More specifically, the present invention resides in a process for cleaning with minimal associated reagent degradation of the organic reagent used in nickel and cobalt separation processes.

The separation and recovery of nickel and cobalt, both present in ammoniacal media, by the use of solvent extraction procedures is well documented in the literature.

The use of high strength ammoniacal ammonium carbonate solutions to strip nickel from the loaded organic reagent was disclosed in Australian Patent 605867 and a commercial plant now producing 28,000 te of nickel and 1800 te cobalt per annum is currently operating in Australia.

For the process to operate efficiently on a continuous basis, it is necessary that the cobalt present in the ammoniacal media exists as a cobaltic (Co lil) ammine as distinct from a cobaltous (Co II) ammine.

Cobaltic ammines are stable compounds and do not react with, and are therefore not extracted by organic reagents. Cobaltous ammines and nickel 11 ammines are both labile compounds and are rapidly extracted by reagents such as the ß- diketones, acetophenone oximes, benzophenone oximes and salicycaldoximes which are commonly used in solvent extraction processes.

When cobaltous cobalt is loaded onto the organic reagent, the complex can be oxidised to the cobaltic state. This is known to occur rapidly in the presence of an oxygen containing gas. The cobaltic organic complex cannot be stripped with high strength ammoniacal ammonium carbonate liquor used to recover nickel from the organic reagent. Consequently, as the organic reagent is recycled continuously around the circuit, the cobalt content of the organic reagent increases while the nickel transfer capacity of the reagent decreases and unless the contaminated reagent is bled from the process and cleaned or replaced by fresh reagent, the process eventually fails due to reagent poisoning.

In order to separate nickel from cobalt, both being present in ammoniacal media, it is therefore necessary to control the concentration of cobaltous cobalt in the ammoniacal liquor at about the 1-2 ppm level. This may be achieved as disclosed in AU Patent 605867 by oxidation with air and hydrogen peroxide. Even under the best

controlled conditions, it does not necessarily follow that the cobaltous cobalt value will not increase. In an industrial operating process there are many impurities present in the liquors and some of these, such as reduced sulphur compounds, will react with cobaltic ammines to form more cobaltous cobalt. This conversion will be enhanced by both temperature and storage time of the ammoniacal liquor prior to contacting the solvent extraction reagent.

Despite the best operating procedures the organic solvent extraction reagent will gradually become contaminated by cobalt and it will be necessary to treat a bleed stream to reverse this poisoning by a suitable treatment procedure.

US Patent 4,083, 915 lists 30 experiments with metallic and non metallic entities in combination with ammonium carbonate or acidic media to effect reduction of the cobalt poisoned reagent to facilitate the removal of cobalt from the organic complex.

The approach used was to reduce the cobalt III organic complex to a cobalt 11 state followed by stripping with either acidic or ammoniacal media to remove the cobalt, thus freeing up organic reagent for further nickel recovery. The temperature range was from room temperature to 60° C with contact time of 10 minutes under inert gas atmosphere.

The use of metals such as iron or zinc, to effect reduction of Co III to Co II, in the presence of dilute sulphuric acid (20% VN), while reasonably efficient in reversing cobalt poisoning of the organic reagent, can introduce deleterious side effects into the solvent extraction process.

If the pH is not controlled, or if solid liquid separation of the reactants and organic reagent is not extremely efficient, iron in solid or solution form will be transferred to the solvent extraction circuit. In an ammoniacal system, this results in the formation of ferric hydroxide solids which interfere with the phase separation process in the solvent extraction settlers. Consequently, throughput of the system is reduced requiring additional remedial action.

When zinc powder is used as a reductant metal in the presence of sulphuric acid, it is solubilized during the reaction. The spent acid stream containing both zinc and the stripped cobalt is recycled to recover the valuable cobalt and this results in increased zinc concentrations which adversely impact on the cobalt purification and recovery process.

With respect to the use of any reductant metals in small quantities, it is more difficult to meter consistently accurate quantities of powder than it is to pump exact liquid volumes.

US Patent 3,981, 968 recognised the cobalt poisoning effect when extracting nickel from ammonia solutions also containing cobalt and reversed this effect by contacting a bleed stream of nickel depleted organic reagent (following nickel stripping with sulphuric acid at 60°-100° F) with strong sulphuric acid (15% W/W) at temperatures in the range of 210°-250° F to remove 90% of the cobalt from the organic reagent.

This patent discloses a number of oxime type reagents that are thought to be amenable to the above treatment but the reagent 2-hydroxy-5-tertiary nonyl acetophenone oxime (LIX84 type of Cognis) amongst others is not covered by the disclosure.

No information has been provided in either of the above US patents on the stability of the organic reagent when subjected to the cobalt removal procedures disclosed, particularly US Patent 3,981, 968 where strong sulphuric acid (15% W/W) at the boiling point of water is used to effect cobalt removal.

This is an important consideration in any operating process as destruction of the organic reagent combined with cobalt III during removal of the latter or destruction of the uncomplexed organic reagent also subjected to the same conditions can seriously impact on the economics of metal separation and recovery processes.

Degradation of the reagent occurs when the oxime (Structure 1a below) is destroyed by conversion to a ketone (1b) in the first instance and then this is slowly converted to a salicylic acid derivative (1c) in comparatively small quantities. In the structure of these compounds, it should be noted that the ketone (lb) can be reoximated efficiently and economically to the original oxime (la) as disclosed in AU Patent 612528.

The salicylic acid derivative degradation product (Ic) is a more polar compound and may contribute to poor phase separation properties of the reagent mixture. It may form an ammonium salt in the operating circuit and be partially removed from the system as a water soluble entity and represents an irreversible reagent loss.

Structure 1a Structure 1b Structure 1c Reagent losses through degradation are not the only concern as in most organic reactions, other moities may be formed from the parent reactant and the influence of these on the operation of the solvent extraction system may be significant.

One such concern is the formation of organic by-products that exhibit undesirable interfacial surface tension or viscosity effects or result in the accumulation of"crud"at the organic/aqueous interface and adversely affect the phase separation characteristics of the system.

Another issue associated with the operation of an organic cleaning plant at elevated temperatures (200°F + as in US 3, 981, 968), particularly those higher than the flash point of the solvent, generally 70°C-75°C, is that of safety. Although the organic reagent cleaning plant treats only a bleed stream of probably up to 1% of total organic flow rate, it is still an integral part of a much larger process where organic reagent diluted with volatile solvent is circulated at rates of 6000-7500 I/min.

Clearly the operation of any system where flammable material is a concern is safest at a lower rather than a higher temperature irrespective of degradation and reaction rate considerations.

The above discussion of documents, acts, materials, devices, articles and the like is included in this specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all of these matters formed part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed in Australia before the priority date of each claim of this application.

It is an object of the present invention to overcome or at least alleviate one or more of the difficulties associated with the prior art.

The present invention aims to provide a process for treating a cobalt poisoned organic reagent by stripping cobalt from the reagent. The invention preferably aims to

provide a process for treating a cobalt poisoned acetophenone oxime reagent used in a nickel and cobalt separation process, in the absence of metal reductants.

It is a particularly desired aim of the invention to reverse cobalt poisoning of the organic reagent with minimum degradation of the reagent thus maintaining the overall metal transfer capacity of the reagent inventory.

The present invention relates to a process for treating an organic reagent typically used in a nickel and cobalt separation process. The process of the invention is relevant for the treatment of an organic reagent used in separating nickel and cobalt from aqueous ammoniacal solutions. In such processes, some cobaltous (II) cobalt is loaded onto the organic reagent together with the nickel, which cobalt is generally converted to the cobaltic state in the presence of oxygen containing compounds and/or other oxidants. This cobaltic cobalt cannot be stripped from the organic reagent and in time the organic reagent looses effectiveness for the transfer of nickel and becomes poisoned. The present invention aims to provide a process for treating a cobalt poisoned organic reagent while minimizing degradation of the reagent.

Accordingly, the present invention resides in a process for treating an organic reagent containing cobalt in a nickel and cobalt separation process, said process including the step of contacting the organic reagent with dilute sulphuric acid at higher than ambient temperature for a period of at least 1 hour to remove cobalt from the organic reagent while minimizing associated reagent degradation.

In a particular embodiment, the invention resides in a process for treating an organic reagent containing cobalt in a nickel and cobalt separation process, said process including the step of contacting the organic reagent with dilute sulphuric acid at a temperature range of from 50° C to 80° C for a period of at least one hour to remove cobalt from the organic reagent while minimizing associated reagent degradation.

The process of the present invention is particularly applicable when the organic reagent used is a (3-diketone, including alkyl, aryl and halide substituted p-diketones, an acetophenone oxime, benzophenone oxime or a salicycaldoxime. In particular, the invention is most applicable when the organic reagent is an acetophenone oxime, particularly 2-hydroxy-5-tertiary nonyl acetophenone oxime.

The concentration of the sulphuric acid may be any suitable concentration in which to extract the cobalt from the organic reagent with minimal degradation of the reagent.

Preferably, the concentration of the sulphuric acid is from 50 gl-'to 150 gl-', more preferably from 70 to 125 gl-1, with about 75 go-'being most preferred.

The term"minimizing associated reagent degradation"relates to minimizing the formation of"crud"type by-products which are observable in a reaction between the organic reagent and sulphuric acid. In a reaction between sulphuric acid and an acetophenone oxime, degradation can also be measured by the level of formation of the corresponding ketone and salicylic acid. Minimizing the formation of degradation products is intended to mean keeping the formation of these products to a level that will not impair phase separation and operating efficiency in the nickel and cobalt separation process.

It has also been found that in general, as the temperature increases the effectiveness of the sulphuric acid in removing the cobalt also increases. However there is potentially an increase in the degradation rate of the organic reagent as the temperature increases. It is for this reason that it has been found that the process is carried out at preferably a temperature range of from 50° to 80° C and most preferably from 60°C to 70°C when an acetophenone oxime reagent is used, in order to minimize both reversible and irreversible degradation of the organic reagent.

For example, 3 month stability trials gave the following results for organic reagent in contact with aqueous sulphuric acid (75 g/I) at 60° C and 90° C.. SAMPLE CHANGE IN KETONE (g/l) CHANGE IN SUBSTITUTED SALICYLIC ACID (ppm) 60° C 62 40 90° C 155 485 Clearly reagent treated at the higher temperature suffered far more reversible and irreversible degradation.

In order to reverse the formation of Ketones, it is advantageous that the cleaned organic reagent produced by this process be reoximated before returning to the circuit.

The reoximation process may include the steps of reacting the organic reagent, which may have been partially converted to the corresponding ketone, dissolved in an organic

solvent, preferably kerosene, with a hydroxylamine acid salt and aqueous ammonia to form the corresponding oxime. A typical reoximation process is described in Australian Patent 612528. If salicylic acid is formed, the degradation is not generally reversible.

The present invention will now be described with reference to the accompanying drawings and Examples. The drawings and Examples should be considered to be illustrative of the invention and should not be construed as limiting upon the scope of the invention.

Figure 1 is a process flow diagram showing purification of the organic reagent in accordance with the present invention.

Figure 2 is a graph showing the effect of H2 S04 acid strength on organic cobalt stripping rate.

Figure 3 is a graph showing the effect of temperature on cobalt stripping rate at 75 g/l H2S04 at O : A of 2 : 1.

Figure 1 illustrates a continuous organic reagent cleaning plant with first reactor (1) having a sulphuric acid feed line (3) and an organic reagent feed line (5) leading to the reactor. Inert gas, (preferably carbon dioxide) is added through line (7).

The mixture is mixed by impeller (9) and raised to the appropriate temperature through heating coil (11).

The mixture is then transferred to organic/aqueous phase separator (15) via organic/aqueous reaction mixture line (13). The cleaned organic reagent is removed through line (17). The spent acid is discharged through line (19).

Experiments were conducted with respect to the reversal of cobalt poisoning using sulphuric acid, to systematically examine the effects of temperature, reaction time, acid strength and organic to aqueous phase ratios and reported in the Examples.

Example 1 The effect of acid strength on cobalt striping rate was examined over the concentration range of 75 to 125 gl-'H2 S04.

A sample of the organic reagent recovered from the operating solvent extraction circuit and contaminated with approximately 1.6 gel-'cobalt together with other elements Ni, Cu, Fe and Zn was pretreated sequentially with ammoniacal ammonium carbonate strip liquor, deionized water and dilute acid at ambient temperature, discarding the aqueous phase after each treatment in the above sequence. Analysis of the pretreated organic reagent gave the following element concentrations: (ppm) Ni Co Cu Fe Zn 11 1540 128 698 The pretreated organic reagent poisoned by 1.54 gel-'cobalt was preheated to 60° C in the presence of sulphuric acid of specified strength at an organic to aqueous (0 : A) ratio of 2: 1 and then agitated vigorously under a nitrogen atmosphere for a period of 7 hours. During this agitation period samples were withdrawn at selected intervals and after phase separation, the percentage cobalt stripped from the organic reagent was determined, the results are listed in Table 1. TEST ACID gl-% Co STRIPPED (hrs) 0 3. 0 7.0 1 75 0 40. 6 52.8 2 100 0 41. 8 53.4 3 125 0 42. 9 54.2 TABLE 1 The results of this Example are also shown graphically in Figure 2, illustrating a leveling out of the % cobalt stripped after 7 hours, although an increase in the percentage of cobalt stripped would be expected beyond 7 hours.

Example 2 The effect of temperature on cobalt stripping A second portion of the pretreated organic reagent described above in Example 1 was tested at two temperatures, 60° C and 70° C with acid strength 75 gl-'and O : A ratio of 2: 1. The results are given in Table 2. TEST T°C % Co STRIPPED (hrs) 0 3. 0 7.0 1 60 0 40.6 52.8 2 70 0 51.9 62.3

TABLE 2 Clearly a 10°C increase in reaction temperature has reduced the contact time to achieve in excess of 51 % cobalt removal by 4 hours. The results are show graphically in Figure 3.

Example 3 The cobalt removal rate was also examined at the higher temperatures of 80°C and 90°C with sulphuric acid strength of 100g/i'\ The results are given in Table 3. TEST TUC % Co STRIPPED (hrs) 0 3.0 7. 0 1 80 0 68 73 2 90 0 77 83

TABLE 3 Example 4 The effect of reaction time on cobalt stripping A third portion of the pretreated organic reagent described above for Example 1 was tested at an 0 : A of 2: 1 with 75 gl-'sulphuric acid at a reaction temperature of 70° C over a 7 hour period. In addition, a second procedure in which the sulphuric acid solution also contained approximately 14 gel-'cobalt was included for comparison. This was to determine the effect, if any, cobalt may have in an aqueous recycle continuous circuit. The results are listed in Table 4. TEST Co gl''% Co STRIPPED (hrs) 0 0. 25 0. 5 1 2 3 5 7 0 17. 5 25.3 37.0 46.8 51.9 59.1 62. 2 13. 8 0 15. 6 27.9 38.6 47.7 52.6 60.4 61. (

TABLE 4 Clearly the presence of approximately 14 go-'cobalt in the acidic strip liquor has not affected the efficiency of the strip process. Again, this is illustrated graphically in Figure 3, together with the results of Example 2.

Example 5 The results of the above Examples, taking into account the flash point (72°C) of the diluent used in the reagent mixture and other considerations such as degradation, suggested that a continuous trial operating at an organic to aqueous ratio of 2: 1 and acid strength 75 gl-'H2 S04 and temperature 70°C be conducted over a 3 day period.

The results produced are listed in Table 5. TIME POISONED ORGANIC CLEANED ORGANIC % COBALT REMOVED Ni g/i Co g/l Ni (ppm) Cog/) 0300 34 1.4 0900 1. 0 2. 8 42 1.4 50 1500 45 1.5 2100 0.9 2.9 38 1. 5 48 0300 33 1.3 0900 1.1 3.0 40 1. 3 57 1500 43 1.5 2100 1. 0 3.1 35 1. 2 61 0300 38 1.3 0900 1.1 3.2 26 0. 7 78 1500 42 0.7 2100 1.1 2.7 41 1.7 37 TABLE 5 AVERAGE COBALT REMOVAL 55% Given that the flow rate and reaction vessel capacity were controlled to allow a 3.5 hour retention time, the average value of 55% cobalt removal is regarded as very satisfactory and useful for future scale up calculations.

Reagent degradation studies To determine the effects of both temperature and acid strength on reagent stability, organic reagent recovered from the operating solvent extraction plant was subjected to 3 month stability tests at 60°C and 90°C, under an inert C02 atmosphere, with acid concentration of 75 gl-1 H2 S04 and at 70° C with H2 S04 at pH 1.5.

The chemical and physical properties of the organic reagent in each system were monitored monthly providing the following information.

The 70° C system produced the least degradation of the acetophenone oxime to the substituted ketone without"crud"fomation but with diminished phase separation

properties relative to the 60° C system. The 90° C system exhibited the greatest rate of reagent destruction generating both substituted ketone and substituted salicylic acid and also significant amounts of"crud".

The 60°C system while producing some"crud"maintained good physical properties of the reagent.

It was concluded that for a given acidity the undesirable effects on the reagent became more pronounced as the temperature was increased. It would be reasonable to conclude that the operating conditions disclosed in US Patent 3,981, 968 of 15% W/W H2 S04 (-160 gl-1) at 210°-250°F (>100° C) would result in significant irreversible degradation of the organic reagent together with impaired phase separation properties.

While this degradation would only affect a small fraction of the reagent, between 0: 1 and 10.0% depending on the frequency of stripping, the"crud"type by-products returned to the operating circuit would accumulate over time resulting in impaired phase separation and operating efficiency.

It was concluded that an optimum temperature range of 50° to 80° C, preferably 60°C-70°C was in order to minimise both reversible and irreversible degradation product formation and solvent vapour pressure.

The above description is intended to be illustrative of the preferred embodiment of the invention. It should be understood that any modification without departing from the spirit or ambit of the invention described herein should be considered to be incorporated.