A great number of research, innovation and policy actions in the mineral sector have been initiated in recent years within the EU and in Finland1. These actions aim to improve the opportunities for sustainable growth in the mineral sector, for creating more jobs in Europe and for securing the raw material needs of European industries. In addition, Finnish on-going or recent programmes, such as the Tekes Green Growth and Functional Materials programmes and the FIMECC SHOK, support the development of a more sustainable and technologically stronger mineral sector.
What is missing from the arena is a way to link the intensive R&D&I efforts that have been put on the upstream end of the mineral value chain – the mine – and on the downstream end – the end-of-life products. This missing link emerges in part from a natural discontinuity in the mineral value chain: mineral commodities are traded on the global market place, and in practice there is little vertical integration from primary raw materials to final products. The value chain discontinuity applies to commodities, e.g. nickel or copper, which have a long history of volume-intensive end use applications and well defined global specifications. When, however, considering raw materials that in the future are expected to become business critical for a number of industries, the entire innovation ecosystem in the mineral sector requires further development. Downstream in the value chain, substituting business-critical raw materials in applications calls for a completely new design philosophy, which broadens the design problem into understanding the origin and recyclability of potentially cheaper and less critical materials, while ensuring the functionality (i.e. the high technology features) and ecological sustainability of the products with the new substitute materials. When no substitutes exist, increasing attention needs to be paid to the recovery of small concentrations of a great number of valuable metal raw materials (Salminen et.al. 2014). Similarly, at the upstream end, the large volumes of side-streams that are currently being landfilled at mines and around processing sites contain a number of business-critical raw materials. Compared to just a few decades ago, these materials are now being given substantial interest, with efforts to understand their end-use and the concomitant business logic for efficient recovery.
This article addresses the recovery potential of a number of critical and valuable raw materials through a hybrid system combining mechanical separation, bio and hydrometallurgical processes as well as biosorption. For the sake of simplicity, we use the example of a printed circuit assembly (PCA) which contains the major components attached on the surface of the printed circuit board (PCB) for the recovery of gold, copper and plastics. The basic flow sheet (Figure 1), however, can be applied to a number of different flows and end product categories, including consumer goods, industrial intermediates, process side streams and tailings. All recovery processes include pre-treatment stages, concentrate production and extractive metallurgy. Both pyrometallurgy and hydrometallurgy (including biohydrometallurgy) may be applied depending on the composition and mineralogy of the feed.
This paper is constructed so as to follow the logic of the flow sheet in Figure 1. First, we describe the dismantling, crushing and flotation segments of the process. Second, we proceed to describing the recovery of copper through bioleaching from the light fraction, as well as the final treatment of the residual organic and inorganic flow. Third, we describe the process of recovering gold with a non-cyanide leaching process combined with biosorption and solvent extraction. Fourth, and finally, we describe some preliminary concepts of recovering copper from both the heavy and light fractions and conclude with recommendation for the future use of the process model.
Pre-treatment of PCA
Due to their heterogeneous nature, waste materials are rarely suitable for recovery processes as such. Mechanical pre-treatment is needed to produce concentrated fractions that can be processed further. We selected mobile phones as case material because of the high level of metals concentration in their PCAs. The dominant metal in PCA is copper, with concentrations of between 15 % and 27 % depending on the PCA’s host device (Ogunniyi and Vermaak, 2009; UNEP, 2013). In our study, the copper concentration was roughly 25%. In addition, high precious metal concentrations are present in sophisticated and complex applications such as mobile phones and computers (UNEP, 2013). Metals such as silver, gold and palladium are used in connectors in order to increase the conductivity and to reduce oxidation (Luda, 2012).
Mechanical tests included unit processes such as crushing, sieving, magnetic separation, eddy current separation and sensor-based separation as well as manual sorting. As a result of the pre-treatment of mobile phones, a PCA-rich fraction corresponding to slightly above 20% of the feed quantity was produced. During the mobile phone experiments, it was observed that manual sorting produced better quality PCA than mechanical processing. Therefore, further mechanical treatment of PCA by flotation was carried out on the manually sorted PCA fraction. Prior to flotation, PCAs were crushed into a grain size below 250 µm. In addition, the liberation degree of metals was increased by crushing. This is essential for efficient separation. It avoids the risk of hybrid-particle-aggregates with many different mixed materials separating as such and not as individual component fractions. The flotation experiments focused on separating plastics that contain e.g. antimony, bromines and chlorides, which cause corrosion problems later especially in the pyrometallurgy process and which add to the impurities of the hydrometallurgical process.
In the flotation process, hydrophobic particles are separated from hydrophilic particles by blowing air into a sludge containing both types of particles. Hydrophobic particles are picked up by air bubbles and travel to the top of the sludge and can then be skimmed off mechanically. The experiments were conducted with a full factorial design where the variable parameters were agitation speed, aeration rate and pulp density. Based on the full factorial experiment design including three centre points, 11 experiments were carried out. During each experiment lasting 30 minutes, five sub-samples were taken from the froth. The efficiency of flotation was evaluated by enrichment ratios and recoveries of monitored elements (Cu, Au, Cl and Si) over the flotation time.
The froth fractions produced were analysed with a portable XRF analyser in order to determine the efficiency of the flotation by recovery rates and enrichment ratios. The enrichment ratio versus the recovery of copper is presented in Figure 2.
Figure 2. Grade versus recovery of Cu. The enrichment ratio versus the recovery of copper in the sink metal concentrate fraction.
Figure 2 shows that experiments 8 and 6 were able to enrich the copper concentration more than 1.5 times the concentration in the feed. The copper concentrations of experiment 8 were approximately 43% in the metal concentrate and around 9% in the froth. Both experiments 6 and 8 with the highest enrichment ratio had a high pulp density and agitation speed, which indicates a strong effect on the enrichment. The effect of aeration was not as strong as for the enrichment ratio; however, it improved the recovery rate. Experiment number 4 was clearly non-selective in relation to copper compared with the other experiments. The low agitation and high pulp density and aeration probably produced unfavourable circumstances for selective flotation, at which time copper particles ended up both in the froth and metal concentrate. Comparing the copper recovery rates and enrichment ratios of this study with the results obtained by Ogunniyi (Ogunniyi and Vermaak, 2009), the enrichment ratios were slightly lower in this study. However, the recovery rates in this study were considerably higher. One reason for the difference may be explained by the different content, particle size and shapes of the feed as well as the parameters used.
The enrichment ratios and recovery rates for Au, Cl and Si are presented in Table 1.
The flotation experiments showed that metals were enriched with froth flotation to the metal concentrate while most of the plastics, resins and other harmful elements were separated to froth. The full factorial flotation experiments revealed that the most optimal copper enrichments were obtained with a pulp density of 20%, an agitation speed of 1,200 rpm, and an aeration amount of 3,000 ml/min, at which time copper could be enriched from approximately 25% to 45% with a recovery of 85%.
Table 1. Enrichment ratios (ER) and recovery rates for copper, gold, chlorine and silicon.
Bioleaching of copper and side stream treatment
According to elemental analysis, the PCA froth still contained high levels of copper (153 g/kg). Therefore, additional treatment and metal recovery operation were needed. Because of its low-cost and low technology requirement, acid bioleaching, was combined with low hazardous emissions. Moreover, promising results on the bioleaching of printed circuit boards have been presented in the literature (Liang et al. 2013; Liang et al. 2010; Xiang et al. 2010; Yang et al. 2009; Zhu et al. 2011).
Acid bioleaching technology relies on autotrophic bacteria, such as Acidithiobacillus thiooxidans and At. ferrooxidans, which are well known for their ability to oxidize reduced sulphur compounds or elemental sulphur so as to produce sulphuric acid. In addition, At. ferrooxidans and Leptospirillum ferrooxidans are known to oxidize iron from the ferrous to ferric state (Sand et al. 2001; Watling, 2006). Even though WEEE and PCA are very unconventional materials for autotrophic microorganisms, the leaching chemistry shown in equations 1–4 has been observed (Liang et al. 2013; Xiang et al. 2010; Yang et al. 2009; Zhu et al. 2011).
According to these equations, a simple bioleaching monitoring method was established for measuring the pH, oxidation-reduction potential (ORP) and dissolved Fe2+ and Cu2+ concentrations. As PCA froth tends to create a neutral-pH and rather strongly buffered solution with water, the thriving of sulphur oxidizing microorganisms and production of sulphuric acid (equation 1) reduces the pH and keeps the solution acidic. On the other hand, when iron oxidizing microorganisms are thriving, they oxidize Fe2+ to Fe3+ (equation 3), resulting in a severe increase in ORP and a decrease in Fe2+ concentration.
Bioleaching experiments of PCA froth with acidophilic culture proceeded from the adaptation of microorganisms to preliminary experiments, followed by optimisation of leaching parameters and finally a test run with a scaled up bioreactor. In most of the experiments abiotic sulphuric acid leaching experiments were also carried out in order to verify whether microorganisms improved the leaching rate of copper. Mixed acidophilic culture, enriched from a sulphide ore mine site (Halinen et al. 2009) and containing At. ferrooxidans, At. thiooxidans/albertensis, At. caldus, L. ferrooxidans, Sb. thermosulfidooxidans, Sb. thermotolerans and some members of the Alicyclobacillus genus was used.
The adaptation of the culture for PCA froth and subsequent preliminary bioleaching experiments were conducted in the presence of modified K9-media, 10 g/l S0 and 4.5 g/l Fe2+. Due to the unconventional material, strong sulphuric acid was introduced to reach pH 2, which is optimal for the majority of the bioleaching microorganisms (Rawlings, 2002). It was seen that microorganisms were able to maintain the pH or even further acidify the sample and simultaneously cause a rapid increase in ORP and a collapse in Fe2+, illustrating the thriving of both sulphur and iron oxidizers. However, copper concentrations were lower than with abiotic sulphuric acid leaching and the kinetics slower. The maximum tolerated PCA pulp density was about 50 g/l, which is well in line with the literature (Liang et al. 2010; Xiang et al. 2010; Zhu et al. 2011).
To increase the recovery rate and leaching kinetics of copper, bioleaching parameters were optimised. Fe2+ concentrations of 0, 4.5 and 9.0 g/l were introduced to a system with a PCA froth pulp density of 20 g/l in the presence of modified K9 media and 10 g/l S0. The copper concentrations achieved in the bioleaching solution according to 0, 4.5 and 9.0 g/l of Fe2+ and abiotic sulphuric acid leaching were 1.6 g/l, 2.2 g/l, 2.6 g/l and 2.6 g/l respectively. It is clear that ferric iron attack plays a vital role in the rapid dissolution of copper, and bioleaching can outrun the abiotic sulphuric acid leaching when the process is operated at around pH 2. However, as seen from equations 1 and 3, addition of extra Fe2+ also consumes extra sulphuric acid. Optimum process conditions need to be determined so that ferric iron and acid attack give the best results with minimum chemical costs.
The bioleaching system was further optimised with a scaled up bioreactor (3 litre CSTR-type). Production of ferric iron and sulphuric acid was separated from the treatment of PCA froth by cultivating the microorganisms in a two-step mode; the first step favoured the sulphur oxidizers in the presence of modified K9 media, 2.5 g/l S0 and 0.4 g/l Fe2+; and second step produced ferric iron with the addition of 7.8 g/l Fe2+. The new cultivation procedure produced a leaching solution with parameters of pH 1.1, ORP of +865 mV (SHE) and 7.4 g/l Fe3+. This solution was used to treat PCA froth with a pulp density of 50 g/l, resulting in a 3 day copper dissolution of 99%. During the actual leaching, the pH rose to 1.6, where it was maintained with strong sulphuric acid. Bioleaching was found to be a very specific method for copper leaching, as the only major metallic elements in the solution were copper and iron, 6.8 g/l and 7.0 g/l, respectively. Therefore, later recovery of copper from bioleaching solution should be rather simple and economical.
Gold recovery through leaching and biosorption
The heavy fraction from the PCA flotation process contains a high concentration of valuable metals for which there are well proven mainstream recovery processes. The gold process, however, currently requires the use of cyanide, which makes recovery both costly and prone to environmental risk. To complete the PCA recovery route with sustainable alternatives, we ran trials with non-cyanide leaching and biosorption. It has recently been shown that certain biomasses can bind heavy metals and gold due to either active cell defence mechanisms or negatively charged cell walls structures. In spite of these promising results, however, traditional biosorption methods utilising freely-suspended biomass is inadequate for industrial applications, mainly due to challenges in the separation of the biomass from the treated solution. Therefore, research has moved on to immobilisation of the biomass into different kinds of immobilizing matrices (Das, 2010; Khoo & Ting, 2001). Our approach was to take advantage of the ability of certain fungi to produce thick filamentous meshes which can cause self-immobilization of the biomass to a certain form, suitable for industrial applications.
Self-immobilized mats of P. chrysosporium with an average weight of 0.05 g and a diameter of 5 cm were placed on filtration columns and 20 ml of two mine site waters (named as MW1 and MW2, concentrations presented in Table 2) were run freely through the mats. Filtrates were analysed with ICP-MS and ICP-AES for metals and sulphate.
Table 3 shows the treated mine site waters and removal efficiencies. Surprisingly, self-immobilized mats of P. chrysosporium were adsorbing unselectively all the elements studied with an efficiency of 86 – 92 %, despite different initial concentrations and pH values. When treated, mine site waters are compared to Metal Mining Effluent Regulations for effluent discharge limits for metal mines (maximum authorized monthly mean concentration, mg/l) (Environment Canada, 2012), it is seen that MW2 meets the requirements, but MW1 for Cu and Zn does not. Therefore, when treating very badly contaminated waters, some pre-treatment operations are required (e.g. precipitation with lime).
Biosorbents have been reported to be selective for gold and other precious metals (Das, 2010). Encouraged by this, a self-immobilized mat of P. chrysosporium (dry biomass weight 0.53 g) on a fabric bottom layer was placed on filtration columns (d = 5 cm), and 60 ml of artificial chloride-hypochlorite gold process leachate (composition presented in Table 4) was run freely through the mat. Filtrate was analysed with ICP-MS and ICP-AES for elements and sulphate.
Table 2. Concentrations (mg/l) and pH in studied mine site waste waters MW1 and MW2
Table 3. Concentrations (mg/l) in treated mine site waters and removal efficiency (RE %). MMER: Metal Mining Effluent Regulations for effluent discharge limits for metal mines (maximum authorized monthly mean concentration, mg/l) (Environment Canada, 2012)
Figure 3 presents the removal degree of elements when an artificial chloride-hypochlorite leachate was filtered through a self-immobilized mat of P. chrysosporium. Gold was seen to adsorb effectively on the biomass, while all the other elements were poorly adsorbed. The gold adsorption capacity of P. chrysosporium mat was approximately 32 mg/g biomass (0.16 mmol/g), which is in line with the capacities of biosorbents reviewed by Das (2010). However, this particular biosorption application neglects drawbacks observed with freely-suspended systems.
Figure 3. Removal of elements by P. chrysosporium from artificial chloride-hypochlorite leachate.
Conclusions and future steps
The increasing use of secondary raw materials including waste electrical and electronic equipment is an essential feature of sustainability in the metallurgical industry. The use of these secondary raw materials for metal extraction is driven by legislative, political, environmental and economic drivers. Metals have traditionally been a good source of wealth and income for the recycling industry due to their relatively simple treatment processes and high prices. For base metals, the recycling processes are firmly established and are approaching a closed recycling loop. Take the stainless steel of Outokumpu as an example: currently about 80%of the end product is comprised of recycled materials. Critical metals found in consumer products such as phosphors, electrical appliances, touch screens, permanent magnets and glass surfaces are, however, still to find suitable recycling systems.
In this article, we have presented a recycling system that combines new angles on mechanical pre-treatment with bioleaching and bioadsorption. The resulting metal recovery rates are significantly better than those with the traditional pyro-metallurgical pathway. However, with the efficient market processes for precious metals and copper in place, we see small concentration technology metals such as lanthanides as a key target for future hybrid recovery metals. The future need for lanthanide recovery is highlighted by the low substitutability of lanthanides in high-tech applications. A good example is neodymium, without which the permanent magnet in the hard drives of computers would not work nearly as efficiently as it does today. Another example is europium, which makes possible the red colour in flat screen televisions and fluorescent lamps. Neither of these elements can readily be substituted; nor are they currently recycled at all. In fact, a long term study conducted by Graedel et.al (2013) reveals that none of the 62 widely used metals or metalloids can be substituted in a way that would fulfil the requirements of their key applications. The problem is twofold: without a proper recycling system, the supply of raw materials for new high tech products cannot be guaranteed. At the same time, such a recycling system is not feasible without a better product design, collection schemes and process design. The solution to the problem requires systemic eco-innovation along the entire value chain. This includes the development of IPR governance, because leading material combinations are typically heavily protected by companies because of their great value and competitive edge.
The process described in this article is expected to be particularly suitable for the separation of plastics from end-of-life products and form intermediate metal products (like salts) for further treatment in a pyrometallurgical or hydrometallurgical facility. The solution purification and separation of metals from solution, including the selective separation of gold shown in this article, can be applied to different aqueous streams independent of the source. Hydrometallurgical separation methods are increasingly being used in extractive metallurgy. The bioleaching method presented in this paper illustrates the possibilities of continuous reactor leaching systems for treating side streams and wastes still containing rather high residual levels of valuable metals. By utilising biology, there is no need to apply extremely acidic environments with strong oxidizing agents in hydrometallurgy, as the efficient leaching conditions can be achieved by microorganisms from negative-value mineral wastes. As this kind of bioleaching method seems to be very effective for zero-valent metals, it can provide a variety of solutions for treating end-of-life products.
Table 4. Concentrations (g/l) in prepared artificial chloride hypochlorite leachate
1 The EU Raw Materials Initiative (2008), Finnish natural resource strategy (2009), the national mineral strategy (2010), the ad hoc list of critical raw materials for the EU (2010), the EU Roadmap for Resource Efficient Europe (2011), the Tekes Green Mining Programme (2011), the European Innovation Partnership (EIP) on Raw Materials (2012), the ERECON European Rare Earth Competence Network (2013), and the upcoming Knowledge and Innovation Community (KIC) on raw materials. The list is an example and not comprehensive.
European Commission, 2008: COM 699.
Graedel, T. E., Harper, E.M., Nassar, N.T. and Reck, B.K. 2013. On the materials basis of modern society. PNAS: doi: 10.1073/pnas.1312752110
Halinen, A-K., Rahunen, N., Kaksonen, A.H., Puhakka, J.A., 2009. Heap bioleaching of a complex sulfide ore Part I: Effect of pH on metal extraction and microbial composition in pH controlled columns. Hydrometallurgy, 98, 92–100
Liang, G., Mo, Y., Zhou, Q., 2010. Novel strategies of bioleaching metals from printed circuit boards (PCBs) in mixed cultivation of two acidophiles. Enzyme Microb. Technol. 47, 322-326
Liang, G., Tang, J., Liu, W., Zhou, Q., 2013. Optimizing mixed culture of two acidophiles to improve copper recovery from printed circuit boards (PCBs). J. Hazard. Mater. 250-251, 238-245
Luda, M.P., 2011. Recycling of Printed Circuit Boards, in: Kumar, S. (Eds.), Integrated Waste Management II. ISBN: 978-953-307-447-4. < http://www.intechopen.com/books/integrated-waste-management-volume-ii/recycling-of-printed-circuit-boards > (accessed 17.03.14)
Ogunniyi, I. and Vermaak, M. (2009). Investigation of froth flotation for beneficiation of printed circuit board comminution fines. Miner Eng., vol. 22, p. 378–385
Rawlings, D.E., 2002. Heavy Metal Mining Using Microbes. Annu. Rev. Microbiol. 56, 65–91.
Salminen, J., Virolainen, S., Kinnunen, P. and Salmi, O. (2014): Sustainable Mining, Metals Processing and Recovery. In: Chemical Processes for Sustainable Future. Royal Society of Chemistry. In Press.
Sand, W., Gehrke, T., Jozsa, P-G., Schippers, A., 2001. (Bio)chemistry of bacterial leaching – direct vs. indirect bioleaching. Hydrometallurgy, 59, 159–175
UNEP, 2013. Metal Recycling: Opportunities, Limits, Infrastructure. A Report of the Working Group on the Global Metal Flows to the International Resource Panel. Reuter, M. A.; Hudson, C.; van Schaik, A.; Heiskanen, K.; Meskers, C.; Hagelüken, C. ISBN: 978-92-807-3267-2 < http://www.unep.org/resourcepanel/Portals/24102/PDFs/Metal_Recycling_Full_Report.pdf > (accessed 17.03.14)
Watling, H.R., 2006. The bioleaching of sulphide minerals with emphasis on copper sulphides – A review. Hydrometallurgy 84, 81–108
Xiang, Y., Wu, P., Zhu, N., Zhang, T., Liu, W., Wu, J., Li, P., 2010. Bioleaching of copper from waste printed circuit boards by bacterial consortium enriched from acid mine drainage. J. Hazard. Mater. 184, 812–818
Yang, T., Xu, Z., Wen, J., Yang, L., 2009. Factors influencing bioleaching of copper from waste printed circuit boards by Acidithiobacillus ferrooxidans. Hydrometallurgy, 97, 29–32
Zhu, N., Xiang, Y., Zhang, T., Wu, P., Dang, Z., Li, P., Wu, J., 2011. Bioleaching of metal concentrates of waste printed circuit boards by mixed culture of acidophilic bacteria J. Hazard. Mater. 192, 614–619