Repositorio Institucional de Documentos

2014 55 Rosa Adriana Domínguez Vega Exergy Cost Assessment in Global Mining Departamento Director/es Instituto Universitario de Investigación Mixto CIRCE Valero Capilla, Antonio Valero Delgado, Alicia Director/es Tesis Doctoral Autor Repositorio de la Universidad de Zaragoza – Zaguan http://zaguan.unizar.es UNIVERSIDAD DE ZARAGOZA Departamento Director/es Rosa Adriana Domínguez Vega EXERGY COST ASSESSMENT IN GLOBAL MINING Director/es UNIVERSIDAD DE ZARAGOZA Instituto Universitario de Investigación Mixto CIRCE 2014 Valero Capilla, Antonio Valero Delgado, Alicia Tesis Doctoral Autor Repositorio de la Universidad de Zaragoza – Zaguan http://zaguan.unizar.es Departamento Director/es Director/es Tesis Doctoral Autor Repositorio de la Universidad de Zaragoza – Zaguan http://zaguan.unizar.es UNIVERSIDAD DE ZARAGOZA Exergy Cost Assessment in Global Mining By Rosa Adriana Domínguez Vega PhD Dissertation Directed by: Dr. Antonio Valero Capilla Dr. Alicia Valero Delgado Zaragoza, 2014 Acknowledgements I want to express my gratitude to CONACYT (Consejo Nacional de Ciencia y Tecnología) Mexico for the financial support during the development of this dissertation. I want to thank Profr. Antonio Valero and Dra. Alicia Valero for the opportunity to work with them and for their guidance and valuable comments during these years, which made possible the development of this thesis. I would like to thank Prof. Wojciech Stanek from the Institute of Thermal Technology in the Silesian University of Technology, who was always willing to help me during my 3-month stay in his Institute. I want to thank my colleagues and the administrative staff of CIRCE, of the Mechanical Engineering Department, and of the University of Zaragoza for the friendly work environment during these years. Finally, I want to thank my parents for their unconditional support. i Abstract Economic, social and technological development of current society is strongly linked to the extraction of mineral resources. A society which is constantly growing consumes these resources fast and almost unlimited. The continual increase of worldwide demand of mineral resources is especially enhanced by the economic growth of China and other Asian countries, which require large amounts of raw materials in the construction, infrastructure and manufacture sectors. The depletion of non-renewable natural resources is the consequence of this progress and is the greatest challenge that the mining industry has to face. Consequently, future availability of mineral resources is increasingly gaining importance in the strategic government planning. It is well known that once the minerals have been extracted, several processes which consume large amounts of energy are needed in order to produce useful commodities. The energy consumption in mineral extraction and processing depends mainly on the minerals quality and composition. As a consequence of the depletion of mineral deposits, the energy requirements and environmental burden are constantly increasing due to a general reduction of global ore grades. Accordingly, more materials need to be processed to obtain an equivalent amount of metal. In this regard, a critical issue that mineral industry needs to overcome is the energy availability required for mineral extraction and processing. Therefore, it is of imperative importance to analyze and understand the processes involved in the mining industry so as to determine potential improvements while accounting for the depletion factor of raw materials. The first activity can be carried out via a thermoeconomic approach. Thermoeconomics has been traditionally used for the optimization of thermal power plants using exergy as the unit of measure. In this PhD, the thermoeconomic analysis is adapted and modified to cover the complexities of mining and metallurgical processes, where raw materials in addition to energy flows come into play. That said, if the scarcity factor of mineral resources wants to be taken into account in the analysis, an additional variable needs to be included. This is done through the exergoecological approach proposed by Valero et al. (2003). Conceptually, the Exergoecological method allows for an evaluation of mineral resources by means of the Exergy Replacement Cost, which is the exergy required to return to the initial state of concentration and composition found in mines, the minerals that have been totally dispersed throughout the crust once their useful lives have come to an end. Accordingly, the aim of this PhD is to adapt and apply thermoeconomic methodologies so as to perform an absolute Life Cycle Assessment (LCA) of mineral resources: a conventional one from the cradle to the entry gate (production of the refined raw material) and an additional one from the grave to the cradle, thereby accounting for the depletion factor of minerals. In order to perform an exergy analysis of mining and metallurgical processes, a set of objectives were defined. The first specific objective of this PhD thesis was to accomplish a detailed study of technologies and associated energy consumptions in the mining and metallurgical industry. A second objective was to analyze the influence of technological learning and decliniii ing ore grades on the availability of world non-fuel mineral resources, to become acquainted if technological breakthroughs that have occurred can preclude the rising energy demand in the mining industry. With the results obtained in the first and second activities, an important improvement in the exergoecological methodology was presented: exergy replacement costs which had been traditionally assessed in a static way, could be modified taking into account the long-term decline in ore grades. A further improvement carried out in this PhD was to solve the allocation problem among products, by-products and wastes commonly appearing in the mining and metallurgical industry. Accordingly, a new allocation procedure to be used in the Thermoecological analysis applied to mining and metallurgical processes has been proposed based on the newly obtained exergy replacement costs. The fifth objective of this PhD was to merge the Thermoeconomic analysis with the Thermoecological Cost methodology developed by the ITC group in the Silesian University of Technology, so as to combine the strength of both approaches for the analysis of the mineral’s industry. Finally, each of the objectives described above were applied to different case studies. iv Chapter 1 Overview of mineral resources and the mining industry The aim of this starting chapter is to provide an overview of mineral resources and the mining industry. Since this PhD is focused on the assessment of global mining, an overview of world mineral supply and market price evolution is performed. Additionally issues like international reporting, criticality of raw materials, sustainable development in the mining industry and life cycle assessment of abiotic resource depletion are reviewed. 1.1 Minerals as resources Resources are valued for their function in society, whether economic, cultural or physical. Society is highly dependent on minerals and metals. For instance, minerals are valued because they can generate wealth with which goods and services can be purchased. Additionally, they can be used to produce metals which are essential to maintain the standard of living of actual society e.g. transport, construction, infrastructure, health, leisure, defence, etc. According to Sohn (2006), the demand for minerals and metals is related with the society patterns and levels of consumption of goods and services produced in the economy and the evolving mix of raw materials used in each product or service. Equally, Giurco et al. (2010) believe that people’s daily lives is linked directly or indirectly to the availability of mineral resources, mineral production and consumption demand which in turn determine the value that society implicitly or explicitly confer to these resources. Figure 1.1 shows the minerals and metals used everywhere by actual society. Mining activities begin with exploration and evaluation of an area of interest. If the exploration and evaluation is successful, a mine can be developed, and commercial mining production can commence. Bradley (2007) claims that minerals are not simply collected, gathered or mined (e.g. extracted and separated) in the sense of obtaining ready-made supply. Minerals are created and produced, using highly complex, capital-intensive processes that transform matter into economic goods. In fact, the profitability will be affected by several economic, technical, social and geopolitical issues. Hence, a mineral deposit is a concentration of a mineral of sufficient size and grade that might, under the most favorable of circumstances, be considered to have economic potential. Once the mineral deposit has been explored and is known to be of sufficient size, grade, and 1 CONCRETELIMESTONE/CEMENT SAND/GRAVEL SAND/GRAVEL SCHIST QUARTZ GRANITE NATURAL STONE LARVIKITE Ice cream: TITANIUM DIOXIDE WEARING COURSE COVER? } ?WEIGHT FILLER CONCRETE SLEEPERS FROM LIMESTONE/ CEMENT SAND/GRAVEL SAND/GRAVEL AGGREGATE AGGREGATE AGGREGATE AGGREGATE FILTERLAYER BASE BASEMENT Minerals A/S CONCRETE LIMESTONE/CEMENT SAND/GRAVEL ALUMINIUM STEEL, TITANIUM BRICKS FROM CLAY AND OLIVINE A/S CONCRETE A/S CONCRETE LIGHTWEIGHT FILLER (ASPHALT, AGGREGATE OR GRAVEL) LECA Light bulb: QUARTZ? Car/train: IRON ALUMINIUM LEAD,COPPER ZINC MAGNESIUM SAND Glas: FELDSPAR NEPHELINE SYENITE QUARTZ Rubber: DOLOMITE LIMESTONE TALC GRAPHITE Paintwork: LIMESTONE TALC MICA TITANIUM DIOXIDE FROM ILMENITE RUTILE BASE COURSE BEDROCK TILL TILL SAND/GRAVEL OR LECA OR FINE AGGREGATE LECA SAND/GRAVEL AGGREGATE TIGHT WEIGHT FILLER IRON HEMATITE BEDROCK SAND LECA/CLAY ROCKWOOL ROOFING SLATE CONCRETE ROOFTILES Paper: LIMESTONE DOLOMITE TITANIUM TALC KAOLIN Paint: LIMESTONE TALC,KAOLIN TITANIUM FROM ILMENITE/RUTILE Pencil: GRAPHITE CLAY Tiles: Environmental liming Limestone: LIMESTONE DOLOMITE? PC: COPPER Floor: MARBLE Porcelain: LIMESTONE NEPHELINE SYENITE DOLOMITE Peer R. Neeb 2006 ANORTHOSITE FELDSPAR 6 Everyday use of mineral resources. 0 500 1 000 1 500 2 000 2 500 3 000 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Export value (mill. NOK-2008) Industrial minerals Metallic ore Natural stone Construction materials Coal Figure 1.1: Everyday’s uses of minerals and metals. NGU (2008). accessibility to be producible to yield a profit, it becomes an ore deposit. According to Cox & Singer (1992) a mineral deposit model is the systematically arranged information describing the essential attributes (properties) of a class of mineral deposits. The globalization of the mining industry makes imperative the developing of international standards for reporting mineral reserves, mineral resources and exploration results in order to underpin the mining industry and its capability to supply the worldwide increasing demand of mineral resources. The mining industry depends on financial investments, hence it is important to define a common terminology in order to enable investors to understand the risk involved in estimating mineral resources. Weatherstone (2008) accomplished an overview of the international standards for reporting of mineral resources and reserves, claiming that CRIRSCO (Committee for Mineral Reserves International Reporting Standards) was established in 1994 to assess mineral resources taking into account; mining, metallurgical, economic, marketing, legal, environmental, social and governmental issues. According to the CRIRSCO, the Joint Ore Reserves Committee (JORC) defines mineral resource as a concentration or occurrence of material of intrinsic economic interest in such form, quality and quantity that there are reasonable prospects for eventual economic extraction. Mineral resources are subdivided, in order of increasing geological confidence, into inferred,indicated and measured categories. An ore reserve is the economically minable part of a measured or indicated mineral resource. Ore reserves are subdivided in order to increasing confidence into probable ore reserves and proved ore reserves. The U. S. Geological Survey (USGS) use meanwhile the categories reserve and reserve base, which are similar to JORC’s ore reserves and mineral resources definitions, respectively. The mineral resource classification is shown in Fig. 1.2. Reserves are part of the reserve base which could be economically extracted or produced at the time of determination. Marginal Reserves is that part of the reserve base which, at the time of determination, which presents economic uncertainty. Subeconomic Resources is the part of identified resources that does not meet the 2 economic criteria of reserves and marginal reserves. Undiscovered Resources can be classified as hypothetical or speculative resources. Figure 1.2: Mineral Resource Classification (USGS, 1980). Additionally to land mineral deposits, the existence of deep-ocean minerals deposits has been known for more than a century. However, surveys devoted to understanding their genesis, distribution and resource potential began more recently, like those performed by Hein et al. (2013), Massari & Ruberti (2013). The minerals located in these marine deposits are crucial for an assortment of high-tech, green-tech and energy applications. For instance, the rare earth elements are used in plenty of new electronic and advanced components: such as fuel cells, mobile phones, displays, hi-capacity batteries, permanent magnets for wind power generation, green energy devices, etc. Although exploration contracts to develop deep-ocean mineral deposits have been signed in countries such as: China, France, Germany, India, Japan, Korea and Russia, technical challenges as well as environmental considerations need to be overcome. 1.2 Overview of world mineral supply This section presents a general review of production and prices of metals. In this regard, the distribution of global supply and demand of mineral resources has changed significantly in recent years. Humphreys (2013) believes that a new mercantilism is reshaping the world metal supply. The major countries consuming metals used to be also the major countries producing them, hence their interest to promote mine development to provide low cost raw materials. However, over the past fifty years the production of commodities of consuming countries has declined, and other countries like China have emerged over a very short time as the largest producer in the world, using their low cost capital and innovative metallurgical technologies. Now, producing countries are more interested on how to maximise the benefit of metal extraction to their economies rather than on how to supply cheap raw materials. Consequently, recent high metal 3 prices have led to increase the number of countries which are looking to metals production in order to promote their development. 0" 500" 1,000" 1,500" 2,000" 2,500" 3,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" World&produc*on,&metric&tons&[t]& Year& Au" Te" (a) Au and Te 0" 20,000" 40,000" 60,000" 80,000" 100,000" 120,000" 140,000" 160,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" World&produc*on,&metric&tons&[t]& Year& REE" Ag" (b) Ag and REE 0" 50,000" 100,000" 150,000" 200,000" 250,000" 300,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" World&produc*on,&metric&tons&[t]& Year& Cd" Co" Mo" (c) Cd, Co and Mo 0" 1,000" 2,000" 3,000" 4,000" 5,000" 6,000" 7,000" 8,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" World&produc*on,&metric&tons&[t]& Millares& Year& Cr" Pb" Ni" (d) Cr, Pb and Ni 0" 5,000" 10,000" 15,000" 20,000" 25,000" 30,000" 35,000" 40,000" 45,000" 50,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" World&produc*on,&metric&tons&[t]& Millares& Year& Al" Cu" Mn" Zn" (e) Al, Cu, Mn and Zn 0" 200" 400" 600" 800" 1,000" 1,200" 1,400" 1,600" 1943" 1953" 1963" 1973" 1983" 1993" 2003" World&produc*on,&metric&tons&[t]& Millones& Year& Fe" (f) Fe Figure 1.3: World production of metals. Data from U. S. Geological Survey (USGS). 4 According to Ghosh & Hem (1984), the commercial production of metals depends on factors such as: accessibility of ore deposits, richness of ore deposits, nature of extraction and refining process for the metal, physical and chemical properties of the metal as well as the demand for the metal. These factors depend on economics. A metal becomes a common one if it is readily available and easily produced with low processing cost and if it allows development of attractive properties. The world production of the metals analyzed in this thesis, is presented in Fig. 1.3. It can be seen that iron ores as steel is by far the most widely produced metal. This is due to the fact that iron ores are available in abundance in easily accessible deposits, the processing of iron ores is relatively simple and economical and because alloys of iron have a wide range of useful properties. The nonferrous metals which are produced in large quantities include metals such as aluminium, copper, manganese and zinc. In lesser amounts, metals such as lead, nickel, cadmium, cobalt and molybdenum, are produced. Finally, precious metals such as silver and gold, are produced in minor quantities as well as rare earth metals or less common metals such as tellurium. Nevertheless, the general trend of all metals is an abrupt increase (exponentiallike) in their worldwide production. 0" 500" 1,000" 1,500" 2,000" 2,500" 3,000" 3,500" 4,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" Unit%price,%[$/t]% Year% Al" Mn" Zn" Fe" Pb" Cr" (a) Fe,Mn,Pb,Al,Zn and Cr 0" 2,000" 4,000" 6,000" 8,000" 10,000" 12,000" 14,000" 16,000" 18,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" Unit%price,%[$/t]% % Year% Cd" Cu" (b) Cd and Cu 0" 10,000" 20,000" 30,000" 40,000" 50,000" 60,000" 70,000" 80,000" 1900" 1910" 1920" 1930" 1940" 1950" 1960" 1970" 1980" 1990" 2000" 2010" Unit%price,%[$/t]% % Year% Co" Mo" Ni" REE" (c) Ni,REE,Mo and Co (d) Te,Ag and Au Figure 1.4: Unit price of metals [$/t]. Data from U. S. Geological Survey (USGS). The high demand of metals and their consequent depletion has an immediate effect on mar5 ket prices. The unit price of internationally traded metals analyzed in this thesis, is presented in Fig. 1.4. It can be observed that iron ores as steel has the lowest price. The opposite case is that of gold, which is by far the most expensive of all metals analyzed in this thesis. In general Fig. 1.4 shows that metals price volatility increased in the 1980s, reaching record highs in recent times. This fact is driven by the strong demand from emerging economies, such as China and India. The magnitude of price fluctuations has increased drastically and in many cases caused commodity prices to multiply within only few years (e.g. lead or REE). According to Kriechbaumer et al. (2014), metal prices are the result of complex market dynamics and stochastic economic processes. Labys et al. (1999) state that metal price depend on macroeconomic variables such as industrial production, consumer prices, interest rates, stock prices, and exchange rates. Humphreys (2013) based on Goldman Sachs report The Revenge of the Old ’Political’ Economy 1published in 2008, argues that mineral prices increase because companies are making investments in the most accessible deposits instead of those with highest quality. The latter leads higher cost and lower efficiency. In the opinion of Humphreys, the world faces not so much a resource problem but an investment problem. 1.2.1 Critical raw materials One of the most important issues influencing the use of raw materials in the future is technological change. The high-tech metals (e.g. antimony, cobalt, lithium, tantalum, tungsten and molybdenum) are indispensable to new environmentally friendly products. For instance, electric cars require lithium and neodymium, car catalysis platinum, solar panels require indium, gallium, selenium and tellurium, energy efficient high-speed trains require cobalt and samarium, and new fuel-efficient aircraft needs rhenium alloys. Appendix Ashows in Table A.4 the main uses as well as the main driving emerging technologies for the several materials including those considered critical raw materials. A mineral is said to be critical depending on a number of factors such as such as geopolitical and depletion issues. Achzet & Helbig (2013) give and overview about differences and similarities of 15 criticality assessments for metallic raw materials performed by several working groups around the world from 2006 until 2011, concluding that there is a lack of consensus about which indicators give reliable information for raw material supply risk and how these indicators should be aggregated. The indicators identified by Achzet & Helbig (2013) for evaluation of supply risk include: country concentration, country risk, depletion time, by-product dependency, company concentration in mining corporations, demand growth, recycling/recycling potential, substitutability, import dependence, commodity prices, exploration degree, production costs in extraction, stock keeping, market balance, mine/refinery capacity, future market capacity, investment in mining, climate change vulnerability, temporary scarcity, risk of strategic use and abundance in earth’s crust. Other survey is the report performed by the ad-hoc Working Group (2010), which identify a list of critical raw materials for the European Union in regards to the economic importance and supply risk. The survey analyzes 41 metals, concluding that there are 14 critical raw materials for the EU: antimony, beryllium, cobalt, fluorspar, gallium, germanium, graphite, indium, magnesium, niobium, PGMs, REE, tantalum and tungsten, as depicted in Fig. 1.5. It is important to highlight that geological availability was not considered for determining criticality of raw mate1The Goldman Sachs Group, Inc. is a global investment banking, securities and investment management firm. http://www.goldmansachs.com 6 Figure 1.5: Critical raw materials for the European Union. The ad-hoc Working Group (2010). rials because global reserves are not considered as a reliable indicator of long term availability. Instead, according to the aforementioned study, geopolitical and economic changes are issues that impact greatly on the supply and demand of raw materials. It has to be highlighted that slight changes of the parameters of the supply risk metric of materials positioned in the sub-clusters of Fig. 1.5 may lead to a reclassification of these materials as “critical”. The supply risk includes issues such as: political and economic stability of the producing countries, the level of concentration of production, the potential to substitute and the recycling rate. The supply risk is attributed mainly to the fact that countries like China, Russia, the Democratic Republic of Congo and Brazil control the worldwide production, as Fig. 1.6 shows. This production concentration, generally is heightened by low sustainability and low recycling rates. According to Achzet & Helbig (2013) the aim of criticality assessment methods is to analyze driving factors, which makes a raw material critical from an economic, ecological, social or even ethical perspective. The criticality assessment methods constitute a starting point for a better understanding of raw material supply and demand. In regards to demand and supply of metals there is currently only intransparent and limited data available. It becomes therefore urgent to elaborate comprehensive databases of all raw material supply chain levels from the mining, to the manufacturing of products and the end of life phase with dissipation and recycling. 1.2.2 The role of recycling Lee (1998) defines resource depletion as a resource which has been consumed and discarded and can no longer be utilized by human beings. Lee considers that some of the factors that affect resource depletion are: • Reserves: The more reserve the resource has, the less tendency for depletion it will have. 7 Figure 1.6: Production concentration of critical raw materials for the European Union. The ad-hoc Working Group (2010). • Consumption rate: For those resources which have equal reserves, the greater the consumption rate, the more serious the depletion. • Natural replenishment rate: If the consumption rates of renewable resources are smaller than their replenishment rates, then these resources should have no depletion problems. For non-renewable resources, the natural replenishment rate is so low that it can be neglected. • Recycling rate: Resources can be recycled from discarded products and be re-manufactured into new products for human use. Hence, the recycling of a discarded resource can reduce its consumption rate, a high recycling rate can considerably diminish the resource depletion. • Resource substitution: If a particular resource can be easily and economically substituted by other resources, its depletion may be reduced. • Resource distribution: Most of the resources are not distributed equally over earth. Hence, local resource depletion problems depend on the local resource distribution characteristics. • Resource reliability: The reliability of imported resources that some countries need to import from other countries to support their own development influence the local depletion. Of the aforementioned factors, recycling plays a major role. In order to counteract the worldwide demand for primary mineral resources, it is imperative to recycle materials more widely and more effectively. Consequently, once the scarcity of “virgin” material becomes acute and should commodity prices rise, the recovery of secondary resources from technospheric 8 stocks, popularly referred to as urban mining (Brunner & Rechberger 2004) may become a realistic option. Indeed, this possibility is gaining increasing attention and has been addressed as a suitable and necessary alternative by the UNEP International Resource Panel (Graedel et al. 2010) or the Swedish Environmental Protection Agency (SEPA 2012). Accordingly, in the future, metals will be supplied from a combination of primary metal produced from newly mined ores and recycled metals. Brobech (1996) and Bravard et al. (1972) claim that recycling must play an important role in the life cycle of metals production and should involve issues like education, research, proposals and targets. Furthermore, recycling operations requires technological, economic and environmental proficiency. Most metals are recyclable, either easily (e.g. aluminium) or with deliberate programs (e.g. lead, platinum). According to Giurco et al. (2010), the overall stock of recyclable materials is rising, but the availability of various metals for recycling differs widely. In this regard, legislation and product management system like the end of life vehicle legislation often lead to improved recycling rates. Efficient recycling of products as well as all kinds of production residues at various points in the life cycle of a product, reduces significantly the demand of raw materials. Furthermore, in several situations recycling leads to energy savings and reduction of climate change impacts. For instance, to obtain steel or aluminium from scrap, already requires less energy than from primary raw material. However, depending on how concentrated or dispersed the metal is and the transport required, recycling may not always have lower energy requirements. The higher the import dependence on an individual metal, the more important recycling becomes, particularly if substitution possibilities are limited. The ad-hoc Working Group (2010) listed a set of actions to improve the efficient recycling of raw materials. The latter includes the design of products using materials that can be recycled, selecting materials that contain a high percentage of recycled content, reducing the number of different materials within an assembly, marking parts for simple material identification, using compatible materials within an assembly, selecting materials that do not need to be separated for recycling, making products easy to disassemble, identifying discarded products with critical raw material for proper collection instead of stockpiling them in households or discarding them into landfill or incineration, improving overall organisation, logistics and efficiency of recycling chains, preventing illegal exports of discarded products containing critical raw materials and increasing transparency in flow and promoting research on system optimisation and recycling of products and substances. That said, according to Wellmer & Becker-Platen (2002), a recycling rate of 100% is impossible to achieve because depending on the application, a very pure metal could be required. Reaching the quality of primary raw materials, would entail higher energy requirements and environmental impacts. Heighten recycling together with material efficiency improvements will play important roles, but for the foreseeable future it is likely that new primary raw materials will continue to be required. Moreover, access to and extraction of primary raw materials will always be needed, due to market growth or new applications. Even if there is a total recycling the breach between the time of product manufacturing and product end of life phase needs to be overcome. Finally, recycling is determined by economic issues such as the ratio of the prices for recycled and primary raw material. 9 1.2.3 Environmental and social implications in the mining industry As aforementioned, recycling still constitutes (with the exception of a few metals) a minor practice. Consequently, traditional mining will constitute in the short to medium term the main mineral supplier with not insignificant consequences to the environment. Indeed, mining is one of the activities with the greatest environmental impact from a cradle to grave perspective. It is a fact that the worldwide increase in production, use and disposal of minerals and metals has caused harmful environmental impacts, from global warming to local pollution affecting land, air and water. Ayres (2008) states that minerals extracted from the earth and utilized for economic purposes are not literally “consumed” because they become waste residuals that do not disappear and may cause environmental damage. Giurco et al. (2010) claim that these impacts will likely become unsustainable in the medium to long-term. The climate change has brought green house gas emission constraints, which are an effort to internalise major global environmental costs. In the mining industry, fossil fuels are the main source of energy. Accordingly, mechanisms like carbon trading, is likely to cause increased cost for mining production. Other environmental cost at local level include mine site closure and storing of waste rock. In the US and Europe, stricter environmental regulations and greater difficulty obtaining mining permits has made open cut mining economically unattractive. Toxic emissions from mining are harmful to communities and the natural environment. Besides, mine wastes such as the tailings and waste rock remaining after completion of a project, is a challenge for the mining industry due to given declining ore grades and deeper mines. Technology plays a key role in addressing the environmental impacts. For instance, implementing scrubber systems to capture sulfur dioxide emissions from smelters and converting this to sulphuric acid, implementing tall stacks from smelters to ensure adequate dispersion of atmospheric pollutants, methane gas extraction systems prior to coal mining and sulphidic mine waste and acid and metalliferous drainage. However, Giurco et al. (2010) assert that new technologies not always led to lower environmental costs. Hence, it is a challenge for future technologies. A way to counteract environmental but also social problems associated with the mining industry is with sustainable development practices. Indeed, several studies like those performed by Cleveland & Ruth (1997), Shields (1998), Wellmer & Becker-Platen (2002), suggest that sustainable development is currently one of the most complex and challenging issues of the mining industry, because it involves the preservation, rational use and enhancement of natural resources. As is well known, the World Commission on Environment and Development WCED (1987) performed the report “Our Common Future”, where long-term environmental strategies for achieving sustainable development were proposed. In this report, the Commission highlighted that “Humanity has the ability to make development sustainable to ensure that it meets the needs of the present without compromising the ability of future generations to meet their own needs”. The latter is a widespread definition of sustainable development. Although a more specific mining sustainable development definition could be: “sustainability means the design, construction, operation and closure of mines in a manner that respects and responds to the social, environmental and economic needs of the present generations and anticipates those of future generations in the communities and countries where it works” according with gold mining company Placer Dome (now Barrick Company). Nappi & Poulin (1998) assert that the major sustainability issue for metals is the influence on environmental and social conditions. Similarly, Yellishetty et al. (2011) claim that social, environmental and economic objectives of 10 METHODS, TOOLS, AND SOFTWARE Figure 1 Life cycle assessment from grave to grave. Manuf. =manufacturing; conc. =concentration. a matter of precision. For many unrefined analyses we may use both the energy and embodied energy concepts as substitutes for exergy and exergy cost concepts, respectively, but in fact, in no way are they synonymous. Both exergy and exergy cost require precise definitions. As is well known, LCA results are relative to the chosen system’s boundary. Nowadays, no absolute LCA values exist for a given good or service. Notwithstanding that, suppose we start our analysis from a hypothetical cradle in which all the commercial minerals and fossil fuels have been depleted (i.e., Thanatia). This degraded planet serves us both as a boundary limit and as an absolute reference system good enough for calculating the exergy and exergy costs of any commodity at the industry gate. Theoretically speaking, this is the only way to get absolute LCA values, by converging LCA with second law analysis through Thanatia as a reference environment. In the future Thanatia may become the starting point for the assessment of abiotic resource depletion. The exergy measured from Thanatia gives a measure of the quality of the resource and constitutes a universal, objective, and useful tool for classifying resources according to their depletion states. Presented over time, exergy can give an indication of the speed at which degradation is occurring. However, we must state that even considering a consequent baseline, exergy is still insufficient for realistically quantifying resource depletion and we should additionally resort to exergy costs. Hence, from the exergoecological point of view, we propose to introduce a new stage in the LCA’s cradle-to-grave methodology, namely the grave-to-cradle approach, as depicted in figure 1. In our view, there is a lack of theory rather than a lack of indicators. Partial or total cradle-to-grave assessments are only half of the cycle. We call them “over the rainbow” accounting methodologies. They lack the other side: the graveto-cradle assessment. In the same way that imaginary numbers can hardly be explained in real space, some phenomena like depletion may be better explained in the “down the rainbow” approach. It is important to close the whole material’s cycle, as stated by McDonough and Braungart (2002) in their book Cradle to Cradle. Once extracted, concentrated, and refined (in the cradle-to-gate stage), minerals are converted into useful products (entry gate to exit gate). Some of these materials are recycled when their useful life ends and go directly into the entry gate–to–exit gate stage. However, most of them end up as waste—either as pollution or disposed of in landfills—and become degraded and dispersed, arriving at a final depleted state (to the grave)—that is, to Thanatia. In this way, at least four main costs in the life cycle of a product come into play (see figure 1). The first one belongs to the “down the rainbow” part. It is in fact an avoided and imaginary cost, and represents the debt we acquire with future generations as the effort that nature spent in producing minerals in a concentrated state. The remaining costs belong to the “over the rainbow” part and are associated with real consumption. The second one is the cost associated with mining, mineral processing, smelting, and refining. Next, the manufacturing costs result when the already processed raw materials are converted into useful products. And finally, recycling costs will eventually appear if scrap and wastes are recycled. The distances from the horizontal axis of each one of the reservoirs in figure 1 are variable for each material under analysis. The same thing happens with the relative distance between 48 Journal of Industrial Ecology Figure 1.7: Life cycle assessment from grave to grave. Valero & Valero (2012). 1.5 Summary Mineral resources are finite, therefore they will eventually be depleted by ongoing extraction. In this chapter, a general outlook of the mineral situation (world production and price of metals) in the actual society as well as the factors which influence the availability of minerals resources and the environmental consequences of mineral extraction were analyzed. Undoubtedly, sustainable development is one of the most complex and challenging issues because it involves the preservation, rational use and enhancement of natural resources. At the same time social, environmental and economic objectives of sustainability must be fulfilled over the long-term. Resource depletion depends on several factors such as: availability of energy sources, water limitations, climate change, legal restrictions, environmental protection, social disruption, international trade, recycling and reuse, substitution, growth in demand and technology change. In this respect, the life cycle assessment is a promising tool in the assessment of mining and processing minerals. The chapter ended with a critical analysis of current LCA approaches for the evaluation of abiotic resource depletion. A new methodology for addressing a good number for open issues in conventional LCA was proposed, based on the exergoecological approach. Such methodology involves the inclusion of a new stage in the LCA methodology, namely the grave to cradle stage, thereby accounting for the depletion degree of minerals. Consequently, an absolute LCA can be performed: from the cradle to the grave and back again to the cradle. In the next chapter, the first stage of LCA: cradle to entry gate (raw material extraction and refining) is analyzed, as the first step for an absolute life cycle assessment. 17 Chapter 2 Metal Resources and Energy Mining industry provides essential raw materials like coal, metals, minerals, sand, and gravel to the manufacturing and construction industries. There is a wide variety of metallurgical processes in mining industry, which depends on the metal extracted, the ore grade, mining conditions, available technology, economic issues, etcetera. The aim of this chapter is to describe the main physical processes applied in mining and metallurgical industry, in order to describe the metal processing and energy consumption of each commodity analyzed and applied in different case studies in this thesis. The analysis of these processes constitutes the first stage of LCA: cradle to entry gate (raw material extraction and refining). In this chapter, an introduction of the main physical processes applied in mineral processing and extractive metallurgy is presented. The information presented in this section was taken from several sources: Ghosh & Hem (1984), Gupta & Mukherjee (1990), Rosenqvist (1983), Wils (2006). Afterwards, the metal processing and energy requirements of selected metals analyzed in this thesis are given. Energy requirements data for metal processing varies greatly, and increase substantially due to different factors such as: lower ore grades, deeper mines, complex ores and more mine wastes. 2.1 Mineral processing and extractive metallurgy Production of metals from natural ores is mainly a separation process. The separation processes can be classified into two stages.Concentration which is the separation of the compound containing the desired metal from other constituents. This stage includes the mining and beneficiation processes.Extraction and Refining which is the separation of the desired metal from other constituents of the metallic compound and further purification of the metal. There are two basic mining techniques: underground and open-pit mining. The method selected depends on a variety of factors, including the nature and location of the deposit, and the size, depth and grade of the deposit. The most common method is the open-pit mining, due to the lower energy consumption. Underground mining requires more energy than surface mining due to greater requirements for hauling, ventilation, water pumping, and other operations. The mining process can be divided into two general stages, each involving several operations. The first stage is extraction, which includes activities such as blasting and drilling in order to remove material from the mine. The second stage is materials handling, which involves the transportation of ore and waste away from the mine to the mill or disposal area. A 19 brief description of the operations used in mining follows. Drilling is the act or process of making a cylindrical hole with a tool for the purpose of exploration, blasting preparation, or tunneling. Blasting uses explosives to aid in the extraction or removal of mined material by fracturing rock and ore by the energy released during the blast. Digging is to excavate, make a passage into or through, or remove by taking away material from the earth. The goal of digging is to extract as much valuable material as possible and reduce the amount of unwanted materials. Digging equipment includes hydraulic shovels, cable shovels, continuous mining machines, longwall mining machines, and drag lines. Ventilation is the process of bringing fresh air to the underground mine workings while removing stale and/or contaminated air from the mine and also for cooling work areas in deep underground mines. Dewatering is the process of pumping water from the mine workings. Mineral processing (or Beneficiation) follows mining and consists of several operations required to prepare and classify ores before the valuable constituents can be separated or concentrated for its further use or treatment. In order to separate the minerals from gangue (the waste minerals), it is necessary to crush and grind the rock to liberate valuable minerals. This process of size reduction is called comminution. Apart from regulating the size of the ore, it is a process of physically separating the grains of valuable minerals from the gangue minerals, to produce an enriched portion, or concentrate, containing most of the valuable minerals, and a discard, or tailing, containing predominantly the gangue material. There are two fundamental operations in mineral processing: liberation (the release of the valuable minerals from their waste gangue minerals, by means of comminution) and concentration (separation of these values from gangue). According to Wils (2006), grinding is the greatest energy consumer, accounting for up to 50% of a concentrator’s energy consumption. The main physical methods which are used to concentrate ores are: •Sorting is the separation based on optical and other properties. •Gravity concentration is the separation based on differences in density. •Froth flotation is the separation utilising the different surface properties of the minerals. •Magnetic separation is the separation dependent on magnetic properties. •High-tension separation is the separation dependent on electrical conductivity properties. Extractive metallurgy deals with extraction of metals from their ores. Mining and extraction processes consists of some individual sequential steps. For instance, electrolysis is applied to aluminium, zinc, copper and many other metals. Smelting is performed for extraction of iron or lead. The steps can be physical operations like comminution, filtration, casting, distillation, etc., or chemical operations such as leaching, smelting, etc. The extractive metallurgy classified the methods of extraction and refining into:pyrometallurgy which is performed at high temperatures, hydrometallurgy which is carried out in aqueous media at or around environment temperature and electrometallurgy which employs electrolysis for separation at high or environment temperature. 20 2.1.1 Pyrometallurgy Pyrometallurgical methods of metal production are usually cheaper and suited for large scale productions. There are several processes used in pyrometallurgy such as: Calcination is the thermal treatment of an ore to provoke its decomposition as well as the elimination of volatile products (generally carbon dioxide and/or water). Roasting usually involves heating of ores below the fusion point in excess of air. It produces a chemical conversion and makes the raw materials more suitable for subsequent reduction. Some roasting operations includes: oxidizing, volatilizing and chloridizing. Oxidizing roasting: this operation is performed to burn sulphur from sulphides with conversion of sulphides in whole or in part into oxides. For example PbS(s)+3 2O2(g)=PbO(s)+SO2(g) (2.1) Under certain conditions, oxidizing roasting may involve other reactions, leading to formation of sulphates or even the release of the metal itself in elemental form. Volatilizing roasting: this is carried out to eliminate volatile oxides such as As2O3,Sb2O3 and ZnO. Chloridizing roasting: this is done to convert certain metal compounds to chlorides from which the metal may be subsequently obtained by reduction. The reaction is of the type 2NaCl(s)+PbS(s)+2O2(g)=Na2SO4(s)+PbCl2(l) (2.2) Smelting is a process for the reduction of a metal oxide to metal, carbon is by far the most common reducing agent because of its easy availability and low cost. 2.1.2 Hydrometallurgy Hydrometallurgical processes have the objective of isolate and win metals by the use of water or aqueous solutions. It involves a large sort of processes such as cyanidation for treating gold and silver ores, and the Bayer process for the production of alumina. There are several operations used in hydrometallurgy such as: Leaching is a process used to recover metals from ores by dissolving the metal into a solution in contact with the crushed ore. There are different types of leaching, such as: high temperature acid or alkaline leaching, cyanide leaching of gold and bioleaching of sulfide minerals. Precipitation of Impurities is a chemical process in which a dissolved substance separates from solution as a fine suspension of solid particles. Also called crystallisation when the solids formed are crystals. Solvent Extraction (SX) is a process to separate dissolved metals (e.g. uranium, copper, nickel, cobalt, etc.) from impurities or between them. A solution of a metal leached from an ore is mixed with an organic solvent containing specific extractant chemicals. The aqueous and organic phases form an emulsion in which the extractant chemicals in the organic solvent bind to the metal ions and pull the metal into the organic phase. The mixture is left to settle so that the organic and aqueous phases separate. The organic phase containing the concentrated, 21 purified metal is removed. The metal ions are then transferred from the solvent back into an aqueous solution from which they can be recovered in high purity. Adsorption and Ion Exchange is the accumulation or concentration of a substance on a surface. In physical adsorption, molecules are attracted to a surface by intermolecular forces of attraction. In chemical adsorption, molecules, atoms or ions are attached to a surface by chemical bonds. 2.1.3 Electrometallurgy Electrometallurgy is the application of electrolysis to the winning and refining of metals. Electrometallurgy is usually the last stage inmetal production after corresponding pyro or hydrometallurgical operations. As stated by Rosenqvist (1983), electrowinning is important for the very reactive light metals aluminum and magnesium, which are almost always produced by electrolysis of fused salts. Furthermore, electrorefining is fundamental for the recovery of valuable impurities, such as silver and gold from copper. Electrowinning recovers a metal (such as copper, zinc or nickel) from a solution containing the metal ions. An electrowinning production cell is an electrochemical cell with an applied electric current in which metal ions deposit the metal on the cathode. Electrorefining is the electrolytic purification of the crude metal produced by pyrometallurgical operations. The basic metal is dissolved from a positive electrode and further, it is reduced on the cathode where the final product is highly pure metal. 2.2 Metal Processing and Energy Requirements In this section, a brief description of the main uses, physical, chemical and geological characteristics, producing countries and the extractive and production processes as well as the energy requirements of 17 metals used throughout the development of this thesis are presented. Additional information about uses of these and other metal are presented in Appendix A. A summarize table of the energy consumption of each metal is presented, the Emining represents the empirical energy data for mining and concentration processes, whilst the Erefining accounts the additional energy required to obtain the metal in an average refining grade through the smelting and refining processes. 2.2.1 Aluminum Aluminium is an important commodity for modern manufacturing, used in construction and automobile production. It is a lightweight, high-strength, corrosion-resistant metal with high electrical and thermal conductivity, and it is easy to recycle. Primary aluminum is produced globally by mining bauxite ore, refining the ore to alumina, and finally, smelting alumina to produce aluminum. More than 95% of the bauxite today is gained from open mines. Secondary aluminum is produced by sorting, melting and treating recycled aluminum scrap (Classen et al. 2007). The major producers are Australia, Guinea, Jamaica and Brazil. The total energy for the production of alumina from bauxite ore in the ground includes the following main processes: mining, crushing, griding, clarification, filtering and calcination. 22 A simplified overview of the processes is given in Fig. 2.1, followed by a brief description of the production processes. In chapter 6, section 6.3.2 additional information about aluminium production is presented. &Q>&7#+'*.-6+"*#& & 1 Introduction This part deals with primary and secondary aluminium production. Because bauxite, aluminium hydrate and aluminium oxide are mainly consumed in the aluminium production, these chemicals are also included in this report. A considerable part of the aluminium consumed today is secondary material. Thus also the collection of aluminium scrap and the recycling process are included. A simplified overview of the processes is given in Fig. 1.1. mine, bauxite recultivation, bauxite mine aluminium hydroxide, p lant aluminium oxide, plant aluminium hydroxide, at p lant aluminium oxide, at p lant aluminium electrolysis, p lant cathode, aluminium electrol y sis anode, plant anode, aluminium electrol y sis aluminium, primary, at p lant aluminium scrap, new, at p lant aluminium scrap, old, at p lant scrap preparation p lant aluminium, production mix, cast alloy, at plant aluminium, production mix, wrought alloy, at plant aluminium, production mix, at plant bauxite, at mine aluminium, primary, liquid, at plant aluminium, secondary, from new scrap, at plant aluminium melting furnace aluminium casting, plant aluminium, secondary, from old scrap, at plant Fig. 1.1 Simplified overview of the processes for the aluminium production ecoinvent v2.0 report No. 10&E&Q&E& Figure 2.1: Simplified overview of the processes for the aluminium production. Classen et al. (2007). According to Classen et al. (2007), the ore is won in stripes using hydraulic excavators and / or belt loaders. Drilling and blasting is only needed in some mines with exceptionally hard bauxite. The bauxite is transported to the mobile grinder and possibly to a dryer from where it is shipped to the aluminium oxide extraction plant. 98% of the world wide production of aluminium hydroxide and aluminium oxide is based on the Bayer-Process, whilst 98% of the industrial production of aluminium is done by electrolysis of aluminium oxide (HallHeroult process). This process requires a big amount of electricity. As stated by Boustead & Hancock (1979), the concentrating energy requirement in USA summarizes 30.52 GJ/t, additionally it has a high energy use about 283.11 GJ/t for the refining process. The aluminum industry produces ingots of pure aluminum (greater than 99%). Whilst, Yoshiki & Toguri (1993) assert that if mining and ore preparation (crushing, washing or wet 23 screening, and drying) are considered as the concentrating process, the energy consumption is 36.63 GJ/t while 127.38 GJ/t is the energy demanded in the refining process (smelting). Taking into account that for the production of 1 tonne of alumina, which produces about 0.53 tonnes of aluminum, is required approximately 2 tonnes of bauxite, the energy required in this concentrating process is 37.37 GJ/t whilst the refining process requires 53 GJ/t for the best operated anode cells to 61 GJ/t for some traditional cells, on the report of IPPC (2009a). When aluminum is produced from bauxite, Classen et al. (2007) state that the energy required for both concentrating and refining processes is 78.7 GJ/t. If the production of alumina from bauxite ore is done through the Bayer process, Chapman & Roberts (1983) affirm that the energy requirement is about 50 GJ/t of aluminum correspond to the energy consumption for the concentrating process and 228 GJ/t are used in the concentrating process to produce aluminum by the electrolysis of alumina in the Hall cell (Hall-Heroult process is a very electricityintensive). Ayres et al. (2002) state that the energy consumption per tonne of aluminum in the refining process through 1997 was 53.43 GJ/t, in 2002 it was 48.922 GJ/t and during 2005 it was reported as 40.075 GJ/t. These data shows a decrease which is partly due to the closure of some older smelters, according to EAA (2006). Either the amount of 175 GJ/t for this process was reported. The most efficient smelters operate with an energy consumption of about 46.8 GJ/t. Margolis (1997) argues that the average energy consumption of aluminum reduction will continue to decrease as old and obsolete smelters are shut down, existing smelters are retrofitted and modernized, and new cells lines with moderns technology are built. This technology renewal is slow, especially due to the high investment costs of new capacity. Otherwise, is not expected substantial energy savings from the Hall-Heroult electrolytic process. The most accurately data for the aluminum production process involves an energy requirement for the concentration process of gibbsite from the ground as 37.77 GJ/t, reported in IPPC (2009a), while the energy consumption in the refining process of gibbsite to obtain pure aluminum is 55.692 GJ/t. Table 2.1 summarize the energy consumption in mining and refining of aluminium, according to the aforementioned process used to obtain it. Table 2.1: Energy consumption in the aluminium production. Values expressed in GJ/ton of Al. Emining Erefining Etotal Boustead & Hancock (1979) 30.52 283.11 313.63 Chapman & Roberts (1983) 50 228 278 Yoshiki & Toguri (1993) 36.63 127.38 164.01 EAA (2006) 40.07–53.43 Classen et al. (2007) 78.7 IPPC (2009a) 37.37 53 90.37 On the report of DOE (2007), it is asserted that the aluminum industry has large opportunities to further reduce its energy intensity, because is constantly evaluating, adopting, and improving furnace technologies and practices. This provides not only energy and environmental benefits, but also cost savings. According to Margolis (1997), the average energy consumption of aluminium reduction will continue to decrease as old and obsolete smelters are shut down, existing smelters are retrofitted and modernized, and new cells lines with moderns technology are built. This technology renewal is slow, especially due to the high investment costs of new capacity. Otherwise, is not expected substantial energy savings from the Hall-Heroult electrolytic process. 24 2.2.2 Cadmium Cadmium applications has shifted away from the market areas of pigments, stabilisers and coatings to Ni-Cd batteries which have extensive applications in the railroad and aircraft industry, cellular telephones and portable computers (Classen et al. 2007). Cadmium is a very scarce element (0.1 – 0.2 ppm in the earth’s crust) (Ayres & Ayres 1996). It is a by-product of zinc extraction. Nowadays, approximately 80% of world cadmium production was derived from mining, smelting, and refining of zinc, and the remaining 20% came from copper and lead smelting and the recycling of cadmium products USGS (2011a). According to Ayres & Ayres (1996), Japan was the biggest producer of refined metal, followed by Russia, Belgium and USA. The most common production process for cadmium production is the electrolysis from processed cadmium sludge from hydrometallurgical zinc operations. This operation includes three stages; precipitation, oxidation and electrolysis. If cadmium is removed with copper by reduction with zinc dust during the hydrometallurgical zinc refining, the metallic sludge constitutes the most important source for cadmium refining process, Classen et al. (2007) report an energy consumption of 4.5 GJ/t . Otherwise, assuming the concentrating process as a chemical leaching, the energy consumption will be 103.5 GJ/t, according to Kihlstedt (1975). While, Botero (2000) reports the refining energy requirement as 6.48 GJ/t. Table 2.2 summarize the energy consumption in mining and refining of cadmium, according to the aforementioned process used to obtain it. Table 2.2: Energy consumption in the cadmium production. Values expressed in GJ/ton of Cd. Emining Erefining Etotal Kihlstedt (1975) 103.5 Botero (2000) 6.48 Classen et al. (2007) 4.5 2.2.3 Chromium Chromium is one of the modern industry’s essential element and important raw material for the production of stainless steel, and this is the major form in which chromium is recycled. It has a wide range of uses in metals, chemicals, and refractories. Chromium is obtained from chromite ore. Chromite enhances thermal shock and slag resistance, volume stability, strength and is used mainly in the ferrous and non-ferrous industry, cement industry and glass manufacturing. Ayres & Ayres (1996) assert that chromium (along with cobalt) is the archetypical ‘strategic metal’ because of its importance in corrosion and heat resistant alloys, especially in the so-called ‘superalloys’ used in aerospace technology. The major chromite ore and concentrates producing countries are South Africa, India and Kazakhstan, representing 70% of 2008 world production as a whole. South Africa and Zimbabwe hold about 90% of the world’s chromite reserves and resources, according to Murthy et al. (2011). Chromium metal is a metallurgical industry product. It is produced by one of two processes: electrolysis or aluminothermic reduction (Classen et al. 2007), as shown in Fig. 2.2. The metallurgical route involves smelting in an electric furnace to yield ferro-chromium. Chromium is considered to be made by alumiothermic process (75%) and electrolysis (25%). 25 &a>&/51+(R&4F$'$6+('"1$+"*#& 5 System Characterisation This study focuses on the main processes in the chromium production as shown in Fig. 4.1. As shown in chapter 4 Use / Application of Material no secondary production of chromium has to be considered, since the recycling of chromium occurs directly into the foundries in the form of re-melted alloys from chromium containing scrap. %*0L6*$% =\Tm???&+ 75% H(+$%%-'O"6$%&@%$#+ ^mT\=mb??&+ high-coal =m]?Tm???&+ am???&+ (%(6+'*%5+"6& )'*.-6+"*# Tam???&+ $%-R"#*L +F('R"6& '(.-6+"*# 17% Chromite bm=a[m???&+ 4F(R"6$%&@%$#+ C^^m???&+ C^^m???&+ 8% P(,'$6+*'5&7#.-1+'5 ^b?m???&+ '(,'$6+*'"(1 ^b?m???&+ 4F'*R"-R&4F(R"6$%1 =?m???&+ !(''*L 6F'*R"-R Chromium Metal Fig. 5.1 Scheme of the chromium production system considered. The study focuses on the production of highcarbon ferrochromium, which is based on chromite. Chromite itself is used in different other modules in ecoinvent like “portachrom, at plant” Hischier (2003), “sodium dichromate, at plant” Althaus et al. (2003) and in the cement production. Metallic chromium is won from two different processes. Values shown are in tonnes of chromium contained and are based on USGS (2003) and Adelhardt & Antrekowitsch (1998). ecoinvent v2.0 report No. 10&L&T=&L& Figure 2.2: Scheme of the main processes in the chromium production. Classen et al. (2007). According to Boustead & Hancock (1979), the energy requirement for producing chromium was 0.33 GJ/t. Chapman & Roberts (1983) report the energy requirement for mining and concentrating ore to produce chromium as 3.11 GJ/t, whilst the smelting and refining processes is 129.84 GJ/t . Classen et al. (2007) state that the energy needed for the production of metallic chromium is 0.29 GJ/t during the concentrating by aluminothermic process and 0.025 GJ/t for the refining step. The consumption of energy for the production of chrome is 0.5 GJ/t, on the report of IPPC (2009a). Table 2.3 summarize the energy consumption in mining and refining of chromium, according to the aforementioned process used to obtain it. Table 2.3: Energy consumption in the chromium production. Values expressed in GJ/ton of Cr. Emining Erefining Etotal Boustead & Hancock (1979) 0.33 Chapman & Roberts (1983) 3.11 Classen et al. (2007) 0.29 0.025 IPPC (2009a) 0.5 2.2.4 Cobalt Cobalt is a crucial element in many technological applications. The largest end-use of cobalt is for high-temperature resistant ‘superalloys’ in jet engines. Cobalt is mainly produced as a by-product and recovery of other abundant metals such as copper, platinum or nickel. Ayres & Ayres (1996) states the cobalt is undoubtedly a ‘strategic’ mineral, given the small number of producers and the importance of its uses. According to Harper et al. (2012), the three principal 26 deposit usually is in the range of 100 meters. Iron ore is mined mainly in China, Brazil and in Australia. Nearly all iron ore is reduced to iron in blast furnaces, it is usually known as pig iron or “hot metal” if is liquid. In general, blast furnace burden contains lump ore, sinter and pellets, therefore the gross energy consumption include the energy used in the sinter, pellet and coke oven plant plus the energy required in blast furnace, oxygen furnace or electric arc furnaces. Figure 2.6 displays a rough overview of the cradle to gate process of the cast iron and steel production. 5. System Characterisation Ecoinvent v2.0 report No. 10 - 7 - 5 System Characterisation 5.1 General overview A rough overview of the cradle to gate processes of the cast iron and steel production are given in Fig. 5.1. Casting Transport to steel work Mining of iron ore Transport to iron work Sorting Lump ore Benefication Pellets Sinter Pellet feed Sinter feed Transport to iron work Direct reduction Blast furnace reduction Other meltreduction Sponge iron Hot metal/ Pig iron Pig iron Electric steel furnace Basic oxygenfurnace Transport to steel work SteelCast iron Iron scrap Fig. 5.1 Overview of cast iron and steel production (The grey materials and processes are of minor relevance and not considered in the ecoinvent data) 5.2 Iron ore mining and beneficiation This part is mainly based on (Roth et al. (1999)) and (Meiler (1997)). 5.2.1 Production process and infrastructure Most of the iron ore today is gained from open mines. The thickness of the recoverable ore deposit usually is in the range of 100 meters. The cap of the deposit is removed as overburden and deposited nearby to be used for the recultivation when the mine has been depleted. The ore is won in stripes usFigure 2.6: Overview of cast iron and steel production Classen et al. (2007). The process of concentrating iron ore from ore in the ground requires 1.11 GJ/t and the refining process has an average consumption of 24.67 GJ/t based on the fuel energy to operate a blast furnace, according to Boustead & Hancock (1979). Whilst, Botero (2000) reports the total energy requirements for the concentrating process as 1 GJ/t and a high energy use around 28 GJ/t is required for the refining in the blast furnace, considering a concentration upper than 92 %.Classen et al. (2007) state that the energy utilized in the overall mining process is 28.8 GJ/t including electricity and fuels utilized in the sinter, coke and pellet plants as well as in the blast 33 furnace. Sohn (2006) states that technological change in steel making has produced 20-25% increase in efficiency. The energy accounting for all the process is 20 GJ/t, as reported on IPPC (2009b). Table 2.6 summarize the energy consumption in mining and refining of iron. Table 2.6: Energy consumption in the iron production. Values expressed in GJ/ton of Fe. Emining Erefining Etotal Boustead & Hancock (1979) 1.11 24.67 25.78 Classen et al. (2007) 28.8 IPPC (2009b) 20 2.2.8 Lead The main applications of lead are batteries and pigments. The main lead mineral is galena, although it occurs frequently with zinc, copper, arsenic, tin, antimony, silver, gold and bismuth. Mixed lead-zinc ore deposits are the most important. Argentiferous lead and lead-zinc ores, which frequently appear together, are found in almost all silver-producing countries. According to Classen et al. (2007), per tonne of lead mined, 2.3 kg of silver is jointly extracted and refined. United States, Australia, Canada, China and India are the nations with the biggest share of lead mine production. Lead bullion may contain varying amounts of copper, silver, bismuth, antimony, arsenic and tin. There are two methods of refining crude lead: electrolytic refining and pyrometallurgical refining, as depicted in Fig. 2.7. Kihlstedt (1975) reports that in the production oflead from a low-grade ore mined on underground mine, the energy consumptions for concentrating and refining processes were 4.95 GJ/t y 7.56 GJ/t respectively, the concentrating process involves mining, comminution, flotation, tailings disposal, filtration and drying. According to Boustead & Hancock (1979), the production of lead from ore in the ground requires during the concentrating process 11.34 GJ/t. The sinter production and blast furnace operation for the refining process consumes 17.31 GJ/t. Whilst Chapman & Roberts (1983) report the energy requirement for mining and concentrating ore to produce lead is 9.5 GJ/t, whilst the smelting and refining processes require 18.9 GJ/t . There are two basic pyrometallurgical processes available for the production of lead from sulphide concentrates: sinter oxidation/blast furnace reduction route or direct smelting reduction process. Modelling the smelting process as a combination of both processes Classen et al. (2007) state that the energy consumption for this stageis 2.535 GJ/t, 44 % through the older inter oxidation/blast furnace processes and 56 % by the new direct smelting process (QSL, Kivcet). The energy requirement for the sintering/smelting process using a special design blast furnace named the Imperial Smelting Furnace (ISF) is 1.4 GJ/t, whilst there are several processes for direct smelting such as Smelt/Ausmelt furnaces (ISA), Kaldo (TBRC) and QSL integrated process have an average consumption of 1.35 GJ/t, according to data reported on IPPC (2009a). Table 2.1 summarize the energy consumption in mining and refining of lead, according to the aforementioned process used to obtain it. 34 0='L#$&)T ?+")@$&&+-9)A$#',9)!"(-9#&.$9 KOK $V#..#)8D% 9% 0#$.-/"53'1+% 133,6% #$% &'*,2'&'+% 1)+% -)+'&8,'$% (-&/"'&% &'(#)#)8% /,% E&,+-*' 0#$.-/"D aBL;(R9AA48: A*+!-4>!I4> >8:**!H)++/) L:7+)4.4@ ,875B:B; ABL; 9)*./,)8.4@ A 9)*./,)8.4@ I e-266; >)^.42.4@ 9)0.*;6+3K .*.4@!H)++/) #)7.4.4@ H)++/) #B74:B;(ABL; Z<8EE 278<(<BQ=QA4:0> ,9AI69< &4< h4:Q )LAQ49HC HL0:BE49H 'L$K #),)80)8-+:8J 7684-2) 2@=cI<8;9Q5 5<BL5HB:5> aBL; )8IIB<(;<8EE S9HBE(58 RL0689EB d-8>!/)-> 7684-2) &<EB:4QLA(. L:54H8:4LA(E/4HE ,AL0(2<B59<:(58(Q6L<0B(I<BIL<L548:> d:G-8> F8)** aBL; ,4ANB< E/4HE #)+:8+ h4:Q 258(;BE4ANB<4:0> d:G-8> F8)** aBL; ,/4HE )L9E54Q(;<8EE(3456(h:C(,RC(&E(Q8HI89:;E 2<B59<:(58 Q6L<0B(I<BIL<L548:> '6F)/ &4< Z8<e aBL;(8]4;B #)7.4.4@ H)++/) 'L$K &:54H8:4LA 26L<;>(ABL; )8IIB<(<4Q6(HL55B(L:; EIB4EE(258 Q8IIB<(EHBA5B<> #BE4;9LA i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{/.%[H%aSPR%R0%:AAX|D K/1/#*% .,-3+$% 1)+% *,)2'6,&% *1$/#)8% .1*"#)'$% 1&'% -$'+% /,% E&,+-*'% 03,*V$H% $310$% 1)+% #)8,/$D ?,)/#)-,-$%*1$/#)8%.1*"#)'$%1&'%-$'+%/,%E&,+-*'%&,+%(,&%&'+-*/#,)%/,%G#&'D%%U-.'%'7/&1*/#,)%#$ -$'+%1/%/"'%31-)+'&$%1)+%/1EE#)8%E,#)/$D Figure 2.7: Diagram of lead refining processes IPPC (2009a). Table 2.7: Energy consumption in the lead production. Values expressed in GJ/ton of Pb. Emining Erefining Etotal Kihlstedt (1975) 4.95 7.56 12.51 Boustead & Hancock (1979) 11.34 17.31 28.65 Chapman & Roberts (1983) 9.5 18.9 28.4 Classen et al. (2007) 2.535 IPPC (2009a) 1.4 2.2.9 Manganese Manganese metal is hard and very brittle, and its primary uses in a metallic form are as alloying, desulphurising, and deoxidising agent for steel, cast iron, and non-ferrous metals, steel is the main form in which manganese is recycled. Whilst, manganese compounds are used in chemical industry and battery manufacture. Nowadays, about 17% of the magnesium supply is 35 produced by recycling of magnesium wastes. The main producing countries of manganese are: Australia, Brazil, China, Gabon and India. Manganese ore are generally mined in open pits. After mining, the beneficiation step follows with operations such as crushing, gravity concentration, converting, calcining or reducing and sintering. Then, the production of pure manganese metal can be performed through two main options, electrolysis of aqueous manganese salts (this process involves milling, reduction, calcination, leaching, filtration, precipitation and electrolysis) or electrothermal decomposition of manganese ores (this multistage process includes several smelted stages). Figure 2.8 depicts a summary of manganese products and their process routes. & [>&f1(&c&E))%"6$+"*#&*,&H$+('"$%& & ! D.$0'J0E'23**"-4')8'*"#$"#%&%'R-)P3Q+&'"#P'+W%.-'R-)Q%&&'-)3+%&'ST%776%7),%P'%+'"70'SEUUMVV0' ! ! D.$0'J0/'2.*R7.8.%P'P."$-"*'8)-'+W%'R-)P3Q+.)#')8'`%4'*"#$"#%&%'QW%*.Q"7&'Sa)#%&'SEUUJVV0' ! ! ecoinvent v2.0 report No. 10&L&^&L& Figure 2.8: Manganese products and their process routes Classen et al. (2007). Chapman & Roberts (1983) report that the energy requirement for mining and concentrating ore to produce manganese is 1.83 GJ/t, whilst the smelting and refining processes requires around 116.93 GJ/t using either electric arc or blast furnace. Whilst, Boustead & Hancock (1979) report the energy for electrowining of manganese from leach liquor as 148.67 GJ/t. According to Botero (2000), the energy consumptions for concentrating and refining processes are 15.5 GJ/t and 41.4 GJ/t, respectively . Whilst, Classen et al. (2007) report the energy requirement for the production of manganese during the concentration process as 16.5 GJ/t, whilst the refining process consumes 23.22 GJ/t. Table 2.8 summarize the energy consumption in mining and refining of manganese. Table 2.8: Energy consumption in the manganese production. Values expressed in GJ/ton of Mn. Emining Erefining Etotal Boustead & Hancock (1979) 148.67 Chapman & Roberts (1983) 1.83 116.93 118.76 Classen et al. (2007) 16.5 23.22 39.72 36 2.2.10 Molybdenum Molybdenum in its pure stage is a lustrous grey metal that can be used for a wide range of industrial applications. It is mainly used as an alloying element in steel, cast iron, and superalloys to increase hardenability, strength, toughness, and corrosion resistance. About the half of the world-wide produced molybdenum originates as co-product from the copper industry. Molybdenum is obtained commercially almost exclusively from molybdenite, which is either mined and concentrated as the primary product of the mine during the ore processing from a copper mine. Molybdenite concentrate is converted to technical-grade molybdenum trioxide by a roasting operation. The energy requirement for mining and concentrating ore to produce molybdenum is 136 GJ/t, whilst the smelting and refining processes require 12 GJ/t, in accordance to Chapman & Roberts (1983). The consumption of energy for the smelting and refining of molybdenum in an electron beam furnace reported on IPPC (2009a) is 18 GJ/t. Table 2.9 summarize the energy consumption in mining and refining of molybdenum. Table 2.9: Energy consumption in the molybdenum production. Values expressed in GJ/ton of Mo. Emining Erefining Etotal Chapman & Roberts (1983) 136 12 148 IPPC (2009a) 18 2.2.11 Nickel The importance of nickel comes from its capability, when alloyed with other elements, to increase strength, toughness and corrosion resistance of metal over a large temperature range. It thus plays a key role in technology (aerospace, marine, electronics, etc.) and construction applications. Nickel is crucial to the iron and steel industry, predominately stainless steel production and nickel alloying. Global nickel production has increased from 10,000 tonnes in 1900 to 2.1 million tonnes in 2012. More than 60% of the world nickel production during 2011 (1,800,000 ton) was produced mainly in the following five countries: Russia (15.6%), Indonesia (12.8%), Philippines (12.8%), Canada (11.1%), and Australia (10%). The largest reserves (80,000,000 tonnes of nickel) are located in Australia (30%), New Caledonia (15%), Brazil (11%), Russia (7.5%) and Cuba (6.9%), according to data reported by the USGS (2011b). It is expected that its demand will rise along with the increasing consumption trends of China, India and other emerging countries. To satisfy this demand, it is likely that most of future nickel will have to come from laterites. Therefore, assessment of energy and greenhouse emission cost for different routes of Ni production is an imperative task that has received close attention in surveys performed by Eckelman (2010), Mudd (2010), Norgate & Haque (2010). In this thesis, the analysis is performed using exergy in Section 5.2.4. Nickel is found and produced from two types of ores; oxidic (laterite and saprolite) and sulphidic ores. Due to their complex metallurgy, there is a wide range of extraction, concentration and refining processes required. Whilst the nickel content of sulphide ores can be concentrated using economical techniques, laterite processing has a tendency to be more cost-intensive (because of the extensive and complex treatment required to extract nickel), even though mining costs are lower than those for sulphide ores. Approximately 60% of nickel is mined from sulphide deposits and 40% from oxide deposits, 37 although paradoxically approximately 60% of nickel resources are found in laterites and the remaining 40% is contained in sulphides, according to USGS (2011b). The reason behind this relates to the complexity of processing nickel laterites compared to sulphides with the refining of the former implying a substantial amount of energy. Besides, the continuous decline in sulphide ore grades along with the increasing cost of underground mining, will mean that future supply of nickel will come from oxide ores, which are relatively uniform in grade and can be surface mined. Then, nickel derived from laterites will need to increase significantly in the future in order to satisfy the growing demand. During nickel production from sulphides, there are important by-products obtained such as: copper, cobalt, gold, iron and PGM (Platinum Group Metals). Cobalt, chromium and iron are the main by-products meanwhile, associated with nickel production from laterites. In recent years, the idea of using new leaching technologies instead of pyrometallurgical routes has been raised. Nevertheless, Mäkinen & Taskinen (2008) state that almost 90% of the world’s nickel production capacity is still based on pyrometallurgical processes. According to Eckelman (2010), the smelting and refining steps consume most of the primary energy, whilst mining and concentrating account for only 7-35%, depending exactly on the nickel product as well as the source. Notwithstanding, the energy required for nickel ore mining, milling and beneficiation will continue to increase as average ore grades decrease and more materials need to be processed in order to acquire the same amount of metal. Table 2.10 shows the energy consumption associated with the mining and refining processes of nickel production from laterites and sulphides according to different studies. Table 2.10: Energy consumption in the nickel production. Values expressed in GJ/ton of Ni. Sulphides Laterites Emining Erefining Etotal Emining Erefining Etotal Boustead & Hancock (1979) 232.23 696.71 Chapman & Roberts (1983) 67 150 217 6.3 570 576.3 Classen et al. (2007) 20.19 84.88 105.07 Norgate et al. (2007) 114 228 388 IPPC (2009a) 18.5 45 Mudd (2010) 39.46 100.2 139.66 572 The lower energy values presented by the European Commission, in its document Integrated Pollution Prevention and Control – IPPC (2009a), derive from the fact that nickel production comes from sulphide ores containing 4 to 15% nickel, which is a relatively high ore grade compared with those of the other references published in Table 2.10. Furthermore, energy values from an international database such as those presented in Classen et al. (2007) are based on diverse sources and are not only from one particular mine or country, hence the range. 2.2.12 Nickel laterites Laterite ores are generally found with iron oxide or silica compounds and are difficult to upgrade it to a concentrate. Laterite ore concentrations are rarely high, typically having a maximum nickel content of 3% was reported on IPPC (2009a). The general route of producing nickel from laterite ores consists of five linked operations: ore mining, drying, roasting, melting and refining, as shown in Fig. 2.9. 38 0='L#$&)DD SBC ?+")@$&&+-9)A$#',9)!"(-9#&.$9 aL5B<45B(8<B Z<=4:0 &HH8:4LQLA(aBLQ64:0 ,9AI69<4Q(LQ4; )LAQ4:4:0 aBLQ64:0 K=;<80B:(#B;9Q548: ,HBA54:0(4: *4]B;(E9AI64;B !ABQ5<4Q(S9<:LQB 58(EBIL<L5B HL55B(5<BL5HB:5 '4Q/BA(<8:;BAABE SB<<8:4Q/BA )8:NB<5B< '4Q/BA(8]4;B (('4 E4:5B< ((HL55B SB<<4Q()6A8<4;B '4(QL568;BE aBLQ64:0 !ABQ5<834::4:0 @.%-&$)DDID\)W$"$&.5)4,+V)9=$$#)4+&)".5P$,)L&+(-5#.+")4&+7)a'#$&.#$)+&$9 K1E&,3#/'% ,&'$%*1)% 0'% $.'3/'+% G#/"% $-3E"-&% $,% /"1/% /"'% )#*V'3% ,7#+'% #$% *,)2'&/'+% /,% 1% )#*V'3 $-3E"#+'%.1//'%1)+%#&,)%#$%&'.,2'+%1$%1%$318%{/.%:IAH%YWCR%:AAX|D%%!"'%.1//'%#$%/&'1/'+%#)%/"' $1.'%.1))'&%1$%.1//'%E&,+-*'+%(&,.%$-3E"#+'%,&'$D K.'3/#)8%/,%('&&,)#*V'3%1**,-)/$%(,&%1%31&8'%E&,E,&/#,)%,(%)#*V'3%E&,+-*/#,)%(&,.%31/'&#/'%,&'$H /"'$'%E&,*'$$'$%1&'%+#$*-$$'+%-)+'&%('&&,5133,6$D%%N'1*"#)8%,(%31/'&#/'%G#/"%1..,)#1%#$%13$,%-$'+ /,%'7/&1*/%)#*V'3%{/.%=IH%aSPR%W#%:AAX_%/.%]hH%Z-/,V-.E-%:AAh_%/.%A;H%Z-/,V-.E-%:AAe|%1)+ /"#$%E&,*'$$%#$%0'*,.#)8%.,&'%#.E,&/1)/D%%93/",-8"%*,)2'&$#,)%,(%)#*V'3%,7#+'%/,%#.E-&'%)#*V'3 1)+%/"')%/,%)#*V'3%*1&0,)63H%G"#*"%#$%2,31/#3'H%#$%-$'+%/,%E&,+-*'%&'(#)'+%)#*V'3H%/"'%)#*V'3%,7#+' #$%E&,+-*'+%(&,.%/"'%$.'3/#)8%,(%1%$-3E"#+#*%,&'D%%!"'%31/'&#/'%,&'$%8')'&1336%"12'%1%.17#.-. )#*V'3%*,)/')/%,(%Xg%1)+%1&'%/"'&'(,&'%),/%-$'+%+#&'*/36%#)%/"#$%E&,*'$$D !"'%E&'$$-&'%3'1*"#)8%,(%31/'&#/'$%G#/"%$-3E"-&#*%1*#+%#$%E&#)*#E1336%1%$#.E3'%1)+%$/&1#8"/(,&G1&+ E&,*'$$D%%!"'%/'.E'&1/-&'H%E&'$$-&'%1)+%,/"'&%E1&1.'/'&$%.16%21&6%(&,.%*1$'%/,%*1$'%/,%1*"#'2' /"'%0'$/%E,$$#03'%.'/133-&8#*13%*,)+#/#,)$%+'E')+#)8%,)%/"'%,&'%1)+% E&,+-*/$%#)%4-'$/#,)%1)+ ,/"'&%,0F'*/#2'$D%!"'%/'.E'&1/-&'%,(%/"'%3'1*"#)8%1-/,*312'$%#$%-$-1336%0'/G'')%=XI%1)+%=;Ik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h% 5% =]g% W#H% 1&'% E&,+-*'+% G"#*"% .1V'$% (-&/"'& E&,*'$$#)8%'1$#'&D%%!"'% )#*V'3%*,)*')/&1/'$% 1&'% -$-1336% $.'3/'+% -)+'&% ,7#+#$#)8% *,)+#/#,)$% /, ,7#+#$'%/"'%#&,)%$-3E"#+'$H%G"#*"%G#/"%,/"'&%81)8-'%.1/' $%(,&.$%1)%#&,)%$#3#*1/'%$318D%%!"' Figure 2.9: Generic flow sheet for the production of nickel from laterite ores IPPC (2009a). A brief process description follows. Laterite ores mining: Laterite ores are formed near the surface, consequently, laterite mines are mostly open cut. The energy required to concentrate the ore from the mine is very small compared to the energy required to refine it. Drying and Roasting: Since laterite ores are found in tropical climates dotted around the equator, a high moisture content is present. Hence, it is necessary to remove it by drying or calcining. Before smelting, the ore is usually roasted in a rotatory kiln electric furnace, which is a pyro-processing equipment used to acquire high temperatures in ores in a continuous process. These two stages and the next one, are those which require the highest energy input as natural gas, carbon and electricity, respectively. Smelting: An electric furnace is usually used for smelting. Nickel matte is obtained following the addition of sulphur so that the nickel oxide is converted to a nickel sulphide matte. Subsequently, it can be treated with the same processing methods as the matte produced from sulphide ores. Refining: There are several processes in the refining of nickel depending on the final products obtained. For instance, ferronickel is gained from converting processes (oxidation of the iron that is still in the matte through a Peirce-Smith converter by injecting air or oxygen into the molten bath); nickel cathodes are derived from leaching; metallic nickel can be produced by electro-winning; and metal powders can be obtained via hydrogen reduction. In this step, an improved technique used is the pressure leaching of laterites, where conditions such as pressure, temperature and other parameters are set in accordance with the ore properties or desired products in order to achieve the best possible metallurgical conditions. The resultant solution is purified either by modern solvent extraction methods or by traditional precipitation methods. By-products such as cobalt, iron and other minerals are also obtained. 39 2.2.13 Nickel sulphides Sulphide ores are typically derived from volcanic or hydrothermal processes with a typical nickel content ranging from 0.4 to 2%, according to Classen et al. (2007). Nickel production from sulphide ores involves either underground (95%) or open cut mining (5%). The processes to produce nickel from sulphide ores include several steps: ore mining, beneficiation, drying, roasting, smelting, converting, sulphuric acid, leaching, reduction, electrolysis, purification of leachate and carbonyl. Figure 2.10 gives an overview of the process options. 0='L#$&)DD ?+")@$&&+-9)A$#',9)!"(-9#&.$9 SBD Z-/-V-.E-% (31$"% (-&)1*'%#$% -$'+% #)% C-&,E'_% /"'% Z-/-V-.E-% 1)+% PW?Z% U31$"% U-&)1*'$%1)+ '3'*/&#*%,&%$"1(/%(-&)1*'$%1&'%-$'+%'3$'G"'&'%#)%/"'%^,&3+D !"'%)#*V'3%#$%&'*,2'&'+%#)/,%1%$-3E"#+'%.1//'%*,)/1#)#)8%X]%5%hIg%W#H%?,%1)+%?-D%%!"'%.1//' *1)%0'%/&'1/'+%#)%1%R#'&*'%K.#/"%?,)2'&/'&%,&%13/'&)1/#2'36%#/%*1)%0'%8&1)-31/'+%,&%*,,3'+%$3,G36 0'(,&'%1%"6+&,.'/133-&8#*13%&'*,2'&6%$/18'%{/.%:[=H%U#)31)+%W:%:AAA|D%%!"'%*,)2'&/'&%$/18'%#$ ),/%-$'+%#)%C-&,E'%1/%/"'%/#.'%,(%G&#/#)8D P.E,&/1)/%*,.E,)')/$%,(%/"'%)#*V'3%.1//'$%1&'%*,013/H%*,EE'&%1)+% E&'*#,-$%.'/13$D%%!"'%$318 E&,+-*'+%+-&#)8%$.'3/#)8%13$,%*,)/1#)$%&'*,2'&103'%.'/13%1)+%#$%/&'1/'+%#)%1)%'3'*/&#*%(-&)1*'%/, E&,+-*'%.,&'%)#*V'3%.1//'D%%!"#$%*1)%0'%8&1)-31/'+%1)+%/&'1/'+%$'E1&1/'36%{/.%A[H%W#%C7E'&/ Q&,-E%:AAe_%/.%A;H%Z-/,V-.E-%:AAe|D%%K'*,)+1&6%.1/' $%1&'%$,.'/#.'$%&'*,2'&'+%#)%/"' '3'*/&#*%(-&)1*'D !"'%(,33,G#)8%(#8-&'%8#2'$%1)%,2'&2#'G%,(%/"'%E&,*'$$%,E/#,)$D ,9AI64;B(Q8:QB:5<L5B #8LE54:0 !ABQ5<4Q(,HBA54:0 SALE6(,HBA54:0 SALE6(,HBA54:0 &HH8:4LQLA(aBLQ64:0 )8:NB<54:0 )8:NB<54:0 K=;<80B:(#B;9Q548: '4Q/BA(I83;B<.R<4d9B55BE !ABQ5<8A=54Q(#B74:4:0 )L<R8:=A(<B74:4:0 )6A8<4;B(aBLQ64:0 ,9AI6L5B(aBLQ64:0 I<8QBEE !ABQ5<834::4:0 !ABQ5<834::4:0. K=;<80B:(#B;9Q548: '4(QL568;B '4(IBAAB5E.I83;B< '4Q/BA(QL568;B '4(QL568;BE.(I83;B<.(R<4d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igure 2.10: Generic flow sheet for the production of nickel from sulphide concentrates IPPC (2009a). There are assorted processes used to produce nickel and the differences depend on factors such as the grade or the concentrate and the presence of other metals in the material mined. The options for nickel production from sulphide concentrates are classified in Table 2.11. A short description of each process involved in sulphide mining and refining is presented next. Table 2.11: Processes for nickel production from sulphide ores. Option Processes A Mining and beneficiation, drying and roasting, smelting, converting and electrolysis. B Mining and beneficiation, drying and roasting, smelting, converting and carbonyl. C Mining and beneficiation, drying and roasting, smelting, converting and leaching. D Mining and beneficiation, smelting, converting and carbonyl. E Mining and beneficiation, smelting, converting and electrolysis. F Mining and beneficiation, smelting, converting and leaching. G Mining and beneficiation, smelting, converting, leaching and hydrogen reduction. H Mining and beneficiation, smelting and leaching. I Mining and beneficiation, smelting and electrolysis. J Mining and beneficiation, leaching and hydrogen reduction. Sulphide ores mining and beneficiation: Sulphide ores are normally found hundreds of metres below the surface, therefore, they require an underground mining infrastructure. However, the major advantage of sulphide ores is that they can be concentrated easily by flotation, a 40 method which upgrades the Ni content to some 7-25 %. In this process, the ore is mixed with special reagents and agitated by mechanical and pneumatic means. Drying and Roasting: This step is needed to carry out the smelting process which requires dry sulphide ore containing less than 1% moisture. Smelting processes require a roasting step to reduce sulphur content and volatiles. According to Norgate & Haque (2010), an opportunity to improve this step is the use of the emerging bath smelting technology for ferronickel production instead of the rotatory kiln/electric furnace process. Smelting: The smelting process is often achieved in a conventional flash smelting furnace or in an Outokumpu flash furnace (DON process), which is characterized by its low energy consumption. In the DON process, high grade nickel matte of low iron content is produced in the flash smelting furnace directly without subsequent converting. Converting: Nickel is recovered into a sulphide matte containing 35-70% Ni, Co, Cu and precious metals. The matte still contains iron and sulphur that are oxidized in a Pierce-Smith converter to sulphur dioxide and iron oxide by injecting air or oxygen into the molten bath. Refining: The mattes produced by the smelting process must go through a multi-stage process in order to: recover and refine the metal content, reject iron and ultimately recover copper, cobalt and precious metals. Matte can be treated by pyrometallurgical methods but hydrometallurgical processes are more widely used. Leaching: Matte can be leached under different processes as a function of the substance used. For instance, nickel leach with chloride solution using chlorine gas, as an oxidant, allows the acquisition of copper, cobalt, lead and manganese as co-products. Another option is atmospheric leaching in a sulphate base whereby copper and cobalt are obtained as by-products. A further process is ammonia pressure leaching which uses air as an oxidant; this process is followed by hydrogen reduction to produce metallic nickel powder and eventually copper and cobalt. Ferric chloride leaching is an additional method where nickel is electro-won. This process has by-products like iron, cobalt, chrome, aluminum and lead. Lately, a new technology for Ni production is being used, namely that of heap leaching including biological activation. Although it is mainly applied in laterites, recent projects have been developed for sulphide ores. Mudd (2010) claims that heap leaching has an appreciably low capital cost but still there is dubiety regarding ore chemistry and leaching dynamics. Reduction: Hydrogen pressure reduction produces metallic nickel powder and briquettes. Electrolysis: Nickel is placed onto pure nickel cathodes from sulphate or chloride solutions in electrolytic cells. Carbonyl: Nickel carbonyl is formed by the reaction of metal with carbon monoxide at low temperature and pressure. 2.2.14 Rare Earth Elements Rare earth elements are integrated by a group of seventeen chemically similar elements; lanthanide, yttrium and scandium series. There is a classification for light, medium and heavy elements, as shown in Table 2.12. This classification is important in order to determine the suitable separation process. 41 Table 2.12: Rare Earth Elements Classification. Koltun & Tharumarajah (2008) Type Element Symbol Light Lanthanum La Cerium Ce Praseodymium Pr Neodymium Nd Promethium Pm Medium Samarium Sm Europium Eu Gadolinium Gd Heavy Terbium Tb Dysprosium Dy Holmium Ho Erbium Er Thulium Tm Ytterbium Y Lutetium Lu Yttrium Y Scandium Sc Rare earth resources are inequable distributed in the world. Minerals such as bastnasite and monazite, have been recovered for commercial production. Bastnasite deposits are found in China and United States, whilst monazite deposits are in Australia, Brazil, China, India, etc., hence China produce more than 90 % of REE. China is applying export restrictions. Massari & Ruberti (2013) declare that this fact has greatly increased the REEs prices, causing tension and uncertainty among the world hi-tech markets. Then, new mine projects are underway in other countries. Rare earths can be used like a mixture or with various levels of purity, depending on the technical application (phosphors, high refractive glass, lasers, permanent magnets, capacitors, memory systems, magneto-optical recording, oxygen sensors, temperature resistant mater, high temperature superconductivity, decolourising, polishing, deoxidising, pyrophoric properties, alloys, oil refining and catalytic converters). As new products and new applications are devised, the demand will increase and it is possible the exhaustion of the resources in near future. On the other hand, recovery processes have been developed but none of them is currently commercially viable. In regards to substitution, the ad-hoc Working Group (2010) asserts that a lot of applications for rare earth are available but the problem lies on loss of performance. The mineral monazite is a source of mostly light RE elements and contains considerable amount of radioactive elements (thorium and uranium ) and phosphorus. Koltun & Tharumarajah (2008) state that overall production processes of REE can be separated into three stages, as depicted in Fig. 2.11 and described briefly below. 42 Chapter 3 Technical development in the mining industry In the previous chapter, it was seen that factors such as ore grade and technological constraints play an important role when energy requirements in mining and metallurgical industry are evaluated. This chapter analyzes these variables through the learning curves theory, in order to know if energy required to mining ores decreases as technologies improve, if the amount of improvement decreases as time goes on and if technology improvements have more or less influence than ore grade declining in energy consumption. Furthermore, learning curves can be used as a tool to estimate future energy reductions in the mining industry. 3.1 Introduction As described in chapter 1, the mining industry is experiencing groundbreaking changes such as commodity price fluctuating, rising energy demand, water and cyanide consumption, increasing costs, declining ore grades, green-house gas emissions, increasing waste volumes and the challenge to achieve sustainable industry. Thence, sustainability1practices have become important for most major mining companies in order to reach a balance between socio-political, economic and environmental issues. In order to overcome the sustainability challenges in the mining industry, it is important to increase efficiency on resource extraction. One of the most powerful forces influencing the economic importance of natural resources in the future is technological change. Through technological innovations, it is possible to increase material efficiency in manufacturing processes and seek new substitute raw materials. Technology has always played an important role to transform mineral resources into mineral wealth and useful end-products. Nevertheless, technology breakthroughs of mineral extraction had been relatively slow until the Industrial Revolution when it showed a growth demand in commodities like coal, iron or copper. This fact lead to ever greater technological advances that even now are still used such as flotation, the blast furnace, railways, geophysics, drilling, trucks and transport, etc., allowing the mining of lower ore-grade mines. 1For Placer Dome Inc. which is one of the world’s largest gold producers “sustainability means the exploration, design, construction, operation and closure of mines in a manner that respects and responds to the social, environmental and economic needs of present generations and anticipates those of future generations in the communities and countries where it works” Milton (1998). 49 According to the ad-hoc Working Group (2010), the technological progress in exploring, mining and processing mineral raw materials has actually been the key driver that has allowed supply to keep up with demand in the past. Sohn (2006) states that over the last 25 years, powerful advances in exploration techniques in the mining sector, such as remote sensing imagery, have located new and major deposits. Breakthroughs in mining machinery, the extensive use of computers, more effective chemicals and reagents, and better use of explosives in the mines, are some of the examples that allows for cost reductions and energy efficiency. Therefore, more efficient processing methods can have a great impact on future availability of mineral resources. That said, there is a debate about the availability of commodities in the future, because there are two counteracting trends. First, general trends suggest a long-term decline in ore grade, which rises energy consumption per ton of metal extracted, and second technological transitions, which may hamper this trend. Jolliet et al. (2003) argue that if the quality of a mined abiotic resource is reduced over time, the effort to extract the remaining resource will increase. Technological learning has been widely used and it has acquired support in many applications as the energy analysis proposed in this thesis. The aim of this chapter is to become acquainted if technological breakthroughs that have occurred can preclude the rising energy demand for the gold mining industry as a case study. As experience is acquired, material and energy efficiency increase and technical changes can be expressed through the so called learning curves. In this chapter these opposite issues are analyzed through the survey of data sets of 17 major gold producing countries, with the aim to establish relationships among resource extraction and energy use; therefore it allows being aware if actual mining processes are leading the gold mining sector towards sustainability. Nevertheless, technical efficiency cannot improve indefinitely and it can never overcome thermodynamic bounds. 3.2 Ore grade evolution In the mining industry, energy consumption trend is strongly influenced by the decline in ore grades. Declining ore grades are indicative of a shift from “easier and cheaper” to more “complex and expensive” processing, which implies declining in productivity and the consequential rise in the energy intensity of mineral processing. Giurco et al. (2010) state that almost all minerals are being produced today at greater rates than at any time in history. Historical data compiled by different authors such as Mudd (2007a,b,c,2010), Page & Creasey (1975), Skinner (1986), Norgate & Jahanshahi (2010,2011) reveal that ore grades have declined dramatically throughout the last century. The best mines with the highest grades have been already extracted and today mining companies need to go further and deeper to find profitable ores. Consequently, much more energy is required per ton of mineral extracted, as depicted in Fig. 3.1. But not only that, the use of water, chemicals and waste-rock produced is also increasing strongly, leading to serious environmental problems. Hence, production and cost trends show an increment as explained promptly. That said, although there is evidence supporting the long-term decline in gold ore grades, there also exists the possibility that technological learning will overwhelm this fact. Hence, in the next sections the relationship between two issues: the decline in ore grades and the rising in energy consumption per ton of metal extracted, is analyzed. 50 0" 200,000" 400,000" 600,000" 800,000" 1,000,000" 1,200,000" 0" 5" 10" 15" 20" 25" 30" 35" 40" 45" 50" Energy'Consump.on'(GJ/t)' Ore'grade'(g/t)' United"States" Tanzania" South"Africa" Russia" NewZ" PNG" Peru" Namibia" Mexico" Mali" Laos" Indonesia" Guinea" Ghana" Chile" Canada" Brazil" Bolivia" ArgenNna" Australia" Figure 3.1: Energy consumption as a function of ore grade in global gold mining. Data from Mudd (2007b). 3.3 Evolution in technology in the minerals industry Natural resources are the main inputs into mining production. The problem is their nonrenewable character. As mineral and energy deposits are depleted, the quality and accessibility of remaining reserves usually decline, requiring more complex extraction techniques as well as greater costs. Hence, the importance of a deeper understanding of the role of technology in the mining industry. Technology has always been, and remains, a key issue of the mining industry and its ability to transform mineral resources into mineral wealth and useful end-products. According to Giurco et al. (2010), improved technological performance has allowed the industry to: • improve exploration capabilities • decrease costs and increase efficiency when processing complex ores • restrict the cost associated with environmental regulations • reduce cost of labour by increasing mechanisation • diminish input cost due to energy or chemicals • recover resources from wastes The development and widespread use of new technologies has made possible the growing in mineral supply to meet increasing demand. The latter is the main reasoning against mineral depletion. Nonetheless, this historical situation may change due to declining in ore grades. The role of technology in mineral exploration, mining, processing, manufacturing and recycling has been critical in the minerals industry. A major breakthrough in mineral exploration 51 was the emergence of geophysics, besides drilling, remote sensing, bio-prospecting, complex geological modelling tools, field analysis instruments, and so on. The main techniques of mining remain the same, however increasing mechanisation in underground mining allows to reach greater depths. In open cut mining, increasing truck sizes, safer and cheaper explosives and cheap diesel fuel, are the factors that have made possible the long-term growth. Whilst, flotation, gravity and dense media separation, carbon-in-pulp and heap leaching solvent extraction electrowinning (SX-EW), are the technologies developed to reduce energy intensity and improve metals recovery during minerals processing. Technology in manufacturing plays an important role because it allows to reduce the metal used in products as well as to design products in such way that recycling at their end of life be more easy. Technical change is a gradual process that entails technical knowledge and investment, but also an increase in material and energy efficiency. Ruth (1993) claims that both material and energy efficiency increase independently and changes can be led to the learning by doing concept. In accordance to Söderholm & Sundqvist (2007), technical change is introduced by implementing technology learning rates, which specify the quantitative relationship between the cumulative experiences of the technology and cost reductions. 3.4 The learning curve theory applied to the mining industry Learning curves come out as an empirical method to assess the effect of learning on technical change. Learning curves are used to analyze a well known observed fact. Because of the experience, humans become increasingly efficient. As experience is acquired, cost decline, efficiency and quality upgrade and waste is reduced. There is a widespread use of learning curves because of their usefulness to quantify the impact of increased experience and learning of a given technology, allowing to obtain a representation of technical change with a variety of different indicators of technological performance. A good number of studies based on the learning curves theory have been carried out. For instance, Söderholm & Sundqvist (2007) used learning curves for assessing the economic outlook of renewable energy technologies to link future cost developments to current investment in new technology. Other studies about the impact of quality on learning that suggest that learning is the link between quality improvement and productivity increase have been accomplished by various authors like Li & Rajagopalan (1997) and Jaber & Guiffrida (2008). The rate of improvement is not subjective, it is a function of the process itself. Limitations to improve the process often requires a capital investment and remove the limitations inherent in the process. Learning curves describe long-term improvement through the knowledge of • How fast can you improve to a energy saving of x? • What are the limitations to improve the process? • Are forceful goal achievable? Yelle (1979) states that the simplest and most frequently representation of learning curve in energy technology studies is the Wright’s log-linear model: Yx=Y0xb(3.1) 52 where Yxrepresents the energy required to produce the xthunit, Y0is the theoretical energy of the first production unit, xis the sequential number of the unit for which the energy is to be computed and bis a constant reflecting the rate energy decrease from year to year (learning index) and is calculated as: b=lnS ln2(3.2) where Sis the energy slope expressed as a decimal value (learning rate), while (1-S) is defined as the progress ratio which express the fraction to which energy requirements are reduced with cumulated production. However, Ruth (1995b) asserts that there is a limit on the energy use to ore production that cannot be exceed with increasing experience and in this case it is the minimum theoretical energy required to concentrate a substance from an ideal mixture of components. This limit can be calculated through the concentration exergy presented in Chapter 4and expressed in Eq. 4.28. Accordingly, considering the energy limiting value, the learning curve can be expressed as Eq. 3.3 which integrates thermodynamic concepts to the learning curve analysis. Y−bc=Y0xb(3.3) In the mining industry, the technology learning rates state the correspondence between the cumulative experience of the technology and the energy requirement reductions. These reductions are the result of learning by doing. For instance, performance improves as new technologies and mining methods are implemented. Accordingly, learning curves will be used to empirically quantify the impact of accomplishing new mining practices on the energy consumptions of ore mining. Kahouli-Brahmi (2008) has performed an enhancement for the simplest learning curve by applying a factor related with research and development. These extended formulation is known as the two factor learning curve (TFLC) expressed as follows: Y(x,K S)=Y0xb∗K Sc(3.4) where K S is the knowledge stock and cis the elasticity of learning by researching. According to Weiss et al. (2010), both types of learning curves are the most commonly used to assess technology learning rates in the energy sector. However, Jamasb & Kohler (2007) declare that a multiple-factor learning curves that account for other independent variables besides cumulative production represent an approach to get a better understand of technological learning by enhancing the knowledge base. Learning rates depend on the data points that are chosen. Previous surveys, as the one performed by Kahouli-Brahmi (2008) reveals significant variability in estimated rates between different energy technologies, which ranges from 1% to 41.5%. On the other hand, the study developed by McDonald & Schrattenholzer (2001) show that the average value of learning rates for energy technologies is 16-17% and learning rates for manufacturing is 19-20%. Negative learning rates can be interpreted as a consequence of experience depreciation, if no important external factors (such as declining ore grades in mining technologies) are influencing the production process. 53 The estimating of learning curves using econometric techniques has some highlights. For instance, the need to survey the effect of detach single observations especially outliers that may effect the learning rate estimate or the impact of using different variable definitions. Another issue is related with the way in which technology learning is operationalized in order to know if the indicator of learning by doing selected is assuredly capturing the impact of learning by doing activities or if it is following a general trend of technical progress. Also is important to investigate differences in learning rates across different technologies. Söderholm & Sundqvist (2007) assert that a further point is the assumption in the model when the independent variable is defined as well as the influence of other variables on this. The use of learning curves implies that a technology had been already invented and implemented. Ruth (1993) argues that as more awareness is obtained in using materials and energy to produce outputs with a determined technology, changes in its performance occur slowly. This can suggest that a high level of expertise in ore mining with a specific technology will result in a discontinuity in learning. Further interruptions or discontinuities in the learning curve occur when intermittent production is presented, e.g. mine rehabilitation. Afterwards, learning curves may start again with a new technological breakthrough, a radical innovation or a paradigm change, according to Wellmer & Becker-Platen (2002). Also, Schoots et al. (2008) aver that breakthroughs may influence the progress ratio. Wellmer & Becker-Platen (2002) avow that learning curves for mining sector are driven by financial rewards, directly (i.e. by the discovery of a very economic deposit or a price increase of a commodity) or indirectly (by a penalty). 3.5 Learning Curves applied to Global Gold Mining In this section, the analysis of data set on historic gold mining in the main gold producing countries was carried out by means of linking resource extraction with energy use through the learning curves approach. Learning curves were originally developed to evaluate the effect of learning by doing in manufacturing. However, there are new applications such as analysis innovation and technical change in energy technology. This section looks over energy data on gold mining for Australia, North America, Africa and the Asia-Pacific compiled by Mudd (2007b,c), who pointed out the critical aspects of mineral resource sustainability such as resource intensity linked to technology. Learning rates and progress ratios were calculated for each mine using Eq. 3.2 and Eq. 3.3. The information was grouped according to the different mining technologies used, because learning by doing will differ between mines, countries and technologies. Progress ratios are different for each country and even for each mine although they use the same recovery process technology. This is due to inherent factors to each mine such as project age, depth, ore types, etc. Results of the analysis performed for different mining operations and recovery processes are shown in Table 3.1. The assorted configurations of gold mines like open pit (OP), underground (UG) or mixed (MIX), as well as the energy source (diesel, coal, hydro, gas or any mix among them) are factors that influenced the progress ratio. Recovery process technologies show average progress ratios around ±25% as shown in Fig. 3.2. Open pit operations with heap leach technology (HL) as recovery process as well as un54 derground operations using carbon in pulp (CIP)2technology are the mining options with the greatest progress ratios. Figure 3.2: Distribution of average progress ratios for gold mining industry. Negative progress ratios convey that technological learning has been unable to overcome the increase in energy consumption during mining operations due to the declining in ore grade. On the other side, positive progress ratios imply that mining recovery processes have achieved to maintain or decrease the energy consumption during mining operations throughout time. Accordingly, the progress ratio becomes an indicator to identify those mines where mining practices are successful when saving energy. Table 3.1: Progress ratio for global gold mining. Country Mines Operation Recovery Energy consumption Average ore Progress ratio process (GJ/t Au) grade (g/t Au) by mine (%) Argentina Veladero OP HL 476.295 1,2 20% Cerro Vanguardia OP CIL 57.568 7,4 -15% Australia Granny Smith OP CIP 48.010 12 -1% Kidston OP CIP 95.494 10,5 -7% Henty UG CIL 91.984 5,6 -31% Kalgoorlie West MIX CIL 88.191 39,3 -8% Agnew MIX CIP 171.934 21,6 -22% Hill 50 MIX CIL 76.790 12,2 -6% 2The difference between carbon-in-pulp and carbon-in-leach processes is that for the first one the adsorption occurs after the leaching cascade section of the plant, whilst for the second one leaching and adsorption occur simultaneously De Andrade (2007). 55 Country Mines Operation Recovery Energy consumption Average ore Progress ratio process (GJ/t Au) grade (g/t Au) by mine (%) St Ives MIX CIP, HL 47.554 12,9 4% Central Norseman MIX CIL 148.309 8,4 1% Plutonic UG CIP, HL 155.778 3,5 2% Darlot UG CIL 181.927 1,5 14% Lawlers UG CIL 168.441 2 4% SuperPit OP CIL 141.350 2,1 -4% Mt Leyshon OP CIP 117.687 2,9 5% Tanami-Granites MIX CIP/CIL 174.270 1,6 -30% Boddington OP CIL 498.021 1,1 -14% Pajingo UG CIP 112.653 4,3 24% Bronzewing MIX CIL 63.504 7,5 4% Jundee MIX CIL 144.078 3,9 -19% Sunrise Dam OP CIL 180.196 0,4 1% Challenger MIX CIP 210.419 0,5 -26% Ravenswood OP CIP 236.498 1,5 18% Stawell UG CIL 87.669 1,1 -4% Fosterville OP BIOX, CIL 167.234 1,2 -5% Peak (NSW) MIX CIL 162.455 3,3 6% Brazil Mineracao OP HL 73.567 6,9 -4% Serra Grande OP CIL 64.707 6,3 -23% Amapari OP HL 208.293 2,4 29% Morro do Ouro OP CIL 176.444 0,5 1% Paracatu OP CIL 191.369 0,3 -2% Maricunga OP HL 163.089 0,4 24% Canada Dome-Porcupine MIX CIP 113.562 2,9 0% Hemlo MIX HL 102.457 4,6 -6% Musselwhite UG CIP 91.984 5,6 -16% Campbell UG CIL/CIP 67.688 16,8 -2% Eskay Creek UG 88.191 39,3 -16% Red Lake UG CIP 22.401 79,7 0% Chile La Coipa OP CIL 385.254 1,1 -3% Ghana Tarkwa OP CIL 110.248 1 -4% Damang OP CIL 166.186 1,4 -11% Iduapriem OP CIP 130.283 1,8 -22% Obuasi MIX HL 192.560 2,1 -4% Guinea Siguiri OP CIP 189.244 1,1 4% Indonesia Kelian OP CIL 190.161 2,4 0% 56 Country Mines Operation Recovery Energy consumption Average ore Progress ratio process (GJ/t Au) grade (g/t Au) by mine (%) Laos Sepon Au OP CIL 217.951 2,4 0% Mali Sadiola OP CIP 102.837 2,9 -9% Yatela OP CIP 56.497 3,5 1% Morila OP CIL 125.031 4,2 0% Mexico San Dimas UG CIL/CIP 59.814 7,3 -12% Namibia Navachab OP CIP 110.056 1,8 -7% Peru Pierina OP HL 64.597 1,9 -9% Lagunas Norte OP HL, CIP/CIL 37.261 2,2 7% PNG Misima OP CIP 352.899 1,1 -2% Porgera OP CIP 431.600 5,2 2% Lihir OP CIL 374.258 5,4 -3% South Africa Harmony Group UG-tails CIP/CIL 205.291 5 -4% Beatrix UG-tails CIL 195.083 4,5 -1% Driefontein UG-tails CIP 204.009 5,1 -1% Kloof UG-tails CIP 229.860 7 -12% West Wits Field UG-tails CIP 175.272 8,5 0% South Deep UG CIP 424.041 6,5 -3% Vaal River UG-tails CIP, CIL 170.913 3,4 -2% Tanzania Geita OP CIL 257.028 2,3 -18% North Mara OP CIL 201.729 3,5 12% Tulawaka OP CIL 178.024 10,7 11% Bulyanhulu UG FLOTATION 78.621 12 3% United States Round Mountain OP GRAVITY 151.530 0,6 -5% Marigold OP HL 157.725 0,7 -5% Bald Mountain OP HL 129.211 1,1 -33% Cripple Creek-Victor JV OP HL 148.190 0,6 -3% Golden Sunlight OP CIP 131.676 2,6 -13% Cortez OP CIL 124.621 2,5 -5% Fort Knox OP HL 187.460 0,8 1% Wharf OP HL 146.938 1 -1% Goldstrike MIX CIL 138.706 6,8 0% Turquoise Ridge UG CIL 49.936 14,9 10% Pogo UG CIP 111.056 14,6 8% 57 3.5.1 Australia Global demand for Australian minerals and metals continues to rise. However, energy inputs to mining are rising, direct employment from mining is relatively low and productivity is declining. Giurco et al. (2010) avow that being Australia very dependent on mining exports, strategies to assure long-term national benefits from minerals including social, economic and environmental issues should be developed. Australian data was compiled by Mudd (2007b) who asserts that industry will be challenged by declining ore grades and the increase of environmental and social costs or resource intensity. Other authors such as Prior et al. (2012) reveal that mineral production in Australia is currently unsustainable, not because of resources being finite, but because of the impacts associated with its processing and use. Notwithstanding it, commodities mining in Australia has been and continues to be a very important issue of Australian industry, representing approximately 7.7% of Australia’s GDP in 2008-2009, as stated by Giurco et al. (2010). Gold is produced in almost 30 mines, including open pit, underground and mixed. In this chapter an analysis of the influence of technical development and declining ore grades on the availability of Australian gold resources was accomplished. Obtained results suggests that although progress in technology has been made, in most cases energy requirements are increasing, because the main variable is the ore grade. Progress ratios represent the amount of improvement in mining technologies for several mines in Australia, such as Kidston, Henty, Kalgoorlie, Agnew, St Ives, Plutonic, Darlot, Lawers, Superpit, Mt. Leyshon, Tanami, Boddington, Pajingo, Sunrise Dam, Challenger, Ravenswood and Peak. Available data from 1990 through 2008, is shown in Fig. 3.3. !100%% !80%% !60%% !40%% !20%% 0%% 20%% 40%% 60%% 80%% 1991% 1992% 1993% 1994% 1995% 1996% 1997% 1998% 1999% 2000% 2001% 2002% 2003% 2004% 2005% 2006% 2007% 2008% Progress'Ra*o'(%)' Granny%Smith% Kidston% Henty% Kalgoorlie%West% Agnew% Hill%50% St%Ives% Central%Norseman% Plutonic% Darlot% Lawlers% SuperPit% Mt%Leyshon% Tanami!Granites% Boddington% Pajingo% Figure 3.3: Progress ratios for gold mines in Australia. Australia has a rising trend in energy consumption as well as in learning rates from 1857 to 2009, as shown in Figure 3.4. During this period its cumulative production was 11,565 t of gold. The general trend is a rising energy requirement throughout time, but there are some mines such as Agnew, Challenger, Plutonic, Super Pit, Fosterville, Tanami-Granites and Jundee, which show a reduction in its energy consumptions through the time. Another important issue when analyzing the energy consumption is the ore grade. As the concentration of the ore in a deposit tends to zero, the energy required to separate the substance from the mine tends to infinity. This is a consequence of the Second Law of Thermodynamics and is an empirical fact. 58 3.5.13 Peru Pierina belonging to Barrick Company (2013a), is an open-pit mine, with truck and loader operations. Ore is crushed and transported through an overland conveyor to the leach pad area. Run-of-mine ore is trucked directly to a classic valley-fill type of leach pad. Pierina is currently engaged in energy efficiency optimization efforts which lead to decreasing energy consumptions or increasing energy efficiencies as well as reducing greenhouse gas emissions (Barrick Company 2013b). Over the last decade, improvements in the leach pad system as well as in the surface water management system have been made, as reported by Ausenco (2013). Despite the improvements made in the mine, energy consumptions continue growing due to the ore grade declining, leading to negative progress ratios. This can imply that technological efforts adopted by the mine are not enough to bring down the energy increasing trend. Yanacocha Gold Mine is the largest and most profitable gold mine in Latin America. It operates a complex of six open pit gold mines, five leach pads and two processing facilities. Gold is extracted from ore through a cyanide heap leach process, then the solution is treated by the Merrill Crowe process. After recovery, the contained metal is smelted and casted as bars containing 75% gold and 20% silver, according to Mining Technology (2013). Everything about Yanacocha facilities are enormous as well as its wealth, then this mine has the largest gold production with very low energy consumption. Lagunas Norte mine is owned by Barrick Company (2013a). In 2006 began the development of a high grade area with a longer hauling cycle. Gold and silver are recovered in a conventional clarification and zinc precipitation circuit, using the Merrill-Crowe process. For this reason, a decrement in energy consumption from 2006 is observed, leading to a positive progress ratio, regardless the ore grade declining. Available data from 2001 through 2007, is shown in Fig. 3.12. !25%% !20%% !15%% !10%% !5%% 0%% 5%% 10%% 15%% 2002% 2003% 2004% 2005% 2006% 2007% Progress'Ra*o'(%)' Pierina% Lagunas%Norte% Figure 3.12: Progress ratios for gold mines in Peru. 3.5.14 Papua New Guinea Misima gold/silver mine ends its operation in 2001 with stockpile milling anticipated to continue into 2004, as informed by Mining Technology (2013). Gold recovery uses a standard crushing, grinding and carbon-in-pulp (CIP) flowsheet. All data analyzed for this mine indicates that both cumulative production and energy consumption increase. Besides, ore grades are declining, resulting in negative progress ratios. For ore grades varying between 3 and 5 g/t Au, there are two mines that are mined: Porgera and Lihir mines. Almost all data for this range is from Porgera mine, the energy consumption 65 grows as cumulative production does, whilst progress ratios vary according to the ore grade changes. For the interval within 5 and 6 g/t Au of ore grade, Porgera mine reveals a slightly trend of ore grade decrease, whilst cumulative production increase. The result is a decreasing tendency of energy consumption prompting a positive progress ratio. The Porgera gold mine is operated by a Barrick subsidiary. According to Mining Technology (2013), both open-pit and underground mining methods are employed because it was initially an underground operation until 1997, but was resumed in 2002. Additionally an open-pit mining became increasingly important from 1993. A lot of changes have been made into the mining processes. For instance, the open pit has been mined in five stages, with final-stage overburden removal taking place during 2001. The open-pit truck and shovel fleet was expanded in 1995 and 1997. Besides, in 1999 a flotation expansion was installed as well as and additional oxygen capacity to increase autoclave throughput. Run-of-mine ore is crushed and ground, then gold is recovered in a gravity circuit and flotation is used to recover a sulphide concentrate before the applying of CIP cyanide leaching process. The final step is the electrowinning process that produces bars of 88% gold average. The Lihir gold mine is an open-pit mine consisting of two adjacent overlapping pits. Its operations include; crusher, SAG and ball mill circuit, flotation circuit, pressure oxidation and CIL processing facilities, and electrowinning and smelting facilities to produce gold dorè (Global Infomine 2013). Available data from 1997 through 2007, is shown in Fig. 3.13. !10%% !8%% !6%% !4%% !2%% 0%% 2%% 4%% 6%% 8%% 1998% 1999% 2000% 2001% 2002% 2003% 2004% 2005% 2006% 2007% Progress'Ra*o'(%)' Misima% Porgera% Lihir% Figure 3.13: Progress ratios for gold mines in Papua New Guinea. 3.5.15 South Africa In South Africa the data are from the following mines: Harmony Group, Vaal River, Beatrix, Driefontein, South Deep, Kloof and West Field. Gold Fields company owns Beatrix, Driefontein, South Deep and Kloof. Beatrix Gold Mine owned by Gold Fields (2013) consists of four surface operating shafts that mine various gold bearing reefs from open ground and pillars. Ore is processed at two metallurgical plants, where milling, CIL process, elution and gravity circuits, electrowinning and smelting operations are carried out. Driefontein Gold Mine owned by Gold Fields (2013) includes eight shaft systems that mine various gold bearing reefs from open ground and pillars. Ore extracted from the bearing reefs is processed at three metallurgical plants. It has a centralized elution and carbon treatment facility since 2001. The mineral processing technology was based on SAG milling circuit following by a cyanide leaching until the year 2003, when the-se processes were replaced by the CIP plant. 66 Kloof Gold Mine owned by Gold Fields (2013) is composed of five shaft systems and two gold plants, the gold is produced from a combination of underground mining and processing of surface waste rock dump material. For the mineral processing, two operational metallurgical facilities are used, including a central elution and smelting facility. In 2001 and ACC Pump Cell CIP circuit was installed to replace the less efficient drum filtration and zinc precipitation. Also, the upgrade included the installation of continuous electrowinning sludge reactors. South Deep Gold Mine owned by Gold Fields (2013) incorporates two shaft systems that mine various auriferous conglomerates from open ground and pillars. The ore is processed at a central metallurgical plant. The mineral processing includes a milling circuit SAG, a CIP circuit, an elution system to finally recover gold by electrowinning and smelting processes. Great Noligwa underground gold mine is situated close to the Vaal River, it comprises four gold plants, one uranium plant and a sulphuric acid plant. Great Noligwa has its own milling and treatment plant which applies conventional crushing, screening, grinding and CIL processes to threat the ore and extract the gold, as informed by Mining Technology (2013). Harmony (2013) Group in South Africa include the next mines: Bambanani, Doornkop, Kusasalethu, Evander, Joel, Kalgold, Masimong, Phakisa, Phoenix, Target, Tshepong and Virginia . Sometimes the value of ore milled is too large compared with the values of other mines in the same ore grade scope. Therefore, it is probably that these data are referred to several mines. West Wits Operations include Driefontein, Kloof and South Deep. Harmony Group reports the highest energy consumption value, leading to a negative progress ratio. For ore grades from 6 to 7 g/t Au, the only existing mine is Kloof, which shows an increase in the energy consumption for the same ore grade value from 2004 to 2008, therefore a negative progress ratio is shown. Data in the span between 8 and 9 g/t Au is from West Wits Field, even though ore grade decrease, energy consumption decrease too, leading to a null positive progress ratio. Available data from 2003 through 2008, is shown in Fig. 3.14. !35%% !30%% !25%% !20%% !15%% !10%% !5%% 0%% 5%% 10%% 2003% 2004% 2005% 2006% 2007% 2008% Progress'Ra*o'(%)' Harmony%Group% Beatrix% Driefontein% Kloof% West%Wits%Field% South%Deep% Vaal%River% Figure 3.14: Progress ratios for gold mines in SA. 3.5.16 Tanzania North Mara gold mine consists of three open pit deposits and belongs to Placer Dome Company. AngloGold Ashanti (2013) owns Geita gold mine, which began production in 2000. Geita is a multiple open-pit operation with underground potential. For ore grades between 3 and 5 g/t Au, most of the data is from North Mara, during 2003 and 2004 the ore grade is the same but 67 the energy consumption differ greatly, almost by a 50%, the same situation is repeated for years 2005 and 2007, despite of the-se observations, there is a declining trend in energy consumption, resulting in a positive progress ratio of 12%. Tulawaka Gold mine consists of a completed open pit mine with an underground access ramp, an ore stockpile area and crushing plant, a processing plant. The ore processing method includes SAG, gravity recovery and CIL (Global Infomine 2013). It is the only mine which report data in the span of 9 and 14 g/t Au, within this large ore grades the energy consumption falls as cumulative production increase, hence a positive progress ratio of 11% is shown. Bulyanhulu Mine is owned by Barrick Gold Corp. Bulyanhulu is an underground trackless operation using long hole and drift and fill as its principal toping methods (Gold mining in Tanzania 2013). It shows a clear downtrend in ore grade that results in an energy consumption increase as cumulative production rises. However, due to the fact that ore grade vary from one year to another without a clear trend of increase or de-crease and this variation is not significant, the progress ratio is positive. Available data from 2001 through 2007, is shown in Fig. 3.15. !40%% !30%% !20%% !10%% 0%% 10%% 20%% 30%% 2002% 2003% 2004% 2005% 2006% 2007% Progress'Ra*o'(%)' Geita% North%Mara% Tulawaka% Bulyanhulu% Figure 3.15: Progress ratios for gold mines in Tanzania. 3.5.17 United States Barrick Company (2013a) in USA owns several mines. Bald Mountain mine is an open pit, runof mine with conventional heap leaching technology and carbon absorption for ore treatment. Cortez mine is mined by conventional open-pit methods. It employs three different metallurgical processes to recover gold. Lower-grade oxide ore is heap leached, while higher grade ore is treated in a conventional mill using cyanidation and a CIL process. Heap leached ore is hauled directly to leach pads for gold recovery. Golden Sunlight mine is mined by conventional open-pit methods. The ore treatment plant uses conventional CIP technology as well as San Tailing Retreatment (SRT). Goldstrike complex includes an open pit mine and two underground mines. The open pit is a truck and shovel operation using large electric shovels. While one of the underground mines is a high grade ore body which is mined by transverse longhole stoping, underhand drift and fill mining methods, while the other is a trackless operation, using two different underground mining methods: longhole open stoping and drift and fill. Also, it consists of two processing facilities that are used for both the surface and underground operations: (1) an autoclave circuit and (2) the roaster. Marigold mine is an open-pit operation that uses heap-leaching to process its ore. Round Mountain mine is a conventional open-pit operation that uses multiple processing methods 68 including crushed ore leaching, run of mine ore leaching, milling of higher grade ore and the gravity concentration circuit. Turquoise Ridge mine uses underhand cut and fill mining methods. Ore is transported to an external mill for processing. The refractory gold ore is treated by pressure oxidation technology and gold is recovered using conventional CIL technology. Cripple Creek Victor JV gold mine is a low-grade, open pit operation. The ore is treated using a valley-type, heap leach process with activated carbon used to recover the gold. The resulting doré buttons are shipped to a refinery for final processing, according to Mining Technology (2013). Fort Knox mine belonging to Kinross (2013) is an open pit mine, that uses as processing methods CIP mill, heap leach and gravity. According to Mining Technology (2013), production from heap leach began in late 2009. Ridgeway underground mine produces gold and copper. It is in the process of transitioning from the sub-level cave to a block cave beneath the existing mine. Crushed ore from the underground is delivered by conveyor to a surface stockpile; then, gold and copper are recovered in a conventional floatation circuit to produce a copper concentrate containing elevated gold levels. The next step is to pump to the filtration plant where it is dewatered prior to being transported to export to smelters throughout East Asia. Wharf owned by Goldcorp (2013) is an open pit and heap leach mine that has been in operation since 1983. Global Infomine (2013) state that Barneys Canyon gold mine is an open pit mine that started production in 1989. Mining and milling ended in 2001. Gold production from stockpiles continued until 2005. The Kettle River-Buckhorn gold mine belonging to Kinross (2013), was originally conceived as an open pit mine, but then it was redesigned and developed as an underground mine. The primary mining method employed is cut and fill. Its ore is processed through milling, flotation and CIP processes. Pogo gold mine is an underground mine that utilizes a cut and fill drift method. The milling operation includes grinding, sulfide flotation, paste thickening, leach/CIP, cyanide detoxification, tailings filtration and gravity recovery (Global Infomine 2013). Open pit mining in USA is characterized by negative progress ratios, but underground mining using CIL and CIP shows positive progress ratios. Available data from 2002 through 2007, is shown in Fig. 3.16. !80%% !70%% !60%% !50%% !40%% !30%% !20%% !10%% 0%% 10%% 20%% 30%% 2002% 2003% 2004% 2005% 2006% 2007% Progress'Ra*o'(%)' Round%Mountain% Marigold% Bald%Mountain% Cripple%Creek!Victor%JV% Golden%Sunlight% Cortez% Fort%Knox% Wharf% Goldstrike% Turquoise%Ridge% Pogo% Figure 3.16: Progress ratios for gold mines in United States of America. 69 3.6 Progress ratios in the gold mining industry Assuming that all mines of the same kind of operation as well as recovery process for a specific country are comparable, progress ratios can be presented as Fig. 3.17 displays. Average progress ratios obtained between different operation and recovery processes ranged from + 20% to – 22%. Australia has an excellent progress ratio when CIP technology is used either in open pit or underground mines. This can imply that gold mining industry in Australia has overcome the declining in ore grades through technological learning. A compilation of the best practices and all mining process in general would be very useful for mines using the same recovery process around the world. The sharing of operational and technical experiences, with countries such as Papua New Guinea would be an excellent way to improve the efficiency in the gold mining sector. For the United States, positive progress ratios can be observed when underground mining is performed (either when CIP or CIL process are applied). Hence, the relevant fact here is the kind of mining operation used: underground. Performance benchmarking of gold mines in Canada can be a worthwhile action to improve practices in the gold mining industry. In South America, countries like Argentina and Brazil show great positive progress ratio when operations in open pit mines with heap in leach technology as recovery process is employed. Again, it would be very useful an extended compilation of their best practices in order to share this information with countries that have mines with the same geological and technological characteristics, such as mines in Peru. 3.7 Summary In this chapter the influence of technical development and declining ore grades on the availability of world gold resources has been studied, applying the learning curves approach and estimating progress ratios for each country. The latter allowed to identify mines in which mining operations have proved to be successful when the goal is to save energy. Therefore these estimates can be used to point out best mining practices and serve as a reference for other mines with similar conditions. Results for Australia gold mining industry were published in Valero, Valero & Domínguez (2011), afterwards results for global gold mining were published in Domínguez & Valero (2013). It should be pointed out, that the improvement in mining technologies, represented in this survey by the progress ratios calculated for different countries, mines operations and recovery processes are not related to time or cumulative production as it happens to conventional applications (such as manufacturing) when the theory of learning curves is applied. In the mining sector, an additional factor needs to be taken into account, and that is the key variable ore grade change. The learning effect is measured in terms of reduction in the energy requirements of mining operations. This way, an improvement in the energy efficiency of the processes does not necessarily imply an overall energy reduction, since the decrease in the ore grade may dominate. General results suggest that although progress in technology has been made, in most cases energy requirements are increasing, because the main variable is the ore grade. Therefore, it can be asserted that technology cannot in general avert the rising energy demand for gold mining 70 Figure 3.17: Progress ratio for different recovery process and countries in gold mining industry. 71 in the future if no major changes are performed in gold mines around the world. Hence, energy consumptions will continue to increase threatening the capability to fulfill the prospective demand. It is crucial to analyze carefully those countries that are and will be the major gold producers such as Australia, South Africa, Russia, Chile, United States and Indonesia. The data analyzed reveals that South Africa and Australia show the greatest energy consumptions and hence should increase their efforts in improving their mining practices. Additionally, due to the strategic position of China in the gold mining industry, analyzing its data sets in energy consumptions and ore grades would be also very interesting and profitable. This analysis allow us to have a more suitable understanding of the mining sector and the outcomes of technology evolution together with ore grade declining. Furthermore, this analysis lets us understand the general trends in resource consumption in the mining industry by means of identifying best mining practices around the world. Learning models can be used to survey the effect of policies and research and development resources applied in the mining sector in order to accelerate technical progress. Nevertheless, the accuracy of the estimated learning rates and progress ratios remains a major issue. This chapter closes the analysis of energy consumption in the mining industry performed in this thesis. Hence, the next chapter presents an overview of thermodynamics and the methodologies that can be used to analyze mining industry from a thermodynamic point of view. 72 Chapter 4 Thermoeconomic Analysis The aim of this chapter is to provide the thermodynamic background to assess mineral resources and their corresponding mining and metallurgical processes. Particularly, it introduces Thermoeconomics, the Thermoecological cost and the Exergoecological approach, which will be later combined so as to adequately cover the requirements of the mining industry. 4.1 Thermodynamics and the evaluation of natural resources The use of thermodynamics to evaluate natural resources began with Georgescu-Roegen (1971) who draw on the idea of entropic degradation as a fundamental restriction for economic activity. He pointed out that the increasing rates of extraction of natural resources, leads to the entropic degradation of the earth. Afterwards, authors such as Naredo (1987) and Cleveland & Ruth (1997) studied the implications for material and energy use in economic development and the ecological systems, taking into account thermodynamic constraints. Within this research line, in Naredo & Valero (1998), Valero proposed a new approach to calculate the exergy replacement cost of natural resources. This method based on exergy has led several research works which in turn have given birth to two new disciplines: Physical Hydronomics and Physical Geonomics. The former is concerned with water analyzes. The first study on this topic was performed by Zaleta et al. (1998), who applied the methodology to the water of a river. Martinez (2009) refined the methodology for undertaking exergy cost assessments of water bodies. Physical Geonomics meanwhile, is concerned with mineral analysis. In this respect, an exergy cost analysis of the Earth’s mineral wealth was firstly accomplished by Ranz (1999). Later on, Botero (2000) carried out an exergy assessment of natural resources (including minerals, water and fossil fuels) and Valero Delgado (2008) performed a research work focused on the exergy evolution of the mineral capital on Earth which has led to many publications in this field (see section 4.6). This thesis continues with this research line with the aim of addressing unresolved issues that remained open in Valero Delgado (2008) such as the assessment of the exergy cost and ore grade in global mining. A brief review of thermoeconomic concepts taken from Valero & Lozano (1994) are presented in the next section, so as to give an overview of the different tools used in this thesis. 73 4.2 Fundamentals of Exergy Analysis and Thermoeconomics The exergy analysis is based on both the first and second laws of thermodynamics. Its aim is to calculate the useful energy associated with a thermodynamic system or with each flow stream in a given process, thereby identifying and evaluating the inefficiencies of the system. Exergy analysis provides two important messages: one is that it allows quantifying and locating thermodynamic losses. The second is, it allows to concentrate on the relevant part of the energy, namely “useful” energy. These analyzes can solve problems related to complex energy systems that could not be solved by using conventional energy analyzes. Nevertheless, exergy analysis (Thermodynamics) is necessary but not sufficient to determine the origin of losses and the potential for energy saving as will be see in the next sections. Accordingly, it is necessary to include the concept of purpose (Economics) by means of the definition of efficiency. Exergy efficiency compares the performance of a real process to an ideal one of the same type. The conceptual link between these two disciplines is Thermoeconomics, which is a general theory for energy saving, that integrates thermodynamics (exergy analysis) and economics (exergy cost) by means of the Second Law, for the analysis, optimization and diagnosis of complex energy systems Lozano & Valero (1993). In this kind of systems, energy flows are easily converted into exergy ones and their value is an indication of quality. Among other applications, thermoeconomics is used for: • Rational prices assessment of plant products based on physical criteria. • Optimization of specific process unit variables to minimize the final product cost, i.e. global and local optimisation. • Detection of inefficiencies and calculation of their economic effects in operating plants, i.e. plant operation thermoeconomic diagnosis. • Evaluation of various design alternatives or operation decisions and profitability maximization. • Energy audits. 4.2.1 Exergy As already explained, exergy is an adequate thermodynamic property to account for energy quality. The exergy of a thermodynamic flow is the minimum amount of technical work needed for its production, from a given reference environment. Exergy provides a thermodynamic value of any energy stream with respect to reference conditions, and can be rigorously obtained from the laws of thermodynamics, which allows precise measurements. Valero & Torres (2004) argue that exergy may be regarded as a measure of the capacity of a given form of energy to produce work and is reasonable to assess the price of energy products on the basis of its exergy content. Exergy is also a worthwhile concept in Economics. Exergy puts forward an approach to assess resource depletion and environmental destruction. Szargut (1993) pointed out that “the concept of exergy is crucial not only to efficiency studies but also to cost accounting and economic analysis”. Hence, exergy is a pragmatic tool for evaluating: fuels and resources, process, devices and system efficiencies and their costs, as well as the value and cost of systems outputs. 74 cific TEC of a raw material or semi-finished product ith of the considered process, fi j is the coefficient of production of the ith by-product per unit of the jth main product, ai j is the coefficient of consumption of the ith raw material and semi-finished product per unit of the jth main product, bf j is the exergy of the fuel (f) and bm j is the exergy of the mineral (m) immediately extracted from nature per unit of the jth main product, pk j is coefficient of the production of the kth rejected harmful waste product per unit of the jth main product, and ζkis the TEC of compensation of the deleterious impact of the kth rejected waste product. Distinctness of the TEC into a fuel part T EC fand mineral part T ECmallows to identify the kind and amount of exergy consumed in the course of the production process. Accordingly, it is necessary to include a second set of equations used to determine the fuel part of TEC through the expression: zjρj+X i (fi j −ai j )ziρi=X f bf j +X k pk j zkζk(4.24) where zjis the fraction of TEC of a main product jth due to fuel consumption and ziis the fraction of TEC of a raw material or semi-finished product ith due to fuel consumption. The fraction of TEC of a raw material or semi-finished product ith due to mineral consumption does not appear in Eq. 4.24 because this equation is evaluating only the fuel part. Hence, the fraction of TEC of a main product jth due to mineral consumption will be the difference (1-zj), whilst (1-zi) will be the fraction of TEC of a raw material or semi-finished product ith due to mineral consumption. Szargut (1986) proposed that the TEC part due to rejections of harmful substances to the natural environmental can be assessed by means of: ζk=Bwk GDP +X k Pkwk (4.25) where ζkis the Thermo-Ecological Cost due to the emission of a unit of the kth waste product, Bis the exergy extracted per year from the domestic non-renewable natural resources, wk is a monetary index of harmfulness of kth substances, GDP is the Gross Domestic Product and Pkis the nominal flow rate of the kth deleterious waste product rejected to the environment. Finally, the TEC can be the basis for determining the index of sustainability which expresses the ratio between Thermo-Ecological Cost of the useful ith product and its specific chemical exergy: ri=ρi bi (4.26) A lower index of sustainability means fewer cumulative exergy consumption of natural resources per unit of exergy of a given useful product. Accordingly, the lower the index of sustainability, the better the product from an ecological point of view. 81 4.5 Similarities and differences between Thermoecological Cost and Thermoeconmic Input-Output Analysis As aforementioned, the exergy cost proposed by Valero et al. (1986) in their “General Theory of Exergy Saving” can be defined as the amount of exergy required to produce a mass or energy stream. Whilst, the thermoecological cost proposed by Szargut & Morris (1987) almost simultaneously in their “Cumulative Exergy Consumption Theory” can be defined as the cumulative consumption of non-renewable exergy connected with the fabrication of a particular product. Basically, both concepts represent embodied exergy. The Thermoeconomic Input-Output Analysis is in essence the same concept of the Thermoecological Cost, which can be defined as the sum of all resources required to build a product from its component parts, expressed in exergy units. In fact both are isomorphous with the Input-Output analysis (see Appendix B) as described in section 4.3. The Eq. 4.13 from thermoeconomic input-output analysis, expresses the same that Eq. 4.23 from the thermoecological cost. Here, the exergy of the process Piis equivalent to the specific thermoecological cost ρj. Whilst the specific TEC of a raw material or semi-finished product ith of the considered process ρiis analogous to Pj. The exergy of the fuel bf j and the exergy of the mineral bm j in the TEC methodology correspond to the exergy of process icoming from the exergy of environment E0i. Finally, the junction coefficients ri j , are the coefficient of consumption ai j of TEC. Both, the exergy of the process Picalculated through the Thermoeconomic Input-Output analysis and the Thermoecological cost ρj, account for the by-products or wastes obtained in a system. The Thermoecological cost perform this analysis through the coefficients of production of the ith by-product per unit of the jth main product, fi j . The Thermoeconomic InputOutput analysis meanwhile has been applied to develop a method for the allocation of the cost of wastes, presented in Agudelo, Valero & Torres (2011). Besides, both methodologies analyze the emissions related to the use of fossil fuels. The Thermoecological cost has an additional inclusion of the consumption resulting from the necessity of compensation of environmental losses caused by rejection of harmful substances to the environment. The Thermoeconomic Input-Output analyze carbon emissions through the abatement costs (Agudelo, Valero & Uson 2011). The latter methodology allows to perform an exergy decompostion in order to know the contributions of every external resource to each stream of a system, revealing how energy resources are used along productive processes. In the same way, the Thermoecological cost method performs a differentiation between the TEC resulting from fuel or non-fuel mineral resources. However, the TEC can be applied on a regional scale, involving the thermo-ecological cost associated with the imported raw materials and semi-finished products. When the analysis of any system is performed, the matrix system obtained from the Thermoeconomic Input-Output Analysis can be solved using an Excel spreadsheet. Whilst, the system of linear input-output equations derived from Thermoecological Cost Analysis requires a software package used for solution of systems of equations, like EES (Engineering Equation Solver), Matlab, etcetera. 82 4.6 The exergoecology approach and the degraded Earth Thanatia The methodologies presented in the previous sections allow to analyze the path “cradle to grave” in a LCA of mining industry, because processes such as raw material extraction, refining, product manufacturing, recycling and disposal can be evaluated. Nevertheless, the “grave to cradle” path still needs to be assessed for an absolute LCA (see Fig. 4.2). CRADLETOGRAVE Solar energy Thermoecological cost Services of products Exergy Exergy NATURE/CRADLE Resources Life cycle of aproduct Abatement processes Emissions Residues THANATIA/ GRAVE Wastes Effluents Emissions Replacement processes Exergy GRAVETOCRADLE Exergyreplacement cost Figure 4.2: Conceptual diagram of cradle-to-grave and grave-to-cradle methodologies. Valero & Valero D. (2014). The exergoecological approach proposed by Valero (1998), Valero et al. (2003) is used for assessing the exergy of any natural resource from a defined dispersed state of the Earth. Conceptually, the Exergoecological method allows an evaluation of mineral resources through the Exergy Replacement Cost, which is the exergy required to return to the initial state of concentration and composition found in mines, the minerals that have been totally dispersed throughout the crust once their useful lives have come to an end. That is to say, the Exergy Replacement Cost quantifies the amount of exergy that man saves when resources are extracted from a mine, instead of from a pool of materials contained in a hypothetical Earth that has reached the maximum level of deterioration. The dispersed state is assimilated to an Earth where all fossil fuels have been burnt and all mines have been commercially exhausted and dispersed. The model for this “commercial end” of the planet developed by Valero Delgado (2008), is the Crepuscular Earth model of the theoretical Earth state Thanatia. It represents a degraded planet with an exhausted atmosphere, hydrosphere and continental crust. Nevertheless, as opposed to that of the current Earth, the cre83 puscular crust contains no mineral deposits with all non-fuel minerals having been extracted and dispersed and all fossil fuels having been burnt. Valero, Agudelo & Valero (2010) assert that as a consequence, the CO2concentration of the crepuscular atmosphere is much higher than it is at present. Similarly, all water available in the hydrosphere, except for a tiny 2.5% due to the hydrological cycle as it is now, is saline due to all freshwater and saltwater having mixed. The crust model proposed by Valero, Valero & Gomez (2011) is a first approximation of the average mineralogical composition of the upper crust and provides the composition and concentration of the most common 294 minerals currently found on Earth. These concentrations represent the lower limit of ore grades, and once it is achieved, the natural exergy of a mine becomes zero (Fig. 4.3). Although closely related the concept of Thanatia should not be mixed up with that of the well known Reference Environment (R.E.) commonly used for the calculation of chemical exergies. For the calculation of ERC, the concentration of all minerals in a dispersed crust are required, something which is missing in conventional R.E. That said, the Reference Environment is still required for the assessment of the chemical exergy of minerals and in fact Thanatia has chemical exergy with respect to the R.E. (see Fig. 4.3). Figure 4.3: Conceptual diagram of RE and Crepuscular Earth for the evaluation of mineral capital. Valero, Valero & Gomez (2011). In the exergoecology approach, a resource is not only valued according to its mass or chemical composition. Mineral resources are assessed according to its differentiation with the environment. The greater the ore grade of the mineral with respect to the dispersed state, the greater its thermodynamic value. This is because the value of a mineral is related with the amount of exergy needed to return it from the depleted state of Thanatia to the conditions of the mine where it was originally found. Valero & Valero D. (2014) assert that the exergy difference between Thanatia and the mine increases with the mine’s quality (e.g. with its ore grade) and decrease when the mineral deposits become exhausted. At the threshold when all natural resources have been extracted and dispersed, the planet has lost all its natural bonus. Valero, Valero & Gomez (2011) 84 claim that the mining of highly concentrated mines is “penalised” since they have greater associated exergy replacement costs. For instance, mines with high-grade ores do not need so much energy during the mining and concentration processes, however if ores of high exergy content are being depleted, inexorably its exergy replacement costs will be high. On the contrary, mines with low-grade ores require high amounts of energy during the mining and beneficiation processes, but their exergy replacement costs will remain low. And each time a mine is exploited, that exergy bonus gets lost and is therefore unavailable for future generations. According to Valero & Valero (2010a), the exergoecology approach is based on two concepts: exergy and exergy cost. The first is defined as the minimum energy required to produce a natural resource with a specific structure and concentration from common materials in the reference environment, and the second accounts for the actual exergy required for accomplishing the same process with available technology, as shown in Fig. 4.3. By means of these two concepts, the Exergy Replacement Costs (ERC) were preliminary calculated in Valero, Valero & Martínez (2010). The objective of the ERC is thus to determine the exergy bonus that Nature provides by having minerals concentrated in mines instead of having them dispersed throughout the Earth’s crust. It can also be used alongside extraction data, to assess how Nature’s stock is being degraded and dispersed by mankind and at which rate. 4.6.1 Reference Environment As mentioned above, to calculate the chemical exergy of a mineral resource, it is necessary to define a reference environment (RE). Authors such as Ahrendts (1980), Kameyama et al. (1982), Szargut et al. (1988a) have proposed different references environments. The RE must be determined by the natural environment and is fixed by its chemical composition. The differences between standard chemical exergies of the elements obtained from different RE can be very significant. Each RE definition generate different exergies, therefore an appropriate RE is necessary. There are criterion differences between the different RE proposals. For instance, the RE proposed by Ahrendts (1980) is based on the chemical equilibrium, however the natural environment is ar removed from such equilibrium. Kameyama et al. (1982) proposed a reference environment with the criterion of chemical stability, nonetheless the most stable compounds selected by this author are not the more common in the real environment. The RE proposed by Szargut et al. (1988a) is based on partial abundance, this author select the most stable substance among a group of reasonable abundant substances in order to satisfy the “earth similarity criterion”. Further information of these RE and other partial RE has been carried out by Ranz (1999) and Valero Delgado (2008). An appropriate RE to assess natural resources, has been proposed by Valero Delgado (2008). This reference environment is based on that developed by Szargut et al. (1988a) and is an updated of the RE proposed by Ranz (1999). The update takes into account new data of the standard chemical exergy of chemical compounds as well as the gaseous, solid and liquid reference substances. 85 4.6.2 Assessment of mineral resource depletion Exergy of mineral resources The exergy of a mineral resource has at least three components: one associated with its chemical composition (bch), one associated with its concentration (bc) and one associated with comminution processes (bcom). The chemical exergy1bch of the resource can be calculated by means of an exergy balance of the reversible formation reaction, as proposed by Szargut et al. (1988a). bch =X k vkb0 ch,k+∆Gminer al (4.27) Where b0 ch,kis the standard chemical exergy of the elements that compose the mineral, vk is the number of moles of element kin the mineral and ∆Gminer al is the standard normal free energy of formation of the mineral (Gibbs free energy). In a recent survey, Valero & Valero (2012) have obtained a database of enthalpy and Gibbs free energy of formation of minerals in order to calculate thermodynamic properties such as the chemical exergy of about 300 natural substances. The concentration exergy (bc), meanwhile, is the minimum amount of energy associated with the concentration of a substance from an ideal mixture of two components, was defined by Faber (1984) as: bc i =− ¯ RT 0·lnxi+(1−xi) xi ln(1−xi)¸(4.28) where ¯ Ris the universal gas constant (8.314 kJ/kmolK), T0is the temperature of the reference environment (298.15 K) and xiis the concentration of the substance i. The exergy accounting of mineral resources implies to know the ore grade which is the average mineral concentration in a mine xmas well as the average concentration in the Earth’s crust (in Thanatia) xc. The value of x[g/g] in Eq. 4.28 is replaced by xcor xmto obtain their respective exergies, bc(x=xc) and bc(x=xm) whilst the difference between them ∆bc(xc→xm) represents the minimum energy (exergy) required to form the mineral from the concentration in the Earth’s crust to the concentration in the mineral deposits as: ∆bc(xc→xm)=bc(x=xc)−bc(x=xm) (4.29) In the case of concentrating two solids, another term must be added: the variation in cohesion exergy between the final and the initial state. Minerals in the crust are commonly embedded in a sillicate matrix and its cohesion exergy is its comminution exergy (bcom), or minimum exergy needed to comminute the mineral between two given sizes Valero & Valero D. (2012). Considering that the size of the bare rock of the Crepuscular Earth model dθis so large compared to the geometrical mean size dMof the natural fragments found in the mine, the comminution exergy cost (b∗ com) of a mineral in a mine of size dMis defined as: b∗ com =10Wi(1/p(dM)−1/p(dθ)) (4.30) 1Chemical exergy is the work that can be obtained by a substance having the parameters T0and P0to a state of thermodynamic equilibrium with the datum level components of the environment (Szargut et al. (1988b)). 86 Where Wiis the Bond work index [kWh/t][µm]0.5. If the differences in size are not substantial and the mineral is present in the same sillicate matrix, this comminution exergy variation may be neglected in a first approach. Exergy Replacement Cost From the 2nd Law of Thermodynamics, Eq. 4.28 indicates that as the concentration of the substance tends to zero, the exergy required to separate it tends to infinity. So Thermodynamics provides the tendency of the behavior. The real energy required is several orders of magnitude greater than what the Thermodynamics of reversible processes dictates. In fact, mixing and separating are the most irreversible processes. When salt and sugar are mixed, the energy that is liberated in the mixture is almost imperceptible. If the process were reversible, the same amount of energy would be required to separate the mixture. But that is obviously not the case in the real world and when this happens in everyday life, it is easier to throw out the mixture than trying to separate it. Hence, it should be noted that if minerals are assessed solely in exergy terms, the results obtained would be very far removed from “socially accepted” values given to minerals. This is because exergy measures minimum thermodynamic values. Man’s technology is however very far removed from reversibility conditions and this is why we need to resort to the exergy replacement costs, including the so called unit exergy costs (k). The latter factor is defined as the ratio between the exergy consumed in the real mining process (Ereal process) used to obtain the mineral from the ore grade xcto the commercial grade xrand the minimum exergy (∆bminer al ) required to accomplish the same process, expressed as: km=Erealpr ocess ∆bmineral (4.31) Accordingly kis a measure of the irreversibility of man-made processes and amplify minimum exergies by a factor of ten to several thousand times, depending on the commodity analyzed. Consequently, the Exergy Replacement Cost of a mineral would require ktimes the minimum exergy, and can be calculated with the following expression: b∗ t=kch ·bch +kc·bc+b∗ com (4.32) The exergy replacement cost is defined as the total exergy required to concentrate the mineral resources from Thanatia, with the best available technologies. Therefore, these are not absolute and universal values, as opposed to property exergy. Domínguez & Valero (2013) argue that the exergy costs are a function of the extraction and separation technologies, which in turn vary with time, with the type of mineral analyzed, and with man’s ability to extract it, i.e. with its learning curve. See Section 3.4. For most case studies, the chemical component of Eq. 4.32 is zero because there is no need to chemically produce the mineral from Thanatia again, since the Crepuscular Earth model already contains the substance, but at a significant lower concentration. Whilst, the comminution exergy cost of a mineral can be neglected if the differences in size are not substantial, as considered in the analyzes carried out in this paper. Therefore, one only needs to account for the exergy required to concentrate the mineral from the dispersed conditions found in Thanatia, until the original concentration found in the mines. As aforementioned, in Eq.4.33 the unit exergy cost kis the ratio between the real cumulative energy required to accomplish the real process to concentrate the mineral from the ore grade 87 xmto the commercial grade xrand the minimum thermodynamic exergy required to accomplish the same process as, it can be expressed also as: k=E(xm→xr) ∆b(xm→xr)(4.33) The general procedure to calculate the exergy replacement cost of mineral resources is presented in Fig. 4.4. Calculation of the theoretical exergy of the mining and concentration process from the ore grade xm to the refining conditions xr b(xm->xr) Assessment of the real mining and concentration energy xm->xr) Calculation of unit exergy costs k(xm)= E(xm->xr)/b(xm->xr) Extrapolation of unit exergy costs to xc k(xc) Calculation of the theoretical exergy of the concentration process from the crespuscular concentration xc to the mine conditions xm b(xc->xm) Calculation of the exergy replacement costs (mineral exergy bonus) b*= k(xc)b(xc->xm) Figure 4.4: Calculation procedure for obtaining a mineral’s exergy replacement costs. Valero & Valero D. (2014). The energy required for mining is a function of the ore grade of the mine and of the technology used, thence it can be defined as the unit exergy cost. Moreover, both variables have an opposite effect on the energy used, as Ruth (1993,1995a,b,c) states. The lower the ore grade, the more energy is required for mining. On the contrary, technological development usually improves the efficiency of mining processes and hence, decrease the energy consumption. In other words, the unit exergy cost depends on the ore grade (xm) and the time (t) that is considered through the improvements in mining techniques, which in turn are reflected in the real energy consumptions. k=k(x,t) (4.34) 88 Hence, the temporal function kis only definable for the past and for each particular mineral. It is therefore difficult to extrapolate it towards the future for the practical impossibility to predict changes in the scientific and technological knowledge that will eventually appear. The second problem with kis that it is not a continuous function. The technology applied can also vary with the concentration ranges of a particular deposit. And in turn, each mining technique (i.e. underground or open-pit mining), has a particular effect on the energy consumption due to different factors such as ore grade, grind size, nature, depth and processing route. These factors have been analyzed in Norgate & Haque (2010), Norgate & Jahanshahi (2010,2011), for different commodities such as for copper, nickel aluminium and iron through the life cycle assessment methodology. Bearing in mind these limitations and the kind of data available for mining (which is usually very scarce) in this survey it is assumed that the same technology is applied for the range of concentration between the ore grade xmin the mine and the refining grade xr, than between the dispersed state of the crepuscular crust xcand xm. This way, an analysis of the average energy vs. ore grade trends for different minerals is carried out, in order to calculate the corresponding unit exergy cost values and extrapolate them to ore grades equal to those of the dispersed conditions of Thanatia. 4.7 Summary In this chapter a brief review of the main thermoeconomic concepts was presented, in order to introduce the concepts that allow to assess natural resources. In this regard, two methodologies: Thermoeconomic Input-Output and Thermoecological Cost were explained and compared. Both methods will be applied in several case studies of mining and metallurgical industry in Chapters 5and 6of this PhD. These methods are used to analyze the path “cradle to grave” in a LCA of mining industry, because processes such as raw material extraction, refining, product manufacturing, recycling and disposal can be evaluated. Afterwards, the Exergoecology approach and the exergy replacement costs were presented as tools to evaluate the “grave to cradle” path. In this way, the cycle of natural resource assessment is closed and an absolute LCA “cradle-to-cradle” can be performed. The importance of the exergy replacement costs lies in their ability to establish a scarcity factor of mineral resources which should be accounted for, when assessing the sustainability of mining and metallurgical processes. Accordingly, in next chapter an exergy analysis of mineral resources and metallurgical systems will be carried out, and the exergy replacement cost of several metals will be calculated. Additionally, the problem of cost allocation when two or more commodities are produced in the same mining operation will be settled. 89 y"="138.8x)0.28" 0" 100" 200" 300" 400" 500" 600" 700" 800" 900" 0.00" 0.05" 0.10" 0.15" 0.20" 0.25" 0.30" 0.35" Concentra)on*Energy*[GJ/t]* Ore*Grade[%]* Real"data"set" Concentra=on"Energy"[GJ/t]" Figure 5.5: Trends of unit exergy costs and concentration energy in uranium mining (data referred to uranium oxide U3O8). Adapted from Mudd & Diesendorf (2008). 5.2.6 Summary of ERC obtained A summary of all commodities analyzed previously is presented in Table 5.1. It contains the main values for each substance, such as xc,xmand xralong with the equations to calculate the energy required to mining and concentrating a specific ore, the unit exergy cost and the exergy replacement cost. Table 5.1: Exergy Replacement Costs. Values of xc,xmand xrare referred to the assumed mineral that represents the ores from which the metal is extracted. E(x)[G J/t]xc[g/g]xm[g/g]xr[g/g]k(x=xc)k(x=xm) Exergy Replacement Cost Cobalt (Linnaeite) E=2.24x−0.64 5.15E−09 1.90E−03 4.56E−02 10,872 Copper (Chalcopyrite) E=23.81x−0.35 6.64E−05 1.67E−02 8.09E−01 525 170 110 Gold E=135,664x−0.285 1.28E−09 2.24E−06 1.38E−04 6,380,357 2,135,879 583,668 Nickel (Sulphide - Pentlandite) E=17.01x−0.67 5.75E−05 3.36E−02 4.68E−01 13,039 585 761 Nickel (Laterite - Garnierite) E=2.11x−0.5 4.10E−06 4.42E−02 8.04E−02 876 136 167 Uranium (Uraninite) E=138.8x−0.28 1.51E−06 3.18E−03 7.50E−01 13,843 3,697 901 As stated before, the unit exergy cost is the ratio between the real energy required for mining and concentrating a substance and the minimum thermodynamic energy (exergy) required to achieve the same process. Hence, it provides a measure for the irreversibility (or technological ignorance) of the process. Generally, the energy value (nominator) increases more rapidly than the exergy one (denominator). An exception of this fact is found for elevated ore grades, where the opposite happens. Consequently, the lower the ore grade, the more energy is required for mining and concentrating the mineral and the more irreversible is the process. 97 For instance, gold has the highest unit exergy cost value associated to the deposits (when x=xm) of the metals analyzed in this work, attributable to its low concentration in mines and the consequent amount of energy needed to concentrate it. Besides, the actual ore grade in mines is close to that in crepuscular Earth’s crust. The opposite example is copper which has a lower kvalue ascribed to its actual high ore grade in mines. But the state of technology plays also an important role. This fact is highlighted with nickel from sulphide ore. Even if its ore grade is similar to that of copper, the elevated value of kis an indicator of the significant irreversibility of the production process. A particular case is that of nickel and its ores. Historically, the metal was likely obtained from sulphide ores due to the major energy requirement of laterites in the refining process. Nevertheless, more Ni resources are in the form of laterites than of sulphides. But focusing only in the concentration energy, sulphide ores have larger concentration requirements than lateritic ores, as revealed by the larger unit exergy costs. However, continuous sulphide ore grade decline, the increasing cost of underground mining coupled to the estimations that in the future nickel production will be obtained from lateritic ores in order to fulfil the nickel demand, augur an increase of the energy associated with nickel production. Of special interest is the value of k(x=xc). This value multiplied by the minimum exergy required to concentrate the mineral from xcto xmrepresents the amount of energy required to mine and concentrate a substance from the bedrock (from Thanatia) to the current conditions in the mineral deposits and provides a measure of the exergy that Nature provides “for free” thanks of having minerals concentrated in mines instead of dispersed throughout the crust. The value of the crepuscular unit exergy cost of the different minerals is always greater than that of the current mineral deposits. The difference increases generally with the separation between the crepuscular grade and the average grade in the deposits. For instance, the crepuscular kvalue of gold is in the same order of magnitude than the mine k-value, because the grade of gold is close to that found in Thanatia. The opposite happens with uranium, which has a small crepuscular grade compared to current average ore grades. Therefore its crepuscular k-value is considerably much larger compared to its unit exergy cost when the average ore grade in U-mines is taking into account. Finally it is worth to note in Table 5.1 that the energy consumption as a function of the ore grade shows expression varying from x−0.2 to x−0.9. From this observation and recognising that empirical data for most of the minerals produced in the world is very limited, it is proposed the following general expression for the exponential curve applied to estimate the energy consumption as a function of the ore grade: E(xm)=A·x−0.5 m[xm, metal concentration %] (5.1) Coefficient A is determined for each mineral since generally, the average ore grade xmand the energy required for concentrating and extracting the mineral at that grade E(xm) is known. It should be noted that xmvalues are expressed in Eq. 5.1 as mass percentage of the element under consideration. This is a very rough approximation, but it is more in agreement with actual mining behaviour than the equation proposed by Chapman & Roberts (1983), where the energy is inversely proportional to the ore grade. The same analysis was carried out for several commodities, which will be used in further chapters. Hence, Table 5.2 summarize the exergy replacement cost for different commodities. It can be observed that a mineral has a high exergy replacement cost when 1) the concentration 98 of the mineral in the crepuscular crust (in Thanatia) is low and the difference between the ore grade of the current mines (xm) is high and/or 2) when the energy required to extract the mineral and beneficiate it is significant. The exergy replacement cost is greater for critical elements (e.g. gold) than it is for those found in abundance (e.g. chromium). Table 5.2: Exergy Replacement Costs for different commodities. Energy and exergy values are expressed in GJ/ton of mineral. Mineral xc[g/g]xm[g/g]xr[g/g] Concentration Smelting Exergy and beneficiation and refining replacement energy energy cost Aluminium (Gibbsite) 1.38E−03 7.03E−01 9.50E−01 10.55 23.87 627 Antimony (Stibnite) 2.75E−07 5.27E−02 9.00E−01 1.4 12 474 Cadmium (Greenockite) 1.16E−07 1.28E−04 3.86E−03 1.4 263.9 278.5 Chromium (Chromite) 1.98E−04 6.37E−01 8.10E−01 0.01 36.3 5 Iron (Hematite) 9.66E−04 7.3E−01 9.5E−01 0.7 13.4 18 Lead (Galena) 6.67E−06 2.37E−02 6.35E−01 0.9 3.3 37 Manganese (Pyrolusite) 4.90E−05 5.00E−01 6.71E−01 0.2 57.4 16 Molybdenum (Molybdenite) 1.83E−06 5.01E−04 9.18E−01 136 12 908 Phosphate rock (Apatite) 4.03E−04 5.97E−03 9.00E−01 0.3 4.6 0.4 Silver (Argentite) 1.24E−08 4.27E−06 9.00E−01 1281.4 284.8 7371.4 Zinc (Sphalerite) 9.96E−05 6.05E−02 7.90E−01 1.5 40.4 25 Rare Earth Metals Cerium (Monazite) 1.03E−04 3.00E−04 8.0E−01 523.05 56.1 43.47 Gadolinium (Monazite) 1.03E−04 3.00E−04 9.5E−01 3607.25 43.05 115.23 Lanthanum (Monazite) 1.03E−04 3.00E−04 8.0E−01 296.75 53.45 16.77 Neodymium (Monazite) 1.03E−04 3.00E−04 8.0E−01 591.7 53 31.88 Praseodymium (Monazite) 7.10E−06 3.00E−04 8.0E−01 296.28 55 124.37 Yttrium (Monazite) 1.03E−04 3.00E−04 9.5E−01 1198.25 47.6 32.01 5.3 Introduction to Exergy Cost Allocation of by-products in the mining industry The long-term availability of mineral resources is a key factor to satisfy human activities, technology and economic activity, including those metals that do not generally are the primary production of mines, such as copper or nickel, but instead are mined as by-products during the mining of primary ores. In this regard, mining industry is confronted with the difficult and often rather complicated problem of assigning costs to their by-products and joint products, which have highly complex demand/supply and technology and investments requirements. The availability of by-product metals depend on the availability of technology to recover those metals during or after the processing of host metal ores, as well as on the economic profit of by-product metal recovery. Mudd et al. (2013) proposed a set of parameters in order to evaluate by-product metal availability such as: (1) size and type of the by-product metal ore bodies; (2) the characteristics abundances of by-product metals in the host or hosts; (3) the typical recovery efficiencies for these by-product metals (mainly the different technologies used) and (4) models of the relative costs and benefits of by-product metal recovery. However, in Mudd’s proposed parameters there is a missing one related with energy consumption of by-product metals. In this respect, conventional LCA software usually performs allocations among products when one or more by-products come about in a mining or metal99 lurgical process, based either on tonnage or on revenue (commercial prices). Both ways of allocating environmental costs entail many disadvantages, such as introducing subjectivity with price or underestimating burden for certain by-products when the tonnage is low. Specifically in the Ecoinvent database, the by-product allocation problem (the joint production of silver and lead, for instance) is undertaken by a subdivision of the sub-processes. The starting point for the estimation depends on the general profit expectations of the company, considering an arbitrary performance value of 10%. Hence, the allocation factors are based on revenue but these values are corrected by mass in order to keep up with the resource balance in the final commodity. Consequently, allocation methods are confronted with the difficult and often rather complicated problem of assigning costs to their by-products and joint products. Furthermore, LCA uses computer software and database like Ecoinvent (Classen et al. 2007), which contains a very broad data on energy supply, resource extraction, raw material supply like chemicals, metals, explosives or water. However, most of such databases consider regional averages of environmental impacts associated with mineral processing, with arbitrary allocations among those operations using raw materials, hydro, gas, coal or nuclear generated electricity, or those occurring in different countries, or even any variations from process to process. In this way, authors like Yellishetty et al. (2009) have conducted a critical review of existing LCA methods in the minerals and metals sector in relation to allocation issues related to indicators of abiotic resource depletion, concluding that LCA issues of minerals and metals need to be investigated further to get more understanding, to facilitate the future use of LCA as a policy tool in the mining sector and increase objectivity with more scientific validity. There is clearly a need for better data related to mining industry such as, energy consumption, production efficiency at various levels of aggregation, waste disposal, fugitive emissions, among others. Authors such as Brown et al. (2012), McLellan et al. (2012) think that a global perspective on the current energy use and GHG emissions proceeding from minerals production is required. Espí (2009), Sagar & Frosch (1997) assert that this kind of information is crucial to perform analysis which allows a deeper understanding of mining sector. Since mineral extraction and processing constitutes the first stage of any given process, using conventional LCA software may lead to incorrect results. Aware of this problem, in this section, the second objective will be achieved through the development of a new cost allocation methodology based on the exergy replacement costs. The final aim is to estimate the energy consumption and eventually economic, technology and environmental bearings of each commodity produced. The novelty introduced with respect to what is already being done in conventional Thermoeconomic analysis is that when non-fuel minerals come into play, allocation is carried out through the exergy replacement costs instead of through chemical exergy, what was traditionally done in conventional thermoeconomic analyzes. In this way, the scarcity factor of minerals is taken into account. In order to allocate costs among non-fuel minerals through the exergy replacement costs, 33 different mineral deposit models where 12 coupled products are obtained have been analyzed. Additionally, as study cases, exergy cost allocation was applied to nickel, copper, lead and REE production with its respective by-products. The mineral deposits models used to develop the allocation methodology in this section, are taken from the comprehensive study of average ore grades accomplished by Cox & Singer (1992). In their study, a compendium of geologic models was presented, including 85 descriptive models identifying attributes of the deposit type and 60 grade-tonnage models giving estimated pre-mining tonnage’s grades from over 3,900 well-characterized deposits all over the world. The average grade xmof the different mineral deposits analyzed is calculated with Eq. 5.2, taking into account the tonnage (M) and ore grade (xm) of each model and the number of 100 deposits containing the mineral under consideration. Tables 5.3 through 5.9 are calculated with the mean average grade and tonnage of each deposit type. xm=RM 0xmdM RM 0dM (5.2) Such models will serve to demonstrate why the allocation model presented in this thesis is more suitable than conventional approaches using tonnage or market prices. 5.4 Joint products and by-products in the mining industry Products produced at the same time are classified as joint products or by-products; generally driven by the importance of the different products to the viability of the mine. The same metal may be treated differently based on differing grades and quantities of products. The decision as to whether these are joint products or one is only a by-product is important as it may affect the allocation costs. Joint products are metals or minerals within an ore body which each have significant relative sales values. Hansen et al. (2009) state that they are produced simultaneously by the same process up to a “split-off” point and from a common raw material source. Joint products are related to each other such that one joint product cannot be produced without the other. Furthermore, an increase in the output of one increases the output of the others, although not necessarily in the same ratio. Up to the often called split-off point, it cannot be obtained more of one product without getting more of the other(s). The split-off point is the point at which the joint products become separate and identifiable. By-products are metals or minerals within an ore body which have minor sales values when compared with the principal product or products. Is a secondary product recovered in the course of production or processing of a primary product. For instance, by-products resulting from scrap, trimmings, and so forth, of the main products. According to Mudd et al. (2013), these metals are known also as companion metals (e.g., cobalt, molybdenum, rhenium, selenium, germanium, gallium, tellurium and indium). Although these metals often have economic and technological importance, the economic driver for mining here is undoubtedly the major metal. Besides, there are some groups of metals such as PGMs or REE that may occur as “coupled elements” without a real carrier metal. Otherwise, some metals generally produced as by-products may also be mined as target metals on their own if they occur in elevated concentrations (e.g. cobalt, bismuth, molybdenum, gold, silver, PGMs and tantalum) or if demand exceed the supply available from byproduct and joint product output. As previously explained, more than one metal is commonly produced by the same mining and refining processes. Metals such as lead and zinc are commonly found together; silver is often found with gold. These are only two examples of the many joint products that Nature provides. Figure 5.6 depicts that each carrier commodity metal is associated in Nature (geology) by a distinctive mix of valuable minor elements. The latter has led to metallurgical processing being sharpened to effectively recover most elements economically. This complex link of materials can be observed in the “Metal Wheel”, in which each element has a unique position. Therefore, eliminating or disrupting the production of one element will have an effect over all 101 other connected elements Verhoef et al. (2004). Figure 5.6: The metal wheel. Verhoef et al. (2004) 5.5 Approaches to allocating joint cost As aforementioned, the Exergy Life Cycle Assessment provides average exergy costs of all manufactured products as well as the exergy needed for their use, maintenance, repair and disposal, it is focused in obtaining round numbers which could be used as ecological indexes of sustainability. However, the problem comes up when two or more products, by-products and residues are produced simultaneously. How to allocate costs? As stated by Valero & Lozano (1994), it is necessary to look inside the system in order to understand the process of cost formation, by identifying the internal relationships of all the structure components. Indeed, the main problem of allocating costs has been to find a function that adequately characterizes every one of the internal flows in a system and distributes cost proportionally. This function needs to be universal, sensitive and additive. That is, it needs to have an objective value for every possible material manifestations, it needs to be able to vary when these manifestations do so and each internal flow property needs to be represented additively. There is a wide international consensus that the best function, at least for energy systems, is exergy, which can contain in its own analytical structure the flow history. Therefore three conditions are needed to allocate costs. First, the definition of the boundaries of the system. Second, a structure of the system in which all the components or processes are described in terms of black boxes interacting to each other through energy flows (or more generalized: energy, economic or information flows). Third, the definition of the purpose of 102 production for each and every component. The purposes of cost accounting could be stated in broad terms as: determination of the actual cost of products, settlement of a rational basis for pricing products and/or evaluation of their profitability as well as means for controlling costs. Accordingly, cost allocation of mineral resources is not a simple task, due to the fact that an allocation method must allocates the cost on as reasonable basis as possible. As reported on PWC (2012), a systematic and rational basis of cost allocation should be applied when the conversion costs of a product are not separately identifiable. Joint costs are the total of the raw material, labor, and overhead costs incurred up to the initial split-off point. Joint cost allocation is based on the proportional values of the products at the split-off point. Whilst, separable costs are those costs incurred after the split-off point; they can be easily traced to individual products. For instance, if ore contains both iron and zinc, the direct material itself, is a joint product. Since neither zinc nor iron can be produced alone prior to the split-off point, the related processing costs of mining, crushing, and splitting the ore are also joint costs. The variety of products and the mutually beneficial costs, appear either because the material itself is a joint product or because processing results in the simultaneous output of more than one product. By-products are not usually allocated any of the joint costs. Some mining entities assign to by-products only the costs of processing after the split-off point to by-products, because byproducts are processed beyond the split-off point to bring them to a marketable form or to increase their value above their selling prince at the split-off point. Furthermore, the accounting treatment of by-products necessitates a reasonably complete knowledge of the technological factors underlying their manufacture, since the origins of by-products may differ. As mentioned above, cost are either separable or not. Separable cost are easily traced to individual products and offer no particular problem. If not separable, they must be allocated to various products. All joint products benefit from the entire joint cost. The objective in joint cost allocation is to determine the most appropriate way to allocate a cost that is not really separable. Although, there are variety methods for allocating costs to joint and by-products, they can be associated to two approaches, using market-based data or using physical measures. However, the physical measure of the individual products may have no relationship to their respective revenue-generating capacity. For example, in a gold mine that extracts ore containing gold, silver, and lead, the use of a common physical measure (tons) would result in almost all costs being allocated to lead, the product that weighs the most but has the lowest revenue-generating power. The physical method of cost allocation is inconsistent for this case, with the main reason that the mining company earns revenues from gold and silver, not lead. Thereby, Hansen et al. (2009) assert that there is no well-accepted theoretical way to determine which product incurs what part of the joint cost. In this thesis, three different allocation methods are reviewed: • Through Tonnage: For this allocation, the tonnage of each of the product and by-product is considered. External resources such as energy and raw materials are divided into the products according to its mass value. • Through Market Price: The assignment based on price, considers the market price of each product obtained in the process. 103 • Through Exergy Replacement Cost: In this allotment, the exergy replacement cost of each element is taking into account. This allocation method introduces the depletion factor of mineral resources. 5.5.1 Tonnage Allocation of by-products Under the physical units method, joint cost are distributed to products based on some physical measure of the joint products at the split-off point. However, cost allocation based on tonnage may be inappropriate if there is a significant difference between the relative sales values of the joint products, such as in a mine producing lead and silver. The costs allocated to the lower value product may exceed its net realisable value whilst the higher value product would result in “super” profits, as stated by Hansen et al. (2009). Through tonnage allocation, the majority of environmental impacts would be ascribed to the most abundant product mined. The issue with this is, although based on physical phenomena, such a procedure does not reflect the physical value of minerals. To exemplify, a tonnage allocation was done using the data of 33 mineral deposit models. Results presented in Tables 5.3 –5.9, show that allocation based only on tonnage is not very suitable, due to the underestimating charges for certain by-products when the tonnage is low. For instance, Table 5.3 depicts the cost allocation for three mineral deposits developed by Cox & Singer (1992), where copper, gold and silver are found. It can be observed that cost allocation based on tonnage is not a reliable decision, because all cost would be attributed to copper, as in tonnage terms it is the most significant. Otherwise, the amount of silver and gold is very small, and consequently the cost associated with these metals would be almost neglected. Table 5.3: Cost allocation of Cu-Au-Ag deposits as a function of tonnage, price and ERC. Deposit Porphyry Cu-Au Cu skarn Epith. quartz-alunite Au type Ton. Price ERC Ton. Price ERC Ton. Price ERC [%] [%] [%] [%] [%] [%] [%] [%] [%] 1980 2006 1980 2006 1980 2006 Copper 99.96 56.6 81.3 70.3 99.86 43.3 73.3 46.1 98.9 3.1 9.5 5.3 Gold 0.01 38 17.4 28.2 0.01 40.3 21.7 50.0 0.3 90 86.7 92 Silver 0.03 5.4 1.4 1.5 0.13 16.3 5 3.9 0.7 6.9 3.8 2.7 5.5.2 Market Price Allocation of by-products Metal price fluctuations depend on macroeconomic variables like industrial production, consumer prices, interest rates, stock prices and exchange rates, according to Labys et al. (1999). Kriechbaumer et al. (2014) assert that metal prices are the result of complex market and economic dynamics resulting in a great price fluctuation. Metal prices have been analyzed in several surveys by Brunetti & Gilbert (1995), Chen (2010), Dooley & Lenihan (2005), Figuerola & Gilbert (2001), McMillan & Speight (2001), Roberts (2009). Gleich et al. (2013) claim that in many cases commodity prices double or even triple within only few years. Humphreys (2013) argue that the price boom that started around 2004 has led to politicise metal supply. However, authors like Gordon & Tilton (2008), assert that long-run trends in real prices of a mineral commodity provide a better indicator of trends in availability than physical measures indicating how much is left in the ground. This is because trends in real mineral-commodity 104 prices can be upward or downward depending largely on whether new technology with the assistance of mineral substitution offsets the cost-increasing effects of depletion. Furthermore, Gordon & Tilton (2008) argue that if environmental impacts of mining industry are gradually included into cost structure of metal production (for instance, through carbon taxes) the price of metals will increase. The most common allocation methods used to share out joint production costs to joint products at the split-off point, based on market data include: the sales value at split-off method, the net realizable value method, the constant gross-margin percentage method. These methods are preferred over the physical measures, because revenues are, in general, a better indicator of benefits received (PWC 2012). For instance, mining companies receive more benefits from 1 ton of gold than they do from 10 tons of copper. Besides, the physical measure of the individual products may have no relationship to their respective revenue-generating capacity. For example, in a gold mine that extracts ore containing gold, silver, and lead, the use of a common physical measure (tons) would result in almost all costs being allocated to lead, the product that weighs the most but has the lowest revenue-generating power. The physical method of cost allocation is inconsistent for this case, with the main reason that the mining company earns revenues from gold and silver, not lead. In this section a market price allocation was done taking into account the unit value ($/ton) of the different commodities published in U. S. Geological Survey (USGS) from year 1900 to 2011. Chen (2010) asserts that metal prices rise to a peak in the 1980s, and in recent times, the price boom started around 2004, as stated by Humphreys (2013). Therefore, Tables 5.3 –5.9 show the allocation factor for the year 1980 and 2006. Results depict great variations among years, commodities, by-products and models. The latter supports the idea that despite what price allocation supporters claim, market price is not a suitable indicator to assess mineral resources, because of the huge price volatility. Clearly therefore, resource consumption and greenhouse gas emissions need to be related to physical expenditures, not prices. 105 Table 5.4: Cost allocation of Cu-Au-Pb-Ag-Zn deposits as a function of tonnage, price and ERC. Deposit Zn-Pb skarn Polymet. Replacement Polymetallic vein type Ton. Price ERC Ton. Price ERC Ton. Price ERC [%] [%] [%] [%] [%] [%] [%] [%] [%] 1980 2006 1980 2006 1980 2006 Copper 4.79 5.9 9.2 11.9 2.49 2.3 4.9 5.2 1.58 0.6 2.2 1.9 Gold 0.0005 5.1 2.5 6.2 0.001 6.2 4.3 4.0 0.001 1.7 2 3.3 Lead 33.54 17.3 15.9 27.7 54.82 20.9 26.6 37.7 74.58 12 25.6 30 Silver 0.12 43.7 12.4 19.8 0.21 56.4 22.1 29.0 0.72 82.3 54 58.4 Zinc 61.55 28 59.9 34.4 42.47 14.2 42.1 19.8 23.12 3.3 16.2 6.3 Deposit Cyprus massive sulfide Creede epith. vein Comstock epith. vein type Ton. Price ERC Ton. Price ERC Ton. Price ERC [%] [%] [%] [%] [%] [%] [%] [%] [%] 1980 2006 1980 2006 1980 2006 Copper 65.54 51.6 68.5 67.8 6.33 3.9 9.5 8.5 27.7 0.2 0.7 0.42 Gold 0.004 25.9 10.9 20.4 0.004 24.4 18.9 31.8 1 65.6 76.2 83.2 Lead 2.05 0.7 0.5 0.7 53.77 14.0 20 24 13.8 0.04 0.1 0.07 Silver 0.05 12.3 3.0 3.6 0.26 48.7 21.4 23.8 15.9 34 22.4 16.2 Zinc 32.36 9.4 17.1 7.5 39.64 9.1 30.1 12 41.5 0.1 0.6 0.14 Deposit Kuroko mass. sulfide type Ton. Price ERC [%] [%] [%] 1980 2006 Copper 26.13 30.3 38.9 45.9 Gold 0.002 16.6 6.8 15 Lead 15.55 7.6 5.7 9.1 Silver 0.06 20.6 4.8 7 Zinc 58.26 25 43.8 23 Table 5.4 shows the cost allocation for seven mineral deposits developed by Cox & Singer (1992), where copper, gold, lead, silver and zinc are found. In general, the high prices of gold and silver may drive the exploitation of the other commodities. However, it can be observed that cost allocation based on market price, not always attributes the cost to the most expensive commodity (e.g. gold). This is because the quantity of the produced metal is also considered. 106 0%# 10%# 20%# 30%# 40%# 50%# 60%# 70%# 80%# 90%# 1900# 1902# 1904# 1906# 1908# 1910# 1912# 1914# 1916# 1918# 1920# 1922# 1924# 1926# 1928# 1930# 1932# 1934# 1936# 1938# 1940# 1942# 1944# 1946# 1948# 1950# 1952# 1954# 1956# 1958# 1960# 1962# 1964# 1966# 1968# 1970# 1972# 1974# 1976# 1978# 1980# 1982# 1984# 1986# 1988# 1990# 1992# 1994# 1996# 1998# 2000# 2002# 2004# 2006# 2008# 2010# Silver# Zinc# Lead# B*Ag# B*Zn# B*Pb# Pb,$B*$ Zn,$B*$ Ag,$B*$ Pb,$$$ Zn,$$$ Ag,$$$ Figure 5.12: Cost allocation of Pb-Ag-Zn deposits as a function of price and ERC. 5.7 Exergy cost allocation applied to mining and metallurgical processes Once the cost allocation has been done for specific deposits, the methodology can be applied to mining processes in order to allocate inputs of raw materials, utilities, emissions and so on. For this kind of allocation methods, it is necessary a detailed knowledge of the process involved in the commodities production. Data such as energy (fuel and electricity), ore and raw material mass, in each production stage is needed to perform the analysis. In this section, the analysis of four case studies of mineral production is carried out. Due to lack of information about real processes, the case studies used in this section were proposed from information of different sections of Ecoinvent database (Classen et al. 2007) and do not represent a specific mining operation. A detailed analysis of a metallurgical process will be carried out in the following chapter. Hence, in this chapter only theoretical and simplified case studies are presented as a way to show how the allocation procedure works in the mining industry. The first case involves copper production and its by-products: molybdenum, silver and tellurium. The second case assesses nickel production and its by-products: copper and cobalt. The third case analyzes lead production and its by-products: zinc, silver and cadmium. The fourth case evaluates the rare earth elements: lanthanum, cerium, praseodymium, neodymium, gadolinium and yttrium. 5.7.1 Exergy Cost Allocation of Copper and its by-products Copper is always associated with other metals, mainly nickel, molybdenum and platinum group metals. In this case, copper is extracted jointly with molybdenum, silver and tellurium. In order 113 to assess the production process of copper, a Thermoeconomic Input-Output analysis (considering ERC as explained in Chapter 4) was performed. Data of raw material and utilities (e.g. electricity, natural gas and fuel inputs) used during the production process were taken from Ecoinvent database Classen et al. (2007). According to the copper production process shown in Fig. 5.13, molybdenum is obtained right after the mining and beneficiation operations. Pretreatment, reduction and refining processes follow and by-products tellurium and silver are obtained from this second stage. Once the material flows have been identified, exergy cost allocation of inputs of electricity or natural gas among the products of the considered process is performed as depicted in Fig.5.14. In the same way, all inputs such as chemicals, raw materials and utilities are allocated. Figure 5.13: Exergy Cost allocation of Cu and its by-products Mo, Ag and Te. 114 Finally, a thermoeconomic analysis of the process allows to obtain the exergy cost of each of the commodities obtained, which in turn can be converted into monetary units considering the electricity price. Accordingly, one can compare the market price of the commodities with that considering the exergy costs required to produce them. Obviously market prices do not only take into account production costs, but also include other factors alien to the physical reality of commodities. In addition, prices do not take into account the fact that with extraction and the further dispersion of materials, the mineral heritage of the Earth is being destroyed and becoming thus unavailable for future generations. Table 5.10 depicts the results obtained for the copper example. Amount of product obtained, unit price (market price of the commodity for the year 2011 according to U. S. Geological Survey (USGS)), exergy cost and their respective allocation factors (a.f.), as well as an additional allocation factor based on the chemical exergy, are shown. Table 5.10: Allocation factors of copper and its by-products Mo, Ag and Te Commodity Output Exergy cost Exergy Cost Unit price Ton. Price bch product [kg] a.f. [%] [MJ/kg] [$/kg] a.f. [%] a.f. [%] a.f. [%] Copper_1 97 44 99.6 98.5 99.7 Copper_2 1 89 186 9 99.956 94 99.99457 Molybdenum 0.00411 3 545 34 0.4 1.5 0.3 Tellurium 7.17E-06 8 73,324 349 0.040 5 0.00004 Silver 3.26E-04 2 6,419 1,130 0.005 1 0.00539 Figure 5.14 shows the commodity prices vs. the units of exergy required to produce 1 kilogram of metal. It can be seen that costs based on ERC show higher values respect the unit price value. Accordingly, a critical commodity such as tellurium, should have a higher unit price, in order to reflect properly its physical condition in Nature. 0" 200" 400" 600" 800" 1,000" 1,200" 1" 10" 100" 1,000" 10,000" 100,000" Cu Mo Te Ag Unit%price%[$/kg]% Exergy%Cost%[MJ/kg]% Exergy"Cost" Unit"price" Figure 5.14: Cost allocation of Cu by-products, based on price and exergy. 115 5.7.2 Exergy Cost Allocation of Nickel and its by-products Since nickel is always associated with other elements, its production generates other metals like copper and cobalt. A detailed study of the nickel production process is performed in Chapter 6. Thence, the exergy analysis required to obtain the exergy cost used in this section is taken from section 6.1.1. The production system for nickel production is shown in Fig. 5.15. The processing operations have been grouped into four general stages: 1) Mining and Beneficiation and 2) Drying, Roasting and Smelting, 3) Converting and 4)Electrolysis. Figure 5.15 depicts the input-output of utilities (e.g. electricity, natural gas, fuel oil and coal) and raw materials (e.g. ores from which nickel is produced). Figure 5.15: Exergy Cost allocation of Ni and its by-products Cu and Co. 116 Nickel by-products are obtained together with nickel at the last step, in the amounts indicated in Table 5.11. Two issues are shown in Fig. 5.15. First, at mining and beneficiation step (j=1), an input of 1.505 MJ/kg of fuel oil is required. Accordingly, the ERC of minerals involved in this operation are depicted, and the exergy cost allocation factor is calculated (taken into account the mass output of each mineral). This way, the input can be distributed to each mineral in an objective way. Second, the following steps (j=2, j=3, j=4) show all inputs of natural gas, electricity and fuel oil for each stage. Considering the exergy cost allocation, calculated previously in stage 1, these input are alloted to each mineral. This will be the general procedure to analyze any metallurgical system, where main product and by-products are obtained. Table 5.11 shows the variables considered to obtain the exergy cost allocation of each commodity, such as: the output product, the exergy cost allocation factor obtained from exergy replacement cost and the amount of mineral mined, the exergy cost calculated from a thermoeconomic input-output analysis and the unit price provided by U. S. Geological Survey (USGS) in the year 2011. Table 5.11: Allocation factors of nickel and its by-products Cu and Co. Commodity Output Exergy cost Exergy Cost Unit price Ton. Price bch product [kg] a.f. [%] [MJ/kg] [$/kg] a.f. [%] a.f. [%] a.f. [%] Nickel 1 82 1,534 23 65.57 82.2 67.3 Copper 0.515 6 53 9 33.77 16.5 32.2 Cobalt 0.01 12 1,906 36 0.66 1.3 0.4 Figure 5.16 depicts the exergy cost calculated through the thermoeconomic input-output analysis vs. unit price of metal. It can be observed that exergy cost based on ERC and unit price, gives the same appreciation for the three commodities. In first place cobalt is highlighted by price and by exergy content, in second place is nickel and in third place copper. 0" 5" 10" 15" 20" 25" 30" 35" 40" 0" 250" 500" 750" 1,000" 1,250" 1,500" 1,750" 2,000" Ni Cu Co Unit%price%[$/kg]% Exergy%Cost%[MJ/kg]% Exergy"Cost" Unit"price" Figure 5.16: Cost allocation of Ni by-products, based on price and exergy 5.7.3 Exergy Cost Allocation of Lead and its by-products The allocation of the subprocesses lead-silver production is not clear, because the only available data are the final commodity prices. Besides the share of the products in total costs is not easy to estimate. In Ecoinvent, Classen et al. (2007) the by-product allocation problem (the join production of silver and lead) is undertake by a subdivision of the sub-processes. The starting 117 point for the estimation depends on the general profit expectations of the company, considering an arbitrary performance value of 10%. Hence, the allocation factors are based on revenue but these values are corrected by mass in order to keep up with the resource balance in the final commodity. Metal prices are used as the allocation parameter for polymetallic mines in Classen et al. (2007). Figure 5.17: Exergy Cost allocation of Pb and its by-products Zn, Ag and Cd. The exergy cost allocation to lead production and its by-products zinc, cadmium and silver, starts with the definition of the processing operations, which are grouped in two general stages: 1) Mining and Beneficiation and 2) Smelting. Figure 5.17 depicts the input-output of utilities (e.g. electricity, natural gas, fuel oil and coal) and raw materials (e.g. ores from which copper 118 is produced), as well as the point in the process where the by-product are obtained. The same procedure (previously explained for copper and nickel, in Figs. 5.14 –5.16, respectively) is followed to estimate the exergy cost allocation factor. In this manner, the inputs are shared out among the commodities obtained in each process. Table 5.12 depicts the exergy cost allocation factor for each commodity, it can be observed that in the case of lead in the first stage the allocation factor is 70% whilst in the second stage is 1.8%. This situation is presented because in the first case, lead shares the energy inputs with zinc, which has a lower ERC. The opposite occurs in the last stage, where cadmium and silver have a very high ERC. This situation demonstrates that a general exergy cost allocation for a mineral processing operation cannot be established, but rather a detailed analysis of each process is required in order to allocate cost suitably. Table 5.12: Allocation factors of lead and its by-product Zn, Ag and Cd Commodity Output Exergy cost Exergy Cost Unit price Ton. Price bch product [kg] a.f. [%] [MJ/kg] [$/kg] a.f. [%] a.f. [%] a.f. [%] Lead_1 70 120 61.5 65 33.5 Lead_2 1 1.8 252 3 74.6 50 61.3 Zinc 0.626 30 59 2 38.5 35 66.5 Cadmium 0.339 97.7 4,595 3 25.29 17 38.4 Silver 0.00153 0.6 6,659 1,130 0.11 32 0.3 Figure 5.18 depicts the exergy cost calculated from a thermoeconomic input-output analysis and the unit price of commodities provided by U. S. Geological Survey (USGS) in the year 2011. It can be observed that silver is the only metal with high market price and high exergy cost. On the contrary, cadmium for instance, has a very low price considering its high exergy content. 0" 200" 400" 600" 800" 1,000" 1,200" 1" 10" 100" 1,000" 10,000" Pb Zn Ag Cd Unit%price%[$/kg]% Exergy%Cost%[MJ/kg]% Exergy"Cost" Unit"price" Figure 5.18: Cost allocation of Pb by-products, based on price and exergy. 5.7.4 Exergy Cost Allocation of REE The rare earth elements (REE) are a group of seventeen chemically similar elements. Nonetheless in this analysis only six will be taken into account: lanthanum, cerium, praseodymium, neodymium, gadolinium and yttrium. REE are generally obtained from monazite or bastnaesite. This study is focused on the former ore. The production structure of REE processing de119 picted in Fig. 5.19 is divided into three general steps: 1) Mining and Beneficiation, 2) REO (Rare Earth Oxides) separation and 3) Reduction. Stages 2 and 3 are separated for light, medium or heavy REE. (See Chapter 2, section 2.12). In order to perform the exergy analysis, the production process was separated in two stages: oxide production (mining and beneficiation and REO separation) and metal production (separation and reduction of LRE, MRE and HRE). The energy required during these processes are presented in Table 2.13. The oxide production is considered as the concentration step. In fact, with such energy values, the ERC [MJ/kg] were calculated and presented previously in Table 5.2. All data used in this analysis was taken from the study developed by Koltun & Tharumarajah (2008). Figure 5.19: Exergy Cost allocation of REE. 120 Figure 5.19 shows the general scheme to produce REE. Once, the ERC have been calculated, it is possible to calculate the exergy cost allocation factors (% B), for the oxide production processes. Then, the total energy input of 166.175 MJ during the oxide production of 1 kg of a mixture of REO (j=1 and j=2) was considered and allocated to each REE as shown in Fig. 5.19. The subsequent stage of metal production has a total energy input of 92.35 MJ/kg of light REE, 55 MJ/kg of medium REE and 81 MJ/kg of heavy REE. The exergy allocation was performed in the same way than for oxide production. Accordingly, the total energy consumption (Etotal ) which includes the oxide and metal production was calculated for each REE and depicted in Fig. 5.19. Table 5.13 shows the output amount of each REE, the exergy cost allocation factor, the unit price provided by U. S. Geological Survey (USGS) in the year 2011, the price allocation factor and the tonnage allocation factor. It can be observed that although the numbers are quite different for the three allocation factors (ERC, price and tonnage), the order to appraise the REE is similar. The REE with the greatest allocation factors is cerium, whereas yttrium and gadolinium are those with the lowest ones. This is because cerium is the most abundant REE in monazite whilst yttrium and gadolinium are the least abundant. It is important to highlight that the exergy cost allocation factors are based on ERC (which in turn are independent on the mass output product or metal prices) this is why exergy cost allocation factors provide more objective values to perform any distribution of resources among the different products obtained from the same process. Table 5.13: Exergy Cost Allocation of REE. Commodity Output Exergy Cost Unit Price Price Ton. product [kg] a.f. [%] [$/kg] a.f. [%] a.f. [%] Lanthanum 0.228 12.2 38 26.2 27.9 Cerium 0.392 54.2 30 35.6 48 Praseodymium 0.043 17.0 60 7.8 5.3 Neodymium 0.147 14.9 63 27.9 18 Gadolinium 0.004 1.6 165 2.2 0.5 Yttrium 0.002 0.2 50 0.3 0.3 5.8 Summary In this chapter, the tools described in the Ch. 4 have been improved, so as to adapt the methodologies to the analysis of mining and metallurgical systems. Accordingly, an analysis of the energy used in mining as function of ore grade, the exergy replacement cost and the unit exergy costs of different minerals have been performed. Results obtained were published in Domínguez et al. (2013). The general observed trend is that as the ore grade declines, the energy and the exergy replacement costs increase exponentially. A mineral has a high exergy replacement cost when 1) the concentration of the mineral in the crepuscular crust (in Thanatia) is low and the difference between the ore grade of the current mines (xm) is high and/or 2) when the energy required to extract the mineral and beneficiate it is significant. The obtained values are an intermediate step for assessing in a physical way, the free exergy provided by Nature that man is rapidly destroying through the depletion of high-grade ores. This should serve as an assessment tool for decision-makers in the mineral industry. It should be stated that the values obtained are first assessments. Important assumptions have 121 been made, such as assuming that the same technology is applied for the whole range of grades analyzed, including the crepuscular ore grade. One of the major limitations found is the lack of real data over time. So estimation of future trends of this issue without real and reliable information becomes subjective. Therefore, the results and data provided are an attempt to afford indicators for identifying challenges and opportunities in the mining sector and should not be taken as final and closed. An additional activity undertaken in this chapter was to propose a new allocation method. Since the mining and metallurgical industry commonly produces two or more commodities simultaneously and bearing in mind that the use of LCA in these processes will continue to increase, the necessity to look for suitable by-product cost allocation procedures, was presented. In this chapter, three cost allocation approaches through tonnage, price and exergy replacement cost were analyzed and applied to 33 different mineral deposits. The exergy replacement cost allocation procedure was further applied to four metallurgical processes (copper, nickel, lead and REE), using an extended thermoeconomic approach. It was demonstrated that cost allocation through tonnage is not always a suitable option because minor by-products appearing at very low concentrations can be even more valuable than the major product. With market prices used to perform cost allocations, the problem is associated with their high volatility, which in turn depends on macroeconomic variables that commonly do not reflect the physical conditions of non-fuel mineral resources in Nature and the fact Man is destroying the mineral wealth on Earth, which will be unavailable for future generations. The exergy cost allocation using exergy replacement cost provides in turn, objective values to dispersion and offers a “natural” cost of minerals extracted. Furthermore, it measures mineral endowment depletion on a grave to cradle basis. Once the general basis to perform a complete exergy analysis of mineral resources and the metallurgical processes have been presented, in the next chapter a more detailed exergy analysis of the processing of seven metals is carried out. 122 Table 6.3: Main mass and energy flows for the production of 1 kg of nickel from sulphides - option A. Process Mining Beneficiation Drying Roasting Smelting Converting Electrolysis Inputs [kg] Ni ground 1.26 Cu ground 0.65 Explosive 0.17 Water 106.1 152 Sand 45.6 0.54 1.66 Cement 3.62 Lime 0.16 0.54 1.66 Cyanide 0.004 Sulphuric Acid 0.25 Utilities [MJ] Diesel 11.2 Electricity 15.7 0.25 0.78 10.44 5.09 5.1 Natural gas 2.44 2.8 3.05 3.97 Fuel oil 1.8 24.8 Coke 2.2 Outputs [kg] Sulphidic tailings 50.7 65 Tailings 33.3 Slag 10.5 0.39 Sulphur dioxide 1.5 Wastes [kg] Steam 0.78 13.1 2.86 Anode slime 0.07 Spoiled anodes 0.18 Waste heat [MJ] 37.01 2.89 2.628 24.79 6.65 9.14 Table 6.4: Main mass and energy flows for several processes in the production of 1 kg of nickel from sulphides . Process Sulphuric Leaching Reduction Carbonyl Purification Hydrogen acid of leaching reduction Inputs [kg] Off gas 22.6 Leaching agents 0.08 Hydrogen 0.005 Utilities [MJ] Electricity 4.49 2.62 0.06 0.82 0.06 0.06 Natural gas 1.76 Fuel oil 1.21 Wastes [kg] Ammonium sulphat 0.3 Precipitates 0.17 Dross 0.03 0.001 Copper 0.51 Steam 0.24 Waste heat [MJ] 22.98 1.95 0.06 2.35 129 The exergy values presented in Table 6.5, show the same tendency as energy consumption for nickel production presented in Table 2.10 in Chapter 2, since in both cases, energy and exergy analysis, it can be observed that greater values are presented for nickel production coming from laterites than from sulphides, although the chemical exergy of laterites (0.0788 GJ/t) is smaller than for sulphides (8.85 GJ/t). ! Figure 6.3: Exergy flows (MJ) for several processes in the production of 1 kg of nickel from sulphides. Exergy required to obtain 1 kg of nickel from laterites or sulphide ores is 221 GJ/t and 127 GJ/t, respectively, as depicted in Table 6.5. However, different values can be obtained depending on: the system boundary selected (for instance, if direct electricity or the primary energy sources extracted from the environment used to produce electricity, are considered as inputs), the accurate composition of ores (based on information from specific mines), the multi-output allocation problem (which usually takes place in mining and metal production where several metals are produced from the same raw material) and the way in which it is solved. Sulphide ore processing has efficiencies fluctuating from 0.67 to 0.79, depending on the specific technologies utilised. The higher efficiencies are reached when leaching technologies are used. On the contrary if nickel is produced from laterites, the efficiencies achieved are lower on average (0.38) due to the cost-intensive processing. 130 Table 6.5: Exergy and exergy efficiency in nickel production. Values expressed in GJ/ton of Ni. Process Mining and Smelting and Total Exergy beneficiation refining Efficiency ε Laterites 4.45 217.02 221.46 0.39 Sulphides A 61.17 65.71 126.88 0.67 B 61.17 58.60 119.77 0.71 C 61.17 61.07 122.25 0.69 D 61.17 52.68 113.85 0.75 E 61.17 59.79 120.96 0.71 F 61.17 55.01 116.18 0.73 G 61.17 56.83 118 0.72 H 61.17 46.88 108.06 0.79 I 61.17 51.52 112.7 0.76 The obtained results allow to have a unified picture of the overall processes involved in nickel production. The use of the same units for all inputs and outputs facilitates a direct comparison among production routes, the establishment of efficiency ratios including not only energy (in form of electricity, diesel, natural gas or coal), but also water and raw material consumptions and the identification of improvement possibilities. This way for instance, it has been stated that when leaching technology is used to produce nickel, higher efficiencies are obtained than when electrolysis processes are utilised. Nevertheless, there are some aspects in this methodology that are not taken into account. One of the most limiting shortcomings of the exergy analysis is its complexity of calculation and in the understanding of it for non-exergy practitioners, including plant engineers or decision makers. This means that the advantages of using exergy as an accounting tool need to be perfectly justified and in this regard, exergy analysis should provide information and solutions that other more conventional analyzes are unable to do. The total exergy associated with the production of nickel is lower for sulphides than for laterites. But the same conclusion can be drawn if the analysis is carried out regarding only energy values (as revealed by Table 2.10), since the use of other substances is not very relevant. Hence, the use of exergy as it was implemented here is not justified. The same thing happens with many other metals where the dominant factor in the production chain is the consumption of energy and not of water, chemicals, etc. Furthermore, awkward results can be obtained when using only the chemical exergy of substances as an accounting tool. For instance, a common chemical used in the metallurgy sector, sulphuric acid, has a chemical exergy value of 161 kJ/mol, whereas scarce and precious gold has only 60 kJ/mol. Here it can be seen clearly that the chemical exergy component is not always a good accounting tool and should not be used in isolation. In the case study, laterites have an almost negligible chemical exergy, although the energy requirements and exergy accounting results are greater than those obtained for sulphides, which in turn have a greater chemical exergy. There is an important factor missing from the methodology, namely the concentration exergy. As explained previously, laterites are more abundant than sulphides. If only chemical exergy is taken into account, then it is clear that the best way to produce nickel is from sulphides. In the same way, from a chemical exergy point of view, gold is of little value. Therefore, the way in which the minerals are found on the crust is completely ignored with the methodology suggested by Ayres et al. (2006). This shortcoming is solved by integrating the Exergy replacement 131 cost into the calculations, as performed in the next sections. The latter approach implies that the scarcity factor of mineral resources should be accounted for when assessing the sustainability of mining and metallurgical processes. If this is indeed carried out, the advantage of using sulphides over laterites is not so evident. Additional and detailed information in regards to mining processing can be obtained if the boundaries of the analyzed system are expanded. For instance, the analysis of a specific mining operation can be enlarged if the system of energy production (power and refinery plants) required to produce minerals is included. The latter can be performed through the ThermoEcological Cost methodology, which also will be complemented with the Exergy replacement cost, as depicted in next sections. 6.2 Thermo-Ecological Cost applied to metallurgical systems The Thermo-Ecological Cost (TEC) expresses the cumulative consumption of non-renewable exergy per unit of the considered useful product (Szargut 1986,1989,2005) and provides a cradle to market approach. In this section, the analysis of nickel processing is extended with the TEC methodology, broadening the scope from a mineral processing operation to a more complex metallurgical system where the power and refinery plants (required to satisfy the energy consumption of minerals production) are included into the exergy analysis. 6.2.1 TEC analysis: the case of nickel production The first step to calculate the TEC of a process is to identify the inlet and outlet flows of raw materials, semi-finished products and products for each stage. In order to exemplify the methodology, the case of nickel production is analyzed first. In Chapter 2– section 2.2.13, the general route to produce nickel from sulphide ores has been explained. It includes several steps, namely: ore mining, beneficiation, drying, roasting, smelting, converting, sulphuric acid, leaching, reduction, electrolysis, purification of leachate and carbonyl. There are assorted processes used to produce nickel. These variations depend on factors such as the grade or the concentrate and the presence of other metals in the material mined. The options for nickel production from sulphide concentrates are classified in Table 2.11. Nevertheless, Fig. 6.4 depicts the nickel processing route A: mining, beneficiation, drying, roasting, smelting, converting and electrolysis. 132 Figure 6.4: Nickel production system for sulphide ores. Whilst, the way to produce nickel from laterite ores, described in Chapter 2– section 2.2.12, consists mainly of five linked operations: ore mining, drying, roasting, melting and refining, as depicts Fig.6.5. Accordingly, Tables 6.6 and 6.7 contain data of sulphide and laterite ores, respectively. 133 Figure 6.5: Nickel production system for laterite ores. Data for estimation of coefficients of consumption ai j were obtained from several sources such as CFE (2013), Classen et al. (2007), Domínguez et al. (2013), ITP (2002), Stanek (2009), Szargut & Stanek (2012), Valero et al. (2013). For instance, assumptions taken into account to perform the calculations include that production of fuels such as natural gas or coal, requires 0.26 and 0.24 MJ of electricity per kilogram of fuel, respectively, in accordance with Szargut & Stanek (2012). Production of electricity meanwhile, requires 2.48 MJ of natural gas, 2.87 MJ of fuel oil and 2.816 MJ of coal per MJ of electricity produced CFE (2013). It was considered a low heating value of 39.5 MJ/kg of fuel oil, 44 MJ/kg of natural gas and 21.68 MJ/kg of coal. Whilst, the efficiencies to produce energy from fuel oil was assumed as 35 %, from natural gas 36.5 % and from coal 28.5 %. 134 Table 6.6: Consumed products during Ni production from sulphide ore - option A. Considered jth process 1 2 3 4 5 6 7 8 9 10 11 kg_Ni kg_Ni kg_Ni kg_Ni kg_NG MJ_pp kg_R kg_Cu kg_coal kg_lime kg_cem 1 kg_Ni 2.80 2 kg_Ni 1.143 3 kg_Ni 1.246 4 kg_Ni 5 MJ_NG 1.871 2.669 3.178 2.480 0.005 0.005 6 MJ_PP 2.110 4.096 4.454 4.091 0.260 0.080 0.240 0.014 0.450 Consumed 7 MJ_R 1.505 10.290 2.870 0.022 ith product 8 kg_Cu 0.087 ai j [i]/[j] 9 MJ_coal 2.816 0.001 0.151 10 kg_lime 0.021 0.786 11 kg_cem 0.487 12 kg_sil 6.129 0.786 13 kg_NaCN 0.001 14 kg_sulph 0.034 15 kg_CuS 0.01472 Chemical bjMJ/[j] 8.85 45.76 42.265 8.34 23.58 0.2 0.24 Exergy Concent. bjMJ/[j] 260.27 38.22 1.47 Exergy 135 It is important to note that Tables 6.6 and 6.7 are related to a physical structure which depicts the overall connections between each branch as shown in Fig. 6.4 for sulphides ores or in Fig.6.5 for laterites ores. Table 6.7: Consumed products during Ni production from laterite ore. Considered jth process 1 2 3 4 5 6 7 8 kg_Ni kg_Ni kg_Ni kg_NG MJ_pp kg_R kg_coal kg_lime 1 kg_Ni 3.5 2 kg_Ni 1.286 Consumed 3 kg_Ni ith product 4 MJ_NG 8.229 0.777 2.48 0.005 ai j [i]/[j] 5 MJ_PP 0.028 8.486 2.24 0.260 0.08 0.24 0.014 6 MJ_R 0.503 1.071 2.87 0.022 7 MJ_coal 11.43 2.816 0.001 8 kg_lime 0.134 Chemical bjMJ/[j] 0.07 45.76 42.265 23.58 0.2 Exergy Concent. bjMJ/[j] 56.84 1.47 Exergy Subsequently, the TEC balance through Eqs.4.23–4.24 is applied individually for each process jth of nickel production from sulphide ores, in order to obtain two equations for each branch. Hence, it is necessary to define the value for each coefficient aand f, which are related with the real production process shown in Fig. 6.4. The utilities used during the nickel production process through the different stages are gathered in the Ecoinvent database (Classen et al. 2007). Nickel production from sulphide ore using leaching (option B - Table 6.9) leads to obtain copper as a by-product. Thence, the coefficient fis the amount of copper as by-product in the mining step and represents the avoided cost of mine copper because it is mined together with nickel ore. There are three types of equations that can be formulated depending if the process under analysis corresponds to the production of: 1. Mineral. When an ore is mined frequently different minerals are obtained. For this reason, a mineral can be acquired as a main product or as by-product. Hence, both cases have different TEC equations. - Production of a mineral as a main product includes electricity from power plant, fuels, raw minerals, other minerals presented in the same ore where the desired mineral is mined and the exergy of the mineral. For instance, the following Eqs. 6.1–6.2 relate all links between the different processes needed during the first step of nickel production: mining and beneficiation (j=1), shown in Fig. 6.4. It should be noticed that exergy of the mineral appears only in Eq. 6.1, because Eq. 6.2 represents the TEC part due to fuel consumption. ρ1=a6,1 ·ρ6+a7,1 ·ρ7+···+bNi (6.1) z1·ρ1=a6,1 ·ρ6·z6+a7,1 ·ρ7·z7+... (6.2) 136 - Production of a mineral as by-product includes the same inputs as production of a mineral as a main product, but additionally the by-product term must be included. For example, in this study copper is produced as a by-product during the electrolysis process of nickel production from sulphide ore depicted in Fig. 6.4. Then, Eqs. 6.3–6.4 are obtained: ρ4=a3,4 ·ρ3+a5,4 ·ρ5+···−f4,8 ·ρ8(6.3) z4·ρ4=a3,4 ·ρ3·z3+a5,4 ·ρ5·z5+···−f4,8 ·ρ8·z8(6.4) Fuel. In general when a fuel is produced, for instance natural gas, coal or fuel oil, the main contribution to its TEC is electricity and its chemical exergy. For production of natural gas (j= 5) shown in Fig. 6.4, Eqs. 6.5–6.6 are obtained. It is important to mention that the fuel exergy is included in both equations and not only in the first one as in the case of mineral production (Eqs. 6.1–6.2). ρ5=a6,5 ·ρ6+bNG (6.5) z5·ρ5=a6,5 ·ρ6·z6+bNG (6.6) Products which are not obtained directly from nature e.g. electricity. The main inputs for the electricity production in a power plant (j=6) are fuels such as natural gas, oil and coal, as Fig. 6.4 depicts and Eqs. 6.7–6.8 express: ρ6=a5,6 ·ρ5+a7,6 ·ρ7+a9,6 ·ρ9(6.7) z6·ρ6=a5,6 ·ρ5·z5+a7,6 ·ρ7·z7+a9,6 ·ρ9·z9(6.8) Once the TEC methodology has been applied, the ERC are introduced in the set of equations listed previously. This way, the exergy of the mineral denoted by bNi in Eq. 6.1 is substituted by the ERC. The same procedure is performed to those equations which involve mineral resources. The Exergy Replacement Cost of the commodities analyzed in this chapter are presented in Table 5.2, in Chapter 5. The aforementioned fusion of TEC and ERC is explained in the next section. 6.3 Integration of the Thermo-Ecological Cost and Exergy Replacement Cost to assess mineral processing The motivation to join both indicators, Thermo-Ecological Cost and Exergy Replacement Cost, starts when I was in a 3-month research stay at the Institute of Thermal Technology of the Silesian University of Technology in Poland. Professor Stanek explained me their Thermo-Ecological Cost methodology. He had applied the TEC method to a simplified steel work (Szargut & Stanek (2012)) and results showed that the mineral part of TEC was always smaller when compared to the fuel part, due to the fact that in TEC analysis, the mineral part was obtained taken into 137 account only the chemical exergy of mineral resources. However, this assumption is not always sufficient when mineral resources are assessed, as explained in Chapter 5. The value of a mineral is very much associated with its scarcity degree. This is why the concentration exergy component, accounted for through the Exergy Replacement Cost, is very relevant when non-fuel minerals come into play. The original TEC approach leads to results implying that mainly exergy coming from fuel consumption influences the Thermo-Ecological Cost of a particular good. Therefore, it suggests that non-fuel mineral consumption is of minor relevance from the Thermo-Ecological Cost Theory point of view. As this is not necessarily true, the TEC methodology is complemented with ERC, so as to account for the impacts associated with mineral consumption and dispersion. Details of both methodologies can be consulted in chapter 5. As aforementioned, the TEC provides a cradle to market approach. The ERC, meanwhile, enhances the TEC because it assesses the concentration exergy that would be expended in recovering a mineral deposit from the material dispersed in the Earth’s crust with the available technology through a grave to cradle approach developed by Valero & Valero (2010a). Results show that when Exergy Replacement Costs are embedded into the TEC infrastructure, the impacts associated with mineral consumption are significantly greater. Accordingly, the inclusion of ERC into TEC allows for a more comprehensive and fairer weight to the consumption of nonfuel mineral resources, thereby providing better indications as to the achievement of a more sustainable production. TERC methodology proposed in this chapter, means that the Thermo-ecological cost is complemented with the Exergy Replacement Cost, with the main objective to integrate concentration exergy of mineral resources into the cumulative exergy account associated with the production of a particular commodity performed through the TEC method. The Thermo-Ecological and Exergy Replacement Costs of mineral processing are very variable, depending on raw material sources, production process and final products obtained. Both methodologies have been applied to seven different cases of metal production. In order to explain the procedure that integrates the ERC into TEC analysis, the results from the case of nickel processing will be analyzed first. 6.3.1 TERC analysis of nickel production Both methodologies have been applied to three different cases of nickel production: nickel from sulphide ore using electrolysis (option A - Table 6.8), nickel from sulphide ore using leaching obtaining copper as a by-product (option B - Table 6.9), and ferronickel from laterite ore (Tables 6.10 and 6.11). Ni1, Ni2, Ni3, Ni4represent Ni obtained at each of the processes involved in the nickel production chain, which are shown in Figs. 6.4 and Fig.6.5. Results of the exergy analysis of nickel production are presented in four sections, as follows: 1) TEC methodology, 2) TEC complemented with ERC, and 3) sensitivity analysis. TEC methodology In order to explain the results of TEC analysis, results of nickel production from sulphide ores (option A) and shown in Table 6.8 will be explained. TEC increase through the first step of mining and beneficiation (27.29 MJ/kg_Ni) until the last electrolysis step (179.2 MJ/kg_Ni), because in this final step the highest amount of energy consumption (as electricity and natural gas) 138