LIFE CYCLE ASSESSMENT OF COPPER AND NICKEL PRODUCTION
T. E. Norgate and W. J. Rankin
CSIRO Minerals, Clayton, Victoria, Australia
Published in Proceedings, Minprex 2000, International Conference on Minerals Processing and Extractive Metallurgy, September 2000, pp133-138.
A study of refined metallic copper and nickel production was carried out using Life Cycle Assessment methodology to estimate the life cycle emissions of greenhouse and acid rain gases for these metals using data derived from the literature. Various processing routes were investigated which included a pyrometallurgical and a hydrometallurgical route for each metal. The effect of a number of process variables, namely ore grade, fuel type for electricity generation and efficiency of generation were also examined in the study.
The results of the study, expressed in terms of total (or full cycle) energy consumption, Global Warming Potential (GWP) and Acidification Potential (AP), showed that the hydrometallurgical processes involving solvent extraction and electrowinning had higher energy consumptions and GWPs than the pyrometallurgical processes for both metals. Nickel metal production was shown to be several-fold more energy intensive than copper metal production. The significant effect that falling ore grades have on attempts to meet greenhouse emission targets was also highlighted in the study.
The issue of gobal climate change has been a focus of international debate in recent years. While the impact of changes in atmospheric gas composition continue to be debated in scientific circles, the conclusion that they represent a significant risk has been firmly embedded in national and international policies. As a signatory to the 1997 Kyoto Protocol, Australia is committed to limiting the growth of its greenhouse gas emissions to 8% above 1990 levels by 2008 – 2012. Given that the mining, minerals and metal production sector’s contribution to Australia’s total CO2 emissions in 1996 was in the order of 10% (National Greenhouse Gas Inventory Committee, 1996), it is critical that new and existing processes for metal and minerals extraction and refining be evaluated with respect to these issues. It is no longer sufficient to conduct a techno-economic evaluation of a new process to establish its likely economic viability – this evaluation must now include some form of environmental impact analysis.
However, comparing new and existing processes or products may be meaningless, or worse misleading, if only "across-the-fence" plant emissions are considered. If a true picture of the environmental impact of a process is to be obtained, it is essential that all inputs and outputs during the entire life of the process or product be included. Life Cycle Assessment is one of a number of methodologies that can be used for such purposes.
This paper describes a study carried out to determine the life cycle emissions of greenhouse and acid rain gases produced during the mining and production of refined metallic copper and nickel. The purpose of the study was to give first estimates of the relative differences between the various processing routes with regard to these emissions.
LIFE CYCLE ASSESSMENT
Life Cycle Assessment (LCA) is an analytical tool for quantifying the resource consumption and environmental impacts associated with a product, process or activity during its entire life cycle. LCA has also been referred to as "full fuel cycle analysis" or "cradle-to-grave" analysis. The objective of most LCA studies is to find the design option that minimizes the life cycle impact of the process.
The LCA methodology has four distinct stages:
One of the most important issues at the goal definition and scoping stage is the definition of the system to be studied. This is a critical step in ensuring that the LCA is both manageable and meaningful. Attempts have been made to define "decision rules" which can be used to set system boundaries, although these still allow for considerable discretion. These rules are generally based on the contribution of the system components to total energy, total mass or "environmental relevance". Decision rules based on mass contribution are probably the most commonly used for excluding ancillary materials; e.g., exclude materials that contribute less than 5% to the mass of a unit process or those that contribute less than 1% to the overall mass of the system. However, before excluding a material from the inventory it is advisable to repeat the exercise using energy as the criterion. It is also common practice in conducting LCAs to omit material and energy inputs associated with equipment manufacture and plant construction on the grounds that they are negligible. Another reason is that these inputs are quite difficult to calculate accurately and, furthermore, require an iterative approach; e.g., coal production requires inputs of steel and electricity but electricity production requires coal and steel. All inputs that are included in the LCA must be tracked back to naturally occurring materials.
Data availability is a major issue in conducting LCAs. Indeed it is the data collection associated with the inventory stage that is responsible for most of the time and cost of LCA. While the data preferably come from measurements on actual plants, such data are usually confidential to companies and very little is published in a form suitable for use in LCA. Alternatively, mass and energy balance outputs from process simulation models may be used as the source of inputs for LCAs. This is particularly the case for new process designs.
A problem that often arises at the inventory stage of an LCA is the use of allocation rules for estimating inventory data when there is more than one useful product. Simple allocation procedures may be based on mass, volume, energy content or economic value. The most common practice is to allocate on the basis of mass. An improvement on this approach (Azapagic and Clift, 1995) is to allocate on the basis of marginal changes in the co-products; i.e., the effect of individual changes in co-product mass flowrates on process emissions.
The inventory results are classified according to the kind of environmental problems to which they contribute during the impact assessment stage. Attempts are then made to quantify the contributions to each impact category. To do this equivalency factors are used. These factors indicate how much a substance contributes to an environmental impact compared to some reference substance. Some of the environmental impacts commonly considered are:
Each inventory amount is multiplied by its corresponding equivalency factor and an aggregated score for each impact category is obtained. The results from this step are considered to be the environmental profile of the system. The aggregated scores for global warming and acidification are referred to as the Global Warming Potential (GWP) and Acidification Potential (AP) respectively. The GWP and AP per kg of metal produced were used to compare the various processes. The equivalency factors (IEA Greenhouse Gas R&D Programme, 1999; Gediga, J., Florin, H. and Eyerer, 1997) used in this study were:
To assist with the data storage, retrieval and manipulation associated with LCA, a software program has been developed by CSIRO Minerals using MS EXCELTM. This program allows the user to quickly build up an LCA workbook for a process, with separate worksheets for each process step. The emissions from one worksheet are carried over to the next until the final worksheet where the overall process emissions are calculated. The GWP and AP for the process is then calculated and the contributions of the various process steps to the overall result are reported. Program options include fuel type and efficiency for power generation, mode of transport and fuel calorific values.
SCOPE OF WORK
Because of the complexity of the strict "cradle-to-grave" approach of LCA and the lack of suitable data, the system boundary for the study described here was restricted to "cradle-to-gate"; i.e., the processes have only been considered to the point where refined metallic products are made available to the secondary manufacturing sector. This limited form of LCA is quite valid for comparing the environmental impacts of the "process" stage of the life cycle, assuming that the "product" stage of the life cycle is the same in both cases.
The selected processing routes are given in Table 1 and include a pyrometallurgical and hydrometallurgical process for each metal. All routes include a mining stage (underground for all cases with the exception of the nickel laterite ore) and some form of mineral beneficiation or ore preparation stage. Both smelting processes include an acid plant to capture SO2 emissions.
Generic flowsheets of each process were drawn up and used to establish the individual LCA spreadsheet models. Each flowsheet was constructed at a level of detail consistent with publically available process data. This generally resulted in a flowsheet of 3 to 5 process steps; e.g., mining, mineral processing, smelting and refining. The decision rules referred to earlier were used to exclude certain ancillary material inputs from the flowsheets such as flotation reagents in the mineral beneficiation stage. The data used for input to each model were derived solely from the literature, with the data being cross-checked with more than one source where possible. These data and their sources are summarised in Table 2, although for the sake of brevity those inputs that do not have a significant effect on GWP and AP, such as water, have been omitted from the table.
While the LCA software program developed by CSIRO Minerals estimates the life cycle emissions of gaseous, liquid and solid wastes from a process given the appropriate inventory data, the environmental impact categories were limited to greenhouse and acid rain gas emissions in this study.
The following assumptions were made in carrying out the LCA of the processing routes outlined in Table 1.
The results of the study are summarised in Table 3 in terms of total (or full cycle) energy consumption, GWP and AP. These results show that nickel production is several-fold more energy intensive than copper and that the hydrometallurgical processes involving solvent extraction and electrowinning have higher energy consumptions and GWPs than the pyrometallurgical processes for both metals. The contributions of the various process stages to the overall GWP are shown in Figures 1 and 2 for the copper and nickel base cases respectively. The APs of the copper pyrometallurgical and hydrometallurgical processes are similar, while the AP of the nickel hydrometallurgical process is about half that of the pyrometallurgical process because of the oxide ore being treated.
Figure 1. Contributions of process stages to GWP for copper
Figure 2. Contribution of process stage to GWP
for nickel of process stages to GWP for copper
The effect of several process variables on the LCA results were also investigated. Figures 3 and 4 show the effect of ore grade on total energy consumption, GWP and AP for the production of metallic copper by the smelting and electro-refining route and metallic nickel production by the smelting and Sherritt-Gordon refining route respectively. The results of Chapman (1974) and Kellogg (1974) who estimated the total energy consumption for the production of copper by a similar process route to that considered here, are also shown in Figure 3 for comparison. The average ore grade and total energy reported by Chapman and Roberts (1983) for nickel production by smelting and refining is also shown in Figure 3. It is apparent from these figures that falling ore grades have a large effect on the results for both copper and nickel at metal grades below about 1%.
Figure 3. Total energy vs. ore grade (smelting & refining)
Figure 4. GWP and AP vs. ore grade (smelting & refining)
Because of the major contribution of electricity generation and supply to the GWP of all four processes considered (see Table 3), the effect of the type of fuel used for electricity generation as well as the efficiency of generation on the GWP of the processes were investigated and the results are shown in Figures 5 and 6 for copper and nickel metal production respectively. Figure 5 shows that by changing from black coal to natural gas (at the base case efficiency of 35%) the GWP of the copper hydrometallurgical process was reduced from 6.2 to 4.2 while for the pyrometallurgical process it was reduced from 3.3 to 2.4. The corresponding values from Figure 6 for nickel are 16.1 to 13.3 for the hydrometallurgical process and 11.4 to 9.5 for the pyrometallurgical process. Increasing the electricity generation efficiency reduced the GWP of both processes for both copper and nickel as expected, with the decrease being more pronounced for the copper hydrometallurgical process because of the greater electricity contribution to that process. At no point do the hydrometallurgical processes have a lower GWP than the equivalent pyrometallurgical processes for the two fuel types considered here.
Figure 5. Effect of fuel type and efficiency
for electricity generation on GWP for copper.
Figure 6. Effect of fuel type and efficiency
for electricity generation on GWP for nickel.
The above results are based on average process data taken from the literature and as such should be considered as indicative only. Nevertheless they do show good agreement with the limited total energy data reported in the literature (Figure 3 ). The observation that the hydrometallurgical route for copper production is more energy intensive (ie., a higher energy input per unit mass of metal product) than the pyrometallurgical route is also in agreement with the findings of Kellogg (1982).
The use of electricity in the production of copper is shown below, with the data from Table 2 being expressed in terms of per tonne of refined copper: Most electricity consumption in the smelting route is in crushing, grinding and flotation to prepare a copper concentrate for smelting. Little electricity is consumed in the leach /SX/EW route for raw material beneficiation since the leaching is performed on the largely "as-mined" lumpy ore. In the latter route, most electricity is consumed in the energy intensive solvent extraction mixer-settlers (mechanical agitation) and in the thermodynamically-limited energy intensive electrowining process.
The picture for nickel is similar to that for copper as shown by the electricity consumption rates given below, also taken from Table 2, although in this case the hydrometallurgical refining stage is the largest consumer of electricity in the smelting route.
While it may be possible to reduce the energy consumption of the comminution and solvent extraction stages in the above processing routes by improved technology, this will not generally be the case for electrowinning because of the thermodynamic limitations referred to above.
It is apparent from this study that declining ore grades in the future will have a significant effect on the efforts of the mining and metallurgical industry to meet greenhouse gas emission targets. One way of addressing this problem is by metal recycling. Kellogg (1977) has estimated that the production of copper and nickel from secondary metals requires only 16% and 10% respectively of the energy required from their ores. However recycling is only possible for metals used in non-dissipative applications where the metals can be economically reclaimed. Other generic strategies for reducing greenhouse gas emissions from the mining and metallurgical industries have been proposed by Rankin (1990) and include the following:
A reduction in the indirect emission of greenhouse gases from the mining and metallurgical industry will also be achieved by efforts to improve the efficiency of electricity generation using fossil fuels; e.g., co-generation and advanced power cycles.
This study has illustrated that the total environmental impact of a process can be established only if all inputs and outputs during the entire life of the process are included in the evaluation. Comparing the various processing routes for copper and nickel metal production on a life cycle basis has shown :
Finally, the study has demonstrated that while traditional LCAs are expensive to carry out in detail and require access to high integrity data from known sources, they can be undertaken at a lower level of detail using generic data to provide an overall understanding of issues.
Azapagic, A and Clift, R. 1995. Life Cycle Assessment and linear programming – environmental optimisation of product system. Computers in Chemical Enginering, Vol. 19 Suppl., pp. S229 – S234.
Biswas, A K and Davenport, W G. 1994. Extractive Metallurgy of Copper, 3rd edition, Pergamon, UK, p.202-
Blanco, J L and Holliday B M. 1981. Energy swings at Nickel, J of Metals, 50-51.
Bureau of Transport and Communications Economics. 1995. Greenhouse gas emissions from Australian transport. Long term projections. Report 88. Australian Government Publishing Service.
Chapman, P F. 1974. The energy costs of producing copper and aluminium from primary sources, Metals and Materials, 8 (2) pp 107-111.
Chapman, P F and Roberts, F. 1983. Metal Resources and Energy, Butterworths.
Dasher, J. 1976. The energy picture in nickel production, Mining Magazine, 379-389.
Edwards, J S. 1998. Isasmelt – A 250000 tpa smelting furnace, in AusIMM ’98 – The Mining Cycle, pp 395-400.
Fountain, C R, Tuppurainen, J M and Whitworth, N R. 1993. Operation of the copper Isasmelt plants at Mount Isa Mines Limited, Mount Isa, Qld, in Australasian Mining & Metallurgy, pp 695-700 (AusIMM).
Gediga, J, Florin, H and Eyerer, P. 1997. Life-cycle assessment. A case study in zinc production, in Heavy minerals 1997 Johannesburg, pp 87-92 (South African Institute of Mining and Metallurgy).
Hoey, D W, Leahy, G J, Middlin, B and O’Kane, J. 1987. Modern tankhouse design and practice at Copper Refineries Pty Ltd, in Proceedings Electrorefining and Winning of Copper Symposium, pp 271-293, The Metallurgy Society, Inc.
Hoppe, R W. 1977. Amax’s Port Nickel refines the only pure nickel in the US, Eng & Min J, 76-79.
IEA Greenhouse Gas R&D Programme. 1999. Greenhouse Issues, March.
IPCC/OECD/IEA Programme on National Greenhouse Gas Inventories. 1996. Revised 1996 Guidelines for National Greenhouse Gas Inventories, Reference Manual (Volume 3).
Hunt, P R and George C W. 1985. Efficient use of energy mediums in the reduction of nickel metal from ammonium sulphate solutions, Smelting & Refining Operators Symposium, pp 161-170.
Kellogg, H H. 1974. Energy efficiency in the Age of Scarcity, J of Metals, 26(6), 25-29.
Kellogg, H H. 1977. Sizing up the energy requirements for producing primary metals, Eng & Min J, 61-65.
Kellogg, H H. 1982. The state of nonferrous extractive metallurgy, J of Metals, 34(10), 35-42.
Krag, P, Imrie, W and Berkoe, J. 1993. Anode furnace practice for high-sulphur blister, in Converting, Fire Refining and Casting, pp 255-267, The Minerals, Metals & Materials Society.
Krishnan, N I. 1993. Lead-zinc ore concentrator practice by Pasminco Mining – Rosebery, Rosebery, Tas, in Australasian Mining & Metallurgy, pp 519-522 (AusIMM).
Molinia, A B. 1993. Zinc-lead-copper-silver ore mining by Aberfoyle Limited, Hellyer, Tas, in Australasian Mining & Metallurgy, pp 482-485 (AusIMM).
Nashner, S. 1955. The Sherritt Gordon Lynn Lake Project, CMM Bulletin, 396-410.
National Greenhouse Gas Inventory Committee. 1996. National Greenhouse Gas Inventory 1996..
National Greenhouse Gas Inventory Committee. 1996. Workbooks 1.1 – 8.1.
Nilsson, D. 1992. Surface vs. underground methods, in SME Mining Engineering Handbook, 2nd edition, pp 2058-2069, Society for Mining, Metallurgy and Exploration Inc.
Ozberk, E, Gendron S A and Kaiura G. 1986. Review of nickel smelters – responses to questionnaire, in Proceedings Nickel Metallurgy Symposium, pp 304-319 (Metallurgical Society of CIM).
Paschen, P, Langfellner, M and Mori, G. 1991. Productivity increase and energy conservation in copper electrowinning, in Proceedings Copper 91 International Symposium, Vol. 3, pp 575-591.
Rankin, W J. 1990. The greenhouse effect and the metallurgical industry, in Greenhouse and Energy, pp 338-347 (CSIRO Australia).
Rich, D W. 1993. Smelting practice at Olympic Dam, The AusIMM Centenary Conference, Adelaide, pp 207-215
Salinovich, T and Strachan, P. 1995. The Bulong Project Status April 1995, Nickel/Cobalt laterites, ALTA Metallurgical Services.
Slater, P. 1990. Nickel smelting operations at Kalgoorlie, PYROSEM WA, pp 7-22 (Murdoch University Press).
Taylor, A. 1996. Hydrometallurgical options for treatment of copper sulphide ores, Copper hydrometallurgy forum, ALTA Metallurgical Services.
Taylor, A and Cairns, D. 1997. Technical development of the Bulong laterite treatment project, Nickel/Cobalt pressure leaching & hydrometallurgy forum, ALTA Metallurgical Services.
Westcott, P and Hall, R. 1993. Surface mining equipment operating costs, in Cost Estimation Handbook for the Australian Mining Industry, p 279, (AusIMM).
Wright, P J. 1993. Nickel ore concentration at Leinster Nickel Operations of Western Mining Corporation Limited, Leinster, WA, in Australasian Mining & Metallurgy, pp 1189-1193 (AusIMM).
© Copyright CSIRO Australia, 2000. Use of this web site and information available from it is subject to our Legal Notice and Disclaimer.