Maria Matveinen
SELECTIVE LEACHING OF MULTIMETAL TAILINGS
Examiners: Professor Antti Häkkinen D.Sc. (Tech.) Riina Salmimies
Instructor: D.Sc. (Tech.) Riina Salmimies
Lappeenranta University of Technology during the end of year 2013 and beginning of 2014.
To begin with I would like to thank Professor Antti Häkkinen for his trust in me during my work. My greatest gratitudes go for Riina Salmimies, the dedicated instructor of both my Bachelor’s and Master’s thesis. She awakened my interest towards the mining industry and reinforced my skills in academic writing and research. I’m grateful to have had the chance to work in the solid/liquid- separation research group. The inspirational working environment and the great, supportive people made it pleasant to wake up early every weekday.
Studying wouldn’t have been anything without wonderful friends. I will definitely remember the wonderful times I’ve had while singing in Resonanssi and while working out at Huhtari. Thank all you gorgeous girls for giving me unforgettable and instructive moments and for awakening the small choir leader in me.
My family including Ville have a very special place in my heart. Thank you for giving me needed breaks during my study years by arranging the unforgettable trips abroad. Thank you for being there when I needed you and thank you for all your love and support.
Lappeenranta, 15th of April, 2014
Maria Matveinen
Lappeenranta University of Technology School of Technology
Degree Program of Chemical Technology Maria Matveinen
Selective leaching of multimetal tailings
Master’s thesis 2014
78 pages, 23 figures, 13 tables, 1 appendix Examiners: Professor Antti Häkkinen
D.Sc. (Tech.) Riina Salmimies
Keywords: leaching, tailings, sulfide, chalcopyrite, pentlandite, kinetic modeling Depletion of high grade mineral resources, tightening of environmental regulations and the environmental impact of acid mine drainage caused by sulfidic minerals continuously increase the interest in processing tailings and other mine waste. Treating waste requires additional capital and operational input, but the decrease in size and need of tailings ponds and permits decrease the overall costs.
Treatment and utilization of the tailings could also bring added revenue by the recovery of valuables.
Leaching of metal sulfides is very demanding and time consuming and hence process conditions need to be carefully optimized. The leaching of sulfides is affected by for example the choice of leaching agent, its concentration and temperature, pH, the redox potential, pressure, pulp density and particle size distribution. With reference to the mine case study the leaching of nickel and copper sulfides, especially the primary minerals pentlandite and chalcopyrite were investigated.
Leaching behavior and recoveries for nickel, copper and iron were found out by sulfuric and citric acid leaching experiments using tailings samples of high and low sulfur content. Moderate recoveries were obtained and citric acid seemed more attractive. Increase in temperature and decrease in pulp density had positive effect on the recovery and pH was also proven to have a significant effect on the recovery of valuables.
The rate determining step was determined through kinetic modeling in case of all valuables separately. Leaching was controlled by diffusion. The investigated multimetal tailing showed moderate potential in recovering of metal valuables from low grade tailing deposits. The process conditions should however be further optimized.
TIIVISTELMÄ
Lappeenrannan teknillinen yliopisto Teknillinen tiedekunta
Kemiantekniikan koulutusohjelma Maria Matveinen
Monimetallirikastushiekan selektiivinen liuotus
Diplomityö 2014
78 sivua, 23 kuvaa, 13 taulukkoa, 1 liite Tarkastajat: Professori Antti Häkkinen
TkT Riina Salmimies
Hakusanat: liuotus, rikastushiekka, sulfidi, kuparikiisu, pentlandiitti, kineettinen mallinnus
Kaivosjätteiden käsittelyyn liittyvän lainsäädännön tiukentuminen, rikkaiden kaivosesiintymien väheneminen sekä sulfidimineraalien aiheuttamien happovuotojen ympäristövaikutukset kasvattavat jatkuvasti mielenkiintoa rikastushiekan ja muun kaivosjätteen hyödyntämiseen. Liuottamalla arvometalleja rikastushiekkajätteestä voidaan altaiden määrää ja kokoa pienentää sekä rikastushiekkajätteen käsittelystä tehdä mahdollisesti taloudellisesti kannattavaa.
Metallien liukenemiseen vaikuttavat muun muassa liuotin ja sen konsentraatio, prosessiolosuhteet (pH, lämpötila, paine, hapetus-pelkistyspotentiaali) sekä kiintoainepitoisuus ja liuotettavan aineen partikkelikoko. Sulfidimineraalit ovat yleisesti ottaen hitaita liukenemaan sekä arvometallit vaikeasti talteenotettavissa.
Liuotusolosuhteiden optimoinnilla voidaan arvometallien talteenottoon vaikuttaa.
Esimerkkikaivoksesta löytyvien sulfidien mukaisesti työssä keskitytään nikkeli- ja kuparisulfidien, erityisesti kuparikiisun ja pentlandiitin liuotukseen.
Nikkelin, kuparin ja raudan liuotusta tutkittiin kahdelle rikkipitoisuudeltaan eroavalle rikastushiekkanäytteelle rikkihapossa ja sitruunahapossa, huoneenlämmössä ja ilmanpaineessa. Työn tuloksena nikkeliä ja kuparia saatiin liuotettua sekä rikkihappoon että sitruunahappoon kohtalaisesti, sitruunahapon vaikuttaessa lupaavammalta. Lämpötilan noston ja kiintoainepitoisuuden laskun osoitettiin vaikuttavan liukenemiseen positiivisesti ja pH:lla osoitettiin olevan suuri vaikutus liukenemiseen.
Kineettisellä mallinnuksella osoitettiin liukenemisen rajoittavan tekijän olevan diffuusio mineraalipartikkelin pinnalla. Tutkitulla monimetallirikastushiekalla on selkeää potentiaalia arvometallien talteenoton suhteen, kunhan prosessiolosuhteet optimoidaan tarkasti.
A pre-exponential factor in Equation 1 1/s
EA activation energy J/mol
k reaction rate constant 1/s
ko reaction rate coefficient taking the radius of a particle into consideration
1/sm
n order of diffraction, constant -
R universal gas constant J/molK
ro radius of a particle m
T temperature K
t time s
α extent of reaction -
ABBREVIATIONS
AAS atomic absorption spectrometry
Ar arsenopyrite
Cat catierite
Cp chalcopyrite
HS high sulfur tailings sample
ICP-OES inductively coupled plasma optical emission spectrometry
Lin linnacite
LS low sulfur tailings sample
Mil millerite
Pe pentlandite
PGE platinum group elements
Pr pyrrhotite
Py pyrite
RO reverse osmosis
SEM scanning electron microscope
MINERALS
arsenopyrite FeAsS bornite Cu5FeS4
catierite CoS2 chalcocite Cu2S chalcopyrite CuFeS2 covellite CuS cubanite CuFe2S3 enargite Cu3AsS4
galena PbS
gersdorffite NiAsS heazlewoodite Ni3S2
hematite Fe2O3
linnacite Co3S4
magnetite Fe3O4
maucherite Ni11As8
millerite NiS molybdenite MoS2
nickeline NiAs
olivine (Mg,Fe)2SiO4
pentlandite (Fe,Ni)9S8
pyrite FeS2
pyrrhotite Fe1-x(x=0-0.17)S tennantite Cu12As4S13
CONTENTS
1 INTRODUCTION ... 2
LITERATURE PART ... 3
2 FACTORS THAT AFFECT LEACHING ... 4
2.1 Particle size distribution ... 4
2.2 Leaching agent ... 6
2.2.1 Concentration of leaching agent ... 8
2.3 Temperature ... 10
2.4 Pulp density ... 13
2.5 pH ... 15
3 LEACHING OF SULFIDES ... 16
3.1 Sulfur formation ... 17
3.2 Leaching in presence of oxidizing agent... 19
3.2.1 Oxidative dissolution ... 21
3.2.2 Protonation ... 22
3.2.3 Reductive/oxidative dissolution ... 23
3.3 Pentlandite ... 24
3.4 Chalcopyrite ... 26
4 KINETICS OF SULFIDE LEACHING ... 28
4.1 Modeling ... 29
5 PROCESSING AT THE CASE MINE ... 33
EXPERIMENTAL PART ... 37
6 TAILINGS CHARACTERIZATION ... 37
7 LEACHING EXPERIMENTS ... 39
7.1 Design of experiments ... 44
7.2 Experimental setup ... 45
8 RESULTS AND DISCUSSION ... 46
8.1 Sulfuric acid leaching ... 46
8.2 Citric acid leaching ... 52
8.3 Comparison of leaching agents ... 56
8.4 Temperature effects... 59
8.5 Pulp density effects ... 62
9 KINETIC MODELING ... 64
10CONCLUSIONS ... 68
REFERENCES ... 71
APPENDICES ... 78
1 INTRODUCTION
With increasing environmental concerns and with a decreasing amount of high grade mineral deposits, tailings have become more attractive sources of valuable minerals. Even though mineral concentration methods improve continuously, some metal valuables are inevitably lost to mineral tailings. Processes having tailings deposits as raw materials consume less energy during the production stage since the material is already quarried and grinded. Hence tailings can in some cases be seen as more beneficial resources than low grade ores.
Focus of this thesis was to find out a way of selective leaching of valuable metals found in tailings. Emphasis was made on the leaching of sulfides since at the chosen case study nickel and copper sulfides are the primarily recovered products.
Platinum group elements, especially palladium, platinum and gold, would also be of great interest, but in this thesis focus was only laid on recovering nickel and copper sulfides.
In the literature part factors affecting leaching processes are introduced after which the leaching of sulfides is described. Special focus is laid on the leaching of pentlandite ((Fe,Ni)9S8) and chalcopyrite (CuFeS2) which are the main sulfides found at the mine case study. Previous studies on leaching of pentlandite and chalcopyrite are reviewed and possible leaching mechanisms are described. The empirical modeling of the kinetics of sulfide leaching is also introduced.
In the experimental part leaching of two different tailings samples are studied.
Preliminary leaching tests are conducted after which sulfuric acid and citric acid are chosen for further research. The kinetics of sulfide leaching is examined by empirical modeling and the rate controlling step is found out.
LITERATURE PART
Extracting or dissolving a required constituent, in this case a metal, in a suitable solvent is called leaching. Leaching can be performed in order to achieve two different things. Either the leaching agent is applied to open the concentrates, metallurgical products or ores in order to reach the soluble metals or then a leaching agent is applied to dissolve gangue minerals in order to have the mineral in a more concentrated form. (Habashi, 1999a)
In this work focus is laid on sulfides, especially copper and nickel sulfides.
Pentlandite, chalcopyrite and pyrite (FeS2) are the main copper and nickel sulfides found in the case study. This mine was chosen as a case study since it is a rather young mine. The mine also contains sectors in which the platinum group element (in this case palladium, platinum and gold, further on referred as PGE) content is rather high and hence economic value can be found. The recovery of PGE is however not included in this study. In addition, possible implementations to new process structures can still be done in order to optimize the process.
Leaching of sulfides especially with respect to leaching of chalcopyrite and pentlandite will be presented in this literature part. In this literature part investigations are made if the recovery of the valuables lost to tailings could be increased by the use of selective leaching. Many factors affect leaching behavior and hence these factors will shortly be discussed. In order to present experimental results, the kinetics of sulfide leaching will be described and empirical modeling will be used. The models used to describe sulfide leaching will briefly be presented. In the end the case study of processing at the chosen mine and the tailings found at the mine will serve as an introduction to the experimental part.
2 FACTORS THAT AFFECT LEACHING
When hydrometallurgical reactions take place between a solid and a liquid phase, the processes are heterogenous. According to Habashi (1999a) the dissolution of solids depends on the nature of the bonds present. In the case of chemical processes (neutralization, complex formation, oxidation and protonation), solids are either mainly covalent or partly ionic and partly covalent. Sulfides, being of major concern in this study, belong to the latter group. Alkali and alkali earth metal sulfides have ionic character whereas other metal sulfides have mainly covalent bonds (Burkin, 2001).
The character of sulfides largely affects the leaching behavior. Many other factors also contribute on how effective leaching processes are. These factors include particle size distribution, initial acid concentration, agitation and temperature (Lamya and Lorenzen, 2009). Also pulp density, pH value and the leaching time affect the leaching behavior (Antonijević et al., 2008). Some of these factors will be discussed in more detail in the following paragraphs.
2.1 Particle size distribution
Particle size distribution and the microstructure of minerals have a great effect on leaching kinetics. The shape of the minerals and the way how they are attached to each other affect the rate at which the valuable constituents can be extracted from the non-valuable gangue. The microstructure of minerals and how easily the minerals may be separated is represented in Figure 1. Minerals with easy liberation have the valuable constituent situated almost separately from the worthless gangue. The valuable mineral might however be distributed throughout the non-valuable gangue or even inside it making the liberation more difficult.
Liberation of valuable constituents from chalcopyrite is difficult or even very difficult (Amstutz, 1961). Pyrrhotite-pentlandite samples are often of layered structure, also called lamella structure. These minerals belong to the group of difficult liberation. Minerals that are relatively easy to liberate can be separated by the use of mechanical processes whereas the minerals that are harder to liberate
need to be extracted by leaching. Also the distribution of valuable minerals in the bedrock affects the recovery.
FIGURE 1. Mineral microstructures (Amstutz, 1961, with permission from Colorado School of Mines Press).
In order to optimize the recovery of metals and the leaching process overall, some pretreatment to the minerals may be necessary to perform. The particles may undergo different size reduction steps and particles that contain the metal can thereafter be separated from the total mixture. Reducing the size of the ore particles increases the surface area available for leaching and more of the minerals are liberated easier from the gangue. Several researches have been made on the effect of particle size distribution of which a couple are introduced next.
Fu et al. (2010) studied the effect of particle size distribution on the selective leaching of nickel from a low-sulfur Ni-Cu matte containing a small quantity of sulfides. The studied grain sizes were <38, 38-45, 45-74, 74-150, 150-380 and
>380 µm. Three different matte samples, which were grinded for 0 min, 15 min and 240 min were compared. The matte was leached by a mixture of copper sulfate and sulfuric acid in the presence of air as an oxidizing agent. A diffusion layer was said to be formed during the leaching process and hence the smaller particles the smaller diffusion layers and hence a higher recovery of the valuable metal.
Hansen et al. (2005) also studied the influence of particle size on the leachability of copper. The mine tailing sample having a copper content of 1150 mg kg-1 dry matter was divided into different size fractions: <38, 38-45, 45-53, 53-75, 75-106, 106-150, 150-212 and >212 µm. The main minerals in this tailing sample were chalcopyrite, CuFeS2, (86-90 %) and bornite, Cu5FeS4, (6-9 %). For the divided size fragments it was found out that the smallest sized particles contained 50-60 % copper that was leachable in weak acid whereas the largest particles contained only 32 % weak acid leachable copper. As leaching acids HCl, HNO3, H2SO4 and citric acid were used. Again, a smaller particle size gave a higher recovery.
By first dividing the tailings into different size fractions and thereby concentrating the feed before leaching, the volume of leaching reagent could be diminished.
However, reducing the particle size of the mineral might not always be economically profitable because of the high need of energy to crush and grind the particles. An advantage in leaching is that only part of the valuable metal surface needs to be available for the leaching reagent. The metal can be separated progressively as long as there is some direct contact to the valuable metal.
2.2 Leaching agent
Care should be taken when choosing the right leaching reagent since often several metals are dissolved into the same solvent. Ideally a reagent only dissolves the wanted metals into the solution. This is rarely reached since leaching reagents also take many other metals that are present in the ore into the solution at the same time. There are several aspects to take into consideration when choosing a leaching agent. The desired constituent has to be soluble in the leaching agent, the agent has to be relatively cheap and the metal has to be economically recoverable from the solution. The corrosive nature of the leaching agent has to be taken into consideration as well. The capital cost of the equipment is higher if the leaching agent is corrosive, because tanks and other equipment need to be made of stainless steel, titanium or Hastelloy (Habashi, 1999a). With higher environmental concerns, the regeneration of the leaching agent for recycle is also important (Ray and Ghosh, 1991).
Water is the most ideal leaching reagent because it is cheap and not corrosive.
However, according to Habashi (1999a), using water as a leaching reagent is limited to only a few cases. Naturally occurring salts, flue dusts, calcines and some sulfide concentrates are leached using water. Some sulfides are dissolved and converted to sulfates by water leaching under pressure and at 200 °C in the presence of air. According to Ray and Ghosh (1991) low-grade copper sulfide ores can be leached under atmospheric conditions by water since they slowly form water soluble sulfates. Sulfides are most commonly leached by sulfuric acid (Habashi, 1999a). Commonly used leaching agents, others than water, can be divided into three categories: acids, bases and aqueous salt solutions (Habashi, 1999a). In addition to these categories, aqueous chlorine and hypochlorite are used in minor extents. The most common leaching agents used for different minerals are shown in Table I.
Sulfuric acid is commonly used when leaching metals because of its rather low cost (Noyes, 1994). In addition sulfuric acid only has minor corrosion problems and sulfuric acid can dissolve many metal compounds. Dissolving many metal compounds might however not be wanted since also the worthless metals might be dissolved into the solution.
Ammonia is also used for dissolving nickel and copper (Noyes, 1994; Moore, 1981). Ammonia and ammonium carbonate are however more expensive than for example the commonly used sulfuric acid and hence the recycling of especially ammonia and ammonium carbonate is essential.
TABLE I Summary of commonly used leaching reagents for various minerals. 1Habashi (1999a), 2Moore (1981).
Mineral Leaching reagent Metal sulfates Water2, H2SO42
Metal sulfides H2SO42
, Fe2(SO4)3 solution1,
NH4OH + air (nickel sulfide concentrates)1 Metal oxides H2SO42
Cu/Ni minerals NH32, FeCl32
Gold and silver ores NaCN + air1, Cl2 (aq)1, Aqua regia (noble metals)1
In addition to the importance on the choice of the reagent also the concentration has an effect on the leaching of sulfides. This will be discussed in the following section.
2.2.1 Concentration of leaching agent
Rates of reactions are usually increased by increasing reactant concentration (Habashi, 1999b). Increasing reactant concentration is however not economical beyond stoichiometric concentrations. When designing leaching processes the minimum concentration should be found out in order not to waste any valuable reagent even though leaching agents are reused inside the process.
Sokić et al. (2009) investigated the effect of concentration of sulfuric acid during the leaching of a chalcopyrite flotation concentrate. Sodium nitrate was added as an oxidizing agent. By keeping all the other variables constant and by investigating four different sulfuric acid concentrations (0.6, 1.0, 1.5 and 2.0 mol/dm3) the recovery of copper was determined. After a 240 min leaching time the reaction had not reached equilibrium. At 240 min a copper extraction of 47 % and 75 % was achieved by having acid concentrations of 0.6 mol/dm3 and 2.0 mol/dm3, respectively. This can be explained through the oxidizing agent used.
Increasing leaching acid concentration also increases the concentration of H+ ions which affects the oxidizing potential of NO3-
ions. The higher the concentration of H+ ions, the higher the oxidizing potential of NO3-
ions.
Koleini et al. (2011) concluded in their study that the initial acid concentration has little effect on the recoveries of copper from chalcopyrite. This has been explained by the difficulties in controlling the redox potential. Sulfuric acid concentrations used during the study were 7.5, 15 and 30 g/dm3 (0.075, 0.15 and 0.3 mol/dm3).
Similar recoveries were observed with 7.5 and 30 g/dm3 whereas the highest recovery was obtained with 15 g/dm3. The recovery with an initial acid concentration of 15 g/dm3 was over 80 % whereas with the other acid concentrations the recovery was about 70 %. The changes have been discussed by Koleini et al. (2011) and occur presumably because of the different increases in pH and the difficulty in controlling the redox potential. Oxygen or nitrogen gas was injected continuously to control the redox potential. The increase in pH indicates the use of protons from sulfuric acid when dissolving valuables. At the beginning of the experiments the pH was about 0.8 and at the end a pH of 1.8-1.9 (which was the final pH of 0.15 mol/dm3 sulfuric acid dissolution reaction) was reached. This value of pH is a limit pH when ferric ion might precipitate as jarosite. Jarosite has been discussed to cause passivation of chalcopyrite.
According to Dutrizac et al. (1969) the acid should prevent ferric ions from hydrolyzing. However an acid concentration of 0.1 mol/dm3 should be sufficient to prevent hydrolysis and hence the reason of little change in recovery when increasing the acid concentration presumably lies in the difficulty of controlling the redox potential.
Aydogan et al. (2006) came to similar conclusions as Sokić et al. (2009). They both concluded that copper recoveries were increased when the concentration of sulfuric acid was increased. In the study made by Aydogan et al. (2006) a chalcopyrite sample was leached in sulfuric acid in five different concentrations, 0.1, 0.2, 0.3, 0.4 and 0.5 mol/dm3. Strongly oxidizing conditions were formed by the addition of potassium dichromate. After 150 min leaching a copper extraction of 20 % and 54 % was achieved by having acid concentrations of 0.1 mol/dm3 and 0.4 mol/dm3 respectively. However, there was no significant change in the copper extraction when comparing 0.4 mol/dm3 and 0.5 mol/dm3 acid. The reason for an increase in recovery when increasing the acid concentration occurs of the same reasons as for the study made by Sokić et al. (2009). Increasing the acid
concentration and hence the H+ ion concentration also increases the reduction potential of dichromate ion which was used to obtain strongly oxidizing leaching conditions.
The results of the investigations described above show that increasing acid concentration affects the redox potential through which the recovery of valuables is affected. In addition there should be a sufficient acid concentration in order to prevent the hydrolysis of ferric ions that form products that hinder the dissolution of chalcopyrite.
2.3 Temperature
Also the temperature at which leaching is performed has a great effect on recoveries of valuable metals. This has been studied by Aydogan et al. (2006), Córdoba et al. (2008) and Koleini et al. (2011). As a result of all the investigations an increase in temperature was concluded to increase the recoveries. However an optimum temperature should always be found out since increasing temperature also means an increase in heating costs. Economic aspects and sustainable development should always be taken into consideration and hence operating at high temperatures might not be the best way in terms of economy.
Aydogan et al. (2006) studied the kinetics of dissolving a chalcopyrite concentrate in acidic potassium dichromate. The concentrate was obtained from a flotation circuit and contained a complex ore CuFeS2-PbS-ZnS. Leaching was done at temperatures from 50 °C to 97 °C and the temperatures in each test were kept constant by the use of a thermostatically controlled water bath. At a temperature of 50 °C the amount of extracted copper was 54 % whereas at 92 °C it was increased to 82 %. The reaction has however not reached the state of equilibrium and hence comparison of the recoveries of copper at equilibrium can not be made.
The increase in temperature can thus only be shown to speed up the kinetics of the reaction.
Córdoba et al. (2008) also concluded in their study on the leaching of chalcopyrite using ferric ions that an increase in temperature increases the recovery of copper.
In four different temperatures between 35 and 68 °C and after 13 days of leaching they experienced an increase in the extraction of copper from less than 3 % to more than 80 %, respectively. In these experiments the reactions done at 57 °C and 68 °C seemed to reach equilibrium whereas the two reactions at 35 °C and 46
°C were still under linear growth after 13 days of leaching. Equilibrium at the two higher temperatures was reached approximately after seven days with a temperature of 68 °C and after ten days with a temperature of 57 °C.
Also Koleini et al. (2011) came to a conclusion that increasing temperature increases copper recovery. The chalcopyrite concentrate was leached in sulfuric acid in the presence of pyrite in three different temperatures, 48 °C, 68 °C and 85
°C. A copper recovery of higher than 80 % was achieved at the highest temperature after 24 hours of leaching whereas the recovery with a temperature of 68 °C was about 15 percentage points less. Both of these higher temperature recoveries were more than four times greater than the recovery obtained in 48 °C leaching. The recovery of copper showed a linear dependency with respect to temperature after 24 hours of leaching.
Baláž et al. (2000) studied the effect of temperature on the leaching behavior of a pentlandite concentrate. Also in this study an increase in temperature gave a higher yield of the compound to be extracted. The main components in the concentrate were pentlandite, chalcopyrite, pyrrhotite (Fe1-x(x=0-0.17)S) and pyrite.
First the concentrate was leached in water after which leaching in Fe2(SO4)3
solution was performed. Leaching was done at two different temperatures, 45 °C and 90 °C, for an as received concentrate sample and a ground (referred by Baláž et al. (2000) as a mechanically activated) sample. Grinding was carried out in order to accelerate the leaching process (Maurice and Hawk, 1998). When grinding a sample the surface area of the sample increases which leads to an increased reactivity at the surface and to structural changes in the sample. For example for a pentlandite sample, grinding increases the surface area and
decreases the amount of the crystalline phase hence increasing the degree of amorphization.
In the study made by Baláž et al. (2000) nickel, copper and cobalt were extracted.
Recoveries were presented for the second step of the leaching reaction, which was the leaching in Fe2(SO4)3. The recoveries at the beginning of this second step were not zero since some water soluble sulfides were formed in the grinding stage.
These water soluble sulfides were leached already in the first leaching stage. The recoveries presented graphically by Baláž et al. (2000) are shown in Table II.
TABLE II Recovered values at two different temperatures from a sample as received and a sample after mechanically activating (grinding) it for 60 min after 120 min leaching in Fe2(SO4)3. Data read from graphs presented by Baláž et al. (2000).
Sample As received Grinded Temperature (°C) 45 90 45 90 Recovered Ni (%) 6 13 26 73 Recovered Cu (%) 9 19 33 100 Recovered Co (%) 9 36 30 89
As shown in Table II, the recovered values of valuable constituents are higher for the ground sample when compared to the as received sample. A higher temperature also gave a significantly higher recovery in the case of mechanically activated samples.
The effect of temperature on the rate of reaction can be explained by the Arrhenius equation shown in equation 1.
RT EA
Ae
k= − (1)
In which k is the reaction rate constant, A the pre-exponential factor, EA the activation energy, R the universal gas constant and T the absolute temperature in Kelvin. Activation energy does however not reveal any significant information on the kinetics or mechanisms of dissolution (Crundwell, 2013). It only gives you two values: if the activation energy, EA, is lower than 20 kJ/mol the reaction is
diffusion controlled in the aqueous phase and if it is above 40 kJ/mol the reaction is chemically controlled. However even these values are generalized and changes might occur depending on temperatures used (Córdoba, 2008).
Activation energies were studied by several authors (Biswas et al., 2014; Córdoba et al., 2008; Kaplun et al., 2011; Li et al., 2013; Kimball et al., 2010; Nicol et al., 2010; Sokić et al., 2009; Vračar et al., 2003) in order to get suggestions for the controlling factor of the leaching process. Vračar et al. (2003) for example used activation energy to prove that the leaching of copper (I) sulfide with sulfuric acid and added sodium nitrate was chemically controlled (the value of activation energy obtained was 60 ±0.7 kJ/mol). High activation energies would indicate a need for high temperatures since the bonds in the mineral need to be broken down in order for leaching to occur. Hence finding out activation energies gives some direction on how high temperatures are optimal.
As a conclusion of the studies shown above, an increase in temperature seems to increase the amount of valuable constituents recovered and speed up the kinetics of the reaction. In the study by Córdoba et al. (2008) the equilibrium concentration of dissolved copper increased with increasing temperature, but for studies made by Aydogan et al. (2006), Baláž et al. (2000) and Koleini et al.
(2011) no conclusions on the effect of temperature on the equilibrium concentration could be made. The reaction kinetics were however shown to speed up by increasing the temperature. Note should be made to the fact that increasing temperature is not always the most economical way to increase the recovery. An optimum temperature should be found out instead of using the temperature giving the highest recovery.
2.4 Pulp density
The effect of pulp density on the recovery of copper from chalcopyrite concentrates was studied by Koleini et al. (2011). The copper concentrate consisted mainly of chalcopyrite with small amounts of chalcocite, Cu2S, and covellite, CuS. Sulfuric acid was used as leaching media and pyrite was added as
a catalyst. The effect of pulp density was investigated by measuring the copper recovery at three mass concentrations, 2.5 %, 7.8 % and 14.5 %. After 24 hours of leaching, equilibrium was not reached. The recovery of copper after 24 hours was about 5 percentage points higher when leaching with 7.8 % pulp density compared to 2.5 %. At a mass concentration of 14.5 % the copper recovery was about 20 % lower than at a mass concentration of 7.8 %. A smaller pulp density thus gave a higher copper recovery. Comparing the pulp densities is however somewhat questionable since equilibrium was not reached.
A similar behavior was discovered in the study made by Demir et al. (2004), in which spherical copper particles were leached by nitric acid. The ratio between solid and liquid was investigated in the range of 0.0025-0.02 g/cm3. By increasing the solid-liquid ratio the dissolution rate was decreased and hence at higher pulp density a longer leaching time was needed to recover more Cu. Also in this case a smaller pulp density gave a higher copper recovery faster.
Also Antonijević and Bogdanović (2004) studied the effect of pulp density through changing the solid liquid ratio of the experiments conducted. Leaching was investigated in columns by percolating sulfuric acid solutions of different pH through the solid layer of an ore sample containing mostly chalcopyrite and pyrite sulfides. Solid liquid ratios of 1:1, 2:1 and 4:1 were used. At a solid liquid ratio of 4:1 the recovery was best of the three ratios, however the recovery of copper being quite poor. Less than 0.5 g/dm3 was recovered in 90 days. This is presumably caused by the large grain size (about 80 % of the particles were
>1mm) and hence the lack of contact between the leaching solution and the copper minerals.
Having a smaller solid-liquid ratio would mean that there is more reactive agent per solid to leach the valuable constituents from the solid. A minimum amount of leaching agent in order to leach all the leachable constituents should be found out.
Optimum solid to liquid ratio should be found out in order not to have excess leaching agent or lost valuable constituents.
2.5 pH
Working inside a suitable pH range is essential for optimal recovery of valuable constituents. Chalcopyrite leaching consumes acid and hence enough acid and an enough low pH should be utilized (Koleini et al., 2011). A low pH is also needed to prevent the hydrolysis of ferric ions to insoluble iron products which deteriorate the leaching of chalcopyrite (Koleini et al., 2011; Bartlett, 1992). Possible small changes in pH that occur while the reaction goes on do not seem to have a big effect on the dissolution of chalcopyrite when pH is low enough (Antonijević and Bogdanović, 2004; Córdoba et al., 2008).
Antonijević and Bogdanović (2004) studied the effect of having different pH on the recovery of copper from a chalcopyritic ore in acidic solutions. The experiments were done in PVC columns for 90 days by percolating the leaching solution through a layer of the ore. The ore sample of size smaller than 5 mm consisted mainly of chalcopyrite and pyrite. The recovery of Cu was studied in five different pH values 0.5, 0.7, 1.0, 1.3 and 2.0, which were kept constant by adding sodium hydroxide or sulfuric acid solution. It was concluded that the change in pH in this range had a small effect on the dissolution rate of chalcopyrite. The changes might however be this small due to the small amount of copper in the initial sample and the small content of dissolved copper after 90 days of leaching. The lowest pH, 0.5, gave the lowest content of dissolved copper.
This has been explained by Lu et al. (2000a,c) and Antonijević and Bogdanović (2004) by the fact that chalcopyrite might be in a passive state because of the formation of a hindering layer on top of the mineral. At higher pH values chalcopyrite oxidises easier increasing the recovery of copper after 90 days of leaching. Again the importance of leaching under a suitable pH is emphasized.
In the review by Córdoba et al. (2008) the dissolution of chalcopyrite when ferric sulfate is used as a leaching agent is diminished with decreasing pH in the range from 2.0 to 0.5. The effect of pH on the dissolution cannot be seen until after six days of leaching after which increasing the pH also shows an increase in the content of dissolved copper. Although the dissolution of copper at higher pH has
been said to be hindered by the hydrolysis of ferric ions, the higher the pH the higher the recovery also was in the study by Córdoba et al. (2008). They have described that the increase in recovery with increasing pH would occur because of the increase in concentration of Fe(SO4)2- which is responsible for chalcopyrite oxidation.
Antonijević et al. (2008) investigated the effect of pH on the leaching kinetics of copper and iron from flotation tailings. The flotation tailings of Bor Copper mine in Serbia were leached with distilled water in which a solution of sulfuric acid and tap water was added depending on the desired pH value. The tailings sample consisted of 21 % of sulfidic minerals of which 20.81 % were pyrite minerals.
Solutions of various pH values (0.0, 1.0, 1.5, 2.0, 3.0, 5.7, 7.2) were investigated and the pH value of each test was adjusted by adding sulfuric acid. Some dissolution could be seen already at neutral conditions with no added sulfuric acid, which indicates that some copper sulfate was present. Copper sulfate may have been formed when the flotation tailings were exposed to oxidation through weathering. By increasing the H+ concentration the dissolution of iron and copper increased as well. This is in line with the research made on changing the concentration of leaching agent. Changing pH affects the redox potentials through which the dissolution behavior is also affected. In order to dissolve as much valuables as possible, the most essential however seems to be working under a suitable pH range.
3 LEACHING OF SULFIDES
Hydrometallurgical processes have become more attractive in metal extraction and production in comparison to pyrometallurgical processes. In pyrometallurgical processes the energy consumption is higher than in hydrometallurgical processes and hence energy consumption costs are diminished by using hydrometallurgical processes. By optimizing hydrometallurgical processes, more economical and environmentally safe processes can be obtained.
Several studies have been made regarding hydrometallurgical leaching methods, such as biochemical leaching (Aromaa et al., 2013) and acidic dissolution (Akcil
and Ciftci, 2002; Antonijević et al., 2008). Interest in generating processes to recover metals from low grade ores without the use of high pressures or temperatures has been shown by many researchers (Akcil and Ciftci, 2002; Xie, 2005; Fu et al., (2010)).
Low grade ores, as is the case of tailings, are most often treated by hydrometallurgical processes (Habashi, 1999a). Pyrometallurgy is often unsuitable since there is a large amount of gangue to be melted and hence a large amount of energy would be consumed when extracting the gangue. By choosing a suitable leaching agent, low grade ores can be treated more effectively. As mentioned in Table I, metal sulfides are leached for example by using sulfuric acid, iron(III)sulfate or ammonium hydroxide together with air. In addition studies have shown that ferric chloride solutions are also commonly used when leaching sulfide ores (Maurice and Hawk, 1999; Akcil and Ciftci, 2002).
Leaching can be carried out either in the absence or presence of an oxidizing agent. The solid sulfide is dissociated in the aqueous phase as is described in equation 2.
MS M2+ + S2- (2)
Acid soluble, alkali soluble and complex forming sulfides may be leached in the absence of oxidizing agents, i.e. under reducing conditions (Habashi, 1999a).
Depending if the sulfides dissolve in acids or in bases either H2S or S2- is formed respectively. NiS, FeS and CoS are all acid soluble sulfides whereas PbS is an alkali soluble sulfide. ZnS is both soluble in acid and alkali.
3.1 Sulfur formation
When leaching is carried out under oxidizing conditions elemental sulfur usually forms (Habashi, 1999a). Formation of elemental sulfur is preferred over other sulfur products, while it is easily stored and shipped for further use or if desired H2SO4 or SO2 can be produced. Elemental sulfur is very unstable in basic media
and in the presence of strong oxidizing agents and hence sulfates are easily formed. Figure 2 shows the narrow region where elemental sulfur is formed at equilibrium. It should be emphasized that this diagram only holds in equilibrium thus it gives no indication on the kinetics of the reaction (Moore, 1981). Only at the state of equilibrium elemental sulfur is formed in the region shown in Figure 2.
FIGURE 2. The potential-pH diagram for sulfur at 100 °C showing the narrow region in which elemental sulfur can form (Habashi, 1999a).
At ambient conditions the chemical leaching reactions, with oxidizing agents present, are very slow and sulfates are usually formed. The reaction can be accelerated by increasing the temperature. The leaching conditions, e.g.
temperature and pH, also affect the formation of elemental sulfur in electrochemical oxidative leaching processes. The elemental sulfur may be oxidized further and hence prevent the formation of elemental sulfur which would be the preferred product. Acid insoluble sulfides are commonly present in electrochemical processes and the electrochemical processes are described by equations 3 and 4 (Habashi, 1999a). Equation 3 describes the oxidation reaction while equation 4 is the reduction equation.
MS M2+ + S + 2e- (3)
Oxidizing agent + ne- Reduced species (4)
Leaching processes under oxidizing conditions are either chemical or electrochemical reactions, and may lead to the formation of sulfates or elemental sulfur as shown in equations 5 and 6. Sulfates are formed in processes that take place in neutral medium while elemental sulfur is formed in the case of acid soluble sulfides or in electrochemical processes with acid insoluble sulfides.
S2- + 2O2 SO42- (5)
S2- + 2H+ + ½ O2 S + H2O (6)
Generally leaching of sulfides occurs in the presence of oxidizing agents (Lundström, 2009). The following section will deal with the leaching of chalcopyrite and pentlandite in the presence of an oxidizing agent. The mechanisms of dissolution are explained by chalcopyrite being an example.
3.2 Leaching in presence of oxidizing agent
The most commonly used oxidizing agents are oxygen, nitric acid, ferric ion, concentrated H2SO4, chlorine water and sodium hypochlorite (Habashi, 1999a).
Also cupric ions (Cu2+) are used as oxidizing agents (Lundström, 2009). Copper, nickel and cobalt can form ammine complexes when oxygen is used as an oxidant in a solution of water, ammonium hydroxide and a dilute acid. This aqueous oxidation is a very slow process under ambient conditions and hence increased pressure is needed to accelerate the process. Pressure oxidation leaching processes are however beyond the scope of this study and focus is laid on leaching at ambient pressure. The leaching mechanisms in different leaching conditions for different chalcopyrite samples are collected in Table III.
TABLE III
Sulfide*Dissolving mediumT/°CpHMechanismExtra notesSource CpH2SO4, Cr2O72- 50-97-Oxidative dissolutionatmospheric pressureAydogan et al. (2006) CpH2SO4, NaNO380-Oxidative dissolutionatmospheric pressureSokić et al. (2009) CpH2SO4, Fe2(SO4)3850.8-1.9Oxidationatmosphericpressure,redoxpotential controlled by oxygen or nitrogen additionKoleini et al. (2011) CpH2SO4, Fe2(SO4)3751-2OxidationLi et al. (2010) CpHClO4851Oxidationatmospheric pressureHarmer et al. (2006) Cp, Ar, Pe, Pr, Py
Ammonium thiosulfate, ammonium sulfate, ammonia water, cupric sulfate, hydrogen peroxide, HCl
25-Complexformation and oxidationatmospheric pressureFeng & Van Deventer (2002) CpH2SO4,FeSO4,CuSO4 Fe2(SO4)340-Reductive/oxidativeatmospheric pressureHiroyoshi et al. (2000)
TABLE III Leaching mechanisms for different sulfides in different leaching solutions. * Sulfides: Ar=arsenopyrite (FeAsS), Cp=chalcopyrite, Pe=pentlandite, Pr=pyrrhotite, Py=pyrite
Based on this table and a research made on chalcopyrite leaching (Lundström, 2009) focus is laid on the leaching of sulfides in the presence of an oxidizing agent. Many of the studies shown in Table III have used ferric ions as oxidizing agents. In the presence of an oxidizing agent dissolution takes place through transferring of electrons. The dissolution of chalcopyrite takes place through protonation, oxidation, a combination of reduction and oxidation or through complex formation.
Nicol et al. (2010) have done an extensive review on the dissolution of chalcopyrite in chloride solutions. The dissolution mechanism of chalcopyrite depends on the leaching media used. The mechanism can be a redox, a non- oxidative or a mechanism in which first reduction occurs and secondly oxidation (from hereon called reductive/oxidative). The non-oxidative dissolution model can be extended to cover the whole reaction and in this case the model is called a non-oxidative/oxidative model. Most of these mechanisms mentioned by Nicol et al. (2010) are also represented in Table III and hence also apply for non chloride leaching agents.
3.2.1 Oxidative dissolution
Oxidative dissolution by ferric or cupric ions can be represented by an electrochemical model composing of an anodic reaction and a cathodic reaction (Gómez et al., 1996; Nicol et al., 2010). Oxidative dissolution can also be called a redox reaction since both an anodic and a cathodic reaction take place simultaneously. These both reactions are represented by equations 7 and 8, respectively. The ions of chalcopyrite dissolve into the solvent in which ferric ions or cupric ions are used as reducing agents.
CuFeS2 → Cu2+ + Fe3+ + 2 So + 5e- (7) 4Fe3+ + 4e = 4Fe2+ or 4Cu2+ + 4e = 4Cu+ (8)
Aydogan et al. (2006) presented that the oxidative dissolution of chalcopyrite in sulfuric acid with dichromate ion addition produces, presumably due to different
stoichiometry of the equations, either elemental sulfur as shown in equation 9 or sulfate according to equation 10. In the oxidative dissolution of chalcopyrite either elemental sulfur or sulfate is formed. The chromate ion acts as a reducing agent.
6CuFeS2 + 5Cr2O72- + 70H+↔ 6Cu2+ + 6Fe3+ + 12S + 10Cr3+ + 35 H2O (9) 6CuFeS2+ 17Cr2O72-
+ 142H+↔6Cu2+ + 6Fe3+ + 12SO42-
+ 34Cr3+ + 71 H2O (10)
When sulfides are leached using nitrate as leaching agent elemental sulfur is formed. The sulfides undergo oxidative dissolution according to equations 11 or 12 (Sokić et al., 2009). The study by Vračar et al. (2003) proposes that equation 12 would be the more probable reaction to occur. This has been explained by a more negative value for the change in Gibbs energy.
3MeS + 2NO3-
+8H+ = 3Me2+ + 3S + 2NO + 4H2O (11) MeS + 2NO3-
+4H+ = Me2+ + S + 2NO2 + 2H2O (12)
3.2.2 Protonation
The opposite of oxidative dissolution namely non-oxidative dissolution can also be called protonation. In protonation copper is detached from the chalcopyrite through proton attacks as described in equation 13 (Nicol et al., 2010; Kimball et al., 2010).
CuFeS2 + 4H+ = Cu2+ + Fe2+ + 2H2S (13)
However, as described in Chapter 3.1 the formation of elemental sulfur is preferred and hence reaction 13 can be extended in order to obtain the overall oxidative dissolution process. The overall reaction can be obtained by combining equation 13 with equation 14 (Nicol et al., 2010).
H2S + 2Fe3+ = S + 2Fe2+ + 2H+ (14)
3.2.3 Reductive/oxidative dissolution
The reductive/oxidative dissolution model mentioned by Nicol et al. (2010) is a model proposed by Hiroyoshi et al. (2000). This model is for ferrous-promoted chalcopyrite leaching. The dissolution of copper occurs in sulfuric acid in the presence of cupric and ferrous ions through two steps shown in equation 15. The first step is the reduction of chalcopyrite by ferrous ions to form Cu2S as shown in equation 16. Cu2S can then be oxidized by dissolved oxygen or ferric ions to cupric ions as shown in equation 17.
CuFeS2 Cu2S Cu2+ (15)
CuFeS2 + 3Cu2+ + 3Fe2+ = 2Cu2S + 4Fe3+ (16) 2Cu2S + 8Fe3+ = 4Cu2+ + 8Fe2+ + 2S (17)
With reference to the chosen case study, pentlandite and chalcopyrite are studied in more detail. The main sulfide minerals found at the case study are pentlandite, pyrite and chalcopyrite. Much research has been made on the recovery of chalcopyrite, while pentlandite has not been subjected to as many detailed studies.
Since copper and nickel are the valuable metals to be recovered from the tailings samples, following chapters deal with leaching of these sulfidic minerals in more detail. A summary of the leaching studies made for pentlandite and chalcopyrite are shown in Table IV. These studies will be used as references when choosing leaching reagents for the experimental part of this work.
TABLE IV Leaching studies conducted for pentlandite and chalcopyrite.
Metal sulfide Leaching reagent Reference pentlandite 1 mol/dm3 FeCl3 +
1 mol/dm3 NaCl + 0.25 mol/dm3 HCl
Maurice and Hawk (1999)
1. H2O, 2. Fe2(SO4)3
Baláž et al.(2000) H2SO4 + NaCl Lu et al. (2000b) chalcopyrite 1 mol/dm3 FeCl3 +
1 mol/dm3 NaCl + 0.25 mol/dm3 HCl
Maurice and Hawk, 1999
H2SO4 Li et al. (2010), Hansen et al.
(2005)
HClO4 Li et al. (2010), Harmer et al.
(2006)
HCl Li et al. (2010), Cai et al.
(2012), Hansen et al. (2005) H2SO4 + 0.25 NaCl Li et al. (2010)
H2SO4 + Fe2(SO4)3 Li et al. (2010) H2SO4 + K2Cr2O7 Aydogan et al. (2006)
HNO3 Hansen et al. (2005)
citric acid Hansen et al. (2005)
3.3 Pentlandite
One of the most important sulfides from where nickel is recovered is pentlandite.
Pentlandite is the most abundant nickel sulfide mineral. As many other minerals, pentlandite has usually been recovered by pyrometallurgical processes, but pyrometallurgical processes require large amounts of energy and hence the interest in hydrometallurgical processes has increased. Also the formation of sulfur oxides and hence acid rain and air pollution, have increased the interest in hydrometallurgical processes. Nowadays the gases can however be purified in order to decrease the harmful release to the environment. In hydrometallurgical treatment elemental sulfur or oxy-sulfur anions are formed instead of the gaseous sulfur oxides (Lu et al., 2000b).
Lu et al. (2000b) have studied the leaching of a pentlandite mineral concentrate in 0.4 mol/dm3 H2SO4 in the presence of oxygen at a temperature of 85 °C. The
concentrate studied composed of pentlandite (54.1 %), pyrite (20.6 %), pyrrhotite (15.1 %) and chalcopyrite (2.7 %). In addition to leaching experiments, also the amount of elemental sulfur produced was determined. The leaching of pentlandite is an electrochemical process and hence electrochemical studies were performed by Lu et al. (2000b). Electrochemical processes are processes that involve an oxidation-reduction couple and these redox reactions occur simultaneously. The electrochemical studies show that at the initial stages of the leaching reaction, a sulfur rich surface layer is formed when the iron is firstly removed from the sulfide mineral. This layer is an unwanted protective coating on the mineral and makes the oxidation process slower. However, this only happens at low oxidative conditions. At higher oxidative conditions the sulfur on the surface of the mineral is oxidized to sulfate which does not hinder the leaching of nickel from pentlandite. Sulfate is however harder to treat than elemental sulfur.
The effect of chloride ions on the leaching of pentlandite was also investigated by Lu et al. (2000b) by the addition of NaCl to the leaching solution. Some investigations included the use of chloride. Even though some of them are described here the use of chloride will be overlooked in this study since addition of a corrosive substance would mean more strict demands on the process equipment in terms of material of construction. In the absence of chloride ions the leaching was slower and less nickel was extracted. This is because chloride ions reduce the blocking of the surface of pentlandite mineral. The surface of the mineral was studied using scanning electron microscopy (SEM). The layer formed when chloride ions were present was highly porous and hence leaching of nickel was possible already at earlier stages of the process.
Maurice and Hawk (1999) studied the effect of grinding a concentrate before leaching in ferric chloride solution. Both a concentrate and an ore (having both disseminated and massive forms) were examined. The disseminated form of the ore had only at maximum 25 % sulfides whereas the massive ore contained 50-95
% or more of sulfide minerals. The sulfides found in the ore were 40-60 % pyrrhotite, 15-25 % chalcopyrite, 10-15 % pentlandite and small amounts of pyrite and cubanite (CuFe2S3). The milled concentrate was leached in a solution
