Search Documents

By M. Pelser, Lorenzen L.

"The presence of calcium and magnesium in the hydrometallurgical processing of base metals, results in a number of process difficulties. These problems range from the contamination of the final product (where the final product is a precipitate) to high energy consumptions and large bleed streams during electrowinning. Calcium poses an additional problem in sulphate solutions owing to the low solubility of its sulphate salts. This can easily lead to a process stream that is supersaturated with calcium sulphate, resulting in unwanted solids formation within the process equipment.A review of literature has proven that there is no conventional method for the control of calcium and or magnesium, although a number of potential processes have been identified. The technologies investigated included precipitation, ion exchange and solvent extraction, for which at least one method was found under each. To make the reader aware of these processes a short description was provided of each within the text.Of these processes, fluoride precipitation was selected for experimental investigation based on the low solubility of calcium and magnesium compared to the base metals. This process involves the addition of a soluble fluoride to precipitate CaF2 and MgF2. The experiments were designed to investigate the equilibrium of the system at varying parameters of which the level of fluoride addition, the ratio of calcium to magnesium, and the identity of the soluble fluoride was reported. In all of the instances the process showed high selectivity towards calcium and magnesium over the base metals. The experiments proofed the process to have a higher affinity towards the removal of magnesium at compositions where the molar ratio of calcium to magnesium was higher than 1:3. It is often the case that the calcium to magnesium ratio is in the region of 1: 10. From this it follows that the fluoride addition has to be stoichiometric or greater for calcium and magnesium combined, if the calcium is to be precipitated. This was demonstrated as well. On the identity of the soluble fluoride hydrogen fluoride was compared to ammonium fluoride, of which the latter was found to be preferred."

Jan 1, 2003

By Francisco Vicuña, Miguel Becerra, Nelson Iglesias

"The following article presents the development of a dynamic tool for the estimation of Jb/Jc rate (ball filling level / total load filling level) of a SAG mil operating in a single stage mode. Using the methodology detailed in this article, it is now possible to have a dynamic estimation tool in the operation, which enables strategic decision making and achieving a continuous development in process optimization. The estimating of the load filling level (Jc) was conducted using a power model, which in this case corresponds to the C-Model developed by (Morrell, 1993) that was calibrated by measuring the load filling levels (Jc) and balls (Jb) using the FARO laser scanning technology by means of Grind-Out and Crash-Stop procedures during the year 2013 and 2014. Finally, to estimate the dynamic filling level of full charge, Jc, three independent models were developed that allowed to feed the adjusted power model: •,SAG mill discharge grate slot wear model •,SAG mill internal volume wear model •,Ball filling level estimation model."

Jan 1, 2014

By M. H. Moys, M. M. Bwalya

The use of the Discrete Element Method (DEM) to simulate the interaction of particles in dynamic environments is growing rapidly. This technique is also being applied in the study of milling extensively. However, to date more focus has been on the validation of the model. In spite of the advantages of the detailed information that is available from the simulator, appropriate application of this information remains a daunting task. In this paper a technique that combines particle breakage probability data with the energy spectra generated from the DEM is discussed.

Jan 1, 2003

By Takashi Furuyama, Sukeyuki Mori, Tetsuo Ikejiri, Abel Bissombolo, Yoshitsugu Koga, Masamichi Hata

In Japan, the increasing number of automobiles newly put into circulation is followed by a larger amount of end-of-life vehicles from which valuable metals are being recovered in shredder and related plants. Nevertheless, the processing of these scrapped automobiles which has reach a recycling rate of about 75%, generates a huge volume of waste, locally known as .shredder dust., which is commonly dumped on landfills. The saturation and depletion of valuable disposal sites in the near future, as well as the social and environmental problems related to the accumulation of these residues disposals are a major challenge for the automobile shredding industry. To promote an alternative way for the management of these disposals, the authors have initiated and developed a system for their processing, on the basis that more than 90% of these residues are combustible materials that could be used as refuses derived fuel. The developed processing system points out the mechanical/physical separation of shredded automobiles residues into two streams by means of an ascending airflow passing through a vertical zigzag separator and, the compaction of the light-density fraction of materials (that is about 60% of auto shredder residue recovered as float products and mainly composed of cotton, sponges and light plastics) to form a refuse derived fuel (RDF), the density, calorific value and ignition loss of which are 1.19 g/cm3, 4470 kcal/kg and 60.7%, respectively.

Jan 1, 2003

By G. Merazchiev

In the selection of copper ? molybdenum concentrates with the use of reagents containing sulfide ion ? sodium sulfide, sodium hydrosulfide etc for depression of copper minerals and pyrites, big part of depressor is oxidized by oxygen, which is dissolving in pulp during aeration with air in flotation machines and it is expended unproductively. This is the main reason for the big expenditures of depressor, reaching up 15~20 kg/t collective concentrate and the high exploitation expenditures for the copper ? molybdenum selection. Effective direction for lowering expenditure of depressor and increasing of the technical ? economical effectiveness of moly production is the conducting of the selection in nitrogen medium. In the paper are shown the results, obtained by applying of this direction in selection of collective copper ? molybdenum concentrate from the porphyry copper deposit ?Ellatzite? ? Bulgaria, in which the ratio Mo:Cu is over 1:120. In laboratory conditions, through comparative experiments in medium of air and nitrogen are defined: expenditure and initial concentration of NaHS in pulp for effective depression of copper minerals and pyrites; kinetics of flotation process of molybdenite; influence of pulp density; emf (negative milivolts) of pulp and the expenditure of nitrogen on tech-nological indexes of molybdenum flotation. When working in nitrogen medium decrease of NaHS in rougher molybdenum flotation from 40% to 45% is settled. When nitrogen is also fed in cleaner stages, the total expenditure of depressor is lowered by 64%. The quality of end molybdenum con-centrate is raised with over 5%. On the ground of the results of laboratory studies and of world practice data, an industrial installation for selection of copper ? molybdenum concentrate in nitrogen medium in Wemco inert ? gas machines.

Jan 1, 2014

By Lei Zhang, Xianjun Lu1, Minjie Lian, Liyang Zheng

In order to investigate rheological behaviour of fine tailings backfill pastes, the fine iron ore tailings (size fraction of -0.074mm accounts for 70%) from Fuquan Mining Co., Ltd. of Sinosteel Corporation (Shandong, China) was investigated. The rheological parameters such as slump, viscosity and yield stress of the tailings pastes under different densities were measured in detail, and the effect of paste density on the rheological property and the relationships of different rheological parameters of the pastes were also discussed. According to the results, the rheological behaviour is in accordance with the pseudoplastic fluid in a wide range of paste densities (solid content, w/w) of 65% to 80% and the pastes turn to be Bingham fluids at higher shearing rates. Along with the increase of paste density, the slump drops gradually and the viscosity and shear stress increase gradually. However, the variation extent of the rheological parameters varies at different density ranges. According to the variation pattern of the rheological parameters at different paste densities, the flow ability of the pastes can be classified into three zones. The high fluidity zone was observed at the density range of 65%-69% where the rheological parameters vary slightly along with the paste density, the moderate fluidity zone was observed at the density range of 70%-76% where the rheological parameters vary significantly along with the paste density, and the low fluidity zone was observed at the density range of more than 76% where the rheological parameters vary sharply along with the paste density and the critical density for this sudden variation is about 76%. Furthermore, linear correlations of slump/viscosity and slump/yield stress were observed. It is possible to predict the viscosity and yield stress of the pastes based on the slump measurements.

Jan 1, 2014

By Manisha Singh, Manikanchan Sarkar

"Presently the global steel production is mainly based on iron ore sourced from high grade hematite deposits. India has large deposits of good quality hematite ore but with the prevailing practice, such deposits are depleting fast and getting exhausted. However, apart from such high grade iron ores, India is also endowed with large reserves of low grade iron ore, such as BHQ (Banded Hematite Quartzite), BHJ (Banded Hematite Jasper), BMQ (Banded Magnetite Quartzite) and Magnetite. In the present situation, it has become necessary to focus attention on upgradation of these low grade ores to produce useable concentrate. It augurs well that already substantial investigation work has been carried out in this respect and a few plants have been set up in India and abroad. One important aspect in this regard is the process route options with optimum yield and grade. Some of these low grade ores, particularly magnetite, are found to contain high phosphorous level, removal of which poses a major challenge. Since the grade is low, substantial upgradation necessitates fine grinding for appropriate liberation. Fine grinding is energy intensive and needs attention to achieve economy in this regard. Large quantity of tailings is generated in the process, wet disposal of which gives rise to a lot of hassles and hence, possibility of dry disposal of the same needs to be explored. The experience of the authors, with regard to aforesaid aspects which has been gained in course of carrying out relevant assignments is presented in this paper."

Jan 1, 2014

By Joseph Kelly, Christopher Kelsey

This revolutionary super-fine crusher has the unique capacity to generate fine and ultra-fine products from single stage reductions of coarse or fine mineral feeds. The crusher consists of a rotating compression chamber and an internal gyrating mandrel. The axis of rotation of the shell is displaced relative to the axis of the vertically mounted mandrel. The inner surface of the compression chamber is machined as double cone frustums at an angle equal to the displacement angle. Crusher feed is accelerated into a compacted particle bed, which lines the inside of the compression chamber. The depth of the compacted bed progressively increases until the mandrel is engaged. The mandrel is then free to rotate with the compression chamber. Continued feeding increases power draw and triggers the counter rotating mandrel gyration drive. The gyrating mandrel delivers multiple compressions at high pressure to the rotating particle bed. High concentrations of energy are harnessed within a small active volume over short retention times. The combination of extraordinarily high power intensity and extreme compressive force results in primary compressive fracture and induced secondary tensile failure. The axially displaced rotation of the compression chamber effectively distributes the breakage forces throughout the particle bed. This paper covers open circuit, pilot trials on magnetite, perlite, garnet, marble and a copper/pyrite concentrate. Particular emphasis is placed on the extraordinary results achieved with trials on zircon and highly refractory fly ash. The extreme duties were deliberately chosen to demonstrate the crusher’s application scope and potential to impact capital operating and maintenance costs as the technology matures. Super-fine crushing, once considered impracticable has the potential to offer compelling options for the generation of ultra-fine products.

Jan 1, 2014

By Janusz S. Laskowski

"Frothers offered by different manufacturers are marketed with some information on their chemical and physical properties (e.g. density, viscosity, color, water-solubility, safety/toxicity, etc.). But frothers are used in flotation to generate fine bubbles and to stabilize the forth, and this information is only provided in a very qualitative description of the visual appearance of bubbles and froths. The terms “powerful” and “selective”, which have been commonly used to characterize frothers, have intuitive rather than scientific meaning.This paper introduces the DFI (Dynamic Foamability Index) plotted vs. CCC (Critical Coalescence Concentration) diagrams to classify flotation frothers (Fig. 1). The frothers characterized by low CCC and high DFI values have properties of strong flotation frothers, while those with high CCC and low DFI values are more selective. The proposed frother classification system, based on simple CCC and DFI measurements, provides, for the first time, a solid ground-work for characterization of the flotation frothers."

Jan 1, 2003

By V. Avzian Belozersky reg 453524, Vasily V. Lodeischikov, Adolf F. Panchenko, Til E. Vidusov, Olga D. Khmelnitskaya, Vladimir E. Dementyev

"Presently, there is a number of gold deposits which cannot be developed using the conventional technologies with acceptable economic indices due to low gold content, high depth of ore occurence and other conditions. For such ores, an innovation technology of underground gold leaching in the place of their occurrence is of special interest. The technology excludes ore mining and hoisting to the surface. This enables the production cost of 1 g of gold to be considerably reduced in comparison with conventional methods of gold recovery.The Maminskoye deposit oxidized ores were chosen as the subject of underground leaching because they are ideally suited for this purpose due to their composition and hydrological parameters. The ore reserves within the underground leaching block are 90, 000 t with 0.7 g/t average gold grade.An experimental system for underground leaching with a throughput of 15 cubic m solutions per hour was developed. A combined filtration-and-infiltration system of the ore massif development by holedrillingwhere pregnant solutions are air-lifted was worked out.The underground leaching flowsheet includes Au dissolution by chlorine-containing solutions, Au adsorption from solution onto active carbon, carbon concentrate ashing and calcine melting to produce Dore metal."

Jan 1, 2003

By Sergey Gudkov, Yuri Yemelianov, Stanislav Balikov, Andrey Boldyrev, Andrey Bogorodsky

"The results of POX of refractory gold-bearing concentrates using halide-ion as a lixiviant for precious metals and activated carbon as an adsorbent are given. It was found that gold did not recover to the solution when halide was added during POX. Silver partially dissolved in the solution. When halide-based reagents and activated carbon were added during POX, gold dissolved and loaded onto the activated carbon. The degree of sulfides oxidation was 99.3-99.6%. The recovery of gold onto granulated carbon was 94.7-94.8% when using chlorine-bearing lixiviants: NaCL, CaCL and carnalite and 98-99% when using KJ and KBr. This treatment process of refractory gold-bearing concentrates combines sulfide oxidation, dissolution and adsorption of precious metals. The results of the metals recovery achieved using this process can be compared with the recovery during POX-cyanidation. In this case, sodium cyanide was not used. This allows the reduction of capital and operational costs due to the absence of cyanidation, CIL and detoxification of leaching tailings circuits."

Jan 1, 2014

By K. W. Stöckelhuber, A. Wenger, H. J. Schulze

"An important elementary step of the flotation process is the forming, drainage and rupture of a thin intervening liquid film between bubble and mineral particle.After the encounter of the bubble with the solid surface a wetting film is formed. The initial thinning of this film is driven by the capillary pressure inside the bubble. When the film has reached a certain thickness rupture can take place. There are two possible mechanisms describing this process:(a) Gas nucleation, caused by nano bubbles inside the film.(b) Growing fluctuation waves (spinodal dewetting) on fluid interfaces under the influence of any kind of attractive forces (electrostatic, van der Waals).Metastable wetting films on negatively charged, hydrophobic glass surfaces (gaseous phase methylated), or on hydrophilic, positively charged glass surfaces (with Al3+ ions), have been analysed by means of the kinetics of film thinning according to the Reynolds law. These experiments demonstrate that both mechanisms can be responsible for the rupture of a thin wetting film."

Jan 1, 2003

By Leena Vesa, Antti Remes

Design of thickeners is based on laboratory test work, characterizing the material and determining the optimal reagents, dosages and settling regimes. Several methods for dimensioning plant scale equipment exist. Kynch and Talmage-Fitch design procedures has been applied widely for a long time. Nowadays, besides them, a dynamic laboratory test work is applied for design of high rate thickeners. In that the laboratory unit is operated with known solids load per area, feed and flocculation conditions and with raking of the underflow. All these test methods afford useful data for setting up and calibrating a dynamic simulation model of a plant scale thickener. This paper presents a flocculation kinetics based dynamic thickener simulation model and a procedure to calibrate the model parameters. With the model, the underflow stream composition can be simulated over time as a function of the thickener operating parameters, which are: the feed rate and composition, underflow pumping rate, flocculant dosage and dilution of the feed. In addition, the model incorporates calculation of the slurry bed level, the cone pressure, the yield stress and the rake torque. Such a dynamic modeling scheme enables setting up simulation aided operator trainings and developing and testing of advanced control strategies, for example. The unit model can be integrated as a part of concentrator-wide flow sheet simulation model as well. Here, calibration of the kinetic flocculation model based on a laboratory test case data and further the operation of a simulated plant scale thickener are presented.

Jan 1, 2014

By B. McFadzean, K. Moller

"The mechanism whereby thiol collectors adsorb onto a mineral surface has been widely studied. Most of these studies examine the mineral surface at equilibrium state and, often, well above monolayer coverage. However, the formation of species at the mineral surface is a complex process that is both kinetically and thermodynamically controlled. In a typical industrial application, the dosage of collector will be around monolayer level. This study uses isothermal titration microcalorimetry to measure the heat evolved during the reaction of sodium ethyl xanthate (SEX) and sodium isobutyl dithiocarbamate (SIBDTC) and their mixtures with galena and chalcopyrite. The sensitivity of the instrument allows heats of reaction of sub-monolayer quantities of thiol collector to be measured.The study found that both the magnitude and kinetics of the reaction between a collector and a mineral surface was dependent on the type of mineral and the type of collector. Proposals are made for the reactions based on literature and the evidence from this investigation. It was found that when collectors were reacted with mixtures of galena and chalcopyrite, that the magnitude of the reaction was lower compared to that measured for the pure minerals alone. This may be due to galvanic interactions between the minerals, resulting in a passivating effect on one or both of the minerals."

Jan 1, 2014

By Isaak I. Maksimov, Tatyana E. Koltunova, Mikhail F. Yemelyanov, Liudmila A. Otrozhdyonnova

"INTRODUCTIONMetals losses amount up to 50% in coarse fractions in (>100 µm) and composite particles, and to 20- 25% in fine fractions (<10 µm) at non-ferrous floatation concentrating plants of the CIS. Therefore, reduction of metals losses in difficult fractions is of high significance.IMPROVEMENT OF FLOATABILITY OF COARSE AND COMPOSITE PARTICLESReagent regimes with application of strong collectors – xanthates ?5-?8, combination of butyl xanthates and aerofloats ?10-?12, butyl xanthates with hydrocarbons are proposed for this purpose. Application of such reagent regimes for copper and copper-molybdenum ores permitted to increase copper and molybdenum recovery by 0.5-3.9%.Studies, performed by means of specially designed non-frothing flotation machine with independently controlled agitation intensity and narrow size-ranges of air bubbles, established that the best results are achieved with employment of bubbles of 0.5-0.8 mm size regarding particles of any size, with small amounts of up to 2 mm size “transport” bubbles to be expediently used for floatation of coarse particles.Energy dissipation in the main volume of floatation cells must be lower than 0.2-0.25 m2/s3.Performed experiments permitted to formulate requirements to development of machine for floatation of coarse and composite particles. Such floatation machine with volume of 8.5 m3 provided for weak agitation in the main volume of floatation cell and uniform distribution of coarse particles throughout the depth owing to establishing of pulp circulation with upward flows. In addition, in the upper part of floatation cell where agitation is considerably weaker, favorable conditions are created for attachment of particles to bubbles. Testing of this floatation machine demonstrated a 7.5% increase in recovery of +0.2 mm particles in pyrite floatation and a 12-14% increase in recovery of +0.2 mm particles in apatite floatation as compared with a reference floatation machine."

Jan 1, 2003

By Andrew T. McBride, Malcolm S. Powell, Indresan Govender

"Current empirical Semi Autogneous Grinding (SAG) mill models contain assumptions concerning the amount of energy available for particle breakage, how this energy is distributed in the mill media and the rate at which the energy is transferred. Despite these assumptions the JK SAG model is highly functional with a proven track record. DEM provides a numerical framework to greatly improve on these assumptions that were necessary due to a lack of available input data.This paper identifies key outputs of the Discrete Element Method (DEM), such as energy spectra and impact probability distribution. It describes how they can be incorporated into current mill models utilising practical rock breakage data. The rate at which energy is transferred is shown to be dependent on particle position and mill rotational speed and not once per cycle as previously assumed. Based upon this an improved energy transfer methodology is presented. The fundamental outputs of the DEM are validated against experimental three-dimensional particle tracking data, so as to establish whether the DEM produces correct energy transfer predictions on the inter-particle level, which is crucial to this work. • Effective Experimental ValidationAn automated 3D tracking technique is employed to study the motion of plastic particles deep within the charge of a 142mm diameter experimental mill. A Biplanar Angioscope is used to film the mill with X-rays in two planes simultaneously at 50 frames/s, with a shutter speed of 1/3000s. A marked ball within the charge is tracked for over 2 minutes and the resulting digital images processed by a fully automated imaging technique, which locates the 3D coordinates of the particles to within 0.2mm. With over 250 circulations, this provides statistically rigorous and accurate data for verification in the numerical modelling of mills.A series of validation techniques have been developed to compare the experimental trajectory data with the outputs of both 2D and 3D DEM simulations. The techniques describe the shape of the charge and the particle motion within mill, allowing objective comparisons to be made of key features of the charge, such as the equilibrium surface and the centre-of-circulation."

Jan 1, 2003

By Katsunori Okaya, Toshio Inoue

"Movement of mill charge in the centrifugal mill is essentially dissimilar from that in the vibration mill with circular path despite the apparently similar mode in the motion of mill body. This dissimilarity is the reason for the substantial advantage regarding the comminution intensity of the centrifugal mill over the conventional vibration mill accounting approximately to one order of magnitude.Bradley et al. (1983) developed a math model regarding the mill charge as a lump. This model is presented in a simple equation with which motion of the mill charge can be predicted in terms of the parameters describing the specification and the operation of the centrifugal mill. This equation can be used to explain the power draft of the mill as well as the stability of mill operation. More recently, on the other hand, a numerical simulation technology called discrete element method (DEM) has become a vital vehicle for estimating the motion of the grinding media in the mill. The authors developed a mill simulator called ‘Mill Solver’. Mill Solver is designed to simulate various types of mills including centrifugal and vibration mills. The principle of this simulator is to calculate individual kinetic events of interaction of balls in the mill in a minute time interval, hence the mill performance can be constructed for estimation."

Jan 1, 2003

By J. D. Miller, C. L. Lin

In general, x-ray computed tomographic (CT) techniques are able to provide 3D images of the internal structure of opaque materials in a nondestructive manner. The unique cone beam geometry allows acquisition of all 2D projections with only one rotation of the sample thus providing for fast data acquisition and better x-ray utilization, as a complete 2-D detector array receives the cone-shaped flux of rays. Thus, an isotropic 3D volume can be reconstructed without the mechanical translation and the stacking of sequential slices as is the case for more conventional CT scanners. In this regard, a state-of-theart, custom designed x-ray microtomography facility to provide very detailed 3D spatial analysis of multiphase particles has been installed and is in operation at the University of Utah. The reconstructed 3D tomographic volume allows for an adjustable spatial resolution from 5 to 20 microns for sample dimensions from 10 to 40 mm in diameter, respectively. For example, a packed bed of coal particles is shown below in which different crystalline froms of pyrite are distinguished in 3D.

Jan 1, 2003

By Sebastian Fredes, Williams Calderon, Mauricio Toledo, Gonzalo Montes-Atenas

The influence of the temperature on leaching kinetics of copper-bearing minerals has been widely acknowledged: the leaching of copper oxide ores is generally described as exothermic reactions while copper sulfide ores are associated with endothermic reactions. Reports studying the influence of temperature on leaching kinetics at laboratory scale frequently involve the implementation of either fully flooded isothermal systems or jacketed column setups. Nevertheless, at industrial scale controlling or monitoring the temperature or temperature changes is almost impractical. In fact, leaching process is usually a large tonnage operation and changing the temperature of vast amounts of material would require an enormous amount of energy and highly specific heat transfer equipment. Preliminary computations on the amount of energy required to increase the temperature of a heap or dump in 1°C indicates that it would be impossible to find an energy source able to accomplish the task. In order to provide a solution to this matter it is suggested to create an energy managing protocol (EMP) that permits heat up or down the porous body locally taking advantage from the temperature gradients and from the reactivity of different regions within the material as a function of the leaching cycles. The aim of this study is to evaluate different heat transfer regimes in a 1 ton pilot column leach operation. It was found that best results of copper recovery are obtained when heating up the column in horizontal bands starting from the top of the column moving downwards as the leach cycle evolves. The system is expected to be suitable for industrial applications, but further development of the technique is required, particularly in the design of the heat transfer dispositive.

Jan 1, 2014

By Sashnee Raja, Mehdi Nemati, Ashley Sissing, Susan T. L. Harrison

"Extreme thermophiles, operating at 65 - 90°C, are finding increasing application in biohydrometallurgy. These micro-organisms have the advantage of exhibiting better oxidation kinetics in terms of reaction rate and extent of mineral solubilisation (Duarte et al., 1993; Norris and Barr, 1988; Konishi et al., 1995). These extreme thermophilic Archae, characterised by a different cell envelope structure (König and Stetter 1986, König 1988, Michel et al. 1980) perceived to provide reduced structural strength in comparison to bacterial envelopes, and customised metabolic pathwaysThe active tank bioleaching process, used for high rate leaching, is conducted in a slurry bioreactor. Here sufficient energy dissipation is required to ensure suspension of the mineral phase as well as to sustain high rates of gas liquid mass transfer. It is well known that subjection to hydrodynamic stress of sufficient magnitude may alter the physiological state of a range of micro-organisms through changed viability (ability of the microbes to grow), metabolic activity and integrity. While limited study on the influence of the solids phase on thermophilic leaching performance is reported, initial reports suggest that the thermophiles used for bio-oxidation appear to be sensitive to hydrodynamic conditions (Jordan et al 1993, Clark and Norris, 1996) and the presence of solids (Norris and Bar 1988, Le Roux and Wakerley, 1988, Escobar et al. 1993, Torres et al. 1995, Nemati and Harrison, 2000)."

Jan 1, 2003