Artículo 5. Economic Feasibility of the Co-production of Indium From Zinc Sulphide Using Bioleaching Extraction in Germany
Artículo 5. Economic Feasibility of the Co-production of Indium From Zinc Sulphide Using Bioleaching Extraction in Germany
Artículo 5. Economic Feasibility of the Co-production of Indium From Zinc Sulphide Using Bioleaching Extraction in Germany
Hydrometallurgy
journal homepage: www.elsevier.com/locate/hydromet
A R T I C L E I N F O A B S T R A C T
Keywords: Bioleaching has been a well-established technique for the extraction of various metals since the 1990s. In
Bioleaching particular, leaching in continuously stirred tank reactors (CSTR) facilitates the rapid and environmentally
CSTR friendly extraction of different metals. This work investigates the economic feasibility of this technology when
Zinc
carried out using a suitable process and sulphide concentrate from a mining operation in the Erzgebirge region of
Indium
Economic feasibility
Germany. The process is based on assumptions and tests are pending. The focus is on recovering zinc, copper and,
in particular, indium from zinc sulphide. Based on the defined process characteristics of the bioleaching step as
well as down- and upstream processes, investment costs and operating costs of the necessary plant structure were
determined. After determining the potential profits based on given metal concentrations and current market
prices of zinc, copper and indium, a net present value (NPV) and the corresponding internal rate of return (IRR)
were calculated to ensure comparability with other investments. Both values are positive, showing that an in
vestment in the proposed bioleaching plant pays off even with low process volumes.
* Corresponding author.
E-mail addresses: [email protected] (N. Irrgang), [email protected] (M. Schlömann), [email protected]
(M. Höck).
https://doi.org/10.1016/j.hydromet.2021.105566
Received 17 August 2020; Received in revised form 16 November 2020; Accepted 27 January 2021
Available online 30 January 2021
0304-386X/© 2021 Elsevier B.V. All rights reserved.
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
deposit as part of the Hämmerlein-Tellerhäuser polymetallic deposit. Its requires energy and it may be an economic consideration whether and to
elementary composition was determined by inductively coupled plasma which size to grind. An agitator bead mill is suitable for achieving
mass spectrometry (ICPMS) after mineral digestion using aqua regia. comminution to a size of 20 μm.
The mineralogical investigations were carried out by X-ray diffraction In the bioleaching step, bacteria dissolve the different minerals.
(XRD) analysis. Cassiterite – tin oxide - is the main valuable mineral of Sphalerite is of particular interest for the planned process, as it is present
this ore and will be separated from the sulphides at an early stage, fol in high concentrations in the flotated material. Because of the copper
lowed by separation of magnetite from the sulphides. The sulphidic and the enclosed indium-containing sphalerite, the relatively small
concentrate produced and considered as representative was mainly amount of chalcopyrite is also of interest.
composed of sphalerite, i.e., approx. 65%, followed by arsenopyrite, After leaching, the solution from the last reactor would be forwarded
pyrite and chalcopyrite (see Table 1). In addition to tin, zinc and copper, to the extraction part, which in our model consists of thickening and
concentrations of arsenic, indium, bismuth, cadmium and silver were filtration of precipitated solids, like jarosite and arsenate, in a solid/
also found in this ore. Marble, mica schist and granite have been listed as liquid separation step, a liquid/liquid separation step and the solvent
wall rocks (Schuppan et al., 2012). According to the work of Matthias extraction itself including the stripping. Solvent extraction may be
et al. (2017), the indium is concentrated at the interface of the chalco performed in separate steps for copper, zinc and indium. It is followed by
pyrite and an unusual “sphalerite II" which occurs in addition to more the separate electrowinning of zinc, copper and indium in the next step.
usual “sphalerite I". Indium seems to have migrated from the chalco In all extraction parts, solid/liquid and liquid/liquid separation as well
pyrite into the sphalerite II, leading to very high-grade indium enclo as solvent extraction, recycling of the sulphuric acid for further bio
sure. Consequently, it is necessary to leach both chalcopyrite and leaching is possible.
sphalerite to maximize the recovery of indium.
Assuming 350,000 t of ore mined per year, about 3300 t of tin 2.2.2. Bioleaching process
concentrate, 63,000 t of iron concentrate, 52,000 t of road gravel and Bioleaching may be defined as the extraction of metals from ore,
17,500 t of sulphide concentrate could be obtained by processing concentrate or other solids using microbial activity. Microorganisms
(Saxore Bergbau GmbH, 2017). The remaining 60% of the unusable ore naturally occurring in mines, tailings and mining effluents are capable of
could be completely backfilled underground, with neither waste rock oxidizing iron and sulphur and of using them as sources of energy.
piles nor tailings ponds planned for above ground. The processing of Biogenic Fe3+ ions are powerful leaching agents, which further react
sulphide concentrate may be assumed to reach about 40 t per day with the sulphidic minerals in the ore (Schippers et al., 2014).
(Saxore Bergbau GmbH, 2017), which thus would be the amount that Industrial bioleaching is usually carried out as heap leaching or in
would be used in the planned bioleaching process. Assuming this daily continuous stirred tank reactors (CSTR). In the present case, CSTR op
quantity of 40 t of ore concentrate, a total of 6220 t of metal could be erations seem to be a more suitable method because of their high yields
extracted annually. The mass of indium at maximum process efficiency and short retention times compared to heap bioleaching. In fact, CSTRs
would be around 6.6 t per year (Matthias et al., 2017; Schuppan et al., are usually preferred for the processing of smaller volumes of valuable
2012). material (e.g. gold ore) at higher rates.
The composition (mineral as well as chemical) of the leaching ma
terial is crucial for the successful creation of a leaching process and the
2.2. Process description
definition of process parameters. After determining the material
composition, the selection of suitable microorganisms is an important
2.2.1. Process structure
step. As a rule, consortia containing at least an iron and a sulphur
To produce the sulphidic concentrate investigated the mined ore had
oxidizing microorganism are used. A more detailed selection of micro
been crushed classified and washed. In a next step, sensor based sorting,
organism depends largely on the process temperature and pH aimed at.
drying and comminution were carried out as well as separated using a
The hydraulic retention time will depend largely on the grain size of the
spiral separator. A wet ball milling circuit as well as magnetic separa
material which is also influencing the efficiency of the chosen micro
tion, gravity separation, hydro-classification, de-sliming dewatering and
organism. (Neale et al., 2009). Once a process with optimal parameters
sulphide flotation had then been used (Morgenroth et al., 2019). After
has been defined, the reactor (including agitators) can finally be plan
these steps the sulphidic concentrate may have a grain size of <300 μm.
ned. Process planning, however, is a very complex procedure – espe
As shown in Fig. 1, the further processing of the sulphidic concen
cially for large reactors. Most notably, the optimal distribution of the
trate can be separated into different steps, starting with further
introduced gases depends on a number of factors, such as the density and
comminution. Before bioleaching, the ore concentrate has to be ground
viscosity of the mixture, the size of the rising air bubbles and the reactor
to a sufficiently small particle size to ensure liberation of as much of the
geometry. Experiments carried out by, e.g., Neale et al. (2009), Gericke
minerals as possible. Here an average particle size of 20 μm was chosen,
et al. (2010), and Lotfalian et al. (2015) show the effectiveness of bio
in part as a result from other experiments in the literature (Carranza and
leaching in three-stage, continuously operated reactor systems with
Iglesias, 1998; Gericke et al., 2008; Norris et al., 2013). In addition,
solid-material contents of 10% and more. Higher solid-material contents
according to Matthias et al. (2017), MLA analysis has shown that the
could decrease the extraction rate and at the same time could tend to
sphalerite inclusions mentioned above have an average size of 20 μm.
lower the activity of the bacteria and inhibit leaching. In their research
Such grinding processes enhance ore suspension in the tank and the
into the bioleaching of chalcopyrite for the extraction of copper, Gericke
available surface area for reaction thus making enclosed minerals
et al. (2010) stated that stable leaching processes with solid contents of
accessible and speeding up the leaching process. However, grinding
up to 16% were possible. The solid content for industrial bioleaching, for
example technologies and processes used in copper bioleaching (Min
Table 1
tek/Bachtech) and Biox ®, is about 15% (Gericke et al., 2009). Other
Results of XRD analysis and the calculated proportions of elements contained in
experiments on smaller, laboratory scales – including those of Ahmadi
the sulphide concentrate.
et al. (2012) and Sandström and Petersson (1997) – processed even
Ore composition
higher solid fractions. Rawlings et al. (2003) and Morin (2007)
Sphalerite Arsenopyrite Quartz Chalcopyrite Pyrite Other mentioned a limit of 20% of solids for mesophilic bacteria. For calcu
62% 9% 6.6% 3.8% 1.6% 17% lation of the possible extraction quantities, the solid-material content in
Calculated proportion of elements the bioreactor in this study was set at 10%. This value should be
Zinc Copper Indium Arsenic applicable in practice over a longer period of time. With the definition of
45% 1.5% 500 ppm 1.6 %
the solids content, the hydraulic retention time, a theoretical bulk
2
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
3
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
Fig. 2. Bioleaching reactor setup of six connected CSTRs with two reactors in primary stage (PR 1–2) and four reactors in secondary stage (SR3 – SR6).
may damage the microbial cells (Gericke et al., 2008; Neale et al., 2009; 3. Economic feasibility
Norris et al., 2013).
The conditions in the primary stage (PR1 + PR2) probably will In the feasibility study presented in the following, the bioleaching
depend primarily on the amount of chalcopyrite in the concentrate step as well as all of the associated upstream and downstream processes
produced. If the process is to be carried out with suitable species at about were considered and calculated. All results of economic efficiency in this
45 ◦ C, the reactors of the primary stage must be operated with a chapter are based on a theoretically assumed process. The process steps
controllable redox potential to prevent passivation of the chalcopyrite in the flow sheet (Fig. 1) formed the basis for the economic evaluation.
mineral surface at low temperatures (Ahmadi et al., 2010; Wang et al., Calculations were carried out not only for the bioleaching process, but
2016; Zhao et al., 2015). In the other reactors, the redox potential will be also for the grinding of the concentrate and for the extraction and
higher from step to step. The pulp is transferred steadily by gravity from electrowinning plant in the downstream section. Cost estimates for some
one reactor to the next during operation. Due to the low process tem of the equipment such as the bioreactors, the agitators, the measuring
perature, only small quantities of evaporated water are expected – even technology and the agitator bead mill for grinding the concentrate were
in open reactors. In the first vessel, the addition of H2SO4 will be obtained from different companies located in Germany. Prices for the
necessary to maintain the pH value at the required low level. H2SO4 can construction of buildings and halls as well as for other specific plant and
be recycled from the raffinate after the separation and solvent extraction processing components were determined by means of online searches of
step. Further, the reactors are expected to be cooled by a cooling tower the websites of the respective manufacturers or were taken from Info
and heat exchanger (K). It is also possible to use hot and cold water – fed Mine USA Inc. (2015). However, the costs for conventional plants for
automatically into all reactors via measuring and control systems – to solvent extraction and electrowinning of the target metals as well as for
control the temperature of each individual reactor. dewatering the mixture were taken from the literature (Schlesinger et al.
In experimental descriptions from Falagán et al. (2017), a separate 2011a; Sinclair, 2005). Researched euro prices were converted into US
tank with only a small part of the total process volume is used for the dollars (USD) at a rate of 1:1.13 USD. The next chapter will provide an
initial cultivation and reproduction of the bacteria. However, adaptation overview of the necessary investments.
of the bacteria is necessary to ensure a functioning process. The reactors To calculate the required plant size for the solvent extraction and
are filled with the necessary liquids including a culture medium, and electrowinning section, a bioleaching efficiency of 90% was assumed.
adjusted to a pH value of about 2.0 or lower (Liang et al., 2014). The Furthermore, it was anticipated that the bioleaching plant was compa
time required for growth, unfortunately, cannot be predicted exactly. rable to similar chemical plants. This resulted in a 90 to 95% operating
In the primary stage, mainly sphalerite along with chalcopyrite is time over the calendar year and corresponded to 8000 working hours or
expected to be dissolved, and the highest extraction rate is likely to be 334 days of production per year (Towler and Sinnott, 2007). The cal
achieved. One important factor is a low redox potential, which can be culations were performed using data from Table 1 (the results of XRD
controlled in the reactors, for example, by varying the air supply and analysis and the fraction of elements contained in the sulphide ore)
agitator speed or by hydraulic retention time. Redox control at this scale, resulting in a total quantity of 5411 t of zinc, 180 t of copper and 6,01 t
especially the effects on reactors and gearboxes, is not tested yet. Control of indium of per year. Since the extraction and electrowinning processes
via the addition of chemicals is also a possibility, but would mean for zinc, copper and indium vary in their methodology and scale,
further purchases and higher costs. different electrowinning plants would be required which must be
4
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
calculated separately (Beukes and Badenhorst, 2009). The reference costs, therefore, a mark-up of 15% of the value of the direct costs was
values used were 1000 USD per tonne of copper and 1500 USD per tonne added. The rate of this additional charge is well established for con
of zinc per year (Schlesinger et al., 2011a; Sinclair, 2005). There was no struction services in various industries (Sinclair, 2005).
reference value for the extraction of indium. It was assumed that the cost The bioleaching step includes the reactors with the necessary heating
of indium electrowinning is similar to that of zinc production and that / cooling systems with piping, measuring and analysis technology as
both could be extracted in a similar plant (Fuller and Harb, 2018). well as a production hall. Prices for the bioleaching equipment were
In addition to the costs for acquisition and construction of all parts of requested directly from specialized companies. In this process, the
the plant, operational costs for regular operation also had to be taken calculated volume is 445 m3 for each reactor. The reactors would need
into account. Firstly, electricity and water would be required for the to be made of chrome‑nickel steel (AISI 304 L / AISI 316 L) as described
operation of all 4 sections. A large quantity of the electric power would by Morin (2007). Chrome‑nickel steel reactors designed by a German
be needed for the downstream extraction and electrowinning of the company with a volume of approx. 200 m3 could be sourced at a cost of
metals – processes that are also present in conventional, chemical-only 1,130,000 USD. Their extension to the required volume of 445 m3 was
extraction. Furthermore, a considerable amount of power would be considered to be feasible, but would require more precise planning, and
consumed by continuous stirring with agitators. The larger the reactor, would be associated with a price premium of 50%. The planned cost of a
the larger the dimensions of the agitator and, thus, the drive power single reactor is therefore approx. 1,695,000 USD. The overall cost for
required. Heating and cooling of the reactor contents, pumps, control six necessary reactors including all peripheral equipment represents
and measuring instruments, IT requirements and general consumers approx. 40% of the total direct costs. When the costs of the extraction
such as lighting and ventilation would also contribute to electricity and electrowinning steps – which account for about 43% of overall in
consumption. vestment – are considered, it becomes clear that bioleaching itself is the
The stirring, the supply of the concentrate and the pumping of the second most costly component.
mixture should take place automatically in order to constantly maintain The cost and volume of the bioleaching plant as well as the extraction
an optimal mixture with the necessary parameters. Associated human and electrowinning components depend directly on the amount of
resources would accordingly be limited to checking and adjusting the concentrate that is to be processed. The technical designs of the plants
parameters, maintenance work, process supervision and, perhaps, for recovery electrolysis differ slightly for the deposition of zinc, copper
manual transportation of materials and metals. The amount of human and other metals (Forner et al., 2018). However, the individual metals
labour required to operate the equipment in the extraction phase is only require different compositions of the cathodes and different
comparatively high. Schlesinger et al. (2011b) have stated that chemicals. In addition, the required volumes of the vessels differ
personnel costs for operation, maintenance and supervision account for considerably. For the separation of metals from impurities, it is neces
20% of the total operating costs of an electrowinning plant. sary to follow certain pre-processing steps. These include, for example,
filtration processes and separation by solid/liquid and liquid/liquid
3.1. Capital expenditures (CAPEX) separation. The capital cost of a solvent extraction plant with the rele
vant equipment is estimated at 600 USD per tonne of metal extracted per
Based on the process specifications, costs for all necessary equipment year for liquid/liquid separation and 70 USD per tonne of metal
were researched and are listed in Table 3. The calculations were divided extracted per year for solid/liquid separation. As already mentioned, the
into direct and indirect costs for the plant. In addition to the direct costs cost of a suitable plant for electrowinning is about 1000 USD per
for plant components, indirect costs for the construction and planning of annually produced tonne of copper and 1500 USD per annually pro
the plant were also included. Since indirect costs are always both indi duced tonne of zinc (Schlesinger et al., 2011b; Sinclair, 2005).
vidually and regionally dependent, they cannot be calculated with ab Together with average costs of 2054 USD per tonne of metal pro
solute accuracy. To provide a more realistic picture of the total CAPEX duced for the bioleaching plant and all other positions, the overall
Table 3
Capital expenditures for equipment in each of the process steps.
List of Requirements Size, length, volume, other Quantity Percentage of Cost in total
costs
(direct)
Milling Agitator bead mill 40 t concentrate/day, 300 μm to 20 μm incl. Periphery 2 2.02% 565,000 USD
Bioleaching Bioreactors 450 m3, steel (AISI 304 L/316 L), stirrer (210 kW) incl. 6 36.33% 10,170,000
USD
Measurement/Sensor pH, temperature, redox control, flow rate for reactors and 6 0.41% 115,300 USD
technology pumps
Cooling Adiabatic cooling, external 8 4.04% 1,130,000 USD
Aeration External fans 13 0.30% 83,800 USD
Extraction and Electrowinning Extraction Whole plant (solid/liquid + liquid/liquid) 1 13.36% 3,738,300 USD
1,2,3
Electrowinning Copper 1 0.64% 180,360 USD
Electrowinning Zinc / Indium 1 29.03% 8,125,220 USD
Logistics and Buildings3 Buildings Bioleaching, extraction, electrowinning (+ administration, 2 11.14% 3,119,140 USD
personnel, lab)
Transport Forklifts 6 1.30% 363,000 USD
Conveyer systems About 30 m length TBD 0.54% 150,080 USD
Electricity distribution Transformation, rectifier TBD 0.90% 251,150 USD
Direct costs (total) 100 27,991,295
USD
Indirect costs 4,198,694 USD
Construction, planning and execution, 15% of total direct costs
Total costs 32,189,989
USD
1
Schlesinger et al., 2011b.
2
Sinclair, 2005.
3
InfoMine USA Inc., 2015
5
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
investment costs add up to 5751 USD per tonne of metal produced per Table 5
year. Capital costs have to be added to the calculation, if capital is Electricity demand by process step.
borrowed. Depending on the chosen financing method and repayment Process step kWh per Tonne kWh per year Electricity
time, in reality the costs could increase at different rates. tonne per year costs per
In addition to varying operating costs, the estimated investment costs year
could change. While the costs of building an electrowinning plant are Grinding 1 60 kWh 13,360 t 801,600 kWh $ 108,857
easy to plan due to the widespread use of such plants in industry, bio per tonne
reactors suitable for metal leaching, on the other hand, and especially in of ore
Bioleaching 1 2525 5597.2 t 14,135,014 $ 1,919,535
Germany, are still niche products. The costs vary accordingly between a kWh per kWh
small number of suppliers, and reliable cost planning for the targeted tonne of
volumes cannot be guaranteed. A more precise conception and calcu metal
lation of the reactor size is only possible with further experiments in Extraction 2 160 kWh 5597.2 t 895,547 kWh $ 121,615
per tonne
pilot plants. With the help of further tests, it can be assessed whether the
of metal
planned volume of the reactors can be further reduced, thus reducing Electrowinning 714 kWh 180.36 t 128,777 kWh $ 17,488
investment costs. A reduction in volume can be achieved, for example, Copper 3 per t
through a higher proportion of solids in the leaching mixture. metal
Electrowinning 3338 5416.8 t 18,081,318 $ 2,455,443
Zinc/Indium 2 kWh per t kWh
3.2. Operating expenditures (OPEX) metal
“EEG-levy” on first – – 1000,000 kWh $ 61,500
In addition to costs for electricity and water consumption, operating MWh at 0.0615
USD per kWh
costs also include personnel costs as well as maintenance costs and
Total – – 34,042,257.00 $ 4,684,438
consumables. As detailed in Table 4, the cost estimation for the pro kWh
duction of metals in our scenario is about 1580 USD / t of metal pro 1
Power consumption given by producer.
duced or around 8.3 million USD per year when production reaches 2
Power consumption from Sinclair, 2005; Donati and Sand, 2007.
5600 t. 3
Power consumption from Schlesinger et al. 2011.
Personnel costs add up to 1.125 million USD per year and are based
on assumptions in the literature about the costs associated with oper
maintenance only change, if the entire plant is enlarged or reduced due
ating a conventional smelter (Schlesinger et al., 2011b). With average
to a change of capacity. The amount of particularly toxic waste to be
salaries of 45,000 USD, about 25 jobs could be created. The maintenance
disposed of also depends on the amount of concentrate processed. It
of the entire plant is expected to cost approx. 2 to 5% of the investment
could be based on the percentage of arsenopyrite. The exact amount of
costs per year (Towler and Sinnott, 2007). A factor of 5% was chosen
waste can currently not be estimated. The costs for proper disposal vary
and results in estimated maintenance costs of 1.6 million USD per year.
depending on region and company. Solidification/ stabilization was
Budget positions for water consumption (170,500 USD) and consum
selected as treatment method and costs were set to 200 USD per tonne of
ables as well as disposal services (667,000 USD) are much lower than the
arsenopyrite resulting in about 270,000 USD per year. There are a range
other positions. The largest component of the operating costs of around
of options available for disposal, such as disposal in landfills or stabili
4.7 million USD is attributed to electricity consumption. For a better
zation/solidification technologies (Clancy et al., 2013; Sullivan et al.,
overview and the calculation of estimated electricity demand, it is
2010).
helpful to separate consumption and costs for each process step (see
Table 5). In addition to a high electricity price in Germany, there are
special regulations and reduced energy tax levels for metal-producing 3.3. Cash flow calculation
companies that require some explanation. The average basic price of
one kWh in this calculation is 0.19 Euro. The tax reduction for metal- An important postulation for the calculation of the profits from the
producing companies leads to a price reduction of 0.0698 Euro per bioleaching and extraction steps is the respective efficiency of the pro
kWh after using the first GWh at the base price (Bundesamt für Justiz, cesses. For bioleaching, it was assumed that 90% of the metals are
2017). For every kWh after the first GWh, a price of 0.1358 USD (0,1202 brought into solution by the microorganisms (Brierley, 2008). For the
euro) is used. As this is only the case in Germany, electricity prices in extraction process and for electrowinning in particular, an efficiency of
other countries may vary widely and will probably lower the periodic between 95 and 97% was assumed over a longer period of time (Beukes
costs. and Badenhorst, 2009; Saba and Elsherief, 2000; Sinclair, 2005). To
While electricity and water costs as well as the quantity of consum facilitate simplified calculation, 100% is assumed for the latter part.
ables depend on the amount of ore processed, the costs for personnel and Because of the direct connection between the process efficiency and the
mass of extracted metals that can be sold at the end, efficiency is a
Table 4 critical factor when calculating possible earnings. At the end of the
Operational costs. extraction process, all of the metal will be sold for a fixed market price.
As for this calculation, all earnings were calculated with respect to a
Position Details Costs/ Costs/ t
year metal quality of 99.99% or higher and average prices over a period of 4 years
(2015 to 2019) for zinc and copper are used. Because of the high price
Electricity For detailed consumption see $ $ 945.00
Table 5 4,684,440
volatility associated with indium, however, which is partly due to the
Water 86,700 m3 per year $ 170,513 $ 30.50 monopolistic position of China, the price calculations are based on
Maintenance and 5% of investment sum per year $ $ 287.50 average market prices between February 2019 and January 2020
wear 1,609,499 (Bundesanstalt für Geowissenschaften und Rohstoffe, 2020). The
Personnel costs 25 positions, incl. Labour, $ $ 201.00
possible annual revenues from the sale of copper, zinc and indium –
laboratory, administration 1,125,000
Consumables and For example: nutrients, calcium $ 667,200 $ 119.00 calculated with an overall process efficiency of 90% – are shown in
services carbonate, laboratory equipment, Table 6.
disposal service Two values are considered to facilitate better comparison of the in
Total – $ $ vestment and its return. Firstly, the net present value (NPV) and, sec
8,256,651 1583.00
ondly, the internal rate of return (IRR) resulting from the investment.
6
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
Table 6 revenues by a further 2%. The NPV would increase to 68.1 million USD
Calculation of profits from metal sales (90% overall process efficiency, average and the IRR would be around 24%. A different high-price scenario, this
market prices for zinc, copper from 2015 to 2019; average indium price from time with indium prices exhibiting a stronger increase of, for example,
January 2019 to February 2020 from Bundesanstalt für Geowissenschaften und between 200 and 300% – resulting in price levels of around 600 USD/k,
Rohstoffe (2020)). as occurred in 2013 (Marscheider-Weidemann et al., 2016) – also shows
Metal Tonnes/year USD/t Revenue/year % of revenue the influence of the relatively low output of indium on revenue levels. In
Copper 180.36 $ 5811 $ 1,048,071 6.8% this case, the revenues from indium sales would reach between 2.9
Zinc 5410.8 $ 2477 $ 13,402,551 86.9% million USD/year (200% increase to 480 USD/k) to 3.9 million USD/
Indium 6.012 $ 160,310 $ 963,783 6.3% year (300% increase to 640 USD/k). Indium’s share of total revenues
Total 5597.17 $ 15,414,422 100,0%
would increase to between 19% and 25%, respectively.
However, a 20% drop in the market prices for the metals produced
General data for the calculation of exemplary figures for NPV and IRR, would result in earnings before interest, taxes, depreciation and amor
based on average market price data from 2015 to 2019 for zinc and tization (EBITDA) of approximately 4.1 million USD after the deduction
copper or 2019 to 2020 for indium, respectively, are shown along with of all operating costs. In this case, the NPV drops to 2.5 million USD and
the results in Table 7. The period of production was set to 20 years the IRR is equal to 1%. Because of the very low IRR, if other deductions
(Saxore Bergbau GmbH, 2017) and the interest rate, based on a high risk for real investments are taken into account, a possible investment would
investment, is set to 10%. certainly not be considered if compared with other alternative invest
Varying operational costs will have a considerable influence on the ment opportunities with lower risk.
cash flow as well as on the NPV and IRR. The cost breakdown in this
work offers a limited overview. On the one hand, costs have been ob 4.2. Low process-efficiency scenario
tained from companies based in Germany and may vary significantly in
other countries. On the other hand, some costs were only partially sur Due to the uncertainty surrounding the long-term process efficiency
veyed due to a lack of information or significant deviations from the of bioleaching in a plant of this size, it is important to know how varying
data. For example, the costs for the disposal of arsenic-containing ma efficiency could affect the economic outcome. Both the efficiency of
terials could change. At the time of writing this work, neither the exact metal extraction by microorganisms in the bioleaching process as well as
amount of hazardous waste nor a planned course of action for its the efficiencies of the solvent extraction and electrowinning processes
disposal had been determined. will influence revenues. Since the solvent extraction of various metals
and subsequent electrowinning are a common and proven processes
(Weigel, 1964), only the effect of the bioleaching process is discussed in
4. Scenarios
the low process-efficiency scenario.
A varying efficiency does have a direct impact on the amount of
In order to evaluate the robustness of the results, it is appropriate to
metal that is transferred into solution by the microorganisms. The much
analyze the effects of various relevant factors. In particular, the sales
higher volume of an industrial plant – in contrast to experimental or
prices and the overall efficiency of the planned process have a direct
pilot plants – is one of the main reasons why a process efficiency of more
influence on the profit. Hence, high- and low-price scenarios can help to
than 90% is difficult to achieve. One problem can be that the efficiency
assess the profitability of a plant when market prices are volatile. In
of the microorganisms working in the reactors can be lower because of
addition, it may also be helpful to analyze fluctuations in process effi
higher shear forces, which can be triggered by the greater effective ve
ciency and how they could affect the level of potential revenues. By
locity at the tips of the larger impellers. Other problems can result from
combining the two scenarios and taking into account the variable fac
the fact that it is much more difficult to control oxygen levels, solid
tors, approximate profit margins can be given.
contents, pH-values and other process factors in a higher volume and,
thus, to achieve a homogeneous mixture (Acevedo, 2000; Brucato and
4.1. High- and low-price scenarios Brucato, 1998).
With a theoretical efficiency of 100% for each of the process steps,
In order to explain price sensitivity, it is appropriate to calculate the one would extract 6220 t of metal consisting of 96.7% zinc, 3.2% copper
differences in the level of revenue due to fluctuations in the price of zinc, and 0.1% indium, with possible revenues of about 17.1 million USD. The
copper and indium. Because zinc accounts for more than 86% of pro assumption that the overall efficiency of the bioleaching process does
jected profit, changes in its market price will have the greatest influence not decrease with increasing plant size means that the proportion of
on revenues as well as on the final result. extracted metals remains relatively constant. In reality, however, the
The price changes in Table 8 are based on the metal prices in Table 6 extraction rates of the various metals may differ, which also would
(i.e., the calculation of profits from metal sales, Chapter 3.3) and an change the composition and the amount of revenue achievable. Among
amount of 5597 t of metal sold at 90% process efficiency. In a high-price the three metals of interest zinc is the one which probably is most easily
scenario – i.e., with a 30% increase in the metal price – zinc in particular leached (Natarajan, 2018), thus providing a solid basis for possible
would increase revenues by around 26% to approx. 20 million USD/ yields and revenues. If the calculated operating costs of 8.3 million USD
year. Price increases of 30% for indium and copper would each increase are used as the lower revenue limit, the break-even point is reached at
about 55% overall efficiency. The limit at which NPV and IRR tend to
Table 7 ward 0 is at around 70% efficiency when calculated at an interest rate of
General data for the calculation of the NPV and IRR. 10% over 20 years.
Periods/ Investment Operational Revenues/ Interest
However, if extraction of indium and copper would be low because of
duration costs costs/yearb year rate decreased leaching efficiency of chalcopyrite, it would reduce the
annual revenues. A leaching efficiency drop for copper and indium
20 years $ 32,189,989 $ 8,256,651b $ 10%
a
15,414,422c alone, for example to 60%, would reduce the IRR to 8.7%. If no copper
NPV $ 28,748,145 or indium can be extracted, revenues would be reduced to 13.4 million
IRR 10.72% USD/year. In this case, IRR drops to 4.5%.
a
Taken from Table 3.
b
Taken from Table 4.
c
Taken from Table 6.
7
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
Table 8
Revenues resulting from price changes of the metals produced in the range of - 30% to +30%.
Revenue (in USD, rounded) from respective metal assuming a change of metal prices by:
4.3. Combination of scenarios copper, and indium from a sulphidic concentrate obtained as by-product
of winning of tin. In order to plan and calculate the cost of a suitable
An overview of the changes in revenue at different levels of efficiency plant, it was necessary to define and describe the properties of an
combined with price changes of all metals from – 30% to +30% shows appropriate bioleaching process, which would, in theory, extract metal
the range of possible income from sales (Fig. 3). Depending on the ef with high efficiency. By defining the process and its inputs and outputs,
ficiency achieved, price changes would have a greater or lesser impact important data such as the reactor volume were obtained, which formed
on the total amount of revenue. In general, with lower process effi the basis for the design and cost planning of the entire plant. The
ciency, the break-even point will be reached on a much lower level if combination of capital expenditure and operating costs as well as the
prices for metals are higher. The corresponding NPV rises and falls in the expected profits allowed the NPV and IRR to be calculated. The costs for
same context and will reach a maximum, if both efficiency and market the bioleaching plant, including peripheral equipment, were procured
prices are high. through direct company contacts, and supplemented with costs for
To provide realistic examples of the variations, the NPV was calcu further plant components, such as electrowinning technology and
lated for two combined scenarios: additional logistics, using information available in the literature. Since
the plant calculation was based on very specific process parameters and
1. An overall process efficiency of 80% and a price change for all metals specifications, a direct transfer to plant designs in other countries and to
of +10%. This would lead to revenue of 15.1 million USD and an the processing of other mineral types may not be possible because of
NPV of 25.8 million USD. The IRR would be about 9.7%. differences in the regional market.
2. An overall efficiency of 70% and a price change for all metals of - However, the positive result - in particular the IRR of about 11% -
10%. This would lead to total revenues of 10.8 million USD and an shows that a bioleaching plant can be operated economically for pro
NPV of – 10.6 million USD. The IRR would fall below 0 to – 4.8%. cessing smaller quantities of concentrate on the European continent –
and even in Germany with its relatively high energy prices. Companies
The examples show that even small changes in process efficiency – normally have close to zero opportunities to exert an influence on either
combined with possible price fluctuations in the market – can have a the market or the resources in the ground. However, adjusting all other
major impact on projected profits and the NPV, which is a decisive factor steps in the production process can lead to a maximization of the NPV.
for investors. The best way to insulate the project from market fluctu As shown in the process scenario section, the efficiency of the bio
ations is to ensure high and stable process efficiency. leaching process and the variable prices of metals influence possible
revenues the most. Since the level of operating costs is largely influenced
5. Conclusion and outlook by the price of electricity, it would also make sense to consider the
possibility of a company producing its own electricity using wind or
This work focussed on the economic feasibility of a projected bio- solar power in a further research step.
hydrometallurgical processing plant for the production of zinc, To ensure a high metal extraction rate, experiments with higher
Fig. 3. Possible profit margin extrapolated from the data used for the calculation of the bioleaching plant with a combination of different efficiency levels and market
price changes for the produced metals with price decreases/increases from - 30% to +30%.
8
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
reactor volumes are of particular importance, as they would help to Brucato, Alberto, Brucato, Valerio, 1998. Unsuspended mass of solid particles in stirred
tanks. Can. J. Chem. Eng. 76 (3), 420–427. https://doi.org/10.1002/
confirm the efficiency of the bioleaching step with the factors selected
cjce.5450760311.
for the process defined in this work. Further studies to determine Carranza, F., Iglesias, N., 1998. Application of IBES process to a zn sulphide concentrate:
changes in the cost structure should also be carried out. Alternative effect of Cu2+ ion. Miner. Eng. 11 (4), 385–390. https://doi.org/10.1016/S0892-
energy concepts such as the use of renewable energy sources or the use 6875(98)00016-8.
Clancy, Tara M., Hayes, Kim F., Raskin, Lutgarde, 2013. Arsenic waste management: a
of waste heat from the reactors could reduce the estimated costs. A more critical review of testing and disposal of arsenic-bearing solid wastes generated
detailed analysis of cost components such as taxes, electricity and ma during arsenic removal from drinking water. Environ. Sci. Technol. 47 (19),
terial costs – which differ from location to location – can assist in 10799–10812. https://doi.org/10.1021/es401749b.
Conic, Vesna T., Vujasinovic, Mirjana M., Trujic, Vlastimir K., Cvetkovski, Vladimir B.,
obtaining a more realistic view of the possible realization of a bio- 2014. Copper, zinc, and iron bioleaching from polymetallic sulphide concentrate.
hydrometallurgical plant in Eastern Germany. Transactions of Nonferrous Metals Society of China 24 (11). https://doi.org/
Future research could provide a great deal of beneficial information, 10.1016/S1003-6326(14)63516-0. S. 3688–3695.
Falagán, Carmen, Grail, Barry M., Johnson, D. Barrie, 2017. New approaches for
such as, for example, a market analysis of possible demand and price extracting and recovering metals from mine tailings. Miner. Eng. 106, 71–78.
potential for indium produced in Germany or in the European Union in https://doi.org/10.1016/j.mineng.2016.10.008.
general. There is considerable potential surrounding the creation of in Forner, E.L., Scheepers, J., Du Toit, A.J., Miller, G.M., 2018. Copper electrowinning
circuit design: optimized costing as a function of cell arrangement, productivity,
centives for a pilot project to investigate the environmentally friendly rectiformer size, and throughput. J. S. Afr. Inst. Min. Metall. 118 (11) https://doi.
production of primary and secondary raw materials throughout org/10.17159/2411-9717/2018//v118n11a3. S. 1141–1148.
Germany. Fuller, T.F., Harb, J.N., 2018. Electrochemical Engineering. Wiley.
Gericke, M., Muller, H.H., van Staden, P.J., Pinches, A., 2008. Development of a tank
bioleaching process for the treatment of complex cu-polymetallic concentrates.
Declaration of Competing Interest Hydrometallurgy 94 (1–4), 23–28. https://doi.org/10.1016/j.
hydromet.2008.05.018.
The authors declare that they have no known competing financial Gericke, M., Neale, J.W., Van Staden, P.J., 2009. A Mintek perspective of the past 25
years in minerals bioleaching. J. South. Afr. Inst. Min. Metall. 109 (10), 567–585.
interests or personal relationships that could have appeared to influence Gericke, M., Govender, Y., Pinches, A., 2010. Tank bioleaching of low-grade chalcopyrite
the work reported in this paper. concentrates using redox control. Hydrometallurgy 104 (3–4), 414–419. https://doi.
org/10.1016/j.hydromet.2010.02.024.
InfoMine USA Inc., 2015. Minig Cost Service. Hg. V. InfoMine USA Inc, Washington,
Acknowledgement USA.
Justiz, Bundesamt für, 2017. Stromsteuergesetz § 9a Erlass, Erstattung oder Vergütung
This research did not receive any specific grant from funding der Steuer für bestimmte Prozesse und Verfahren. StromStG.
Liang, Chang-li, Xia, Jin-lan, Nie, Zhen-yuan, Xu, Bao-quan, Yu, Shui-jing, 2014. Effect of
agencies in the public, commercial, or not-for-profit sectors. initial pH on chalcopyrite oxidation dissolution in the presence of extreme
thermophile Acidianus manzaensis. Trans. Nonferrous Metals Soc. China 24 (6),
References 1890–1897. https://doi.org/10.1016/S1003-6326(14)63268-4.
Lotfalian, M., Ranjbar, M., Fazaelipoor, M.H., Schaffie, M., Manafi, Z., 2015. The effect of
redox control on the continuous bioleaching of chalcopyrite concentrate. Miner. Eng.
Acevedo, Fernando, 2000. The use of reactors in biomining processes. In: Electronic
81, 52–57. https://doi.org/10.1016/j.mineng.2015.07.006.
Journal of Biotechnology (Chile) Num.3 Vol.3 3. https://doi.org/10.2225/vol3-issue3-
Marscheider-Weidemann, Frank, Langkau, Sabine, Hummen, Torsten, Erdmann, Lorenz,
fulltext-4.
Espinoza, Tercero, Alberto, Luis, Angerer, Gerhard, et al., 2016. Rohstoffe für
Donati, E.R., Sand, W., 2007. Microbial Processing of Metal Sulfides. Springer,
Zukunftstechnologien 2016. Auftragsstudie. Hannover: DERA (DERA
Netherlands.
Rohstoffinformationen, p. 28.
Zhao, Hongbo, Wang, Jun, Yang, Congren, Hu, Minghao, Gan, Xiaowen, Tao, Lang, et al.,
Miller, P.C., Rhodes, M.K., Winby, R., Pinches, A., van Staden, P.J., 1999.
2015. Effect of redox potential on bioleaching of chalcopyrite by moderately
Commercialization of bioleaching for base-metal extraction. Mining, Metallurgy &
thermophilic bacteria. An emphasis on solution compositions. Hydrometallurgy 151.
Exploration 16 (4). https://doi.org/10.1007/BF03403233. S. 42–50.
https://doi.org/10.1016/j.hydromet.2014.11.009. S. 141–150.
Morgenroth, Henning, Bremerstein, Irina, Roscher, Marco, 2019. Eine Pilotanlage für die
Matthias, Bauer, Thomas, Seifert, Mathias, Burisch, Joachim, Krause, Nancy, Richter,
Aufbereitung eines Komplexerzes der Lagerstätte Hämmerlein/Tellerhäuser -
Jens, Gutzmer, 2017. Indium-bearing sulfides from the Hämmerlein skarn deposit,
Technik und Ergebnisse. In: Gesellschaft für Verfahrenstechnik UVR-FIA e.V.
Erzgebirge, Germany: evidence for late-stage diffusion of indium into sphalerite.
Freiberg, Helmholtz-Institut Freiberg für Ressourcentechnologie (HIF) am HZDR und
Miner Deposita 27 (7). https://doi.org/10.1007/s00126-017-0773-1. S. 18.
TU Bergakademie Freiberg (Hg.): Tagung 2019 “Aufbereitung und Recycling”.
Saxore Bergbau GmbH, 2017. Das neue Berggeschrey im Erzgebirge aus der Sicht der
Freiberg, 07.11. - 08.11, vol. 2019, p. 6.
Saxore Bergbau GmbH. Die Lagerstätte Hämmerlein/Tellerhäuser als neues
Morin, Dominique Henri Roger, 2007. Bioleaching of Sulfide minerals in continuous
Zinnerzbergwerk? https://www.vbgu.de/fileadmin/news/2017-10/Jahrestagung/S
stirredtanks. In: Donati, Edgardo R., Sand, W. (Eds.), (Hg.): Microbial processing of
axore_VBGU2017b.pdf (last access: 23.07.20201).
metal sulfides. Dordrecht. Springer, London. https://doi.org/10.1007/1–4020-5589-
Yunhua, Xiao, Xueduan, Liu, Jun, Fang, Yili, Liang, Xian, Zhang, Delong, Meng,
7_7. , S. 133–150. Online verfügbar unter.
Huaqun, Yin, 2017. Responses of zinc recovery to temperature and mineral
Natarajan, K.A. (Ed.), 2018. Biotechnology of Metals. Principles, Recovery Methods and
composition during sphalerite bioleaching process. AMB Express 7 (1). https://doi.
Environmental Concerns. Elsevier, Amsterdam, Netherlands.
org/10.1186/s13568-017-0491-1.
Neale, J.W., Robertson, S.W., Muller, H.H., Gericke, M., 2009. Integrated piloting of a
Bundesanstalt für Geowissenschaften und Rohstoffe, 2020. Preismonitor. Januar 2020.
thermophilic bioleaching process for the treatment of a low-grade nickel-copper
Hg. v. Bundesanstalt für Geowissenschaften und Rohstoffe.
sulphide concentrate. J. South. Afr. Inst. Min. Metall. 109 (5), 273–293.
Ahmadi, A., Schaffie, M., Manafi, Z., Ranjbar, M., 2010. Electrochemical bioleaching of
Norris, Paul R., Burton, Nicolas P., Clark, Darren A., 2013. Mineral sulfide concentrate
high grade chalcopyrite flotation concentrates in a stirred bioreactor.
leaching in high temperature bioreactors. Miner. Eng. 48, 10–19. https://doi.org/
Hydrometallurgy 104 (1), 99–105. https://doi.org/10.1016/j.
10.1016/j.mineng.2013.01.001.
hydromet.2010.05.001.
Rawlings, Douglas E., 1997. Biomining. Theory, microbes and industrial processes.
Ahmadi, Ali, Ranjbar, Mohammad, Schaffie, Mahin, Petersen, Jochen, 2012. Kinetic
Springer (Biotechnology intelligence unit), Berlin, New York.
modeling of bioleaching of copper sulfide concentrates in conventional and
Rawlings, Douglas E., Dew, David, Plessis, Du, Chris, 2003. Biomineralization of metal-
electrochemically controlled systems. Hydrometallurgy 127-128, 16–23. https://doi.
containing ores and concentrates. Trends Biotechnol. 21 (1), 38–44. https://doi.org/
org/10.1016/j.hydromet.2012.06.010.
10.1016/S0167-7799(02)00004-5.
Ballester, A., González, F., Blázquez, M.L., Gómez, C., Mier, J.L., 1992. The use of
Bernhardt, David, Reilly, I.I., James, F., 2019. Mineral Commodity Summaries 2019. Hg.
catalytic ions in bioleaching. Hydrometallurgy 29 (1–3), 145–160. https://doi.org/
v. Department of the Interior (DOI), U.S. Geological Survey (USGS). Online. https
10.1016/0304-386X(92)90010-W.
://prd-wret.s3-us-west-2.amazonaws.com/assets/palladium/production/atoms/files
Beukes, N.T., Badenhorst, J., 2009. Copper electrowinning: theoretical and practical
/mcs2019_all.pdf.
design. In: Heavy Minerals Conference - What Next? 20 - 23 September 2009,
Saba, A.E., Elsherief, A.E., 2000. Continuous electrowinning of zinc. Hydrometallurgy 54
Champagne Sports Resort, 57. Southern African Inst. of Mining and Metallurgy
(2–3). https://doi.org/10.1016/S0304-386X(99)00061-4. S. 91–106.
Symposium series/The South African Institute of Mining and Metallurgy,
Sandström, Åke, Petersson, Stig, 1997. Bioleaching of a complex sulphide ore with
Drakensberg. Johannesburg, pp. 213–240.
moderate thermophilic and extreme thermophilic microorganisms. Hydrometallurgy
Brierley, C.L., 2008. How will biomining be applied in future? Trans. Nonferrous Metals
46 (1–2), 181–190. https://doi.org/10.1016/S0304-386X(97)00012-1.
Soc. China 18 (6). https://doi.org/10.1016/S1003-6326(09)60002-9. S. 1302–1310.
Schippers, Axel, Hedrich, Sabrina, Vasters, Jürgen, Drobe, Malte, Sand, Wolfgang,
British Geological Survey, Bureau de Recherches Géologiques et Minières, Deloitte
Willscher, Sabine, 2014. Biomining: metal recovery from ores with microorganisms.
Sustainability, Directorate-General for Internal Market, Industry, Entrepreneurship
Adv. Biochem. Eng. Biotechnol. 141, 1–47. https://doi.org/10.1007/10_2013_216.
and SMEs (European Commission), TNO, 2017. Study on the Review of the List of
Critical Raw Materials. Final Report. Publications Office of the European Union,
Luxembourg.
9
N. Irrgang et al. Hydrometallurgy 200 (2021) 105566
Schlesinger, Mark E., King, Matthew J., Sole, Kathryn C., Davenport, William G., 2011b. Towler, G., Sinnott, R., 2007. Chemical Engineering Design: Principles, Practice and
Costs of copper production. In: Schlesinger, Mark E. (Ed.), Extractive Metallurgy of Economics of Plant and Process Design. Elsevier Science.
Copper, 5. ed. Elsevier, Amsterdam, pp. 427–439. Wang, Jun, Gan, Xiaowen, Zhao, Hongbo, Hu, Minghao, Li, Kaiyun, Qin, Wenqing,
Schlesinger, M.E., King, M.J., Sole, K.C., Davenport, W.G., 2011a. Extractive Metallurgy Qiu, Guanzhou, 2016. Dissolution and passivation mechanisms of chalcopyrite
of Copper. Elsevier Science. during bioleaching. DFT calculation, XPS and electrochemistry analysis. Minerals
Schuppan, Werner, Hiller, Axel, Krejny, Inge, 2012. Die Komplexlagerstätten Engineering 98. https://doi.org/10.1016/j.mineng.2016.09.008. S. 264–278.
Tellerhäuser und Hämmerlein. Uranbergbau und Zinnerkundung in der Grube Pöhla Weigel, H., 1964. Zink-Elektrolyse. In: Chemie Ing. Techn, 36 (6). https://doi.org/
der SDAG Wismut. Freiberg. Sächsisches Landesamt für Umwelt, Landwirtschaft und 10.1002/cite.330360610. S. 656–659.
Geologie (Bergbau in Sachsen, 17). Zapata, Dioni Mabel, Márquez, M.A., Ossa, D.M.H., 2007. Sulfur product layer in
Sinclair, Roderick J., 2005. The extractive metallurgy of zinc, , 1. ed.13. AusIMM sphalerite biooxidation: evidences for a mechanism of formation. In: AMR 20-21,
(Spectrum series/Australasian institute of Mining and Metallurgy, Carlton, Vic. pp. 134–138. https://doi.org/10.4028/www.scientific.net/AMR.20-21.134.
Sullivan, Colin, Tyrer, Mark, Cheeseman, Christopher R., Graham, Nigel J.D., 2010.
Disposal of water treatment wastes containing arsenic - a review. Sci. Total Environ.
408 (8), 1770–1778. https://doi.org/10.1016/j.scitotenv.2010.01.010.
10