University
stringclasses
19 values
Text
stringlengths
458
20.7k
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry and environmental groups. The climate in the region is in general sunny and mild all year around with summer temperatures ranging between 15 and 30°C (Infomine: Palabora 2005). The Palabora is an underground mine. The recovery process is made up of concentrator, a copper smelter and refinery. Copper anodes from the smelter, with a purity of 99.5%, are treated with electrolysis to form 99.99% pure copper cathodes. From the electrolytic refining, precious metals are removed from the copper, which can be retained and sold to refineries (Infomine: Palabora 2005). Palabora was one of the first mining companies in South Africa that implemented ISO 14001 environmental management system, which they did as early as 1998 (Infomine: Palabora 2005). 7.3 Gold Mines As in the case of the copper mines, a number of gold mines, in regions near the uranium mines, have been selected for the comparative study. 7.3.1 Goldstrike The Goldstrike property consists of an open pit and two underground gold mines located in Nevada, USA. The site is owned by Barrick Gold Corporation and the property was first mined in 1976. The climate of the area, which is located at an altitude of 1700 metres, is relatively dry all year round with warm summers and cold winter (Infomine: Goldstrike 2006). The gold, which is found in sulphide ores at an average grade of 0.143oz/ton (4 ppm) is processed into doré at site and shipped to an outside refinery for processing into gold bullion (Infomine: Goldstrike 2006). Production in 2007 amounted to 1.63 million ounces (Barrick: Global Operations - Goldstrike Property 2007), and approximately 1800 people work on site (Infomine: Goldstrike 2006). Goldstrike has two processing facilities: a roaster to treat carbonaceous ore and one autoclave installation to treat non-carbonaceous sulphide ore (Infomine: Goldstrike 2006). 7.3.2 Kalgoorlie Super Pit – Fimiston Operations The Kalgoorlie operations are situated about 600 km east of Perth in Western Australia, at the south- east corner of the city Kalgoorlie-Boulder (KCGM 2008). It is an equally shared joint venture between Barrick gold of Australia and Newmont Australia Ltd, and consists of the Fimiston open pit (more commonly called the Super Pit), Mt Charlotte underground mine, Fimiston Mill and the Gidji Roaster. On behalf of the owners, the Kalgoorlie operations are managed by the company Kalgoorlie Consolidated Gold mines, KCGM. KCGM produces approximately 600 000 ounces of gold every year, which makes it the largest gold producing company in Australia (KCGM 2008) and most of it is mined at the Super Pit (Infomine: Kalgoorlie 2006). The Super Pit will, when completed, be 3.5km long, 1.5 km wide and as much as 640 meters deep at some places (KCGM 2008). The Kalgoorlie area has been mined since 1893 and KCGM has proven reserves that are calculated to last until 2013 (Infomine: Kalgoorlie 2006). Kalgoorlie is situated in the Western Australian desert. The climate in the area is dry, with an average and evenly distributed rainfall of 260 mm. The summers are hot and the winters are cool (Infomine: Kalgoorlie 2006). 35
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry The goldfield in Kalgoorlie is made up of several isolated deposits of sulphide ore, of which the Gold Mile is the largest deposit, accounting for about 90 percent of the gold production from the field. The mineral grade of the ore mined was 2007 2.01 g/t4. Ore is treated at the Fimiston Mill. The resulting sulphide concentrate is transported 20 km to the Gidji roaster where, after treatment, the golden-laden carbon is shipped once again back to the mill for processing. Finally, doré is shipped for refinement into gold bullion in Perth (Infomine: Kalgoorlie 2006). There are no records of an implemented management system. However, according to the report Barrick Responsibility – 2007 Environmental, Health, Safety and Social Performance (2008), the company is working with the implementation of an environmental management system at all their operations. 7.3.3 Navachab Navachab is situated on the southwest coast of Namibia, 10 km southwest of Karibib and 170 km northwest of Windhoek. The operation is fully owned by AngloGold Ashanti and has been in production since 1989. The Navachab gold deposit consists of greenschist-amphibolite facies, calc- silicate, marbles and volcano-clastics. The gold grains are very fine and associated with pyrrhotite and some traces of pyrite, chalcopyrite, maldonite and bismuthinite (AngloGold Ashanti: Country Report Namibia 07 2008). The Navachab operation includes an open pit mine and its processing plant. The processing plants consist of mills, carbon-in-pulp and electro-winning facilities (AngloGold Ashanti: Country Report Namibia 07 2008). All of AngloGold Ashanti’s operations have ISO 14001-based Environmental Management System in place, with external auditing and certification (AngloGold Ashanti: Country Report Namibia 07 2008). 7.3.4 Vaal River The Vaal River operations are fully owned by AngloGold Ashanti and include the four underground mines: Great Noligwa, Kopanang, Tao Lekoa and Moab Khotsong. The operations are located near the towns of Klerksdorp and Orkney on the border of the North West and Free State province and the gold occurs in three different reefs, namely Vaal Reef, the Ventersdorp Contact Reef (VCR) and the Crystalkop Reef. The Vaal River complex also includes four gold plants, one uranium plant and one sulphuric acid plant. The uranium diuranate, or yellow cake, produced at the operations is a bi- product to the gold production5. All together the operations produced 35.3 tonnes of gold in 2007 and the ore grade ranges between 3.62 and 7.94 g/tonnes (ppm) (AngloGold Ashanti: Country Report South Africa Vaal River Operations 07 2008). 4 Louise Crogan, Communty Relations, Kalgoorlie Consolidated Gold Mines, personal communication (e-mail) 2008-08-08 5 Tony Da Cruz, Environmental Manager, Anglogold Ashanti Ltd, personal contact (e-mail) 2008-07-17 36
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 8 Companies Operating the Mines The following companies are responsible for the operation of the mines described in chapter seven. 8.1 AngloGold Ashanti AngloGold Ashanti, with headquarters in Johannesburg, South Africa, has operations situated in South Africa, Australia, Ghana, Mali, Brazil, Tanzania, the United States, Guinea, Argentina and Namibia. The company produced 5.5 Moz gold in 2007 employed the same year approximately 62000 people, including contractors (AngloGold Ashanti: Country Report Namibia 07 2008). 8.2 Areva Areva has manufacturing facilities in 43 countries and over 65000 employees (Areva: In brief 2008). It has operations in all steps of the nuclear energy life cycle from the mining of the uranium to treatment and recycling of used fuel, including reactor construction and electricity transmission (Areva: Operations 2008). 8.3 Barrick Gold Corporation Barrick Gold Corporation is the leading gold mining company of the world with 27 operating mines and 10 development projects in five continents. The company was founded in 1983 (Barrick: History 2007). 8.4 BHP Billiton BHP Billiton is the world’s largest diversified resource company with 41000 employees working in 100 operations in 25 different countries. Commodity businesses included in the company’s operations are aluminium, energy coal and metallurgical coal, copper, manganese, iron ore, uranium, nickel, silver and titanium minerals, and it has interests in oil, gas, liquefied natural gas and diamonds. Its global headquarters are located in Melbourne with major offices in London (BHP Billiton: Company Profile 2008). 8.5 Cameco Corporation Cameco is the world’s largest uranium producer accounting for 19% of global production. It has four operating mines located in Canada and in the USA. Furthermore, Cameco is one of three uranium conversion suppliers in the western world and it also owns or controls 35% of the western world’s uraniumhexaflouride production capacity (Cameco: Company Profile – Overview 2007). Cameco was formed in 1988 by the merger of Saskatchewan Mining Development Corporation and Eldorado Nuclear Limited (Cameco: Company Profile – History 2007). Cameco is also involved in the Gold mining industry by owning 53% of the shares in Canadian based gold mining and exploration company Centerra Gold (Cameco: Other Investments 2008). 8.6 Heathgate Resources Pty Ltd Heathgate Resources, with its head office in Adelaide, South Australia, is an affiliate of General Atomics. It was formed in 1990 when the former owners offered the Beverley mine for sale (Heathgate Resources: Who is Heathgate 2006). A workforce of about 150 people works at the Beverley mine (Heathgate Resources 2008). 37
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 8.7 Kennecott Utah Copper Kennecott Utah Copper Corporation is a subsidiary company to Rio Tinto that operates the mine, the smelter and the concentrator at Bingham Canyon. Kennecott Utah Copper became the sole operator of the property year 1936 (Rio Tinto: Kennecott Utah Copper 2008) and since 1959, it has been a fully integrated copper producer with all four production steps; mining, concentration, smelting and refining (Rio Tinto: Kennecott Mine in Utah History 2008). 8.8 Rio Tinto Rio Tinto is one of the world’s leading mining and exploration companies. Their products include aluminium, copper, diamonds, energy products, gold, industrial minerals and iron ore (Rio Tinto: Who we are 2008) Rio Tinto is a combination of the companies Rio Tinto plc from UK and Australian Rio Tinto Limited. The British branch was founded already in 1873 (Rio Tinto: Timeline 2008). Rio Tinto operations are found in all five continents (Rio Tinto: Our Operations 2008). 8.9 Xstrata Plc Xstrata is a global mining group with headquarters in Zug, Switzerland, and operations and projects in 18 countries around the world. The company employs 56000 people, including contractors and had 2007 sales of 28542 million USD. Xstrata is active in several mineral markets, including e.g. copper, coal, ferrochrome and nickel (Xstrata: Mount Isa Mine Sustainability Report 2007 2008). 38
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry the product compared to the total operating cost. Operating costs of the different process steps in copper cathode production are described in Davenport et al (2002). Prices: USD/tonne Copper – concentrate 1 285 Copper – anode 6 393 Copper – cathode 7 139 Gold 22 374 055 Uranium - U O 99 206 3 8 Tab. 3: Mineral prices used in this study. 9.1 Impact Category (parameter) Definition The impact categories finally used for this study are based on a selection of parameters for which comparable data was available. It includes resources, human health and ecological consequences. The data is presented in the following parameters for all sites: • Energy usage • Water usage • Greenhouse gas emissions • Emissions to air of sulphur dioxide, nitrogen dioxide and particulate matter • Injury frequency rate For operations located in Australia and the US, the extensive data available in the NPI and TRI databases respectively allow for an in depth analysis. The data from NPI and TRI have been classified into emissions to air, water and soil and characterised with Life Cycle Impact Assessment (LCIA) characterisation factors from Guinée (ed.) (2002). This was performed in order to reduce the number of parameters to simplify comparison. The characterisation resulted in the following additional parameters used in the analysis: • Human toxicity potential • Freshwater aquatic ecotoxicity potential • Freshwater sediment ecotoxicity potential • Terrestrial ecotoxicity potential • Photo-oxidant formation potential • Acidification potential • Eutrophication potential All emissions to water were treated as if the recipient is a freshwater ecosystem and all emissions to soil were treated as emissions to agricultural soil. 9.2 Characterisation Characterisation is a quantitative way of translating environmental loads into a specified environmental impact using pre defined characterisation factors for all emissions. After characterisation, all emissions giving rise to a certain impact are expressed as one equivalent emission of a specified unit. With knowledge from scientific methods within environmental chemistry, toxicology, ecology etc, characterisation combines physicochemical properties of the pollutants and modelling of how pollutants react in the environment in order to assess the impact of 40
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry methodology was originally developed by researchers in the radioactive waste management area. The aim of the performance assessment method is to systematically study all relevant processes and mechanisms influencing release rates from mining waste (Allard & Herbert eds. 2006). In performance assessment, a number of conceptual and mathematical models are used to describe and quantify the main processes that govern the release and transport of different contaminants, as well as chemical and biochemical reactions from source to final recipient. For evaluations of long- term effects of different processes, studies of natural counterparts can be used as a support. The analysis presented in this report does not include any calculations due to limited amounts of data available but the focus will instead be on general discussions. The enormous amounts of waste and tailings generated by the mining industry contain toxic materials that are connected to a number of environmental issues as well as health hazards. There is little specific data available within this field and the studies that have been performed do not cover a long enough period of time to allow for a long-term analysis. Since samples never or rarely exist from before the mining operations started, it is often extremely difficult to separate what is caused by mining operations from what is caused by geological and environmental conditions in the region.7 Consequently, in a very simplified form, the performance assessment methodology has been used to discuss a number of parameters that are of interest and consequences of changes within these parameters have been compared. Since the ore at Beverley Mine is extracted with in situ leaching, it is not included in the following analysis. Six parameters were mentioned in chapter 5.3 as most relevant for tailings related issues and processes: • The ore: source term and water chemistry • Age of the operations • Tailings management • Metals and substances related to the different minerals and ores • Process chemicals and complexes • Geography: precipitation and hydrology Since this study focuses on the current performance of different operations, the age of the operations and tailings will not be treated in the comparative discussion. The same goes for tailings management. Factors affecting the success of tailings management facilities are several and of interest but due to lack of data and inability to assess the performance of the different operations, this area will not be included. The type of data required to perform a comparison is mostly unavailable. Therefore, the following comparison will only deal with some of the parameters of importance. The parameters that are discussed are amounts of tailings generated, the type of ore, the metals usually connected to the specific type of ore, process chemicals and water conditions (mainly in the form of precipitation). All parameters are directly related to the specific operations, except contained metals and substances that instead are derived from the nature of the source term. 7Bert Allard, Professor in Chemistry, Örebro University, meeting at Örebro University November 3, 2008 42
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 10 Results and Analysis This part will present the results from the different investigated parameters. When no result is displayed for a specific mine, the information is not available and it does not contribute to the average values displayed at the right hand side of the graphs. 10.1 Energy Use Fig. 5 shows energy usage at each site expressed in GJ per million US dollar of product. Where such data is available, the energy usage is separated into direct energy and indirect energy. Direct energy refers to all energy consumed at site except electric energy. Indirect energy refers to electric energy consumed at site. For Beverley and Ranger mines, indirect energy constitutes the total primary energy in the natural gas required for on-site electricity production. Consequently, the electrical energy consumed is considerably lower than the indirect energy indicated on the graph. Energy use 10 000 8 000 6 000 4 000 2 000 0 Total Direct Indirect Fig. 5: Energy usage in GJ per MUSD of product. The graph clearly shows that gold mines generally consume more energy per economic value of product produced. This could be explained by the very low grade of gold in the ore. This requires larger amounts of ore to be extracted in order to produce a given amount of mineral and consequently more energy is needed in the process. Uranium mines have the lowest average energy consumption per million US dollar of product. 10.2 Greenhouse Gas Emissions Emissions of greenhouse gases from the investigated sites follow the same general trend as the energy usage and considering average greenhouse gas emissions per million US dollar of product, gold mines have the largest emissions and uranium mines the lowest emissions as can be observed in Fig. 6. When investigating relative individual trends however, there are some deviations from the reported total energy usage. This can result from differences in the way emissions from electricity 43 tcudorp fo DSUM rep JG
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry production are accounted for or due to different electricity production mixes. Although this study aimed to exclude variations from national or regional electricity production mixes, such a distinction was impossible to make due the fact that the sources of the emissions were not specified at all sites. GWP100 2 000 1 500 1 000 500 0 Total Direct Indirect Fig. 6: Global warming potential in tonnes CO -equivalents per MUSD of product. 2 10.3 Water Usage The average water usage, illustrated in Fig. 7, shows the same trend as for both energy usage and greenhouse gas emissions with gold mines having the highest and uranium mines having the lowest average water consumption per million US dollar of product. Water usage trends may be explained to a large extent by regional availability. Rabbit Lake uranium mine for example, located among several lakes in the north of Canada, have the highest consumption of water. However, this figure includes inflow to the mine or precipitation that has to be treated before it is released. While water may represent a problem due to its abundance in one site, it may be a problem due to its scarcity at other sites. It should be noted that the gold mines, having the largest water use, all operates in dry climates where water is a scarcity. 44 tcudorp fo DSUM rep .qe-2OC sennot
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Water usage 25 000 20 000 15 000 10 000 5 000 0 Fig. 7: Water usage in m3 per MUSD of product. 10.4 Emissions to Air Fig. 8 shows emissions to air of sulphur dioxide, nitrogen oxides and particulate matter. Some copper mines have emissions of sulphur dioxide that are several factors higher than for other mines. This is explained by the fact that the copper is contained in sulphide ores giving rise to large emissions of sulphur dioxide when the ore is roasted and smelted. The emissions of nitrogen oxides and particulate matter follow earlier trends connected to energy consumption and carbon dioxide emissions, i.e. gold have the highest emissions, followed by copper and last uranium mining. Combustion engines, power production and explosives are examples of sources to nitrogen oxides and particulate matter and are all closely connected to resource consumption that can be explained by a low ore grade. Hence, even if the data of nitrogen oxide emissions and particulate matter from gold mining is only available from KCGM and thus constitute the average value, these results appear reasonable. 45 tcudorp fo DSUM rep 3m
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Injury Frequency Rate per 200 000h worked 3,5 3,0 2,5 2,0 1,5 1,0 0,5 0,0 LTIFR (Lost Time Injury Frequency Rate) FIFR (Fatal Injury Frequency Rate) Fig. 9: Lost time, and fatal injury frequency rate per 200000h worked. The highest value is surprisingly from Beverly mine, which is not a traditional mine but an in situ operation. As well, the range between the highest lost time injury frequency rates of different mines producing the same mineral is large for all minerals. The extent to which the LTIFR figures reflect the dangers of the work at the mine or is a subject to corporate policies or national legislation remains unknown. 10.6 Human Toxicity Potential Human toxicity, as expressed in kg of 1,4-dichlorobenzene (DCB) equivalents in Fig. 10, shows very high levels for uranium mines. This is explained by the large emissions of polyaromatic hydrocarbons from the Olympic Dam operations. It should be noted that Olympic Dam is a multi mineral mine treated as a uranium mine in this study. 20% of the sales originate from uranium production and hence, 20% of the emissions are allocated to the uranium production. Gold mines show the lowest levels. The very low levels reported by the Goldstrike and Kennecott operations can be explained by varying reporting regulations between the NPI and TRI systems in Australia and the United States, as well. 47
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry mines is the dose reported in the “Groups of people receiving the highest dose”-column in Tab. 4. Thus, doses reported in that column are not directly comparable since the figures for designated workers include a larger group of people. Average figures sometimes also include office workers, which significantly lower the average. These reported figures can be compared with other figures of effective occupational doses as in Tab. 5, below and with the figure of worldwide annual average background sources of 2.4 mSv (UNSCEAR 2000). Source/ practice Number of Average monitored annual workers effective (thousands) dose (mSv) Man made Nuclear fuel cycle 800 1.8 sources (incl. uranium) Industrial uses of 700 0.5 radiation Defence activities 420 0.2 Medical uses of 2 320 0.3 radiation Education/veterinary 360 0.1 Total man made 4 600 0.6 sources Enhanced Air travel (crew) 250 3 natural sources Mining (other than 760 2.7 coal) Coal mining 3 910 0.7 Mineral processing 300 1 Above ground 1 250 4.8 workplaces (radon) Total natural 6 500 1.8 sources Tab. 5: Background radiation sources (adopted from UNSCEAR (2000)) With a few exceptions, the reported doses are in comparison relatively low, especially doses to the public. Relating this with epidemiological studies on groups exposed to ionizing radiation we can only attempt to assess the health impacts from uranium mining activities. Such studies have estimated that people exposed to a total dose of 100 mSv have a 10 percent higher risk of dying of non leukaemia cancer and a 19 percent higher risk of dying of leukaemia than people not exposed to that dose (Analysgruppen vid KSU 2006). Consequently, in the worst case scenario, if the same person is exposed to doses in the region of the maximum reported individual dose each year, the worker would be exposed to a total dose of 100 mSv after ten years. This hypothetical dose of 100 mSv is then achieved during a period of time that is the double of those recommendations made by ICRP, 54
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry which recommends that the total dose shall not exceed 100 mSv over a period of five years (ICRP 1991). 10.12 Tailings A general discussion of the tailings dilemma connected to the different operations is presented below. By using a simplified reasoning, the amounts of tailings generated, the type of ore, the metals usually connected to the specific type of ore, process chemicals and water conditions (mainly in the form of precipitation) will be compared. 10.12.1 Waste Quantities Generated Not all mining companies report the amount of mine waste generated. The waste was therefore calculated using the production quantities and ore grades. The result is visualized in Fig. 17. As expected, gold mines produce the most waste owing to the low ore grade. The gold grade of the mines in this study ranges between 0.00016% at Navachab to 0.00066% at Vaal River operations. If not considering McArthur’s remarkably high ore grade, the copper mines have the highest average ore grade of the compared objects, ranging from 0.54 to 3.42%. The variation in mineral grade for the uranium mines in this report is extremely large. Rössing, with the lowest ore grade, has a grade of 0.028%, while the figure of MacArthur is 19 %. However, the ore grade at McArthur is somewhat remarkable, since the remaining mines have an ore grade between 0.028-0.69%. The extremely high ore grade at McArthur is the reason why the uranium mines produce less waste than copper operations. Amount of Tailings 40 000 35 000 30 000 25 000 20 000 15 000 10 000 5 000 0 Fig. 17: Amounts of tailings in tonnes per MUSD of product. 55 tcudorp fo DSUM rep sennoT
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 10.12.2 Performance Assessment The parameters chosen for the analysis are specified for each operation in Tab. 6. Operations Mineral Ore Waste Main ore Water assets Main leach mined grade quantities minerals and chemical (%) (t/MUSD) precipitation Kennecott copper 0.54 15 000 sulphides High (flotation chemicals) Mount Isa copper 3.42 2 200 sulphides, Low (flotation (carbonates) chemicals) Northparkes copper 1.4 16 000 sulphides Average (flotation chemicals) Palabora copper 0.7 4 500 sulphides, Average (flotation (oxides, chemicals) phosphates, silicates) Goldstrike gold 0.0004 11 000 sulphides Low cyanide KCGM gold 0.0002 22 000 sulphides Low cyanide Navachab gold 0.00016 29 000 schist, Low cyanide sulphides Vaal River gold 0.00066 6 300 sulphides Low cyanide McArthur- uranium 19.05 43 oxides High acid Key Lake McClean Lake Uranium 0.45 1 900 ? High sulphuric acid Olympic Dam uranium 0.07 2 800 oxides, Low sulphuric sulphides acid Rabbit Lake uranium 0.69 1 450 oxides, High sulphuric (silicates) acid Ranger uranium 0.12 8 400 schist, High sulphuric silicified acid carbonates Rössing uranium 0.0289 35 000 oxides, Low sulphuric silicates acid Tab. 6: Summary of Performance Assessment parameters. The most common combinations of parameters for each mineral are presented in Tab. 7 to serve as a base for the following reasoning. The column for the type of mineral describes the minerals most commonly associated to Au, Cu and U in the operations investigated. The sulphide minerals at Olympic Dam ore will also be discussed; due to the strong connection between sulphide minerals and acid mine drainage. 56
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Mineral mined Average Type of Water Contained Process Environmental annual ore abundance metals and chemicals impact tailings minerals substances quantities (t/MUSD) - Rel. low AMD potential. Gold 17 000 sulphide low Cyanide - Cyanide - Heavy Metals As, Cd, Pb, Zn - Rel. low AMD low etc potential (flotation Copper 9 300 sulphide chemicals) - High AMD high potential - Heavy metals As, Cd, Pb, Zn, - Rel. low AMD, sulphide low U, Ra, Th etc -Radiation - No AMD low Sulphuric potential, Uranium 8 200 acid - Radiation oxide U, Ra, Th etc - No AMD high potential, - Radiation Tab. 7: General Performance Assessment approach. 10.12.2.1 Copper Copper is exclusively extracted from sulphide ores and the precipitation is similar to Scandinavian conditions, except for Mount Isa. Thus, there is a great risk of acid mine drainage and related releases of heavy metals. Metals commonly present in sulphide ores are primarily arsenic, cadmium, copper, iron and zinc. With less precipitation and as well as buffering carbonates present in the ore, the acid mine drainage potential of the mine waste from Mount Isa Mine, should be lower than for the other copper mines. The Palabora ore contains silicates that might also have buffering capacity that can mitigate acid mine drainage. 10.12.2.2 Gold Similarly to copper, sulphide minerals host most of the gold of this study and it is connected to the same heavy metals as the copper ores. The acid mine drainage connected to the investigated gold mining operations is expected to be of lower gravity due to the relatively dry climate at all sites. A greater problem is instead the use of cyanide. The free form of cyanide will decay in the tailings but WAD complexes with metals common in sulphide minerals, e.g. copper, cadmium and zinc, can be released, poising potentially severe environmental threats. In dry areas, cyanide in tailings is more problematic than otherwise because animals are more attracted to the water present in the tailings management facility. 57
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 10.12.2.3 Uranium Uranium is generally mined from oxide ores. At Olympic Dam, the uranium is extracted from sulphide ore and at Ranger, schist is the mineral connected to the uranium mined. A majority of the mines are located in areas with a good access to water but for example Rössing in Namibia operates in an arid area. Besides uranium, radioactive elements, such as radon and radium, can be hazardous to both environment and humans, which has been treated in chapter 5. Oxide minerals are neutral in water, hence problems of acid mine drainage are not applicable in the same way for uranium. The water conditions are therefore not as crucial in this case. Nevertheless, radiological concerns connected to tailings issues remain a problem. At Olympic Dam, sulphide ores are mined. Hence, there is a risk of acid mine drainage and the subsequent leaching and transportation of heavy metals. However, the risk is somewhat decreased due to the dry climate. 10.13 Summarised Results The two following graphs (Fig. 18 & Fig. 19) summarise the average results for most of the parameters in this investigation. Summarised results 2007 per MUSD of product 20 000 18 000 16 000 14 000 12 000 10 000 8 000 6 000 4 000 2 000 0 Energy use Water use Tailings CO2 SO2 (kg) NO2 (kg) PM (kg) (GJ) (m3) (tonnes) (tonnes) Copper 5 845,53 12 935,30 9 311,72 859,22 42 951,01 1 145,16 6 020,87 Gold 8 504,03 18 871,28 17 100,44 1 267,09 208,61 4 254,20 12 512,36 Uranium 3 858,50 10 367,97 8 236,08 398,64 287,43 1 084,34 1 188,27 Fig. 18: Summarised results I for year 2007. 58 159 24
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Summarised results 2007 per MUSD of product 45 000 40 000 35 000 30 000 25 000 20 000 15 000 10 000 5 000 0 HTP POCP Air FAETP (kg FSETP (kg TETP (kg EP Air (kg (tonnes (kg AP Air (kg 1,4-DCB- 1,4-DCB- 1,4-DCB- PO4(3-)- 1,4-DCB- ethylene- SO2-eq) eq) eq) eq) eq) eq) eq) Copper avg. 1 152,75 38 357,68 91 303,50 36 375,56 2 102,42 43 437,07 148,93 Gold avg. 84,75 2 865,84 6 362,50 26 597,99 160,56 1 508,94 278,77 Uranium avg. 3 609,21 3 028,25 8 063,93 4 040,78 105,03 1 104,24 148,26 Fig. 19: Summarised results II for year 2007. 10.14 Robustness of Results Since the economic value of the production is the functional unit of this study, the robustness of the results primarily needs to be analysed with respect to this parameter. Prices on the investigated minerals have increased during the past five years and using the Metal Prices page on infomine (Infomine: Metal Prices 2008) the metal prices five years ago were determined. Tab. 8 below shows approximate mineral prices five years ago and the prices used in this study and the relative change during this period. As can be observed, the prices of copper and gold have witnessed a similar increase while the increase of gold price has been somewhat more modest. Mineral Approximate Price used in Unit Increase Price November this study 2003 (avg. 2007) Copper 2 000 7 139 USD/tonne 257% Gold 400 696 USD/troy ounce 74% Uranium 13 45 USD/lb 246% Tab. 8: Mineral prices in a five year perspective. Using price levels five years ago, the results are quite different, which can be observed in Fig. 20 and Fig. 21. In general, the uranium mining industry still performs well but the results are most significantly changed for the gold mining industry, which shows better performances in relation to current results. This is most probably related to the fact that the price on gold has been more stable. Five years ago, the value of current production was relatively smaller for copper and uranium compared to today and consequently, all impacts are more severe per economic value. However, since the mineral prices most likely reflect the production, it is not reasonable to believe that the current production would be the same five years ago. 59 303 19
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Summarised results per MUSD (Price level Nov. 2003) 160 000 140 000 120 000 100 000 80 000 60 000 40 000 20 000 0 GWP100 HTP Tot FAETP Tot FSETP Tot TETP Tot (kg POCP Air (kg AP Air (kg EP Air (kg (tonnes CO2- (tonnes 1,4- (kg 1,4-DCB- (kg 1,4-DCB- 1,4-DCB-eq) ethylene-eq) SO2-eq) PO4(3-)-eq) eq) DCB-eq) eq) eq) Copper avg. 2 297 4 106 135 667 322 843 129 775 7 447 153 735 396 Gold avg. 2 277 147 4 986 11 069 46 275 279 2 625 485 Uranium avg. 1 134 12 493 10 482 27 914 13 987 364 3 822 513 Fig. 20: Summarised results I per MUSD of product with November 2003 price levels. Summarised results per MUSD (Price level Nov. 2003) 50 40 30 20 10 0 TJ ML ktonnes CO2 tonnes SO2 tonnes NO2 tonnes Energy use (tot) Water usage GHG (tot) SOx NOx PM tot Copper avg. 18,13 45,44 2,82 152,88 3,04 20,91 Gold avg. 14,92 33,13 2,24 0,37 7,40 21,77 Uranium avg. 13,01 35,89 1,44 0,99 3,75 4,11 Fig. 21: Summarised results II per MUSD of product with November 2003 price levels. One uncertainty about the results concerns the missing characterisation factors. But even after setting relatively high hypothetical values for the missing characterisation factors, the relative results remain unchanged. 60
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 11 Conclusion The purpose of this investigation is to determine whether or not uranium mines, from an environmental point of view generally perform better or worse than any other mineral mining industry. Comparing environmental performance and health aspects in the mining industry is a broad and extensive task and based on the findings of this report it is impossible to come to an unequivocal conclusion that is valid for the entire industry. However, based on the parameters investigated in this study, the relative performances of the three minerals can be used to conclude something in this specific case. Considering all parameters subject to investigation, human toxicology, radiation and lost time injury frequency rate issues are the only ones showing a negative result for the uranium mining industry, in relative terms. These three areas are typically categorized under health aspects, which is one of the topics of this study. However, the main effects contributing to the unfavourable result in the human toxicology impact for uranium mines originate from Olympic Dam operations, which is a multi mineral operation where copper accounts for about 75% of total sales. This study has not been able to determine whether the operations responsible for the release of the toxic substances at Olympic Dam can be allocated to the entire operations or to any specific mineral. Radiation is a more unambiguous parameter that is solely related to uranium mining in this study. It is estimated to be primarily an occupational issue since the reported doses to the public are very low. In all other investigated aspects, the uranium mining industry shows a very good relative performance, generally positioned as the best or tied leader of the investigated minerals. All these areas of analysis can be categorized as contributing to environmental impacts. Finally, the most complex but also among the most important areas concern mine waste and tailings issues. Based on the highly restricted amount of data and information available about the area, only some general conclusions can be drawn. The three different minerals can generally be connected to some specific impacts of special importance to each mine. Acid mine drainage generally appears to be a problem primarily for the copper mining industry but also in the case of gold due to their occurrence in sulphide mineral ores. For the mining and extraction of gold, one issue of special importance is that of cyanide management. The uranium mining industry has to deal with long-term radiological issues from tailings management facilities. Furthermore the amount of tailings, and consequently also the land area required for its storage is highly related to the mineral grade. Therefore, the highest amounts of tailings are commonly encountered in the gold mining industry. With the knowledge and information gained and collected during the course of this thesis, and based on the parameters included in this study, it can be concluded that the uranium mining industry is connected to more adverse potential impacts to the human health compared with other mines of this investigation. From an environmental point of view however, including resource consumption, the uranium mining industry is associated with impacts that are significantly lower than the worst average mineral, in each parameter, of this study. 61
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 12 Discussion The choice of an appropriate functional unit for this study, when comparing completely different commodities such as gold, copper and uranium, is not obvious. Besides the functional unit chosen, that is the value of the product when crossing the system boundaries, the amount of ore mined and the amount of product produced were also discussed and analysed briefly. When the functional unit was changed to tonnes of product, gold with the lowest ore grade became far worse than the other minerals in all categories. Uranium, associated with the highest average ore grade benefited from the change in functional unit. When comparing performance using the amounts of ore mined, gold improved vastly while the competitiveness of uranium decreased. However, most important when looking at resource consumption and impacts on society and environment should be to compare it with the value provided by the products and consequently by the operations. There is not always a clear distinction between what is classified as ore mined and waste rock, resulting in large uncertainties regarding reported data. Further, the degree of refinement varies between the different minerals but also between different sites. Hence, the economic value of the product is judged to represent the most useful functional unit. A large number of data has been interpreted and analysed throughout the project. The quality and amount of data has influenced the results and conclusions. Companies report different data in different ways and in varying detail. Additionally, the databases from NPI and NPRI, providing data of emissions do not have the same criteria for reporting. Further, the amount and type of data available has limited what aspects to study. There are more impact categories of interest when comparing the different operations, such as ozone depletion potential and land use to mention some of them. The lack of some characterisation factors could also have influenced the calculations of impact categories. Cyanide is a substance that likely would have changed the results if a characterisation factor existed. Human toxicity and ecotoxicity potential are controversial areas that are often the subject to discussions and disagreement. A dilemma worth discussing more explicitly is the allocation of resource consumption and emissions from operations producing more than one type of mineral and the subsequent analysis of the results. Especially Olympic Dam is troublesome, due to the fact that the largest part of the production and sales is connected to copper production and not uranium. At all other operations, only minor amounts of additional minerals are produced. At Olympic Dam, copper represents as much as 75% of sales. Allocation with respect to share of sales was considered the most appropriate since specific and detailed data connected to the different mineral operations did not exist. The same method has been used by Mudd and Diesendorf (2007), investigating the sustainability of uranium mining and milling. Even though one might discuss the appropriateness in treating Olympic Dam as a uranium mine, it was included since it is the largest uranium deposit in the world. Tailings and tailings management are important areas that were only covered briefly in this study. The discussions and conclusions were drawn using an extremely simplified picture of a complex interaction between several factors. There are more aspects to include in an analysis of this sort and the actual performance and properties of the objects studied should be extended beyond the current system boundaries. The original goal was to be able to draw general conclusions of the impact of mining uranium, copper and gold, using data from specific operation. This study can hopefully offer an indication of such a comparison. However, for general conclusions to be made, the number of objects investigated 62
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry 13 Recommendations on Further Studies The allocation of emissions and resource consumption to a specific mineral in multi mineral mines has been discussed and it is an area which can be studied further. The current approach of economic allocation does not take into account processes that are linked to one mineral only. Based on the limited amount of information available for this study it was judged to be the most appropriate solution but a more in depth analysis of the different mining methods and processes and their contribution to the environmental loads of the different operations would be interesting. This is particularly interesting for Olympic Dam and its extremely high emissions of polycyclic aromatic hydrocarbon, contributing to the unfavourable results connected to human toxicity potential. The problems connected to tailings are very important and complex and they could be studied far more in depth than what was possible in this report. Even with a more thorough analysis however, the problem to compare and value problems such as acid mine drainage and radioactive tailings leachate remains. A tool for an evaluation and comparative study of that sort is recommended. Depending on where the different operations are located, the companies struggle with different environmental and social problems. For example HIV and AIDS are extremely important issues in the African countries and thus also important for the companies with operations in this part of the world. These companies have acknowledged the severity of the disease and appear to work actively with prevention and medical aid. How mining operations affect indigenous people, e.g. the aborigines in Australia, is another local aspect. For future studies, a wider sustainability perspective can be used in order to incorporate these aspects. 64
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry KCGM - Barrrick & Newmont Raw data Unit Adapted Changed unit figures Production Total movement (entire KGCM) 78,71 Mtonnes 78710000 tonnes of which ore 12,68 Mtonnes 12680000 tonnes waste rock 66,03 Mtonnes 66030000 tonnes Mineral produced in Fimiston 0,63 Mounces gold bullion 17,860185 tonnes gold Mill bullions Mineral grade 2,01 g/t 0,000201 % Energy use Purchased Electricity (Natural 384 556 MWh 1384401,6 GJ Gas) LPG (non-transport) 3247 kL 83447,9 GJ Automotive Gasoline (Petrol) 50,05 kL 1711,71 GJ Automotive Diesel 72 230 kL 2788078 GJ Employees 588 # 588 # Additional contractors 465 # 465 # Water 8769 ML water is used, see 8769000 m3 table below for type and amouns of water. Water consumed by mining and 6026000 m3 6026000 m3 processing Chemicals Explosives used 18304 tonnes 18304 tonnes Cyanide use 783 tonnes 783 tonnes Emissions GHG-emissions 584 690 tonnes CO2-eq 584 690 tonnes CO2- eq LTFIR 0,1 Lost-Time Injury Rate – 0,1 LTI per 200 the number of work- 000 h related injuries that result in days away from work for every 200,000 hours worked. Land disturbance Used area in total 1600 ha in total land 1600 ha disturbed and rehabilitated by KCGM since 1989 (approx figurefrom graph) Used area - pit area 2,7 km2 Rehabilitated land 36,1 ha rehab. In 2007 36,1 ha Total amount of rehabilitated 700 ha, approx. Figure of 700 ha land from start rebab. land since 1989. Area 460 ha 460 ha Waste rock 66,03 tonnes 66,03 tonnes 76
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry KENNECOTT UTAH COPPER - Rio Tinto Raw data Unit Adapted figures Changed unit Reference unit Sales - total sales 3546 $million 3546000000 2017917434 USD - net profit 1625 $million 1625000000 924736556,9 USD Production Ore mined (calculated value 49185185,2 tonnes 49185185,19 27989746,95 tonnes based on ore grade of Cu and copper cathode production) Grade of ore 0,54 % Cu 0,54 0,54 % 0,32 g/t Au 0,000032 0,000032 % 2,59 g/t Ag 0,000259 0,000259 % Mineral produced - copper 265600 metric tonnes cathodes 265600 265600 tonnes - gold 522800 troy oz 16,26089767 16,26089767 tonnes - silver 4365000 troy oz 135,7666762 135,7666762 tonnes - molybdenum 14906 metric tonnes cathodes 14906 14906 tonnes Energy use 21040000 GJ onsite usage and 21040000 11973204,4 GJ purchased power Employees - employees 1815 # 1815 1815 # - contractors 1200 # approximations (try 1200 1200 # homepage for more details) Total at site 3015 3015 3015 # Freshwater consumption 4255 ML 4255000 2421387,107 m3 Emissions GHG-emissions 1,9 total million tonnes of 1900000 1081230,436 tonnes CO2 -eq, based on the total onsite and purchased power Sox 3695 short tons 3352,0301 1907,535245 tonnes Nox 6654 short tons 6036,37572 3435,112184 tonnes Particulate matter 3682 short tons (estimated 3340,23676 1900,824025 tonnes emissions - see the Utah division of Air Quality for precise numbers) Health LTFIR 0,41 occurence of LTI per 200 0,41 0,41 /200000h 000 hours worked Land disturbance Used area 5262 ha (land disturbed forv 5262 2994,439238 ha mining, processing and related activitis that is not currently rehabilitated) 78
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry MCARTHUR - CAMECO KEY LAKE - CAMECO TOTAL Raw Data Unit Raw Data Unit Total Unit Production Ore mined 59642 tonnes of ore with 59642 tonnes average grade of 14.24% U3O8 produced Ore milled 44038 dry tonnes ore 44038 tonnes milled Ore shipped 44367 dry tonnes ore 44367 tonnes shipped to Key lake Grade of ore 19,05 % U3O8 19,05 % Mineral produced 18718337 lbs of U3O8 8488740 kg U3O8 8488,74 tonnes (U3O8) Mineral grade 19,35 % high grade ore reamed 0,53 % low grade ore reamed 0,97 % development low grade 14,24 % average treated Uranium produced 7198452 kg U Ore leached (diluted 211806 dry tonnes ore 211806 tonnes with stockpiled ore leached from mined out pits) Grade of ore 4,07 %U3O8 (mill 4,07 % leached (mill head head grade) grade) Ammonium sulphate 12991 tonnes 12991 tonnes fertilizer ammonium sulphate fertilizer Energy use Electricity 350265,1169 GJ (2006) multiply 504095 GJ (2006) 826147,85 GJ with production multiply with factor below to production get data for 2007 factor below to get data for 2007 Fossil Fuel 235788,336 GJ (2006) multiply 229918 GJ (2006) 450328,006 GJ with production multiply with factor below to production get data for 2007 factor below to get data for 2007 80
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Total 586053,4529 GJ (2006) multiply 734013 GJ (2006) 1276475,86 GJ with production multiply with factor below to production get data for 2007 factor below to get data for 2007 Employees 715 # 1102 # 750 # 750 # (McArthur River 750 # (McArthur + Key Lake total) River + Key Lake total) Water Amount of water 2399616 In 2007, the 3952484 m3 released from McArthur River treatment plant Operation discharged 2,399,616 m3 of treated effluent to the receiving environment. Water usage 122793 m3 from toby lake 10468961 m3 (total inflow 10591754 m3 potable water to Key Lake) intake Treated mill effluent 1552868 m3 (treated mill effluent released) Chemicals mera finns i table 3.2 Barium Chloride 85200 kg 367200 kg 452400 kg Ferric Sulphate 429560 kg 429560 kg Lime 35907 kg 20583020 kg lime 20618927 kg Sulphuric Acid 51253 kg 51253 kg Sulfur 11162952 kg sulphur 11162952 kg Acid 46231000 kg acid 46231000 kg Emissions Particulate matter 4,2 tonnes PM10 33 tonnes PM10 37,2 tonnes PM10 PM2,5 0,99 tonnes PM2.5 0,209 tonnes PM2.5 1,199 tonnes Total 6,5 tonnes total 250 tonnes total 256,5 tonnes particulate matter particulate matter GHG-emissions 81855 tonnes CO2-ekv 89450,9 tonnes CO2-ekv 165649,119 tonnes CO2 (2006) multiply (2006) multiply with production with production factor below to factor below to get data for 2007 get data for 2007 81
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Total direct 21991 tonnes CO2-ekv 31077,3 tonnes CO2-ekv 51315,9042 tonnes CO2 (2006) multiply (2006) multiply with production with production factor below to factor below to get data for 2007 get data for 2007 Total indirect 59864 tonnes CO2-ekv 58373,6 tonnes CO2-ekv 114333,215 tonnes CO2 (2006) multiply (2006) multiply with production with production factor below to factor below to get data for 2007 get data for 2007 Sox 258 tonnes SO2 258 tonnes SO2 Nox 32 Tonnes NO2 64 Tonnes NO2 96 tonnes NO2 Ammonia 84 tonnes 84 tonnes ammonia VOC 481 tonnes volatile 481 tonnes organic carbon Health Radiation (average 1,5 mSv (average) 0,58 mSv (average 1,04 mSv both sites) effective dose) Highest average 2,5 mSv (underground 1,35 mSv (highest 2,5 mSv dose (underground miner, highest average miner) average dose) effective dose, mill operations) Highest maximum 10,2 mSv (underground 5,24 mSv (max 10,2 mSv dose (undeground support, maximum effective dose, support) dose) mill maintenance) LTFIR (average) 1 LTI 2007 per how 0,49 /200000 h 0,745 /200000h many hours worked Urine test 1 of 3387 samples had U conc above 5 µg/l Land disturbance rehabilitated land 81,1 "ha (Agronomic and Native Grass Revegetation 1998-2005 McArthur 81,1 ha River Operation)" 2,5 ha (Tree/Shrub 2,5 ha Seedling Revegetation 2007 McArthur River Operation) Tailings area 45 ha (AGTMF) 45 ha 82
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry MCCLEAN LAKE - Areva Raw data Unit Adapted Changed unit figures Sales 1900000 Cpd Production Total movement 2973436 m3 2973436 m3 Ore mined - Total from Sue 193503 tonnes 193503 tonnes ore - from Sue E ore 190921 tonnes 190921 tonnes U-grade in ore 0,49 % 0,49 % mined art Sue E Mineral recovered 933918 kg 933,92 tonnes from mining Mill feed 170083 tonnes 170083 tonnes mineral grade in mill 0,45 % U 0,45 % U feed Recovery rate mill 94,7 % U 94,7 % U Mineral recovered 732952 kg U in U3O8 864,36 tonnes U3O8 after processing Mineral shipped 732097 kg U in U3O8 863,36 tonnes U3O8 shipped Employees 314 # 314 # Water usage 1393575 m3 1393575 m3 Amount of water "JEB WTP: 1 released from 333 435 m3, treatment plant Sue WTP: 1 091 884 2425319 m3 in total m3 " Energy use From fossil fuels 572 906 GJ 572906,07 GJ Power (electric 40625910 kWh 146253,28 GJ energy) Total energy GJ 719159,34 GJ Emissions Particulate matter 223,633 tonnes of total 223,63 tonnes particulate matter. Includes PM10 and PM2,5 GHG-emissions (tot) 48403 tCo2-eq 48403,35 tonnes of CO2-eq. GHG-emissions 39459 tCo2-eq 39459,41 tonnes of (direct) CO2-eq. GHG-emissions 8944 tCo2-eq 8943,95 tonnes of (indirect) CO2-eq. Sox 11,346 tonnes (see 11,35 tonnes comments by table of what is included) 84
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry MOUNT ISA MINE - Xstrata Raw data Unit Allocated to Unit Adapted Changed Cu- figures unit production Sales - Total sales Mount 2353859852 USD USD 2,4E+09 USD Isa Operations - copper 1274041835 USD 1274041835 USD 1,3E+09 USD Share of sales - Copper 54,12564533 % % - Zinc 27,27450211 % % - Lead 14,04447005 % % - Silver 4,5553825 % % Production Ore mined, - copper 6000000 tonnes ore mined 6000000 tonnes ore mined 6000000 tonnes - Zn-Ag-Pb-ore 5,12 Mtonnes Mtonnes 5,12 tonnes mineral produced - copper 172552 tonnes of copper 172552 tonnes of copper 172552 tonnes in concentrate in concentrate copper in concentrate 217907 tonnes copper in 140129,464 tonnes copper in 140129 tonnes anodes at smelter, anodes copper including Ernest anode Henry - ZN 226529 tonnes concentrate - Pb 125195 tonnes bullion Mineral grade 3,42 i.e. copper head 3,42 i.e. copper head 3,42 % grade grade Energy use Electricity (indirect 1100 kWh/tonne 153371699 kWh (kWh per 552 GJ energy) contained metal tonne contained 138,11 (mount isa op.) metal*tonnes Cu- (estimated from anode*0,995 graph) (99,5% Cu in anodes)) Diesel 45655000 litres 24711063,4 litres 953 GJ 847,05 LPG (non-transport) 343000 litres 185650,964 litres 4 771,23 GJ Unleaded petrol 138000 litres 74693,3906 litres 2 554,51 GJ Total direct energy 961 GJ 172,79 86
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry Total ENERGY 1 513 GJ 310,90 Employees 4376 total Mount Isa, i.e. total Mount Isa, also zinc and etc i.e. also zinc and etc 2600 # for copper 2600 # for copper 2600 # employees operations, operations, including including contractors, contractors, approximate approximate number number Water Water usage 8023 megalitres of 4342,50053 megalitres of 4342501 m3 freshwater (total freshwater (total MIM) MIM) Chemicals/Materials Cement 99708 tonnes 53967,5985 tonnes 5,4E+07 kg Coke 39075 tonnes 21149,5959 tonnes 2,1E+07 kg Copper sulphate 4040 tonnes 2186,67607 tonnes 2186676 kg Explosives 12059 tonnes 6527,01157 tonnes 6527,01 tonnes Grinding media 21291 tonnes 11523,8912 tonnes 1,2E+07 kg Lime 12179 tonnes 6591,96235 tonnes 6591962 kg Limestone 68353 tonnes 36996,5024 tonnes 3,7E+07 kg Lubricating oils 2894000 litres 1566396,18 litres 1566396 litres Sodium carbonate 4731 tonnes 2560,68428 tonnes 2560684 kg Sodium hydroxide 101 tonnes 54,6669018 tonnes 54666,9 kg Silica 75987 tonnes 41128,4541 tonnes 4,1E+07 kg Tyres (heavy vehicles) 375 # 202,97117 # 202,971 # Tyres (light vehicles) 2946 # 1594,54151 # 1594,54 # Wood 1867 tonnes 1010,5258 tonnes 1010526 kg Xanathe 2089 tonnes 1130,68473 tonnes 1130685 kg 87
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry OLYMPIC DAM - BHP Billington Raw Unit Adapted Allocation to Changed unit data figures, all U- operations production, 20% Revenue 20 % from U 75 % from Cu 5 % from Au/Ag Production Ore mined, alt 9200000 tonnes 9200000 1840000 tonnes mineral produced 3985 tonnes U3O8 3985 3985 tonnes U3O8 204000 tonnes coppper 204000 tonnes cu cathodes cathodes approximately mineral grade 0,7 kg U3O8/tonnes 0,07 0,07 % Energy use 5,3 PJ (Total) 5300000 1060000 GJ Coal and coke 0,3 PJ 300000 60000 GJ purchased electricity 3,1 PJ 3100000 620000 GJ Distillate 1 PJ 1000000 200000 GJ Fuel Oil 0,2 PJ 200000 40000 GJ Other 0,8 PJ 800000 160000 GJ Employees 2500 # (about) 2500 2500 # Water usage 12740 Ml (high quality 12740000 2548000 m3 water, ground water from wells) 320 Ml (low quality water, 320000 64000 m3 saline) 440 Ml (recycled water) 440000 88000 m3 Emissions GHG-emissions 1030000 tonnes CO2-ekv 1030000 206000 tonnes CO2-ekv SOx 1280 tonnes 1280 256 tonnes SO2-ekv NOx 1180 tonnes 1180 236 tonnes NO2-ekv Health Radiation 4,1 mSv/year (average 4,1 4,1 mSv/year designated worker) (average designated worker) 8,1 mSv/year (maximum 8,1 8,1 mSv/year dose) (maximum dose) Dust Chart 16-1 TRIFR 13,2 injuries/million man 2,64 2,64 injuries/200000h hours worked Land disturbance used area 1216110 ha (Total footprint) 1216110 243222 ha infrastrucure 340 ha 340 68 ha 92
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry PALABORA, Rio Tinto Raw data Unit Adapted Changed unit figures Production Ore mined 11850000 tonnes 11850000 tonnes - copper concentrate delivered 70443 tonnes 70443 "tonnes to smelter - Supplementing copper 43605 tonnes concentrate, puchased from Chambasi and Fosker - copper anode 90700 tonnes 90700 tonnes recovery rate mill 93,6 % 93,6 % mineral recovered after 91659 tonnes copper 91659 tonnes copper processing cathod cathod ore grade 0,7 % copper in ore 0,7 % hoisted sales 1861000000 rand 13128832214 USD earnings/profit 58 milion US$ 58000000 USD Employees 1955 permanent 1955 # permanent employees 126 fixed term 126 # fixed term employees Water Water usage 10444 Ml 10444000 m3 (intake+rainfall+from dewatering and seepage capturing) recycled water 25168 Ml 25168000 m3 Energy use tot 6482888 GJ 6482888 GJ indirect (electricity) 678608 MWh 2442988,8 GJ direct (fossil) substraction 4039899,2 GJ Emission GHG-emissions tot 985706 tonnes CO2-eq 985706 tonnes CO2-eq GHG-emissions (direct) tonnes CO2-eq 342854,26 tonnes CO2-eq GHG-emissions (indirect) tonnes CO2-eq 642851,74 tonnes CO2-eq Sox 22839 tonnes 22839 tonnes Health Dust 21 microgram/m3/year 21 microgram/m3/year LTFIR 0,32 per 200 000 h 0,32 per 200 000 h worked worked Land disturbance used area 119000 m2 11,9 ha rehab. And decommissioning 312000000 rand 2201,07 MUSD fund/liability 94
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry RABBIT LAKE - Cameco Raw data Unit Adapted Changed unit figures Production Ore mined 191470 tonnes 191470 tonnes total haulage 282016 tonnes 282016 tonnes (ore+waste+special waste) recovery rate mill 96,7 % 96,7 % mineral recovered after 4000000 lbs U3O8 1814,36948 tonnes processing mineral recovered after 1814360 kg U3O8 1814,36 tonnes processing "mineral grade in mill feed Average mill head grade (% 0,69 % 0,69 % U3O8)" Employees 466 # 466 839, acc to 6,9, 466, p 8-1 922, table 6.9.1.1-1 Water Water usage 4171622 m3 (total inflow) 4171622 m3 The total quantity of water 3783667 m3 (5-2, see also table 5.3.1 for water 3783667 m3 discharged to the balance) environment via Weir No.3 (Station 2.3.3) was 3,783,667 m3 in 2007. water pumped from pit 1421387 m3 1421387 m3 Energy use Electricity 493763,9 GJ (2006) multiply with production 430470,5406 GJ factor below to get data for 2007 Fossil Fuel 293009,3 GJ (2006) multiply with production 255449,7641 GJ factor below to get data for 2007 Total 786773,2 GJ (2006) multiply with production 685920,3047 GJ factor below to get data for 2007 Emissions GHG-emissions 105435,3 tonnes CO2-ekv (2006) multiply with 91920,02613 tonnes production factor below to get data for 2007 total direct 31043,5 tonnes CO2-ekv (2006) multiply with 27064,17425 tonnes production factor below to get data for 2007 total indirect 74391,8 tonnes CO2-ekv (2006) multiply with 64855,85188 tonnes production factor below to get data for 2007 Health Radiation 1,5 mSv (average) 1,5 mSv 5,5 mSv (underground miner, highest 5,5 mSv average dose) 12,6 mSv (underground miner, maximum 12,6 mSv dose) 96
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry RANGER - Rio Tinto & ERA Raw data Unit Adapted Changed unit figures Sales Total sales 357080000 $ AUD 426903185,1 USD Net profit 76089000 $ AUD 90967392,33 USD Production Ore mined, alt 2900000 tonnes ore mined 2900000 tonnes ore processed 1900000 tonnes ore milled 1900000 tonnes mineral produced 5412 tonnes U3O8 5412 tonnes mineral grade 0,12 % 0,12 % Energy use (tot) 1222980 GJ 1222980 GJ Direct (in report) 17709 GJ 17709 GJ Indirect (as stated for 20934662 litres diesel fuel 829012,6152 GJ 2006) Employees 419 419 # Water usage 650 ML freshwater 650000 m3 Chemicals Emissions GHG-emissions 17,7 tonnes CO2 ekv/tonnes UO produced 2006 741 tonnes additional GHG emissions 2007 4748 tonnes U3O8 production 2006 84039,6 total tonnes CO2-ekv 2006 (17.7*4748) Direct 84780,6 total tonnes CO2-ekv 84780,6 tonnes CO2- 2007 (84039.6+741) ekv Indirect 60767,99256 tonnes CO2 (GJ 60767,99256 tonnes CO2- diesel*73,67*0,995) ekv Total 83109 tonnes CO2 83109 tonnes CO2- ekv Sox 120 tonnes SO2 120 tonnes SO2 Health Radiation 1,3 mSv (240 designated 1,3 mSv workers, 0.6 mSv for the rest) Health 5,1 mSv max 2006 5,1 mSv LTFIR 0,3 /200000h (estimated 0,3 /200000h from graph) AIFR 1 /200000h (estimated 1 /200000h from graph) Land disturbance used area 520 ha (79 km2 total land 520 ha lease) 98
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry RÖSSING - Rio Tinto 69% Raw data Unit Adapted Changed unit figures Net profit 979000000 N$ (after/tax) 6900791517 USD Production Ore mined, alt 21396000 tonnes mined rock 21396000 tonnes ore processed 12613000 tonnes ore processed 12613000 tonnes mineral produced 3046 tonnes U3O8 3046 tonnes mineral grade 0,289 kg/tonnes of U3O8 0,0289 % Energy use, total 1533981 GJ 1533981 GJ Diesel for mobile 769810 GJ 769810 GJ equipment Diesel for roaster 19968 GJ 19968 GJ Petrol 17853 GJ 17853 GJ Purchsed electricity 726350 GJ 726350 GJ Employees 1175 # 1175 # Water amount of water 3050000 m3 seepage water 3050000 m3 released from collected treatment plant Water usage 3299980 m3 3299980 m3 0,26 m3/tonnes of ore milled 900 m3/tonnes of U3O8 produced (estimated from graph) GHG-emissions total 197029 tonnes 197029 tonnes CO2- ekv total excl. Electricity 90296 tonnes 90296 tonnes CO2- ekv Diesel from mobile 57047 tonnes 57047 tonnes CO2- equipment ekv Diesel used at 1480 tonnes 1480 tonnes CO2- roaster ekv Petrol 1176 tonnes 1176 tonnes CO2- ekv Extraction process 27748 tonnes 27748 tonnes CO2- ekv Blasting of 2573 tonnes 2573 tonnes CO2- explosives ekv Sewage plant 227 tonnes 227 tonnes CO2- ekv Sodium carbonate 21 tonnes 21 tonnes CO2- ekv 100
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry NPI - Facility ERA Ranger Mine Heathgate Kalgoorlie CGM Super Pit/Fimiston Open Pit Time period 01-Jan-2006 to 31-Dec-2006 01-Jul-2006 to 30-Jun-2007 01-Jul-2006 to 30-Jun-2007 Substance Total Air Land Water Total Air Land Water Total Air Land Water (kg) (kg) (kg) Ammonia (total) Antimony & Compounds 28 28 0 Arsenic & compounds 11 8,3 0,81 1,6 14 6 8 410 400 2,2 Benzene 250 250 300 300 Beryllium & compounds 18 17 0,49 0,69 0 0 0 4,4 4,4 0 Boron & compounds 81 81 0 Cadmium & compounds 0,57 0,13 0,19 0,25 0 0 0 2 2 0 Carbon disulfide Carbon monoxide 3E+05 3E+05 54000 54000 6E+05 6E+05 Chlorine Chromium (III) compounds 200 190 2 3,1 110 69 41 1500 1500 0 Chromium (VI) compounds 0 0 Cobalt & compounds 62 57 4 0,62 340 340 0 Copper & compounds 1200 1200 11 4,5 46 30 16 2300 760 1600 Cresol Cumene (1- 3,3 3,3 methylethylbenzene) Cyanide (inorganic) 26000 25000 320 compounds Dioxine Ethylbenzene 4,4 4,4 9,1 9,1 Ethylene glycol Fluoride compounds 0 0 290 200 90 4800 4800 0 Formaldehyde (methyl 27000 27000 aldehyde) Hydrochloric acid 0 0 Lead & compounds 51 30 19 1,9 76 35 41 67 67 0 Magnesium oxide fume 0 0 Manganese & compounds 560 200 230 130 14000 14000 0 Mercury & compounds 0,44 0 0,19 0,25 0 0 0 570 560 1,7 Nickel & compounds 190 180 7 1,8 140 49 88 550 550 0 Nickel carbonyl 0 0 Nickel subsulfide 0 0 Nitrate Nitric acid Oxides of Nitrogen 8E+05 8E+05 88000 88000 2E+06 2E+06 Particulate Matter 10.0 um 8E+05 8E+05 3E+05 3E+05 5E+06 5E+06 Polychlorinated dioxins and 0,3 0,3 0 0 furans Polycyclic aromatic 3,9 3,9 2,5 2,5 19 19 hydrocarbons Selenium & compounds 2,3 2,3 0 107
Chalmers University of Technology
Environmental Impacts and Health Aspects in the Mining Industry NPI - Facility Mount Isa Mine Northparkes Olympic Dam Time period 01-Jul-2006 to 30-Jun-2007 01-Jul-2006 to 30-Jun-2007 01-Jul-2006 to 30-Jun-2007 Substance Total Air Land Water Total Air Land Water Total Air Land Water (kg) (kg) (kg) Ammonia (total) 8800 0,01 8700 58 15 15 Antimony & Compounds 26000 26000 13 0,27 7 7 Arsenic & compounds 75000 74000 59 12 74 74 1100 1100 Benzene 220 220 210 210 Beryllium & compounds 980 960 22 4 4 9,5 9,5 Boron & compounds 13000 7400 5800 170 170 Cadmium & compounds 4200 4000 230 3,6 3,4 3,4 37 37 Carbon disulfide 1200 1200 15 400 400 Carbon monoxide 8E+05 8E+05 31000 31000 4E+05 4E+05 Chlorine 8 8 Chromium (III) compounds 91 90 0,34 490 490 1800 1800 Chromium (VI) compounds 0 0 0,41 0,41 Cobalt & compounds 9500 8800 650 130 130 88 88 Copper & compounds 4E+05 4E+05 53000 490 15000 15000 11000 11000 Cresol Cumene (1- 0,11 0,11 0,03 0,03 methylethylbenzene) Cyanide (inorganic) 10000 10000 17 670 670 compounds Dioxine Ethylbenzene 0,75 0,75 0,54 0,54 Ethylene glycol Fluoride compounds 30000 5200 25000 2200 2200 2100 2100 Formaldehyde (methyl aldehyde) Hydrochloric acid 4500 4500 Lead & compounds 3E+05 3E+05 450 170 290 290 4800 4800 Magnesium oxide fume 0 0 68 68 Manganese & compounds 30000 7800 23000 32 7800 7800 4000 4000 Mercury & compounds 940 940 0,01 0,53 0,53 16 16 Nickel & compounds 3300 3200 180 1 270 270 870 870 109
Chalmers University of Technology
Abstract This master thesis will address the topic of rare earth metals and their product chain from early mining processes, through refining to final products and recycling. The analysis will cover the global product chain and its different actors. Rare earth metals consist of 17 elements of the periodic table; the 15 lanthanides, scandium and yttrium. Despite their name, these elements are not rare, but quite abundant all over the earth’s crust. Nevertheless, their concentration in the deposits is often very low and they are difficult to extract and refine. The rare earth metals are similar in their characteristics and can be used for many different applications. Rare earths are commonly used in green technology since their unique properties can save weight and increase efficiency in products such as light bulbs, generators in wind power plants, and in different parts of hybrid and electric vehicles. The focus in this report will be on the product chains of the rare earth metals used in permanent magnets and batteries used in hybrid and electric vehicles, these are: neodymium, praseodymium, lanthanum, dysprosium, cerium and terbium. In nature these elements do not occur as metals and many complicated processes are needed to convert them into metals. The product chain starts with the mining and separation processes, next comes further refining and separation processes, and then production of rare earth metals and alloys before they can be shaped into parts in a final product. In total China has more than 95% of the global market for mining and separation processes for rare earth metals, also the production of rare earth metals and alloys occur to a large extent in China. The production of most permanent magnets also takes place in China, but there is also production in Japan, Finland and Germany. Japanese car manufacturers have most of their parts produced within Japanese borders, which means that most permanent magnets used in hybrid and electric vehicles are produced in Japan. The same is true for the NiHM batteries, about 95 % of those that are inside these vehicles today are produced in Japan by Japanese manufacturers. The markets for green technology seem to expand, creating an increasing demand for rare earth metals. To meet the demand with a larger supply there is need for solutions such as new mining projects, reopening of closed mines, or recycling and recovering of rare earths from scrap products and mine tailings. The urge for more supply of rare earth metals and also rising prices presents a good incentive for recycling and recovering different rare earth metals from discarded products. These recycling and recovering processes are technically and often economically very demanding. Today there are processes that could recover rare earths with a high level of purity, but the problem is often the huge costs for the processes as well as handling of waste chemicals and materials that might be toxic. New processes will have to be developed. There has been research for new processes, and recycling/recovering plants are planned to be constructed in Japan as well as other countries. This report will also look at the potential for having parts of the rare earth product chain in Sweden, since the country has both the physical resources and also metallurgical knowledge. With this in mind we discussed the possibilities of eventually operating the processes in the product chain in Sweden, and aspects of this scenario. 1
Chalmers University of Technology
Preface This master thesis was written at ESA, Environmental Systems Analysis, at Chalmers University of Technology. Henrikke Baumann, associate professor at the department of ESA has been our supervisor. The making of this master thesis has been the final task of the Master program Industrial Ecology – for a sustainable society. Therese Eriksson has previously studied the bachelor program Industrial Economy at Chalmers University of Technology, and David Olsson has a Bachelor of Science in Mechanical Engineering also at Chalmers. The idea of this master thesis came from Magnus Karlström who writes a daily newsletter to people within Swedish automotive industry. The newsletter covers a lot of news about the vehicle manufacturing industry and rare earth metals is a hot topic among vehicle manufacturers today since hybrid technology uses them frequently in different applications. The thesis outline was addressed to map the global product chain for these rare earths frequently used in hybrid vehicles and describe the different steps from mining to potential recycling. Also a future scenario with possible stages of the rare earth processes located in Sweden would be discussed. In order to map the product chain our goal has been to collect and analyze information from different available and trustworthy sources and analyze it to get a perspective over the whole product chain globally. 2
Chalmers University of Technology
1 Introduction Once upon a time there was a Chinese fishing boat. One day in September 2010 the fishing boat was out fishing on the South China Sea, when suddenly it collided with a Japanese coast guard vessel. The Japanese coast guards told the Chinese fishermen that they were trespassing Japanese water. Then the Japanese coast guards took the captain of the fishing boat into detention. This made the Chinese government upset. They told the Japanese to let go of the captain. But the Japanese would not listen. Then China decided to punish Japan by stopping all exports of rare earth elements to Japan. Since Japan is a country with much industry and little natural resources they got really scared. How would they now get raw material for their companies producing things like hybrid cars and electronics? The moral of this story is: do not buy all your eggs from one basket! The story above is perhaps not exactly what happened, but an event like this was said to be the reason why China stopped exporting rare earth elements to Japan for some time in the fall of 2010. That export stop worked as a wakeup call to the world, because China controls more than 95% of the global supply of rare earths today. Rare earths are being used in many applications today. Several “green” technologies, such as hybrid electric vehicles, wind power, low energy light bulbs, biofuels, etc., depend on them. Today China stands for more than 95% of the world supply. This has not always been the case. USA used to be the dominant supplier between 1950s and 1980s, but because of lax regulations and low costs for industry in China they have now taken over the dominance. This has become a power-broker for China, who is “threatening” to restrict and raise taxes on rare earth export. Leaving the rest of the world in anxiousness for future supply to, what seems to be, a growing demand. Though in all fairness, China will probably need more rare earths for domestic use since they are planning on e.g. expanding wind power and having more electric vehicles on the roads, also there is concern that the resources will be exhausted within a few decades with recent production rate (in case of no new discovery of deposits). However, the concern of supply shortages has compelled many nations and companies to start planning ahead and think of strategies to avoid deficits. Rare earths are actually not rare, they can be found all over the earth’s crust, but they are difficult to extract and separate, and are usually only found in low concentrations. There are many mining projects exploring feasibility of production outside of China, and the question is if they will be operable in time, before a supply crisis might strike. There are also large amounts of rare earths in- use, which are potential sources of supply. Today there is practically no recycling of these elements, since the technology for it is missing. 5
Chalmers University of Technology
1.1 Aim and objectives The aim of this report is to analyze the global product chain of the rare earth elements that are used in permanent magnets and NiMH batteries in hybrid and electric vehicles. This analysis will follow the elements from the procurement of raw material to the end in the final product and then possibly their looping back through recycling. The rare earth market has been a hot topic in the media lately, due to the global trade and political issues around these elements. The outcome of this report is supposed to be used as information basis of a daily newsletter that is sent out to people within Swedish commercial and industrial life. The report should provide an overview of the present situation, and to some extent speculate about the future, but it will be up to the reader to make up her/his mind about the future of the rare earth industry and market. The research questions that have lead this report are:  What processes and actors are involved in the product chain?  Where are those processes and actors located in the world?  Could the rare earth elements be a constraining factor for the hybrid or electric vehicle? 1.2 Thesis outline The structure of this report is described below. This is meant to give guidance to the reader for what to expect from each chapter. Chapter 2 explains how the methodology of this report was configured, combining four tools from environmental science. The reader will also be introduced to the rare earths that will be the focus in this report. And finally, a description of how the information and data for this report was collected Chapter 3 will give the reader background information to the subject of this report: what rare earths are, why they are important, in what applications they are used and why there is concern for supply shortage of them. Chapter 4 reviews the product chain for rare earths both for permanent magnets and NiMH batteries. The chapter will deal with the processes involved in the product chain. Chapter 5 presents the organization of the product chain; its included actors, their geographical locations and connections will be mapped. To some extent mass flows will be presented in this chapter. Chapter 6 will discuss the possibility of having the whole product chain in Sweden. Chapter 7 will discuss the results from chapter 4, 5 and 6 regarding the aim of the report. Chapter 8 will conclude chapter 7 in short sentences. 6
Chalmers University of Technology
1.3 Abbreviations and concepts This subchapter will explain some of the abbreviations and concepts that will be used in this report. The chapter can be used as a dictionary when reading the following text. BGS – British Geological Survey – a public sector organization responsible for advising the UK government CIS – Commonwealth of Independent States EV – electric vehicle HEV – hybrid electric vehicle HREE – heavy rare earth elements Hydrometallurgical processes – chemical processes to recover metals from ore or recycled products LREE – light rare earth elements MEP – ministry of environmental protection MOFCOM – ministry of commerce in China Misch metals – a mixture or alloy of different rare earth metals PEV – pure electric vehicle RE – rare earths – short for the 17 rare earth elements REE – rare earth elements – the elements in their natural form REM – rare earth metal – after processing the elements into metals REO – rare earth oxide – after separating the metals from the host mineral, the oxide usually contains 60-70 % pure elements, still a mix of several rare earth elements Reserve – the part of a (natural) resource that is economically recoverable with existing technology Resource – an entity of limited availability TREO – total rare earth oxides – the term includes all rare earth elements in oxide form, commonly used when talking about mining and production USGS – US Geological Survey – a governmental science organization 7
Chalmers University of Technology
2 Method 2.1 Combining four tools into one method For this report inspiration was taken from four different tools: life cycle analysis, material flow analysis, commodity chain analysis, and technology assessment. These are different tools that has contributed in different ways and helped us form a method exactly fit for our purpose. None of the tools have been used outright, therefore this part will explain how the tools have been used and in what part of the work they have been used, see also figure 1. • Product chain • Mass flows Material Life cycle flow assessment analysis Commodity Technology chain Assessment analysis • Strategic • Organization analysis of the chain Figure 1 A visual description of how our method is composed 2.1.1 Life cycle analysis In a life cycle analysis (LCA) the environmental impacts during the entire life cycle of a product is assessed, from procurement of the raw material to disposal of the used product. The idea is to measure the amount of used natural resources and pollution a product is responsible for and then to assess the consequent environmental impacts. An LCA can be used either to see where in a product's life cycle there might be environmental problems, or to compare products that provide the same function. In an LCA there needs to be clear boundaries to the system that will be assessed and what processes to include. It is not necessary to look at the whole life cycle of a product, in some LCAs the user and disposal phase might be excluded, but there needs to be definition of the boundaries for the results to be useful. The system boundaries depend on the purpose of making the LCA and the question that is to be answered. However, the system boundaries in this case only include different processes, geography and time is usually not included. (Baumann and Tillman, 2004) The life cycle thinking was used in the initial stage of making this report; the perspective was used when mapping all processes in the life of rare earths, from mining to disposal. However, the aim of the report is not at all to assess the environmental impacts of these processes. Therefore there will not be many resemblances with an LCA and the results of this report, except the life cycle 8
Chalmers University of Technology
perspective. This report, as opposed to LCA, includes the geographical pathways, and their mapping is of greater importance than the environmental impacts, which will only briefly be mentioned. 2.1.2 Materiel flow analysis MFA is a tool to systematically assess the flows and stocks of a material or substance in a system defined with boundaries in space and time. The idea is based on conservation of matter within the system; all that goes in will either stay in the system or exit from it. Within the system the pathways that connect all the intermediate and final sinks are mapped to track all the material that goes in, and detect where it ends up. Flows that enter the system, passing its boundaries, are called inputs, and those exiting are called outputs. An MFA is a quantitative analysis that uses mass balance to check that all inputs correspond to the stocks and the output. (Brunner & Rechberger, 2004) Our approach to map the product chain of rare earth metals was largely influenced by the Material Flow Analysis (MFA). The aim of this thesis was to map the global flows of rare earth metals. However, in a complete MFA, all the flows of the material is mapped, whereas this report will only follow those elements' pathways that end up in permanent magnets and NiMH batteries for electric vehicles. The whole chain of processes involved in the life cycle of these products, from mining to final product (and also recycling possibilities), will be mapped to the extent possible. But since there are too many different manufacturers, in some parts of the chains, those that are located in the same country, e.g. China will be “lumped together”. This will be clear to the reader in the parts of the report that this concerns. Also, an MFA is a quantitative tool, whereas here the analysis will be of a more qualitative nature since it would be too difficult and time consuming (not to say impossible) to find the exact numbers and pathways. A full quantification would also not be required to fulfill the aim of this report, but some quantification will be made to show the dimensions of some of the flows and stocks (e.g. mining rate and the in-use stock). The unit that will be used is rare earth oxide (REO), which is the total amount of the rare earth elements in a mineral. REO does not say what share an individual rare earth elements has in the oxide, that differs depending on location of the mine and type of mineral mined. Our system's spatial boundary will be the earth. The temporal boundary will be a year; this is mostly because data on supply and demand is usually given on a yearly basis. Different years might be reviewed to enable comparison and trace trends. 2.1.3 Commodity chain analysis The commodity chain analysis (CCA) is an extension of supply chain analysis that also takes social and societal aspects into account. In a CCA the task is to understand what social values a commodity has across its chain of production, from raw materials extraction to final product and what it is that drives each actor to be a part of that chain (Boons and Howard-Grenville, 2009). In this report we will not look at social values, but we will look at different actors along the product chain. This will be a complement to the features that we have used from MFA, since we also want to know who are participating along the chain, driving the material flow, and how the chain is organized. 2.1.4 Technology assessment The method for this report will to a large extent resemble a technology assessment (TA), in the sense that a technology with possible materials constraints will be analyzed. A TA is a concept which can use many tools to assess concerns of the development of a technology: economical, environmental and social. It is used to support decisions on strategies and policies. The concept is quite broad and there are several types of TA's, but the overall aim is to forecast the impacts and effects of 9
Chalmers University of Technology
technological change, so that they can be included in decision making. Van den Ende et al (1998) distinguishes four different types of TA:  Awareness TA – a forecasting approach to give warning about impacts and consequences of technological developments.  Strategic TA – to be used as a support for actors to make decisions about strategies or policies in regard to specific developing technologies.  Constructive TA – to influence the development of a technology in a direction that is desirable for the society.  Back casting – to create a desired future scenario and then develop technology that will lead to that scenario. The approach that is most fit for this report is the strategic TA. The use of such an assessment is to see how sustainable a technology would be in a longer term perspective, with developments, extensions and/or modifications. This report aims at investigating the question of possible materials constraint for the currently most used technology in the motors and batteries of electrical vehicles. In order to do so, the report needs to look at resources, other applications, and recycling of the materials in question. However, the aim of this report is not to predict the future and give an answer for the sustainability of these technologies, but rather to provide a view on the present state of the supply and demand of materials used in the technologies the way they are today. It will then be up to the reader to judge if sustainable. 2.1.5 The method of this report Working on this report has involved a lot of mapping. As in LCA we have mapped all the processes in the making of the specific products, from “cradle to grave”, as in MFA we have mapped how the elements flow on a global scale, and as in CCA we have mapped the actors that are involved in the product chain. This will then make up basic data for a TA inspired discussion about selected rare earths. The product chains used in this report is made up of eight respectively seven steps, as shown in chapter 4. For each step in these chains the involved processes and actors will be described. The quantitative flows and stocks in the chains will not be balanced; rather they will in some cases be estimated, to give the reader an appreciation of the volumes. Geographical sites and flows will also be described, to provide an understanding of where in the world the different processes in the chain takes place. The actors in the chain will be briefly presented as well as the relations and cooperation between them. The perspective in this report is on the product chain. This report will provide a snapshot of a product chain that potentially is about to change. By using the information that is provided the reader will have some knowledge about the present situation to easier understand the implications for future changes. 2.2 The selection of rare earths in this report In this report we are focusing on a selection of rare earth elements that are used for two specific applications. Permanent magnets can contain the elements: neodymium, praseodymium, 10
Chalmers University of Technology
dysprosium and terbium, and the NiMH batteries contain lanthanum, praseodymium, neodymium and cerium. However, in the first parts of the product chains it is very difficult to distinguish individual elements, since rare earths always exist together in a host mineral. They are quite difficult to separate, so this is made in the third step of the product chains (presented in chapter 4). Any data on amounts of REO does not specify the amount of individual elements, since this differs between locations and minerals. Distribution of the individual elements in the ore from some of the different mines that are in or close to production can be found in the appendix. 2.3 “Speculations” about a possible product chain in Sweden Chapter 6 of this report is of less scientific degree, as it contains speculations from us along with consultations with some experts within the area of metallurgical industry and recycling in Sweden. The chapter prospects future possible scenarios for Sweden as a participator in the global and domestic product chain for rare earth metals. Most scenarios are based on the fact that Sweden is a well-developed country when it comes to mining of metals such as, iron, copper and silver, and also the metallurgical industry with separation, refining, handling and recycling of such metals. For these reasons it could be interesting to speculate in the possibilities of future participation in the product chain. The information and speculations in this chapter is gathered through telephone conversations with people at Rönnskärsverket (which is a recycling plant for steel, iron and copper and a part of the company Boliden AB), Höganäs AB (which is a company specialized in iron and metal powders) as well as professors at University of Luleå, LTU, and Chalmers University of Technology, Gothenburg. The future scenarios mentioned in the chapter is therefore only suggestions of possible acts of the product chain that Sweden might be able to participate in since we do have knowledge within many other metallurgical areas with similar aspects as for the product chain of rare earth elements. 2.4 Collection of data and information For this report different sources of information have been used in different phases of study. Firstly, to get a sense of the field, information was sought in newspapers, blogs written by experts in the area, governmental websites, home pages of nongovernmental organizations, and websites of companies all the way from mining companies to car manufacturers, encyclopedias and also searches on youtube. This gave a wide perspective on the area, including basic facts, political issues, technological developments and the race for production of rare earth elements. In a second phase the information search was more focused as the aim of the report had become clearer. But still, the same media was used. However, starting to know the area and different actors in and around the chain of rare earths, the sources were evaluated more thoroughly for reliability. Some of the companies we were looking at provided useful information on their websites and in their annual reports; others were contacted by e-mail or phone calls. A few interviews have been held, in order to enable some discussion with valuable contacts. For chapter 6 there was not much published information to be found, instead experts (see chapter 2.3) was contacted to tell about their thoughts, opinions and the knowledge they had in their area. The aim of this report is to provide a picture of the current status of the product chain of rare earths. In order to get the most updated information and latest news, internet sources were used to a large extent. The reliability of internet sources is always dangerous ground, which we are well aware of. Therefore it is important that the reader is aware of this and also reads this report with that in mind. We have valuated our sources and only used the ones that we have found reliable according to the 11
Chalmers University of Technology
3 Description of rare earth elements This chapter will explain shortly what rare earth elements are, how they are produced, and in which applications they are needed. There will also be a part concerning the environmental impacts of their production; since they are used in so many green technologies it is important to also be aware that there are environmental consequences associated with them as well. The end of the chapter will lead the reader to the focus of this report: rare earth metals used in electric vehicles. Note that when this report talks about electric vehicles, hybrids are also included. The product chains that will later be analyzed in the report will be presented, and also the concerns related to their functioning. 3.1 Context Since the first rare earth elements were discovered outside Stockholm in the late 18th century their unique qualities has become important in many technologies. Several of the “green” technologies today use rare earths, because they can increase energy efficiency and decrease weight in many applications. When they are used in windmills they help produce CO free electricity, and when that 2 electricity is used to drive electric vehicles we can avoid using fossil fuels. This report will focus on those elements that are used in the motor and battery of hybrid and electric vehicles. Later in the report they will both be referred to as electric vehicles, since they partly of fully are driven by electricity. The rare earth production has, up until now, had a rising trend. The world demand for rare earth elements was in 2010 approximately 134 000 tons (Humphries, 2010). That is an 8% increase from the demanded 124 000 tons in 2008, which in turn is a 45% increase since 2003, the biggest importers are Japan, USA, Germany, Austria and France (China can still supply their own needs as well as others') (BGS, 2010). The applications are many: magnets, ceramics, phosphors, fluorescent lamps, lasers, glass coloring, catalysts, and more. Products containing rare metals are for example: electric vehicles, hard drives, wind turbines, nuclear reactors, laptops, mobile phones, and TVs. They are also important for military uses, e.g. smart bombs, missile guidance systems, jet fighter engines, mine detection systems, and communication systems (Humphries, 2010). Figure 2 shows some of the other applications for the REEs that are of significance for the aim of this report. These other technologies can be considered as competing technologies; competing for the same material. When forecasting future demand and availability of the materials needed in permanent magnets and NiMH batteries the growth (or decline) of these other technologies must be considered. Lanthanum Neodymium Praseodymium Dysprosium Terbium Cerium •Catalytic •Hard disks •Motors •Lasers •Fuel cells •Catalytic converters •Mobile •Fibre optics •Nuclear •Flourescent converters •Fuel additives phones •Coloring glass reactor rods lamps •Glass •Glass •Loudspeakers •Carbon arc •Hard disks •Low energy production •Fuel cells •Magnetic lights •Dosimeters lighting •Steel •Steel refridgeration •Solid state •Carbon arc •Capacitors •Lasers devices lights Figure 2 Different applications for the rare earths focused on in this report 13
Chalmers University of Technology
Rare earth elements have been a hot topic in politics and media lately, especially after the temporary Chinese export halt to Japan in the fall of 2010. The demand for these elements have increased substantially in the last 15 years, and there are strong indications that it will continue to rise as the technologies using rare earths, perhaps most importantly hybrid- and electric vehicles and wind power turbines, are growing (BGS, 2010). However, it is uncertain if the supply will meet the demand in the future. The fact that China dominates the production of these metals so strongly makes the rest of the world absolutely dependent on their exports today. Lately China has cut their export quotas; in 2010 the export had decreased by 9.3% compared to 2009, and in the first half 2011 the export is supposed to be 35% less than the same period in 2010 (China Mining, 2011). This cannot be explained by decreased production in China, because the output quota is set 5% higher in 2011 than previous year (china.org.cn, 2011). There are perhaps many reasons why China is cutting the export, and it is not within the scope of this report to deeply investigate this matter. However, there are indications that China will be using larger amounts of rare earths themselves in the future, which would mean less to export. The Ministry of Commerce (MOFCOM) in China recently posted news on a drafted plan, made jointly by four Chinese ministries, that suggests that during 2011-2015 China should have the capacity to produce one million non-fossil fueled vehicles, out of which 50% should be hybrid or electric vehicles. Until 2020 this number should increase to 5 million, since the statement says that China aims to be the leading nation in the “new-energy vehicle” sector. (MOFCOM, 2011a) Assuming that the same technology will be used in these vehicles as today, this means that an increased production of electric vehicles will consume large amounts of rare earth metals. China also plans to have more wind power in the future, which is a technology that in some cases uses large amounts of the same rare earths as electric vehicles. There are no certain estimations for future demand for rare earths. Mark Smith, the CEO of Molycorp Minerals, anticipates a demand exceeding 200 000 tons by 2015 (Smith, 2010), which is an estimation that he shares with several exerts in the field, who suggests roughly the same number. The Chinese rare earth expert Chen Zhanheng, The Chinese Society of Rare Earths, predicts a 15% increased demand annually in good economic conditions. The demand outside of China would be 80 000 tons, according to him (Zhanheng, 2011b). Something that is certain is that the demand for the individual elements will vary a lot. Smith forecasts that the elements most likely to be in short supply vis-à-vis the demand will be neodymium, europium, terbium, and dysprosium. On an average the prices of rare earths has increased significantly during the last years. There is no recognized exchange for rare earth elements, they are usually bought through trading companies and their prices are highly influenced by the producers (BGS, 2010). The price of rare earths is affected by several factors, e.g.: the different elements are valued individually (the HREE are usually more expensive since they are less abundant), the prices fluctuates depending on the prevailing demand, and the elements are difficult to separate and therefore the purity level also affects the price. According to MOFCOM (2011b) the prices for two of the metals used both in permanent magnets and NiMH-batteries, neodymium and praseodymium, have increased by 125% of previous year in 2011. For dysprosium, also used in the same applications, the increase was 104%. China will also put a higher tax on mined rare earths, starting in April 2011, which will mean higher production costs. 14
Chalmers University of Technology
3.2 Elementary information “Rare earth elements” is a collective name for 17 chemical elements including scandium, yttrium and the 15 lanthanides occupying numbers 57 to 71 in the periodic table (see figure 3). These elements often occur in the same ore and have similar properties: soft, malleable, ductile and reactive (especially at elevated temperatures) (Gupta & Krishnamurthy, 2005). Most of the elements are also magnetic. The lanthanides are usually divided into two groups: light rare earth elements (LREE) and heavy rare earth elements (HREE). The LREEs are La-Eu, and the HREEs are Gd-Lu (BGS, 2010). Yttrium is usually considered to be a HREE (Gupta & Krishnamurthy, 2005 ). Scandium is a bit more uncertain since it is not always considered a rare earth at all (to confuse the reader even further, the opinions on which elements are considered rare earth elements also differ between experts, but in this report we have chosen to believe the opinion of the majority, and the latest information). The definition of which elements belong to which subgroup differs in the literature, and sometimes the elements are divided into three subgroups, in that case samarium to dysprosium are called medium rare earths (Gupta & Krishnamurthy, 2005 ). The lighter elements are usually more abundant than the heavy since they are more incompatible, also elements with even atomic numbers tend to be more abundant in the earth's crust. (BGS, 2010) Figure 3 The periodic table with the rare earth elements highlighted (BPC, 2010) Despite their name these elements are not really “rare”. They can be found in various places around the Earth's crust. Cerium is the most abundant rare earth, also yttrium, neodymium and lanthanum has abundance comparable with common industrial metals such as chromium, copper and nickel. The least abundant rare earths (tellurium and lutetium) are still nearly 200 times more abundant than gold. (USGS, 2011) The problem in producing these metals is that they are usually found in very low concentrations, which often makes the findings economically unfeasible for mining. Since they are chemically similar to each other they can easily substitute for one another in crystal structures, they usually occur together in the same mineral and are relatively difficult to separate. (BGS, 2010) 15
Chalmers University of Technology
So, in spite of their name, there are plenty of rare earths and they do exist in many places all over the world. In nature the rare earths do not occur as metals, but as a part of host minerals where they substitute for ions. There are at least 200 minerals that contain rare earths, but only a few are used for rare earth production, the most common ones are monazite, bastnäsite, and xenotime. (BGS, 2010) These three usually contain approximately 65-70% REO. Monazite is the single most common rare earth mineral (USGS, 2002). It has the formula (Ce, La, Nd, Th)PO . As can be seen in its formula it contains 4 radioactive thorium, and also uranium, which can present a problem for mining if the weight percent is high. It can be found in many different environments, from rocks to beach sands. Monazite contains mostly LREE, and low concentrations of HREE. Bastnäsite also contains mostly LREE, but very low concentrations of radioactive elements, which is good for mining. This is the main rare earth mineral in e.g. the large deposits in Bayan Obo, China, and Mountain Pass in California, USA. Xenotime is richer in HREE than the other two, and has low to no content of the radioactive elements. It is found in heavy mineral sands in e.g. Malaysia and Brazil. (Gupta, 2005) Table 1 below shows the content of rare earth oxides, thorium and uranium in these three minerals. Rare earth elements are also produced from ion adsorption clays. These clays are especially rich in HREE, but they are only known to exist in China and Kazakhstan, since they require special weather and environmental conditions to form, which is the reason why China is the main supplier of HREE today. (USGS, 2010) Mineral Formula REO (wt%) ThO (wt%) UO (wt%) 2 2 Monazite (Ce,La,Nd,Th)PO 35-71 0-20 0-16 4 Bastnäsite (Ce,La)(CO )F 70-74 0-0.03 0.09 3 Xenotime YPO 52-67 -- 0-5 4 Table 1 The content of REO, ThO2 and UO2 in the three most commonly mined RE minerals (source: USGS, 2010) Since the concentration of rare earth elements in the host minerals are usually very low, the feasibility of mining them economically is often a problem. However, the rare earth minerals are often found near other minerals containing e.g. iron, copper, uranium, tantalum, and zirconium. The rare earths can then be mined as a co-product (the costs of mining and production are shared by the main and co-product) or a by-product (the costs of mining and production is covered by the main product). (Gupta, 2005) As prices for rare earth metals are increasing, more mining companies are starting to realize that their tailings, containing rare earths, could be valuable. That is the case for LKAB in Sweden, who are now investigating possible future processing of tailings from their iron mining. 3.3 Environmental issues The environmental consequences of the rare earth metal production are many, ranging from mining to separation: e.g. energy use, chemical use, land destruction, waste production and radioactive waste. One of the main reasons why China was able to take over and to dominate the world production of rare earths is that the regulations was lacking or was so much slacker than in other countries that historically has been large producers of rare earths (e.g. USA and Australia), which meant that they could offer lower prices (Houses of Parliament, 2010). In China problems such as soil erosion, water loss, and pollution of soil and ground water has led to destruction of much land area around the mines because of insufficient control. The extraction of rare earth metals is energy intense, and since most of it takes place in China where coal power plants produces a large share of 16
Chalmers University of Technology
the energy it creates emissions. Up till now there have not been limits for emissions or production, and in the south of China there have also been problems with illegal mining. (Zhanheng, 2010) The smuggling of rare earths has been a huge problem in China. In 2008 as much as 20 000 tons of rare earths was reported to have been smuggled out of the country illegally, which was a third of the total exported rare earths from China that year (Hurst, 2010), numbers for illegal mining are naturally just estimations, not facts. Illegal mining means that the resources will be depleted faster, and that this mining is operated without any regulation at all which has large implications on the environment and also affects the prices of rare earths. In 2011 the Chinese government started to take action for protecting the environment and the resources, and thwart smuggling. This includes setting new standards and rules for the rare earth industry, e.g. lowering the limits for the allowed emission levels, limit production quotas, encouraging mergers and acquisitions, and reducing the number of mines and processing plants. Over all China wants to increase the governmental control over the rare earth industry and consolidate it; in Baotou, where the largest rare earth reserves in China are, the number of producers have been reduced from 150 to 18 during the last years. New licenses for rare earth mining has not been issued since 2006, and to fight illegal production those who produce less than 8000 tons a year will have to close or merge with another producer. (MEP, 2010) The new and stricter environmental standards in China will come in force in the second half of 2011, but the companies will have two years’ time to upgrade their technology before they are banned from the market. This will force many companies to invest more in environmental protection, and smaller companies might be forced to merge with bigger ones to cope with the costs. The market is concerned that the prices of rare earths from China will rise because of this. (China Daily, 2011) 3.4 Rare earth elements in hybrid and electric vehicles Rare earth elements are used in many parts of an electric vehicle, e.g.: motors, NiMH batteries, windows, glass, catalytic converters, electronic equipment and more. However, this report will focus on the permanent magnet motors and the NiMH battery. The world's most sold hybrid vehicle model, Toyota Prius, uses somewhere around 10 kilos of rare earths, see table 2. Permanent Magnet NiMH Battery Lanthanum - 5400 g Neodymium 300-600 g 900 g Praseodymium 75-150 g 1800 g Dysprosium Smaller amounts - Cerium - 900 g Terbium Smaller amounts - Table 2 Rare earth content in the NdFeB permanent magnet and NiMH battery in a Prius (Sources: Bubar, 2011 and Maruo, 2011) The permanent magnet in electric vehicles weights between 1 and 2 kilos, approximately 30% is neodymium (Bubar, 2011). The neodymium can partly be replaced with praseodymium; the alloy is then called didymium and contains ca 75% Nd and 25% Pr. To increase the temperature stability and coercivity of the magnet, smaller amounts of dysprosium or terbium can be added. The NiMH battery in Prius vehicles has been improved over time and nowadays they are significantly smaller and lighter than they were when it was first launched in 1997. Approximately the battery pack weighs 37 kg today, the amount of rare earths in the anode material might be a little more than a quarter of that. The material in the battery is a lanthanum misch metal, the ratio between the included elements is: 60% La, 20% Pr, 10% Nd, and 10% Ce. (Maruo, 2011) 17
Chalmers University of Technology
The Toyota Prius was the first hybrid vehicle to be mass produced. Since its lunching in 1997, more than 3 million Prius cars have been sold globally (Toyota, 2011). The Prius has been far more successful than any other car manufacturer’s hybrid car model, so to estimate the number of hybrids sold globally since 1997, a little more than 4 million might be enough. The future outlook for hybrids and pure electric cars seems bright. However, it is difficult to say how many of these will use NiMH batteries, since Li-ion are expected to be the more popular choice in the future. Even Toyota will have to use Li-ion in their plug-ins, since NiMH is not good for PEVs. Most commonly there are two types of permanent magnets. One of them is the Nd-Fe-B magnet, often referred to as neodymium magnet. The second type is the Sm-Co type magnet. The samarium- cobalt magnets were introduced in the 1970’s and this lead to an expansion in the usage of permanent magnets. The intensified use of magnets increased knowledge and helped to develop another compound with even better magnetic characteristics, the neodymium type (Sastri, et. al, 2003). The characteristics of these two types are somewhat similar but the neodymium magnets have a higher magnetic energy density which makes it a better choice in wind turbines and hybrid vehicles. Due to the high energy density in permanent magnets, they often replace electromagnets in many types of applications (Sastri, et. al, 2003). This is due to the fact that no external energy source is needed to provide the magnetism. 18
Chalmers University of Technology
4 The structure of the product chain In this chapter the technical approach of the rare earth product chain is presented in order to explain the different stages. The first three steps are shared for both permanent magnets and NiMH batteries, and will be dealt with in subchapters 4.1 and 4.2. After that the two chains part and are dealt with in separate subchapters 4.3 and 4.4. Finally, they merge again in the last step, recycling and recovery, which is chapter 4.5. Alloys and Final Recycling Mining & separating Oxides Metals Magnet Magnets Components Technologies and recovery Powders Mining and Battery Incorporation Recycling and Oxides Metals Components separating assembly in vehicle recovery 4.1 Mining and Separation The rare earth deposits around the world differ in nature which calls for different methods of mining and extraction of the elements. In general, the extraction can be either open pit mining or underground mining. The mined ore usually has a very low grade of rare earth content, sometimes as low as just a few wt%, therefore it is common that there is a facility (sometimes called a concentration plant) close to the mine to concentrate the REE and separate it from the gangue minerals, and possibly other products (BGS, 2010). In Bayan Obo, the district that has the world’s largest production today, the open pit method is used for mining. The separation process contains grinding of the ore and separating through flotation with different chemicals (alkalis and acids). The concentrate then contains a little more than 60 % REO. The separation processes are usually quite complicated, and as mentioned different ores containing different minerals require specific methods. Leaching, electrostatic and magnetic separation, and filtration are other processes that can be included. The beach sands in India already have a very high concentration of REE and do not need any treatment in this step (Gupta & Krishnamurthy, 2005). 4.2 Oxides and metals The oxides must be treated further in order to increase the RE concentration or to be used as metals. This procedure is very difficult since the oxides are very stable. In order to obtain more pure forms of rare earth metals the oxides have to be reduced, which can be done in different ways. The basic principle is to use a chemical reducing agent to remove the impurities. These processes are technically difficult and demands large amounts of electricity. The metals can after the initial refining stages reach a purity of around 98-99%, and then be further refined to increase purity with another half percent. The metals are often formed and sold as ingots which at later stages can be grinded into powders for further usage. (Gupta & Krishnamurthy, 2005) 4.3 Permanent Magnets This subchapter will continue along the product chain and explain the technical steps from metal to magnet. Alloys and Final Recycling Mining & separating Oxides Metals Magnet Magnets Components Technologies and recovery Powders 19
Chalmers University of Technology
There are two different types of neodymium magnets. The ways of producing these two types of magnets are somewhat similar but have some differences. The sintered magnets are made from metallic powder which is processed and grained from metal ingots into a fine powder with a particle size smaller than ten µm. The powder is then magnetically aligned by a magnetic field and pressed in a vessel and then sintered in a sintering oven, where the powder melts and get the wanted shape of a solid magnet (Kuhrt, 1995). When producing bonded neodymium magnets, the metallic powder is mixed or capsuled with a binder of some kind and then molded into shape. The binder can be of many types, for instance polymers, nitrile rubber, vinyl, nylon, Teflon, polyester or thermoset epoxies (Buschow, 2001). This bonding makes the magnet more versatile according to shape and size and can then be used in smaller applications and in various shapes (Bauer, et al., 2010). However, the added binding material also decreases the magnetic energy density which makes bonded magnets less attractive in larger applications (Buschow, 2001). The produced magnets are when finalized assembled in a larger product, for instance a vehicle, motor, wind power plant or other devices. 4.4 NiMH batteries This subchapter will continue along the chain from metals to NiMH batteries. Mining and Battery Incorporation Recycling and Oxides Metals Components separating assembly in vehicle recovery The product chain of NiMH batteries continues, after the metal production step, with production of components. The metal used in NiMH batteries is a misch metal containing lanthanum (60%), praseodymium (20%), neodymium (10%), and cerium (10%) (Maruo, 2011). The next step is the assembly of the battery, which in the sixth step is incorporated into the vehicle. The exact processes for how to produce a battery will not be explained in this report. 4.5 Recycling and recovery The last step in both product chains is the recycling and recovery step. Future-wise, this might be one of the most important steps since only small amounts of these products are recycled or recovered today. Alloys and Final Recycling Mining & separating Oxides Metals Magnet Magnets Components Technologies and recovery Powders Mining and Battery Incorporation Recycling and Oxides Metals Components separating assembly in vehicle recovery The recycling processes regarding rare earth metals are quite new and still in the developing stages. Most processes are chemical but there are also possibilities to extract them mechanically and/or combinations of both mechanical and chemical processes. There are also possibilities to even reuse some products, like permanent magnets, with minor modifications (Lehner, 2011 and Ekberg, 2011). Regarding NiMH batteries commonly used in cars, these types of batteries contain misch metals which for these batteries are a mixture of nickel and rare earths lanthanum, cerium, neodymium and praseodymium. These rare earths may be recycled through a procedure of mechanical and chemical separations (Ito, et. al., 2009). Ito et. al. along with Japan Oil, Gas and Metals National Corporation has come up with a method starting with mechanical separation of the battery to recover all the 20
Chalmers University of Technology
individual components after being crushed, and thereafter the components containing rare earths and other metals are chemically treated with dissolution and melting in order to separate the metals. The method was performed on two different types of batteries, cylindrical and prismatic shaped batteries, and includes many different steps that are quite advanced. The parts of the battery that are to be recycled are the anode and cathode parts. The method shows that it is possible to recover some of the rare earths in the batteries and that the purity of the recovered product was high and may be treated further in chemical stages in order to increase purity. However, the shape of the battery and how it is assembled can affect the efficiency of the recovery processes (Ito, et. al., 2009). When producing neodymium magnets, especially for computer hard drives, approximately 50% of the used material is discarded as waste or scrap. This depends much on the fact that neodymium easily forms very stable compounds with other elements, of which oxygen is an example, and therefore the scrap cannot be recovered or reused directly without being treated first (Takeda, et. al., 2005). According to Takeda, et. al., 2005, there is a process to extract neodymium by using magnesium as an extraction medium. This liquor that contains the rare earths can later be treated with hydrometallurgical methods to extract the rare earths from the liquid. Similar methods as for the permanent magnets in hard drives and the NiMH batteries, can be used to recover rare earths from used computer monitors (Resende and Morais, 2010). There are some different methods for recovering rare earths from spent products but the basic method is to first perform mechanical separation and then further processing with chemicals through dissolution and refining processes. 21
Chalmers University of Technology
5 The organization of the product chain The product chain for the permanent magnets and the NiMH batteries are joined in the first three steps, since rare earth elements occur mixed together in their host mineral. But after the separation and refinement steps they part. This chapter follows this by presenting the first three steps in separate subchapters and then there will be two subchapters, one presenting the rest of the product chain for permanent magnets and one for NiMH batteries. The chapter will end with a subchapter about recycling, where the elements are again joined for a loop back into the product chains. 5.1 Mining and separation This subchapter will provide an overview of the mining situation in the world: today, in the next three years and longer term. Beginning with the estimated world reserves and their locations, continuing in current production and existing mines, the chapter will later mention some projects for new mines. As mentioned in chapter 4.1 it is most common that separation takes place near the mine, by the same mining company. 5.1.1 Reserves and current production According to the USGS (2010) the world industrial reserves of total rare earth oxides were an estimated 99 million metric tons in 2009. The country with the largest reserves is China, with about 36 million metric tons of REO. On second and third place comes the CIS and the USA, with 19 and 13 million tons respectively (see figure 4). However, the precisely accurate size of the world reserves is uncertain and changing. More recent estimations imply that they could be larger in 2010. Zhanheng (2011a) writes that Brazil, which according to the USGS only has a small share of the reserves, actually has the world’s largest, about 52.6 million metric tons of REO. Nevertheless, USGS (2011) has not changed their figure for the Brazilian reserves in 2010. The figure for China’s reserves has increased to 55 million metric tons. Figures for the other countries are unchanged or, as for Australia lowered in 2010 (see figure 5). The new figure for world total reserves of REO is then almost 114 million tons. One reason for widely differentiating figures is how the reserves are defined. Figure 4 World reserves of REO in 2009 (source: USGS, 2010) 22
Chalmers University of Technology
Figure 5 World reserves of REO in 2010 (source: USGS, 2011) In the figures above, the “other” category includes Canada, Greenland, South Africa, Namibia, Mauretania, Burundi, Malawi, Vietnam, Thailand and Indonesia. The reserves are, in this data from USGS, defined as that part of the reserve base that could be economically extracted or produced with existing technology. Also these are the reserves known today, reserves are however not fixed; their size can change depending on prices and costs, technological development and new discoveries. There might still be undiscovered deposits in the world that could provide future production. One example of this is a newly discovered rare earth and niobium deposit in Afghanistan with an estimated value of $89 billion (Najafizada, 2011). Other examples will be presented in 5.1.3. The production of rare earths is today mostly concentrated in China, where about 95% are mined, but historically the domination has changed between several countries. Until the 1940s the main suppliers of rare earths were India and Brazil, after that production also started in Australia and Malaysia. The Mountain Pass mine in California, USA, started producing in the 1950s, and during the 1960s to the 1980s USA were the main producers in the world. (BGS, 2010) China did not begin to produce rare earths until the 1980s, but because of lower production costs and less strict regulations they could sell their products much cheaper and possessed dominance as RE producer in just a few years (Haxell, 2002). The historical world production of rare earth oxides can be seen in figure 6 (Haxell, 2002). 23
Chalmers University of Technology
Figure 6 Historical global production of REO (source: Haxell, 2002) The latest published numbers for world production comes from 2009, shown in figure 7. The main producer is China, followed by India, the Commonwealth of Independent States, Brazil, and Malaysia. The total production that year was 126 230 tons according to the USGS (2010). Figure 7 World production of REO in 2009 (source: USGS, 2010) 5.1.1.1 China Even though China only has about 36% of the world's rare earth reserves, according to 2009’s numbers, it has about 95% of the production (USGS, 2010). The mines are distributed in ten of the provinces: Inner Mongolia, Shandong, Sichuan, Zhejiang, Jiangxi, Hunan, Fujian, Yunnan, Guangxi, and Guandong (Buchert et al, 2011). The Bayan Obo area in the north has China's largest reserves of REO, about 90 % (Baotou National Rare Earth Hi-Tech Industrial Development Zone, n/d). The rare earth minerals in this area are monazite and bastnäsite (Buchert et al., 2011), which is rich in light rare earth elements (see table in appendix). Sichuan and Shandong also produces mostly LREE from 24
Chalmers University of Technology
bastnäsite (USGS, 2010). The reserves in the seven provinces in the south are found in ion adsorption deposits; these are especially rich in the less common heavy rare earth elements (BSG, 2010). According to Vulcan (2008), the ion adsorption clays of southern China contains about 80% of the world reserves of HREE. The province Inner Mongolia in the north of China includes the Baotou city area, which is a highly industrialized area where most of the world's rare earth metals are mined. Rare earths are not the main product here, but mainly produced as a by-product to other minerals such as iron, niobium, gold, magnesium, copper and coal. The foundation of the Baotou industry is iron and steel, but the key industry is rare earths (Baotou-China, n/d). The largest company is Baoutou Steel Rare-Earth Hi- Tech Co. Ltd. (Shen, 2011); in excess of their main product, steel, the company also produces rare earth metals. Exactly how many mining companies there are in total in this area is difficult to say, and with new regulations and policies in China that will come into force 2011 the industry might change, according to plans the number of producers will decrease. In the south of China there are many companies of different sizes in the above mentioned provinces, some of the biggest are: Jianxi Copper, China Minmetals, and Chinalco, but the industry is poorly controlled and reportedly there is also a lot of illegal mining operations (Chegwidden, J. and Kingsnorth, D., n/d). In February this year (2011) the Chinese Ministry of Land and Resources announced that, in order to protect resources and the environment, the mining in the Jiangxi province will be regulated into 11 state owned mining zones, and in the Sichuan province there will be two mining zones. (MOFCOM, 2011c) The mining industry has been causing much pollution and environmental damage in China. To address this the Ministry of Environmental Protection (MEP) has set emission standards for the rare earth mining industry, these will be the first standards especially targeted at this industry, meant to solve the problems of illegal mining, waste of resources, outdated technology, pollution and soil erosion. The emission standards will go into effect in October 2011 and there will be a grace period of two years for existing companies, but the costs might be too high for smaller companies, which might have to close down or join with bigger companies. Also, in order to protect resources, the government will introduce quotas for production and export. The effect of this could be higher prices on rare earth metals from China, as the prices would then reflect the environmental costs and restricted production. (MEP, 2011) During 2010 the Chinese government set guidelines that encouraged consolidation and merger in the mining sector. The guidelines are said to aim at reducing the current 90 rare earth firms to 20 by 2015. Also, the licensing for mining production was stopped in 2006, and since then many producers have shut down. In a publication from MEP (2010) it is said that in Inner Mongolia there is now only 18 producers remaining from what used to be 150. 5.1.1.2 India In India there are a couple mining companies that produce rare earth elements, e.g. the governmental undertaking Indian Rare Earths Limited that has mines in Chavara, Manavalakurichi, Oscom and Aluva (www.irel.gov.in), and the Kerala Minerals and Metals Ltd. that extracts monazite from mineral rich beach sands. However, the reliance on India as a producer of rare earth metals is uncertain. According to an article in Wall Street Journal (Roy, 2010) India lacks technology to process the ore into metals, but according to the national Indian newspaper The Hindu (Dikshit, 2011) India is considering to start export rare earth chloride to the Japanese company Toyota Tsusho instead of 25
Chalmers University of Technology
processing it themselves, which would keep them in business. India is the world's second largest producer of rare earths, but has only 3% of the world total reserves. (USGS, 2010) 5.1.1.3 Commonwealth of Independent States The current production of rare earth elements is located in the Lovozero district in Russia (USGS, 2010). The mineral containing the rare earths is Loparite. The Loparite ore is cracked in Solikamsk, the rare earth carbonate is then shipped to the Silmet plant in Estonia for separation (Serova, 2011). Silmet converts the carbonate into oxides, chlorides, alloys or metals (Vereschagin, YU.A. et al., 2006). The CEO at Silmet, David O'Brock, confirms that they “produce pure elements for the companies who make [electric vehicle permanent magnet and NiMH battery] components” (Serove, 2011). However, Silmet also buys rare earth raw materials from “elsewhere”, which indicates that the CIS production is not enough. 5.1.1.4 Brazil There is no current production of rare earths in Brazil (Cilene, 2011). Indústrias Nucleares do Brasil's main business is to produce uranium fuel for nuclear plants, but they also used to produce iron titanate, titanium oxide, zircon silicate and rare earth products from mineral sands (INB, 2011). Now INB are exporting monazite to China (Cilene, 2011). The only mine in production is Buena in Rio de Janeiro with a modest annual production of 650 metric tons of REO (USGS, 2010). According to USGS (2010) data from 2009 Brazil would only have 0,05% of the world reserves of total rare earth metals. Chen (2011) however, notes that this might be a vast underestimation, as data for 2010 suggest that Brazil might actually have the largest reserves in the world. Of course, the size also depends on how reserves are defined. USGS defines reserves as the part of the reserve base that could be economically mined or produced with existing technology. 5.1.1.5 Malaysia The REO reserves in Malaysia is an estimated 30 000 tons, they consist in the light rare earth mineral monazite, and xenotime that contains more of the heavy rare earth elements like yttrium. Both these minerals contain some radioactive elements (see chapter 3.2). The rare earth minerals mined in Malaysia are mainly byproducts from tin production. There used to be two processing plants in Malaysia: Asian Rare Earth (ARE) and MAREC. ARE was established by a joint venture between Mitsubishi Chemical Industries and BEH Minerals. The rare earth containing minerals were processed into a concentrate of about 50%, which was then shipped to Japan for further purification. MAREC produced yttrium oxide from xenotime. However, both these plants were shut down because of problems with the radioactive waste from the production. The clean-up is still ongoing. (Latifah, 2002) A new processing plant is currently being built by the Australian rare earth company Lynas Corp., production is supposed to start in the second half of 2011. This plant will process the rare earth concentrate from the Lynas’ Mount Weld rare earth mine in Australia. Taken history into account the Malaysian government now puts strict regulations for the plant, to avoid radiation problems in the future. Lynas counteracts by saying that the new plant uses advanced, safe, modern technology, and also that the rare earth minerals at Mount Weld contains much less radioactivity than those from Malaysia. (AP, 2011) 26
Chalmers University of Technology
5.1.1.6 USA The world’s third largest REO reserves are found in the United States. Historically the US used to be the dominant supplier of REE. The country’s largest deposit is the Mountain Pass, where mining started in 1952. The peak production, 20 000 metric tons, was in 1990. Soon after, in 1992, the mine was closed for reasons e.g. environmental problems and competition from China. (USGS, 2010) Currently there is no mining in the US, but some processing of stockpiled ore from the mine has occurred since then. There are several projects for mining rare earths in the US, in chapters 5.1.2 and 5.1.3 two of these will be described more closely with planned output and time lines. 5.1.1.7 Australia Rare earth production started in Australia in the late 1940s where the source was mostly beach sand (Gupta & Krishnamurthy, 2005). There are significantly large rare earth deposits with high concentration of minerals. According to BGS (2010) the Mount Weld deposit has a resource of 12.24 million tons with a concentration of 9.7% REO, which can be compared to Bayan Obo’s concentration of 6%. Alkane Resources Ltd. (2009), owner of the Dubbo deposit, also reports that there are high concentrations of the less common heavy rare earth elements. There are a couple of far come projects in Australia, with planned production within the next few years, more about this in the next part of this chapter. 5.1.2 Mining expected within 3 years This chapter will introduce some of the most developed projects for new rare earth oxide mines. The projects will be presented with planned production capacity, timeline for production start, and owner. Most information is taken from the respective company's webpage. With increasing demand and prices for rare earth metals there are several projects exploring new RE deposits with feasible mining. The projects are spread over the world, including at least four continents. A few projects have already come a long way and are expected to start production within the next three years. These projects are listed in table 3 with some comments in the following text. Name Country Company Earliest production Production plan (t/y) Mount Weld Australia Lynas 2011 11.000-22.000 Dubbo Australia Alkane Resources 2013 2.600 Nolans Australia Arafura Resources 2013 20.000 Mountain Pass USA Molycorp Mining 2012 20.000 Orissa India Toyota Tsusho + India 2011 3.000-4.000 Rare Earth Dong Pao Vietnam Toyota Tsusho + Sojitz 2012 2.000-3.000 + Vietnamese Government Steenkampskraal South Great Western 2013 2.700 Africa Mineral Group Total Production 75.000 (rounded up) Table 3 Mining projects with expected production within 3 years 5.1.2.1 Australia Australia has a few interesting projects with far come plans. The one closest to production is Mount Weld, owned by Lynas Corporation, with an expected start in 2011. Lynas claims to have the richest source of rare earth element oxides in the world, the mineral is monazite with mostly LREE (see appendix). The production is planned to start in two phases: in the first phase the amount will be 27
Chalmers University of Technology
11 000 TREO, and in the second phase it will be twice as much. The ore from Mount Weld will firstly be processed in a concentration plant 1.5 km from the mine. Here a concentrate of 40% REO will be the product from four times as much ore, which will thereafter be shipped to Malaysia for separation into individual metals in a processing plant. The processing plant, Lynas Advanced Materials Plant, in Malaysia is planned to receive its first feed in the third quarter of 2011. (Lynas Corporation, 2011) The Dubbo Zirconia Project outside of New Wales, Australia, is held by Australian Zirconia Ltd., a subsidiary wholly owned by Alkane Resources Ltd. This project started to explore the Toongi intrusive in 2000, which contains several elements: zirconium, niobium, tantalum, and rare earth elements. The timeline is to start production during 2012. Initially the production plan is to mine 400 000 tons of ore annually, which equals about 2600 tons of REO. The production could increase with increased market demands and prices. With the initial yearly production rate the mine could have a life of 200 years or more according to Alkane (2009), however that might not hold for all the different elements in the deposit. Compared to many other deposits, Dubbo has a large share of heavy rare earth elements, about 23% of TREO. However, there is no data published on individual elements. (Alkane Resources Ltd., 2009) A third project in Australia that has come far in planning is the Nolans Project, owned by Arafura Resources Ltd. Nolans Bore is a deposit of not only rare earths, but also phosphoric acid, gypsum and uranium. In comparison, Nolans is richer in neodymium than many other deposits (see appendix). The life time of the mine is expected to be at least 20 years with a yearly production of 20 000 tons of REO. Close to the mine Arafura plans to build a processing plant, where the concentrate from the mine will be recovered into oxides. But as the project still has not got the needed environmental approvals for construction, production is estimated to be possible in 2013. (Arafura Resources Ltd, n/d) 5.1.2.2 USA The Mountain Pass is an old, shut down mine that used to be in production in the second part of the 20th century. This rare earth deposit was discovered in 1949, and in 1952 the Molybdenum Corporation of America (now Molycorp) started to mine its minerals and process them in small scale. (USGS, 2010) The production increased in the 1960’s; between 1965 and 1995 the Mountain Pass mine was the largest supplier of rare earth elements in the world (Castor 2008). The mine had its peak years around 1990 when the production was in the scale of 22 000 tons of rare earth oxides per year (USGS, 2010). However, as China started their production in the 1980’s competition became difficult to manage, and the sales from Mountain Pass started to decline. The mine was closed in 2002 (not only because of competition from China, but also because of environmental problems), since then there have been shipments with materials from its stockpiles to e.g. Japan. (Castor, 2008) With the rising demand and prices of REE and the restricted Chinese export, Molycorp has decided to start production again. The environmental problems are said to have been solved and production is expected to start in 2012 and amount to 20 000 tons annually. (Molycorp Minerals, 2009) The rare earth elements will be processed into oxides in an on-site facility, and also in 2011 Molycorp acquired the Silmet Plant in Estonia (mentioned in 5.1.1.3) to increase production. The acquisition of Silmet was also a strategic move in order to get manufacturing abilities and intellectual capital for a planned future production of metals. Indeed, Molycorp’s intentions are ambitious; they also have a mine-to- magnet plan (to produce neodymium magnets that could be used in EV motors) which includes the 28
Chalmers University of Technology
acquisition of Santoku America (an NdFeB alloy producer), and a joint venture with Hitachi (because they have the license for magnet production, more about this in 5.3). (Molycorp, 2010) 5.1.2.3 India and Vietnam Toyota Tsusho plans to start producing rare earth elements in cooperation with India Rare Earths Ltd., who has extracted uranium and thorium from monazite minerals. A byproduct from the extraction process is rare earth chloride mixtures, which will be used as raw materials for REE production. The construction of a manufacturing plant was planned to start in 2011, and also production in the same year. The expected yearly production is 3000-4000 tons. Toyota Tsusho has also entered an agreement with the Vietnamese government to explore the possibilities for rare earth mining in Dong Pao, Vietnam. (Toyota Tsusho Corp.) Detailed information about this project was difficult to gather, which is why the source for the information in the table above comes from Roskill (Chegwidden, J. and Kingsnorth, D., n/d) instead of the company. 5.1.2.4 South Africa The monazite mine Steenkampskraal in South Africa is mostly owned by the Canadian company Great Western Minerals Group (GWMG), who has a supply agreement on 100% of the rare earth supply of the mine. Steenkampskraal is an abandoned mine with high concentration of REO. However, the annual production plan is quite low, only 2 600 tons. Since the recoverable resource is estimated to 30 000 tons of TREO, this would give an expected lifetime of just about ten years. In this case the rare earths would be the primary product, but thorium will be stored for possible future requirements. (GWMG, 2011a) The project is currently somewhere in between pre-feasibility study and bankable feasibility study. In a press release from February 24 2011 GMWG (2011b) are still discussing different alternatives for where to separate the ore into individual element oxides, one alternative in to construct a separation plant on site. GWMG cooperates with British company Less Common Metals Ltd, who produces rare earth alloys, with a mine to market strategy. 5.1.3 Mining possible in longer term The projects in 5.1.2 are all close to possible start-up, but there are an estimated 200 projects with a longer time to production (Zhanheng, 2011b). All projects cannot be presented here, but a few, that we found the most interesting, will be listed below (table 4). Also, to keep in mind is that many projects are only in exploration phase, with a very uncertain future. The economic feasibility of a rare earth mining project often depends on if it they are being mined as a primary product or byproduct. If they are being mined as a byproduct the mine is supported economically by the primary product, but if the rare earths are the primary product the requirements on the quality and concentration of the deposit is of high importance. In 2010 the production from the ion adsorption clays of southern China was a primary product. But for the rest of the world production, rare earths were mostly a byproduct. (USGS, 2010) What this means is that rare earth mining, due to low concentration of the elements, is often dependent on another co-product in order to pay off. 29
Chalmers University of Technology
Name Country Company Earliest Production Production plan (t/y) Ulba Kazakhstan Sumitomo 2015 15.000 +KazAtomProm Kutessay Kyrgyzstan Stans Energy Not Announced 1.230 Kvanefjeld Greenland Greenland 2016 Not Announced Minerals & Energy Hoidas Lake Canada GWMG Not Announced Not Announced Nechalacho Canada Avalon Rare Metals 2014 5.000-10.000 Bear Lodge USA Rare Elements 2015 12.000 Resources Zandkopsdrift South Africa Frontier Rare 2015 20.000 Earths Kangankunde Malawi Lynas Not Announced Not Announced Taboca Pitinga Brazil Neo Material Not Announced Not Announced Technologies + Mitsubishi Norra Kärr Sweden Tasman Metals Not Announced Not Announced Table 4 Mining projects with possible production in longer term 5.1.3.1 Kazakhstan Sumitomo Corporation has an agreement on joint venture with KazAtomProm to start production of rare earth elements in the existing facility in Ulba, Kazakhstan. KazAtomProm is the world's third largest uranium producer; they also produce other metals like beryllium, tantalum, and niobium. They will supply Sumitomo with excess ore from their production as raw material for RE extraction. They will also supply infrastructure. Production still has some way left to go since it is expected to start as late as 2015. (Sumitomo Corp., 2009) 5.1.3.2 Kyrgyzstan The Canadian company Stans Energy Corporation has purchased the, since 1991 closed, Kyrgyz mine Kutessay II. They have also agreed with the current owner, Kyrgyz Chemical Metallurgical Plant, to purchase the nearby processing plant with a private railway station. This enables Stans Energy to mine, process and transport oxides for selling. According to historical information published on the company webpage, the rare earth oxide reserve in Kutessay is approximately 88 000 tons. With a yearly mining of 300 000 tons of ore and a rare earth mineral content of 0,41 % that would mean an annual production of 1230 tons of rare earths. The approximate lifetime of the mine would be 72 years. Stans Energy also publishes data of the distribution of individual rare earth metals in the minerals (see appendix). These data implies a high share of the less common heavy rare earth elements. For example dysprosium has a share of 6.47%, which is comparable to the deposits in south of China and higher than most deposits in the world (see appendix). Overall the estimation is that the ratio between LREE and HREE in Kutessay II is about 50/50, which is very high compared to other reserves. (www.stansenergy.com) Starting date is not announced. 5.1.3.3 Greenland Greenland Minerals and Energy Ltd are planning to finish their pre-feasible study on a REO deposit in Kvanefjeld, Greenland, in 2011. The interim pre-feasibility report states that TREO production from Kvanefjeld could be 43,700 ton per year and that the life time of the mine would be 23 years. Greenland Minerals and Energy has a time line for production start in 2015/2016, environmental and social impact assessments are due in 2013. Besides the rare earth elements uranium and zinc is also 30
Chalmers University of Technology
present in Kvanefjeld. Greenland Minerals and Energy wants to produce uranium oxide as a co- product to the rare earth concentrate, with an annual production of 3,895 tons. This requires mining approval from the government, since there is risk involved in uranium mining and there is currently a ban for uranium mining on Greenland. Zinc is also present, but is not included in the pre-feasibility report. (Greenland Minerals and Energy Ltd, 2011) 5.1.3.4 Canada There are several companies conducting exploration projects for rare earth mines in Canada, e.g. Great Western Minerals Group Ltd., Avalon Rare Metals Inc., Quest Rare Minerals Ltd., Matamec Explorations Inc., and Stans Energy. Many projects are still only in exploration phase, but Canada could be a promising supplier of rare earths in the future. Great Western Minerals Group is the owner of the Hoidas Lake Rare Earth Project in Canada. The project webpage does not reveal much information or data, but it does say that the share of neodymium in the deposit is relatively large, which is why GWMG consider it a strategic investment for expected future increased demand in neodymium for permanent magnets. The challenges for GWMG in this project lie in extracting the rare earth metals from the bastnäsite and they also report some transportation issues. (GWMG, 2010) Beyond the Steenkampskraal and Hoidas Lake GWMG also have the rare earth projects Benjamin River and Douglas River in Canada, and Deep Sands in USA, but these are only in exploration phase Avalon Rare Metals Inc. has a project at the Nechalacho deposit in Thor Lake, Canada. The project has not yet finished the bankable feasibility study and has not yet gotten all the approvals and licenses that are needed, but time line is to start construction in 2013 and start production in 2014. Initially the production of ore would be 1000 tonnes per day, which would equal 5000 tonnes of REO per year. During the fourth year the production rate is expected to double. The expected lifetime of the mine is 18 years. The rare earth elements will be the primary product, sold as oxides, and byproducts will be zirconium oxide, niobium oxide, and tantalum oxide, which are commonly found along with rare earths. (Avalon Rare Metals Inc., 2010) 5.1.3.5 USA Beside Mountain Pass there are a few more projects in the US. The one that is perhaps the most promising on is that in Bear Lodge, owned by Rare Element Resources Ltd. The exact amount of reserves is not stated in the interim feasibility report, but in a conceptual plan the figure for annual production is 325 000 tonnes of ore. How much REO that corresponds to depends on the concentration of rare earths in the ore, the report shows an average figure of 3.62 % REO, which would then be almost 12 000 tonnes of REO per year. However this is just a conceptual figure, future reports will have to give more certain figures. The life time of the mine is expected to be 15 years beginning in 2015. Other elements found in the deposit are gold copper, uranium and thorium. (Richardson, 2010) 5.1.3.6 South Africa Africa does not have any recorded production of rare earths today. But there are several deposits and exploration projects running. The project that seems to be closest to production is Zandkopsdrift, held by the Canadian company Frontier Rare Earth Limited. The project plans are: target scope study finished 2011, pre-feasibility study finished 2012, and bankable feasibility study also in the end of 2012. This could lead to possible production in 2015, and the announced annual 31
Chalmers University of Technology
production is 20 000 tons of separated REO. The mineral from Zandkopsdrift is monazite with low average levels of thorium (0.0225%) and uranium (0.0065%). The size of the deposit or the expected lifetime of the mine is not determined. Distribution of the elements is shown in the table in the appendix (Frontier Rare Earths Limited, 2011) 5.1.3.7 Malawi In late 2010 Lynas got the approval from the Malawi government for their acquisition of Kangankunde Rare Earth Resource in Malawi. The project is in feasibility study phase. The size of the resource depends on the cutoff grade, with a cutoff grade at 3.5 % the size of the resource is 107 000 tonnes at an average grade of 4.25% REO. The thorium content is reported to be very low. Since Lynas are about to construct a processing plant in Malaysia, the ore from Kangankunde could either be fully processed in Malaysia, or it could be concentrated in Africa and then shipped to Malaysia. There are today no figures for production rate or startup date. (Lynas Corporation, 2011) 5.1.3.8 Brazil In 2009 Neo Material Technologies (“Neo”) signed an agreement with the Brazilian mining company Mineracão Taboca for the allowance to be free to investigate the possibilities of rare earth production in their Taboca Pitinga mine. The primary product from this mine is tin concentrate, byproducts are niobium and tantalum. Neo will have access to the tailings from the tin production, which are thought to have relatively high concentration of the heavy rare earths dysprosium and terbium, and other resources. The mineral in Taboca is xenotime. (Neo Material Technologies, 2009a) That same year Neo signed a memorandum of understanding with Mitsubishi Corporation to establish a “strategic partnership” for developing a rare earth supply outside China. The partnership is not limited to the Taboca mine, but will also look for other opportunities around the world to produce rare earths as byproducts. (Neo Material Technologies, 2009b) 5.1.3.9 Sweden There is no history of rare earth mining industry in Sweden, even though this is the country where the elements were discovered. However, there are two projects exploring the possibilities of future production. One is LKAB’s investigation of possible production from tailings from their mine in Malmberget. LKAB is working together with Luleå University of Technology on this. The other project is pursued by the Canadian mineral exploration company Tasman Metals Ltd in Norra Kärr in southern Sweden. The project is still in exploration phase and drillings are being made to determine feasibility of mining. There is no announced date for possible production start at this stage. (Tasman Metals, 2011) 5.1.4 Issues related to starting new mines As prices and demand for rare earths are rising, more and more projects are started to explore new opportunities for producing them. The time from exploration to production can be many years though, and there are many steps and procedures to be handled on the way. Most nations have rules and guidelines for mining projects, e.g. the Canadian NI 43-101, the Australasian JORC Code, and the South African SAMREC Code. These guidelines usually demands feasibility studies and technical reports. The Canadian mineral exploration and development company Avalon Rare Metals Inc. (London, 2010) presents a list of 8 stages before production is possible: 32
Chalmers University of Technology
1. Grassroots: 1 year 2. Target generation and drilling: 1-2 years 3. Discovery delineation: 1-2 years 4. Infill drilling: 1-2 years 5. Bulk sample and metallurgy: 1 year 6. Prefeasibility: 1-2 years 7. Permitting, marketing and feasibility: 1-2 years 8. Construction: 1-3 years This means that a time line of around 10 years is required from finding the reserve to actual production. There are also some other difficulties in producing rare earths, the Organization for Economic Co-operation and Development (Kim and Korinek, 2010) lists a few obstacles to enter the rare earth market:  To process the ore into concentrates, oxides or even metals require special technology, and the technology depends on the ore and needs to be specially adapted.  There is a high capital cost to the production of rare earths, OECD estimates USD 30 000 per ton.  There is no official trade market for rare earths; the companies must have their own costumers.  The production knowledge is limited outside of China  The input costs in China are very low and difficult to compete with. One way of handling the capital costs can be to have the rare earth production as by- or co- production with another product, such as niobium, tantalum or iron. Today there are many joint ventures and collaborations between mining enterprises, trading houses (e.g. Sojitz) and producers (e.g. Toyota). This is a way for both sellers and buyers to secure costumers/supply. The last bullet might be about to change as the Chinese government intends to put stricter regulations on the industry to protect the country's resources and environment, which could make the rare earths produced in China more expensive as the companies will have to invest in upgrading their technology (see Chapter 3.3). 5.1.5 Separation After mining the ores have to go through separation processes in order to sort out tailings and obtain the rare earth oxides. In the initial phase of producing the rare earth metals by refining and separating oxides, approximately 25% of the ore is lost as waste and added to that is another 5% lost as slag product (Du and Graedel, 2010). Since the majority of all production of sintered neodymium magnets are produced in Asia large parts of the world are today dependent on the imports of magnets from Asia and specifically China. The Chinese production of rare earth materials and magnets is totally dominating the global market. China produces more than 95% of the rare earth metals worldwide and is also the only exporter of commercial quantities to other countries. There are some production of rare earth metals in other countries as well, Japan for instance, but that production is not for export but for use within the country (Humphries, 2010). 33
Chalmers University of Technology
5.2 Oxides and metals Alloys and Final Recycling Mining & separating Oxides Metals Magnet Magnets Components Technologies and recovery Powders Mining and Battery Incorporation Recycling and Oxides Metals Components separating assembly in vehicle recovery As mentioned above the most of the rare earth mining occurs in China at present, the same is true for the production of the oxides, metals and alloys (Humphries, 2010). The exports of rare earth elements from China can be in the form of oxides that have to be processed further (to increase the rare earth content in the oxide), or as rare earth metals or alloys. China also has significant amounts of production of rare earth metal alloys, around 90% of global total (Humphries, 2010). This indicates that some countries buy REO and refines them into metals themselves. During the second phase of fabrication and manufacturing, where metals and alloys are made, there are other losses in the form of scrap and industrial waste, and according to Du and Graedel the amount of losses in this phase is quite vague but an estimation of approximately 10% of material is lost in this phase. 5.3 Permanent magnets This chapter will further explain the product chain for permanent magnets from alloys and magnet powders to components and final technologies. Alloys and Final Recycling Mining & separating Oxides Metals Magnet Magnets Components Technologies and recovery Powders 5.3.1 Alloys and magnet powders Japan is a large importer of rare earths for production of permanent magnets for the vehicle industry and Hitachi is one of the producers of such motors, especially for Toyota (Clenfield, et. al., 2010). So far Japan has mostly imported raw material, rare earth in the form of ores and intermediate raw materials, to produce their own permanent magnets instead of buying finished magnets from China. For the Japanese companies using neodymium, the raw material can be bought either in the form of pure neodymium or didymium which is a mixture of neodymium and praseodymium (Cordier and Hedrick, 2010). This indicates that countries that produce their own rare earth products import pure rare earth metals and alloys from China or produces their own metals and alloys after buying intermediate materials such as REOs from China. 5.3.2 Magnets The production of rare earth magnets is also substantially located in Asia and mostly in China. Regarding the production of neodymium magnets, China produces 75% of the global total. Even for the production of the other type of permanent magnet, the Samarium-Cobalt magnet, China has the major part of 60% of the global total production (Humphries, 2010). Basically, there are two different types of neodymium magnets, which most commonly have the formula Nd Fe B, sintered and bonded. The global manufacturing, sales and import of neodymium 2 14 permanent magnets are today restricted by patents. The master patents are held by two groups: 34
Chalmers University of Technology
Hitachi Metals and the Magnequench consortium (Hitachi, 2007 and Bauer, et al., 2010). The Hitachi group (formerly Sumitomo Special Metals), and Magnequench (formerly part of General Motors, now belonging to Neo Material Technologies), have cross-licensed each other. This means that Hitachi has the licenses to produce and sell sintered NdFeB magnets while Magnequench has similar patents for bonded NdFeB magnets (Kroll, 2011). Hitachi Group has entered license agreements with 10 companies to manufacture the Nd-Fe-B magnets and these licensed manufacturers work under Hitachi’s patents (Bauer, et al., 2010). Magnequench on the other hand does not have any licensed manufacturers but only provides the metallic powder to their customers. However, when buying Magnequench’s powder, the customers are allowed to work under Magnequench’s patents and thereby produce and sell bonded NdFeB magnets (Kroll, 2011). Bonded magnets will not be mentioned much more in this report, but in order to understand the differences they are mentioned above. Sintered magnets have higher performance, in comparison to bonded, and are therefore mostly used in larger applications as vehicles and wind turbines. Therefore, most focus in this report will be on neodymium magnets and not samarium-cobalt magnets. Hitachi’s licensed producers are mostly located in China and around the south-western parts of Asia. There are also some production facilities in European countries, Germany and Finland. Most of the Chinese producers like Ningbo Yunsheng Co. Ltd., Thinova Co. Ltd., AT&M Co. Ltd., Beijing Jingci Magnetism Technology Co. and Beijing Zhong Ke San Huan High-Tech Co. Ltd. have their production of neodymium permanent magnets located in China (Ningbo, 2005, Thinova, 2009, AT&M, 2011, BJMT, 2005, San Huan, 2009). The Finnish company Neorem Magnets OY have some production in Ulvila, Finland, as well as in Ningbo, China (Neorem, 2009). The German companies Schramberg Magnet- und Kunststofftechnik GmbH and Vacuumschmelze GmbH have production of magnets in Schramberg and Hanau, respectively. Vacuumschmelze also have some production in Beijing, China (Schramberg, 2011 and Vacuumschmelze, 2010). For the Japanese magnets producers the scenario is somewhat different as the production facilities are more spread out over Asia. TDK Corporation has a production plant in Narita, Japan, along with two in Hong Kong, China, and one in Rojana/Wagnei in Thailand (TDK, 2011). The last company on Hitachi’s list of licensed Nd-Fe-B magnet producers is Japanese ShinEtsu Chemical Co. Ltd. which have production facilities in Takefu, Japan, and also in Malaysia, Philippines and on the Indonesian island Batam (ShinEtsu, 2007). In a global perspective there has been very little or no activity in production of sintered neodymium magnets on the American continents until lately. Mostly, due to lower production costs, the production and manufacturing of sintered neodymium magnets is located in Asia except for some factories in Europe (Humphries, 2010). As seen in figure 8 below, the stars on the map indicates facilities where production of sintered neodymium magnets is located under license of Hitachi Metals. The larger stars located in China and Japan indicates that larger quantities are produced there. 35
Chalmers University of Technology
Figure 8 Global locations for production of sintered neodymium permanent magnets The American rare earth company Molycorp Minerals has established a mine to magnet strategy for starting production of permanent magnets in the USA. For this strategy to be realizable Molycorp has acquired Estonian Silmet processing facility, and the rare earth metal/alloy producer Santoku America. Silmet was acquired in order to double Molycorp’s oxide production capacity and to provide knowledge in rare earth metal production. The plan is to produce rare earth oxides, metal alloys and magnets for applications such as wind power turbines, hybrid and electric vehicles and defense system products. The permits to produce neodymium magnets are provided by Hitachi Metals. (Molycorp, 2010) The sintered neodymium permanent magnets that are used in the automotive industry are to a large extent produced outside of China. Today China produces huge amounts of neodymium magnets, however, the net shape of the magnets does not fulfill the high standards of many certain applications and may therefore be treated before assembled into a product. This is one possible reason for Japan to produce much of their used permanent magnets themselves. The common focus for outcome of Chinas permanent magnet production is quantity and not highest quality. It is more common to produce larger blocks which have to be machined to remove extensive material. This also result in large amounts of scrap material which has to be handled (Humphries, 2010). In 2005 China produced ca 30 000 tons of sintered neodymium magnets. That same year the global production of sintered neodymium magnets outside of China totaled around 9000 tons, of which 8500 tons were produced in Japan and 450 tons in Europe (Feng, 2006). 5.3.3 Components and Final technologies The neodymium type permanent magnet is the type most frequently used in automotive motors today since they have higher magnetic energy density than samarium-cobalt magnets as well as lower price per weight which is preferable in vehicles (Sastri, et. al., 2003). Added to the neodymium magnets are commonly other rare earths, such as praseodymium, dysprosium or terbium in order to increase the operating temperature of the magnets. This is often needed for vehicle motors that operate in higher temperatures (Bauer, et al., 2010). By using permanent magnets, space and weight 36
Chalmers University of Technology
is saved which is very profitable in most applications as it can reduce the price with fewer components and make products more attractive due to low weight and size. 5.4 NiMH batteries Mining and Battery Incorporation Recycling and Oxides Metals Components separating assembly in vehicle recovery The chain of actors seems to be quite short for the NiMH batteries for hybrid vehicles. The components are made from alloys and the battery is then assembled with those components by the same battery manufacturer. There also seems to be a limited number of actors, with close relationships between battery supplier and car manufacturer. Most batteries are manufactured in Japan. The first successful NiMH battery using rare earth metals was patented by Stanford Ovshinsky, who, together with his wife Iris, co-founded Ovonic Battery Company, a subsidiary of Energy Conversion Devices (ECD) (Zelinsky, 2010). ECD still holds the patent, but has licensed it to a number of companies. Out of these companies only a few produces batteries for use in electric or hybrid vehicles. Today Primearth EV Energy and Sanyo Electric are the largest producers, both located in Japan. Primearth, former Panasonic Electric Vehicle Energy, is a joint venture between Matsushita Electric Industrial Co., Ltd. and Toyota Motor Corporation; they produce all batteries for Toyota's vehicles. Sanyo supplies, among others Honda and Ford. GS-Yuasa Group is a joint venture between Mitsubishi and Mitsubishi Motors, the group also includes a Honda joint venture, and Toyota is one of the principal shareholders. For hybrid and electric automotives GS-Yuasa seems to be an example of a producer that is chosing lithium-ion technology instead of NiMH: “the GS-Yuasa Group will make further efforts at (…) focus on Lithium-ion Batteries for automotives to establish it as a core for our new business” (GS-Yuasa Corporation, 2009). They report to have a planned annual production of batteries for 67 8000 electric vehicles in 2012, and in the future exceed 100 000 (LEJ, 2010). Other companies choosing Li-ion for hybrid and electric vehicles is e.g. Hitatchi, Saft, SB LiMotive and Toshiba. The point of the part above is that there are many manufacturers that have the patent to make NiMH batteries, but two producers have become the successful suppliers for the vehicle industry. These two companies have produced more than 95% of all batteries in hybrid and electric vehicles in use today (Schreffler, 2010), and all of their production is located in Japan. 5.5 Recycling Alloys and Final Recycling Mining & separating Oxides Metals Magnet Magnets Components Technologies and recovery Powders Mining and Battery Incorporation Recycling and Oxides Metals Components separating assembly in vehicle recovery Constant development of new technology creates a demand for new goods and products which generate huge quantities of obsolete products that will not be used anymore (Resende and Morais, 2010). There are huge quantities of waste, scrap and obsolete products which contain valuable 37
Chalmers University of Technology
materials and from a resource effective and economic perspective those materials can be turned into resources if handled in a proper way. Recycling of used materials is something of great concern in the development of a sustainable and growing society. In some parts of the world, at least among some different materials (paper, metals, some polymers etc.) and technological devices (mobile phones, computer monitors etc.), recycling is performed to extract valuable materials. However, most recycling is only performed if it is economically feasible without too significant losses of the materials’ purity and economical value. The common saying when it comes to recycling is that what is profitable to recycle is recycled (Lehner, 2011). Regarding the rare earth elements, the recycling and after usage stages of the product chain are not very well developed. This depends greatly on the physical and technological difficulties in the processes in which different rare earth elements may be treated and recycled. The problem depends basically on two aspects, economy and technology. The alloys of rare earth elements mixed with other metals and chemicals are very complex and often contain several different compounds. This complexity makes it very difficult to use a simple process to separate the material in a sufficient way. With the occurring situation in China with decreased exports of rare earth elements, demands for new methods of extracting rare earths are urged for. Since the rare earths usually are very stable and mixed in ores they demand large amounts of energy to obtain rare earth metals of high purity. Even if some rare earths are not that rare, they still need a lot of energy and processing, as well as there being a lot of generated waste (Takeda, et. al., 2005). Therefore, recycling could be a good alternative to provide REMs. However, the recycling of rare earth metals is not very common yet, but still there are some different methods available to extract rare earths from discarded products. There were in the year of 2005 approximately one billion of used NiMH batteries which totally contains light rare earth metals, like lanthanum, cerium, praseodymium and neodymium, to an amount of 2500 tons (Li, et. al., 2009). To address this problem there have been many hydrometallurgical methods developed in order to extract rare earths from scrapped batteries. These methods can basically be described as chemical dissolution of the compounds of the batteries, through the processes of leaching, solvent extraction, evaporation and crystallization. In various steps the rare earths are solved and extracted by using chemicals, water and energy (Li, et. al., 2009). Several studies have been performed where different types of acids are used as leaching agents in order to recover europium and yttrium from old color TV screens (Resende and Morais, 2010). This example along with the mentioned processes above and in chapter 4 indicated that there are possible ways to recover and recycle rare earths from different applications, and that this can be done with a high purity of the outcome. China has the major role of what is happening in the rare earth industry today. Since China has the dominating position of production and exportation of rare earths, it also affects the market and afterlife processes of products containing rare earth elements. The increasing demand for rare earths in various products around the world along with the increasing use of such elements within China has created difficult situation where the demand for rare earths might exceed the availability. The expansive market of rare earths has in the last two decade generated a stock of rare earths bound in products. By recovering these rare earths or at least parts of it, might be a reliable source for future demands of rare earth elements (Du and Graedel, 2010). 38
Chalmers University of Technology
Still, the situation is not easy since the different stages of the material flow of rare earths are not completed. Looking at a Material Flow Analysis (MFA) for these elements, the primary stages of production, fabrication and manufacturing, and the use phase is well known but the final phase, waste management and recycling is not as well developed (Du and Graedel, 2010). When looking at the flows of material within these different phases it becomes clear that a lot of material, rare earths that is, “disappear” during the material flow stages. As mentioned earlier there are losses in the separation and refining stages of the production cycle but. The third phase, use stage, is the phase where the rare earths are put into products and this is very difficult to estimate numbers of losses since rare earths often are used in alloys in various products, for instance permanent magnets that may contain neodymium, praseodymium, dysprosium, gadolinium and terbium (Du and Graedel, 2010). Most of these applications contain alloys or rare earth compounds that are very stable and/or assembled in a way where leaching or loss of material is not very likely to happen. However, in figure 9 and 10 below numbers of in-use stocks are visible, which indicates the amount or rare earth metals that might be recovered or seen as a resource when the product has served its purpose and turns into waste. The last phase of the material flow stages is recycling which is the least covered phase. This depends mostly on economic and technical issues. The technical issues are a problem due to complexity and advanced procedures in recovering the elements and this also becomes an economical issue since energy, chemicals, equipment, technology, disposal and/or treatment of wastewater generates high costs. If the amount of recovered rare earth metal is too small, the value of the outcome cannot cover the expenses of the recovering process and thereby not being feasible (Ekberg, 2011, and Lehner, 2011). On the other hand, if the demand for rare earths increases even more, which is not unlikely, cost effective processes of recovering rare earth elements have a high potential. According to Du and Graedel, 2010, the in-use stock of cerium, neodymium, lanthanum and praseodymium was in the year of 2007 four times the annual extraction rate for these four elements in that very same year. This is a good example that shows the amount or rare earths that are in circulation in current technology, and a possible stock that may be able for recovery and recycling. In figures 9 and 10 below, the in-use stock use by the year of 2007 can be seen for various rare earth metals in different applications. A possible increase of recovery and recycling or rare earth metals could result in a decrease of mining for virgin material which is today is the most common source for rare earth elements (Du and Graedel, 2010). To make recycling of rare earth elements sufficient and economically feasible the amounts of recycled outcome has to be large enough. This makes the recycling of larger applications more likely to be the choice of possible future recycling of rare earth elements (Du and Graedel, 2010). Possible element to recycle are rare earths such as neodymium, lanthanum, cerium and praseodymium, which all are used in fairly large quantities and in large applications like for instance in automobile catalysts, permanent magnets in car engines and wind turbines. Due to the large quantities, the recycling can be performed more efficiently (Du and Graedel, 2010). Smaller applications demand for very accurate and effective recycling processes and are therefore not likely to be recycled in the nearest future (Du and Graedel, 2010). 39
Chalmers University of Technology
Already there are companies that have started to look at recycling of rare earths mainly to supply for domestic use and to cover some of the own usage. One of the companies that more or less have started to recycle rare earth metals is Japan’s second largest rare earth magnet producer Shin-Etsu Chemical Co. that has plans to start recycling of rare earth metals in used air conditioners and thereafter reuse the metals in permanent magnets (Nikkei, 2010 and Sadden, 2011). Shin-Etsu has the metallurgical knowledge of handling rare earths and how to extract them from ores and oxides, and this knowledge can be used to extract the rare earths from the air conditioner motors (Nikkei, 2010). From a magnet of 50-100 grams from the motors, approximately 30% can be extracted for reuse. Shin-Etsu already has a plant that extracts rare earths from their own production scrap. There are also possibilities to recycle rare earths from other applications and products like hybrid-vehicle motors etc. (Nikkei, 2010). Stated on the Green Car Congress homepage, there is collaboration between Toyota Motor Corporation, Toyota Chemical Engineering, Sumitomo Metal Mining and Primearth EV Energy Co., Ltd. where they have come up with a process to collect and recover NiMH batteries. The process can be described as battery-to-battery recycling where used batteries are collected, dismantled, reduced, sorted, refined and finally processed into new NiMH batteries. The process chain can be seen in figure 11. There are also investigations going on by Toyota Motor Corporation to see the possibilities of using this process in America (GCC, 2010) Toyota New NiMH batteries Dealers Further processeing Customers and assembly Collection of spent Refining process batteries Reduction, crushing Dismantling and and sorting removal proceses Figure 11 Description of battery recovering process for Toyota (GCC, 2010) Another company that has started to look for recycling options is the biggest rare earth company of Japan, namely Hitachi Ltd., who produces motors for the Toyota Prius hybrid car. Hitachi uses approximately 600 tons of rare earth metals every year and is therefore dependent of a stable supply or rare earth resources (Clenfield et. al., 2010). Since China has cut the export quotas of rare earth elements, Hitachi has plans of using recycling as a resource to cover some of the company’s need for rare earths. Today the recycling is more or less nothing. However, a quite realistic future perspective is that 10% of the Hitachi’s need for rare earth metals shall come from recycled metals (Clenfield et. 41
Chalmers University of Technology
al., 2010). Similar to Shin-Etsu’s processes of recycling rare earths, Hitachi also uses spent air conditioner motors which are opened in order to recover the permanent magnets inside of it. This sudden boost of plans to recycle rare earths has settled new recycling projects in other companies as well. The Japanese chemical company Showa Denko KK has opened a plant to recycle dysprosium and didymium, a mixture of neodymium and praseodymium, in Vietnam. Also Mitsubishi Materials Corp. along with Panasonic Corp. and Sharp Corp. has projects going on to investigate the costs of recovering neodymium and dysprosium from products like washing machines and air conditioners (Clenfield et. al., 2010 and Sadden, 2011). Increasing of conventional and green technology increases needs of REMs and therefore recycling is needed can be seen as a good alternative supply. So far the recycling has been held back, mostly due to economic and technical issues, where the technical issues greatly affect the economic issues due to, efficiency, expensive equipment and technology, energy and waste treatment (Ekberg, 2011, Lehner, 2011 and Weihed, 2011). There are also other projects going on worldwide where other sources are investigated for future supply of rare earth elements. An example of such projects is recovering of rare earth metals from tailings of old mines. For instance in northern Sweden, LKAB (Luossavaara-Kiirunavaara Aktiebolag) has started a project along with Luleå University of Technology (LTU) where they look at the possibilities of extracting and recovering rare earth metals from apatite (Lehner, 2011). Tests of the apatite in Kiruna and in Vitåfors, Malmberget show that it contains 15 different rare earth metals, which if mined have the possibility of gaining production values in the magnitude of billion Swedish crowns (kronor). Right now, LKAB has collaboration with LTU in order to test different methods for extraction of these rare earth metals (Magnusson, 2011). If the project will result in production of rare earth metals this might be realized in the year of 2015. As of today the rare earth metals in the apatite is treated as impurities. According to LKAB a yearly production of 400000 tons of apatite would be possible during a period of 14 years. The apatite contains rare earth metals to an extent of 0,5 % which during the time period would result in approximately 28000 tons (Magnusson, 2011). The rare earth metals that exist in highest amounts in Kiruna are yttrium, cerium, europium, terbium and neodymium (Magnusson, 2011). According to this example in Kiruna there are possibilities to start extracting rare earth metals from many different resources than regular virgin mining. Today and in the future alternatives like above mentioned recovery from spent products and scrap metals, extraction from mined tailings like apatite and effective recycling might be available in an economic and technically available manner. According to Toyota, 2010, the company had by October 2010 produced more than two million Prius hybrid cars. Each Prius car contains approximately 1200-1500 grams of neodymium, 5400 grams of lanthanum, approximately 1900 grams of praseodymium and 900 grams of Cerium. This means that those two million produced Prius hybrids by the year of 2010 contain approximately 2400-3000 tons of neodymium, 10800 tons of lanthanum, 3800 tons of praseodymium and 1800 tons of cerium. Sooner or later the lifetime of these two million Priuses will come to an end which if recycled could generate a significant amount of recoverable rare earth metals. These two million hybrid vehicles alone could generate thousands of tons of rare earth metals if recycled efficiently. 42
Chalmers University of Technology
6 Speculations about a possible product chain in Sweden This chapter will discuss the possibilities of having parts of or the complete product chain for production of permanent magnets and NiMH batteries, containing rare earth elements, in Sweden. The text will be speculative in its nature and include features of discussion since it is talking about the possibility of a future scenario. Alloys and Final Recycling Mining & separating Oxides Metals Magnet Magnets Components Technologies and recovery Powders Mining and Battery Incorporation Recycling and Oxides Metals Components separating assembly in vehicle recovery Sweden is a country with a well-developed metallurgical industry. For instance companies such as LKAB, Höganäs AB and Boliden AB possess knowledge within the metallurgical science and Sweden as a country is also well developed from a recycling point of view. The future possible Swedish participation of the rare earth product chain depend much on how far the mining companies want to go in the purification of metals and if the rare earth elements could be processed under economically feasible circumstances (Weihed, 2011). If Sweden has highly effective processes it could be used for competition with developing countries (Weihed, 2011). 6.1 Mining and Separation Sweden has a long history of producing different types of metals, such as iron, copper, silver and gold. The knowledge and efficiency within the processes is well developed and even with high salaries compared to less developed countries it is possible to mine and separate metals profitable from an economic point of view (Weihed, 2011). Since Sweden has known sources of rare earth elements at various locations in Sweden it would be technically possible to mine those sources for rare earths. This can be done from virgin mining in Norra Kärr if that project will be realized. Other solutions than virgin mining is as mentioned in earlier chapter 5.5 for instance extraction from apatite from the mines around Kiruna. From an economic point of view the availability and percentage of rare earths in the ore are of great significance when considering realization of mining in Sweden. If the content in the ores is high enough it could be mined and separated. The knowledge from mining and separation among other metals in Sweden adds to the advantage of how to handle metals and what processes to use for separation. (Tengzelius, 2011 and Weihed, 2011) 6.2 Oxides and metals The stages of refining the separated ores further into oxides and metals could also possibly take place within the Swedish borders. Again the knowledge within refining other metals would help to provide the know-how in order to process the separated ores into oxides and later metals. The processing of ores in order to produce rare earth oxides and metals demand wet processing with acids and chemicals which creates large costs especially from an environmental perspective and laws for handling of toxic chemicals, water and waste after the processes. However, if the regulations and laws regarding such issues were the same globally, the costs would not be that varying across the world depending on location and/or lack of environmental laws, as it is today (Tengzelius). 43
Chalmers University of Technology
6.3 Alloys and Magnet Powders The powders and alloys could be produced in Sweden under the right circumstances. Höganäs already has significant knowledge in producing ferrous powders and such metallurgical knowledge could be very useful also in the production of rare earth alloys and magnet powders. As of today Höganäs is producing ferrous powders competitively in comparison with China. This depends much on the high level of automation within the production stages. If the processes could be performed with a high level of automation the production of rare earth metals could be possible as long as the level of material is sufficient and not too difficult to enrich. With a high level of automation the limiting factors turns out to be energy and raw material. Tengzelius at Höganäs estimated a possible annual production of rare earth metals in Sweden to the magnitude of 20 000 tons in order to be economically feasible, but this number is still very uncertain and may vary due to uncertainties in costs, processes, wages, availability etc. 6.4 Magnets, components and final technologies This part of the product chain is more uncertain since Sweden today does not have that many industries that use rare earth magnets, except for maybe some electronic companies, possibly among car manufacturers and wind power companies. The Swedish interest in the final stages of the REE product chain depends quite much on how the development within Swedish industry will be the coming years. Sweden has two semi-large car manufacturers: Volvo and SAAB, which might provide some potential market interest in hybrid vehicles and therefore permanent magnets and NiMH batteries. Possibly there could be an interest in producing rare earth metals and alloys for production of batteries and permanent magnets in case of a growing market and demand for such products within the industry. The market for wind turbines and hybrid vehicles seems to grow but if it is economically feasible to produce permanent magnets and NiMH batteries is very hard to tell since these processes are very well established in Asia and the competition is tough. 6.5 Recycling The recycling situation in Sweden today is quite well developed and many different products and materials are being recycled; from light bulbs, batteries and refrigerators to materials such as iron, copper, glass and cardboard. Today car engines are recycled in separated processes for steel and copper. Often the whole engine is grinded to powder and then the steel and copper is separated in order to recycle them individually (Tengzelius). 44
Chalmers University of Technology
7 Analysis and discussion This chapter will both analyze and discuss the results from chapters 3, 4 and 5. In this chapter we will allow ourselves to discuss from our own perspective, which we have gotten after studying the topic for almost six months. 7.1 Mining At present there seems to be a rare earth race going on in the world, with more than 200 exploration projects for rare earth mining. Rare earth production could turn out to be a lucrative business if demand continues to rise. Increasing prices for REO makes them an interesting investment opportunity and mining projects will have a better chance for financing. Also, for companies like Toyota and Sojitz it is an urgent matter to find ways to secure supply in the future, which has lead them to invest in mines of their own. China’s dominance in the market is about to meet challenge, as new mining projects has started on almost every continent of the world. This gives the impression that we need not worry about the supply of rare earths in the future. And that might be correct. But, there are a few things to keep in mind; firstly, most of these projects are still in exploration phase. This means that they are not yet feasible for mining, and even if they will be, it takes years to come into production considering all the obstacles and bureaucracy associated with rare earth production (see chapter 5.1.4). So the question is if the new production will start in time, before there might be a supply shortage. Secondly, the demand is not the same for all rare earths and they do not occur in the same percentages in the REO, so even if the total REO produced will meet the predicted demand there could still be deficits for some of the elements (as well as excess of others). China is today trying to consolidate its industry, mitigate environmental destruction and illegal mining operation by closing down smaller mines, cutting export quotas, introducing stricter regulations and higher taxes. In the heated debate it is sometimes implied that China came to dominate the RE production with unfair means of competition, such as lax regulations and low production costs. This might be quite true, that they did have less regulation (that was not enforced) and lower costs. But the money from REs has come with a price, that of destroyed environment and human health. From a Chinese perspective one could wonder if the world RE dominance has been worth it. Up until recently, there has been many RE miners in China which means high competition, and lax regulations which means low control. And since there is no official trade for REs it has been a buyers’ market in which the price has not reflected true costs (if including e.g. environment and health). When listening to the debate on China’s export of REs it is very difficult not to become biased, it is sort of China against the rest of the world. And the rest of the world speaks more often and louder. Some are implicating that China is using their RE resources to exert control and power, that they got the dominance on the market because they could sell low cost REO and now when the rest of the world is dependent they cut exports. Another event that could strengthen the opinion that China is keen on keeping the RE dominance is that the Chinese government tried to buy the Mount Weld mine in Australia, but the Australian government said no. However, if China comes through with the stricter regulations, consolidations, policies, taxes and production quotas it could protect their environment and resources as well as setting a fair price that other non-Chinese producers could compete with. There might also be other explanations why China wants to cut exports; in chapter 3.1 we mentioned that China has plans to build 500 000 hybrid or electric vehicles between 2011 and 2015. This means that the domestic demand for rare earths in China will increase substantially. By 45
Chalmers University of Technology
how much, depends on which technology will be used. In the pure electric vehicles the batteries will probably be Li-ion, but in the hybrids NiMH could be used. However, permanent magnets will probably be used in all the engines, if no new technology is invented. This will mean that China will, themselves, need more of the rare earths that are most susceptible to deficit. The political issues related to rare earths change all the time. Even though China sets stricter regulations, they are yet to be enforced. The export was to be set 35% lower in the first half of 2011, yet a recent article in Wall Street Journal (Yap, 2011) reports a 33% rise on the year for the exports in the first four months of 2011. This is one of the most important reasons why this report should not tell the reader about what will happen in the future when it comes to rare earths. The aim of this report is to give some background information and sketch a picture of the current situation, so that the reader can follow the news and him/herself work out the future. The mines listed in 5.1.2 could together produce as much as 75 000 tons REO annually by 2014. If Mark Smith at Molycorp would be right in his anticipation of a world demand of more than 200 000 tons REO by 2015, this new production would cover the approximate 65 000 tons that would be missing compared to the 2009 production of 126 000 tons. If China’s production would decrease or their domestic demand would increase to the extent that they would stop their export totally by 2015, and Zhanheng would be right in his estimations of an outside of China demand of 80 000 tons, there could be a supply shortage. There are a lot of speculations circulating in the rare earth “blogosphere” about future demands, China’s actions, and mining projects. To draw any conclusions from this is like predicting the weather for a month ahead; there are large uncertainties within limits. The future predictions depend a lot on world economic situation, technological developments, and politics. Just comparing numbers for future demand and supply of REO can be misleading since the demand for the individual elements will vary significantly depending on the technologies that uses them, and the supply will depend on where the REO is mined since the distribution of individual elements differ in different deposits and minerals. If the market for hybrids and electric vehicles, and windmills increase there might be a future shortage of neodymium, dysprosium and terbium. The latter ones are not used in large shares in these applications (compared to the other rare earths in the same applications), but they are HREEs which are rarer and usually occur only in small amounts in the REO. Neodymium is a LREE, but since there are several applications competing for it there could still be a risk of shortage. If looking at the occurrence of the individual elements in the oxide of the existing and new mines (see appendix), it turns out that China probably will play a big part in the supply of HREEs in the future as well. Important to say is that the numbers for the rare earth content in the REO in the new mines, collected in the table in appendix comes from the companies who owns the mines, which makes them less reliable than those numbers from USGS. Also, not all companies would reveal the distribution of elements in their REO, so there is much data missing in this report. One example of this is the Kvanefjeld project, owned by Greenland Minerals & Energy. They claim that the deposit has a very high share of heavy rare earths, but no exact numbers for individual elements are published. 7. 2 Permanent magnets Rare earth metals are and most likely will be a very important piece of the foundation of future technology and green development of the industry. The increasing sales of hybrid vehicles and wind 46
Chalmers University of Technology
power plants will increase the demand for permanent magnets since that is the best technology serving these purposes so far. As of today the market for hybrid vehicles and wind power plants seem to be increasing and the production of the Toyota Prius has increased almost every year since it first was put in production. As shown in table 5 below, the Toyota hybrid vehicle Prius has increased its sales annually every year since initial production except for the year of 2002 when sales decreased a little. This might be an indication that people find hybrid vehicles appealing and a good sustainable technology (Toyota, 2010). Year 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009Jan-sep 2010 Total Japan 0,3 17,7 15,2 12,5 11,0 6,7 17,0 59,8 43,7 48,6 58,3 73,1 208,9 254,2 826,9 North America 0,0 0,0 0,0 5,8 16,0 20,3 24,9 55,9 109,9 109,0 183,8 163,3 144,3 105,9 939,1 Europe 0,0 0,0 0,0 0,7 2,3 0,8 0,9 8,1 18,8 22,8 32,2 41,5 42,6 35,5 206,1 Other 0,0 0,0 0,0 0,01 0,2 0,2 0,4 1,9 2,9 5,3 7,0 7,7 8,4 5,8 39,7 Total 0,3 17,7 15,2 19,0 29,5 28,1 43,2 125,7 175,2 185,6 281,3 285,7 404,2 401,3 2011,8 Table 5 Annual sales of Toyota Prius 1997 - September 2010 (source: Toyota, 2010) The growing markets for hybrid vehicles and wind turbines will most likely increase the production of permanent magnets since that technology is the most efficient and cost efficient alternative so far. Increasing production of permanent magnets will result in increasing demands for rare earth elements such as neodymium, dysprosium, praseodymium and terbium, which are the most important ingredients in neodymium permanent magnets. The numbers shown for the sold Prius hybrids are only for the model Prius and it might be good to consider that there are other car manufacturers than Toyota that also produces hybrid vehicles with similar technology using permanent magnets. Other car manufacturers that produce hybrid vehicles with combined combustion and electric engines are for instance Honda, Nissan and Mercedes to name a few. As long as the stocks of REMs do not become too scarce the technology of permanent magnets will probably stick around for quite some time. The future of neodymium permanent magnets is from one point of view hard to predict. This is due to the fact that many of the patents that restrict the production of such magnets expire in the coming years. What will happen to the market then is difficult to foretell. Possibly if the patents do restrict who can produce the magnets there might be greater competition and possibly more competitors on the market. There are also possibilities of prolonging some of the patents to keep the restrictions of who can produce the Nd-Fe-B magnets. However, there is very little published information about this and what might happen within the coming years. 7.3 NiMH batteries The NiMH battery is the most commonly used in the hybrid vehicles on the roads today. This is a lot thanks to Toyota who decided to use these in their very popular Prius. There are other kinds of batteries that can be used in hybrids, but Toyota managed to find a way to mass produce NiMH batteries cost effectively. Chapter 5.4 estimates that a modern battery used in the Prius contains about 10 kg rare earths, of which at least two (dysprosium and neodymium) are forecasted to be in deficit in the future. However, for the future of hybrids and plug-in electric vehicles this does not have to present a problem. The reason is that many battery manufacturers are producing and/or developing Li-ion batteries instead. Even Toyota will be using Li-ion batteries in a coming plug-in version of Prius. 47
Chalmers University of Technology
Studying battery manufacturers show that many companies has the patent from Ovonics to produce the NiMH battery, and many has also put research into developing a battery for vehicles. However, most companies do not seem to have been successful, many are now switching to Li-ion instead. The companies that have succeeded have been those with a close partnership with a car manufacturer, e.g. Primearth (former Panasonic) and Sanyo. A guess is that this partly has to do with the cost effectiveness of mass production, made possible with reliable demand. The question is then if the NiMH battery is an obsolete technology. Our answer is: maybe. The NiMH will probably stay on the market for years to come, since it is an established technology that is successful. But for the future the Li-ion battery will probably be a preferred alternative. For plug-ins it is definitely the preferred choice since its qualities are better suited for that application. But depending on how the price curve for rare earths will develop in the future, Li-ion could perhaps also be the more cost effective choice. The conclusion is that the NiMH battery using rare earths do not seem to be a restricting factor for the growth of the hybrid or plug-in car technology. 7.4 Recycling The demand for rare earths might also increase the possibilities of recovering rare earths from scrap metal and discarded products. As indicated in chapter 5 there are huge quantities of rare earth metals in use that at some point will reach the end of their lifecycle. This opens up for recycling instead of mining for virgin materials. In a future perspective it is likely that the mining of virgin rare earth elements will increase as well as recycling and recovering processes probably will be further developed and put into use. The material recycling and recovery from scrap and discarded products might be a very valuable asset for countries without much natural resources, like Japan, who has an extensive industry and a need for raw materials. It would also make such resource poor countries less dependent on imports from other countries. According to articles and other published documents there seems to be quite a lot of activity in the area of recycling of rare earths currently. There has been a boost of interest in recycling processes that might be a result of the decreased exports quotas of REMs from China and rising prices. The demand for secure deliveries of REMs and stable stocks pushes the development of recycling methods that could recover large amounts of REMs in an efficient and economically feasible way. Products that at some point in the “end-of-life” stage could be recycled or recovered could be permanent magnets in wind turbines, motors, catalysts, air conditioners, computer monitors and other larger products. When it comes to hybrid vehicles, the ones produced by non-Toyota manufacturers added to the two million sold Prius cars till September of 2010 can be seen as a valuable stock of potential recyclable and recoverable resource since there are REMs in the motor and the batteries. Worth noting is that there is only about one kilogram of rare earth elements in a hybrid vehicle motor and around ten kilograms in the battery while there might be several hundred kilograms in a wind power plant which could be an even more important asset of valuable rare earth elements in the future. Most likely the recycling and recovering of REMs will in the initial stages only handle larger applications that contain fairly large amounts of REMs. Too small contents of REMs are in the initial phase not going to be economically feasible to recover. There will probably be a lot of development in the area of rare earth recycling and recovering in the coming years. 48
Chalmers University of Technology
7.5 Sweden perspective There are both possibilities and disadvantages of having parts of the rare earth product chain in Sweden, but worth mentioning is that it is very hard to estimate the likelihood and to what extent Sweden can take part of the product chain. Still, it is an interesting topic to speculate about. The greatest disadvantage from a Swedish perspective might be if there is a need for very large investments to small resources, and this insecurity might be a long term problem as investors prefer security. Many different perspectives can be taken into consideration when assessing the probability of having parts of the production chain in Sweden. It very much depends on economic and technical issues. Considering the Swedish knowledge within metallurgical processes it would be likely to obtain the know how in order to process REMs. The main issue would in that case be the economic aspect. It is difficult to estimate how much of an issue this might be but since Sweden does not have any parts of the product chain today it would most likely demand large investments in order to obtain processes in Sweden. Contributing factors that might affect the outcome would be the possible benefits for Swedish industry, possibility of exports, demand and market request. In an initial phase the mining for virgin materials or recovering of REMs from tailings would be possible activities since these are being evaluated and researched right now. Regarding the refining processes of REMs it is more difficult to estimate the Swedish participation. The part of the product chain that is closest to realization is probably virgin mining, possibly in Norra Kärr, or perhaps extraction of rare earths from apatite deposits. It would have been interesting to discuss this topic with more people within the metallurgical industry in Sweden but it is also difficult to find people with knowledge about an industry that does not yet exist in Sweden. A contributing factor that also is an obstacle for the possibilities of producing rare earth permanent magnets in Sweden is the restricting patent laws. If the restrictions are dismissed or permissions granted it would be legal to produce neodymium permanent magnets in Sweden. However, the feasibility of producing the magnets in Sweden could be questioned due to labor costs that might be higher in Sweden than in Asia. From this perspective the level of automation in the production stages also becomes very important. With more automatized processes the cost for employees decreases and makes it easier to compete with low wage countries. Also the efficiency of the processes becomes an important factor. Highly efficient processes could add to the benefits as high efficiency increases the output of material in comparison to the input which result in less waste and scrapped material. Under right circumstances and with a market for rare earths in Sweden and Europe the likelihood of having parts of the production stages of rare earth permanent magnets is a possibility. Sweden has great knowledge within metallurgical industry as well as knowledge about highly technological processes which speaks for the rare earth industry taking place in Sweden. 7.6 Method When the work on this report first started there was a reasonably clear idea of what it was supposed to contain and what the outcome was going to be. The tools that were used in this report we were already acquainted with from earlier in the master program Industrial Ecology here at Chalmers. Therefore they were sort of conjointly used instinctively from the beginning without there being a defined method yet. This meant that we had to sort out which tools we were using, how and in what stage of the work. 49
Chalmers University of Technology
The perspective on the product chain was taken from LCA, where the whole life cycle with all its processes are included. There was however no interest in performing the impact assessment of the LCA, this tool was just used as a way of thinking about the life cycle of the rare earth elements. Chapter 3 briefly mentions some of the environmental impacts from the mining, since this is of course an important matter when considering sustainability of rare earth production, but this is just to inform the reader about the issue. It would have been interesting to look more closely at this, especially since rare earths are used in “green” technologies. Perhaps to compare impacts from the production of the rare earth raw material, and the impacts the products containing rare earths would save (compared to other products with the same functions, e.g. 100 km driven with a Prius versus gasoline combustion engine car). However, this was not the purpose or aim for this report. The perhaps most influencing tool was the MFA. The flow of material in the product chain has been mapped on a global scale to the extent possible. The interest was to see where this raw material was produced and where it goes from there. The thought behind this was to prepare for the TA, where one of the research questions was if rare earths will be a constraining material in the future because of insufficient supply. The exact quantities for the flows have not been balanced, that would have been very difficult to do because it would have been difficult to find data for it for several reasons. One of the reasons is that this report was limited to six elements that are used in specified applications, but rare earth elements are often sold as oxides (in which they are mixed). Therefore it is difficult to find out which flows will end up in hybrid vehicles. Some numbers that are relevant for discussing the research questions have been presented: reserves, production, in-use stocks, and demand. The parts of the MFA that were excluded were replaced with CCA. Instead of tracking the exact flow quantities, the actors using the material were mapped. In some stages there were many actors, more than could be individually listed, in those cases they have been dealt with as a group with common features. One example of this is all the miners in China; they have been dealt with as a group since they are all found in the same nation, producing the same raw material. In later steps it was easier to distinct different actors, e.g. the actors that had licenses to produce sintered permanent magnets. Inspired by CCA the report has also tried to distinguish the relations between the actors in the chain; where in the chain can e.g. joint ventures be found and why. The reason for this is that when it comes to rare earth elements the numbers for reserves and demand is not enough to explain why there might be a risk of supply shortages in the future. Only comparing that, there does not seem to be a problem. The issue lies to a large extent in geopolitics, technology development, police targets and economic factors. The TA was used as an inspiration for the view of the discussion. Many nations nowadays see rare earths as strategic materials, not the least because they are often used in military applications. The interest in this matter was mainly to investigate if rare earths could constrain the industry of electric vehicles in the future. This report provides quite a broad perspective. Since a few tools have been combined none of them have been executed thoroughly. In a way this has been both good and bad. It has been good in the way that the snapshot picture the report was aimed to provide is quite inclusive when it comes to information about the product chains. The bad thing is that the report sometimes lacks depth in information about specific topics, it is up to the reader to further investigate special interests. This 50
Chalmers University of Technology
report will not provide any new information that did not already exist; it has rather collected information from many sources to provide an overall picture of the current situation. However, that is in line with the aim of this report. So in all the authors of this report feel that the right method was used in order to fulfill the purpose and it has been easy to work with the method as it is customized by combining several tools which all had fitting characteristics. Some problems in the research had to do with finding information. Since there were no existing works with the same purpose as this report, information had to be collected from many different sources to complete the picture that it wanted to provide. This is why so many different sources have been used: scientific papers, books, news media, organizational information and reports, governmental information, company information, interviews with selected people, and blogs. The internet has been a very useful tool to get in contact with the information sources. E-mail correspondence has been frequently used to acquire information. This method has worked very well with people from Europe and the Americas, and not at all with people from Asia. This probably has a lot to do with cultural differences, and how we do contact. It could have been useful to actually travel to China or Japan, or visit conferences where we could have gotten in touch with people in person. However, that was not a possibility since there was no finance for this. And of course the actors involved in the chain would not provide confidential company information, which left us with annual reports or other public information. Another problem was constant stream of new information. A lot is happening within this area at the moment, so to have the latest updated information meant that text that was written in the beginning of the project of this report has to be revised and corrected later. It was also difficult to screen the information flow for accurate information. 51
Chalmers University of Technology
8 Conclusions Demand for rare earths will most likely keep growing in the next coming years. There are many applications for rare earths due to unique qualities; they are especially suitable for green technologies where they can improve performance and reduce weight. There is a risk of some rare earth elements running short in the future. If large scale wind power stations increase in number along with other technologies using the same elements, risks are that not even with a growing mining industry the demand will be met. The elements that are most likely to become deficit are neodymium, dysprosium and terbium. This would probably be an effect much related to a growing production of permanent magnets using these elements, and also to the fact that these elements only occur in small percentages in the REO of many deposits in the world. Today there are few producers outside of China. Most of the intellectual capital is therefore also in China today. Nevertheless, there will probably be outside-China production within next three years. Intellectual capital is attained through merger and acquisition between companies with different competency. Since no official trade market exists today it is common that there are joint ventures, part ownerships or other agreements between companies in the different steps of the production chain. This is a way to secure business. For the rare earth producers this is important since their operations are very costly. For the manufacturers of rare earth containing products this is important because it would be very costly if they could not get their supply. With the mining projects that are underway today the predicted demand of TREO is likely to be met. The issue of deficit will probably just apply to some of the elements; those that are used in the NdFeB permanent magnets are among those. The reserves known today are likely to increase in the future. According to the definition of reserves they only include those parts of a resource that can be economically recoverable with existing technology. Since the price of some rare earth elements have escalated during the last years the profitability of mining has improved, which might in the future weigh up for high extraction costs. There are also uncertain signs that discoveries of new deposits have been made in countries like Afghanistan and Pakistan, which could add to the reserves. There are many co-operations and joint ventures in the product chains that were analyzed in this report. This has to do with the lack of an official trade of rare earths today and the restrictions of patent rights for production. Rare earths will probably not be a constraining factor for battery production for electric vehicles (HEVs and PEVs) since the lithium ion battery is predicted to be the preferred/dominating alternative in the future (the question whether there is enough lithium that can be produced in a sustainable way is outside the scope of this report). However, if the NiMH battery production for hybrids and electric vehicles would increase, which is plausible with an expected growth in the HEV/PEV market, the elements at risk would probably be neodymium (because of competition from permanent magnets) and praseodymium (because of low concentration in the REO of most deposits). Also, the transition to Li-ion batteries will take decades. 52
Chalmers University of Technology
Small Earthfill Dams Operating in the Mining Industry) Conceptual Design Master of Science Thesis in the Master’s Programme JOHANNA LUNDIN Department of Civil and Environmental Engineering Division of GeoEngineering Geology Chalmers University of Technology ABSTRACT The mining industry produces large volumes of residue water and it is necessary to have systems to contain and treat the water. In the water treatment process small dam constructions, forming reservoirs, take a large part. This thesis has intended to investigate the present small dams in five mines within the area of Boliden, Sweden. The considered constructions have been earthfill dams with heights between one to five meters, constructed with material found close to the mine. The investigation has concerned factors that contribute to the function and stability of the dam. Parameters determined to affect the stability has been slope stability and hydraulic gradient. The conceptual model consists of two dam heights, three and five meters, and two different slopes, 1:2 and 1:2.5. There are four operating scenarios that have been analysed according to slope stability, these are: steady state, steady state with load, rapid drawdown and rapid drawdown and load. Conclusions made are that for small earthfill dams it is essential for the stability of the dam to verify impacts of different material properties. Furthermore, a safe dam environment for small dams is similar with large dam and depended on clear directives regarding dam operation procedures. Key words: Earthfill Dams, Mining Industry, Water Storage, Hydraulic gradients, Slope Stability. I
Chalmers University of Technology
Preface This study has taken place in the area of Boliden County and the facilities of the mining company Boliden Mineral. The effort with this thesis has been carried out from June 2013 to December 2013. Together with Boliden Mineral and the author the work was initiated as a master thesis concluding the studies for a Master´s degree in Civil Engineering, within the department of Civil and Environmental Engineering and the division of GeoEngineering at Chalmers University of Technology. Supervisor from the department of Civil and Environmental Engineering has been Johan Funehag. Johan have shown patience when introducing the world of technical writing for me. This study has been carried out together with Boliden Mineral with Camilla Årebäck as supervisor. Special thanks for making me feel comfortable, welcome and for interesting conversations concerning application areas of dam engineering. Also a special thanks to Johan’s colleagues at the Tyréns office in Skellefteå for an office space and contributing with experience. The extensive part of the field study as wells as reflection concerning dams and the operating in mining industry has been essential for the thesis. Special thanks to Camilla Årebäck and Michael Sandberg at Boliden Mineral for guiding and describing the process of construction of small dams within the mining industry. Finally, it should be noted that this thesis would not have been possible without all the kind, wise and always cheering family and friends I have the opportunity to have in my life. Boliden November 2013 Johanna Lundin CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 V
Chalmers University of Technology
1 Introduction Dam constructions are found in civil engineering activities all over the world and have a long history within infrastructure development. Construction of dams is diverse and dependent on the purpose with the retention of water. Application areas in minor extent are water irrigation and agriculture purposes, and in larger extent hydropower or reservoir storage. Novak et al. (1990) define that the primary purpose of a dam construction is to provide “a safe retention and storage of water”. Appearances of dams show diversity in geometry, material and application area (Vick, 1990). Dam constructions may be divided into two groups, dependent on the material used: - Embankment dams. Embankment dams are constructed from earthfill and/or rockfill. Angular slopes forming a rather wide construction. - Concrete dams. Dam constructed from concrete, construction appearance may vary from steep downstream sides and vertical upstream. It is possible to create more slender constructions. In the Skellefteå area the metal company Boliden Mineral has been exploring and mining ore since the early 20th century. Five different mines are in operation today; Kristineberg, Kankberg, Maurliden, Maurliden Östra and Renström mines. Current production considers complex ore containing zinc, copper, lead, gold and silver. In total the production per year is 1.9 Mtonnes of ore and there are around 500 employees today. Within the mining industry there are large tailings dams, reservoirs that contain residue tailings from the processing of ore. But there are also other application areas for dam construction within the industry. Newly constructed dams within the area of Boliden are earthfill dams from natural materials, preferable from the area close to the construction site. Soil material with different fractions and characteristics will form a united construction without any binding, only by compaction. The mining activity causes a lot of residue water. It is necessary to contain and treat the water, which is carried out by various methods. Minor dam construction forming reservoirs takes a large part of these processes. The water contained may be from the mine activity and activities in connection to the mine. Examples could be short time storage of residue water, sedimentation basins both before and after treatment facilities and collection of surface runoff. The systems of the dams are continuously constructed due to mine production demands. One frequent problem has been extended seepage when the dam is taken into production. Problems may be due to variations in the material and performance during construction. Due to production processes there are frequently changing reservoir levels and the dams are exposed to various loading conditions over the entire life span. It is essential that the dam should provide a reliable system for the water treatment process during normal operating conditions. There is also a need to have evaluated the stability and reliability during extreme conditions. For larger dam constructions there are clear regulations and legislation for construction and operating, it has not been evaluated whether there is possible to apply the same regulation for these in relation small dams. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 1
Chalmers University of Technology
1.1 Aim of this Thesis The aim of this thesis is to investigate the current constructions of small dams in the area of Boliden. According to the present constructions and materials, the investigation will concern function and stability of small dams operating in the mining industry. 1.1.1 Objectives of this Thesis Variations in reservoir level have been considered to affect the stability of the slope and the hydraulic gradients within the dam. The following two objectives will be further investigated: - The slope of the dam will be exposed to various pressures when the reservoir level is frequently changed. What material properties will govern the stability and conditions at which the slope is stable will be analysed. - The hydraulic gradient within the embankment is affected by the variations in the reservoir level. The investigation will consider which material properties are governing the internal stability and when there may be a risk of failure. 1.1.2 Limitations This thesis will only consider constructions with a height between one to five meters. The material properties used will be according to an extensive investigation of a till deposit by CM Tracing and Christer Mattsson (2007). In the case when there is a lack of information concerning material properties, present standards and characteristic values will be applied. The areas of interest are assumed to have ground conditions with large depths of till. The characteristics of the foundation need to be determined for each individual construction site. From the visits to the different sites the extent and conditions of small dam constructions have been obtained. The thesis have not contained any on-site testing, there have only been visible control and interviews with staff in the mines of interest. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 2
Chalmers University of Technology
To further understand what the different systems of a dam are and how they work, different features are described in Figure 2. Dam construction is to a large extent controlled by the interaction between different properties of dam features. Characteristics of certain interest have then been described more in detail below. 1Core 2Fine Filter 3Intermediary Filter 4Coarse Filter 5Support Fill 6Grouting 7Surface Liner 8Erosion Protection Material 9Dam crest 10Toe Reinforcement Figure 2 Characteristic and details of an embankment dam (SveMin, 2012). Core The core consists of a less permeable material than the surrounding support fill to prevent seepage through the dam (Fell et al., 2005). The permeability will be governed by the material in the core. A dam consisting of core, support fill and other design features are referred to as a zoned dam. Filter systems The usages of filter and drainage systems have several purposes. They are used to prevent erosion of material within the construction, but also to collect eroded material due to seepage. The system will also function as a draining system decreasing porewater pressures and containing seepage within the construction (Fell et al., 2005). There are different kinds of filter solutions and the material properties will be diverse depending on the position in the construction. Fine filters are found close to the core preventing seepage and collecting particles. Coarser filter will be placed outside the fine filter, collecting particles that manage to get through. The filter will have more permeable properties than the surrounding soil and contain drainage characteristics (SveMin, 2012). Support fill The material will function as the stability factor of the construction as well as surrounding and protecting other design features. Depending on the properties of the material, the support fill also prevents transport of eroded material (Fell et al., 2005). A dam consisting of only support fill is referred to as a homogenous dam. Erosion protection material The dam construction needs to be protected from damages caused by wave actions, freezing and surface runoff (SveMin, 2012). The extent of wave actions is dependent on the wind speed, reservoir length, the wave duration and the allocation of depth in the reservoir. The most common material for erosion protection material is a boulder sized material sometimes called riprap. There are two methods of application systems. The erosion protection material is either dumped from the crest down on the slope. Or the material is dumped and later applied with the help of excavators. This is called an arranged erosion protection (SveMin, 2012). CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 4
Chalmers University of Technology
Dam crest The dam crest will be dependent on the composition of the dam. There should be a sufficient distance to the core and other frost sensitive materials to avoid problems with freezing. The width of the crest should provide space for the required transportation during construction and the life span of the dam (SveMin, 2012). Toe bank To improve the stability in a downstream slope sometimes a reinforcement of the toe is suitable (SveMin, 2012). The reinforcement consists of a boulder sized material and the intention is that it should work as a resisting force on the slope. It is important that the material has draining properties so the porewater pressure will not be built up in the dam. Freeboard The freeboard is defined as the distance between the reservoir level and the dam crest. The dimension of a freeboard needs to be sufficient to prevent damages on the dam caused by waves in combination with winds. The freeboard need to prevent overtopping, which means that water will be running over the crest. This may with high probability lead to failure of the dam (ICOLD, 1995). Figure 3 displays an illustration of an upstream slope where the wave run up height R is shown. Figure 3 The freeboard represents the distance between reservoir level and the dam crest (ICOLD, 2005-2010). Discharge system The outlet from a reservoir is called discharge system. The capacity should be dimensioned to take care of the expected spillwater from the reservoir. But also to discharge additional water causing a potential safety risk due to high levels in the reservoir (ICOLD, 2005-2010). Discharge systems may consist of different solutions, for example: overflow thresholds, natural overflows dependent on the surrounding environment, pipes or conduits. Foundation According to SwedenergyAB (2012) the foundation of the dam must be “designed for safe interaction with the foundation” and “be drained to eliminate the risk of leakage, internal erosion and instability”. To determine what needs that have to be met considering foundation, the permeability of the material in the foundation has to be verified. GruvRIDAS (2012) states that if the foundation contains equivalent qualities as the embankment, the ground may be smoothened and compacted in the same manners as the embankment itself. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 5
Chalmers University of Technology
Independently if the embankment is to be founded on soil or rock, it is recommended to investigate the properties of the material (SveMin, 2012). Some construction examples of when a dam is founded on soil are presented below (Fell et al., 2005). - General foundation excavation. The weak soils are removed in order to strengthen the bedrock until sufficient strength is reached. - Cut off excavation foundation. Materials of high permeability below foundation level are removed to avoid uncontrolled seepage below the dam construction. - Curtain grouting. Grout is injected into the rock foundation to decrease the permeability of the bedrock. - Consolidation grouting. Injection with grout in the cutoff of the foundation is grouted to reduce the permeability. This solution is also called Blanket Grouting. Transition Zone Transition zones are found between for example downstream filter and support fill, support fill and coarser material at the downstream toe and erosion material as well as upstream support filling. The risk with these areas is the differential in settlements causing possible areas for internal erosion or deformations in the outer areas of the dam (Fell et al., 2005). Liner System Even if the dam is designed to be stable and fulfilling the construction features, there will always be seepage through the construction (Fell et al., 2005). To decrease or prevent this, a lining system may be used on the bottom and slopes of the construction. Consisting of a system of material preventing, collecting or decreasing leakage of both natural and synthetically materials. Geomembranes may be used to control and prevent leakage of residue water. Geomembranes consist of a rubber blanket and are within the mining industry used when there is a need for waterproof impoundments. Regulations for disposals of waste are found in the Swedish Environmental code and further recommendations to this legislation are found in Handbook 2004:2. Geological barriers are without active measures and maintenance supposed to function during a long lifespan. When natural conditions do not fulfil this, counteractions must be taken. Recommendations from regulation according to Naturvårdsverket (2008) a liner system should consist of at least one meter thick layer and permeability according to Table 1. Table 1 Limitation for permeability of a liner system according to regulation from Naturvårdsverket (Rättsnätet, 2001). Deposited Material Permeability Thickness Hazardous waste <1,0*10-9 [m/s] > 5 [m] Non - Hazardous waste <1,0*10-9 [m/s] > 1 [m] Inconvenient waste <1,0*10-7 [m/s] > 1 [m] CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 6
Chalmers University of Technology
1.2.2 Loading Conditions to be considered for an Embankment Dam When constructing a dam there are some parameters that need to be considered before choosing what kind of constructions to use. Novak et al (1990) claims that environmental aspects and regulations certainly need to be respected. There are certain loads that affect the stability of the dam construction these may be divided into three groups; primary, secondary and extreme. Figure 4 displays an illustration of the different loading conditions on a dam construction. The proportions of the three loading categories will be unique for each construction. Figure 4 Example of loading conditions on a dam with reservoir levels on both sides. Loading conditions according to Novak et al (1990) are presented below. - Primary loads. The load of most importance for dam safety independent of type: o Water load. Hydrostatic distribution of pressure with resultant force, L 1. o Self-weight load. The unit weight of material, forming the resultant L , 2 considered operating from the centre of a given section. o Seepage load. The unavoidable seepage through a dam in voids and discontinuities, will contribute to vertical loads. This may be seen as both internal and external loads, L and L . 3 4 - Secondary loads. Universally applied loads from environment and surroundings: o Sediment loads. An additional hydrostatical load will be the accumulated materials or sediments will give the resultant L . 5 o Hydrodynamic loads. Transient load formed by waves acting on the dam, L . 6 o Ice load. Loads due to formation ice floe, L , during winter time. 7 Freezing may have different consequences for concrete and embankment dams. o Thermal load. Internal changes due to cement hydration and cooling. Important when dealing with concrete dams. o Interactive effects. Possible internal changes in relative stiffness and diverse deformations within the construction. - Exceptional loads: o Seismic loads. o Tectonic effects. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 7
Chalmers University of Technology
1.2.3 Hydraulic Behaviour in Soils Dam construction is dependent on soil properties and the hydraulic behaviour of the soil. The governing factor for an embankment dam construction and a potential failure are to a large extent dependent on the porewater pressure. This needs to be considered through the process of construction and finally the lifespan of the dam. Terzaghi et al. (1996) describes three critical stages of certain interest to follow the porewater pressure conditions: - Construction. During construction and immediately after completion. - Full reservoir. With a full reservoir and when steady state conditions have been reach in the dam and foundation. - Drawdown states. During and immediately after lowering of the reservoir level. The total normal stress through a saturated soil consists of two parts. First the porewater pressure or what may be termed as the neutral stress, working in all directions with equal intensity (Terzaghi et al., 1996). Secondly this represents the excess over the normal stress and will consider only the solid phase of the soil, called the effective stress. Cedergren (1988) states that a permeable material is one that is capable of being penetrated by another substances, usually a gas or a liquid. Jantzer (2009) describes the phenomena of the permeability of a soil by: “…a material ability to transmit a fluid..”. A soil with voids filled with water is referred to as a saturated soil also called an undrained soil (Johnson & DeGraff, 1988). The opposite scenario would be if the voids of the soil are filled with air, called drained conditions. The movement or flow of water through the soil will be dependent on the permeability, called the coefficient of permeability. Darcy’s Law are used to determine the coefficient, see equation 1.1. (1.1) An applied load on a soil will contribute to both compression and shear forces. Water will only provide resistance towards compression forces and resistance for shear forces will be dependent on the effective stress of the material (Johnson & DeGraff, 1988). The effective stress are dependent on the pore pressure, so when raising the pore pressure within the soil the pore pressure will increase causing an decrease in the effective stress. The critical point will be at the state where the effective stress is getting closer to zero. At that state the pressure within the voids will push the particles and lose contact and the resistance to shear forces are the least. The porewater pressure may be determined be the equation 1.2. (1.2) The hydrostatic pressure in a soil represents the energy level at that certain point when the water is motionless (Johnson & DeGraff, 1988). The energy level may be determined by using piezometers. If the piezometer indicates different energy levels, pressure head, it is called hydraulic gradient. The flow of water will contribute to seepage pressure due to the difference in total energy between two points. It needs to be determined whether the seepage pressure may potentially be high enough to move individual particles. This may potentially led to greater seepage pressure and higher flow velocities. Seepage pressure is determined by multiplying the hydraulic gradient and the unit weight of water. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 8
Chalmers University of Technology
1.2.4 Compaction of Soils Johnson & DeGraff (1988) describes compressibility of the soil is the decrease in volume of a soil mass due to natural and artificial means. With artificial means one often indicates compaction. It might be performed by vibrating on top of a soil mass or loading and unloading of material. The intention is to decrease the volume of voids in the soil. The amount of compaction possible will to a great extent depend on the moisture content of the soil. Compaction is applied in all construction areas using soil material. The intention is to reduce permeability and increase the overall strength of the soil. During construction the compaction is investigated to verify that the design specifications are met. Johnson & Degraff (1988) describes the present of water in the soil as an increase in resistance for particles to reorientation due to surface tension between the grains. But the frictional resistance may reduce due to water content in the soil. There is a need to find a level of water content where the maximum dry density are found, this level is called the optimum water content. Verifications may be performed by using standard Proctor compaction or modified Proctor compaction. This method will give the responding dry density in terms of different levels of water content and specified compaction energy. An example may be seen in Figure 5 where S represents the compaction energy. Figure 5 The responding dry density in relation to the moisture content of the soil and the compaction energy. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 9
Chalmers University of Technology
1.2.5 Shear Stresses in Soils and Till as Dam Construction Material General assumption and main advantage of till as a construction material is relatively low permeability, high shear strength and low deformability in comparison with other fraction materials (Sherard, 1986). These qualities may be determined due to the broad range of grain size distribution together with a high content of fines. The composition of different grains depends on the rock of origin and the quality of the grains. Shear strength are the largest shear stress possible to be obtained on a surface of a material (Vägverket, 1986). In friction soils, the maximum shear stresses are proportional to the normal stress towards the surface and the tangent of the friction angle. For friction soils the friction angle will be governing for the shear strength. The friction angle is not only material-dependent; it will also be affected by the stress levels in the soil. When the effective stress increases, the friction angle will decrease. Slope stability is controlled by the shear stresses of the materials in the dam body and the foundation. For a homogenous dam the slope itself is the critical slip surface. To evaluate the failure criterion of a soil material, the Mohr-Coulomb failure criterion is used and the shear strength is determined according to equation 1.4 (Terzaghi et al., 1996). 𝛷 (1.4) Stress calculations in soils normally assume that the Mohr – Coulombs failure hypothesis is valid, see Figure 6. The failure criterion is determined by the effective strength parameters cohesion and frictional angle and the effective normal strength (Terzaghi et al., 1996). The intersection with the axis where σ=0 represents the cohesion of the soil. The line formed is the strength envelope for the material, which depends on σ and σ . Using Mohr circles, the increasing σ will gradually develop 1 3 1 towards the strength envelop at the point where the circle tangent to the line failure occurs. Figure 6 Mohr rupture diagram, the straight line represents the failure envelope (Terzaghi et al., 1996). In relation to this the factor of safety may be determined according to the relation between, accessible shear stress appearing at maximum friction coefficient and the mobilized shear stress (Sällfors, 2001), seen in equation 1.5. (1.5) CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 10
Chalmers University of Technology
Soil materials of silt, fine- and medium sand and silty tills have sometimes been found to be stable in more steeper slopes than the responding friction angle of the soil (Vägverket, 1986). This is termed as false cohesion and may be explained with the capillary tensile stresses which increase the pressure between the soil particles. Capillary stress in the water in connection with the particle´s contact surfaces causes a false cohesion in the material. When the soil is saturate, the capillary stresses disappear and the soil recovers its original friction angle. According to ICOLD (1989), construction with till material is prohibited during freezing seasons. Even if there are measures taken to protect the material, the hauling, initial temperature, wind velocities are hard to predict. Till that has been frozen and then melted may lead to softening of the compaction, a reduction in shear strength and an increase in permeability. According to ICOLD (1989) there are two approaches that may be taken when constructing a dam in a till material and having winter conditions. - All production will be completed when the ground freezes and material already on place should be covered to avoid frost penetration. - If the work continuous even during cold period’s actions needs to be taken, such as: adding chemicals, till in stockpiles, heated or stored water. 1.2.6 Critical Scenarios for Embankment Dams According to ICOLD, (1995) the term failure is: ”Collapse or movement of part of a dam or its foundation, so that the dam cannot retain water”. In general, a failure results in the release of large quantities of water, imposing a risk on the people or property downstream. There are two types of failures categorized by ICOLD (1974): - Type1 - “A major failure involving the complete abandonment of the dam” - Type2 - “A failure which at the time may have been severe, but yet has permitted the extent of damage to successfully be repaired, and brought into use again” The most common mode for failure of a dam is overtopping of reservoir water (Fell et al., 2005). According to statistics the second most common reason for failure in earthfill dams is internal erosion by for example piping, a seepage path forming within the embankment. Piping may be initiated by backward erosion, concentrated leak or suffusion. In Figure 7 backward erosion are illustrated, to the left erosion on the downstream side of the core. To the right the process has continued and the formation of a pipe is displayed forming a path from the reservoir through the embankment. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 11
Chalmers University of Technology
Figure 7 Example of backward erosion initiated in the core and backwards increasing (Fell et al., 2005). According to Terzaghi et al. (1996) the critical states for an embankment dam do not only depend on the material, but also the boundary conditions for drainage when considering porewater pressure conditions. Concerning the effects on porewater pressure due to drawdown of reservoir level, called Rapid DrawDown (RDD), the reason for changes will be dependent on the compressibility of the different parts of the construction. In Figure 8 two situations for reservoir levels are illustrated, together with the responding flow net for each scenario. The figure to the left describes the flow of water through the dam and responding point in the filter toe. The scenario to the right describes when the reservoir has been rapidly emptied and the flow nets are divided between the upstream and the downstream slope. Figure 8 Different scenarios illustrating change in flow net due to rapid loss of resisting load (Terzaghi et al., 1996). The rapid loss of resisting load on the reservoir side will cause high porewater pressures in the upstream slope and increase the risk of failure depending on the extent of the effective stress. The dam consists of a fine, clean, well-compacted sand and foundation of an impermeable material. When the construction is exposed of a RDD of reservoir level, the behaviour of the flow towards the downstream flow will continue as before. But the upstream side will cause potential instability problems on the slope side. The pressure will sequentially decrease from the top line and downwards. 1.2.7 Regulation Concerning Construction of Dams In Sweden today there are several different provisions applied to create a safe dam environment. The Swedish environmental code is one of the dominant ones regarding dam safety. The Swedish Association of Power Plants, Swedenergy, accepted the Hydropower industry’s Dam Safety Guidelines, RIDAS in 1997 (SwedenergyAB, 2012). The latest revision was published in 2012. It is clarified that RIDAS should not be seen as regulation but as guidelines to a safer dam performance. The guidelines consist of two parts, first Guidance and secondly Application Instructions. According to the guidance dam safety is defined as: “safety against the uncontrolled outflow of water from the reservoir (dam failure) that could result in damage or injury”. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 12
Chalmers University of Technology
This demands that the additional water is drained or pumped away. This is a quite time-consuming and extensive process, but still a necessary part of the treatment process. Reservoirs for storage will often have, as mentioned, systems where the water is pumped away. The design of discharge systems is dependent on the reservoir level, as well as the amount of freeboard. Example when guidance in GruvRIDAS (SveMin, 2012) are hard to apply are the demand for a minimum freeboard of 2 meters. From the visits on site and interviews, some areas of problem have been identified. For example there have been problems with large amounts of seepage through the dam in the earlier stages of production requiring reinforcement. In one of the areas by ice lenses have been discovered within the dam, caused due to production during wintertime. Stability problems in slopes have caused slope failure on the upstream side of the reservoir. The instability of the slope has appeared when the level in the reservoir has been lowered. The source of the problem has not been investigated, but solution has been to reinforce the slope and decrease the slope angle. 2.2 Literature Study The aim of the literature study has been to understand the governing properties for a dam construction and the boundary conditions for a small dam operating in the mining industry. The literature study is presented in the introduction chapter and has together with the field study formed the foundation of the thesis. 2.3 Conceptual Model The process of evaluating the present constructions in the area has been carried out in an iterative way. A general model has been applied to be able to verify the function of the dam and the numbers are based upon the field study. It is stated that the small earthfill dams consist of natural material with the following characteristics: - Dam height of one to five meters - Width of bottom 20 meters - Length of bottom 100 meters - Crest width five meters The recurrent changes in reservoir level are applying high stress on the construction and will demand stability of the soil both internally and externally. To be able to evaluate the situation, a characteristic construction has been presented with water from the reservoir as the major loading condition, L and internal stresses due to 1, hydraulic conditions within the construction, L . The reservoir levels are assumed to 2 be full on both sides of the construction, see Figure 9. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 15
Chalmers University of Technology
Initially the cohesion was set to zero, representing the worst conditions for the slope stability. Then, by using various properties for the cohesion, the intention will be to get closer to an actual value of the present material. The stability controls have been carried out with two different dam heights, of three and five meters, assumed to be representative for the entire height interval. Slope angles have been determined to be 1:2 and 1:2.5 in the different sites and the focus have been to analyse the stability of these two geometries. Results from the analysis will be presented according to the factor of safety. There will be three material simulations, with different material properties. The first one have been based on recommendations of material properties from the consultant (Carlsson, 2013). In the second simulation some cohesion was applied in the material, assumed to be favourable for the stability of the slope. The third case was general values for the friction angle according to reference literature. The freeboard levels impact on stability was initially investigated. Figure 10 display a comparison of two freeboard heights and the different geometries. The normal operation scenarios, Steady state and steady state with load, and the extreme conditions, RDD and RDD and load, defined in the conceptual model are also shown. The vertical axis represents the factor of safety and the horizontal the operation scenario and slope angles. Blue and green bars represent freeboard height of 1.0 meter and red and purple freeboard height of 0.5 meters. As can be seen freeboard height of 0.5 meters will provide a more unsafe solution, especially for the extreme conditions and the dam with slope angles 1:2. Hence, the following simulations are based on a freeboard level of 0.5 meters, simulating the worst case scenario for the dam. Figure 10 Variations in factor of safety according to the two freeboard heights. 2.5 Internal Seepage Load To evaluate the conditions of the hydraulic gradient of the conceptual dam, the material investigation by Mattsson (2007) have been applied. Verified material properties together with standard values have together been used to determine critical hydraulic conditions. To be able to verify if there is a risk for internal erosion within the dam potentially leading to failure, one important parameter is that in all embankment dams there will be a flow of seepage. The amounts depend on the permeability of the soil material and the performance of the construction. Despite, in most cases, the flow of water does not contribute to dam failure. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 17
Chalmers University of Technology
There needs to be a balance between the total stress concept, where the total stress is pushing the particles together, and porewater pressure, pushing them apart (Johnson & DeGraff, 1988). The critical hydraulic gradient i is defined as the state where the c average seepage pressure is equal to the weight of the sand. Saying that the average seepage pressure will be equal to weight of the soil and the ratio will represent the critical gradient, equation 2.1 are used for evaluation. Input data and results of evaluation of the critical gradients are seen in Table 2. (2.1) Table 2 Calculated properties according to evaluation of material sampling. Minimum MaximumAverage Unit Porosity n 24% 21% 22% [%] (Trafikverket, 2010) Bulk density ρ 2,16 2,36 2,27 [t/m3] (Mattsson, 2007) Compact density ρ 2,65 2,70 2,68 [t/m3] (Trafikverket, 2010) s Density water ρ 1,00 1,00 1,00 [t/m3] (Trafikverket, 2010) w Critical hydraulic gradient i 1,26 1,35 1,31 [-] (Johnson & Degraff, 1988) c The stability of the soil would be fulfilled if the factor of safety is minimum 2, see equation 2.2. F (2.2) 2.6 Evaluation of Embankment Dam Features The conditions of features on the constructions will contribute to the overall dam safety. Some of the characteristics will be evaluated and their function will be compared to the present conditions found in the field study. 2.6.1 Dam Crest The dimensions have been based on a crest width of five meter, which was found to be sufficient for operating vehicles. Further parameters influencing the appearance of the crest is for example the geographical location, since there are several months of winter climate in the area of Boliden. There will be a risk of freezing of the material especially when the snow is removed from the crest. To avoid risk of damages when freezing there should be a sufficient distance between the dam crest and sensitive construction features like core, filter or drainage systems. The frost depth will be dependent on the material in the dam and the number of freezing hours in the area. In Figure 11, the variation in freezing hours in Sweden is presented. The freezing depth is evaluated according to the equation; √ . The factor k freezing represent material properties. In the area of the existing mines there are three different zones for freezing hours, noted with the black circles. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 18
Chalmers University of Technology
Figure 12 Illustration of parameters considered when determining the extent of freeboard (ICOLD, 2005-2010) Alternative methods to determine the freeboard of a dam have been used by ICOLD (2005-2010). In this equation the total height of the dam and the reservoir volume are taken into account, see equation 2.4. √ (2.4) Both of the methods have been applied and their respective results have been compared and evaluated together with other aspects considered relevant for the stability of the dam. 2.6.3 Discharge system Design parameters for discharge systems will be considered according to experiences from the field trip. Design need to be both functional and provide a safe dam environment. Installation may be in both natural material so as in concrete or by different pipe constructions. The capacity needs to be fulfilled for both normal operating conditions so as in case of additional inflow due to extensive run off or defects in the primary spillway systems. 2.6.4 Erosion Protection Material Due to the terrain in the area of interest, wind velocity on ten meters height above the reservoir level is assumed. The area is determined to be bare mountain and wind velocities of 25 m/s and the wind return time is set to 50-100 years are stated (SveMin, 2012). Dimensions of the erosion protection material have been determined by Vattenfall (1988) by using the significant wave height and the angle of the slope. Recommended dimension of the erosion protection material will be two times the material diameter. 2.6.5 Lining System Design of a liner system need to consider limitations concerning leakage as well as the processes connected to the function of the dam. Methods for evaluation have been practical once as well as technical aspects, to find the most suitable system according to production demands and current constructions. CHALMERS, Civil and Environmental Engineering, Master’s Thesis 2014:18 20