In many industries the final product, or the raw material at somestage of the manufacturing process, is in powdered form and in consequence the rapid and cheap preparation of powdered materials is a matter of considerable economic importance.
In some cases the powdered material may be prepared directly; for example by precipitation from solution, a process which is used in the preparation of certain types of pigments and drugs, or by the vacuum drying of a fine spray of the material, a process which is widely adopted for the preparation of milk powder, soluble coffee extracts and similar products. Such processes are, however, of limited applicability and in by far the greatest number of industrial applications the powdered materials are prepared by the reduction, in some form of mill, of the grain size of the material having an initial size larger than that required in the final product. These processes for the reduction of the particle size of a granular material are known as milling or grinding and it appears that these names are used interchangeably, there being no accepted technical differentiation between the two.
Examples of the first two classes occur in mineral dressing, in which size reduction is used to liberate the desired ore from the gangue and also to reduce the ore to a form in which it presents a large surface to the leaching reagents.
Under the third heading may be classed many medicinal and pharmaceutical products, foodstuffs, fertilizers, insecticides, etc., and under the fourth heading falls the size reduction of mineral ores, etc.; these materials often being reduced to particles of moderate size for ease in handling, storing and loading into trucks and into the holds of ships.
The quantity of powder to be subjected to such processes of size reduction varies widely according to the industries involved, for example in the pharmaceutical industries the quantities involved per annum, can be measured in terms of a few tons, or in the case of certain drugs, possibly a few pounds; whereas in the cement industry the quantities involved run into tens of millions of tons; the British cement industry alone having produced, in round figures, 12 million tons of Portland Cement.
For the preparation of small quantities of powder many types of mill are available but, even so, the ball mill is frequently used. For the grinding of the largest quantities of material however, the ball, tube or rod mill is used almost exclusively, since these are the only types of mill which possess throughput capacity of the required magnitude.
The great range of sizes covered by industrial ball mills is well exemplified by Fig. 1.1 and Fig. 1.2. In the first illustration is shown a laboratory batch mill of about 1-litre capacity, whilst in Fig. 1.2 is shown a tube mill used in the cement industry the tube having a diameter of about 8 ft and length of about 45 ft.
In Fig. 1.3 is shown a large ball mill, designed for the dry grinding of limestone, dolomite, quartz, refractory and similar materials; this type of mill being made in a series of sizes having diameters ranging from about 26 in. to 108 in., with the corresponding lengths of drum ranging from about 15 in. to 55 in.
At this point it is perhaps of value to study the nomenclature used in connection with the mills under consideration, but it must be emphasized that the lines of demarcation between the types to which the names are applied are not very definite.
The term ball mill is usually applied to a mill in which the grinding media are bodies of spherical form (balls) and in which the length of the mill is of the same order as the diameter of the mill body; in rough figures the length is, say, one to three times the diameter of the mill.
The tube mill is a mill in which the grinding bodies are spherical but in which the length of the mill body is greater in proportion to the diameter than is the case of the ball mill; in fact the length to diameter ratio is often of the order of ten to one.
The rod mill is a mill in which the grinding bodies are circular rods instead of balls, and, in order to avoid tangling of the rods, the length to diameter ratio of such mills is usually within the range of about 15 to 1 and 5 to 1.
It will be noticed that the differentiation between ball mill and the tube mill arises only from the different length to diameter ratios involved, and not from any difference in fundamental principles. The rod mill, however, differs in principle in that the grinding bodies are rods instead of spheres whilst a pebble mill is a ball mill in which the grinding bodies are of natural stone or of ceramic material.
As the name implies, in the batch mills, Fig. 1.4a, the charge of powder to be ground is loaded into the mill in a batch and, after the grinding process is completed, is removed in a batch. Clearly such a mode of operation can only be applied to mills of small or moderate sizes; say to mills of up to about 7 ft diameter by about 7 ft long.
In the grate discharge mill, Fig. 1.4b, a diaphragm in the form of a grating confines the ball charge to one end of the mill and the space between the diaphragm and the other end of the mill houses a scoop for the removal of the ground material. The raw material is fed in through a hollow trunnion at the entrance end of the mill and during grinding traverses the ball charge; after which it passes through the grating and is picked up and removed by the discharge scoop or is discharged through peripheral ports. In this connection, it is relevant to mention that scoops are sometimes referred to as lifters in the literature. In the present work, the use of the term lifter will be confined to the description of a certain form of mill liner construction, fitted with lifter bars in order to promote the tumbling of the charge, which will be described in a later section.
In the trunnion overflow mill, Fig. 1.4c the raw material is fed in through a hollow trunnion at one end of the milland the ground product overflows at the other end. In this case, therefore, the grating and discharge scoop are eliminated.
A variant of the grate discharge mill is shown in Fig. 1.4d, in which the discharge scoop is eliminated by the provision of peripheral discharge ports, with a suitable dust hood, at the exit end of the mill.
Within the classes of mills enumerated above there are a number of variations; for example there occur in practice mills in which the shell is divided into a number of chambers by means of perforated diaphragms and it is arranged that the mean diameter of the balls in the various chambers shall decrease towards the discharge end of the mill; such an arrangement being shown in Fig. 1.6. The reason for this distribution of ball size is that, for optimum grinding conditions, the ratio of ball diameter to particle diameter should be approximately constant. In consequence smaller balls should be used for the later stages of the grinding process, where the powder is finer, and by the adoption of a number of chambers in each of which the mean ball diameter is suitably chosen an approximation is made towards the desired constancy in the ratio of the ball size to the particle size.
The problem of the optimum distribution of ball size within a mill will be dealt with more fully in a later chapter, but at this point it is relevant to mention a mill in which the segregation of the balls is brought about by an ingenious method; especially as the mill carries a distinctive name, even though no principles which place it outside the classification given previously are involved.
The Hardinge mill, Fig. 1.7, uses spheres as a grinding agent but the body is of cylindro-conical form and usually has a length to diameter ratio intermediate between those associated with the ball mill and the tube mill. The reason for this form of construction is that it is found that, during, the operation of the mill, the largest balls accumulate at the large end of the cone and the smallest balls at the small end; there being a continuous gradation of size along the cone. If then the raw material is fed in at the large end of the mill and the ground product removed at the smaller end, the powder in its progression through the mill is ground by progressively small balls and in consequence the theoretical ideal of a constant ratio between ball size and particle size during grinding is, to some extent, attained.
The type of ball mill illustrated in Fig. 1.3, incorporates a peripheral discharge through line screens lining the cylindrical part of the mill. Heavy perforated plates protect the screens from injury and act as a lining for the tumbling charge; sometimes also the fine screen is further protected by coarse screens mounted directly inside it. This type of mill, which is often known as the Krupp mill, is of interest since it represents a very early type of mill which, with modifications, has retained its popularity. The Krupp mill is particularly suited to the grinding of soft materials since the rate of wear of the perforated liners is then not excessive. At this point it will perhaps be useful to discussthe factors upon which the choice between a ball a tube or a rod mill depends.
When a mill is used as a batch mill, the capacity of the mill is clearly limited to the quantity which can be handled manually; furthermore the mill is, as far as useful work is concerned, idle during the time required for loading and unloading the machine: the load factor thus being adversely affected. Clearly then, there will be a considerable gain in throughput, a saving in handling costs and improved load factor, if the mill operation is made continuous by feeding the material into the mill through one trunnion and withdrawing it either through the other trunnion or through discharge ports at the exit end of the mill body.
Since, however, the flow of powder through the mill is now continuous, it is necessary that the mill body is of such a length that the powder is in the mill for a time sufficiently long for the grinding to be carried to the required degree of fineness. This, in general, demands a mill body of considerable length, or continuous circulation with a classifier, and it is increased length which gives rise to the tube mill.
In the metallurgical industries very large tonnages have to be handled and, furthermore, an excess of fine material is undesirable since it often complicates subsequent treatment processes. In such applications a single-stage tube mill in circuit with a product classifier, by means of which the material which has reached optimum fineness is removed for transport to the subsequent processing and the oversize is returned to the mill for further grinding, is an obvious solution. Once continuous feed and a long mill body have been accepted, however, the overall grinding efficiency of the mill may be improved by fairly simple modifications.
As has already been mentioned; for optimum grinding conditions there is a fairly definite ratio of ball size to particle size and so the most efficient grinding process cannot be attained when a product with a large size range is present in the mill. If, however, a tube mill is divided into a number of compartments and the mean ball size of the grinding media decreases in each succeeding compartment; then the optimum ratio between ball size and particle size is more nearly maintained, and a better overall performance of the mill is achieved; this giving rise to the compartment mill shown in Fig. 1.6. The tube mill has the further advantage that, to some extent, the grinding characteristics of the mill are under control; for example, an increase in the size of the balls in the final chamber will reduce the rate of grinding of the finer fractions but will leave the rate of grinding of the coarser fractions sensibly unchanged and so the amount of coarse material in the final product will be reduced without any excessive overall increase in fineness.
The principal field of application of the rod mill is probably as an intermediate stage between the crushing plant and the ball mills, in the metallurgical industries. Thus, material between about 1-in. and 2-in. size may be reduced to about 6 mesh for feeding to the ball mills. Rod mills are, however, being used in closed circuit with a classifier to produce a product of less than about 48-mesh size, but such applications are unusual.
EF2 Open Circuit Grinding when grinding in open circuit ball mills, the amount of extra power required, compared to closed circuit ball milling, is a function of the degree of control required on the product produced. The inefficiency factors for open circuit grinding are given in Table VIII.
EF4 Oversized Feed when being fed a coarser than optimum feed, this factor applies to rod milling and ball milling. However, the most frequent use is found in conjunction with single stage ball milling. This is the one efficiency factor that is related to Work Index as is seen in the following equation:
When available, use the Work Index from a grindability test at the desired grind for Wi in equation 9. For equation 11, use either the Work Index from an impact test or a rod mill grindability test, whichever is higher. For equation 12, use the Work Index from a rod mill grindability test, since this more represents the coarse fraction of the feed; if not available then use the ball mill grindability test results.
This factor always applies to low ratios of reduction but its application to high ratios of reduction is not always needed, but should be used for mill size selection whenever Wi from the rod mill and ball mill grindability tests exceed 7.0.
EF7 Low Ratio of Reduction Ball Hill the need to use this factor does not occur very often as it only applies to ball milling when the Ratio of Reduction is less than 6. This shows up particularly in regrinding concentrates and tailings. The equation for this is:
EF8 Rod Milling a study of rod mill operations shows that rod mill performance is affected by the attention given to preparation and feeding a uniform top size feed size to the mill and the care given to maintaining the rod charge. This efficiency factor has not been definitely determined. In selecting rod mills based upon power calculated from grindability tests, the following procedure has been recommended:
EF8 The rod mill feed will be prepared by closed circuit crushing and the rod mill will be in a rod mill-ball mill (or pebble mill) circuit with no intermediate concentration stage so no EF8 factor need be applied. If it were just a rod milling circuit or if there were an intermediate concentration stage between the rod and the mill a 1.2 factor would apply.
Referring to Table V two mills will be required. The preliminary rod mill selection would be a 3.66 meter (12 foot) inside shell 3.46 meter (11.35 foot) diameter inside new shell liners. Referring to Table IX the EF3 (Diameter Efficiency) factor is 0.931.
Referring to Table V the 3.66 m x 4.88 m rod mill with 4.72 m (15.5 ft.) long rods calculates to draw 972 HP when carrying a 40 percent rod charge with a worn-in bulk density of 5606 kg per cubic meter (350 pounds per cubic foot). 1031 HP is required. Therefore, increase mill length by 0.3 meters (1 foot).
Therefore, use two 3.66 meter (12 foot) diameter inside shell 3.46 meter (11.35 foot) diameter inside new shell liners by 5.18 meter (17.0 foot) long overflow rod mills with a 40 percent by mill volume rod charge with 5.02 meter (16.5 foot) long rods.
These mills are required to prepare ball mill feed. With pebble milling the pebble portion of the product does not go thru the rod mill thus the rod mill feed rate is reduced by 30 metric tonnes per hour (6% of 500 metric tonnes per hour).
Therefore, use two 3.66 meter (12 foot) diameter inside shell 3.46 meter (11.35 foot) inside new shell liner by 4.88 meter (16 foot) long overflow rod mills with a 40 percent by mill volume rod charge with 4.72 meter (15.5 foot) long rods.
AG milling extends itself to many applications due to the range of mill sizes available. AG mills can accomplish the same size reduction work as two or three stages of crushing and screening, a rod mill and some or all of the work done by a ball mill.
The range of mill sizes and versatile applications allow AG milling to be accomplished with fewer lines than conventional set-ups. This, in turn, contributes to lower capital and maintenance costs for an AG mill circuit.
The Metso Outotec Qdx4TM mill drive provides the next step in the evolution of change in mill drive architecture, while allowing the system to be built with components that are within current manufacturing capabilities. We are essentially providing up to twice the power transmission of a conventional dual pinion drive.
Vertical stirred mills are usually charged with media occupying 80% of the mill volume which is in sharp contrast to tumbling mills that are seldom charged more than 40% of their volumes to allow space for the tumbling action to develop. The stirred mills are charged with a media size of 1012mm and operated at a maximum tip speed of 38m/s. The Metso detritor mill, however, has a maximum tip speed of 1112m/s. Even though a finer ground product is obtained with higher speeds, a limit has to be imposed. This is to allow the separation of media and mineral particles at the top of the mill where a settling zone develops. The ultrafine ground product is usually discharged as it passes through the separating screen. It also takes with it fine media product. The Metso detritor mill uses a screen size of 300m to retain sand when used as a media.
During operation a small amount of heat is generated. This affects the viscosity of the slurry. According to Kwade , if the viscosity of the slurry is too high the grinding efficiency is reduced as activity in the grinding chamber is inhibited and the contact between the grinding media and the particles is decreased resulting in less abrasive action on the mineral particles. Thus, the heat generated during media stirring could assist in lowering the slurry viscosity and hence benefit grinding efficiency.
The stress intensity and the number of stress events determine the specific energy to achieve a certain product fineness. The number of stress events is a function of mill operating parameters such as grind time, stirrer speed, percent solids and media size. The relationship between stress intensity and specific energy was shown by Kwade as indicated in Figure10.10.
The stress energy available to particle breakage is distributed in different sections of the mill, being maximum near the tip of the stirrer. For a satisfactory grind and size reduction of all particles, the residence time of the slurry in the mill is the prime factor. Experience so far indicates that about 3060seconds of travel time through a mill is adequate.
It may be noted that all the stress energy generated are not transferred to the mineral particles and the media hardness and media size affects the product size. The portion of the stress energy transferred to the particles depends on the Youngs modulus of both the grinding media and the particles .
The Vertimill is a vertical stirred mill that uses a helical screw to impart motion in the charge. Mechanically, the Vertimill is a very simple machine with an agitating screw suspended into the grinding chamber, supported by spherical roller bearings, and driven by a fixed speed motor through a planetary gearbox. Figure 8.14 presents a schematic diagram of the Vertimill. The screw rotates slowly such that the ball charge and slurry are not fluidized, but settle under gravity. The screw action pulls the ball charge up the center of the mill, and the charge eventually cascades over the edge of the screw, creating a general downward flow pattern at the mill perimeter. This pattern of flow, coupled with the low velocities involved, ensures that the grinding media and particles stay in contact with one another, thus enabling the efficient transfer of the drive energy into attrition and abrasion breakage throughout the charge. The operating conditions of the Vertimill are very similar to those of the conventional ball mill in the sense that the percentage solids of the feed should be kept in the range of between 65% and 75% by mass. The power draw of the mill is directly linked to the mass of balls within the mill. The distinct advantage that the VTM has over the conventional ball mill is its ability to use media smaller than 25mm more effectively.
In the past 20 years, more than 60 Vertimill machines have been sold for iron ore applications, with more than 20 of those in the last 2 years alone. Table 8.2 lists examples of the Vertimill in Brazilian iron ore applications. The Metso Vertimill is very common in regrind applications, but the industry quickly realized its potential in coarser applications as well.
where P is the shear power (W), the viscosity of the mill contents (N.s m2) (for which a relationship was given), the angular velocity (rad s1), and V the shear volume (referring to all the shear surface pairs between an impeller and the mill chamber). With appropriate calibration the predicted power matched the measured power for the data of Gao et al. (1996) and Jankovic (1998).
The IsaMill milling technology is a large-scale commercial high-speed stirred mill that is currently under development for coal micronising applications. The technology achieves very high energy efficiencies by using small milling balls in a high intensity configuration (>300 kWmm3). It is claimed that the technology (Fig.15.7) can produce coal with 90% of the particles less than 20m.
IsaMill technology is receiving serious consideration as part of a CSIRO investigation into preparing micronised coal for coal/water fuel in a diesel engine to deliver base-load power (see Section15.10.1).
Vibratory mills use oscillatory motion of the mill shell to agitate the media. As for the stirred mills, the active grinding zone encompasses the entire mill volume. The grinding energy is supplied by the inertia of the media and is not limited by gravity. In principle, high energy can be supplied to quite fine media, making these devices attractive for ultrafine grinding applications. By very careful matching of media size, powder size, and energy input (based on vibrational amplitude and frequency) it should be possible to achieve quite high grinding efficiencies. Unfortunately, mechanical design for reliability and low maintenance is not simple. Problems in these areas have tended to limit their large-scale application.
Different kind of mills are suitable for grinding, mechanical alloying and mechanical milling such as horizontal mills (tumbler ball mill), stirred mill (attritor, e.g. Szegvari attrition mill1), planetary ball mill, vibrating mill (tube vibrating mill, Sweco vibrating mill and shaker vibrating mill (e.g. Spex is a lab-scale mill3)). Their working principles and operating conditions are summarised in Fig. 12.1 and Table 12.1. The classification on a scale of increasing mill energy is: horizontal ball mill, attritor, planetary ball mill and vibrating ball mill. For example, a process that takes only a few minutes in the SPEX mill can take hours in an attritor and a few days in a commercial tumbler ball mill.
The choice of a milling technique is determined by many parameters. For example, attrition mills are more efficient than tumbler ball mills for mixing and blending WC-Co cutting tool powders because of short milling time, production of fine particle size (submicron sized) and enhanced smearing of Co onto carbide particles. However, as the product output is relatively low with attrition mills, tumbler ball mills are usually used for production runs of over 100kg/day. Moreover, powder contamination, which is an important criterion for many applications, can be due to the initial purity of the powder, the milling equipment (design), the milling operating conditions (mill speed, balls size and material, atmosphere) and/or the use of process control agent. It increases with milling time, milling intensity and with the reduction of the difference of strength/hardness between the powder and the milling balls.
There are in general two methods by which nanoparticles can be produced using high-energy milling: (i) milling alone and (ii) combining chemistry and milling (referred to as chemical-mechanical milling or mechanochemical processing). It is suggested that these methods offer the advantage of being easily scaled. References [19 and 20] are good starting points for further reading.
High-energy milling processes involve the comminution of bulk materials. The principle of comminution is centred on applying physical forces to bulk material so as to effect breakage into smaller sizes. The forces required to effect breakage are usually a combination of either impact or shear. Material is introduced into a milling chamber in which grinding (milling) media are contained. Milling occurs when the media is made to move either by stirring (using a rotor) or by shaking/vibrating the chamber and contacts the bulk material thus imparting, depending on the milling parameters, either impact or shear forces on it. Breakage can occur through a variety of mechanisms and are generally described as attrition, abrasion, fragmentation or chipping and occur both at the macro and microscopic level . This is illustrated in Figs 1.3 and 1.4
The rate at which comminution occurs is dependent on the size and frequency with which forces are applied. Breakage is influenced by both extrinsic and intrinsic factors. Intrinsic factors include such things as material properties (hardness, density, size) whilst extrinsic factors are determined by the amount of energy put into the system and the efficiency with which that energy is translated to the milling process. The latter is determined by variables such as vibrating frequency (in a rotorless mill), rotor speed (in a stirred mill), mill design, media size and loading, solids loading and whether the milling is performed dry or wet. These variables dictate which force regime predominates (i.e. shear or impact) which in turn dictate milling rate and efficiency. In high-speed stirred mills the effect of mill tip speed, media size and density can be evaluated simultaneously using the grinding media 'stress intensity' approach and an illustration of this is summarised in Fig. 1.5 which shows a plot of product particles size (starting size 45 m, product size 26 m) versus stress intensity for a pin mill using a zinc concentrate.
A variety of mills are commercially available and range from tumbling, shaker, vibratory, planetary and stirred ball mills. Production of nanoparticles using this technique is sometimes limited by the need for extended milling times, material properties and contamination issues. Attrition methods allow the production of alloys and composites that cannot be synthesised by conventional casting methods. They have also gained attention as non-equilibrium processes resulting in solid state alloying beyond the equilibrium solubility limit and the formation.
The types of nanoparticles produced by the attrition milling technique are generally alloys or single-phase powders. When a single-phase elemental powder (or intermetallic) is milled, grain size asymptotically reduces to a range of 330 nm . For alloys produced by this method, unstable intermediate substances are formed, from mixing and diffusion as a consequence of repeated deformation and folding of the different metals. These intermediates allow the chemical reactions necessary for alloy formation to occur .
For non-metallic compounds (carbides, oxides, etc.) the reduction in grain size is consequent on fracturing and cold welding and the limit to minimum grain size is determined by the minimum size that does not support nucleation and propagation of cracks within the grain. For metallics, on the other hand, it is thought that the reduction in grain size is a process where localised plastic deformation is induced, subgrains are formed (by eradication of dislocations) which combine (through intimate mechanical contact) to form discrete grains. The latter process is analogous to recrystallisation observed during hot forming of metals and alloys but in these circumstances at low temperatures. In intermetallics, the process is thought to be different again in that grain formation is due to nucleation (on a nanoscale) followed by a limited growth of the generated phase [20,23]. There are numerous examples in the literature of alloy and mixed metal oxide production using this technique [20,28]. Few examples can be found where single-phase powders or particles are produced at the nanoscale level [24,25].
In this technique, a large number of small grinding media are agitated by impellers, screws, or disks in a chamber. Breakage occurs mainly by the collision of the media. There are designs of the mill with a vertical or horizontal rotating shaft with wet or dry systems (Fig. 2.19).
The medium agitating mill is one of the most efficient devices for micronizing materials and has been in regular use for the production of fine particles and mechanical alloying processes. A typical stirred mill with a vertical rotating shaft and horizontal arms is shown in Fig. 2.20. This stirring action causes a differential movement between balls and the material being milled, thus a substantially higher degree of surface contact than is achieved in tumbling or vibratory mills is ensured. Milling takes place by impact and shear forces. The rotating charge of balls and milling product form a whirl where the milling product is impacted by balls moving in various trajectories.
The attrition mill agitator rotates at speeds ranging from 60rpm for production units to 300rpm for laboratory units and uses media that range from 3 to 6mm while ball mills use large grinding media, usually 12mm or larger, and run at low rotational speeds of 1050rpm. Power input to attrition mills is used to agitate the medium, not to rotate or vibrate heavy drums. Therefore, specific energy consumption of attrition mills is significantly less than with ball mills. Table 2.4 offers a comparison of grinding mills by rotational velocity. In the attrition grinding process, grinding time is related to medium diameter and agitator speeds , within given limits, as:
where t is grinding time required to reach a certain medium particle size; k is a constant that varies with the suspension being processed and the type of medium and mill being used; d is the diameter of the grinding medium; and n is shaft movement, in rpm.
Attrittion mills are classified as batch-, continuous-, or circulation-type mills. In the batch mill, material is loaded into the chamber and ground until the desired dispersion and particle size are achieved. Chamber walls are jacketed so that either hot or cold water can be circulated to control and maintain the temperature of the charging. The batch attrition mills can process high-density material, such as tungsten carbide, as well as viscous materials. They are also suitable for dry grinding and for producing dispersion-strengthened metals by means of mechanical alloying.
Continuous attrition mills, more appropriate for large production output, consist of a tall, jacketed chamber through which previously prepared slurry is pumped in at the bottom and discharged at the top. Grids located at the bottom and top retain the grinding medium, as shown in Fig. 2.21.
The circulation grinding system comprises an attrition mill and a large holding tank, generally 10 times the volume of the grinding unit. The attrition mill contains grids that restrain the medium while the slurry passes through. Usually, the contents of the holding tank are passed through the system at a rate of 10 times per hour. The slurry can be monitored continuously and processing is stopped when the desired particle size dispersion is achieved.
Dry milling can provide reduced transportation costs compared to wet grinding because 50% of the gross weight is liquid in many wet slurry processes. Because the removal of the liquid from a wet grinding process involves not only another process step but also requires large amounts of energy, dry grinding can provide reduced energy costs. Another advantage is the elimination of waste liquid disposal.
Attrition mills find application for hard materials such as carbides and hard metals where conventional tumbling and vibratory ball mills are less efficient. The principal advantages of attrition mills for mixing and blending tungsten carbide with cobalt as binder metal cutting tool powders include a short milling time, the production of fine particle size (submicron sizes), and the enhanced coating of cobalt onto the surface of tungsten carbide particles . Attrition mills effectively grind metals in inert atmospheres, such as in solid state or mechanical alloying.
Previous work (Heitzmann ) performed with coloured tracer experiments in a glass body version of a four blades Dyno mill showed that the action of the stirrer - beads system was first to delimit four perfectly mixed cells centered on each of the four blades. Further, it has been shown that classical models (plug flow, cascade of perfectly mixed cells, dispersion models) were unable to correctly represent RTD experiments. An internal recirculation loop model, with a single adjustable parameter R (see Figure 1), was considered and gave very good results in continuous milling of suspensions of known grinding kinetics (Berthiaux et al. ).
Another important conclusion of this work was concerned with the physical meaning of the recirculation ratio R which is undoubtedly linked with the local hydrodynamic conditions, such as porosity, stirrer speed of rotation N, and perhaps mill flow-rate Q. It was also suggested that there exists an optimum value of R that leads to the best continuous grinding conditions (see Figure 2). For example, low values of R benefits the flow itself as it approaches plug flow through tanks in series, while it clearly slows down the kinetics of grinding because the bead - particle collisions are of a lower intensity. In the absence of kinetic data, typical R-values should then range between 0.5 and 5.
The procedure followed to obtain these RTD curves becomes tedious when dealing with a greater number of perfectly mixed cells, or better said a greater number of stirring blades, as it is the case for other types of mills. In general, the analytical or numerical derivation of the RTD from any complex model is in fact highly subjected to errors when done by the classical transfer function method. Particularly, many problems can be incurred when hypotheses are made to simplify the mathematical equations, which may lead to unrealistic dynamic behaviour (see Gibilaro et al. ). The advantage of using the Markov chain approach lies in the fact that it is systematic, and its application does not depend on the complexity of the flow scheme.
A Markov chain is a system which can occupy various states, and whose evolution is defined once an initial state and the probability transitions between the states are fixed. It can therefore be said that a Markov chain does not have memory. In the case of flow problems (Fan et al. ), the system is a fluid element, the states are the perfectly mixed cells of the flow model (as plug flow can be represented by a series of such cells), and the probability transitions are fixed by elementary mass balances.
For example (Figure 3), the probability pii of remaining in cell i is exp(t/i), where t is the time interval under which the system is observed, and i is the geometric residence time corresponding to cell i. The other transitions pij depend then on the flow rate ratios and on the value of 1- exp(t/j), which is the probability of getting out of cell i during t. All these information are then collected in a probability transitions matrix P, whose rows (i) and columns (j) are the pijs. Further, the initial state of the system is represented by a single row E0, being En the state of system after n transitions (steps of duration t), which is available from the following matrix product (Eq.1):
The last element of En, which is the collecting cell or outlet of the network, represents therefore the dynamic response of the system to a perturbation that may be a tracer impulse: E0=[1 0 0]. Simulation of the RTD curve of the model is further performed by letting t become smaller and smaller until the stability of the solution is ensured.
A rod mill is an ore grinding mechanism that uses a number of loose steel rods within a rotating drum to provide its attrition or grinding action. An ore charge is added to the drum, and as it rotates, friction between the tumbling rods breaks the ore down into finer particles. Although similar in operation, a rod mill is often more effective than a ball mill as it requires lower rotational speeds and less steel to achieve the same results. It is, however, limited to maximum rod and drum lengths of approximately 20 feet (6 meters) and is generally only used for wet grinding processes. The rod mill also tends to suffer from accelerated drum liner and lifter wear due to the increased weight of the rods.
Mills of various types have been used for centuries to break solids or coarse particulate materials down into finer finished products. From the humble mortar and pestle through animal, wind, and water driven mills to the giant electrically driven versions common in modern industrial applications, all share one common characteristic: mechanical attrition or grinding. All mill types utilize a grinding process of one or another description to gradually reduce the size of the initial charge of material. In older mills, for example, this action was achieved by placing the coarse material between two mill stones and turning one against the other to produce a finer end product.
Modern rotary mills make apply the same principle by tumbling loose grinding elements around in a closed drum to which the charge material is added. Common examples are rod and ball mills, both of which are of the rotary drum type which rely on internal grinding agents to achieve their milling action. Unlike the ball mill which utilizes a large number of hardened steel balls to impart the grinding action, the rod mill uses steel rods lying within the drum and parallel to its axis. When the drum rotates, these rods roll around inside it, thereby crushing the feed material between them.
The rod mill is generally more efficient than the ball mill due to its more effective cascading action and the greater bearing surface offered by the rods. This means it can operate at lower speeds and with less grinding agents and producing less undesirable slimes byproduct. Rod mills do, however, require more attention during operation to prevent rod tangles and are generally ineffective at dry milling operations. They are also limited to a maximum rod length of approximately 20 feet (6 meters) which means they are generally smaller than ball mills. Rod mills also exhibit more liner and lifter wear than other mill types due to the comparatively high weights of the rods.
The following equation is used to determine the power that a rod mill should draw.The BottomTable lists many of the common size rod mills giving speed, loading and power data. The rod mill motor power is in horsepower at the mill pinion-shaft. For different length rod mills power varies directly as rod length.For difference between new and worn liners increase power draw by 6%, and adjust for bulk density per Table A.
Wet grinding rod mills are normally used in minerals processing plants. Experience with dry grinding generally indicates many difficult problems and should be avoided except where absolutely necessary; in which case the problem should be referred to the mill manufacturers for recommendations.
The relative efficiency for size reduction of a given material from a constant feed size to a constant product size can readily be seen as linearly proportional to the tonnage treated (directly) or the power consumed (inversely). Bond (1961) observed that material size distributions as naturally produced through stages of comminution were highly consistent, and therefore could be represented by a single value such as the 80% passing size. Secondly, from numerous observations, he derived the following approximate relationship between the per unit work applied by the comminution device and the amount of size reduction achieved, commonly referred to as the Bond Law of Comminution.
The constant in this equation is tensed the Bond Operating Work Index, Wio. Assuming Wio does not change as a function of the feed or product sire, and the ore itself has not changed, it is a direct measure of circuit efficiency, variations in circuit feed and product size taken into account.
Bond Laboratory Work Index determinations (one for each type of comminution machine) are widely used for mill sizing, but may also be used to monitor ore grindability characteristics in circuit evaluation work. In the latter case, correction factors needed for reasonably accurate scale-up to industrial size mills under various design conditions may be factored out of the basic Bond Operating Work Index, to derive the Corrected Operating Work Index, for direct comparison with the Laboratory Work Index for the same ore.
Note, however, that this corrected figure is much less meaningful than the basic Bond Operating Work Index for circuit performance evaluation purposes because it no longer pertains to the actual power efficiency that the circuit, as designed and operated, is providing. Within the limitations of the accuracy of the assumptions previously mentioned, the uncorrected Bond Operating Work Index is the true indicator of basic relative circuit efficiency, and is therefore used throughout this paper.
Precise descriptions of procedures for data acquisition are not generally included in reports on plant operating experiences, and neither is ore grindability normally monitored during plant testwork. However, it is nevertheless reasonable to reach clear conclusions about changes in relative circuit efficiency, as defined by the Bond Operating Work Index, by sensible interpretation of consistently, if not accurately, collected and reported operating data.
A table of estimates of rod mill power draw is published in Rowlands work (1982). Figures in this table are generally slightly low compared to some reported plant data or other manufacturers estimates, and are based on an empirical formula initially proposed by Bond (1961) designed to cover a wide range of mill dimensions, and the normal operating range of mill load (Vp = .35 to .50) and speed (Cs = .50 to .80). Please see the reference article for S.I. equivalents.
Note that the above equation (3), was derived to empirically relate power consumption over a wide range of mill sizes, on various slurries, as well as over a fairly broad range of mill speed and load levels, all combined into a single expression. Effects such as slurry density, media size, liner condition, and trunnion opening size are ignored. Although it still provides a close approximation, the limits of accuracy, easily plus or minus 5 to 10%, should be recognized. As well, interpretation of the concise effect of any single variable taken out of the context of the complete equation may be misleading.
Refractory ore processing methods almost always serve only one purpose, to treat ores that will not liberate their values by conventional cyanide leaching. The refractory ore treatment process is then followed by a conventional cyanidation step. Refractory ore processing methods include:
Today, cyanide leaching is the method of choice for the recovery of most of the worlds gold production. There are however, many other chemical leaching processes that have been sporadically or historically used. In most instances, cyanide leaching will provide a more technologically effective and cost efficient method. Alternative lixiviants include:
Amalgamation is one of the oldest processes available. It relies upon the contact of ore with mercury to form a gold-mercury amalgam. This process is strongly out of favor with the major mining companies, due to the extremely toxic nature of mercury and the processes inferior performance when compared to the available alternatives. The process is still used extensively by artesian mines in third world countries and at small mom and pop mines, due to its simplicity.
Gravity concentration processes rely on the principal that gold contained within an ore body is higher in specific gravity than the host rocks that contain the gold. Elemental gold has a specific gravity of 19.3, and typical ore has a specific gravity of about 2.6. All gravity concentration devices create movement between the gold and host rock particles in a manner to separate the heavy pieces from the lighter pieces of material.
The prospectors gold pan is the most familiar gravity concentration device. To function properly, the ore must be broken down to particles small enough to provide a significant specific gravity difference among the particles.
Placer mining has generally been where gravity concentrates have been most widely applied. In a placer deposit, there has generally been a pre-concentration of gold made naturally by gravity concentration due to ore particles being transported by water. Mechanical concentration is used to continue the process until sufficient concentration is obtained.
The flotation process consists of producing a mineral concentrate through the use of chemical conditioning agents followed by intense agitation and air sparging of the agitated ore slurry to produce a mineral rich foam concentrate. The process is said to have been invented by a miner who watched the process happening while washing dirty work clothing in his home washing machine.
Specific chemicals are added to either float (foam off) specific minerals or to depress the flotation of other minerals. Several stages of processing are generally involved with rough bulk flotation products being subjected to additional flotation steps to increase product purity.
The flotation process in general does not float free gold particles but is particularly effective when gold is associated with sulfide minerals such as pyrites. In a typical pyrytic gold ore, the gold is encapsulated within an iron sulfide crystal structure. Highly oxidized ores generally do not respond well to flotation.
Advantages of the flotation process are that gold values are generally liberated at a fairly coarse particle size (28 mesh) which means that ore grinding costs are minimized. The reagents used for flotation are generally not toxic, which means that tailings disposal costs are low.
Flotation will frequently be used when gold is recovered in conjunction with other metals such as copper, lead, or zinc. Flotation concentrates are usually sent to an off-site smelting facility for recovery of gold and base metals.
Cyanide leaching is the standard method used for recovering most of the gold throughout the world today. The process originated around 1890 and quickly replaced all competing technologies. The reason was strictly economical in nature. Where amalgamation plants could recover about 60% of the gold present, cyanide could recover about 90%. Because of the improved recovery, many of the old tailings piles from other processes have been economically reprocessed by cyanide leaching. Cyanide is as close to a universal solvent for gold as has been developed. Other leaching reagents will only work on very specific types of ore.
The standard cyanide leach process consists of grinding the ore to about 80% 200 mesh, mixing the ore/water grinding slurry with about 2 pounds per ton of sodium cyanide and enough quick lime to keep the pH of the solution at about 11.0. At a slurry concentration of 50% solids, the slurry passes through a series of agitated mixing tanks with a residence time of 24 hours. The gold bearing liquid is then separated from the leached solids in thickener tanks or vacuum filters, and the tailings are washed to remove gold and cyanide prior to disposal. The separation and washing take place in a series of units by a process referred to as counter current decantation (CCD). Gold is then recovered from the pregnant solution by zinc precipitation and the solution is recycled for reuse in leaching and grinding.
One major category of refractory ores are gold values contained within the crystalline structure of sulfide minerals such as pyrite and arsenopyrite. For cyanide to leach gold, the cyanide solution must come into direct contact with gold molecules. With many sulfide ores, the ore cannot practically be ground down fine enough to expose the gold particles. The objective of pretreatment for these ores is to remove enough of the sulfide so that at least a small portion of all gold particles are directly exposed to the elements. Processes available for treatment all involve oxidation of sulfur to form water soluble sulfates or sulfur dioxide. The main sulfur oxidation processes include:
Heap leaching was introduced in the 1970s as a means to drastically reduce gold recovery costs. This process has literally made many mines by taking low grade geological resources and transforming them to the proven ore category. Ore grades as low as 0.01 oz Au per ton have been economically processed by heap leaching.
Heap leaching involves placing crushed or run of mine ore in a pile built upon an impervious liner. Cyanide solution is distributed across the top of the pile and the solution percolates down through the pile and leaches out the gold. The gold laden pregnant solution drains out from the bottom of the pile and is collected for gold recovery by either carbon adsorption or zinc precipitation. The barren solution is then recycled to the pile.
Heap leaching generally requires 60 to 90 days for processing ore that could be leached in 24 hours in a conventional agitated leach process. Gold recovery is typically 70% as compared with 90% in an agitated leach plant. Even with this inferior performance, the process has found wide favor, due to the vastly reduced processing costs compared with agitated leaching.
Quite frequently, mines will use agitated leaching for high grade ore and heap leaching for marginal grade ores that otherwise would be considered waste rock. A common recovery plant is often employed for both operations.
The traditional method for gold recovery from pregnant cyanide solutions is zinc precipitation. Originally, solutions were passed through boxes containing zinc metal shavings. Gold and silver would precipitate out of solution by a simple replacement reaction procedure. Around 1920, zinc shaving precipitation was replaced by the Merrill-Crowe method of zinc precipitation.
The Merrill-Crowe process starts with the filtration of pregnant solution in media filters. Filter types used include pressure leaf filters, filter presses, and vacuum leaf filters. Generally, a precoat of diatomaceous earth is used to produce a sparkling clear solution.
Granular coconut shell activated carbon, is widely used for recovery of gold from cyanide solutions. The process can be applied to clean solutions through fluidized bed adsorption columns, or directly to leached ore slurries by the addition of carbon to agitated slurry tanks, followed by separation of the carbon from the slurry by coarse screening methods.
Gold cyanide is adsorbed into the pores of activated carbon, resulting in a process solution that is devoid of gold. The loaded carbon is heated by a strong solution of hot caustic and cyanide to reverse the adsorption process and strip the carbon of gold. Gold is then removed from the solution by electrowinning. Stripped carbon is returned to adsorption for reuse.
The major advantage of carbon-in-pulp recovery over Merrill Crowe recovery is the elimination of the leached ore solids and liquid separation unit operation. The separation step typically involves a series of expensive gravity separation thickeners or continuous filters arranged for countercurrent washing or filtration of the solids. For ores exhibiting slow settling or filtration rates, such as ores with high clay content, the countercurrent decantation (CCD) step can become cost prohibitive.
Ores with high silver content will generally suggest that Merrill-Crowe recovery be used. This is because of the very large carbon stripping and electrowinning systems required for processing large quantities of silver. The typical rule of thumb states that economic silver to gold ratios of greater than 4 to 1, will favor installation of a Merrill-Crowe system, but this decision can be altered if the ore exhibits very slow settling rates.
1. Carbon-In-Column (CIC): With carbon-in-column operation, solution flows through a series of fluidized bed columns in an upflow direction. Columns are most frequently open topped, but closed top pressurized columns are occasionally used.
Carbon columns are most commonly used to recover gold and silver from heap leach solutions. The major advantage of fluidized bed carbon columns is their ability to process solutions that contain as much as 2 to 3 wt% solids. Heap leach solutions are frequently high in solids due to fine particle washing from heaps. Down flow carbon columns are rarely used for gold recovery, because they act like sand filters and are subsequently subject to frequent plugging.
2. Carbon-In-Pulp (CIP): Carbon-in-pulp operation is a variation of the conventional cyanidation process. Ore is crushed, finely ground, and cyanide leached in a series of agitated tanks to solubilize the gold values. Instead of separating solids from the pregnant solution, as in the traditional cyanidation process, granular activated carbon is added to the leached slurry.
The carbon adsorbs the gold from the slurry solution and is removed from the slurry by coarse screening. In practice, this is accomplished by a series of five or six agitated tanks where carbon and ore slurry are contacted in a staged countercurrent manner.
This greatly increases the possible gold loading onto the carbon while maintaining a high recovery percentage. Carbon is retained within the individual CIP tanks by CIP tank screens. The opening size of the CIP tank screens is such that the finely ground ore particles will pass through the screens, but the coarse carbon will not. Almost every imaginable type of screen has been tried for this application, with some types being much more successful than the rest.
3. Carbon-In-Leach (CIL): The carbon-in-leach process integrates leaching and carbon-in-pulp into a single unit process operation. Leach tanks are fitted with carbon retention screens and the CIP tanks are eliminated. Carbon is added in leach so that the gold is adsorbed onto carbon almost as soon as it is dissolved by the cyanide solution. The CIL process is frequently used when native carbon is present in the gold ore. This native carbon will adsorb the leached gold and prevent its recovery. This phenomenon is referred to commonly as preg-robbing. The carbon added in CIL is more active than native carbon, so the gold will be preferentially adsorbed by carbon that can be recovered for stripping. The CIL process will frequently be used in small cyanide mills to reduce the complexity and cost of the circuit.
There are several disadvantages to CIL compared with CIP. Carbon loading will be 20 to 30% less than with CIP, which means more carbon has to be stripped. (This disadvantage may be overcome by a hybrid circuit, incorporating a cross between CIL and CIP.) The CIL process requires a larger carbon inventory in the circuit, which results in a larger in-process tie up of gold. The larger carbon inventory can also result in higher carbon (and gold) losses through carbon attrition.
Denver Mineral Engineers has had extensive experience with all of the commercially viable gold and silver recovery mining processes. We can suggest the optimal process and equipment for virtually any ore. Although we are not a testing laboratory, we can design and coordinate your testing program. If we dont have the answers, our network of industry experts can be utilized.