December 12, 2006

Continuous Revolving Barrel Bioreactor Tailored to the Bioleaching Microorganisms

By Loi, G; Rossi, A; Trois, P; Rossi, G


Until recently, two types of bioreactors were used for bioleaching: the stirred tank reactor (STR), borrowed from chemical engineering, and the so-called Pachuca tank airlift reactor (ALR), borrowed from hydrometallurgy. These reactors perform very satisfactorily in the fields for which they were developed, but they have a number of major shortcomings when used for bioleaching. These drawbacks stem from the stressful conditions that the mechanics and fluid dynamics of STRs and ALRs create for the microorganisms. A rotating barrel batch bioreactor, designed, built and tested in the authors' laboratory in the 1990s, was considered to be the most attractive option for incorporating into a continuous machine to overcome these drawbacks. This paper describes a novel type of continuous bioreactor that was tailored to suit certain physiological requirements of bioleaching microorganisms established by recent microbiological research. The original batch reactor was modified to enable integration with the continuous machine. With the continuous bioreactor, 4 g/L of net volume per hour of a 40% solids mineral suspension, consisting of a sulfide flotation concentrate that was predominantly pyrite, was completely leached in 24 hours. The final leach liquor contained concentrations of as much as 53 g/ L of ferric ion, complexed with fragments of extracellular polymeric substance (EPS). Just three units arranged in series are sufficient to ensure complete leaching of the feed. The improved performance of this innovative bioreactor vis--vis STRs and ALRs, is interpreted on the grounds of experimental evidence and of the most recent advances in bacterial physiology.

Key words: Biotechnology, Bioleaching, Extracellular polymeric substance (EPS), Mineral sulfides, Bioreactors

State of the art of bioleaching in bioreactors Current knowledge of bioleaching process. The mechanism by which microorganisms enhance metal sulfides oxidation and leaching kinetics is still surrounded by controversy (Rossi, 1990). One school of thought argues that the bacterial action is a consequence of a physicochemical alteration of the mineral surface produced by physical contact with the bacteria, the so-called "direct attack." The other school maintains that bioleaching is simply a chemical process, the dominating factor being the presence in the mineral suspension of Fe(III), which in an acidic environment oxidizes the mineral crystals. The bacteria should serve merely to enhance the oxidation kinetics of the Fe(II) that is formed continuously during the mineral oxidation reactions. There does, however, appear to be consensus about the current lack of evidence as to the existence of an enzyme that justifies the theory of the direct attack.

A better understanding of the bacterial glycocalyx has no doubt fuelled this as yet unresolved controversy (Rossi, 1990). In effect, recent research has thrown new light on this part of the microbial cell. It has been demonstrated that the glycocalyx in Acidithiobacillus ferrooxidans and Leptospirillum ferrooxidans consists of a mechanically weak EPS believed to form a sort of "bridge" between the bacterium and the crystal surface. It creates a microenvironment where most of the bioleaching process takes place. It should be stressed, however, that this model does not by any means rule out the important contribution of ferric iron (Gehrke, 1998). It is highly significant that the composition of the EPS is mineral specific, as recent research has confirmed (Gehrke, 1998; Sand et al., 2001; Rohwerder et al., 2003; Sharma and Hanumantha Rao, 2005). From the point of view of the leaching process, undeniably the integrity of EPS is of paramount importance for the effectiveness of the bacterial action.

Requisites for the effectiveness of the bioleaching process. Bioleaching processes take place in three-phase systems consisting of:

* an aqueous phase: i.e., a solution of salts providing nutrients for the microflora;

* a solid phase: composed of the finely ground ore containing a mixture of minor amounts of waste rock ("gangue") and metal values combined with sulfur to form metal sulfides; and

* a gaseous phase: consisting of atmospheric oxygen and carbon dioxide.

The aqueous phase is the suspending medium where several elementary processes occur. The most important of these are:

* biological: the growth of microorganisms and the interactions of the various microbial strains present;

* transport: the maximum probability of the microorganisms coming into contact with solid particles and nutrient molecules; the maximum probabiliy of the solid particles coming into contact with chemically active molecules; the rapid removal of the microbial metabolic products and the dissolved species from the mineral particles surface; and the uniform distribution and effective dissolution of oxygen and carbon dioxide; and

* chemical and biochemical: adaptation to the toxic elements released by the mineral substrate and to changing pH and Eh conditions of the aqueous phase.

The solid sulfide mineral phase provides the energy necessary for microbial biosynthesis, i.e., for microbial growth, and continuously releases metal and sulfur ions in the oxidized form (Rohwerder et al., 2003). The gaseous phase supplies the oxygen required, as final electron acceptor, for the oxidation processes as well as the carbon dioxide required by the microflora for biosynthesis (Rossi, 1990; Rawlings, 2002).

Effective mixing is essential for all matter transport processes. Bioleaching attains its maximum effectiveness when all these processes are optimized in the bioreactor.

Performance of conventional bioreactors. For the bioleaching process to perform effectively the reactor needs to meet all of the above requirements, which consist essentially in adequate suspension aeration and mixing.

The reactors still used in most biohydrometallurgical laboratories and in continuously operating commercial plants are the stirred tank reactor (STR) and the air-lift (ALR) or pneumatic reactor (PR) of the Pachuca type. Aeration in these reactors depends on suspension agitation, and "flooding" may cause some limitations, while very strong agitation results in cell damage. Details on the engineering and biotechnological features of these machines are provided in Acevedo (2000) and in Rossi, G. (2001).

Performance limitations of both types of bioreactors are well documented and date back to the early days of biohydrometallurgy, be they used in the laboratory as batch machines (Torma et al., 1970; Sakaguchi et al., 1976) or in continuous-flow commercial operations (Dew, 1993; van Answegen, 1993; Loi et al., 1994; Nicholson et al., 1994; Bell and Quan, 1997; Dew et al., 1997 ;Stephenson and Kelson, 1997; van Ans wegen and Marais, 2001). This limitation is related to the pronounced deterioration in bioleaching efficiency for solids concentrations (denned as percent ratio of mass of dry solids to mass of suspension it forms) of 16% to 20%. The findings of a recent study (Gentile et al., 2003) confirmed this observation.

Two of the most serious implications of this drawback are the considerable tank volume required per unit mass of dry solid throughput and the power requirements per kilogram of dry solid. For a residence time of six days, which is customary for the kinetics of commercial bioleaching processes, it can easily be shown that for a sphalerite concentrate (density 4 g/cm^sup 3^) and a raw coal (density 1.5 g/cm^sup 3^) a 50% decrease in solids concentration, e.g., from 40% to 20%, entails an increase in total tank volume of 143% for the sphalerite concentrate and of 115% for the raw coal for the continuous-flow stirred tank reactor (CFSTR).

Proposed explanations for the shortcomings of conventional bioreactors. Several explanations have been advanced for the shortcomings of the conventional bioreactors described above. Some workers (Torma et al., 1970;Beyeretal., 1986) claimed, although without being able to provide theoretical evidence at the time, that beyond a solids concentration of around 20%, the frequency of interparticle collision and friction increases significantly, causing the violent detachment of the microbial cells adhering to the mineral surface and damaging them. It is significant, in this regard, that Nienow and Conti (1978) demonstrated that, for suspension concentrations of more than 20%, the velocity of interparticle abrasion in an STR is proportional to the solids concentration squared. The hypothesis advanced by Torma and Beyer for bioleaching with A. ferrooxidans was also borne out indirectly by Ragusa (1991), who produced evidence that they appear to lose viability or, at least the ability to enhance leaching, when they become detached from the surface.

Other workers (Lee et al., 1982; Albal et al., 1983; Ledacoviczetal., 1983;Brehmetal., 1985;Liuetal., 1989)observed that the decrease in the mass transfer capacity coefficient, k^sub L^ a (where k^sub L^ is the mass transfer coefficient and a is the gas/ liquid interfacial area per unit liquid volume), of the gaseous mixture (air+CO2) required for microflora biosynthesis, produced by increasing solids concentrations to mor\e than about 20% (by volume), contributes significantly to the decline in bioleaching efficiency.

The discovery that the microorganisms adhere to the sulfide mineral surface by means of the EPS and that EPS plays a role in the oxidation and solubilization of the mineral constituents has added another piece to the puzzle. From a practical processengineering standpoint, five properties of EPS's are particularly important (Gehrke, 1998; Sand and Gehrke, 1999; Sand et al., 2001; Rohwerder et al., 2003):

* they mediate the contact;

* they form a special, enlarged reaction space for the microbial cells;

* should the microorganism be deprived, for any reason, of the EPS, it loses its ability to attach and, thus, its "catalytic" action;

* the ability of the cells to replenish any loss in glycocalyx material (forinstance, owing to mechanical action) within the space of a few hours; and

* their chemical composition and surface activity are influenced by the substrate.

From a practical viewpoint, the third property appears to be significant. In STRs and ALRs with high suspended-solids concentration, abrasion may seriously affect microbial action by tearing the EPS from the cells. These findings fully confirm the earlier observations. Therefore, a consensus on the important role played by the EPS is of paramount importance as it provides vital information for proper bioreactor design.

Revolving-drum bioreactor. The main objective of the work reported in this paper was to design a continuous bioreactor that overcame the shortcomings of STRs and ALRs, such as the limitation of solids concentration in the mineral suspension. Other objectives included increasing throughput capacity, better and easier regulation of the atmospheric oxygen supply irrespective of suspension agitation and mixing, gentle mixing, pulp transfer without the need for interstage pumps so as to avoid shear stresses and, lastly, flexibility of the bioreactor operation, which can readily adapt to variations in feed composition, solids concentration and throughput.

Based on experience, the authors conceived, designed and developed an innovative machine that was better tailored to the following bioleaching requirements: higher solids concentration, thorough mixing, an easily adjustable k^sub L^a for oxygen mass transfer from atmosphere to suspension so as not to inhibit bacterial growth, and very gentle suspension agitation to practically eliminate shear stresses that would prevent the cells from adhering to the mineral surfaces. A first prototype, consisting of a batch machine (the batch Biorotor) was tested in the early 1990s with very encouraging results (Loi et al., 1995).

In 1994, a three-stage, continuously operating bench-scale bioreactor was set up and thoroughly tested during the following years. The bioreactor was recently granted a patent (Rossi et al., 2005).

Each stage consists of a cylindrical barrel basically similar in design to the batch machine mentioned above (Loi et al., 1995). A spur split gear is located near the center bolted to flanges on the outer shell. The gears of the three barrels are driven by spur pinions all keyed to a shaft that is direct connected to a variable speed reducer coupled to an electric motor. The three barrels thus revolve with the same speed, and in this sense they form one single system consisting of a cascade of three units. Heads are machined and drilled to fit barrel flanges.

All the heads on the feed side are fitted with a scoop feeder. The scoop feeders pick up the suspension flowing into the respective boxes and convey it to the barrels. This innovative feature ensures transfer from one barrel to the next without the need for pumps, thus avoiding any additional stress on the microbial cells. The center of each head on the feed side is fitted with a roller bearing and a double air seal provided with a stainless steel pipe through which the gas mixture can be blown when required. The discharge pipe at the opposite end of each barrel also acts as an exhaust outlet, thus preventing undesired pressure build-up inside the barrel. The net useful volume of each barrel slightly exceeds a half-barrel geometric volume.

The suspension circulation per re volution corresponds to a volume slightly exceeding 1.5 times the useful barrel volume. Thus, at 10 rpm (about 1 rad/s) this corresponds to a circulation, inside each barrel, 15 times the useful barrel volume.


All tests were carried out in the three-stage bioreactor.

Materials and methods.

The mineral: The case history of the Fairview Plant, located in South Africa, is well documented (Rossi, 1990; van Answegen and Marais, 2001). This mineral concentrate was chosen because it is well characterized and currently used in a commercial CFSTR operation of proven performance. This was considered the best procedure for properly evaluating the advantages of the bioreactor over the CFSTR. For this reason, it was decided to purchase 100 kg of the concentrate processed at the Fairview flotation plant.

The mineralogical components of the concentrate, determined by X- ray diffractometry, are in order of abundance: quartz, pyrite, arsenopyrite and illite. The concentrate arrived at the University of Cagliari in polythene bags and was very moist. It was dried in a thermostated oven at 30C and dry ground in a ceramic ball mill to - 75 m. The chemical composition (weight percentages) is 22.12% Fe, 4.39% As, 1.23% Mg, 0.85% Ca, 0.289% Ni, 0.35% Al and 33.47% SiO^sub 2^. The mineralogical composition (weight percentages) is 31 % quartz, 28.1% pyrite, 17.8% clinochlore, 10.2% muscovite, 9.1% arsenopyrite and 2.6% gypsum (Lichen, 2003).

The inoculum: A mixed microflora, consisting of A.ferrooxidans, L. ferrooxidans and A. thiooxidans strains, isolated from the drainage of the complex sulfide ores mine of Fenice Capanne, Tuscany, Italy, conventionally called "FC" (Rossi, 1971 ) was adapted to the Fairview concentrate in a conventional 2-dm^sup 3^ Pyrex glass STR fitted with four baffles, a Rushtontype turbine and air sparger located beneath the turbine. The suspension consisted of 1.5 dm^sup 3^ 9K medium (Silverman and Lundgren, 1959) without iron sulfate, pH 2.40, and 25 g of Fairview concentrate. The A. ferrooxidans strain of this microflora was found to be well suited for adaptation to metals, owing to its DNA characteristics (Buonfiglio, 1992; Polidoro, 1992). Adaptation was very slow, requiring 15 transfers, and in accordance with the well-documented procedure (Rossi, 1990) the strain was considered successfully adapted when the exponential branches of the final two growth curves exhibited the same maximum slope. The supernatant from the tank containing the adapted microflora was centrifuged in a Sorvall SS-3 bench supercentrifuge at 5,400 g. The pellet so obtained was dispersed in 1,000 cm^sup 3^ of 9K medium without iron sulfate at pH 1.5, and the suspension was then poured into the feed box of the first stage. No further inocula were needed for the entire duration of machine operation.

Monitoring: pH and Eh of pulp samples were determined daily using a conventional potentiometer.

Power consumption was also recorded on a daily basis, simply to check that the operation was running smoothly, as this parameter is insignificant in the comparison with commercial machines mainly because of the relatively high power losses caused by the speed variator and the gear.

Mode of bioreactor operation: The three-stage facility consisted of three equal-size barrels arranged in series and numbered 1, 2 and 3 from the feed to the discharge end (Fig. 1). Each barrel is 540- mm long with an inner diameter of 280 mm and with 12 equally spaced lifters. The rotation speed was set in the 1 to 1.5 rad/s range corresponding to a k^sub L^a value in water of from 100 to 320 h^sup -1^ (Loi et al., 1995).

To assess the flexibility of this bioreactor system, three series of six runs each were performed for throughputs of 2,3 and 4 g/ dm^sup 3^ of net volume (3 x 15 = 45 dm^sup 3^) per hour, corresponding to throughputs of 90, 135 and 180 g/h, respectively. For all the runs, the flow rate of suspending liquid (9K medium (Silverman and Lundgren, 1959) without ferrous sulfate, pH 2.40) was regulated so as to form a 40% solids suspension.

Analytical techniques: This paper is concerned with the analytical results. Details of the techniques used along with specific interpretation procedures are to be published elsewhere.

Figure 1 - The three-stage bioreactor with three barrels arranged in series.

The following analyses were performed:

* Chemical analysis of minerals and solutions: The treated ferrous and ferric iron, complexed by the EPS fragments in the leach liquors, were determined using the α-α'dypiridil colorimetric method (Snell and Snell, 1959). The solid phases (feed and bioleaching residues) were investigated by means of X-ray diffractometry, and their composition was determined by quantitative chemical analysis using standard methods.

* Mineralogical analysis by X-ray diffractometry: Diffractometric analysis of granular samples was performed using a θθ geometry diffractometer manufactured by Seifert, Model 3000TT, with Bragg-Brentano focusing geometry and Cu anode and W cathode X-ray tube, 50 kV accelerating voltage and 35 mA current.

* Surface analysis by X-ray Electron Spectroscopy (XPS): Surface analysis of samples was carried out with an ESCALAB MKII spectrometer manufactured by Vacuum Generator Ltd., East Grinstead, U.K., mounting the ESCALAB 200 analyzer. A detailed description of the spectrometer and the calibration procedure are given in (Rossi, A., 2001).

* Chemical analysis of bioleaching products by inductively coupled plasma (ICP) spectroscopy: Analysis of the cations released by the Fairview concentrate after chemical attack was performed with an ICP instrument model ASH ATOM-SCAN 25 manufactured by Thermo- Jarrell.

* Biochem\ical analysis of liquid phases (Harneit, 2005): The samples were homogenized for three minutes at 12,000 rpm with an Ultra Turrax machine to separate cells from mineral particles and to strip off EPS from cells and particles. The homogenized suspension was then centrifuged at 100 g for 5 minutes to separate the solids from the cells. Consecutively, the supernatant suspension was centrifuged at 10,000 g to sediment the remaining cells. The supernatant containing the cell-free EPS was collected separately and subjected to freeze-drying for EPS-winning. The isolation, partial purification and analysis of EPS were carried out according to the procedure described by Gehrke (1998). The EPS of 10(10) cells grown on pyrite was determined (Gehrke, 1998) to be 2,760301 g and can be accurately characterized by its sugar content. Therefore, the EPS of the samples was analyzed for sugar content.

* Sampling and statistical analysis: Two days after steady-state operation had been attained for a given feed throughput, sampling of the suspension flowing out of each stage commenced for operation monitoring of each six-day run. Samples were collected every 24 hours, making a total of six samples for each feed throughput. Each sample was centrifuged at 230 to 240 g to separate the residual solids from the liquid phase and, after pH and Eh measurement of the supernatant, part of the solids was assayed. The remaining suspension was stored in a freezer box at -10C. Results of each six- day run are plotted as arithmetic mean values. Standard deviations (SD) are plotted as error bars. The maximum standard deviations were 15% for iron plots, 2.7% for Eh plots and 3.33% for pH plots.

Results of continuous operation tests. Figures 2 through 4 show the pH and Eh plots vs. stage effluent, while Figs. 5 through 7 show the total iron, Fe(III) and Fe(II) plots vs. stage effluent for the same runs. Suspension pH (mean values) ranged from 1.30 to 1.41 in the first-stage effluent, from 1.15 to 1.18 in the second-stage effluent and from 0.98 to 1.06 in the final-stage effluent.

The Eh plots showed practically the same trend, with the exception of the lowest throughput, and the mean values did not vary substantially from the first to the third stage. Significantly, the final Eh ranged from 560 to 570 mV, much lower than the values (about 700 mV) obtained in pure pyrite bioleaching tests carried out in a batch STR in the authors' laboratory.

Figure 2 - Graph of the pH and Eh values of the suspensions flowing out of the bioreactor stages; throughput: 2 g (dm3 hour)^sup -1^ solids; suspension concentration: 40%.

Figure 3 - Graph of the pH and Eh values of the suspensions flowing out of the bioreactor stages; throughput: 3 g (dm3 hour)^sup -1^ solids; suspension concentration: 40%.

Figure 4 - Graph of the pH and Eh values of the suspensions flowing out of the bioreactor stages; throughput: 4 g (dm3 hour)^sup -1^ solids; suspension concentration: 40%.

Complete bioleaching was always achieved, the solids in the effluent of the final stage consisting of quartz and illite with only traces of iron and arsenic.

The composition of the effluent liquor from the final stage, once steady-state operation had been attained, was rather unusual and especially interesting. Ferric ion concentrations were of the same order of magnitude as shown by Figs. 5, 6 and 7, with their mean values ranging from 38 g/dm^sup 3^ to 53 g/dm^sup 3^. These are far higher than the values compatible with the electrochemistry of aqueous solution having that pH (Sand et al., 2001).

The daily power consumption was 9 0.8 kWh, irrespective of the rotation speed range considered and of throughput, demonstrating that most of the power is absorbed by the mechanical losses of the speed variator and gear that were necessarily oversized with respect to machine scale.

For the throughput range tested, the performance of the bioreactor system remained apparently unchanged. This result differs somewhat from the values observed in the final supernatant of the STR where adaptation had been carried out. After 85% pyrite and 89% arsenopyrite bioleaching, a pH of 1.83, an Eh of 690 and a total iron in solution of 2.4 g/dm^sup 3^ were determined, substantially less than the 3.57 g/dm^sup 3^ expected for bioleaching efficiency comparable to that of the three-stage bioreactor. In effect, some precipitated iron was observed in the solid residue. In addition, the supernatant of the STR was light yellow-green in color and transparent, whereas the supernatant of the suspension flowing out of the bioreactor units was dark brown and opaque.

Analysis of the liquid phase revealed very large amounts of sugar components, ranging from 58 to 76 mg/cm^sup 3^, attributable to the EPS of A. ferrooxidans and L. ferrooxidans (Gehrke, 1998), corresponding to a number of bacteria in the order of magnitude of 1013 cells per cm^sup 3^ (Harneit, 2005).

XPS analyses of the solid phase revealed the presence of Fe, As, S, C, K, Mg, Ca, Si, Al and O on the surface of the feed particles. The solids in the first-stage effluent already exhibited weaker signals for S, As and Fe, which completely disappeared in the final stage where, on the other hand the N1 s signal and a very strong O1s signal were detected. Particularly significant are the data for oxygen. XPS analysis shows a high binding energy peak (534.5 eV) accounting for 37% of total signal. The other two components at 531.6 0.2 eV and at 532.5 eV can be assigned to the oxygen involved in hydroxide-type bonds with a metallic ion and/or to As oxide, where present, and to the oxygen of the Si-O-Si bridges, respectively (Moulder et al., 1995). As far as the present research is concerned, it should be noted that signals with such a high binding energy were only determined in polyphosphate molecules where phosphor atoms are bridged by oxygen atoms forming polymer-type chains.

Discussion and conclusions

For throughputs ranging from 2 to 4 g of dry solids (dm^sup 3^ net volume per hour), the three-stage bioreactor with 40% solids suspension outperformed conventional bioreactors, at least in terms of solids concentration, and its potential capacity is probably even greater because under all test conditions the Fairview mine flotation concentrate was completely bioleached. It is well documented (Rossi, 1990) that in CFSTRs residence time of part of the feed is shorter than that required to complete the reaction, but this drawback is overcome in commercial plants using up to six CFSTRs arranged in series (Dew, 1997; van Answegen and Marais, 2001; Brown et al., 1994; Nicholson et al., 1994). The continuous flow bioreactor system apparently requires a smaller number of units arranged in series. In effect, the capacity of most commercial CFSTRs described in the literature, always consisting of several units arranged in series and employed for bioleaching with 20% solid concentration (Brown et al., 1994;Nicholsonetal., 1994;Dew, 1997; van Answegen and Marais, 2001), is lower than 2.0 g dry solids (dm3 net volume)/h. Very similar pH and Eh trends were observed for the higher feed rates (Figs. 3 and 4) that provide, for the same net volume, larger amounts of iron and sulfur. The low initial Eh value at the lowest feed rate (Fig. 2) is to be attributed to the relatively small solids throughput.

The plots shown in Figs. 5 to 7 provide some significant indications about the characteristics of the process taking place in the continuous bioreactor. Already in the first stage, ferrous iron is lower than 20 g/dm^sup 3^ and shows a similar trend under all operating conditions. The high iron concentrations in the liquid effluent from the first stage shown in Fig. 7 may be attributed to the high solids throughput. Ferric iron can be considered the real novelty of reactor bioleaching, concentrations exhibiting an increasing trend from the first through to the final stage.

On the other hand, a specific procedure had to be devised for the analytical determinations, owing to the fact that iron was complexed by the EPS (Gehrke, 1998) present in the suspension.

These observations agree with the high EPS concentrations detected and justify the following interpretation of three-barrel bioreactor performance.

The suspension in the revolving barrels undergoes a very thorough mixing and because of the very gentle movement, shear stresses are practically absent. In spite of the relatively high solids concentration, this gentle mixing action apparently does not produce any significant abrasion effects on the particle surfaces and the atmospheric oxygen availability, that can be easily adapted to requirements simply by adjusting rotation speed to attain the required k^sub L^a, favors the growth of the microflora.

The absence of shear stresses and of abrasion effects creates a very favorable environment, in which microbial cells should be able to adhere undisturbed to the solid surfaces by means of the EPS.

The recently proposed "indirect-attack" model (Gehrke, 1998; Gehrke et al., 1998; Gehrke et al., 2001) suggests that the EPS bridges formed between cells and mineral surfaces create microenvironments where concentrations of ferric ions of as much as several tens of g/dm^sup 3^ can exist, thus pronouncedly enhancing mineral dissolution kinetics.

In conclusion, the bioreactor system design ensures that the suspended solids are handled compatibly with the physiological requirements of the microbial population and the device is thus able to fully exploit its potential for maximizing bioleaching efficiency using suspended solids concentrations and throughputs per unit net volume that appear never to have been attained so far with conventional bioreactors.


The invaluable suggestions of Prof. Wolfgang Sand (Universities of Hamburg and Duisburg) and the cooperation of Dr. Harneit, who analyzed the EPS contained in the effluent, are gratefully acknowledged. The au\thors wish to express their appreciation to Dr. Cristina Trois of the University of Natal, Durban, South Africa, for her kind assistance with the purchase and shipment of the Fairview Mine concentrate and to Ms. Cristina Licheri and Ms. Marzia Fantauzzi for the analytical work painstakingly carried out within their graduate thesis project. The authors also acknowledge the management of Fairview Mine for granting permission to purchase the concentrate. The authors are indebted to Mr. Migoni, Manager of the Migoni & Co., a mechanical engineering firm, and to Mr. Marco Loi, Manager of Centroplast Inc., for their patient and helpful cooperation in constructing and assembling the bioreactor system prototype. This work was carried out with the financial support of the Italian Ministry for Education, Universities and Research within the framework of the Research Project of National Importance "Unconventional Bioreactors."

Figure 5 - Graph of the total-, ironlll- and iron II concentrations in the suspensions flowing out of the bioreactor stages; throughput: 2 g (dm^sup 3^ hour)^sup -1^ solids; suspension concentration: 40%.

Figure 6 - Graphs of the total-, ironlll- and iron II concentrations in the suspensions flowing out of the bioreactor stages; throughput: 3 g (dm^sup 3^ hour)^sup -1^ solids; suspension concentration: 40%.

Figure 7 - Graph of the total-, ironlll- and iron II concentrations in the suspensions flowing out of the bioreactor stages; throughput: 4 g (dm^sup 3^ hour)^sup -1^ solids; suspension concentration: 40%.

Paper number 05-315. Original manuscript submitted May 2005. Revised manuscript received and accepted for publication January 2006. Discussion of this peer-reviewed and approved paper is invited and must be submitted to SME Publications Dept prior to May. 31, 2007. Copyright 2006, Society for Mining, Metallurgy, and Exploration, Inc.


Acevedo, F, 2000, "The use of reactors in biomining processes," Electronic J. Biotechnol., Vol. 3, No. 3, pp. 1-11.

Albal, R.S., Shah, Y.T., and Schumpe, A., 1983, "Mass transfer in multiphase agitated contactors," The Chem. Eng. J., Vol. 27, pp. 61- 80.

Bell, N., and Quan, L., 1997, "The application of BacTech (Australia) Ltd. technology for processing refractory gold ores at Youanmi Gold Mines, " in Biomine '97 - "Biotechnology Comes of Age, " International Biohydrometallurgical Symposium IBS'97, Glenside, S.A., Australian Mineral Foundation, Sydney, Australia pp.M2.3.1- M2.3.9.

Beyer, M., Ebner, H.G., and Klein, J., 1986, "Influence of pulp density and bioreactor design on microbial desulphurization of coal," Appl. Microbiol. Biotechnol., Vol. 24, pp. 342-346.

Brehm, A., Oguz, H., and Kisakuerek, B., 1985, "Gas-Liquid mass transfer data in a three phase stirred vessel" in 5th European Conference on MIXING, Wurzburg, West Germany, (Bedford, England, Bhra Fluid Engineering Centre), pp. 419-425.

Brown, Allan R.G., Irvine, W., and Odd, Paul, A.R., 1994, "Bioleaching-Wiluna operating experience," In Biomine '94, Australian Mineral Foundation, Glenside, SA, Paper XVI, pp. 16.1- 16.8

Buonfiglio, V., 1992, "Ossidazione dello solfo in Thiobacillus ferrooxidans "(Sulphur Oxidation in Thiobacillus ferrooxidans), Ph.D. Thesis, in Italian, Istituto di Microbiologia - Universit di Roma "La Sapienza," 92 pp.

Dew, D.W., 1993, "Comparison of performance for continuous biooxidation of refractory gold ore flotation concentrates " in Biohydrometallurgical Processing, J.V. Jerez, H. Toledo, J.V. Wiertz, eds., University of Santiago, Chile, Vol. 1, pp. 239-251.

Dew, D.W., Lawson, E.N., and Broadhurst, J.L., 1997, "The Biox process for gold-bearing ores or concentrates," in Biomining, Rawlings, D.E., ed., Springer, Berlin, pp. 45-80.

Gehrke, T, 1998, " Bedeutung extrazellulrer PolymererSubstanzen von Thiobacillus ferrooxidans fr die Mikrobielle Besiedelung und Laugung von Pyrit und Schwefel," in German, Dissertation, University of Hamburg, Hamburg, 100 pp.

Gehrke, T, Telegdi, J., Thierry, D., and Sand, W., 1998, "Importance of extracellular polymeric substances from Thiobacillus ferroxidanstor bioleaching, " Applied and Environmental Microbiology, Vol. 64, No. 7, pp. 2743-2747.

Gehrke, T, Hallmann, R., Kinzler, K., and Sand, W., 2001, "The EPS of Acidithiobacillus ferrooxidans - a model for structure- function relationships of attached bacteria and their physiology," Water Science and Technology, Vol. 43, No.6, pp. 159-167.

Gentile, F, Oleschko, H., Veverka, P., Machon, V, Paglianti, A., Bujalski, W., Etchells, A.W., and Nienow, A.W., 2003, "Some effects of particle wettability in agitated solid-gas-liquid systems: Gas- liquid mass transfer and the dispersion of floating solids, " in 6th International Conference on Gas-Liquid and Gas-Liquid -Solid Reactor Engineering, Aug. 17-20, 2003.

Harneit, K., 2005, personal written communications.

Ledacovicz, S., Kokuun, R., Brehm, A., and Dekwer, W.-D., 1983, "Einfluss suspendierter inerten Partikeln auf die Stoffbergang- Charakteristik eines Rhrkessels," Che. Ing. Tech., Vol. 55, No. 8, pp. 648-649.

Lee, J.C., Ali, S.S., and Tasakorn, P., 1982, "Influence of suspended solids on gas-liquid mass transfer in an agitated tank, " in Fourth European Conference on Mixing. BHHA Fluid Engineering, pp. 399-415.

Licheri, C., 2003, "Analisi e caratterizzazione di solfuri minerali misti di ferro e arsenico sottoposti a biolisciviazione," (Analysis and characterization of bioleached mixed iron and arsenic mineral sulfides), Thesis for the degree in chemistry, in Italian, Department of Inorganic and Analytical Chemistry, University of Cagliari, 97 pp.

Liu, M.S., Branion, R.M.R., and Duncan, D.W., 1989, "Oxygen transfer to Thiobacillus ferrooxidans cultures," in Biohydrometallurgy, Proceedings of the International Symposium, Warwick 1987, Norris, Paul, R., and Kelly, Don, P., eds., Science and Technology Letters, Kew, Surrey, U.K., pp. 375-384.

Loi, G.,Trois, P. and Rossi, G., 1995, "Biorotor: a new development for biohydrometallurgical processing," in Biohydrometallurgical Processing. Vargas, T., Jerez, C.A., and Wiertz, J.Y., eds., Vol. 1, The University of Chile, Santiago, pp. 253-271.

Loi, G., Mura, A., Trois, P., and Rossi, G., 1994, "The Porto Torres biodepyritization pilot plant: light and shade of one year operation, " Fuel Processes and Technology, Vol. 40, pp. 261-268.

Moulder J.F., Stickle W.F., Sobol P.E., Bomben K.D., 1995, Handbook of X-Ray Photoelectron Spectroscopy: A Reference Book of Standard Spectra for Identification and Interpretation of XPS, Physical Electronics, USA.

Nicholson, H.M., Smith, G.R., Stewart, RJ., Kock, F.W., and Marais, HJ., 1994, "Design and Commissioning of Ashanti's Sansu Biox Plant Glenside," in Biomine '94, International Conference and Workshop - Applications of Biotechnology to the Mining Industry, Australian Mineral Foundation, Perth, Western Australia, pp. 2.1- 2.8.

Nienow, A.W., and Conti, R., 1978, "Particle abrasion at high solids concentration in stirred vessels," Chemical Engineering Science, Vol. 33, pp. 1077-1086.

Polidoro, M., 1992, "Analisi plasmidica in Thiobacillus ferrooxidans. Caratterizzazione di un plasmide di 20 kb presente in ceppi isolati da different! aree geografiche e geochimiche, " (Plasmid analysis in T. ferrooxidans. Characterization of 20 kb plasmids found in strains isolated from different geographic and geochemicalk areas), Ph.D. Thesis, in Italian, lstituto di Microbiologia - Universit di Roma "La Sapienza," 86 pp.

Ragusa, S., 1991, "The enumeration of the viable autotrophic leaching bacteria Thiobacillus ferrooxidans, " Ph. D. Thesis, The University of New South Wales, School of Biological Technologies, Sydney, Australia, 254 pp.

Rawlings, D.E., "Heavy metal mining using microbes," 2002, Ann. Rev. Microbiol., Vol. 56, pp. 65-91.

Rohwerder, T, Gehrke, T, Kinzler, K., and Sand, W., 2003, "Bioleaching review Part A: Progress in bioleaching: fundamentals and mechanisms of bacterial metal sulfide oxidation," Appl. Microbiol. Biotechnol., Vol. 63, pp. 239-248.

Rossi A., Atzei D., Da Pelo S., Frau F., Lattanzi P., England K.E.R., and Vaughan, DJ., 2001, "Quantitative X-ray photoelectron spectroscopy study of enargite (Cu^sub 3^AsS^sub 4^) surface," Surface Interface Analysis, Vol. 31, pp. 465-470.

Rossi, G., 1971, "La lisciviazione microbica dei minerali. I.:L'azione dei microrganismi reperiti in acqueacide di miniere italiane sulla pirite, " Resoconti dell'Associazione Mineraria Sarda - Iglesias, Iglesias, Italy, Vol. 76, pp. 5-23.

Rossi, G., Biohydrometallurgy, 1990, McGraw-Hill Book Company, G.m.b.H, Hamburg, 609 pp.

Rossi, G., 2001, "The design of bioreactors," Hydrometallurgy, Vol. 59, Nos. 2-3, pp. 217-231.

Rossi, G., Loi, G., and Trois, P., 2005, "Reattore a tamburo rotante per idrometallurgia, bioidrometallurgia e trattamento delle acque di rifiuto per esercizio continue" (Revolving barrel reactor for continuous operation for hydrometallurgy, biohydrometallurgy and water treatment), Italian Patent No. 0001329859; November 21, 2005.

Sakaguchi, H., Silver, M., and Torma, A.E., 1976, "Microbiological leaching of chalcopyrite concentrate by Thiobacillus ferrooxidans," Biotechnol. Bioeng., Vol. 18, No. 8, August 1976, pp. 1091-1101.

Sand, W., and Gehrke, T, 1999, "Analysis and function of the EPS from the strong acidophile Thiobacillus ferrooxidans," in Microbial Extracellular Polymeric Substances, Wingender, J., Neu, T.R., and Flemming H.-K., eds., Chapter 7, pp. 127-141.

Sand, W., Gehrke, T, Jozsa, R-G., and Schippers, A., 2001, "Biochemistry of bacterial leaching," Hydrometallurgy, Vol. 59, Nos. 2-3, pp. 159-175.

Sharma, P.K., and Hanumantha Rao, K., 2005, "Surface characterization of bacterial cells relevant to the mineral industry," Minerals & Metallurgical Processing, Vol. 22, No. 1, pp. 31-37.

Silverman, P., and Lundgren, D.G., 1959, "Studies on the chemoautotrophic iron bacterium Ferrobacillus ferrooxidans. I. An improved medium anda harvesting procedure for securing high cell yields," J. Bact., Vol. 77, pp. 642-647.

Snell, RD., and Snell, C.T, 1959, Colohmetric Methods of Analysis, 3rd Edition, Vol. II A, D. Van Nostrand, New York.

Stephenson, D., and Kelson, R., 1997, "Wiluna Biox plant expansion and new developments, Biomine '97. Australian Mineral Foundation, Glenside, South Australia, pp. M4.1.1-M4.1.8

Torma, A.E., Walden, C.C., and Branion, R.M.R., 1970, "Microbiological leaching of a ZnS concentrate," Biotechnol. Bioeng., Vol. 12, pp. 501-517.

van Answegen, P.C., 1993, "Bio-oxidation of refractory gold ores- the Genmin experience," Biomine '93, Australian Mineral Foundation, Adelaide, South Australia, pp. 15.1-15.14.

van Answegen, P.C., and Marais, H.J., 2001, "Advances in the Application of the BIOX Process for Refractory Gold Ores," In Mineral Biotechnology, Kawatra, S.K. and Natarajan, K.A., eds., SME, Littleton, Colorado, USA, pp. 121-134.

G. Loi

Department of Medical Sciences "Mario Aresu," University of Cagliari, Italy

A. Rossi

Professor of analytical chemistry. Department of Inorganic and Analytical Chemistry, University of Cagliari, Italy

P. Trois and G. Rossi

Professor of drilling technology and professor of mining engineering (retired), respectively, Geoengineering and Environmental Technologies Department, University of Cagliari, Italy

Copyright Society for Mining, Metallurgy, and Exploration, Inc. Nov 2006

(c) 2006 Minerals & Metallurgical Processing. Provided by ProQuest Information and Learning. All rights Reserved.