ENERGY AND GREENHOUSE GAS IMPLICATIONS OF DETERIORATING QUALITY ORE RESERVES Terry Norgate and Sharif Jahanshahi CSIRO Minerals/Centre for Sustainable Resource Processing URL : http://www.minerals.csiro.au E-mail :
[email protected] ABSTRACT Despite increased levels of recycling, there will be a need for primary metals well into the future. However, reserves of metallic ores around the world are generally deteriorating in grade. Furthermore, many of these low grade ores are fine-grained, requiring additional grinding in order to achieve mineral liberation. Both of these effects, either combined or in isolation, will increase the embodied energy of primary metals. Given that most energy inputs into primary metal production processes are fossil fuel-based, there will also be an increase in the greenhouse gas emissions associated with these processes. Life cycle assessment methodology has been used in the paper to show how these two aspects of deteriorating ore reserves impact on embodied energy and related greenhouse gas emissions for primary metal production, using copper and nickel as examples. The results show that the effect of falling ore grades will be significant at grades below 1% because of the additional energy that must be consumed in the mining and mineral processing stages to move and treat the additional gangue (waste) material. The effect will also be more significant for those metals where the mining and mineral processing stages make substantial contributions to the cradle-to-gate part of the metals life cycle, eg. copper produced pyrometallurgically. Some possible approaches for addressing these issues are discussed in the paper, eg. eliminating the grinding of these ores by smelting them directly. Keywords: ore reserves, ore grade, grind size, embodied energy, greenhouse gas emissions, primary metal production, life cycle assessment 1. INTRODUCTION The anticipated growth in the economies of the developing countries as they strive to improve their standard of living means that there will be an on-going need for primary metals for at least many decades, even with increased levels of dematerialisation (ie. the reduction in the amount of energy and materials required for the production of consumer goods or the provision of services) and recycling. Indeed, it has been reported [1] that under a high growth scenario over the next fifty years, assuming no major politico-social, economic or environmental discontinuities, the consumption of aluminium, copper and nickel in Brazil, Russia, India and China combined will increase to 536%, 249% and 353% respectively of 2002 world consumption of these metals. However, the grades (ie. metal content) of metallic ores are falling globally as the higher grade reserves1 are exploited first and are progressively depleted. Furthermore, many of these newer reserves are fine-grained, requiring finer grind sizes in order to achieve mineral liberation. Both of these issues, either together or in isolation, have significant energy and greenhouse implications for primary metal production in the future. 1
A resource is a concentration of naturally occurring material in the earth s crust in such form and amount that extraction of a commodity from the concentration is currently or potentially feasible. A reserve is that part of an identified resource which could be economically extracted or produced at the time of determination. 5th Australian Conference on Life Cycle Assessment Achieving business benefits from managing life cycle impacts Melbourne, 22-24 November 2006
1
Amount
From a geological perspective, metals, along with other elements, are classified as either being geochemically abundant or geochemically scarce. The first group consists of 12 elements, of which 4 are widely used metals (aluminium, iron, magnesium and manganese) that accounts for 99.2% of the mass of the earth s continental crust. The other elements, including all other metals, account for the remaining 0.8% of the crustal mass. It has been suggested that certain geochemically scarce elements tend to have a bimodal distribution as shown in Fig. 1, in which the smaller peak (corresponding to relatively high concentrations) reflects geochemical mineralization, while the main peak reflects atomic substitution in more common minerals [2, 3]. Using copper as an example, Fig. 2 shows how the average grade of copper ore mined in the United States has fallen over the last century to a current value of about 0.5% (globally 0.8%). However, copper present as atomic substitutions in commom crustal rocks has an average grade of around 0.006% [2]. Separating the copper atoms from the surrounding mineral matrix would require significantly more energy than current extraction processes. Thus to mine the eath s crustal rock for copper, after the reserves of mineralised copper are exhausted, would increase energy requirements (per tonne of copper metal extracted) by a factor of hundreds or even thousands [3]. This has been referred to as the mineralogical barrier as shown in Fig. 3, where the dashed and solid lines refer to geochemically scarce and abundant metals respectively This paper is largely concerned with the effect of ore grade on energy consumption and greenhouse gas emissions at grades above this mineralogical barrier as indicated in Fig. 3.
Current mining
Grade (%)
Fig 1. Probable distribution of geochemically scarce metal in the Earth s crust [3]. 2.0 1.8
Ore grade (% Cu)
1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 1900
1920
1940
1960
1980
2000
Ye ar
Fig 2. Decline in average copper ore grade in the United States. 2. EFFECT OF ORE GRADE The energy consumed over the cradle-to-gate stage of the life cycle of primary metal production (ie. embodied energy or cumulative energy demand) and the related greenhouse gas emissions for various metals and processing routes have
2
Energy consumed/unit mass metal recovered
Metals trapped by atomic substitution Ge oche m ically s carce m e tals "M ine ralogical barrie r"
Metals concentrated in separate ore minerals
Ge oche m ically abundant m e tals
This w ork
Grade (%)
Fig 3. The effect of ore grade and the mineralogical barrier [3].
been reported previously [4, 5] based on earlier life cycle assessments (LCAs) of these processes. Figs. 4 and 5 show the effect of ore grade on the stage-by-stage contributions to the embodied energy2 of copper and nickel respectively, produced by pyrometallurgical processing [5]. The increase in embodied energy in the mining and mineral processing stages with decreasing ore grade occurs because of the additional energy that must be consumed in these stages to move and treat the additional gangue (waste) material. As the copper and nickel concentrates produced at the end of the mineral processing stage have a relatively constant grade (assumed here to be 27.3% Cu and 13% Ni respectively) independent of initial ore grade, the subsequent smelting and refining stages are essentially unaffected by ore grade. Fig. 6 shows stage-by-stage contributions from Figs. 4 and 5 totalled to give the overall embodied energy for ore grades of 3.0% and 0.5% for copper, and 2.3% and 0.5 % for nickel. This figure shows that for typical copper ore (sulphide) grades in Australia (viz. 3.0%), the mining and mineral processing stages account for about 60% of the overall embodied energy, while for an ore grade of 0.5% these stages account for about 90% of the total. On the other hand, for typical nickel ore (sulphide) grades in Australia (viz. 2.3%), the mining and mineral processing stages account for only 27% of the total, increasing to 63% at 0.5% ore grade. Fig. 7 shows the corresponding results for 80 0.5%
Embodied energy (MJ/kg Cu)
70 60 50
0.5%
40
1.0%
30 1.0% 2.0%
20 10
2.0% 3.0%
3.0%
0 Mining
2
Mineral processing
Smelting
Ref ining
Electricity component assumed to be black coal-based at a generation efficiency of 35%. 3
Fig 4. Effect of ore grade on stage-by-stage energy consumption for copper production (pyrometallurgical). 100 0.5%
90
Embodied energy (MJ/kg Ni)
80 70 60 0.5% 50
1.0
40 30
1.0% 2.0%
20 10
2.0%
2.3%
2.3%
0 Mining
Mineral processing
Smelting
Refining
Fig 5. Effect of ore grade on stage-by-stage energy consumption for nickel production (pyrometallurgical).
Embodied energy (MJ/kg Ni or Cu)
250 Smelting & ref ining 200 Mining & mineral processing 150
100
50
0 2.3% Ni
0.5% Ni
3.0% Cu
0.5% Cu
Fig 6. Embodied energy for copper and nickel production (pyrometallurgical).
25
GWP (kg CO2e/kg Ni or Cu)
Smelting & ref ining 20 Mining & mineral processing 15
10
5
0 2.3% Ni
0.5% Ni
3.0% Cu
0.5% Cu
0.5% Cu Giurco et al
4
Fig 7. Greenhouse gas emissions for copper and nickel production (pyrometallurgical). greenhouse gas emissions,3 including some reported data on copper from the literature. While our result for the smelting and refining stage for 0.5% copper ore grade is lower than that of Giurco et al [9], it is similar to that reported by Lunt et al [10] adjusted to the same concentrate grade, viz. 1.40 cf. 1.56. The greenhouse gas results in Fig. 7 largely reflect the embodied energy results in Fig. 6 as the energy inputs are fossil fuel-based. Figs. 8 and 9 show how the overall embodied energy and greenhouse gas emissions for copper and nickel increase as ore grade progressively decreases. Some reported data from the literature is also included in Fig. 8 and show good agreement with the current results.
Embodied energy (MJ/kg Ni or Cu)
400 Ni (pyro) - present w ork Chapman & Roberts (1983) Cu (pyro) - present w ork Chapman (1974) Kellogg (1974)
350 300 250 200 150 100 50 0 0
1
2
3
4
Ore grade (%)
Fig 8. Effect of ore grade on total embodied energy consumption for copper and nickel production.
GWP (kg CO2 equiv/kg Ni or Cu)
40 Ni (pyro) 30
Cu (pyro)
20
10
0 0
1
2
3
Ore grade (%)
Fig 9. Effect of ore grade on greenhouse gas emissions for copper and nickel production. 3. EFFECT OF GRAIN SIZE The particle size to which an ore must be crushed or ground4 to produce separate particles of either value mineral or gangue that can be removed from the ore (as concentrate or tailings) with an acceptable efficiency by a commercial 3
Greenhouse gas emissions are reported as Global Warming Potential (GWP) which is the aggregated contributions of the various gases expressed in terms of the equivalent amount of CO2 emissions. 4 Crushing produces material typically coarser than 5 mm and consumes relatively low levels of energy, while grinding (or milling) produces very fine products (often below 0.1 mm) and is very energy intensive. 5
unit process is referred to as the liberation size. Liberation size does not imply pure mineral species, but rather an economic trade-off between grade and recovery. Obviously the finer the liberation size for a particular mineral, the finer the ore must be ground, resulting in higher energy consumption. The Bond equation is widely used to estimate the energy required for grinding, and has the form: E = WI(10/ P80 where
10/ F80)
E = grinding energy (kWh/t) WI = Bond grinding Work Index (kWh/t) P80, F80 = 80% passing size of product and feed respectively ( m)
Bond (ball mill) Work Indices for copper and nickel sulphide ores are typically in the order of 15 kWh/t. Fig. 10 shows how the energy required to grind these ores increases as the liberation or grind size (P80) decreases, based on the Bond equation above with an F80 of 5000 m (5 mm).
Grinding energy (kWh/t ore)
70 60 50 40 30 20 10 0 1
10
100
1000
10000
80% pas s ing s ize (um )
Fig 10. Effect of mineral grind (liberation) size on grinding energy.
Grind sizes for copper and nickel sulphide ores in Australian mineral processing plants are currently in the order of 75100 m [11, 12], so it was assumed that the energy consumption included in the mineral processing stage of the earlier LCAs for each of these metals [5] corresponded to a grind size of 75 m. Fig. 10 was then used to estimate the increase in the energy consumption of this stage as the grind size was progessively reduced from 75 m to 5 m, and the revised energy estimate was included in the respective LCAs. Figs. 11 and 12 show how the embodied energy and greenhouse gas emissions increase as both ore grade and grind size decrease for copper metal produced pyrometallurgically, while Figs. 13 and 14 show the corresponding results for nickel production. It should be noted that fine-grained ores are not necessarily low grade. On the contrary, some high grade ores are fine-grained, such as the McArthur River lead-zinc deposit in the Northern Territory.
Embodied energy (MJ/kg Cu)
500 450
75 um
400
25 um
350
10 um
300
5 um
250 200 150 100 50 0 0
0.5
1
1.5
2
Ore grade (%)
6
2.5
3
3.5
Fig 11. Effect of ore grade and grind size on embodied energy for copper production.
GWP (kg CO2 equivalent/kg Cu)
50 75 um 40
25 um 10 um
30
5 um
20
10
0 0
0.5
1
1.5
2
2.5
3
3.5
Ore grade (%)
Fig 12. Effect of ore grade and grind size on greenhouse gas emissions for copper production.
Embodied energy (MJ/kg Ni)
700 75 um
600
25 um 500
10 um 5 um
400 300 200 100 0 0
0.5
1
1.5
2
2.5
Ore grade (%)
Fig 13. Effect of ore grade and grind size on embodied energy for nickel production.
GWP (kg CO2 equivalent/kg Ni)
70 75 um
60
25 um 50
10 um 5 um
40 30 20 10 0 0
0.5
1
1.5
2
2.5
Ore grade (%)
Fig 14. Effect of ore grade and grind size on greenhouse gas emissions for nickel production. 7
Figs. 11-14 show that as reserves of these metallic ores deteriorate in both grade and grind (liberation) size, the embodied energy and greenhouse gas emissions associated with the production of copper and nickel by pyrometallurgical processing will increase significantly. For example, a decrease in ore grade from 3.0% to 0.5% and grind size from 75 m to 10 m will increase both the embodied energy and greenhouse gas emissions for copper production by about 500% (to 199 MJ/kg and 19.1 kg CO2e/kg respectively). A similar reduction in grind size for nickel ore with a reduction in grade from 2.3% to 0.5% will increase both the embodied energy and greenhouse gas emissions for nickel production by about 160% (to 300 MJ/kg and 29.3 kg CO2e/kg respectively). The proportional effect of ore grade and grind size is greater for copper than nickel because of the greater contribution of the mining and mineral processing stages for copper compared to nickel as shown in Figs. 6 and 7. While not considered in this paper, the hydrometallurgical production of copper and nickel metal was previously shown [5] to have higher embodied energies and greenhouse gas emissions than pyrometallurgical production for typical grades of ores treated by these processes in Australia. However, given the different relative contributions of the various processing stages for the hydrometallurgical route compared to that shown above for the pyrometallurgical route, the effect of ore grade and grind size are likely to be numerically different to that shown above, although the trends will be broadly similar. 4. MITIGATING THE EFFECT OF DETERIORATING ORE RESERVES While it is inevitable that ore reserves will deteriorate (both in ore grade and grind size) over time, there are a number of possible approaches to help mitigate the energy and greenhouse gas impacts expected from such a change. Some of these approaches are aimed at reducing the embodied energy of metal production, which in turn will reduce greenhouse gas emissions as the energy inputs to these processes are largely fossil fuel-based. Other approaches are aimed at reducing greenhouse gas emissions without necessarily reducing embodied energy. One obvious way of addressing the problem of deteriorating ore reserves is to reduce the demand for primary metal to be produced from these reserves in the first place. Dematerialisation and recycling (ie. secondary metal production) as mentioned earlier, will both help in achieving this goal, however recycling is only possible for metals used in nondissipative applications where the metals can be economically reclaimed. The approaches outlined below are focussed mainly on primary metal production.
4.1 Reducing energy consumption Given the significant contributions of the mining and mineral processing stages to the overall embodied energy and greenhouse gas emissions as ore grade and grind size decrease, with comminution (ie. size reduction primarily grinding) accounting for the major part of this contribution, substantial reductions in these impacts can be achieved by reducing comminution energy. Some possible approches are: comminute less material
ore sorting, pre-concentration, improved mining practices to reduce dilution by waste;
comminute more efficiently optimising the design of comminution circuits including process control, use of more energy-efficient comminution equipment, eg. high pressure grinding rolls, stirred mills; doing less comminution less liberation of the value mineral to achieve higher recovery at the expense of lower concentrate grade (ie. more metal and gangue to be separated in smelting and refining stages); direct smelting of ore done;
this is the extreme case of the preceding approach where no (or very little) comminution is
more comminution in the blasting (mining) stage; improved energy efficiencies (eg. waste heat recovery) of existing metal extraction and refining processes; development of more energy-efficient metal extraction and refining processes.
4.2 Reducing greenhouse gas emissions It is possible to reduce greenhouse gas emissions (both direct and indirect) even with no reduction in energy (nonrenewable) consumption. There are a number of ways in which this can be achieved, some of which are: improved efficiency of electricity generation using fossil fuels, eg. co-generation and advanced power cycles; electricity generation from renewable energy sources, eg. biomass, hydroelectricity; 8
use of renewable energy sources such as biomass as direct fuel inputs into processes; sequestration of carbon dioxide emissions. 5. CONCLUSIONS The anticipated on-going demand for primary metal production well into the future means that the world s ore reserves will continue to be depleted. While the potential for the discovery of new high grade reserves exists, it is almost inevitable that ore reserves will deteriorate over time as the higher grade reserves are exploited first. As ore grades fall, the embodied energy of primary metal production and the associated greenhouse gas emissions will increase, particularly at ore grades below 1%, as a result of the additional energy that must be consumed in the mining and mineral processing stages to move and treat the additional gangue material. The downstream metal extraction and refining stages are virtually unaffected by ore grade as a relatively constant grade concentrate is produced in the mineral processing stage, irrespective of initial ore grade. This increase in embodied energy and greenhouse gas emissions with falling ore grades will be more significant for those metals where the mining and mineral processing stages make substantial contributions to the cradle-to-gate part of the metals life cycle, eg. copper produced pyrometallurgically. Another potential consequence of deteriorating ore reserves is need for finer grind sizes in order to achieve mineral liberation, as many of these lower grade reserves are fine-grained. This will increase the energy required for comminution in the mining and mineral processing stages, and will in turn exacerbate the effect of falling ore grades. There are a number of possible approaches that can be taken to help mitigate the above impacts of deteriorating ore reserves. The most promising approaches are aimed at reducing the embodied energy of metal production, as they will also reduce greenhouse gas emissions, while other approaches are aimed at reducing greenhouse gas emissions without necessarily reducing embodied energy. The former largely involve reducing the amount of material and energy going into comminution, while the latter involve improving electricity generation efficiences, the use of renewable energy and carbon dioxide sequestration. An extreme case of the former is to smelt the ore directly, with no (or very little) comminution beforehand. This will obviously involve a trade-off between the additional energy required for smelting and the reduced or eliminated energy for comminution. Other economic issues will also influence the feasibility of this approach. The direct smelting of low grade/fine-grained ores is the subject of on-going research at CSIRO Minerals. 6. ACKNOWLEDGEMENT The work presented in this paper was undertaken under the auspices of the Centre of Sustainable Resource Processing which is established and supported under the Australian Government Co-operative Research Centre program. 7. REFERENCES 1 Rankin, W. J. The future of metals in sustainable development. Proceedings John Floyd International Symposium on Sustainable Developments in Metals Processing, Melbourne July 2005, pp. 449-463. 2 Ayres, R. Resources, scarcity, (http://ec.europa.eu/environment/enveco/waste/ayres.pdf)
growth
and
the
environment
3 Skinner, B. A second iron age ahead ?. American Scientist, May-June 1976, 258-269. 4 Norgate, T. E., Jahanshahi, S and Rankin, W. J. Assessing the environmental impact of metal production processes. Submitted to Journal of Cleaner Production. 5 Norgate, T. E. and Rankin, W. J. Life cycle assessment of copper and nickel production. Proceedings MINPREX 2000, Melbourne, 2000, pp. 133-138. 6 Chapman, P F and Roberts, F. 1983. Metal Resources and Energy, Butterworths. 7 Chapman, P F. 1974. The energy costs of producing copper and aluminium from primary sources, Metals and Materials, 8 (2) pp 107-111. 8 Kellogg, H H. 1974. Energy efficiency in the Age of Scarcity, Journal of Metals, 26(6), 25-29. 9 Giurco, D., Stewart, M. and Petrie, J. Decision making to support sustainability in the copper industry: technology selection. Proceedings 6th World Congress of Chemical Engineering, Melbourne, 2001. 9
10 Lunt, D., Zhuang, Y and La Brooy, S. Life cycle assessment of process options for copper production. Proceedings Green Processing Conference, Cairns, 2002, pp. 185-193. 11 Lumsdaine, I and O Hare, S. Copper concentrator practice at Mount Isa Mines Limited, Mt Isa, Qld. Australasian Mining and Metallurgy (J Woodcock and J Hamilton, Editors), 1993, AusIMM, pp. 649-651. 12 Wright, P. Nickel ore concentration at Leinster Nickel Operations of Western Mining Corporation Limited, Leinster, WA. Australasian Mining and Metallurgy (J Woodcock and J Hamilton, Editors), 1993, AusIMM, pp. 1189-1192.
10
