Cross Reference to Related Application [0001] This application claims the benefit of U.S. provisional patent application No. 60/592,375 filed on August 2, 2004, the disclosure of which is hereby incorporated herein by reference. Field of the invention [0002] The present invention relates to the hydrometallurgical processing of nickeliferous laterite ore, and in particular to a method for acid leaching both the limonite fraction and the saprolite fraction of such ores in a single process. Background of the invention [0003] Laterite ores are formed by the in-situ weathering of nickel- bearing ultramafic rocks near or at the surface of the earth in tropical environments by the action of naturally acidic meteoric waters over geologic time. They consist of a variety of clay, oxide and silicate minerals, some enriched in nickel and/or cobalt, and this distinguishes them from the other major class of nickel ores, the sulfide ores. The latter consist typically of sulfide minerals of iron, nickel and cobalt, often with copper and minor precious metals, and are associated with mafic-ultramafic magmatic intrusions in the earth's crust.

[0004] The weathering process typically creates a layered deposit, with the products of complete or most extensive weathering occurring near the surface and these grading into the products of lesser degrees of weathering as depth is increased and finally terminating in unweathered rock at some greater depth. The highly weathered layer usually contains most of its contained nickel microscopically distributed within very finely divided goethite particles. Goethite is an oxyhydroxide of ferric iron with the chemical formula FeOOH. This layer is usually given the name limonite, and typically contains a high proportion of iron. [0005] Cobalt is usually associated with the limonite layer and is usually predominantly associated with oxidized manganese minerals (Mn(III) and/or Mn(IV) containing oxides and hydroxides), often called asbolane or manganese wad.

[0006] The lesser weathered layers typically contain increasing proportions of their contained nickel in various magnesium silicate minerals, such as, for example, serpentine. Serpentine is a silicate mineral of magnesium which has the chemical formula 3MgO*2Siθ2*2H2O. It is believed that nickel substitutes for some of the magnesium in serpentine. Magnesium may also be substituted by other divalent metals, for example, ferrous iron (Fe2+). There may be many other silicate minerals that also host nickel in the incompletely weathered zones. The partially weathered, high-magnesium bearing zone is often given the name saprolite, or garnierite. ("Garnierite" is also used to describe a particular apple-green colored magnesium-nickel silicate mineral of variable composition.)

[0007] In some deposits there is another zone typically located between the limonite and saprolite that consists predominantly of nontronite clays; these are silicates of magnesium, iron and aluminum that may also be nickeliferous. In most deposits located in the (current) tropics, the nontronite zone is largely absent.

[0008] It should be noted also that none of the weathering zones are homogeneous in mineralogical or chemical composition, nor is the boundary between the zones parallel to the earth's surface. However, there is usually a fairly sharp transition from ore of high iron and relatively low magnesium contents to ore of a relatively high magnesium content and lower, although variable, iron content, which occurs over vertical distances of 1 to 3 m within a laterite deposit.

[0009] For illustration purposes only, typical chemical compositions of limonite and saprolite are as follows: [0010] Limonite: 1.0-1.8 % Ni, 0.05-0.3 % Co, 35-50 % Fe, 0.2-3.5 % Mg

[0011] Saprolite: 1.2-3.5 % Ni, 0.02-0.07 % Co, 7-20 % Fe, 10-20 % Mg

[0012] Each zone also contains typically significant concentrations of aluminum, manganese and chromium, as well as trace concentrations of other heavy metals such as copper and zinc in a variety of other minerals.

[0013] A challenging aspect of nickel recovery from laterite ores is that the nickel values typically can not be concentrated substantially by physical means, that is, so-called ore dressing techniques, prior to chemical processing to separate the metal values. This renders the processing of laterites expensive, and means to lower the costs of processing laterites have been sought for many decades.

[0014] Also, because of the distinct mineralogical and chemical composition of limonite and saprolite ores, these ores usually are not amenable to processing by the same process technique.

[0015] One known acid leaching process for nickel laterites is the so- called High Pressure Acid Leaching (HPAL) process (see, for example pages 437-453 in "The Winning of Nickel Its Geology, Mining and Extractive Metallurgy," by Joseph R. Boldt, Longmans Canada Ltd. 1967). This process was first employed at Moa Bay in Cuba in the late 1950s and additional plants were constructed in Western Australia in the late 1990s.

[0016] The process utilizes sulfuric acid leaching at high temperature, typically 250 0C, and high pressure; the associated steam pressure at 250 0C is approx. 570 psi or 39 atmospheres. At this temperature, the nickeliferous minerals in the ore are nearly completely solubilized. The dissolved iron is rapidly precipitated as hematite (Fe∑Oβ) at the high temperature employed because this compound is largely insoluble even in slightly acidic solutions at this temperature. The nickel remains in solution and after cooling, the leach residue containing iron is separated from the nickel-bearing solution by thickening in a series of wash thickeners, a so-called counter-current decantation (CCD) circuit. Thus, the primary objective of the leaching process, which is separation of nickel from iron, is achieved.

[0017] A major disadvantage of the HPAL process is that it requires sophisticated high-temperature, high-pressure autoclaves and associated equipment which are expensive, both to install and to maintain. In addition, the HPAL process also consumes more sulfuric acid than is required to stoichiometrically dissolve the non-ferrous metals content of the ore because at high temperature most of the sulfate ions provided by sulfuric acid are tied up as bisulfate ions (HSO4"). In other words, sulfuric acid (H2SO4) only dissociates to release a single proton (H+) at high temperature. On cooling and neutralization of the leach liquor the bisulfate ions decompose to sulfate (SO42") and another proton. The latter proton (acid) is therefore not utilized fully for leaching and results in excess sulfuric acid which must be neutralized, for example with limestone.

[0018] Another disadvantage of the HPAL process is that it is limited to treating largely limonite-type feeds because the presence of saprolite will cause a large, and often uneconomic, increase in sulfuric acid consumption due to the leaching of magnesium from saprolite. This is exacerbated by the bisulfate "shift" problem at high temperature, which is described above.

[0019] U.S. Pat. No. 4,097,575 describes an improvement to the HPAL process which constitutes roasting saprolite ore below about 820 0C in order to render the ore more reactive with sulfuric acid and then using the roaster calcine to neutralize excess acid in the discharge of an autoclave wherein pressure leaching of limonite ore occurs. Nickel contained in the saprolite ore is largely dissolved during this neutralization. The reported advantages of this process are that it better utilizes the sulfuric acid added during pressure leaching of limonite, it reduces the consumption of limestone or other costly neutralizing agents to treat the autoclave discharge liquor, and it achieves the capability of treating both the limonite and saprolite fractions of a typical nickel laterite ore body. Disadvantages of the process are that it still requires the use of expensive autoclaves for leaching limonite, and it requires a roasting process for saprolite ore, which is expensive both in capital and operating cost terms.

[0020] U.S. Pat. No. 6,379,636 B2 describes a further improvement to the process described in U.S. Pat. No. 4,097,575 wherein the saprolite roasting step is eliminated and the saprolite in "raw" form is used to neutralize the excess acid in the autoclave discharge solution. In addition, more acid could be added to the discharge to increase the amount of saprolite that could be leached. However, this process still requires the use of expensive autoclaves.

[0021] Several processes have also been described that utilize acid leaching at atmospheric pressure only, eliminating the disadvantages of pressure leaching described above. U.S. Patent No. 3,793,432 describes an atmospheric leaching process for laterite ore, in which the ore is reacted with sulfuric acid at or below the boiling point and the precipitation of dissolved iron is achieved by the addition of an iron precipitating agent such as ammonium, sodium, potassium or lithium ions. Although not stated explicitly, all of the examples cited in the specification employed limonite ore samples, as evidenced by the high iron content and low magnesium content of the feed ore. While this process overcomes the disadvantages of pressure leaching, it has other disadvantages. First, the precipitation of iron is as a jarosite compound, which is a thermodynamically unstable compound of iron that decomposes over time to release sulfuric acid, thus causing environmental problems. (Although jarosite is not stated explicitly it would be apparent to one skilled in the art that jarosite will precipitate at the conditions outlined in the examples). Jarosite contains two moles of sulfate for every three moles of iron and thus this compound represents substantial excess consumption of sulfuric acid to provide the necessary sulfate ions.

[0022] Second, the nickel extractions from the ore were apparently relatively low. While extractions were not stated explicitly, based on the nickel content of the residue and the fact that the residue weight must be more than the weight of the original ore because jarosite contains a lower percentage of iron than the original ore and virtually all of the iron remained in the residue, nickel extractions were usually in the 60-65 % range.

[0023] Third, there is a requirement for very long leach times, of the order of 4-5 days. Fourth, there is a need to add relatively expensive iron precipitating agents such as potassium carbonate, sodium carbonate or the like.

[0024] U.S. Patent Nos. 6,261 ,527 B1 and 6,680,035 B2 describe another atmospheric leaching process in which limonite ore is first "totally" leached with strong sulfuric acid, i.e. both nickel and iron are substantially dissolved from goethite, and then saprolite ore is leached in the resulting limonite leach slurry while simultaneously precipitating iron as jarosite by the addition of a jarosite precipitating agent. This process also has the disadvantage of producing jarosite.

[0025] WO 03/093517 A1 describes an improvement to this process, which constitutes eliminating the addition of a jarosite-forming ion such as sodium, potassium or ammonium, and causing the iron to precipitate as a compound other than jarosite, such as goethite. The process overcomes the disadvantages of jarosite, but sulfuric acid consumption was 0.59 to 0.87 tonnes per tonne of ore in the examples cited, and was over 0.72 tonnes per tonne of ore in nine of the eleven examples cited.

[0026] The processes described in U.S. Patent Nos. 6,261 ,527 B1 and 6,680,035 B2 and WO 03/093517 A1 are based on the fact that goethite is more refractory to acid leaching than typical saprolite minerals, such as serpentine. This has been demonstrated by other researchers (see, for example: John H. Canterford, "Leaching of Some Australian Nickeliferous Laterites with Sulfuric Acid at Atmospheric Pressure," Proc. Australasian Inst. Min. Metall., 265 (1978), 19-26; N. M. Rice and L.W. Strong, "The Leaching of Lateritic Nickel Ores in Hydrochloric Acid," Canadian Metallurgical Quarterly, 13(3)(1974), 485-493; and Figure 5 of U.S. Patent No. 5,571,308). Thus, only saprolite can be used effectively in the second stage of leaching where iron precipitation occurs simultaneously. This is because the acidity of the solution must be relatively low to enable the precipitation of jarosite and even lower to enable the precipitation of goethite or other hydrolysis products of ferric iron. The goethite contained in limonite would leach very slowly under these conditions. Limonite (principally goethite) is thus leached in an initial stage with a relatively high acid concentration and both iron and nickel are brought into solution.

[0027] Another disadvantage of the processes described in U.S. Patent Nos. 6,261 ,527 B1 and 6,680,035 B2 and WO 03/093517 A1 is that the leach process is slow. Greater than 10 hours leach retention time is required to complete the reactions. Thus, many large leach reactors are required to carry out the process and this increases the capital and operating costs of the process compared to a leach process which has a much shorter retention time.

[0028] The atmospheric leach processes described above address the disadvantages of high pressure leaching but have significantly lower nickel extraction (typically 80-85 % for atmospheric leaching versus 90-97 % for high pressure acid leaching). The object of the present invention is to obviate or mitigate the disadvantages of high pressure acid leaching processes while achieving higher and faster recoveries of nickel and cobalt than the known atmospheric leach processes.

Summary of the invention [0029] The present invention provides a process for the efficient leaching of nickel and cobalt from limonite and saprolite ores in two stages, the first stage consisting of mixing and reacting a slurry of the limonite ore with concentrated mineral acid at atmospheric pressure, and the second stage consisting of adding saprolite ore to the resulting leach slurry and then leaching at a moderately elevated temperature and pressure. Iron is efficiently separated from nickel and cobalt in the solid leach residue primarily as an oxide and/or hydroxide of ferric iron other than jarosite. [0030] The limonite leach step should be carried out at a temperature close to the boiling point. The quantity of acid added should be approximately that required to stoichiometrically dissolve the soluble non-ferrous metals in the ore as well as the soluble iron, and a small excess of acid may advantageously be used to drive the dissolution reactions. Preferably, a reductant, such as sulfur dioxide or ferrous sulfate, is added to the limonite leach slurry in order to enhance the dissolution of nickel, and particularly cobalt.

[0031] In the saprolite leaching stage, the temperature should be high enough to achieve a rapid rate of reaction and satisfactory nickel (and cobalt) extraction, but low enough that the resulting working pressure is within the tolerance of a simple, low cost autoclave. The working pressure of the autoclave is approximately equal to the pressure of saturated steam at the working temperature employed. This pressure increases very rapidly with increasing temperature, especially much above about 150 0C. To avoid the complexity and difficulty of operating at excessive pressures, an appropriate range of temperature for carrying out the second stage of leaching in the present invention is from about 120 to 160 0C, and the temperature should be kept as low as reasonably possible consistent with good process performance. It has been found that excellent results are achieved, for example, at 150 0C. The associated pressure at this temperature (approximately 70 psi) is low enough to enable a simple autoclave system to be used for carrying out the leach.

[0032] In one preferred embodiment, an iron-bearing "seed" material is added to the leach slurry at the start of the saprolite pressure leach stage to accelerate the precipitation of dissolved ferric iron and the extraction of remaining nickel and cobalt from the solid phases.

[0033] Following the saprolite leaching step, the leach solution is preferably neutralized prior to nickel and cobalt recovery.

[0034] The laterite leaching method of the present invention can surprisingly achieve high levels of nickel extraction while avoiding the high cost of high-pressure autoclaves, and avoiding the production of environmentally unfriendly, high-acid consuming jarosite compounds. While the process does require an autoclave, the operating conditions are relatively benign compared to the high pressure acid leaching processes (nearly ten times or more the operating pressure required for the latter autoclave processes). Consequently, the process of this invention permits much simpler equipment design, and operations and maintenance are also easier than in high pressure acid leach processes. In addition, the saprolite leach and iron precipitation reactions occur much faster at moderately elevated temperature and a much shorter leach retention time, compared to the previously described atmospheric leach processes, is required. Thus, the autoclave required to carry out the process of the current invention is much smaller than the atmospheric leach reactors required for the processes of U.S. Patent Nos. 6,261 ,527 B1 and 6,680,035 B2 and WO 03/093517 A1. In addition, the process of the current invention achieves higher nickel extraction than can be achieved with the aforementioned atmospheric leach processes.

[0035] It has been found that the present invention can achieve at least about 90 %, and as high as 95 % or more, nickel extraction and as much as 95 % or more cobalt extraction, with about 5 to 10 % iron extraction.

Brief description of the drawings [0036] FIG. 1 is a flow sheet showing in simplified form one embodiment of the process of the present invention.

[0037] FIG. 2 is a flow sheet showing another embodiment of the process of the present invention in which some of the leach residue is recycled in order to provide seed for iron precipitation.

Detailed description of the invention [0038] Referring to FIG. 1 , a slurry of limonite ore is mixed with a concentrated mineral acid, chosen from the group sulfuric, hydrochloric and nitric, or a mixture of any of these acids, in a suitable device such as a stirred tank reactor, or in the case of continuous processing, a series of stirred tank reactors. The limonite ore slurry is produced by conventional means, such as screening and pulping the ore in a drum scrubber, which will be apparent to those skilled in the art.

[0039] The quantity of acid added is approximately that required to stoichiometrically dissolve the soluble non-ferrous metals in the ore as well as the soluble iron, i.e. most of the nickel, cobalt, magnesium, aluminum, copper, zinc, iron present in goethite and other soluble iron hydroxyl-oxide minerals, and a small portion of the chromium, which is usually contained primarily in the relatively insoluble mineral chromite. A small excess of acid may be added to drive the dissolution reactions as far as possible to completion and to ensure maximum extraction of nickel and cobalt from the limonite ore. In some cases, some of the magnesium and aluminum may be insoluble and this should be taken into account to determine the precise acid addition.

[0040] The addition of concentrated acid to limonite ore slurry results in the generation of substantial heat, raising the temperature of the mixture as high as the boiling point of the leach solution. The atmospheric limonite leach step is preferably carried out close to the boiling point of the solution to maximize the rates of the leaching reactions and thereby minimize the residence time required to complete the reactions. Additional steam or other energy may be added to the leach reactors in order to maintain the leach temperature as close to the boiling point as possible. It is preferable that the limonite ore slurry density be as high as possible consistent with good mixing in order to minimize the need for additional energy and to minimize the volume of pregnant liquor that needs to be treated subsequently for nickel and cobalt recovery.

[0041] The limonite ore slurry is leached for sufficient time to complete the reactions. This is typically 1 to 4 hours if the limonite leach is conducted at approx. 95 to 105 0C.

[0042] In a preferred embodiment of the process, a reductant is added to the limonite leach slurry in order to enhance the dissolution of nickel, and particularly cobalt, from the ore. Most of the cobalt and a smaller proportion of the nickel present in limonite ore is contained typically in a variety of oxidized manganese minerals, referred to collectively as "manganese wad." Manganese is typically in the tetravalent or trivalent state in these minerals and is refractory to acid leaching unless a reductant is added to cause the manganese to dissolve in the divalent form. The dissolution of manganese is necessary to allow the contained cobalt and nickel to dissolve as well, as will be apparent to those skilled in the art. Suitable reductants include sulfur dioxide (SO2), either as a gas or aqueous solution of SO2 and ferrous iron, as a soluble salt, e.g. ferrous sulfate, although many other reducing species will also react with oxidized manganese compounds.

[0043] The resulting limonite leach slurry is injected into an autoclave along with the saprolite ore. The saprolite ore typically will be prepared by crushing, grinding and screening or cycloning of the run-of-mine saprolite ore to achieve a particle size that allows the saprolite ore particles to be suspended in the autoclave reactor during the leaching process. The resulting slurry is heated to the desired reaction temperature, for example in the range of 120 to 160 0C, by any appropriate means, for example by direct injection of intermediate pressure steam into the autoclave, or by direct or indirect steam heating of the leach slurry prior to injection into the autoclave. The autoclave retention time is sufficient to allow most of the iron in solution after the limonite atmospheric leach to hydrolyze and precipitate, and the acid regenerated by iron hydrolysis to react with the saprolite ore, thus extracting most of the contained nickel and cobalt, as well as magnesium and other impurity metals.

[0044] At 150 0C, typically only 1 to 2 hours is required to complete the reactions. This compares to 10 to 11 hours typically in the atmospheric pressure leach step described in WO 03/093517 A1. Thus, although the process of the present invention requires an autoclave, the much reduced retention time means that this reactor will be much smaller than the atmospheric leach reactors required in the process of WO 03/093517 A1. In addition, .since the working pressure of an autoclave operating at up to 160 0C is equal to or less than 90 psi, this reactor is relatively simple compared to the autoclaves used in high pressure leaching processes at temperatures of from approx. 250 to 270 0C and pressures of 580 to 800 psi. Overall, the moderate- pressure autoclave systems required in the current process can be of comparable cost to the atmospheric leach reactors required in the process of WO 03/093517 A1 and much simpler to operate and maintain than the typical high pressure acid leach autoclaves.

[0045] It has been found that the nickel extraction obtained in the current process is up to 10 or more percentage points greater than the nickel extraction obtained in the process of WO 03/093517 A1 , using similar acid/ore and saprolite/limonite ratios. This is a very significant advantage of the current process as compared to the processes described in U.S. Patent Nos. 6,261 ,527 B1 and 6,680,035 B2 and WO 03/093517 Al

[0046] The use of a temperature above the boiling point for the saprolite leaching stage may also provide a higher nickel/iron ratio in solution, which is advantageous with respect to downstream processing of the leach solution. This is because in most cases virtually all iron must be removed from solution before effecting nickel and cobalt recovery. Usually, the residual iron in solution is removed by adding a base, for example calcium carbonate, to the leach slurry and precipitating iron oxyhydroxide compounds. Some nickel will co-precipitate with the iron resulting in losses of the pay metal. The neutralizing agent also represents an additional operating cost of the process.

[0047] A further advantage of the use of higher temperature is an improvement in the solid/liquid separation properties of the final leach slurry, with higher settling rates being achieved with a higher leaching temperature.

[0048] After leaching, the leach slurry is subjected to solid/liquid separation by filtration or thickening to produce a pregnant leach solution containing most of the nickel and cobalt contained in the ore and a solid residue containing most of the iron in the ore. Advantageously, thickening is carried out in a series of thickeners with counter-current flow of a wash water stream and the leach slurry in order to wash most of the entrained metal values out of the leach residue, a method called counter-current decantation (CCD). The metal values report preponderantly to the thickener overflow of the first thickener, which is the pregnant leach solution.

[0049] The pregnant leach solution proceeds to nickel and cobalt recovery by methods known to those skilled in the art, such as solvent extraction, ion exchange, sulfide precipitation using sulfiding agents, e.g. hydrogen sulfide, or hydroxide precipitation, using for example magnesia as the precipitating agent.

[0050] The nickel and cobalt can also be recovered from the leach slurry without prior solid/liquid separation, using the resin-in-pulp process. In this process, an ion exchange resin which extracts nickel and possibly cobalt is added directly to the leach slurry. After the extraction is complete, the resin is separated from the nickel-depleted leach slurry by screening. After washing the resin to remove solids, the nickel can be eluted from the resin with a fresh acid solution.

[0051] Prior to or during nickel and cobalt recovery by one of the aforementioned methods, the leach solution may be neutralized with a base, such as calcium carbonate, magnesium oxide, sodium carbonate or the like, to neutralize free acidity remaining from the leach process and precipitate small amounts of ferric iron, aluminum, and chromium, while minimizing co- precipitation of nickel and cobalt. This process may be carried out in a single or multiple steps separated by intermediate solid/liquid separations.

[0052] In one embodiment of the invention, the first stage of neutralization may be carried out prior to separating the leach residue from the leach solution. The combined leach and neutralization residue may then be separated from the partially neutralized leach solution by filtration or thickening, as described above. A second stage of neutralization may then still be desirable, depending on the method selected for nickel and cobalt recovery from the pregnant leach solution. After this second stage of neutralization, the resulting neutralization residue may be separated from the neutralized leach solution by filtration or thickening. This second-stage neutralization residue is ideally returned to the first stage neutralization to re- dissolve any co-precipitated nickel and cobalt.

[0053] In an alternative preferred embodiment of the invention, an iron- bearing "seed" material is added to the leach slurry at the start of the saprolite pressure leach stage, as shown in FIG. 2. The purpose is to accelerate the precipitation of dissolved ferric iron and the extraction of remaining nickel and cobalt from the solid phases. The surfaces of the seed particles provide low- activation energy sites for hydrolysis and precipitation of iron, for example as ferric hydroxide, goethite, or hematite. This seed material is ideally a portion of the final leach residue itself, which contains precipitated iron compounds.

[0054] The following examples illustrate the method of the present invention. The ore used in these examples came from a Central American laterite deposit. The limonite and saprolite fractions of the ore had the compositions given in Table 1. The saprolite ore was crushed and ground to approx. -100 mesh before use in the tests.

TABLE 1

[0055] EXAMPLE 1: An acid solution was prepared by adding 287.6 g of 96 % sulfuric acid and 48.0 g of 37 % HCI (both mineral acids being reagent grade) to 702 mL of water. The acid solution was transferred to a 2- liter cylindrical reaction kettle equipped with a mechanical stirrer, 4 plastic baffles, and a tight-fitting lid equipped with a water condenser open to the atmosphere. The reaction kettle was heated by an external, electrical heating mantle. 381.7 g (wet) of the limonite ore described in Table 1 were added to the acid solution while heating and stirring the mixture. The temperature was controlled at 94 to 105 0C and the limonite ore was leached for 5 hours. Liquid samples were taken from the leach slurry at 1 , 2 and 5 hours. [0056] After 5 hours of leaching, the leach slurry was transferred to a 2- liter titanium autoclave, along with 344.8 g (wet) of the saprolite ore described in Table 1. The saprolite ore had first been wet ground to approximately -100 mesh and then filtered to form a cake containing 27.5 % moisture. 128.6 g of technical grade hematite also was added to the autoclave to "seed" the precipitation of iron. The autoclave was equipped with a mechanical stirrer, thermocouple, cooling coil and external heating mantle. The leach slurry was heated to 150 0C and held at that temperature for 2 hours. A sample was removed from the autoclave after 1 hour (6 hours total leach time, including the limonite atmospheric leach).

[0057] The slurry was then rapidly cooled to approx. 50 0C using the water cooling coil and discharged from the autoclave. The entire slurry was filtered. The filtercake was repulped twice in fresh water to wash out the entrained metal values. The cake was then dried and weighed. The dry solids, filtrate and the combined wash water were assayed separately. Based on the weights, volumes and assays of the final residue and solutions, the extractions of the various metals were calculated. The assays of the liquid samples taken during the atmospheric limonite leach step and moderate pressure saprolite leach step are shown in Table 2 and the final solution and residue assays, as well as the calculated metal extractions, are given in Table 3.

[0058] The limonite/saprolite weight ratio (on a dry solids basis) was 1.0 and the overall acid/ore ratio was 600 kg equivalent H2SO4 per tonne of dry solids.

TABLE 2

TABLE 3

* Extractions based on residue and final solutions assays, weights and volumes.

[0059] The assays in Table 2 indicate that the nickel and iron in limonite ore were dissolved substantially during the atmospheric leach stage and that 1 hour was sufficient to carry the leaching reactions almost to completion, although the further reduction in free acid at 2 and 5 hours indicates that some further reaction occurred. The 6-hour solution assays (Table 2, 1 hour of pressure leaching) and the final solution assays (Table 3, 2 hours of pressure leaching) indicate that the iron in solution was rapidly hydrolyzed and precipitated at the moderate pressure and temperature conditions in the autoclave, while dissolving most of the nickel contained in the saprolite ore. The final solution nickel and iron assays were significantly higher than the 1-hour solution assays due to substantial evaporation that occurred during vacuum filtration of the final leach liquor.

[0060] The results indicate that very high nickel extraction and low iron extraction are features of the process of this invention. The low sulfur content of the final residue indicates that most of the iron could not have been precipitated as jarosite, which has a theoretical Fe/S ratio of approx. 2.6 by weight.

[0061] EXAMPLE 2: Another test was carried out similarly to that described in Example 1 with the following exceptions. The limonite ore used had the following composition: 1.55 % Ni, 48.4 % Fe, 0.47 % Mg, and 37.2 % moisture. 398.1 g (wet) of this ore was used in the test, along with 573.7 g water, 288.2 g of 96 % H2SO4, 46.9 g of 37 % HCI, and 285.7 g of MgSO4*7H2θ. No hematite seed material was used in the test. The MgSO4*7H2θ was dissolved in the water prior to adding the acid to the solution. The purpose of adding soluble magnesium to the solution was to simulate the recycle of a magnesium-rich solution which would remain after nickel and cobalt recovery from the pregnant leach solution. The atmospheric leach was carried out with a temperature of 96 to 101 0C. No sample was taken from the autoclave during the pressure leach stage.

[0062] The assays of the liquid samples taken during the atmospheric limonite leach are shown in Table 4 and the final solution and residue assays, as well as the calculated metal extractions, are given in Table 5.

[0063] The limonite/saprolite weight ratio (on a dry solids basis) was 1.0 and the overall acid/ore ratio was 600 kg equivalent H2SCM per tonne of dry solids.

TABLE 4

TABLE 5

* Extractions based on residue and final solutions assays, weights and volumes.

[0064] The results of this test are fairly similar to the results given in Example 1 and illustrate that seeding is not required to achieve effective iron precipitation. Also, the presence of dissolved magnesium in solution initially does not appear to impact negatively on the extraction of nickel or precipitation of iron. [0065] EXAMPLE 3: For comparison, another test was done to

simulate the conditions of the process described in WO 03/093517 A1. The

atmospheric limonite leaching stage was carried out similarly to the limonite

leach stage described in Example 1. 398.1 g (wet) of the same limonite ore

used in Example 2 was added to a solution comprised of 719.9 g water, 288.2

g 96 % H2SO4 and 46.9 g of 37 % HCI. Leaching was carried out for 4 hours

at 101-104 0C. After 4 hours, 310.6 g of ground saprolite containing 20.0 %

H2O and 128.6 g hematite seed were added to the leach slurry and leaching

was continued at 98-102 0C for 10 additional hours. Liquid samples were

taken periodically to follow the course of leaching. The final leach slurry was

filtered, the filtercake repulped twice with fresh water and the filtrate, wash

solution and final washed residue were assayed as in the previous examples.

[0066] The conditions of this test were thus essentially identical to

those of Example 1 , except that instead of pressure leaching at 150 0C for 2

hours, atmospheric leaching at approx. 100 0C for 10 hours was carried out

after the addition of saprolite ore.

[0067] The limonite/saprolite weight ratio (on a dry solids basis) was

1.0 and the overall acid/ore ratio was 600 kg equivalent H2SO4 per tonne of

dry solids.

[0068] The results of this test are given in Tables 6 and 7.

TABLE 6

TABLE 7

* Extractions based on residue and final solutions assays, weights and volumes.

[0069] The results given in Tables 6 and 7 are similar to those described in the examples of WO 03/093517 A1 , as expected. However, the nickel extraction obtained with the atmospheric leach process of WO 03/093517 A1 was 8 to 9 percentage points lower than that obtained with the combined atmospheric and pressure leaching process of the current invention. This clearly demonstrates the advantage of the current process compared to that of WO 03/093517 A1.

[0070] EXAMPLE 4: Another test was done to compare the results obtained from using a combination of atmospheric and moderate pressure leaching, as prescribed in the current invention, with the results from using moderate pressure leaching alone. In this test, 381.7 g (wet) of the limonite described in Table 1 , 312.5 g of the ground saprolite at 20.0 % moisture, 734.4 mL of water, 288.3 g of 96 % H2SO4 and 46.8 g of 37 % HCI were charged to a 2-liter titanium autoclave, heated to 150 0C, and leached for 2 hours. After rapid cooling, the leach slurry was filtered and the leach residue was repulp washed as in the previous examples. The filtrate, wash solution and solid residue were assayed as in the other examples. Metals extractions were calculated based on the solution volumes, residue weight and assays.

[0071] The limonite/saprolite weight ratio (on a dry solids basis) was 1.0 and the overall acid/ore ratio was 600 kg equivalent H2SO4 per tonne of dry solids.

[0072] The results are shown in Table 8. TABLE 8

* Extractions based on residue and final solutions assays, weights and volumes.

[0073] The very low nickel extraction achieved in this test illustrates

that direct pressure leaching of mixed limonite/saprolite ore at moderate

temperature and pressure is unlikely to result in a viable process for nickel

extraction. Whereas, by combining atmospheric leaching and moderate

pressure leaching, as in the process of the current invention, leach extractions

approaching those of the high pressure leach processes are achieved without

the complexity of high pressure associated with the latter processes.

[0074] EXAMPLE 5: The limonite and saprolite fractions of the ore

used in this example had the compositions given in Table 9:

TABLE 9

[0075] In this case, 238 g (dry basis) of the limonite ore was slurried at

35 % solids in water and placed in a reaction kettle similar to that used in

Example 1. 338.5 g of 96 % sulfuric acid was slowly added to the reactor over

about 10 minutes. The heat of dissolution of the acid raised the temperature

to 99 0C within 5 minutes of adding all of the acid. The limonite was leached at

a temperature of 94 -102 0C for 4 hours. Sulfur dioxide gas was bubbled into

the leach slurry during the limonite leaching period to control the leach slurry

oxidation reduction potential at about 620 mV (vs. saturated Ag/AgCI

reference electrode). A sample was taken at the end of this initial leaching

period. [0076] The saprolite was wet ground to -100 mesh and filtered to

produce a filtercake. 262 g (dry basis) of this filtercake was added to the

limonite leach slurry, which was then transferred to an autoclave. The

autoclave was sealed and heated to 150 0C, at which temperature leaching

was allowed to continue for an additional hour before rapidly cooling the

autoclave. The slurry was sampled at the completion of the additional one-

hour leach period. The saprolite to limonite ratio was 1.1 and the sulfuric acid

addition was 650 kg H2SO4 per tonne of ore in this test.

[0077] The assays of the leach liquor and the solid residue at the

completion of each stage of leaching are given in Table 10:

TABLE 10

[0078] Also shown in Table 10 are the calculated extractions of Ni, Co,

Fe and Mg. Ni, Co, and Mg extractions were calculated using a "silicon-tie"

method in which the weight of solid residue was calculated using ore and

residue silicon assays on the assumption that none of the silicon leached.

These weights and ore and residue assays were then used to calculate

extractions.

[0079] This test differs from Example 1 mainly in that only 4 hours of

limonite leaching were employed, sulfur dioxide gas was added during the limonite leach, no seed was added during the saprolite leaching phase, and only one hour of autoclave leaching was used. The results still indicate a very high nickel extraction with minimal iron extraction and very good leaching of cobalt. The latter was due to the efficacy of sulfur dioxide as a reductant for the manganese wad material present in the ore.

[0080] It will of course be appreciated by those skilled in the art that many variations of the process would be possible within the broad scope of the present invention. Those skilled in the art will appreciate that the invention upon which the description is based may be utilized in other embodiments that carry out the purposes and fulfill the objects of the present invention. The above disclosure is intended to be illustrative while the scope of the invention is defined by the following claims.