A METHOD FOR SEPARATING METALS FROM ALLOYS
The invention relates to a method for separating a hypo eutectic molten alloy comprising a base metal A and an alloying element B into two melts, namely a melt comprising solely A and a melt comprising an alloy enriched with B in an amount up to the eutectic composition. The method according to the invention is particularly suitable for use when refining metal with respect to elements more noble than the metal in question, the elements forming with the metal a eutectic, preferably a eutectic of relatively low melting point, and being present in the metal alloy in amounts smaller than those which correspond to the eutectic composition. As is well known, noble elements cannot be separated from metals by oxidation, which is otherwise a simple method of separation much used when refining iron and non-ferrous materials. Consequently, when refining metals in order to recover the noble elements contained therein, it is necessary to find other, more sophisticated methods. Some of these methods are based on separating crystals from molten baths by cooling.
It is a well known fact, and one applied in practice, that those crystals separated from a hypoeutectic melt of a metal alloy contain less of the alloying metal, and thus afford a possibility of purifying the base metal or of concentrating the alloying metal in a molten bath, up to the eutectic composition. For example, it is possible to separate pure lead from silver, or to concentrate silver from a lead bath using the Pattison process, in which a molten bath of silver-containing lead is partially solidified to separate lead in a purer form from the bath, while concentrating the silver in the bath residue. Subsequent to tapping-off the lead, the process is repeated until there is obtained a eutectic lead-silver-alloy containing about 2.5% silver, from which the silver can be recovered in a pure form after driving off the lead. The lead
crystals, wliich are of a purer quality, are melted down and treated in a simple manner, and the process is repeated a number of times to obtain a lead which is free from silver. This process, however, is extremely uneconomic and impracticable, because of the large quantities of energy consumed and the large work force required. Consequently, the Pattison process has long been abandoned in favour of the Parke s process for desnlvering lead, in which process zinc is added to the molten bath. This process, however, is also particularly compli cated and difficult to adapt to present day requirements with regard to productivity and a healthy working environment.
Zone refining is another example of the use to which the aforementioned principle can be put. In zone refining processes, a melted zone is caused to pass slowly along a length of solid metal. As the zone moves along the. solid metal, a pure metal is constantly separated on the side of the zone where the metal solidifies, while substantially all the impurities remain In the molten zone and accompany the zone as it moves along said length of metal. Although this method is very effective, it is also expensive and time-consuming, and consequently the method has been limited to the super- refinement of such elements as silicon and germanium for electronic purposes.
A theoretically possible method has been proposed for continously dividing an alloy in one such manner. The principles on which this theoretical process is based are set forth hereinafter. Assume that there is produced an alloy comprising a base metal A and an alloying metal B and that these metals form an eutectic phase diagram of the kind illustrated in Figure 1. When a molten bath of the alloying metal having composition l1 is cooled, a small amount of crystals of composition s1 will separate at temperature t1. When these
crystals are transferred to a bath of composition l2 having a lower content of B than l1 and haying a temperature t2 somewhat higher than t1, the crystals will find equilibrium with the bath and obtain the composition s2. The crystals therefore contain less of the alloying metal B than the earlier separated crystals of composition s1. When these new crystals are transferred to a bath of composition l3 at a temperature t3, the reaction course taken will be the same as that taken previously, to pro vide crystals of composition s3. Subsequent to repeating the procedure several times, constantly transferring the crystals in the aforesaid fashion, such that said crystals find equilibrium wfbh baths of decreasing B-content and increasing temperature, the crystals will e-yentually comprise pure A and can be removed from the system at the temperature ta, either as a solid phase or a liquid phase, since the surrounding melt also comprises pure A.
After the first crystallization of A, the bath residue can be treated in a corresponding fashion. In this respect, the bath residue can be brought into contact with crystals of composition s12, and having a temperature of t12. In order for equilibrium to be achieved, the R-content l12 of the bath must be slightly higher than l1. This procedure can also be repeated a number of times, such that the bath constantly obtains equilibrium with, crystals successively richer in B, until the bath, finally obtains the- desired B-content, which at most can correspond to the composition of the eutectic; the temperature falling, at the same time, to the lowest melting point teut of the eutectic.
In Australia (CSIRO) tests have been carried out with a method based on the principles set forth in the intro auction, in which a metal smelt is held in a vertical vessel, where the temperature varies from tA at the lower end of the vessel to tAB, the melting point of the desired alloy, at the upper end of said vessel, with
a constant temperature gradient through the vessel. According to the method, crystals forming in the melt fall slowly gravi tationally towards the warmer end (tA) of the vessel, where the composition of the crystals changes in the aforedescribed fashion. At the same time, the melt in equilibrium with the metal crystals moves towards the colder end (tAB) of the vessel and therewith determines the desired concentration gradient. The end products can be taken out at respective ends of the vessel. The flow of alloy to the vessel must take place at that location in said vessel where theraelt has a corresponding composition. By using gravity as the driving force for moving the crystals, it should be possible when applying the theoretically proposed principles discussed above, to carry a smooth, continuous method into effect in a simple and ready fashion, since metal crystals separation from the melt are normally heavier than the mother raelt. It is also possible to maintain with ease a downwardly decreasing temperature gradient, for exaraple in a cylindrical crucible having a vertical axis. When the crystals fall to the bottom of the crucible, the melt should be displaced upwardly therein. The method is disclosed in the US patent publications US,A, 4043802, 4 133 517 and 4 138 247.
When putting this method into practice, however, difficulttes were encountered in obtaining systematic movement of the metal crystals. The relatively small difference in relative density was probably not sufficent to move a mass of minute crystals in counterflow with the melt. Neither was it possible to influence the quantity and magnitude of the crystals which need to form in order for the desired course of events to take place, since it is not possible to cool the crucible extern-ally without forming crystals on the crucible walls, which must be prevented.
An object of the invention is to provide a method in which the aforementioned disadvantages are eliminated. Accor
dingly the invention proposes a method of separation in which the movement of the crystals, their growth and the quantity in whi ch they are formed can be controlled by moving the crystals mechanically through the melt. The method according to the invention is characterized by the procedural steps set forth in the accompanying claims. As will be seen, an important characteristic of the invention resides in effecting separation in an elongate reactor vessel in which the melt can be cooled internally, i.e. not through the walls of the reactor vessel, and the crystals are moved mechanically in a desired direction and at a desired speed. In a preferred embodiment of the invention, the crystals are caused to form on a substrate comprising a bunch of narrow pipes through- passed by a temperature-controlling medium. The melt is located in an elongate, preferably horizontal reactor vessel, and one end of the melt is maintained at a temperature tA and the ottier end of said melt at a temperature tB by supplying heat from without, with a uniform terapera ture fall within the vessel.
The invention will be described in more detail hereinafter with reference to a preferred .embodiment employing bunches of pipes for cooling the melt and transporting the crystals. The pipe bunches are caused to move through the melt or bath in a direction away from the colder end to the warmer end. By adjusting the surface temperature of the pipes, for example by passing different, adjusted flows of coolant therethrough, there is formed a thin coating of crystals on the surface of the pipes. The remainder of the process is similar to that described in the introduction. A given portion of the pure A-metal must be allowed to flow back at the warmer end, in order to obtain a back-flow of melt in counterflow to the crystals.
In addition to the controlled movement of the crystals
through the melt of varying composition, the invention also affords the advantage whereby the crystals can be maintained at a temperature slightly beneath the temperature of the melt, owing to the fact that the temperature of the pipe bunches can be controlled and regulated. As a result of their constant movement through the vessel, the crystals will always be surrounded by a melt whose equilibrium solidus has a slightly lower B-content than the crystals, causing the melt to dissolve metal, preferably B-metal, from the crystals. Because the crystals are maintained at a temperature which is slightly lower than the temperature of the melt, a further quantity of metal poor in B will be separated from the melt, this quantity corresponding to the amount dissolved from the crystals. It is possible in this way to maintain a crystal layer of constant thickness. Since the heat-consuming dissolution of metal from the crystals takes place simultaneously with the heat-emitting crystali zati on process, a thermal balance is obtained. The temperature of the vessel walls must lie immediately above the liquidus. temperature, so as to prevent crystals deposing on the walls; the unavoidable excess of heat is compensated here by the .cox.! ant flowing through the pipes.
In order to obtain effective division of the melt it has been found necessary to obtain a given back-flow of molten material. This back-flow should be at least 20% of the amount of metal advanced in crystal form, which means that it is not possible to enrich a metal more than five times.
A preferred embodiment of the invention carried out in a reactor vessel having the form of a circular trough for refining lead with respect to silver will be discussed hereinafter in detail with reference to Figure 2, whi ch is a schematic top-plan view of a horizontal trough. The molten
bath of silver-containing lead, heated to an equilibrium temperature t, for separating crystals of composition s1, is introduced into a trough 1, as shown by the arrow labelled " ING.Pb,t1 oC Located in the trough 1 is a bath 2 of molten lead. The temperature of the bath 2 is controlled and regulated by means of heating elements (not shown), connected to the trough, such that the temperature at one end 3 of the-trough 1, called the colder end, is about 305° C, while the temperature at the other, warmer end 4 of the trough is somewhat higher than the melting point of lead, namely about 328°C, said trough ends being mutually separated by a bath-impermeable heat-insulating partition 5. The temperature between the ends 3 and 4 is controlled so as to obtain a substantially linear temperature gradient therebetween. Crystals of composition s1 corresponding to the solidus at at temperature t1 are separated from the bath onto a cooling means 6, having, for example, the form of a suitable helical metal pipe, which is moved continuosly or intermittently around the trough towards the warmer end 4, as indicated by the arrow 7. A coolant, such as air, is passed to the helical pipe through stationary, flexible supply means 8a, 8, 9. Although only the cooling means 6 are shown, it will be understood that a greater number of such cooling means are distributed over the entire periphery of the trough in substantially uniform spaced relationship. When a cooling means 6 reaches the warmer end 4, the cooling means is heated to at least 328οC and then lifted from the bath, thus being liberated from the crystals being formed on said cooling means, and which are of pure lead practically free from silver, having melted and having been tapped-off from the trough, as indicated by arrow 10. When the cooling means 5 has been freed from its coating of crystals, said means is immersed in the cold end 3 of the trough to begin a further crystallization cycle. Because the crystals
separated from the bath are forced to move clockwise by the action of the cooling means 6, the melt in which the silver is concentrated is forced to move counter clockwise in the trough, as shown by the arrows labelled (1). The silver-enriched melt is removed from the trough at the cold end 3, as shown by the arrow 11.
EXAMPLE
Experiments in which lead was delivered were carried out in an apparatus having the form of a metal trough, such as that illustrated in Figure 2. Lead containing 857 grams of silver per ton was used in the experiments. It was found that the lead could be desilvered satisfactory by regulating the cooling of the pipe bunches so as to form a crystal layer having a thickness of about 0.5mm, which corresponds to about 5 kg of lead per m 2 of the operative surface of the pipe bunches. The pipe bunches were moved at a vertical velocity of 0.8m/min. With a pipe-surface area of 20 m2 per meter of length of the metal trough it is possible to introduce into the trough 100kg of silver- containing lead per minute, and to remove from said trough 80% = 80kg of lead per minute, i.e. the apparatus has a capacity of 4.8 tons per hour or 115 tons per day.
Division of the silγer-containing lead was effected in two stages. In the first stage relatively pure lead containing about 1 gram of silver per ton was recovered, together with a silver-enriched alloy containing 4055 grams of silver per ton. This alloy was treated in the second stage, in which there was obtained a silverdepleted alloy containing 910 grams of silver per ton, together with an alloy rich in silver, containing 21500 grams of silver per ton.
The energy consumed in this division of the lead-silver alloy was estimated to be between 20 and 30 KWh per ton of
l ead at each s tage .
Although the division of metal alloys in accordance with the invention does not enable complete separation of the alloy constituents, it does afford a relatively simple method of concentrating the alloying metals in a minor portion of the amount of base metal contained; this enables complete separation processes, such as electrolysis, which are relatively expensive, to be restricted to this minor part of the base metal content.
In addition to the lead-silver alloys discussed above, where complete separation of the ingredients is expensive and difficult to carry out, the present invention also affords the possibility of separating the ingredients of other binary alloys. Examples of such binary alloys which can be effectively treated in accordance with the invention are given below:
Base Metal Alloying Metal (s) Lead Bismuth
Aluminium Silicon, iron, magnesium
Copper Silver, gold
Zinc Cadmium