figure 2.1 Hydrothermal fluids venting from chimneys on the seafloor (Pacific Ring of Fire 2004 Expedition— NOAA Office of Ocean Exploration; Dr. Bob Embley, NOAA PMEL, Chief Scientist) ...... 23 Figure 2.2 Typical copper orebody with secondary minerals resulting from weathering processes ...... 24 Figure 2.3 The Ring of Fire (courtesy U.S. geological survey) ...... 25 Figure 2.4 Atacamite from the oxide zone at Chuquicamata mine, Chile ...... 26 Figure 2.5 Native copper infilling cavities in basalt, Wolverine Mine, Michigan, USA (courtesy of James St John—CC-BY-2). . . . 28 Figure 2.6 Giant tube worms surrounding hydrothermal vents (NOAA Okeanos Explorer Program, Galapagos Rift Expedition 2011) ...... 28 Figure 2.7 Burial, compaction and coalification forms coal from peat (courtesy of Stephen Greb, Kentucky Geological Survey and University of Kentucky) ...... 31 Figure 2.8 The genus Glossopteris was a seed bearing plant with fern-like foliage; the name refers to the fossilised leaves that are common in Permian coal deposits (courtesy Daderot CC-zero—Exhibit in the Houston Museum of Natural Science, Texas, USA) ...... 32

xi xii List of Figures

Figure 2.9 Macroscopic classification of bituminous coal showing the Australian (SA) and International Committee for Coal and Organic Petrology (ICCP) schemes (courtesy Joan Esterle, University of Queensland) ...... 33 Figure 3.1 See Hansen (2013) for pictures of axes found at Mersin (image from Google Earth) ...... 40 Figure 3.2 Malachite (Golding 1999)...... 41 Figure 3.3 Chalcopyrite (Golding 1999)...... 43 Figure 3.4 Cassiterite (Golding 1999) ...... 44 Figure 3.5 Secondary copper mineralisation—image provided courtesy of Eurasian Minerals Inc...... 46 Figure 3.6 Metal workers using blowpipes (Duell 1938). Courtesy of the Oriental Institute of the University of Chicago ...... 48 Figure 3.7 Timna and Feynan locations (image from Google Earth) ...... 50 Figure 3.8 Crown and a sceptre from the Nahal Mishmar hoard Photographs by Clara Amit © Israel Antiquities Authority and Hecht Museum in Haifa. Credit John Bedell ...... 51 Figure 3.9 Mining method at Timna (with permission of the Dead Sea & Arava Science Centre) ...... 53 Figure 3.10 Timna shaft showing metal tool marks and protruding steps ...... 54 Figure 3.11 Recently re-excavated shaft in the foreground with plates behind ...... 55 Figure 3.12 Near horizontal hat demonstrates the strong wind on a hill overlooking the modern mine...... 56 Figure 3.13 An example of the smelting furnaces excavated at Timna . . . 57 Figure 3.14 Doors for the temple of Amen at Karnak modified from Plate XVIII (Newberry 1900) ...... 58 Figure 3.15 Estimated annual copper production from Hong et al. (1996) ...... 62 Figure 3.16 Rückseite des Annaberger Bergaltars by Hans Hesse 1522 Wikimedia Commons, the free media repository ...... 63 Figure 3.17 Fire-setting (Agricola 1556) credit Dover Publications, Inc...... 68 Figure 3.18 Ming Dynasty three pot furnace after Zhang (1986) ...... 73 Figure 3.19 Interior of one of the copper mines on the Paris Mountain: 1792 by John Warwick Smith. By kind permission of the National Library of Wales ...... 74 Figure 3.20 Anhalt-Dessau 1694 silver medal...... 76 Figure 3.21 Drilling an upper at East Pool Mine c. 1895—photographer J.C. Burrow: © The British Library Board, 7105.e.21, item 24 ...... 77 List of Figures xiii

Figure 3.22 Reverberatory furnace ...... 78 Figure 3.23 The Welsh copper smelting process (Symons 2003)...... 80 Figure 3.24 Parys or Anglesey Penny ...... 81 Figure 3.25 Annual copper production (Schmitz 1979; Symons 2003) . . . 83 Figure 3.26 The staithes at Wallsend (Hair and Ross 1844) ...... 90 Figure 3.27 Colliery Gin by Edward Whymper c. 1790s ...... 91 Figure 3.28 Roman water wheel, found in South Lode (Rio Tinto Mines, Spain), 1919, courtesy of the Historic Archive of Río Tinto Foundation...... 95 Figure 3.29 Rio Tinto Roman water wheels—after Palmer (1928)...... 96 Figure 3.30 Map of the Rio Tinto catchment—after Olías and Nieto (2015) ...... 97 Figure 3.31 Open air roasting at Rio Tinto (Nash 1904) ...... 99 Figure 3.32 Rio Tinto near the mine and near to the Gulf of Cadiz. . . . . 107 Figure 3.33 Replicas of Columbus’s ships and the Muelle de Minerales in Huelva ...... 107 Figure 3.34 World coal production 1800–1900—(Rutledge 2011; Day 1904; Walcott 1901) ...... 109 Figure 3.35 The ancient mine-pits of Point Keweenaw, Michigan (Whittlesey 1863) ...... 111 Figure 3.36 Copper fish hooks, courtesy of the Milwaukee Public Museum ...... 112 Figure 3.37 Bingham and Salt Lake City (USGS 1885) ...... 115 Figure 3.38 Bingham and Carr Fork Canyons (USGS 1900) ...... 116 Figure 3.39 Tipping some 300 tonnes of ore into the in-pit crusher. Courtesy Kennecott Utah Copper...... 124 Figure 3.40 Flash smelting furnace—after Mäkinen (2006), King (2007), Firdu (2009) ...... 125 Figure 3.41 Primary energy usage (Rio Tinto 2006) ...... 127 Figure 3.42 Concentrator energy inputs (Rio Tinto 2006)...... 127 Figure 3.43 Block caving (McGraw-Hill 2007)...... 128 Figure 3.44 2013 Manefay fault landslide (Sutherlin 2014) ...... 129 Figure 3.45 Copper price and world production (Porter et al. 2015) . . . . 130 Figure 3.46 SX-EW flow sheet with permission of Professor James Myers (Myers 2010)...... 131 Figure 3.47 Major coal-producing countries in the twentieth century . . . . 132 Figure 3.48 UK coal production and employment 1914–2000...... 133 Figure 3.49 World coal, oil and gas production in million tonnes coal equivalent. Sources Rutledge (2011), BP (2014a), Benichou (2015), U.S. Energy Information Administration (2015) ...... 134 Figure 3.50 US, Chile and total world copper mined...... 135 Figure 3.51 US copper mining and concentrating—energy used per kilogram (Golding and Campbell 2014) ...... 136 xiv List of Figures

Figure 3.52 Moche smiths smelting—by kind permission of Nathan Benn who photographed pottery from the collection of the National Museum of Archaeology and Anthropology in Lima, Peru ...... 138 Figure 3.53 War of the Pacific: adapted from Wikipedia (2014b) ...... 139 Figure 3.54 Chilean independence leader Bernardo O’Higgins ...... 142 Figure 3.55 Copper cathodes passing through Calama en route to port ...... 145 Figure 3.56 Energy consumption versus copper ore grade (Comision Chilena del Cobre 2014, 2015) ...... 146 Figure 4.1 World population, copper and coal consumption 1940–2015 ...... 159 Figure 4.2 World coal and copper consumption per person...... 159 Figure 4.3 U.S. crude oil production (EIA 2016)...... 164 Figure 4.4 World oil production and intensity (BP 2016b) ...... 165 Figure 4.5 Fossil fuel energy consumption (The Shift Project 2016; Rutledge 2011; BP 2016b) ...... 165 Figure 4.6 Oil and coal price in 2015 dollars (BP 2015; The World Bank 2016)...... 166 Figure 4.7 Coal consumption by region (BP 2016b) ...... 175 Figure 4.8 World coal production and forecast (EWG 2007)...... 176 Figure 4.9 Actual energy to concentrate copper in U.S. copper mines 1954–2002 (U.S. Census Bureau 2004; Edelstein 1998) ...... 178 List of Tables

Table 1.1 Comparative properties of copper, iron and aluminium ...... 9 Table 1.2 Comparative properties of copper, iron and aluminium alloys ...... 10 Table 1.3 Copper consumption by sector in 2015 ...... 12 Table 1.4 Selected typical coal attributes ...... 16 Table 1.5 Energy per kilogram of coal in various units ...... 18 Table 2.1 Commercially important copper minerals...... 23 Table 2.2 Summary geological time scale with division boundaries based on recommendations in USGS (2010)—Fact Sheet 3059; the Tertiary Period is not recognised in some modern time scales but is still widely used on geological maps and in reports ...... 31 Table 3.1 2015 ranking of the world’s most highly valued metals...... 38 Table 3.2 Analyses of copper alloy from the Bowden and Bounty . . . . . 82 Table 3.3 Varieties of regulus (Raymond 1881) ...... 85 Table 3.4 Copper production costs 1869 ...... 99 Table 3.5 Rio Tinto mine production and sales 1890 (Harvey 1981) . . . . 103 Table 4.1 A selection of events and works on sustainability ...... 160 Table 4.2 Selected resource reserves from The Limits to Growth ...... 162 Table 4.3 Scientists and engineers on metals and energy resources . . . . . 163 Table 4.4 Abundance of aluminium, iron and copper ...... 168 Table 4.5 2015 R/P ratio for the major coal producing countries (BP 2016b) ...... 174 Table 4.6 Transcendental logarithmic cost function ...... 177 Table 4.7 Ehrlich–Simon sustainability wager 35 years on ...... 181

xv Units, Conversions and Abbreviations

Units

$ United States dollar unless otherwise stated £ British Pound Sterling unless otherwise stated Billion 109 Btu British thermal unit Gt Giga tonne—one billion tonnes GWh Gigawatt-hours—one billion watt-hours hp Horsepower HV Vickers hardness J Joules kg Kilogram—one thousand grams kWh Kilowatt-hours—one thousand watt-hours Mt Million tonnes Mtce Million tonnes of coal equivalent MW Megawatt—one million watts MWh Megawatt-hours—one million watt-hours tonne Metric tonne = 1000 kg t tonne W Watt

System International (SI) Units

Measure Unit Symbol Distance Metre m Mass Kilogram kg Force Newton N Energy Joule J Work Joule J Power Watt W

xvii xviii Units, Conversions and Abbreviations

Deﬁnitions of Work and Power

Work = Force × Distance Energy is the capacity to do work Energy and work are expressed in the same units, e.g., joules

Power = Work/Time Power is the amount of work available per unit time Power units include watts and horsepower

Selected Conversions Between Units

Energy 1 Btu = 1055 joules 1 calorie = 4.187 joules

Power 1 watt = 1 joule/second 746 watts = 1 horsepower 1 newton of force will accelerate 1 kilogram at 1 m/sec2 Applying 1 newton over 1 metre expends 1 joule of energy Expending 1 joule/sec requires 1 watt of power

Acronyms

AMD Acid mine drainage EIA The Energy Information Administration EPA US Environmental Protection Agency ESD Ecologically sustainable development EWG Energy Watch Group GFC The global ﬁnancial crisis IEA The International Energy Agency LTG Limits to growth OPEC Organization of Petroleum Exporting Countries PEV Plug in electric vehicle R/C Reserve to consumption ratio R/P Reserve to production ratio Units, Conversions and Abbreviations xix

SD Sustainable development SX-EW Solvent extraction-electrowinning UN United Nations US United States of America. Names may have U.S. USGS U.S. Geological Survey WWI First World War Doctor Copper and King Coal 1

Abstract Copper was the ﬁrst industrial metal and for over 7000 years has been one of our most essential metals. Coal resources have only been developed in the last few hundred years although the contribution of coal to human welfare has been just as signiﬁcant. This introductory chapter explains the title of the book and the essential properties of copper and coal that stimulated our demand for both. When steel hulled ships were introduced, there was no longer a need for copper sheathing of wooden hulled ships; however, the Age of Electricity was beginning and copper was in even greater demand. Today there is increasing demand for copper in wind farms and electric cars.

1.1 Why Doctor Copper and King Coal

Copper is a ubiquitous metal. It was the first industrial metal and, for over 7000 years, copper has been one of our most essential metals. Today it is the conduit for energy, information and water in our homes and offices. The benefits that copper and coal have provided are fairly obvious; however, the connections between the two are not self-evident. The following chapters will explore the evolution of our use of both copper and coal and explain the connection. The reader will get a glimpse of civilisations’ past as the story of copper unfolds from the Stone Age to the Computer Age. At the end of the last Glacial Maximum, around 10,000 BC, humans in Europe began to emerge from the more habitable areas (known as refugia) where they had survived through the Ice Age. Conversely, the humans living on the Sunda and Sahul shelves surrounding modern day Indonesia and Australia appear to have thrived during the Ice Age in a verdant land. Unfortunately, little evidence of their occupation of the area survives because sea level rose an estimated 120 m as the

© Springer International Publishing AG 2017 1 B. Golding and S.D. Golding, Metals, Energy and Sustainability, DOI 10.1007/978-3-319-51175-7_1 2 1 Doctor Copper and King Coal ice retreated. Their hunting grounds were inundated and their pleasant lifestyle, as evidenced in the many Bradshaw1 paintings in Northern Australia, was no longer sustainable. The average of the numerous estimates of the number of humans on Earth at this time is around four million. By the end of the nineteenth century, as the Industrial Revolution gave way to the Age of Electricity, world population had increased to over 1.6 billion. Our book explains the significant contribution copper and coal made in the evolution of the technology that enabled this population increase and that helps support our current population of over 7.4 billion. Nevertheless, there are many who argue that resources such as copper and coal are limited. The proposition that the production of oil had peaked has resulted in a volume of literature on the subject of Peak oil. Likewise, there are supporters of the concept of Peak copper and Peak coal. These concepts will be explored in the final chapter of the book; however, in order to explain the issue of sustainability of copper and coal and the development of these resources, we will first describe the nature of copper and coal in Chap. 2. Commencing with the earliest evidence, Chap. 3 explains how the resources have been exploited at significant mining areas throughout time. Copper is credited with having a Ph.D. in economics because the price of copper is considered a leading indicator of turning points in the global economy. Copper is required by almost all sectors of the economy, so this is not an unreasonable thesis. The counterargument is that copper price is a lagging rather than a leading indicator of impending upturns or downturns in the global economy. Modern copper mining often requires the extraction and treatment of large volumes of low-grade copper ore. Such mines require investments of billions of dollars to develop new resources that are often many years in the planning and approval process. A sudden increase in demand for copper such as the one that occurred in China around the beginning of this century will lead to a significant increase in the price of copper and stimulate the large investments in new mines or mine expansions necessary to bring on new supplies of copper. Hence, it would seem that copper price may be a lead indicator. Nevertheless, a lead indicator should predict downturns as well as upturns in the global economy. If a recession is looming, then the price of copper should have fallen, pre-empting the downturn. When the Global Financial Crisis (GFC) recession began in July 2007, the price of copper increased from $7980/tonne in July to $8020/tonne in October. Copper price was not a lead indicator in this instance. Rather than a lead indicator of the health of the world economy, copper price was an indicator of demand, mostly from China, exceeding world supply. The impact of the GFC and the subsequent recession was still evident in some countries in 2016. Unemployment in the United States (U.S.) remained above 8% in 2012; however, the price of copper peaked at over $9800/tonne in February 2011 and then began to decline as supply increased. These examples indicate that an increase in the price of

1Pastoralist Joseph Bradshaw observed this unusual type of rock art in the northwest Kimberley region of Western Australia while searching for suitable pastoral land in 1891. 1.1 Why Doctor Copper and King Coal 3 copper, although an indicator of the lag in developing new copper resources, is not necessarily an indicator of the state of the economy. Nevertheless, the unprece- dented increase in demand in China confounded many economic forecasters and Dr. Copper was no exception. Dr. Copper may be a harbinger of turning points in the global economy, although economic indicators such as GDP and employment are possibly better indicators. However, the long-term copper price is an indicator of the sustainability of copper mining and this will be considered in the ﬁnal chapter. Compared to our copper resources that have been exploited for millennia, coal resources have only been developed in the last few hundred years, although the contribution of coal to human welfare has been just as signiﬁcant. When George Orwell described the life of miners and conditions in a British coal mine in his book The Road to Wigan Pier, there were some 750,000 coal miners in Britain. Orwell recognised the important contribution coal made to society in the following paragraph.

Practically everything we do, from eating an ice to crossing the Atlantic, and from baking a loaf to writing a novel, involves the use of coal, directly or indirectly. For all the arts of peace coal is needed; if war breaks out it is needed all the more (Orwell 1937). An early reference to King Coal may be found in an 1881 edition of Punch magazine where a cartoon with the title What will he grow to? depicts King Steam and King Coal contemplating the new and portentous infant Electricity. The cartoon in Fig. 1.1 was on the cover of Puck magazine published on 17 September 1902. A dejected Old King Coal is wearing kingly robes and a crown that he holds on his head, whilst in the background factories are burning oil for fuel and spewing thick black smoke that drifts, in the shape of a hand, towards the coal sculpture with the intent of snatching the crown from his head. King Coal is also the title of a novel by Upton Sinclair published in 1917 that describes the poor working conditions in the coal mining industry in the western U.S. during the 1910s. Almost all cartoons show King Coal as either forlorn or tyrannical. Very few recognise the great beneﬁts that coal has delivered, although in Scotland they have a saying Lang may yer lum reek! (Long may your chimney smoke!) Wi’ ither folks coal! (With other people’s coal!), indicating how much coal was appreciated. The last phrase may refer to an old Scottish New Year’s Eve custom of bringing a piece of coal into each home visited to add warmth to the ﬁre of a friend.

1.2 Copper: A Most Useful Metal

Through the ages, copper has been there to meet our needs. At ﬁrst it was for tools as we moved out of the Stone Age. One of the most impressive examples of the technological evolution is the copper axe shown in Fig. 1.2 that belonged to the 4 1 Doctor Copper and King Coal

‘Iceman’2 dating from around 3300 BC. His tools also included a flint knife, indicating copper had not replaced sharp flint when it came to cutting. The addition of arsenic or tin to copper improved the hardness of the metal creating bronze. Flint tools were displaced by bronze and the Bronze Age was born. Once we discovered how to make iron, which was harder and cheaper than bronze, copper was no longer needed to make most tools. Nevertheless, copper remained a valued metal in the Roman Empire. The Pan- theon in Rome, built by Emperor Hadrian in the years from 117 to 138, is one of the finest surviving examples of Roman architecture. The dome, which is almost 43 m in diameter, was originally covered with copper plates with an outside covering of copper and bronze tiles. The tiles were stolen by Constans II in 663 and were on route to Constantinople when they were in turn stolen by Saracens (Smith 1965). Pope Urban VIII (1623–1644) removed the copper plates from the roof that reportedly yielded some 200 tonnes of copper sheets, in addition to four tonnes of copper nails. Evidently the copper was mostly used to make cannon.3 One of the more intriguing uses of copper was as flattened sheets for writing. In March 1952, Henri de Contenson, an archaeologist, discovered two lumps of what is now known as the ‘Copper Scroll’ in a hillside cave close to Qumran. The Copper Scroll was found with other Dead Sea Scrolls4 and is thought to date from the later of the dates attributed to the Scrolls. Copper plate also made possible the distribution of one of the first recognisable world maps. In 1477, the maps of Ptolemy5 were recreated in Italy by engraving onto and printing from copper plates. Bronze continued to be used for making weapons such as cannon guns up until the seventeenth century. Copper also continued to be used for making coins and kitchen utensils; however, in the eighteenth century, it was found that copper sheathing protected the wooden hulls of naval vessels from shipworm and various marine weeds. A First rate ‘ship of the line’6 required some 36.5 tonnes of copper bolts, 17 tonnes of sheathing and 1.5 tonnes of nails or 55 tonnes of copper in total (Derrick 1806). By 1782, 82 ‘ships of the line’ and some 267 other Royal Navy ships had been copper sheathed. Lord Sandwich, First Lord of the Admiralty, who considered coppering of the fleet one of his greatest achievements, wrote ‘copper bottoms fear

2In 1991, tourists hiking in the southern Tyrol, on the border between Austria and Italy, found a desiccated corpse protruding from a receding glacier. Nick-named Ötzi after the Ötztal region, the ‘Iceman’ had been lying frozen for over 5300 years. 3A contemporary Roman satirist is credited with saying ‘What the barbarians did not do the Barberinis (Urban VIII’s family name) did’. 4The Dead Sea Scrolls were found in caves in the vicinity of Khirbet Qumran inland from the northwest shore of the Dead Sea. The Dead Sea or Qumran Caves Scrolls are believed to date from between the last three centuries BC and the ﬁrst century AD. 5Claudius Ptolemy was a geographer who worked in Alexandria around 150. He is credited with inventing the concept of longitude and latitude. Christopher Columbus studied Ptolemy’s maps ahead of his 1492 expedition that discovered the Americas. 6Ships-of-the-line were the British Royal Navy’s largest ships and First rate was the largest class in that category carrying at least 100 guns. 1.2 Copper: A Most Useful Metal 5

Fig. 1.1 Old King Coal’s crown in danger—Library of Congress nothing’ (Knight 1973). By 1790, all the ships of the line, some 93, were copper-bottomed, which alone would have required over 5000 tonnes of copper. Additionally, many other merchant and foreign vessels ships were copper-bottomed. Thomas Williams, a dominant figure in the copper market, stated that his Liverpool office had sheathed 105 ships and repaired the coppering of 33 more in 1799 (Harris 1966). According to Symons (2003), between 1793 and 1799, the Royal Navy consumed some 7640 tonnes of copper. In addition to sheathing for both navel and merchant shipping, British copper was used for coinage and household goods as well as being exported. Great Britain produced almost three-quarters of the world’s copper in 1790. Copper ore from Cornish mines accounted for about 5000 tonnes of refined copper, and the Welsh mines of Anglesea over 2000 tonnes. By way of comparison, the great Mansfeld Mine in Germany made only 380 tonnes in that year (Stevens 1905). 6 1 Doctor Copper and King Coal

Fig. 1.2 The Iceman’s copper axe: Photo credit: South Tyrol Museum of Archaeology—www. iceman.it

The transition from wooden hulled ships to steel hulled ships commenced with cladding iron over a traditional wooden hull. In 1860, the Royal Navy commissioned its first fully iron hulled ship, the HMS Warrior, and twenty years later its first steel hulled ship, HMS Iris, was commissioned. In 1890, the Royal Navy launched its last wooden hulled ship, the training brig Mayflower. Soon there was no longer a need for copper sheathing to protect hulls from shipworm. Nevertheless, as Fig. 1.3 shows, the absence of the copper sheathing trade had little impact on the world production of copper according to the data reported by Schmitz (1979). The graph below does not include China’s production as this is not included in the data that Schmitz painstakingly compiled. During the nineteenth century, the Birmingham brass industry expanded enor- mously. According to Hughes (2000), in 1866, Birmingham industries consumed 3000 tonnes of copper to make brass for domestic use, and 7000 tonnes were used for the specialist lower copper, high zinc, i.e. 60% Cu, 40% Zn brass alloy for sheathing wooden hulled sailing vessels. In addition, 10,000 tonnes were used for engineering of which 6000 tonnes was consumed for making copper and brass tubes and 4000 tonnes for making copper wire. The incandescent light bulb made its appearance in 1879 and the first electric power station was built three years later. There was a need for a malleable metal to conduct electricity and once again copper was there to meet our need. Copper production was about to increase exponentially. In fact, world copper production increased from just over 100,000 tonnes in 1870 to almost 500,000 tonnes by 1900. 1.2 Copper: A Most Useful Metal 7

UK Spain US Chile World 120

100

40 Copper (000' production tonnes) 20

0 1750 1760 1770 1780 1790 1800 1810 1820 1830 1840 1850 1860 1870

Fig. 1.3 Known world copper production—After (Schmitz 1979)

Stevens in his tome A Manual of the Copper Industry of the World listed the many uses of copper in 1904. Top of his list was the need for copper in all electrical installations, and for power and telegraph transmission lines.

The iron wires of pioneer telegraph and telephone lines are rapidly giving way to copper strands. Iron is low in electrical conductivity, making it an inefﬁcient and costly medium for transmission. It is also subject to rust, and its lack of the ductility, which is such a prominent characteristic of copper, causes iron wires to break from winds (Stevens 1905). He noted that, whereas electric light was once a luxury, it is now a necessity. Telephone lines now stretched across America, and he reported that Stockholm had one phone for every family of its population. Next to its electrical uses, copper is most extensively employed in engineering, where brass, gunmetal7 and bronze are a necessity. Copper and brass boiler tubes are used in locomotive and other boilers, having the ability to withstand enormous pressures, and being excellent conductors of heat. Cooking utensils also required much copper; however, Stevens anticipated that copper would give way to aluminium because aluminium was lighter and tarnishes less easily.

7Gunmetal, is an alloy of copper, tin and zinc in proportions approximating 88% Cu, 9% Sn and 3% Zn. 8 1 Doctor Copper and King Coal

Manufacturers of scientiﬁc instruments are also excellent patrons of the brass foundry where copper alloys in many forms, such as sheets, tubes, rods, wires and castings for the construction of microscopes, telescopes and surveying equipment may be obtained. Copper sulfate is one of the most important chemical agents known to science and industry. As an insecticide it stands without an equal. Dilute solutions of copper stayed the ravages of the phylloxera when the vineyards of France seemed doomed.

It is probable that not less than one hundred thousand tons of copper sulfate, containing a quarter of its weight in metallic copper, is consumed every year in spraying the vines and fruit trees of Europe and America, and thus it may be said that it is to copper that we owe the sparkling wines of France, the peerless American apple, and the blushing peach that reaches perfection on every continent (Stevens 1905). Copper sulfate, blue stone, blue vitriol are all common names for pentahydrate cupric sulfate, CuSO4.5H2O, the best known and the most widely used of the copper salts. In 1974, the Copper Development Association reported that some 200,000 tonnes of copper sulfate were consumed annually, about 75% of which was used in agriculture, principally as a fungicide (CDA 1974).

1.3 Properties that Make Copper Desirable

There are many quantitative and qualitative measures that might be selected to show why copper remains a very desirable metal. The measures in Table 1.1 have been selected to compare the properties of the three most commonly used industrial metals copper, iron and aluminium. The data were obtained from multiple sources that often gave differing values for the measures chosen so should be considered as an aid for relative comparison rather than being precisely accurate. For example, the value for the melting point of copper, which one would have thought would be definitively established, varies in the literature from 1083 to 1086°C. Tests, such as the hardness of a metal, are dependent on the testing equipment used as well as the patience and subjective judgement of the technician carrying out the test. Conse- quently, values found in the literature often vary above and below the values reported in Table 1.1. The first three columns of the table show the values for the pure metals copper, iron and aluminium; however, only copper is commonly found in its pure form on Earth. Pure iron in its elemental form is rarely found on Earth. It is soft, malleable, ductile and strongly magnetic. Because iron combines readily with oxygen, sulfur and other anions, it is rarely found in the native state. Native iron does occur as minute pellets in basaltic rocks in many areas; however, the only significant occurrence of large masses of native iron is found in basalts on Disko Island, Greenland. Both native iron and meteoritic iron are quite rare, so the technique of smelting iron oxide ores had to be developed before iron became a widely used metal (Rapp 2009). Iron exists on Earth mostly in the form of oxides, the most 1.3 Properties that Make Copper Desirable 9

Table 1.1 Comparative properties of copper, iron and aluminium Copper Iron Aluminium Atomic symbol Cu Fe Al Abundance in earth’s crust % 0.006 5.3 8.2 Speciﬁc gravity 8.9 7.9 2.7 Atomic number 29 26 13 Atomic mass 65.5 55.9 27.0 Melting point °C 1085 1535 660 Electric resistivity nΩ m (at 20 °C) 17 89 27 Thermal conductivity (W/m-K) 400 80 205 Vickers hardness 50 65 15 Vickers hardness work hardened 100 * 25 Tensile strength ultimate MPa 210 270 70 Elongation at break % 60 50 60 *Although pure iron is a not commonly used in industry, values are reported in Table 1.1 where they were available. No data were found for work hardened pure iron

mined being hematite (Fe2O3). Iron is the major component of steel that contains minor amounts of carbon and other elements. Before the advent of modern steel-making, wrought iron was the most commonly used form of malleable iron. Wrought iron, like mild steel, contains a small amount of carbon, as well as other impurities and therefore is not pure iron. Likewise, although aluminium is the most abundant element in the Earth’s crust after oxygen and silicon, it is rarely found as a free element in nature.8 Pure aluminium is soft and lacks strength, but alloyed with small amounts of copper and other elements it has become one of our most useful metals. Aluminium is reﬁned from bauxite, which contains aluminium hydroxides, using the Bayer Process invented and patented in 1887 by Karl Josef Bayer. Whilst the properties of pure copper, iron and aluminium are listed in Table 1.1, only copper is used in industry in its pure form. Therefore, it is more appropriate to compare the properties of the alloys of copper, iron and aluminium such as those alloys selected in Table 1.2. Given that the abundance of copper in the Earth’s crust is just 0.006%, it is remarkable that copper ranks with iron and aluminium as one of the three major industrial metals. The reasons for copper becoming our ﬁrst industrial metal will become evident in Chap. 3. The reason why copper has maintained its prominence is partly explained by its electrical properties. Electric resistivity is the most commonly reported measure of the electrical properties for metals. It is the recip- rocal of the material’s ability to conduct electricity and so, the lower the value the better the electrical conductivity. Pure copper is twice as good as the aluminium

8Aluminium’s highly reactive nature prevents the occurrence of elemental aluminium in natural specimens. Native aluminium grains have been found in some highly reduced (low oxygen) volcanic muds. 10 1 Doctor Copper and King Coal

Table 1.2 Comparative properties of copper, iron and aluminium alloys Bronze Low carbon steel Aluminium 90% Cu 10% Sn (Mild steel) Alloy (6061) 0.05–0.25% Carbon (C) Electric resistivity nΩ m 16 150 37 Thermal conductivity (W/m-K) 110 60 180 Vickers hardness 80 100 30 Vickers hardness work hardened 220 200 100 Tensile strength ultimate MPa 310 440 300 Elongation at break % 25 20 25 alloy and many times better than mild steel in its ability to conduct electricity. Additionally, the thermal conductivity of copper far exceeds that of its rival metals. Hardness is measured on many scales and Vickers Hardness (HV)9 has been chosen because it is widely reported. Steel has the edge over copper when it comes to hardness but as can be seen work hardened bronze is a match for cold worked mild steel; however, there are other alloys of steel that are considerably harder than mild steel. Ultimate tensile strength is a measure of the force required to pull a material apart. As the longitudinal force is applied, the material will tend to stretch or elongate. Ductile materials elongate more than brittle materials. The tensile strength for mild steel reported is the highest of all the metals; however, it shows the smallest elongation indicating it is the least ductile. Comparing the conductivity, malleability and ductility of copper in Tables 1.1 and 1.2 with its major competitors, aluminium and steel, it is easy to understand how copper has maintained its prominence. Over the ages, as civilisations and technology developed, copper was there to meet our requirements. In the ﬁrst millennium, tin and zinc were alloyed with copper to produce bronze and brass respectively. Today many more elements are alloyed with copper. Cad- mium, chromium, lead, manganese, nickel and silver are just some of the metals alloyed with copper to improve strength, corrosion and wear resistance, machin- ability and colour. Davis (2001) lists the major groups of copper and its alloys including:

coppers that contain a minimum of 99.3% Cu; copper–zinc alloys (brasses) that contain up to 40% Zn; copper–tin alloys (phosphor bronzes) that contain up to 10% Sn and 0.2% P; copper–aluminium alloys (aluminium bronzes) that contain up to 10% Al; copper–silicon alloys (silicon bronzes) that contain up to 3% Si; and copper–nickel alloys that contain up to 30% Ni.

9The Vickers Hardness test was developed in 1921 by Robert L. Smith and George E. Sandland at Vickers Ltd. 1.3 Properties that Make Copper Desirable 11

The warm colours of copper and its alloys are prized for architectural and consumer items as well as objects of art. The colour of the alloy varies with the percentage of the alloying element and the six alloys in Fig. 1.4 have been selected to demonstrate the range of colours available. The bacteria count on copper surfaces reduces quickly in comparison to other surfaces including stainless steel. Consequently, antimicrobial copper alloys are often selected in hospitals for surfaces such as bedrails, tables, medical equipment, bathroom furniture and other items in close proximity to the patient. Given copper’s many attributes it is unsurprising that almost 23 million tonnes of copper were consumed in 2015. In 1980, there were three major copper consuming countries, the U.S., Japan and Germany. Although these countries consume copper domestically, much is included in manufactured products for export. In the 1990s, the sleeping giant China awoke as illustrated in Fig. 1.5 that shows China’s copper consumption increasing exponentially. In 2000, China’s copper consumption surpassed Germany and Japan as China continued to increase copper production for both the domestic market and in materials for export. China accounted for about half of the world’s copper consumption in 2015. Comparing the uses of copper in China, a developing economy, with the U.S. and the European Union (EU) in Table 1.3 reveals a changing pattern of usage. The categories reported for each zone are not identical; however, they do give an approximate idea of how copper is currently used, most especially the high infrastructure usage in China compared to the more developed world. The U.S. figures do not include a category for power and telecom. This may be because there are not so many high voltage transmission lines being built in the U.S. and when they are built aluminium is the main material used. Similarly, many telecommu- nication cables are now being constructed of optical fibre10 although copper is still mostly used within buildings. However, each time a material is found to replace copper, there appears another technology that requires the attributes of Dr. Copper. Wind farms generated 3% of world energy in 2015 (GWEC 2015), about one-third of which was generated in China. The United States Geological Survey (USGS) estimated that three tonnes of copper were required per Megawatt (MW) of wind turbine capacity (Wilburn 2011). However, Falconer (2009) estimated that the copper required for off-shore wind farms was as high as nine tonnes per MW of installed capacity. A coal fired power station on the other hand requires about one tonne of copper per MW of installed capacity (BBF Associates and Kundig 2011). Additionally, the capacity factor11 for conventional coal fired power stations is about 60% compared to wind

10Optical fibre is a flexible, transparent fibre made by drawing glass (silica) to a diameter slightly thicker than that of a human hair. Fibre optic technology uses pulses of light to carry the signal, whereas on traditional copper wires the message is transmitted by electrical currents. 11Capacity factor equals the actual energy produced by an energy generating unit divided by the maximum possible energy that the unit can produce. 12 1 Doctor Copper and King Coal

Fig. 1.4 Various copper alloy colours—courtesy of the International Copper Association, Australia

Annual copper consumption 14.0 25.0

12.0 20.0 10.0

8.0 15.0

6.0 10.0 4.0

5.0 World consumption 2.0

0.0 0.0 Major consuming countries Mtpa 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020

China UnitedStates Germany Japan World

Fig. 1.5 Consumption of copper by the major consuming countries

Table 1.3 Copper consumption by sector in 2015 World (%) China (%) U.S. (%) EU (%) Building construction 29 21 43 36 Infrastructure (power and telecom) 14 27 10 Industrial (motors, transformers etc.) 13 11 7 15 Transport 12 11 19 14 Appliances 16 15 12 25 Electronics 4 7 19 Other 11 8 1.3 Properties that Make Copper Desirable 13 farms of 30% in the U.S. (EIA 2016a). Thus, the need for copper is several times greater when electricity is generated by wind farms compared to conventional coal ﬁred power stations. Similarly, replacing conventional cars with Plug in Electric Vehicles (PEVs) will increase the demand for copper. Sales of PEVs have increased from 45,000 in 2011 to over 550,000 in 2015 and sales in China surpassed the U.S. in 2015. While a conventional car contains about 19 kg of copper (Smith et al. 2015), a PEV contains some 64 kg of copper. 30,000 electric vehicles were reportedly sold in China in April 2016, implying upward of 360,000 vehicles may be sold in 2016, which would require some 23,000 tonnes of copper. However, PEV production appears to be expanding exponentially and by 2020 the PEV demand for copper could be over 100,000 tonnes of copper for China alone.

No surprise, China leads the world in the consumption of steel, copper, aluminium, lead, stainless steel, gold, silver, palladium, zinc, platinum, rare earth compounds, and pretty much anything else labelled ‘metal’. But China is desperately short of metal resources of its own. For example, in 2012 China produced 5.6 million tons of copper, of which 2.75 million tons was made from scrap. Of that scrap copper, 70% was imported, with most coming from the United States. In other words, just under half of China’s copper supply is imported as scrap metal. That’s not a trivial matter: copper, more than any other metal, is essential to modern life. It is the means by which we transmit power and information (Minter 2013).

1.4 Coal: The Energy for the Industrial Revolution

Compared to copper, the world’s coal resources have only recently been exploited. Dodson et al. (2014) studied sites located near coal deposits in China and they postulate that the people there made use of coal from at least Bronze Age times. They suggest that the removal of woody vegetation may have created an energy shortage for smelting, cooking and heating, and people living close by the coal deposits used coal as an alternative to timber and charcoal. They acknowledge that their work is preliminary; however, they hoped it would encourage others to research the earliest relationship between humans and their use of coal, which is a much neglected topic. There are reports that coal was used in blast furnaces during the Han Dynasty (200 BC to AD 9); however, Bronson (1999) suggests that the coal was used solely in kilns for ﬁring tiles. Marco Polo, who reputedly travelled to China in the later part of the thirteenth century, describes a black stone that was dug from the mountains in China, where it ran in veins. When lighted, the stone burnt like charcoal, and retained the ﬁre much better than wood; insomuch that it was preserved during the night, and in the morning would be found still burning (Polo et al. 2004). As will be explained in Chap. 3, during Marco Polo’s time, wood and charcoal were the preferred options for heating in Europe, although coal was burnt in Britain during the Roman occupation. 14 1 Doctor Copper and King Coal

1.5 Our Current Use of Coal

Coal is the fossilised remains of plants that lived many millions of years ago as will be more fully described in Chap. 2. Given the various sources of the organic matter and its subsequent burial and transformation, coal is not a homogeneous material and is found in even more forms than copper. Consequently, there have been many attempts to classify coal. Because coal is predominantly used as a fuel, classifica- tions have generally been based on the behaviour of coal as it is burns. Therefore, the use of coal will first be described before explaining some of the methods of classification. Like copper, the use of coal has changed through the ages as will be described in Chap. 3, although like modern day copper use, coal consumption is strongly linked to the Age of Electricity. Almost eight billion tonnes of coal were consumed in 2013 and as Fig. 1.6 shows more than half was used to produce electricity. The U.S. and China are the two largest producers and consumers of coal. The U.S. consumed over 900 million tonne of coal in 2014, and of that mass about 92% was consumed by the electricity sector. China’s industrial sector accounted for 95% of its coal consumption in 2012, half of which was consumed by thermal power producers to generate electricity. China consumed 3.6 billion tonnes of coal in 2014 and close to 80% of electricity was generated from thermal power. Global crude steel production reached a high of 1.7 billion tonnes in 2014. China has been the driving force in the expansion of the

World coal consumpƟon 9,000

8,000

7,000

6,000 on Mtpa

Ɵ 5,000 Electricity 4,000 Steel 3,000

Coal consump HeaƟng 2,000 Industry 1,000 Other

1973 1983 1993 2004 2013

Fig. 1.6 World coal consumption by sector 1.5 Our Current Use of Coal 15 global steel industry over the past decade and in 2014, China’s crude steel output reached 822 million tonnes, about half the global production. Approximately 75% of world steel is produced by pouring iron from Blast Furnaces (BF) into Basic Oxygen Furnaces (BOF), commonly called BF-BOF technology. The remainder is mostly produced in Electric Arc Furnaces (ACF) that largely depend on scrap steel. The BF-BOF process consumes about 0.8 tonne of coal per tonne of steel produced, whereas the ACF process, assuming the electricity is produced from coal, requires about 0.15 tonne of coal per tonne of steel produced. Based on these assumptions over 1.1 billion tonnes of coal, about 14% of world coal consumption, were used to produce steel in 2014.

1.6 The Essential Properties of Coal

Coal fired power stations work on much the same principles developed for our earliest steam engines, the only difference being that the steam generated in the boiler is used to drive a turbine to generate electricity rather than directly driving a machine such as a pump or a locomotive. A modern thermal power station can be fired with a wide range of coal, whereas other applications may require particular types of coal. The complex nature of coals is evidenced by the British Royal Navy’s painstaking selection of coal for its ship’s boilers. No two coals are the same and even seams in the same mine may produce coals with different properties, ranging from rock-hard anthracite through to bituminous coals. Admiralty Welsh steam coal was a semi-bituminous coal, which burnt not too hot and was not too hard that it would damage furnaces, and not too friable that it burnt too fast and disintegrated into dust. It contained few impurities and produced a more complete combustion keeping smoke and ash to a minimum and had a high stowage of calorific value per volume. The best steam coal lay deep below the Rhondda valleys in Wales, and mining it required significant capital investment. For example, Nixon’s Navigation Colliery’s main shaft took seven years to sink before production commenced in 1860 (Brown 2003). The Admiralty’s use of Welsh coal enhanced its reputation for quality and significantly boosted investment and production. South Wales’ coal exports boomed from some 63,000 tonnes in 1840 to nearly 4 million tonnes in 1874. Given the numerous species of plants that become coal and the millions of years of geological disturbance since the plants were deposited, attempting to classify coal is almost as challenging as attempting to classify all the species of plants on Earth today. The best known method is according to the coals’ rank that will be more fully explained in Chap. 2. The rank of coal is determined by the amount of alteration or metamorphism that has occurred since the organic material was deposited. 16 1 Doctor Copper and King Coal

Table 1.4 Selected typical coal attributes Attribute Unit Basis Lignite Bituminous Bituminous Anthracite thermal coking Total moisture % As 55 15 10 7 received Volatile matter % Dry ash 55 27 27 8 free Fixed carbon % As 30 55 65 80 received Speciﬁc energy MJ/kg As 823 2932 received Hardgrove grindability 40 100 100 50 index (HGI) Crucible swelling 0.5 6 0.5 number (CSN)

Lignite, the lowest rank of coal, is soft and crumbly with high moisture content relative to older and more metamorphosed coals. Because it is not so dark in colour it is commonly described as brown coal. Anthracite, which is black and brittle with a glassy appearance, is the highest rank of coal. Between these two extremes is a wide band of bituminous coals, the most abundant form of coal. The attributes of coal in Table 1.4 have been selected to differentiate the categories of coal based on rank. However, although the rank classification is based on the amount of metamorphism, coal is mostly classified according to its behaviour when burnt and so bituminous coal has been divided into thermal and coking. Whilst a coking coal may be used in power stations, thermal coals do not coke and therefore are not used in steel making. Numerous test procedures have been developed in an attempt to predict how a coal will behave in a particular use. The most common test procedure is the proximate analysis that quantifies the percentage of the four components obtained when the coal is heated. These are: moisture; volatile matter, consisting of gases driven off during pyrolysis12; fixed carbon and the ash remaining after combustion. These coal properties may be reported on several bases. The two chosen in Table 1.4 are as received (ar) and dry ash free (daf). The percentage of moisture reported on an as received basis is the percentage of moisture in the coal in the sample as it arrived at the laboratory. Providing the sample was well sealed when sampled, the as received moisture is equivalent to the moisture as sampled. If the sample is immediately taken from a core of coal from an exploration drill hole, it will be similar to the in situ state of the coal. A similar logic applies to the other attributes reported on an as received basis in Table 1.4.

12Pyrolysis is a thermochemical decomposition of organic material at elevated temperatures in the absence of oxygen. The word is derived from the Greek pyro ‘ﬁre’ and lysis ‘separating’. 1.6 The Essential Properties of Coal 17

Dry ash-free basis (daf) is a means of expressing the analytical results based on a hypothetical condition in which the solid mineral fuel is considered to be free from both moisture and ash. The values in Table 1.4 are typical and not meant to represent any particular coal but rather to demonstrate the difference between the coal categories. As the table shows, the total moisture decreases significantly from lignite to bituminous coal but not so much from bituminous coal to anthracite. Volatile matter includes methane, hydrocarbons, hydrogen and carbon monoxide, and incombustible gases like carbon dioxide. Volatile matter is a measure of the gaseous fuels present in coal. Volatile matter expressed on a daf basis is high for lignite and low for anthracite. This is logical given that the moisture and ash have been removed from the calculation and whilst the volatile matter is naturally high in lignite, the amount is exaggerated when the moisture and ash are removed. Fixed carbon is the solid fuel left after the volatile matter is driven off. It consists mostly of carbon but also contains some hydrogen, oxygen, sulfur and nitrogen. Fixed carbon provides an estimate of the heating value of coal, hence coals with high fixed carbon are favoured in thermal power plants. Counter intuitively, although anthracite has the highest fixed carbon content, it is not a favoured power plant coal. Anthracite has a higher heat content than bituminous coals and usually a lower sulfur content. However, anthracite is more difficult to ignite due to the lower volatile matter content. Additionally, after initial ignition is achieved, combustion is more difficult to maintain (Charmbury 1975). Specific energy is the energy per unit mass of the fuel. Energy may be reported in many different units including British thermal units, calories, joules and kilowatt-hours. Converting from one set of units to another was once a nightmare for engineers; however, there are now many free conversion programmes on the World Wide Web (WWW). Where possible, the International System of Units (SI) is used throughout this book and therefore specific energy is reported in million (mega) Joules per kilogram (MJ/kg) of coal in Table 1.5. For a power plant, specific energy is one of the most important properties as this determines the amount of electricity the plant my eventually generate. As an aid for those more familiar with other units, Table 1.5 shows the energy in other units equivalent to 25 MJ. A typical bituminous coal contains around 25 MJ/kg; however, some of the units are often reported per pound (lb) in which case 1 kg ≈ 2.2 lb. The value of coal to our ancestors may best be understood by the labour coal replaced. Assuming a Cornish miner13 in the twentieth century consumed 3600 calories a day and was 18% efficient at converting that energy into work, his work effort would light a 30 W bulb for 24 h. Assuming a power plant efficiency of 45%, and line losses of 5% (EIA 2016b), the miner’s labour would equate to about 260 grams of coal feed to the power plant or less than the weight of a can of Coke.

13Miners in Cornwall traditionally ate meat and potato pasties underground. Assuming a traditional Cornish pasty contains 500 calories, miner’s crib tins (lunchboxes) could be quite large. 18 1 Doctor Copper and King Coal

Table 1.5 Energy per kilogram of coal in various units MJ Kilocalories BTU Watt hours 25 5970 23,700 6940

Several other coal properties are important when purchasing coal, one of the most important being the percentage of sulfur in the coal. In the U.S., the Clean Air Act Amendments of 1990 mandated reductions in sulfur dioxide emissions from electric power plants. Many power plants opted to substitute coal from the east of the U.S. with Powder River Basin coals from the western U.S. that are comparatively low in sulfur even though this involved transporting the coal over much greater distances (Considine 2013). The Hardgrove grindability index (HGI) indicates how easy a coal is to grind. This is particularly important if coal is to be burnt in a pulverised state as it is in most power plants. The HGI is calculated by applying a standard amount of work in a crushing mill and measuring the fraction of coal passing through a mesh after milling. Power plants favour coals with a high HGI. The final attribute selected in Table 1.4 is the crucible swelling number (CSN) that is important to steel manufacturers. Coke manufactured from coal is the most important raw material feed into a blast furnace in terms of its effect on blast furnace operation. The coke must be strong enough to support a smooth descent of the iron ore and limestone in the blast furnace with as little degradation as possible enabling the optimum permeability for the flow of gaseous and molten products. The CSN indicates the capacity of the coal to expand when subjected to a stan- dardised heat and is used to evaluate the coking properties of the coal. CSN ranges from 0 to 9 and a CSN of 9 implies the coal has very good coking properties. The four coal categories listed in Table 1.4 have been simplified for ease of explanation and there are several other sub-categories including sub-bituminous coal. However, the International Coal Classification of the Economic Commission for Europe recognises just two broad categories of coal. Hard coal includes all coals having a specific energy greater than 23.9 MJ/kg on an ash free but moist basis. Coals with less energy are classified as brown coal. The International Energy Agency (IEA) Coal Information Statistics for World Coal Supply also has these two major categories: hard coal, which is the sum of anthracite, coking coal and other bituminous coals, and brown coal, which is the sum of lignite and sub-bituminous coal (IEA 2016a). Consequently, the data set includes the U.S. Powder River Basin coals as brown coal (IEA 2016b). In addition to coal’s use in power plants and steel production, the World Coal Association (2016) lists several other current day uses of coal. Refined coal tar is used in the manufacture of chemicals, such as creosote oil, naphthalene, phenol and benzene. Ammonia gas recovered from coke ovens is used to manufacture ammonia salts, nitric acid and agricultural fertilisers. Many products including soap, solvents, dyes, plastics and fibres, such as rayon and nylon, contain coal by-products. Coal is also an essential ingredient in the production of specialist products such as activated carbon that is used in water filters and carbon fibre that is 1.6 The Essential Properties of Coal 19 used in the construction of bikes because it is strong but light weight. Sulfur captured from modern coal fired power plants is used to manufacture sulfuric acid, a by-product coal has in common with copper smelting. Much of the electric power required in the copper mining industry is also produced in coal fired power plants; however, as will be explained in Chap. 3, coal was essential to the development of the modern copper industry. The final chapter will catalogue production of copper and coal in the twenty-first century and speculate on the sustainability of the copper and coal industries.

References

BBF Associates, Kundig KJA (2011) Market study: current and projected wind and solar renewable electric generating capacity and resulting copper demand. Copper Development Association Inc., New York Bronson B (1999) The transition to iron in ancient China. In: Pigott VC (ed) The archaeometallurgy of the asian old world. The University of Pennsylvania Museum of Archaeology and Anthropology, Philadelphia, pp 177–193 Brown WM (2003) The Royal Navy’s fuel supplies, 1898–1939; the transition from coal to oil. University of London, London CDA (1974) Uses of copper compounds. CDA technical note. Copper Development Association, Hemel Hempstead Charmbury HB (1975) Potential markets and economic constraints. In: Burti UH, Kalatut JR (eds) Advanced anthracite technology and research, University of Scranton, 1975. Pennsyl- vania State University, Scranton, p 13 Considine TJ (2013) Powder river basin coal: powering America. Nat Resour 4(8) Davis JR (ed) (2001) ASM specialty handbook: copper and copper alloys. ASM International, Materials Park, Ohio Derrick C (1806) Memoirs of the rise and progress of the Royal Navy. H. Teape, London Dodson J, li X, Sun N, Atahan P, Zhou X, Liu H, Zhao K, Hu S, Yang Z (2014) Use of coal in the Bronze Age in China. The Holocene 24(5): 525–530 EIA (2016a) Electric power monthly with data for February 2016. U.S. Energy Information Administration, Washington, DC EIA (2016b) How much electricity is lost in transmission and distribution in the United States? Frequently Asked Questions. U.S. Energy Information Administration (EIA), Washington, DC Falconer IK (2009) Metals required for the UK’s low carbon energy system: the case of copper usage in wind farms. University of Exeter, Exeter GWEC (2015) Wind in numbers. Global Wind Energy Council (GWEC), Brussels Harris JR (1966) Copper and shipping in the eighteenth century. Econ Hist Rev 19(3):550–568 Hughes S (2000) Copperopolis: landscapes of the early industrial period in Swansea. Royal Commission on the Ancient and Historical Monuments of Wales, Aberystwyth IEA (2016a) Coal information 2016—preliminary edition documentation for online data service. Paris IEA (2016b) World coal supply. IEA Coal Information Statistics, Paris Knight RJB (1973) Introduction of copper sheathing into the Royal Navy, 1779–1786. Mariner’s Mirror 59(3):299–309 Minter A (2013) Junkyard planet: travels in the billion-dollar trash trade. Bloomsbury Press, New York Orwell G (1937) The road to wigan pier. Victor Gollancz, London Polo M, Benedetto F, Ricci A (2004) The travels of marco polo. Psychology Press, Abingdon Rapp GR (2009) Archaeomineralogy. Natural Science in Archaeology, Springer, Berlin 20 1 Doctor Copper and King Coal

Schmitz CJ (1979) World non-ferrous metal production and prices, 1700–1976. Frank Cass, London Smith BW (1965) 60 centuries of copper. history of copper. Copper Development Association, New York Smith KS, Plumlee GS, Hageman PL (2015) Mining for metals in society’s waste. The Conversation, Boston Stevens HJ (1905) The copper handbook, vol IV. Houghton, Michigan Symons JC (2003) The mining and smelting of copper in England and wales, 1760–1820. Coventry University, Coventry WCA (2016) Uses of coal. World Coal Association, London Wilburn DR (2011) Wind energy in the United States and materials required for the land-based wind turbine industry from 2010 through 2030. U.S. History of copper, VA Copper and Coal Resources 2

Abstract Copper deposits occur in rocks of all ages and are formed as a result of geological processes that concentrated copper initially dispersed through large volumes of magma or rock. The majority of copper deposits were created by hydrothermal processes when metal sulfides were precipitated from hot waters in fractures and permeable rocks in the subsurface and at seafloor hydrothermal vents. Copper minerals can also crystallise in magma chambers or form as a result of secondary enrichment processes when primary copper deposits are weathered. Porphyry copper and associated skarn, vein and replacement deposits are the most important type of copper deposits accounting for some 60% of current world copper production. Sediment-hosted stratiform copper deposits in sedimentary basins account for some 20% of historic world copper production and were some of the earliest copper ores mined. Volcanic-hosted massive sulfide deposits occur in submarine volcanic rocks and are observed forming today at seafloor hydrothermal vents. Magmatic nickel-copper sulfide deposits in igneous rocks have a very different origin than the hydrothermal copper deposits that dominate current and historic world copper production. This type of deposit forms when mafic and ultramafic magmas separate a metal-sulfide magma that sinks to the bottom of the magma chamber or flow conduit. Copper sulfide orebodies exposed at the surface are subject to weathering processes and typically show a progression from an iron-rich cap through leached rock to oxidised ore that contains copper minerals such as malachite, azurite, cuprite and chrysocolla. Copper liberated from the breakdown of sulfide minerals may also be precipitated as native copper and secondary sulfides such as covellite and chalcocite in the vicinity of the water table. The most common copper minerals in primary ores are chalcopyrite, bornite and tetrahedrite- tennantite. The earliest metalsmiths exploited near surface deposits of copper oxide and carbonate minerals, which were easier to mine and not as difficult to smelt as the underlying sulfide ores.

Coal is a combustible sedimentary rock composed of the altered remains of plants that accumulated in vast swamps and peat bogs. Increasing temperature and pressure as a result of gradual burial beneath overlying sediments subsequently transformed the plant remains into coal. Significant coal formation first occurred some 360 million years ago during the Carboniferous Period. Coal varies greatly in its physical and chemical properties because coal type and rank are independent variables that reflect depositional environment and coalification history, respectively. Humic coals are the most common coal type that form largely from woody plant remains. Sapropelic coals are less common and dominated by non-woody plant materials. Coalification is the process that produces coals of different rank, with higher rank coals having a higher carbon content and higher calorific value than low rank coals.

2.1 Copper Mineralogy and Formation

The average concentration of copper in the Earth’s crust is some 0.006%, whereas currently mined copper deposits contain at least 0.4% copper and possibly minor amounts of gold, silver and molybdenum. Deposits of copper occur in rocks of all ages and formed through a variety of geological processes that concentrated copper initially dispersed through large volumes of magma or rock. The most common types of copper deposits formed as a result of hydrothermal processes when copper and copper–iron sulfides were precipitated from hot waters in fractures and permeable rocks in the subsurface and on the seafloor at hydrothermal vents (Fig. 2.1). These deposits are often associated with intrusion of magma but can form wherever there is an elevated geothermal gradient and so are most commonly located in areas where tectonic plates collide or diverge.1 Copper minerals can also crystallise in magma chambers or form as a result of secondary enrichment processes when primary copper deposits are weathered. The most important deposit types are porphyry copper and associated skarn, vein and replacement deposits, sediment-hosted copper deposits, volcanic-hosted sulfide deposits and magmatic sulfide deposits in mafic and ultramafic rocks (Kesler 1994). From a metallurgical perspective, copper ores are divided into native copper, and oxide and sulfide ores that require very different levels of technology to extract the copper from the ore. The earliest metalsmiths exploited high grade, near surface deposits of copper oxide and carbonate minerals, which were easier to mine and not as difficult to smelt as the underlying sulfide ores. The most common copper minerals in these different ore types are shown in Table 2.1. The mineral

1Plate tectonic theory is a unifying theory like Charles Darwin’s theory of evolution that explains how the Earth works. It is based on the observation that the outer most layer of the Earth (lithosphere) is made up of a number of rigid pieces or plates that move with respect to each other. Plates collide in subduction zones, pull apart at mid-ocean ridges and can also slide past each other at transform boundaries. Volcanic activity and earthquakes mostly occur at plate boundaries. 2.1 Copper Mineralogy and Formation 23

Fig. 2.1 Hydrothermal fluids venting from chimneys on the seafloor (Pacific Ring of Fire 2004 Expedition—NOAA Office of Ocean Exploration; Dr. Bob Embley, NOAA PMEL, Chief Scientist)

Table 2.1 Commercially important copper minerals Ore type Formation Name Composition % Cu Native ores Native Copper Cu 100.0

Oxide ores Cuprite Cu2O 88.8

Malachite Cu2CO3(OH)2 57.5

Azurite Cu2(CO3)2(OH)2 55.3 Sulﬁde ores Secondary Covellitea CuS 66.5 a Chalcocite Cu2S 79.8

Primary Chalcopyrite CuFeS2 34.6 b Bornite Cu5FeS4 63.3

Tetrahedrite Cu6[Cu4(Fe,Zn)2]Sb4S13 35.8

Tennantite Cu6[Cu4(Fe,Zn)2]As4S13 47.5 aSometimes primary bSometimes secondary assemblage of a copper deposit depends on the conditions of formation including the host rock and solution chemistry in the case of hydrothermal ores as well as temperature and pressure. Copper ores exposed to air in weathering environments tend to be oxidised, whereas those in oxygen-poor subsurface environments are dominated by sulﬁde minerals. The upper parts of copper deposits exposed at the surface typically show a progression from an iron-rich cap or gossan through leached rock to oxidised ore that contains copper minerals such as malachite, azurite, cuprite and chrysocolla (Fig. 2.2). 24 2 Copper and Coal Resources

Gossan Limonite

Oxidised Malachite, zone azurite, cuprite native copper, chrysocolla. WaterTable Secondary Covellite, enriched chalcocite, zone bornite Primary Chalcopyrite zone

Fig. 2.2 Typical copper orebody with secondary minerals resulting from weathering processes

Copper carbonate and oxide minerals form more or less in situ in the near surface oxidised zone, whereas copper liberated from the breakdown of copper sulfide minerals may also migrate downwards and be precipitated as native copper and secondary copper and copper–iron sulfides such as covellite, chalcocite and bornite in the reducing environment in the vicinity of the water table (Fig. 2.2). As a result, the copper grade of the secondary enrichment zone may be significantly higher than the primary mineralisation. The most common copper minerals in primary ores are chalcopyrite, bornite and tetrahedrite-tennantite (Table 2.1). Porphyry copper and associated skarn, vein and replacement deposits are the most important type of copper deposits mined today accounting for some 60% of world production (Johnson et al. 2014). This deposit type was exploited in Utah at the mine now known as Bingham Canyon where large-scale mining began in 1906. Northern Chile currently has the largest number of major porphyry copper deposits that include Chuquicamata, El Teniente and Escondida. The majority of porphyry copper deposits occur in relatively young volcanic belts around the margins of the Pacific Ocean in the Ring of Fire (Fig. 2.3). They formed when hydrothermal fluids were expelled from cooling igneous intrusions producing a shattered and miner- alised volume of rock that can be mined by open pit or underground mass mining methods. The name comes from the characteristic porphyry texture of the intrusions that have large early formed crystals in a finer matrix. This texture is common in igneous intrusions that crystallise at shallow levels beneath volcanoes. 2.1 Copper Mineralogy and Formation 25

Fig. 2.3 The Ring of Fire (courtesy U.S. geological survey)

Copper sulfides and copper–iron sulfides occur with quartz in veinlets and disseminated along fractures in the host porphyry and the surrounding country rocks, which are variably altered as a result of chemical reaction with the hydrothermal fluids. Copper grades in the primary copper sulfide ores typically range between 0.4 and 1.0%; however, many porphyry copper deposits particularly those in the western U.S. and Chile have well-developed zones of oxidised and supergene copper ore that facilitated their exploitation by early miners. At Chuquicamata, near surface oxidised copper ores containing copper sulfate and copper chloride minerals are underlain by partially oxidised and supergene-enriched copper sulfide ore, with primary sulfide mineralisation at deeper levels (Park and MacDiarmid 1964). Atacamite, a hydrous copper chloride (Cu2Cl(OH)3), is a common phase in the upper part of the oxide zone at Chuquicamata (Fig. 2.4), and occurs with other oxide minerals in a number of porphyry copper deposits in northern Chile but is otherwise quite rare. Antlerite, a hydrous copper sulfate (Cu3(SO4)(OH)4), is the principle ore mineral at deeper levels in the oxide zone at Chuquicamata and occurs in the oxidised zone of copper deposits in arid regions around the world. Atacamite is soluble in fresh water so it likely formed by replacement of pre-existing oxide minerals as a result of interaction with migrating saline waters since the onset of hyperaridity in northern Chile (Cameron et al. 2007). Dating of the mineral gypsum (CaSO4Á2H20) that occurs with atacamite at 26 2 Copper and Coal Resources

Fig. 2.4 Atacamite from the oxide zone at Chuquicamata mine, Chile

Chuquicamata suggests that atacamite has formed and been preserved only over the last 1.5 million years reflecting the modern hyperarid climate in the Atacama Desert (Reich et al. 2009). Higher grade skarn ores form where hot metal-rich fluids released from the cooling igneous intrusion react with limestones producing rock rich in calcium silicate minerals and metal oxides and sulfides. Copper minerals in vein and replacement deposits are the product of hydrothermal fluids that were focused through fractures and permeable zones in the subsurface. Vein type deposits comprise the minerals precipitated from the hydrothermal fluid mainly sulfides with quartz and carbonate minerals, whereas replacement bodies form where hydrothermal fluids react with the host rock and locally replace it with sulfide minerals. Skarns are a specific type of replacement body that form at the contact of igneous intrusions, but igneous intrusions are not essential for the formation of replacement deposits that can also occur in sedimentary and metamorphic rocks. The classic vein deposits in Cornwall mined variously for copper, tin, arsenic and lead formed some 350 million years ago when granitic magmas intruded the slates that make up much of Cornwall. Skarn, vein and replacement deposits of copper are more widely distributed than porphyry copper deposits reflecting the range of geological settings in which hydrothermal ore deposits may form. They are less important volumetrically than porphyry copper deposits but remain an important 2.1 Copper Mineralogy and Formation 27 source of copper because of their high copper grades. One such ore deposit is the Enterprise Mine at Mount Isa in Queensland where the ore contains some 2.9% copper (Glencore 2015). Sediment-hosted stratiform copper deposits in sedimentary basins account for some 20% of historic world copper production and were some of the earliest copper ores mined in Europe and the Middle East (Zientek et al. 2010). The copper ores are said to be stratiform because the copper minerals are concentrated in dark-coloured sedimentary layers within or underlain by red beds. These copper ores formed when copper-bearing oxidising sedimentary brines reacted with organic matter-rich rock types or mixed with reduced low salinity water in sandstones. Important examples include the White Pine copper deposit in Michigan, the Kupferschiefer deposits in Germany and Poland, and the African Copper Belt that runs through the southern Democratic Republic of Congo into northern Zambia. The ore mineralogy of these deposits is quite complex and may include disseminated and veinlet chalcocite, bornite and chalcopyrite as well as native copper. At the historic copper deposits of the Timna Valley in southern Israel, copper sulfide nodules in sandstone are partially or completely altered to malachite, copper silicates and halides including paratacamite ((Cu,Zn)2Cl(OH)3)), which may reflect lowering of the water table (Asael et al. 2012). Volcanic red bed deposits such as the copper deposits of the Keweenaw Peninsula in Michigan formed in a similar way to the sediment-hosted copper deposits when native copper and silver were deposited from migrating brines in cavities in terrestrial basalt flows and conglomerates (Fig. 2.5). Volcanic-hosted massive sulfide (VHMS) deposits occur in submarine volcanic rocks and are observed forming today at or near the seafloor where hydrothermal fluids vent at mid-ocean ridges and in rift zones (Shanks III and Koski 2012). Metal sulfides precipitate as a result of mixing between the hydrothermal fluids that commonly have temperatures exceeding 300°C and near freezing seawater. The first active seafloor hydrothermal vents were discovered by the submersible Alvin in 1977 in the eastern Pacific Ocean. They were termed ‘black smokers’ because dark-coloured, particle-rich water was seen issuing from chimneys that were found to be composed of sulfide and sulfate minerals. What was even more surprising was the presence of a variety of strange life forms such as tube worms, giant clams, long-necked barnacles and spider crabs (Fig. 2.6). Life is only possible in such an extreme environment because microbes metabolise hydrogen sulfide and methane in the vent fluids and carbon dioxide in seawater to form carbohydrates through a process called chemosynthesis. Some of the exotic life forms feed directly on these carbohydrates, whereas others are scavengers or predators. A typical VHMS deposit consists of lenses of pyrite-chalcopyrite-sphalerite- galena (massive sulfide ores) overlying a pipe-shaped replacement body with veinlets of chalcopyrite and pyrite that crosscut the underlying volcanics (footwall stringer ores). The stringer zone represents the pathway through which the metal-bearing hydrothermal fluids reached the seafloor, whereas the massive sulfide lenses formed on the seafloor. VHMS deposits are mostly polymetallic and may be Cu, Cu–Zn or Zn–Pb–Cu dominated (Large 1992). They are widely distributed although the majority of deposits occur in the northern hemisphere and range in size 28 2 Copper and Coal Resources

Fig. 2.5 Native copper inﬁlling cavities in basalt, Wolverine Mine, Michigan, USA (courtesy of James St John—CC-BY-2)

Fig. 2.6 Giant tube worms surrounding hydrothermal vents (NOAA Okeanos Explorer Program, Galapagos Rift Expedition 2011) 2.1 Copper Mineralogy and Formation 29 from a few million tonnes to world class deposits like Rio Tinto and Tharsis (Spain), Brunswick No. 12 (Canada), Ducktown (USA) and Mt Lyell (Australia). Rio Tinto, Tharsis and other VHMS deposits in the Iberian pyrite belt differ from the majority of VHMS deposits in that the bulk of the material mined was pyrite used to produce sulfuric acid for the chemical industry. Copper was produced in significant quantity, however, from low grade copper ores and cuprous pyrite initially using hydrometallurgical methods described in the next chapter. In addition, the host rocks overlying the pyrite lenses are typically of sedimentary origin such that the Rio Tinto and Tharsis deposits have hybrid characteristics intermediate between sediment-hosted copper and VHMS deposits. Magmatic sulfide deposits in ultramafic and mafic igneous rocks have a very different origin than the hydrothermal copper deposits that dominate current and historic world copper production. Different silicate minerals crystallise over a range of temperatures in magma chambers that is the process of fractional crystallisation. This enriches the residual magma in metals and gases that can escape through the top and sides of the magma chamber forming hydrothermal ore deposits. In certain situations, the magma can separate into two different immiscible magmas. This happens most commonly with mafic and ultramafic2 magmas that separate an immiscible metal-sulfide magma, which sinks to the bottom of the magma chamber to form metal-sulfide orebodies (Kesler 1994). Sulfide immiscibility results from contamination of the magma with sulfur as a result of interaction with sulfur-rich crustal rocks. Economically important nickel–copper deposits with copper grades higher than 0.5% such as Voisey’s Bay (Canada), Noril’sk (Russia) and Jinchuan (China) occur in magma conduits rather than layered igneous intrusions (Song et al. 2011). The giant Sudbury deposit (Canada) is an exception that formed when a catastrophic meteorite impact melted crust and mantle rocks forming a layered igneous intrusion with nickel–copper orebodies at its base. The majority of production of particular commodities including copper comes from a small number of giant to supergiant deposits that are defined on the basis of contained metal normalised to standard continental crust (Laznicka 2014). This author identified 26 supergiants and 236 giant copper deposits using these criteria, the largest of which were porphyry copper, sediment-hosted or magmatic sulfide deposits. In a similar vein, the world’s 20 largest copper mines accounted for almost 40% of world copper capacity in 2014 (ICSG 2015). The historical importance of specific copper deposits does not always correlate with size, however, as location, depth and ore type were major factors that governed whether a deposit could be economically exploited. Near surface native and oxide ores were exploited first followed by high-grade sulfide ores once smelting techniques had been developed that is the subject of the next chapter.

2Ultramafic and mafic magmas have relatively low silica and high magnesium and iron contents, and form by partial melting of the mantle that is the layer between the crust and core of the Earth. Felsic magmas have relatively high silica and low magnesium and iron contents and can form by partial melting of crustal rocks or the process of fractional crystallisation of mafic magma. 30 2 Copper and Coal Resources

2.2 Coal Composition and Formation

Coal is an organic matter rich, combustible sedimentary rock composed of the altered remains of plants that accumulated in vast swamps and peat bogs. Plant debris accumulates in such settings because the swamp waters are deficient in oxygen that limits decay of organic matter. The water level needs to remain relatively constant over many thousands of years to accumulate sufficient plant debris to form a coal seam as some ten metres of peat are required to form one metre of black coal (Diessel 2012). Increasing temperature and pressure resulting from gradual burial beneath overlying sediments subsequently modified the plant remains transforming them into coal (Fig. 2.7). This process is called coalification and is a major factor determining coal composition. Water and volatile matter are progressively removed from coal during coalification that produces coals of increasing rank. The chemical and physical properties of coal vary systematically with coal rank as the coal changes from peat to lignite, sub-bituminous coal, bituminous coal and finally anthracite. Lignite popularly called brown coal has a lower fixed carbon content and higher moisture and volatile content than higher rank black coals so is lighter coloured and less dense. The mineral matter content of coal is also important and varies from a few percent to 49%; sedimentary rocks with less than 50% organic matter are by definition carbonaceous mudstones. Significant coal formation first occurred some 360 million years ago during the Carboniferous Period (Table 2.2), which was named for the large deposits of coal in rocks of this age in England and Wales.3 The first land plants appeared some 130 million years earlier in the Ordovician Period but lacked vascular tissue that restricted their size and distribution. By the late Devonian, plants with leaves and roots had evolved and trees with true wood grew together in the world’s first forests. Plant life proliferated in the Carboniferous in the tropical lowland swamp forests of Euramerica that straddled the equator and included terranes later to become part of North America and Europe through the process of continent assembly and breakup. The supercontinent Gondwana lay to the south at this time close to the South Pole and was subject to periodic glaciation. Giant club mosses and tree ferns dominated the coal producing Carboniferous swamps that represent the most extensive tropical mire systems in Earth history (Greb et al. 2003). Car- boniferous coals occur in basins in the USA, Canada, Britain, Europe and East Asia. Geographical proximity to Carboniferous coal deposits was a major factor influencing industrial development in Britain and Europe in the eighteenth and nineteenth centuries through the development of steam power and coke for metal smelting.

3The geological time scale is divided into a number of deﬁned stratigraphic units based on actual sequences of rocks and the fossils contained in them (Table 2.2). The boundary ages of these units are described in terms of years before the present day based on radiometric dating of suitable rock types. Absolute age dating of rocks was developed in the early twentieth century. Prior to that time the geological time scale was relative although geologists and paleontologists recognised the great antiquity of the Earth. 2.2 Coal Composition and Formation 31

Fig. 2.7 Burial, compaction and coaliﬁcation forms coal from peat (courtesy of Stephen Greb, Kentucky Geological Survey and University of Kentucky)

Table 2.2 Summary geological time scale with division boundaries based on recommendations in USGS (2010)—Fact Sheet 3059; the Tertiary Period is not recognised in some modern time scales but is still widely used on geological maps and in reports Eon/Era Period Epoch Years before present Hadean 4.6–4.0 billion years Archean 4.0–2.5 billion years Proterozoic 2.5 billion–542 million years Paleozoic Cambrian 542–488 million years Ordovician 488–444 million years Silurian 444–416 million years Devonian 416–359 million years Carboniferous 359–299 million years Permian 299–251 million years Mesozoic Triassic 251–200 million years Jurassic 200–145 million years Cretaceous 145–65 million years Cenozoic Tertiary Palaeocene 65–56 million years Eocene 56–34 million years Oligocene 34–23 million years Miocene 23–5.3 million years Pliocene 5.3–2.6 million years Quaternary Pleistocene 2.6 million–11,700 years Holocene 11,700 years to the present 32 2 Copper and Coal Resources

Coal formation has continued since the Carboniferous with three main episodes of coal accumulation (Thompson 2012). Coals formed during the Carboniferous and Permian periods comprise the bulk of the world’s black coal resources. They include the tropical Carboniferous coals of Euramerica and the temperate Permian coals of Gondwana that formed under varied cool to cold conditions from the distinctive glossopteris ﬂora. Glossopteris was a new type of plant with fern-like fronds but seeds rather than spores (Fig. 2.8), which dominated the deciduous forests of the southern hemisphere during the Permian Period. Identiﬁcation of glossopteris fossils in Australia, Antarctica, Southern Africa, India and South America provided evidence that these land masses were once joined in the supercontinent Gondwana. The second episode of coal accumulation occurred during the Jurassic-Cretaceous with black coals of this age in both the northern and southern hemispheres. Primitive conifers were the dominant trees in the humid coal forming swamps of this period with a diverse understory of cycads, ferns, mosses and liverworts. The third major period of coal accumulation was the Tertiary Period. Younger coals are typically lower rank than older coals and the majority of the world’s brown coal resources are of Tertiary age.

Fig. 2.8 The genus Glossopteris was a seed bearing plant with fern-like foliage; the name refers to the fossilised leaves that are common in Permian coal deposits (courtesy Daderot CC-zero—Exhibit in the Houston Museum of Natural Science, Texas, USA) 2.2 Coal Composition and Formation 33

Coal types are distinguished on the basis of the constituent plant materials. The majority of black coals are banded and vary from dull banded to bright banded, with four main lithotypes (vitrain, clarain, durain, fusain) distinguished at the macroscopic scale (Fig. 2.9). Vitrain has a bright lustre and breaks smoothly; it occurs mostly in thin bands or lenses. Clarain has a semi-bright to silky appearance and commonly contains thin vitrain bands. Vitrain and clarain are the major constituents of bright coal. Durain has a dull, granular appearance and is harder than vitrain. Dull coals are dominated by durain bands. Fusain is a dull, sooty black with a charcoal like appearance and occurs in thin irregular bands. This system developed from the work of Dr. Marie Stopes, a paleobotanist, who is probably better known as a pioneer campaigner for family planning in the United Kingdom. Microscopic examination of coal reveals the presence of different components that are called macerals. Macerals are grouped into three subdivisions based on their physical and chemical characteristics, which are vitrinite, liptinite and inertinite. Vitrinite is the most common maceral that forms from plant cell wall material and woody tissue. The brightness (reflectance) of vitrinite increases with increasing temperature and is used to determine the level of maturity or rank of coal and other organic matter. Liptinite forms from waxy or oily plant parts such as algae, spores and resin, and is enriched in hydrogen relative to vitrinite and inertinite. Coal oil also known as kerosene was sourced historically from coal rich in liptinite; kerosene was subsequently distilled from petroleum. Inertinite is partially oxidised plant material that may have been burnt in periodic wildfires or subject to oxidation when water levels dropped in the mire. Apart from the rare cannel coals rich in liptinite, coals range in maceral composition from mostly vitrinite to mostly inertinite reflecting the extent of oxidation of the plant materials. In this context, Permian Gondwana coals show a greater range of maceral compositions and a higher inertinite content than the Carboniferous Euramerica coals that indicates a more variable and seasonal climate (Diessel 2012). On the other hand, some coals of the Jurassic and Cretaceous periods have higher liptinite contents because peat mire communities at this time included conifer species that were rich in resinous material.

Fig. 2.9 Macroscopic classiﬁcation of bituminous coal showing the Australian (SA) and International Committee for Coal and Organic Petrology (ICCP) schemes (courtesy Joan Esterle, University of Queensland) 34 2 Copper and Coal Resources

The mineral matter content of coal is another important characteristic that influences how a coal behaves when it is combusted or made into coke. Mineral grains may be washed into the peat mire or ash deposited in the aftermath of volcanic eruptions. Minerals can also form after the peat has been buried as a result of groundwater moving through the coal bed. The most common minerals in coal are clays, quartz, pyrite and calcite that can occur as separate grains or fill fractures and smaller voids in the coal. In general, Gondwana coals have higher mineral matter content but lower sulfur content than the coals of Euramerica (Thompson 2012). Differences between coals reflect plant evolutionary changes, the environmental conditions of formation and geological factors particularly age and coalification history. Floral communities vary even within the same bog-lake complex that is evident from study of peats formed since the end of the Ice Age some 12,000 years ago. The most extensive modern peatlands occur in northern, cool temperate to subpolar latitudes, whereas the thickest modern peats are located at low latitude in Indonesia and Malaysia (Greb et al. 2003). The Vasyugan Bog Complex is part of a much larger system of raised peat bogs, lakes and river channels in Western Siberia that is comparable in size to the Permian Gondwana coal basins. It shows considerable plant diversity with swamp and forest landscapes and a range of peat deposit types. The Indonesian and Malaysian peats are limited in extent by the coastal geography and not as extensive as the vast Carboniferous coal deposits of Euramerica (Greb et al. 2003). Nevertheless, a comparison between modern temperate and tropical peatlands shows that the major factors influencing coal type are water table level and the degree to which this fluctuated (Moore and Shearer 2012). Coal varies greatly in its physical and chemical properties because coal type and rank are independent variables that reflect depositional environment and coalification history, respectively (O’Keefe et al. 2013). Humic coals are the most common coal type that form largely from woody plant remains and show a range from dull to bright lithotypes as previously discussed. Sapropelic coals are less common and dominated by non woody plant materials. Vitrinite and inertinite are the dominant macerals in humic coals, whereas liptinite is the dominant maceral in sapropelic coals. Coalification is the process that produces coals of different rank, with higher rank coals having a higher carbon content and higher calorific value than low rank coals. Coal grade refers to the overall proportion of mineral matter in the coal that is a product of the depositional environment and processes affecting the coal after burial. Coal quality is a more nebulous concept as it relates to how inorganic impurities in the coal impact on utilisation. Historically, sulfide and organic sulfur were considered the most deleterious impurities as they produced sulfurous fumes that impacted coal use for heating in the home, industrial drying as in brewing and various metallurgical smelting applications. The development of the coking process detailed in the next chapter solved many of these problems, but there are still issues and opportunities today in relation to the trace metal content of different coals. References 35

References

Asael D, Matthews A, Bar-Matthews M, Harlavan Y, Segal I (2012) Tracking redox controls and sources of sedimentary mineralization using copper and lead isotopes. Chem Geol 310–311:23–35 Cameron EM, Leybourne MI, Palacios C (2007) Atacamite in the oxide zone of copper deposits in northern Chile: involvement of deep formation waters? Miner Deposita 42:205–218 Diessel CFK (2012) Coal-bearing depositional systems. Springer, Berlin Glencore (2015) Resources & reserves as at 31 December 2015. Glencore plc, Baar, Switzerland Greb SF, Andrews WM, Eble CR, DiMichele W, Cecil CB, Hower JC (2003) Desmoinesian coal beds of the Eastern Interior and surrounding basins; The largest tropical peat mires in Earth history. In: Chan MA, Archer AW (eds) Extreme depositional environments: mega end members in geologic time. Geological Society of America Special Paper 370, Boulder, CO, pp 127–150 ICSG (ed) (2015) The world copper fact book 2015. International Copper Study Group (ICSG), Lisbon Johnson KM, Hammarstrom JM, Zientek ML, Dicken CL (2014) Estimate of undiscovered copper resources of the world, 2013. Fact sheet 2014-3004. U.S. Department of the Interior, U.S. Geological Survey, Reston, VA Kesler SE (1994) Mineral resources, economics and the environment. Macmillan College Publishing Company Inc., New York Large RR (1992) Australian volcanic-hosted massive sulfide deposits: features, styles and genetic models. Econ Geol 87:471–510 Laznicka P (2014) Giant metallic deposits—a century of progress. Ore Geol Rev 62:259–314 Moore TA, Shearer JC (2012) Peat/coal type and depositional environment—are they related? Int J Coal Geol 56:233–252 O’Keefe JMK, Bechtel A, Christanis K, Dai S, DiMichelee WA, Eble CF, Esterle JS, Mastalerz M, Raymond AL, Valentim BV, Wagner NJ, Ward CR, Hower JC (2013) On the fundamental difference between coal rank and type. Int J Coal Geol 118:58–87 Park CF Jr, MacDiarmid RA (1964) Ore deposits. W. H. Freeman & Co, San Francisco Reich M, Palacios C, Vargas G, Luo S, Cameron EM, Leybourne MI, Parada M, Zúñiga A, You C (2009) Supergene enrichment of copper deposits since the onset of modern hyperaridity in the Atacama Desert, Chile. Miner Deposita 44:497–504 Shanks III WCP, Koski RA (2012) Chapter 1. Introduction, volcanogenic massive sulfide occurrence model. In: Scientific investigations report 2010–5070–C. U.S. Department of the Interior, U.S. Geological Survey, Reston, VA, pp 5–8 Song X, Wang Y, Chen L (2011) Magmatic Ni–Cu–(PGE) deposits in magma plumbing systems: features, formation and exploration. Geosci Front 2:375–384 Thompson LJ (2012) Coal geology, 2nd edn. Wiley, New Jersey USGS (2010) Divisions of geologic time—major chronostratigraphic and geochronologic units. Fact Sheet 2010–3059. U.S. Department of the Interior, U.S. Geological Survey, Reston, VA Zientek ML, Hayes TS, Hammarstrom JM (2010) Chapter 1. Overview of a New descriptive model for sediment hosted stratabound copper deposits. In: Descriptive models, grade-tonnage relations, and databases for the assessment of sediment hosted copper deposits—with emphasis on deposits in the Central African Copperbelt, Democratic Republic of the Congo and Zambia, Scientific investigations report 2010–5090–J. U.S. Department of the Interior, U.S. Geological Survey, Reston, VA, pp 2–16 Copper and Coal Through the Ages 3

Abstract We will never know who first smelted copper. Nevertheless, archaeometallurgy has made considerable progress in identifying where, when and how our first industrial metal was made into tools. Stone tools had served us well; however, copper proved to be more versatile especially when combined with tin to make bronze. The technology for mining copper ores and extracting copper progressed slowly for the first five thousand years but gathered pace in the first century AD as the Roman Empire expanded and introduced new technology. As Europe emerged from the Dark Ages, production of copper began to increase, firstly from Mansfeld Land in Germany. By 1650, the largest European production was coming from the Falun Mine in Sweden. In the 1780s, the reverberatory furnace and Welsh coal enabled Swansea, known as ‘Copper- opolis’ at this time, to become the world’s leading copper producer. North of Huelva on the Iberian Peninsula, the mines at the headwaters of the Rio Tinto had drawn Phoenician merchants to Spain and were a major sources of copper for the Roman Empire. There was little mining activity in the Rio Tinto region after Roman mining ceased around the year 400 until 1725 when Liebert Wolters, a native of Stockholm, formed one of the first joint stock companies in Spain to develop the mines of Guadalcanal, Cazalla, Aracena, Galaroza and Rio Tinto. In 1873, after many unsuccessful attempts to make the Rio Tinto Mine profitable, it was bought by a syndicate led by Hugh Matheson for the equivalent of £3,850,000. Rio Tinto Company developed one of the first modern mines. In the latter half of the nineteenth century, rich copper ore bodies were discovered in the United States, firstly around Lake Superior in the east and then from Arizona in the south to Alaska in the north. The U.S. became the dominant world copper producer and remained so for almost 100 years until surpassed by Chile. The U.S. surpassed Britain as the major coal producer in the last years of the nineteenth century and remained the dominant producer until surpassed by China in the 1980s. At the centre of the copper mining story was Bingham Canyon.

The War of the Paciﬁc between Chile, Peru and Bolivia that erupted in 1879 was primarily fought over the right to mine saltpetre in the Atacama Desert. However, the Copper Man is testament to copper mining in the region some 1500 years ago. The mega-mine Escondida produces over one million tonnes of copper annually mostly as concentrate, although around 320,000 tonnes were in cathode form in 2015. Chile produces some 30% of the world’s copper, and one mine Chuquicamata holds the record for total copper produced. The story of Chuquicamata, ‘Chuqui’ to the local population, encapsulates the history of copper mining in Chile.

3.1 Earliest Metalsmiths

I keep six honest serving-men (They taught me all I knew); Their names are What and Why and When And How and Where and Who. I send them over land and sea, I send them east and west; But after they have worked for me, I give them all a rest. From The Elephant’s Child by Rudyard Kipling Who first learned to extract copper from ore is probably unknowable since we had gained the knowledge some eight thousand years ago, well before the Sumerians invented writing around 3300 BC. Copper was our first industrial metal and as Table 3.1 shows, copper remains one of our most valuable metals based on annual production and price. Although we will never know who first extracted copper from rock, research in the fields of archaeology, archaeometallurgy and geology has made considerable progress in answering the questions when, where and how we discovered the technology. Australian Aboriginals have probably been using fire to straighten their spears for 50,000 years. Most likely we have been using heat to alter materials (pyrotechnology) since our migration out of Africa and before. We learnt to make

Table 3.1 2015 ranking of Metal Primary Average Value ’ the world s most highly production price valued metals Mt $US/t $billion Copper 21 $5510 115 Aluminium 58 $1664 96 Steel 1212 $460 558 3.1 Earliest Metalsmiths 39

fired clay pots in the Neolithic1 Period. Pottery fragments found in a south China cave have been confirmed to be 20,000 years old, making them the oldest known pottery in the world (Wu et al. 2012). Pottery fragments from about 6500 BC have been found at Catal Huyuk in Turkey, and Hansen Streily (2000) reports the earliest known pottery kiln found in the Middle East is the one at Yarim Tepe in southern Turkey that can be dated to the second half of the 7th millennium BC. Rather incongruously, the first fired bricks only appear in Mesopotamia around 3500 BC (Goffer 2007; Woods and Woods 2011); however, the first documented appearance of a lime kiln is at the Natufian site, Hayonim Cave, in the southern part of the western Galilee dated around 10,000 BC (Al-Bashaireh 2008). Therefore, it appears that outside of China, lime-burning may have been the first pyrotechnology that made a significant impact on society. In this process, limestone is broken into small pieces, and fired in a kiln made of stone blocks, with layers of wood between the layers of crushed stone. At temperatures of 800–900°C, the limestone is ‘calcined’, that is, it breaks down as follows:

CaCO3 ) CaO þ CO2

Once the ﬁre has died down and the calcined lime dragged out, the pile of powder is mixed with water to produce calcium hydroxide or slaked lime:

CaO þ H2O ) CaðÞ OH 2

Over time, calcium hydroxide is unstable, and breaks down by reacting with carbon dioxide in the air, expelling water vapour invisibly as it does so:

CaðÞ OH 2 þ CO2 ) CaCO3 þ H2O

This last reaction allowed people to make plaster. Damp calcium hydroxide would be mixed with sand as a cheap filler to make mortar that could be used to bind, fill, and eventually strengthen a stone wall, or it could be mixed with limestone chips (or other stones) to make a terrazzo floor. Lime-burning and plaster technology appears to precede pottery making. Around 6500 BC, the people of Çayönü Tepesi in eastern Anatolia, which is present day Turkey, were laying terrazzo floors in their houses. An estimated 500 kg of plaster was required per house for the flooring at Çayönü Tepesi that would have required a considerable amount of wood fuel to burn the lime (Cowen 2009). Pyrotechnology was next extended to pottery. In the reactions inside the kiln, the clay minerals react to form compounds that are much harder and stronger, heat resistant, and, with a glaze, air tight. Potters would have found that the quality of pottery increases when firing at higher temperatures and built improved kilns to achieve those higher temperatures. Potters would also have found that charcoal

1The Neolithic Age, or New Stone Age, began about 10,000 BC in the Middle East and ended about 4500 BC. 40 3 Copper and Coal Through the Ages gave more heat than dry wood and took up a much smaller volume in the kiln. Once pyrotechnology had reached a stage where there was almost universal use of pottery and pottery kilns, plaster and terrazzo, there would have been considerable demand on local woods and forests (Cowen 2009). Some of the earliest known copper objects are hooks and awls from Çayönü Tepesi that date from around 7000 BC. The Çayönü Tepesi site contains some of the very earliest attempts at working native copper with ﬁre. According to Hansen (2013), by the second half of the 7th millennium BC, the command of pyrotechnology had developed in Anatolia through work with pottery kilns to such an extent that attempts at smelting metal might be expected. However, the earliest casting yet discovered are attested to after 5000 BC (Hansen 2013). Unlike objects made from native copper that are almost pure copper, the metal objects from Mersin (Fig. 3.1) in southern Turkey, dating from the beginning of the 5th millennium BC, show signiﬁcant amounts of impurities such as arsenic and tin indicating the smelting of polymetallic ores (Roberts and Thornton 2014).

Fig. 3.1 See Hansen (2013) for pictures of axes found at Mersin (image from Google Earth) 3.1 Earliest Metalsmiths 41

Fig. 3.2 Malachite (Golding 1999)

The discovery that copper metal could be extracted from a seemingly unrelated rock was one of the most signiﬁcant technological advances in human history, and the discoverers would eventually have an epoch named after them, the Chalcolithic2 or Copper Age. One explanation for the discovery may be deduced from the chemical reactions of heated copper minerals and the temperature that may be achieved in pottery kilns at the time of the discovery. Copper melts at 1085°C and today’s copper smelting furnaces reach temperatures of 1500°C. Firing clay into useful pottery demands the careful production of high temperatures inside a kiln. Firing to about 450°C makes pottery hard and waterproof. Firing to 1400°C makes the pot shiny and even harder, and at this temperature copper metal may be smelted out of a copper ore. Pottery kilns were therefore capable of melting copper ores but why would anyone have placed copper ores that bear no resemblance to copper metal into a pottery kiln? Perhaps a perspicacious potter (Cowen 2009) was experimenting with copper-bearing minerals for ornamental glazes. The copper ores malachite and azurite are bright green and bright blue, respectively. As Fig. 3.2 shows malachite

2Chalcolithic is derived from the ancient Greek khalkós for copper and líthos for stone. In the Near East and Europe it is the period after the Neolithic Age and before the Bronze Age, roughly between about 4500 and 3500 BC (Burton and Levy 2011). 42 3 Copper and Coal Through the Ages may be found with native copper and so the link between malachite and copper may seem obvious to us now. A potter experimenting with powdered malachite and azurite as pigments for glazing pots might have accidentally produced small droplets of molten copper, even if the kiln did not reach much over 1000°C. Malachite and azurite are copper carbonate minerals, Cu2CO3(OH)2 and Cu3(CO3)2(OH)2, respectively. As the kiln heats up, the carbonates break down releasing water vapour and carbon dioxide, and copper oxide is produced:

Cu2CO3ðÞOH 2 ) 2CuO þ CO2 þ H2O

Cu3ðÞCO3 2ðÞOH 2 ) 3CuO þ 2CO2 þ H2O

This may have intrigued the potter as copper oxide is red to black, rather than the blue and green of the original powders. If copper oxide is heated to high temperature using charcoal fuel in an oxygen-starved kiln, large quantities of carbon monoxide are produced in the kiln as charcoal burns to CO rather than CO2. Carbon monoxide attacks oxides, stripping off oxygen atoms to form CO2. At about 1100°C copper oxide forms copper metal:

CuO þ CO ) Cu þ CO2

Copper was already known and in use from sources of native copper. Once malachite and azurite were identified as sources of the already known metal copper, potters may have deliberately built kilns dedicated to smelting copper ores, and it would not be long before the basic design of the kiln was modified to make an effective smelting furnace, operated by a new kind of craftsman who was a spe- cialised smelter, our first metalsmith. The metalsmith, or smith for short, would pack charcoal in intimate contact with crushed ore to ensure more complete chemical reactions in more reliably oxygen-starved conditions immediately around the ore. The working temperature could have been raised by blowing air into the furnace through specially designed pipes called tuyeres, with or without bellows (Cowen 2009). The early worked copper carbonate ores, malachite and azurite produced relatively pure copper. However, when these were exhausted miners may have encountered copper sulfide ores below the carbonate ores and deduced that that these ores also contained copper even though sulfide ores bear no resemblance to either carbonate ores or to copper as Fig. 3.3 attests. Once smiths could extract copper from sulfide ores, copper metal became much more plentiful. However, smiths would have inadvertently smelted batches of metal that were not pure copper. Almost all copper ores contain small amounts of other elements such as arsenic, tin, zinc, antimony and nickel that combine with the copper during smelting producing numerous subtly different alloys. The alloys, although predominantly copper, have a lower melting point than pure copper, which makes melting and casting easier. The castings are better quality, and the alloy is harder than pure copper that has been worked by hammering. Paradoxically, 3.1 Earliest Metalsmiths 43

Fig. 3.3 Chalcopyrite (Golding 1999) the less pure copper ore that was available, the greater the variety of alloys the smith would produce. By trial and error, early smiths would soon come to associate a particular mixture of ores in the furnace with a particular result. In time, a skilled smith would be able to have some control over the end product, producing a specific alloy as required for the task at hand. The Bronze Age marks the time at which smiths became metallurgists, makers of magic, and were considered as heroes and gods. Bronze Age smiths were often buried with the tools of their trade such as hammers, anvils, knives and moulds (Cowen 2009). With the exception of Egypt, where arsenic was used until 2000 BC, tin replaced arsenic as the major non-copper ingredient of bronze in the 3rd millennium BC and tin bronze was the dominant metal of advanced civilisations in the western world for 2000 years. This was a fortuitous improvement for smiths because working with arsenic bronze may have resulted in arsenic poisoning that would have developed slowly, usually over a period of years. The most obvious symptoms are gradual nerve damage in the limbs. The agony of so many Bronze Age smiths has come down to us in legends of the Greek smith god Hephaestus and his Roman coun- terpart Vulcan who were lame (Cowen 2009). The proposition that some early smiths first extracted copper from the copper carbonate ores malachite and azurite and progressed to extracting copper from copper sulfide ore such as chalcopyrite is plausible. The sulfide ores could have been encountered as the miners mined below the carbonate ores. Nevertheless, the recognition that minerals like chalcopyrite that bear no resemblance to malachite would produce copper was an inspired deduction. However, the smith who first added tin to copper to form bronze was either a genius or incredibly lucky. The 44 3 Copper and Coal Through the Ages

Fig. 3.4 Cassiterite (Golding 1999)

most prevalent tin ore, cassiterite (SnO2), shown in Fig. 3.4, is a relatively scarce commodity, not commonly found in association with chalcopyrite or malachite and bears no resemblance to either. Cassiterite occurs as distinctive black grains in alluvial sands and also in mineral veins. One possibility proposed by Cowen (2009) is that cassiterite grains may have been caught up in potters’ clay. Since cassiterite melts at only 600°C, it would perhaps be accidentally smelted to tin. Alternatively, a curious smith may have wondered what would happen if these small heavy dark stones found in the river bed were added to the copper ore and so started the Bronze Age. Archaeologist had originally assumed that the tin for making early tin bronze was imported into Turkey possibly from Afghanistan as there appeared to be few sources of tin in the region. However, archaeological excavation by Aslhan Yener and her colleagues found evidence of Bronze Age tin mining at the Kestel Mine in the Central Taurus Mountains in ancient Anatolia, now southern Turkey (Yener 2009). Cassiterite crystals were seen in a stream in the Taurus foothills and nearby in a hill called Kestel there was a tin mine. Fragments of Bronze Age pottery were found in and near the mine and inside, there were veins of bright purple tin ore. The Kestel Mine has some three kilometres of tunnels, many of which are only about sixty centimetres wide, just large enough to allow children to do the mining work. Skeletons of children twelve to fifteen years of age were found in the mine, sup- porting Yener’s view that children were the miners at the site (Harms 1994). Yener also discovered industrial debris, including many tin-slagged crucibles with 30% tin content, at the adjacent mining village of Goltepe. The discovery provided clues about how the tin was smelted and established beyond doubt that tin metal had been produced and was the motivation for the mining and smelting industry there. In recreating the production process, the researchers deduced that the process began with washing the ore, much the way gold is panned. The ore was ground and then smelted in covered crucibles, into which workers blew air through reeds. Droplets of tin became encased in molten slag, which was ground out, rewashed 3.1 Earliest Metalsmiths 45 and resmelted in a labour intensive process. The numerous stone tools found on the site suggest that the ore and slag was crushed to release the tin globules. Radio- carbon dating showed that Goltepe was occupied from c. 4375–3750 BC to 2880– 2175 BC (Lehner and Yener 2014). Tin bronze is produced by alloying from 5 to 15% tin with copper. This range produces an alloy that is harder than copper even though it melts at a lower temperature. A lower amount of tin does not improve the copper enough, and a higher amount makes the alloy brittle. Kaufman (2011) suggests tin, if available, would have been preferred for bronze smelting because it is more fuel efficient than smelting arsenic bronze. However, although tin bronze smelting was independently developed in both the eastern and western hemispheres, the incentives may have differed. In South America, this alloy may have been chosen for aesthetic purposes rather than fuel efficiency because tin bronzes are gold in colour compared to arsenical copper that is silver in colour (Kaufman 2011). Tin could have been put into a crucible with the pieces of copper under a charcoal cover and both heated together. The tin would have melted at 232°C and some would have diffused into the copper and lowered the melting point of the mix to about 950°C thus eliminating the need to raise the temperature in the crucible to 1085°C, the melting point of copper (Tylecote 1976). Additionally, tin bronze would have been easier to cast than either pure copper or arsenic bronze. Tin bronze as well as its predecessor arsenic bronze makes strong hard tools and weapons that retain an edge as well or better than stone and can be resharpened. A 10% tin bronze, by a combination of heating and hammering, may achieve a Vickers Hardness (HV) of 250 or more compared to the hardness of pure copper which may only reach a HV of 100 (Shalev 2004; Scott 1992). Copper can also be alloyed with zinc to make brass and also with nickel to make cupronickel. The latter has been made by the Chinese since about the third century BC and described as ‘white copper’. However, these alloys were probably not available to the Bronze Age smiths of Anatolia (Cowen 2009). A tin bronze foil, dated to around 4650 BC, was found in Pločnik in Serbia and pictured in Radivojević et al. (2013) that contained 11.7% tin. Radivojević and colleagues postulate that malachite copper ores commonly being smelted at Pločnik were tainted with stannite, a copper, iron and tin sulfide (Cu2FeSnS4) that provided the necessary tin. This confounding discovery implies that around Pločnik, in the 5th millennium BC, an independent, very complex bronze metallurgy developed that left no traces in the metallurgy of the region after the disappearance of the Vinča and Varna cultures. It appears that tin bronze technology disappeared for over a millennium, before emerging again in the fertile crescent between the Tigris and Euphrates in the Sumerian culture around 3500 BC, some 2500 km to the SE or as already referenced in ancient Anatolia, some 1300 km to the SE. Although the proposition that copper metallurgy developed through pottery has support it may not be necessarily so. Craddock (2000) presents a compelling argument that copper smelting was an independent development and the technological roots of metallurgy have nothing in common with pottery kilns. Moreover, charcoal was necessary to provide reducing conditions during firing. 46 3 Copper and Coal Through the Ages

Fig. 3.5 Secondary copper mineralisation—image provided courtesy of Eurasian Minerals Inc.

For Chalcolithic smiths, copper oxide ores such as malachite and cuprite would not have been as difficult to smelt as the copper sulfide ores. Additionally, the oxide ores such as malachite were relatively accessible to mine since they are often found near to the Earth’s surface as depicted in Fig. 3.5. When these oxide ores are heated to a temperature in the range of 1100–1200°C, the copper is freed from its compounds and merges into droplets. If the slag is thick rather than free flowing these droplets are trapped inside and can be recovered only by breaking up the cooled slag and picking out the hardened droplets called prills. Adding a flux such as iron oxide, sand or ash helps melt the slag so that the copper droplets can sink below the slag to the bottom of the furnace where it may be removed by tapping it while hot or breaking it away when it cools. Without our knowledge of chemistry, Chalcolithic smiths worked out the requirements for a successful smelt as described by Horne (1982). They would have selected the shallow and richest oxide ores first and probably enriched the ore by crushing and selecting the portions with high copper content. The deeper sulfide ores would have been converted to oxides by roasting in an open fire. Roasting does not require charcoal, and needs only an open fire. It uses relatively little fuel because, with certain ores, once the process commences it is self-sustaining. A furnace of just the right size and a good draught was necessary because the smelting temperature is too high to be achieved on an open hearth. Finally, charcoal was required as it burns hotter than wood. Moreover, charcoal provided the 3.1 Earliest Metalsmiths 47 necessary reducing atmosphere during combustion. However, combustion is an oxidation process and oxygen must combine with the carbon, hydrogen and hydrocarbons, of which fuels are made, in order for the fuel to burn. In many pyrotechnical processes, such as ceramic production and copper smelting, the nature of the atmosphere in which combustion occurs is critical. A reducing atmosphere is produced when the oxygen level is insufficient for complete combustion to take place. An oxidising atmosphere is produced when the draught in the furnace is strong enough to provide more air than is necessary for the fuel to burn, and the excess oxygen will combine with any other suitable substances it finds such as the metal being melted. Roasting ores in an open container over a wood fire is an oxidising process. The aim is to drive off the sulfur or other impurities in the ore and replace them with oxides. However, when smelting copper oxide ores, the aim is to remove oxygen from the copper-bearing compounds. Closing the draught during smelting would create a reducing atmosphere although it would probably lower the furnace temperature below the point where melting takes place. Charcoal, a reducing fuel, provided the solution. Charcoal is produced by a process called pyrolysis whereby the chemical structure of wood breaks down under high temperature in the absence of air. Many methods of producing charcoal have been developed and the following method, used in Britain in the eighteenth century, is posited to provide an understanding of the basic process. Firstly, timber was cut and gathered into billets. A collier3 would supervise the construction of a mound of timber. Longer pieces of timber, up to two metres long, were stacked vertically in the centre and progressively shorter timber was placed vertically towards the periphery so as to develop a dome like profile. Gaps between logs were packed with small wood to make the pile as dense as possible. The dome was sealed firstly with coarse grass and then earth was spread over the grass. The pile was then set alight under the careful supervision of the Collier who ensured that timber did not flame during the 10–14 days required to complete the process. Colliers required great skill to properly build a charcoal mound and to recognise the various stages of the burn by the colour and smell of the smoke. If the wood was allowed to burn instead of smoulder, a pile of ashes rather than charcoal resulted. From experience, and by studying the changing colour of the smoke, the collier knew when coaling4 was finished and the charcoal ready. Unlike wood, charcoal is composed largely of pure carbon, the other elements having been burned off in the charring process. The main impurity in charcoal is ash, which coincidentally has fluxing properties and was a desirable addition in early copper smelting. Charcoal has a calorific value of about 30 MJ/kg, almost twice that of wood, and is about half the weight of wood. It produces carbon

3Collier is derived from the Middle English word coal. The original sense was ‘maker of charcoal’; however, it came to mean a miner who worked in a colliery or coal mine and also a ship that carried coal. The Endeavour, which Captain Cook sailed on his ﬁrst voyage that included surveying the east coast of Australia in 1770, was a collier. 4To burn wood to a charcoal. 48 3 Copper and Coal Through the Ages

Fig. 3.6 Metal workers using blowpipes (Duell 1938). Courtesy of the Oriental Institute of the University of Chicago monoxide gas when burned that creates an oxygen-starved atmosphere. Coal is also a carbon rich reducing fuel, and it might seem equally suitable; however, coal mining was hard work with primitive tools and not generally available locally. Moreover, coal contains a number of impurities that are damaging to metals. Coal did not replace charcoal as a major industrial fuel until a process was found to remove the harmful impurities. Coke was produced in Britain by charring coal in a manner analogous to the way charcoal was produced from wood and used to roast malt for the manufacture of beer in the seventeenth century. However, it was not until 1709, when Abraham Darby established a coke-fired blast furnace at Coal- brookdale in Shropshire, England to produce cast iron that coke began to replace charcoal for metallurgical processes. Coke’s superior crushing strength allowed blast furnaces to become taller and larger. Coke quickly became the preferred fuel in metallurgical processes replacing charcoal that had been the fuel of choice for at least 7000 years. The controlled use of fire in metallurgy dates from at least the 8th millennium BC when native copper was worked to produce sheets of metal that would have required extensive and prolonged hammering, and frequent annealing5 of the copper to red heat. Reduction of ores was carried out in crucibles as exemplified by finds such as those at Goltepe in Anatolia. Blowpipes or tuyeres, similar to that depicted in Fig. 3.6 in the Mastaba of Mereruka (5th Dynasty c. 2494–2345 BC), at Saqqara, Egypt may have been used to control the temperature of smelting.

5Metals are recrystallized by annealing to improve ductility prior to further working. Generally, the metal is heated until glowing and then allowed to cool to room temperature; however, copper and brass may be quenched in water. 3.1 Earliest Metalsmiths 49

Craddock (2000) proposes that the introduction of charcoal as the fuel by Chalcolithic smiths was an essential step in the evolution of copper smelting. Its high calorific value enabled the necessary high temperatures to be attained in a relatively small space, and it also produced the highly reducing carbon monoxide gas essential for the reduction of the copper oxide ores. Additionally, charcoal is a very clean fuel, free of harmful impurities such as sulfur. Consequently, charcoal remained the copper smelting fuel for millennia until it was replaced by the stronger and cheaper coke from coal during the Industrial Revolution. The first copper smelting furnaces seem to have been carefully constructed and positioned to take advantage of the prevailing local winds. Furnaces such as those found at Timna and Feynan in Israel were built into the slopes of hills just beneath the summit facing the prevailing wind. From the remaining remnants, Craddock estimates that these furnaces were approximately 25–50 cm in diameter and between 30 and 50 cm in height. Inside such structures the charcoal and ore could be contained in close proximity, and the requisite temperatures and reducing conditions obtained. However, winds are generally seasonal and even when blowing often capricious and a lull of only a few minutes would be disastrous for a slag-forming process. The introduction of bellows along with the necessary tuyeres6 in the Bronze Age overcame this problem. Designs allowing for tapping the slags followed. By the end of the Bronze Age, shaft furnaces up to one metre diameter and two metres tall were in use. The restrictions in size were imposed by the limitations of the air blast from manually operated bellows. This limitation was overcome by the adoption of water-powered bellows in the Medieval Period, thereby allowing furnaces to grow considerably, particularly for the smelting of iron in Europe. The next constraint was the physical weakness of charcoal as there was a tendency for the charcoal in the combustion zone to be crushed by the weight of the furnace charge above, thus restricting the air supply. This was overcome in turn by the use of coke from coal as fuel in eighteenth century Europe. With this constraint removed, and with the adoption of steam-driven blowing engines, the way was open for the enormous blast furnaces of the present day. However, despite their huge size and the sophistication of smelting furnaces today, the principles upon which they operate were developed thousands of years previously when we first learnt to smelt copper (Craddock 2000). Egyptians mined copper from the lower Levant7 in the 1st millennium BC; however, archaeologists have dated the first mining in the area to as early as the 5th millennium BC. Figure 3.7 shows the Chalcolithic copper mining districts of Feynan (also spelt Faynan or Feinan) and Timna in the southern Levant that were

6An opening through which a blast of air enters a furnace in order to facilitate combustion. The tuyere may have been a reed some 2 cm in diameter with a pieced clay bulb at the far end so the reed would not burn in the heat of the smelting furnace. 7The term Levant ﬁrst appeared in English in 1497 to describe the East in general, derived from the French Levant ‘rising’, referring to the point where the sun rises. In modern usage, the region west of the Syrian Desert to the Mediterranean Sea, from Aleppo in the north to the Sinai in the south roughly delineates the Levant. 50 3 Copper and Coal Through the Ages

Fig. 3.7 Timna and Feynan locations (image from Google Earth) expanded in the Bronze Age. The ancient Timna valley mining area (Timna), some 30 km north of Eilat has been extensively investigated following on the work of Professor Beno Rothenberg, who in 1959 lead the Arabah8 Expedition, sponsored by the Eretz Israel Museum, and the Tel Aviv University Institute of Archaeology. Consequently, it has been possible to reconstruct Timna’s complex history of copper production, from the Chalcolithic Period to the Middle Ages. The name Timna in derived from one of the Biblical chiefs named in Genesis 39. Professor Beno Rothenberg’s Arabah Expedition to Timna coincided with the beginning of advanced archaeometallurgical research. As this expedition and ongoing investigations have revealed, Timna has attracted miners from the Chal- colithic Period (Rothenberg and Merkel 1998) to the present day. Artefacts from what has become known as the Nahal Mishmar hoard dating from the 5th

8The Arabah referred to the section of the Jordan Rift Valley running between the Sea of Galilee through the Dead Sea to the Gulf of Aqaba. Modern geographers deﬁne only the section from the Dead Sea south as the Arabah. 3.1 Earliest Metalsmiths 51

Fig. 3.8 Crown and a sceptre from the Nahal Mishmar hoard Photographs by Clara Amit © Israel Antiquities Authority and Hecht Museum in Haifa. Credit John Bedell

Millennium BC (Gilead and Gošić 2014) show the highly developed skills of the Chalcolithic smiths. In 1961, archaeologist Pessah Bar-Adon was exploring a cave overlooking the Mishmar stream (Nahal Mishmar) that ﬂows into the Dead Sea when he discovered a hoard of 429 objects wrapped in a reed mat. The bronze, copper, ivory and stone objects, dating from the Chalcolithic Period, included prestige items such as the crown and sceptre shown in Fig. 3.8 as well as what appeared to be more utilitarian objects such as adzes and chisels. The provenance of the hoard is debated; however, the chemical composition of the objects indicates at least two different sources. The utilitarian objects are made from nearly pure copper possibly derived from ores of the kind found at the ancient mines at Timna and Faynan. However, the more elaborate prestige objects were made with a copper containing up to 20% alloy including arsenic, antimony and nickel indicating the ore originated from another source (Shalev and Northover 1993). Copper ores in eastern Anatolia and the Caucasus were capable of producing such an alloy and if so, it would show that connections between the eastern Mediterranean and the Caucasus go back into the 4th millennium BC (Muhly 1999). The shape of some of the adzes and chisels suggests these artefacts may not have been used as tools. That would explain why, after the casting and cold working some of the adzes were annealed again. Annealing as the last step in the production process would have eliminated the previously achieved results from hardening by cold hammering, an illogical and unnecessary procedure if the items were meant to be used as actual tools, suggesting these artefacts were prestige items or ingots (Hauptmann 2007). 52 3 Copper and Coal Through the Ages

The copper oxide ores contained in Nubian Sandstone at Timna required only a relatively simple pyrotechnical process to extract the copper. Local iron and manganese mineral deposits, together with silica and/or limestone either as gangue or retrieved as flux from adjoining geological formations, presented an almost ideal feed for a straightforward one-step copper smelting process. Additionally, the prevailing winds on the surrounding ridges made smelting with natural ventilation possible from when smelting commenced up until the 4th millennium BC. Forced draught through bellows and tuyeres in hole-in-the-ground smelting hearths, or at least partly above ground bowl or shaft furnaces followed. Timna offered maximum metal output with minimum energy input. Energy inputs included human effort for mining, beneficiation and smelting including powering the bellows, as well as the production of charcoal that was necessary in order to supply sufficient heat to smelt the copper. As previously described, malachite and azurite break down when heated releasing water vapour and carbon dioxide, and copper oxide is produced. As heating continues, and with the help of a flux and the reducing atmosphere provided by the charcoal, copper is formed:

4CuO ) 2Cu2O þ O2

Cu2O þ CO ) 2Cu þ CO2

ðÞflux Fe2O3 þ CO ) 2FeO þ CO2

ðÞgangue SiO2 þ 2FeO ) 2FeOÁSiO2 ðfayalite smelting slagÞ

During the Chalcolithic Period, smelting without forced air did not achieve separation of the copper metal, and the fayalite slag had to be crushed to extract the enclosed copper (Bamberger et al. 1986). Fortuitously for archaeologists, the dry climate and isolated location of Timna helped to prevent large-scale destruction and deterioration of the mining sites. Furthermore, the relatively sparse vegetation and, consequently, the lack of fuel rather than the exhaustion of the ore deposits may have been the cause of the periodic interruptions of metal production. Ben-Yosef (2010), however, proposed a military campaign of a neighbouring power probably caused the abrupt end of copper exploitation at Timna towards the end of the ninth century BC. Whatever the cause of the disruptions in exploitation, Timna offers an almost unique opportunity to observe the introduction, development and improvement of mining and smelting methods over ﬁve millennia (Rothenberg 1990). The research suggests that the ﬁrst miners at Timna mined malachite nodules by digging pits into gravel in the valleys below sandstone cliffs. The sandstone cliffs contained copper ores. As the cliffs eroded, the copper ore nodules were deposited into the silt and gravel conglomerate valleys over time. When these sources were exhausted, a form of shaft and gallery mining technology evolved that can be 3.1 Earliest Metalsmiths 53

Illustration of ancient copper mining at Timna

Depth (metres)

Eroded conglomerate layer

Mining in white sandstone layer

Distance (metres)

Fig. 3.9 Mining method at Timna (with permission of the Dead Sea & Arava Science Centre) identified by the marks left by the stone tools used to carve into the copper-bearing sandstone. Finally, a more advanced form of shaft and gallery mining technology, as depicted in Fig. 3.9, may be identified by marks left from metal tools (Smitheram et al. 2013). Numerous shafts at Timna are associated with this type of mining technology. The shaft in Fig. 3.10 shows the distinct indents from metal tools and steps protruding from the shaft wall. The Arabah Expedition dated this form of mining technology to the Egyptian New Kingdom in particular the twelfth century BC (Wilson 1977). Malachite and paratacamite were identified as the main copper minerals containing up to 36% copper (Segal et al. 1998). Numerous ‘plates’ such as those shown in Fig. 3.11 are found throughout Timna. They are the remains of shafts that have been either intentionally filled by the original miners or filled by sediment over the millennia. Smitheram et al. (2013) used optically stimulated luminescence (OSL) to confirm the dates of the various mining technologies at Timna. OSL measures the amount of energy stored in fine-sand quartz grains from the time they were last exposed to light. Thus the date the quartz grains were covered, which corresponds to the time mining ceased, may be calculated. The dates calculated using OSL confirmed that the open-pit mining took place in the Chalcolithic Period and that a mineshaft excavated using stone tools also dated from the Chalcolithic Period, affirming the dates estimated by the Arabah Expedition. 54 3 Copper and Coal Through the Ages

Approximately 80 cm diameter

Fig. 3.10 Timna shaft showing metal tool marks and protruding steps

Smelting furnaces were initially located high on the surrounding ridges, so that the wind would help raise the furnace temperatures by forced draught. The strength of the wind is demonstrated in Fig. 3.12 on a hill adjacent to one of the ancient smelting sites. The furnaces themselves are bowl-shaped as shown in Fig. 3.13. Pieces of slag are scattered around the furnaces, some containing little blobs of metallic copper created in the furnaces. The fuel was possibly collected from the local acacia trees. Today the region is very sparsely vegetated; however, examination of watercourse geomorphology indicates that, at the beginning of the 3rd millennium BC, the southern Levant was wetter than the current climate (Gallo 2014). Perhaps there was a greater abundance of vegetation for the early smelters, or perhaps wood or charcoal was carried in along with food, water and other goods for the mines. The highlands of Feynan were the closest suitable agricultural and pastoral land and supplies may have come from there. Mules or donkeys would have been the beast of burden, however, as research by Sapir-Hen and Ben-Yosef (2013) indicates that the oldest camel bones found in Israel are those from the copper mines in Timna and these dates from the late tenth century BC. Assuming acacia trees were the major fuel for smelting copper ores at Timna, when the fuel source was depleted acacia from the eastern Araba or from the Ain Ghadian– Yotvata oasis 15 km to the northeast may have been used (Ben-Yosef 2012). 3.1 Earliest Metalsmiths 55

Fig. 3.11 Recently re-excavated shaft in the foreground with plates behind

Some eighty cubic metres of charcoal would have been required to smelt one cubic metre of high grade malachite ore.9 Why the fuel would be transported to the ore at Timna and not vice versa is a mystery. Additionally, water would have had to be transported given the lack of water at Timna. Transporting the ore to the site of the fuel appears more practical and analysis of Bronze Age-Iron Age slag from copper smelting at Yotvata (Ben-Yosef et al. 2010), suggests that copper ore from Timna, as well as copper-manganese ore from the region of the manganese deposits north of Timna, were used by the prehistoric smelters of Yotvata.

9According to the few estimates available, about 8 kg of charcoal was required to smelt a kilogram of high grade Timna ore containing 32% copper (Merkel 1990; Bamberger et al. 1986), which equates to an ore to charcoal ratio of about 80:1 by volume. 56 3 Copper and Coal Through the Ages

Fig. 3.12 Near horizontal hat demonstrates the strong wind on a hill overlooking the modern mine

The Timna furnaces produced rather impure masses of metal that needed further cold working by hammering, or a further firing to produce much purer copper ingots. Later technology, with larger furnaces and hotter fires, allowed smelting to be carried out in one step (Cowen 2009). The best documented history of technological advances in early smelting methods comes from Egypt, where pictorial images flesh out the written record. The Egyptians brought new technology to Timna in the period between 1300 BC and 1100 BC, where Amalekites and later Midianites worked along with Egyptians using bronze tools (Rothenberg 1972). The smelting centres were highly organised, with areas for ore crushing, storage pits for ore, charcoal and iron oxide flux. Furnaces were clustered close together at this time on the valley floor rather than on the ridges, because the Egyptians now relied entirely on bellows to pump air through the furnaces. Each furnace was sunk into the desert sand, and was lined 3.1 Earliest Metalsmiths 57

Fig. 3.13 An example of the smelting furnaces excavated at Timna with cement. Each had one or more tuyeres in it to which bellows were connected (Cowen 2009). The Timna copper mineralisation occurs as irregularly shaped nodules consisting of malachite, chalcocite, paratacamite and cuprite with accompanying azurite, limonite, goethite, hematite, calcite and gypsum. After hand sorting, the ore to be smelted may have had a copper content exceeding 30%. Ingots found at Timna dating from the Egyptian New Kingdom contained some 94% copper. The copper ingots were sent to Egypt to be made into bronze: tin bronze by this time. Copper and tin ingots were melted together in the right proportion, and the molten bronze was used to make objects sometimes as large as the temple door depicted in the tomb of Rekh-Mi-Re (Rekhmire). Rekh-Mi-Re (c. 1479–1400 BC) was Vizier of southern Egypt during the reigns of Thutmosis III and Amenhotep II. His tomb shows well preserved scenes of daily life including the foundry scene in Fig. 3.14. The scene depicts the making of two bronze doors for the temple of Amen at Karnak. To the right of the bottom row labourers, behind whom is an overseer with a rod in his right hand, carry ingots of metal to the four groups of smelters seen at 58 3 Copper and Coal Through the Ages

Fig. 3.14 Doors for the temple of Amen at Karnak modiﬁed from Plate XVIII (Newberry 1900) the left end of the top row. The metal is melted in crucibles over charcoal ﬁres, which are blown by bellows worked by the feet. On the bottom right of the top row, two metal workers collect the molten metal in a crucible using tongs and on the left side of the bottom row pour the molten copper into the mould. In the centre of the bottom row to the right of the mould, a labourer empties a basket of charcoal. Over the heads of the three men carrying the metal is an inscription reading ‘Bringing the ingots of copper, which were brought by his victorious Majesty from the land of Syria for making the two doors of the temple of Amen in Karnak’ (Newberry 1900). Mining continued sporadically at Timna until today. Evidence for Roman mining is more evident in Wadi Amram (Willies 1991), some nine kilometres to the south of Timna and at Feynan to the north. Feynan was one of the largest copper mines in the Roman Empire and during the early fourth century persecuted Christians, mainly from Gaza and Egypt, were exiled there (Weisgerber 2006; Watts-Plumpkin 2005). Skeletons excavated in the late Roman to early Byzantine cemetery revealed that part of the population suffered from severe osteoarthritis, most likely as a result of heavy physical work. Notwithstanding, Rome had access to many sources of copper including Spain and Britain and its advances in mining 3.1 Earliest Metalsmiths 59 technology enabled deeper mining. Traders provided tin for bronze making to Rome, sourced from Cornwall and Devon, prior to the Roman invasion of Britain. Some postulate the Roman conquest of Britain was partly to access the tin mined in Cornwall and Devon; however, although Julia Caesar recorded the presence of tin after his two unsuccessful invasion in 54 and 55 BC in his Commentaries on the Gallic War, he appears more impressed by the number of cattle in Britain (Caesar 1869). Given this assessment of Britain, the successful invasion and conquest by Claudius in AD 54 may be attributed to a desire for prestige rather than resources (Faulkner 2011). In the 1st millennium BC, iron superseded bronze as the predominant industrial metal as technology developed to produce iron more cheaply from the more abundant iron ores. As explained earlier, bronze containing some 10% tin has a Vickers hardness of about 80 HV; however, after annealing a hardness of 250 HV may be achieved. The average hardness of wrought iron is 105 HV; however, two tempered Egyptian axes c. 900 BC achieved over 250 HV (Snodgrass 2000; Buchwald 2005). Once smiths mastered the annealing process for wrought iron, the Bronze Age gave way to the Iron Age. Most ancient iron in the Western world contained some 0.3% to 0.6% carbon and iron tools dating from before 1000 BC would most likely be wrought iron. The smith would have taken a mass of spongy iron from the smelting furnace, reheated it and squeezed out the combination of ferrous oxide and silica gangue by hammering it (Moorey 1999). Based on smelting experiments and metallographic investigation of iron pieces found at archaeological sites, Friede and Steel (1997) contend that primitive iron smelting and forging was conducted at temperatures above 800°C but not exceeding 1200°C. Iron melts at 1535°C and copper at 1085°C. If higher furnace temperatures were required for iron, then intuitively copper smelting should precede iron smelting. Nevertheless, it is necessary to understand how the metals are extracted from their ores before concluding that the melting point determined precedence. Copper may be smelted from malachite or other copper carbonate ores at a temperature of 800°C. However, the prills of copper extracted from the spongy cake were then heated to some 1200°C to obtain the copper to produce tools. The reduction of iron oxide ores such as hematite (Fe2O3) also occurs at about 800°C (Lucas and Harris 1948). At a smelting temperature of about 1200°C, a slag of silicates and aluminates forms that can be separated from the iron. Charcoal present as fuel provides the reducing agent as shown in the following formula for smelting of hematite ore:

Fe2O3 þ 3CO ) 2Fe þ 3CO2

The product is a spongy mass resembling a ﬂower and hence called a bloom. Pieces of concentrated iron can be broken from the bloom and forged by the smith by heating and hot hammering. The eventual product after several cycles of forging was wrought iron containing less than 0.8% carbon (Lambert 1997). On analysis, the reason the smelting of copper preceded iron may not be attributed to either their 60 3 Copper and Coal Through the Ages respective melting points or hardness. However, bronze containing 10% tin melts at 1005°C, almost 200°C below the temperature required to manufacture iron bloom and perhaps this difference explains why copper tools preceded iron tools. By the Middle Bronze Age, in the 2nd millennium BC, bronze smiths had mastered the art of obtaining maximum effective hardness of bronze by alloying copper with tin and cold working. However, smiths had far less control over the composition of iron, affected as it was by its carbon and phosphorous content. A hard edge could be made on a copper or bronze tool or weapon by cold workhardening with a hammer and a tin bronze can be work-hardened to a hardness of over 250 HV. Presumably, early ironworkers would have recognised that iron air-cooled after forging was harder and stronger than cast bronze, but not as hard as hardened bronze. They would have experimented until they found the best method for work-hardening iron to produce a wrought iron considerably harder than the best work-hardened bronze. Bronze has only a limited capacity for plastic deformation before it breaks and cannot be repaired quickly and effectively, whereas iron tends to bend and notch rather than break and can be straighten and resharpened fairly rapidly. Forging an object of iron probably required not much more effort than that involved in making it in bronze when the preparation needed for casting is taken into account; however, repair of the iron object is easier. Moreover, an iron object may easily be changed from one form to another, whereas a bronze one would have to be recast (Moorey 1999). Based on the attributes of manufacturing effort, hardness and malleability, wrought iron tools are superior to bronze tools so why did the Bronze Age precede the Iron Age? Perhaps it was only because pure copper in the form of native copper was found in association with the copper ores and hence early smiths deduced that copper metal could be extracted from the copper ores. In contrast, pure iron found on Earth, thanks to passing iron meteorites,10 is rare and not naturally associated with the ores of iron. The start of the Iron Age is often attributed to the discovery of iron smelting and smithing techniques in Anatolia or the Caucasus and Balkans around 1300 BC. This thesis is challenged, however, by scholars who place the emphasis on the price and availability of iron rather than the development of iron smelting technology on its own. Excavations at Tell Hammeh, Jordan, have revealed perhaps the earliest evidence of bloomery11 iron smelting dating to 930 BC, based on radiocarbon dating (Blakelock et al. 2009). Bloomery smelting was the main method of smelting iron ores in premodern times. Iron ore is heated using charcoal to a temperature of around 1200°C, which although below 1530°C, the melting point of iron, is suf- ﬁcient to reduce the iron oxides that coalesce as a solid mass of metal or ‘bloom’.

10Meteoritic iron is found in meteorites and made from the elements iron and nickel. Apart from minor amounts of telluric iron or native iron, meteoritic iron is the only naturally occurring form of the element iron on the Earth’s surface. 11The bloomery was the earliest form of smelter capable of smelting iron. A bloomery’s product is a porous mass of iron and slag called a bloom. 3.1 Earliest Metalsmiths 61

This bloom can subsequently be refined and shaped into the required forms by hot working or smithing. Some scholars suggest iron weapons gave the Hittites a great advantage over the bronze weapons of Egypt during the battle of Kadesh in 1274 BC. The Battle of Kadesh (also Qadesh) took place between the Egyptian army under Ramesses II that included some 2000 chariots and a similar sized Hittite army under the command of Hattusilis III. Kadesh is on the Orontes River, just upstream of Lake Homs12 near the modern Syrian-Lebanese border. The prevalence of iron weapons in 1274 BC is, however, disputed. A letter from the then King Hattusilis III (c. 1267–1237 BC), to an Assyrian King written some years later apologises for his inability to provide the Assyrians with iron, suggesting that iron was a scarce commodity (Buchwald 2005). Nevertheless, by the ninth century BC, iron had become common enough in Assyria to be used for tools and weapons and was recorded in war booty and tribute. Although iron became the most used industrial metal, copper production increased in the Roman Period, not only for coinage but also for plumbing. The name ‘copper’ is probably derived from the Latin ‘aes cyprium’, meaning ‘metal of Cyprus’. However, some speculate that the name Cyprus may have been derived from an older word for copper. With the collapse of the Roman Empire, Europe entered the Medieval or Middle Age spanning from the fifth to the fifteenth century. Bronze had been supplanted by iron for industrial use and, with the loss of Roman mining skills that had enabled deep mining, copper production decreased substantially. Hong et al. (1996) used estimates obtained from various sources of annual copper production over the past seven thousand years to support their paper on copper traces found in Greenland ice cores. Based on these estimates, Fig. 3.15 shows a peak in production of copper during the Roman Period of some fifteen thousand tonnes per annum in the first century. Production fell to around two thousand tonnes per annum following the collapse of the Roman Empire and did not increase in Europe until the European Renaissance. In the east, however, production increases in the Song Dynasty accounted for most of the peak of some fourteen thousand tonnes around 1000. Ebrey et al. (2009) report that during the Song Dynasty in China some six billion copper coins were produced in 1085. Assuming these were the Yanfeng Tongbao copper coins and based on the analysis in Wang et al. (2005), the copper content of the coins would amount to some 21,000 tonnes indicating that the estimate of 14,000 tonnes shown in Fig. 3.15 is low or perhaps the copper used to make the coins came from stockpiles mined in previous years. Conversely, the estimate of coins may be exaggerated. Nevertheless, the estimates of copper produced are of a similar magnitude and support the proposition that copper production increased significantly in China whilst Europe was in the Middle Ages. Although the use of copper decreased in Europe, bronze was still sought after and some indication of the

12Lake Homs is an artiﬁcial lake formed by a dam built around 284 by the Roman emperor Diocletian for irrigation purposes. 62 3 Copper and Coal Through the Ages

World copper production 18,000

Industrial Mostly Song 16,000 Revolution Dynasty commences

14,000 20,000 18,000 16,000 Annual world copper production 14,000 12,000 (anno Domini) 12,000 10,000 8,000 10,000 6,000 4,000 2,000 Copper propduction (t'000) 8,000 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Years AD 6,000 Copper producedCopper tonnes per annum

4,000

Battle of Roman 2,000 Kadesh occupation of Britain Renaissance begins

5,500 5,000 4,500 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 Years before 2015

Fig. 3.15 Estimated annual copper production from Hong et al. (1996) demand for bronze may possibly be gained from the production of the tin required to manufacture bronze. Cornwall and Devon possessed a virtual monopoly of European tin production for over a millennium, following the cessation of mining in northern Spain in the mid-third century. Diodorus Siculus, writing in about 8 BC, describes the export of tin from the island of Ictis (possibly St. Michael’s Mount) off Cornwall across the Strait to Gaul and on through Gaul to the mouth of the river Rhone (Hatcher 1973). Although access to tin is sometimes given as a reason for the Roman invasion of Britain, the evidence does not support this thesis. Firstly, the tin ﬁelds of northern Spain supplied easily worked surface deposits of readily accessible tin to Rome. More importantly, Roman occupation of Cornwall did not occur for almost 200 years after the Roman invasion. It is likely the invasion of Spain by the Moors in 171, which caused the temporary loss of Roman administrative control (Edmondson 1989), was the catalyst for the construction of roads in the mining regions of Cornwall and the expansion of the tin mines in the third century. Objects of tin and pewter became common in many parts of Britain, and the increased number of late third and fourth century Roman coins found in middle and west Cornwall suggest a measure of prosperity. Increased tin production in the thirteenth century from the Erzgebirge (Ore Mountains) that formed a natural border between Saxony and Bohemia did not dislodge England from her position as the main supplier of tin to the known world. 3.1 Earliest Metalsmiths 63

Fig. 3.16 Rückseite des Annaberger Bergaltars by Hans Hesse 1522 Wikimedia Commons, the free media repository

Mining around Annaberg-Buchholz in the Ore Mountains is beautifully depicted in a 1522 picture by Hans Hesse, a portion of which is shown in Fig. 3.16. The miners in Fig. 3.16 appear to be mining individual shafts in contrast to the industrial scale mining operations of the Romans ﬁfteen hundred years earlier. Even after the development of deposits in the Far East, English tin retained an almost complete monopoly of the international markets of Europe and the Middle East well into the seventeenth century. The abundance and incomparable quality of English tin, together with the indispensable role of the metal in society, helped ensure its remarkably wide distribution throughout the known world (Hatcher 1973). 64 3 Copper and Coal Through the Ages

The Crown, seeking to increase its revenues by stimulating tin production, introduced the laws of the stannaries that were expansionist as well as egalitarian. Whereas other industries were organised into craft guilds whose ordinances sought to restrict entry and output by limiting admission, tin mining laws encouraged entry and competition. The earliest known stannary charter is that of 1201, when King John granted a charter to the Cornish stannaries.13 The tinners were confirmed in their privileges of digging tin, and turfs for smelting it, at all times freely and peaceably and without hindrance from any man, everywhere in moors and in the fees14 of bishops, abbots and counts, and of buying faggots to smelt the tin without waste of forest, and diverting streams for their works. Furthermore, the stannary charter of 1305 granted stannaries exemption from many forms of general taxation; however, the price of these privileges and exemptions was a heavy tax on tin production called the coinage duty, which, in the course of the thirteenth century, came to be fixed at the rate of forty shillings per thousandweight (1000 lb) of tin in Cornwall (Hatcher 1973). Hatcher (1973) lists the annual production of tin on which the stannaries paid coinage15 in Devon and Cornwall that show a peak production from 1515 through to 1547 of on average 800 tonne per annum. Whilst the lists are no doubt an accurate record of the tin officially mined, Hatcher cautions that the volumes may be underestimated given that Cornwall and Devon were notorious haunts of smugglers and pirates.16 Tin could be melted down into small easily transportable and negotiable bars, locally called pocket tin and easily smuggled out of Cornwall’s many coves. It was not so much the customs duties that stimulated smuggling, for they were never exorbitant and did not rise above 5% on native exports, but rather the coinage paid to the Duke of Cornwall amounting at the very least to 15% of the final selling price. Unfortunately, it is not possible to equate the volume of tin mined with the quantity of bronze produced and hence the volume of copper mined because a good percentage of tin mined was used to make pewter. Pewterware was gaining acceptance in the eleventh century and an ecclesiastical Synod sitting at Rouen in 1074 and a council held at Winchester in 1076 forbade the use of wood for chalices in churches, and resolved that pewter was to be allowed as the sole alternative to gold or silver when the poverty of the congregation rendered the use of these costlier metals impossible. One hundred years later at the Council of Westminster

13Stannary is derived from the Medieval Latin stannaria (tin mine), ultimately from Late Latin stannum (tin) and the origin of the symbol (Sn) for the chemical element. 14Land held by a man from his lord for a fee. 15Tin coinage was a tax on refined tin, payable to the Duke of Cornwall in the Stannary Towns. The oldest surviving records of coinage show that it was collected in 1156. It was abolished in 1838. 16Penzance, on the western tip of Cornwall, made famous by Gilbert and Sullivan’s comic opera, knew press gangs and privateers, Barbary pirates, Spanish raiders and adventurous smugglers. However, its most famous son was Humphry Davy who invented the Davey Safety lamp and chose not to patent his lamp so that miners could use it as widely as possible (Penzance Town Council 2015). 3.1 Earliest Metalsmiths 65 of 1175, bishops were commanded not to consecrate pewter chalices; however, pewter continued to play a significant part in ecclesiastical furnishings (Hatcher 1973). Furthermore, the melting down of vast quantities of church plate to contribute towards the ransom17 of Richard 1 in 1194 necessitated the restoration of pewter to its hitherto prohibited place upon English altars. The Worshipful Company of Pewterers of London specified the composition of the pewter alloy in their 1348 Pewterers’ Ordinances. Fine pewter containing only tin and as much copper as could be absorbed, typically about 2%, was to be used for plates (sadware) as well as ecclesiastical flagons. Lay metal for pots (hollowware) was to be an alloy containing tin and some 20% lead (Weinstein 2011). Although pewterware became a household good that consumed a considerable amount of the tin produced, bronze remained in demand for both ecclesiastical and military uses as attested by the following castings. Bells made of copper and 20% tin, called bell metal, were found to be long lasting and produced a favourable sound that travels. In 1367, three great bells with a combined weight of some six tonnes were cast for Edward III, and one destined for Westminster had a weight of some four tonnes. England possessed a substantial number of bellfounders in the later Middle Ages and bells were sometimes exported (Hatcher 1973). The bronze doors each weighing four tonnes, created by Ghiberti and dubbed The ‘Gates of Paradise’ by Michelangelo, were installed in the Florence Baptistery in 1452. This was at the time of Cosimo Medici, and Florence was at the heart of the Renaissance18 that began in Italy during the fourteenth century and lasted until the sixteenth century, marking the transition between Medieval and Early Modern Europe. Towards the end of the Renaissance in Britain, the cannon of Henry VIII’s Navy Royal warship, the Mary Rose, which sank in 1545, were all bronze. Conversely, the cannon recovered from an Elizabethan shipwreck discovered off the Island of Alderney, which sank around 1592, were all cast iron. Nevertheless, the Royal Navy did not abandon bronze cannon until the eighteenth century.19 Copper production in Europe begins to increase in the sixteenth century. The gold trade of West Africa and the spice trade of the East Indies were intimately linked to the rise in the German copper industry and the rise of Antwerp. In the first half of the sixteenth century, copper from Thuringia, Saxony Tirol and Hungary poured into the port of Antwerp. The merchants of southern Germany, in particular the Fugger family of Augsburg profited from the copper mining in central Europe. In 1527 the Fugger stockpiles at Antwerp amounted to some 1850 tonnes. Some of

17The ransom paid to Henry IV (Holy Roman Emperor) was 150,000 Marks, over twice the annual income of the crown of England, hence the term a King’s Ransom (Farmer 2011). 18One proposed starting date for the Renaissance is 1401, when Lorenzo Ghiberti and Filippo Brunelleschi competed for the contract to build the bronze doors for the Baptistery of Florence. 19HMS Victory, the predecessor to Lord Nelson’s Victory, that sank in 1744, was armed with bronze cannon. HMS Victory, Lord Nelson’s ship at the Battle of Trafalgar in 1805, was armed with cast iron cannon. 66 3 Copper and Coal Through the Ages the copper was used to manufacture manillas. The name manilla is said to derive from the Spanish for a bracelet. Copper bracelets and leg bands were the principal currency in West Africa and they were usually worn by women to display their husband’s wealth. Early Portuguese traders found a pre-existing willingness of their West African counterparts to accept manillas for trade of gold and sadly also slaves.20 Copper became known as the red gold of Africa (Herbert 1984). Timber or charcoal was the commonly used fuel use in the Roman Empire, no more so than in Britain where timber was readily available. Nevertheless, coal begins to make an appearance in Roman Britain and a little coal has been found on Roman sites. The Romans built baths and temples at Bath in Somerset, south west England around AD 60. Solinus, writing in the early third century states in his Collection of Curiosities, that, in the temple of Minerva21, there are everlasting fires that never whitened into ashes, but, as the flame fades, turn into rocky balls. Historians suggest that the rocky balls were the cinders of the Somersetshire coal, which outcrops very close to Bath (Mortimer Wheeler et al. 1930). As so far described, copper production commenced some 7000 years ago and two periods of history, the Chalcolithic and the Bronze Age testify to copper’s importance. Coal was not mined in significant quantities until the nineteenth century. In the early Modern Era, mining activity increases in Europe and for a period Britain became the largest producer of copper and coal provided the necessary energy. How and why this happened is the subject of the next section.

3.2 Out of the Dark Ages

As Europe emerged from the Dark Ages,22 production of copper began to increase, ﬁrstly from Mansfeld Land in Germany, where according to Tylecote (1976) annual output was some 2000 tonnes in 1530. By 1650, the largest European production was coming from the Falun Mine in Sweden, which supplied some 3000 tonnes of copper that year. In the 1800s, Britain became Europe’s major producer with an annual output of some 22,000 tonnes by 1850. Visitors to the ancient Great Orme copper mine can walk through tunnels mined some 3500 years ago and experience the environment that prehistoric miners faced in their search for copper. The mine is located on the Great Orme headland, a

20In Benin and along the slave rivers of the Niger Delta, the price of a slave rose from 15 manillas in 1506 to 57 in 1517 (Falola and Warnock 2007). 21A bronze head of Sulis Minerva was found at Bath. Sulis was a Celtic goddess of the thermal springs and the Romans adopted the name Sulis Minerva recognising the significance of the place to the local inhabitants. 22The Dark Ages was originally a synonym for the Middle Ages; however, later historians considered only the period from the fifth century to the eleventh century as the Dark Ages. 3.2 Out of the Dark Ages 67 spectacular mass of layered limestone in North Wales that juts upwards and out- wards into the Irish Sea (Zhuwakinyu 2013). Bronze Age miners would have been attracted to the site owing to the relative ease with which the copper ore could be extracted. Because the limestone encasing the copper ore veins had been subjected to dolomitisation, the rock had become soft and brittle and would have been easily scratched or dug away. Archaeological excavations of the ancient workings have uncovered numerous tools that provide an indication of the mining methods of the Bronze Age miners. Over 8000 bone scrapers and picks used to extract the copper ore have been found and some on display are stained green by the copper ores. Stone hammers were used to break the rock and many of the more than 900 found show signs of heavy hammering on one end. Additionally, fires were set to weaken recalcitrant rock (Zhuwakinyu 2013). Fire-setting was a method of mining used from prehistoric times and was continued throughout the Middle Ages,23 and up until the introduction of gunpowder rock blasting in the seventeenth century in Europe or somewhat earlier in China. Pliny the Elder (AD 23–AD 79), in his Naturalis Historia Book XXXIII, describes fire setting in the section on gold mining as follows:

masses of flint are encountered, which are burst asunder by means of fire and vinegar, though more often, as this method makes the tunnels suffocating through heat and smoke, they (the flint) are broken to pieces with crushing-machines carrying 150 lbs. of iron (Secundus 1952). Georg Bauer, whose pen name was the Latinized Georgius Agricola, describes the art of mining, refining and smelting metals in De Re Metallica. Production of the excellent and much copied woodcuts for the book, such as Fig. 3.17 depicting fire-setting, delayed its publication until 1556, a year after his death. The first English translation of De Re Metallica was privately published in London in 1912. The translators were Herbert Hoover, a mining engineer (and later President of the United States), and his wife, Lou Henry Hoover, a geologist and Latinist.24 Agricola (1556) made the following observations of fire-setting:

if the excavation is low, only one pile of logs is placed in it, if high, there are two, one placed above the other, by which plan the lower bundle being kindled sets alight the upper one; and the ﬁre being driven by the draught into the vein, separates it from the rock which, however hard it may be, often becomes so softened as to be the most easily breakable of all. Applying this principle, Hannibal, the Carthaginian General, imitating the Spanish miners, overcame the hardness of the Alps by the use of vinegar and ﬁre.

23The Middle Ages or Medieval Period delineates the period beginning with the collapse of the Western Roman Empire in the ﬁfth century until the Renaissance in the ﬁfteenth century. 24Herbert Hoover, U.S. President from 1929 to 1933, was a mining engineer who worked in Coolgardie and Kalgoorlie, Australia in 1897/98 and subsequently in China from 1899/1900 during the Boxer Rebellion. 68 3 Copper and Coal Through the Ages

Fig. 3.17 Fire-setting (Agricola 1556) credit Dover Publications, Inc.

Fire-setting at the Great Orme Mine may also have included quenching of the hot rock with water intended to increase the brittleness. Having made the rock brittle, miners used stone hammers to break the rock from the face, and wooden or antler picks, exploiting the natural rock cleavage, to extract the copper ore. Despite such primitive methods, archaeologists estimate that some 1700 tonnes of pure copper were extracted from the rich complex during the Bronze Age, sufficient to forge perhaps millions of bronze axes. A maze of tunnels and caverns has been explored to 70 m below the surface. Archaeological excavations uncovered over seven kilometres of Bronze Age underground tunnels, and it is believed that over eight kilometres of passages remain to be discovered (Zhuwakinyu 2013). Some 180 km to the east of the Great Orme lies Ecton Hill, which is surrounded by the spectacular landscapes of the Peak District National Park, in the Midlands of England. Copper mineralisation around the Deep Ecton and Dutchman mines, 3.2 Out of the Dark Ages 69 which were developed between the seventeenth and nineteenth century, is contained within folded beds. Thin layers of shale have been washed out leaving joints that acted as conduits for the mineralising fluids. The open structure of these joints would have made mining to some depth relatively easy with primitive tools without the need for fire-setting (Barnatt and Timberlake 2013). An antler mining tool found in a small partly choked underground passage in the late 1990s was radiocarbon dated to between 1880 BC and 1630 BC, prompting further investigations in 2000 and 2009. Subsequently, ten hammerstones and five split cattle long bones, possibly used as scrapers, were found at the Dutchman Mine. Radiocarbon dating of the bone tools suggests mining started sometime after 2000 BC and ended before 1570 BC. The nearby copper deposits at The Lumb were not rich; however, Barnatt and Timberlake (2013) estimate that some 500 kg of ore were removed in prehistory, whereas ten times as much may have been recovered from the richer Dutchman Mine and Deep Ecton Mine. Bronze Age mining at these two mines may have been up to 30 m deep. Based on these estimates of the ore mined, the total amount of copper extracted from all the ancient mining in the area was likely to be in the order of a few tonnes. Recent reanalysis of the dates of eleven Bronze Age mines in Britain indicates an eastward shift in the start date of mining for copper, moving from the middle of Wales to the north of Wales and then into the northwest of England over a three-hundred-year period. This progression in working dates, and the similarity in mining method and tools, points to a shared knowledge and experience linking the miners of Ecton with those of the Great Orme that was worked at the same time with very similar sets of bone tools. There is little evidence of ancient copper mining in Britain after 1400 BC; however, mining appears to have continued at Great Orme until 1000 BC (O’Brien 1999). Although the Romans mined some copper-bearing outcrops, they appear to have sunk no new mines. Little working of copper appears to have occurred following the period of Roman rule and during the reign of Elizabeth I, Britain depended on imports of copper from the Continent. Likewise, there was only limited mining for copper during the Middle Ages in Europe; however, the Stora Kopparberg Mine near Falun in Sweden was active (Rowlands 1996). Elizabeth I helped re-establish mining by granting a patent to work copper ore in Britain to German industrialists in 1564. The company of the Mines Royal was established in 1568, largely due to the influence of Elizabeth’s Secretary of State, William Cecil. The company brought many skilled copper workers to England from Germany and was given the sole right to mine copper in most of England and in all parts of Wales. As well as developing the mining of copper, the Mines Royal Company also set up one of the earliest smelting works of the modern period, at Neath in South Wales. The Stuart Kings looked on metals almost purely as producers of royal revenue; however, the Government did help the industry in 1625 by imposing a duty on imported copper, especially from Sweden which at that time monopolised the 70 3 Copper and Coal Through the Ages

European copper market. Copper mining was virtually at a standstill in Britain during the reigns of Charles II and James II. Acts of Parliament in 1689 and 1693 abolished the restrictive rights of the Company of Mines Royal and the Mineral and Battery Company, enabling anyone prepared to risk their capital to search freely for copper ore previously the monopoly of the two companies (Symons 2003). Competition was further encouraged during the same period by the Government’s action in permitting the exportation of copper duty free, in using British copper for making coins, and in levying additional protective duties to discourage the importation of Swedish copper. There was also an increased demand for copper in the early eighteenth century for war purposes as well as for manufacturing many domestic articles. The growing demand encouraged the search for new ore deposits resulting in discoveries in Cornwall in the early part of the century and by the mid-eighteenth century prospectors were also active in Anglesey, Wales. The medical virtues of the copper containing liquid ﬂowing down the slopes of Parys Mountain, in north east Anglesey, were acclaimed by Dr. John Rutty in his 1760 address to the Royal Society. He listed the liquids many medical beneﬁts including the treatment of ulcers and mange on the skin and by mouth for the cure of diarrhoea, and worms. Waters on the summit of Parys Mountain were known as the Mine Pool, and a hearth for smelting lead and some pieces of lead and charcoal and a plate of copper weighing some 14 kg are possible evidence that the Romans mined on Parys Mountain (Long 1833). However, no further metallurgical activity seems to have taken place there until 1779, when there was an unsuccessful commercial venture to recover copper out of the liquid emanating from the mountain by cementation—the precipitation of copper using iron. The cementation reaction is:

CuSO4 þ Fe ) Cu þ FeSO4

The renewed activity and the search for deposits of copper ore on Parys Mountain in the second half of the eighteenth century was partly caused by naval demand for copper sheathing of British warships that provided a good market for high quality copper (Rowlands 1996). Copper sheathing was placed over the under-water portions of a ship’s hull, in order to prevent attacks of teredo worm and the fouling of the bottom.25 Below the water line of Admiral Lord Nelson’s Flagship HMS Victory, the oak hull was covered with 3923 sheets of copper 1.2 m by 0.3 m weighing some seventeen tonnes. This sheeting was added in 1780, twenty-ﬁve years before the Battle of Trafalgar. On the day of the battle the Royal Sovereign was the ﬁrst British ship into battle. She had recently had a new copper bottom, which had increased her sailing speed, and she hit the allied line at noon, at least ten minutes before the next ship (Antill and Rickard 2014).

25‘Copper-bottomed’ became synonymous with reliability and trustworthiness. 3.2 Out of the Dark Ages 71

Sir Nicholas Bayly owned a portion of land on Parys Mountain and had leased the mining to a Messrs Roe and Co. One account of the first discovery of a significant lode of copper is that in a final desperate attempt to strike a rich vein, Messrs Roe and Co sent a Derbyshire miner called Jonathan Roose to Parys Mountain. Within two days on 2 March 1768, he discovered a rich vein of copper ore a few feet below the surface. The following lines on the tombstone of Jonathan Roose in Amlwch parish churchyard support this version of the tale.

Here lived one who’s mind had long to bear A toilsome task of industry and care. He first yon mountain wondrous riches found First drew it’s mineral blessing from the ground He heard the miners first exhulting shout Then toiled for over 50 years to guide it’s treasures out. Although an entertaining tale, according to Rowland (1992), the discovery is unlikely to have been as sudden or unheralded as the tale portrays. It is difficult to understand why, if Parys Mountain looked like being a failure, Roe and Co. erected smelting works at Liverpool in 1767. It is also significant that a fortnight before the supposed great discovery, Bayly sent his agent, William Elliott, to Macclesfield to demand that he be made a partner with Roe and Co. so that he could share in the direct profit of the mine. Although there were prolonged discussions, Bayly did not become a shareholder. Nevertheless, such last minute discoveries are not unknown in mining,26 and perhaps the epitaph on Roose’s grave should be the final word on the 1768 discovery on Parys Mountain. The orebody on Bayly’s land extended onto the neighbouring Parys Farm and in 1774, Edward Hughes began to mine Parys Farm in partnership with Thomas Williams, an Anglesey lawyer who had been employed to help him in the legal struggle with Bayly over the mining of the ore. The great Parys Mine Company was formed in 1778 (Rowlands 1996). In 1780 Williams erected rolling mills and works at Greenfield in North Wales. A partnership was formed with John Westwood of Birmingham who had patented a cold-rolling method that would help provide copper sheet and copper nails for naval sheathing. Williams also set up the Stanley Smelting Company, with refineries at St Helens and Swansea. An article in The ‘The Penny Cyclopædia’ (Long 1833) describes the ore as a sulfurate of copper intermixed with black copper, blue and green carbonate and some strings of native copper. This is a reasonable description of what is now classified as a polymetallic deposit that has been partly oxidised. Since then numerous minerals have been identified at Parys Mountain including chalcopyrite, sphalerite, covellite, galena and pyrite. Parys Mountain is the type locality for

26William Knox D’Arcy (11 October 1849–1 May 1917) made a fortune from the Mount Morgan Mine in Central Queensland, Australia. On his return to England he began investing in the exploration for oil in Persia in 1901. In 1908 with his finances running out, Mr. Reynolds the drilling supervisor received a telegraph ‘drill to 1600 feet and give up’. Fortuitously, on the 26 May 1908, when the rig drilling near the village of Masjid-i-Sulaiman in south west Iran reached 1180 feet, a fountain of oil burst out into the dawn sky. Within a year, the Anglo-Persian Oil Company, which would one day become BP, was in business (BP 2014b). 72 3 Copper and Coal Through the Ages anglesite (PbSO4). The primary minerals were deposited from hot fluids exhaling onto the seabed, much in the same way as is observed today in deep sea ‘black smokers’ as described in Chap. 2. This hot water was rich in sulfur and consequently most of the minerals formed contain sulfur. Chalcopyrite is reported to have been the principal source of copper at Parys Mountain (Greenly 1919). After being quarried from the open cast mine, the richer ores were initially sent to smelters in Swansea and Stanley. Workers, mainly women and children, crushed the remaining ores, containing between 1 and 2.5% copper, into small pieces about the size of walnuts using hammers. The ore was piled into oblong heaps about 1.5 m high and becoming narrow by degrees till it peaked like the roof of a house. The pile was set on fire at both ends with coal, and burnt for six to ten months, with no additional coal as the great quantities of sulfur in the ore kept the pile burning until the inflammable parts were consumed. This process was commonly known as calcination; however, it is now more correctly called roasting. The process of calcination derives its name from the Latin calcinare (to burn lime) due to its most common application. Originally, calcination referred to the heating of limestone above 900°C to drive off the CO2 and produce lime:

CaCO3 ) CaO þ CO2

Typically, calcination refers to any process where the material is heated to drive off volatile organics, CO2, chemically bound water or similar compounds. In contrast, roasting generally refers to exothermic reactions that provide sufficient heat to complete the reactions after initial heating. In the following simplified example for chalcopyrite, about one third of the sulfide in the ore is oxidised, producing a mixture of copper and iron sulfides, sulfates and oxides. Roasting involves not only heating, but also reaction with a gas, most often air because it is free:

CuFeS2 þ 4O2 ) CuSO4 þ FeSO4 13 2CuFeS þ O ) 2CuO þ Fe O þ 4SO 2 2 2 2 3 2

The early roasting kilns were at Amlwch located on the coast three kilometres to the north of Parys Mountain. In 1778, a new company interested in the preparation of brimstone27 put forward proposals to build better kilns on Parys Mountain to remove the prodigious quantity of sulfur in the ore (Rowlands 1996). The extraction of the copper from the pyritic Parys Mountain ore was thence forth subsidised by the extraction of sulfur. Likewise, the production of copper in Europe would often be subsidised by and sometimes be a by-product of sulfur extraction. Sulfur was known to the Chinese as early as the Zhou Dynasty (sixth century BC). Early in the Ming Dynasty (1368–1644), a method of extracting sulfur util- ising three pots as depicted in Fig. 3.18 was adopted in Shanxi. The upper pot,

27Brimstone is the archaic name for sulfur; see Revelation 19:20 King James Version. 3.2 Out of the Dark Ages 73

Fig. 3.18 Ming Dynasty three pot furnace after Zhang (1986)

(1) Furnace wall; (2) Coal fire; (3) Upper pot containing pyrite; and (4) the two lower pots, into which the sulfur vapour condensed. containing pyrite, was placed upside down. After heating, sulfur vapour entered the two pots buried in the earth and condensed (Zhang 1986). The Chinese invented gunpowder, perhaps as early the Tang Dynasty (618–907), but certainly by the Song Dynasty (960–1279). The contemporary gunpowder recipe called for ‘Jin Zhou sulfur’ both because it was pure and because it was easily transported. Jin Zhou is near Kaifeng, which was the capital of the Northern Song Dynasty, and was the centre of gunpowder production. Great quantities of sulfur were required to supply the gunpowder needed to sustain the continual ﬁghting against foreign countries. This need led to the manufacture of sulfur from pyrite in Shanxi. A purer sulfur was obtained from pyrite than obtained from natural sulfur. Using purer sulfur (along with purer potassium nitrate) led to an improved gunpowder that was an explosive rather than an incendiary weapon. The number of areas producing pyrite-derived sulfur greatly increased in the Ming Dynasty, and documents from the period mention that the emperor allowed the central and four local governments to buy about ﬁve tonne of sulfur per year to replenish their supplies for gunpowder manufacture (Zhang 1986). On Parys Mountain, the new company paid for the construction of the kilns on the mountainside and for a fee of £50 per annum burnt all the ore. In return the new company owned the sulfur that was extracted. The heating took place over a period of six months and the copper ore was reduced to a quarter of its original mass containing a much higher percentage metal (Rowlands 1996). Nevertheless, cementation using the water that accumulated at the bottom of the great hole where mining took place produced a sludge containing a greater percentage of copper than that of the richer copper ore, or lower grade ore after roasting. The copper cementation reaction may be expressed as:

Cu2 þ ðaqÞþFeðÞ) s CuðÞþ s Fe2 þ ðÞaq ; where (s) signiﬁes solid and (aq) aqueous. 74 3 Copper and Coal Through the Ages

Copper ions in solution (aqueous) are precipitated out of solution in the presence of solid iron because copper is higher on the galvanic series than iron (Agrawal and Kapoor 1982). Ironically, this same galvanic reaction occurred when the British Navy commenced copper sheathing vessels. In 1761, the frigate Alarm was sheathed with copper prior to a two-year voyage to the West Indies. Although the sheathing proved to be successful in keeping the hull clean, the iron bolts which secured the frame and planking were found to be corroded by the galvanic action of copper on the iron bolts. The Navy Board solved this problem in 1783 by using copper/zinc alloy bolts (Staniforth 1985). In modern industry, the galvanic process is often employed in many situations to protect more important or expensive metal parts, e.g. sacrificial anodes attached to outboard motors. The water, which was saturated with copper ions, was drawn up from the mine in buckets raised by whimseys as depicted in Fig. 3.19. It was then transferred into specially prepared ponds or pits of about ten metres long, five metres wide and half a metre deep. There were many of these pits set at about two metres apart and at different levels. This enabled the pits to be drained whenever the sludge containing copper had to be removed. Large quantities of scrap iron were placed in the pits and left there for the galvanic reaction to take place. The scrap iron was regularly turned until it finally dissolved leaving copper precipitate mixed with the mud at the bottom of the pit. After the pits had been drained, the copper ‘mud’ was removed, dried, baked and smelted. The precipitate mud contained between 20 and 30% copper (Rowlands 1996).

Fig. 3.19 Interior of one of the copper mines on the Paris Mountain: 1792 by John Warwick Smith. By kind permission of the National Library of Wales 3.2 Out of the Dark Ages 75

The mining processes of the nineteenth century mine are imaginatively described by Enoch Jones in an 1848 pamphlet.

Gunpowder makes its way much further; the manner in which it is used in blasting at these mines is the best and most effectual ever discovered….When the ore is thus blasted it is conveyed in barrows to the mouth of the shaft, there put into large wooden trestles, called kippies, and drawn to the surface by a whimsey of two horse power, from the various depths of 100 to 200 yards. After the ore has been brought to the surface, it is wheeled to a commodious section to be broken; for this operation the miners use the phrase ‘rapscaling’ this being done, it is conveyed to tents, each containing 10 to 20 ‘copper ladies’ whose occupation it is to break the ore into lumps of about one inch in size, at the same time collecting as much waste as possible from the ore. The appearance of these women called ‘copper ladies’ is very singular; they sit in a row before a square block of iron, on which they break the copper ore; the fingers of the hand which grasps the ore are covered with iron, while the other gaily handles a hammer of about 4 lbs in weight, and thus they merrily toil. The copper thus broken is carried to the kilns for calcining as before mentioned. The copper waste which is thrown aside by these ‘ladies’ is washed by numerous groups of boys, whose lynx-eyed quickness in selecting the copper from the waste is truly astonishing (Jones 1848). The horse driven whimseys described by Jones are not captured by Smith in Fig. 3.19; however, two men are operating a counter balanced windless28 simultaneously rising and lowering the large buckets (kippies or kibbles) containing the copper ore. The mining of the relatively low grade ore on Parys Mountain was made possible by the technological advancement of blasting in mining. Although there is evidence of gunpowder use at Italian quarries in the sixteenth century, according to Hollister-Short (1994), the first certain date for the use of gunpowder in underground mining is furnished by an official report of a public demonstration of blasting carried out in 1627 by Caspar Weindl in the workings of Oberbieberstollen Mine, adjacent to the silver and gold mining town of Banská Štiavnica (Schemnitz), in Slovakia. Weindl’s technique travelled fast and within 50 years virtually every ore mining region in Europe was familiar with the technique of ‘boring and shooting’ rock. The 1694 Anhalt-Dessau or Anhalt-Harzgerode silver medal depicts mining in the principality. Shot firing is illustrated on the left side. The enlarged oval section of the coin in Fig. 3.20 shows the shot firer lighting the shot and making a hasty retreat from the gallery. He is shown wearing a leather jacket with a deeply cut tail suited to working in confined rough rocky spaces often in a lying position (Krehl 2009). Gunpowder offered a much more effective way of breaking rock than fire-setting; however, a method of drilling holes into the rock was required to make efficient use of the gunpowder. Hammer and tap or rock drilling by hand uses the same method of penetration as modern rock drills. When a chisel bit hits a rock

28An excellent example of counter balancing is the Lynton and Lynmouth Cliff Railway, a water-powered funicular railway joining the twin towns of Lynton and Lynmouth on the rugged coast of North Devon. Built in 1900 and still in service. 76 3 Copper and Coal Through the Ages

Fig. 3.20 Anhalt-Dessau 1694 silver medal surface, the induced stress causes a shatter zone around the bit. If the bit is then continuously rotated a few degrees and hit, a circle of crushed rock will result and consequently a hole in the rock can be created to take the gunpowder. Hollister provides the following description based on later records from the Rohe Birke Mine in Saxony and other records, to reconstruct the technique Weindl used in 1627. The two-man team’s first task was to make a small hole with a hammer and wedge in the face to be bored in order to provide a firm key for the borer so as to avoid its wandering offline as it was struck. Deep boreholes, one metre or over, were drilled using successively short, medium and long drills the latter about 1.5 m long and about 5 cm thick. The worker holding the borer turned the borer a quarter turn after every blow from the hammer to allow the flanged tip of the borer to bite into fresh rock. During drilling, water was introduced into the hole if the rock was dry so as to speed up the action. The ‘stone flour’ produced by boring was cleared from the hole from time to time using a tool with a scoop at one end and a slot for a rag at the other. When the hole was complete, a quantity of loose powder was poured in and then began the work of blocking up the remaining space in the hole. This was done with plugs of hardwood about 75 cm long and 5 cm in diameter. The plug was prepared by first boring a hole centrally through its length to provide space for the powder train. The cylinder was then sawn diagonally lengthwise to produce two cone-shaped wedges. These were driven into the shot-hole with a heavy sledge hammer weighing some ten kilograms. These plugs on their own were insufficient to confine the blast and it was necessary to place a 15 cm square iron plate some 3 cm thick with a central hole for the shot line over the mouth of the hole. The plate was secured in position by wooden props wedged into the gallery (tunnel) walls. After firing, mining could be resumed almost immediately and the report on Weindl’s demonstration blast noted that the smoke had cleared after fifteen minutes. As described earlier, it was many hours before mining could be recommenced using 3.2 Out of the Dark Ages 77

Fig. 3.21 Drilling an upper at East Pool Mine c. 1895—photographer J.C. Burrow: © The British Library Board, 7105.e.21, item 24 the fire-setting technique. Powder blasting commenced in the Peak District of Britain in the 1670s (Barnatt et al. 1997). Rock drills powered by compressed air only began to replace drilling by hand held hammer late in the nineteenth century. Lees (2009) describes the drilling method employed in Cornish mines just prior to the introduction of modern rock drills. An octagonal in section steel drill was hit by a sledge hammer of between 2.3 and 4.5 kg ‘hammer and tap’ as shown in Fig. 3.21 until a hole of sufficient length for the explosive was made. By the sixteenth century, as the ravages of the Black Death subsided, Europe’s population had rebounded to the mid-fourteenth century level. By the end of the sixteenth century with increasing population and industrialisation, Europe’s forests were under pressure from the increasing demand for timber both for construction and fuel. Gunpowder blasting eliminated the need for profligate wood consuming fire-setting, reducing the pressure on local forests. Whether due to the price signals brought about by timber scarcity or prohibitions, changes occurred to reduce the pressure on forests. Mining of coal from Europe’s Carboniferous basins that stretched from Russia in the east to Scotland in the west was increasing. Besides replacing wood fuel, cheap coal enabled the manufacture of bricks that replaced timber as a building material. Additionally, forest management regimes sought to maximise wood yields by systematising the harvesting of forests, and canals were constructed that enabled wood from remote but still forested regions to flow to wood-deficient regions. Jones’s(1848) pamphlet also includes an equally descriptive passage on the metallurgical process at the Parys Mine that for brevity is summarised here. Twenty reverberatory furnaces are designed so that the ore is melted, not through coming into immediate contact with the fuel, but by the reverberations of the flame upon it 78 3 Copper and Coal Through the Ages

Reverberated or reflected heat

Coal Copper ore

Fig. 3.22 Reverberatory furnace

(Fig. 3.22). These furnaces are divided into: 6 roasters, 6 ore furnaces, 3 calciners, 3 precipitates and 2 refiners. The process commences with calcining (roasting) of approximately three tonnes of copper ore for twelve hours. The calcined ore is then melted in the ore furnace and the melted matter is let out at a hole opened in the side of the furnace into adjoining sand pits, where it becomes granulated as it cools to form coarse pigs. This granulated metal is subjected to calcinations and fusions alternately, until it contains from 80 to 90% of pure metal. In this state, the bars or pigs are put into the refining furnace and gradually melted. The surface of the metal is covered with charcoal, and a pole, commonly birch wood, is then held in the liquid metal which causes considerable ebullition, owing to the evolution of the gaseous matter and this operation of poling is continued until the refiner ascertains, by various trials, that the copper is in the proper state of purity and malleability. Jones (1848), reported that ‘The copper sold from these works commands in the market fully £5 per tonne above the market price, on account of its extreme purity and malleability.’ The reverberatory furnace enabled Swansea, nickname at the time ‘Copper- opolis’, to become the world’s leading copper producer. The coal seams, outcropping on the coast in the southwest of Wales provided the fuel for the ‘Welsh process’, which used coal-fired reverberatory furnaces rather than charcoal or coke fuelled blast furnaces, at that time the default process outside of Wales, to smelt copper (Evans and Saunders 2015). The Penny Cyclopædia in 1838 describes how the cupol or cupola, sometimes referred to as a reverberatory furnace, is employed for the smelting of copper. The ore and fuel do not come in contact but the furnace is so contrived that the flame only passes over the ore as shown in Fig. 3.22 (Long 1838). 3.2 Out of the Dark Ages 79

Thus coal, a cheaper and abundant reducing fuel, could be used without any of the impurities within the coal coming in direct contact with the ore. Consequently, coal replaced charcoal as the dominant fuel for copper smelting and forests of timber were no longer required, although charcoal was not eliminated from the process altogether. Nevertheless, the Welsh method required a prodigious amount of coal. An estimated thirty tonnes of coal were required in the 1730s to treat ten tonne of ore that yielded one tonne of copper (Symons 2003). Records of imports and exports for Swansea in 1850 indicate that the coal to ore ratio was still approximately 2:1 for imported ores containing on average 9% copper (Hughes 2000). Even with the improved thermal efficiency, coal still accounted for almost half of the total smelting cost (Newell 1990). Mansfeld Land, encompassing the eastern foothills of the Harz Mountains in Germany, was producing copper contemporaneously with Swansea. There, copper was smelted in blast furnaces where the copper ore was placed in the furnace along with charcoal or coke. Scoffern et al. (1857) reported that the Mansfeld process was characterised by the expenditure of a large amount of time and labour, with an economy of fuel, while the Welsh process was distinguished by great economy of time and labour, with a comparatively large expenditure of fuel. In Mansfeld, economy of fuel was of more importance than that of labour or of time while in Wales, fuel was so cheap that economy of labour and of time was of very much greater importance. Nevertheless, according to Vivian (1881), although the Mansfeld process required less coke, coke was about two and a half times more costly than coal, hence the reverberatory furnace was still more economical per tonne of copper produced. Sadly, in the 1730s, some of the leading slaving houses in Bristol were interested in expanding their production of copper in order to produce the manillas mentioned previously that facilitated the slave trade. The first known print of the White Rock works in Swansea from 1744 identifies one of the structures as the Manilla House, where these objects for the slave trade were produced or stored.29 Symons (2003) provides a more succinct description of the Welsh process of smelting that evolved during the eighteenth century and he includes the process flow chart depicted in Fig. 3.23. Jones (1848), includes a paragraph on the importance of coal to the process:

As the produce of our mines requires ﬂuxes for melting, ores from all parts of the world are extensively bought to assist the fusion of our native production. A faint idea as to the extent of these works may be estimated when we say that upwards of 30,000 tonnes of coal are consumed annually. The copper smelting process required a plentiful supply of coal. Quantities varied from three tonnes for each tonne of ore during the ﬁrst half of the eighteenth century, reducing to one or two by the beginning of the nineteenth century. This

29Thanks to the relentless efforts of John Newton, once a slave ship captain, William Wilberforce and many likeminded abolitionists, the Slave Trade Act was passed by the British Parliament on 25 March 1807, making the slave trade illegal throughout the British Empire. John Newton’s hymn Amazing Grace is a reminder of the shame he felt having been party to slavery. 80 3 Copper and Coal Through the Ages

Fig. 3.23 The Welsh copper smelting process (Symons 2003)

was one reason for the concentration of smelting in South Wales. In the eighteenth century, as it is today, the most cost-effective means for the bulk transport of commodities like coal and copper ore was by sea, hence smelters were generally located adjacent to ports. Nevertheless, horses and mules were burdened with hauling ores between the mines and the ports (Symons 2003). Most of the Parys Mountain opencast was worked by the Parys Mine, the smaller ‘Hillside opencast’ to the east being worked by the Mona Mine. The opencast mines represented only a small proportion of the mining as later extraction occurred through shafts that reached depths of 300 m, some 130 m below sea level and therefore now flooded (Anglesey Mining plc 2014). During the years either side of 1789, Long (1833) estimated that from sixty to eighty thousand tonnes of ore averaging around 5% copper was mined from Parys Mountain. This would have yielded upward of three thousand tonnes of copper. In the 1770s, there were insufficient coins in circulation to satisfy Britain’s growing industrial economy. Mechanisation of cotton spinning and looms had heralded in Britain’s Industrial Revolution in the 1760s, and by the 1780s coke from coal was being employed in iron manufacture. Abraham Darby cast and commenced building the world’s first iron bridge at Coalbrookdale in 1779. Completed in 1781, and still standing today, the Iron Bridge across the River Seven is recognised as a symbol of the Industrial Revolution. The new workforce necessitated new coins and merchants and miners contracted with modern factories to strike their own money. In response to a national shortage of small value currency, the Parys Mine Company produced its own coinage. 3.2 Out of the Dark Ages 81

Fig. 3.24 Parys or Anglesey Penny

The Parys Mine Company was ﬁrst to issue tokens, for payment to their workers, and general circulation, ahead of John Wilkinson, the Shropshire Ironmaster. The ﬁrst issue of 1787 Parys Pennies (Fig. 3.24), also known as the Anglesey Penny, featured the bust of a hooded Druid on one side and ‘promise to pay the bearer one penny’ on the reverse side. In 1797, the Government authorised Matthew Boulton to strike copper pennies and twopences or tuppences at his Soho Mint in Birmingham. The face value of a coin was to be equal to the value of the copper contained in the coin. The 1797 penny was 36 mm in diameter and contained one ounce (28.3 g) of pure copper. From this information we can glean the quality of the copper used in copper sheathing Royal Navy Ships. In a letter to the Navy Board, responding to their questioning the quality of copper he had provided, Boulton responded:

I have always been of opinion that the durability of copper sheathing depended on the purity of that metal, and I assure you the whole quantity (of copper sheathing) I had the honour to furnish you with was doubly reﬁned for it was made from the scissel, or scrap copper remaining after cutting out the halfpence I made by order of our Government….But as all theories and reasonings are liable to prove fallacious when put in competition with actual experiment, I think it would be well if your Honourable Board were to order a regular series of experiments to be made on this subject (Dickinson 2010). Boulton’s suggestion did eventually lead to the addition of zinc for sheathing as is demonstrated from two analyses carried out on the wrecks of the Bounty and the Bowden. The mutinous crew of the Bounty30 arrived at Pitcairn Island in January 1790. The Bounty had been on an expedition to collect breadfruit in Tahiti for transport to British plantations in the West Indies. The Bowden was on voyage from San Francisco to Falmouth with a cargo of about 2500 tonnes of wheat, when she struck a reef west of Pitcairn Island on April 1893.

30The mutiny, led by Fletcher Christian against Captain William Bligh, is chronicled in books, ﬁlms, and songs. With Bligh navigating, he and his loyal crew travelled some 6000 km in the Bounty’s open launch from Tonga to Timor in 47 days. 82 3 Copper and Coal Through the Ages

Table 3.2 Analyses of copper alloy from the Bowden and Bounty Vessel Ag Bi Cu Fe Ni Pb Sb Sn Zn As Total Bowden 0.02 0.02 61.90 0.15 0.04 0.42 <0.01 0.02 37.30 0.04 99.9 Bounty 0.02 0.10 96.90 <0.01 0.02 0.12 0.02 <0.005 0.00 0.53 97.7

Viduka and Ness (2004) had samples of the sheathing from the wrecks of the Bounty and Bowden analysed and the results from the two best preserved samples are shown in Table 3.2. These analyses demonstrate the evolution in sheathing from pure copper to an alloy of approximately 60% copper and 40% zinc (Muntz metal) that by the 1850s was used by British and foreign shipping. Muntz metal is a form of brass named after George Fredrick Muntz, a metal-roller of Birmingham, England, who commercialised the alloy based on his 1832 patent. Because zinc melts at 420°C and vaporises at 950°C, well below the melting point of copper, in the early eighteenth century brass was made by mixing ground calamine ore with copper and heating the mixture.31 The heat was sufficient to reduce the ore to its metallic state but not to melt the copper. The vapour from the zinc permeated the copper to form brass that could then be melted to give a uniform alloy. By 1766, William Champion’s copper works, near Bristol was making brass from sphalerite (ZnS) also known as zinc blende or black-jack. The sphalerite was washed and ground fine before being calcined, and the product was then mixed with charcoal and made into spelter32 that was used as a substitute for calamine in the manufacture of brass (Percy 1875). The annual naval demand for copper rose from 300 to 1000 tonnes between 1780 and 1790 and, as Fig. 3.25 shows, the Parys Mountain mines production increased at an opportune time to meet the increased demand (Jones 1848). Although the Parys Mountain mines represented the largest single source of European copper production in the 1780s, Cornwall continually produced more copper. The growth in Cornish production was made possible by the advances in engine and pumping technology. In 1763, whilst employed as an instrument maker at Glasgow University, James Watt (1736–1819) developed a solution to the thermal inefficiency of the Newcomen atmospheric engine. His solution was to conserve the cylinder’s high temperature by enclosing it within a steam jacket, whilst the steam injected below the piston was carried off to be condensed in a separate vessel, appropriately named the condenser. Matthew Boulton (1728–1809) appreciated the potential of Watt’s modification to the Newcomen engine, and in 1775 Boulton negotiated an extension of Watt’s patent for a further 25 years. The first Watt-Boulton partnership engine was installed at the Ting Tang Mine in Gwennap

31 Hemimorphite, Zn4Si2O7(OH)2ÁH2O, is found in the upper parts of zinc orebodies, along with smithsonite, ZnCO3. Because both minerals were found in close association they were assumed to be the same mineral and both were called calamine. 32Spelter was an earlier name given to zinc. 3.2 Out of the Dark Ages 83

Cornwall Anglesey World

18,000 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2,000 Annual copper production - tonnes

Fig. 3.25 Annual copper production (Schmitz 1979; Symons 2003) in 1777 (Symons 2003).33 This engine and many more like it, pumped water from the copper mines, enabling Cornwall to become the dominant copper producing area in Britain. By the end of the eighteenth century, the easily available ores at Parys Mountain had been exhausted although mining recommenced in the nineteenth century. Some 130,000 tonnes of copper metal were produced from the Parys Mountain mines that in large part may be attributed to the efforts of Thomas Williams (Callcut 2005). His compatriot, Matthew Boulton, nicknamed Williams the ‘copper king’ and considered him to be ‘the despotic sovereign of the copper trade’ During the last half of the eighteenth century, Britain was the largest producer of copper in the world; however, mines were beginning to open all over the world often due to the expertise gained in mines like those on Parys Mountain. Some had high grade ores that could be extracted cheaply and imported to the Swansea smelters economically. Copper mining developments in Spain, North and South

33James Watt’s modified Newcomen atmospheric engine helped bring about the Industrial Revolution. However, it would be another 60 years before George Stevenson’s locomotive would haul the first coal on a public railroad from Darlington to the port of Stockton that significantly reduced the price of coal delivered to Stockton. Watt estimated that a typical brewery horse, attached to a mill that ground the mash for making beer, pulled with a force of 180 lb. Rounding for convenience, he set the unit of one horsepower at 33,000 foot-pounds of work every minute. For example, a horse exerting one horsepower would raise 330 lb of coal 100 feet in a minute. The International System of Units (SI) unit of power is the watt and one horsepower equals about 746 watts. A healthy human can sustain about 0.1 horsepower. Providing this human can sustain this effort for nine hours then almost enough energy will be produced to light a 30 W bulb for 24 h. A Boulton and Watt engine rated at 10 horsepower that was built in 1785 for the London Brewery of Samuel Whitbread to drive the malt crushing mill can be seen at the Powerhouse Museum in Sydney, Australia. 84 3 Copper and Coal Through the Ages

America, Australia, Africa and Asia would soon reduce Britain’s share of world copper output. The range of coals in the Swansea area were ideally suited to the needs of the reverberatory furnaces giving the Swansea copper smelters a natural advantage of the cheapest and best fuel in the world (Roberts 1956). Additionally, ships bringing high grade copper ores to Swansea smelters could take on coal as return cargo. Although this copper and coal trade was a boon to smelting it ushered in the demise of the local copper mining industry. According to Newell (1990), two factors explain the increase in copper imports to Swansea in the 1820s. Firstly, the breakup of the Spanish Empire led to the growth of British interest in metal mining in Chile and elsewhere. Secondly, new trade regulations provided strong incentives for overseas producers to export ore to Britain. British merchants, seeking new markets, quickly became established in South America. In addition to selling their wares, these merchants sought to generate exchange by importing goods from South America to Britain and minerals were an obvious possibility. News of the mineral wealth in Latin America transmitted to London drew the attention of speculators. Only a century after the South Sea Bubble,34 British investors, assuming British technology and management would improve the proﬁtability of mining rushed to invest in Latin America. Investors were lead to believe that Latin America abounded in gold, silver, copper and other metals. The value of shares in mining companies founded in 1824, doubled by early 1825; however, as the year wore on news of the true situation of the mining prospects ﬁltered back to Britain and investors lost faith in both the mining companies and the loans to Latin American governments. The bubble burst and many of the mining associations collapsed before even a hole was dug (Hoffman et al. 2009). However, some British companies survived by mining copper, which was found in abundance, rather than silver or gold. Links between South American mines and British mining and ancillary industries developed, and

34The South Sea Company (SSC) inveigled high profile investors into holding shares in the SSC and, with the sole right to all trade in the South Seas and since King George I of Great Britain was governor of the company, investors were keen to invest. In 1720 the share price increased from some £130 in January to more than £1000 in August. In September SSC shares collapsed, and by December 1720 were down to £124. Investors lost heavily, including Sir Isaac Newton who supposedly lost £20,000 and reputedly said ‘I can calculate the movement of heavenly bodies but not the madness of men’ (O’Farrell 2007). Concurrently in France, Scottish financier John Law, who had infiltrated the upper echelons of French public finance partly through his friendship with the Duke of Orléans, was also generating a bubble. The Mississippi Company gained a monopoly on the development of France’s Mississippi Territory in North America. In 1718, Jean Baptiste le Moyne, Sieur de Bienville, Governor of Louisiana set about building a capital, which, in honour of the Regent, he called New Orleans. The Mississippi Company, which had been renamed the Compagnie des Indes, also collapsed in 1720. The collapse of Banque Générale and the Compagnie des Indes, which coincided with the popping of Britain’s South Sea Bubble, plunged France and other European countries into a severe economic depression that contributed towards the French Revolution (Colombo 2012). 3.2 Out of the Dark Ages 85

Table 3.3 Varieties of Regulus Percentage of contained regulus (Raymond 1881) copper Coarse 20–40 Red 48 Blue 60 Sparkle 74 White 77 Pimple 79 skilled labour and equipment were exported to both British and South American owned mines. Smelting in the early nineteenth century may be considered as a three step process. The first step, the production of regulus35 was relatively simple. The next series of smelters produced copper bars and finally the bars were refined to almost pure copper. Valenzuela (1992) reports that when Charles Darwin visited Chile in 1836 he observed that few smelting furnaces were working in any part of Chile as it was found to be more profitable, on account of the extreme scarcity of firewood and because the Chilean method of reduction was so unskilful, to ship the ore to Swansea. Estimates of the amount of coal required to produce one tonne of copper in the mid-nineteenth century vary greatly. Percy (1875) reports that 18 tonnes of coal were required and estimates of up to 30 tonnes can be found in the literature. This discrepancy may in part be explained by the quality of the input to and the output from the smelter. Ores could contain as low as a few percent copper and as high as over 30% copper. The imprecise term regulus is often used to describe the input to the smelting process; however, there were several varieties of regulus as shown in Table 3.3. Fine metal includes the latter four varieties in the table. Additionally, the copper bars could contain varying percentages of copper. The last stage of the smelting process generally produced blister copper bars containing some 90–95% copper. This was the feed to the refining process, which produced refined copper that contained some 99.5% copper in the early nineteenth century. However, the refining stage could commence with a lower grade feed. Russell (2000) states that in 1750 the Swansea area was responsible for 50% of British copper production. At that time about three tonnes of coal were required per tonne of copper ore smelted. Therefore, smelting would be most economical near the coalfield rather than near ore mining areas that lacked coal. This remained the case up to the 1880s, when on average, smelting techniques required two tonnes of coal to smelt one tonne of copper ore, coal accounting for almost 45% of the total smelting cost.

35The more or less impure mass of metal that sinks to the bottom of a furnace, separating itself by gravity from the supernatant slag (Merriam-Webster 2015). 86 3 Copper and Coal Through the Ages

The accounts for Vivian & Sons’ Hafod works show that the cost of reducing one tonne of ore containing 10% copper to regulus containing 50% copper using the Welsh process was about 30% of the cost of fully smelting the ore to copper. Refining was the costliest of the many reduction processes amounting to over 28% of the total costs because a large amount of coal was required to remove relatively small quantities of impurities (Newell 1990). Although estimates of the quantity of coal required to produce a tonne of copper vary, multiple sources would indicate that around 1850 in Swansea approximately 20 tonnes of coal were required to produce a tonne of copper (Hughes 2000; Percy 1875). Assuming, based on Newell (1990), that 30% of the coal is required to produce a regulus containing 50% copper, then six tonnes of coal is required to produce two tonnes of regulus containing one tonne of copper. Assuming the ore contained 10% copper, then 10 tonnes of ore would be required. Two tonnes of regulus are produced and therefore for each six tonnes of coal, the mass to be transported from the mine site is reduced by two tonnes. Ceteris paribus, economics would dictate that the coal be transported to the mine to produce regulus containing 50% copper. If the facilities to smelt the regulus already existed at the coalfield and since the smelting and refining a regulus of 50% copper only reduces the mass input by half, then based on the similar logic smelting at the coalfield would be preferred especially since these processes required more skill. In summary, if coal was not available locally, it would appear logical to transport coal to the mine to convert the ore to regulus and employ the same vessels to transport the regulus to smelters at the coalfield. This proposition is supported by Percy (1875) who references a 1797 report that the poorest ores of the Parys Mine were partially smelted so as to produce a regulus containing 50% copper that together with the rich ores, was exported to Swansea. In Britain, the 1827 Customs Act that enabled copper ores to be imported free of duty for subsequent re-export gave copper merchants an incentive to import ore. Additionally, copper ore trade became exempt from the Navigation Laws enabling copper ore to be imported in vessels of any nationality. Consequently, copper ores and regulus imports into Britain increased from almost zero in 1826 to some 190,000 tonnes in 1886 of which some 100,000 tonnes was imported to Swansea (Newell 1990). Consolidation of the Swansea copper smelting industry meant that, by the early 1840s, the four largest firms accounted for over seventy percent of ore purchases. In 1842 a new firm, the English Copper Co, entered the industry provoking a price war that failed to force the new entrant out of the market. Instead this self-inflicted period of reduced prices led to formation of a secret cartel, the Copper Trade Association in 1844 that attempted to restore profitability to the industry. The cartel’s oligopolistic36 behaviour enabled its members to generate abnormal profits by increasing the margin between the price of metal and ore. This behaviour and knowledge of the existence of a cartel exercising price control may have

36A market dominated by a small number of producers or sellers is described as an oligopoly. 3.2 Out of the Dark Ages 87 hastened the development of smelting overseas. According to Valenzuela (1992), the formation of a copper monopsony37 in Britain created the impetus for developing the archaic Chilean copper smelting industry, and for stimulating smelting in the United States, France and Australia. Of the three major providers of copper ore to the United Kingdom during the 1840s, Chile (23%), Australia (8%) and Cuba (62%), only the latter failed to create a copper smelting industry. Messrs. Penny and Owen constructed one of the first smelters in South Australia in 1848 at Apoinga, 124 km north of Adelaide and 32 km south of the Burra mining area, where timber was plentiful. Ore was purchased from Burra and the first smelted copper was produced in 1849. Evidently a load of copper ingots, weighing nearly three tonnes, was paraded through Adelaide streets with a flag proudly proclaiming ‘Swansea Monopoly Defeated in South Australia’ (Drew 2011). The growth of regulus production in Chile was a harbinger that Swansea smelters were under threat from international competition. By 1886, twenty-five percent of the imports into Swansea were regulus. As coal became more readily available in Chile through the growth of the return trade in coal from Swansea and the development of a domestic coal mining industry, it was realised that savings in transport costs could be made by reducing the material to be exported by converting ores of 10–20% copper to regulus of 40–60% copper. Shortly after the emergence of the regulus trade, unrefined Chilean copper, i.e. bars containing 90–95% copper, also began to be exported to Britain. By the 1860s unrefined copper had overtaken regulus as Chile’s main export and only minimal amounts of ore were exported (Newell 1990). It had taken decades to create a Chilean smelting industry with a sizable capacity and even then it was unable fully to capture the country’s mine output, and substantial exports of regulus continued to be sent to Britain. In other countries the development of smelting was even slower. The slow development of smelting overseas and the way in which the production of regulus preceded that of unrefined copper bars reflects partly the advantage established smelters in Swansea held in the final stages of smelting required to produce near pure copper. Aside from difficulties in raising capital and providing adequate transport facilities, prospective smelters faced two fundamental technical problems which Swansea-based smelters had overcome. The first was the inability to obtain a sufficient variety of ores for blending in order to minimise fuel costs and produce good quality copper when treating principally sulfide ores. The second was the high cost of fuel in the latter stages of smelting. As well as its lower fuel requirement, regulus production was far less capital-intensive than the following stages of smelting and refining copper. A 50% regulus could be produced using the most primitive furnaces. In contrast, the full reduction of sulfide ores was technically difficult requiring sophisticated plant of a scale sufficiently large to produce copper competitively in the international market. Additionally, lack of capital was a further constraint in the introduction of smelting overseas (Newell 1990).

37A monopsony means there is one buyer and many sellers and the term may be applied to the cartel. 88 3 Copper and Coal Through the Ages

Roberts (1956) records that in 1840 the average grade of the ore arriving from the Cobre Mine in Cuba was more than 27% copper compared to the native mines of Cornwall, where an ore containing 8% copper was considered economical. Furthermore, Roberts reported that Australian ores reaching Britain in the period 1843–46 yielded 40% copper and those of Chile from 20 to 60%. British capital and citizens played a prominent role in developing copper mining throughout the world. For example, the engineer Richard Trevithick, who was a contemporary of James Watt and Robert Stephenson, developed copper mines in South America, and British loans enabled the development of Chilean copper mining (Roberts 1956). The ore grades reported by Roberts may have been exaggerated. The Australian ores would have mostly come from Burra in South Australia. According to Lawrence and Davies (2012), from the discovery of the Burra Mine in 1844 until a smelter was built in 1849, ore from the mine was hauled to the surface, sorted and bagged and hauled by bullock wagons to Adelaide, where it was shipped to Wales for smelting. ‘Only the richness of the Burra ores, containing on average 20% copper allowed such expensive handling.’ The British copper sector localised in a corner of South Wales, assumed an unprecedentedly global character in the 1830s. Founded on the abundant local coal ideal for use in the reverberatory furnaces and originally local copper ores, it sucked in cupreous materials from Australia, Africa, the Americas, the Caribbean and many parts of Europe. By doing so, according to Evans and Saunders (2015), it created the first truly global industry which by 1840 was supplying 40% of the world’s output of smelted copper. A third important change in this period was the development of copper production from pyrites. In 1838, a French sulfuric acid manufacturer was given a monopsony over the entire output from the Sicilian brimstone industry, Britain’s principal supplier of sulfur. The resulting hiatus in the sulfur trade caused a sharp rise in the commodity’s price and stimulated interest in the recovery of sulfur from cupreous pyrites, a potential substitute for brimstone as a source of sulfur. Because of its low copper content, pyrites was not normally purchased by copper smelters. However, sulfuric acid manufacturers were interested in the recovery of copper as a by-product from burnt pyrites, although it was not until the following decade that this became commercially viable. In 1865, a method patented by William Henderson was used in newly constructed works at Hebburn-on-Tyne, and following this advance copper extraction became a common feature of sulfuric acid manufacture. On the supply side, the increased use of this new method of copper production owed much to the development of open-cast pyrites mining in southern Spain and Portugal by the British Rio Tinto and Tharsis companies from the 1860s, which became the main suppliers of pyrites to the British chemical industry. With the expansion of the trade, new copper producers emerged in areas where the chemical industry was growing (Newell 1990). Paradoxically, the exercising of monopolistic power by the brimstone monopsony in Sicily and the copper producers’ cartel in Swansea stimulated competition and contributed to the eventual demise of the Swansea copper smelting industry. However, as the copper industry expanded to meet demand and smelting 3.2 Out of the Dark Ages 89 technology and transportation developed, it was inevitable that Swansea would lose its position as the world’s largest copper producer. In addition to copper industry expansions in both South and North America, the technologies developed in Britain enabled Scottish industrialist Hugh Matheson to develop the Rio Tinto Mine in Spain and the Scottish chemist and industrialist Sir Charles Tennant to develop the nearby sulfur and copper mines of Tharsis. The challenge of developing the mines in Spain is the topic of next section; however, before proceeding to Spain, the important contribution coal had made to the expansion of the copper industry will be briefly described. By the thirteenth century, coal mining was well established along the River Tyne that flows through Newcastle. Coal began to be exported from the River Tyne from the mid-thirteenth century onwards. Royalty were keen users as witnessed by a 1264 order from Henry III for a boat load of coal for Windsor Castle most probably for the burning of limestone to make lime (Weissenbacher 2009). Nef (1966) records that some 6700 tonnes of coal were being shipped from Newcastle in 1378 and by 1658 this had increased to 367,000 tonnes. Such was Newcastle’s importance that it began to rival London in its wealth, as evidenced by the following extract from a poem published in John Cleveland’s 1651 News from Newcastle attributed to Thomas Winnard::

We shall exhaust their chamber and devour. Their treasures of Guildhall, the Mint, the Tower. Our staiths their mortgaged streets will soon divide. Seventeenth century colliery railways, called ‘Newcastle Roads’, enabled the coal mines to be opened further away from the rivers Tyne and Wear. The New- castle Roads were built first of wood and later of iron. Horses pulled wagons of coal known as chaldrons along the rails that are credited as being the world’s first railways. The first recorded railway, ‘The Whickham Grand Lease Way’ of 1620, ran from Whickham to Dunston on Tyne. The Tanfield Railway in northwest Durham dates from 1725 and now claims to be the oldest existing railway in the world. The historic stone bridge, known as Causey Arch, which crosses the Causey Burn (stream) was part of the Tanfield Railway, and is considered the world’s oldest surviving railway bridge. The Tyne was shallow near the banks preventing easy access by the colliers38 that carried the coal for export. Consequently, shallow-draught keels transported coal to the colliers. In 1266, the standard load of a keel was set at 20 chaldrons (wagonloads) or approximately 17 tonnes, assuming a chaldron of coal weighed 850 kg. By 1700, there were some 400 keels operating on the Tyne. The weight of a Newcastle chaldron was gradually increased by coal traders due to the taxes on coal (which were charged per chaldron) until 1678 when its weight was fixed by law at some 2.7 tonnes.

38James Cook (later Captain Cook) worked on Whitby colliers shipping coal from the Tyne and Wear to London. 90 3 Copper and Coal Through the Ages

Fig. 3.26 The staithes at Wallsend (Hair and Ross 1844)

As new coal mining shafts or pits were sunk further away from the river and coal was transported to the riverbank on railways, it was a natural progression to build short piers called staithes or staiths, as depicted in Fig. 3.26, out into deeper parts of the river enabling coal to be dropped directly into the holds of the colliers without the need for keels. This was the beginning of the end for the keelmen. In 1794, the Tyneside keelmen went on strike against the use of staithes for loading coal; however, theirs were just one among many labour intensive jobs that were to be replaced by mechanisation as the Industrial Revolution got underway. Steam engines would soon replace many back breaking jobs including that of the horses that drove the gins39 that raised the coal from of the pits. George Stephenson was employed to drive the gin at Black Callerton Colliery, as depicted in Fig. 3.27, taken from The life of George Stephenson railway engineer by Smiles (1959). Hetton Colliery south of Newcastle was one of the ﬁrst to use locomotives in 1822. The Hetton Colliery Railway was the ﬁrst railway designed from the start to be operated without animal power. Designed by George Stephenson,40 it ran for some thirteen kilometres to a staithe on the River Wear. At that time the Hetton Colliery Railway was the largest in the world and was partly operated using sta- tionary engines and partly by locomotives. Stephenson’s locomotives and railways at Hetton Colliery served as models for the Stockton and Darlington Railway, the

39Gin gang refers to the shed that housed the gin (engine) and the horse that did the gang or going around in a circle to drive the gin. 40His Locomotion No. 1 was the ﬁrst steam locomotive to carry passengers on a public rail line, the Stockton and Darlington Railway. 3.2 Out of the Dark Ages 91

Fig. 3.27 Colliery Gin by Edward Whymper c. 1790s world’s first public railway. It is interesting to note that the gauge Stephenson chose 1 for his railways (4′ 8 /2″) is now the standard gauge for railways throughout the world. Similarly, the Bowes Railway was built to carry coal mainly from pits in North West Durham to the Tyne. The earliest section was also designed by George Stephenson and opened in 1826. It was some 24 km long when completed in 1855. Locomotives hauled the wagons at each end; however, in the middle steeply inclined section, the wagons were raised and lowered by rope haulage. At its peak, the railway handled over one million tonnes of coal annually and remained virtually intact until 1968. The Bowes Railway is credited to be the only surviving standard gauge rope-hauled railway in the world (Young 2012). English manufacturing led Europe in its widespread use of coal in the seventeenth century. Steam engines, too, were used widely before James Watt and Matthew Boulton installed their first large engine to pump water from Bloomfield Colliery west of Birmingham in Staffordshire in 1776 (Uglow 2002). Steam-powered beam engines had been employed in draining mines from around 1715 when there were seven or eight Newcomen engines operating from Cornwall to Newcastle. By 1733 there were sixty and by 1780 around 300; however, thanks to Watt and Boulton, there may well have been 1200 steam engines, at work in 1800 (Rule 1992). Most of the estimated 2.6 million tonnes of coal mined in 1700 was used for domestic heating. In 1750, the output of Great Britain was some four million tonnes and London was receiving over 700,000 tonnes (Nef 1966; Pollard 1980). By 1850, British coal output had reached sixty million tonnes and the iron industry required 92 3 Copper and Coal Through the Ages in the order of one million tonnes of coke.41 However, by 1880, the iron industry required some seven million tonnes of coke of which about five million were produced in Durham County, one million tonnes in the South Wales Coalfield, and the remainder in Yorkshire and Derbyshire (Beaver 1951). Although primacy in the use of coal in metal smelting is usually attributed to iron smelting, in Britain the non-ferrous metals lead, tin and copper lead the way. Lead, tin and copper smelted using coal added more than basic iron to the national income for most of the eighteenth century. Copper mining and smelting boomed in the 1780s, and in 1788 annual output was twice as valuable as that of pig iron. The technological breakthrough into coal smelting occurred before the end of the seventeenth century. Fuel economy was achieved through the use of the reverberatory furnace which, by separation of the fuel from the ore, overcame the problem of pollutants associated with the use of coal as opposed to smelting with almost pollutant free charcoal. Copper smelting works were established first at Bristol and then at Swansea and Liverpool. When ores from Anglesey came on to the market in the last third of the eighteenth century, some were smelted at Swansea although initially a large proportion went to the Lancashire Coalfield (Rule 1992). Manufacturing rather than domestic demand drove Welsh coal production. The share of Britain’s coal coming from South Wales increased from some two percent in 1750 to eleven percent in 1800. On its coalfields, the tin-plate industry was established and returning ships carried coal across the Bristol Channel for Cornwall’s steam engines returning laden with copper ore for smelting at Swansea. The readily available coal enabled Swansea to become the world leading copper smelting locality and these smelters would initially treat ore from the Iberian Peninsula, the subject of the next section.

3.3 The First Modern Mines

The Rio Tinto Mine is one of the oldest mines in the world. The Iberian Peninsula’s mineral wealth drew Phoenician merchants to Spain and it was one of the major sources of copper for the Roman Empire. The mines were rediscovered in 1556 but it was not until 1724, when the availability of metals from the New World declined, that there was sufficient interest in reopening the mines. Unfortunately for the Spanish Government, inefficiency dogged the mining and transport of copper that had to be carried to the port of Seville by mule and cart. Frustrated by the lack of progress, the Spanish government decided to sell the mines. The mines were bought at auction in 1873 by a British syndicate led by Hugh Matheson’s Matheson and Company, which ultimately formed a syndicate consisting of Deutsche Bank (56% ownership), Matheson (24%), and railway firm Clark, Punchard and Company (20%). The winning bid was 92,800,000 Pesetas plus 1,195,912 Pesetas for plant, buildings and mineral stocks. This was the equivalent of £3,850,000 (Nash 1904;

41Approximately three tonnes of coal were required to produce two tonnes of coke (Beaver 1951). 3.3 The First Modern Mines 93

Salkield and Cahalan 1987). The bid specified that Spain permanently relinquish any right to claim royalties on the mine’s production. The syndicate launched the Rio Tinto Company (RTC), registering it on 29 March 1873 (Chaplow 2015). The process of opening a mine in the twenty-first century generally requires at least a pre-feasibility study followed by a full feasibility study. The latter will likely include several tomes on environmental impacts, risk assessments using bewil- dering statistical methods and financial analyses including a net present value analysis with an assumption of such factors as exchange rates twenty years hence. Providing the appropriate rate of return is likely to be achieved, the probability of success is high enough, and the appropriate environmental approvals obtained, the project may receive the seal of approval and after many years of analysis construction commences. Such scrutiny may not have supported the perilous journey in 1873 by William Macfarlane, a trusted servant of Matheson and Company, from Paris over the Pyrenees by bullock cart to Madrid. His cargo was gold to the value of some four hundred and twenty thousand pound sterling. This was the down payment for a mine that Hugh Matheson wished to purchase on the strength of a recommendation from Heinrich Doetsch and Wilhelm Sundheim, two Germans with mining interests in the Huelva Province. Sovereign risk alone would have ruled out investing in Spain where the socialists, the anarchists and the supporters of the Carlist42 claimant to the throne were threatening the government of King Amadeo. Where others saw risk Math- eson saw opportunity. The insurrection and the Spanish government’s need for money gave him the opportunity to reduce the purchase price and gain perpetual freehold title of all the land owned by the Spanish Government within the Rio Tinto boundaries as well as the compulsory acquisition of the land to build the railway to Huelva. On 11 February 1873, King Amadeo declared the people of Spain to be ungovernable and abdicated. The first Spanish republic was proclaimed with Estanislao Figueras as its president and three days later the sale of the royal mines of Rio Tinto to the consortium led by Hugh Matheson was announced (Avery 1974). Matheson became Chairman of the newly formed Rio Tinto Company (RTC) and remained involved until his death 25 years later. By the end of the 1880s, Rio Tinto was the world’s leading copper producer. RTC provided its Spanish workers with social services unavailable elsewhere in Spain including elementary education at a time when only twenty percent of the population could read and write (Frago 1990). The staff of the company are also credited with introducing soccer to Spain and the formation of the first Spanish football club in Huelva. The Parque Minero Riotinto (Rio Tinto Mining Park) some 100 km NW of Seville is a must visit for those interested in history or technology. The Tartessians are generally credited with the earliest large-scale mining at Rio Tinto. Phoenician traders came next from the eighth to the sixth century BC followed by the Romans

42The Carlists supported Carlos VII, claimant to the throne of Spain. 94 3 Copper and Coal Through the Ages with their advanced hydraulic engineering. To combat flooding, a major risk in deep mining, they used norias (waterwheels) operated in pairs by slaves, raising water from 100 m below ground. The museum has an excellent reconstruction of working conditions in a Roman mine. Palmer (1928) observed that where the ore was at a shallow depth, the Roman miners sunk shafts or possibly a pair of shafts into the ore zone, extracting all the ore that could easily be reached from these shafts, and then sinking other shafts in the immediate vicinity. Shafts were about a metre diameter and there was evidence that firesetting was used to break the rock, and also the use of both large and small hammer gads or chisels. This is similar to the mining method previously described at Timna. When the depth of the ore was greater than the shaft system warranted, timbered galleries were mined that radiated in all directions from the bottom of a shaft, and the miners extracted the ore that could be reached from these. This system was apparently effective, as not much of any value appears to have been left in those places where the gallery system was complete (Palmer 1928). Pliny the Elder43 writing in the first century describes shafts opened by Han- nibal, and still being worked in the Spanish provinces, which furnished Hannibal with three hundred pounds weight of silver per day. ‘The mountain is already excavated for a distance of fifteen hundred paces; and throughout the whole of this distance there are water-bearers standing night and day, baling out the water in turns, regulated by the light of torches, and so forming quite a river’ (Bostock and Riley 1885). More impressive; however, were the Roman waterwheels one of which is shown in Fig. 3.28. A system of sixteen such wheels working in pairs was discovered at Rio Tinto and is depicted in Fig. 3.29. Each pair of wheels was about 3.5 m in diameter, so giving a total lift of some thirty metres. A slave rotated the wheel by treading on slats at the side of each wheel so that each wheel rotated in the opposite direction to its pair. Scott (2010) surmises that Rio Tinto was the mine cited by Pliny because the Phoenicians, Carthaginians and Romans had worked the jarosite deposits there, which could have contained as much as 1000 ppm of silver. Cupriferous pyrite ores, such as those at Rio Tinto, consist mainly of pyrite containing some copper, most commonly as chalcopyrite. Silver, lead and other elements may also be present. Archaeological excavation of the ancient slag heap at Rio Tinto revealed two distinct phases of mining spanning from the late Bronze Age to the Roman era. The Bronze Age slags were the product of mining the jarosite silver ore located in the lower parts of the gossan.44 The copper ores came from the secondary enrichment zone between the gossan, which varied between ten

43Pliny the Elder (Gaius Plinius Secundusr) died on August 25, AD 79, while attempting the rescue by ship of a friend and his family from the eruption of Mount Vesuvius that had just destroyed Pompeii. 44Sulﬁde ore deposits exposed at the surface may through the ages gain a thick cap or gossan of iron oxide minerals. The word “gossan” comes from the Cornish language and refers to the red colour of the oxidised iron minerals. 3.3 The First Modern Mines 95

Fig. 3.28 Roman water wheel, found in South Lode (Rio Tinto Mines, Spain), 1919, courtesy of the Historic Archive of Río Tinto Foundation and seventy metres in thickness, and the underlying massive sulfide deposit (Velasco et al. 2013; Rothenberg and Palomero 1986). Early investigators estimated that there were over fifteen million tonnes of lead/silver slag and one million tonnes of copper slag, a total of sixteen million tonnes of ancient slag at Rio Tinto. Modern surveys, however, indicted that there was only ever about three million tonnes of jarosite ore of which two million tonnes were mined in ancient times. Rothenberg re-estimated the total amount of slag to be about six million tonnes. Copper had, by Roman times, become a main product of the Rio Tinto workings so it is assumed that the one million tonnes of copper slag relates to Roman mining as the Romans had developed the technology to extract copper from copper sulfide ores. Jones (1980) surmises that the Moorish incursions into Spain in the 170s resulted in a breakdown in silver and copper production from Rio Tinto, probably the largest mine in antiquity that had a drastic effect on the production of silver coinage throughout the Roman Empire in the 180s. Some authors have suggested that the availability of timber may have been a restriction on the production of copper in Roman times. This thesis is not necessarily supported by Rehder (2000) in his excellent book Mastery and Uses of Fire in Antiquity that has a detailed calculation of the timber requirements for ancient copper mining on Cyprus. Based on a conservative estimate that 200 tonne of timber was required to produce one tonne of copper, Rehder calculated that an area of only about 0.15% of the total area of Cyprus was required to supply sufficient timber given the volume of slag produced over a 3000 year period. It therefore 96 3 Copper and Coal Through the Ages

Fig. 3.29 Rio Tinto Roman water wheels—after Palmer (1928) seems unlikely that in Cyprus pyrotechnology was a major factor in deforestation. Habitation, food and fibre were probably much more important factors in land clearance. Following the departure of the Romans around the year 400, there was little mining activity in the Rio Tinto region until 1725 when Liebert Wolters, a native of Stockholm, formed one of the first joint stock companies in Spain to develop the mines of Guadalcanal, Cazalla, Aracena, Galaroza and Rio Tinto. Vagarious cir- cumstance contributed to the leases being worked next by the industrious English woman, Lady Mary Herbert. In 1742, a Royal order placed Lady Herbert in pos- session of the Rio Tinto Mine shown in Fig. 3.30 as well as the mines of Guadalcanal, Galaroza and Aracena to the north. Management of Rio Tinto subsequently returned to Samuel Tiquet, a nephew of Wolters. In the twelve years prior to Tiquet’s death in 1758, some 160 tonnes of copper were produced; however, sales revenue did not cover the cost of production. Approximately twenty-five tonnes were produced annually in the two years prior to Tiquet’s death. Management next passed to Francisco Sanz, a tailor from Valencia. Sanz also failed to make the mine a financial success and in 1778 the mines were declared vacant and returned to the Crown although Sanz continued to direct affairs 3.3 The First Modern Mines 97

Rio Tinto Mine Huelva Province

Map courtesy of Manuel Olías and José Miguel Nieto

Fig. 3.30 Map of the Rio Tinto catchment—after Olías and Nieto (2015) at the mine until he was pensioned off in 1783 (Nash 1904). To Sanz’s credit, the mine was producing approximately 100 tonnes of copper annually when his lease expired in 1776. Despite the best efforts of Wolters, Lady Herbert, Tiquet and Sanz, the mines remained unproﬁtable. Government administrators fared little better. Between 1800 and 1809, Nash (1904) records that 834 tonnes of copper were produced at an average cost of about £149/tonne, somewhat more than its selling price. 98 3 Copper and Coal Through the Ages

By 1817 the mine was in ruins. From 1829 to 1849 about 5000 tonnes of copper were produced during the concession to the Marquis de Remisa. However, some 460 tonnes of fine copper were produced in 1848 and there were technological advances. The first reverberatory furnace was built and the open air roasting of lower grade ores was introduced. Large triangular shaped heaps of ore, called teleras because of their resemblance to a bread baked in Andalucía, were constructed of large pieces of ore and topped with timber which, when burnt, ignited the sulfurous ore (Nash 1904). Each telera was about eleven metres by six metres at the base and contained some 200 tonnes of ore. Some twelve tonnes of wood were then placed on top followed by one tonne of kindling used to ignite the telera. Roasting lasted until the fire died out some six months later. The remaining ore was placed in troughs each holding about twenty-five tonnes of ore that were continually filled and drained over nine days with mine water. The water was drawn off into cementation tanks containing pig iron, where the copper replaced the iron as previously described. The scales that formed on the pig iron contained about 25% copper. Each tonne of copper produced required approximately two and half tonnes of iron. The water drawn off from the cementation tanks containing the pig iron went into a third tank where the residue was allowed to settle to the bottom. The resulting mud-like deposit, containing about 10% copper, was finally treated in the reverberatory furnace that required about twenty tonnes of pinewood to treat four tonnes of crude copper. The refined metal produced was run off into basins and then ladled into moulds (Avery 1974). Unfortunately, the open air roasting caused considerable damage to the surrounding vegetation. This process depicted in Fig. 3.31 was still active when the mine was sold to Matheson in 1873 (Nash 1904). Production of copper increased under Government control. During the last ten years of Government administration up until 1872, annual production exceeded one thousand tonnes of fine copper in some years. Table 3.4 lists the production costs for 1869, when 233 teléras were roasting and 822 tonnes of copper were recovered by cementation (Salkield and Cahalan 1987). Assuming an exchange rate of some 25 pesetas to the pound sterling this equates to £83/tonne of copper. At that time the price realised for copper in Seville was 1400 Pesetas per tonne and the Treasury concluded that it would not be wise for the State to finance further development of Rio Tinto so the mine was put up for auction. Nevertheless, it would be unfair to criticise the mine operators for these disappointing results when, as Salkield and Cahalan (1987) report, the percentage of copper in the ore mined decreased from over 10% in 1849 to under 2% in 1869. At that time the input to the Swansea smelters would have contained at least 10% copper. Additionally, rich copper orebodies were being discovered throughout the world. In the U.S., increased production from the great copper mines of Lake Superior caused the copper price to fall in 1867 and 1868 and their owners came before Congress and asked for an increase of duties. The copper act of 1869 imposed a duty on imported ores of $66/tonne for each tonne of pure copper. The ingot copper duty was increased from $55/tonne to $110/tonne. Under the pre-1869 3.3 The First Modern Mines 99

Fig. 3.31 Open air roasting at Rio Tinto (Nash 1904)

Table 3.4 Copper production costs 1869 Pesetas Mining cost per tonne of ore 7.15 Roasting, cementation and reﬁning per tonne of ore 15.02 Administration and other per tonne of ore 2.17 Total cost of treating one tonne ore 24.34 Cost per kg Cu assuming 11.7 kg of copper extracted per tonne 2.08 Cost per tonne 2080 duty on copper ores, a large copper industry had grown up in Boston and Baltimore. The effect of the higher duty was to accelerate the closing of the smelting estab- lishments that had treated imported ores, and increase the proﬁts for the domestic producers of copper. The displacement of the imported copper by the Lake Superior product would have come in any case; for, as events proved, the sources of supply in the U.S. were rich enough not only to oust foreign competitors at home, but soon to invade their markets abroad. When it was impossible to dispose of the entire product within the U.S., large quantities were sent abroad and sold at whatever price that could be got, lower in any case than the domestic price. Thus, for a series of years the great natural resources of the country became a cause not of abundance and cheapness, but of curtailment of supply and dearness. The duty, far from stimulating the fall in price, checked it (Taussig 1905). 100 3 Copper and Coal Through the Ages

In Britain, copper price averaged £86/tonne from 1846–1850 but fell to £81/tonne in 1870–1874. Sharply rising imports caused the copper price to collapse during the next two decades and fell to £55/tonne in the years 1886–1890 (Burt and Kudo 2014). Assuming a conversion rate of five U.S. dollars to a Pound Sterling this equates to about $276/tonne. How was it that from 1873 onward, the successful bidder, Matheson’s RTC, was able to transform a mine already many thousand years old that had been a drain on the public purse into one of the most profitable mining enterprises in Europe and the largest single employer in Spain? Fortuitously, the introduction of dynamite, which Alfred Bernhard Nobel patented in 1867, significantly increased mining productivity. Nobel found that when nitroglycerin was absorbed into inert diatomaceous earth it became relatively stable and more convenient and safer to handle. Nobel demonstrated his explosive for the first time that year, at a quarry in Redhill, Surrey, England. However, it was the increased scale and efficiency of production along with the demand for pyrites for sulfur manufacture that made the mine profitable. Steam engines and the readily available coal to drive them enabled the increase in the volume of saleable produce whilst simultaneously reducing the cost per tonne of the material sold. Annual copper production increased from under 1000 tonnes in 1872 to some 20,000 tonnes in 1890. The railway was the essential new development that made it possible to rail the copper and pyrites produced to the newly built port at Huelva and to transport the necessary coal, machinery and other supplies to the mine site. Once the railway was built, the cost of delivering pig iron from England by ship and rail was about £5 per tonne, half the delivered cost from the Seville merchants in 1875 (Avery 1974). The new railway, pier and railway rolling stock were estimated to cost one million pounds and a further one million pounds would be required to convert the mine to opencast mining. The purchase had cost almost four million pounds, which meant the total cash requirements were about six million pounds. RTC offered 200,000 ten pound shares. The Board gambled that, providing two million pounds were raised from the sale of shares, the remaining four million could be amortised over ten years using the profits from sales. Directors of the Tharsis Sulfur and Copper Company of Glasgow, owners of the nearby Tharsis Mine, considered that the two million pounds would be insufficient to finance the mine. They may have been correct as additional shares had to be placed and by 1878 RTC’s share price had decreased to just above two pounds. A Belgium mining engineer, Julien Deby, without visiting the site, restated the ore reserves to be significantly larger than previously stated. Over estimating reserves is not uncommon in the history of mining, although estimates of reserves by a mining engineer sight unseen is unusual.45 Mark Car, the general manager, resigned in protest and Peter Denny and

45The overestimate of gold reserves in a gold ﬁnd at Busang, in Indonesian Borneo by a Canadian company in 1995 is considered one of the largest scams in the history of world mining. As was the case in Charles Dickens’s Bleak House, there was no money left for distribution to shareholders when the company involved was wound up. 3.3 The First Modern Mines 101

Martin Ridley-Smith resigned from the board (Avery 1974). However, none made any public protest and the share price rapidly rose above the ten pound issue price as a result of Deby’s report. Nevertheless, the claims made in the original prospectus proved true and in the years 1879–1913 the average annual dividend on ordinary shares was over 38%. By 1892 approximately 1.4 million tonnes of pyritic ore were being mined with an average copper content of 2.8%. Almost one million tonnes were treated on site. In addition to the pyrites, some twenty thousand tonnes of copper were produced by the 10,260 employees (Salkield and Cahalan 1987). Nevertheless, the percentage profit from the sale of pyrites exceeded the profit from the sale of copper. According to Harvey (1981), the increase in the margin between the products sold from RTC and the price of refined copper in the 1880s, prompted the company to build their own refining plant. RTC requested their metallurgical adviser, Tho- mas Angove, to investigate establishing a plant capable of producing 20,000 tonnes of refined copper per annum. Angove recommended the plant be built in the Swansea district of South Wales. Not only were the western areas of South Wales favoured by a ready supply of high quality coal, but the coastal towns were well placed for the transhipment of bulky materials. Moreover, since approximately three tonnes of coal were needed to smelt and refine one tonne of matte copper, it followed that transport costs would be lower if the works were located in the Swansea area rather than at the mine. Transport cost favoured the Welsh coalfields. When Cornwall emerged as a major producer of copper during the eighteenth century, it had proved cheaper to ship the ore to South Wales for smelting than to bring fuel to the mining areas. Likewise, when Chile and Cuba became substantial producers in the 1830s, the absence of domestic coal supplies favoured shipping these ores to Swansea for smelting. Consequently, by 1850, nearly 50% of the world’s copper was being shipped to South Wales for smelting. Luckily for RTC, the Cwm Avon copper smelting works situated a few miles inland from Port Talbot, which had ranked amongst the largest copper smelters in the world for many decades, was idle following the liquidation of the English Copper Company in 1876. In October 1883, the Cm Avon copper refining plant passed into the hands of RTC and production recommenced in 1885. The local collieries, to which the plant was linked by tramway, stood ready to provide a regular supply of fuel, while the Rhondda and Swansea Bay Railway Company whose branch line linked Cwm Avon with the Port Talbot dock, was eager to regain the smelter traffic. The Cwm Avon plant was soon recognised in metallurgical circles as one of the most efficient smelters in the world, and RTC copper was being sought throughout Europe. Thereafter, production rose steadily to a peak of 23,150 tonnes in 1892 with 67,470 tonnes of coal and coke being consumed that year, indicating 2.9 tonnes of coal and coke were required to refine one tonne of matte. The regeneration of Cwm Avon stimulated the British smelting industry, and between 1884 and 1889 production increased from 75,400 to 102,200 tonnes and Britain became a net exporter of refined copper once more (Harvey 1981). The proposition presented in the previous section that it was economical to carry out preliminary smelting at the mine site is supported by the actions of RTC. Records 102 3 Copper and Coal Through the Ages from Salkield and Cahalan (1987) show that in 1885 some 155,000 tonnes of ore were smelted at the mine site to produce about 18,000 tonnes of matte containing some 6000 tonnes of copper. Smelting required approximately 24,000 tonnes of coke or 72,000 tonnes of coal, since approximately three tonnes of coal were required to produce one tonne of coke. The reduction in volume by converting the ore to matte was some 137,000 tonnes, far greater than the volume of fuel imported to the mine. Therefore, producing matte at the mine containing approximately 35% copper that was exported for further smelting was economical. Additionally, ships carrying matte to the smelter in Wales could take on coal for the return voyage. The scale of mining at Rio Tinto was made possible by the railway and existing technology. Coal could be imported to drive the many steam engines required at the mine although there is no record of coal replacing wood in the teleras. The Com- pany continued to use the previously described method of open air roasting of the ore until 1907 despite a riot over air pollution in 1888 that resulted in serious loss of life when the military fired on the crowd (Salkield and Cahalan 1987). A similar lack of early adoption of new technology was evident in mining where the hand held hammer and tap rock drilling method previously explained was not fully replaced by compressed air drilling until 1912 even though compressed air rock drills were in production in the late nineteenth century. Evidently, a three-man crew using hammer and tap were expected to drill between five and eight metres in a six-hour shift. The drilling and blasting crews would have been very busy breaking enough material to match the steam shovels that were first introduced in 1904. Harvey (1981) suggests that RTC’s slow adoption of new technologies and lack of technological innovation before 1914 could be due to the background and per- sonal interests of Rio Tinto board members. Prior to 1914, no director possessed scientific or engineering training at an advanced level. Rather, most had experience in either finance or commerce. Consequently, technical matters were considered at the highest level only in respect of their financial aspects. Correspondingly, a large amount of responsibility for such matters was devolved to senior technical per- sonnel, which in practice meant the general manager and his departmental chiefs. However, an article in The Engineer of 1887 states that boring at Rio Tinto was done partly by hand, but chiefly by means of rock drills driven by compressed air (Anon. 1987) so perhaps the criticism is not entirely fair. The cementation process developed at the Parys Mine was employed on a grand scale at the Rio Tinto Mine. The roasting process was similar to that prior to 1873 but on a larger scale. Broken ore containing between 1 and 2.5% copper was heaped up to form teleras, each telera containing up to 1500 tonnes of broken ore. The heaps were set alight with as much as four tonnes of wood and then burnt for up to six months. The roasted ore whilst still hot was manually loaded into rail wagons, transported to and tipped directly into concrete tanks with a false bottom of square timber, and water introduced to the tanks to leach out the copper. The leach water then flowed into settling tanks, where the copper was precipitated on pig iron. To obtain the maximum extraction of the copper from the roasted ore, the partially leached ore was taken out and piled in heaps called terreros, and allowed to weather before re-leaching. This process was repeated up to six times. When Eissler 3.3 The First Modern Mines 103

(1902) recorded the process he estimated that there was about seven million tonnes of low grade ore in terreros at the Rio Tinto Mine. There were several classiﬁcations for the ores mined at Rio Tinto based on both analysis and physical condition. In 1888, the main products were:

• pyrites containing less than 0.5% Cu sold as poor ore; • ore containing from 1 to 2.5% Cu initially treated on teleras; • ore containing up to 3.75% Cu that was exported; • high copper content ore, usually around 4%, occasionally as rich as 12–14% that was smelted at Rio Tinto; • quartzitic copper ore containing 3–4% Cu that was used as ﬂux in copper smelting or for heap leaching; and • soft decomposed copper rich black ore, known as ‘negrillo’, that was sent to the smelter.

For export ore, sulfur content was critical and the ore was normally sold on the basis of 48% sulfur, with premiums and penalties for departures (Salkield and Cahalan 1987). Initially the ore was mined both by open cut and underground pillar and stall mining. The latter method of mining extracted less than thirty percent of the ore and was by 1890 replaced by cut and ﬁll mining whereby waste rock was introduced to the mine to support the old workings so that the pillars could be removed. The cementation process previously described continued at a much increased scale and cement copper represented about two thirds of the copper produced. The remainder was sold in the form of matte from the reverberatory furnace the company had installed (Salkield and Cahalan 1987). In 1890, Rio Tinto produced 11% of the world’s supply of copper. Nevertheless, as Table 3.5 reveals, sales of pyrite accounted for 52% of gross proﬁt that year and RTC continued to dominate the world pyrites industry until the outbreak of war in 1914. In 1887, Hyacinthe Secrétan, head of the Société Industrielle et Commerciale des Métaux, the largest manufacturer of brass and copper products in France proposed a scheme to corner the world copper market. He formed a syndicate backed by French banks including the Paris Rothschilds that undertook to provide advances and guarantees for the scheme.

Table 3.5 Rio Tinto mine production and sales 1890 (Harvey 1981) Revenue/tonne Tonnes produced Sales revenue £ Cu Pyrites Cu Pyrites Cu Pyrites 57.38 2.02 20,317 396,949 1,100,721 801,837 Production cost Gross proﬁt £ % of gross proﬁt £/tonne Cu Pyrites Cu Pyrites Pyrites Copper 37.37 0.91 406,543 440,613 52% 48% 104 3 Copper and Coal Through the Ages

Secrétan and his associates, observing the extraordinarily low copper price of £40 per tonne and perceiving increasing demand for copper, proceeded to buy all the available stock of copper to hold for a rise. The total copper in Europe and afloat in 1887 was 49,000 tonnes although world production was some 230,000 tonnes. Secrétan’s scheme required him to reach agreements with the American copper producers who supplied some 40% of world copper output as well as with RTC, the most lucrative copper property in the world. The American mine owners were to receive £61 per tonne for three years. Thus the syndicate gained control of almost eighty-five percent of the world’s copper product (Andrews 1889). As soon as the syndicate’s attempt to corner the market became known, prices leaped from under £40 a tonne to over £80 a tonne. World output increased to 254,000 tonnes in 1888, the bulk of the increase coming from the United States that produced over 107,000 tonnes, some twenty-seven percent more than in 1887. The output of Anaconda alone reached 28,000 tonnes, a quantity which only Rio Tinto exceeded. While the high prices encouraged production, they naturally lowered consumption. The United States consumed some 14,000 less tonnes of copper in 1888 than in 1887 and England and France, apparently, some 38,000 tonnes less. By March 1888, the increasing stock of copper was an ominous warning. As the price of copper rose, smaller companies outside the control of the syndicate responded by producing more copper. At the same time, many of the industrialists buying the metal allowed their stocks to decrease and held off buying as long as they could in the hope of the price falling once again. Secrétan was forced to increase his stockpile by mopping up the extra production and the syndicate was forced to buy not less than 163,000 tonnes out of a world production of 254,000 tonnes during 1888. He was compelled by the end of 1888 to borrow £6,880,000, half of which was in the form of unsecured loans (Avery 1974). An impending financial disaster was only averted by the Bank of France taking on the debt, and though hasty and much criticised the bank’s action was justified as it localised the crisis. The collateral that the Bank of France received consisted mainly of copper warrants made over to it by banks that supported the syndicate. Copper that was selling for up to £75 per tonne in March 1889 fell to a price of £36 per tonne by April. When the price collapsed, the large copper producers and dealers attempted to prevent the hoard gathered by the syndicate from being thrown upon the market. Although their efforts were partly successful, the price of copper did not rise substantially until the end of the century. Unsuccessful attempts to control commodity prices are not restricted to metals. In 1991, the collapse of the reserve price scheme for wool in Australia lead to one of that country’s largest financial failures.46

46When the Reserve Price Scheme collapsed in 1991, the Australian Wool Corporation Board was left with a stockpile of 4.8 million bales (about 830,000 tonnes) of wool and a debt of over $3 billion (Massy 2011). 3.3 The First Modern Mines 105

At RTC, Matheson who respected Secrétan’s abilities had agreed to participate in the scheme. In 1888, despite a major dislocation caused at the Rio Tinto Mine by the riot over air pollution previously mentioned, proﬁts on sales of copper jumped by 40%, and it was possible to increase the dividend that year from 10 to 17%. By September of that year, the market price of copper stood at almost £117 per tonne, and though the company was only receiving about £73 a tonne under its agreement with Secrétan, this was almost double what it was getting a year earlier in the free market. Avery (1974) and Andrews (1889) postulates that it was during this period that the French branch of the Rothschild family acquired its controlling interest in RTC. However, RTC continued to prosper and the annual production of copper from the mine did not fall below 20,000 tonnes until 1932. A two third share of the mine was sold to the newly formed Compañía Española de Minas Río Tinto in 1935, eighty-two years after Matheson’s purchase. The visitor to the Rio Tinto Mining Park can observe the now abandoned Rio Tinto Mine workings and the red wine coloured river that is a major tourist attraction rather than a liability. Almost all mining is ephemeral and sustainable mining of a single mine is an almost implausible concept. However, if the concept can be applied to any mine then Rio Tinto, with its millennial long mining history, is an ideal mine on which to test the concept of sustainability in relation to mining. Although the word sustainability is a new addition to the English language, the concept is ancient. In his sixteenth century text De Re Metallica, Georgio Agri- cola’s considers the environmental impacts of mining.

The strongest argument of the detractors is that the fields are devastated by mining operations. Also, they argue that the woods and groves are cut down, for there is need for an endless amount of wood for timbers, machines, and the smelting of metals. And when the woods and groves are felled, then are exterminated the beasts and birds, very many of which furnish an agreeable and pleasant food for man. Further, when the ores are washed, the water which has been used poisons the brooks and streams, and either destroys the fish or drives them away. Therefore, the inhabitants of these regions, on account of the dev- astation of their fields, woods, groves, brooks and rivers find great difficulty in procuring the necessaries of life, and by reason of the destruction of the timber they are forced to greater expense in erecting buildings. Thus it is said, it is clear to all that there is greater detriment from mining than the value of the metals which the mining produces (Agricola 1556). Agricola refutes the argument that there is greater detriment from mining than the value of the metals that the mining produces, explaining that no civilised society is possible without metals, and; ‘Moreover, as miners dig almost exclusively in mountains otherwise unproductive, and in valleys invested in gloom, they do either slight damage to the fields or none at all. Lastly, where woods and glades are cut down, they may be sown with grain after they have been cleared from the roots of shrubs and trees. These new fields soon produce rich crops so that they repair the losses which inhabitants suffer from increased cost of timber. Moreover, with the metals which are melted from the ore, birds without number, edible beasts and fish can be purchased elsewhere and brought to these mountainous regions’ (Agricola 1556). 106 3 Copper and Coal Through the Ages

Agricola did not acknowledge the often transitory and short-lived nature of individual mines. Fish and fowl may be bought from the proceeds of mining while mining continues to be profitable; however, when mining ceases, those remaining on surrounding lands may not survive without the resources of natural capital such as unpolluted rivers. The Rio Tinto (red wine river) of Spain is so named because of pollution from the Iberian Pyrite Belt, one of the largest sulfide mineral occurrences in the world. The resource has been mined from around 2500 BC until the present day. As a consequence of weathering, exposed pyrite reacts with oxygen and water and with the help of microbial processes acid water is produced (Gomez et al. 1999). The resulting brown stained water is often described as acid mine drainage (AMD). However, the staining of the Rio Tinto is thought to be largely natural as a result of exposure of sulfide ores to the elements (Sánchez et al. 2011). As a result of the low pH, varying from 1.5 to 2.5, and heavy metal contamination, the Rio Tinto does not support aquatic life apart from the bacteria that can survive in the low pH conditions. Nevertheless, a sample core drilled in the Rio Tinto estuary shows the lead concentration increased from a few parts per million to over a 1000 parts per million around 2500 BC. Lead levels higher in the core, representing more recent times, dropped but then increased again around 190 BC corresponding to the time the Romans commenced their mining activity on the Phoenician mining sites (Leblanc et al. 2000). Given that copper levels of 20 g/tonne were found in the estuary from deposition that occurred before mining commenced around 2500 BC, the high metal levels and low pH of the Rio Tinto may be considered as partly anthropogenic. Whether the Rio Tinto sustained aquatic life other than bacteria prior to mining is as yet unknown. Metals are an important component of our welfare; however, pollution from mining may have long-term impacts on the local environment. Anthropogenic flows of metals may remain mobilised for millennia in estuarine sediments (Leblanc et al. 2000). Whether mining, such as has occurred in the Rio Tinto valley, can be classified as Ecologically Sustainable Development will be discussed in the final chapter. Nevertheless, the Rio Tinto Mining Park with its excellent museum and the tourist train trip along the Rio Tinto are today much visited tourist attractions. Figure 3.32 includes a photograph taken from the tourist train, which tracks along the river on the old train line to Huelva that shows the red stained river. The Muelle de Minerales depicted in Fig. 3.33, built under the direction of Thomas Gibson, was designed so that copper ore could be dumped from wagons directly into sailing ships. Gibson (1878) gave a detailed account of the construction of the Huelva Pier of the Rio Tinto Railway from which the following description has been extracted. A straight section of rail on trestles extended some 180 m into the river. The remains of this section are now on reclaimed land. The pier then curves for 236 m and the remaining length of 160 m is straight. The 20 m wide pier has three floors. The lowest level was used for traffic other than ore. The second floor carried two rails having a continuous falling gradient from the pier head end to the shore end. The third floor carried two lines of rail on a rising grade from the shore to the summit before descending 240 m to the station at the head end of the pier. Some 3.3 The First Modern Mines 107

Fig. 3.32 Rio Tinto near the mine and near to the Gulf of Cadiz

Fig. 3.33 Replicas of Columbus’s ships and the Muelle de Minerales in Huelva

3470 tonnes of cast and wrought iron were used to construct the pier structure that supported the wooden decks and railway. The loaded wagons were pushed by an engine up to the summit of the pier and then descended by gravity to a station at the end of the pier. The wagons then were allowed to descend to the second floor and positioned over one of four spouts. Each wagon, containing seven tonnes of minerals, was discharged through an opening in the bottom of the wagon into the spout and thence onto the waiting vessel. Once unloaded, the wagons returned under gravity to the shore. It took on average one minute to discharge a wagon. The Muelle de Minerales was restored by the Huelva Town Hall in 2003. The matte from the reverberatory furnace at the mine, containing about 35% copper, was cast in moulds, and when cold, broken up and examined before being railed to Huelva and shipped from the Muelle de Minerales. As previously 108 3 Copper and Coal Through the Ages explained, after 1885, most was sent to the RTC smelting works at Cwm Avon, in South Wales for refining. The symbiotic relationship between coal and copper in the nineteenth century is most evident in the development of the Welsh coal mines. By the 1840s, improvements in shaft sinking technology including wire ropes and hoisting and ventilation equipment enabled miners to extract coal from the coal seams over 600 m deep, found in the central section of the basin-shaped South Wales coalfield. The need for copper to clad ships was previously described. As steel hulled ships replaced wooden hulled ships the need for copper sheathing decreased; however, as steam engines replaced sail the demand for Welsh coal increased. In the 1850s, the British Admiralty declared South Wales steam coal the best available for naval requirements stimulating demand. Miners and other industrial workers were required and in the 1880s thousands abandoned rural Wales for the industrialising coal valleys. The South Wales coalfield was developed into one of the world’s largest coal-producing regions. By the turn of the twentieth century, more than five hundred mines were operating, shipping coal by rail to the docks at Swansea, Newport and Cardiff. The population of Wales grew from about half a million in 1800, eighty percent of whom depended on agriculture, to over 2.5 million in 1914, when eighty percent were employed in non-agricultural pursuits. By 1921, King Coal employed 270,000 miners, and one in every four Welshmen depended on the pits for a livelihood. Coal production increased from 4.6 million tonnes in 1840 to 8.6 million tonnes in 1854. By 1900, South Wales was the leading coal exporting region in the United King- dom, and, from 1890 to 1910, South Wales shipped nearly a third of the world’s total coal exports. In 1913, Wales produced 57.7 million of the 292 million tonnes of coal brought to the surface in Britain. As well as contributing knowledge of smelting to the world, Wales also provided mining skills. It is estimated that by 1890, more than 100,000 Welsh-born immigrants resided in the United States. A majority of them were skilled labourers from the coal mines of Wales who had been recruited by American mining companies (Lewis 2008). As Fig. 3.34 shows, the United States surpassed Britain as the world leading coal-producing nation in 1900. The Industrial Revolution is often described as the period of transition to new manufacturing processes in the period from about 1760 to around 1830. However, if coal consumption is a measure of industrialisation, the pace of change accelerates after 1830. In 1800, water wheels were still the major source of energy in Britain and most factories were driven by water power until the 1840s (Allen 2011). Once the cost of coal-fired steam engines was reduced, they quickly became the major source of power and coal the major source of energy. Nevertheless, the famous Mississippi steamboats burnt mainly wood until the 1860s. Although the world’s first electric light was made by English scientist Humphry Davy in 1800 and the first electric power station was designed and built by Lord Armstrong at Cragside, England in 1868, it was the humble light bulb that ushered in the electric age. The first street to be lit by an incandescent light bulb was Mosley 3.3 The First Modern Mines 109

United States Great Britain World

800

700

600

500

400

300

Coal production Mt production Coal 200

100

0 1800 1810 1820 1830 1840 1850 1860 1870 1880 1890 1900

Fig. 3.34 World coal production 1800–1900—(Rutledge 2011; Day 1904; Walcott 1901)

Street in Newcastle upon Tyne with Joseph Swan’s incandescent lamps in 1879. Sir Joseph Wilson Swan and Thomas Alva Edison are credited with independently inventing the incandescent light bulb. The Edison Electric Illuminating Company of New York began marketing their light bulbs in 1880 (Palermo 2014). The first public steam-driven electricity generating power station was the Holborn Viaduct Power Station in London, constructed by Edison in 1882. A Babcock and Wilcox boiler provided steam to a 125 hp steam engine linked to the generator that supplied electricity to premises in the area including the City Temple Nonconformist Church (Harris 1982). The Age of Electricity had begun and electricity would be in demand both for households and industry. Copper would be needed for turbines, engines and in households, and coal would supply the energy for the power stations to produce the electricity. As the Rio Tinto approaches the waters of the Atlantic Ocean, the river’s colour changes from red to the light green, captured in Fig. 3.32. Near Huelva, on the Río Tinto estuary, you can see life-size replicas of Columbus’s three ships, Niña, Pinta and Santa María, moored in a recreation of the port as it would have been when the three ships set sail on 2 August 1492 for the East Indies, only to discover the Americas. And it was also to the Americas that the centre of copper production was going to move in the twentieth century. As the cost of producing copper from the mines of Tharsis and Rio in the Iberian Peninsula increased, across the world in North America rich copper deposits were being discovered. However, as the high grade copper deposits were depleted and with the discovery of the froth flotation process to concentrate low grade copper ores, miners searched for super-sized lower grade ore bodies. One of the most significant of these was found in Bingham Canyon in Utah. 110 3 Copper and Coal Through the Ages

3.4 The Age of Electricity

In the last year of the eighteenth century, Great Britain produced about 80% of the world’s output of copper. The Cornish miners supplied most of the ore, and the Swansea smelters extracted the metal. By 1913, the United States of America (U.S.) was producing some 56% of the world’s copper, mostly from ores mined in the U.S., and Great Britain refined barely 6% (Carpenter 1918). Several authors have captured the characters, good bad and ugly, who made the expansion in U.S. copper production possible. Others have described the colourful events including underground wars and claim jumping. Still others have explained the technological changes made in extracting and treating the various copper ores. The following section focusses on just some of the mines, their creators and the major technological advances. If at the end of this section you wish to read more on this interesting period in U.S. history, Ira Joralemon’s book Copper: The encompassing story of mankind’s first metal is highly recommended. Large deposits of native copper were to be found on the Canadian North Shore of Lake Superior, on Isle Royale and on the Keweenaw Peninsula of Michigan. Throughout the Keweenaw Peninsula and Isle Royale, pits and trenches dug into the rock by prehistoric miners were discovered by the European settlers. These showed evidence of copper having been removed from the rock, and in some cases copper was found partially worked out of the rock but still in place. Hammerstones were found in association with the pits along with finished copper knives, spear points and other copper artefacts as well as unworked cached copper. Although it is believed that humans migrated from north to south in the Americas, anachronistically, the extraction of copper in North America by the original inhabitants was less sophisticated than that of their southern neighbours. Nevertheless, the ancient copper pits found along the Copper Range of the Keweenaw Peninsula represent the earliest evidence of copper mining in the Americas. Figure 3.35 shows Whittlesey’s map of ancient mining pits on Point Keweenaw. He describes how mining pits up to three metres deep were filled up by natural processes and scarcely visible. Larger pits, up to five metres deep and thirty-five metres in diameter, were more obvious to the explorer. These were also filled with up to three metres of decayed vegetation; however, they contained no broken rock or gravel. Martin’s(1995) paper, about ancient copper mining in Michigan, builds on Griffin’s(1961) seminal work, Lake Superior copper and the Indians to provide an account of the ancient copper mining on the peninsula and dispels some of the myths surrounding the copper pits. Firstly, she debunks the suggestion, that the copper was exploited by some alien race. Martin explains that the scientific evidence shows that the people responsible for the prehistoric copper exploitation of the native copper were the indigenous American Indians. The National Park Service supported five years of research on Isle Royale from 1985 to 1990 that tested many 3.4 The Age of Electricity 111

Fig. 3.35 The ancient mine-pits of Point Keweenaw, Michigan (Whittlesey 1863) sites and also expanded the site location data base. This extensive research turned up no evidence of non-aboriginal exploitation of the copper deposits. Another misunderstanding surrounds the duration of the prehistoric mining era, which is now known to have been much longer than originally thought, and mining is now estimated to have commenced some seven thousand years ago. Copper working continued up until aboriginal contact with Europeans in the seventeenth century. According to Martin, the campsites of indigenous peoples of the Upper Great Lakes contain everything consistent with a long-lived, continuous, regional hunting/gathering/fishing adaptation, and contain nothing attributable to European cultures until the seventeenth century. During the last glacial period that lasted some one hundred thousand years and ended around 10,000 BC, glaciers scoured the rocks containing the native copper exposing veins of copper, as well as shearing off pieces of copper and transporting them sometimes for many kilometres. The first indigenous peoples who mined and utilised the copper were labelled ‘Old Copper Complex’. The evidence suggests that native copper was used to produce a wide variety of tools beginning around 4000 BC. Investigations by the University of Michigan radiocarbon dated mining pits on Isle Royale to around 2470 BC. Miners may have burnt timber on top of the rock adjacent to the copper vein until the rock became very hot and then quenched the heated rock with water causing the rock to spall. If so, then this method would have been invented concurrently with firesetting previously described in Europe. A small amount of copper would thus be exposed and stone hammers could be used to break off small pieces. 112 3 Copper and Coal Through the Ages

Fig. 3.36 Copper ﬁsh hooks, courtesy of the Milwaukee Public Museum

There is no evidence that the Old Copper Complex possessed the technology to cast copper, and artefacts exhibit obvious signs of layering, caused by hammering and folding to produce the finished product (Milwaukee Public Museum 2015). Copper becomes brittle when hammered repeatedly and must be heated to the recrystallisation temperatures, which is in excess of 300°C (Chandler 1996), to make it malleable before it can be worked further. Some artefacts appear to have been formed by hammering and annealing a nodule of copper. Others appear to have been formed by first hammering and annealing into sheets of even thickness and then rolling to consolidate the copper into a thicker artefact. Final annealing temperatures may have exceeded 600°C (Martin 2008). Once formed, tools requiring a sharp edge or point could have been honed on a stone surface. Fish was an important component of the diet of the Old Copper Complex people as evidenced by the abundance of fishhooks attributed to them some of which are shown in Fig. 3.36. Native copper orebodies the size of that on the Keweenaw Peninsula, which has produced some five million tonnes of copper (Bornhorst and Barron 2011), are rare. Local geological events placed this large native copper deposit in its relatively shallow and accessible position, thereby allowing prehistoric miners to recover it. The copper-bearing strata outcrop is in a narrow band some five kilometres wide along the Keweenaw Peninsula and Isle Royale. The beds also outcrop at the northern and eastern shores of Lake Superior and extend, deeply buried, further to the west into Minnesota (Martin 2008). The 1842 treaty of La Pointe between the United States and the Ojibwe (Chippewa) ceded the Keweenaw Peninsula to the United States. The Old Copper Complex mining pits guided European prospectors to the rich copper orebodies. Miners flocked to the region that became the nation’s leading producer of copper from 1845 until 1887. In 1847, S.O. Knapp found the old Indian workings and mining these rich copper masses proved very profitable. In the 1850s, copper was found in amygdaloidal or almond shaped cavities in lava beds and although these amygdaloid ore 3.4 The Age of Electricity 113 bodies were not as rich as the initial copper masses they were much more extensive. Finally, a pig supposedly led E.J. Hulbert to an old Indian cache of copper at the bottom of a shaft where he found high grade copper ore in conglomerate. A more plausible explanation can be found in an article in the Michigan Geological Survey, Bulletin No. 1 (Van Pelt 1964). The Calumet conglomerate was one of the great copper ore bodies. Between 1871 and 1921, the Calumet and Helca Mining Company paid over $152 million in dividends and Boston grew prosperous as a financial centre for mining (Joralemon 1973). By 1929, some 3,735,000 tonnes of copper had been produced in the Michigan copper district of which 1,644,000 tonnes had been mined from the Calumet conglomerate. Most of the other copper was produced from amygdaloid deposits of which the Kearsarge Amygdaloid Lode yielded just over one million tonnes of copper (Broderick 1931). In the 1880s, rich veins of copper were discovered in the Anaconda Mine located in the Butte district of southwest Montana. Butte became known as ‘the Richest Hill on Earth’. Copper kings, W.A. Clark, M. Daly and F.A. Heinze attempted to control the resource. Battles for ownership of the rich ore were fought in the mines and in the courts. In the end, the Anaconda Company came to dominate the mining of copper at Butte (Joralemon 1973). Since 1880, Butte has yielded about ten million tonnes of copper as well as significant amounts of silver and gold and other metals. In the far south of Arizona, almost on the Mexican border, the Copper Queen Mine at Bisbee was enjoying early success. Between 1880 and 1884, the Copper Queen produced over 81,000 tonnes of ore that yielded 19% copper. Over 15,000 tonnes of the ore was ‘black copper’ assaying 65% copper. The black copper ore probably included the copper mineral chalcocite that contains almost 80% copper by weight. Stockholders received $1,225,000 in dividends. Unfortu- nately, the ore was apparently running out. Dr. James Douglas suggested that Phelps Dodge partners purchased the adjacent Atlanta Claim in 1881. This led to the purchase of the neighbouring Copper Queen Mine. Wes Howell, who Dr. Douglas had employed as foreman, evidently disobeying orders, mined a tunnel that hit rich ore and saved the mine. Dr. Douglas gradually built up the Copper Queen into a model mining operation and became president and general manager of the Copper Queen Consolidated Mining Co. According to Joralemon (1973), the Copper Queen paid the Phelps Dodge Company one hundred million dollars in profits and produced over 900,000 tonnes of copper. For decades it was one of the richest copper mines in the world. His engineering colleagues recognised Dr. Douglas’ accomplishments by electing him president of the American Institute of Mining Engineers. Whether he had made the Copper Queen or the Copper Queen had made him, the combination of man and mine was one of the most fortunate in the history of copper (Joralemon 1973). Way up north in the Yukon region of Canada adjacent to the Alaska border, the Klondike Gold Rush of 1896 brought thousands of prospectors to Alaska; however, the Bonanza Lode of copper proved to be more enduring than the alluvial gold. In 1899, prospectors Jack Smith and Clarence Warner found rich copper ore on the Bonanza Mountain about five kilometres east of the Kennicott glacier. Below the outcrop were thousands of tonnes of fragments of ore, torn from the mountain by 114 3 Copper and Coal Through the Ages the ice just waiting to be shipped out. Isolation largely protected the lode until 1907 when prospects of a copper shortage prompted the Guggenheims to buy and develop the Bonanza Lode. They formed the Kennecott Copper Company and persuaded J.P. Morgan to finance the 314 km railway that took three years to build and cost thirteen million dollars (Joralemon 1973). Construction of the single track standard gauge railroad running 210 km from Cordova on the coast to Chitina, then branching 104 km to the Kennecott mill proved challenging. Apart from the cold conditions, the contractors had to bridge glacial rivers down which large boulders tumbled as the winter snows retreated. The ore treatment mill was located nearer the Kennicott glacier 610 m above sea level; however, the Bonanza Mine was at 1710 m and workers either walked the six kilometres up 1100 m to the mine or rode in the bucket on the aerial tramway. The increase in demand for copper during World War I (WWI) pushed up the price of copper encouraging production. The Kennecott mines and mill were operating around the clock and annual production of copper peaked in 1916 at 54,000 tonnes. Up until the mill closed in 1938, some 600,000 tonnes of copper were extracted from the Kennecott mines averaging about 13% copper and Ken- necott made a profit of one hundred million dollars (Hunt 1990). One of the minerals in the ores initially extracted was chalcocite (Cu2S) that as previous explained contains almost 80% copper. Although the total amount of copper produced from the Kennecott mines was less than from the mines of Montana and Arizona, according to Hunt, no other mine has surpassed it in the high metal content of its ore. These newly found copper mines stretching from Arizona in the south to Alaska in the north are just a selection of those developed shortly after the mines of Michigan. Their geological and geographical extents explain why the U.S. became the major producer of copper in the nineteenth century. In 1869, Michigan produced more than 95% of the country’s copper (Stevens 1908). The Calumet and Hecla Mine closed in 1969 and the White Pine Mine, the last major copper mine in Michigan, closed in 1995. However, there would be a very different outcome for the mining field discovered around the same time in Utah where technology and larger orebodies have enabled the mine known colloquially as ‘Old Reliable’ to continue producing copper economically from an ore containing less than 1% copper compared to the native copper ore of the Keweenaw Peninsula, which contained up to 99% copper. When Mormon pioneers entered the Salt Lake Valley in 1847, they soon recognised the economic value of the Oquirrh Mountains for grazing and farming. The mineral wealth of the mountains was not so obvious and in any event Brigham Young, their leader, did not countenance mining for fear it would bring an influx of miners to the area. Two Mormon brothers, Sanford and Thomas Bingham, who grazed cattle in Bingham Canyon, reported their find of ores to their leader, Brigham Young. Given the proximity of the find to Salt Lake City as shown in Fig. 3.37, Young advised against pursuing mining operations because the survival and establishment of agricultural settlements was of paramount importance at that time (Bailey 1988; Powell 1994). 3.4 The Age of Electricity 115

Fig. 3.37 Bingham and Salt Lake City (USGS 1885)

Brigham Young’s fears were realised in the 1860s when the discovery of gold brought the predicted influx of miners to Bingham Canyon. Silver and lead mining soon followed; however, the presence of copper was visible in most claims and the first ore of any kind shipped from Utah was a carload of copper ore from Bingham Canyon, hauled to Uintah by Union Pacific, and forwarded by the Walker brothers to Baltimore in 1868 (Bancroft 1889). However, the presence of copper was not always appreciated as evidenced at the Highland Boy Mine located at the head of Car Fork Canyon. Although the oxidised zone was rich in gold, the copper present in the ore caused higher than normal consumption of cyanide when extracting the gold. Fortunately, it was the dawn of the electric age and copper was the desired conductor of electricity. The price of copper rose to some six cents a kilogram and Samuel Newhouse, the mine’s owner, saw the opportunity rather than the problem that copper presented. Newhouse, whose British contacts included the then Prince of Wales (later to become King 116 3 Copper and Coal Through the Ages

Fig. 3.38 Bingham and Carr Fork Canyons (USGS 1900)

Edward VII), floated the Highland Boy Mine Gold Mining Company47 on the London Stock Market and Highland Boy made its first shipment of copper in 1896. When the Highland Boy Mine Gold Mining Company merged with Utah Consolidated in 1899, Newhouse used his proceeds to acquire prospective copper claims including one on top of the mountain separating Bingham Canyon and Carr Fork Canyon. The canyons are evident in Fig. 3.38 and although many features shown in the map, including Bingham Canyon town, were subsumed by the open cut mine, if you search on Google Earth you will find features surrounding the mine that are on the map from 1900. Newhouse’s prospecting team identified some 290 million tonnes of ore containing 2% copper (Bailey 1988). He formed the Boston Consolidated Copper and Gold Mining Company (Boston Consolidated) in 1898 and once again raised funds through his British colleagues. The company constructed a large concentrator

47It is not unusual for a gold mining company to produce copper. The Mount Morgan Gold Mining Company, later Mount Morgan Limited in Queensland, Australia, was a registered gold mine that produced some 250 tonnes of gold; however, it also produced approximately 360,000 tonnes of copper. 3.4 The Age of Electricity 117 designed in six units each capable of handling 500 tonnes per day of the porphyry ore. Newhouse entered into a contract with the American Smelting and Refining Company (ASARCO) to smelt seventy-five tonnes of concentrated ore daily. ASARCO completed the Garfield smelter to the north of the mine close to the Great Salt Lake in 1906. Two carloads of dynamite48 were detonated in 1906, considered at the time to be one of the largest blasts in Utah history, to make way for steam shovels (Bailey 1988) and by 1907 four shovels were eating away the ironstone capping overburden. Newhouse demonstrated that mining the lower grade porphyry copper was feasible on the top of the mountain known colloquially as ‘Copper Mountain’. The next major entrant would start at the base of the same mountain at the instigation of an ambitious young metallurgist named Daniel Cowan Jackling. At the age of nineteen, Jackling’s ambition was to become a school teacher in order to gain a much more lucrative job clerking and thus save enough money to buy land to farm. However, after observing engineers at work on a building site where his uncle had an excavation contract and ascertaining their remuneration, engineering seemed more appealing and he enrolled at the Missouri School of Mines, funding his course by working for a railroad survey party during vacations. He completed the four year course in three years graduating with a bachelor degree in science and metallurgy. After graduating he worked in the mining industry in several roles including as a miner and a metallurgical mill superintendent (Bailey 1988). His mining experience led him to identify a potential porphyry copper deposit at the base of Copper Mountain, and he persuaded a group of investors to launch the Utah Copper Company (UCC) in 1903. In 1905, the Guggenheims, after procuring a report by Seely W. Mudd and A. Chester Beatty (Bailey 1988) that identified forty million tonnes of copper ore, invested in UCC, partly to ensure a supply of ore for reputedly the world’s largest smelter that was under construction at Garfield. Concurrently, the Magma Mill was constructed some six kilometres to the east of the smelter. When completed in 1908, the mill had capacity to receive some 6000 tonnes of ore per day. The steep terrain of the lease was not conducive to steam shovels and Jackling, now General Manager of UCC, did not attempt to introduce shovels until 1906. As was the case on the top of the hill, the shovels were used to remove the ironstone capping, which at the UCC Mine was some sixty metres thick. By 1909, eight steam shovels were removing the overburden; however, the company was in trouble. UCC was constrained to a lease of just 200 acres (81 ha) at the bottom of the mountain, whereas their competitor, Boston Consolidated, was mining 370 acres (150 ha) at the top of Copper Mountain, a lease that was much more amenable to opencast mining. The alternative for Jackling and the Guggenheims was to commence more expensive underground mining or merge with Boston Consolidated.

48Two carloads may have been as much as three tonnes of explosive. Although this may have seemed huge at the time, it does not compare with the 450 tonnes of explosive placed by Canadian and Australian miners under Messines Ridge in WWI and detonated on 7 June 1917. 118 3 Copper and Coal Through the Ages

In 1910, UCC shareholders approved a share issue for a merger with Boston Consolidated. Samuel Untermeyer, who supposedly was acting as legal counsel for both companies, was reputed to have received over $776,000 (Bailey 1988) for arranging the merger, probably shared in part with at least one director of Boston Consolidated according to Col. Enos Wall who had resigned his directorship in UCC in 1908 (Colonel was an honorary title bestowed on Wall because of his dignified manner). Untermeyer on the other hand successfully asserted that Wall bribed the litigant in a suit in 1910 (New York Times 1910). Samuel Newhouse had prospered once again and he used his considerable proceeds from the merger to invest in real estate including conceiving and building the Flatiron Building in New York City. The way was now clear for the UCC to fully exploit the Copper Mountain. The Bingham and Garfield Railway was completed in 1911. A twice-daily passenger train was operated between Salt Lake City and Bingham and was reported to have carried 671,004 passengers in the year 1916. With its steep grade, spectacular views and the volume of traffic carried, it was considered the most marvellous little railroad in the world (Finay 1920). Built with a maximum grade of 2.5% leading up to the mine, forty empty wagons and sixty full wagons could be hauled between the expanding open cut and the Magma Mill (Bailey 1988). Additional feeder lines were built to the many mining benches on which the steam shovels removed the overburden and ore. By 1918, there were 23 benches connected by switchbacks, the uppermost being some 450 m above the bottommost level. The UCC also built a coal-fired power plant to generate electricity for the smelter, mill and mining operations. Samuel Newhouse and Daniel Jackling had demonstrated that large lower grade porphyry copper deposits could be mined profitably. This was made possible by the use of steam-powered shovels, railways and an abundant source of coal both for the steam engines and the electric power station. There was, however, one other major technological breakthrough; froth flotation had conveniently evolved to efficiently separate metal sulfides from low grade ores. Froth flotation,49 as the name implies, refers to the floating of metallic minerals to the surface in froth leaving the waste at the bottom. At least two discoveries were required to master the process. Firstly, the discovery that oily substances attract metallic minerals. Secondly, and somewhat counterintuitively, that air bubbles introduced to a tank containing fine ore and water along with the oil will produce oily bubbles that attract the heavier metallic minerals to their surface as they rise while the lighter gangue50 or waste material sinks to the bottom of the tank. Much of the history of the evolution of the flotation process can be gleaned from the litigation over patents rather than technical papers recording the progress of the invention. Hoover (1912) lamented in his seminal historic overview of flotation that: ‘Constructive work, may be entirely suspended while the courts adjudicate the rights of rival claimants; the completion of the structure is delayed; the time and means of the builders are wasted; and the needs of the industry suffer.’

49Flotsam as in ﬂotsam and jetsam has the same word origin, from Anglo-French ﬂoteson. 50The worthless rock in which valuable minerals are found is called gangue. 3.4 The Age of Electricity 119

Herodotus (c. 484–c. 425 BC), records that according to the Carthaginians there is an island called Cyraunis full of olive trees and vines and where girls collected gold dust by means of feathers daubed with pitch (Laurent 1846). It is unlikely, however, that this ancient reference to extracting metals using pitch influenced the many inventors involved in the evolution of current day froth flotation; nevertheless, some of the modern legends surrounding flotation appear just as fanciful. One legend is that a young schoolteacher in Colorado who was washing oil stained ore sacks in her brother’s assay office noticed that the pyrite floated on the water contaminated by the oil. The tale appears to be as mythical as the girls of Cyraunis. Evidently, Carrie Everson fabricated the myth and her husband, Dr. William K. Everson, was probably the originator of the method that was patented. Nevertheless, other female inventors of the period had been discredited or had their ideas stolen.51 Unfortunately, Dr. Everson died before fully developing the process and Mrs. Everson obtained a patent in 1885 based on ‘the chemical affinity of oils and fatty substances for mineral particles’. She and her husband had ascertained that acidification of the ore pulp was necessary for the sharp oil differentiation of mineral from gangue. Unfortunately for her, the method she patented in 1885 was a failure as a metallurgical process, although it probably did serve to assist some of the later investigations and was used freely in the attempt to disprove the originality of subsequent inventions (Rickard and Ralston 1917). In 1894, George Robson and Samuel Crowder who were working at a mine in Wales patented a process for the separation of the metallic matter from the gangue by the mixture of oils alone. The proportion of oil was large, as much as three times the weight of ore. It was a method of buoying the sulfide particles with oil. The process was tried on a working scale at the Glasdir Copper and Gold Mine in Wales and although unsuccessful led the way for the participation of the Elmore brothers. Francis Elmore was an engineer with an inventive mind. His father, William Elmore, bought the Glasdir Mine from Samuel Crowder in 1896. Recovery of metal out of the gold bearing chalcopyrite ore from the mine using a conventional concentrating plant of jigs and shaking tables was poor. William Elmore sent his two sons, Francis and Stanley, to investigate. Stanley Elmore claimed, that on the occasion of one of their visits to the mill, his brother Francis noticed copper adhering to the oil that had dropped from a shaft bearing, and thus obtained the idea of his invention.52 Finely divided wet copper sulfides would adhere to a greasy surface, whereas finely divided wet rock would not. In 1898, Francis Elmore obtained his first patent. In it, he describes the process as a ‘Process for separating the metallic from the rocky constituents of ore by mixing the pulverised ore first with water in considerable quantity, then adding to the mixture an oil of the kind described, which adheres to the metallic constituents

51Margaret Knight who invented the flat bottomed paper bag had her idea stolen by Charles Annan. Annan argued that a woman could never design such an innovative machine. Knight displayed actual evidence that the invention indeed belonged to her and consequently received her patent in 1871. 52Note that the Mrs. Everson 1885 patent had recognised this property. 120 3 Copper and Coal Through the Ages but not to the wet rocky constituents, allowing the water carrying the rocky material to subside while the oil carrying metallic constituents floats above’ (Elmore 1898). He used thick oil and introduced the idea of the freely flowing mixture of pulverized ore and water as against the mixing of oil with crushed ore in the presence of only a small proportion of water. By using more water, he inadvertently entrained more air, so essential to success. In the first plant, at the Glasdir Mine, the mixture of crushed ore and water was fed at the upper end of a slowly revolving drum, provided with annular helical ribs and transverse blades, so as to mix the pulp and oil without producing emulsifi- cation. The oil was introduced through a separate pipe. The mixture was discharged into a V-shaped vessel, where the water and sand subsided, while the oil buoyed the sulfide particles, which floated to the top. Rickard (1916) cites a paper by C.M. Rolker read before the Institution of Mining and Metallurgy in 1900 describing how an oil mixture of 0.89 specific gravity was used in equal parts by weight with the ore. An ore concentrate that was about 7% of the mass of ore input was achieved and 70% of the copper, 69% of the gold and 65% of the silver was recovered from the chalcopyritic ore assaying 1.12% copper, 0.49 oz. gold, and 0.8 oz. silver per tonne. Rickard surmises that, from the discussion of Rolker’s paper recorded in the Transactions of the Institution, nobody at that time recognised the part played by air in the process of flotation. On the same occasion Mr. Rolker said: ‘The viscosity of the oil is the all-important point.’ Subsequent discussions made no reference to the agency of air, which was entrained with the ore and water while they were being mixed in the revolving drum. As late as 1903, Stanley Elmore took out a patent for an improved apparatus that did not include the addition of air into his process of concentration by oil. It appears that in 1900 the agency of air was not understood by any of the exponents of flotation; however, that did not preclude the Elmores from claiming they were well aware of the role air played in future patent litigation. A number of plants were built using the Elmore bulk-oil process at mines including the Tywarnhaile in Cornwall, the Sygun in Wales and at the Boston Consolidated Mining Company that operated mines adjacent to UCC (Hoover 1912). Unfortu- nately, according to Rickard, not one of them was an unquestioned metallurgical success. Around the same period, the treatment of low grade complex zinc-lead ore at Broken Hill,53 and more particularly the beneficiation of dumps of similar material discarded in the course of milling operations, stimulated efforts to introduce some form of flotation to the conventional concentration process. Hence the next chapter in the evolution of flotation concerns itself mainly with the work of a group of Australian metallurgists. An effort was made to employ flotation for the purpose of treating the huge accumulations of tailings at Broken Hill, which contained some 18% zinc, 7.5% lead and 10 oz. silver per tonne. In 1902, Charles Potter, an Australian engineer,

53Before mining commenced soon after the discovery of the orebody in 1883, Broken Hill in Australia was one of the world’s largest lead-zinc-silver orebodies, containing some 300 million tonnes of high grade ore (Dresher 2001). 3.4 The Age of Electricity 121 obtained a British patent for the flotation of sulfides in a hot acid solution. He used a stirrer and claimed that the solution would react on the soluble sulfides present to form bubbles of sulfuretted hydrogen on the ore particles and thereby raise them to the surface. Hoover (1912) notes that it was clear that Potter had recognised a surface tension process. Later that year, Guillaume Delprat, manager of the Broken Hill Proprietary Mine, applied for a similar patent, except that he used salt cake (anhydrous sodium sulfate Na2SO4) instead of sulfuric acid. In his first American patent, No. 735,071, filed in 1903, Delprat states that his process ‘depends upon the ore particles being attacked by the acid to form a gas. Each ore particle so attacked will have a bubble or bubbles of gas adhering to it, by means of which it will be floated and can be skimmed or floated off the solution.’ In another place he says specifically that ‘The sulfides in the ore are rapidly attacked by the acid and gas bubbles formed on them that quickly carry them to the surface.’ In this and in Potter’s patent we have the earliest recognition of bubble levitation (Rickard and Ralston 1917). The Potter and Delprat interests soon became involved in litigation ending in a compromise, and so the method was generally described as the Potter-Delprat process (Hoover 1912). The processes of Potter and Delprat were labelled under ‘acid flotation’ and ‘surface tension’ methods. In their original form they did not include the use of oil. No substance other than acid was used in the processes; however, there may have been organic substances in the ore which, upon the addition of acid, yielded gummy organic compounds that selectively adhered to the ore (Hoover 1912). According to Rickard (1916), Potter and Delprat were mistaken in the reactions that were supposed to follow the introduction of the acid, whether it was the sulfuric acid, the nitric acid, or the sodium bisulfate that they variously used. At that time, it was believed that the sulfuric acid reacted with the sulfides to form hydrogen sulfide without attacking the gangue. Then it was suggested that carbon dioxide was generated by decomposition of a carbonate encrustation on the sulfides that formed due to weathering of the ore. Such explanations overlooked the simple fact that the Broken Hill ore contains a considerable proportion of carbonates, notably calcite, siderite and rhodocrosite. From any of these a warm acid solution would release the carbon dioxide gas that promptly attached itself to the surface of the metallic particles. The evolution of flotation next shifts from Australia to Italy. At the time when Potter and Delprat introduced their methods at Broken Hill, another investigator was about to contribute the missing ingredient in the development of flotation. The Elmore bulk-oil method had been seen by Alcide Froment at the Traversella Mine, in Italy, where he was engaged as an engineer in 1901, when he invented what he himself termed a modification of what was then known as the oil process of concentration. His modification, patented in June 1902, was to introduce a gas into the freely flowing oiled pulp. He argued, in his patent, that when a gas of any kind is introduced into the pulp, the bubbles of the gas become coated with an envelope of sulfide and thus rise readily to the surface of the liquid where they form a kind of metallic froth. The phrase ‘gas of any kind’ is important, for, although he generated his bubbles of gas by the reaction between sulfuric acid and the carbonates of the 122 3 Copper and Coal Through the Ages gangue or between the acid and the limestone that he added to the pulp, he hit upon the fundamental missing ingredient for froth flotation. If he had specified air as the particular gas to be used, he would have been acknowledged as the pioneer of present-day flotation. An abstract of Froment’s British patent was published in the Journal of the Society of Chemical Industry and was seen by Mr. Sulman, a director of Minerals Separation Ltd., whereupon negotiations for the purchase of Froment’s patent were opened by Mr. Ballot, also a Director of Minerals Separation. He went to Milan to meet Froment and purchased his patent rights in 1903. Froment sent some draw- ings, with descriptions and instructions explaining his mode of operation. His patent had mentioned a thin layer of ordinary oil; however, in his instructions to Minerals Separation, he specified as little as 1% of oil for ore containing up to 5% of metals and up to 3% for ore containing 50% of metallic lead (Rickard and Ralston 1917). Early in 1904 a small plant designed by him was forwarded to London, but the apparatus was discarded, and destroyed subsequently, by Minerals Separation. Froment was in poor health at that time, and he died soon afterward. His patents had been taken out in Great Britain and Italy, but not in the United States, and when Mr. Ballot acquired them it was too late to obtain American rights, more than a year having elapsed since the grant of the 1902 British patent. So the Froment patent was set aside as of no immediate value. Much of the early experimental work of Minerals Separation was done in the laboratory of Sulman and Picard in London. Sulman and Picard were experimenting with another method and trying to develop a workable process at the time when their attention was called to Froment’s patent. They made experiments in accord with the specifications and the later instructions sent by Froment; however, this was not the story told in the courts of law (Rickard and Ralston 1917). They testified that they had discarded Froment’s patent and his instructions, having found them worthless, and were trying various modifications of another method when suddenly they happened upon the particular combination essential to the froth-agitation process. Messrs. Ballot, Sulman and Picard stated that protracted experiments were being conducted in their London laboratory under the immediate charge of Arthur Higgins, who had been instructed to try all sorts of variations in temperature, acidulation, oiling and mixing. Nothing noteworthy happened until the proportion of oil was reduced, whereupon the granules began to rise instead of sinking and the quantity of floating material increased rapidly when the oil was reduced below 0.62% of oleic acid, according to the testimony of Mr. Ballot. Thus, according to W.H. Ballantyne, Mr. Ballot’s patent lawyer, the froth-agitation process was discovered. British patent No. 7803 was granted along with the American duplicate, No. 835,120 in 1905. The two important elements of froth flotation, oil and bubbles had now been recognised and the scene was now set for full scale introduction into ore milling (Rickard and Ralston 1917). Nevertheless, Hoover (1912) warned that ores in which the valuable minerals were wholly or partly bornite or chalcocite, as those of Bingham Canyon, may give trouble to the flotation processes. However, by 1913, Minerals Separation staff had demonstrated a 90% recovery of copper from ore 3.4 The Age of Electricity 123 containing 2% copper within both bornite and chalcocite (Rickard and Ralston 1917). Joralemon (1973) credits Dr. Rickets with the construction of the first mill in the U.S. to use the flotation process at the Inspiration Consolidated Copper Com- pany mine in Arizona that commenced in 1915 and resulted in a 50% improvement in recovery. Froth flotation soon became the established method of milling porphyry copper ores and Kennecott Copper Mines, the new owner of the Utah Copper Mountain, was able to take full advantage of the new technology as were other large copper producers such as Anaconda. In a very short period the U.S. would become the major world copper producer. Kennecott Copper Mines was formed in 1910 after a merger of the Utah Copper and Kennecott mining companies. The Kennecott Copper Corporation fully acquired the company in 1936. In the 1970s, the township of Bingham was subsumed by the Bingham Canyon Mine. British Petroleum acquired Kennecott in 1987; however, ownership of the Kennecott Utah Copper unit of BP Minerals America was sold to RTZ Corporation in 1989. The name of BP Minerals America, Kennecott Utah Copper was changed to Kennecott Corporation (Salt Lake Tribune, July 6, 1989). RTZ Corporation changed its name to Rio Tinto in 1997. The Copper Mountain is now the Bingham Canyon Mine open cut, over four kilometres long, almost two and a half kilometres wide and more than a kilometre deep, and the mountain has delivered over 20 million tonnes of copper. Copper ore is hauled to the in-pit crusher (Fig. 3.39) and conveyed eight kilometres to the Copperton Concentrator.54 Here the ore is finely ground and a very sophisticated version of the froth flotation process previously described produces a concentrate containing in the order of 28% copper (pers. comm.). The concentrate then flows to an Outokumpu flash smelter through a 27 km pipeline. Our ability to extract copper economically from deeper and lower grade ore can be attributed to many technological advances, including the water wheel, dynamite, steam-driven shovels and froth flotation. Flash smelting was the most significant technological advance following flotation. Until the first flash smelter was built in 1949, copper concentrates were generally smelted in reverberatory furnaces followed by Peirce-Smith converters. These technologies were refinements on the techniques developed in Wales in the eighteenth century and further refined at Rio Tinto in the nineteenth century. The reverberatory furnace required an external source of heat, most commonly coal. The flash smelter significantly reduced the need for an external source of heat, by using the exothermic oxidation reaction of the copper sulfide concentrate to melt the feed material. In 1908, a large deposit of copper ore was discovered at Outokumpu, in Northern Karelia, Finland. The Outokumpu Company (OTK) was established to develop the now exhausted mine. The flash smelting process, which was developed

54Assuming the open-pit walls slope at 45°, then the top bench length will be 2π times, or approximately six times the depth of the pit. At some stage it becomes more economical to drive a tunnel into the pit and install a conveyor and in-pit crusher to convey the ore from the mine. 124 3 Copper and Coal Through the Ages

Fig. 3.39 Tipping some 300 tonnes of ore into the in-pit crusher. Courtesy Kennecott Utah Copper by OTK in response to the electric energy shortage in Finland, is considered one of the most important innovations in copper metallurgy in the twentieth century (Kojo et al. 2000). Semi-autogenous sulfide smelting (i.e. with a significantly reduced need for an external heat source) was thermodynamically possible; however, its practical application had proven elusive. OTK finally demonstrated that it could be commercially achieved at its first flash smelter at Harjavalta in 1949. The process was progressively refined and marketed globally and in 1956 an Outokumpu flash smelter was built at Ashio in Japan. By 1972, plants had also been built in Romania, India, Turkey and two in Australia, one at Mount Morgan and another at Tennant Creek (Mäkinen 2006). In the reaction shaft reactions, like the one described below, take place after spontaneous ignition at 400–600°C:

2CuFeS2 (chalcopyrite) + 5/2O2 + SiO2 (ﬂux) ⇒ Cu2SÁFeS (matte) + 2 FeOÁSiO2 (slag) + 2SO2 (gas) + heat (Firdu 2009)

Changing from reverberatory to flash furnaces cut total smelting and refining energy requirements by about a third (Gibbons 1988). The flash smelting process generates significant heat from the oxidation of copper and iron sulfides in the concentrate feed so that the amount of fuel required in the reaction shaft is considerably reduced. Furthermore, by using oxygen-enriched air, the smelting process in the reaction shaft can become fully autogenous for certain grades and types of 3.4 The Age of Electricity 125

Fig. 3.40 Flash smelting furnace—after Mäkinen (2006), King (2007), Firdu (2009) sulfide concentrates. An added benefit of using oxygen enrichment is that the quality of the sulfur dioxide gas is sufficient to warrant its capture and conversion into sulfuric acid. The molten products from the reaction shaft, as depicted in Fig. 3.40, are a high grade copper iron sulfide matte that requires further processing to extract the copper metal and an iron silicate slag that is discarded. The oil shocks of the 1970s and the subsequent oil price rises triggered widespread interest in Outokumpu’s flash smelting process. In the U.S., however, where energy was abundant and prices relatively low, it was the 1970 Federal Clean Air Act Amendments, requiring recovery of 90% of the sulfur input to copper smelters that prompted the introduction of flash smelting. The existing reverberatory furnace smelters were able to capture only 50–60% of sulfur if equipped with an acid plant but most smelters were not (Gundiler 2000). The Hidalgo Smelter in New Mexico was the first flash furnace smelter in the U.S.. Built for the Phelps Dodge Corpo- ration in New Mexico in 1976, the smelter incorporated two sulfuric acid plants capable of capturing 96% of sulfur input and consequently was the first U.S. smelter able to meet the new air quality standards. At the Bingham Canyon Mine, Kennecott commissioned an Outokumpu flash smelter and flash converter system in 1995. The smelter routinely captures in excess of 99% of the sulfur introduced by employing two flash furnaces (Newman et al. 1999). One is for the production of matte while the second converts this matte into blister copper. The second furnace was built to avoid the fugitive emissions that are inevitable with Pierce-Smith converters. Copper concentrates, silica flux and recycle material are mixed and dried before entering the flash smelting furnace reaction shaft, where oxygen-enriched air containing 80–85% oxygen is fed to the burner located in the top of the shaft. Slag is tapped at a target temperature of 1315°C (Newman et al. 1999) while the matte 126 3 Copper and Coal Through the Ages containing about 69% copper, mainly as CuS, is conveyed down heated launders to a water-granulation facility. The ground matte is then conveyed to the flash converting furnace where burned lime and recycled dust are fed to a burner on top of the furnace reaction shaft together with oxygen-enriched air containing 75–85% oxygen. The copper matte is converted to blister copper and the residual iron combines with lime to form a calcium ferrite slag. As in the case of the flash smelting furnace, the gases produced that contain 35–40% sulfur dioxide are cooled and cleaned before being sent to the sulfuric acid plant. Blister copper from the flash converting furnace is collected in anode furnaces where the remaining sulfur is removed by injecting air into the molten bath. The finished refined copper is cast into plates or anodes of some 320 kg containing 99.6% copper that are railed to the refinery (Rio Tinto 2006). Steam, produced in the furnace waste-heat boilers, is used to drive the acid plant blowers that move the gases through the plant.55 Excess heat from the reactions in the acid plant is used to generate low pressure steam that is blended with the residual steam from the blowers and directed to a turbo generator at the powerhouse. Approximately 24 MW56 of electrical power is available in this manner representing some 60% of the smelter’s electrical requirements (Rio Tinto Ken- necott 2015). At the refinery, 47 anodes are interleaved with 46 stainless steel blanks and placed in an automatically guided vehicle that transports the cell load to the tank house where they are lowered into an acidic solution. An electric current is sent between the anode and the cathode, causing the copper ions to migrate to the stainless steel cathode while impurities, including gold and silver, fall into the bottom of the cell and are recovered in the Precious Metals plant. One anode will produce two plates of 99.99% pure copper each weighing some 140 kg that are Bingham Canyon Mine’s finished product (Bleiwas 2011). The remaining spent anodes are returned to the smelter. In addition to the energy supplied by the smelter, energy to feed this highly efficient and integrated copper production process also comes from a 175 MW power plant that is fuelled by coal delivered via rail. As Fig. 3.41 shows, the concentration process consumes the majority of the energy required to produce copper. Figure 3.42 shows that 85% of this energy comes from electricity. In 2014, some forty one million tonnes of ore was mined with an average grade of 0.58% copper (Rio Tinto 2015) compared to the 2% grade that Newhouse and Jackling believed could profitably be mined at the end of the nineteenth century. Although their detractors considered their respective ventures at the top and the

55The sulfuric acid produced has many applications including the desiccation of potato vines, allowing an easier harvest as well as hardening the potato skin. The solvent extraction–electro winning (SW-EX) plants described later also beneﬁt from the ready supply of sulfuric acid. 56A megawatt (MW) is one million watts and a kilowatt (kW) is one thousand watts. An average size room heater rated at 1 kW consumes one thousand Watt-hours in an hour. 3.4 The Age of Electricity 127

Fig. 3.41 Primary energy usage (Rio Tinto 2006)

Fig. 3.42 Concentrator energy inputs (Rio Tinto 2006)

bottom of the Copper Mountain risky, the mine known locally as ‘Old Reliable’ remains among the world’s largest producing copper mines. Copper consumption in China increased from some two million tonnes in 2000 to an estimated 8.7 million tonnes in 2014 (Yam 2005), although some estimates are as high as 10 million tonnes (Campbell and Lavier 2014). In comparison the U.S. consumed some 1.8 million tonnes in 2014. Although the Bingham Canyon open pit was approaching its economic limits,57 the increase in demand and resulting increase in the price of copper made extending the life of the open pit to 2029 economic. Additionally, an underground mine is being planned that could extend the mine’s life to 2050 and beyond. Many of the world’s large open pits are approaching their economic limits. Some, including Grasberg (Indonesia), Oyu Tolgoi (Mongolia), and Chuquicamata, the Chilean pit similar to Bingham in scale and age, are planning to go underground. Block caving is one of the mining methods being considered (Heffernan 2013; Priyadarshi and Caldwell 2012). This method requires significant underground mining development prior to extracting the first ore but little thereafter. All the ore passes and one slice of the block of ore is mined prior to the initial blast. Finally, the ore above the initial slice is blasted, and the shattered ore will flow down the ore passes for removal. After the initial blast the production of ore is self-sustaining. The ore body simply collapses under its own weight, and as one load of broken ore is trucked out, more flows in to

57Assuming a 45° pit slope and a cylindrical vertical ore body the ratio of waste to ore increases in proportion to the depth cubed. 128 3 Copper and Coal Through the Ages

Fig. 3.43 Block caving (McGraw-Hill 2007) replace it. The force of gravity replaces the need for any further drilling and blasting as depicted in Fig. 3.43. There is, however, a similar pattern of development for each block of ore and the pattern may be repeated below the ﬁrst level if the ore body continues at a greater depth. Block caving, like all underground mining, is not without risk and danger. In November 1999, at Northparkes in Australia, the ore body above the cave collapsed unexpectedly into the void, creating an air blast that travelled through the underground workings of the mine. The force of the air blast was such that roof bolts and metal mesh were bent, motor vehicles destroyed and four miners, including the mine manager, were killed. To reduce risk, robotic trucks and scoops may be used to remove the ore. Once again Bingham Canyon is challenging the old paradigms in technology to underground mine relatively low grade ore on a scale never before achieved. The plan is to mine over ﬁfty million tonnes of ore each year, as much as the pit currently produces. Although ambitious in its size, the mining method is not dissimilar to the underground mining method used by Boston Consolidated as described in the thesis of Boucher (1914). Figure 3.44 captures the moment of a large landslide at Bingham Canyon Mine in April 2013 (Sutherlin 2014). Monitors showed an increasing rate of strain in the hours before failure indicated that a collapse was imminent. The mining company released a warning about the landslide in advance and no one was injured by the slide. This was an impressive example of landslide prediction using several monitoring methods. Such an incident, although impacting mine production, indicates how far mine subsidence monitoring and prediction has improved and augurs well for the safe use of block caving. University of Utah scientists reported that the 3.4 The Age of Electricity 129

Fig. 3.44 2013 Manefay fault landslide (Sutherlin 2014) landslide on the Manefay Fault, estimated to be over 140 million tonnes, was the biggest non-volcanic slide in North America’s modern history (Carter 2014). Before moving the discussion on to Chile, currently the world’s largest producer of copper, one major innovation in copper production has yet to be addressed. As Fig. 3.45 shows, the price of copper reached its lowest level in real terms towards the end of the twentieth century forcing producers to investigate lower cost production methods. The pyrometallurgical58 advances in copper extraction so far described were followed by possibly even more signiﬁcant hydrometallurgical59 extraction methods. Beginning in the mid-1980s, a new technology, commonly known as the heap leach-solvent extraction-electrowinning process or SX-EW was adopted by mines seeking to reduce the cost of production. The SX-EW process uses weak sulfuric acid at ambient temperature and pressure to extract copper from copper oxide and certain types of low grade sulﬁde ores in existing mine dumps, prepared ore heaps or in situ ore (Myers 2010). As earlier described, copper has been extracted by leaching and then recovered by precipitation with iron for centuries. There are numerous patents60 relating to the process and until the twentieth century this was the only hydrometallurgical method

58Ore reﬁning by heating, such as smelting. 59Ore reﬁning by liquid processes, such as leaching. 60See: Recovery of copper by cementation U.S. 3930847 A at http://www.google.com/patents/ US3930847. 130 3 Copper and Coal Through the Ages

$11,000 20,000 Nominal price $/tonne $10,000 Real price 1998 $/tonne 18,000 $9,000 Annual world production (tonnes' 000) 16,000 $8,000 14,000 $7,000 12,000 $6,000 10,000 $5,000 thousandtonnes $/tonne 8,000 $4,000 6,000 $3,000 $2,000 4,000 Worldproduction $1,000 2,000 $ 1905 1920 1930 1945 1955 1960 1970 1980 1995 1900 1910 1915 1925 1935 1940 1950 1965 1975 1985 1990 2000 2005 2010

Fig. 3.45 Copper price and world production (Porter et al. 2015) available to produce copper metal. With the commercial development of electricity production, the direct electrowinning of copper became possible and an early application of electrowinning took place in 1915 at Chuquicamata Mine in Chile (Tilton and Landsberg 1999). These processes did not produce copper of sufficiently high purity to compete with electrorefined copper; however, solvent extraction technology, developed to recover uranium during and after World War II, provided the needed breakthrough. The LX/SX/EW (leaching/solvent extraction/electrowinning) process is often described as the SX-EW process. The process shown in Fig. 3.46 commences by leaching existing mine dumps, prepared ore heaps, or in situ ore with a weak acidic solution. The copper rich solution is then mixed with an organic solvent in the extraction stage (SX) that selectively removes the copper. The copper-loaded extractant is then mixed with an aqueous acid solution that strips it of the copper. The resulting electrolyte is highly concentrated and relatively pure, and is processed into high quality copper in the third and final stage by electrowinning (EW). Although the 1915 application of electrowinning in Chile was unsuccessful, in 2013 SX-EW cathodes accounted for 70% of the refined copper produced in Chile In 1990, the SX-EW process accounted for only 6.4% of the world copper production (Ayres et al. 2002); however, by 2013, the SX-EW accounted for 18% of world output (ICSG 2014) and 33% of the copper produced in Chile (Simpson et al. 2014; Fagerström 2015). Nevertheless, concentrates still accounted for 45% of the copper exported from Chile. Producing concentrates and SX-EW are, however, electricity intensive processes and it was relatively cheap energy from fossil fuels that enabled the production of copper to increase from under one million tonnes in 1900 to almost nineteen million tonnes in 2014. Contemporaneously, the need for copper in all parts of the 3.4 The Age of Electricity 131

Fig. 3.46 SX-EW flow sheet with permission of Professor James Myers (Myers 2010) electricity industry drove up the demand for copper. The contribution coal made to world energy demand including from copper production will be discussed before proceeding to the expansion of mining in Chile, the country that is currently the dominant copper producer. Figure 3.47 shows the fivefold increase in coal production in the twentieth century and the contribution by the major producing countries. Annual world coal production grew from under a billion tonnes at the beginning of the century to almost five billion tonnes by the end of the century. Examining the line representing Europe and the UK, we can see the impacts on coal production from both world wars and the reduction during the great depression from 1929 through to 1936. The latter is also reflected in U.S. production. Although coal production decreased in both Germany and Britain during the wars, the reductions occurred at different times. During WWI, British coal production dropped at the beginning of the war as miners enlisted in the armed forces. Having learnt from this experience the British Government conscripted young men to work in the mines in the Second World War.61 In Germany coal production

61Between 1943 and 1948, 48,000 men, known as the ‘Bevin Boys’ were directed to work in British coal mines. One in ten of those conscripted into military service were balloted to work in the mines. Approximately 43% of the Bevin Boys were ‘ballotees’, the remainder were those who volunteered to work in the mines after being called up. 132 3 Copper and Coal Through the Ages

Coal production in the 20th century

5,000 Europe & UK

US 4,000 Other including Australia & South Africa Asia 3,000 Russia

Total 2,000

Production million tonnes 1,000

0 1901 1911 1921 1931 1941 1951 1961 1971 1981 1991

Fig. 3.47 Major coal-producing countries in the twentieth century dropped dramatically in 1945 because the transport infrastructure had been destroyed. The Attlee Labour government nationalised the British coal industry in 1946. A sign posted at every colliery stated that ‘This colliery is now managed by the National Coal Board on behalf of the people’. The National Coal Board (NCB) acquired 958 collieries, the property of about 800 companies. Compensation of £164,660,000 was paid to the owners for the collieries (Hill 2001). Coal accounted for 90% of UK energy consumption in 1950; however, with the relatively lower price of petroleum and natural gas, coal’s share had dropped to 31% by 1980. During that period the workforce was reduced from some 700,000 to 240,000. As Fig. 3.48 shows, strikes caused significant dips in production, and the 1984 strike was no exception. Given that coal production and the workforce had been steady declining since 1950 and the rate of decline in production appears to be no greater after the strike, the consternation caused by strike of 1984 is difficult to comprehend. The strike was reportedly triggered by the NCB’s intention to close five uneconomic pits. The average selling price from British collieries reported by Robinson (1988) was some £42/tonne in 1983. Coal was being delivered to Rot- terdam from South Africa and Australia for under £32/tonne. Operating costs provided by Minford and Kung (1988) show that, for the 24 most uneconomic collieries, the average operating cost was some £78/tonne. Davis and Metcalf (1988) cite the Monopoly and Mergers Commission Report that estimated that 3.4 The Age of Electricity 133

Coal produced million tonnes Colliery output million tonnes Coal imports million tonnes Employees thousands Tonnes per employee 300 6,000

250 5,000

200 4,000 1921 Miners strike

150 3,000 1926 General strike Million tonnes 100 2,000

50 1984 Miners strike 1,000 Employees and tonnes peer employee

0 0 1980 1929 1950 1962 1965 1995 1998 1914 1944 1923 1926 1989 1992 1935 1938 1941 1956 1971 1977 1920 1983 1986 1932 1947 1953 1968 1974 1917 1959

Fig. 3.48 UK coal production and employment 1914–2000 seventy collieries were operating with losses of greater than £10/tonne in 1982. In 1983, taxpayer-funded subsidy to coal mining was some £1.3 billion. This ﬁgure did not include the cost to taxpayer-funded industries such as steel and electricity that were obliged to buy British coal (Phelan 2012). The miners’ strike of 1984 and the subsequent impact on mining families may be attributed to the willingness of governments to introduce subsidies, and their subsequent reluctance to reduce the subsidies until forced to when the amount of the subsidy becomes unsustainable.62 Although coal remained the dominant source of world energy until 1960, oil production had been increasing from 1914. The ﬁrst Model T Ford was produced in 1908, and by using assembly line techniques, reputedly one could be bought for the price of four months’ wages by a worker on Henry Ford’s assembly line. Mass production of motor vehicles stimulated demand in America and worldwide for gasoline (petrol). Britain’s Royal Navy commenced building oil powered light cruisers in 1912, a high risk strategy given that England had abundant coal reserves but no known oil reserves. In 1901, the Shah of Persia granted William Knox D’Arcy63 a concession

62The U.S. motorists’ subsidy to the ethanol industry was an estimated $83 billion between 2007 and 2014 as a consequence of the U.S. Government’s ethanol mandate (Bryce 2015). Paradoxically, the oil subsidy to citizens of Iran and Saudi Arabia was estimated to be almost $90 billion in 2013 (IEA 2015). 63D’Arcy migrated to Rockhampton, Australia from England in 1866. In 1882, he joined a syndicate to develop the Mount Morgan gold and copper mine. Supported by a rich dividend stream and the sale of some of his shares in the mine, D’Arcy, then extremely wealthy, returned to England in 1887. In part, it was the proceeds of copper mining that funded the discovery of oil in Persia. 134 3 Copper and Coal Through the Ages

World coal oil and gas production 18,000 16,000 14,000 12,000 10,000 Gas 8,000 Oil 6,000 Coal 4,000 2,000 Million tonnes coal equivalent coal tonnes Million - 1915 1970 1920 1975 1980 1925 1930 1955 1985 1990 1900 1905 1910 1935 1940 1945 1950 1960 1965 1995 2000

Fig. 3.49 World coal, oil and gas production in million tonnes coal equivalent. Sources Rutledge (2011), BP (2014a), Benichou (2015), U.S. Energy Information Administration (2015) to explore, drill, produce, and export petroleum in Persia for a period of 60 years. In 1908, thanks to the persistence of geologist George Reynold, oil was discovered near Masjed Soleyman in modern day Iran close to the ruins of an ancient Zoroastrain ﬁre temple. In 1909, a new company, the Anglo-Persian Oil Company (APOC), replaced the Concessions Syndicate Ltd. D’Arcy remained its director until his death in 1917. In 1914, the British government, in which Winston Churchill served as First Lord of the Admiralty, bought 51% of APOC in order to secure oil supply for the Royal Navy (Sorkhabi 2008). In 1935, APOC was renamed the Anglo-Iranian Oil Company (AIOC), and in 1954 it became the British Petroleum Company (BP). The production data for oil and gas shown in Fig. 3.49 have been converted to million tonnes of coal equivalent (Mtce) for convenient comparison. The conversion is based on the average gross energy content of each fuel and the conversion multipliers were 2.0 and 2.6 for oil and gas, respectively. In 1900, the world’s main oil producers were Russia and the United States. Some eighty percent of the oil produced was reﬁned into kerosene to provide illuminating oil.64 After 1920, oil replaced coal in engines in Britain and America putting downward pressure on the price of coal, and there were industrial disputes in coal mines as owners tried to cut costs by cutting wages. In the U.S. coal was replaced

64The thesis that cheaper kerosene saved the whales from extinction is not entirely supported by the evidence. After World War II, fats were rationed in Europe and new factories and more powerful ships were built to hunt and process whales for edible oil. The Tangalooma Whaling Station, just 75 min from Brisbane, Australia, operated from 1952 until 1962. In total, 6277 humpback whales were killed. In 1951, whale oil sold for around £140 per tonne. However, with the availability of other cheaper vegetable and ﬁsh oil, the price fell to about £60 per tonne in 1961. Additionally, the quota of 700 whales was not met. A lack of whales and a low price for whale oil caused the demise of the whaling industry in Australia. Visitors can now view whales not far from the old Tangalooma Whaling Station. 3.4 The Age of Electricity 135

Copper mined in the 20th century 14,000 12,000 US Chile World

10,000 8,000 6,000

4,000 2,000

Copper mined annually (t'000) - 1975 1945 1915 1925 1955 1960 1970 1980 1990 2000 1900 1905 1940 1995 1910 1920 1935 1950 1930 1965 1985

Fig. 3.50 US, Chile and total world copper mined by oil as the main source of energy by the late 1930s. The Texan field, discovered in 1930, was within three years providing half of American output. Between 1920 and 1950, the United States never produced less than half the world’s oil. In Europe, coal still provided nearly ninety percent of energy needs in 1950. However, with the development of oil in the Middle East, Western Europe tran- sitioned from coal to oil in the 1950s, which resulted in many coal mine closures. Only in Eastern Europe was coal still providing over eighty percent of energy in the late 1960s. In 1950, the Middle East region provided less than 10% of the world’s oil; however, by the 1970s, the figure was over 30% (Ponting 1999). As depicted in Fig. 3.49, coal remained the dominant energy source until surpassed by oil in the 1960s and by the end of the twentieth century, the shares for coal, oil and gas production were 28, 45 and 27%, respectively. The United States was the dominant world producer of coal for much of the twentieth century until surpassed by China in the 1980s. Also in the 1980s, Chile overtook the United States as the primary source of mined copper. Figure 3.50 shows that copper production began to increase almost exponentially from 1945 onward as new mines were developed in many countries including Australia, Canada, New Guinea, Philippines South Africa and in the Central African Copperbelt.65 Nevertheless, the U.S. remained a major producer of both coal and copper throughout the century. The U.S. had abundant oil and coal resources and copper miners were able to adopt the most cost-effective energy source. Additionally, efficiencies in mining and concentrating the copper ores were able to offset the cost

65The Central African Copperbelt, one of the largest resources of copper in the world, straddles the border between Zambia and the Democratic Republic of Congo. 136 3 Copper and Coal Through the Ages

Ore grade vs. energy to concentrate a kilogram of copper 14.0 0.90 0.80 12.0 Energy to concentrate ore trendline 0.70 10.0 Copper ore grade 0.60

8.0 0.50

6.0 0.40 kWh/kg 0.30 4.0 0.20

2.0 0.10 Copper ore grade percent

0.0 0.00 1950 1960 1970 1980 1990 2000 2010

Fig. 3.51 US copper mining and concentrating—energy used per kilogram (Golding and Campbell 2014) of treating the reducing grade of copper ore. Figure 3.51 depicts the reduction in energy achieved even as grade decreased. Although Chile has almost 30% of the world copper reserves and mines over 30% of the worlds copper (USGS 2015), copper production has been hampered by the lack of energy resources. By the end of the twentieth century, Chile was importing more than 65% of its energy needs (The World Bank 2015). How Chile overcame the lack of indigenous energy resource to become the world’s largest producer of copper and home to the ﬁrst mine to produce over one million tonnes of copper in a single year is the subject of the next section.

3.5 The Mega Mines

Given the success in mining massive low grade copper ore bodies in North America, prospectors began looking for similar ore bodies elsewhere. The presence of mineral wealth in South America was known to the outside world following conquistador Francisco Pizarro’s entry into northern Peru in 1528. It was the cre- ation of beautiful works in gold, silver and copper by the Inca that was to be their downfall. Nevertheless, it was not until large reserves of low grade porphyry deposits were identiﬁed in the early twentieth century that Chile became a signiﬁcant copper producer. Today Chile produces some 30% of the world’s copper, and one mine Chuquicamata holds the record for the total volume of copper produced although the Escondida Mine to the south now holds the annual production record. 3.5 The Mega Mines 137

The story of Chuquicamata, ‘Chuqui’ to the local population, encapsulates the history of copper mining in Chile. The mine located in the north of Chile, just outside of Calama and some 215 km northeast of Antofagasta is by excavated volume the biggest open-pit copper mine on Earth. How this region, once part of the pre-Columbian Tiwanaku culture and later part of Bolivia, was annexed by Chile is an important and intriguing part of Chile’s history. Evidence of early mining of copper in Chile was provided in 1899 when the mummified body of a miner was found at Chuquicamata. The corpse, along with his mining tools, was reportedly discovered at a depth of two metres inside an ancient mining tunnel where the young man appeared to have been trapped by a cave-in. He was covered in a film of copper dust and its antibacterial properties prevented bacteria from attacking his flesh. The hot, dry, anaerobic conditions of his tomb and the copper dust both contributed to the preservation of the mummy (Sapse and Kobilinsky 2011). The accidentally mummified body, colloquially known as the ‘Copper Man’, was acquired by J.P. Morgan for the American Museum of Natural History in New York City where his body is on display. In 1978, carbon-14 analyses dated the clothes and body to 550 ± 40. Fuller (2004) surmises that the metallic copper produced from the ore Copper Man mined was carried by caravans that circulated between the high plain of what is currently Bolivia and the coast where the current Chilean city of Antofagasta is located. Furthermore, some of the copper fishing hooks, which were then in use along the whole of the coast of Chile, were possibly the product of the labours of Copper Man or his compatriots. How the copper was smelted remains a mystery? Bellows were not used in the region until after the arrival of Spanish following the Battle of Cajamarca in 1532 when Pizarro captured the Inca ruler Atahualpa. Air may have been forced through the smelting structure to increase the temperature of the fire by locating the furnace on a hillside or in other locations where strong winds would circulate the air naturally as was previously described at Timna. Alternately, smiths may have blown into the furnace using wooden or ceramic tubes known as toberas as depicted in Fig. 3.52. Adding to the smiths’ difficulty, the most common near surface copper mineral was chalcanthite, accompanied by minor quantities of brochantite and atacamite. These copper sulfates and chlorides are more difficult to process than the copper oxides and carbonates worked by ancient smiths in areas such as Timna because reduction of the sulfate and chloride ores requires temperatures greater than those necessary to smelt copper oxides, and carbonates. Fuller (2004) demonstrated that copper metal could be produced by adding sodium nitrate, which is abundant in local soils, to create an exothermic reaction in which combustion reached a temperature of more than 1200°C. Fuller concluded that by adding sodium nitrate, it was necessary to reach only 771°C and when this temperature is reached, the reaction inside the hearth activates spontaneously, surpassing 1085°C and generating liquid copper. Although not explored in the literature, paratacamite found at Timna and in the Atacama Desert contains zinc and perhaps the zinc lowered the melting point of the copper-zinc ore. 138 3 Copper and Coal Through the Ages

Fig. 3.52 Moche smiths smelting—by kind permission of Nathan Benn who photographed pottery from the collection of the National Museum of Archaeology and Anthropology in Lima, Peru

In 1912, a well-known German archaeologist, Max Uhle described the remains of an ancient indigenous metallurgical site in the area of Calama, where metallic copper had been produced. He estimated that the remains discovered in Calama must have been from approximately 1000 years before his visit. This statement, along with the presence of Copper Man in Chuquicamata lead Fuller to conclude that copper smelting in the area occurred during the period of the Tiwanaku culture expansion. The loincloth and artefacts found with the copper also indicate that the miner was probably related to the Tiwanaku culture. The Tiwanaku (Spanish: ‘Tiahuanaco’), centred on Lake Titicaca in Bolivia, flourished between approximately 200 and 1100, expanding their influence into northern Chile (Sapse and Kobilinsky 2011). Although the presence of copper was known at least since 550, as evidence by Copper Man, the resource was not exploited to any great extent until after the ‘War of the Pacific’. In 1866, Chile and Bolivia agreed to a border on the 24th parallel. The treaty gave Chilean and Bolivian interests equal right to exploit the resources between the 23rd and 25th parallels and guaranteed their Governments half of the tax revenues from the minerals exported from the region. 3.5 The Mega Mines 139

Fig. 3.53 War of the Paciﬁc: adapted from Wikipedia (2014b)

Consequently, Anglo-Chilean capital was invested in the region and mining expanded and mining practices advanced. Sodium nitrate was in demand as a fertiliser as well as for the manufacture of explosives, and nitrate exports from the Atacama Desert doubled from 1865 to 1875. A second treaty in 1874 left the Chile-Bolivia border at the 24th parallel (Fig. 3.53); however, Chile relinquished the rights to taxes north of the 24th parallel and received a 25 year guarantee that taxes on Chilean enterprise operating in the Bolivian province of Antofagasta would not increase. Nevertheless, four years later Bolivia increased the taxes on the nitrate exports from Antofagasta, violating the 1874 treaty. Chilean companies operating in Antofagasta refused to pay the increased taxes, and the Bolivian government ordered the confiscation of the Chilean properties. On 14 February 1879, the day set for the seizure and sale of the Chilean properties, Chilean troops occupied the Port of Antofagasta. Peru attempted to mediate a peace; however, when Chile became aware of a secret treaty between Peru and Bolivia that required either country to support the other in the case of a war with Chile, Chile declared war on both countries on 5 April 1879. So commenced the ‘War of the Pacific’ also known as the ‘Saltpetre War’ (Sater 2007; Wikipedia 2014b). The near lifeless salt pans of the Atacama region, which were formed from the desiccated remains of an ancient inland sea and minerals washed down from the Andes, are particularly rich in sodium nitrate, a salt also known as Chile saltpetre. Ironically, Peru was in financial difficulties because of large investments made on 140 3 Copper and Coal Through the Ages the strength of the income from guano, then one of the major sources of potassium nitrate also known as saltpetre. The reserves of guano had not been as great as Peru had anticipated. Guano exports dropped from 580,000 tonnes in 1869 to less than 355,000 tonnes in 1873 as the guano on Chincha and other islands were depleted. All three countries had serious economic problems; however, Chile’s financial woes were not as desperate as those of Peru and Bolivia. Chile also had the support of English capital interests and a superior navy. Following sea and land battles, the Peruvian capital Lima was captured by Chile in January 1881 and the Peruvians finally conceded defeat in October 1883. Bolivia did not sue for peace until 1884. The Chilean victory gave Chile control of Bolivia’s former province of Antofa- gasta, Peru’s former provinces of Tacna, Carapace and Arica, and vast deposits of nitrate and the undeveloped copper resources (Keen and Haynes 2013). A nitrate boom followed that gave Chile forty years of prosperity. Additionally, miners with ‘Red Gold Fever’ arrived searching for high grade copper ores. Soon the area surrounding the current mine of Chuquicamata was covered with mines and mining claims. Mining was unruly and disorganised, and title to claims were disputed due in part to the defective 1873 Mining Code and also because rebels confiscated mines belonging to loyalists following the capture of Calama in 1891 during the Chilean Civil War. Many of the miners lived in lawless shanty towns around the mines that provided alcohol, gambling and prostitution and where murder was a regular occurrence. As late as 1918, the army had to be sent into keep order. These early miners focused on the rich veins containing up to 15% copper. One attempt was made to process the low grade ore in 1900 by Norman Walker, a partner in La Compañia de Cobres de Antofagasta, but it failed leaving the company deeply in debt. Nevertheless, larger mining companies eventually emerged, organised as commercial rather than mining operations to avoid the imperfections of the Mining Code, and started to buy up and consolidate the small mines and claims (Wikipedia 2014a). Exploitation of the lower grade copper ores was made possible by metallurgical developments in the United States. The Guggenheim Brothers examined claims around Chuquicamata owned by the lawyer and industrialist Albert Burrage and estimated reserves of some 690 million tonnes averaging some 2.6% copper. They created the Chile Exploration Company (Chilex) in January 1912 and eventually bought out Burrage for $25 million in Chilex stock. The Guggenheims were familiar with mining in Chile having completed the development of the El Teniente copper resource. El Teniente (The Lieutenant) is a porphyry/breccia style copper-molybdenum ore deposit that rims a dormant volcano on the western slope of the Andes Mountains situated 75 km southeast of Santiago and 2300 m above sea level. The Guggenheim Exploration Co. (Guggenex) bought a controlling interest in the developing mine in 1909 and completed the facility, including a smelter and refinery, at a cost of $25 million. By 1915 it was turning out over 22,000 tonnes of copper a year at a production cost of less than eighteen cents per kilogram (Codelco 2014). 3.5 The Mega Mines 141

E.A. Cappelen Smith, consulting metallurgist for the Guggenheims, worked out the first process for the treatment of Chuquicamata oxidised copper ore. Chilex then developed a mine on the eastern section of the Chuquicamata field and gradually acquired the remainder of the field over the next 15 years. The leaching plant was planned to produce some 50,000 tonnes of electrolytic copper annually. Amongst the equipment purchased were steam shovels from the Panama Canal. A port and an oil-fired power plant were built at Tocopilla, 145 km west on the Pacific coast and an aqueduct was constructed to bring water in from the Andes. Production started in 1915 and rose from 4300 tonnes in the first year to 50,400 tonnes in 1920. Chuquicamata became one of the world’s most productive and profitable copper mines. In 1923, refined copper was being produced for as little as thirteen cents a kilogram, amongst the lowest cost in the world. Even so, in the commodities slump that followed WWI, it was losing money. Anaconda Copper Co. bought 51% of the mine in 1923 for $70 million and the remainder in 1929. That year the mine produced 135,890 tonnes of copper before the Depression hit and demand fell. Production for many years came from the oxidised capping of the orebody that only required leaching followed by electrowinning of the copper; however, by 1951, the oxidised reserves were largely exhausted and Anaconda built a mill, flotation plant and smelter to treat the huge reserves of underlying supergene copper sulfides. These secondary sulfides arise from the leaching of the overlying ore and its re-deposition on and replacement of the deeper primary (hypogene) sulfides. Chuquicamata was developed into the world’s largest open-pit copper mine and by the end of 1955, it had yielded almost 5 million tonnes of copper. The Chilean copper operations of Anaconda were lucrative for many years. The Chilean government introduced a 12% income tax in 1922; however, by 1952 the total tax burden was over 70% of income; nevertheless, the Anaconda mines remained profitable. Production did not rise significantly during the 1950s, as copper began to lose market share to aluminium. Moreover, the percentage of Chilean copper ore from all Chile’s mines that was refined in Chile dropped from 89 to 45%. Anaconda agreed to raise production markedly in return for a tax rate cut; however, in 1969 Anaconda sold 51% to the Chilean government (Codelco 2014). Anaconda built a town adjacent to the mine equipped with a railroad, schools, soccer fields and social clubs. The town of some 20,000 people was evacuated to nearby Calama in 2007 for health reasons and to make way for mining. The seven-floor hospital Anaconda built in the 1960s is now buried under waste rock from the expanding mine (Henao 2012). In 2014, the statue of Chilean independence leader Bernardo O’Higgins66 shown in Fig. 3.54, stood in the deserted town defiantly facing the encroaching mine. In 1971, Chile’s newly elected socialist president, Salvador Allende, expropriated Anaconda’s Chilean copper mines. The Allende government was overthrown in 1973, and the new military government agreed to pay Anaconda some

66Bernardo O’Higgins (1778–1842) was a Chilean independence leader who, together with José de San Martín, freed Chile from Spanish rule in the Chilean War of Independence. 142 3 Copper and Coal Through the Ages

Fig. 3.54 Chilean independence leader Bernardo O’Higgins

$250 million for its expropriated mines (Encyclopaedia Britannica 2014). Corpo- racion Nacional del Cobre de Chile (Codelco) was established in 1976 to operate the Andina, Chuquicamata, El Salvador, and El Teniente mines. In the following year, Codelco’s mines produced about 890,000 tonnes of copper and Chuquicamata accounted for about 579,000 tonnes, a then world record for an individual mine. Codelco’s production had risen to 1.1 million tonnes by 1983, when its sales reached $1.8 billion. With the costs of some 97 cents a kilogram, amongst the lowest in the world, Codelco achieved a net income of $220.6 million and paid 3.5 The Mega Mines 143

$679 million in taxes in 1983 providing 46% of Chile’s foreign exchange. For most of the 1980s, however, world copper prices slumped because of the development of fibre optics and superconductors and the substitution in many cases of aluminium and plastics. In 1994, Codelco discovered that an employee had cost the company at least $175 million between 1989 and 1994 by making bad commodities trades in the futures market. The scandal led to the resignation of Codelco’s chief executive and seven other senior executives and a net loss for 1993. Codelco opened the Radomiro Tomic Mine adjacent to Chuquicamata in 1998. Unfortunately, world copper prices fell to an average low of $1.54/kg the next year, the lowest in the previous twelve years and the lowest, in real terms, in sixty years. Fortunately, Codelco produced a record of 1.69 Mt of copper in 1999, Chuquicamata accounting for 650,000 tonnes (Codelco 2014). World demand for copper increased by over 50% from 2000 to 2014, mainly due to Chinese demand that rose from just 12% of world refined copper consumption to 50%. The copper price reached an all-time high of almost $10/kg in early 2011 and Codelco made a pre-tax profit of just over seven billion dollars (Hernández 2012). That year the combined copper output of Chuquicamata and adjacent Radomiro Tomic was some 913,000 tonnes. For many years, Chuquicamata was the mine with the largest annual production in the world but was recently overtaken by Minera Escondida. Nevertheless, it remains the mine with by far the largest total production, estimated in 2013 by Mudd et el. (2013) to be 55 Mt of copper including the adjacent Radomiro Tomić Mine. Despite over 90 years of intensive exploitation, it remains one of the largest known copper resources. Electricity for the Chuquicamata Mine continued to be supplied from Tocopilla. In 2011, Codelco sold its 40% stake to E-CL, the principal electricity generator in Chile’s Northern Electricity Grid (SING), for $1.04 billion. E-CL is constructing a 375 MW pulverized coal-fired unit at nearby Mejillones that is scheduled for completion in 2018 (E-ECL GDF-SUEZ 2014; Baker 2015). In 2012, the combined copper production from Codelco’s Chuquicamata and adjacent Radomiro Tomic Mine was reported to be 784,000 tonnes of copper (Dewison and Hinde 2014) approximately half of which was in cathode copper and the remainder was concentrates containing around 30% copper. Some 2500 GWh of electricity were consumed (Fagerström 2015) that equates to approximately 3.2 kWh/kg of contained copper produced. This figure is similar to the Escondida Mine to the south that reported electricity consumption of approximately 3.3 kWh/kg of copper in 2011 (Mackenzie 2012). Electrical energy may be only half the total energy required to produce a kilogram of copper as much of the mobile equipment in the mine is powered by diesel fuel. Energy currently accounts for about 20% of production cost and Chilean mines pay some of the highest prices in the world for energy according to an article in Nueva Mineria y Energia (nme 2012). The article referenced a report from Feller Rate, a subsidiary of Standard & Poor’s, which estimated that the kilowatt-hours of energy required to produce one tonne of copper had increased 31% between 2002 and 2011 and that in 2011 energy 144 3 Copper and Coal Through the Ages represented between 15 and 20% of cash cost and may reach a value of ‘between 35 and 45% of C167 costs by 2020’. Wind turbines now dot the horizon as one approaches the Calama airport and CODELCO inaugurated Calama Solar 3 in June 2012, the country’s first industrial solar power plant with overall power of 1 MW to supply energy to Codelco’s Chuquicamata Mine. Nevertheless, according to the Centre for Studies on Copper and Mining (CESCO), the solution for the mining industry’s energy shortages is not likely to come from sunlight. It was the mining industry’s unsatisfied demand for electricity that drove up the costs, and both problems can only be solved with large-scale solutions that offer an abundant supply of energy. The cost of energy in Chile has risen sevenfold in the last decade. Simultaneously, the mining industry is required to show a reduction in greenhouse gas emissions. Generating electricity using thermal, coal or diesel-fired plants, or hydropower, could provide an answer; however, environmental and social conflicts are hindering, or entirely blocking these kinds of initiatives (Jarroud 2013) The viewing platform overlooking the Chuquicamata open pit provides an awesome view of possibly the world’s largest human construction. Comparisons are difficult. At 2.5 km wide it is twice the width of the Dardanelles68 which is 1.2 km wide at the narrowest point and only 55 m deep, whereas Chuquicamata is approaching one kilometre in depth. The Great Pyramid of Giza, which is 140 m high and 230 m wide at the base, would fit easily in one corner of the open pit. Trucks, winding their way out of the pit carrying over 360 tonnes of ore each load on tyres more than twice the height of their driver, look like miniature toys. The ore, containing less than 1% copper, is crushed and then concentrated in froth flotation cells to provide a concentrate containing approximately 33% copper that feeds the Outotec Flash Smelter. Copper anodes containing approximately 98% copper are sent to the electrolytic refinery with a capacity to produce 855,000 tonnes of copper cathodes per annum that are 99.99% copper. Finally, the cathodes are railed to the port of Mejillones for export (Fig. 3.55). The humble shaft, two metres below the ground, that became Copper Man’s tomb over 1500 years ago, evolved into the Chuquicamata Mine producing over 55 million tonnes of copper, more than any other mine on Earth. Caravans of railcars, with each railcar carrying approximately 46 tonne of copper, have replaced the lama caravans that may have transported Copper Man’s hard won copper to the coast. As discussed in the previous section, Chile has been disadvantaged by its relatively small fossil fuel resources. Geography and geology have been determining factors in electric power generation and distribution. Chile stretches some 4300 km

67The cost of producing the metal, excluding royalties, interest payments and depreciation. 68The Dardanelles, formerly known as the Hellespont, is a narrow strait in Turkey connecting the Aegean Sea to the Sea of Marmara. During the second Persian invasion of Greece in 480 BC, the Persian King Xerxes constructed pontoon bridges to traverse the Hellespont. In World War I, the Allies in their attempt to capture Constantinople failed to break through the Ottoman mine ﬁeld laid across the Dardanelles. A land invasion that included Australians and New Zealanders (the original ANZACS) was attempted and also failed. 3.5 The Mega Mines 145

Fig. 3.55 Copper cathodes passing through Calama en route to port from its borders with Peru and Bolivia above the Tropic of Capricorn in the north to Cape Horn in the south where it is the closest country to the Antarctic. Its average width is about 180 km, bordered by the Andes mountains on the east and the Paciﬁc Ocean on the west. Chile contains some of the most spectacular and contrasting landscapes in the world. In the south penguins are more prevalent than people and in the north is the Atacama Desert, one of the driest deserts in the world, where some of the world’s largest copper mines may be found. Unfortunately, the relatively small coal resources are located in the southern most region where the Mina Invierno (literally Winter Mine) coal mining project began production in 2013 and is expected to meet 30% of Chile’s coal demand. Chile’s power generation is organised around four grid systems. The two most important are the northern grid, Sistema Interconectado del Norte Grande (SING) that accounts for about 19% of national generation and the Central Interconnected System (SIC) that accounts for some 69% of national generation. SING supplies the northern regions, where approximately 90% of customers are large industrial and mining operations (GENI 2015). SING had an installed capacity of approximately 4055 MW in 2014. Coal-ﬁred power stations provided some 80% of the electricity generated (Olave and Garcia 2014). Chile imported over nine million tonnes of thermal coal in 2014 of which SING consumed about six million tonnes. Some two million tonnes of coal were exported from Chile’s single operating coal mine Mina Invierno (EQUUS 2015). With the need for energy to mine and produce copper, and relatively small indigenous energy resources, Chile takes a particular interest in energy use in the copper industry. The Comision Chilena del Cobre (Chilean Copper Commission) within the Ministerio de Minería (Ministry of Mining) tracks energy usage in the mining industry. Additionally, the Comision Chilena del Cobre tracks the grade of copper mined and by combining this information we can see from Fig. 3.56 that as the grade of copper has decreased since 2001, the amount of energy consumed has increased. Only power consumed in mining, concentrating and LX/SX/EW (Leaching/ Solvent Extraction/Electrowinning) has been included in the data to produce Fig. 3.56 so as to make the graph comparable with Fig. 3.51 in the previous section. Since the concentrating process produces a concentrate containing some 30% copper, no matter what the ore grade (head grade), a decreasing grade means a 146 3 Copper and Coal Through the Ages

Energy consumed in mining concentrating and electrowinning copper 7.0 1.4

6.0 1.2

5.0 1.0

4.0 0.8

3.0 0.6

2.0 0.4

Energy consumed KWh/kg 1.0 0.2 Copper ore grade mined % copper

- - 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

kWh/kg Copper ore grade %

Fig. 3.56 Energy consumption versus copper ore grade (Comision Chilena del Cobre 2014, 2015) greater volume of ore has to be crushed and concentrated. However, decreasing grade is not the sole reason for the increase in energy required and the Comision Chilena del Cobre cites the increasing depth of open-pit mining along with the construction of desalination plants as additional causes of the increase in energy consumed per tonne of copper. Somewhat forebodingly, the data for the period from 2001 to 2014 indicate that improvements in technology have not been sufﬁcient to counterbalance the impact of decreasing grade on energy requirements. This is an important factor when attempting to predict the future cost of producing copper and the sustainability of the metal that is the subject of the ﬁnal chapter.

References

Agrawal RD, Kapoor ML (1982) Theoretical considerations of the cementation of copper with iron. J South Afr Inst Min Metall 6 Agricola G (1556) De Re Metallica (trans: Hoover HC, Hoover LH). Dover Publications, New York Al-Bashaireh KS (2008) Chronology and technological production styles of Nabatean and Roman plasters and mortars at Petra (Jordan). The University of Arizona Allen RC (2011) Global economic history: a very short introduction. Very short introductions, vol 282. Oxford University Press, Oxford Andrews EB (1889) The late copper syndicate. Q J Econ 3(4):508–516 References 147

Anglesey Mining plc (2014) Parys mountain—a special place. Amlwch Industrial Heritage Trust, Amlwch Anon (1987) Rio Tinto mines. The engineer 16 Sept 1887 Antill PD, Rickard J (2014) Battle of Trafalgar. Military history encyclopedia on the web. http:// www.historyofwar.org/articles/battles_trafalgar2.html Avery D (1974) Not on Queen Victoria’s birthday: the story of the Rio Tinto mines. Collins, London Ayres RU, Ayres LW, Råde I (2002) The life cycle of copper, its co-products and by-products. Mining, minerals and sustainable development—MMSD working paper series. International Institute for Environment and Development (IIED), London Bailey LR (1988) A history of Bingham Canyon, Utah—Old reliable. Westernlore Press, Tucson Baker A (2015) Chile’s E-CL to shelve 375 MW coal project. BNamericas. Business News Americas, Santiago Bamberger M, Wincierz P, Bachmann HG, Rothenberg B (1986) Ancient smelting of oxide copper ore: archaeological evidence at Timna and experimental approach. Metallwissenschaft Technik 1166–1174 Bancroft HH (1889) History of Utah, 1540–1886. The History Company, San Francisco Barnatt J, Rieuwerts J, Thomas GH (1997) Early use of Gunpowder in the Peak District. Bull Peak District Mines Hist Soc 13(4) Barnatt J, Timberlake S (2013) The Ecton mines: four millennia of extracting copper. Br Archaeol Beaver SH (1951) Coke manufacture in great Britain: a study in industrial geography. Trans Pap Inst Br Geogr 17 Ben-Yosef E (2010) Technology and social process: oscillations in Iron Age copper production and power in southern Jordan. University of California, San Diego Ben-Yosef E (2012) Environmental constraints on ancient copper production in the Aravah Valley. Tel Aviv 39 Ben-Yosef E, Tauxe L, Levy TE (2010) Archaeomagnetic dating of copper smelting site F2 in the Timna Valley (Israel) and its implications for the modelling of ancient technological developments. Archaeometry 52(6) Benichou L (2015) The shift project. Paris Blakelock E, Martino M, Veldhuijzen HA, Young T (2009) Slag inclusions in iron objects and the quest for provenance: an experiment and a case study. J Archaeol Sci 36(8):1745–1757 Bleiwas DI (2011) Estimates of Electricity requirements for the recovery of mineral commodities, with examples applied to Sub-Saharan Africa. U.S. Department of the Interior and U.S. Geological Survey, Reston, Virginia Bornhorst TJ, Barron RJ (2011) Copper deposits of the western Upper Peninsula of Michigan. Geol Soc Am Field Guide 24 Bostock J, Riley HT (eds) (1885) Pliny the Elder, the natural history, vol XXXIII. H. G. Bohn, London Boucher JL (1914) Mining methods and geology of the Boston consolidated mine at Bingham Utah. University of Missouri, Rolla BP (2014a) BP statistical review of world energy. British Petroleum, London BP (2014b) First oil. British Petroleum, London Broderick TM (1931) Fissure vein and lode relations in Michigan copper deposits. Econ Geol 26 (8):840–856 Bryce R (2015) End the ethanol rip-off. New York Times, New York Buchwald VF (2005) Iron and steel in ancient times. Det Kongelige Danske Videnskabernes Selskab, Copenhagen Burt R, Kudo N (2014) Tin and global capitalism, 1850–2000: a history of “the Devil’s Metal”. In: Ingulstad M, Perchard A, Storli E (eds) Routledge international studies in business history, vol 25. Routledge, Abingdon, pp 22–45 148 3 Copper and Coal Through the Ages

Burton MM, Levy TE (2011) The end of the Chalcolithic period (4500–3600) in the northern Negev Desert, Israel. In: Lovell J, Rowan Y (eds) Culture, chronology and the chalcolithic. Theory and transition vol 9. Levant supplementary Oxbow Books, Oxford, pp 178–191 Caesar J (1869) Book 5. Commentaries on the Gallic War. Harper & Brothers, New York Callcut V (2005) John Wilkinson—Copper King—a review of the activities of John Wilkinson in connection with the copper industry. Wilkinson Lecture, Broseley Campbell E, Lavier AJ (2014) Copper, dubbed by traders as the metal with the economics Ph.D., may need a new nickname. Bloomberg, New York Carpenter HCH (1918) Progress in the metallurgy of copper. J Roy Soc Arts 66(3398):114–123 Carter RA (2014) Recovering from Bingham Canyon’s record-setting 2013 Slide. Eng Min J Chandler H (ed) (1996) Heat treater’s guide: practices and procedures for nonferrous alloys. ASM International, Ohio Chaplow C (2015) Huelva Province—Rio Tinto Mines. Andalucia.Com, Malaga Codelco (2014) Corporacion Nacional del Cobre de Chile—company profile, information, business description, history, background information on Corporacion Nacional del Cobre de Chile. Codelco, Santiago Colombo J (2012) The Mississippi bubble of 1718–1720 http://www.thebubblebubble.com/ mississippi-bubble/ Comision Chilena del Cobre (2014) Análisis de Variables Claves para la Sustentabilidad de la Minería en Chile. Santiago Comision Chilena del Cobre (2015) Update report on energy usage in copper mining for 2014. Santiago Cowen R (2009) Essays on geology, history, and people, chapter 3: “fire and metals: copper”. University of California, Davis Craddock PT (2000) From hearth to furnace: evidences for the earliest metal smelting technologies in the Eastern Mediterranean. Paléorient 26(2):151–165 Davies G, Metcalf D (1988) Pit closures: some economics. In: Cooper D, Hopper T (eds) Debating coal closures—economic calculations in the coal dispute 1984–5. Cambridge University Press, Cambridge, pp 25–45 Day DT (1904) Mineral resources of the United States 1902. Government Printing Office, Washington Dewison P, Hinde C (2014) The copper king. Mining Magazine Dickinson HW (2010) Matthew Boulton. Cambridge University Press, Cambridge Dresher WH (2001) How hydrometallurgy and the SX/EW process made copper the “green” metal. Copperorg Drew GJ (2011) Australia’s first mining era. Mineral resources and energy resources divisions. MESA J. 61 Duell P (1938) The Mastaba of Mereruka, Part I vol 1. OIP 31, vol XXXI. Oriental Institute Publications, Chicago E-ECL GDF-SUEZ (2014) www.e-cl.cl/prontus_ecl Ebrey P, Walthall A, Palais J (2009) East Asia: a cultural, social, and political history. Wadsworth, Belmont, CA Edmondson JC (1989) Mining in the later Roman Empire and beyond: continuity or disruption? J Rom Stud 79 Eissler M (1902) The hydro-metallurgy of copper. Crosby, Lockwood & Son, London Elmore FE (1898) Improvements in separating metallic from rocky constituents of ores and apparatus therefor. UK Patent, 1900 Encyclopaedia Britannica (2014) Anaconda Company https://www.britannica.com/topic/ Anaconda-Company EQUUS (2015) Quarterly activities report March 2015. EQUUS Mining Limited, Sydney Evans C, Saunders O (2015) Copper ore: an unlikely global commodity. The Open University’s History Department References 149

Fagerström C (2015) Copper mining in Chile and its electric power demand. Novia University of Applied Sciences, Vaasa Falola T, Warnock A (eds) (2007) Encyclopaedia of the middle passage. Greenwood Press, Westport Farmer M (2011) A King’s Ransom. http://merryfarmer.net/tag/richard-the-lionheart/ Faulkner N (2011) Roman Britain, 43-410 AD. British Broadcasting Corporation, London Finay JR (1920) The cost of mining, 3rd edn. McGraw-Hill, New York Firdu FT (2009) Kinetics of copper reduction from molten slags. Helsinki University of Technology, Helsinki Frago AF (1990) The history of literacy in Spain: evolution, traits, and questions. Hist Educ Q 30 (4) Friede HM, Steel RH (1997) An experimental study of iron-smelting techniques used in the South African Iron Age. J South Afr Inst Min Metall Fuller DF (2004) The production of copper in 6th century Chile’s Chuquicamata mine. J Miner Met Mater Soc 56(11):62–66 Gallo E (2014) Destruction in early Bronze Age Southern Levant. Rome “La Sapienza” Stud Archaeol Palestine and Transjordan 11:141 GENI (2015) Chile electricity—grid summary. Global Energy Network Institute, San Diego, California Gibbons JH (1988) Energy use in the copper industry. In: US Congress Office of Technology Assessment (ed) Copper: technology and competitiveness. US Government Printing Office, Washington, DC, pp 151–158 Gibson T (1878) The Huelva Pier of the Rio Tinto railway. Inst Civil Eng 53(3) Gilead I, Gošić M (2014) Fifty years later: a critical review of the stratigraphy, chronology and context of the Nahal Mishmar hoard. J Isr Prehistoric Soc 44:226–239 Goffer Z (2007) Archaeological chemistry, 2nd edn. Wiley, Somerset, NJ Golding B, Campbell HF (2014) Sustainability of metals production: the case of copper. Min Econ 26(3) Golding SD (1999) Key to minerals. Knowledge Books & Software, Varsity Lakes Gomez DJC, Martinez GA, Macias ER (1999) The Tinto River: a very old acid river of the Iberian Pyrite Belt. In: Mine, Water & Environment—1999 IMWA Congress, Sevilla Greenly E (1919) The geology of Anglesey. Memoirs of the geological survey of Great Britain. J. Truscott and son, Limited, London Griffin JB (1961) Lake Superior Copper and the Indians, vol 17. Lake Superior Copper and the Indians: Miscellaneous Studies of Great Lakes Prehistory, Anthropological Papers, Museum of Anthropology. University of Michigan, Michigan Gundiler IH (2000) The Hidalgo smelter. In: Lawton TF, McMillan NJ, McLemore VT (eds) 51st field conference southwest passage-A trip through the Phanerozoic, New Mexico, New Mexico Geological Society, Socorro, pp 263–266 Hair TH, Ross M (1844) A series of views of the collieries in the counties of Northumberland and Durham. J. Madden & Company, London Hansen S (2013) Innovative metals: copper, gold and silver in the Black Sea Region and the Carpathian Basin during the 5th and 4th millennium BC. In: Burmeister S, Hansen S, Kunst M, Müller-Scheeßel N (eds) Metal matters; innovative technologies and social change in prehistory and antiquity. Vml Verlag Marie Leidorf Rahden/Westf., pp 137–167 Hansen Streily A (2000) Early Pottery Kilns in the Middle East. Paléorient 69–81 Harms W (1994) Bronze Age source of tin discovered. Univ Chicago Chronicle 13(9) Harris J (1982) The electricity of Holborn. New Scientist vol 14 Jan. Reed Business Information Ltd, London Harvey CE (1981) The Rio Tinto company: an economic history of a leading international mining concern 1873–1954. Alison Hodge, Penzance, Cornwall Hatcher J (1973) English tin production and trade before 1550. Clarendon Press, Oxford 150 3 Copper and Coal Through the Ages

Hauptmann A (2007) The Archaeometallurgy of copper—evidence from Faynan, Jordan. Natural Science in Archaeology. Springer, Berlin Heffernan T (2013) The new Bronze Age. Pacific Standard, North Hollywood, California Henao LA (2012) Chile’s huge open-pit copper mine goes underground. Associated Press, New York Herbert EW (1984) Red gold of Africa: copper in precolonial history and culture. University of Wisconsin Press, London Hernández D (2012) Codelco financial results January–December 2011. Codelco, Santiago Hill A (2001) The South Yorkshire Coalfield: a history and development. Tempus, Brimscombe Port Hoffman PT, Postel-Vinay G, Rosenthal J (2009) Surviving large losses: financial crises, the middle class, and the development of capital markets. Harvard University Press, Cambridge, MA Hollister-Short G (1994) The introduction of powder. Bull Peak District Min Hist Soc 12(3) Hong S, Candelone J, Soutifd M, Boutron CF (1996) A reconstruction of changes in copper production and copper emissions to the atmosphere during the past 7000 years. Sci Total Environ 188(2–3):183–193 Hoover TJ (1912) Concentrating ores by flotation; being a description and history of a recent metallurgical development, together with a summary of patents and litigation. The Mining Magazine, London Horne L (1982) Fuel for the metal worker: the role of charcoal and charcoal production in ancient metallurgy. Mag Archaeol Anthropol 25:6–13 Hughes S (2000) Copperopolis: landscapes of the early industrial period in Swansea. Royal Commission on the Ancient and Historical Monuments of Wales, Aberystwyth Hunt WR (1990) Golden places—the history of Alaska-Yukon mining. National Park Service, Anchorage ICSG (ed) (2014) The world copper facebook 2014. International Copper Study Group (ICSG), Lisbon IEA (2015) IEA fossil-fuel subsidies database. International Energy Agency, Paris Jarroud M (2013) Chile’s mining industry turns to sunlight to ease energy shortage. Tierra America. United Nations Environment Programme, Regional Office for Latin America and the Caribbean, Panama City Jones E (1848) Amlwch and the celebrated Mona and Parys copper mines. Enoch Jones Beaumaris Jones GDB (1980) The Roman mines at Riotinto. J Rom Stud 70:146–165 Joralemon IB (1973) Copper: the encompassing story of mankind’s first metal. Howell-North Books, Berkeley Kaufman B (2011) Metallurgy and archaeological change in the ancient near east annual review of the Cotsen Institute of Archaeology at UCLA. CIoA Press Los Angeles Keen B, Haynes K (2013) A history of Latin America, Ninth edn, vol 1: ancient America to 1910, vol 1. Ninth edn. Wadsworth, Boston King MG (2007) The evolution of technology for extractive metallurgy over the last 50 years—Is the best yet to come? J Miner Met Mater Soc 59(2) Kojo IV, Jokilaakso A, Hanniala P (2000) Flash smelting and converting furnaces: a 50 year retrospect. JOM 53(2):57–61 Krehl POK (2009) History of shock waves, explosions and impact: a chronological and biographical reference. Springer, Berlin, Heidelberg Lambert JB (1997) Traces of the past: unravelling the secrets of archaeology through chemistry. Addison-Wesley, Reading, MA Laurent PE (1846) The nine books of the history of Herodotus. Book IV Melpomene, 3rd edn. Henry Slatter, Oxford Lawrence S, Davies P (2012) An archaeology of Australia since 1788. Springer, New York References 151

Leblanc MJA, Morales JM, Elbaz-Poulichet F (2000) 4,500 year-old mining pollution in south-western Spain: long-term implications for modern mining pollution. Econ Geol 95:655– 662 Lees D (2009) The history of drill and blast. Sydney Lehner JW, Yener KA (2014) Organization and specialization of early mining and metal technologies in Anatolia. In: Roberts BW, Thornton CP (eds) Archaeometallurgy in global perspective: methods and syntheses. Springer, New York, pp 529–557 Lewis RL (2008) Welsh Americans: a history of assimilation in the coalfields. University of North Carolina Press, Chapel Hill Long G (1833) Amlwch copper mines. The Penny Cyclopædia of the society for the diffusion of useful knowledge, vol 1. Charles Knight and Co, London Long G (1838) Founding. The Penny cyclopædia of the society for the diffusion of useful knowledge, vol 10. Charles Knight and Co, London Lucas A, Harris JR (1948) Ancient Egyptian materials and industries. Arnold, London Mackenzie A (2012) Escondida—base metals briefing and Chilean site tour—securing our power needs for the medium term http://www.bhpbilliton.com/*/media/bhp/documents/investors/ reports/2012/120930_bhpbillitonbasemetalsbriefingsitetour.pdf?la=en Mäkinen J (2006) Eco-efficient solutions in the Finnish metallurgical industry. VTT Business Solutions Martin SR (1995) The state of our knowledge about ancient copper mining in Michigan. Mich Archaeologist 41(2/3) Martin SR (2008) Copper mining in the Great Lakes USA. In: Selin H (ed) Encyclopaedia of the history of science, technology, and medicine in non-western cultures. Springer Netherlands, Verlag Berlin Heidelberg New York, pp 1680–1685 Massy C (2011) Breaking the sheep’s back. University of Queensland Press, St Lucia, QLD McGraw-Hill (2007) McGraw-Hill encyclopedia of science and technology, vol 29, 10th edn. Mark D. Licker, Columbus, OH Merkel JF (1990) Experimental reconstruction of Bronze Age copper smelting based on archaeological evidence from Timna. In: Rothenberg B (ed) The ancient metallurgy of copper. Researches in the Arabah, 1959–1984. Institute for Archaeo-Metallurgical Studies, London, pp 78–122 Merriam-Webster (2015) Regulus. Merriam-Webster. Springfield, MA Milwaukee Public Museum (2015) Old copper culture. Milwaukee Public Museum, Milwaukee Minford P, Kung p (1988) The costs and benefits of coal pit closures. In: Cooper D, Hopper T (eds) Debating coal closures—economic calculations in the coal dispute 1984-5. Cambridge University Press, Cambridge, pp 46–56 Moorey PRS (1999) Ancient mesopotamian materials and industries: the archaeological evidence. Eisenbrauns Winona Lake, Indiana Mortimer Wheeler RE, Clapham AW, Godfrey WG (1930) Prefatory note on Bath. In: Proceedings at meetings of the Royal Archaeological Institute the summer meeting at Bath Mudd GM, Weng Z, Jowitt SM (2013) A detailed assessment of global Cu resource trends and endowments. Econ Geol 108:1163–1183 Muhly JD (1999) Copper and bronze in Cyprus and the Eastern Mediterranean. In: Pigott VC (ed) The archaeometallurgy of the Asian Old World. The University of Pennsylvania Museum of Archaeology and Anthropology, Philadelphia, PA, pp 15–25 Myers J (2010) Production: ore processing: low grade—SX-EW. Web Gallery—Metals: Scarce: Base—Copper. University of Wyoming, Laramie Nash WG (1904) The Rio Tinto mine; its history and romance Simpkin, Marshall, Hamilton. Kent & Co Ltd, London Nef JU (1966) The rise of the British coal industry, vol 2. Frank Cass & Co., London New York Times (1910) Bribery charge in copper merger suit. The New York Times, 25 Jan 1910 Newberry PE (1900) The life of Rekhmara, Vezir of Upper Egypt under Thothmes III and Amenhetep II (c. 1471–1448 BC). Constable, London 152 3 Copper and Coal Through the Ages

Newell E (1990) “Copperopolis”: the rise and fall of the copper industry in the Swansea district, 1826–1921. In: Harvey C, Press J (ed) International competition and industrial change: essays in the history of mining and metallurgy 1800–1950. Routledge, pp 75–97 Newman CJ, Collins DN, Weddick AJ (1999) Recent operation and environmental control in the Kennecott Smelter. In: George DB, Chen WJ, Mackay PJ, Weddick AJ (eds) Fourth international conference copper 99 volume V, Pointe Hilton Resort on South Mountain Phoenix, Arizona, USA. TMS The Minerals, Metals and Minerals Society, Warrendale, PA, pp 29–45 nme (2012) Chilean copper mining costs triple during last decade. Nueva Mineria y Energia Nueva Mineria y Energia. http://www.nuevamineria.com/revista/chilean-copper-mining-costs-triple- during-last-decade/ O’Brien W (1999) Bronze Age copper mining in Britain and Ireland. Shire Publications Ltd, Princes Risborough O’Farrell J (2007) An Utterly impartial history of Britain: Or 2000 years of upper class idiots. Random House, London Olave X, Garcia P (2014) E.CL S.A. Full Rating Report. Fitch Ratings Olías M, Nieto JM (2015) Background conditions and mining pollution throughout history in the Río Tinto (SW Spain). Environments 2 Palermo E (2014) Who invented the light bulb? http://www.livescience.com/43424-who-invented- the-light-bulb.html, New York Palmer RE (1928) Notes on some ancient mine equipment and systems. Trans Inst Min Metall 36:299–336 Penzance Town Council (2015) Penzance a town with a history. Penzance Town Council, Penzance Percy J (1875) Metallurgy: the art of extracting metals from their ores. J. Murray, London Phelan J (2012) Thatcher’s achievements will long outlive the spite of Sheffield’s sons and daughters. The commentator. Robin Shepherd, London Pollard S (1980) A new estimate of British coal production, 1750–1850. Econ Hist Rev New Ser 33(2) Ponting C (1999) The Pimlico history of the twentieth century. Random House Porter KE, Edelstein DL, Brininstool M (2015) Copper statistics USGS. Washington Powell K (ed) (1994) Utah history encyclopedia. University of Utah Press, Salt Lake City Priyadarshi H, Caldwell J (2012) Block caving. InfoMine http://technology.infomine.com/reviews/ Blockcaving/welcome.asp?view=full Radivojević M, Rehren T, Kuzmanović-Cvetković J, Jovanović M, Northover P (2013) Tainted ores and the rise of tin bronzes in Eurasia, c. 6500 years ago. Antiquity 87:1030–1045 Raymond RW (1881) A glossary of mining and metallurgical terms. Transactions of the American Institute of Mine Engineers IX (May 1880 to February 1881) Rehder JE (2000) Rehder 2000 mastery and uses of fire in antiquity. McGill-Queen’s Press, Montreal Richard TA (ed) (1916) The flotation process. Mining and Scientific Press Rickard TA, Ralston OC (1917) Flotation. Dewey Publishing Co., San Francisco Rio Tinto (2006) Kennecott Utah Copper—copper environmental profile life cycle assessment. Kennecott Utah Copper Salt Lake City Rio Tinto (2015) Rio Tinto delivers strong first quarter production. Media Release, London Rio Tinto Kennecott (2015) Visitors centre virtual tour. Rio Tinto Kennecott. http://www. kennecott.com/visitors-center Roberts BW, Thornton C (2014) Archaeometallurgy in global perspective: methods and syntheses. Springer, New York Roberts RO (1956) The development and decline of the copper and other non-ferrous metal industries in South Wales. Transactions of the Honourable Society of Cymmrodorion, pp 78– 115 References 153

Robinson PW (1988) Economic background to the coal dispute. In: Cooper D, Hopper T (eds) Debating coal closures—economic calculations in the coal dispute 1984–5. Cambridge University Press, Cambridge, pp 107–118 Rothenberg B (1972) Timna: valley of the biblical copper mines (new aspects of antiquity). Thames & Hudson Ltd, London Rothenberg B (ed) (1990) The ancient metallurgy of copper archaeology-experiment-theory, vol 2. Researches in Arabah 1959–1984. The Institute for Archaeo-Metallurgical Studies (IAMS), London Rothenberg B, Merkel J (1998) Chalcolithic, 5th millennium BC, copper smelting at Timna—New radiocarbon dating evidence for Timna Site 39. Inst Archaeo-Metallurgical Stud (IAMS) 20:1– 3 (University College London) Rothenberg B, Palomero FG (1986) The Rio Tinto Enigma—no more. Inst Archaeo-Metallurgical Stud (IAMS) Newsl 8:1–5 Rowland P (1992) Integrated catchment management in Queensland; projected impacts on the mining industry. In: Council AMI (ed) Seventh annual environmental workshop, Yeppoon. Australian Mining Industry Council, p 36 Rowlands J (1996) Copper mountain. Anglesey Antiquarian Society, Llangefni Rule J (1992) The vital century: England’s developing economy, 1714–1815. Longman, London Russell CA (2000) Chemistry, society and environment: a new history of the British chemical industry. Royal Society of Chemistry, Cambridge Rutledge D (2011) Estimating long-term world coal production with logit and probit transforms. Int J Coal Geol 85(1) Salkield LU, Cahalan MJ (1987) A technical history of the Rio Tinto mines: some notes on exploitation from pre-Phoenician times to the 1950s. Institution of Mining and Metallurgy, London Sánchez AI, Rodriguez N, Amalis R, Sans JL (2011) Microbial diversity in anaerobic sediments at Río Tinto, a naturally acidic environment with a high heavy metal content. Appl Environ Microbiol 77(17) Sapir-Hen L, Ben-Yosef E (2013) The introduction of domestic camels to the southern levant: evidence from the Aravah Valley. Tel Aviv 40:277–285 Sapse D, Kobilinsky L (eds) (2011) Forensic science advances and their application in the judiciary system. Taylor & Francis Inc, Bosa Roca Sater FS (2007) Andean tragedy: fighting the war of the Pacific, 1879–1884. Nebraska Press, Lincoln Schmitz CJ (1979) World non-ferrous metal production and prices, 1700–1976. Frank Cass, London Scoffern J, Oxland R, Truran WC, Fairburn W, Clay W, Atkin WC, Pickett WV (1857) The useful metals and their alloys. Houlston and Wright, London Scott DA (1992) Metallography and microstructure in ancient and historic metals. Getty Trust Publications, Los Angeles Scott DA (2010) Ancient metals: microstructure and metallurgy, vol 1. Conservation Science Press, CSP, Los Angeles Secundus GP (1952) Pliny the Elder—Natural History (trans: Rackham H). Harvard University Press, Cambridge, MA Segal I, Rothenberg B, Bar-Matthews M (1998) Smelting slag from prehistoric sites F2 and N3 in Timna, SW Arabah, Israel, vol 8. Metallurgica Antiqua, Deutsches Bergbau-Museum, Bochum Shalev S (2004) Swords and daggers in late Bronze Canaan. Franz Steiner Verlag Shalev S, Northover JP (1993) The metallurgy of the Nahal Mishmar hoard reconsidered. Archaeometry 35:35–47 Simpson M, Aravena E, Deverell J (2014) www.csiro.au/en/Research/Mining-manufacturing/ CSIRO-Chile/Future-of-Mining. Commonwealth Scientific and Industrial Research Organisa- tion (CSIRO), Clayton South, Victoria Smiles S (1959) The life of George Stephenson railway engineer John Murray, London 154 3 Copper and Coal Through the Ages

Smitheram C, Ben-Yosef E, Porat N, Faereshtein G (2013) The central Timna Valley project— optically stimulated luminescence (OSL), and its application in the dating of ancient copper mines. Central Timna Valley Project (CTV), Tel Aviv Snodgrass AM (2000) The dark age of Greece: an archaeological survey of the eleventh to the eighth centuries BC. Edinburgh University Press, Edinburgh Sorkhabi R (2008) The centenary of the first oil well in the middle east. GEO ExPro 5 Staniforth M (1985) The introduction and use of copper sheathing—a history. Bull Aust Inst Marit Archaeol 9(1/2) Stevens HJ (1908) The copper handbook, vol VIII. Houghton, Michigan Sutherlin C (2014) Lessons from Kennecott Utah Copper’s Manefay Slide. Paper presented at the 22nd annual mine design, operations & closure conference, Butte, Montana, 29 Apr 2014 Symons JC (2003) The mining and smelting of copper in England and Wales, 1760–1820. Coventry University, Coventry Taussig FW (1905) The tariff history of the United States, 6th edn. G. P. Putnam’s Sons, New York The World Bank (2015) World development indicator. Energy imports, net (% of energy use). The World Bank, Washington, D.C Tilton JE, Landsberg HH (1999) Innovation, productivity growth, and the survival of the U.S. copper industry. In: Simpson RD (ed) Productivity in natural resource industries: improvement through innovation. Resources for the Future, New York, pp 109–139 Tylecote RF (1976) A history of metallurgy, 1st edn. The Metals Society, London U.S. Energy Information Administration (2015) International energy statistics https://www.eia. gov/beta/international/data/browser/#/?pa=000000001&c=ruvvvvvfvtvnvv1urvvvvfvvvvvvfvv vou20evvvvvvvvvnvvuvo&ct=0&vs=INTL.44-2-AFG-QBTU.A&ord=SA&vo=0&v=H&start =1980&end=2014 Uglow JS (2002) The lunar men: the friends who made the future, 1730–1810. Faber, London USGS (1885) Salt lake and Tooele. Perry-Castañeda Library, University of Texas at Austin USGS (1900) Bingham. Perry-Castañeda Library, University of Texas at Austin USGS (2015) Mineral commodity summaries. Copper statistics and information. U.S. Geological Survey, Reston, VA Valenzuela L (1992) The Chilean copper smelting industry in the mid-nineteenth century: phases of expansion and stagnation, 1834–58. J Latin Am Stud 24(3) Van Pelt JR (1964) Boston and Keweenaw—An Etching in Copper. Geolo Surv Bull 1 Velasco F, Herrero JM, Suárez S, Yusta I, Alvaro A, Tornos F (2013) Supergene features and evolution of gossans capping massive sulphide deposits in the Iberian Pyrite Belt. Ore Geol Rev Viduka A, Ness S (2004) Analysis of some copper-alloy items from HMAV Bounty wrecked at Pitcairn Island in 1790. In: Metal 04: international conference on conservation of metals, Canberra. National Museum of Australia, pp 160–172 Vivian HH (1881) Copper smelting: its history and processes. The Scientific Publishing Company, New York Walcott CD (1901) Twenty first annual report of the United States geological survey 1899–1900. Government Printing Office, Washington Wang H, Cowell M, Cribb J, Bowman S (eds) (2005) Metallurgical analysis of Chinese coins at the British Museum, vol British Museum Research Publication Number 152. The British Museum, London Watts-Plumpkin E (2005) Wadi Faynan, copper mine. Curr World Archaeol 13 Weinstein R (2011) The archaeology of Pewter vessels in England 1200–1700: a study of form and usage. Durham University Durham E-Theses Online Weisgerber G (2006) The mineral wealth of ancient Arabia and its use I: copper mining and smelting at Feinan and Timna—comparison and evaluation of techniques, production, and strategies. Arabian Archaeol Epigr 17(1) References 155

Weissenbacher M (2009) Sources of power: how energy forges human history. Praeger, Santa Barbara, CA Whittlesey C (1863) Ancient mining on the shores of Lake Superior. D. Appleton & Co, New York Wikipedia (2014a) Chuquicamata. Wikipedia. http://en.wikipedia.org/wiki/Chuquicamata Wikipedia (2014b) War in the Paciﬁc. Wikipedia. http://en.wikipedia.org/wiki/File:Map_of_the_ War_of_the_Paciﬁc.en.svg Willies L (1991) Ancient copper mining at Wadi Amram, Israel. Bull. Peak District Min Hist Soc 11(3) Wilson AJ (1977) Timna; cradle of the world’s copper mining industry. Min Mag 136:262–271 Woods M, Woods MB (2011) Ancient construction technology: from pyramids to fortresses. Twenty-First Century Books, Minneapolis Wu X, Zhang C, Goldberg P, Cohen D, Pan Y, Arpin T, Bar-Yosef O (2012) Early pottery at 20,000 years ago in Xianrendong Cave, China. Science 336(6089):1696–1700 Yam P (2005) China grid investment to boost copper consumption Reuters, 19 Jan 1 2015 Yener KA (2009) Strategic industries and Tin in the ancient near east: Anatolia updated. Academiaedu. https://www.academia.edu/1581326/Strategic_Industries_and_Tin_in_the_ Ancient_NE_Anatolia_Updated Young J (2012) About Bowes Railway. pers. comm., Newcastle upon Tyne Zhang Y (1986) Ancient Chinese sulfur manufacturing processes. Isis 77(3) Zhuwakinyu M (2013) A glimpse into Bronze Age copper mining. Mining Weekly October 11 The Future for Copper and Coal 4

Abstract Although sustainability was popularised in the late twentieth century, it was also a concern in the late eighteenth century as evidenced in the writings of Thomas Malthus. In his sixteenth century text, De Re Metallica, Georgio Agricola also addressed the environmental impacts of mining. Although copper mining on the Iberian Peninsula took place over thousands of years, no single copper ore body will last forever. The question ‘is copper mining sustainable?’ is addressed by examining previous research into the question and exploring various methods researchers have used. The statement that an activity is sustainable if it enhances or at least does not decrease human welfare now or in the future implies that the relative cost of producing copper will not increase. More precisely, ‘will the future cost of producing copper including environmental cost increase or decrease?’ U.S. data indicate that in the past technological advances have reduced the energy required to mine and concentrate copper ores even though ore grades have decreased. In 2002, the cost of producing copper was less than in 1954. An econometric model predicts that the unit cost of producing copper in the U.S. in 2020 will be less than in 2002. What happens after 2020 is less predictable.

4.1 Consumption of Copper and Coal Since 1940

The ten most important mineral products in the United States, in the order of value of annual output, are coal, iron, copper, clay products, petroleum, gold, stone, cement, natural gas. Any attempt to estimate the country’s mineral wealth must convince us of the vital importance of utilizing most efﬁciently these resources, which are so essential to the nation’s welfare.

Only as geologic explorations and mining operations uncover new deposits and block out known reserves can the United States…avoid facing the sure exhaustion of the supply of certain important minerals within this century (Smith 1909). Although the quotes above were speciﬁc to the U.S. at the beginning of the twentieth century, they apply generally to our Earth today. Advances in science, technology and governance have enabled the human population to grow from a few million at the end of the last Ice Age to over 7 billion in 2015. Copper and coal have made a signiﬁcant contribution to enabling population growth, especially the exponential increase from 1940 captured in Fig. 4.1. More importantly, as Fig. 4.2 shows, the volume of copper and coal consumed per person has increased from 1.0 and 734 kg in 1940 to 3.1 and 1061 kg, respectively, in 2015.

4.2 Defining Sustainability

There are many weighty tomes addressing the sustainability of the Earth’s human population, just a few of which are listed in Table 4.1. Although sustainability was popularised in the late twentieth century, it was also a concern in the late eighteenth century. Malthus (1798) argued that population increases geometrically while food supply increases arithmetically1 and consequently starvation was imminent for many. However, increases in food production have enabled a more than sixfold increase in population since Malthus made his prediction that population increase would outstrip food production. Researchers have attempted to ascertain the carrying capacity of particular environments, although their predictions have not always been appreciated by the powers of the day. In the 1920s, the academic geographer, Thomas Grifﬁth Taylor, of the University of Sydney attempted to dissuade the utopians who believed that the power of human ingenuity would enable Australia to support more than 60 million people. Taylor’s controversial view that the Australian desert was almost useless for settlement was condemned as unpatriotic. Taylor gathered what data were available of Australia’s water and soil resources, and argued that Australia could never support more than 60 million and predicted that Australia’s population would be no more than 20 million by the end of the century. Australia’s population was about 19 million in 2000. Taylor was attacked as a croaking pessimist and a prophet of environmental determinism. Taylor responded by demanding that geography be taught in schools in the place of classics, so that the next generation could analyse the situation for themselves. After a stormy decade of debate, Taylor

1For a geometric sequence, the ratio between two consecutive terms is constant. In an arithmetic sequence, the difference between two consecutive terms is constant. Our age is an arithmetic series, each year we get a year older. Our savings in the bank, assuming they earn a ﬁxed interest rate, grow geometrically. 4.2 Defining Sustainability 159

World population vs. copper and coal annual consumption 8,000 30 7,000 25 6,000 20 5,000 4,000 15 3,000 10 2,000 World population 5 1,000 - 0 1940 1950 1960 1970 1980 1990 2000 2010 Copper and coal annual consumption Population million Copper million tonnes Coal billion tonnes

Fig. 4.1 World population, copper and coal consumption 1940–2015

Coal kg /person Cu kg/person 1,200 3.5 1,000 3.0 2.5 800 2.0 600 1.5 400 1.0 Kilograms of coal 200 0.5 Kilograms of copper 0 0.0 1980 1940 1945 1950 1960 1965 1970 1975 1985 1995 2000 1955 2005 2010 2015 1990

Fig. 4.2 World coal and copper consumption per person shook the dust of Australia from his boots, and left for Canada, still under attack, still arguing cogently, persistently, undiplomatically his conservationist concerns (Stone 1996). Malthus would have been sceptical of any suggestion that the Earth’s human population would exceed six billion at the end of the second millennium, six times greater than the one billion humans who were alive when he made his prediction. Nevertheless, in 2012, almost one billion people lived below what is considered the 160 4 The Future for Copper and Coal

Table 4.1 A selection of events and works on sustainability 1798 Malthus—An Essay On the Principle of Population 1908 U.S. President Theodore Roosevelt sets up a National Conservation Commission to investigate the use, wastage and conservation of natural resources 1930s Hotelling examined the use of exhaustible resources and impact on future generations. The Hotelling rule—‘Competitive market forces will lead to efficient allocation over time’ 1969 U.S. creates the first national agency for environmental protection—the EPA 1972 The Club of Rome publishes The Limits to Growth and The LTG model 1973 OPEC oil crisis fuels the limits-to-growth debate 1980 Economist Julian Simon and biologist Paul Ehrlich bet on the price of copper, chrome, nickel, tin and tungsten in 1990 1982 Slade—Trends in Natural-Resource Commodity Prices. Resource prices appear to follow a U-shaped curve 1984 David Pearce appointed first Professor of Environmental Economics, University College London (UCL) 1987 Our Common Future (Brundtland Report) published. Defines ‘sustainable development’ 1992 UN Conference on Environment and Development (UNCED) Rio de Janeiro. The Earth Summit produces a treaty to reduce greenhouse gas emissions 1992 Beyond the Limits to Growth. Original authors revisit The Limits to Growth 1997 The Kyoto Protocol for the implementation of the Framework Convention on Climate Change is negotiated 2002 Rio+10 held in Johannesburg. Mining industry presents Breaking New Ground 2007 The Economics of Climate Change: The Stern Review. The costs of inaction on climate change will be equivalent to losing 5% of global GDP each year forever 2008 The Garnaut Climate Change Review. Proposes carbon trading 2015 UN Climate Change Conference, twenty-first yearly session of the Conference of the Parties (COP 21). Each country that ratifies the Paris Agreement is required to set its own target for emission reduction extreme poverty line of less than $1.90 a day. However, the number living below the poverty line in 1981 was almost twice the number in 2012. China alone accounted for most of the decline in extreme poverty over the past three decades where some 750 million people moved above the $1.90-a-day poverty threshold, thanks in part to China’s one child policy. The Malthusian proposition may also be evaluated by examining resources other than food supply, such as copper and coal. Figure 4.1 shows world population increased threefold from 2.3 billion in 1940 to 7.3 billion in 2015. At the same time, copper production increased ninefold allowing the annual consumption of copper per person to triple from one kilogram per person to over three. Similarly, coal consumption increased although not so dramatically. As with agricultural production, the data on copper and coal do not support the Malthusian proposition. 4.3 Sustainability Predictions Based on Reserves 161

4.3 Sustainability Predictions Based on Reserves

The Limits to Growth was written as a report for The Club of Rome’s project on the predicament of humanity. Donella Meadows and her colleagues from the Mas- sachusetts Institute of Technology developed World models encompassing population, capital, food, non-renewable resources and pollution. The authors’ primary conclusion from these models was that, with present growth trends in world population and consumption, the limits to growth would be reached in the next 100 years. Meadows et al. (1972) calculated how many years the reserves of metals and fossil fuels would last, a selection of which is shown in Table 4.2. Whilst the authors acknowledged many reservations about the necessary sim- plifications in the World model, they were led to the conclusion that, ‘The basic behaviour mode of the world system is exponential growth of population and capital, followed by collapse’ (Meadows et al. 1972). Critics of The Limits to Growth were numerous. The first chapter of Models of Doom by Cole et al. is titled ‘Malthus with a Computer’, portraying the obvious similarity between the conclusions drawn by Malthus and those of Meadows et al. One of the main modes of collapse in the model is resource depletion due to fixed levels of economically available resources and diminishing returns from technology change. Meadows et al. (1992) revisited The Limits to Growth in 1992 and concluded that, from the World 3 model, and what they had learned in the intervening 20 years, the primary conclusions they drew in The Limits to Growth were still valid. Donella Meadows died unexpectedly in 2001; however, out of respect, she was cited as principal author of Facing the Limits to Growth (Meadows et al. 2004). Three decades after The Limits to Growth was published, and within 20 years of the time when their scenarios suggest that growth would near its end, the authors’ basic conclusions were unchanged. Their model still suggested overshoot and collapse. Whereas Meadows et al. suggested copper reserves would last only 48 years, they were more optimistic about coal reserves estimating they would last 150 years. The science and engineering literature relating to sustainability presents a contrary view that is often pessimistic with regard to the availability of fossil fuels but optimistic regarding the long-term availability of minerals (Table 4.3). Given the history of human development, Cole et al. considered the Meadows et al. model assumptions were invalid. Although they saw no natural law barriers to producing metals, they noted one possible exception could be energy on which processing technology increasingly depends. They asserted that the then current prices of fuels did not reflect their long-run supply costs. They recommended that government sponsor research and development of new energy sources and technologies because the time frame for their development was beyond the normal planning horizons of individual firms. Stanley Jevons, who studied chemistry and was an assayer at the Sydney Mint predicted that Britain’s coal reserves would be uneconomic within 100 years and that world economic activity would decline because of the exhaustion of coal as an energy source (Jevons 1865). However, his advice to the government was not to 6 h uuefrCpe n Coal and Copper for Future The 4 162

Table 4.2 Selected resource reserves from The Limits to Growth Known Annual Years remaining at current Average Years remaining growing Years remaining at reserves consumption consumption and current consumption consumption current ﬁve times current reserves growth reserves reserves Million Million Years % Years Years tonnes tonnes Aluminium 1170 12 100 6.4 31 55 Chromium 775 2 420 2.6 95 155 Coal 5,000,000 2174 2300 4.1 111 150 Copper 308 9 36 4.6 21 48 Iron 100,000 417 240 1.8 93 173 Petroleum 65,000 2097 31 3.9 20 50 4.3 Sustainability Predictions Based on Reserves 163

Table 4.3 Scientists and engineers on metals and energy resources 1865 Jevons—The Coal Question, predicting England would run out of coal in 100 years 1956 Hubbert predicts U.S. oil production would peak around 1972. Hubbert based his estimates on oil production fitting a Normal Bell Curve 1974 Manne presents a case for the U.S. spending up to seven billion dollars on research to have the breeder reactor available by 1995 1983 Chapman—Metal Resources and Energy. Fuel cost will come to dominate the total cost of metal production 1993 Skinner—Finding Mineral Resources and the Consequences of Using Them: Major Challenges in the 21st Century. The global resources of minerals mined for purposes other than energy are such that exhaustion will not be a problem a century hence 1998 Lomborg—The Skeptical Environmentalist. New oil fields will be continually added as price rises 1998 Colin Campbell and Jean Laherrere publish an article in Scientific American entitled The end of cheap oil. They argued that world oil production will most likely peak in volume before 2010 2001 Deffeyes publishes Hubbert’s Peak—The Impending World Oil Shortage. Predicts world production will peak in 2005 2002 Steen and Gunnar—An Estimation of the Sustainable Production of Metal Concentrates from the Earth’s Crust. They see no theoretical constraint on metal production given sufficient energy conserve coal because as it became scarcer, the economic system would respond by bidding up its price. Although his prediction that coal mining in the United Kingdom would be uneconomic in 100 years may have seemed premature, many British coal mines may have been uneconomic in 1965, as Jevons predicted. The British Government nationalised the coal industry in 1947 and then subsidised coal production up until the miner’s strike of 1984. Prime Minister Thatcher closed the heavily subsidised mines after the miners’ strike and annual coal production in the United Kingdom continued to fall from its 1913 peak of 287 Mt. Kellingley Col- liery, the last operating deep coal mine in the UK began closing in December 2015. Kellingley still had vast reserves of coal; however, the mine could not compete with cheaper imports, a lack of investment and increases in the carbon tax on generating electricity from coal (Jones 2015). UK mines produced 8.5 Mt of coal in 2015 and imported 25.5 Mt; of the total coal supplied, electrical generation consumed 29.4 Mt coal (UK Government 2016). King Hubbert (1903–1989), an American geophysicist predicted that U.S. oil production would peak in the early 1970s. Although oil production from the lower 48 states did have a peak around 1970 as Fig. 4.3 shows, oil production in 2015 was almost equal to the 1970 peak production, thanks to shale oil extraction.2

2Shale oil is extracted out of oil-bearing shales from which the oil will not naturally ﬂow freely. Shale oil has become more accessible due to advances in horizontal drilling and hydraulic fracturing or ‘fracking’ technology. 164 4 The Future for Copper and Coal

600

500

400

300

200 Million tonnes

100

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

Fig. 4.3 U.S. crude oil production (EIA 2016)

In Hubbert’s last paper in 1982, he estimated world oil reserves of 2 trillion barrels and, using his 1956 method, predicted world oil production would peak in 2005. Deffeyes (2001) surmised that most of the Earth’s endowment of oil that had been accumulated over millions of years would be consumed within 100 years. The volume of oil currently consumed each year took on average three million years to form. The short bump of oil exploitation on the geological timeline became known as ‘Hubbert’s Peak’. Fifteen years after Hubbert’s prediction, Trainer (1997) quoted the 1995 report, World Oil Supply 1930–2050, by Campbell and Laherrere, Petroconsultants Pty. Ltd. predicting oil production would plateau in 2000. Duncan (1997) described the progress of Industrial Society as a single-pulse waveform measured by the average energy used per person per year through time. Duncan’s Olduvai3 theory states that the life expectancy of Industrial Civilization is some 100 years from 1930 to 2030 based on the ratio of world energy production to population. Energy consumption use per capita did appear to peak in 1979. Fortunately for consumers of oil, technological advances in oil extraction has delivered a sufﬁcient increase in oil production to keep pace with population growth and, as Fig. 4.4 shows, the volume of oil consumed per head of population has been stable since 1985. Although the peak oil consumption of 700 kg/person achieved in 1979 has not yet been surpassed, overall fossil fuel energy consumed per person is now above the 1979 level as Fig. 4.5 shows.

3The Olduvai Gorge is a picturesque valley that extends east to west within the Serengeti National Park in northern Tanzania. Duncan (2001) chose the name ‘Olduvai’ because it is famous for the myriad of hominid fossils and stone tools discovered there and it is a good metaphor for the Stone Age way of life. 4.3 Sustainability Predictions Based on Reserves 165

World oil production Kg of oil/person

5,000 800

4,500 700 4,000 600 3,500 3,000 500 2,500 400

2,000 300

Million tonnes 1,500 200 1,000 kilograms of oil per person 500 100 - - 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Fig. 4.4 World oil production and intensity (BP 2016b)

Fossil fuel energy Mtce Population million tce/person

25,000 3.5

3.0 20,000 2.5

15,000 2.0

10,000 1.5 1.0 5,000 0.5

- - Tonnes of coal equivalent per person Population and fossil fuels consumed 1995 2005 1920 1960 1980 1990 2000 1930 1940 1945 1950 1900 1905 1910 1915 1965 1970 1985 1925 1935 1955 1975 2010 2015

Fig. 4.5 Fossil fuel energy consumption (The Shift Project 2016; Rutledge 2011;BP2016b)

In hindsight, the Peak oil predictions made in the late twentieth century sound apocalyptic now; however, they were not without reason. The 1970s oil price hike and similar increase in coal price, as shown in Fig. 4.6, created an atmosphere of apprehension about the availability of fossil fuels and the world’s industrial future. In his 1976 paper, Salant (1976) considered the impact of industrial structure on exhaustible resources, more especially the impact of the Organization of Petroleum Exporting Countries (OPEC) cartel on the oil market. He predicted that a dispro- portional share of the increased proﬁts resulting from the formation of an oil producers’ cartel would go to non-members and that the cartel’s restriction on sales 166 4 The Future for Copper and Coal

Coal price Oil price

160

140

120

100

0 Coal price/tonne and oil price/barrel 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Fig. 4.6 Oil and coal price in 2015 dollars (BP 2015; The World Bank 2016) would eventually leave it as the sole supplier of oil. This proposition assumes no other reserves of oil were discovered, and is similar to the assumptions that Hya- cinthe Secrétan made when he attempted to control the world copper market as described in Chap. 3. The high oil prices encouraged the search for oil and as the newly found oil fields were developed the price of oil fell as Fig. 4.6 shows. Supply from OPEC members was below their capacity to supply through the last 20 years of the twentieth century and although prices were low, they may have been even lower if OPEC had produced at capacity. Contrary to the supply position taken by OPEC members in the later part of the twentieth century, when the U.S. began producing significant volumes of oil from oil shale, OPEC did not initially reduce supply and Saudi Arabia increased production from 538 Mt in 2013 to 569 Mt in 2015 (BP 2016a). Simmons (2000) revisited The Limits to Growth and concluded that energy limits must be a genuine concern and he foreshadowed the risk of serious energy shortages possibly as early as 2010 and noted that the issue has been largely ignored over the past 30 years. Simmons considered that the challenge of guiding the world to true global prosperity by 2050 or 2070 should be taxing all the world’s best minds. ‘Instead of rolling up their collective sleeves to begin addressing serious energy issues, these kibitzers spent their precious hours attacking the few voices of energy sanity. Over the years, the energy economists’ incorrect dismissal of The Limits to Growth was not only a mistake but their criticism also turned somewhat mean-spirited and at times even shrill!’ (Simmons 2000). In contrast to the predictions of oil scarcity, Lomborg (1998) quoted the Energy Information Agency and U.S. Geological Survey to infer that new fields would be continually added as demand rose and that the oil price per barrel will decline to the low $20’s until 2020. 4.3 Sustainability Predictions Based on Reserves 167

Whilst there has been vigorous debate regarding the reserves of oil, the pro- jections of metal scarcity raised in The Limits to Growth have scarcely received a mention, even though exhaustion times for copper listed in Table 4.2 are similar to that of oil. There are two possible reasons for this lack of concern regards metal scarcity. First, reserves of metals have generally increased in the intervening period and, second, it may be feasible to extract metals from ubiquitous crustal rocks implying that, in principle, metal reserves are inexhaustible. The First Law of Thermodynamics is often called the Law of Conservation of Energy. The First Law states that energy can never be created or destroyed; it can only be transformed from one form to another form. For example, when petrol is ignited in an engine, part of it is transformed into work, some of it changes into heat and some part goes out as exhaust products. The Second Law of Thermodynamics, the Entropy Law or Law of Entropy, states that every time energy is transformed from one state to another, there is a loss in the amount of that form of energy that becomes available to perform work of some kind. The loss in the amount of ‘available energy’ is known as ‘entropy’ or conversely entropy is a measure of the unavailable energy in a thermodynamic system. Entropy can be considered as a negative kind of quantity, the opposite of available energy. When a piece of coal burns the total amount of energy remains the same; however, in the process of burning, the coal is transformed into gases and ash that can no longer generate the same work as was generated in burning the coal. The amount of thermal energy available to do work has decreased and the amount of entropy has increased. The Second Law of Thermodynamics explains that the total entropy in the world is constantly increasing. An entropy increase, therefore, means a decrease in available energy. This is the principle of the ‘degradation of energy’.4 Georgescu-Roegen focused on closed systems, analogous to the Earth, where energy such as radiation can enter or leave the system, but materials cannot. He concluded that the quality or grade of mineral resources would limit our ability to extract metals. Furthermore, he disagreed with the assumption that, with sufficient energy, it would be possible to extract the metals from low-grade mineral deposits. Georgescu-Roegen rejected the argument that mineral resources are more or less infinite because the planet is composed of minerals, and therefore even our back- yards are potential mines (Georgescu-Roegen 1979). He presented the analogy of a pearl dissolved in acid and spread over the oceans. Even if the energy was available, it would take a practically infinite amount of time to reassemble the pearl (Georgescu-Roegen 1977). The choice of a pearl may not have been the best analogy because the pearl is of organic origin and as long as oysters exist, new pearls will be ‘reassembled’ in fairly short time. Nevertheless, his observation that all mineral and energy discoveries include a substantial proportion of easily accessible resources that cost less to extract than the more inaccessible resources is historically correct (Georgescu-Roegen 1975).

4Flanders and Swann present a humorous explanation of entropy that may be found on the WWW under ‘First and Second Law’. 168 4 The Future for Copper and Coal

Energy became cheaper as the most easily available fossil fuels were exploited and innovations made use of the cheaper energy and, consequently, the ratio of labour to output declined. In concert, capital was directed towards innovations, which delivered the required service at less cost, but that used more non-renewable energy and less human energy to achieve the same result. For example, the sailing ships, which transported wool from Australia to Great Britain, were replaced by coal-ﬁred steam ships that reduced the number of sailors and replaced renewable wind energy with fossil fuel energy. However, Georgescu-Roegen asserted that we could not assume that the cost of producing a unit of metal will always follow a declining trend or that the continuous progress of technology renders accessible resources almost inexhaustible. He questions the Barnett and Morse (1963) vision of energy becoming available in unlimited quantities at constant cost perhaps from nuclear fusion or solar energy, thus enabling ever increasing volumes of resources with ever decreasing grades to be processed at reasonable cost. Ayres and Miller (1980) were not convinced by Georgescu-Roegen’s argument that metals would be made unavailable through mining because of increasing entropy. Instead, they argued that human activity can never change the amounts of each element in the Earth’s crust and the physical distribution of each element can never result in a distribution worse than in the regolith in which every element is present in approximately the same proportions as its average crustal abundance. Answering the question, can the elements be extracted from such a regolith, Ayres and Miller say ‘yes’ providing there is enough available energy. Numerous researchers, e.g. Dunham (1978), Brookins (1990) and Tilton (2003) have estimated the abundance of metal in the Earth’s crust. Table 4.4 draws on these data to show the abundance of three important metals present in the top 300 m of the crust, a depth easily accessible by conventional mining methods. The estimated remaining 752 years of available copper assumes a 1.1% annual increase in production to match the current growth in human population. However, if world population were to continue to increase at its current rate of 1.1% per annum, our population of approximately 7.4 billion would have increased to 27 trillion over the 752 years. The land area available per person would be reduced from the present 20,000 m2 per person to 5.4 m2 per person, which would make open pit mining to 300 m depth difﬁcult. Conversely, if per capita consumption were to increase at 1.1% annually and the population level remained constant, then the per capita consumption would be about 12 tonnes in 752 years time compared to the current global average of 3.1 kg per person.

Table 4.4 Abundance of aluminium, iron and copper Metal in surface 300 Annual production from mines Remaining years of supply metres with no increase annual increase of 1.1% t t years years Aluminium 9.88E+15 39,900,000 247,678,809 1353 Iron 6.76E+15 1,083,243,243 6,240,858 1017 Copper 6.60E+12 19,308,000 342,048 752 4.3 Sustainability Predictions Based on Reserves 169

In 2015, the U.S. per capita consumption was 5.5 kg compared to China at 8.3 kg and Germany at 15.1 kg per person. Perhaps his early years in Australia influenced Brian Skinner to suggest that, while global resources of minerals mined for purposes other than energy are such that exhaustion will not be a constraint on development a century hence, the availability of freshwater may be (Skinner 1993). Continuing this theme, Steen and Gunnar (2002) have shown that it is technically possible to extract metals from granite, granodiorite and basalt rock. These igneous rock types are abundant on the Earth’s surface. The authors assume current technology for extracting the metal and sustainable electricity production from solar radiation and direct heating of oil which, in turn, is used to generate steam for a turbine. The cost of extracting copper was approximately 90 times greater than current production costs. This research sets an upper boundary for metal production costs. While the reserves of fossil fuels are finite, being limited by the original volume of buried organic material, metallic elements are abundant; nevertheless, ore deposits containing high concentrations of metals are scarce. Georgescu-Roegen (1979) argued that the quality of energy was also important. For example, the oceans contain enough energy to support economic activity for millennia; however, for all practical purposes, the energy is presently unavailable for conversion into mechanical work by an engine. While it may be possible to substitute more energy to compensate for lower grade mineral deposits, this increases the consumption of high-quality energy resources such as oil and coal resulting in more rapid depletion of these fossil energy resources. The arguments of Georgescu-Roegen and Steen and Gunnar are the juxtaposi- tions regarding the scarcity or abundance of minerals and energy, respectively, and they help frame the boundaries of this topic that is so important to future human welfare. At a symposium titled the Economics of Exhaustible Resources held in 1974, the theme of optimal depletion of exhaustible resources, especially energy resources, was explored within a framework of intergenerational equity. In his introduction to the symposium, Heal (1974) listed two questions ‘How fast should a resource stock be depleted?’ and ‘Will market-determined depletion rates be acceptable?’. These questions were asked within the context of an equitable balance between the needs of present and future generations. Heal also noted the intellectual heterogeneity surrounding exhaustible resources. Opinions are possibly more diverse today. One paper in particular presented at the symposium, ‘Waiting for the Breeder’,is as topical now as it was in 1974. Manne (1974) disparaged the doomsday prophets who envision depletion of oil and gas reserves by 2004 and uranium 30 years later. He argued that these doomsday prophets did not consider energy from solar or nuclear fusion or nuclear fission via the fast breeder reactor5 as possible alternatives

5The fast breeder reactor produces fissionable plutonium-239 from non-fissionable uranium-238. Non-fissionable uranium-238 is over 100 times more abundant than fissionable U-235. Breeder reactors produce more plutonium-239 than current conventional reactors. Plutonium-239 has a half-life of 24,000 years and can be used to make nuclear weapons. The report ‘The Future of 170 4 The Future for Copper and Coal to oil and gas. Manne presented a case for the U.S. spending up to seven billion dollars on research to have the breeder reactor (nuclear fission) available by 1995. The breeder reactor and fusion reactor6 remain in the experimental stage, and some of the doomsday prophets of today are saying conventional oil production has already peaked (Lewis 2010). Dasgupta and Heal (1974) also presented a paper at the symposium, which investigated the optimal depletion of exhaustible resources. They noted that resource production was dependent on investment in both exploration and technology and achieving any optimal rate of depletion requires a matching optimal rate of investment in resources. In analysing the depletion rate, they assumed a constant level of population, whilst noting the need for an optimal population policy together with an optimal resource consumption and depletion policy. There is a revealing insight into the optimism of the period in the author’s statement that ‘energy experts…have in mind the possibility of discovering a source that will provide … an unlimited flow of energy.’ If such an energy form were found, then this would set the maximum price for energy from conventional sources. The source of that unlimited flow of energy remains to be found.

4.4 Evaluating the Sustainability of Copper and Coal

Whilst in theory all forms of capital are substitutable, the elasticity of substitution7 may be so low that it is impractical. For example, capital can provide humans with some of the resources previously provided by a river system killed by acid mine drainage when the mine is producing sufficient profits to provide capital or if funds are provided from other sources. Technically these resources can continue to be provided when mining is no longer profitable; however, such altruism is unsustainable even for governments of wealthy nations. Once a mine from which acid mine water drains is unprofitable, capital will no longer be provided by that mine. Such capital is as ephemeral as the mining that sustained it and in the long term will not be a substitute for natural capital.

(Footnote 5 continued) Nuclear Power’ (Beckjord 2003) recommended giving priority to the once-through fuel cycle (conventional reactors) rather than the development of more expensive fast breeder reactors. 6The project to build a nuclear fusion reactor, estimated to cost €5bn, was announced in 2005. Construction of the International Thermonuclear Experimental Reactor (ITER) commenced at Cadarache in Southern France in 2008. In 2010, the European Union and six member states reached a deal on the financing and timetable for the experimental nuclear fusion reactor, which was then estimated to cost €16bn (McGrath 2010). Sadly, the first full-power fusion is not expected before 2035 and the current cost estimate is some €19bn (De Clercq 2016). 7Elasticity of substitution measures how easy it is to substitute one product for another. For the precise economic definition see Pigou (1934). 4.4 Evaluating the Sustainability of Copper and Coal 171

Nevertheless, we cannot be morally obligated to do something that is not feasible. The obligation to leave the world as we found it in every detail is not feasible. Ecologically sustainable development (ESD) does, however, obligate us to leave future generations the option or capacity to be as well off as we are and not to satisfy ourselves by impoverishing our successors. The obligation is to improve rather than to reduce the opportunity for future generations to have increasing or at worst non-decreasing well-being over time. Two philosophical questions need to be answered in order for decision makers to act on these obligations. The first relates to the substitutability between the endowments to be left for future generations. Some argue that it is the total level of saving provided for the future generations that is important. They contend that it is possible to substitute human capital and knowledge for natural resources in the production process for goods and services. The second question relates to whose well-being is to be measured. An anthropocentric view implies that quality of life refers to human life. If all life is included, then achieving the ESD goal is a much greater challenge. Preisler and Brown (2003) proposed a more succinct definition of sustainable development than that found in the Brundtland report listed in Table 4.1, submit- ting that to be sustainable, economic activities should not jeopardise future generations’ ability to provide the same or better standard of living as ourselves. Even from an anthropocentric perspective, however, well-being has many attributes. Well-being depends on individual tastes. For people of an urban nature, another hypermarket or shopping centre may produce a definite improvement in well-being. For such an urban dweller, the suggestion of camping in the bush may conjure up fearful visions of snakes, spiders, rampaging fires and flooding streams. Reducing the risk of harm from any of these events would constitute a possible increase in well-being. For this person, human capital and knowledge may be substitutable for the natural environment. Conversely, for those with a deep green persuasion, any shopping centre is one to many and having to live surrounded by concrete significantly decreases well-being. Human cultures are similarly different. The value a community places on environmental attributes is strongly influenced by the culture, wealth and population density of that community. The value any individual places on environmental attributes is often endogenous to the culture in which they have been raised. Nevertheless, tastes change and exposure to new ideas and to the natural environment may change the value a society places on environmental attributes and the preferences of its members. Economists and politicians often assume that degradation of natural resources can be ameliorated by increase in human capital, including ideas. If we require a high rate of substitutability among the endowments to be provided for future generations, we need to first focus on expanding opportunities for the current generation, while understanding that irreparable ecological destruction reduces the option for substitutability in the future. This leads onto the question as to whose well-being is to be measured. Econ- omists have always argued the need to focus on the present rather than the long run for as John Maynard Keynes wrote in 1923 ‘In the long run we’re all dead’. Golding (1972) proposed that it is a natural human instinct to put much greater 172 4 The Future for Copper and Coal emphasis on the well-being of those we can identify with most closely. In economics, this principle is partly recognised by applying a discount rate to future utilities thus placing greater importance on current welfare. Combining the earlier statements, to be sustainable, economic activities should not jeopardise future generations’ ability to provide per capita human well-being or utility at least equal to that of the current generation. This chapter explores the question of metal and energy sustainability, more specifically copper and coal, assuming that an activity is sustainable if it enhances or at least does not decrease human welfare now or in the future within the constraint of ESD. This statement implies that the relative cost of producing copper will not increase. More precisely, this chapter addresses the question ‘will the future cost of producing copper including environmental cost increase or decrease?’ U.S. copper production has been chosen because the data are more readily available than in other countries. The U.S. copper industry has met the cost of complying with environmental regulation since the U.S. created the first national agency for environmental protection the Environmental Protection Agency (EPA) in 1969. The cost of complying with the EPA regulations relating directly to copper mining and coal mining should therefore be reflected in the data sourced for this book. If the cost of producing copper rises, in the absence of a perfect substitute for copper at its future cost of production, and assuming tastes and technology unchanged, then the economic welfare of the future generation will be impaired. Historic energy prices include the cost of pollution regulation; however, fore- casts of future energy price need to take account of the possibility of future events such as increasingly scarce fossil fuel supplies and greenhouse gas mitigation measures. Rapid increases in oil and metal prices like those that immediately preceded the GFC in 2008 have in the past prompted researchers to re-examine mineral and fossil fuel availability. However, in the twenty-first century the literature does not indicate an increased interest in resource availability. Instead, interest and research in the social and geographic sciences have been directed towards the issue of global warming as evidenced by the references to the Stern Review, The Garnaut Climate Change Review and COP 21 in Table 4.1. Many authors have set out to evaluate the hypothesis of resource scarcity proposed by Malthus and Ricardo. Malthus envisaged that available resources had fixed limits and as these limits were approached, so the returns to economic inputs would decrease. Ricardian arguments assume a range in the quality of resources rather than fixed limits. If the resources that require the least effort to extract the desired elements were used first, then there would be a diminishing return to effort as lower quality resources are exploited in the future. Barnett and Morse (1963) examined the inputs to production and the outputs of natural resources in the U.S. from 1870 to 1957 and found that the trend of real mineral prices, which had been level since the last quarter of the nineteenth century, did not support the scarcity hypothesis. They concluded that this was due to technological advances in the extractive process and consequential redefining of the available resources. Chap- man and Roberts (1983) argued that improvements in technology cannot continue 4.4 Evaluating the Sustainability of Copper and Coal 173 to compensate for lower quality resources because technology cannot continue to improve at the rate it has in the recent past. Slade (1982) extended the data used by Barnett and Morse and showed that the prices of most metals and energy resources followed a predictable U-shaped curve. Twenty years on Brown and Wolk (2000) found no such evidence. Berndt and Wood (1975) observed that to their knowledge no empirical study had explicitly investigated cross-substitution possibilities between energy and non-energy inputs and they noted that the lack of such studies was distressingly apparent for anyone trying to understand the implications of increasingly scarce and higher priced energy inputs. They developed the (KLEMS) production function adding in energy (E) and services (S) to the commonly included factors of capital (K), labour (L) and materials (M). Others including Allen (1979), Hannon and Joyce (1981) and Jorgenson (1988) also included energy in their models. The latter author concluded that in order to understand the sources of productivity change it is necessary to disaggregate the sources of economic growth to the sector level and decompose sector output growth between productivity growth and the growth of capital, labour, energy and materials inputs. This chapter takes up that challenge by examining copper production to answer the question ‘how much will a rise in energy costs impact the cost and sustainability of metals production?’

4.5 Sustainability of Coal Production

The Berlin based Energy Watch Group (EWG) was founded in 2006 with the intention of acquiring the necessary scientiﬁc information, independent of outside economic interests, to support the development of a sustainable energy policy. They required this information to support policies that will secure long-term energy supply at affordable prices. Furthermore, they propose that such policies should avoid conﬂicts over energy and ensure effective climatic and environmental protection (EWG 2010). EWG in their estimate of world coal resources and future production raised doubts regarding the quality of coal reserve assessments, e.g. Germany and UK had downgraded proven recoverable reserves by more than 90%. EWG noted that Poland’s hard coal reserves had been reduced by 50% compared to a 1997 estimate. However, the worst example of the poor quality of coal reserves data cited by EWG was the reduction without explanation of proven German hard coal reserves by 99% from 23 billion tonnes to 0.18 billion tonnes in 2004 (EWG 2007). In 2014, Germany produced 186 Mt of coal of which some 178 Mt was lignite. According to Van der Burg and Pickard (2016), the hard coal production of under 10 Mt received a subsidy of $1.5 billion in 2014, which if correct, amounts to a subsidy of over $150/t. Germany is scheduled to phase out hard coal production in 2018 so some of those funds could be for assisting the phase out, rather than being a subsidy to make the hard coal mines competitive with imported coal. 174 4 The Future for Copper and Coal

The U.S. reserves also appear to have been overstated. The USGS 1996 Fact Sheet FS-157-96 (USGS 1996) reported that, in 1994, coal production in the United States reached over 909 million tonnes. The Energy Information Administration estimated that the United States had enough coal to last 250 years. Based on the annual production reported by the USGS, the estimated recoverable reserves would therefore have been approximately 225 billion tonnes. However, the USGS did caution that although coal reserves were large, during coming centuries, coal production rates will likely decline as mining thinner, deeper, and less desirable coal beds become necessary. Nevertheless, the USGS did not anticipate an overall decline in U.S. coal production for many years. In The National Coal Resource Assessment of U.S. reserves, Luppens et al. (2009) estimated that the economically recoverable coal resource was 26 billion tonnes. An article by Selvans (2014) comparing the EIA and USGS estimates suggest that we might expect to see a large downward revision in the EIA estimated U.S. coal reserve numbers in the near future, from some 230 billion tonnes of estimated recoverable reserves to 26 billion tonnes of economically recoverable resources, an almost tenfold reduction. Such reductions in recoverable reserves support the EWG view that world coal reserves data are inaccurate and overstated. BP reports separate reserves for anthracite/bituminous coal, sub-bituminous coal and lignite in million tonnes of oil equivalent. Some 72% of global coal reserves are concentrated in the ﬁve countries listed in Table 4.5,which in descending order of reserves are: USA, Russia, China, Australia and India. The U.S. reserves are listed as 237 billion million tonnes of coal, which is signiﬁcantly higher than the economically recoverable coal resource of 26 billion tonnes reported in The National Coal Resource Assessment. The U.S. Reserves to Production (R/P) ratio would be reduced from 292 years to just 32 years if based on the recoverable coal resource of 26 billion tonnes. The Australian reserves presented in Table 4.5 may also be misleading. Com- bining sub-bituminous and lignite with anthracite and bituminous coal when cal- culating an R/P ratio indicates the reserves will not be exhausted for 158 years.

Table 4.5 2015 R/P ratio for the major coal producing countries (BP 2016b) Reserves-million tonnes Production Reserves to Anthracite and Sub-bituminous Total million tonnes production (R/P) bituminous and lignite ratio U.S. 108,501 128,794 237,295 813 292 Russia 49,088 107,922 157,010 373 421 China 62,200 52,300 114,500 3747 31 Australia 37,100 39,300 76,400 485 158 India 56,100 4500 60,600 677 89 Subtotal 312,989 332,816 645,805 6095 106 %of 77.6 68.2 72.4 77.5 World total World total 403,199 488,332 891,531 7861 113 4.5 Sustainability of Coal Production 175

Although brown coal reserves (lignite) may last for 700 years at current rates of production, it is unlikely that this coal will be mined in the foreseeable future due to environmental constraints. The brown coal reserves are found predominantly in the state of Victoria and the State Government has committed to setting a carbon dioxide (CO2) emissions target of zero by 2050 in legislation (State Government of Victoria 2016). Without Carbon Capture and Storage (CCS) of CO2 emitted, it is unlikely any new brown coal power stations will be constructed in Victoria and the existing brown coal plants will close when they reach the end of their commercial life. The concept of an R/P ratio becomes meaningless if the reserves will not be extracted. Assuming that the estimate of Australian black coal reserves is correct, then at the current production the estimate of 37 Gt of anthracite and bituminous coals will be exhausted in about 60 years; however, this estimate may also prove optimistic as opening new mines in Australia is being contested by environmental activists. The ﬁve largest coal producers account for over 80% of global coal production. China, U.S. and India consume over 60% of the total global coal production. Figure 4.7 shows the remarkable increase in China’s consumption and China alone accounted for 47% of world consumption in 2015. Just as there are exponents of a Peak oil thesis, there is also support for the concept of Peak coal. Figure 4.8, prepared by EWG, indicates that coal production will peak around 2020. This estimate of when the production of coal will peak may not be contradictory to Table 4.5, which shows many more years of production, because Fig. 4.8 indicates only that production will peak, not that production will cease. Nevertheless, ultimately the rate of decline of coal consumption will depend on the cost of mining coal as has been evident in the ﬁrst major coal producing countries, the UK and Germany.

China India US EU Russia World Total

5,000 9,000 4,500 8,000 4,000 7,000 3,500 6,000 3,000 5,000 2,500 4,000 2,000 3,000 1,500 1,000 2,000 World cpmsomption Mt Regional consumption Mt 500 1,000 - - 2004 2007 2008 2010 2013 2014 2015 2001 2002 2009 2011 2000 2003 2005 2006 2012

Fig. 4.7 Coal consumption by region (BP 2016b) 176 4 The Future for Copper and Coal

Fig. 4.8 World coal production and forecast (EWG 2007)

Since this chapter examines the sustainability of copper production in the U.S., the reserves of coal important to this examination are those for the U.S. and even the lowest estimate of reserves indicates more than sufficient energy from coal is available within the time frame of a forecast of copper price out to 2020. As shown in Fig. 4.8, coal production from the Western U.S. increases through to 2035, but at a slower rate than in the past. Both new and existing electric power plants are major sources of additional demand for Western U.S. coal. Low-cost supplies of coal from the Western U.S. satisfy most of the additional fuel needs at coal-fired power plants both west and east of the Mississippi River. Coal production in the interior region primarily supplants more expensive coal from Central Appalachia that is currently consumed at coal-fired power plants in the south-east.

4.6 Sustainability of Copper Production

Our Common Future (the Brundtland Report) stated that ‘sustainable development (SD) is development that meets the needs of the present without compromising the ability of future generations to meet their own needs’ (Brundtland 1987). In the context of the exploitation of exhaustible resources, this goal can be interpreted as requiring that the resource endowment bequeathed to future generations should be sufficient, in combination with their own efforts, for them to maintain a standard of living at least equal to that which we enjoy today. One test of SD, therefore, is 4.6 Sustainability of Copper Production 177 whether metals can be produced in the future at unit costs similar in real terms to current costs. This is a strong test because it takes no account of consumers’ ability to substitute away from materials that experience a rise in relative price. Never- theless, as a test, it is a useful starting point in the analysis of sustainability as it can readily be executed. The unit cost of copper production in the future will depend mainly on the future prices of energy and other inputs, the available reserves and grade of copper ore, and improvements in extraction and processing technology. Golding and Campbell (2014) developed an economic model using price data for copper production in the United States from 1954 to 2002 to predict the unit cost of producing copper in the future. Since energy prices are forecast to rise and ore grades to fall, the copper industry will need to rely on improvements in technology to prevent a rise in the unit cost of production, even if there were no changes in the real prices of other inputs. Golding and Campbell also present an engineering model examining actual and predicted amounts of energy consumed in the production of copper to support the findings flowing from the economic model. The economic model used for the analysis was an industry-level transcendental logarithmic cost function as shown in Table 4.6. The n input prices are those of capital (K), labour (L), energy (E) and materials (M). Ore grade (g) is included in natural log form as a proxy for the price of reserves, and time (t), is included as a proxy for technical change. The engineering model was based on past and current production technology and reveals some important insights, not offered by the economic model, into changes in the required energy input per unit of output in the face of declining ore grades. Furthermore, a comparison of predicted energy input requirements with the actual record can be used to infer the rate of energy-saving due to technical change and adjustments in the levels of other inputs per unit output of the production process. The percentage of copper contained in mill concentrates delivered to smelters varies within a relatively narrow band around 30% copper content and consequently the smelting energy per tonne of concentrate is fairly constant, assuming the mineralogy of the ore does not change. During the period studied, flash smelting

Table 4.6 Transcendental logarithmic cost function Pn Pn Pn ÀÁ 1 lnðÞ¼C=Q b0 þ bi ln Pi þ 2 bijðÞln Pi ln Pj þ btt þ bg ln g i¼1 i¼1 j¼1 The total cost of production is designated by C, Q represents quantity and C/Q is the unit cost β terms are first and second-order coefficients Pi. Denotes the prices of the four inputs K, L, E, M The ore grade variable (g) and time (t) are included in the cost function as first-order terms only, on the assumption that grade and technology changes affect the cost of production without leading to changes in factor proportions 178 4 The Future for Copper and Coal was introduced and the energy consumed in smelting would have decreased; however, the increase in energy required to produce a unit of copper, as grade decreases, is observed predominantly in the mining and milling process. Chapman (1974) extended the insight of the relationship between grade and tonnes mined by estimating the energy required to produce one tonne of copper metal. Assuming the following percentages of copper are recovered at the stages of production: crushing, grinding and floatation—85%; smelting—92.5% and refining—99%, these values generate a recovery factor of 0.778. Chapman applies this recovery factor to show that 1.285 units of copper ore are required in order to produce the amount of copper equivalent to that contained in one unit of ore. The U.S. Congress Office of Technology Assessment (U.S. Congress Office of Technology Assessment 1988) lists the common values for milling and smelter operating parameters as: Refined Grade (99.99%), Refinery Recovery (99%); Blister Grade (98–99%), Smelter Recovery (95–98%); Concentrate Grade (25–40%) and Mill Recovery (75–95%). These recovery factors imply that between 1.418 and 1.085 units of copper ore is required to produce the amount of copper contained in a unit of ore. The figure of 1.285 used by Chapman is close to the middle of this range. Applying the logic developed by Chapman and using the actual U.S. ore grade values for the period, a trend line for the expected energy requirements was calculated. When the actual quantity of energy required to produce copper concentrates is plotted, the energy efficiency gains over the period of the sample are made apparent. As Fig. 4.9 shows, the quantity of energy required to mine, crush and concentrate copper ore to produce one kilogram of refined copper in the United States decreased

Eusa g US energy (Eusa) trendline grade (g) trendline

14.0 1.00

EUSa= -0.057835t + 11.472678 0.90 12.0 R² = 0.287756 0.80 10.0 0.70

8.0 0.60 0.50 6.0 0.40 kWh/kg 4.0 g = -0.128lnt + 0.9469 0.30 R² = 0.8707 0.20

2.0 gradeCopper percent 0.10 0.0 0.00 1954… 1957… 1996… 2002… 1987… 1999… 1963… 1969… 1990… 1960… 1966… 1975… 1981… 1993… 1978… 1984… 1972… Year

Fig. 4.9 Actual energy to concentrate copper in U.S. copper mines 1954–2002 (U.S. Census Bureau 2004; Edelstein 1998) 4.6 Sustainability of Copper Production 179 over the period from 1954 to 2002, despite the falling ore grade that is also shown on the graph. This graph reveals that, contrary to the predicted energy per unit of production increasing as grade decreases, the actual energy used per kilogram of copper in the U.S. (Eusa) decreased in the period studied. Assuming the concentrate grade has remained within the narrow band described earlier, the most likely explanation for the divergence between the expected energy requirements and the actual results reported in Fig. 4.9 is energy-saving technical change over the period. These results along with the econometric model results were used to predict future price trends for copper. The economic model relies on the assumption that the copper market was competitive in the period studied. Agostini (2006) observed that the U.S. copper industry has few sellers, large sunk costs, inelastic demand and supply, and uniform technologies and that this is a type of industry in which one may expect to find evidence of market power. Surprisingly, he found that the results of his modelling were consistent with prices being close to those predicted by a competitive model of the industry. This result may have been more a consequence of foreign rather than domestic competition as U.S. production as a percentage of world production steadily fell from 25% in 1954 to 15% in 2002. Furthermore, U.S. copper imports rose relative to primary production and by 2002 approximately 1.1 Mt of copper was imported compared to 1.4 Mt of primary production from U.S. mines. Pro- viding U.S. manufacturers had the option of purchasing either domestic or imported copper, then a competitive market should have existed and price should reflect unit cost of production. Based on this evidence, the proposition that the U.S. copper market was reasonably competitive between 1954 and 2002 cannot be rejected. Notwithstanding the arguments that the production of conventional oil has peaked and that the production of coal will peak, there appears to be adequate remaining fossil fuel reserves in the U.S. out to and well beyond 2020. The average U.S. price of a barrel of oil was $26.18 in 2002; however, the price increased to a high of $97.98 in 2013 but fell back to $48.66 in 2015. Although the price of a tonne of Wyoming coal increased from $7.02 in 2002 to $15.32 in 2014, it fell to $9.59 in 2016. Distillate is the primary fuel used in mining, and electricity is the primary energy source used in milling. A ratio of 60% electricity and 40% fuel was used in constructing the price of energy in the economic model presented by Golding and Campbell. The prediction in the model of a doubling of the real price of energy between 2002 and 2020 now appears to err on the high side. The coefficients obtained from the model together with the predicted 2020 values of the input prices and ore grade were applied to calculate the increase in unit cost of production. Capital, labour and materials prices were held constant in real terms and only energy price, ore grade and the technology proxy, time, are varied, assuming the observed rate of technology improvement is maintained. The estimated and predicted values of unit cost, C/Q, were then calculated. 180 4 The Future for Copper and Coal

The combined effect of the doubling of the real energy price, the predicted reduction of ore grade and the predicted technical change delivers a 21% reduction in the value of unit cost between 2002 and 2020. Since the forecast increase in the price of energy is less than twofold, the econometric model indicated that the U.S. cost of producing copper is likely to be lower in real terms in 2020 than it was in 2002, provided there are no significant increases in the real prices of other inputs. The engineering model predicted that although ore grade is predicted to fall from 0.45% copper in 2002 to approximately 0.41% in 2020, the increased amount of energy required to counter this decline is more than offset by the estimated improvement in energy-saving technology over the period, and the predicted amount of energy required to produce a kilogram of copper continues to fall up until 2020. Nevertheless, the engineering model indicates that after 2020 the energy required begins to increase as the extra energy required to produce concentrate from lower grade ore outweighs the gains from technology. However, the engineering model was not designed to predict the future cost of copper but rather the amount of energy required to produce a kilogram of copper. The cost of producing a kilogram of copper will depend on producers’ ability to adjust the factor inputs so as to produce copper at the lowest cost. Although the amount of energy in kilowatt hours required to produce a kilogram of copper has decreased over the period 1954–2002, the cost share of energy increased by 11%. The labour share decreased over the period suggesting that energy has substituted for labour throughout the period studied. How producers will adapt to a doubling of energy price, if it were to occur, relative to the price of labour is impossible to judge from the engineering model. As noted earlier, the engineering model does not take account of the factor substitution possibilities available to producers or of the more general technical change occurring beyond the ore beneficiation process. This highlights the advantage of the economic model, which, by taking an all-inclusive view of the technology, enables the incorporation of all technology improvements. While the engineering model cannot be used to forecast future cost, it does answer the important question of the energy requirements for exploiting lower grade ores and goes part of the way in capturing the significant technology improvements that occurred in the period under review. Whilst the analysis by Golding and Campbell may prove true for the U.S., the data presented by the Comision Chilena del Cobre in the previous chapter would indicate that since 2001 the amount of energy consumed per kilogram of copper in Chile has increased as the grade of copper has decreased. Skinner (1993) stated ‘The global resources of minerals mined for purposes other than energy are such that exhaustion will not be a problem a century hence.’ However, an increase in the relative cost of production of a metal would be evidence of increasing scarcity. Tilton and Lagos (2007) concluded that given the many unknowns, there is simply no way to know the real cost of production of copper decades in advance. Although it may not be feasible to model the estimated cost of producing copper in 2093 in order to test Skinner’s thesis that exhaustion of copper will not be a problem a century hence, the two models presented by Golding 4.6 Sustainability of Copper Production 181 and Campbell suggest that it is feasible to estimate the cost of producing copper at the end of this decade. The econometric model predicts that the unit cost of producing copper in the U.S. in 2020 will be below the 2002 cost in relative terms even if there were a twofold increase in the price of energy, holding the prices of other inputs constant in real terms. Time, representing the technology trend, was introduced into the economic model as a natural number, implying that the rate of technical change is constant; however, when time is introduced as the natural log of time, implying a decreasing rate of technological improvement, the increase in energy cost is no longer offset by the technology cost reduction, and a twofold increase in the cost of energy delivers a forecast 5% increase in the relative cost of producing copper in 2020. The results of the engineering model suggest that while the energy required to produce a kilogram of copper will continue to decrease out to 2020, even though ore grade decreases, there will be a modest increase in cost of production as a result of a doubling of energy price. The finding that the unit cost of producing copper in the U.S. in 2020 will not be significantly higher than the 2002 cost in real terms, and may be significantly lower if the past rate of technical change is maintained, is premised on industries adhering to existing environmental regulations; however, new environmental and other tax and regulatory changes may contribute to an increase in the cost of production (Stollery 1985) over and above the level predicted in this paper. As mentioned in Table 4.1, economist Julian Simon and biologist Paul Ehrlich made a wager on the future price of a basket of metals. The wager was based on the premise that the price for the metals was an indicator of sustainability. Ehrlich predicted shortages would result in price increases. Ehrlich and his colleagues picked five metals that they thought would increase in price: chromium, copper, nickel, tin and tungsten. Then they calculated the mass of each metal that could be bought for $200 in 1980. September 29, 1990, 10 years after the bet, was the agreed settlement date. If the inflation-adjusted prices of the various metals rose in the interim, Simon would pay Ehrlich the combined difference and if the prices fell, Ehrlich would pay Simon. The price of each of the metals in the basket of metals fell in real terms and Julian Simon won. Ehrlich sent Simon a check for the basket of metals differential value of $567.07 (Fitzpatrick and Spohn 2009).

Table 4.7 Ehrlich–Simon sustainability wager 35 years on Basket of metals 1980 2015 Ehrlich selected $ (1980 dollars) Chromium 200 177 Copper 200 144 Nickel 200 127 Tin 200 67 Tungsten 200 152 Total 1000 666 182 4 The Future for Copper and Coal

Table 4.7 replicates the wager, showing the 2015 average price for the basket of metals in 1980 dollars. As was the case in 1990, the price of each metal has fallen. Nevertheless, at times during the intervening 35 years, the sustainability wager would have fallen in favour of Ehrlich. The truth of sustainability probably lies somewhere between the cornucopian optimists and the pessimistic Malthusians; however, in the case of metal prices, the result of the wager still supports Simon’s view.

4.7 Conclusions

Any model that assumes increasing human population and matching copper consumption must conclude that copper production is unsustainable. This is an inevitability of compound growth that is often touted by investment gurus. Com- pound money growth is easily destroyed by Governments using methods such as quantitative easing or printing money that reduces the value of the currency. Compound population growth is not so easily addressed. We have set out in this book the history of the development of copper and coal, showing how technology has enabled the production of copper at continually lower real prices. Hopefully that will be the case for generations to come. Nevertheless, the world has experienced catastrophic collapse in the past from wars, famine and plague. The collapse of the Roman Empire and the following Dark Ages resulted in the loss of technology that was most evident in the fall in the production of copper. In Europe copper production did not reach Roman levels until well after the Renaissance. Fortunately, the world was yet to reveal the vast reserves of high-grade copper ores that were unavailable in the Roman Empire. These rich copper deposits along with the provision of what appeared to be almost limitless energy from coal has enabled today’s level of copper production and our current living standard. With most of the habitable globe now fully populated and explored, it is unlikely that there are many large near surface copper deposits yet to be discovered. Many of the world’s richest copper reserves were mined in the previous two centuries. No doubt new resources will be found at depth; however, this will require more sophisticated exploration techniques and expenditure on exploration. The ore grades from available near surface deposits will continually reduce over time as the better quality resources are exploited and, without technological improvements, the energy required to extract copper will increase. Should the world experience a collapse similar to that of the Roman Empire, the remaining inhabitants will not have the luxury of ﬁnding high-grade copper ore deposits that our forefathers had. The answer to whether copper mining is sustainable will ultimately depend on the world’s ﬁnancial and political stability and human population levels. Water may be a constraint on increasing human welfare long before the availability of copper. Desalination plants that require more than three times the amount of energy per 4.7 Conclusions 183 megalitre of water compared to conventional water treatment plants (Cooley and Heberger 2013) may not improve human welfare. In Australia, the state governments of Queensland, NSW, Victoria, South Australia and Western Australia built desalination plants costing over $10 billion during a period of drought. Even though most are now idle they still cost over one billion dollars a year to maintain (Ferguson 2014). Worldwide 338 GW of new coal capacity was in construction in January 2015 compared to 330 GW a year earlier. In addition, 1086 GW was in various stages of planning compared to 1083 GW the year before (Beckman 2016). Nevertheless, as coal becomes more expensive, a cheaper source of energy is needed. Nuclear fusion may one day provide abundant cheap and safe electricity for both copper production and desalination.

References

Agostini CA (2006) Estimating market power in the U.S. copper industry. Rev Industrial Organ 28(1):17–39 Allen EL (1979) Energy and economic growth in the United States. MIT Press, Cambridge (MA) Ayres RU, Miller SM (1980) The role of technological change. J Environ Econ Manag 7:353–371 Barnett HJ, Morse C (1963) Scarcity and growth. John Hopkins University Press, Baltimore Beckjord ES (2003) The future of nuclear power. Massachusetts Institute of Technology, Massachusetts Beckman K (2016) Global coal power: capacity keeps going up, utilisation goes down. Energy Post, Amsterdam Berndt ER, Wood DO (1975) Technology, prices and the derived demand for energy. Rev Econ Stat 57 (3) BP (2015) BP statistical review of world energy 2015 workbook. Oil: Crude oil prices 1861–2014. British Petroleum, London BP (2016a) bp statistical review of world energy. 65 Edition. British Petroleum, London BP (2016b) BP statistical review of world energy June 2016. British Petroleum, London Brookins DG (1990) Mineral and energy deposits. Merrill, New York Brown SPA, Wolk D (2000) Natural resource scarcity and technological change. Econ Financ Rev Q1:2–13 Brundtland HG (ed) (1987) Our common future. Oxford University Press, Oxford Chapman PF (1974) The energy cost of producing copper and aluminium from primary sources. Met Mater 8(2):107–111 Chapman PF, Roberts F (1983) Metal resources and energy. Butterworth & Co (Publishers) Ltd, Thetford Cooley H, Heberger M (2013) Key issues for seawater desalination in California energy and greenhouse gas emissions. Paciﬁc Institute, Oakland Dasgupta P, Heal G (1974) The optimal depletion of exhaustible resources. Rev Econ Stud 41:3–28 (Symposium on the Exhaustible Resources) De Clercq G (2016) ITER nuclear fusion project faces new delay, cost overrun. Science News Deffeyes KS (2001) Hubbert’s peak. Princeton University Press, Princeton Duncan RC (1997) Olduvai theory: sliding towards a post-industrial age. In: Campbell CJ (ed) The coming oil crisis. Multi-science Publishing Co., Ltd., Brentwood, pp 106–108 Duncan RC (2001) World energy production, population growth, and the road to the Olduvai Gorge. Popul Environ 22(5):503–510 Dunham (1978) World supply of no fuel minerals. Resour Policy 4(2):92–99 184 4 The Future for Copper and Coal

Edelstein DL (1998) Salient copper statistics. Copper Statistical Compendium EIA (2016) U.S. ﬁeld production of crude oil U.S. Energy Information Administration, Washington EWG (2007) Coal: resources and future production. Energy Watch Group, Berlin EWG (2010) Energy policy needs objective Information. Energy Watch Group, Berlin Ferguson J (2014) Billions in desalination costs for not a drop of water. The Australian, Sydney Fitzpatrick TM, Spohn K (2009) A 25th anniversary Redux of the Simon and Ehrlich global sustainability wager. J Int Bus Cultural Stud 1 Georgescu-Roegen N (1975) Energy and economic myths. South Econ J 41(3):347–381 Georgescu-Roegen N (1977) The steady state and ecological salvation: a thermodynamic analysis. Bioscience 27(4):266–270 Georgescu-Roegen N (1979) Energy analysis and economic valuation. South Econ J 45(4): 1023–1058 Golding MP (1972) Obligations to future generations. The Monist 56:85–99 Golding B, Campbell HF (2014) Sustainability of metals production: the case of copper. Mineral Econ 26(3) Hannon BM, Joyce J (1981) Energy and technical progress. Energy 6:187–195 Heal G (1974) Introduction to symposium on the economics of exhaustible resources. Rev Econ Stud 41:1–2 (Symposium on the Economics of Exhaustible Resources) Jevons WS (1865) The coal question: an inquiry concerning the progress of the nation, and the probable exhaustion of our coal mines. 3rd reprinted 1965 edn. Augustus M. Kelley, New York Jones N (2015) The UK’s last deep pit coal mines. BBC News Magazine, British Broadcasting Corporation, London Jorgenson DW (1988) Productivity and postwar U.S. economic growth. J Econ Perspect 2(4): 23–41 Lewis B (2010) Peak oil man shifts focus to peak price, demand. Reuters 6 April Lomborg B (1998) The skeptical environmentalist—measuring the real state of the world. Cambridge University Press, Cambridge Luppens JA, Rohrbacher TJ, Osmonson M, Carter MD (eds) (2009) Coal resource availability, recoverability, and economic evaluations in the United States—a summary. The National Coal Resource Assessment Overview. USGS, Reston Malthus T (1798) An essay on the principle of population, as it affects the future improvement of society. J. Johnson, London Manne AS (1974) Waiting for the breeder. Rev Econ Stud 41:46–65 (Symposium on the Economics of Exhaustible Resources) McGrath M (2010) Deal ﬁnalised on fusion reactor. News Science and Environment 29 July Meadows DH, Meadows DI, Randers J, Behrens WW (1972) The limits to growth. Universe Books, New York Meadows DH, Meadows DL, Randers J (1992) Beyond the limits to growth. In Context 32 (Summer 1992) Meadows DH, Randers J, Meadows D (2004) Facing the limits to growth. Chelsea Green Publishing, Brattleboro Pigou AC (1934) The elasticity of substitution. Econ J 44(174):232–241 Preisler F, Brown RPC (2003) Evaluating Sustainability. In: Brown R, Hanahan C (eds) National environment conference 2003, Brisbane, 2003. Environmental Engineering Society, Queens- land Chapter, p 66 Rutledge D (2011) Estimating long-term world coal production with logit and probit transforms. Int J Coal Geol 85(1) Salant SW (1976) Exhaustible resources and industrial structure: a Nash-Cournot approach to the world oil market. J Polit Econ 84(5):1079–1094 Selvans Z (2014) A long time coming: revising U.S. coal reserves. Clean Energy Action, Boulder Simmons MR (2000) Revisiting the limits to growth: could the club of rome have been correct, after all? An Energy White Paper. 30 Sept References 185

Skinner BJ (1993) Finding mineral resources and the consequences of using them: major challenges in the 21st century. In: The Australasian Institute of Mining and Metallurgy Centenary conference, Adelaide, 1993. The Australasian Institute of Mining and Metallurgy, pp 1–8 Slade ME (1982) Trends in natural-resource commodity prices: an analysis of the frequency domain. J Environ Econ Manag 9:138–148 Smith OS (1909) Our mineral resources. Ann Am Acad Polit Soc Sci 33(3):195–201 State Government of Victoria (2016) Victorian Government Response to the Independent Review of the Climate Change Act 2010. Melbourne Steen B, Gunnar B (2002) An estimation of the cost of sustainable production of metal concentrates from the earth’s crust. Ecol Econ 42(3):401–413 Stollery KR (1985) Productivity change in Canadian mining 1957-1979. Appl Econ 17(3): 543–558 Stone J (1996) Empty or full? the debate over the population of Australia. Annual general meeting of the Australian Academy of Science (37) The Shift Project (2016) Historical energy production statistics The World Bank (2016) Commodity prices. Commodity Markets Tilton JE (2003) On borrowed time. Resources for the Future, Washington Tilton JE, Lagos G (2007) Assessing the long-run availability of copper. Resources Policy 32: 19–23 Trainer T (1997) The death of the oil economy. Earth Island 12(2):25 U.S. Census Bureau (2004) 2002 Economic census mining. U.S. Department of Commerce, Washington, DC U.S. Congress Ofﬁce of Technology Assessment (1988) Copper, technology and competitiveness. U.S. Government Printing Ofﬁce, Washington, DC UK Government (2016) Energy trends: June 2016. London USGS (1996) U.S. geological survey energy resource surveys program USGS Fact Sheet FS-157-96. U.S. Geological Survey, Washington DC Van der Burg L, Pickard S (2016) G20 subsidies to oil, gas and coal production: Germany. The Overseas Development Institute, London Glossary

Acid mine drainage Acidic water, usually from the oxidation of pyrite, that drains from areas disturbed by mining Alloy A material composed of two or more metals (or a metal and non-metal) Anode copper The product of fire refining, termed anode because it is the positive terminal in the electrolytic cell for electrorefining Autogenous Arising from within or from a thing itself Blast furnace Furnace used for smelting metals, generally iron, but also others such as lead or copper Blister copper The product of smelting, called “blister” because the residual sulfur and oxygen form bubbles on the surface as the metal cools Brass An alloy of copper and zinc Bronze An alloy of copper and tin Calcine To heat and drive off a gas that is part of a compound; often used for driving off CO2 from limestone (CaCO3) to form lime (CaO). Also the partially oxidized copper produced after roasting Cathode copper The product of electrorefining copper Concentrate A product in which the proportion of ore mineral has been increased significantly above its concentration in ore; copper concentrate is typically produced by crushing, grinding, and separation by flotation Electric arc furnace steelmaking furnace that is used largely with scrap iron or sponge iron Electrometallurgy The branch of process metallurgy dealing with the use of electricity for refining of metals Electrorefining A purification process in which an impure metal anode is dissolved electrochemically in a solution and the pure metal is recovered by electrodeposition on the cathode

Electrowinning Concentration of one or more dissolved ions onto electrodes placed in a solution; basis for hydrometallurgy; commonly referred to as solvent extraction–electrowinning (SX–EW) Flash furnace Smelter in which injections of concentrate, preheated air, and flux are heated rapidly to high temperatures Flux Substance added to lower the melting temperature of a material Flotation Method used to make a concentrate from a slurry of pulverized ore by coating one or more minerals with an organic liquid that causes them to attach to bubbles and float to the surface Gangue The valueless rock or that part of an ore that is not economically desirable but cannot be avoided in mining. It is separated from the ore minerals during concentration Gossan Reddish or ferruginous iron-bearing weathered product overlying a sulfide deposit. Gossan is formed by the oxidation of sulfides and the leaching out of the sulfur and most metals. Derived from the Cornish for blood Hydrometallurgy The separation of a desired metal from an ore or concentrate by dissolution and later precipitation or electrowinning Matte Molten metal sulfide, formed by melting sulfide minerals in the early stages of smelting, usually involving loss of some S as SO2 Ore Rock that contains minerals that can be economically extracted from the Earth Overburden Rock or regolith covering a mineral or ore deposit Oxidation The addition of oxygen to a compound or the removal of an electron from an atom, ion, or element (opposite of reduction) Porphyry An igneous rock of any composition that contains conspicuous phenocrysts (larger grains) in a fine-grained groundmass Porphyry copper deposit A large body of rock, typically porphyry, that contains disseminated chalcopyrite and other sulfide minerals. Such deposits are mined in bulk on a large scale, generally in open pits, for copper and other metals Pyrotechnology Is the intentional use and control of fire by humans. Pyrometallurgy Smelting processes based on thermal decomposition of ore minerals Reduction Chemical process in which valence electrons are added to elements; commonly characterized by the scarcity of free oxygen Reverberatory furnace A furnace that isolates the material being processed from contact with the fuel and any impurities contained therein, but not from contact with the combustion gases. The term reverberation is used here in a generic sense of rebounding or reflecting Glossary 189

Roasting The treatment of ore or concentrates to dry and/or heat the material prior to smelting, and/or to partially oxidize the sulfur content to sulfur dioxide Smelt To separate metal from its ore mineral by pyrometallurgy Smith In this book smith is short for metalsmith: one who separates metal from its ore mineral by pyrometallurgy Solvent extraction (SX) Concentration of a desired element (such as gold) from a primary solution by dissolving it in a smaller volume of a second solution that is mixed with the primary solution but separates from it because it is immiscible; ﬁrst step in solvent extraction–electrowinning (SX–EW) process Steel Iron-based alloy containing up to 2% carbon Index

A Bingham Canyon, 24, 38, 109, 114–116, 122, Acacia trees, 54 125–128 African Copper Belt, 27 Bituminous coal, 16, 17, 30, 33, 174 Agricola, 67, 105 Black smokers, 27, 72 Agricola, Georgio, 157 Block caving, 127, 128 Allende, Salvador, 141 Bloom, 59, 60 American Smelting and Refining Company Bloomery iron smelting, 60 (ASARCO), 117 Bolivia, 38, 137–140, 145 Amygdaloidal, 112 Bonanza Lode, 113 Anaconda Mine, 113, 141 Bone scrapers, 67 Anglesey Penny, 81 Bornite, 21, 23, 24, 27, 122 Anglesite, 71 Boston Consolidated, 116, 117, 120, 128 Anglo-Persian Oil Company (APOC), 134 Bounty, 81 Anhalt-Dessau 1694 silver medal, 76 Boulton, Matthew, 81–83, 91 Annaberg-Buchholz, 63 Bowden, 81 Anode, 74, 126, 144 Breeder reactor, 163, 169, 170 Anthracite, 15–18, 30, 174 Brimstone, 72, 88 Antlerite, 25 British Petroleum Company (BP), 123, 134, Arabah expedition, 50, 53 174 Archaeometallurgy, 37, 38 Broken Hill, 120, 121 Arsenic, 4, 26, 40, 42, 43, 45, 51 Bronze, 4, 7, 10, 13, 37, 43, 45, 51, 56, 59–61, Arsenic bronze, 43, 45 64, 65, 68 Arsenic poisoning, 43 Bronze Age, 4, 43–45, 49, 50, 59, 60, 66–69, Atacama desert, 26, 38, 137, 139, 145 94 Atacamite, 25, 26, 137 Brundtland Report, 160, 171, 176 Atahualpa, 137 Burra, 87, 88 Australia, 1, 12, 32, 84, 87, 88, 104, 121, 124, Busang, 100 128, 132, 135, 158, 168, 169, 174, 175, Butte, 113 183 Australian Wool Corporation Board, 104 C Azurite, 21, 23, 41–43, 52, 57 Caesar, Julia, 59 China, 2, 3, 6, 11–15, 29, 37, 39, 61, 67, 127, B 135, 160, 169, 174, 175 Battle of Cajamarca, 137 Calama, 137, 138, 140, 141, 144, 145 Bell metal, 65 Calamine, 82 Bellows, 42, 49, 52, 56, 58, 137 Calcination, 72, 78 Berndt, 173 Calcined, 39, 78, 82 Bingham and Garfield Railway, 118 Calorific value, 15, 22, 34, 47, 49

Calumet and Helca Mining Company, 113 Copper monopsony, 87 Calumet conglomerate, 113 Copper Mountain, 117, 118, 123, 127 Campbell, Colin, 163 Copperopolis, 37, 78 Capital (K), labour (L), energy (E) and Copper Queen Mine, 113 materials (M), 177 Copper sulfates and chlorides, 137 Captain William Bligh, 81 Copper sulfide, 21, 24, 25, 27, 42, 43, 46, 95, Carbonate ores, 42, 43, 59 119, 123 Carbon capture and storage, 175 Cornwall, 26, 59, 62, 64, 70, 82, 83, 88, 91, Carboniferous Period, 22, 30 101, 120 Car Fork Canyon, 115 Corporacion Nacional del Cobre de Chile Cassiterite, 44 (Codelco), 142 Catal Huyuk, 39 Covellite, 21, 23, 24, 71 Cathodes, 38, 126, 130, 143–145 Cuba, 87, 88, 101 Çayönü Tepesi, 39 Customs act, 86 Cementation, 70, 73, 98, 99, 102, 103 Cut and fill mining, 103 Chalcocite, 21, 23, 24, 27, 57, 113, 114, 122 Cwm Avon copper smelting works, 101 Chalcolithic, 41, 46, 49, 50, 52, 53, 66 Chalcopyrite, 21, 23, 24, 27, 43, 71, 72, 94, D 119 D’Arcy, William Knox, 133 Chaldron, 89 Dardanelles, 144 Chapman, 163, 173, 178 Dark ages, 37, 66, 182 Charcoal, 13, 33, 39, 42, 45–47, 49, 52, 54–56, Darlington, 90 59, 66, 78, 79, 92 Darwin, Charles, 85 Chile, 24, 25, 37, 38, 84, 85, 87, 88, 129–131, Deby, Julien, 100 135, 136, 138, 140, 141, 144, 180 Deep Ecton Mine, 69 Chile’s Northern Electricity Grid (SING), 143 Deffeyes, 163 Chilean copper smelting industry, 87 Delprat, 121 Chile Exploration Company (Chilex), 140 De Re Metallica, 67, 105, 157 Christians, 58 Desalination, 146, 182, 183 Chuquicamata, 24–26, 38, 130, 136, 138, 140, Devon, 59, 62, 64 141, 143, 144 Duke of Cornwall, 64 Claudius, 59 Durham County, 92 Clean air act, 18, 125 Dutchman Mine, 69 Coalbrookdale, 48, 80 Dynamite, 100, 117, 123 Coal grade, 34 Coalification, 22, 30, 31, 34 E Coalification history, 34 Ecologically sustainable development, 106, Coal quality, 34 171 Coal rank, 30 Egypt, 43, 48, 56, 57, 58, 61 Coal type, 22, 33, 34 Egypt Econometric model, 157, 179–181 Coinage duty, 64 Egyptians, 49, 56 Coke, 16–18, 30, 34, 48, 49, 78–80, 92, 101, Ehrlich–Simon sustainability wager, 181 102 Electric light, 108 Colliers, 47, 89 Electric power station, 108 Colliery Gin, 91 Elizabeth I, 69 Columbus’s three ships, Niña, Pinta and Santa Elmore, 119–121 María, 109 Energy consumed per tonne of copper, 146 Compressed air rock drills, 102 Energy required to mine, crush and concentrate Conference of the parties, 160 copper ore, 178 Consecrate pewter chalices, 65 Energy watch group, 173 Copper-bottomed, 5 Engineering model, 177, 180, 181 Copper King, 83 England, 30, 48, 62, 65, 68, 69, 71, 82, 100, Copper Man, 38, 137, 138, 144 104, 108, 133, 163 Index 193

English Copper Co, 86 Hammerstones, 69, 110 Entropy law, 167 Hattusilis III, 61 Environmental impacts of mining, 105 Henry III, 89 Environmental Protection Agency (EPA), 172 Henry IV (Holy Roman Emperor), 65 Erzgebirge (Ore Mountains), 62 Hetton Colliery, 90 Escondida, 24, 38, 136, 143 Hidalgo Smelter, 125 Euramerica, 30, 32–34 Highland Boy Mine, 115, 116 Everson, Carrie, 119 HMS Victory, 70 Hoover, Herbert, 67 F Hoover, Lou Henry, 67 Factor substitution, 180 Hotelling, 160 Falun, 66, 69 Hubbert, 163, 164 Falun Mine, 37, 66 Hubbert’s Peak, 163, 164 Fayalite, 52 Huelva, 37, 93, 106, 107 Fayalite smelting slag, 52 Humic coals, 22, 34 Feynan, 49, 50, 54, 58 Hyacinthe Secrétan, 103, 166 Fired bricks, 39 Hydrometallurgical, 29, 129 Fire setting, 67 Hydrothermal processes, 21, 22 Fishing hooks, 137 Fletcher Christian, 81 I Florence Baptistery, 65 Iberian Peninsula, 92, 109, 157 Flotation, oil and bubbles, 122 Iberian pyrite belt, 29, 106 Fossil fuel energy consumed per person, 164 Immiscible magmas, 29 Fractional crystallisation, 29 Inertinite, 33, 34 Francisco Sanz, 96 In-pit crusher, 123, 124 Froment, 121, 122 International thermonuclear experimental Froth ﬂotation, 109, 118, 122, 123, 144 reactor, 170 Fugger, 65 Iron Age, 55, 59, 60 Iron and manganese mineral deposits, 52 G Iron Bridge, 80 Galena, 27, 71 Ironstone capping, 117 Garnaut Climate Change Review, The, 160, Isle Royale, 110–112 172 Gates of Paradise, 65 J Georgescu-Roegen, 167–169 Jackling, 117, 118, 126 Germany, 5, 11, 12, 27, 37, 65, 66, 69, 79, 131, Jarosite, 94, 95 169, 173, 175 Jevons, 161, 163 Ghiberti, 65 Jinchuan, 29 Glacial period, 111 Glossopteris, 32 K Goltepe, 44, 45, 48 Kadesh, 61 Gondwana, 30, 32–34 Kearsarge Amygdaloid Lode, 113 Gossan, 23, 94, 188 Keelmen, 90 Grasberg, 127 Kellingley Colliery, 163 Great Orme, 66, 68, 69 Kennecott Copper Company, 114 Guadalcanal, 37, 96 Kennecott copper mines, 123 Guggenheim Exploration Co. (Guggenex), 140 Kennicott glacier, 113 Gunnar, 163, 169 Kerosene, 33, 134 Gunpowder, 67, 73, 75–77 Kestel Mine, 44 Keweenaw Peninsula, 27, 110, 112, 114 H King’s Ransom, 65 Hammer and tap, 75, 77, 102 King Amadeo, 93 Hammering and annealing, 112 King Coal, 1, 3, 5, 16, 18, 108 194 Index

King John, 64 Natufian site, 39 Klondike Gold Rush, 113 Navigation laws, 86 Kupferschiefer deposits, 27 Newcastle Roads, 89 Newhouse, 115, 116, 118, 126 L Newton, John, 79 Lady Herbert, 96 Nobel, 100 Levant, 49, 54 Noril’sk, 29 Liebert Wolters, 37, 96 Nuclear fission, 169, 170 Light bulb, 6, 109 Nuclear fusion, 168, 169, 183 Lignite, 16–18, 30, 173, 174 Lime kiln, 39 O Limits to Growth, The, 160–162, 166 Oberbieberstollen Mine, 75 Liptinite, 33, 34 O’Higgins, Bernardo, 141, 142 Lithotypes (vitrain, clarain, durain, fusain), 33 Old Copper Complex, 111, 112 Lomborg, 163, 166 Old Reliable, 114, 127 OPEC, 160, 165 M Open cut, 103, 116, 118, 123 Macerals, 33, 34 Optically stimulated luminescence, 53 Magmatic nickel-copper sulfide deposits, 21 Outokumpu flash smelter, 123, 125 Malachite, 21, 23, 27, 41–43, 45, 46, 52, 53, Oxide and sulfide ores, 22 57, 59 Oyu Tolgoi, 127 Malthus, 157–161, 172 Manilla, 66, 79 P Mansfeld Land, 37, 66, 79 Paratacamite, 27, 53, 57, 137 Margaret Knight, 119 Parque Minero Riotinto, 93 Mark Car, 100 Parys Mountain, 70–73, 75, 80, 82, 83 Marquis de Remisa, 98 Peak coal, 2, 175 Mary Rose, 65 Peak oil, 2, 164, 165, 175 Masjed Soleyman, 134 Peat, 21, 30, 31, 33, 34 Mastaba of Mereruka, 48 Peirce-Smith converters, 123 Matheson, Hugh, 37, 89, 93 Peru, 38, 136, 139, 145 Matte, 101, 103, 107, 125, 188 Pewter, 62, 64, 65 Meadows, 161 Phelps Dodge Company, 113 Medici, Cosimo, 65 Phoenician, 37, 92–94, 106 Medieval or Middle Age, 61 Pig iron, 92, 98, 100, 102 Mersin, 40 Pillar and stall mining, 103 Metalsmith, 21, 42, 189 Pizarro, 136, 137 Mina Invierno, 145 Plimer, 163 Mineral matter, 30, 34 Pocket tin, 64 Minerva, 66 Porphyry copper, 21, 22, 24–26, 29, 117, 118, Mines Royal, 69 123 Ming Dynasty, 72, 73 Potassium nitrate, 73, 140 Mississippi Company, 84 Potter, 41, 42, 121 Model T Ford, 133 Pottery kiln, 39–41, 45 Morgan, J.P., 137 Primary copper deposits, 21, 22 Mount Morgan Mine, 71 Pyrite, 27, 34, 73, 94, 106, 187 Muelle de Minerales, 106 Pyrolysis, 16, 47 Muntz metal, 82 Pyrometallurgical, 129 Pyrotechnology, 38–40, 96, 188 N Nahal Mishmar hoard, 50, 51 Q National Coal Board (NCB), 132 Quality of coal reserve assessments, 173 Native Copper, 21–24, 27, 28, 40, 42, 48, 60, 71, 110–112, 114 Index 195

R Song Dynasty, 61, 73 Radomiro Tomić Mine, 143 South Sea Bubble, 84 Railway, 75, 89–93, 100–102, 106, 107, 114, South Wales coalfield, 92, 108 118 Spain, 7, 29, 37, 58, 62, 83, 88, 89, 92, 93, 95, Recoverable reserves, 173 96, 100, 106 Recovery factor, 178 Spelter, 82 Regulus, 85, 86 Sphalerite, 27, 71, 82 Rekh-Mi-Re, 57 St. Michael’s Mount, 62 Renaissance, 61, 65, 67 Staithes, 90 Reserves to production (R/P) ratio, 174 Stannaries, 64 Reverberatory furnace, 78, 79, 98, 103, 107, Stannary charter, 64 123, 125 Steam engines, 15, 90, 91, 100, 102, 108, 118 Rhondda and Swansea Bay Railway Company, Steam-powered beam engines, 91 101 Steam shovels, 102, 117, 141 Ricardo, 172 Steen, 163, 169 Ricardo 1, 65 Stephenson, George, 90, 91 Rio Tinto, 29, 37, 88, 89, 93, 95–99, 102–107, Stern Review, The, 160, 172 123 Stockton, 90 Rio Tinto Company, 93 Stockton and Darlington Railway, 90 Rio Tinto Mine, 37, 89, 92, 96, 102, 103, 105 Stora Kopparberg Mine, 69 Rio Tinto Mining Park, 93, 105, 106 Sub-bituminous coal, 18, 30, 174 River Tyne, 89 Subsidy, 133, 173 Roasting, 46, 47, 92, 72, 73, 78, 98, 99, 102 Sudbury deposit, 29 Roose, Jonathan, 71 Sulfur, 8, 17, 18, 34, 47, 49, 72, 73, 88, 100, Roman Empire, 37, 58, 61, 66, 67, 92, 95 103, 126, 187–189 Royal Navy, 81, 133, 134 Sulfur from cupreous pyrites, 88 Sulfuric acid, 19, 29, 121, 125, 126, 129 S Sulis Minerva, 66 Sacrificial anodes, 74 Sulman, 122 Saltpetre War, 139 Sumerians invented writing around 3300 BC, Samuel Tiquet, 96 38 Sapropelic coals, 22, 34 Supergene copper ore, 25 Scrap iron, 74, 187 Swansea, 37, 71, 78, 79, 83, 85–89, 92, 101, Seafloor hydrothermal vents, 21, 27 110 Secondary enrichment processes, 21, 22 Sweden, 37, 66, 69 Second Law of Thermodynamics, 167 Sediment-hosted copper deposits, 22, 27 T Shaft and gallery mining, 52 Tanfield Railway, 89 Shale oil, 163 Tangalooma Whaling Station, 134 Sheathing, 1, 4, 6, 70, 74, 81, 108 Taylor, Thomas Griffith, 158 Silver, 10, 13, 27, 64, 75, 76, 84, 94, 95, 115, Telera, 98, 102 120, 126, 136 Teniente, 24, 140, 142 Skarn, vein and replacement deposits, 21, 22, Teredo worm, 70 24, 26 Terrazzo floor, 39 Skinner, 163, 169, 180 Terreros, 102 Slaves, 66, 94 Tetrahedrite-tennantite, 21, 24 Slave trade, 79 Tharsis, 29, 88, 89, 100, 109 Smith, 42, 43, 46, 49, 51, 59, 75, 101, 123, Timber, 13, 47, 66, 77, 87, 95, 98, 102, 105, 141, 189 111 Sodium nitrate, 137, 139 Timna, 27, 49, 50, 52–55, 57, 58, 94, 137 Solar energy, 168 Timna valley, 27, 50 Solinus, 66 Tin, 4, 10, 26, 40, 42–45, 57, 59, 60, 62–64, Solvent extraction-electrowinning process or 92, 160, 187 SX-EW, 129 Tin bronze, 44, 45, 57 196 Index

Tiwanaku culture, 137, 138 Welsh coal, 15, 37, 92, 101, 108 Toberas, 137 Welsh process, 78, 79, 86 Transcendental logarithmic cost function, 177 Whales, 134 Treaty of La Pointe, 112 Whimsey, 74, 75 Tuyeres, 42, 48, 49, 57 White Pine copper deposit, 27 White Pine Mine, 114 U William Cecil, 69 Untermeyer, 118 Williams, Thomas, 5, 71, 83 USA, 28–30, 32, 174 William Wilberforce, 79 Wire ropes, 108 V Wood, 13, 30, 39, 46–48, 64, 77, 78, 98, 102, Vickers hardness, 9, 10, 59 105, 108 Vitrinite, 33, 34 World’s largest open-pit copper mine, 141 Voisey’s Bay, 29 World population, copper and coal Volatile matter, 16, 17, 30 consumption, 159 Volcanic-hosted sulﬁde deposits, 22 World War I, 141 Wrought iron, 9, 59, 107 W Wales, 15, 30, 67, 69–71, 74, 78–80, 88, 92, X 101, 102, 108, 115, 119, 120, 123 Xerxes, 144 War of the Paciﬁc, 138 Water-powered bellows, 49 Y Waterwheels, 94 Yanfeng Tongbao copper coins, 61 Watt-Boulton partnership engine, 82 Yarim Tepe, 39, 107, 173 Watt, James, 82, 91 Young, Brigham, 114, 115 Weindl, Caspar, 75