CONTROLLED REDOX POTENTIAL MICROBIOLOGICAL LEACHING OF CHALCOPYRITE by MARTHA ALICIA BLANCARTE-ZURITA IBQ Institute Politecnico Nacional 1981 MASc Chemical Engineering UBC 1984 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Chemical Engineering We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1988 ® Martha Alicia Blancarte-Zurita, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Chesntc&L &n Fe atoms Cu atoms S atoms Fig. 1. Space Lattice of Chalcopyrite LITERATURE REVIEW / 29 2.5.2. STANDARD BIOLOGICAL LEACHING Leaching of chalcopyrite can be represented by the following set of reactions: 2CuFeS 2 + 8.502 + H 2 S 0 4 •» 2CuS0 4 + F e 2 ( S 0 4 ) 3 + H 2 0 ..(15) Chalcopyrite is attacked by the bacteria, producing soluble ferric sulfate and soluble copper sulfate. Ferric sulfate undergoes hydrolysis in the medium according to: 3Fe 2(S0 4) 3 + 14H 20 = 2(H 30)Fe 3(S0 4) 2(0H) 6 + 5 H 2 S 0 4 .. (16) As the ferric iron concentration in the medium increases and the redox potential increases, ferric iron can directly attack the sulfide. CuFeS 2 + 2 F e 2 ( S 0 4 ) 3 = C u S 0 4 + 5FeS0 4 + 2S° ..(17) Elemental sulfur produced by reaction 17 undergoes oxidation in a bacterially mediated reaction. bacteria 2S° + 3 0 2 + 2H 20 • 2H2S04..(18) Reaction 18 converts the process to an acid producing process, from an acid consuming process (consider reaction 15 and acid consumption due to gangue associated with chalcopyrite such as C a C 0 3 and the precipitation of basic ferric sulfates (Krauskopf, 1967)). F e 2 ( S 0 4 ) 3 + 2 H 20 = 2Fe(OH)S04 + H2S04..(19) LITERATURE REVIEW / 30 A secondary reaction is the formation of jarosite, a basic ferric sulfate formed below pH 3 resulting from precipitation reactions of ferric sulfate (McGoran et al, 1969; Duncan and Walden, 1972 and Sakaguchi, 1976). 3Fe 2(S0 4)3 + 14H 20 = 2(H 30)Fe 3(S0 4) 2(OH) 6 + 5H2SO4..(20) The leaching of chalcopyrite usually results in 50-60% copper extraction (Bruynesteyn and Duncan, 1970). This low extraction is the result of changes that modify the mineral surface to render it impervious to attack by T. ferrooxidans and to the precipitation of basic ferric sulfates and jarosites. Regrinding of the mineral and re-leaching are necessary to further increase the copper extraction yields (Ivarson, 1973; Torma and Legault, 1973; Torma, 1977; Mehta and Murr, 1982; Keller and Murr, 1982) 2.5.3. CONTROLLED REDOX POTENTIAL LEACHING Leaching of copper concentrates under controlled redox conditions has been studied by Bruynesteyn et al, (1983, 1986) and Lawrence, et al (1984, 1985). The main differences with the standard leach are that the redox conditions of the leaching medium are controlled by the addition of sodium thiosulfate and cupric ions, and that a "catalytic" amount of silver is present in the aqueous acid leaching medium. These changes result in changing the biological leaching of chalcopyrite from an acid producing process to an acid consuming process, generating elemental sulfur and producing high copper recoveries at high biological leach rates. Another effect of leaching under controlled redox conditions is the selective leaching of chalcopyrite over pyrite and other sulphide minerals. LITERATURE REVIEW / 31 The thiosulfate added apparently serves to complex silver and dissolved copper and promotes leaching. (Bruynesteyn et al, 1986). The following stoichiometry has been proposed (Lawrence et al, 1984): CuFeS 2 + 0 2 + 2 H 2 S 0 4 = CuS0 4 + F e S 0 4 + 2S° + 2H 20 ..(21) F e S 0 4 + l/40 2 + 1/2H 2S0 4 = l/2Fe 2(S0 4) 3 + 1/2H20..(22) l/2Fe 2(S0 4) 3 + 7/3H20 = l/3(H 30)Fe 3(OH) 6(S0 4) 2 + 5/6H 2S0 4 ..(23) Ferrous iron produced by reaction 22 is oxidized by the bacteria to ferric iron in reaction 20 and then hydrolyzed to form jarosite. The reactions that take place to produce the low redox conditions have been grouped under the term of chemical activation of the concentrate, because addition of thiosulfate and cupric ions gives more consistent results when added at the beginning of the leach, thus "activating" the concentrate. 2.5.4. CHEMICAL ACTIVATION This stage consists of mixing an excess of silver complexed with thiosulfate, copper sulfate solution and the concentrate. At Eh values lower than 600 mV (SHE), under which conditions leaching takes place, the ferric iron is not "an effective oxidant. In such conditions the cupric-cuprous couple acts in the oxidation process. Cupric ions are reduced and cuprous thiosulfate is formed according to: LITERATURE REVIEW / 32 C u 2 + + S 2 0 3 2 - = C u + + 1/2S 40 6 2" ..(24) 2 C u + + S 2 0 3 2 - = C u 2 S 2 0 3 ..(25) As cuprous thiosulfate is formed, the excess thiosulfate protecting the silver thiosulfate is consumed, and precipitation of silver sulfide, cuprous and cupric sulfide occurs. C u 2 S 2 0 3 + H 2 0 = Cu 2S + H 2 S 0 4 ..(26) A g 2 S 2 0 3 + H 2 0 = Ag 2S + H 2 S 0 4 ..(27) It is thought that these sulfides are oxidized at the mineral surface with corresponding acid consumption and elemental sulfur production. Then chalcopyrite dissolves to replace the copper sulfide and re-establish the equilibrium. Thermodynamically, it can be shown that Cu 2S cannot coexist with elemental sulfur, resulting in the formation of CuS (Hiskey and Wadsworth, 1981); Cu 2S + S° = 2CuS ..(28) which is then oxidized: CuS + l/20 2 + 2H +- = C u 2 + + H 2 0 + S0.. (29) The equilibrium is re-established by the following silver mediated reaction (Lawrence et al, 1984): CuFeS 2 + C u 2 + = 2CuS + F e 2 + ..(30) LITERATURE REVIEW / 33 At this stage, an active culture of T. ferrooxidans is introduced. It is postulated that bacteria oxidize the ferrous iron which forms at the solid-liquid interface as a result of reaction 30 (Bruynesteyn et al, 1983). Following leaching the solid residue typically contains unleached chalcopyrite, unattacked pyrite, basic iron sulphates and elemental sulfur (Bruynesteyn et al, 1986). The biological and biochemical aspects of the controlled redox potential leach are not well undestood. Difficulties in reproducing this process in terms of leaching rates have been encountered by Cooper (1985) and Ballester (1985) leaching chalcopyrite concentrate. Their problems have been attributed to surface oxidation of the concentrate, and to the particular culture of T. ferrooxidans used. During the course of his studies Ballester (1985) embedded chalcopyrite particles in epoxy and then polished the surface. These polished grain surfaces were exposed to a leaching environment. The only continuous deposits of Ag 2S that were observed after leaching occurred in cavities that were present in the chalcopyrite samples before leaching. In his chalcopyrite study he used a A g 2 S 0 4 concentration of 0.65g/L; the exposed surface area of the chlcopyrite particles in the epoxy probe was 1.2 x 10^ Mm2. In the sort of concentrate leaching, which is the subject of the present thesis, the silver concentration was similar (0.58g Ag2S04./L) but the exposed chalcopyrite surface area was of the order of 9 x 10 1 2 Mm2. LITERATURE REVIEW / 34 2.6. FACTORS AFFECTING MICROBIAL LEACHING 2.6.1. METAL TOLERANCE T. ferrooxidans tolerance for metals has been widely recognized but specific limits are difficult to establish due to variations of tolerance among different strains. Metal toxicity depends on the physiological state of the microorganism, on the chemical form of the metals and on the degree of interaction of the metals with the environment (Norris and Kelly, 1978). Tuovinen and Kelly (1972) reviewed the limits of tolerance for a variety of metals, with the limit for copper being set at lOg/L, this seems very low since T. ferrooxidans has frequently been reported to grow at much higher copper concentrations see Bruynesteyn et al, 1986. With respect to silver the limit reported by Tuovinen and Kelly (1972) was 50-100mg/L. Ehrlich (1986) reported that some strains tolerated silver concentrations of up to 170mg/L. In the presence of metals T. ferrooxidans exhibits a lag which is presumably an adaptation or selection period, before iron oxidation can occur (Tuovinen et al, 1971). T. ferrooxidans is more tolerant to heavy metals when it is growing in ferrous iron as opposed to thiosulfate solutions. This suggests that certain enzymes may be more succeptible to heavy metals. Silver forms complexes with cysteine, consequently it will affect enzymes that depend on thiol groups for their activity (Gruen, 1975). Silver also accumulates in T. ferrooxidans cells, this accumulation results in Mg and K losses. LITERATURE REVIEW / 35 Interactions of silver with the medium, will decrease the effective cation concentration, thus explaining some of the differences in the toxic concentrations reported. This is the case of the increase in resistance to silver reported for T. ferrooxidans when cultured on sulfides. Silver has affinity for sulfur and will bind to it forming Ag2S, reducing the effective concentration hence reducing the toxicity of the cation. A similar pattern has been observed for the case of copper toxicity. Copper in concentrations of 0.1-1.0 M inhibits iron oxidation and CO2 fixation (Tuovinen and Kelly, 1974). In this study cell growth occurred even when copper concentrations, increasing throughout the leach, reached 1.5M. For controlled-redox leaching initial C u + + concentrations of 0.45M (copper added during conditioning of the concentrate) had no effect on iron oxidation. 2.6.2. NUTRIENTS T. ferrooxidans, being an autotroph, utilizes carbon dioxide as a carbon source. This is generally supplied in the form of carbon dioxide enriched air, because carbon dioxide consumption by active cultures could exceed the rate at which CO2 can be supplied to the media (Tuovinen, 1972) due to its relatively low solubility. In addition to a carbon source, a nitrogen source, usually ammonium ion is needed, other nutrients such as potassium, calcium and magnesium are supplied by the medium which is usually the 9K medium of Silverman (1959). Magnesium is necessary for carbon dioxide fixation and also for energy fixation, and is a factor in a variety of metabolic reactions. Phophorus also participates in energy metabolism. (Tuovinen et al, 1971). LITERATURE REVIEW / 36 2.6.3. PARTICLE SIZE AND SURFACE AREA Leaching of mineral concentrates involves the surface of the concentrate particles. There are two ways in which we can study the effects of changes in the amount of surface area per unit volume on the leaching process. Increased surface area can be the result of decreasing the size of the solid particles for a fixed amount of mass of solids (eg. by grinding the concentrate), or by increasing the mass of solids per unit volume maintaining the same particle size (increasing the pulp density). In the leaching of sulfide minerals the rate of bacterial oxidation increases when the particle size of the mineral decreases (Duncan et al, 1966; Bruynesteyn and Duncan, 1970; Torma et a l / 1972; Sakaguchi et al, 1976). Copper leach rates increase linearly with increasing solids concentrations up to 60g/L, above this value other factors such as particle sedimentation and limited mixing come into effect. This is a pulp density effect (Pinches et al, 1976). Gormely et al (1975) found that for zinc sulfide, the leach rates were first order with respect to surface area per unit volume. Pinches et al (1976), working with Zambian concentrate of chalcopyrite, found that copper yield was proportional to the external surface area for particles larger than 7 /un. They found that leaching rates increased linearly with decreasing particle size up to 10 Mm, for this size the copper extraction rate was 102mg/L h. Below 10 Mm the dependence of the copper extraction rate with particle size became non linear. For particles of 5.4 Mm the copper extraction rate was 121mg/L h. Below 4 Mm the leach rate was less dependent on particle size and for 2.1 Mm particles the rate was 134mg/L h. They suggested that this complex behaviour may be the LITERATURE REVIEW / 37 result in changes in the active leaching volume which would be a constant for particles larger than 7 um, but decrease for smaller particles. Alternatively, they suggested that this behaviour might be the result of changes in the interaction of the particles with the bacterial cells and flocculation of the smaller particles. Available area for the reaction is a function of particle size, and for a fixed amount of solid material, increases with the fineness of the material. In theory the optimum particle for leaching would be a single crystal (Tuovinen, 1972). Leaches of Newmont concentrate of chalcopyrite under standard conditions conducted with monosize particles of less than 7 t^m, showed that the maximum copper extraction rate (28mg/L. h), and maximum extent of leaching (98%) were obtained when the particles had a size comparable to the bacterial size (1 nm) (Blancarte—Zurita, 1983). This effect could only be characteristic of biological leaching. Jones and Peters (1974) found that for abiotic Caching of chalcopyrite, copper extraction did not improve for particles below 150 fim because of the mechanism of attack of ferric iron lixiviant. Leaching takes place along grain boundaries and fissures, so particle reduction stops being beneficial when no more grain boundaries can be exposed by grinding. The effect of grinding in abiotic leaching was also studied by Beckstead et al (1976) who found extractions improved with particle size reductions down to 0.5 ixm. 2.7. MODELLING BACTERIAL LEACHING 2.7.1. MODELLING AND KINETICS LITERATURE REVIEW / 38 The search for a pattern of substrate utilization and product formation that adequately describes bacterial leaching has been seriously limited because of difficulties in characterization and quantification of the bacterial populations, and their involvement in the leaching of metal values. Knowledge of the mechanism of bacterial attack on a sulfide concentrate would be useful in producing better models ..for the design and control of leaching systems. Initially attempts were, for the most part, constrained to investigations of operating parameters and their effect on metal leaching. Roman and Brenner (1973) discuss the effects of particle size, stirring speed, oxygen concentration, temperature and reagent concentrations on the rate of dissolution of different copper concentrates for both biological leaching and abiotic leaching. The complexity of the kinetic descriptions that are necessary to describe leaching of minerals can be estimated by reviewing the literature of models for electrochemical processes and non-biological leaching. Dutrizac et al (1971), proposed parabolic kinetics for abiotic leaching of chalcopyrite. They found the rate controlling step to be diffusion of ferric sulfate through a thickening sulfur layer formed on the surface of the chalcopyrite. Dutrizac and MacDonald (1974) reported that there was no general agreement with respect to the type of kinetics for abiotic leaching of chalcopyrite, since some investigations had shown linear kinetics (Lowe, 1970). Sepulveda and Herbst (1978) proposed a population balance equation with kinetics based on topochemical leaching (a shrinking core rate expression) for leaching of chalcopyrite. They found the leaching reaction to LITERATURE REVIEW / 39 be electrochemical in nature with the conduction of electrons through the sulfur layer as the rate controlling step. The kinetics followed a shrinking core model. 2.7.2. MODELS IN BIOLOGICAL LEACHING Difficulties in modelling abiotic leaching are only compounded by the presence of cells in biological leaching. Incorporation of all the different relationships among the components of the leaching medium in a theoretical description that includes all the interactions between the cells and the substrate is for the moment unattainable. This difficulty can be explained if we consider a typical leach medium where several processes are occurring concurrently, these involve reactions in solution, . acid-base equilibria, precipitation reactions, changes in pH and rheology of the medium, dissolution of the mineral, gas-liquid exchanges, etc. Also, from the cell point of view, in batch growth the following processes are taking place -induction, repression, growth, death, lysis, etc. Sanmugasunderam et al (1985) applied a model proposed by Gormely et al (1975) which was capable of predicting the rate of growth of T. ferrooxidans on ZnS concentrate but which was not successful in predicting the rate of metal release. Attempts to simplify the description of a leaching system have been made by grouping all the variables derived from the interactions of the cell with the solid substrate in an experimentally measured leach rate. It is possible then to study the process from the substrate point of view, gaining insight into the controlling variables and the general mechanisms. LITERATURE REVIEW / 40 Models developed from experimental, microbiological leaching data have been described for various substrates. Two main approaches have been followed. One approach considers the solid subject to uniform conversion, the other approach considers the solid to be attacked at loci which can be randomly distributed on the solid surface. Both approaches are valid depending on the substrate being considered. For locus conversion, models of pore propagation have been proposed. Locus conversion is clearly applicable for coal desulphurization where coal particles could be seen as containing nucleii, in the form of pyrite inclusions in coal. For this case leaching takes place through moving boundaries around these nucleii, that meet and coalesce, removing pyrite from the coal matrix. Another example is leaching of refractory gold—bearing pyrite where penetrating pores develop and leaching takes place through pore propagation. Selective leaching in the gold rich regions of the concentrate lead to high gold liberation by subsequent cyanidation of the leaching residues. For 30% sulfide oxidation, gold liberation was 90% or higher. Leach rates depended on surface area concentrations but not on particle size (Hansford and Drossou, 1987). Southwood and Southwood (1986) proposed a hybrid model based both on propagating pores and shrinking cores for the leaching of auriferous pyrite. They found that biological leaching of auriferous pyrite, caused the development of pores that coincided with preferred dislocation planes in the crystal lattice. These dislocations were caused by gold inclusions. The pores represented an increase in the surface area that was not accounted for by using the shrinking core model exclusively. For the case of waste ores containing chalcopyrite, sphalerite and pyrite (average metal contents of 0.82% Cu LITERATURE REVIEW / 41 and 0.82 % Zn), Bruynesteyn and Duncan (1970) proposed that the controlling factor in biological leaching was the active leaching volume, made up of the particle surface area and the depth of penetration of the bacteria and the leach liquor into the particle. The active leaching volume became an increasingly large proportion of the total volume as the particle size decreased, thus explaining the exponential increase in leach rates obtained as the particle size decreased. Waste ores could also be considered a special case of the locus conversion with the locus being the minerological phases on the host rock. The number of loci per unit mass, increases for small particle sizes due to the increase in surface area exposed, thus explaining the increased leaching rates for small particle sizes. For chalcopyrite and sphalerite concentrates leaching is believed to occur through uniform conversion. Until now no evidence of pore formation has been found for leaching of chalcopyrite concentrates (Blancarte-Zurita, 1983). The pores observed by Southwood and Southwood (1986), had diameters of the order of 5-25 jim. Pores observed in pyrite by Duncan and Drummond (1973) were around 70 in diameter. These pores are much larger than the pits observed on the surface of leached concentrate particles which are of the order of 2 Mm (Blancarte-Zurita, 1983). For uniform conversion, a shrinking particle mechanism has been proposed. 2.7.2.1. SHRINKING CORE, SHRINKING PARTICLE MODELS Gormely et al (1975) proposed the use of a shrinking core model for leaching of zinc sulfide concentrate. They estimated from experimental data, that the complete mineral surface would be covered by the bacteria during leaching and that a LITERATURE REVIEW / 42 model based on Levenspiel's model (1972) for a shrinking core, spherical particle when chemical reaction at the surface controls the process could be applied. Using a similar zinc sulfide concentrate Sanmugasunderam (1981) pursued the application of a shrinking core model. Using CSTRs in series he achieved steady state in two stage reactors run in straight through mode and found that the amount of leaching could be approximated by a shrinking core model. In previous papers (Blancarte-Zurita et al, 1986a, 1986b, see Appendix 8) we have proposed a variation on the shrinking core model to describe the microbiological leaching of sulfide mineral concentrates. This model is based on the whole particle shrinking rather than on a constant diameter particle which has a shrinking reactive core. The standard leaching of sulfides has been extensively studied for a variety of ores. Trends can be compared among different systems, but quantitative results tend to depend on the ore mineralogy and the history of the microorganism used. Our work in conventional leaching was focused on obtaining some reliable kinetic data on which to base a comparison with the controlled-redox process. At the same time we used techniques such as x ray analysis to investigate cell-substrate interactions. With respect to controlled-redox leaching, until very recently all work done on this subject was carried out at B.C. Research. Their work is considered the foundation for the present study. Recently Cooper (1985) and Ballester (1985) have done some experiments trying to duplicate B.C. Research work. Most of their work was carried out at low pulp density and it was not successful (Ballester, 1987). This can be explained by the portion of this work that examines pulp density effects (appendix 8). CHAPTER 3. THEORY 3.1. SHRINKING PARTICLE MODEL FOR CONCENTRATE LEACHING 3.1.1. DEVELOPMENT OF THE MODEL During leaching, material is being solubilized from the concentrate particles. It is assumed that the particle gets smaller. This would not be the case if there were enough inert material associated with the material being leached so as to maintain the physical integrity of the particle. It also ignores the deposition of any solids on the particle surface such as jarosite, which would tend to increase the particle size. It also assumes that the particle composition remains unchaged as the particle shrinks and its components are solubilized. The rate of decrease in particle size is given by — dn/dt where n is the particle mass and t time. Assuming that the particle has a constant density (p) Where V is the particle volume. We can represent V by V = adp^ where dp = particle diameter, and a = shape factor. Now the change in particle size is dn dt P dt dV ...(31) p dt d[ad p 3] which, if we assume a constant shape factor, becomes 3 a d n 2 43 THEORY / 44 Experimentally the rate of appearance of the metal in solution can be measured; call this rate r which is expressed in unit of mass/(volume x time). To convert the rate of metal dissolution to a rate of particle dissolution we have to divide r by the weight fraction of metal in the concentrate. Let this weight fraction be f. So the rate of particle dissolution is r/f. , 9 d [ d P ] - 3 a d n 2 p — ^ y dt represents the dissolution rate of a single particle; the measurement of r is made in a system containing many particles per unit volume. Assume that no particles disappear during leaching, then the number of particles is constant and is equal to Co/(ap dp 0 3) where Co is the initial mass concentration of particles and dp 0 is the initial particle diameter. So: 3 p a d n 2 Co d[dn] r -( ?—7— ) — ? = ...(32) a p d p o 3 dt f or d [dp] r d 3 = ( P n ) -(33) dt f 3 C 0 d p 2 Integration of equation 33 requires a knowledge of how r varies with respect to particle size and time. This requires experimental measurements. These can be done in shake flask studies or other lab scale reactors by leaching suspensions of several, more or less monosize fractions of concentrate particles, and measuring metal ion concentrations in solution as a function of time. The slope of a plot of metal concentration vs time gives a value for r. THEORY / 45 Solution of equation 33 gives concentrate particle diameter as a function of time. Experimental support for the validity of equation 33 has been provided by Blancarte-Zurita et al (1986a,b). See Appendix 8 and section 5.2. These papers in Appendix 8 should be considered to be a part of the work of this thesis. They are appended rather than included in the main text only to avoid reproducing material already printed. The fractional extraction (X) of metal, assuming it is uniformly distributed throughout the concentrate particle, is given by d p 3 X = 1 £- ... (34) d 6 upo Levenspiel (1972) has shown that the mean fraction ( X m ) not extracted in a continuous, stirred tank reactor is 1 - X m = Jo (1 - X) Edt ... (35) where E is the age distribution of the particles in the reactor outlet. For a perfectly mixed reactor 1 t E = - exp [- - ] ... (36) T T Where T = mean residence time. Since particles can disappear as a result of leaching in a finite time equation 35 should be rewritten, after combining with 34 and 36, as THEORY / 46 1 - X, m exp [ — t 7 ] dt ... (37) where T is the time it takes for particle diameter to go to 0. A digital computer program has been written to solve equation 37. The required input data for this program are 1. concentrate particle size distribution 2. an empirical equation or equations, derived from experiment, relating r to dp or some other relationship such as tables of values of r at various values of time and particle size. 3. feed pulp density 4. mean residence time in the reactor 5. fraction of metal in the concentrate 6. an arbitrary maximum time ( T m a x ) to leave the program when retention times exceed practical levels for the conversion specified 7. the number of tanks in series if multiple stage simulation is contemplated. For concentrate having a particle size distribution, the procedure is discussed on Blancarte-Zurita, 1986a (appendix 8). The particle size distribution puts particles into various size categories. All particle diameters in each size category are represented by the mid-range particle diameter. The program then starts with one fraction and does its calculations. Then it repeats them for the next size range etc. After a time interval At the particle size is calculated using equation 33 and the information obtained from the lab scale experiments which give r as a function of d p. Then the calculated particle diameter is compared with the THEORY / 47 diameter range of applicability of the particular r(dp) relation being used to see if the same one should be used in the next step or if a different one is required. If the same one is to be used the calculation is repeated and the time elapsed compared to an arbitrary, maximum time ( T m a x ) . If the time is less than T m a x the calculations are repeated. When t = T m a x the average particle size is printed. The range of validity of each equation is included as part of the program (see appendix 4). If at any time the particle shrinks to such an extent that it moves to a different size class so that a different r(dp) equation is needed, the time has to be set = 0 again to get the proper values of dp in subsequent calculations. This is corrected for when T m a x is reached. Next conversion is calculated using equation 34. The mean conversion is calculated using equation 37. In the simulation of a single stage of leaching if the % extraction is less than a preassigned value, T m a x is increased and the calculations continued or the run is abandoned with the conclusion that the retention time required is too long for any practical consideration. In the simulation of multistage leaches if the specified % extraction is not achieved in the first stage another tank in added to the system and the calculations repeated. When a second or subsequent tank is added a new feed pulp density has to be calculated. If % extraction is to equal or greater than the preassigned value the program prints the % extraction. All of the above is explained in the computer program logic diagram of Fig 2. THEORY / 48 START \-DATA ^FRACTIONS SIZE DISTRIBUTION MEAN RESIDENCE TIME NUMBER OF TANKS READ FOR FRACTION N Zn PERCENTAGE IN CONCENTRATE INITIAL PARTICLE SIZE dpo | 'f LESS THAN NT f ELSE CALCULATE NEW SIZE DISTRIBUTION CALCULATE INITIAL UASS CONCENTRATION I'.NF'JT REACTOR N*\| CALL LEACH [ C A L C U L A T E C O N V E R S I O N ] INCREASE NUUBCR or TANKS j i r EQUAL TO NT \ t VERSION I [If CQNVEI RACTIONS | 1 EQUAL * I N C R E A S E I S T O R E T I ' M T S E A R C H F O R S I Z E C L A S S C A L C U L A T E R A T E F O R O p o C A L C U L A T E S H R I N K I N G N E W O p T E S T O P E I S N E W O p L E S S T H A N L O W E R L I M I T O F C L A S S T E S T T I M E L A R G E R T H A N T M A X , r L C A L C U L A T E A V E R A G E P A R T I C L E S I Z E D I S P L A Y V A L U E O F T I M E A N D D p I N I T I A L I Z E T I M E r C A L C U L A T E E L A P S E O T I M E S I N C E L E A C H B E G A N -j END | Fig. 2 Logic diagram of the computer simulation In order to test the relevance of this computer program some comparisons were made between % extractions of Zn from ZnS concentrate, as measured experimentally by Sanmugasunderam et al (1985), and % extractions predicted by THEORY / 49 the program for the conditions of those experiments. The calculated and measured results are presented in Table 1 and Fig. 3. Agreement is within about +20%. The results presented in Table 1 are for runs 1-13 made by Sanmugasunderam (1985) for one stage leaching of ZnS concentrate. The availability of the program described above allows one to predict the effects of changes in the operating parameters of the leach reactor on such things as the % metal extraction. For example Fig. 4 plots % Zn extraction as a function of mean residence time for a single stage reactor with a pulp density of lOOg/L. Fig. 5 shows the effect of adding a second stage with the same residence time. Percentage zinc values are the overall extraction of the concentrate after passing through the two reactors as a function of residence time in each tank. Reactor #2 feed pulp density depends on the levels of extraction of reactor #1, and varies between 59g/L for 10 h residence time in the first reactor, to 24g/L for 300 h. Reactor #2 conversions go from 40—60%. Total conversion froth reactors combined) as a function of residence time (Fig. 5) shows that the advantage of using Reactor #1 and Reactor #2 in series is a +40% increase in conversion over a single tank. THEORY / Run r FEED PULP ZINC % ZINC % DENSITY EXP CALC (hours) (g/L) 1 2 3 4 5 6 7 8 9 10 11 12 13 83.7 102.0 102.0 102.0 117.6 76.9 76.9 76.9 76.9 76.9 66.6 66.6 66.6 57 123 159 206 311 213 143 177 205 280 73 145 76 54.4 48.7 51.8 51.8 41.0 52.1 43.3 37.1 42.0 33.7 57.6 55.1 57.1 66.5 51.8 45.6 39.7 36.4 38.5 46.9 42.3 39.2 25.6 60.5 44.8 40.7 Table 1 Experimental and calculated percentage of zinc extraction THEORY / 51 • E X P E R I M E N T A L O C A L C U L A T E D O a o O o o O ' i 1 1 1 , r— 50 100 150 200 250 300 350 Feed pulp density (g / l ) Fig. 3 Zn extraction as function of pulp density THEORY / 52 R e a c t o r o n e 100 - r — OH 20H 1 I 1 i f— 100 ISO 2 0 0 250 3 0 0 M e a n res idence t ime (hours) Fig. 4 Zn extraction reactor #1 of a series of 2 THEORY / 53 R e a c t o r t w o 100 -1 — 20-i 1 1 1 40 60 M e a n res idence time (hours) 80 Fig. 5 Zn extraction reactor #2 of a series of 2 CHAPTER 4. MATERIALS AND METHODS 4.1. COPPER CONCENTRATES Leach experiments were carried out on concentrates, obtained from the B.C. Newmont Mines Limited. two different commercial flotation copper mines of Gibraltar Mines Limited and 4.1.1. S A M P L E PREPARATION The Newmont concentrate had previously been wet ballmiiled, as a lot of 200Kg, at 55% solids for 1 h to 91.8 % -400 Tyler mesh, filtered, and dried at 60°C. Gibraltar concentrate was a lot of 50Kg and was wet ballmiiled at 30% solids for 70 minutes to attain a similar size reduction to that of Newmont concentrate. It was wet sieved to 100% -400 mesh, filtered and dried at 50°C. 4.1.2. SCREEN ANALYSES A sieve analysis of the as-received Gibraltar concentrate was performed on a 200 g sample obtained using a chute splitter, from the contents of an original lot of 60 kg. It showed 45.8% -400 Tyler mesh; the results of a sieve analysis of this concentrate are shown in Table 2. 54 MATERIALS AND METHODS / 55 Tyler Mesh + 150 11.90 13.71 13.66 5.39 9.50 45.82 -150 +200 -200 +270 -270 +325 -325 +400 -400 Table 2 Sieve Analysis of Gibraltar Concentrate 4.1.3. ASSAYS Both concentrates were assayed for copper, iron and sulfur (for the procedures see appendix 1). Results of these analyses are tabulated in Table 3. 4.1.4. PARTICLE SIZE 4.1.4.1. CONCENTRATE FRACTIONATION A simple sedimentation method was used to separate the concentrate into "monosized" fractions. The beaker decantation method (Pryor, 1965) is based on differences in the terminal velocities of spherical particles falling through a fluid at such rates that the Reynolds' number is less than 0.2. This terminal velocity can be calculated using Stokes' law for any given particle size, and from the velocity a sedimentation time can be estimated. MATERIALS AND METHODS / 56 ELEMENT (% by weight) NEWMONT GIBRALTAR COPPER IRON SULFUR INSOLUBLE 27.8 28.0 31.1 5.5 25.7 26.2 31.3 16.8 Table 3 Concentrate Elemental Analyses For the chalcopyrite concentrates the actual particle size of the "monosized" fractions obtained by this method, differs from those calculated by Stokes' law, because of the non-spherical character of the mineral particles and the entrapment of small particles during the settling of the larger ones. To minimize this last problem and obtain homogeneous dispersions a dispersant, sodium hexametaphosphate was added to the particle suspension and subsequently removed from the settled "monosized" fractions by filtering and resuspending the cake in water two or three times. A complete description of the procedure is found in Blancarte-Zurita (1983). The particle sizes of the "monosized" fractions were measured by image analysis as described below. 4.1.4.2. PARTICLE SIZE ANALYSES Particle size distributions were measured by image analysis, using a Leitz Image Analyzer. In order to analyze the concentrate particles obtained from the leach tests, MATERIALS AND METHODS / 57 samples of the suspensions were filtered through millipore membranes (0.2 Mm) and the solids were washed with distilled water, to remove soluble salts. For samples of elemental sulfur producing leaches, the solids were also washed with ethanol and carbon disulfide, to remove sulfur particles. The cake was finally resuspended in ethanol. Solid particles from the dry concentrate were directly suspended in ethanol. Using an ultrasonic bath the samples were dispersed and a portion of the suspension was withdrawn using a Pasteur pipette with a broken tip and transferred to a microscope slide; the slide was then allowed to dry and was mounted on a microscope stand. A television camera captured the image and transferred it to a TV imaging device. The TV screen was calibrated according to the magnification used in the microscope. After editing the image and adjusting contrast and focus, the image was processed by means of a program called TASSEP.ENG (software, Leitz-image analyzer). In this program, size measurement is based on continuous erosion of the two-dimensional image of each individual particle within a field. The number of step erosions are counted and the equivalent diameter of the particle calculated. A meander is pre-programmed into the computer which allows the examination of the slide mounted on the microscope stand, field by field automatically. The meander consists of four parameters specifying width and length of field of examination and number of fields to examine to the right or left, up or down from the initial position of the microscope ocular in the sample. For size analyses a double blind procedure was used; ihe samples, drawn sequentially from the experiments, were number coded and analyzed randomly, so as not to influence the selection of MATERIALS AND METHODS / 58 magnification. In every group of samples analyzed at least one duplicate was examined. Duplicates from the same sample were analyzed on alternate days, to ensure that the apparatus was performing optimally. The average particle sizes measured using the image analyzer on the monosize fractions together with their sedimentation times are shown for as received Gibraltar concentrate in Table 4a, and for -400 mesh Newmont concentrate in Table 4b. The median particle size is often employed as a characteristic size for particulate systems. It is defined as the particle size for which the amount of particles equals 50% of the total. For number distributions it is also called the number median size. (Stockman and Fochtman, 1978) 4.1.4.3. REDOX POTENTIAL A reaction involving chemical substances, neutral molecules and/or positively or negatively charged ions and free electric charges (i.e. electrons arising from a metal) is an electrochemical reaction. For any given electrochemical reaction taking place under fixed physico-chemical conditions, there exists a fixed value of the electrode potential at which the equilibrium state of the reaction is obtained. The redox potential (Eh) is the value of the equilibrium potential difference of the cell formed by coupling the electrode on which the reaction is taking place with a reference electrode (hydrogen electrode, calomel electrode, silver chloride electrode) i.e. the value of the potential difference when the whole of the system is in electrochemical equilibrium (Pourbaix, 1974). In this study the combination of electrodes used was platinum/calomel. MATERIALS AND METHODS / 59 Settling time Mean size 50% value (sec) (Mm) (Mm) 7.56 11.8 21.0 47 189 832 46.8 43.7 38.2 30.4 14.3 8.6 2.5 47.6 43.8 41.2 30.9 14.4 8.6 2.1 oo Table 4a Size Fractionation for as received Gibraltar Concentrate 4.1.4.4. DENSITY Density of the concentrates was determined in a liquid pyknometer by measuring the mass and volume of a sample of particles suspended in distilled water plus sodium hexametaphosphate as dispersing agent. The procedure is described in Allen, (1981). The average density was found to be 3.83 ± 0.07 g/cra3 at 23°C. 4.1.4.5. SPECIFIC SURFACE AREA Specific surface area (ssa), expressed as square centimeters per gram of concentrate, was determined using the Standard Test Method for Fineness of Portland Cement by Air Permeability Apparatus (ASTM, Designation C204-81) following the plateau principle suggested by Price, (1976). The Blaine Air Permeablity Apparatus allows a definite quantity of air to be MATERIALS AND METHODS / 60 Settling time M e a n size 5 0 % value (sec) (Mm) 21.2 37.8 84.9 340.9 1500 6.4 5.9 3.5 2.8 2.7 0.2 4.00 4.10 2.20 2.10 1.80 0.18 Table 4b Size Fractionation for -400 mesh Newmont Concentrate passed through a fixed bed of compressed powder with porosity of 0.5 + 0.005. The size of the particles present in the bed will determine the number and size of the pores in the bed and consequently the rate of airflow through the bed. For each measurement, several beds of concentrate were prepared using different amounts of material in 1 g intervals and the ssa was determined. In general small variations of ssa around one particular value occurred, indicating that this weight produced a bed of the required compactness. For better accuracy using this weight, beds were prepared with the exact weight and that weight ±0.05 g and the rates of airflow through the three beds were measured in triplicate, the ssa for a given sample was the average value calculated from the measured rates. Table 5 presents the average values of ssa for the "monosize" fractions of Gibraltar concentrate. MATERIALS AND METHODS / 61 Table 5 Specific Surface Area for Monosize Fractions Mean Diameter SSA (Mm) (cmz/g) 46.8 43.7 30.4 14.3 8.6 2.5 453 637 808 1398 1816 4132 4.2. BIOLEACH EQUIPMENT AND METHODS 4.2.1. S H A K E F L A S K S Shake flasks were used for culture maintenance and for the preparation of bacterial inocula. In all cases, 7.5 g of concentrate were placed in bottom—baffled, 250 ml Erlenmeyer flasks; 70 mL of iron—free 9K medium solution (Silverman and Lundgren, 1959) were added and the pH of the suspension adjusted to 2.0 with IN H2SO4. The flasks were stoppered with a plastic foam plug and incubated overnight (to allow for acid consumption by gangue) in a dark room with a carbon dioxide enriched atmosphere (dry carbon dioxide, Canadian Liquid Air Ltd.). After incubation, the flasks were removed from the shaker, the pH was adjusted back to 2.0 and the flask was sealed using parafilm and stored under refrigeration at 4°C until used. The temperature of the incubation room was controlled at 35°C by means of a temperature controller (Honeywell type RP 908). MATERIALS AND METHODS / 62 4.2.2. BEAKERS Preliminary bioleach tests on "monosized" fractions were carried out in 100 mL beakers which were agitated by a magnetic stirrer. The pH of the pulps was controlled by using a pH meter—controller, to activate a peristaltic pump which delivered 12N H2SO4. The beakers were covered with acrylic plates to avoid losses due to evaporation. These plates were perforated to insert the pH electrode, which was immersed into the agitated pulp, and to insert the acid feed line for pH control. 4.2.3. 1L TANKS A Virtis Omni—Culture Benchtop laboratory fermenter (No.6700-0060) was used for small scale, tank leach tests. The fermenter consisted of a 1 liter borosilicate, baffled, glass culture vessel, a stainless steel perforated headplate and two six bladed (4.9 cm in diameter), shaft mounted turbine impellers. The impellers were bottom-driven by a magnetically coupled drive. The working volume of the fermentor was 350mL. Temperature was controlled by cooling coils. A mixture of air—carbon dioxide (1% v/v) was introduced through a sparger situated between the impellers. Temperature, agitation and air supply were controlled from a panel located under the fermentor. The vessel contents were maintained in the dark using a removable exterior cloth cover. Visible light has been shown to be inhibitory to some species of Thiobacilli (LeRoux and Marshall, 1977). For controlled-redox leaches the pH of the pulps was maintained at 2 .2 as MATERIALS AND METHODS / 63 described for the beaker experiments. A diagram of the system is shown in Fig. 6. 4.2.4. 3L TANKS Baffled, plexiglass tanks equipped with 4 blade, turbine impellers were used for large scale tests. The working volume for these tanks was 2L. The impeller motor was located on top of the reactor, a perforated headplate made of plexiglass was used to cover the reactor, the perforation was in the form of a radial cut so the headplate could be removed, while the impeller was operating, to scrape the walls of the tank, and clean the top of the electrode from material carried out of the liquid with the foam. A mixture of air—carbon dioxide (1% v/v) was metered in with a rotameter and introduced under the impeller. The tanks were maintained in the dark in a controlled temperature room at 35°C. For catalyzed leaches, the pH of the pulps was controlled as described above. The operation diagram is the same as for the IL fermentor. 4.2.5. CONTROLLED REDOX LEACH TESTS The procedure for preparing 2L of medium for standard catalyzed leaches was as follows. A solution containing 1.16g A g 2 S 0 4 in 400mL of distilled H 2 0 was mixed with lOOmL of a solution containing 12.4g Na 2S203.5H20. Then 400g of concentrate were added and the resulting suspension mixed for 20 minutes. Next IL of a solution containing 142.8g of Cu 2S04.5H 20 was added to the mixed suspension. After a further 15 minutes of mixing, 20mL of each of the nutrient MATERIALS AND METHODS / 64 3 — K D 4 0 10 1 SULPHURIC ACID RESERVOIR 2 pH M E T E R CONTROLLER 3 CO, HUMIDIFIER < AIR HUMIDIFIER 5 SAMPLING PORT 6 CONDENSATION FLASKS 7 pH ELECTRODE 8 STIRRER 9-10 CONSTANT TEMPERATURE RECIRCULATION SYSTEM II LEACH TANK Fig. 6 Leaching system diagram MATERIALS AND METHODS / 65 salt solutions required to produce iron free, 9K medium, were added. 4.2.6. LEACHING TESTS In all cases leaching tests were set up according to the procedures described above. After inoculation the sampling schedule was the same for controlled-redox and standard leaching tests. Every day water losses due to evaporation would be made up with distilled water, and the walls of the reactors and the surfaces of the electrodes, baffles, etc would be freed from froth. Froth accumulated when foam formed as a result of intense aeration and mixing. At this point measurements of temperature, acid consumption (for controlled redox leach only), pH, dissolved oxygen and Eh would be carried out. Every other day 2—8mL samples, depending on the total leach volume, would be drawn and processed according to the analyses to be carried out. For total organic carbon a 1ml sample was transferred to a vial and 2mL of 5N H2SO4 were added, and the sample was stored at 4°C until analyzed. For size analyses a lmL slurry sample was used and processed as described under size analyses. For dissolved metal analysis the sample was allowed to sediment and a lmL portion of the clear supernatant was diluted with 0.01 N H2SO4. Leaches were terminated after no more acid consumption occurred for catalyzed leaches and when copper extraction stopped for standard leaches. At that time 20mL samples of the residue were taken and dried. These samples were used for elemental sulfur analyses. 4.3. BACTERIAL CULTURES MATERIALS AND METHODS / 66 4.3.1. SOURCE OF CULTURES The bacterial culture was a strain of T. ferrooxidans originally isolated from the Britannia Mine near Vancouver (Razzell and Trussell, 1963), and routinely maintained on copper concentrate at B.C. Research. Subculturing on different concentrates in non sterile conditions could in time give rise to mixed cultures. See Appendix 2. 4.3.2. MAINTENANCE OF BACTERIAL CULTURES Cultures of T. ferrooxidans were routinely maintained on medium 9K nutrient solution (Silverman and Lundgren, 1959) in which the ferrous iron energy source was replaced by the mineral concentrate. The schedule of transfer of cells to a new medium was found to be a parameter of paramount importance in assuring that the culture would perform optimally. A frequency of transfer within a 4—6 day period was set as a limit. A complete discussion of culture maintenance is found in Appendix 2. 4.3.3. BACTERIAL GROWTH Bacterial growth was estimated using organic carbon measurements. Since all carbon present in the sample is of bacterial origin, the measurements were made directly on the slurry samples. The apparatus used to make the determination MATERIALS AND METHODS / 67 was a laboratory total organic carbon analyzer (Astro model 1850). The method of analysis is based on ultra—violet promoted, chemical oxidation of the carbon present in the sample by sodium persulfate to carbon dioxide which is transported by an oxygen carrier gas stream to an infrared detector. The system is microprocessor controlled. Standard solutions of ethylene glycol of 10, 50 or 1000 ppm of carbon were used for calibration of the apparatus depending on the scale range used. For the analysis of samples of slurry a 1 to 5mL portion, depending on the carbon concentration, was diluted with IN H2SO4 to a lOmL volume and stored at 4°C until it was analyzed. The portion was then transferred to a syringe and diluted to 20mL. A series of tests was carried out to verify the accuracy of the TOC measurements using a standard provided by the U.S. Environmental Protection Agency (water pollution quality control check samples). The TOC results were within the 95% confidence interval reported; for the low range (0—50 ppm), ± 4 ppm, and for the high range (0—100 ppm), ± 17 ppm. Another test was also carried out to check for interference by the high metal and salt concentrations present in the slurry samples. No effect on the TOC measured was detected for the concentrations studied (up to 61.8g/L Cu, up to 5.3g/L Na and up to 0.50g/L Ag). 4.4. ANALYSES MATERIALS AND METHODS / 68 4.4.1. COPPER Copper in solution was determined by atomic absorption spectrophotometry (Atomic Absorption Spectrophotometer, Perkin Elmer 306). Portions of lmL, after the necessary dilutions with 0.0 IN sulfuric acid, were analysed following standard practices. 4.4.2. IRON Ferric iron in solution was determined by the sulfosalicylic acid colorimetric method. Total iron was measured by the same method following oxidation of the ferrous iron with sodium persulfate; ferrous iron was determined by difference. The method is described in Appendix 1. 4.4.3. ELEMENTAL SULFUR Elemental sulfur was determined gravimetrically. A 20mL sample of leach suspension was dried at 95°C and ground in a mortar. Approximately 2g of the solids were transferred to a Buchner funnel where the elemental sulfur was dissolved using carbon disulfide. After evaporation of the carbon disulfide the sulfur was dried and weighed. The amount of sulfur present in the sample was expressed as percentage on a dry basis. MATERIALS AND METHODS / 69 4.4.4. SCANNING ELECTRON MICROSCOPE TECHNIQUES A scanning electron microscope (S.E.M. autoscan number 26, Etec Corporation) operated at 20kV in the secondary electron emission mode was used to examine the specimens of concentrate. For dry concentrate the samples were mounted directly on aluminum stubs (14mm x 14mm) over a thin layer of graphite in ethanol (20% graphite) and coated with gold, using a Hummer gold spotter coater for 4 minutes at 170 millitorr vacuum and 9 milliamperes D.C. For slurry samples, a portion of the culture suspension was filtered, washed and dried at 65°C, and maintained in a dessicator until analyzed. Samples for bacterial, analyses were freeze—dried and stored frozen in a dessicator until analyzed. The electron beam was used to scan the surface of the sample in a square raster pattern and the image signal displayed on a cathode ray tube. The signal presented an image of the sample surface topography. These images were used to examine the type and extent of attack on the mineral surface during leaching and the presence of bacteria and secondary reaction products at the mineral surface. 4.4.5. X-RAY ANALYSES The primary beam of the S.E.M. generates X—rays of characteristic wavelength and energy which can be detected by a solid—state energy dispersive X — ray detector. X—ray spectra for samples were gathered and used to identify the elements present in the solid sample and give an indication of their relative MATERIALS AND METHODS / 70 amounts and distribution. To determine the spatial distribution (map) of the elements within a sample, X—ray peaks were selected and the count rate of the peak plotted over a scanned area as a series of dots, with the density of the dots corresponding to the relative amount of the element present. These maps can then be correlated with the corresponding images of the mineral surface obtained by SEM, and are useful to indicate the existence of preferential areas of attack on the mineral surface. CHAPTER 5. RESULTS AND DISCUSSION 5.1. STANDARD BIOLEACHING 5.1.1. WITHOUT BACTERIAL INOCULUM In order to investigate if copper leaching takes place in the absence of bacteria, under the initial conditions of the leaching medium, an experiment using —400 mesh Newmont concentrate was set up following the standard procedure, but without the addition of inoculum. Under these conditions copper extraction and bacterial carbon concentration were zero for the duration of the experiment (282 h) and iron extraction was 6.9%. This indicates that under initial conditions of low Eh potential (under 605 mV SHE), chemical leaching was negligible. 5.1.2. AS RECEIVED CONCENTRATE Experiments were carried out using "as received" Gibraltar concentrate at pulp densities of 107g/L in 1L tanks as described in Chapter 3. The number in brackets following the figure number indicates the source of data as table numbers from appendix 6. Two separate experiments, using 11 day and 6 day old inocula, resulted in copper extractions of 4 and 7% respectively. Figures 7(69) and 8(68) present the profiles of copper extraction and cell growth, together with pH and total iron concentration during these leaches. Figure 7 indicates that copper dissolution, during the first 100 hours following inoculation with an 11 day old culture, was related to a modest increase in cell concentration; but after 71 RESULTS AND DISCUSSION / 72 10n • - G --ET . a . - 0 ^ - - G ' " 100 150 200 250 T I M E ( h o u r s ) 300 350 r 3 0 0 N h-200 bo S u o h 100 400 Fig. 7 Standard leaching as received concentrate. (11 day old inoculum) RESULTS AND DISCUSSION / 73 H H Fig. 8 Standard leaching as received concentrate. (6 day old inoculum) RESULTS AND DISCUSSION / 74 this period copper dissolution and cell growth declined. The iron profde of Figure 7, indicates that iron dissolution continued in the absence of copper solubilization. This may be explained by considering that inoculum coming from a spent medium might be a mixed culture, the population balance of which shifted in favor of the iron oxidizing, but non-sulfide oxidizing bacteria. Discussion of such mixed cultures is presented in Appendix 2. Alternatively, iron release could also be the result of chemical leaching. If we compare the amount of iron leached in Figure 7 with the 6.9% iron leached in the absence of cell inoculum (see Section 5.1.1.) the two amounts are of the same order of magnitude. Viable cells were present in the culture as shown by microscopic examination and they could have been utilizing soluble iron for energy. The results presented in Figure 8 show that the use of a 6 day old inoculum resulted in a lag phase lasting over 100 hours. This period was followed by an increase in cell numbers associated with slow, but simultaneous, copper and iron solubilization. In both cases, for the 11 day and 6 day old inoculum experiments, very low copper extraction levels and the long lag phases obtained, can be attributed to the large particle sizes found in the "as-received" concentrate. Extraction levels improved considerably and the lag phase was shortened when the concentrate was ballmilled to 90% -400 mesh as will be shown in the following section. 5.1.3. -400 MESH CONCENTRATE Figure 9(89,93) presents leaching profiles for Newmont and Gibraltar concentrates leached under the same conditions. The profiles are very similar in terms of TOC and Fe. Newmont concentrate seems to have a greater amount of initially RESULTS AND DISCUSSION / 75 soluble Cu than Gibraltar concentrate otherwise the Cu leaching profiles are similar too. Both concentrates are similar in composition and mineralogy. 5.1.3.1. BACTERIAL GROWTH Leach experiments using —400 mesh Gibraltar concentrate were carried out in IL reactors at 107g/L pulp density. Copper extractions of 29 and 34% were measured in two separate experiments. Figures 10(70) and 11(84) show the profiles of copper extraction and cell growth for these experiments. Cell yields were 48 and 47mg TOC/g Cu leached respectively. The corresponding overall copper extraction rates were of 0.013 and 0.0125g Cu/L h. Figure 10 shows a typical profile for a conventional leach: the lag phase for microbial growth was under 20h; after this phase simultaneous release of copper and iron took place, associated with a decrease in the pH, and a net increase in the cell population. Figure 11 presents a duplicate of the experiment reported in Figure 10, extended a further 300 hours; this experiment was carried out to investigate the behavior of the culture at the decline of the growth phase. Figure 11 indicates that cell growth and metal release follow similar patterns for the duration of the active leaching phase, and stop after 650 hours. The difference between the copper and iron curves can be explained by taking into account the redox conditions in the medium; when a high redox potential was reached, at around 400h, iron precipitated from solution, forming jarosites on the particle surfaces thus passivating the residual chalcopyrite; eventually (at around 600h) this process stopped any further mineral dissolution. After 620h of leaching, there was a RESULTS AND DISCUSSION / 9 Newmont and Gibraltar leaching profile comparison RESULTS AND DISCUSSION / 77 short period of about 24h where an increase in cell concentration was measured in the medium and was not accompanied by a metal concentration increase. This indicates that the cells were probably oxidizing the F e + + ions in solution, a possibility examined in the following section. After this 24h period, the culture entered the stationary phase and no further growth or metal release occurred. 5.1.3.2. FERROUS/FERRIC IRON PROFILES To investigate the relationship between the oxidation state of the iron ions in solution and cell growth, two experiments were carried out. Iron, in its two oxidation states, copper, cell carbon and redox potential were measured during the leach. Fig. 12-13(93) show the results of an experiment lasting 150 h. Comparing the total iron curve of Fig 13 with the copper curve of Figure 12, it is evident that copper and iron were being leached simultaneously from the chalcopyrite (molar ratio 1:1.01). TOC concentration rose as the metals were released into solution. The cell yield for this experiment was 35.6mg TOC/g Cu leached. Of the total iron in solution, 75% was found as ferrous iron. This indicates that T. ferrooxidans was obtaining energy from the breakdown of the chalcopyrite crystal, by oxidizing the sulfide component of the mineral, and that it preferred this source to ferrous iron ions in solution. This mechanism predominated at Eh potentials below 420mV. Since ferric iron is not an effective oxidant at this redox potential, and since the ferric iron concentration is low, 1.8g F e + + + compared to the ferric iron concentrations normally used in abiotic leaching (16.6-55.8g F e + + +/L) (Miller and Portillo, 1979), then one could speculate that RESULTS AND DISCUSSION / 78 Fig. 10 Ballmiiled concentrate. Standard leaching profile RESULTS AND DISCUSSION / 79 Fig. 11 Ballmiiled concentrate. Standard leaching profile RESULTS AND DISCUSSION / 80 1 0 n r 3 0 0 T I M E (hours) Fig. 12 Cell growth and copper leaching in ballmiiled concentrate RESULTS AND DISCUSSION / 81 Fig. 13 Iron leaching in ballmilled concentrate RESULTS AND DISCUSSION / 82 in standard leaching there could be a direct attack by the bacteria on the chalcopyrite lattice. To find out what happened at higher redox potentials an experiment was set up and extended to 300h. The results of this experiment are shown in Figures 14-15(91). Fig. 14 shows that 50% of the copper was extracted at a rate of 0.048g Cu/L h from the chalcopyrite in 283h (a longer lag phase in this experiment was due to the use of an older inoculum). Cell growth slowed down from 0.76mg TOC/L h to 0.15mg TOC/L h during the last 48 hours of the copper extraction period. Examination of Fig. 15 indicates that during this same 48 hours, the redox potential rose from 420 to 535mV. Hydrolysis of ferric iron resulted in the formation of jarosites and iron hydroxides. Upon examination of the particles using reflectance microcopy (Sinclair, 1987) some of the particles appeared to be coated with unidentifiable matter that could be jarosite. Passivation of the chalcopyrite also occurred rendering it impervious to further attack, as was the case for the experiment presented in Fig 13. It appears that the cells turned to the use of the available iron in solution when the chalcopyrite became passivated, and the energy obtained from chalcopyrite oxidation was no longer enough to sustain the large cell population and maintain its high growth rate (0.76mg TOC/L h). The oxidation of the soluble ferrous iron present in the medium increased from 0.007 to 0.063g Fe+^/L h. This iron oxidation provided some energy but not enough to sustain the growth rate at 0.76mg TOC/L h which slowed down considerably at this point to 0.15mg TOC/L h. The ferric iron concentration in solution did not exceed 4.4g/L. RESULTS AND DISCUSSION / 83 Redox potential induced transition states in cell growth RESULTS AND DISCUSSION / 84 Fig. 15 Iron leaching under increased redox potential RESULTS AND DISCUSSION / 85 5.1.3.3. MINERAL SURFACE CHANGES To identify the changes taking place at the mineral surface during conventional leaching, electron micrographs were taken of samples from a Newmont concentrate leach. To determine the element distribution of the concentrate, x—ray line scans were also done. Figure 16 is a photograph of an energy dispersive x—ray analysis (EDXA) of a sample of —400 mesh Newmont concentrate before leaching. The three highest peaks from right to left correspond to ka Cu, ka Fe and sulfur; peaks for kj3 Cu and Fe are situated to the right of the corresponding ka peaks. Another element which was present in the sample in significant amounts was silicon and its peak is situated to the left of the sulfur peak. These results correspond to the composition found by chemical analysis (see Table 2), which shows the concentrate to contain copper, iron, sulfur and an insoluble fraction, which corresponds in the x ray analysis to silicon. Several samples of the concentrate were analyzed by EDXA and in all cases the same composition was found. Figure 17 presents a scanning electron microscopy (SEM) photograph of the same sample. Examination of the particle surface reveals that the surface for the most part was featureless and smooth, except at the edges of the particles which were roughened by milling. After 144 hours of leaching an x—ray line scan (Figure 18) showed that the relative proportions of copper, iron and sulfur were maintained in the concentrate undergoing leaching; consequently all three elements were being leached simultaneously. RESULTS AND DISCUSSION / 86 Fig. 16 EDXA analyses of concentrate before leaching RESULTS AND DISCUSSION / 87 Fig. 17 SEM photograph of concentrate before leaching RESULTS AND DISCUSSION / 88 :Fe: !SbSb i s b S b b b b b :Cm: 0 . 8 0 0 R a. n g e = 20. 4 6 0 k e V I n t e q r a. 1 0 20.140 18022 Fig. 18 X— r a y line scan of freeze-dried, leached concentrate Figure 19 is a SEM photograph of a chalcopyrite particle after 144 hours of leaching. The slurry sample was freeze—dried, a procedure which permits one to observe bacteria attached to the surfaces. This attachment is difficult to observe, RESULTS AND DISCUSSION / 89 RESULTS AND DISCUSSION / 90 when the particles are oven dried, because of heat damage. Figure 20 is the same sample (144 hours of leaching), but at a lower magnification. This sample shows the pitted surface resulting from leaching; leaching seems to have taken place all over the surface with shallow pits joined to each other. The net result was an erosion of the particle surface, which should cause the particle to shrink with time. In Figure 20 we can identify jarosite precipitate; at this time the deposition was limited to a small area. Extensive deposition of jarosites has been previously shown to take place. For example, after 200 hours of leaching, chalcopyrite particle surfaces from Newmont concentrate were almost completely covered by jarosite precipitates. (Blancarte—Zurita, 1983). These observations support the view of a direct attack mechanism for conventional leaching of chalcopyrite as has been described elsewhere. The x ray analyzes tend to confirm that in conventional leaching of chalcopyrite simultaneous release of copper, iron and sulfur is taking place. 5.1.4. "MONOSIZE" LE A C H E S In order to test the validity of the concept of a shrinking particle as the mechanism for the leaching of copper from sulfides a model based on this mechanism was tested using experimental data. Copper extraction rates were measured for 5 "monosize" fractions of Gibraltar concentrate in separate experiments carried out in 1L reactors. Particle shrinkage was monitored with leaching time for the duration of the experiments. Other parameters measured during the leach experiments with "monosize" fractions were pH, Eh, T O C , ferrous, ferric and total iron. Fig. 20 SEM photograph of freeze-dried leached concentrate (800 X) RESULTS AND DISCUSSION / 92 5.1.4.1. PARTICLE SIZE MEASUREMENT "Monosize" fractions, obtained by sedimentation comprised not just one particle size, but a narrow size distribution; consequently particle size measured by image analyses generated a number size distribution, which for modelling purposes was represented by the median size of the number distribution. This choice of central tendency for the distribution was based on the experimental procedure used to obtain the fractions and is explained as follows. Most fine particle systems formed by comminution of a bulk material have particle size distributions that obey the log—normal distribution and can be adequately represented by two values, the mean and the standard deviation (Stockham and Fochtman, 1978). Log—normal distributions produce a straight line when particle size vs cumulative percentage oversize are plotted on log—probability paper. If the particles belonging to such a distribution are subjected to sieving this procedure will remove all particles greater or smaller than a certain particle size, producing a distribution that becomes asymptotic towards the sizes that have been removed. (Allen, 1981). Figure 21 is a plot of particle size vs cumulative percentage oversize for —400 mesh concentrate and shows how the distribution deviates from linearity as the larger sizes are approached. When —400 mesh concentrate is separated into "monosize" fractions, using a sedimentation method, the distribution is subdivided, starting from the largest size end, into fractions whose lower limit is given by their sedimentation time (i.e. those particles that because of their small size did not sediment in a given time, remained in suspension and were not included in that particle size RESULTS AND DISCUSSION / 93 Fig. 21 Size distribution of -400 mesh concentrate RESULTS AND DISCUSSION / 94 Fig. 22 Size distribution of concentrate fractions RESULTS AND DISCUSSION / 95 fraction). The particle size distribution of these "monosize" fractions also deviates from linearity but towards the smaller sizes as shown in Figure 22. The presence of small particles with sedimentation times below that of the fraction being separated is due to entrapment, and for such a fraction, a particle size distribution will extend to the right and will result in extreme values on the small side of the distribution. If the mean size of a number distribution is used to represent it, its value will be greatly influenced by the extreme values of the distribution because the main characteristic of the mean is that it is affected by all values actually observed. A better choice for representing the central tendency of the distribution would then be the median size, because in monosize leaches, small particles, even when present in large numbers, represent only a small mass fraction of the total concentrate being leached while the median, which divides the frequency distribution into two equal areas and, is less affected by the extreme values, therefore provides a better estimate of the central tendency of the distribution. Stockam and Fochman (1978) also have found that for skewed frequency distributions, the central tendency is more adequately represented by the median than the mean. The size distribution, mean, and median size values are presented in Appendix 3, for the monosize fractions and for the slurry samples taken during leaching of each particle fraction. RESULTS AND DISCUSSION / 96 5.1.4.2. PARTICLE COMPOSITION When the concentrate was fractionated, the particles reported to a fraction according to their size. The spread of particle sizes found in a powder depends on the milling procedure. In our case the concentrate was wet ballmiiled. During milling, mineral particles were broken up mechanically and the relative strength of the crystal structure determined the extent of the size reduction for a given force applied. It follows that differences in composition of the mineral, hence different crystal strengths, would produce particles with a preferred size. Particle fracture is caused by crack growth, and cracks can be stopped by inclusions and grain boundaries, so that heterogeneous materials can be stronger than their components (Perry, 1973). If this were the case, the result would be the abundance of one particular phase in certain fractions. For example if the silicon fraction were to report to the largest size fraction, then the comparison of leaching rates among the fractions would become a function of the chemical composition of the fraction as well as the size of the particles. To investigate if the composition of the fractionated material changed with the particle size, samples of different sizes were examined mineralogically. Polished sections of concentrate samples embedded in epoxy resin were prepared and examined using a reflectance microscope. Examination of unfractionated Gibraltar concentrate showed angular particles of chalcopyrite with some chalcocite being present and some gangue—like material in very low proportion (= 1%). The pyrite content was estimated around 5%, mostly as individual particles; no locked particles were observed. RESULTS AND DISCUSSION / 97 When the fractionated concentrate was analyzed using the same technique, it was observed that the low proportion of non—-metallic material (quartz or feldspar) — 1%, was maintained, but that fraction 2 with particle size of 9.3 jim was enriched in pyrite. From the original estimate of 5% pyrite the content was estimated to be 15—20% pyrite. 5.1.4.3. LEACHING EXPERIMENTS Leach profiles are presented in Figures 23-32 for fractions of decreasing initial, nominal, median particle sizes from 52.1 ^nti to 2.8 /um. See Section 5.1.4.4 for further discussion on the true particle sizes of these fractions. Figures 23-24(110) are the leaching profiles for the fraction containing particles with initial size of 52.1 fim. Copper extraction and cell growth seemed to be simultaneous processes; ferrous iron in solution was oxidized after 200 hours of leaching, at the same time some iron precipitation occured when the Eh potential surpassed 500 mV. The same tendency was observed when Figures 25—26(112) (for 33.8 nm), 27-28(114) (for 15.9 /nm), 29(116)-30(117) (for 9.3 (im) and 31—32(106) (for 2.8 fira) were analysed. As was shown in Section 4.1.3, this was the same sort of behavior exhibited by —400 mesh concentrate. The difference found among the different size fractions was in the extent of copper extraction, which unexpectedly decreased when the particle size decreased. This decrease in copper extraction was accompanied by lower levels of microbial growth, and by an also unexpected decrease in copper extraction rate. This decrease in copper extraction rate, as particle size decreases, can be observed by RESULTS AND DISCUSSION / 98 Fig. 23 Cell growth and copper leaching of 52.1 Mm material RESULTS AND DISCUSSION / 99 15n T I M E (hours) Fig. 24 Eh and iron leaching profiles for 52.1 Mm material RESULTS AND DISCUSSION / 100 2 0 n Fraction 4 T I M E (hours ) Fig. 25 Cell growth and copper leaching of 33.8 ym material RESULTS AND DISCUSSION / 101 15-i T I M E (hours) Fig. 26 Eh and iron leaching profiles for 33.8 um material RESULTS AND DISCUSSION / 102 I6-1 Fraction 3 T I M E (hours ) Fig. 27 Cell growth and copper leaching of 15.9 n m material RESULTS AND DISCUSSION / 103 15-i T I M E (hours) Fig. 28 Eh and iron leaching profiles for 15.9 um material RESULTS AND DISCUSSION / 104 16n Fraction 2 T I M E (hours) Fig. 29 Cell growth and copper leaching of 9.3 /nm material RESULTS AND DISCUSSION / 105 700 600 o a 500 ^ 400 W L300 • F « 2 ~ A F « 3 ~ o To t a 1 F o • E h 100 1 D T 300 350 Fraction 2 150 200 250 T I M E ( h o u r s ) 400 450 500 Fig. 30 Eh and iron leaching profiles for 9.3 um material RESULTS AND DISCUSSION / 106 I6-1 Fraction 1 T I M E (hours) Fig. 31 Cell growth and copper leaching of 2.8 M m material RESULTS AND DISCUSSION / 107 Fig. 32 Eh and iron leaching profiles for 2.8 Mm material RESULTS AND DISCUSSION / 108 comparing the slopes of the copper extraction curves for the different size fractions of Figure 33. The reproducibility of the experiments was checked by performing a leach in duplicate. The results for both runs are plotted together in Figures 29 and 30. The data spread is reasonable for this type of experiment. The decrease in copper extraction rate as the particle size decreased could not be explained on the basis of previous knowledge which would suggest that extraction rate increased as particle size decreased (Blancarte-Zurita, 1983). Examination of particle size, as measured by image analysis, as a function of leaching time in these "monosized" leaches showed that for large particles (>5jtmi) the median size decreased rapidly before any significant amount of copper had been extracted. So the initial measurement of the particle size made on the dry concentrate, apparently did not represent the particle size when leaching started. When the particle sizes initially measured were replaced by the particle diameters as observed immediately before copper extraction began, the correlation between particle diameter and extraction rate changed. See Fig. 34 and discussion below. 5.1.4.4. SHEAR AND PARTICLE ATTRITION In order to achieve reasonable mixing of the solid suspension in the tanks, the criteria of complete off—bottom suspension was followed. Visual observation indicated that this condition was achieved at an impeller speed of around 300 rpm. The speed of the impeller could be an important parameter, since the higher the speed, the greater the proportion of the power introduced that goes RESULTS AND DISCUSSION / 109 20 -i T I M E (hours) Fig. 33 Cu leaching in monosize fractions (nominal size) RESULTS AND DISCUSSION / 110 20-1 T I M E (hours) Fig. 34 Cu leaching in monosize fractions (true size) RESULTS AND DISCUSSION / 111 into developing fluid velocity, and consequently fluid shear (Oldshue, 1983). High levels of fluid shear and turbulence may give rise to autogenous milling, due to collisions between particles and impact of the particles with the impeller, baffles, etc. For the purposes of this work it became important to differentiate particle shrinkage due to mechanical forces (attrition) from the size reduction caused by particle dissolution due to leaching. Figure 35 shows the changes found in the particle sizes for "as received" concentrate over a 23 hour period of stirring a concentrate suspension of 107g/L in iron-free 9K medium without inoculum, at 300 rpm in the Virtis fermentor. Figure 35 indicates that particles decreased in size with time causing the distribution to shift towards the left. The results of this experiment indicated that the larger sizes were the most affected, so we proceeded to study a monosize fraction to investigate this possibility. Figure 36 presents the particle size distribution changes as a function of agitation time for samples from a non—inoculated monosize leach at 10% pulp density. A 32 Mm monosize fraction was used. Particle size was measured after 22, 50, 77, 167 and 264 hours of agitation at 300 rpm. The results indicate that a significant reduction in the number of particles larger than 10 nm occurred. For particles larger than 20 (im this breakdown was more severe and resulted in the disappearance of all particles larger than 40 /ira and most of those larger than 20 /xm over a 100 hour period. 5.1.4.5. AUTOGENOUS MILLING IN MONOSIZE LEACHES Because of such size reduction due to autogenous milling, when comparing different monosize leaches in terms of copper extraction rate, the starting RESULTS AND DISCUSSION / 112 P A R T I C L E S I Z E (micrometer s ) Fig. 35 Particle fracture in <32 um material (6 um median size) RESULTS AND DISCUSSION / 113 100-1 80-60-40-20-Legend LZ3 0 h 50 h CD 77 h W 167 h CS 264 h 10 ~r~ 20 i 30 40 I 50 Size (micrometers) Fig. 36 Size distribution after mechanical fracture 32 um material RESULTS AND DISCUSSION / 114 diameter was taken not as the original size measured for a particular fraction in the dry powder, but as the size of the slurry sample at the onset of the rapid copper solubilization (generally around 100 hours). Copper extraction rates as a function of that diameter are shown in Figure 37. The changes in median particle diameters observed were as follows: fraction 1 decreased from a nominal size 2.8 to true 2.36 Mm, fraction 2 from 9.3 to 5.94 /xm, this fraction was somewhat resistant to autogenous milling due to its high pyrite content, pyrite being a much harder mineral than chalcopyrite; fraction 3 went from 15.9 to 2.23 n m ; fraction 4 from 33.8 to 2.67 Mm and; fraction 5 from 52.1 to 4.85 Mm. For the largest particle sizes significant reductions in particle size due to autogenous milling occurred. 5.2. MODELLING THE CONVENTIONAL LEACH 5.2.1. INITIAL CONSIDERATIONS 5.2.1.1. SOFTWARE A Fortran computer program was written to predict particle shrinking during leaching. The program version for conventional leaching is presented in Appendix 4. The input data for the program was: initial particle size, leaching time, pulp density and copper content in the concentrate. The calculation of the shrinking diameter that is the focus of the program, was RESULTS AND DISCUSSION / 115 ~! I 1 ' 1 1 2 , 3 4 5 M E D I A N D I A M E T E R (micrometer s ) Fig. 37 Cu extraction rate as function of size RESULTS AND DISCUSSION / 116 based on the experimental copper extraction rates, which were in turn functions of the particle diameter. An equation relating particle size to copper extraction rate was included in the program and was used to calculate the rate of copper extraction at the start of the leach and every time the particle shrank in size. For practical purposes the rate calculation was carried out at intervals of 50 hours of leaching, this interval was selected on the basis that rate changes for such an interval were small enough so as not to affect the continuity of the shrinking process. A logic diagram for the program is presented in Appendix 4. There are several possible ways of using experimental copper extraction data from monosize batch leaches to produce a rate function that relates copper extraction rate to particle size for use in the program. As in any batch leach experiment, the rate of reaction changes with time during bioleaching. The copper extraction rate increased from the induction phase, which lasted for about 20 hours and during which' time the copper extraction was very low, to a constant rate phase lasting for about 280 hours where the bulk of the copper was extracted (depending on the initial particle size, the type of process used to leach, and the concentrate). After the bulk of the copper extraction took place, it was followed by a phase of decreasing extraction rate. Any batch leach experiment can be characterized by the several copper extraction rates mentioned above, plus an overall extraction rate calculated over the total leach time. To select the rate to be used in the calculation of the shrinking diameter, it became necessary to consider how the model used this data to calculate the particle diameter. RESULTS AND DISCUSSION / 117 Ideally three shrinking rates could be incorporated in a model accounting for the three phases the culture goes through during a leaching experiment. These are induction, rapid growth and stationary phase. Difficulties would arise if such a model were to be used to predict shrinking of particles in leaches other than monosize leaches, because for each particle size the induction time and the length of the rapid copper extraction phase would be different, with the induction time depending on the conditions of cultivation and harvesting of the inoculum. In addition, copper solubilization in the induction phase is minimal; consequently particle size reduction due to copper solubilization, would be very small, and changes in particle size would be below the detection limits of the available methods for measuring particle size. Alternatively, use of the rate obtained from the linear portion of the copper extraction vs time curve, would better represent the process of particle shrinking due to copper solubilization and if changes in particle size occur due to copper solubilization, they could be easily measured. Since the leaching rate was obtained from the portion of the leach where rapid growth occurred, then the comparison of the shrinking predicted by the model to the particle size measured experimentally, should be made starting from the onset of the rapid growth phase, after the induction phase has ended. 5.2.1.2. SHRINKING RATES FOR CONVENTIONAL LEACHING Straight lines were fitted by least squares to the portion indicated by the arrows of the copper vs times curves of Figures 23, 25, 27, 29 and 31 (Cu = g/L and time(t) = hours). RESULTS AND DISCUSSION / 118 Fraction 1 Cu = 0.0325t + 1.8786 ..(38) Fraction 2 Cu = 0.0295t - 2.7180 ..(39) Fraction 3 Cu = 0.0372t - 4.8223 ..(40) Fraction 4 Cu = 0.0584t - 4.6231 ..(41) Fraction 5 Cu = 0.0803t - 6.4880 ..(42) A plot of copper extraction rate, for the different monosize leaches, as a function of median particle diameter measured after 100 h is shown in Fig. 37. The copper extraction rate data was fitted with equation 43, an exponential equation, where copper extraction rate is given in g/L h and the size dp is in p.m. The dashed line in Fig. 37 is the curve fit of the experimental data with equation 43. It has a correlation coefficient of 0.963. Cu extraction rate = 11.94 e - 0 - 3 2 (dP) ... (43) This equation shows that the copper extraction rate (r) (g/L h x 102) increases with decreasing particle size (p.). 5.2.2. APPLICATION OF THE MODEL TO MONOSIZE LEACHES 5.2.2.1. PARTICLE SHRINKAGE AND COPPER SOLUBILIZATION A comparison of particle size measured during active copper leaching and the particle shrinkage predicted by the model is depicted in Fig. 38. The scattered RESULTS AND DISCUSSION / 119 points are the experimental measurements, and the lines the predictions of the model using equation 44 to calculate particle shrinkage. - P - = [ 1 - - - ] 0 - 3 3 ... (44) dpo ^ ^o A reasonable fit was obtained for all fractions except the 5.94 fim fraction, where the predicted sizes decreased faster than experimental sizes indicating that particle integrity was preserved to a greater extent compared to the other fractions under the same conditions. This fraction was also the fraction more resistant to autogenous milling, and due to its high pyrite content was also the one with the lowest copper solubilization and lowest copper extraction rate. 5.3. CONTROLLED REDOX POTENTIAL LEACHING 5 . 3.1. PULP DENSITY Initially it was planned that the catalyzed leach experiments would be carried out at the same pulp density as the conventional leaches: 107 g/L. For experiments using monosize concentrate, using this low pulp density would have meant that lower amounts of fractionated concentrate would be needed; this would have been advantageous because the procedure used to obtain fractionated concentrate was very lengthy. But even more significantly, controlled-redox leaches at 107g/L, did not produce elemental sulfur and in all cases had erratic acid consumption with variable copper extraction. Although redox potential was low at the begining of the leach and was maintained, through the addition of thiosulfate in the chemical conditioning stage, these leaches did not produce elemental sulfur. RESULTS AND DISCUSSION / 120 Fig. 38 Model predictions of particle shrinkage for standard leaching RESULTS AND DISCUSSION / 121 So an extensive study was undertaken to search for conditions under which catalyzed leaching at such low pulp densities was feasible. A detailed discussion of the experiments carried out at a pulp density of 107g/L is presented in Appendix 7. The results indicated that there was a lower limit to pulp density below which catalyzed leaching did not take place. This limit has not been precisely determined, but it was somewhere in the range 107— 180g/L. 5.3.2. INITIAL CONDITIONS Chemical conditioning of the concentrate as described in section 2.5.0.4, was the key step in setting up a catalyzed leach. Such pre-treatment of the concentrate established the conditions of low Eh potential and excess cupric ions which seem to be necessary for acid consumption and elemental sulfur production during the oxidation of chalcopyrite concentrates. After the initial conditioning step was completed, inoculum was introduced. This was a culture of T. ferrooxidans grown for 6—8 days in the conventional mode on the concentrate of interest. Refer to Chapter 3 for culture maintenance. 5.3.3. B A C T E R I A L I N O C U L U M 5.3.3.1. REDOX-CONTROLLED LEACHING WITH AND WITHOUT INOCULUM The importance of the bacterial culture in the catalyzed leaching process was studied in a series of experiments in which different inocula were tested. First, it RESULTS AND DISCUSSION / 122 was necessary to test if the presence of bacteria was a requisite for chemically conditioned controlled-redox leaching. Fig. 39(102) shows the results of an attempt to leach the concentrate without inoculum. For this experiment the carbon concentration was zero for the duration of the experiment, copper concentration was also zero and only 4% of the iron present in the concentrate was extracted. The Eh rose slowly from 310mV to 360mV over 280h, and 4.7mL of 12N H 2 S O 4 were consumed to maintain the pH at 2.0. Typical leaching curves for a leach of -400 mesh Newmont concentrate inoculated with a Newmont grown bacterial culture are presented in Figs. 40—41(120). Elemental sulfur was produced, the maximum cell concentration was 101 mg/L, copper extraction was 100% and 410 ml of 12N acid were consumed to give a molar acid:copper ratio of 1:1.1 (the theoretical ratio being 1:1.6 based on CuFeS 2). The results of these two experiments (Figures 39,40,41) demonstrate that inoculation with a bacterial culture is necessary, in addition to low redox potential conditions and silver complexed with an excess of thiosulfate (Bruynesteyn et al, 1983), for controlled-redox copper leaching to occur. To inoculate an experiment an actively growing bacterial culture in the form of a cell suspension obtained from a 6—8 day old culture grown in —400 mesh concentrate was used. Such a cell suspension contained part of the leached solids from the culture medium from which the inoculum had been taken. These solid particles had a variety of sizes. When studying leaching of monosize fractions, the introduction of such solids with the inoculum was not desirable since it would RESULTS AND DISCUSSION / 123 Fig. 39 Redox-controlled leaching without bacterial inoculum RESULTS AND DISCUSSION / 124 75-, T I M E ( h o u r s ) Fig. 40 Cell growth when leaching under controlled redox conditions RESULTS AND DISCUSSION / 125 Fig. 41 Iron leaching profile under controlled redox conditions RESULTS AND DISCUSSION / 126 modify the particle size distribution of the monosize fraction being studied. For this reason the possibility of obtaining a particle free inoculum was contemplated. 5.3.3.2. PARTICLE FREE INOCULUM Microscopic observations of the medium had indicated that during the first days of leaching following inoculation, most cells were to be found at the mineral surface with freely suspended cells appearing only after 4 or 5 days. The 6—8 day old culture, normally used as inoculum, would then contain freely suspended cells, so it was decided to investigate the possibility of recovering these freely suspended cells, for use as inoculum, by centrifugation. Cell yields obtained from an inoculation medium by centrifugation were estimated by measuring TOC. The first attempt was made with 65mL of slurry from a 12 day old culture. After 12 days of cell growth, the culture medium when observed under the microscope, typically contains large amounts of cells non-associated to the mineral particles. The suspension was allowed to sediment for 30 minutes. The two fractions, sediment and supernatant, were treated separately: fraction 1, the supernatant, was centrifuged to recover free cells at 600 x g for 10 minutes (Dauve, 1983; Parry and Pedersen, 1983; Bailey and Ollis, 1986). The cells were washed by resuspending the precipitate in 0.0 IN H2SO4 and recovered by centrifugation. The washed cell pellet was finally suspended in 10 mL of 0.01N H2SO4 and was found to contain 2.8 mg of carbon. Fraction 2, the sediment from the initial settling of the medium was resuspended in 0.01N RESULTS AND DISCUSSION / 127 H 2 S O 4 and agitated for 30 minutes to remove any originally, freely suspended cells that may have been entrapped during sedimentation of the mineral particles, and if possible recover some cells from those attached to the mineral surface. After the 30 minutes of agitation the suspension was centrifuged at 50 x g to remove the mineral solids, and the supernatant was again centrifuged at 600 x g to recover the cells. No cell packet appeared. This indicated that most of those freely suspended cells, observed under the microscope in the original culture, were collected in fraction 1 and that cells attached to the mineral surface could not be removed by the method used. Figures 42—43(98) present the results of a leaching experiment inoculated with a particle—free cell suspension obtained by removing the solids from a 6 day old culture, by centrifugation for 5 minutes at 50 x g, and using the supernatant to obtain a cell packet by centrifugation at 600 x g for 10 minutes. The inoculum was 2 ml of a cell suspension containing 0.8 mg of carbon. For this experiment copper extraction was only 5% over 287 hours, acid consumption was zero and no elemental sulfur was produced. Since the initial carbon concentration in the medium, following inoculation, was below the detection limits for the carbon analysis method (the concentration was calculated to be 0.8mg/L), the experiment was repeated using a larger inoculum. Figs. 44—45(101) present data for an experiment using inoculum containing 1.4mg C. Copper extraction was 5.4% over 300 hours, 0.4mL of acid were consumed and no elemental sulfur was produced. The initial carbon concentration measured in the medium was 3mg/L, which was the same initial concentration RESULTS AND DISCUSSION / 128 Fig. 42 Cell packet inoculum effects on the iron profile RESULTS AND DISCUSSION / 129 Fig. 43 Cell packet inoculum effects on growth and Cu extraction RESULTS AND DISCUSSION / 130 used in the experiment shown in Figures 40—41 for which copper extraction was 100%. This indicated that although inoculum size was adequate, processing the inoculum to remove the mineral solids had somehow changed its characteristics. Alternatively, it could also mean that freely suspended cells were not equivalent to cells attached to the mineral surface, and that attached cells were responsible for starting catalyzed leaching. This second alternative was tested in the experiment reported in Figures 46—47(103) for which the inoculum consisted of the solids obtained by centrifuging a cell suspension for 10 minutes at 600g. The pellet resulting from this centrifugation was resuspended in 0.0 IN H 2 S O 4 and used as inoculum. For this experiment copper extraction was 9% over 140 hours, no acid was consumed and no elemental sulfur was produced. Upon review of the procedure used to obtain the particle free cells it was recalled that material belonging to the turbid, blue supernatant, obtained after centrifugation at 600 x g, had been discarded on every occasion. This fraction contained soluble materials: those first introduced when the leach was set up ie. 9K medium salts. It also contained products of leaching: metal ions, acid, etc. and would also contain any soluble extracellular material produced by the cells. Figures 48—49(109) show the results of an experiment inoculated with a particle free cell suspension obtained by centrifuging the original cell suspension for 5 minutes at 12 x g. In this case when the supernatant, containing cells, metal ions, acid, and soluble cell by—products, was used as inoculum a successful catalyzed leach resulted: elemental sulfur was produced; 1.74 Moles of acid were consumed; and 1.27 Moles of copper were extracted, to give a molar acid:copper RESULTS AND DISCUSSION / 13 Fig. 44 Inoculum cell carbon effect on iron leaching RESULTS AND DISCUSSION / 132 10n Inoculum cell carbon effect on growth and copper extraction RESULTS AND DISCUSSION / 133 Fig. 46 Iron leaching for inoculum drained of the soluble components RESULTS AND DISCUSSION / 134 Fig. 47 Copper leaching for inoculum drained of the soluble components RESULTS AND DISCUSSION / 135 ratio of 1:1.37. Of those materials present in this inoculum, 9K medium salts and metal ions were also present in the new culture medium. It follows then, that the fraction containing the soluble materials (supernatant, 600 x g centrifugation), also contained an extracellular material, factor X. This factor was needed for controlled-redox leaching to occur. Its nature has not been determined. This factor could be a surfactant produced by the cells, which is eliminated when the cells are washed and renders them inactive (Duncan, 1988). Some studies have been made on culture filtrates of T. ferrooxidans concerned with the type of compounds found after cell growth. It was reported by Schnaitman and Lundgren (1965) that F. ferrooxidans growing on a mineral salts medium with ferrous iron as an energy source, released carbon compounds into the medium, 20% of which were found to be strongly associated with ferric iron precipitates, but were not characterized. Agate et al (1968) characterized some of the carbon compounds unaccounted for by Schnaitman and Lundgren, as being an extracellular complex containing phospholipid and lipopolysaccharide. They examined both iron and sulfur grown cells and found that the composition of the extracellular complex changed depending on the substrate used. For iron grown cells the lipid portion of the complex contained 40% phospholipid and for sulfur grown cells 70%. They suggest that the complex may be playing a different role depending on the substrate. For iron grown cells they proposed that the complex was involved in the cell's electron transfer processes, since the quantities of complex found in the medium leveled off as the substrate became completely oxidized. Additionally they found that for iron grown cells the extracellular complex contained 3% F e + + and 0.5% F e + + + where none was present in the sulfur grown cells. The higher phospholipid content of the complex for sulfur grown cells and the RESULTS AND DISCUSSION / 136 Fig. 48 Iron leaching. Solid-free inoculum soluble fraction included RESULTS AND DISCUSSION / 137 3 5 - 1 T I M E ( h o u r s ) Cu leaching. Solid-free inoculum soluble fraction included RESULTS AND DISCUSSION / 138 absence of iron associated with it, as well as its lower protein content (6.4%) compared to the complex from iron grown cells (16.3%), led Agate et al to suggest that the extracellular complex played a role as a wetting agent for the sulfur grown cells. A similar role has been suggested for surface active phospholipids in T. thiooxidans by Schaeffer and Umbreit (1963). There have not been any similar studies made on spent growth medium contents for T. ferrooxidans growing on mineral concentrates, but Duncan et al (1964) working with washed cell suspensions of T. ferrooxidans showed the need of surfactants for the leaching of chalcopyrite to occur. The use of washed cells in our experiment may be related to the elimination of biological surfactants present normally on the spent medium (Duncan, 1988) If factor X corresponded to a surfactant like the extracellular complex proposed for sulfur grown cells, then the need for its presence upon transfer to new medium would indicate that the synthesis of such a complex was affected by exposure to the new medium. For a^ culture growing actively, as in continuous leaching, the critical stage would be the inoculum which would introduce factor X to the leach, it has been demonstrated that controlled-redox leaching can be maintained under continuous culture (Lawrence et al, 1984), then the cells must be able to produce factor X once a certain concentration is present in the medium. 5.3.4. -400 MESH CONCENTRATE RESULTS AND DISCUSSION / 139 5.3.4.1. BACTERIAL GROWTH Fig. 50(118) shows copper extraction and cell growth as a function of time for a — 400 mesh, redox-controlled leach at pulp density of 200g/L. For this leach copper extraction was 73%. The relationship between bacterial growth pattern and metal extraction pattern was different than for the conventional process. Compare these results with those shown in Figure 11 for conventional leaching. Lawrence et al (1982,1984 and 1985) have reported maximum copper extraction rates of 0.27-0.45g/L h. Our maximum copper extraction rates for -400 mesh concentrate were of 0.27-0.35mg/L h, thus being somewhat lower. This difference may be attributed to variations in between concentrate lots and ballmilling procedures. Lawrence et al, (1984) have suggested that there would be lower levels of bacterial populations present in the controlled-redox leach due to a reduction in the energy obtainable from the oxidation of sulfide when it stopped at the elemental sulfur state as opposed to sulfate. When we compare the cell yields obtained for conventional leaching with those obtained for controlled redox leaching, we do in fact find a lower cell yield in controlled redox leaching. See Table 6. Now lets speculate about the reasons for this. For conventional leaching a total of 17 electrons are available from the oxidation of one chalcopyrite molecule. For controlled redox leaching only 1-5 electrons are available for each molecule of chalcopyrite oxidized. 1 electron - if only iron oxidation is biologically mediated, and 4 electrons more if the conversion of S"2 to S° is also biologically mediated. For the case where only iron oxidation is biologically mediated, we would expect for controlled redox leaching of chalcopyrite that a reduction of 1/17 or 0.058 in energy should be reflected in a similar reduction in bacterial yields. RESULTS AND DISCUSSION / 140 50-i T I M E ( h o u r s ) Cell growth and Cu leaching with Eh control RESULTS AND DISCUSSION / 141 If the oxidation of sulfide is also cell mediated the reduction in energy would be 5/17 or 0.29, and we would expect a similar reduction in the cell yield. Table 6 supports the view that the reduction in bacterial yield in controlled-redox leaching, corresponds to the sole use of ferrous iron as energy source in this leach, and that the sulfide conversion to elemental sulfur is not bacterially mediated. In all three cases the ratio of cell yield is closer to the calculated ratio of cell yield reduction of 0.058. In terms of free energy decrease we find similarly that AF for oxidation of sulfide to sulfate is -215.14Kcal/g mol, AF for oxidation of SH" to elemental sulfur is -57.8Kcal/g mol, and that AF for F e + + oxidation is -17.8Kcal/g mol. The ratios for oxidation of chalcopyrite under standard conditions to oxidation under redox-controlled conditions are, if both iron and sulfide oxidation are biologically mediated, 0.25 and if only iron is biologically oxidized 0.04. Fig. 50 indicates that growth of T. ferrooxidans continued after copper extraction stopped in the controlled-redox bioleach process. A longer experiment was necessary to investigate what happened to the cells after the controlled-redox leaching of copper stopped. At this point in the leach, the conditions were redox potential around 500 mV, copper in solution, presumably as C U S O 4 , iron and jarosite precipitates. The remaining sulfide (if any) could be available as an energy source for continuing bacterial growth. Other possibilities for supplying bacterial energy requirements were the oxidation of ferrous iron or elemental sulfur. These two alternatives are considered in the following two sections. RESULTS AND DISCUSSION / 142 Fraction Leach Cell Yield mg TOC/g Ratio Cu extracted 5 C-R Conv 2.27 18.35 0.12 4 C-R Conv 0.87 12.30 0.07 2 C-R Conv 3.12 35.20 0.08 Table 6 Cell Yields in Conventional (Conv) and Controlled-Redox (C-R) Leaching 5.3.4.2. FERROUS/FERRIC PROFILES UNDER REDOX CONTROL According to the stoichiometry for controlled-redox leaching discussed in Chapter 2, the net sulfuric acid consumed in the leaching of copper is a function of the levels of oxidation of ferrous sulfate and hydrolysis of ferric sulfate. The net acid consumption due to these reactions can be estimated if the amounts of ferrous and ferric iron in solution are known. Fig. 51(118) presents the iron profiles for the leach presented in Fig. 50. Only 19% of the iron contained in the concentrate was found in solution, compared to 73% of the copper. Iron was being oxidized throughout the leach, and by the end of the leach seemed to occur mostly in the ferric form, only 0.3g/L of a total of 4.2g/L were in the ferrous state. Cell concentration and ferric iron concentration, expressed as % of the soluble iron, vs time are presented in Fig. 52(118). The shaded area represents the chemical oxidation of iron that occurs in the absence of bacteria. Bacterial growth seems to be related to ferrous iron oxidation because for ferric RESULTS AND DISCUSSION / 143 it has to come from oxidation of ferrous iron. r700 r 6 0 0 9 T I M E ( h o u r s ) Fig. 51 Iron leaching under Eh control RESULTS AND DISCUSSION / 144 Fig. 52 Soluble iron oxidized under Eh control RESULTS AND DISCUSSION / 145 Fig. 53 Sulfur and cell growth under Eh control RESULTS AND DISCUSSION / 146 5.3.4.3. CELL GROWTH AND ELEMENTAL SULFUR PRODUCTION Fig. 53(118) shows the relationship between copper extraction, Eh, consumption of H 2 S O 4 , S^ and cell growth. To maintain a constant pH required H 2 S O 4 addition from the start of the leach. This demand continued after the leaching of copper stopped. This later acid was used for oxidation of ferrous iron via equation 1. After an initial, unproductive period elemental sulfur production and copper extraction started increasing with time. After about 280 h copper extraction ceased. After 240 h, the amount of S° tended to decrease while the bacterial population was still increasing Two possibilities come to mind to explain the decrease of S®. Either S^ was oxidized by dissolved oxygen (Dutrizac and Macdonald, 1974) or bacterial oxidation thereof was occurring. Another experiment was carried out to investigate if bacterial oxidation of S^ was taking place. The results are presented in figures 54—55(119). The portion of Fig. 54 between 0—310 hours shows that cell concentration increases as does the concentration of ferric iron implying that ferrous iron is decreasing. Bacterial growth declines as all the available iron has been oxidized. Thus it seems that even when there is S° present, T. ferrooxidans preferentially oxidizes ferrous iron over elemental sulfur. Inhibition of cell mediated S° oxidation, could be caused by the presence of rree thiosulfate in the culture medium since an excess of thiosulfate was used in the conditioning step (for the formation of the silver thiosulfate complex during RESULTS AND DISCUSSION / 147 O r—< X (fi (fi r_,- • Q"~°-' ----A^—O--: A^>Q- - Q — Q - - O - Q — Q cr A — A -o © H 2 S O 4 A S s -• TOC 100 80 o a, o, o O •a a 60 « 40 ^ 0 100 200 300 400 500 600 700 800 T ime (hours) -200 160 -1 \ 120 5? a -80 -40 Fig. 54 Iron oxidation and cell growth under Eh control RESULTS AND DISCUSSION / 148 0.12 0.10 0.08 0.06 .0.04H 0.02 0.00 « — O — © -0 100 - Q - B - - • — B — B -i 1 r 200 300 40' 600 500 W 400 8 300 chalcopyrite ' s o l i d s IS d e n s i t y g/l x 10" D E h m v 400 500 T i m e ( h o u r s ) 600 700 800 Fig. 55 Sulfur profiles for Eh controlled leaching RESULTS AND DISCUSSION / 149 pre—treatment of the concentrate). Free thiosulfate has been reported to inhibit so oxidation (Landesman et al, 1966). Given the acid conditions of the medium it does not seem likely that any free thiosulfate is in existence in solution. Alternatively if thiosulfate decomposed it would liberate silver, which has also been reported to inhibit sulfur oxidation by interfering with glutathione dependent enzymes or with enzymes containing cysteine near their active sites (Kulpa and Mjoli, 1987). Fig. 55 is the time profile of S° expressed as g S°/g initially added chalcopyrite and g S°/g solids. The difference between these two curves is due to the increase in the pulp density caused by the formation of basic ferric sulfates and jarosites. Since S° concentration was expressed on a mass basis based on solids present in the culture medium, taking this increase in mass of solids in the medium into account makes the decrease of S° less significant, indicating that S° was in fact stable in this system. This also means that S° was not being oxidized by oxygen. Refer to the S° profile of Fig. 55. After 600h of leaching the cell growth and oxidation of chalcopyrite has ceased. Even though air was continually injected to the medium, the S° concentration remained constant. The dependence of cell growth on ferrous iron oxidation can be further analyzed by calculating an overall yield of bacteria based on ferrous iron oxidation. A very high yield relative to expected values for growth on F e + + could indicate that an energy»source other than ferrous iron was available for bacterial growth. The yield based on iron oxidation for the period between 0—310 hours of Fig. 54 was calculated to be 12 mg of cell—carbon/g F e + + oxidized. Kelly and RESULTS AND DISCUSSION / 150 Jones (1978,1983) reported 166 mg CI g F e + + oxidized and 632 mg C/g F e + + oxidized. Liu et al (1988) reported 2.23 mg TOC / g F e + + oxidized. This indicates that the amounts of cells present in our experiment could be produced solely by obtaining energy from ferrous iron oxidation. These results suggest that for the controlled-redox leach process cell growth depends on iron oxidation. Now let's consider some mechanisms through which this can bring about copper release. It could be that iron oxidation shifts the equilibrium of the reaction: CuFeS 2 + C u 2 + — - > 2CuS + F e 2 + ..(30) towards the right. By consuming ferrous iron, the bacterial cells would create a driving force which causes the dissolution of chalcopyrite. This equlibrium shift could be the major role played by the cells in the controlled-redox leach. The formation of elemental sulfur could be the result of the controlled redox, chemical oxidation and not necessarily a direct result of any biochemical process. Possibly bacterial sulfide oxidation and sulfur oxidation were inhibited. Additional support for this mechanism is discussed in section 5.6.2. for cell growth in monosize leaches. The presence of a lag phase in the leaches under redox controlled conditions, before cell growth occurs, may be due to an induction period necessary for the cells to gear up their metabolic reactions for iron oxidation. This could be likened to the pattern of diauxic growth wherein cells, in the presence of two substrates, rapidly metabolize the one which yields more energy and then undergo a lag phase before reassuming growth using the second substrate. In our case the first RESULTS AND DISCUSSION / 151 substrate is the sulfide being oxidized in the inoculum from the stock culture growing under standard conditions, and the second substrate is the soluble iron product of chalcopyrite oxidation. Following this reasoning the length of the lag phase would be controlled by the ferrous iron concentration in the medium, with induction of the necessary iron oxidizing enzyme systems in the bacteria starting only after a certain, amount of ferrous iron was present in the medium. Ferrous iron is released into the medium as a result of the chemical reactions between the Cu +/Cu +" 1" couple and the sulfide see equations 24 et seq for dissolution of chalcopyrite. Support for the proposed mechanism was also obtained from an analysis of the outer membrane preparation of cells grown under redox-controlled conditions in — 400 mesh Newmont concentrate by sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (McBride, 1988). The results are shown in Fig. 56. The protein bands show the presence of a protein unique to sulfur grown cells at 45-50K (Kulpa et al, 1986; Kulpa and Mjoli, 1987). This indicates that inhibition of sulfur oxidation may be at the metabolic regulation level, since synthesis of the protein is not inhibited. Additionally an outermembrane protein 92K, needed for iron oxidation (Kulpa and Mjoli, 1987) was found in cells grown under redox—controlled conditions. This protein is not found in cells grown under standard conditions on sulfides, which points to the fact that the iron oxidation mechanism is not normally functioning when the cells are oxidizing sulfides and needs to be induced before synthesis of the enzymes can proceed. The proteins patterns of the cells of the microorganism involved in this study, 1 2 RESULTS AND DISCUSSION / 152 3 4 5 66 K 45 K 27 K B ft FT 120 K 92 K s 18 K -+ * 14 K — ^ line 1 molecular weight size markers line 2 T. ferrooxidans ATCC strain membrane fraction line 3 T. ferrooxidans controlled-redox purified membrane fraction line 4 T. ferrooxidans controlled-redox membrane fraction line 5 molecular weight size markers Fig. 56 Gel electrophoresis of outer membrane proteins RESULTS AND DISCUSSION / 153 cells grown under redox-controlled conditions, shown in lines 3 and 4 of Figure 56 are very similar to those of the ATCC strain of T. ferrooxidans (line 2) implying that the organism involved is in fact T. ferrooxidans. 5.4. AERATION 5.4.1. SUSPENSIONS The leaching medium was typically an aerated mineral concentrate suspension containing 10—20% solids; it contained particles in the micrometer range and for this reason could be expected to behave in a non—Newtonian mode. The mechanics of mass transfer in non—Newtonian fluids are complicated by the presence of solid particles that interfere with the generation of gas—liquid interfacial area of the system, thus reducing the value of kj^a associated with oxygen transfer from air to liquid by as much as 28% for a 20% solids concentration compared to a medium with no solids (Liu, 1973). Changes in the "viscosity" of the medium with time may also take place during leaching because of the dissolution of the mineral particles and the formation of precipitates; such time dependent responses are factors to consider when determining mass—transfer coefficients in such a system. Other factors include: agitator and tank geometry, air bubble and cell dimensions, fluid characteristics and power input. An in depth study of the mechanics of mass transfer in leaching systems was beyond the scope of this work, but it was necessary nevertheless to insure an adequate oxygen supply to the medium, in order to avoid a limiting oxygen level RESULTS AND DISCUSSION / 154 which would affect metal extraction rates and cell growth. For this purpose metal extraction rates were determined at several aeration levels in 1L reactors, and a level was established as the minimum aeration rate above which non oxygen dependent metal extraction rates were obtained. Larger scale experiments behaved differently, 2L tanks have higher superficial gas velocities for the same gas rates (vvm). For this reason similar metal extraction rates to those achieved in 1L tanks were achieved at lower aeration rates in 2L tanks. This could be the result of increased surface gas reincorporation in the larger tanks due to changes in the tank geometry. To study the effects of aeration levels and establish a minimum aeration level, a series of leaches using —400 mesh concentrate was carried out. The results of catalyzed leaches in terms of copper extraction rates (overall and maximum) for various aeration levels expressed as volumes of air/volume of medium (wm) are presented in Table 7. The leaching profiles are presented in Fig. 57—58(94) for 0.8vvm, Fig. 59-60(121) for 1.5vvm, Fig. 61-62(109) for 2.99vvm and Fig. 63-64(97) for 5.98vvm. Based on the results presented in Table 7 an aeration level of 2.99vvm was selected as the minimum aeration rate for obtaining leaching rates that would be independent of the air flow through the reactors. Below 2.99vvm copper extraction rates were dependent on the air flow through the reactor. Examining the figures corresponding to aeration rates of 0.8vvm (Figs. 57—58) and 1.5vvm (Figs. 59—60) it is evident that the air supplied was not enough to support the oxidation reactions. This was confirmed by the low elemental sulfur levels RESULTS AND DISCUSSION / 155 Table 7 Aeration Effects REACTOR SIZE AERATION COPPER EXTRACTION RATE overall-max ELEMENTAL SULFUR (L) (vvm) (g/L h) % dry residue 1 1 1 1 0.80 1.50 2.99 5.98 0.06-0.12 0.08-0.13 0.09-0.32 0.09-0.31 0.12 0.00 2.98 2.06 2 2 0.68 0.68 0.12-0.35 0.19-0.33 1.22 3.24 measured in the medium at the end of the leach, indicating low conversion of sulfide-sulfur to elemental sulfur (see Table 7), and the low acid consumption. The molar acid:copper ratio was 0 for 0.8wm and 0.75 for 1.5vvm. For an aeration rate of 2.99vvm the profiles of Figs. 61—62 indicate that the oxygen supplied in this case was enough to carry out the oxidation reactions, and acid was consumed, to give a molar acid:copper ratio of 1.8. Figs. 63—64 present the case of excess air supply, the copper extraction rates did not increase above the rates obtained for 2.99wm aeration and ferrous iron was completely oxidized by the end of the leach. Elemental sulfur was produced and acid consumed to give a molar acid:copper ratio of 1.53. RESULTS AND DISCUSSION / 156 12-1 T I M E ( h o u r s ) Fig. Effect of 0.8 vvm on growth and Cu leached under Eh control RESULTS AND DISCUSSION / 157 Fig. 58 Effect of 0.8 vvm on iron leaching under Eh control RESULTS AND DISCUSSION / 158 Fig. 59 Effect of 1.5 vvm on growth and Cu extraction under Eh control RESULTS AND DISCUSSION / 159 Fig. 60 Effect of 1.5 vvm on iron leaching under Eh control RESULTS AND DISCUSSION / 160 35 - i T I M E ( h o u r s ) Fig. 61 Effect of 2.99 vvm on growth and Cu extraction under Eh control RESULTS AND DISCUSSION / 161 Fig. 62 Effect of 2.99 vvm on iron leaching under Eh control RESULTS AND DISCUSSION / 162 45-i 0 50 100 150 200 250 300 350 400 450 T I M E ( h o u r s ) Fig. 63 Effect of 5.98 vvm on growth and Cu extraction under Eh control RESULTS AND DISCUSSION / 163 T I M E ( h o u r s ) Fig. 64 Effect of 5.98 vvm on iron leaching under Eh control RESULTS AND DISCUSSION / 164 5.5. MINERAL SURFACE CHANGES 5.5.1. BEFORE LEACHING During the chemical activation stage of the concentrate, silver is found in a complexed form as Ag2S203 (Lawrence et al, 1984) resulting from a reaction between Ag2S04 with Na 2S 203- With an initial concentration of 400mg/L of silver (Lawrence, personal communication) less than O.lmg/L remained in solution. Therefore it was believed that the initial concentration of 0.58mg/L used in our tests would be in a complexed form and thus not in a free, soluble form. Free, soluble silver has been reported to inhibit the growth of T. ferrooxidans at concentrations higher than lmg/L. (Hoffman and Hendrix, 1976). It was not known whether this silver—thiosulfate complex would remain in solution or if it would associate with the mineral. To investigate if there were any changes taking place at the mineral surface during the chemical activation stage, samples of concentrate were taken from a catalyzed leach at different stages during the chemical activation procedure and analyzed by electron diffraction X—ray analysis (EDXA). Fig. " 65 presents the results of EDXA of chalcopyrite particles before treatment, Fig. 66 after addition of silver sulfate and sodium thiosulfate (first stage of the activation) and after addition of cupric sulfate and 9K medium salts (second stage). In all cases the a and /3 peaks for copper and iron and the sulfur peak are present, as well as the silicon peak. If the silver—thiosulfate complex precipitated on the mineral surface, a signal for silver would be present in the EDXA (Fig. 6 6 ) after the first stage of the activation. An arrow in Figure 6 6 indicates the expected position of a silver R E S U L T S A N D D I S C U S S I O N / 165 peak. Since no si lver could be detected the si lver—thiosulfate complex must have remained in solution. The E D X A analys is of stage 2 (concentrate at the end of the chemical act ivat ion stage) also indicated that no si lver was associated wi th the minera l surface. 5.5.2. A F T E R L E A C H I N G F ig . 67(5) presents the leaching profiles for Gibra l tar concentrate: copper concentration, cel l growth and acid consumption (67a) and i ron extract ion and E h potential (67b) measured dur ing leaching. Copper extraction for this leach was 100% and i ron extraction was 22%; chemical analysis of the solid residue at the end of the leach showed it contained 16% sulfur. Ac id was consumed to give an acid:copper rat io of 1.25. F i g . 68 presents the X — r a y analyses of the concentrate at the star t of the leach (68a) and after 93h of leaching (68b). F igure 68b was obtained by focusing the S E M into a pit. The analyses indicate that the solid mat r i x has changed its composition and that pi ts, where active leaching is tak ing place, have different composition compared to the rest of the minera l . This is due to sequential release of the elements of the mat r i x , and contrasts sharp ly wi th the simultaneous element release in conventional leaching. F ig . 69 presents the results of X — r a y analyses of the concentrate after 136 hours of leaching, wi th the corresponding element maps for iron and sulfur. Copper mapping was attempted but the concentration in this sample was below the detection l imits. The peaks corresponding to copper a and are considerably RESULTS AND DISCUSSION / 166 ,S Fig. 65 EDXA of concentrate before activation. RESULTS AND DISCUSSION / 167 1 A9 K Ca. Kot.Ca K^Ga Fig. 66 EDXA of concentrate during activation. Stage 1 and 2. Fig. 67 Leaching profiles EDXA specimen RESULTS AND DISCUSSION / 169 Vert« 290 counts lM Sp= 1 •l l l l i i l l l i i i l i i i i l i i i i i-.^ E 1 apsed • 18 s e c s i \Cu: ;:;;;;:Fe:;::::;::: V V y . : : : : : : : : : • :::::::::: 4- 0.008 Range" 10.230 keV 10.110 -f 12669 Vert= 350 cou <- 0.000 Range- 10.238 keV Fig. 68 EDXA of solids after 93 h of Eh controlled leaching 9.958 -f 24194 RESULTS AND DISCUSSION / 170 reduced compared to the initial sample. This reduction results from the leaching of this element from the mineral. The a and /3 peaks for iron are not reduced by the same proportion as the copper peaks. This can be explained by considering a small peak (see arrow) that indicates the presence of potassium. Iron at the mineral surface, in excess of the stoichiometric equivalent of copper, could then be in the form of potassium jarosites, resulting from the hydrolysis of ferric iron. The peak corresponding to sulfur indicates that a lot of sulfur was present at the mineral surface at the end of the leach. This was confirmed by comparing the intensity of the signal in the sulfur map, relative to the signal of the iron map. After extraction with carbon disulfide the EDXA shown in Fig. 70 was obtained indicating that most of the S° was easily removed by the solvent. Silver was detected during active controlled-redox leaching. Fig. 71 shows an EDXA for concentrate after 187 h of leaching under redox controlled conditions; copper extraction at this point was 24%. The presence of silver at the surface, indicates that it participates actively during leaching, even though none was detected during the activation stage. After 234 h of leaching, copper extraction was 84%. As shown in Fig.72, an EDXA of the leached surface also showed silver to be present at the mineral surface indicating that silver remained associated with the solids and that it reports to the tailings at the end of the leach. RESULTS AND DISCUSSION / RESULTS AND DISCUSSION / 172 ,". Fe / C a Fig. 70 EDXA of concentrate residue after solvent extraction R E S U L T S A N D DISCUSSION / 173 ::H;;:;::;;:S : c i ; :::::::::::::::::::::::-:A9i;:;::;;;;;::;;;:::;;;; :Fe:::ii::::Cu:::::::::::::::::::::::::::::::::::::::::::::::::::"""::::::::::::":::::::::::j 0.000 Range= 20.460 keV 20.140 I n t e g r a l 0 = 10633 Fig. 71 EDXA of concentrate 187 h leaching 5.5.3. SPATIAL DISTRIBUTION R E S U L T S A N D D I S C U S S I O N / 174 In order to veri fy the homogeneity of the concentrate in terms of spat ia l distribution of the elements, X — r a y maps were used to search for the existence of regions within a concentrate that were part icular ly r ich in si l icon or pyr i te. Fo r the two concentrates used in this study, maps of element distr ibution did not show any measurable differences in the proportions of copper, i ron and sul fur present i n the concentrate, (a lower proportion of copper w i th respect to iron and sulfur could indicate the presence of pyr i te regions). F i g . 73 is such a map of i ron, copper and su l fur for Gibra l ta r concentrate before leaching. X — r a y maps were also used to search for regions that remained unattacked dur ing leaching, as i t had been reported (Bruynesteyn et a l , 1986) that pyr i te remains unattacked dur ing cata lyzed leaching. A t tempts to find part icles of leached concentrate, that contained regions of pyr i te , were not successful because of the presence. of jarosites covering the surfaces of the residual minera l . A micrograph of a jarosite—covered, Newmont concentrate particle is shown in F i g . 74 together w i th the corresponding copper, iron and sulfur X — r a y maps after 234 h of leaching. The distr ibut ion of jarosites is homogeneous, as indicated by the iron map. I ron, i n the fo rm of jarosi te, can be estimated by considering the copper map and subtract ing the density of the s ignal of the copper map f rom that of the iron map. Sul fur also seems to be homogeneously distr ibuted. RESULTS AND DISCUSSION / 175 i Fe ::::::..:::::.::::::::::.::::: pe::::::::: ::::::::::r>. :::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::: eCu iT::' :;Fei::^;;::cu:::i;:::;:;;:::;i!::;::;::::i;::::i:i::;;: ::::::::::::::::::::::::::::::::::::::: 0.000 Range= 20.460 keV 20.220 I n t e g r a l 8 = 81522lj Fig. 72 EDXA of concentrate 234 h leaching RESULTS AND DISCUSSION / 176 Fig. 73 Concentrate element distribution by x—ray mapping RESULTS AND DISCUSSION 5.6. MONOSIZE LEACHES RESULTS AND DISCUSSION / 178 5.6 .1. A D D I T I V E C O N C E N T R A T I O N Since the controlled-redox process didn't always work, i.e., did not always produce S° and consume acid, it was postulated that the leach was somehow dependent on an activation of the concentrate particle surface and that perhaps the ratio of Ag, and/or thiosulfate, and/or Cu"*" ions to particle surface area had to be within a narrow range. Miller and Portillo (1979) found, for non—biological leaching of chalcopyrite, that the copper extraction rate increased, as initial silver concentration was increased from 17—36g/kg of concentrate. Beyond this level, additional silver did not enhance the rate of reaction but reduced it slightly in leaches with a pulp density of 3g/L. For controlled-redox bioleaching, Bruynesteyn et al, (1983) proposed a range of 0.1 — 4.0gAg/kg of concentrate in order to generate elemental sulfur from the leaching of chalcopyrite in leaches with pulp densities of 200g/L. To try to elucidate the relationship between silver concentration, copper extraction rate and mineral surface area, a series of catalyzed bioleaches at pulp densities of 180g/L was carried out using monosized fractions of Gibraltar concentrate with specific surface areas (s.s.a.) ranging from 454—4132 cm2/g as measured on dry concentrate. Figure 75 plots maximum copper extraction rates and S° concentration at the end of the leach as function of s.s.a. RESULTS AND DISCUSSION / 179 S . S . A . (cmVg concentrate) Fig. 75 Sulfur production and maximum extraction rate as function of ssa The maximum rate of copper extraction and the production of S° increased with increments of s.s.a. up to 1390cm2/g. Subsequent increases of s.s.a. resulted in decreasing extraction rates and decreasing amounts of S° all for a constant RESULTS AND DISCUSSION / 180 amount of catalyst (3.1 x 1 0 — 3 g Ag2S04/g concentrate). As shown in Fig. 75, leaching of the finest concentrate fraction (ssa = 4132cm2/g) did not produce any elemental sulfur. To determine if this was due to insufficient silver addition, a series of leaches was carried out in which the amount of silver added gave varying silver to mineral surface area ratios (Table 8). In this series the amount of copper sulfate and sodium thiosulfate was varied in a similar manner. The ratios of silvensurface area to be tested were selected based on knowledge that a leach of a monosize concentrate fraction having a nominal median particle diameter of 30.4jxm produced elemental sulfur and consumed acid. For that leach the amount of Ag2S04 added per gram of concentrate was 2.9mg which was equivalent to 3xl0~~ 3mg Ag2S04/cm 2 of particle surface area. The concentrations of thiosulfate and copper in terms of surface area were 3.8xl0"2mg Na2S 203/cm 2 and 4.4xl0~4g Cu +/cm 2. The results are shown in Table 8. In run 2 the ratio of Ag2S04 to particle surface area was kept constant at 3 x l 0 — 3mg/cm2. ]\T0 s u i f u r W a s produced. In run 5 the same ratio of silver addition to surface area was maintained but the amount of copper added was increased; there was still no elemental sulfur generated. Runs 1 and 4 had lower silver to surface area ratios than the base case (run 2) but no S^ was produced. Run 3 had a higher silver to surface area ratio than run 3 but still no so appeared. The results show that the increase in s.s.a. corresponding to a decrease of particle size from 30 fim (base case) to 2.5 nm could not be compensated for by increasing the concentration of silver sulfate or other reagents to maintain a constant ratio of reagent to surface area. Since the ratio of the preconditioning additives ( A g + , thiosulfate, C u + + ) to particle surface area seemed RESULTS AND DISCUSSION / 181 Table 8 Effects of Catalyst / Specific Surface Area Ratio Run mg mg S° moles mg mg CUSO4 number A g 2 S 0 4 A g 2 S 0 4 H 2 S 0 4 N a 2 S 2 0 3 /g cone. /cm2 /moles Cu /g cone. /g cone. 1 8.7 2x i r r 3 0 12.96 93.0 357 2 14.8 3xl0" 3 0 17.76 158.5 357 3 22.2 5xl0" 3 0 13.93 237.5 357 4 8.7 2xl( r 3 0 9.79 93.0 911 5 14.8 3xl0" 3 0 2.50 158.5 1822 to have no effect and since no Ag was ever observed after conditioning, but prior to leaching, on the surface by EDXA one concludes that the ratio of additive to surface, within the range studied, is not significant. It is strange that a 2.5 um particle cannot be leached by the controlled-redox process but an initially larger particle continues to be leached even when it gets smaller than 2.5 um. No explanation for this can be provided at this time. Grinding the concentrate to sizes below 2.5 (im seems to have resulted in a change in the leaching characteristics of the concentrate, with some copper being leached but no elemental sulfur produced. These results indicate that, at the pulp density of the experiment, there is a limiting s.s.a. above which the catalyzed reaction is not initiated. Once the reaction was initiated extractions of 70—95 + % of the copper from —400 mesh concentrate were not uncommon, and indicated that all particles would eventually dissolve. This finding contrasts with the effects of increased s.s.a. for non—biological leaching (Miller and Portillo, 1972) where RESULTS AND DISCUSSION / 182 increased rates of reaction were found when the s.s.a. increased, when the concentrate was ground, and the particle size went from 50% minus 20 um to 50% minus 0.5 Mm. This difference suggests that the mechanisms for these two processes are different and that ssa or particle diameter plays some important role in establishing the controlled-redox bioleach reaction. This finding also supports the postulate of section 5.3.4.3. regarding the mechanism of sulfur production, because if sulfur were a bacterial product as opposed to being produced as a result of a chemical reaction involving the particle surface then, its production should not be affected by the particle size of the concentrate. 5.6.2. CELL GROWTH IN MONOSIZE LEACHES Leach profiles are presented in Figures 76—85 for fractions of decreasing initial nominal, median particle size from 52.1 Mm to 2.8 Mm. Figures 76 and 77(107) are the leaching profiles for the fraction containing particles with initial size of 52.1 Mm. After 526 hours 36.7% of the copper was extracted, and acid was consumed to give a molar acid:copper ratio of 1:1.97. Copper extraction and cell growth were not simultaneous processes; cell growth was related to ferrous iron oxidation and only reached significant levels after the bulk of the copper extraction had taken place at around 425 hours. Only 15% of the iron in solution in the ferrous form was oxidized. This general pattern was followed by the other particle sizes as observed in Figs. 78—79(111) (for 33.8 Mm), 80-81(113) (for 15.9 Mm), 82-83(115) (for 9.3 Mm). The only exception was the 2.8 Mm fraction which did not produce elemental sulfur (Fig. 84—85(105)). This monosize fraction was discussed in section 5.6.1. Table 9 RESULTS AND DISCUSSION / 183 T I M E (hours) Growth and copper extraction from 52.1 /xm material RESULTS AND DISCUSSION / 184 Fig. 77 Iron profiles for fraction 52.1 RESULTS AND DISCUSSION / 185 36-) T I M E (hours) Fig. 78 Growth and copper extraction from 33.8 urn material RESULTS AND DISCUSSION / 186 Fig. 79 Iron profiles for fraction of 33.8 /im RESULTS AND DISCUSSION / 187 T I M E (hours ) Fig. 80 Growth and copper extraction from 15.9 nm material RESULTS AND DISCUSSION / 188 Fig. 81 Iron profiles for fraction of 15.9 urn R E S U L T S A N D DISCUSSION / 189 48-j 4 4 -' I " i " i " i 1 1 r"—i =i—=i—=h 1 1 1 ^ 0 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 T I M E (hours) Fig. 82 Growth and copper extraction from 9.3 (im material RESULTS AND DISCUSSION / 190 Fig. 83 Iron profiles for fraction of 9.3 um RESULTS AND DISCUSSION / 191 T I M E (hours) Fig. 84 Growth and copper extraction from 2.8 um material RESULTS AND DISCUSSION / 192 Fig. 85 Iron profiles for fraction of 2.8 Mm RESULTS AND DISCUSSION / 193 summarizes the results from these runs which produced elemental sulfur in terms of copper extraction, moles of iron oxidized and moles of cell—carbon produced. As was shown in section 5.3.4. for —400 mesh concentrate, copper leaching in controlled-redox leaches is dependent on ferrous iron oxidation by the bacterial cells. There is a lag in the growth of T. ferrooxidans until a certain concentration of ferrous iron is released into the medium and induction of the iron oxidation pathway starts. The length of the lag phase increased with increasing particle size, for 2.23 um it lasted 260 h, for 2.67 um 338 h, for 4.85 fim 425 h and for 5.94 um 450 h. This effect is probably due to the dependence of iron solubilization rates on particle size, since the substrate needed for induction is soluble ferrous iron. The time required for the iron concentration to reach a threshold that starts induction and growth of the cells will depend on the particle size. 5.6.3. SOFTWARE The Fortran program version used to predict particle shrinking during leaching is presented in Appendix 5. The input data for the program was: initial particle size, leaching time, pulp density and copper content in the concentrate. In this case the initial particle size was the particle diameter after lOOh of leaching, and the leach time started at lOOh. The calculation of the shrinking diameter was done in the same way as for the conventional leaching case, and was based on the experimental copper extraction rates. For the controlled-redox leach the rates were calculated from the portion RESULTS AND DISCUSSION / 194 Table 9 Controlled Redox Monosize Leaches FRACTION CELL-TOC COPPER EXTRACTION IRON OXIDIZED s° MOLES % MOLES 5 4 2 3 0.001 0.002 0.010 0.012 36.4 44.2 62.0 91.6 0.006 0.015 0.023 0.031 2.6 2.9 5.3 5.6 of the copper vs times curves indicated by the arrows. The rates in g Cu/L h were for 2.36 um 0.061, for 5.94 um 0.071, for 2.23 urn 0.155, for 2.67 (im 0.054 and for 4.85 iim 0.060. The experimental copper extraction data are presented in Fig. 86. The copper extraction rates estimated from the "linear" portion of the curves of Fig. 86 are only an approximation of the curve behaviour. The copper extraction rate exhibits a complex pattern. 5.6.4. APPLICATION OF THE MODEL TO CONTROLLED REDOX LEACHING Comparisons of particle sizes measured during active copper leaching and the particle shrinkage predicted by the model are depicted in Figs. 87—91 for the 5 fractions of concentrate having nominal sizes 52.1, 33.8, 15.9, 9.3 and 2.8 um. These nominal sizes correspond to the following sizes measured after 100 hours: 2.34, 2.90, 3.15, 2.61 and 2.33 um. All fractions are plotted together in Fig: 92. The scattered points are the experimental measurements, and the lines the RESULTS AND DISCUSSION / 195 • / O A X -/ / X / / / / Size ia micrometers / / / / X A F R A C T I O N 1 2.36 S / H - X - / X F R A C T I O N 2 5.94 o F R A C T I O N 3 2.23 t ' H F R A C T I O N A 2.67 •' 2 F R A C T I O N 5 4.85 300 400 500 T I M E (hours) 600 700 Fig. 86 Copper extraction rate data for Eh controlled leaching RESULTS AND DISCUSSION / 196 predictions of the model. As in the case of conventional monosize leaches a significant particle size reduction occurred during the first 100 hours due to autogenous milling, for this reason the comparison in particle shrinking was made after 100 hours. The model underpredicts the shrinking taking place at the end of the leach where the particle sizes were below 2.0 p.m and the particles seemed to be shrinking at an accelerated rate. This particle shrinkage indicates that particle size is a critical parameter to start the controlled-redox leach reaction, but once the reaction is underway, it will proceed even for particles below 2.0 (im. The model fails because we assumed simultaneous leaching of copper and iron. However the results show that significant amounts of iron are leached before any copper is leached thus our assumption is not correct. Nevertheless the particles get smaller as the particle size measurements show so there is some justification for using a shrinking particle model. The electron micrographs indicate that shallow pits form on the surface of particles during leaching so, in addition to to the shrinking particle concept, one should also consider adapting the propagating pore model of Hansford and Drossou (1987) to produce a combined shrinking particle, propagating pore model for bioleaching. 5.7. CONTROLLED REDOX LEACHING From the studies reported in sections 5.3 to 5.5 we obtained a clearer picture of the controlled-redox leach. It requires a minimum pulp density, ca. 200g/l, to produce elemental sulfur. Some factor produced by the inoculum cells when they RESULTS AND DISCUSSION / 197 u 4) O 5 s o u a w W 3H 2 H A 2.8 micrometers A i i 1 | 1 | 100 200 300 400 500 600 T I M E (hours) Fig. 87 Curve fit monosize fraction 1 leach RESULTS AND DISCUSSION / 198 100 200 300 400 T I M E ( h o u r s ) 500 600 Fig. 88 Curve fit monosize fraction 2 leach RESULTS AND DISCUSSION / 199 Fig. 89 Curve fit monosize fraction 3 leach RESULTS AND DISCUSSION / 200 33.8 micrometers 1°0 200 300 400 500 600 T I M E (hours ) Fig. 90 Curve fit monosize fraction 4 leach RESULTS AND DISCUSSION / 201 *" 5 4 2-i 52.] micrometers A A A 1 0 0 200 300 400 T I M E (hours) 500 600 Fig. 91 Curve fit monosize fraction 5 leach RESULTS AND DISCUSSION / 202" u Time = 285.23 h Number of particles = 1369 Mean diameter number percent 0.74 654 48 2.23 447 33 3.71 162 12 5.20 60 4 6.68 26 2 8.16 10 1 MEAN: 1.98 MEDIAN 2.31 STANDARD DEVIATION: 1.5868 Table 37. Monosize leach 46.8 Time= 382.73 h Number of particles = 1433 Mean diameter number percent 0.74 . 715 50 2.23 468 33 3.71 173 12 5.20 46 3 6.68 18 1 MEAN: 1.85 MEDIAN: 2.20 STANDARD DEVIATION: 1.4307 v. 1 2 6 6 Table 40. Monosize leach 2.58 jim Time= 162.57 h Number of particles = 1240 Mean diameter number percent 0.74 689 56 2.23 412 33 3.71 106 9 5.20 17 1 MEAN: 1.59 MEDIAN: 2.04 STANDARD DEVIATION: 1.1371 / 267 Table 41. Monosize leach 2.58 um Time= 261.31 h Number of particles = 1281 Mean diameter number percent 0.74 495 39 2.23 391 31 3.71 199 16 5.20 109 9 6.68 38 3 8.16 23 2 9.65 12 1 MEAN: 2.48 MEDIAN: 2.8 STANDARD DEVIATION: 2.0223 / 268 Table 42. Monosize leach 2.58 Mm Time= 357.31 h Number of particles = 1623 Mean diameter number percent 1.45 1008 62 4.34 504 31 7.23 98 6 10.12 15 1 MEAN: 2.70 MEDIAN: 3.78 STANDARD DEVIATION: 1.8846 4 / 269 Table 43. Monosize leach 8.63 um Time= 22.5 H Number of particles= 1216 Mean diameter number percent 1.45 435 36 4.34 272 22 7.23 236 19 10.12 157 13 13.01 72 6 15.90 30 2 MEAN: 5.50 MEDIAN: 5.91 STANDARD DEVIATION: 4.1294 / 270 Table 44. Monosize leach 8.63 um Time= 120.16 h Number of particles = 1181 Mean diameter number percent 0.74 496 42 2.23 322 27 3.71 151 13 5.20 71 6 6.68 47 4 8.16 39 3 9.65 27 2 11.13 15 1 12.62 11 1 MEAN: 2.76 MEDIAN: 2.61 STANDARD DEVIATION: 2.6549 / 271 Table 45. Monosize leach 8.63 um Time= 223.07 h Number of particles = 1120 Mean diameter number percent 0.74 416 37 2.23 299 27 3.71 154 14 5.20 94 8 6.68 61 5 8.16 53 5 9.65 20 2 11.13 13 1 MEAN: 2.94 MEDIAN: 2.89 STANDARD DEVIATION: 2.5878 / 272 Table 46. Monosize leach 8.63 Time= 459.56 h Number of particles = 1164 Mean diameter number percent 0.74 552 47 2.23 265 23 3.71 126 11 5.20 73 6 6.68 61 5 8.16 37 3 9.65 26 2 11.13 9 1 12.62 11 1 MEAN: 2.67 MEDIAN: 2.36 STANDARD DEVIATION: 2.6755 / 273 Table 47. Monosize leach 8.63 jum Time= 444.82 h Number of particles = 1253 Mean diameter number percent 0.74 556 44 2.23 371 30 3.71 182 15 5.20 80 6 6.68 35 3 8.16 16 1 MEAN: 2.20 MEDIAN: 2.476 STANDARD DEVIATION: 1.7921 / 274 Table 48. Monosize leach 8.63 um Time= 535.98 h Number of particles = 1353 Mean diameter number percent 0.27 854 63 0.82 399 29 1.37 64 5 1.91 14 1 MEAN: 0.50 MEDIAN: 0.689 STANDARD DEVIATION: 0.35531 / 275 Table 49. Monosize leach 14.3 um Time= 0 h Number of particles = 1074 Mean diameter number percent 1.45 603 56 4.34 211 20 7.23 82 8 10.12 72 7 13.01 53 5 15.90 23 2 18.79 16 1 21.68 10 1 MEAN: 4.50 MEDIAN: 4.03 STANDARD DEVIATION: 4.7390 / 276 Table 50. Monosize leach 14.3 um Time= 165.73 h Number of particles = 1337 Mean diameter number percent 0.74 402 30 2.23 392 29 3.71 208 16 5.20 142 11 6.68 70 5 8.16 43 3 9.65 31 2 11.13 17 1 12.62 13 1 14.10 11 1 MEAN: 3.31 MEDIAN: 3.15 STANDARD DEVIATION: 2.9836 / 277 Table 51. Monosize leach 14.3 p.m Time= 262.48 h Number of particles = 1198 Mean diameter number percent 0.74 382 32 2.23 323 27 3.71 196 16 5.20 103 9 6.68 73 6 8.16 42 4 9.65 35 3 11.13 14 1 12.62 14 1 14.10 9 1 MEAN: 3.30 MEDIAN: 3.21 STANDARD DEVIATION: 2.9366 / 278 Table 52. Monosize leach 14.3 um Time= 333.73 Number of particles = 1268 Mean diameter number percent 0.74 562 44 2.23 361 28 3.71 169 13 5.20 74 6 6.68 37 3 8.16 18 1 9.65 22 2 11.13 13 1 MEAN: 2.46 MEDIAN: 2.44 STANDARD DEVIATION: 2.3702 / 279 Table 53. Monosize leach 14.3 fim Time= 435.48 h Number of particles = 1256 Mean diameter number percent 0.74 670 53 2.23 403 32 3.71 82 7 5.20 34 3 6.68 29 2 8.16 10 1 9.65 11 1 11.13 7 1 MEAN: 1.91 MEDIAN: 2.14 STANDARD DEVIATION: 1.9016 / 280 Table 54. Monosize leach 14.3 i^m Time= 479.31 h Number of particles = 2850 Mean diameter number percent 0.74 1655 58 2.23 906 32 3.71 185 6 5.20 75 3 6.68 24 1 8.16 15 1 MEAN: 1.66 MEDIAN: 2.05 STANDARD DEVIATION: 1.4152 / 281 Table 55. Monosize leach 14.3 |im Time= 575.30 Number of particles = 1951 Mean diameter number percent 0.74 1202 62 2.23 626 32 3.71 101 5 5.20 19 1 MEAN: 1.45 MEDIAN: 1.94 STANDARD DEVIATION: 1.0360 / 282 Table 56. Monosize leach 30.4 / im Time= 96.92 h Number of particles = 452 Mean diameter number percent 0.74 187 41 2.23 98 22 3.71 66 15 5.20 40 9 6.68 14 3 8.16 10 2 9.65 9 2 11.13 8 2 12.62 7 2 14.10 3 1 15.59 4 1 17.07 5 1 MEAN: 3.18 MEDIAN: 2.90 STANDARD DEVIATION: 3.4104 / 283 Table 57. Monosize leach 30.4 um Time= 195.91 h Number of particles = 1398 Mean diameter number percent 0.74 541 39 2.23 424 30 3.71 187 13 5.20 98 7 6.68 59 4 8.16 34 2 9.65 14 1 11.13 19 1 12.62 8 1 MEAN: 2.34 MEDIAN: 2.38 STANDARD DEVIATION: 2.3159 / 284 Table 58. Monosize leach 30.4 um Time= 292.32 h Number of particles = 1174 Mean diameter number percent 0.74 417 36 2.23 321 27 3.71 180 15 5.20 113 10 6.68 61 5 8.16 29 2 9.65 20 2 11.13 14 1 MEAN: 2.93 MEDIAN: 2.88 STANDARD DEVIATION: 2.6003 / 285 Table 59. Monosize leach 30.4 um Time= 384.57 h Number of particles = 1071 Mean diameter number percent 0.74 448 42 2.23 298 28 3.71 153 14 5.20 69 6 6.68 53 5 8.16 30 3 9.65 7 1 11.13 14 1 12.62 6 1 MEAN: 2.67 MEDIAN: 2.7 STANDARD DEVIATION: 2.4488 / 286 Table 60. Monosize leach 30.4 nm Time= 431.93 h Number of particles = 1311 Mean diameter number percent 0.74 535 41 2.23 456 35 3.71 212 16 5.20 62 5 6.68 33 3 8.16 14 1 MEAN: 2.27 MEDIAN: 2.65 STANDARD DEVIATION: 1.7919 / 287 Table 61. Monosize leach 46.8 Mm Time= 0 h Number of particles= 1036 Mean diameter number percent 2.93 270 26 8.79 239 23 14.65 143 14 20.51 103 10 26.37 79 8 32.23 60 6 38.09 69 7 43.95 16 2 49.80 - 36 3 55.66 13 1 61.52 12 1 MEAN: 17.49 MEDIAN: 15.48 STANDARD DEVIATION: 15.0303 / 288 Table 62. Monosize leach 46.8 urn Time = 98.58 h Number of particles= 1178 Mean diameter number percent 0.74 536 46 2.23 320 27 3.71 121 10 5.20 68 6 6.68 46 4 8.16 25 2 9.65 17 1 11.13 9 1 12.62 6 1 MEAN: 2.73 MEDIAN: 2.34 STANDARD DEVIATION: 3.2553 / 289 Table 63. Monosize leach 46.8 urn Time= 190.24 h Number of particles = 931 Mean diameter number percent 0.74 502 54 2.23 231 25 3.71 104 11 5.20 39 4 6.68 15 2 8.16 15 2 11.13 9 1 14.10 5 1 MEAN: 2.11 MEDIAN: 2.12 STANDARD DEVIATION: 2.3064 / 290 Table 64. Monosize leach 46.8 /um Time= 287.90 h Number of particles = 1077 Mean diameter number percent 0.74 510 47 2.23 293 27 3.71 127 12 5.20 65 6 6.68 37 3 8.16 15 1 9.65 15 1 MEAN: 2.32 MEDIAN: 2.23 STANDARD DEVIATION: 2.2480 / 291 Table 65. Monosize leach 46.8 um Time= 382.31 h Number of particles = 1144 Mean diameter number percent 0.74 557 49 2.23 295 26 3.71 134 12 5.20 59 5 6.68 36 3 8.16 20 2 9.65 14 1 11.13 15 1 MEAN: 2.37 MEDIAN: 2.23 STANDARD DEVIATION: 2.4280 / 292 Table 66. Monosize leach 46.8 um Time= 478.81 Number of particles = 1188 Mean diameter number percent 0.74 821 69 2.23 203 17 3.71 78 7 5.20 34 3 6.68 25 2 8.16 9 1 9.65 9 1 MEAN: 1.64 MEDIAN: 1.82 STANDARD DEVIATION: 1.8227 A P P E N D I X 4 P R O G R A M S T A N D A R D L E A C H I N G START ENTER DATA dp SEARCH FOR SIZE CLASS CALCULATE RATE FOR dp CALCULATE SHRINKING NEW dp -NO INCREASE TIME TEST dp IS NEW dp LESS THAN LOWER LIMIT OF CLASS YES OISPLAY VALUE OF TIME ANO OP STORE TIME •NO TEST TIME LARGER THAN TMAX INITIALIZE TIME YES CALCULATE AVERAGE PARTICLE SIZE CALCULATE ELAPSED TIME SINCE LEACH BEGAN STOP Fig. 96 Program logical diagram for shrinking diameters 2 9 3 / 294 C THIS PROGRAM CALCULATES SHRINKING DIAMETERS C FOR CONVENTIONAL LEACHING OF MONOSIZE C PARTICLES AS A FUNCTION OF LEACHING TIME C IMPLICIT REAL*4(A-H,0-Z) REAL*4EXP,PA,PB,PC,PD,PE 33 READ(5,20) DPE,TMAX,DELT,F,C0,DP0, *DP1,DP2, DP3,DP4 20 FORMAT(9(D13.5,/),D13.5) WRITE(6,10) 10 F0RMAT(1H1, 'PARTICLE CHANGES SIZE CLASS',/) WRITE(6,11) 11 FORMAT(2X,'TIME',6X,'DIAMETER',3X,//) C C SEARCHING FOR THE SIZE CLASS CORRESPONDING C TO THE INITIAL DIAMETER AND INITIALIZING C THE COUNTERS FOR TIME AND PARTICLE DIAMETER C TTE0 = 0.0 TTE1 = 0.0 TTE2 = 0.0 TTE3 = 0.0 TTE4 = 0.0 PA=0.0 PB = 0.0 PC = 0.0 PD = 0.0 PE = 0.0 T = 0.D0 T=T+DELT PA=DPE IF(DPE.GE.DPO.OR.DPE.GT.DP1)GO TO 50 PB = DPE IF(DPE.LE.DPl.AND.DPE.GT.DP2)GO TO 150 PC = DPE IF(DPE.LE.DP2.AND.DPE.GT.DP3)GO TO 250 PD = DPE IF(DPE.LE.DP3.AND.DPE.GT.DP4)GO TO 275 PE=DPE IF(DPE.LE.DP4)GO TO 300 C C CALCULATION OF PARTICLE SHRINKING RATE C 50 R = 0.01*(11.944*EXP(-0.32123*DPE)) DP = ((DPE**3.D0)-(T*R*(DPE**3.D0))/(F*C0))**0.3333D0 C C THE SHRINKING RATE IS EVALUATED AS A FUNCTION OF C THE PARTICLE DIAMETER C PA = DP IF(DP.LE.DPl)GO TO 51 T=T+DELT TTE0=T WRITE(7,77)TTE0,PA 77 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0.GT.TMAX-DELT)GO TO 1000 GO TO 50 51 WRITE(6,100)T,DP 100 FORMAT(2X,F7.2,2X,F7.2) T = 0.0 T=T+DELT PB=PA 150 R=0.01*(11.944*EXP(-0.32123*PA)) DP=((PA**3.D0)-(T*R*(PA**3.D0))/(F*C0))**0.3333D0 PB=DP IF(DP.LE.0.01)GO TO 500 IF(DP.LE.DP2) GO TO 151 T=T+DELT TTE1=T TT=TTE0+TTE1 WRITE(7,78)TT,PB 78 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0+TTE l.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 150 151 WRITE(6,101)T,DP 101 FORMAT(2X,F7.2,2X,F7.2) T=0.0 T=T+DELT PC=PB 250 R=0.01*(11.944*EXP(-0.32123*PB)) DP=((PB**3.D0)-(T*R*(PB**3.DO))/(F*CO))**O.3333D0 PC=DP IF(DP.LE.0.01)GO TO 500 IF(DP.LE.DP3)GO TO 251 T=T+DELT TTE2=T TT=TTE0+TTE 1+TTE2 WRITE(7,79)TT,PC 79 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0+TTE1+TTE2.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 250 251 WRITE(6,102)T,DP 102 FORMAT(2X,F7.2,2X,F7.2) T=0.0 T=T+DELT PD = PC / 296 275 R = 0.01*(11.944*EXP(-0.32123*PQ) DP = ((PC**3.D0)-(T*R*(PC**3.D0))/(F*C0))**0.3333D0 PD = DP IF(DP.LE.0.01)GO TO 500 IF(DP.LE.DP4)GO TO 276 T = T + DELT TTE3=T TT = TTEO+TTE1 + TTE2+TTE3 WRITE(7,80)TT,PD 80 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0 + TTE 1+TTE2+TTE3.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 275 276 WRITE(6,103)T,DP 103 FORMAT(2X,F7.2,2X,F7.2) T=0.0 T=T+DELT PE = PD 300 R=0.01*(11.944*(EXP(-0.32123*PD))) DP = ((PD**3.D0)-(T*R*(PD**3.D0))/(F*C0))**0.3333D0 PE = DP IF(DP.LE.0.01)GO TO 500 T=T+DELT TTE4=T TT=TTEO + TTE 1 + TTE 2+TTE3+TTE 4 WRITE(7,81)TT,PE 81 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0+TTE 1+TTE2+TTE3 + TTE4.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 300 1000 CONTINUE C C RECORDING THE PARTICLE SIZE AT THE END OF THE C DELTA TIME INTERVAL C 500 WRITE(6,108) 108 F0RMAT(1H2,2X,' TIME', *2X,' DIAMETER',//) WRITE(6,109) T,DP 109 F0RMAT(2X,F7.2,6X,F13.5) C C C CALCULATION OF ELAPSED TIME SINCE LEACH BEGAN C TTE = TTEO + TTE 1 + TTE 2 + TTE 3 + TTE4 WRITE(6,110) 110 F0RMAT(1H2,2X,'LEACHING TIME',//) WRITE(6,111)TTE I l l FORMAT(3X,F7.2) IF (DP4.LE.0.01) GO TO 34 GO TO 33 34 STOP END APPENDIX 5 FORTRAN PROGRAM CONTROLLED REDOX LEACHING C THIS PROGRAM CALCULATES SHRINKING DIAMETERS C FOR CONTROLLED REDOX LEACHING OF MONOSIZE C PARTICLES AS A FUNCTION OF LEACHING TIME C IMPLICIT REAL*4(A-H,0-Z) REAL*4EXP,PA)PB,PC,PD,PE 33 READ(5,20) DPE,TMAX,DELT,F,CO,DP0, *DPl,DP2, DP3.DP4 20 FORMAT(9(D13.5,/),D13.5) WRITE(6,10) 10 FORMATQHl, 'PARTICLE CHANGES SIZE CLASS',/) WRITE(6,11) 11 FORMAT(2X,'TIME',6X)'DIAMETER',3X,//) C C C THE SHRINKING RATE IS EVALUATED AS A C FUNCTION OF THE PARTICLE DIAMETER C TTEO = 0.0 TTE1 = 0.0 TTE2 = 0.0 TTE3 = 0.0 TTE4 = 0.0 PA=0.0 PB = 0.0 PC = 0.0 PD = 0.0 PE = 0.0 T=0.D0 T=T+DELT PA = DPE IF(DPE.GE.DP0.OR.DPE.GT.DPl)GO TO 50 PB=DPE IF(DPE.LE.DPl.AND.DPE.GT.DP2)GO TO 150 PC=DPE IF(DPE.LE.DP2.AND.DPE.GT.DP3)GO TO 250 PD = DPE IF(DPE.LE.DP3.AND.DPE.GT.DP4)GO TO 275 PE = DPE IF(DPE.LE.DP4)GO TO 300 C C CALCULATION OF PARTICLE SHRINKAGE C 50 R=0.022 DP = ((DPE**3.D0)-(T*R*(DPE**3.D0))/(F*C0))**0.3333D0 298 PA = DP IF(DP.LE.DP1)G0 TO 51 T=T+DELT TTE0=T WRITE(7J7)TTE0,PA 77 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0.GT.TMAX-DELT)GO TO 1000 GO TO 50 51 WRITE(6,100)T,DP 100 FORMAT(2X,F7.2,2X,F7.2) T=0.0 T=T+DELT PB = PA 150 R=0.0519 DP = ((PA**3.D0)-(T*R*(PA**3.D0))/(F*C0))**O.3333DO PB=DP IF(DP.LE.0.01)GO TO 500 IF(DP.LE.DP2) GO TO 151 T=T+DELT T T E l = T TT=TTEO+TTE 1 WRITE(7,78)TT,PB 78 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0+TTEl.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 150 151 WRITE(6,101)T,DP 101 FORMAT(2X,F7.2,2X,F7.2) T=0.0 T=T+DELT PC = PB 250 R=0.06955 DP = ((PB**3.D0)-(T*R*(PB**3.DO))/(F*C0))**O.3333D0 PC=DP IF(DP.LE.0.01)GO TO 500 IF(DP.LE.DP3)GO TO 251 T=T+DELT TTE2=T TT=TTEO+TTE1+TTE2 WRITE(7,79)TT,PC 79 FORMAT(2X,F7.2,2X,F7.2) IF(TTE0 + TTE 1+TTE2.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 250 251 WRITE(6,102)T,DP 102 FORMAT(2X,F7.2,2X,F7.2) T = 0.0 T=T+DELT / 300 PD = PC 275 R=0.0326 DP = ((PC**3.D0)-(T*R*(PC**3.D0))/(F*CO))**0.3333DO PD = DP IF(DP.LE.0.01)GO TO 500 IF(DP.LE.DP4)GO TO 276 T=T+DELT TTE3=T TT=TTEO+TTE 1 + TTE2 + TTE3 WRITE(7,80)TT,PD 80 FORMAT(2X,F7.2,2X,F7.2) IF(TTEO+TTE1+TTE2+TTE3.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 275 276 WRITE(6,103)T,DP 103 FORMAT(2X,F7.2,2X,F7.2) T=0.0 T=T+DELT PE = PD 300 R=0.0439 DP = ((PD**3.D0)-(T*R*(PD**3.D0))/(F*C0))**0.3333D0 PE = DP IF(DP.LE.0.01)GO TO 500 T=T+DELT TTE4=T TT=TTEO+TTE 1+TTE2 + TTE3 + TTE4 WRITE(7,81)TT,PE 81 FORMAT(2X,F7.2,2X,F7.2) IF(TTEO+TTE1+TTE2 + TTE3 + TTE4.GT.TMAX-DELT)GO TO 1000 IF(T.GT.TMAX-DELT)GO TO 1000 GO TO 300 1000 CONTINUE 500 WRITE(6,108) 108 F0RMAT(1H2,2X,' TIME', *2X,' DIAMETER',//) WRITE(6,109) T,DP 109 FORMAT(2X,F7.2,6X,F13.5) C C C CALCULATION OF ELAPSED TIME SINCE LEACH BEGAN C TTE=TTEO + TTE 1 + TTE 2 + TTE3 + TTE 4 WRITE(6,110) 110 F0RMAT(1H2,2X,'LEACHING TIME',//) WRITE(6,111)TTE 111 F0RMAT(3X,F7.2) IF (DP4.LE.0.01) GO TO 34 GO TO 33 34 S T O P E N D APPENDIX 6 GENERAL LEACHING DATA Table 68 Experiment 1 as received, st, 6 day inoculum TIME pH COPPER T-IRON TOC TEMP SIZE Hours g/L g/L ppm °C U m 0.00 2.00 0.0 0.0 0 34 3.14 24.75 1.98 1.4 1.2 0 34 2.97 72.08 2.20 1.7 1.4 0 38 3.06 115.91 2.40 1.7 1.4 0 35 3.35 167.57 2.12 1.8 1.6 31 34 2.52 217.57 2.35 1.8 1.6 26 32 2.47 260.98 2.14 1.9 1.6 57 35 2.59 Table 69 Experiment 2 as received, st, 11 day inoculum TIME pH COPPER T-IRON TOC TEMP SIZE Hours g/L g/L ppm o c Um 0.00 1.82 0.3 0.4 39 35 2.46 49.16 2.42 0.5 0.7 38 29 2.62 96.16 2.21 0.8 0.9 47 28 3.03 145.16 2.37 0.9 1.0 23 28 2.58 192.82 2.20 0.8 1.3 5 45 3.23 243.32 2.80 0.9 1.5 13 36 2.54 288.15 2.30 1.0 2.0 27 38 1.72 338.48 2.20 1.0 2.6 19 40 3.06 385.56 2.56 1.0 2.5 20 36 4.03 302 Table 70 Experiment 3 ballmiiled, st / 303 TIME PH COPPER T-IRON TOC TEMP SIZE Hours g/L g/L ppm °C Mm 21.50 2.12 0.6 0.7 19 31 2.69 69.66 2.05 1.2 0.7 20 26 3.73 118.16 1.67 1.5 1.0 38 26 3.07 168.99 1.42 2.0 2.3 47 28 3.50 215.10 1.30 2.2 3.0 123 30 4.69 261.46 1.30 2.5 3.8 80 28 3.34 312.04 1.10 3.0 4.6 80 34 7.76 360.12 1.10 3.5 5.1 155 33 4.20 407.20 1.30 4.3 5.9 130 39 452.20 1.90 5.1 6.6 250 35 524.61 1.15 6.8 8.4 307 35 572.81 1.00 7.7 9.8 230 32 Table 71 Experiment 4 ballmiiled, c-r TIME pH COPPER T-IRON TOC ACID TEMP Hours g/L g/L ppm mL °C 23.00 2.04 1.5 0.7 23 0.8 35 68.00 1.84 2.4 1.0 15 2.4 35 117.41 2.09 2.8 1.8 14 3.7 32 162.91 2.08 2.4 1.8 23 4.4 33 212.41 2.02 3.5 2.2 24 4.4 39 261.32 1.90 4.8 2.6 22 4.4 32 310.98 1.87 4.8 2.6 4.4 34 361.98 1.94 5.8 3.0 4.4 30 408.59 1.93 5.8 3.0 4.4 34 456.22 1.72 6.0 3.6 4.4 34 Table 72 Experiment 5 ballmiiled, c-r TIME pH COPPER T-IRON TOC ACID SIZE Hours g/L g/L ppm mL Mm 0 1.92 0 0 9 0 6.50 48.16 2.15 10.6 3.0 12 3.3 6.86 87.49 2.16 23.1 6.2 22 16.8 2.98 135.99 2.16 48.4 3.3 51 28.5 2.34 184.49 1.96 55.2 5.4 57 30.0 2.77 226.49 1.87 64.3 4.5 81 31.6 TIME TEMP OXYGEN Hours °C ppm 0 35 48.16 33 2.3 87.49 28 2.0 135.99 28 1.6 184.49 28 2.8 226.49 28 3.4 Table 73 Experiment 6 38um, c-r, low pulp / 305 TIME pH COPPER T-IRON TOC ACID SIZE Hours g/L g/L ppm mL (xm 0 1.86 0 0 20 0 3.20 39.16 2.06 2.2 1.0 30 1.6 2.66 90.66 2.09 4.8 2.4 93 4.3 2.37 135.57 2.10 5.0 2.4 42 4.3 2.89 186.98 2.15 4.7 2.5 52 4.3 3.02 233.31 2.07 4.6 2.4 22 4.9 2.00 278.81 2.13 4.4 2.7 78 5.5 2.81 326.14 2.12 5.8 3.3 30 6.3 2.58 374.89 2.07 7.3 3.5 37 7.1 6.25 426.39 2.10 8.3 4.5 49 8.0 2.36 493.14 2.15 9.3 5.0 53 9.2 4.82 542.30 1.51 9.1 5.6 82 11.5 3.90 595.80 1.89 11.9 6.8 81 11.6 2.05 642.80 2.16 10.8 7.0 66 12.0 1.98 TIME Eh TEMP OXYGEN Hours mV °c ppm 0 355 35 5.0 39.16 340 32 5.8 90.66 400 27 5.4 135.57 380 27 3.2 186.98 340 28 3.6 233.31 380 28 3.6 278.81 395 25 5.7 326.14 365 29 5.3 374.89 370 32 4.4 426.39 395 30 2.7 493.14 390 32 4.2 542.30 375 30 6.0 595.80 360 28 4.6 642.80 350 26 6.4 Table 74 Experiment 7 28um, st TIME PH COPPER T-IRON TOC SIZE Hours g/L g/L ppm um 0 1.98 0 0 50 2.73 44.50 2.06 3.4 4.0 53 2.52 99.50 1.78 11.5 8.9 438 2.08 165.00 1.64 12.5 7.8 596 2.83 214.25 2.09 13.6 9.8 155 1.18 265.50 1.64 14.0 9.8 97 1.07 310.91 1.64 14.0 11.3 137 0.95 357.91 1.61 14.0 10.8 122 1.09 404.75 1.62 14.2 10.3 153 1.28 TIME Eh TEMP OXYGEN Hours mV °C ppm 0 315 38 6.0 44.5 355 38 5.4 99.5 530 40 165.00 485 36 214.25 520 39 265.50 665 36 310.91 680 36 6.2 357.91 660 37 4.6 404.75 675 36 6.4 Table 75 Experiment 8 ballmilled, c-r, low pulp / 307 TIME pH COPPER Hours g/L 44.5 1.75 0.3 93.34 1.68 140.34 1.70 3.0 187.92 1.79 10.2 TIME Eh TEMP Hours mV °C 44.5 355 30 93.34 350 30 140.34 405 32 187.92 465 31 T-IRON TOC ACID g/L ppm mL 0.5 0 0.5 0.6 0 0.5 3.3 20 0.5 3.2 309 0.5 OXYGEN ppm 7.2 7.0 3.4 3.0 Table 76 Experiment 9 ballmiiled, c--r, low pulp TIME pH COPPER T-IRON TOC SIZE Hours g/L g/L ppm /xm 0 2.0 0 0 66 3.4 45.75 2.1 4.6 3.0 240 3.94 91.0 2.0 14.4 0.1 309 138.33 1.9 21.4 T 5>296 189.99 1.8 35.9 1.7 509 237.24 1.4 8.9 0.5 256 288.49 1.4 11.1 1.0 175 330.15 1.4 11.4 1.3 192 TIME Eh TEMP OXYGEN Hours mV °C ppm 0 360 38 6.2 45.5 420 32 4.0 91.0 565 31 4.7 138.33 575 31 5.8 189.99 570 31 4.6 237.24 445 36 7.2 288.49 420 33 6.6 / 309 Table 77 Experiment 10 ballmilled, c-r, low pulp TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.0 0 0 41 0 20 2.2 0.6 0.6 64 1.2 65.6 1.6 7.2 4.7 123 2.3 109.6 1.4 6.7 6.6 262 2.3 153.6 1.4 11.5 7.7 402 2.3 179.4 1.4 11.8 7.6 444 2.5 252.4 1.4 14.5 7.5 530 2.5 297.6 1.4 15.7 6.8 487 2.5 TIME Eh TEMP OXYGEN Hours mV °C ppm 0 35 20.0 395 34 3.4 65.6 405 35 4.0 109.6 35 3.8 153.6 435 34 3.8 179.4 445 33 3.5 252.4 460 34 3.8 297.6 615 33 3.0 Table 78 Experiment 11 ballmiiled, c-r, low pulp / 310 TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.0 0 0 15 0 43.0 1.4 2.2 1.6 113 0 96.75 1.4 7.1 4.9 268 0 138.5 1.3 8.4 5.7 281 0 193.0 1.3 10.5 8.5 293 0 235.0 1.3 11.7 7.6 284 0 289.25 1.2 17.6 7.7 339 0 336.25 1.2 24.4 9.7 557 0 382.75 1.2 25.1 9.6 595 0 431.25 1.2 21.9 8.0 512 0 TIME Eh TEMP OXYGEN Hours mV °C ppm 0 355 24 7.8 43.0 395 37 6.3 96.75 400 36 6.0 138.5 410 35 6.6 193.0 420 35 5.6 235.0 430 34 3.6 289.25 445 34 336.25 460 35 382.75 505 35 431.25 545 35 / 311 Table 79 Experiment TIME pH COPPER T-IRON Hours g/L g/L 0 2.0 0 0 55.16 2.2 0 1.3 100.41 2.1 0.7 1.5 146.91 2.0 1.1 2.1 195.41 2.0 1.7 2.6 242.82 2.0 2.9 3.0 ballmilled, c-r, low pulp TOC ACID Eh TEMP ppm mL mV °C 30 0 325 25 34 3.35 335 32 57 3.75 365 35 33 3.75 360 36 41 3.75 365 38 76 3.75 360 36 Table 80 Experiment 13 ballmiiled, c-r, low pulp TIME pH COPPER Hours g/L 0 1.8 0 45.91 1.5 1.4 93.24 2.0 2.4 141.57 1.4 3.9 209.20 1.8 5.7 285.48 1.0 7.6 334.56 1.0 6.8 382.39 1.0 8.7 429.98 1.0 8.7 TIME Eh TEMP Hours mV °C 0 325 32 45.91 375 38 93.24 375 38 141.57 405 34 209.20 400 25 285.48 400 26 334.56 410 26 382.39 410 26 429.98 415 26 T-IRON TOC ACID g/L ppm mL 0 18 0 1.5 62 7.0 4.6 156 7.0 6.0 114 17.6 7.3 198 17.6 8.7 120 17.6 8.8 176 17.6 8.7 153 17.6 9.0 164 17.6 Table 81 Experiment 14 ballmiiled, c-r, low pulp TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.0 0 0 23 0 46.3 1.7 0 1.7 16 0 96.38 1.8 1.6 3.8 131 0 193.79 1.6 9.0 7.9 339 0 264.95 1.6 13.2 9.5 357 0 TIME Eh TEMP Hours mV °C 0 320 31 46.3 360 34 96.38 380 34 193.79 390 34 264.95 400 34 Table 82 Experiment 15 ballmilled, c-r, low pulp TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.0 0 0 20 0 45.5 2.4 3.7 2.7 24 5.1 93.5 2.0 14.2 3.4 98 9.9 143.75 1.6 13.5 7.0 251 9.9 185.75 1.4 12.1 7.9 389 9.9 TIME Eh TEMP Hours mV ° C 0 325 33 45.5 390 34 93.5 390 33 143.75 395 33 185.75 385 33 Table 83 Experiment 16 ballmilled, c-r, low pulp TIME PH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.0 0 0 34 0 46.5 1.7 1.6 1.8 178 0 92.83 1.8 5.3 4.6 172 0 116.83 1.8 7.4 6.2 223 0 139.49 1.8 6.9 6.9 225 0 TIME Eh TEMP Hours mV °C 0 330 31 46.5 375 36 92,83 380 35 116.83 385 35 139.49 390 35 / 316 Table 84 Experiment 17 ballmiiled, st TIME pH COPPER T-IRON TOC EH TEMP Hours g/L g/L ppm mv °C 0 1.99 0 0 24 470 35 44.5 2.56 0.5 0.2 36 500 35 117.25 2.64 1.4 0 60 500 35 163.83 2.56 2.2 0.2 100 515 35 211.99 2.17 2.6 0 72 550 35 256.49 2.00 2.5 0.3 91 580 35 308.49 1.96 2.8 0.4 121 590 35 356.15 1.93 3.2 1.0 110 590 35 404.15 1.80 3.8 1.7 140 600 35 452.23 1.99 4.8 2.7 141 550 35 500.48 1.86 6.6 3.0 169 500 35 548.48 1.81 7.8 3.8 220 480 35 620.14 1.86 8.8 4.8 344 520 35 644.72 1.90 7.7 4.7 403 510 35 693.63 2.02 8.5 4.6 300 510 35 739.46 1.91 9.2 4.7 313 505 35 789.46 1.94 8.6 5.2 495 35 839.46 1.97 7.4 4.3 313 480 35 891.96 1.94 8.6 4.3 306 500 35 / 317 Table 85 Experiment 18 as received, c-r,low pulp TIME pH COPPER T-IRON TOC TEMP EH Hours g/L g/L ppm °C mV 0 1.93 0 0 16 35 385 48.0 1.82 2.0 0.4 11 35 365 94.58 2.14 1.2 0.9 12 34 375 142.74 1.85 1.4 2.0 67 36 370 187.24 1.93 4.2 3.6 20 35 380 239.24 1.98 8.2 8.6 292 32 410 286.90 1.82 17.4 9.6 727 32 455 334.90 1.69 23.0 11.4 738 31 590 382.98 1.58 22.2 13.7 686 30 610 431.23 1.55 23.2 16.0 714 31 630 479.23 1.56 23.7 16.0 666 31 650 550.89 1.57 27.1 17.6 723 30 660 575.47 1.58 28.4 16.0 754 30 660 624.38 1.55 32.0 17.0 745 32 650 670.21 1.45 34.0 18.0 715 31 630 / 318 Table 86 Experiment 19 ballmilled, c-r, low pulp TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.05 0 0.5 0 0 43.75 1.64 0.6 1.1 7 0 93.5 1.67 0.8 2.3 7 0 142.25 1.70 0.8 2.9 0 0 TIME Eh TEMP OXYGEN Hours mV °C ppm 0 360 29 43.75 385 35 93.5 370 35 2.2 142.25 385 35 2.4 Table 87 Experiment 20 ballmiiled, c-r, low pulp TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.0 0 4.4 2 0 46.5 2.12 0.3 1.2 0 1.1 89.25 2.09 2.2 2.3 0 3.7 136.92 2.15 3.2 4.2 0 7.9 187.75 2.29 6.7 5.3 0 12.5 234.08 2.09 23.2 6.9 0 17.8 285.99 2.15 11.3 9.2 7 22.5 334.82 2.24 19.0 9.9 0 25.8 380.98 2.14 14.8 8.3 0 29.4 427.48 2.06 16.5 8.7 8 32.2 TIME Eh TEMP Hours mV °C 0 345 35 46.5 385 35 89.25 410 35 136.92 395 35 187.75 415 35 234.08 415 35 285.99 395 35 334.82 415 35 380.98 420 35 427.48 420 35 Table 88 Exper iment 21 bal lmi l led, c-r , low pulp / 320 T I M E p H C O P P E R T - I R O N T O C A C I D Hours g /L g/L ppm m L 0 2.32 0 0.3 0 0 45.75 2.30 0.7 0.9 0 0.8 96.75 2.07 1.1 1.8 0 3.0 144.66 2.13 3.1 2.9 0 3.5 192.74 2.19 1.7 3.1 0 4.1 241.32 2.04 2.0 3.4 0 5.2 287.48 2.09 2.1 3.9 0 5.2 335.64 1.79 1.4 4.0 0 5.7 387.30 2.08 1.2 4.5 0 5.7 431.38 2.10 0 4.9 0 6.6 T I M E E h T E M P O X Y G E N Hours m V °C ppm 0 315 31 0.2 45.75 400 35 1.4 96.75 370 35 3.2 144.66 365 35 3.3 192.74 370 35 241.32 370 35 287.48 370 35 335.64 370 35 387.30 380 35 431.38 375 35 Table 89 Experiment 22 ballmiiled, st, freeze dried TIME Hours 0 47.5 96.16 142.24 TIME Hours 0 47.5 96.16 142.24 P H COPPER T-IRON TOC g/L g/L ppm 2.50 1.5 0.7 3 2.04 5.8 4.2 74 2.06 8.5 6.1 64 2.03 9.4 7.0 216 Eh TEMP OXYGEN mV °C ppm 335 36 4.7 360 35 375 35 385 35 / 322 Table 90 Experiment 23 ballmiiled, c-r TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.08 0 0.9 0 0 43.33 2.22 0.7 2.2 0 1.3 96.16 2.14 2.2 3.7 0 5.7 142.57 2.13 2.8 5.0 0 11.1 188.98 2.22 1.7 5.6 0 15.0 236.56 2.11 4.0 6.8 0 22.7 286.47 2.25 10.4 8.1 0 28.1 335.05 2.32 10.8 7.6 0 32.9 381.05 2.51 13.6 5.7 0 35.4 430.30 2.41 13.0 5.4 0 38.8 483.21 2.09 12.6 6.4 0 43.9 535.87 2.08 15.7 8.2 0 48.2 583.37 2.13 6.3 7.6 4 51.5 620.20 2.10 10.1 9.0 1 54.2 677.86 2.17 20.2 8.8 0 56.8 725.86 2.14 20.0 9.6 9 59.8 TIME Eh FERROUS FERRIC TEMP Hours mV g/L g/L °C 0 300 1.2 0 35 43.3 355 2.3 0.2 35 96.16 360 3.5 0.3 35 142.57 370 5.8 0.7 35 188.98 370 6.6 0.9 35 236.56 380 8.1 1.0 35 286.47 380 7.8 1.2 35 335.05 385 7.9 1.9 35 381.05 385 5.9 1.2 35 430.30 395 6.2 1.2 35 483.21 400 7.8 1.3 35 535.8.7 400 10.3 3.4 35 583.37 400 8.2 2.1 35 630.20 405 9.5 2.2 35 677.86 • 405 10.0 2.6 35 725.86 410 9.5 3.8 35 Table 91 Experiment 24 ballmilled, st, inoculum age TIME P H COPPER T-IRON TOC Hours g/L g/L p p m 0 2.08 0 1.2 0 46.6 2.50 0.1 1.3 0 94.32 2.39 0.8 2.2 0 142.32 2.25 4.4 3 2 3 192.57 2.28 6.2 4.2 18 235.57 2.16 7.3 4.9 108 283.15 1.81 13.8 4.4 115 TIME Eh FERROUS FERRIC TEMP Hours mV g/L g/L °C 0 380 0.7 0.5 35 46.6 350 1.2 0.1 35 94.32 365 2.1 0.2 35 142.32 390 2.7 0.4 35 192.57 405 3.4 0.8 35 235.57 420 3.5 1.4 35 283.15 535 0 4.4 34 Table 92 Experiment 25 ballmiiled, c-r, aeration / 324 TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.10 0 1.5 0 0 25.25 2.13 1.4 2.1 0 0.2 68.25 2.23 0 2.8 0 1.9 115.83 2.23 15.4 5.6 0 11.2 167.25 2.28 15.0 9.0 0 19.4 214.0 2.24 21.2 10.5 5 28.7 262.25 2.26 25.4 11.4 18 32.9 308.3 2.16 26.7 11.0 70 38.9 359.7 2.15 28.5 11.6 12 42.1 405.4 2.26 26.3 9.4 2 43.8 TIME Eh FERROUS FERRIC TEMP Hours mV g/L g/L °C 0 1.1 0.4 35 25.25 355 1.9 0.1 35 68.25 360 2.6 0.2 34 115.83 380 4.9 0.7 35 167.25 390 7.8 1.1 35 214.0 390 9.2 1.3 35 262.25 395 9.8 1.6 35 308.3 400 9.0 2.0 35 359.7 395 9.4 1.6 34 405.4 395 7.8 1.6 35 Table 93 Experiment 26 ballmilled, st, inoculum age TIME Hours 0 45.25 92.0 140.25 TIME Hours 0 45.25 92.0 140-25 pH COPPER T-IRON g/L g/L 2.01 0 0.8 1.84 3.0 3.9 1-90 6.2 5.6 1.95 8.1 73 TOC ppm 0 24 86 285 Eh FERROUS FERRIC TEMP mV g/L g/L °C 410 0.8 0.1 35 380 3.5 0.4 34 400 4.8 0.8 35 420 5.5 1.8 35 / 326 Table 94 Experiment 27 ballmiiled, c-r, aeration TIME PH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.02 0 1.1 7 0 22.33 2.09 0 1.0 0 0 68.99 1.97 1.8 2.9 12 0 117.19 2.08 7.5 5.5 175 0 163.19 2.02 10.3 6.4 63 0 TIME Eh FERROUS FERRIC TEMP Hours mV g/L g/L °C 0 315 1.0 0.1 35 22.33 370 0.9 0.1 35 68.99 385 2.5 0.4 35 117.19 395 4.6 0.9 35 163.19 415 5.1 1.3 35 Table 95 Experiment 28 ballmilled, c-r, aeration TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.05 0 1.0 0 0 45.75 1.84 10.3 1.1 0 0 89.75 1.95 12.0 2.7 0 0 139.66 1.96 15.0 4.8 24 0 TIME Eh FERROUS FERRIC Hours mV g/L g/L 0 320 0.8 0.2 45.75 370 1.1 0 89.75 380 2.4 0.3 139.66 385 4.2 0.6 Table 96 Experiment 29 9K iron, st TIME Hours 2.66 18.83 65.33 115.74 163.57 210.98 TIME Hours 2.66 18.83 65.33 115.74 163.57 210.98 pH T-IRON TOC g/L ppm 1-98 7.9 0 2.03 7.9 12 2.12 7.3 0 2.22 6.3 0 2.26 5.0 0 2.18 4.5 0 Eh FERROUS FERRIC TEMP mV g/L g/L °C 335 7.2 0.6 35 355 7.5 0.4 35 390 6.2 1.0 35 420 4.0 2.2 35 590 0.1 4.8 35 570 0 4.5 35 Table 97 Experiment 30 ballmilled, c-r, aeration TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 2.66 2.17 0 1.9 0 0 18.33 2.10 0 2.1 0 0 65.33 2.30 0 2.8 0 0.6 115.74 2.12 0 4.0 0 3.8 163.57 2.07 3.1 6.8 0 11.7 210.98 . 2.06 17.8 13.0 0 27.4 260.31 2.12 23.5 10.0 0 38.8 306.31 2.14 29.7 12.2 2 49.1 355.56 • 2.22 37.4 15.2 5 56.6 400.31 2.10 40.0 15.2 0 56.6 TIME Eh FERROUS FERRIC TEMP Hours mV g/L g/L °C 2.66 315 1.7 0.1 27 18.33 340 1.9 0.2 32 65.33 365 2.4 0.4 35 115.74 370 3.6 0.4 35 163.57 385 5.8 1.0 35 210.98 400 11.0 1.9 35 260.31 410 7.7 2.3 35 306.31 425 7.2 4.9 35 355.56 450 4.2 11.0 35 400.31 525 0 15.2 35 Table 98 Experiment 31 c-r, solid free inculum / 330 TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.12 0 1.6 0 0 48.33 1.90 0.5 1.6 0 0 94.66 2.09 0 1.8 0 0 143.66 2.07 0.2 3.3 78 0 188.33 2.10 2.0 5.1 94 0 235.99 2.07 2.5 5.5 0 0 287.15 1.98 1.4 5.6 0 0 TIME Eh FERROUS FERRIC Hours mV g/L g/L 0 320 1.4 0.2 48.3 365 1.5 0.1 94.66 365 1.7 0.1 143.66 385 2.9 0.4 188.33 380 4.5 0.6 235.99 380 4.9 0.6 287.15 390 5.0 0.6 Table 99 Experiment 32 c-r, solid free inoculum TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.25 0 1.3 0 0 48.33 2.01 0 1.6 0 0 94.66 2.10 0 1.8 0 0 143.66 2.10 0.5 1.7 87 1.9 188.33 2.18 3.9 5.6 190 1.9 235.99 2.09 5.0 7.0 213 1.9 287.15 2.01 9.5 7.8 0 1.9 TIME Eh FERROUS FERRIC Hours mV g/L g/L 0 315 1.2 0.1 48.33 360 1.5 0.1 94.66 360 1.7 0.1 143.66 380 1.6 0.1 188.33 395 4.7 0.9 235.99 400 5.9 1.1 287.15 400 6.5 1.3 / 332 Table 100 Experiment 33 9K iron, cell composition TIME Hours 0 22.3 71.3 116.3 167.3 pH 2.04 1.99 2.10 2.13 2.13 T-IRON g/L 7.6 7.9 7.6 9.1 7.4 TOC ppm 7 12 21 TIME Hours 0 22.3 71.3 116.3 167.3 Eh mV 340 370 390 410 440 FERROUS g/L 7.2 7.4 6.6 6.9 3.5 FERRIC g/L 0.3 0.5 1.0 2.2 3.9 TEMP °C 35 35 35 35 Table 101 Experiment 34 c-r, high aeration TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.32 0 1.1 3 0 43.16 1.99 0 1.9 21 0.1 94.49 2.06 0 2.6 0 0.1 142.24 2.15 0 2.7 0 0.2 192.32 2.11 0.7 4.2 0 0.3 261.98 2.12 2.8 4.6 0 0.3 307.81 2.06 3.0 6.4 0 0.4 TIME Eh FERROUS FERRIC TEMP Hours mV g/L g/L °C 0 305 1.1 0 35 43.16 350 1.8 0.1 35 94.49 360 2.3 0.3 35 142.24 375 2.5 0.2 35 192.32 375 3.8 0.4 35 261.98 380 4.3 0.4 35 307.81 380 5.7 0.6 35 Table 102 Experiment 35 c-r, w.o. inoculum TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.20 0 1.4 0 0 17.66 2.07 0 2.0 0 1.9 68.99 2.24 0 2.0 0 2.3 116.74 2.37 0 2.6 0 2.7 166.82 2.16 0 3.3 0 4.3 236.48 2.28 0 3.9 0 4.6 282.31 2.29 0 3.0 0 4.6 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0 310 35 1.4 0 17.66 350 35 1.8 0.1 68.99 355 35 1.8 0.2 116.74 355 35 2.5 0.1 166.82 355 35 3.0 0.3 236.42 360 35 3.5 0.4 282.31 360 35 2.9 0.1 / 335 Table 103 Experiment 36 c-r, solids inoculum TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0.25 2.09 0 1.5 0 0 47.25 1.83 1.8 1.7 0 0 94.66 1.93 4.7 3.5 0 0 140.49 1.99 4.3 4.3 0 0 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0.25 320 35 1.4 0.1 47.25 370 35 1.5 0.2 94.66 380 35 3.1 0.4 140-49 380 35 3.8 0.5 / 336 Table 104 Experiment 37 c-r, liquid supernatant inoculum TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0.25 2.25 0 1.8 0 0 47.25 2.02 1.8 1.7 0 0 94.66 2.12 0 1.3 0 0 140.49 2.05 0 1.8 0 0 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0.25 310 35 1.7 0.1 47.25 360 35 1.5 0.2 94.66 370 35 1.2 0.1 140.49 370 35 1.6 0.2 / 337 Table 105 Experiment 38 2.58 um, c-r TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.25 0 1.7 0 0 20.83 1.83 0.8 1.9 0 0 65.83 2.22 1.3 1.6 0 0.2 119.49 2.10 1.1 1.6 0 0.6 162.57 2.13 3.2 3.4 0 0.6 210.48 2.16 9.6 6.8 2 0.6 261.31 2.27 9.1 9.1 16 1.5 310.31 1.99 13.6 7.8 11 1.5 357.31 1.81 15.7 10.0 56 1.5 TIME Eh FERROUS FERRIC MEAN SIZE Hours mV g/L g/L um 0 315 1.6 0.1 1.98 20.83 360 1.6 0.2 65.83 355 1.6 0.0 1.84 119.49 365 1.4 0.3 162.57 360 3.1 0.3 1.61 210.48 375 5.9 0.9 261.31 385 7.6 1.5 2.01 310.31 390 6.2 1.5 357.31 390 8.0 2.0 2.33 / 338 Table 106 Experiment 39 2.58 Mm, st TIME pH COPPER T-IRON TOC TEMP Hours g/L g/L PPm °C 0 2.00 0 2.0 0 35 20.83 1.85 0.6 2.6 0 35 65.83 1.88 4.0 4.6 0 35 119.49 2.07 5.7 7.3 15 36 162.57 2.26 7.2 8.2 0 33 210.48 2.02 7.5 10.6 14 35 261.31 2.01 9.1 8.8 38 35 310.31 1.80 8.6 8.7 0 35 357.31 1.56 6.7 8.5 8 35 TIME Eh FERROUS FERRIC 50% SIZE Hours mV g/L g/L Mm 0 340 1.9 0.0 2.50 20.83 350 2.4 0.2 65.83 360 4.3 0.3 0.94 119.49 380 6.2 1.1 162.57 385 7.2 1.1 1.12 210.48 400 8.3 2.3 261.31 415 6.3 2.5 0.56 310.31 530 3.1 5.6 357.31 540 1.5 7.0 0.57 / 339 Table 107 Experiment 40 46.8 um, c-r TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.31 0 0.6 0 0 50.58 1.97 0 0.5 01 0 98.58 2.26 0 0.6 0 0.2 145.83 2.30 0 1.3 0 1.1 190.24 2.22 0 1.6 0 1.9 236.74 2.12 0 2.4 0 3.3 287.90 2.19 0 3.8 0 5.6 335.15 2.05 4.7 6.3 0 12.3 382.31 2.21 8.4 7.8 0 21.7 430.31 2.20 10.4 13.0 0 27.7 478.81 2.07 16.1 10.0 6 31.5 526.06 2.28 18.9 9.1 12 32.2 575.06 2.17 15.0 6.9 33.9 TIME Eh FERROUS FERRIC TEMP 50% SIZE Hours mV g/L g/L °C um 0 295 0.3 0.3 35 12.5 50.58 350 0.3 0.2 35 98.58 350 0.6 0.0 35 4.93 145.83 350 1.2 0.1 35 190.24 350 1.6 0.0 35 4.56 236.74 350 2.2 0.2 35 287.90 350 3.5 0.3 35 3.34 335.15 360 5.6 0.7 24 382.31 370 6.6 1.0 35 2.53 430.31 385 11.1 1.9 35 478.81 390 8.8 1.3 35 1.55 526.06 390 8.0 1.2 35 575.06 390 6.0 1.0 35 1.83 / 340 Table 108 Experiment 41 46.8 Mm, st TIME pH COPPER T-IRON TOC 50% Hours g/L g/L ppm Mm 0 2.14 0 1.5 0 1.68 50.58 2.04 0.7 2.0 0 98.58 2.00 1.1 3.5 0 1.53 145.83 2.16 4.2 5.4 0 190.24 2.26 6.4 7.1 0 1.44 236.74 2.12 8.1 7.1 0 287.90 1.75 9.5 6.7 41 1.24 335.15 1.54 10.4 9.3 211 382.31 1.46 10.1 11.0 377 1.38 430.31 1.60 17.0 10.0 338 478.81 1.55 14.0 9.4 310 1.23 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0 335 35 1.3 0.2 50.58 330 35 2.0 0.2 98.58 355 34 3.2 0.3 145.83 375 35 4.9 0.5 190.24 390 35 6.0 1.2 236.74 400 34 5.6 1.5 287.90 535 35 0.6 6.1 335.15 560 35 0.8 8.5 382.31 580 35 0.4 10.6 430.31 555 32 0.2 10.0 478.81 550 35 0.6 8.8 / 341 Table 109 Experiment 42 ballmiiled, c-r TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.35 0 0.5 0 0 44.33 2.17 0 1.3 0 92.74 2.21 0 1.7 0 1.1 139.07 2.07 0 1.9 2.4 189.57 1.94 0 3.0 0 6.8 237.98 2.18 7.5 4.8 15.4 285.23 2.17 18.2 10.6 0 27.35 332.23 2.30 31.0 12.2 38.65 382.73 2.04 31.0 12.2 49.5 426.73 2.16 28.3 14.1 51.6 TIME Eh TEMP FERROUS FERRIC 50% Hours mV °C g/L g/L Mm 0 300 35 0.4 0.1 7.01 44.33 355 35 0.8 0.4 92.74 355 35 1.3 0.4 3.39 139.07 360 35 1.6 0.3 189.57 375 35 2.7 0.4 1.95 237.98 385 35 4.0 0.7 285.23 395 35 8.8 1.9 2.21 332.23 400 35 9.8 2.4 382.73 410 35 8.8 3.3 1.31 426.73 410 35 10.3 3.8 1.77 / 342 Table 110 Experiment 43 46.8 um, s t TIME pH COPPER T-IRON TOC 50% Hours g/L g/L ppm um 0 2.18 0 1.5 0 3.28 44.50 1.96 0.7 1.6 0 93.25 2.35 1.0 2.2 0 1.79 139.25 1.79 4.7 5.0 2 189.66 2.01 8.7 9.9 32 1.59 237.99 1.88 16.4 8.7 209 285.23 1.66 16.7 10.5 255 1.57 332.23 1.69 18.8 12.6 345 382.73 1.64 17.0 11.4 313 1.48 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0 365 35 1.4 0.1 44.56 350 35 1.4 0.2 93.25 335 35 2.1 0.1 139.25 380 35 4.4 0.6 189.66 415 35 6.9 2.9 237.99 550 35 0.2 8.5 285.23 585 35 0.4 10.0 332.23 635 35 0.3 12.3 382.73 630 35 0.0 11.4 / 343 Table 111 Experiment 44 30.4 um, c-r TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 1.0 2.46 0 0.8 0 0 50.92 2.16 0 1.2 0 0.4 96.92 2.19 0 1.5 0 1.1 145.08 2.18 0 2.6 0 2.2 195.91 2.15 0.8 3.7 0 5.1 244.57 2.02 7.7 7.4 0 15.9 292.32 2.25 15.2 11.8 0 28.6 339.07 2.21 18.7 10.0 0 35.4 384.57 2.19 21.3 10.5 5 38.0 431.93 2.11 22.8 10.6 20 40.0 TIME Eh TEMP FERROUS FERRIC 50% i Hours mV °C G/L G/L Um 1.0 305 35 0.7 0.3 32.29 50.92 355 35 1.0 0.2 96.92 355 35 1.4 0.1 2.03 145.08 360 34 2.1 0.4 195.91 365 35 3.4 0.4 2.03 244.57 385 35 6.5 1.0 292.32 395 35 9.8 2.0 2.24 339.07 400 35 8.4 1.6 384.57 390 35, 8.4 2.1 1.97 431.93 395 35 8.2 2.4 1.92 / 344 Table 112 Experiment 45 30.4Mm, st TIME pH COPPER T-IRON TOC SIZE Hours g/L g/L ppm Mm 0 2.03 0 2.3 0 32.29 51.80 3.31 1.0 2.5 0 97.16 2.81 1.7 3.4 0 1.67 145.24 1.74 3.8 7.0 6 196.15 1.84 7.0 6.5 30 1.87 244.73 2.17 9.6 8.4 102 292.48 1.63 10.1 8.8 134 1.71 339.23 1.82 11.9 10.2 122 1.31 384.56 1.80 13.1 11.0 141 1.13 432.09 1.86 12.8 9.4 161 1.0 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0 350 35 2.0 0.3 51.80 350 32 2.4 0.0 97.16 340 34 3.1 0.3 145.24 360 34 6.2 0.7 196.15 390 34 5.3 1.3 244.73 530 33 0.4 7.9 292.48 590 34 0.2 8.7 339.23 600 35 0.2 10. 384.56 610 34 0.0 11. 432.09 600 34 0.2 9.3 / 345 Table 113 Experiment 46 14.3 um, c-r TIME pH COPPER T-IRON TOC ACID SIZE Hours g/L g/L ppm mL um 0 2.20 0 1.0 0 0 2.58 21.66 1.83 0 0.8 0 0 71.82 2.37 0 1.4 0 0.6 119.73 2.20 0 1.4 0 1.2 165.73 2.38 0 1.5 0 1.8 1.99 210.98 1.85 0 2.7 0 4.0 262.48 2.32 0 3.6 0 5.5 1.42 333.73 2.13 9.3 9.1 6 18.2 1.21 380.98 2.16 19.7 4.4 32 23.7 425.48 2.19 24.8 3.3 23 32.6 1.37 479.31 2.14 33.3 4.1 62 42.0 1.68 524.97 2.34 42.0 7.3 110 46.1 575.30 47.1 149 46.1 1.21 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0 310 35 0.8 0.1 21.66 350 35 0.5 0.3 71.82 380 35 1.2 0.2 119.73 355 35 1.1 0.3 165.73 350 35 1.2 0.3 210.98 355 35 2.5 0.2 262.48 365 35 2.5 1.0 333.73 390 35 6.3 2.9 380.98 395 35 3.8 0.6 425.48 410 35 2.8 0.6 479.31 425 35 2.8 1.3 524.97 430 35 2.3 5.0 575.30 435 35 3.9 / 346 Table 114 Experiment 47 14.3 izm, st TIME pH COPPER T-IRON TOC SIZE Hours g/L g/L ppm Mm 48.08 2.48 0 1.0 10 1.61 93.75 3.05 0.3 1.6 10 140.41 2.56 0.0 7.4 11 2.05 185.49 1.85 1.1 2.2 5 230.90 1.56 3.8 4.6 18 1.98 281.23 1.65 5.7 6.2 17 325.23 1.35 7.3 8.2 15 1.41 376.23 1.55 8.4 9.1 42 424.23 1.85 7.7 11.4 37 1.07 472.98 1.65 7.2 13.3 52 574.39 1.51 7.9 13.7 53 1.28 616.23 1.52 8.4 13.7 37 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 48.08 355 33 0.7 0.3 93.75 340 35 1.1 0,4 140.41 335 35 7.4 0.0 185.49 350 35 2.0 0.3 230.90 370 35 4.2 0.4 281.23 385 35 5.3 1.0 325.23 415 35 6.1 2.1 376.23 485 35 1.1 8.1 424.23 600 35 0.0 11.4 472.98 660 35 0.4 12.9 574.39 670 35 0.0 13.7 616.23 680 35 0.0 13.7 / 347 Table 115 Experiment 48 8.63 /nm, c-r TIME pH COPPER T-IRON TOC ACID 50% Hours g/L g/L PPm mL /nm 22.5 2.56 0 0.8 0 0.6 4.68 72.16 1.85 0 1.8 0 0.9 120.16 2.40 0 1.6 0 1.3 7.03 171.07 1.95 0.8 2.2 0 1.7 223.07 2.11 2.2 3.5 2 2.0 2.96 269.32 2.20 5.2 5.0 8 3.2 316.82 2.11 4.3 6.4 10 3.3 2.19 364.82 1.96 5.0 7.6 2 3.3 444.82 2.10 6.4 10.6 3 3.35 4.45 486.65 2.02 12.3 11.4 19 3.35 535.98 1.81 17.5 10.5 31 3.35 2.89 605.31 2.13 19.8 4.6 73 4.65 652.31 2.20 23.8 4.2 116 9.75 2.21 700.81 2.17 30.3 2.4 71 14.15 748.81 2.07 31.9 2.1 75 16.55 796.31 2.16 30.5 1.7 180 16.75 1.24 844.47 2.15 37.1 1.9 180 17.0 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 22.5 345 35 0.7 0 72.16 365 35 1.3 0.5 120.16 355 35 1.3 0.3 171.07 360 35 2.0 0.2 223.07 380 35 3.0 0.5 269.32 375 35 4.2 0.8 316.82 375 35 5.6 0.8 364.82 390 35 6.5 1.1 444.82 400 35 7.8 2.9 486.85 415 35 8.5 2.9 535.98 425 35 6.8 3.7 605.31 425 35 3.1 1.5 652.31 440 35 1.9 2.2 700.81 475 35 0.5 2.0 748.81 475 35 0.5 1.6 796.31 470 35 0.4 1.3 844.47 435 35 1.1 0.7 Table 116 Experiment 49 8.63 Mm, st TIME pH COPPER T-IRON TOC SIZE Hours g/L g/L ppm Mm 0 1.90 0.4 1.9 3 44 1.66 0.8 2.0 8 8.33 48 2.01 0 1.9 2 9.07 92.97 2.00 1.5 1.2 0 240.43 1.94 3.0 4.1 0 7.37 288.76 1.45 4.0 6.8 0 7.40 336.16 1.66 3.9 6.5 19 432.32 1.37 4.2 9.0 8 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 0 345 35 0.8 1.0 44 370 35 1.8 0.2 48 415 35 1.4 0.5 92.97 385 35 2.7 0.4 240.73 420 35 3.1 0.9 288.76 685 35 0.2 6.7 336.16 655 35 0 6.5 432.32 685 35 0 9.0 Table 117 Experiment 49 8.63 um, st TIME pH COPPER T-IRON TOC SIZE Hours g/L g/L ppm Mm 0.0 2.00 0 1.0 0 44.80 1.96 1.1 2.2 22 48.14 1.98 0.6 1.1 0 8.15 92.30 2.05 2.7 2.7 0 6.03 240.31 2.11 4.5 4.1 0 280.81 2.18 6.5 5.7 31 8.29 336.14 1.37 4.0 8.7 38 8.27 TIME Eh TEMP Hours mV °C 0 340 35 44.80 370 35 48.14 350 35 92.30 405 35 240.31 400 35 280.81 405 35 336.14 680 35 / 350 Table 118 Experiment T2 2L tank, c-r TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.21 0 1.6 0 0 48.16 2.17 2.0 2.5 3 15 94.74 1.99 2.3 5.3 7 53 142.90 2.16 5.7 10.3 10 133 191.56 2.20 22.9 10.5 33 193 238.64 2.47 31.1 7.8 20 . 245 287.64 2.43 40.4 4.4 54 275 337.00 2.26 40.4 4.2 78 292.5 TIME Eh TEMP SULPHUR FERROUS FERRIC Hours mV °C % g/L g/L 0 315 35 0 1.5 0.1 48.16 390 35 0 1.9 0.6 94.74 390 35 0 3.7 1.6 142.90 410 35 0.29 6.9 3.5 191.56 430 35 1.33 5.2 5.3 238.64 450 35 1.75 2.4 5.3 287.64 465 35 1.20 0.8 3.6 337.00 515 35 1.23 0.3 3.9 Table 119 Experiment T3 2L tank, c-/ 351 TIME pH COPPER T-IRON TOC ACID so Hours g/L g/L ppm mL g/L 0.58 1.98 0 2.7 12 0 0 47.58 2.05 0 4.7 9 18 0 92.91 2.10 3.6 5.6 31 65 0 138.41 2.18 16.5 7.6 183 178 6.44 166.41 2.23 25.8 7.4 80 252.5 7.34 216.41 2.14 38.4 10.0 70 375 19.45 264.41 2.04 51.3 15.2 130 435 21.03 312.41 2.23 58.9 15.8 196 440 13.72 357.66 1.89 47.0 14.9 161 445 14.16 411.16 1.95 57.7 17.5 150 445 13.42 482.16 1.94 43.4 15.2 166 450 528.58 1.83 46.0 17.4 . 143 452.5 17.89 573.32 1.87 47.5 13.7 151 452.5 15.89 627.32 1.81 43.1 15.8 132 457.5 17.18 672.98 1.93 45.8 16.9 117 457.5 16.90 718.64 2.01 42.2 16.9 143 460 16.38 763.64 2.08 44.3 17.4 64 460 814.97 2.05 40.6 14.5 76 460 TIME Eh TEMP DENSITY FERROUS FERRIC Hours mV °C g/L g/L g/L 0.58 330 35 278.7 1.9 0.8 47.58 385 35 268.4 4.1 0.6 92.91 390 35 271.3 4.4 1.2 138.41 440 35 307.9 3.5 4.1 166.41 450 35 325.0 2.5 5.0 216.41 35 372.5 1.4 8.7 264.41 500 35 415.2 0.8 14.5 312.41 580 35 422.7 0 15.8 357.66 350 35 413.5 0 14.8 411.16 590 35 425.9 0.4 17.1 482.16 590 35 381.3 0 15.2 528.58 610 35 394.7 0.5 16.8 573.32 610 35 396.8 0 13.7 627.32 610 35 393.0 0 15.8 672.98 610 35 392.4 0 16.8 718.64 610 35 371.6 0 16.8 763.64 600 35 381.4 0 17.4 814.97 600 35 0 14.5 Table 120 Experiment RT 2L tank, c-r TIME Hours 0 46.5 97.75 143.41 194.82 243.71 289.90 340.40 pH 2.14 2.42 1.94 2.33 2.21 2.11 2.12 2.18 COPPER g/L 0 21.9 29.9 70.2 51.7 59.7 60.4 67.1 T-IRON g/L 0 1.0 9.9 18.5 4.9 3.1 1.4 1.2 TIME Hours 0 46.5 97.75 143.41 194.82 243.71 289.90 340.40 TOC ppm 4 0 5 61 84 79 101 ACID mL 0 0 115 220 280 353 380 410 Eh mV 405 420 495 435 475 490 500 Table 121 Experiment b3 c-r, aeration TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 1 2.13 0 1.0 0 0 73.83 2.05 6.1 4.9 0.1 119.33 1.98 2.0 5.0 37 1.3 169.74 1.68 2.8 6.2 2.4 214.24 1.54 2.4 7.4 72 2.4 265.99 1.51 0.6 8.5 150 2.4 312.32 1.54 4.0 8.8 144 2.4 TIME Eh TEMP FERROUS FERRIC Hours mV °C g/L g/L 1 325 1.0 0 73.83 370 4.3 0.6 119.33 390 4.2 0.8 169.74 375 5.6 0.6 214.24 385 7.3 1.0 265.99 390 7.1 1.4 312.32 400 7.0 1.8 Table 122 Experiment b5 c-r, Ag concentration TIME pH COPPER T-IRON TOC ACID Hours g/L g/L ppm mL 0 2.06 0 1.3 0 44.83 2.18 0 1.5 0 0.35 92.33 1.66 0 4.5 22 2.4 142.24 1.53 4.5 8.2 95 2.5 190.74 1.69 9.1 12.9 151 2.5 236.74 1.66 8.3 14.5 347 2.6 287.90 1.99 7.6 14.5 250 2.6 341.23 1.96 13.6 16.3 569 2.7 393.81 2.08 15.4 16.3 797 2.8 TIME Eh FERROUS FERRIC Hours mV g/L g/L 0 325 1.3 0.0 44.83 370 1.4 0.1 92.33 380 4.1 0.4 142.24 380 6.0 1.2 190.74 400 9.8 3.1 236.74 410 9.8 4.7 287.90 445 5.1 9.4 341.23 475 3.0 13.3 393.81 570 1.1 15.2 APPENDIX 7 LOW PULP DENSITY REDOX CONTROLLED LEACHING In catalyzed leach experiments there is a correlation between elemental sulfur production and acid consumption, one depends on the other. In some cases acid can be consumed without any elemental sulfur being measured, this may be attributed to acid consumption due to gangue in the concentrate, which increases the pH above the control level of 2.2. Controlled-redox leaching of both Newmont and Gibraltar concentrates at a pulp density of 107 g/L did not produce S° and/or did not consume acid. Leaches with copper extractions between 48-91 % produced very little elemental sulfur (0—0.4 % of the residue was S°) and consumed acid well below the stoichiometric ratio, (see experiments 1,2,3 in table 1). Chemical activation of the concentrate was done in the standard way, described in the material and methods section. After repeated attempts which were unsuccessful in obtaining a controlled-redox leach that produced elemental sulfur and consumed acid, the standard conditions were altered in search of appropiate conditions for controlled-redox leaching. Changes in aeration level and inoculum age and size were made, the description of these experiments follows. All the results are presented in table 123 and for comparison purposes the results of an elemental sulfur producing leach at 200g/L pulp density are also presented. Inoculum age. Experiments 4,5,6 test different inoculum ages between 4—17 days, for all cases elemental sulfur was not produced, copper extraction was low between 11—44 % and acid was well below the stoichiometric ratio. 355 / 356 Inoculum size. Inoculum size was varied from a low of 10.7 ml inoculum/L medium to a high of 142.8 ml inoculum/L medium, with the standard inoculum being 20 ml/L. Experiments 7,8,9,10,11 and 12 indicate that between 11—55 % of the copper was leached without elemental sulfur being produced, acid was also well below the stoichiometric ratio. Aeration. Various aeration levels were studied between 0.13 vvm and 3.14 wm, in all cases experiments 13,14 and 15 (Table 123) indicate that not only was no elemental sulfur produced but copper leaching was limited by low aeration levels (only 3 % copper was extracted for an aeration level of 0.13 vvm). / 357 Table 123. Controlled—Redox monosized leaches Run AIR INOC Cu S° ACID T Y P E No. FLOW AGE-SIZE vvm days-** % % *** * 1 2.85 6-71.4 91 0.1 0 G 2 2.85 7-71.4 76 0.4 0 G 3 2.85 6-71.4 48 0.0 2.8 N 4 2.85 4-71.4 11 0.0 4.7 G 5 2.85 14-71.4 23 0.0 12.6 G 6 2.85 5-71.4 44 0.0 0 N 7 0.80 7-20.0 35 0.0 0 N 8 0.80 6-10.7 50 0.0 0 N 9 0.80 6-20.0 43 0.0 1.0 N 10 0.80 6-20.0 55 0.0 0 N 11 2.85 17-142.8 38 0.0 0 G 12 3.14 7-71.4 58 0.0 0.6 G 13 3.14 6-71.4 27 0.0 0 G 14 0.13 6-71.4 3 0.0 0 G ~ 0.75 6-37.5 73 1.22 1.38 N * N = Newmont G = Gibraltar ** Inoculum size = ml/1 medium *** moles acid / moles copper ~ Pulp density = 200g/L APPENDIX 8 PUBLICATIONS Particle Size Effects in the Microbiological Leaching of Sulfide Concentrates by Thiobacillus Ferrooxidans M . A . Blancarte-Zurita and R. M . R. Branion Department of Chemical Engineering, University of British Columbia, Vancouver, British Columbia, Canada V6T 1W5 R. W . Lawrence B.C. Research, 3650 Wesbrook Mall, Vancouver, British Columbia, Canada V6S 2L2 Accepted (or publication July 9, 1985 INTRODUCTION The microbiological leaching of metals from sulfide minerals has attracted much attention from researchers and metal producers. An excellent critical review of this topic has been published by Brierley.' One of the foci of such research is the leaching of sulfide mineral concentrates, which is the concern of the present work. Specifically, this communication deals with the effects of leaching on particle size when Cu and Fe are leached from a chalcopyrite concentrate and Zn is leached from a ZnS concentrate by Thiobacillus ferrooxidans. M A T E R I A L S A N D M E T H O D S Microorgan ism The microorganism used in this study was originally isolated by Razzell and Trussell2 from a copper mine located at Britannia, B.C. It has been routinely main-tained on copper concentrate suspended in the 9 K medium, but without the iron described by Silverman and Lundgren.3 Il was determined'' that to minimize the time taken to reach maximum copper concentration and to max-imize the rate of copper extraction an inoculum age of 8 days was optimal. Thus transfers to new media were made every 8 days. Substrate The chalcopyrite concentrate used in this study was a commercial flotation concentrate supplied by New-mont Mines Ltd., Similkameen Div., Princeton, B.C. This concentrate was wet ball milled at 55% solids Biolechnology and Bioengineering. Vol. XXVIII. Pp. 751-755 (19861 © 1986 John Wiley & Sons. Inc. for I h to give 91.8% smaller than 400 Tyler mesh. It was then recovered by filtration and dried at 60°C. The elemental analysis of the concentrate was 27.8% cop-per, 28.0% iron, 31.1% sulfur, and 5.5% insoluble. Culture Technique All experiments were carried out in 250-mL bottom-baffled Erlenmeyer flasks containing 7.5 g of concen-trate and 70 mL of the iron-free 9 K medium of Sil-verman and Lundgren.3 The pH was adjusted to 2.0 with IN H2S0<. The flasks were loosely stoppered with a cotton plug and mounted on a gyratory shaker. The shaker was located in a dark room, maintained at 35°C with a C02-enriched atmosphere. Before inoculation, but after the adjustment of pH to 2, the flasks were held for 24 h to allow for acid consumption by alkaline gangue present in the con-centrate. After this the pH was readjusted to 2.0 and the flasks inoculated with 5 mL of T. ferrooxidans suspension. • Evaporative water losses were made up periodically by adding distilled water. In the sterile control flasks 1 mL of 1:70 v/v phenol solution was added instead of inoculum. Analyses The copper and iron contents of the medium were determined at various times by stopping the shaker, allowing the solids to settle for 10-15 min, then with-drawing a I-mL sample of the supernatant. The metal analyses were done using an atomic absorption spectrophotometer. C C C 0006-3592/86/05075)-OSS04.00 Concentrate Size Fractionation In some of our studies we wanted lo follow the ef-fects of leaching on particle size and vice versa; thus we needed to get fractions of chalcopyrite having dif-ferent particle sizes. The beaker decantation method5 is a simple way of doing this. Concentrate is added lo water in a large beaker or any other suitable container. The slurry is stirred to mix it well then allowed to settle for a predetermined time. The larger particles settle while the smaller ones remain in suspension. The su-pernatant suspension is carefully poured off into an-other beaker and the process repeated using a longer settling time, etc. This fractionation procedure requires the use of a defloculating agent to obtain uniform dispersions. Tween 40 (polyoxyethylene sorbitan monopalmitate) was used. It was subsequently removed from the fractionated particles by solvent extraction with isopropanol in the presence of K 2 C O 3 . 6 Tween 40 inhibited leaching but after removal by the solvent extraction technique no such inhibition was observed. 99% removal of the dis-persant could be achieved.4 Particle Size Ana lys i s Dilute suspensions of the various fractions were made in 10 mL of ethanol using 0.1% aerosol GPG (sodium dioctyl sulfosuccinale in ethanol and water) as a dis-persant in an ultrasonic bath. Microscope slides were prepared from these suspensions and size distributions were measured using an eyepiece micrometer in a mi-croscope to measure the Martin's diameters of 100 particles. After the leaching phase of this study was completed use of a Leitz image analyzer became possible so the unleached fractions were reanalyzed using this instru-ment counting at least 1000 particles. THEORETICAL The following shrinking particle model is proposed to describe the leaching behavior. Since material is being solubilized fromthe particles the particle mass decreases and it is assumed that as leaching occurs the particles get smaller. The rate of particle size decrease is given by — dNIdt, where N is the particle mass and t is time. Assuming that the particles have a constant density p, we have dN ' dt dV dt (I) where V is particle volume, which can be represented by a dp* (2) where dp is the particle diameter and a the particle shape factor. Now substituting eq. (2) into eq. (1) and assuming that the particle shape factor is constant gives for the change in particle mass (3) dN ddp - — = -Ipadp2 — The rate of appearance of Cu in solution can be measured; call this rate r which is expressed in terms of mass x volume - 1 x time"1. To convert the rate of Cu dissolution to a rate of particle dissolution we must divide r by the weight fraction f of Cu in the concentrate. Thus the rate of particle dissolution is rlf. The term — 3a dp1 (d dpldt) represents the dissolution rate of a single particle. The measurement of r is made in a suspension containing many particles per unit vol-ume. Let n be the number of particles per unit volume. Assume that no particles disappear during leaching; then n is a constant and is equal to CJpa dp03, where C 0 is the initial mass of concentrate per unit volume in the leach system and dp0 is the initial particle diameter. Then pa dp, which reduces to - 3C0 dp2 d dp r (4) (5) dpj dt f r, f, and dpa can be measured, and C 0 can be set by the experimenter; thus eq. (5) can be integrated to give particle diameter as a function of time. If necessary the assumption of no total disappear-ance of particles during leaching can be relaxed but then the integration must be done, numerically; r is a function of the amount of mineral surface per unit vol-ume available to the organism and thus a function of dp since surface area is proportional to dp2. For batch leaches r is also a function of time. R E S U L T S A N D D I S C U S S I O N Table 1 provides.data about the initial sizes of the various fractions used in the leaching experiments as well as the mean size of the unfxactionated concentrate. Table I. Properties of particle size fraction's. Mean particle diameter (/mi) Fraction number Microscope Image analyzer I 2 3 4 5 6 Unfractionated concentrate 0.38 1.81 1.67 3.10 5.64 6.74 1.86 0.20 2.68 2.76 3.53 5.98 6.45 2.86 752 BIOTECHNOLOGY ANO BIOENGINEERING. VOL. 28. MAY 19SC s e t • « U. COMMUNICATION TO THE EDITOR 753 13 11 1« -z t -o 0 < 7 (-z Ul « o z S • o u 4 « U J -a. a. o •" o leg* ^fraction | Ofroction 2 * (f o 12 um (13) 100 T I M E (h) Figure 3: Effect of particle size on zinc extraction A Fortran program was written to integrate equation 3 incorporating equations 12 and 13 and recalculating Co if any particles disappeared during the leach period. The logic algorithm for the program is presented in Fig. 6. Comparison of the cumulative weight percent particle size distribution as measured by Torma et al. (ref 8) and as calculated using the shrinking particle model is shown in Fig. 7 using Torma et al's. data after 212 h of leaching. Fig. 8, again based on Torma et al's. data (ref 8) compares calculated and measured distributions after 338 h of leaching. Agreement between the calculated and measured results is quite good thus providing support for the use of 250 equation 3 as a basis for a kinetic model. Work on application of this model to continuous leaching system, the data for which are provided in (ref 9), is in progress. z o < oi H z w o z o o o z ES «.-<»<=> T I M E (h) Figure 4: Effect of particle size on zinc extraction 10 20 30 D I A M E T E R (micrometers) Figure 5: Effect of particle size on zinc extraction, the logarithm of extraction rate plotted against average particle diameter 25L I SIDtC T«*C) I S K w • » U S S O f C t - ' S S TGSI T « E LARGCR TMAN TMAX CALCOLAIC C L A ^ S C O LCACH BCCAM Figure 6. Computer logic diagram ACKNOWLEDGEMENTS Financial support for this work was provided by El Consejo Nacional de Ciencia y Technologia de Mexico, B.C. Research and the Natural Sciences and Engineering Research Council of Canada. N O M E N C L A T U R E a constant b constant Co initial number of particles in suspension per unit volume [Cu] copper concentration in solution dp particle diameter dpQ initial particle diameter E age distribution of particles in reactor outlet stream f mass fraction of metal component in concentrate n panicle mass Q flow rate of suspension into (and out of) reactor r rate of change of metal ion concentration in solution t time t mean residence time in reactor V particle volume Vol volume occupied by suspension in reactor X fractional extraction (solubilization) of metal X mean fractional extraction a panicle shape factor p panicle density / 372 252 P A R T I C L E D I A M E T E R (am) Figure 7: Particle size distribution of zinc concentrate after 212 h of leaching. Given as cumulative weight percent. P A R T I C L E D I A M E T E R (um) Figure 8: Panicle size distribution of zinc concentrate after 338 h of leaching. Given as cumulative weight percent. 253 REFERENCES 1. EJ.Pryor, Mineral Processing, Elsevier Publishing Co. Ltd., Appendix B, (1965). 2. H.P.Silverman, D.G. Lundgren, J. Bacteriol 77, 642 (1959). 3. W.E.Razzell, P.C.Trussell, J. Bacteriol 85, 595 (1963). 4. M.A.Blancarte-Zurita, Effect of Particle Size on the Kinetics of Microbiological Leaching of Chalcopyrite, MASc Thesis, UBC, Vancouver, (1983). 5. A.E.Torma, Microbiological Leaching of a Zinc Sulphide Concentrate, Ph.D. Thesis. UBC, Vancouver, (1970). 6. V.Sanmugasunderam, Kinetic Studies on the Biological Leaching of a Zinc Sulphide Concentrate in Two Stage Continuous Stirred Tank Reactors, Ph.D. Thesis, UBC, Vancouver, (1981). 7. O.Levenspiel, Chemical Reaction Engineering, John Wiley & Sons (1972). 8. A.E.Torma, CC.Walden, D.W.Duncan, and R.M.R.Branion, Biotechnology and Bioengineering 14 (1972) 777. 9. V.Sanmugasunderam, R.M.R.Branion, and D.W.Duncan, 77iZs volume. MICROBIOLOGICAL LEACHING OF CHALCOPYRITE CONCENTRATES BY THIOBACILLUS FERROOXIDANS. A COMPARATIVE STUDY OF A CONVENTIONAL PROCESS AND A CATALYZED PROCESS. M.A BLANCARTE-ZURITA Chemical Engineering, University of British Columbia and BC Research, Vancouver BC Canada R.M.R. BRANION Chemical Engineering, University of British Columbia, Vancouver BC Canada R.W. LAWRENCE. Coastech Research Inc., N. Vancouver BC Canada SUMMARY Limits on pulp density and specific surface area for chalcopyrite concentrate leaching are established and data regarding bacterial growth in batch reactors are discussed for an elemental sulfur producing, acid consuming, catalyzed bioleach. The role of T. ferrooxidans in the catalyzed process is discussed on the basis of a changing growth medium during batch leaches. The effects of aeration, inoculum size and inoculum age are also considered. INTRODUCTION In the biological leaching of chalcopyrite concentrate, copper recovery has been improved through the use of silver as catalyst The catalyzed process and its advantages over the conventional sulfate producing bioleach have been described elsewhere (Bruynesteyn et al, 1983) (Lawrence et al, 1985) (Bruynesteyn et al, 1986). The advantages are quantitative conversion of sulfide sulfur to the elemental form in a low temperature, ambient pressure process and the increase of one-pass copper extraction from chalcopyrite. There is still need to adequately describe limits for the conditions of the process and the role of Thiobacillus ferrooxidans. This work compares the main features of both processes and studies the relationships between iron oxidation, cell growth, elemental sulfur production and copper release for the silver catalyzed leaching of chalcopyrite. MATERIALS AND METHODS The microorganism used for inoculum preparation and the culture techniques are described in Blancarte-Zurita et al, (1986). The two substrates used in these studies; were -400 mesh flotation concentrates supplied by Newmont Mines Ltd. and Gibraltar Mines Ltd., B.C The compositions are, for Newmont concentrate, 27.8% Cu, 28% Fe and 31.1% S, and for Cibraltar concentrate 25.76% Cu, 26.23% Fe and 31.3% S. Specific surface area was measured by air permeability and particle size by image analysis. EQUIPMENT Leach tests were conducted in 100 ml, magnetically agitated beakers, Il baffled glass reactors and 31 baffled, plexiglass tanks equipped with 6 blade turbine impellers. For the 1 & 31 tanks, air enriched in C O 2 (1%) was introduced under the impeller. For catalyzed leaches, the pH of the pulps was controlled at 2.2 by using a pH meter-controller, to activate a peristaltic pump which delivered 12N sulfuric add. The temperature of the reactors was maintained at 35 °C. LEACH TESTS The procedure for preparing 21 of medium for standard catalyzed leaches was as follows. A solution containing 1.16 g Ag2S0 4 in 400 ml of distilled was mixed with 100 ml of a solution containing 12.4 g Na 2S20 35H20. Then 400g of concentrate was added and the resulting suspension mixed for 20 minutes. Next 11 of a solution containing 142.8 g of Cu 2S0 4SH 20 was added to the mixed suspension. After a further 15 minutes of mixing 20 ml of each of the nutrient salt solutions to produce iron free, 9K media (Silverman and Lundgren, 1959) were added. Conventional leaches were carried out in II glass reactors at pulp densities of 107 g concentrate/l; the concentrate was added to 350 ml of iron free 9K media and agitated overnight to allow for gangue acid consumption, the pH was adjusted to 2.0 with 12 N H2SO4 and 25 ml of a 6 day old inoculum were added, air containing 1% C 0 2 was sparged under the turbine agitator at a rate of 2.85 wm (ml air/ml media per minute). ANALYSES Cell growth was estimated b y measuring total organic carbon in samples of pulp with a TOC analyzer. Cu and Fe concentrations were measured using atomic absorption spectrophotometry. Elemental sulfur was measured gravimetrically after extraction with carbon disulfide, and its expressed as grams of elemental sulfur per gram of solids in suspension or per gram of chalcopyrite (g of substrate). Ferric iron was determined by the sulfosalicylic acid colorimetric method. Total iron was measured by the same method following oxidation of the ferrous iron with sodium persulfate; ferrous iron was determined b y difference. THEORY Conventional bioleaching of chalcopyrite can be represented b y the following reaction: 4CuFeS 2 + 170 2 + 4H + = 4 C u 2 + + 4 F e 3 + + 8S0 4 2" + 2H20... (1) Sulfide sulfur is oxidized completely to sulfate and ferrous iron to feme At pH's favoured b y T. ferrooxidans, feme iron tends to hydrolyze and precipitate as jarosite, releasing acid according to equation 2. 3 F e 3 + + 2S0 4 2- + 7H 20 = H 3OFe 3(SO 4) 2(0H) 6 + SH + ... (2) Thus, the overall reaction is a net acid producer 4CuFeS2 + 1702 + 7 1/ 3H20=' , / 3(H30 Fe 3(S0 4) 2(OH) 6] +2 2/ 3 H + + S*/3 S0 4 2"+4Cu 2 +...(3) Catalyzed bioleaching of chalcopyrite has been described by the following set of reactions (Lawrence et al, 1984): OiFeS2+02 + 2H2S04 = CuS04 + FeS04 + 2S° + 2H20...(4) FeS04 + '/,02 + '/2H2S04 = '/2Fe2(S04)3 + '/jh^ O.-.'S) V2Fe2 (S04)3 + 2H20=,/tH2Fe6(0H)12(S04)4 + 5/6H2S04...(6) The overall reaction can be given by: CuFeS 2 + 1'/,02 +1 2/ 3 H 2SO 4 = CuSO 4 + 2S 0 +'/ 2H 20+ V 3 H 2 Fe6(OH)l2(S04)4..<7) RESULTS AND DISCUSSION CATALYST CONCENTRATION Miller and Portillo (1979) found that for non-biological leaching of chalcopyrite the copper extraction rate increased with an increase in initial silver concentration from 17-36 g/kg concentrate. Beyond this level, additional silver did not enhance the rate of reaction or, even reduced it slightly, in leaches with a pulp density of 3 g/1. For catalyzed bioleaching, Bruynesteyn et al, (1983) proposed a range of 0.1-4.0 g Ag/kg concentrate in order to obtain elemental sulfur from the leaching . of chalcopyrite in leaches with pulp densities of 200 g/I. In the culture medium silver was initially found in a complexed form as AgfSjOj^ (Lawrence et al, 1984), resulting from a reaction between Ag 2S0 4 with Na2S2C>3 in the first step of the procedure used to set up leach tests. For an initial concentration of 400 ppm (unpublished data), less than 0.1 ppm of silver remained in solution. Therefore the initial concentration of 0.58 ppm used in our tests will most likely be in complexed form and not in soluble form. Soluble silver was reported to cause inhibition in the growth of T. ferrooxidans at concentrations higher than 1.0 ppm. (Hoffman and Hendrix, 1976). In order to elucidate the relationship between silver concentration, copper extraction rate and mineral surface area, a series of catalyzed bioleaches at pulp densities of 180 g cbncentrate/1 was carried out using monosized fractions of Gibraltar concentrate with specific surface areas (s.s.a.) between 454 - 4132 cm^/g. Figure 1 shows maximum copper extraction rates and S° concentration as function of s.s.a. The maximum attainable rate of copper extraction and the production of S° increased with increments of s.s.a. up to 1390 cm^/g. Subsequent increases of s.s.a. resulted in decreasing extraction rates and decreasing amounts of S° ' for constant amount of catalyst (3.1 x 10'^g Ag 2S0 4/g concentrate). Figure 2 show's moles H2SC>4 consumed/ moles of Cu extracted as a function of s.s.a. A stoichiometric ratio of 1.5-2.0 moles H 2S0 4/mol Cu leached (dashed line), depending on the level of ferrous iron oxidation, was obtained for a s.s.a. of 12S0 cm2/g or less. Table 1 shows data for leaches of the concentrate fraction having a s.s.a.=4132 crn^/g. In these leaches the amount of Ag 2S0 4 added was proportional to the amount of surface area per unit volume. This amount was based on a ratio of 2.9 mg Ag 2S0 4/g of concentrate (standard addition), equivalent to 3 x 10"^ mg Ag 2S0 4/cm 2 for a monosized concentrate fraction of 30.4 um, which produced an elemental sulfur and acid consuming leach. .355-1 r-100 H 1 1 1 1 , 1 1 r-« 1-0 0 500 1000 1500 2000 2500 3000 3500 4000 4500 S.S.A. (cm'/g concentrate) Fig. 1 S and Cu extraction vs s.s.a. - TABLE 1 EFFECTS OF CATALYST/SPECIFIC SURFACE AREA RATIO mg A g 2 S 0 4 m g A g 2 S ° 4 S° moles H 2 S O 4 mg Na 2S 203 mg CuS0 4 /g conc /cm2 /moles Cu /g cone. /g cone. 3.19 7.7x1 cr4 0 0.108 34.1 393 8.7 2x10-3 0 12.96 93 357 14.8 3x10-3 0 17.76 158.5 357 22.2 5x10-3 0 13.93 237.5 357 8.7 2x10-3 0 9.79 93 911 14.8 3x10-3 0 2.5 1S8.5 1822 As shown in Fig. 1, leaching of the finest concentrate fraction (ssa=4132 cm2/g) did not produce any elemental sulfur. To determine if this was due to insufficient silver addition, a series of leaches was carried out in which silver was added in amounts to give varying silver to mineral surface ratios. In this series, the amounts of thiosulfate and copper sulfate were similarly varied. The results are given in Table 1. The results presented in Table 1 show that the increase in s.s.a. corresponding to a decrease of particle size from 30/im to 2.5 fim could not be compensated for by increasing the concentration of silver sulfate or other reagents to maintain a constant ratio of reagent to surface area. This resulted in a change in the leaching characteristics of the concentrate, with some copper being leached but no elemental sulfur produced. Tlie Tesults indicate that, at the pulp / 378 3 1000 1500 2000 2500 3000 3500 S.S.A. (cm'/g c o n c e n t r a t e ) 4000 4500 Fig. 2 Acid consumption vs s.s.a. density of the experiment, there is a limiting S-S.a. above which the catalyzed reaction is not initiated. Once the reaction is initiated extractions of 70-95+% of copper from -400m concentrate were not uncommon, and indicated that all particles would eventually dissolve. This finding contrasts with the effects of increased s.s.a. for non-biological leaching (Miller and Portillo, 1972) where increased rates of reaction were found when the s.s.a. increased, when the concentrate was ground, and the partide size went from 50% minus 20tim to 50% minus 0.S;zm. This difference suggests that the mechanisms for these two processes are different and points to the critical nature of s.s.a. to establish the catalyzed bioleach reaction. This finding also allows some postulation regarding the mechanism of sulfur production. Sulfur is more likely being formed in the catalyzed leach at the mineral surface than within the bacterial cell since the later should not be affected by partide size. COPPER EXTRACTION AND CELL GROWTH. figure 3 is a plot of copper concentration and cell growth (as organic carbon) for conventional leaching of -400 mesh Gibraltar concentrate. Copper extraction was 34%. Cell growth and copper release are simultaneous processes and cell growth stops after copper leaching stops. Similar results were obtained for Newmont concentrate (Blancarte-Zurita, 1983). figure 4 shows copper extraction and cell growth as a function of time for a -400 mesh Newmont catalyzed leach; copper extraction was 73%. As expected, the relationship between bacterial growth and metal extraction is different between the catalyzed process and the conventional process. Lawrence et al, (1984) have suggested that the lower bacterial populations present in the catalyzed leach are due to a reduction in the energy obtainable from the oxidation of sulfide when it stops at the elemental sulfur state. Figure 4 indicates that growth of T. ferrooxidans continues after copper extraction stops for the catalyzed bioleach. / 379 \ be a o o O 100 200 300 400 500 600 700 800 900' Time ( h o u r s ) Fig- 3 Cell growth in conventional leaching bo a o o h 30 O. O, o O 200 Time ( h o u r s ) Fig. 4 Cell growth in catalyzed leaching 12 bo / -r 100 • C o n r e n t i o n a l E C a t a l y z e d o C o n t r o l « \ o ! v -/ 200 300 400 S00 Time (hours) Fig. 5 Ferric iron profiles 600 —I— roo A longer experiment was necessary to investigate what happens to the cells after the catalyzed leaching of copper stops. At this point in the leach, the conditions were: a redox potential around 500 mV, some ferrous iron in solution, the appearance of elemental sulfur and copper sulfate resulting from the oxidation of the sulfide, as well as some secondary products like jarosites and ferric sulfate. The remaining sulfide (if any) could be available as an energy source for continuing bacterial growth. Other possibilities for obtaining energy are the oxidation of ferrous iron or elemental sulfur. These two alternatives are considered in the following two sections. FERROUS I FERRIC PROFILES FOR CATALYZED LEACHES. According to the stoichiometry of reactions 3-6, the net sulfuric acid consumed by the leaching of copper is a function of the level of oxidation of ferrous sulfate and hydrolysis of ferric sulfate. The net acid consumption due to these reactions can be estimated if the amounts of ferrous and ferric iron in solution are known. Rg 5 shows the percentage of iron in solution as feme vs time for a catalyzed, a non-inoculated catalyzed, and a conventional leach. The profiles of the conventional and catalyzed leaches show that the different redox conditions in solution result in different profiles of iron oxidation and hydrolysis. In the catalyzed leach, low redox values prevail throught most of the leach. In the conventional leach redox values are higher. Thus, jarosites resulting from hydrolysis of ferric iron are being produced in larger amounts and at an earlier stage lor the conventional leach causing the leaching of copper to stop due to the accumulation of precipitate on the chalcopyrite surface. SEM (scanning electron microscope) studies support this theory (Blancarte-Zurita, 1983). Fig 6 presents % iron and copper leached as functions of time for conventional and catalyzed leaches. Cell concentrations for catalyzed and conventional leaches are plotted in figs 7 and 8, together with ferric iron concentrations expressed as a % of the soluble iron vs time. / 381 too T3 a, a, o U 60 • 40 -°Co coarcatioail «Fe coounlioo'l • Co C l t l l jK l f HFc c«t«ljK(i 0 •a 100 200 Time (hours) 400 Fig. 6 Metal extractions a o, A ^ d 6 a K o a o 3 K 100 90 80 70 60 S0-\ 40 30 H 20 10 0 0 Fc"" H T . O . C / 8 / 40 80 120 160 200 240 Time (hours) 280 320 Fig. 7 Cell growth and ferric iron concentration (catalyzed) no S o, 1 1 0 -. 1 0 0 -O 6 » 0 -•a 8 0 -a i ro-•o ** «o--o " K 5 0 -o a 4 0 -o i~. *** 3 0 -o 3 2 0 -*o