AU2018363879A1

AU2018363879A1 - Production of high purity nickel sulfate

Info

Publication number: AU2018363879A1
Application number: AU2018363879A
Authority: AU
Inventors: Richard Clout; John Stewart
Original assignee: BHP Billiton Nickel West Pty Ltd
Current assignee: BHP Billiton Nickel West Pty Ltd
Priority date: 2017-11-10
Filing date: 2018-11-08
Publication date: 2020-05-21
Also published as: WO2019090389A1

Abstract

A process for producing high purity nickel sulfate, preferably having a purity of ≥99.8%, more preferably ≥99.98%, suitable for use in battery manufacture or nickel plating is described. The process involves selectively removing non-nickel metal impurities from a nickel sulfate solution, preferably a sub-saturated nickel sulfate solution, obtained for example from nickel powder, by ion exchange using a nickel pre-loaded ion exchange (IX) resin which adsorbs non-nickel metal impurities from the solution to form a substantially non-nickel metal impurities free nickel sulfate solution from which the high purity nickel sulfate can be recovered. The recovered nickel sulfate can be crystallised to remove the high purity product.

Description

Title of invention

Production of High Purity Nickel Sulfate

Technical Field

The invention relates to a process for the production of high purity nickel sulfate from a nickel powder leach in sulfuric acid.

Background of Invention

Nickel sulfate is used in electroplating and electroless plating as well as being a material for secondary batteries. There is a need for high purity nickel sulfate that is substantially free of impurities including copper, iron and cobalt for such uses.

Currently, purification of nickel sulfate is by organic solvent extraction techniques which typically involve an acid extractant and neutralising agent, for example, sodium hydroxide, to facilitate extraction of the impurities. However, the recovered nickel sulfate product is contaminated with sodium which is difficult to remove. Thus alternative methods for nickel sulfate purification are desirable.

A reference herein to a patent document or any other matter identified as prior art, is not to be taken as an admission that the document or other matter was known or that the information it contains was part of the common general knowledge as at the priority date of any of the claims.

Where any or all of the terms comprise, comprises, comprised or comprising are used in this specification (including the claims) they are to be interpreted as specifying the presence of the stated features, integers, steps or components, but not precluding the presence of one or more other features, integers, steps or components.

Summary of Invention

In one aspect, the invention provides a process for producing high purity nickel sulfate, preferably having a purity of >99,8%, more preferably >99.98%, suitable for use in a battery or nickel plating, comprising the step of:

selectively removing ποπ-nickel metal impurities from a nickel sulfate solution, preferably a sub-saturated acidic nickel sulfate solution for example obtained from nickel powder, by ion exchange using a nickel pre-loaded ion exchange (IX) resin which adsorbs non-nickel metal impurities from the solution to form a substantially non-nickei metal impurities free nickel sulfate solution from which the high purity nickel sulfate can be recovered.

WO 2019/090389

PCT/AU2018/051203

Desirably, the nickel sulfate solution comprises trace amounts of non-nickel metal impurities.

Suitably, the nickel sulfate solution is a sub-saturated nickel sulfate solution.

More suitably, the nickel sulfate solution is non-nickel metal impurity depleted subsaturated nickel sulfate solution.

More suitably still the nickel sulfate solution is acidic. Preferably, the nickel sulfate solution is an acidic non-nickel metal impurity depleted sub-saturated nickel sulfate solution which comprises trace amounts of non-nickel metal impurities.

Preferably, the step of selectively removing the trace amounts of non-nickel metal impurities involves buffering the nickel sulfate solution prior to ion exchange (IX) with one or more basic nickel compounds to a pH optimised for selectivity and stability of non-nickel metal impurity loading onto the nickel pre-loaded ion exchange resin.

Desirably, the process further comprises the step of recovering the high purity nickel sulfate from the substantially non-nickel metal impurities free nickel sulfate solution, preferably in the form of crystalline alpha nickel sulfate hexahydrate.

Preferably, the nickel sulfate solution is an acidic sub-saturated nickel sulfate solution, and the method comprises prior to the selective removal step, the additional steps of:

generating the acidic sub-saturated nickel sulfate solution which comprises the one or more non-nickel metal impurities; and removing bulk non-nickel metal impurities from the acidic sub-saturated nickel sulfate solution to form the non-nickel metal depleted nickel sulfate solution comprising a subsaturated concentration of nickel sulfate and trace amounts of one or more non-nickel metal impurities.

In a preferred embodiment, the process is preferably a batch process, comprising the steps of:

(i) generating an acidic sub-saturated nickel sulfate solution which comprises one or more non-nickel metal impurities;

(ii) removing bulk non-nickei metal impurities from the acidic sub-saturated nickel sulfate solution to form an acidic sub-saturated non-nickel metal impurity depleted nickel sulfate solution comprising a sub-saturated concentration of nickel sulfate and trace amounts of the one or more non-nickel metal impurities;

(ill) selectively removing the trace amounts of non-nickel metal impurities from the non-nickel metal impurity depleted nickel sulfate solution using a nickel pre-loaded ion exchange (IX) resin to remove the non-nickel metal impurity from the solution to form a substantially non-nickel metal impurities free sub-saturated nickel sulfate solution;

WO 2019/090389

PCT/AU2018/051203 (iv) recovering the high purity nickel sulfate from the substantially non-nickel metal impurities free sub-saturated nickel sulfate solution, preferably in the form of alpha nickel sulfate hexahydrate, particularly crystalline alpha nickel sulfate hexahydrate.

Suitably, selective removal of the trace amounts of non-nickel metal impurities involves buffering the acidic sub-saturated non-nickel metal depleted nickel sulfate solution prior to ion exchange (IX) with one or more basic nickel compounds to a pH optimised for selectivity and stability of non-nickel metal impurity loading onto a nickel pre-loaded ion exchange resin. Leaching

In another aspect, the invention provides a process, preferably a batch process, for leaching nickel sulfate from nickel powder comprising the steps of:

(i) leaching a stoichiometric excess of the nickel powder with sulfuric acid to form an acidic sub-saturated solution of dissolved nickel sulfate and one or more non-nickel metal impurities, together with unleached nickel powder;

(ii) separating the acidic sub-saturated nickel sulfate solution from the unleached nickel powder to provide a discharge solution which is a substantially solid-free acidic subsaturated nickel sulfate solution; and optionally repeating steps (I) and (ii) one or more times, wherein the one or more additional leaching steps (I) are carried out with sulfuric acid, preferably using the unleached nickel solid separated in step (ii).

Preferably, the acidic sub-saturated solution of dissolved nickel sulfate and one or more non-nickel metal impurities, together with unleached nickel powder, is a pregnant leach solution. It should be understood that the pregnant leach solution is a solution is a solution of metal laden water generated from stockpile leaching and heap leaching for example. A pregnant leach solution is an acidic solution and may comprise one or more organic and/or inorganic acids.

Preferably, the process is carried out as a batch process.

The nickel powder thought to be leached by the sulfuric acid in accordance with the following equation:

^,v T ^7X^4 (aq) ^2(g) +

Other non-nickel metal impurities are co-leached by the acid, together with the nickel sulfate.Typically, the one or more non-nickel metal impurities include one or more of Ca, Al, Na, P, Si, K, Mg, Mn, Se, Cr, Co, Fe, Cu, Zn, As, Ru, Pb, Hr, Pd, Ag, Cd, Sb, Ir, Pt, Au and Bi. The most significant impurities tend to be varying amounts of Co, Cu, Cr, K, Ca, Na, Zn and/or Fe. However, the impurities in most appreciable quantities are usually Co, Cu and/or

WO 2019/090389

PCT/AU2018/051203

Fe. Where chemically possible, various metal oxidation states species can be present in the leach solution, for example, iron can be present in the ferric or ferrous form.

In order to reduce external metal ion contamination, preferably, the sulfuric acid used in the process for leaching is prepared using demineralised water. This is thought to reduce the contaminant burden to impurities arising from sulfuric acid leaching of nickel powder. This means that metal contaminants that are found in mains water are avoided, in one embodiment, the nickel powder for leaching may be provided directly from a nickel powder leach plant. Thus, it is desirable that the nickel sulfate process is carried out in proximity to a nickel powder leach plant. In such a case, the raw material nickel powder can conveniently been provided to the nickel sulfate processing plant directly from the discharge of wet metals driers in the nickel powder leach plant. Alternatively, the nickel powder raw material can be provided by transport from an alternative source; however, this would be expected to add to the overall processing cost.

Preferably, the nickel powder used in the process described has an average particle size of from about 1 micron to about 1200 microns, more preferably from 10 microns to about 1000 microns, more preferably still from about 100 microns to about 900 microns. In some embodiment, it is believed that coarser particles leach more slowly such that there is a preference for less coarse particles where faster leaching is desirable.

The nickel powder raw material may have a purity of from about 98% to about 100%. However, the nickel powder preferably has a purity of >99.8% nickel. Such nickel powder may be obtainable by any industrial process capable of generating nickel powder having such purity. One example of a suitable process is nickel powder production from a process involving hydrogen pressure reduction.

Preferably, the leaching process is carried out at a temperature of from about 50 °C to about 100 °C, most preferably from about 70 °C to about 95 °C, most preferably still at about 80 °C. Suitably, the process temperature is achieved and/or maintained by steam heating, for example, by a steam heating coil provided in the leach tank. Steam may be conveniently provided from a close by refinery.

A sub-saturated nickel sulfate solution comprises dissolved nickel in a concentration which is about 70%to about 99% of the theoretical nickel sulfate saturation concentration under the particular leaching conditions used. Preferred sub-saturated solutions are of from about 90% to about 98% nickel sulfate, more preferred sub-saturated solutions are of from about 92% to about 97%, with solutions of about 95% of the theoretical nickel sulfate saturation concentration being most preferred. Suitably, the pregnant leach solution is a subsaturated nickel sulfate solution that comprises dissolved nickel in a concentration that is about 70% to about 99%, more preferably from about 92% to about 97%, most preferably about 95% of the theoretical nickel sulfate saturation concentration. Suitably, the sub

WO 2019/090389

PCT/AU2018/051203 saturated nickel sulfate solution is acidic. The degree of nickel sulfate saturation in the leach solution is dependent on factors including the pH of the acid used, the temperature of the leach solution and/or the rate of evaporation from the leaching solution. It should be understood that the leaching sulfuric acid concentration and/or the process temperature may be controlled to in order to produce a discharge solution which is sub-saturated at about 95% of the saturation limit of nickel sulfate at under the processing conditions used. For example, at a leaching process temperature of about 80°C, a nickel sulfate saturation of about 95% is readily achievable using an optimised acid concentration and processing conditions as described herein. Advantageously, operating close to saturation, particularly at higher temperatures, for example, 80°C, minimises evaporation duty in downstream process crystallisation stages, and eliminates the need for a pre-evaporator reducing capital and processing costs.

In preferred embodiments, the initial concentration of sulfuric acid used in leaching step (I) is from about 150 g/L and about 350 g/L, more preferably from about 200 g/L and about 300g/L, more preferably still from about 250 g/L to about 290g/L, most preferably about 280 g/L (which corresponds to about 150 ml/L of sulfuric acid/water).

Preferably nickel powder is preferably provided/maintained at a loading of between about 850 g/L and 1200 g/L, more preferably, from about 950 g/L to about 1100 g/L, most preferably about 1000 g/L (mass of nickel per litre of sulfuric acid).

Suitably, the nickel in the sulfuric acid has a pulp density, that is amount of solids in a pulp of nickel in acid, in the range of about 500 to about 1500 g/l, more preferably from about 600 to about 200 g/L, more preferably still from about 750 g/L to about 850 g/L, with a particularly preferred pulp density being about 800 g/L, for example at a pH of between about 0 and about 3.5.

In preferred embodiments, the process occurs at about atmospheric pressure or at a positive gauge pressure. In particular, at least leaching step (i) may be carried out at about atmospheric pressure or more preferably af a positive gauge pressure.

In a preferred embodiment, during step (i) the nickel solid and sulfuric acid are agitated.

In one embodiment, the leaching and/or separation, that is, steps (i) and/or (li), may be carried out under aeration, for example, under a compressed air atmosphere, for example, having a flow rate of about 0.1 to about 9.5 L/min, more preferably 1 - 5 L/min, most preferably about 1 L/min. Lower flow rates, for example, < 3.5 L/min, are preferred as higher flow rates tend to result in increasing amounts of evaporation of the leaching acid which results in premature saturation of nickel sulfate in the discharge solution.

In some embodiments, the process is carried out anaerobically, for example using N₂sparging.

WO 2019/090389

PCT/AU2018/051203

As hydrogen is evolved as a side product of the leaching process, the process desirably further comprises the step of removing the evolved hydrogen gas, for example, by flushing the local environment around the leaching location with steam, for example, a snuffing steam, and/or by operating the leaching process at a positive gauge pressure to prevent air ingress. It will be understood that the flush is maintained in the vicinity of at least the leach tank vapour space. As hydrogen gas evolution may be ongoing until the leaching reactions reach equilibrium it. may be beneficial to provide the steam flush and/or positive gauge pressure to further steps of the process, particularly, toe filtering and bulk impurities removal steps also described herein. A preferred steam flush is a steam/air mixture comprising from about 50% to about 90% steam. A steam flush providing a minimum of 70 vol% steam atmosphere is particularly desirable.

In embodiments including the step of removing the evolved hydrogen gas, it is desirable, that the process further comprises the step of scrubbing the hydrogen steam flush offgas generated to remove acid mist and particulates prior to atmospheric discharge.

Further desirably, the process includes the step of carrying out at least the leaching step under a nitrogen blanket to prevent a potentially explosive hydrogen/air mixture forming.

In some embodiments, where hydrogen evolution is ongoing, the separating step (ii) and/or additional bulk impurity removal steps also described herein can further be carried out using a steam flush and/or under a nitrogen blanket to minimise the risk of formation of an explosive environment.

Desirably, when leaching step (i) has been completed, the unleached solid nickel is separated from the sub-saturated pregnant leach solution, for example, by filtering or by decanting. Decanting is a particularly preferred separation method as conveniently leaves the unreached nickel in place for easy commencement of a subsequent leach batch where fresh acid is simply to be added in the next batch. Thus, it is preferable that discharge solution decanting takes place after a suitable period for settling has elapsed. Completion of the leaching step can be identified by observation that the pregnant leach solution has a predetermined pH, or that, the ieaching step has been allowed to process fora predetermined period of time. The leaching process can be terminated at. any point by addition of a suitable amount of a neutralisation agent which is suitable to neutralise the excess free acid to provide a solution of a desired pH. Thus, in some embodiment, step (i) is allowed to proceed until the pregnant leach solution has a pH of from about 0 to about 4, more preferably a pH of about 1 to about 3.5, most preferably a pH of about 3. In other embodiments, the leaching step (i) may proceed for a period of from about 2 to about 30 hours, more preferably from about 8 to about 24 hours, more preferably still from about 9 to about 12 hours, most preferably for about 10 hours.

WO 2019/090389

PCT/AU2018/051203

Desirably, on termination, the pregnant leach solution comprises from about 130 g/L to about 210 g/L nickel, more preferably from about 140 g/L to about 200 g/L nickel, more preferably still from about 150 g/L to about 195 g/L nickel, and most preferably about 192 g/L.

Typically under the processing conditions described herein, for example, when leaching is carried out at 80°C, a solution pH 3 generally corresponds to a nickel sulfate solution concentration of 192 g/L. Terminating the leaching process at about pH 3 is preferred as it provides a balance between (i) terminating at a lower pH, but requiring a much great amount of neutralisation agent to neutralise the excess free acid, and (ii) terminating at a higher pH, which means the reaction is far slower and would take much longer and so would require increased capital for larger reaction tanks where the total process throughput is fixed. As used herein, the term about represents a pius/minus deviation on a value corresponding to 1% in the context of pH measurements and 5% of concentration values. Bulk impurity removal in a related aspect, the method further comprises the step of removing the bulk nonnickel metal impurities from the discharge solution.

Thus, in one embodiment, the process further comprises the steps of:

(i) precipitating substantially all of the non-nickel metal impurities as an insoluble non-nickel metal impurity precipitate; and (ii) removing the precipitate from the discharge solution to form a non-nickel metal impurity-depleted discharge solution comprising a sub-saturated solution of nickel sulfate and trace amounts of the one or more non-nickel metal impurities.

Suitably, the precipitation step may include the step of oxidising one or more nonnickel metal impurities in the discharge solution. Such an oxidation step is preferably carried out prior to the precipitation step. Oxidation of one or more of the metal impurities can assist in formation of metal species which precipitate at a more desirable pH, for example, the pH of operation. For example, oxidation converts iron from ferrous to ferric whereby ferric hydroxide and other impurities can be more conveniently precipitated at about pH 5, Suitably, the precipitated non-nickel metal impurities comprise non-nickel metal hydroxides which are formed when the pH of the discharge solution is sufficiently increased to a level where such hydroxides can form. Suitably, the sulfuric acid may comprise one or more oxidising agents, for example, oxygen, preferably supplied by sparging the sulfuric acid with air. Where an oxidising agent is used, preferably it comprises a source of oxygen, for example, air, such as an air sparge. For example, wherein the oxidising step is carried out in an aeration tank, air can be sparged through the discharge solution.

if necessary, the precipitating step can be executed by increasing the pH of the discharge solution to a pH of about 3 to about 7, preferably to a pH of about 4.5 to about 5.5, more preferably to a pH of about 5, A particularly preferred precipitation pH is about 5.

WO 2019/090389

PCT/AU2018/051203

Suitably, the pH of the discharge solution may be increased by the addition of one or more basic compounds, preferably one or more basic nickel compounds, most preferably nickel hydroxide. Nickel hydroxide is suitable to raise the pH of the acidic sub-saturated solution (e.g. a pregnant leach solution) or the discharge solution to about pH 5 to about pH 5.5, with reasonable neutralisation kinetics and efficiency. This is achievable with sufficient leach residence time, if a higher pH is required for impurity precipitation, preferably, an ammonia solution is used to increase the pH further. However, in preferred embodiment, alkali bases, in particular, sodium or potassium hydroxide bases are avoided as these bases typically leads to alkali contamination in the final product which can be burdensome to remove.

Preferably, the process further includes the step of separating the non-nickel metal impurities precipitate from the discharge solution to form a non-nickel metal impurity-depleted discharge solution, in one embodiment, the non-nickel metal impurity-depleted discharge solution is a non-nickel metal hydroxide-depleted discharge solution. It will be understood that non-hydroxide solid materials, such as solid nickel powder, are also advantageously removed in the separating step. Means for separation include any suitable separation means, for example, decanting, centrifuging or filtering. Preferably, the non-nickel metal impuritydepleted discharge solution is filtered to remove any remaining solids prior to storage in the ion exchange (IX) feed tank and/or prior to ion exchange (IX). Suitably, the non-nickel metal impurity-depleted discharge solution is held in the ion exchange (IX) feed tank prior to the ion exchange processing. Suitably, filter sludge recovered from filtering is preferably transferred to a refinery final thickener for further processing or disposal.

Preferably, after non-metal impurity separation, the discharge solution has a pH of from about 3 to about 7, more preferably about 4 to 6, most preferably a pH of about 5.

Preferably, the bulk impurity removal process is associated the additional step of acid mist scrubbing to scrub off or remove any hydrogen gas evolved at this stage of the process. Most preferably, this scrubbing step is completely separate to the off-gas scrubbing used during leaching thereby ensuring hydrogen and oxygen do not mix in corresponding off-gas scrubbers.

If necessary, the steps of the bulk impurity removal process may be carried out by flushing the local environment around the leaching location with snuffing steam and/or a positive gauge pressure to prevent air ingress as described above. If necessary, the bulk impurity removal process may also include the step of carrying out the leaching step under a nitrogen blanket to prevent the hydrogen mixing with air to form an explosive environment as described above.

Ion exchange purification

WO 2019/090389

PCT/AU2018/051203 in another aspect, the invention provides a process for producing high purity nickel suitable for use in battery manufacture or nickel plating, preferably having a purity of >99.98% nickel sulfate, comprising the steps of:

selectively removing non-nickel metal impurities from a nickel sulfate solution by ion exchange (IX) by subjecting a solution of nickel sulfate comprising dissolved nickel sulfate and one or more non-nickel metal impurities to a first round of ion exchange using a resin preloaded with nickel, whereby non-nickel metal impurities in the nickel solution are selectively retained by the resin to generate a first IX discharge solution which is a clean nickel sulfate solution substantially free of non-nickel metal.

in a particularly preferred embodiment, the invention provides a process for producing a high purity nickel sulfate solution suitable for crystallisation of high purity, preferably >99.98% purity nickel sulfate hexahydrate, comprising the steps of:

(i) providing a nickel sulfate solution, preferably being a substantially solid-free acidic sub-saturated nickel sulfate solution, comprising dissolved nickel sulfate and one or more non-nickel metal impurities, preferably a pregnant leach solution, more preferably obtainable by a process as defined in any one of claims 5 to 19;

(ii) selectively removing non-nickel metal impurities from a nickel sulfate solution by ion exchange (IX) by subjecting a solution of nickel sulfate, comprising dissolved nickel sulfate and one or more non-nickel metal impurities to a first round of ion exchange using a resin pre-loaded with nickel, (ill) whereby non-nickel metal impurities in the nickel solution are selectively retained by the resin to generate a first IX discharge solution which is a clean nickel sulfate solution substantially free of non-nickel metal.

Preferably, the nickel sulfate solution is an acidic sub-saturated nickel sulfate, more preferably an acidic sub-saturated non-nickel metal impurity-depleted discharge solution, more particularly, such as one which has been subjected to a bulk non-nickel metal impurity removal process as described herein. In a particularly preferred embodiment, the nickel sulfate solution is a pregnant leach solution which may be obtainable by the leaching processes described herein. Suitably, the nickel sulfate solution from which the non-nickel metal impurities are removed is a subsaturated nickel sulfate solution, preferably having a nickel concentration of about 100 g/L to about 215 g/L nickel. Preferably, the sub-saturated nickel sulfate solution has an initial concentration of nickel in the range of from about 130 g/L to about 210 g/L nickel, more preferably from about 150 g/L to about 200 g/L nickel, more preferably still from about 175 g/L to about 195 g/L nickel, and most preferably about 190 g/L.

Typically, the non-nickel metal impurities are present, for example, in trace amounts, particularly if the nickel sulfate solution has been subject to a bulk non-nickel metal impurity

WO 2019/090389

PCT/AU2018/051203 removal process. Suitably, the non-nickel metal impurities include divalent metal cations, for example, divalent cobalt, iron and/or copper ions, preferably cobalt ions.

Suitably, the non-nickel metal impurities in the nickel sulfate solution are selectively retained by the resin in exchange for pre-loaded nickel and/or hydrogen ions.

Preferably, the subjecting step includes the step of buffering the nickel sulfate solution prior to ion exchange (IX) with one or more basic nickel compounds, preferably to a pH optimised for selectivity and stability of non-nickel metal impurity loading onto the nickel preloaded ion exchange resin. In particular, a preferred amount of nickel hydroxide is sufficient to neutralise acid released by the resin in exchange for nickel during nickel preloading so that the pH of the pre-IX and post-IX solutions are substantially equivalent, preferably within 10% of the relevant pH unit. The precise optimised pH may depend on factors including the chemistry of the resin used but will be readily be determinable as shown in the examples provided herein.

Preferably, prior to ion exchange, the nickel sulfate solution has an initial pH of pH < 6, preferably a pH of from about 4.5 to about 6, more preferably, a pH of about 5 to about 5.5, most preferably, a pH of about 5.

Suitably, after ion exchange, the nickel sulfate solution has a final pH of pH < 6, preferably a pH of from about 4.5 to about 6, more preferably, a pH of about 5 to about 5.5, most preferably, a pH of about 5.

Desirably, the first and one or more additional rounds of ion exchange are carried out at a temperature of from about 50 °C to about 95 °C, more preferably from about 70 °C to about 90 °C, most preferably at about 80°C.

Suitably the resin is capable of extracting metal ions from solution. More suitably still, the resin is capable of selective removal of metal ions from solution. Selective removal means one or more metal ions are removed from the solution in preference to different metal ions which remain in solution at a given pH/operating conditions. Desirably, the resin is a cation exchange resin. The resin may be comprised of one or more polymers based on at least one monovinylaromatic compound and/or at least one polyvinylaromatic compound. It is preferable when chelating resins for the purposes of the invention are polymers composed of for example, styrene, divinylbenzene and ethylstyrene. The resin is preferably in bead form. In a particularly preferred embodiment, the resin is a macroporous crosslinked polystyrene based resin, which is preferably acidic or otherwise capable of extracting metal ions from solution such as the nickel sulfate solutions, as described herein. In one embodiment, the resin is a metal chelating resin which comprises one or more metal chelating functional groups. Preferably, the resin comprises organophosphorus functional groups which are capable of selectively complexing with certain metal ions, for example, under certain pH conditions. Desirably such functional groups include phosphoric, phosphonic or phosphinic

WO 2019/090389

PCT/AU2018/051203 acid based functional groups. An example of a phosphoric acid based resin is one comprising di-2-ethylhexyl-phosphat (D2EHPA) functional groups (Lewatit VP OC 1026®). An example of a phosphonic acid based resin is one comprising aminomethyl phosphonic acid functional groups (Lewatit TP-260®). An example of a phosphinic acid based resin is one comprising bis-(2,4,4-trimethylpentyl-) phosphinic acid functional groups (Lewatit TP-272®). Preferably, the resin is not a TP-207 resin.

Preferably, the resin is provided in two or more columns arranged in a lead/lag configuration, whereby the lag column can act as (i) a polishing column when the impurity breakthrough of the lead column becomes undesirably, or as a new lead column where the original led column is subject to regeneration thereafter becoming the lag column.

Desirably, the first IX discharge solution is transferred to a crystallisation plant. Resin preloading

Control of solution pH is critical to the optimum selectively and efficiency of the ion exchange resin. Increasing the pH of the preloading feed solution typically results in increased nickel loading. However, if the pH is too high, undesirable species precipitation may occur and/or impurity loading may cease or decrease to unacceptably low levels. It should be understood that fresh and preloaded resin has available hydrogen ions for exchange with suitable metal ions. As the hydrogen ion exchange reaction releases two hydrogen ions for each divalent metal (II) ion taken up by the resin, the acidity of discharge/effluent from the ion exchange column tends towards increased acidity overtime such that the equilibrium pH tends to be greater than the initial pH. In some embodiment, a pronounced increase in acidity may be undesirable as this may have a negative effect on the resin’s ability to selectively uptake the non-nickel metal impurities, for example, cobalt in the case where a TP272 resin is used and/or may inhibit nickel preloading/uptake by the resin.

Therefore, methods for controlling and/or buffering the solutions are important.

In a preferred embodiment, the pre-loading step involves introducing nickel ions to the resin in fresh form (protonated) under conditions whereby hydrogen ions on the fresh resin are exchanged for the introduced nickel ions. The nickel ions are typically provided in the form of a nickel pre-load solution. Preferably, the process for selectively removing non-nickel metal impurities from a nickel sulfate solution further comprises the step of preloading the resin of the first round of ion exchange (IX) with nickel ions. Pre-loading with nickel reduces the hydrogen ion availability and thus assists in preventing as great an increase in equilibrium pH compared to that observed where fresh resin is used.

Desirably, as further pH control may be desirable, one or more buffering agents can be used to control the pH at a value that ensures impurity, particularly Co impurity, loading stability and optimum selectively. Preferably, the ion exchange process includes the step of buffering the acidity of the solution passing through the ion exchange resin which increases

WO 2019/090389

PCT/AU2018/051203 with time. Using nickel hydroxide as base/buffer advantageously avoids pH decrease during ion exchange avoids and avoids the introduction of alkali or ammonium contaminant during pH adjustment. Furthermore, the formation of ammoniacal nickel complexes is avoided during the pre-loading by avoiding ammonia pH adjustment; avoiding formation of nickel hydroxide precipitates at higher pH (pH > 6); minimising degradation and loss of resin functional groups by operating at relatively low pH < 6; and providing for minimum washing requirements. PreloadinQ feed solution & pH control

Suitably, the resin of the first round of ion exchange is preloaded with nickel using a nickel pre-loading feed solution, which is preferably a portion of the first IX discharge solution, or an alternative source of filtered clean nickel sulfate solution, preferably in the presence of nickel hydroxide for pH adjustment/buffering.

Suitably, the pH of the pre-loading feed solution, for example, the first IX discharge solution, may be buffered/increased to a desired pH by the addition of one or more basic compounds, preferably basic nickel compounds, most preferably nickel hydroxide. Nickel hydroxide is suitable to raise the pH of an acidic sub-saturated nickel sulfate solution to from about pH 5 to about pH 5.5. At higher pH, the nickel and/or other species may begin to precipitate. Most preferably, the resin is pre-loaded using clean nickel sulfate in the presence of nickel hydroxide for pH adjustment/buffering.

Preferably, the preloading feed solution pH is from about pH 4.5 to about pH 6, more preferably from about pH 4.0 to about pH 5. In this latter pH range minimal nickel precipitation occurred while maximising nickel loading onto resin. In embodiments using TP 272 resin, the preferred preloading feed pH is from about 4.5 to about 6, more preferably from 4.5 to about 5.5, most preferably the preloading feed pH is about 5. For TP 272 resin in particular, using a pre-loading feed having a pH greater than about 5.5 tends to adversely affect Co impurity removal by the column. It is preferred that alkali bases such as sodium hydroxide, potassium hydroxide or ammonium hydroxide bases are avoided as the may lead to contamination in the final nickel sulfate product

In a preferred embodiment, the pre-loading step involves introducing a dean nickel sulfate solution substantially free of non-nickel metal impurities. In other embodiments, the resin may be preloaded using an alternative source of filtered dean nickel sulfate solution. Preferably, the process involves the step of filtering the first IX discharge solution or alternative source of solution prior to preloading. Filtering prior to preloading advantageously removes residual solids, for example, resin particles.

Desirably, the nickel pre-loading is maximised using a maximum strength nickel preloading feed, preferably, a maximum strength dean nickel sulfate solution as described herein. Preferably, the nickel pre-loading feed solution has an initial concentration of nickel in the range of from about 130 g/L to about 210 g/L nickel, more preferably from about 150 g/L

WO 2019/090389

PCT/AU2018/051203 to about 200 g/L nickel, more preferably still from about 175 g/L to about 195 g/L nickel, and most preferably about 190 g/L.

MlOHli.

Desirably a portion of the first IX discharge solution is used to generate nickel hydroxide and/or to preload the resin of the first round of ion exchange (IX) with nickel. In a preferred embodiment, the nickel hydroxide base is generated from a portion of the IX purified clean nickel sulfate solution, which is the first IX discharge solution. Suitably, the clean nickel sulfate solution substantially free of non-nickel metal impurities may be a portion of the first IX discharge solution as described above. Alternatively, the nickel hydroxide could be produced from a bleed of a nickel powder refinery process, for example, a Sherritt Gordon Process, via ammonia steam stripping process to generate nickel hydroxide.

Desirably, the amount of nickel hydroxide provided is sufficient to neutralise acid released by the resin in exchange for nickel during nickel preloading. A preferred amount of nickel hydroxide is one that can substantially stabilise the pH pre-loading feed solution such that the pH of the solution a column of resin is substantially the same as the pH of the solution discharging from the column of resin. In embodiments using TP 272 resin, the preferred preloading feed pH is from about 4.5 to about 6, more preferably from 4.5 to about 5.5, most preferably the preloading feed pH is about 5. For TP 272 resin, using a pre-loading feed having a pH greater than about 5.5 tends to adversely affect Co impurity removal by the column.

in preferred embodiments, the process for selectively removing non-nickel metal impurities from a nickel sulfate solution further comprises the step of generating the nickel hydroxide for pH adjustment/buffering from clean nickel sulfate solution.

Preferably the process further comprises subjecting the first IX discharge solution to a second round of ion exchange using nickel preloaded resin to provide a second IX discharge solution being a polished IX discharge solution.

Suitably, the second round of ion exchange may be used where impurity break through is observed in the first IX discharge solution.

Resin regeneration

Suitably, the process further comprises the step of regenerating the resin used in ion exchange, preferably on observation that (I) loading of substantially ail of the non-nickel metal impurities onto the resin has been achieved or (ii) break through of the non-nickel metal impurity in the first IX discharge solution. In a preferred embodiment, cobalt breakthrough in the IX column discharge solution can signify that resin regeneration or replacement is required.

WO 2019/090389

PCT/AU2018/051203

Desirably, the method further comprises the step of replacing or regenerating the resin used in the ion exchange rounds. Desirably, the requirement for replacement or regeneration of the resin may be determined for example on observation that loading of substantially all of the non-nickel metal impurities onto the resin has been achieved and/or that the non-nickel metal impurity concentration begins to increase in the first IX discharge solution.

Suitably, regenerating resin which has been used in ion exchange involves the steps of:

(i) washing the resin free of entrained nickel with an aqueous solution of acid, preferably sulfuric acid, preferably having a pH of about pH 4 to about pH 5;

(ii) stripping retained non-nickel metal impurities from the resin using an aqueous acid solution, preferably sulfuric acid, preferably at a concentration of from about 0.25 M to about 2.5 M acid, more preferably from about 1.0 M to about 2.0 M acid, wherein the stripping step exchanges the resin retained non-metal impurities for hydrogen ions in the acid, thereby loading hydrogen ions onto the resin to provide resin in the hydrogen form;

(iii) removing excess acid from the resin by washing with water, preferably demineralised water;

(iv) pre-loading the resin with nickel by flushing the resin with a nickel pre-loading feed solution wherein the pre-loading exchanges at least a portion of the hydrogen ions loaded onto the resin for nickel ions in the nickel pre-load feed, thereby loading nickel ions onto the resin to provide resin in nickel pre-loaded form.

Preferably, step (iv) is carried out as described above.

Suitably, nickel pre-loading feed solution has a pH about pH 4.5 to about pH 6, more preferably from about pH 4.0 to about pH 5. in this latter pH range minimal nickel precipitation occurred while maximising nickel loading onto resin.

As discussed above, it is desirable that the nickel pre-loading feed solution is a clean nickel sulfate solution, more preferably being a portion of the first IX discharge solution. It should be understood that the clean nickel sulfate solution and the first IX discharge solution described herein is an acidic sub-saturated nickel sulfate solution, preferable obtainable by a leaching process as described herein.

in a particularly preferred embodiment, the nickel pre-loading feed solution further comprises a buffering amount of nickel hydroxide. In one embodiment, a buffering amount of nickel hydroxide is one which is sufficient to optimise for selectivity and stability of non-nickel metal impurity loading onto the nickel pre-loaded ion exchange resin

Desirably, the nickel pre-loading feed solution comprising nickel sulfate and/or nickel hydroxide is preferably filtered prior to the pre-loading step.

WO 2019/090389

PCT/AU2018/051203

A slightly acidic pH for the wash, for example, a pH of about 3 to about 6, more preferably from about 4 to about 5, avoids dissolution of the active resin component which would lead to a loss of ion exchange capacity. In the case of TP 272, a pH of about 4 to about 5 is particularly suitable to preserve the resin.

Advantageously, the wash sulfuric acid may be recycled to a leach acid solution for reuse.

Desirably, the strip solution comprising the removed impurities is transferred to a waste collection tank.

Crystallisation

In particularly preferred embodiments, the clean nickel sulfate solution preferably being the first IX discharge solution, is crystallised to form high purity nickel sulfate crystals, preferably, nickel sulfate hexahydrate crystals, for example, in the alpha form. Suitably, in the process of the invention the sub-saturated nickel sulfate solution is an acidic pregnant leach solution, for exampie, obtainable from a nickel sulfate leaching process as defined herein, and which has been subjected to bulk impurity removal and trace impurity removal via ion exchange as described herein. Suitably, the clean nickel sulfate solution may be filtered prior to crystallisation.

In another aspect, the invention provides a process for producing high purity nickel suitable for use in battery manufacture or nickel plating, preferably having a purity of >99.98% nickel sulfate, comprising the steps of:

(I) providing a solution sub-saturated with nickel sulfate and comprising one or more non-nickel metal impurities;

(ii) removing the bulk of the one or more non-nickel metal impurities by raising the pH of the sub-saturated nickel sulfate solution by adding a basic nickel compound, preferably nickel hydroxide, to the sub-saturated nickel sulfate solution and separating the non-nickel metal impurities from the solution in the form of precipitated insoluble non-nickel metal hydroxides;

(ill) selectively removing remaining trace non-nickel metal impurities from the subsaturated nickel sulfate solution using one or more rounds of ion exchange purification involving at least a first round of ion exchange using a resin pre-loaded with nickel to form a first IX discharge solution comprising nickel sulfate and a reduced amount of non-nickel metal impurities which is a clean nickel sulfate solution being a substantially non-nickel metal impurities free sub-saturated nickel sulfate solution;

(iv) recovering high purity nickel sulfate from the substantially non-nickel metal impurities free sub-saturated nickel sulfate solution, preferably in the form of alpha nickel sulfate hexahydrate.

WO 2019/090389

PCT/AU2018/051203

Suitably, there is provided a process for producing high purity nickel suitable for use in battery manufacture or nickel plating, preferably having a purity of >99.98% nickel sulfate, comprising the steps of:

(I) providing a solution sub-saturated with nickel sulfate and comprising one or more non-nickel metal impurities, preferably a pregnant leach solution, more preferably obtainable by a process as defined herein;

(iii) selectively removing remaining trace non-nickel metal impurities from the subsaturated nickel sulfate solution using one or more rounds of ion exchange purification involving at least a first round of ion exchange using a resin pre-loaded with nickel to form a first IX discharge solution comprising nickel sulfate and a reduced amount of non-nickel metal impurities which is a clean nickel sulfate solution being a substantially non-nickel metal impurities free sub-saturated nickel sulfate solution;

Preferably, the process further comprises the optional step of oxidising non-nickel metal impurities in the sub-saturated nickel sulfate solution prior to the step of raising the pH of the sub-saturated nickel sulfate solution to induce precipitation of insoluble non-nickel metal impurities.

Suitably, the insoluble precipitate is formed by increasing the pH of the discharge solution to a pH of about 3 to about 7, preferably to a pH of about 4.5 to about 5,5, more preferably to a pH of about 5, and separating the precipitate to form a non-nickel metal impurities depleted, preferably a non-nickel metal hydroxides-depleted sub-saturated nickel sulfate solution and which has been subjected to bulk impurity removal and trace impurity removal via ion exchange as described herein. Suitably, the clean nickel sulfate solution may be filtered prior to crystallisation.

Desirably, recovering step (iv) involves crystallising the non-nickel metal impurity depleted sub-saturated nickel sulfate solution to form high purity nickel sulfate crystals, preferably nickel sulfate hexahydrate crystals, more preferably alpha nickel sulfate hexahydrate crystals. Crystalline alpha nickel sulfate hexahydrate is particularly desirable as it is the industry standard for applications such as battery manufacture and nickel plating, including electroless and electopiating.

WO 2019/090389

PCT/AU2018/051203 /

In a preferred embodiment, the sub-saturated nickel sulfate solution is an acidic pregnant leach solution, for example, obtainable by leaching nickel powder in sulfuric acid, for example, by a process as defined herein.

Suitably, the selectively removal step (ill) is an ion exchange process as defined herein.

Preferably, the crystallising step is carried out at a temperature of from about 45 °C to about 65 °C, preferably from about 50 °C to about 60 °C, most preferably at about 53°C and/or under a pressure of about 10 kPa absolute, preferably forming alpha crystalline nickel sulfate hexahydrate.

Suitably, the crystallising step is carried out in an evaporative crystailiser, for example, a mechanical vapour recompression type crystailiser, preferably a draft tube baffle crystailiser.

Suitably, the process further comprises the step of separating the crystals from the nickel sulfate solution, for example, dewatering using a filter, centrifuge or cyclone.

Preferably, the process further comprises the step of drying the crystals to provide dry high purity nickel sulfate hexahydrate, for example, in a fluid bed dryer.

Plant

In another aspect, the invention provides a processing plant capable of producing high purity (>99.8%, more preferably >99.98%) nickel sulfate from nickel powder comprising:

(i) an acidic nickel leach module for generating sub-saturated pregnant leach solution comprising nickel sulfate and non-nickel metal impurities, preferably configured for batch leaching;

(ii) downstream of the acidic nickel leach module, a bulk non-nickel metal impurity removal module for precipitating the bulk of the non-nickel metal impurities, preferably configured to oxidise oxidisable non-nickel metal impurities in the pregnant leach solution;

(Hi) downstream of the bulk non-nickel metal impurity removal module, a trace non-nickel metal impurity ion exchange removal module for removing trace amounts of nonnickel metal impurity to form a purified sub-saturated nickel sulfate solution substantially free of non-nickel metal impurity, preferably configured for one or more rounds of ion exchange in iead/lag configuration; and optionally, (iv) downstream of the trace non-nickel metal impurity ion exchange removal module, a nickel sulfate crystallisation module for crystallisation of high purity nickel sulfate crystals from the purified sub-saturated nickel sulfate solution substantially free of non-nickel metal impurity, preferably configured to crystallise alpha-nickel sulfate hexahydrate.

Desirably, the plant further comprises a nickel hydroxide formation module for preparation of nickel hydroxide for use as an acid neutraliser and/or as a pH buffer in the

WO 2019/090389

PCT/AU2018/051203 production of the high purity nickel sulfate, wherein the nickel hydroxide formation module is in communication with the bulk non-nickel metal impurity removal module and/or the trace non-nickel metal impurity ion exchange removal module for providing nickel hydroxide solution thereto.

Preferably, the nickel hydroxide formation module comprises a closed circuit for forming a nickel hydroxide solution such that ammonia and/or NaOH used for neutralisation and pH adjustment during nickel hydroxide preparation is isolated from the bulk non-nickel metal impurity removal module and/or the trace non-nickel metal impurity ion exchange removal module thereby avoiding contamination of nickel sulfate solutions.

Suitably, the bulk non-nickel metal impurity oxidiser module is in communication with a source of oxidant, for example, air, and a source of nickel hydroxide acid neutraliser and/or as a pH buffer preferably from the nickel hydroxide formation module as described herein.

In a preferred embodiment, the plant further comprise a hydrogen gas mitigation module associated with at least module (i) and/or module (ii) for management of hydrogen gas evolved during processing.

Suitably, the hydrogen gas mitigation module comprises a source of a steam flush for management of hydrogen evolved during the process. Preferably, the hydrogen gas mitigation module is configured to generate and maintain a steam flush comprising about 70 vol% steam. The hydrogen gas mitigation module comprises a scrubber for scrubbing acidic mist from the offgas prior to venting to the atmosphere.

In a preferred embodiment, the hydrogen gas mitigation module further comprises means for processing nickel leaching under a positive gauge pressure to prevent air ingress.

Suitably, the hydrogen gas mitigation module further comprises a means for providing a blanket of nitrogen gas over at least module (I) and/or module (ii) to assist in the prevention of the formation of a potentially explosive hydrogen/air mixture.

Preferably, the plant is configured to provide sulfuric acid from a refinery to the plant, wherein the plant is configured to provide sulfuric acid to the acidic nickel leach module for leaching nickel sulfate from nickel powder and/or sulfuric acid ion exchange removal module for ion-exchange stripping during regeneration. The plant is provided with a means for producing and/or providing demineralised water which is used to prepared required acid dilutions and wash for the ion exchange steps.

Preferably, the plant further comprises a nickel sulfate dewatering module for removing the bulk of solute from nickel sulfate crystals downstream of the nickel sulfate crystallisation module.

Suitably, the plant further comprises a nickel sulfate crystal dryer module to dry the crystals to a high purity nickel sulfate hexahydrate product in powder form.

WO 2019/090389

PCT/AU2018/051203

In a preferred embodiment, the plant further comprises a high purity nickel sulfate product packaging module.

Preferably, the plant Is configured to recycle condensates from one or more plant modules, for example, heating coils In the leaching module, the product dryer module, and/or the crystallisation module for reuse.

Suitably, the plant further includes one or more off-gas scrubbers, for example, associated with the acidic nickel leach module, the bulk non-nlckel metal impurity oxidiser module and/or the bulk non-nlckel metal impurity removal module for capture of gases and the nickel sulfate crystallisation module to capture nickel sulfate dust.

In another aspect, the invention provides for a use of nickel hydroxide as an acid neutraliser and/or a pH buffer in a process for preparing high purify nickel sulfate, preferably having a purity of >99.8%, more preferably >99.98% nickel sulfate suitable for use in battery manufacture or nickel plating including electroplating and electroless plating.

In another aspect, the invention provides for a use of nickel hydroxide as an acid neutraliser and/or a pH buffer in an ion exchange purification process for selective removal of non-nickel metal impurities from a nickel sulfate solution, preferable an acidic sub-saturated nickel sulfate solution, more preferably, an acidic sub-saturated nickel sulfate solution from which bulk non-nickel metal impurities have been removed.

In another aspect, the invention provides for a use of nickel hydroxide as an acid neutraliser and/or a pH buffer in nickel sulfate solution, preferable an acidic sub-saturated nickel sulfate solution to precipitate non-nickel metal impurities in as insoluble non-nickel metal impurities, for example, non-nickel metal hydroxides.

In another aspect, the invention provides fora use of a nickel pre-loaded ion exchange resin to selectively remove non-nickel metal impurities from a nickel sulfate solution, preferable an acidic sub-saturated nickel sulfate solution, more preferably, an acidic subsaturated nickel sulfate solution from which bulk non-nickel metal impurities have been removed.

Suitably, the resin is a phosphoric acid, a phosphonic acid or phosphinic acid resin.

In another aspect, the invention provides fora use of a nitrogen gas blanket to prevent formation of explosive air and hydrogen mixtures over an acidic nickel leach evolving hydrogen.

In another aspect, the invention provides for a use of a sub-safurated nickel sulfate solution in a process for the preparation of high purity nickel sulfate, preferably having a purity of >99.8%, more preferably >99.98% nickel sulfate, and preferably wherein the sub-saturated nickel sulfate solution is a pregnant leach solution, for example, obtainable by a process according to the invention.

WO 2019/090389

PCT/AU2018/051203

In another aspect, the invention provides for high purity nickel sulfate obtainable by a process according to the invention.

In another aspect, the invention provides for a use of high purify nickel sulfate as defined in herein in the manufacture of an energy storage device, for example, a battery or capacitor and/or in a nickel plating process including electroplating and electroless plating.

In another aspect, the invention provides for an energy storage device, for example, a battery or capacitor, comprising nickel sulfate obtainable by a process according to the invention.

In another aspect, the invention provides for a product comprising plated nickel derived from comprising nickel sulfate obtainable by a process according to the invention.

Brief Description of Drawings

Figure 1 illustrates a process flow diagram for an exemplary embodiment of the production of high purity of nickel sulfate as described herein;

Figure 2 illustrates the particle size distribution (PSD) of the nickel powder sample used in the bench scale experiments;

Figure 3(A) - (G) illustrate the results of bench scale experiments relating to the effect of pulp density on leaching; (A) Effect of pulp density on nickel powder dissolution in tests maintained at pH 0 and 1. Tests were maintained at 80 °C, with a compressed air aeration rate of 5 L/min. All teste were operated for a period of 10 hours; (B) Impact of Pulp density on teaching rate at pH 1,80°C; (C) Comparison of Nickel concentration for tests operated under anaerobic conditions at 290 g/L sulfuric acid, 80 °C; (D) Impact of Pulp density on Neutralisation Rate under anaerobic conditions; (E) Comparison of particle size distribution for tests operated under anaerobic conditions at 290 g/L sulfuric acid, 80 °C; (F) Impact of Pulp density on Ni Extraction for Anaerobic conditions with batch acid addition; (G) impact of Pulp Density on Ni Extraction Rate for Anaerobic conditions with batch acid addition;

Figure 4(A) - (D) illustrate the results of bench scale experiments relating to the effect of pH on leaching; (A) Effect of operational pH on nickel powder dissolution. Tests were maintained at 80 °C, with a compressed air aeration rate of 5 L/min. All tests were operated for a period of 10 hours with 1000 g/L nickel powder present; (B) Comparison of simple free hydrogen ion concentration and total nickel dissolution; (C) Comparison of total nickel dissolved and total sulfuric acid added over all 31 teste, excluding anomalies of tests 7, 8 and 12; (D) Aerobic Impact of Leaching Time on pH;

Figure 5(A) - (C) illustrate the results of bench scale experiments relating to effect of aeration rate on leaching; (A) impact of aeration rate and nitrogen on nickel powder dissolution at 80 °C, 1000 g/L nickel powder and pH 1.0. Red data point is N₂-sparged test;

WO 2019/090389

PCT/AU2018/051203 (B) Impact of aeration rate on measured solution evaporation; (C) Effect of Aeration rate on Ni Extraction Rate;

Figure 6(A) - (E) illustrate the results of bench scale experiments relating to effect of pH on impurity leaching; (A) Comparison of pH; (B) Comparison of free acidity; (C) Comparison of Nickel; (D), Comparison of Cobalt (E) and Comparison of Copper (F) in tests operated under anaerobic conditions at 237, 256, 275 and 293 g/L sulfuric acid. Note that Cu and Co are reported in mg/L compared to Nickel (g/L);

Figure 7(A) & (B) illustrate the results of bench scale experiments relating to effect of temperature on leaching; (A) impact of temperature on nickel powder dissolution at pH 1.0, 1000 g/L nickel and 5 L/min aeration. Red data pointe represent teste undertaken using uncoated impeliors; (B) Effect of Temperature on Leaching Rate;

Figure 8(A) - (G) illustrate the results of bench scale experiments relating to batch acid leaching; (A) impact of acid loading on nickel powder dissolution at 80 °C with 1000 g/L nickel and 5 L/min aeration. (B) Neutralisation with time for acid loading fests at 80 and 100 °C; (C) Nickel concentration overtime during tests operated under near-optimal conditions. (D) Solution pH over time in near optimal tests. Tests were operated at 80 °C, 1000 g/L nickel and 1 L/min aeration; (E) Aerobic Batch Acid Addition Ni Extraction Rate; (F) Impact of acid loading on nickel extraction rate; (G) Batch acid addition pH profile;

Figure 9(A) - (H) illustrate the results of bench scale experiments relating to the effect of acid concentration on impurity leaching for Co, F, Cr, K, Ca, Na, Cu, Zn respectively; Selected impurity element concentrations throughout experiments. Note that Cu is reported in pg/L (ppb), compared to all other elements in mg/L (ppm);

Figure 10 illustrates a comparison of iron, copper and zinc concentrations to pH. Note that Cu and Zn are reported in pg/L (ppb), Fe reported in mg/L (ppm);

Figure 11(A) & (B) illustrate the results of bench scale experiments relating to the effect of agitation on leaching; (A) Comparison of Nickel concentration; (B) pH for tests operated under anaerobic conditions at 290 g/L sulfuric acid, 80 °C;

Figure 12(A) - (E) illustrates a schematic of the IX process of the invention and results of bench scale experiments relating to the effect of preloading of resin on IX; (A) schematic of the IX process of the invention; (B) pH isotherms with fresh TP272 without pH control showing the effect of utilizing the ion exchange resin TP 272 without pre-loading; (C) Ni pre-loading with Ni sulphate solution (initial 20 g/L Ni); (D) Ni loading pH profile with various nickel concentration in the feed solutions (80 °C); (E) Ni pre-loading kinetics for pH 5 and 6 at 80 °C;

Figure 13(A) - (K) illustrates the results of bench scale experiments relating to the effect of non-preloaded resin and comparison to nickel pre-loaded resin; (A) Comparison of initial solution pH with final pH after equilibrium loading; (B) Metal loading (%) pH isotherms

WO 2019/090389

PCT/AU2018/051203 with fresh TP 272 vs. final pH without pH control; (C) Metal loading (mg/g resin) pH isotherms with fresh TP 272 vs. final pH without pH control; (D) Metal loading (mg/g) pH isotherms with Ni pre-loaded resins at pH 5 with nickel hydroxide; (E) Metal loading (%) pH isotherms with Ni pre-loaded resins at pH 5 with nickel hydroxide; (F) Raffinate metal concentrations; (G) Metals in the neutralised synthetic PLS; (H) Metal loading (mg/g) pH isotherms with Ni preloaded resins at pH 6 with ammonium hydroxide; (I) Metal loading (%) pH isotherms with Ni pre-loaded resins at pH 6 with ammonium hydroxide; (J) Raffinate metal concentrations; (K) Final equilibrium pH vs. initial pH using the Ni preloaded resins and the synthetic PLS with pH pre-adjusted with nickel hydroxide;

Figure 14(A) - (D) illustrates (A) pH isotherm profile for TP 272; (B) pH isotherm profile for Cyanex 272 resin;

Figure 15(A) - (D) illustrates the results of bench scale experiments relating to cobalt stripping kinetics; (A) Ni and Co loading (mg/g) vs. loading time; (B) Ni and Co loading (%) vs. loading time; (C) Mass balance of Ni and Co loading; (D) Metal stripping (%) vs. stripping time;

Figure 16 illustrates the results of bench scale experiments relating to the effect of pH (H₂SO₄ concentration) on % metal stripping;

Figure 17(A) - (H) illustrates the results of bench scale experiments relating to the results based on air dry mass; (A) Metal loadings (%) vs. resin / solution ratio; (B) Metal loading (mg/g) vs. resin / solution ratio (g/mL); (C) Metal in raffinate vs. resin / solution ratio;

(D) Co distribution isotherms on air dry resin basis; (E) Comparison of Co distribution isotherm curves based on air dry and estimated dry resin mass; (F) Langmuir model Fitting 1: linear part by taking 6 data points; (G) Langmuir model Fitting 2: linear part by taking 8 data points; (H) Freundlich isotherm equation fitting: linear part by taking 8 data points; (I) Comparison of Ni and Co loadings and variation of equilibrium pH;

Figure 18 illustrates the results of bench scale experiments relating to the total exchange capacity vs. stability test cycles;

Figure 19(A) - (E) illustrates the results of bench scale experiments relating to the variation of exchange capacity; (A) Metal loadings on the resin vs. cycles; (B) Co behaviour in the stability tests; (C) Ni behaviour in the stability tests; (D) Fe behaviour in the stability tests;

(E) Cu behaviour in the stability tests;

WO 2019/090389

PCT/AU2018/051203

Figure 20(A) 20(E) illustrates the results of bench scale experiments relating to the effect of consecutive loading and sampling; (A) Variation of raffinate metal concentrations with pH; (B) Variation of metal loadings with pH based on aqueous assays; (C) Metal loading efficiency (%) vs. pH with TP 207; (D) Raffinate metal concentrations vs. pH with TP 207; (E) Metal mass loading vs. pH with TP 207;

Figure 21(A) & (B) illustrates the results of bench scale experiments relating to the loading kinetics with TP 207; (A) Raffinate metals vs. time with TP 207; (B) Metal loadings vs. time with TP 207;

Figure 22(A) - 22(D) illustrates the results of bench scale experiments relating to loading distribution isotherms for TP 207; (A) Metal loading (%) vs. solution / resin ratios; (B) Raffinate metal concentration with solution / resin ratios; (C) Metal loading at various AJ Resin (TP 207) ratios; (D) Fe distribution isotherm with resin TP 207;

Figure 23 illustrates the results of bench scale experiments relating to the elution of TP In particular to elution acidity isotherms with TP 207;

Figure 24(A) - 24(C) illustrates the results of bench scale experiments relating to IX Amberlite IRC 748 resin; (A) Metal loading efficiency vs. pH with IRC 748 resins; (B) Raffinate metals vs. pH with IRC 748 resin; (C) Metal loadings on resin vs. pH with IRC 748 resin;

Figure 25 illustrates the results of bench scale experiments relating to IX -Purolite; (A) Metal loading pH isotherms with Purolite S910 resin;

Figure 26(A) - 26(C) illustrates the results of bench scale experiments relating to SX - cyanex272; (A) Metal extraction pH isotherms with 10% Cyanex272 in Shellsol D70 at 80 ° and 1:1 A/O ratio; (B) Metal extraction in organic phase with 10% Cyanex272 in Shellsol D70 at 80 ° and 1:1 A/O ratio; (C) Metals in raffinate vs. pH with 10% Cyanex272 in Shellsol D70 at 80 ° and 1:1 A/O ratio;

Figure 27(A) & 27(B) illustrates the results of bench scale experiments relating to column loading tests; (A) Column 1 metal loading profile; (B) Column metal loading sampled on the column top;

Figure 28 illustrates the results of bench scale experiments relating to polishing/loading experiments Raffinate Co assayed by ICP-MS vs. bed volume for the samples from Column 3 - Polishing Runs with Column 1 raffinate solutions as feed at 3 different flow rates or residence time;

Figure 23 illustrates the results of bench scale experiments relating to the effect of column washing in particular a column Co loading washing profile;

Figure 30 illustrates the results of bench scale experiments relating to column washing in particular the elution profile at different acidities (H₂SO₄ concentrations).

WO 2019/090389

PCT/AU2018/051203

Figure 31 illustrates the results of bench scale experiments relating to column elution washing;

Figure 32(A) - 32(E) illustrates the results of bench scale experiments relating to the effect of column tests; (A) breakthrough point of TP 272 resin; (B) cobalt breakthrough curve comparison.

Detailed Description

Overall description of a preferred embodiment of the process

A flowsheet to produce high purity nickel sulfate is shown in Figure 1.

Nickel leach module A

Nickel powder 100 is leached in a batch process in a nickel leach module A. Nickel leach module A comprises one or more leach acid solution preparation tanks 110, preferably a plurality of tanks, which are in fluid communication with the following sources of consumables: a source of steam and/or nitrogen gas 150, a source of sulfuric acid 130, a source of demineralised water 120 for diluting the sulfuric acid to a desired concentration, and a source of nickel powder 100, for example directly from a nearby nickel powder refinery. The tanks are adapted to include a source of steam for heating, for example, indirect steam heating via heating coils associated with the tanks 110. Agitation is provided by the leach acid solution preparation tank agitators (not shown) which are associated with the tanks 110. The nickel leach module A also comprises a leach gas scrubber 170 for scrubbing hydrogen evolved from leaching from the vapour spaces about the tanks 110. The leach gas scrubber 170 is configured to vent scrubbed off-gas 160 to the atmosphere.

The bulk impurity removal module B comprises one or more solution aeration vessels 190 which is in communication a source of oxygen/air 200 for sparging into the vessels 190 to convert iron impurity in the discharge solution 140 from the ferrous to ferric oxidation state. The aeration vessels 190 are in fluid communication with a source of a nickel hydroxide slurry 390 which is generated in a nickel hydroxide preparation module C. The aeration vessels 190 are equipped with separate acid mist scrubber (not shown) to the leach tanks 110 to ensure hydrogen and oxygen do not mix in the off-gas scrubbers. The scrubber is configured to vent scrubbed off-gas 220 to the atmosphere. The vessels are in communication with one or more solution polishing filters 230 for removing any remaining solids to generate a filtered non-nickel metal impurity depleted discharge solution 250 ready for ion exchange which occurs in ion exchange module C. The polishing filters 230 are ideally configured to operate in a duty/standby arrangement, and can be equipped with a backwash cycle and a solution filter sludge tank (not shown) and transfer pump (not shown). Module B

WO 2019/090389

PCT/AU2018/051203 further comprises an IX feed tank (not shown) for holding filtered non-nickel metal impurity depleted discharge solution 250 before IX.

Ion exchange module C

This module comprises one or more ion exchange (IX) columns 270 which ideally are arranged in a lead/lag configuration, A third column (not shown) may be provided off-line for regeneration which includes washing, stripping and pre-loading. The fourth column (not shown) may be installed as a spare. The columns will be loaded with suitable resin. An ion exchange discharge tank (not shown) is included in module C for holding resultant first IX discharge solution which is a clean nickel sulfate solution 310 substantially free of non-nickel metal impurities. The tank is associated with one or more ion exchange discharge polishing filters 400 for removing any carry-over resin or solids from the IX columns 270. A crystailiser plant feed tank (not shown) holds filtered clean nickel sulfate solution 420 and is in fluid communication with the crystailiser 450 of a crystailiser plant E. The fluid line between the one or more ion exchange discharge polishing filters 400 and the crystailiser 450 configured to divert a small portion of the filtered clean nickel sulfate solution 420, a nickel hydroxide preparation module B and to an ion exchange nickel preload tank (not shown) to be held until IX column 270 regeneration is required. The IX columns 270 adapted to be in fluid communication with a source of acid, demineralised water and nickel preload solution from the nickel hydroxide preparation module B as well as an ion exchange waste tank (not shown). The one or more ion exchange discharge polishing filters 400 are in fluid communication with a source of demineralised water 410 and the crystailiser 450,

Leaching module A

Nickel powder 100 is leached in a batch process in a nickel leach module A which comprises one or more leach acid solution preparation tanks 110, preferably a plurality of tanks. The nickel powder 100 may be provided to the nickel powder teach plant A from a discharge of wet metals driers in a nickel leach plant before the powder is either packaged or converted in to briquettes. The batch process begins with the transfer of nickel powder 100 to the tanks 110. The target concentration of nickel powder in solution is 1000 g/L. This is a significant excess in the amount of nickel than that stoichiometrically needed for the reaction, to ensure high reaction kinetics. Typically only approximately 10% of the nickel powder in the tanks 110 reacts during the batch cycle. The reaction proceeds are per the equation below,

Niy_s) + H₂SO₄ -> + Ni.S0_4(aq) (1)

In the leach acid solution preparation tanks 110, sulfuric acid 130 is continuously mixed with demineralised water 120 to generate acid of a target concentration. The acid dilution is exothermic and this step must be operated with caution. The tanks 110 are then fed

WO 2019/090389

PCT/AU2018/051203 with a batch make-up of sulfuric acid solution 130 at a sulfuric acid concentration of approximately 280 g/L. The sulfuric acid solution 130 is transferred to the batch leach tanks 110 on demand using an acid solution pump (not shown). This acid concentration is controlled by addition of demineralised water 120 in order to have the discharge solution 140 at 95% of the saturation limit of nickel sulphate at 80^!>C. The demineralized water 120 for acid dilution may be supplied by hydrogen plant demineralized water pumps (not shown). Operating close to saturation at 80°C minimises the evaporation duty of the crystalliser 450 in crystallisation module E, and eliminates the need fora pre-evaporator. Heating, preferably steam heating, is applied indirectly via heating coils (not shown) in the tanks 110, to maintain the tanks 110 temperature at 80°C. Agitation is provided by the leach acid solution preparation tank agitators (not shown).

The leaching process evolves hydrogen. To prevent the formation of an explosive environment, the vapour space about the leach acid solution preparation tanks 110 is flushed with flush steam 150 and the tanks are preferably operated at a positive pressure to prevent ingress air. The flush steam (snuffing steam) 150 ideally maintains a minimum 70 voi% steam atmosphere in the vapour space about the tanks 110. The resultant hydrogen containing offgas 160 is processed by a leach off-gas scrubber 170 which captures acid mist and particulates before the scrubbed off-gas 180 is discharged via a stack (not shown). In an emergency, where the tanks 110 are operating but the off-gas scrubber 170 fails, the vapour space about the tanks 110 which are connected to a source of nitrogen (not shown) can be flooded with nitrogen 150 to form a nitrogen blanket which can prevent the formation of an explosive environment.

At the completion of the leach, the residual acid concentration in the tanks 110 correlates to approximately pH 3, and the nickel sulfate in solution concentration approximately 192 g/L. The terminal nickel strength is determined by the amount of acid added to the batch (as there is a vast excess of nickel). The decision to terminate the reaction at pH 3 is a trade-off between (i) terminating at a lower pH, but requiring a much increased amount of nickel hydroxide 390 to neutralise the excess free acid, and (ii) terminating at a higher pH, meaning the reaction takes much longer and so would require increased capital for the larger reaction tank 110, given total throughput is fixed.

At the end of the leaching process, agitation is stopped and the solids are settled. The leach discharge solution 140 is decanted from the tanks 110 via a leach decant pump (not shown). Un-ieached solids remain in the heel of the leach tanks 110 and are reprocessed in the next batch of the leach cycle.

An exemplary leach batch cycle is shown below in Table 1:

WO 2019/090389

PCT/AU2018/051203

Table 1 Step by Step Activity	Vakre	Units
STEP 1 : Load Acid/Demin Mix into Tank		minutes
STEP 2 : Load Nickel into Tank	20	minutes
STEP 3 : Leach	24
STEP 4 : Settling		minutes
STEP 5 : Decant Liquor	45	minutes
Technical time (hold time)	444	minutes
Total Batch Leach Cycle Time	2S-.4	bGUFS

The leach discharge solution 140 from the batch leaching process carried out in module A is discharged to a solution surge tank (not shown), which stores the discharge solution 140 thereby providing a flow buffer between the batch upstream (module A) and continuous downstream processes (modules B to E).

The leach discharge solution 140 is then fed to one or more solution aeration vessels 190 provided in a bulk impurity removal module B, where oxygen/air 200 is sparged into the vessels 190 to convert iron impurity in the discharge solution 140 from the ferrous to ferric oxidation state. Bulk iron and other impurities are then precipitated by the addition of a nickel hydroxide slurry 390 which is generated from a nickel hydroxide preparation module B. Addition of the nickel hydroxide slurry 390 raises the leach discharge solution 140 to about pH 5. At this pH, residual bulk iron in the discharge solution 140 precipitates as ferric hydroxide. The aeration vessels 190 have their own separate acid mist scrubber (not shown) to the leach tanks 110 to ensure hydrogen and oxygen do not mix in the off-gas scrubbers. After scrubbing, generated off-gas 220 is vented to the atmosphere. The impurity depleted solution 210 then filtered in one or more solution polishing filters 230 to remove any solids (ferric hydroxide, residual nickel powder and other precipitate material if present) in solution which generates a filtered non-nickel metal impurity depleted discharge solution 250 in preparation for ion exchange which occurs in ion exchange module C. The polishing filters 230 are ideally configured to operate in a duty/standby arrangement, and can be equipped with a backwash cycle and a solution filter sludge tank (not shown) and transfer pump (not shown). The filtered non-nickel metal impurity depleted discharge solution 250 is then fed into an IX feed tank (not shown) to await IX. The sludge 260 from the polishing filters 230 can be transferred to a refinery final thickener for disposal.

Solution ion exchange module C

The filtered non-nickel metal impurity depleted discharge solution 250 from the IX feed tank (not shown) is fed to one or more ion exchange (IX) columns 270 located in ion exchange module C to primarily remove cobalt and any remaining trace impurities from the filtered impurity depleted solution 250. The IX columns 270 can be arranged in a lead/lag configuration, with the filtered non-nickel metal impurity depleted discharge solution 250 entering an IX column 270 with partially loaded resin first (lead) before proceeding to an IX

WO 2019/090389

PCT/AU2018/051203 column 270 with fresh resin (lag). A third column (not shown) may be provided off-line for regeneration which Includes washing, stripping and pre-loading. The fourth column (not shown) may be installed as a spare. The columns will be loaded with suitable resin.

Cobalt and other impurities (primarily iron and copper) are loaded onto the resin from the filtered non-nickel metal impurity depleted discharge solution 250, with a resultant first IX discharge solution which is a clean nickel sulfate solution 310 substantially free of non-nickel metal impurities discharging to the ion exchange discharge tank (not shown). The clean nickel sulfate solution 310 is then fed to one or more ion exchange discharge polishing filters 400 to remove any carry-over resin or solids from the IX columns 270 which can be removed from the filters as sludge 440. The filtered clean nickel sulfate solution 420 is then held in a crystaliiser plant feed tank (not shown) before being pumped to the crystaliiser plant E. A small portion of the filtered clean nickel sulfate solution 420, diverted filtered clean nickel sulfate solution 430 is transferred to a nickel hydroxide preparation module or to an ion exchange nickel preload tank (not shown) to be held until IX column 270 regeneration is required.

The lead column 270 is subjected to a regeneration cycle once the cobalt loading is achieved and the lead column first IX discharge solution cobalt concentration begins to increase. An exemplary regeneration cycle consists of the following steps as shown in Table 2:

Table 2 Step by Step _________Activity_____________________ ________Lag Stage_________________ ________Lead Stage________________ ________Regeneration Stage:________ _______Wash time_______________ ________Back-wash time____________ _________Elution time__________________ _________Elution rinse time_____________ _________Pre-load time________________ Technical time

During resin regeneration, and initially resin washing, several bed volumes of slightly acidic water (pH 4-5) wash solution (not shown) are used to wash the resin to flush out any entrained nickel solution. The wash solution is prepared by mixing sulfuric acid with demineralized water 280. if the resin is exposed to neutral or alkaline water, the active resin component may be dissolved leading to a loss of ion exchange capacity. The slightly acidic water (pH 4-5) wash solution avoids this problem, particularly for TP 272. The IX waste 330 from the washing and other regeneration reagents may be collected in an IX waste tank (not shown) and pumped to the leach acid solution preparation tanks for re-use. During a stripping phase, reagent stream 290 provide an approximately 1.5 -- 2,0 M sulfuric acid solution which

WO 2019/090389

PCT/AU2018/051203 is prepared in a mixing tank (not shown) and pumped for several bed volumes through the IX column 270, stripping off cobalt and other impurities into the stripping waste solution 330. The stripping process loads protons (H+ ions) onto the resin. The strip IX waste 330 is sent to the IX waste collection tank (not shown). An additional wash stage ot several bed volumes of demineralised water 280 is then used to wash excess sulfuric acid from the column and is also sent to the IX waste collection tank (not shown).

During a pre-load stage of regeneration, several bed volumes of diverted filtered clean nickel sulfate solution 430 are held in the pre-load solution recirculation tank 340 with the addition of nickel hydroxide 380 to buffer the pH of the diverted filtered clean nickel sulfate solution 430 to form a suitably buffered nickel preload solution 300. The nickel preload solution 300 is first pumped through a filter (not shown) to remove undissolved solids and then onto the IX column 270 and the effluent 320 from the pre-load phase is then pumped back into the pre-load recirculation tank 340. During nickel pre-loading, the resin in the IX column 270 exchanges protons (H+ ions) on the resin with nickel ions from the diverted filtered clean nickel sulfate solution 430, This exchange reaction results in increased acidity which is immediately neutralised by the buffered nickel preload solution 300. If this acidity is not neutralised, the pH of the recirculating solution drops and the pre-loading reaction stops. The nickel preloaded column 270 is not washed after pre-loading as it contains diverted filtered clean nickel sulfate solution 430.

The filtered clean nickel sulfate solution 420 then proceeds to a nickel sulfate hexahydrate crystallising, dewatering and drying module E. A crystalliser 450, ideally a draft tube baffle (DTB) type crystalliser is chosen over other crystalliser vessel configurations (e.g. Forced Circulation, Oslo) to ensure good product crystal size to allow easy dewatering and washing and to minimise fines during product handling. The crystalliser 450 operates at about 53°C and/or a vacuum of around 10 kPa to produce alpha form nickel sulfate hexahydrate crystals which are the industry standard. The crystalliser module E performs the following functions: (i) blending of filtered clean nickel sulfate solution 420 with centrate stream 510 and other minor streams in the fines recirculation pump (not shown), this includes a bleed pump (not shown) and pipeline (not shown) to control any impurity build up in the crystalliser 450, and (ii) fines destruction in the crystalliser heater (not shown).

A two-stage mechanical vapour recompression (MVR) type evaporative crystalliser operating at 53°C is preferably used to produce nickel sulphate hexahydrate. An MVR supplies energy to the crystalliser 450 via electricity over steam as th is minimises the opex for the unit, and reduces load on the plant’s steam boiler. A cooling water cooled surface condenser (not shown) is provided prior to a vacuum pump (not shown) which is suitable to maintain the preferred operating vacuum. A nickel sulphate hexahydrate crystal slurry 470

WO 2019/090389

PCT/AU2018/051203 generated in the crystaliiser 450 is pumped to dewatering cyclone (not shown), with cyclone underflow reporting to a centrifuge 480 for further dewatering. The centrate generated 510 is sent to the crystaliiser 450 for blending with filtered clean nickel sulfate solution 420.

Centrifuge solids (dewatered) 500 are transferred to a fluid-bed dryer 520 to produce a dry nickel sulfate product 530 which is ready for bagging. The fluid-bed dryer 520 includes an off-gas scrubber 570 for capture of any nickel sulfate dust 540 whereby scrubbed air/gas 560 is vented to the atmosphere. Collected dust 550 is returned to the crystaliiser 450 for further processing.

Nickel Hydroxide and Preload Solution Preparation

The nickel hydroxide slurry 380 is generated from the diverted filtered clean nickel sulfate solution 430 which is pumped from the crystaliiser feed tank(not shown) to one or more nickel hydroxide precipitation tanks 340. Sufficient sodium hydroxide 350 is added to allow the precipitation of nickel hydroxide to occur. The resultant nickel hydroxide slurry 380 from each precipitation tank 340 is then pumped to a nickel hydroxide filter (not shown) which captures solid nickel hydroxide. The filtered nickel hydroxide solids (not shown) are washed with demineralised water 360 to remove entrained sodium, with the captured nickel hydroxide solids (not shown) being discharged to a repulp tank (not shown) and the filtrate collected in a nickel hydroxide filter filtrate tank (not shown). The nickel hydroxide filter filtrate tank can be being returned (not shown) to a refinery.

The filtered nickel hydroxide solids (not shown) are repulped with diverted filtered clean nickel sulfate solution 430 and a diverted portion of the repulped nickel hydroxide slurry 390 is pumped to the aeration vessels/tanks 190 of module B for pH adjustment for metal hydroxide precipitation. A second diverted portion of the nickel hydroxide slurry 380 is pumped to the ion exchange preload recirculation tank (not shown) where it is mixed with diverted filtered clean nickel sulfate solution 430 and is then sent to the ion exchange column 270 for nickel pre-loading as the pH buffered pre-load solution 300. The diverted filtered clean nickel sulfate solution 430 added to the preload recirculation tank (not shown) completely dissolves the nickel hydroxide solids to produce the pH buffered pre-load solution 300. The pH buffered pre-load solution 300 is pumped via a preload solution filter (not shown) to remove any fine nickel hydroxide solids prior to being sent, to the ion exchange column 270. A portion of the diverted portion of the repulped nickel hydroxide slurry 390 is pumped to the aeration vessels 190 for pH adjustment and in order to prevent contaminant build-up.

Sodium hydroxide 350 may be supplied from a refinery directly to the nickel hydroxide preparation area D, where it is used to precipitate the nickel hydroxide in the preload recirculation tanks 340,

WO 2019/090389

PCT/AU2018/051203

Detailed description of the invention

The flowsheet to produce high purity nickel sulfate is shown in Figure 1. A corresponding bench scale test program was split into three parts (I) nickel powder leach in sulfuric acid (ii) neutralization and nickel hydroxide production, and (iii) ion exchange for impurity removal by IX before crystallisation.

An objective of the powder leach bench scale work was to determine the optimum leaching conditions of nickel powder in sulfuric acid as well as to determine the deportment of impurities in the leach. Conceptually several flowsheets were considered including anaerobic and aerobic leaching of nickel powder in sulfuric acid on a continuous or batch basis. The work was also to evaluate the chemistry and kinetics of nickel powder vat leaching using various anaerobic and aerobic leach conditions, for example sulfuric acid and oxidant (oxygen supplied as air) concentrations, solid liquid ratios, and temperatures. The following studies were considered in particular: impact of pulp density on leaching rate (aerobic conditions); impact of pH on leaching rate (aerobic conditions); impact of aeration rate on leaching rate; impact of temperature on leaching rate; batch acid addition (aerobic and anaerobic conditions); impact of pulp density under anaerobic conditions; impact of agitation rate under anaerobic conditions.

The dissolution of nickel powder takes place readily in sulfuric acid. Depending upon the presence and concentration of soluble oxygen, the products of nickel sulfate and water (Reaction 2) or hydrogen gas (Reaction 1) will be generated.

Ni_(s) + H₂SO_4(aq) -» Ni²⁺ _(aq) + H_2(g) + SO₄rf_aq> (1)

Ni_(s) + H₂SO_4(aq) + % O₂ - Ni^2<'(aq) + SO4²;aq) + H₂O (2)

In a system operating at elevated temperatures at atmospheric pressure, oxygen solubility is limited so that it is predicted that both reactions will take place simultaneously in the bulk solution. Temperature, oxygen availability and solubility, ionic strength, agitation and aeration regime impact the specific contribution of each reaction. Practically, in an efficient process with an excess of nickel powder, it is unlikely that oxidant demand will be met by dissolved oxygen due to the low total solubility of gaseous oxygen such that H₂ gas generation is expected. Indeed in all bench scale tests, gas evolved from condensers contained hydrogen gas, with small amounts also detected exiting from the central shaft through which impellors were inserted. A summary of test conditions is presented below in Table 3.

Table 3 Conditions of leach tests conducted. Variables assessed included temperature, pH, acid loading, pulp density and aeration

WO 2019/090389

PCT/AU2018/051203

Test No.	pH/Acid load (mL, g)	Nickel powder mass (g)	Aeratio n rate (L/min)	T (°C)
\|\|\|\|\|\|	1.0	600	iiiiiiii	80
2	1.0	800	5	80
llllll	1.0	1000	iiiiiiii	80
4	1.0	1200	5	80
5	1.0	700	5	80
6	1.0	1000	5	80
7	0	800	illlill	80
8	0	1000	5	80
llllll	liiiiiiiiii	1000	2(N_?gas)	Illi
10	2.0	1000	5	80
11	3.0	WOO	5	80
12	248 mL (456.3 g) H₂SO₄ made up to 1 L with tap water	woo	5	80
13	1.0	woo	W	80
14	1.0	woo	2 5	80
15	1.0	woo	1.0	80
16	90 mL (165.6 g) H₂SO₄ made up to 1 L. with tap water	woo	5	80
\|\|7\|\|\|	100 mL (134.0 g) H₂SO₄ made up to 1 L with tap ’water	woo	illlill	80
18	130 mL (239.2 g) H₂SO₄ made up to 1 L. with tap water	woo	5	80
llllll	150 mL (276.0 g) H₂SO₄ made up to 1 L with tap water	woo	illlill	80
20	1.0	woo	5	15
\|g\|\|\|\|	10	1000	lillllll	40
22	1.0	woo	5	60
\|g\|\|\|\|	1.0	1000	fiiiiii	100
24	90 mL (165.6 g) H₂SO₄ made up to 1 L with tap water	1000	5	80
25	1.0	1000	iiiiiii	80
26	1.0	woo	5	100
27	150 mL (276.0 g) H₂SO,_; made up to 1 L with tap water	1000	5	100
28	150 mL (276.0 g) plant H₂SO₄ made up to 1 L with plant water	woo	1	80
\|\|f\|\|	Ί80 mL (331.2 g) plant H₂SO₄ made up to 1 L with plant water	ιιΐΐιιιιι	iiiiiiiii	Illi
30	200 mL (368.0 g) plant H₂SO₄ made up to 1 L with plant water	1000	1	80
ιιιιιι	150 mL (276.0 g) plant H₂SO₄ made up to 1 L with plant water	1000	lillllll	80
32	128,8 mL (237,0 g) of plant H₂SO₄ made up to 1 L with plant water	woo	1	80
33	139.1 mL. (256.0 g) of plant H₂SO₄ made up to 1 L. with plant water	woo	Iiiiiiii	80
34	149.5 mL (275.0 g) of plant H₂SO₄ made up to 1 L with plant water	woo	1	80
35	159.2 mL. (293.0 g) of plant H₂SO₄ made up to 1 L. with plant water	1000		80

Materials and methods

WO 2019/090389

PCT/AU2018/051203

Nickel powder used during leach tests was provided by BHP Billiton Nickel West. Preparation of representative sub-samples was undertaken by combining the supplied material, blending three times, and sub-sampling using a stainless steel 6.35 mm riffle. Before subsampling, the riffle was washed with de-ionised water and dried at 80 ^!>C. Following this a 200g sample of nickel powder was removed and passed through the riffle and container boxes to scour them of potential residual contaminants. This nickel sample was not included in the bulk sample preparation and was discarded. A particle size distribution (PSD) of the supplied material is presented in Figure 2. The PSD was acquired using a Malvern Mastersizer 2000 laser particle size analyser with the nickel powder suspended in water. Example 1 - impact of pulp density on nickel powder dissolution Impact of pulp density under aerobic conditions

Test work was conducted in 2 L glass agitated vessels which were maintained at 80°C for 10 hours and pH controlled using sulfuric acid (98%) under aeration. The pulp densify was varied by adjusting the nickel powder to solution ratio. The impact of pulp density was evaluated in the range of 600-1200 g/L at pH 1 (Figure 3A and Figure 3B), and 8001000 g/L at pH 0 (Figure 3A). At either pH, pulp densities in excess of 800 g/L had minimal impact on the total quantity of nickel extracted over the 10 h experimental time period (Figure 3A). At an operational pH of 1,0, pulp densities of 600 and 700 g/L resulted in reduced total nickel powder dissolution, with final masses of 73.5 and 92.1 g dissolved, respectively. Experiments at 800, 1000 and 1200 g/L (pH 1.0) were comparable, with masses of 108.7, 106.7 and 122.0 g dissolved, respectively. While greater total quantities of nickel powder were dissolved at pH 0, (146.0 and 140.2 g at 800 and 1000 g/L), the results demonstrated no significant trend at this pH. However, in general, increasing the pulp density increased the rate of nickel extraction as well as the total metal extracted (Figures 3B and 3A, respectively). The reaction rate may be related to the total metal surface area available whereby if the pulp density increased then the total surface area available also increases the leaching rate.

Impact of pulp density under anaerobic conditions

The impact of pulp density under acid-loaded anaerobic conditions was assessed,. Tests were operated at 700, 800, 900, 1000 and 1300 g/L nickel powder, with an additional 1000 g/L test undertaken to confirm results. It is not possible to compare the results directly with aerobic conditions as the aerobic tests were conducted at a constant pH whereas the acid was added batch wise for the anaerobic tests.

Nickel leaching behaviour was generally similar in all tests (i.e. the majority of nickel was leached over the first 10 hours) (Figure 3C). The repeated 1000 g/L test demonstrated quicker rates of nickel powder dissolution (based on solution assay) (Figure 3F), with pH neutralisation behaviour matching this (Figure 3D). Tests at 800 and 1000 g/L demonstrated

WO 2019/090389

PCT/AU2018/051203 spurious results, with the nickel powder mass measured after leach tests indicating that only 118 and 151 g of powder was dissolved which is well under the theoretical quantity of nickel expected to be solubilised in the presence of 290 g of sulfuric acid. The particle size distribution of nickel used was assessed to determine if the varying results could be attributed to discrepancies in surface area. Particle size distribution demonstrated little variation, with all experiments having a decrease in particle size measured after leach tests were conducted (Figure 3E). Neutralisation of the solution throughout the experiment was broadly consistent (Figure 3D), It is unclear why the 1000 g/L repeat test had a faster extraction rate than the other tests.

Generally speaking, under anaerobic conditions, the results indicate that pulp density does not significantly affect: the nickel leaching behaviour under anaerobic, acid loaded (290 g) conditions (Figure 3F), the nickel extraction rate (Figure 3G) or the neutralization rate (Figure 3D) between a pulp density of 700 - 1300 g/L.

Example 2 - impact of pH on dissolution - effect of add loading Add loaded aerobic tests

Operational pH had an effect on the total quantity of nickel powder dissolved. Set point pH values of 0, 1,2 and 3 were evaluated at 80 °C, with increasing acidity demonstrating greater nickel powder dissolution (Figure 4A). Evaluation of the data with respect to free hydrogen ion concentration (calculated assuming ideality) shows a logarithmic response (Figure 4B). The data from Test 3, wherein pH drifted, is not directly comparable as the drifting pH confounds pH with other variables, but does agree with the general trend of increasing nickel dissolution with lower pH. A comparison of acid addition and nickel powder dissolution across all tests shows a very good correlation between acid and nickel dissolution (Figure 4C; R² of 0.94). This analysis also indicated 3 anomalous tests, namely 7, 8 and 12 (excluded from Figure 4C) indicative of a limiting factor during these tests. Interestingly, tests 7 and 8 were observed to have impeller corrosion, evidenced by elevated iron concentrations in solution (8.6 and 4.9 g/L). Test 12 was operated with an excess of sulfuric acid over a period of 31 hours. In this case, the reaction between sulfuric acid and nickel powder would have been impacted by the solution nickel concentration reaching saturation, precluding full reaction between the acid and the nickel powder by precipitation of nickel sulfate. For tests at a pulp density of 1000 g/L, it was observed that the leach pH has a strong influence on the rate of reaction. The highest rate of reaction was pH 0 achieving 174.9 g/L Ni in 10 hours (Figure 4D). However, as pH probes can have limited accuracy at a pH less than one, the pH zero test may not be directly comparable to the other tests. Furthermore, although the rate of reaction was faster at lower pH, the downstream neutralization demand would have been much higher and therefore, subsequent tests focused on batch acid addition rather than continuous.

WO 2019/090389

PCT/AU2018/051203 impact of aeration rate on nickel powder dissolution/leaching rate

In general, adjusting the aeration rate did not change the extraction rate significantly (Figure 5A and Figure 5C). Indeed, compressed airflow rate had minimal impact under the operational conditions tested. Tests operated at aeration rates of 1.0, 2.5, 5 and 10 L/min had 110.9, 112.5, 106.7 and 90.3 g of nickel dissolved, respectively (Figure 5A). Test 9, operated with 2 L/min of nitrogen gas sparging for comparison, retarded nickel dissolution kinetics, with 82.4 g dissolved over the experimental time period (Figure 5A). Actively sparging with nitrogen gas reduces the dissolved oxygen concentration. The presence of oxidants (other than soluble oxygen) is reported to facilitate the dissolution of nickel powder, consistent with this result. Dissolved oxygen is beneficial to the nickel dissolution reaction (see Equation 1), but not essential. 1.0 L/min of air is entirely adequate under these conditions. It can be seen that without air, the rate of reaction is much slower. It should be noted that despite air being added, hydrogen gas was detected in the off-gas from the reactor. This indicated that although air was being added, all of the dissolution of nickel does not occur via the oxidative mechanism (Equation 2) whereby some nickel dissolves via the anaerobic mechanism (Equation 1) that produces hydrogen.

The evaporation rate at 80 °C is positively correlated with the air flow rate (Figure 5B). High airflow rates contribute to reaching nickel saturation sooner by evaporative concentration, which hinders further nickel dissolution. This is believed to have contributed to the reduced nickel extraction at the highest airflow rate. Acid loaded anaerobic tests

A series of tests (32-35) were carried out in order to assess nickel leaching kinetics, impurity build up and neutralisation rate under acid loaded, anaerobic conditions. These (and subsequent) tests were conducted with Teflon coated impellers to ensure that no contamination (from potential deterioration of the Halar® coating) would impact test results. Anaerobic conditions were proposed to mimic large scale operations where hydrogen buildup during operation may be depressed using steam. Tests were conducted over a 30 hour period, with nickel, impurities, pH and free acidity measured over the duration of the experiment. Neutralisation in the presence of 237 g plant sulfuric acid was obtained after 12.5 hours, with a well-defined inflection point (Figure 6A). Other tests demonstrated a significantly increased period for neutralisation to occur, with tests containing 256 and 293 g/L acid reaching pH 5.0-5.5 between 29-30 hours, and the test with 275 g/L acid reaching 4.72 after 30 hours. Free acid titrations (Figure 6B) indicated that the majority of sulfuric acid present was consumed over the first 6-8 hours in tests at 237, 256 and 275 g/L. The test operated with 293 g of acid reached a stable free acid content at approximately 220-230 g/L, an erroneous result, given the nickel content measured (Figure 6C) would require the consumption of the bulk of acid present.

WO 2019/090389

PCT/AU2018/051203

Nickel concentrations typically reached a plateau after approximately 10 hours (Figure 6C). The test with 256 g/L acid added demonstrated a gentle decline over the period of 15-30 hours. It would be expected that this value would increase slightly over time due to evaporation. Of the elements analysed, only copper, cobalt, thorium and thallium were measured at concentrations above detection limits. Thorium and Thallium concentrations did not reflect changes in pH or nickel powder leaching change over the duration of the experiments and were present in low concentrations (5-20 pg/L). Cobalt, concentrations were strongly consistent with nickel leaching and ranged between 145-170 mg/L). However, copper appeared dependent over the first 10 hours before building up slightly over the remainder of the experiment (Figure 6E). Given the low copper concentrations present (0.21.5 mg/L), it is difficult to attribute a definitive trend (Figure 6E). Iron concentrations were notably below detection limits (<2 ppm). The results would indicate that under anaerobic leaching conditions iron and copper concentrations in the PLS are likely to be dependent on impurities from the local water supply and acid source. Demineralised water should be used in prepare the acid as a result. As expected, cobalt present was directly linked to nickel concentrations, with a PLS of 200 g/L nickel expected to have a cobalt concentration of approximately 175 mg/L based on the Ni/Co leach ratio.

Example 3 - impact of temperature on nickel powder dissolution

The effect of temperature on leaching rate was also studied. Tests were conducted in 2L agitated glass reactors in oil bath and maintained at pH 1 by sulfuric acid addition. A glass condenser with recirculating ice cooled water at 1-3°C was used to minimize the amount of evaporation. It was shown that at ambient temperature, the reaction rate was markedly slower than at elevated temperatures. Indeed, a general trend of increased nickel powder dissolution with respect to temperature was observed (Figure 7 A and Figure 7B). However, at pH 1.0, the impact of temperature between 60-100 °C was only loosely correlated, with the tests at 80 °C out-performing those operated at 100 °C. There was minimal difference in leaching rate between 80 to 100°C. ft should be noted that at elevated temperatures the evaporation rate also increased which may have biased the 100°C result. As previously discussed, evaporation contributing to a saturated nickel sulfate solution will impact the total quantity of nickel powder dissolved. The effect of bare stainless steel in the solution is not fully understood, but appears to influence the nickel dissolution kinetics, ft. has been proposed that at pH >0, the presence of ferric sulfate in solution inhibited the dissolution of nickel from nickel metal powder, which has tentatively ascribed to the formation of a passivating layer of NIFe₂O₄. Iron corrosion may be a source of this.

Example 4 - batch acid loading: impact on nickel dissolution and solution neutralisation

WO 2019/090389

PCT/AU2018/051203

Batchwise processing was considered advantageous over continuous processing as neutralization and leaching could occur in the same tank. It was also thought that it would be difficult to pump any unreacted nickel powder and therefore having all of the nickel powder in a single tank would be advantageous. Up-front bench scale batch addition of acid to tests resulted in improved nickel dissolution compared to those operated at set pH points. Tests were initially conducted aerobically with 1 L/min air sparge. The aeration rate was reduced from previous tests to minimise the amount of evaporation. The high initial acid concentrations (9-25 % v/v) provided excess acid to react with the nickel powder. Acid concentrations of 9, 10, 13, and 15 % (v/v) resulted in the dissolution of 96.1, 103.5, 126.0 and 146.7 g of nickel powder, respectively (Figure 8A). However, the presence of high concentrations of acid also increases the potential for generation of hydrogen gas, relative to other tests where the acid concentration is controlled by gradual addition. The rate of solution neutralisation was dependent on the starting acid content (Figure SB). In Test 24, with 9 % v/v of acid loaded, solution neutralisation (or complete reaction of acid and nickel powder) took place in 5 hours. This took 6 hours in Test 17 where 10 % v/v acid was present. Up to 13 % v/v acid was completely reacted within 10.5 hours. Increasing the temperature from 80 to 100 °C reduced the time required for solution neutralisation significantly, viz. complete reaction of 15 % v/v acid within 10.5 hours at 100 °C.

Example 5 - nickel dissolution and solution neutralisation under near optimal conditions

A set of three teste, viz. tests 28 (150 mis, 276 g acid), 29 (180 mis, 331.2 g acid) & 30 (200 mis, 368 g acid), were undertaken to assess maximum nickel dissolution kinetics and solution neutralisation within the defined experimental time period. Tests 28, 29 and 30 were performed with 150 mis (276 g acid), 180 mis (331.2 g acid) and 200 mis (368 g acid) up-front acid loading respectively, and aeration rates of 1 L/min, 1000 g/L pulp density and a temperature of 80 °C. Tests 28, 29 & 30 had corresponding nickel powder dissolutions of 168.3, 186.6 and 194.9 g, respectively. The rate of increase of nickel in solution was fastest at 150 mL, followed by 180 and 200 mL of acid (Figure 8C). The results may demonstrate the inhibitory impact of elevated ionic strength on the rate of reaction. Final nickel solution concentrations of 175.9, 199.4 and 197.6 g/L were measured in these tests. Evaporation rates were measured between 5.1-10.2 mL/h, indicating that final solution concentrations would be impacted minimally by solution evaporation. Neutralisation of 180 mL (331.2 g acid) and 200 mL (368 g acid) tests was not observed over a 25 hour time period (Figure 8F). As discussed previously, nickel concentrations in solution approaching saturation (approximately 200 g/L nickel sulfate based on experimental data) would result in dissolution rates decreasing and becoming dependent on precipitation of nickel sulfate in the system. The test operated with batch loading of 150 mL (276 g acid) sulfuric acid exceeded a pH of 5 within a

WO 2019/090389

PCT/AU2018/051203 period of 8 hours (Figure 8D). in general if was possible to achieve the target nickel strength within 10 hours (Figure 8B). The initial rate of reaction was accelerated compared to the test conducted at pH 0. In general the terminal nickel strength correlated well with the stoichiometric amount of acid added and the evaporation rate. Figure 8B shows the reaction pH verses time. The higher acid dosages (180 mL (331.2 g acid) and 200 mL (368 g acid)) reached pH 1 after 25 hours (Figure 8D). Tests were then conducted anaerobically (1 L/min nitrogen) with varying amounts of acid addition. In general the terminal nickel strength was a function of the initial acid added. Typically the majority of the nickel had dissolved within five to ten hours. However, the pH generally required up to 24-26 hours to achieve pH 3 (Figure 8G).

Impurity element concentrations under near optimal conditions

Selected impurities were measured in the three near optimal tests (Tests 28, 29 & 30). The impurity suite consisted of Ca, Al, Na, P, Si, K, Mg, Mn, Se, Cr, Co, Fe, Cu, Zn, As, Ru, Pb, Hr, Pd, Ag, Cd, Sb, Ir, Pt, Au, and Bi. Of these elements, only Cr, Co, Fe, K and Na exceeded 5 ppm in the final liquors (Figure 9A -- 9H). Ca and Na were leached almost instantaneously from the nickel powder, possibly originating from surface contaminants on the powder itself. Grand Co leaching followed the same general trend as Ni dissolution, suggesting that these impurities are co-leached from the nickel powder (Figure 9A & 9C). Fe leached gradually, but Fe present in solution dropped markedly when the pH exceeded 5 due to iron precipitation which depends on the redox potential (ratio of Fe(lI) and Fe(lll) in solution) as well as pH. Both valence states of iron will be present in solution, however, the presence of oxygen and elevated temperature will facilitate oxidation of Fe(ll) to Fe(lll). Fe(lll) typically starts to form precipitates at pH above 3 in sulfuric acid. Fe(ll) is stable up to approximately circum-neutral pH (pH 6-7). Copper and zinc values were low throughout the experiment (ppb). Concentrations decreased markedly in test 28 when the pH increased above 5, likely to be indicative of the formation and consequent precipitation of these metals as hydroxides (Figure 10).

Impurity element concentrations replica test

It was noted after evaluation of tests 28-30 that the impurity elements were in excess of the expected head value, particularly iron. To verify the impurity results, a final impurity test (test 31) was performed using a fresh batch of nickel powder under the optimum condition established above. The nickel powder was riffled using a virgin plastic riffle, the stirrer for the test was freshly coated with Halar®, and the pH probe was excluded from the test to minimise potential sources of impurity ingression. The test was performed with 150 mL (276 g acid), of ‘plant’ sulfuric acid. Feed and residue samples were assayed in duplicate, but showed no significant variation. The levels of impurities in this duplicate test were substantially lower than those of test 28. The maximum iron level in solution was 23.4 mg/L

WO 2019/090389

PCT/AU2018/051203 as compared to 50.0 mg/L, Cr reached 7.5 mg/L as compared to 17 mg/L, Zn reached 0.06 mg/L compared with 0.3 mg/L. Cobalt, however, rose to 85 mg/L, higher than the previous value of 71 mg/L. Co in the nickel powder was measured at 470 ppm. This suggests that some of the impurities in test 28-30 may have originated in contaminants, however much of the impurities appear to be intrinsic to the tests (particularly cobalt, as cobalt is present in the nickel powder at detectable levels).

Example 6 - impact of impelior agitation rate under acid loaded anaerobic conditions

The impact of impeller agitation rate was assessed (under anaerobic, acid loaded conditions (290 g/L). Impelior agitation rates were 50, 75, 100 and 150 % of the requirement to completely suspend 1000 g of nickel powder in 1 L of sulfuric acid solution. Nickel leaching and pH behaviour was consistent across tests, with minor variations measured in nickel leaching (Figure 11 A) and pH (Figure 11B) behaviour. The test at 75 % agitation demonstrated slightly quicker nickel leaching and pH neutralisation behaviour, and the test at 50 %. Tests indicated that impelior agitation speed within the range tested had minimal impact on nickel powder dissolution.

Exampie 7 - impact of powder particle size on leach rate

The impact of particle size on leach rate has also been considered, and initial tests indicate that coarser particles leach slower. The following particle size ranges are under consideration: +212 pm, -212 + 125 pm, -125 + 75 pm, and -75 + 45 pm.

Example 8 Lock cycle tests on nickel powder

Lock cycle tests are being considered to determine: if the leaching rate remains constant across multiple cycles with the same starting sample of nickel powder; if the average particle size across each cycle changes; and if impurities accumulate over time due to a leaching-precipitation mechanism.

Leaching results summary

168 g of nickel powder was dissolved and leach solution neutralised in 7 hours when operated with 150 mL (276 g acid) of plant acid at 8Q°C, pulp density of 1000 g/L and an aeration rate of 1 L/min compressed air. The nickel concentration in solution reached 175.9 g/L (indicative of evaporation). Chromium, cobalt and iron impurities increased following the same general trend as nickel. Iron, copper and zinc impurities in solution decreased (precipitated) as the solution pH increased above 5.

Nickel dissolution increased to 186.6 and 194.9 g, respectively in the presence of 180 mL (331.2 g acid) and 200 mL (368 g acid) of plant acid. However, in these tests complete neutralisation was not achieved. In experiments operated for 25 hours, solutions became saturated with nickel sulfate, inhibiting further neutralisation.

WO 2019/090389

PCT/AU2018/051203

Evaporation rates increased with respect to aeration rate and temperature. At a temperature of 80 °C, aeration rates of 1, 2.5, 5 and 10 L/min compressed air had evaporation rates of 31.1, 25.8, 14.3 and 10.9 mL/hr respectively. Increases in temperature also impacted evaporation rates, with evaporation rates reaching up to 51 mL/min at 100 °C.

Pulp density between 800-1200 g/L did not impact nickel dissolution kinetics. At pulp densities of 600 and 700 g/L, nickel powder dissolution was decreased relative to comparative tests.

Aeration rates of compressed air between 1-5 L/min had negligible impact on total nickel powder dissolution. At 10 L/min only 98.9 g of nickel was powder was leached compared to 112.7-117.8 g in comparable tests. The higher evaporation rate at 10 L/min may have reduced dissolution rates as the solution approached nickel saturation. Addition of nitrogen gas (2 L/min) to remove oxygen impacted nickel dissolution, with only 86.3 g of nickel powder leached over 10 hours.

The impact of temperature at pH 1.0 was assessed. Nickel dissolution and temperatures between 60-100 °C were loosely correlated, indicating that a limiting constraint was present, or evaporation effects retarded powder dissolution in high temperature tests. Bench scale tests - ion exchange (IX) purification of pregnant leach solution (PLS)

Ion exchange test work was conducted to assess the ability of a commercial resin, such as TP 272, to selectively remove iron, copper and cobalt form leach PLS. Cobalt and copper can be removed selectively from acidic nickel solutions using a suitable selective metal extractant, for example, bis(2,4,4-trimethylpentyl)phosphinic acid (Cyanex272). The commercial ion exchange resin TP 272 is a macroporous solvent impregnated resin which contains Cyanex 272. Solution pH is a crucial parameter for selectivity and efficiency of ion exchange. A challenging issue associated with use of conventional base reagents NaOH or ammonia solutions to control pH is the contamination of the nickel sulfate product. As an alternative to conventional bases for pH control, the process herein relies on (i) neutralisation of the pregnant leach solution with nickel hydroxide thereby avoiding contamination of the nickel sulfate solution, and (ii) pre-loading of the ion exchange resin with nickel sulfate solution, for example, from the purification circuit together with nickel hydroxide for pH adjustment to ensure ion exchange selectivity and efficiency, if needed.

Nickel pre-loading was proposed to be a critical step to provide a mechanism for obtaining stable pH to ensure efficient loading of Co impurity In this investigation. The objectives were (I) to establish the pH profile and range for Ni pre-loading, (II) to test the effect of nickel concentration of feed PLS on loading efficiency, and (iii) to determine type of reagents for neutralisation and ranges of pH for loading. In particular nickel hydroxide prepared was tested as a desirable neutralisation base reagent.

WO 2019/090389

PCT/AU2018/051203

The scheme of nickel pre-loading is shown in Figure 12A consists of: (i) pre-loading nickel onto resin, using either a Co/Ni effluent bleed or a purified Ni effluent bleed, with solution adjustment by nickel hydroxide Ni(OH)₂, and if needed, ammonia or NaOH, followed by a resin wash to remove impurities, and (ii) displacement of the loaded nickel by cobalt impurity, in a cobalt loading step, followed by elution to regenerate the resin for recycle and re-pre-loading. Use of closed circuit allows for isolated use of ammonia or NaOH for neutralisation and pH adjustment without product contamination.

Optimisation experiments were carried out re the nickel pre-loading, impurity loading and stripping conditions required. The ion exchange (IX) stages are operated at 80 °C with a nickel liquor which is close to saturation to reduce the capital cost of a preferred downstream crystallisation step. Tests include: pre-loading and replacement tests, shake out tests, resin stability tests at 80 °C , isotherm tests, and mini-column tests. The chemical stability of the resin at elevated temperature is also studied. Finally, the suitability of various ion exchange resins to remove Fe, Cu and Co (as well as any other impurities) from a neutralised (pH 5 6) nickel sulfate solution such as a pregnant leach solution, produced from the dissolution of nickel powder in sulfuric acid is investigated.

Experimental

All solution pH was monitored using an lonode electrode (model IJ44C HT) coupled with a TPS portable pH meter (model AquapHZ) with manually adjusted temperature otherwise mentioned. ROSS Sure Flow electrode (model 8172BN) and Cole-Parmer (model 05993-28) electrodes were used as alternative for some tests. IKA WERKE RCT basic hot plate/ magnetic stirrer coupled with IKA WERKE ETS-D4 thermometer and New Brunswick Innova 40 incubator shaker were used for mixing the solid-liquid mixtures.

Ni pre-loading experiments

Nickel pre-loading tests were conducted using synthetic PLS prepared by using AR grade nickel sulfate. The pre-loading pH isotherm tests were carried out at solid to liquid (S/L) ratio of 10 g / 40 mL (1:4) in a 100 mL hexagonal glass jar immersed in a temperature controlled oil bath using hot plate with magnetic stirrer. The solution temperature was maintained at 80 ± 1 °C during the tests. The aqueous solution pH was adjusted with ammonia or sulfuric acid solutions. The system was equilibrated at each pH point for 5 minutes and the mixture was sampled at 0.5 pH intervals over a pH range of 3.0 -- 7.5. The pH readings were recorded at each sampling point and used for the construction of pH isotherm graphs. The slurry mixture was quickly separated using vacuum filtration with 0.45 pm sieve to avoid crystallisation due to lower temperature. The aqueous solution was diluted 4 or 5 times for assay and the resin was stripped with 2 M sulfuric acid solution for assay. Cobalt loading pH isotherms experiments

WO 2019/090389

PCT/AU2018/051203

Cobait impurity loading distribution isotherms were prepared using air-dried Ni preloaded resin at initial solution pH 4.6 and various S/L ratios in a 100 mL hexagonal glass jar shaking in an incubator shaker. The solution temperature was maintained at 80 ± 1 °C during the tests. Each pH point test was separately performed with the PLS of different initial pH, which was adjusted with nickel hydroxide solid and sulfuric acid solution up to pH about 5. The mixtures were mixed by shaking for 1 - 2 hours until the final equilibrium pH. The final pH was recorded and adjusted if needed. The slurry mixture was quickly separated using vacuum filtration with 0,45 pm sieve. The aqueous solution was diluted 4 or 5 times for assay and the resin was stripped with sulfuric acid solution for assay. Maximum loading experiments

The maximum loading of cobalt on the TP 272 resin was investigated. A Langmuir isotherm model was used to fit the data and the maximum loading of the resin was calculated at about 11.5 mg Co/g wsr resin. The equivalent loading is 6 g Co/L wet saturated resin. The loading is lower than the maximum rated capacity of the resin (11 g/L Co) due to the high concentration of nickel and low concentration of cobalt in solution.

Extraction kinetics experiments

The extraction kinetics tests were carried out with air-dried Ni pre-loaded resin at initial aqueous pH 5.0 and S/L ratio of 1 g / 40 mL in a 100 mL hexagonal glass jar immersed in a temperature controlled oil bath. The solution temperature was maintained at 80 ± 1 °C during the tests. The extraction kinetics tests were performed at 0.5, 1,2, 5, 10 and 30 minutes. Timer was started when the resin was added to the solution with stirring. The mixing was stopped at the required time, final pH was recorded and the slurry was quickly separated. The aqueous solution was diluted 4 or 5 times for assay and the resin was stripped with sulfuric acid solution for assay.

Stripping pH isotherms experiments

Tests were conducted using different strength sulfuric acid solutions to strip the loaded metals. Stripping acidity isotherm tests were conducted with the Co loaded resin samples which were mixed with different concentrations of sulfuric acid solution at a S/L ratio of 1 g /10 mL and 80 ± 1 °C for 0.5 hour. The tests were carried out in a 100 mL hexagonal jar in an incubator shaker. The slurry mixture was quickly filtered using vacuum filtration with 0.45 pm sieve. The aqueous solution was sent for assay and the resin was stripped with sulfuric acid solution for assay.

Stripping kinetics experiments

The stripping kinetics testa were carried out with air-dried cobalt loaded resin and 1.5 M H₂SO₄ at a S/L ratio of 2 g /10 mL in a 100 mL hexagonal glass jar immersed in a temperature controlled oil bath. The solution temperature was maintained at 80 ± 1 °C during the tests. The stripping kinetics tests were performed at 0.5, 1,2,5,10, 30 and 60 minutes.

WO 2019/090389

PCT/AU2018/051203

Timer was started when the resin was added to the stirred solution. The mixing was stopped at a required time and the slurry was quickly separated. The aqueous solution was sent for assay and the resin was stripped with sulfuric acid solution for assay.

The concentration of iron in the strip solution was lower than the ICP-OES detection limit and therefore was not reported. It was shown that cobalt could not be selectively stripped from the resin as the nickel stripped at a lower pH (1 M, 95% Ni removal vs 1.5 M 95% Co removal). The full scale plant design will use an acid strip solution concentration of about 1.5 M - about 2 M sulfuric acid.

Resin stability experiments

The resin Ni preloading, Co loading and elution were conducted with a total of 20 cycles over 4 weeks. Aqueous and resin samples was taken from each cycle to determine metal loadings and resin exchange capacity overtime. The tests were conducted in a 500 mL Schott bottle in an incubator shaker at 80 °C. One cycle of the test consisted of nickel preloading, loaded resin washing, cobalt loading, and cobalt loaded resin washing, loaded resin elution, and eluted resin washing.

The nickel preloading test was performed with the synthetic solution containing 190 g/L Ni at a S/L ratio of 50 g / 200 mL with constant shaking for 1 hour. Nickel hydroxide powder (4 - 5 g) was also added to the solution to obtain solution pH 5 - 5.5. The resin was then filtered and washed three times with deionised water (200 mL) for 10 minutes with shaking.

The cobalt loading test was conducted with synthetic solution (200 mL) containing target 190 g/L Ni, 185 mg/L Co, 11 mg/L Fe and 4 mg/L Cu. The mixture was mixed by shaking for 20 hours. At the end, the solution pH was recorded and resin was filtered. The cobalt loaded resin was then washed three times with deionised water (200 mL) for 10 minutes with shaking.

The cobalt elution test was conducted with 2 M sulfuric acid solution (200 mL). The mixture was mixed with shaking for 1 hour. The stripped resin was washed three times with deionised water (200 mL) for 10 minutes shaking times each time.

Solution samples were taken at each step of nickel preloading, cobalt loading and resin elution. Solid sample (~ 1 g) was taken at the end of each cycle.

The exchange capacity measurement was conducted with sodium hydroxide solution and titrated with sulfuric acid solution. Air dried eluted resin from each cycle (0.3 - 0.5 g) was mixed with 0.196 M or 0.0196 M sodium hydroxide solution (50 mL) at 50 °C in an incubator shaker for 12 hours (overnight). The supernatant (5 or 10 mL) was titrated with standard sulfuric acid solution to phenolphthalein end point. The amount of sulfuric acid usage was recorded for calculation of the total exchange capacity (Figure 18).

WO 2019/090389

PCT/AU2018/051203

Column design and set up

Ion exchange test work was conducted to assess the ability of the commercial resin TP 272 to selectively remove iron, copper and cobalt from leach PLS. The ion exchange columns were constructed using wafer jacket condensers. Hot wafer from wafer baths was circulated through water jacket. The column size in diameter varies slightly from column to column and top to bottom. The measurement data for Column 1 are given in Table 6, indicating that the size is in the range of 140 - 150 mL and 145 mL on average. The column was packed with Lewatit TP 272 resin (~145 mL / 64.5 g). The leach feed solution was first heated up and pumped from the bottom of the condenser with the flow rate of 2 or 5 bed volume (BV). The outlet solution was collected in a 2 L plastic bottle. The solution pH was recorded and sample was taken at each BV.

Table 6 Dimension of the IX columns and parameters

Column	Units	Column 1	Measurement
Diameter	mm	End 1	21.2
	mm	End 2	21.8
	mm	Average	21.5
Column height	mm		400
Column volume (average)	mL		145

Resin system			Lewatit TP 272
Resin mass (air dried) packed	G		64.9
Initial Resin bed height	mm		-400
Feed flow mode (direction)			Down - Up
Feed rate	L/ hour		-0.3 L/hour (-2 BV/hour)

Initial test work was conducted on the effect of utilizing the ion exchange resin TP 272 in fresh or hydrogen ion form without pre-loading. It was found that without pre-loading the resin, the pH of the pregnant leach solution significantly dropped due to the release of hydrogen ions as the resin loads cobalt and nickel (Figure 12B).

Example 9 - nickel hydroxide preparation

The nickel hydroxide preparation was conducted in a 3 L beaker on a hot plate with magnetic stirrer. Nickel sulfate hexahydrate (1 kg) was dissolved in 2.0 L deionised water. Temperature was maintained at 40 °C initially. Concentrated ammonia solution (25% v/v) was used to adjust the pH to around pH 8. The solution colour turned blue and some nickel hydroxide powder was observed. The temperature was increased to 80 °C with constant stirring overnight. The nickel hydroxide powder was filtered using vacuum filtration. The blue colour filtrate was further mixed with stirring at 80 °C to produce more nickel hydroxide powder. The filtered nickel hydroxide powder was washed three times with deionised water to about pH 6.5 and left air dried before use.

WO 2019/090389

PCT/AU2018/051203

Table 4 Conditions for preparation of nickel hydroxide.

Feed material	Nickel sulphate (LR)
Neutralisation reagent	Ammonia solution (AR)
Temperature	50 °C

Preparation of nickel hydroxide buffer

Measurement of nickel hydroxide: 0.501 g oven dried at 105 °C overnight and dissolved in 100 mL of dilute sulfuric acid solution for assay of major elements by ICP-AES. The results are given in Table 5.

Table 5 Assay results of typical Ni hydroxide.

Batch ID	Concentration [mg/L]	% (w/w) in Ni(OH)₂ product
	Ni	Fe	Co \| Cu	N*	Ni	N	Others
1	2919	<0.2	<0.2 <0.2	<1	92.0	<0.02	-8
2	2942	<0.2	0.235 i <0.2	<1	92.7	<0.02	-7 3

*Toial nitrogen was below detection limit.

Assay results showed that the total nitrogen (N) content was below the detection limit in the dissolution liquor. This suggests that the contamination of ammonium ion was minimal. Some black stains were observed on the surface of the air dried nickel hydroxide, which could be result from formation of minor nickel oxide. The Ni content was calculated to about 92% with about 8% other components such as water and nickel sulfate.

Example 10 - nickel pre-loading

The optimum pH to maximise nickel loading onto TP 272 resin was investigated using batch tests.

It was found that increasing the pH of the solution increased the loading of nickel metal on the resin (Figure 12C). Partial precipitation of the PLS was observed at a pH greater than 5.5 (Figure 12D). The optimum pre-loading pH was determined to be about pH 4.5 to about 5 as minimal nickel precipitation occurred while maximizing nickel loading. It was theorised that at low pH, the active resin component (Cyanex 272) has a low affinity for cobalt. At higher pH, the large excess of nickel competes with cobalt which reduces the overall loading. For comparison, test work was conducted using Cyanex dissolved in Shellsol D70 (Figure 26A to 26C) which showed a similar behavior in terms of Ni:Co selectivity verses pH.

A: Neutralization with nickel hydroxide

Wide ranges of pH and feed nickel concentrations were investigated to determine their optimum ranges and types of base reagents for application including nickel hydroxide and ammonia and/or sodium hydroxide. The optimum pH to maximise nickel loading onto TP 272 resin was investigated using batch tests. From the neutralisation test in Ni pre-loading investigations, it is known that, nickel hydroxide can raise the pH of the synthetic PLS to about pH 5 or some above pH 5 with reasonable neutralisation kinetics and efficiency. The

WO 2019/090389

PCT/AU2018/051203 pH of real PLS produced through post leach extended residence time with nickel powder can falls in the pH range of 5 - 5.5 with sufficient leach residence time. B: Nickel pre-loading pH isotherms

Nickel pre-loading pH isotherms were carried out with Ni sulfate solutions of different Ni concentrations (20, 48, 100, 145 g/L Ni) were conducted in a continuous pH adjustment and sampling of resin for assay of Ni loading at each equilibrium pH, The procedures are detailed in experimental section.

The test with 20 g/L feed Ni was performed separately for each pH to enable observation of behaviour of Ni hydroxide precipitation and effect on Ni loading as a function of solution pH. in this test, both aqueous and resin strip liquors were assayed for Ni concentration. Atypical loading curve and variation of raffinate Ni at each equilibrium pH are shown in Figure 12C. The Ni loading increased almost linearly with pH up to pH 6 with the aqueous Ni concentration remaining at about 20 g/L (Figure 12C). A significant drop of Ni loading at pH 6.5, which could be attributed to a decrease of more than 70% Ni by precipitation (Figure 12C). Ni loading further increased from pH 6.5, which would be the result of the net effect of increased pH and variations of Ni concentration in the solution. Ni loading curves with other feed Ni concentrations are compared in Figure 12D. The loading of Ni generally increased with solution pH. A significant increase of Ni loading was observed with a higher Ni feed concentration (145 g/L Ni) along the pH range tested. As shown in Figure 12C, the precipitation of Ni may occur at a higher pH, which would also depend on the feed Ni concentration.

C: Ni pre-loading kinetics

The nickel preioading kinetics tests were carried out with fresh resin at initial solution pH of 5.0 or 6.0 and S/L ratio of 17.5 g / 70 mL in a 100 mL hexagonal glass jar immersed in temperature controlled oil bath. The solution temperature was maintained af 80 ± 1° C during the tests. The Ni preloading kinetics tests were performed at 0.5, 1,2,5,10 and 30 minutes. Timer was started when the resin was added to the stirred aqueous solution. The samples were taken at the required time and the slurry was quickly separated. The aqueous solution was diluted 4 or 5 times for assay and the resin was stripped with sulfuric acid solution for assay. Nickel loading was very fast (Figure 12E), approaching equilibrium within one minute using both pH 5 and 6 feed nickel sulfate solution. The loadings then varied slightly overtime which could be caused by variation of pH and other conditions. The final Ni loading at 30 minutes with pH 5 feed was slightly lower than that with pH 6 feed, though the pH 5 feed Ni concentration was about doubled that of the pH 6 feed. This was consistent to the observations for the effect of pH and Ni concentration on Ni pre-loading as discussed above. D: Summary of Ni pre-loading findings

WO 2019/090389

PCT/AU2018/051203

4/

The optimum pre-loading pH was determined to be about pH 4.5 to about 5 as minimal nickel precipitation occurred while maximizing nickel loading.

Discussion

Based on the results, the following conditions were suggested for Ni pre-loading for subsequent cobalt loading and stability tests: using synthetic Ni sulfate solution of 190 g/L Ni for Ni pre-loading which is required to stabilise the increase in discharge solution pH. In the scaled up application, the bleed of the purified nickel sulfate stream within the circuit can be employed for solution making which can be used in multi cycles under optimum pH in the range of Ni concentrations up to more than times dilutions, and using only nickel hydroxide, Ni(OH)₂, for pH adjustment to highest achievable pH in the range of pH 5 - 6 for reasonable nickel loading efficiency.

Example 11 nickel hydroxide neutralisation/buffering

Nickel hydroxide (Ni(OH)₂) can be used for pH adjustment to about pH 4.5 - 6, particularly pH 5, with reasonable loading capacity, indicating Ni(OH)₂ can be used as neutralisation reagent in Ni pre-loading.

In the cobalt IX tests, the resin was bulk pre-loaded using pH 5 feed nickel sulfate with nickel hydroxide used as the neutralisation reagent to take the feed solution to pH 5, if necessary.

Example 12 - experiments using fresh TP 272 and synthetic PLS

A: Loading pH isotherms using fresh TP 272 without Ni pre-loading

Co loading pH isotherms using fresh TP 272 without Ni pre-loading were performed as reference to demonstrate whether Ni pre-loading is a necessary step. In order to eliminate any possible Ni(OH)₂ carrying over to affect the variation of solution pH, the feed solution was adjusted with diluted NaOH solution. As expected, the initial feed pH values decreased significantly after each equilibrium loading (Figure 13A). This resulted from the loading of one mole of a divalent metal cation to exchange equivalent 2 moles of hydrogen ion into the solution based on the reaction:

M²⁺ + 2HA = MA₂ + 2H⁺ (1) where M is metals, HA is an acidic extractant, i.e. the active phosphinic acid functional group, for example, in TP 272. The metal loading isotherms against the final equilibrium pH values were plotted in Figure 13B for percent loading efficiency and Figure 13C for metal loading per gram of resin.

As the initial pH 5 decreased to equilibrium pH 3,5, the efficiency of Co loading was below 35% (Figure 13B) with comparable co-loading of Ni (Figure 13C), An increase in initial pH to pH 5.4 resulted in final equilibrium pH 4.3 with an increased loading of Co to about 70%. However, the co-loading of Ni at increased pH was almost doubled (Figure 13C). ft is

WO 2019/090389

PCT/AU2018/051203 known that the affinity of Co complexation to the organic phosphlnic acid extractant, e.g. Cyanex272, is much higher than for Ni, and thus good selectivity of Co over Ni can be achieved in terms of optimum pH range. The pH isotherms of Co and Ni essentially agreed well to the vendor published values where the aqueous feed solution contained 1.6 g/L Co and 77 g/L Ni, compared to the present PLS tested containing 190 g/L Ni and 0.18 g/L Co. The results suggest that pre-loading of Ni is necessary to minimise change of pH during the loading for acceptable loading efficiency.

B: metal loading pH isotherms using nickel preloaded TP272

Bulk resin samples of 17.5 g each was pre-loaded at 80 °C with two synthetic nickel sulfate solutions neutralised to pH using nickel hydroxide and pH 6 using ammonium hydroxide respectively (Table 7).

Table 7 Results of bulk Ni pre-loading at 80 °C and two pH.

Sample ID Ni PLS at pH 5 Ni Rai at pH 5 Ni PLS at pH 6	Neutralisation Ni(OH)₂	I Ni (g/L) I 240.2 i 189.4 192.9	Resin sample 1 St (g) 1 0.92 i ----------------------------------------------------- 1---------------------	ip liquor Ni (g/L) 3.847	Ni loading (mg/g) 17.73
Ni Raf at pH 6	NH₄OH	I 122 1		2.601	20 48

Loading with the resin pre-loaded at pH with nickel hydroxide

The metal loading pH isotherms with Ni preloaded resins are shown in Figure 13D for %loading and Figure 13E formass loading (mg/g), respectively. The variations of raffinate metals are shown in Figure 13F. Co loading curve shows an inverted V shaped trend with a maxima around pH 4.5 over the range of final pH 4.2 - 5.1 (Figure 13D and Figure 13E). This corresponded to a ‘V’ shaped change of the raffinate Co (Figure 13F). Ni re-equilibrium loading curve mirrors Co loading curve (Figure 13D). The effect of competing loading by Ni against Co became significant at a higher pH. Total metal loading increased at each equilibrium pH and mainly followed the trend of Ni loading. Fe and Cu loading varied over the pH range with raffinate Fe and Cu were about 1 mg/L in most of samples. As the concentrations of Fe and Cu in the feed were very low, likely involving precipitations at higher pH, large errors could be introduced by dilution and assay by ICP-AES. The loading of Ni and Co might also be affected by the variations of their concentrations in the feed. This mainly resulted from the neutralisation with nickel hydroxide and filtration of unreacted fines, during which some loss of metals by possible crystallisation of nickel sulfate at decreased temperature, though an effort was made to shorten filtration time and drop in temperature. The selectivity of Co over Ni, expressed by separation factor SF(Co/Ni), increased from 236 at pH 4.2 to 880 at pH 4.4, and then decreased to 120 at pH 5.1. This is generally related to the competing loading at higher pH range. The pH of maximum selectivity at about 4.5 agreed with that observed for the solvent extraction with Cyanex272.

WO 2019/090389

PCT/AU2018/051203

Loading with the resin pre-loaded at pH 6 with ammonium hydroxide

The behaviour of Co loading using the resin pre-loaded at pH 6 was, in principle, similar to those using the resin pre-loaded at pH 5, though the inverted ‘V’ shaped loading curve shifted slightly to right pH range (Figure 13HI and Figure 131). The trend of Ni reequilibrium loading was different, compared to that using the resin pre-loaded at pH 5. With initial pH 4 feed PLS, about 70% of the pre-loaded Ni was stripped into the solution under reequilibrium pH 4.5, This could, at least partially, attribute to an increased re-equilibrium pH caused by displacement of equivalent amount of hydrogen ion from the solution at Ni^:'‘:2H‘ ratio. At higher initial pH 4.5 and 5, Ni loading increased to 27 mg/g and 30.4 mg/g, respectively, while final equilibrium pH increased considerably as well. This suggested that some other factors might be involved in the mechanism. The selectivity of Co over Ni, SF(Co/Ni), decreased consistently over the range of equilibrium pH 4.5 - 5.1. A significant decrease was observed from pH 4.7 to 5.1, suggesting competing loading become significant. In addition, the effect of competing loading between Co and Ni was evidenced by a decreased Co loading which corresponded to an increased Ni loading, with the total metal loadings were constant at about 32 mg/g (-0.55 mM/g). Contribution of Fe and Cu loading was considered to be negligible in this case.

Discussion of mechanism of stabilising pH by the pre-loaded resin

With the resin of Ni pre-loaded at pH 5 with nickel hydroxide, the variation of equilibrium pH versus the initial pH was small with an increased trend in the final pH (Table 8) proving the proposed concept of using Ni pre-loaded resin to stabilise pH and to ensure Co loading.

Table 8 Final equilibrium pH vs. initial pH using the Ni preloaded resins and the synthetic PLS with pH pre-adjusted with nickel hydroxide.

PLS initial pH	Final equilibrium pH
	Ni pre-loaded resin at pH 5 with Ni(OH)₂	Ni pre-loaded resin at pH 6 with NH₄OH
4.0	4.22	4.53
4.5	4.45	4.72
5.0	5.08	5.11

Evidently the Ni pre-loaded resin provides mechanisms of displacement loadings, including displacement of other metal ions of higher affinity (e.g. Fe³⁺, Cu²⁺ and Co²⁺), by metal ion complexes (e.g. bisulphate), and by metal ions including Ni²’ species. In the case of nickel hydroxide used for pH adjustment in the pre-loading, it is also possible to load some nickel hydroxyl species (e.g. NIOH⁺) or carrying over some aqueous soluble Ni(OH)₂ species

WO 2019/090389

PCT/AU2018/051203 in resin micro pore structure, which may contribute to the net change of solution pH, after reequilibrium.

With the resin of Ni pre-loaded at pH 6 with ammonium hydroxide, the final equilibrium pH increased considerably from each initial pH. For the PLS of initial pH 4, an increase to equilibrium 4,53 could be partially attributed to re-equilibrium of nickel loading and other metal loadings from initial 20 mg/g (Ni) to about 7 mg/g (Ni + Co), which would result in equivalent amount of H⁺ at a ratio of M²⁺:2H⁺ loaded from the solution, and thus an increased pH.

However, with the PLS of initial pH 4.5 and 5, the total loadings of metals significantly increased, while the pH increased considerably as well. A possible mechanism may be associated with the loading of nickel amine species [Ni(NH₃)?⁺j (n= 1 - 6) formed during the pre-loading with ammonium solution for pH adjustment. The re-equilibrium with the synthetic PLS may result in re-loading to release NH₃ into solution to cause an increased pH through the equilibrium reaction:

NH₃ + H₂O = NH₄* + OH (3)

Summary and discussion of metal loading pH isotherm results

Loading with Ni pre-loaded resin stabilised solution pH with a tendency of a slight pH increase after re-equiiibrium. The results proved that Ni pre-loading effects stable pH and warrants efficient Co loading at a desired pH range. Ni loading in competing with Co became significant at a higher pH > 5, and there was an optimum pH for Co loading, which shifted slightly with different pre-loaded resin systems in the range of pH 4.5 - 5. Based on the above results, the following conditions for further tests were used: pre-loading TP 272 with nickel sulfate solution of 180 g/L at about pH 5 using the prepared nickel hydroxide for pH adjustment, and adjusting pH of synthetic PLS with the prepared nickel hydroxide to close to pH 5 (4.6 - 5).

B: loading kinetics

Co loading increased slowly, and took about 30 minutes to approach equilibrium (Figure 31B and Figure 31C), In comparison, additional Ni was loaded rapidly from the preloaded 18 mg/g to about 32 mg/g within one minute, and then stable at about 32 mg/g followed by a slow increase after 30 minutes. Solution pH increased from 4.6 to above 5 almost simultaneously as the additional Ni was rapidly loaded, and then became constant at 5.2 - 5.4. This again demonstrated that the pre-loaded resin functioned well for stabilising the solution pH by a mechanism which could be related to displace some pre-loaded Ni species involving basic components for neutralisation of further metal loading. The accountabilities of Ni and Co were reasonably good (Figure 31C), taking the experimental and assay errors and other factors such as high concentrations of Ni and matrix and multi steps involving Ni preloading and high temperature associated variations.

C: elution kinetics

WO 2019/090389

PCT/AU2018/051203

Metal elution kinetics studies were performed using 1.5 M H₂SO₄ solution at 80 °C. All the metal behaved similarly, approaching equilibrium at within 30 minutes (Figure 15) D: elution acidity isotherms

Elution acidity profile was shown in Figure 15, Stripping of >90% Ni and Co occurred at 1.0 M H₂SO₄ and 1.5 M H₂SO_4; respectively. Near complete stripping of Cu occurred at 2M H₂SO₄. As little Fe was loaded, the stripping of Fe could not assessed. From solvent extraction experience using Cyanex272 with the same function group as Lewatit TP 272, both Ni and Co are easy to strip with much lower acidity. The different behaviour of Ni and Co stripping in terms of high acidity requirement from the loaded TP 272 resin could be attributed to the requirement for high acidity, i.e. strong driving force, to minimise concentration gradient for mass transfer and diffusion of the reactant into the infernal micro structure sites and reaction products away from the sites.

E: loading distribution isotherms

Metal loading distribution isotherms were conducted at a fixed solution volume (50 mL) and varying mass of air dry resin to obtain a wide range of resin / solution (R/S) ratio (g/mL). The detailed procedures are given in Experimental section and conditions summarised in Table 9.

Table 9 Test condition for metal loading distribution isotherms

Resin	Ni-Preloaded Lewatit TP 272
Solution	50 mL synthetic PLS (initial pH 4.6) for each test
Temperature	80 °C
Loading mixing time	2 hours
Loaded resin sample stripping	10 mLof2M H₂SO₄ at 80 °C

Results based on air dry mass

Air dried resin was used for the distribution isotherm tests with original assay data (shaded) in Table 10 and Table 11), and mass balance data in Table 12. Metal loading curves are shown in Figure 17A for loading efficiency (%) and in Figure 17B for mass loading (mg/g resin). Raffinate metals and Co distribution curve are shown in Figure 17C and Figure 17D, respectively.

WO 2019/090389

PCT/AU2018/051203

Table 10 Feed and raffinate data for the distribution isotherm tests.

ID	Tciai resin	IR/illl
	..(g)___________________	(S/mL)	Ni	Co	Cu	Fe	Ni	Co	Cu	Fe
Feed	0.00	0.000	168.9	0.154	0.003S	0.0141
DI-1	0.10	0.002	170 9	0.147	0.0018	0.0004	-1.22	i 4.52	54 23
DI-2	0.25	0.005	169.6	0.119	0.0029	0.0109	-0.42	i 22.61	25.21	22.42
Di-3	0.50	0.010	163.6	0.072	0.0019	0.0105	3.14	: 53.41	50.94	25.61
DI-4	0.67	0.013	155.8	0 051	0.0014	0.0097	7.72	i 67.10	64.00	30.85
DI-5	1.00	0.020	172.4	0.035	0.0015	0.0122	-2.08	i 77.20	61.07	13.42
DI-6	1.25	0.025	162 7	0.029	0.0013	0.0108	3 63	i 81.04	67 28	23 03
DI-7	2.50	0.050	162 5	0.016	0.0009	0.0094	3.78	i 89.67	78.15	33 07
DI-8	3.33	0.067	166.9	0.012	0.0007	0.0083	1.16	i 92.47	32.83	37.70
DI-9	5.00	0.100	163.9	0.009	0.0007	0 0088	2.93	i 94.31	81.62	37.59
DI-10	6.25	0.125	169.9	0.007	0.0007	0.0110	-0.62	i S5.50	81.00	21.83
Di-11	8.33	0.167	169.1	0.006	0.0007	0.0085	-0.17	i 96.25	81 72	39.30
DI-12	12 50	0.250	162 4	0.004	0.0005	0.0076	3 82	i 97.69	86 99	46.03

Table 11 Metal loading on air dry basis (original assay data in blue)

	lOiiiii	IBgllli	samples		basis
	Jg)________	.MtLLh	...(g)................	Ni	Co	Cu	Fe	Ni	Co	Cu	Fe
Feed	0.00	0.000	1.025	1.817	0.000	0.0002	0.0002	17.73	0.00	0.00	0.00
DI-1	0.10	0.002	0.081	0.216	0.059	0.0009	0.0012	26.69	7.24	0.11	0.15
DI-2	0.25	0.005	0.265	0.879	0.135	0.0028	0.0015	33.16	6.99		0.06
DI-3	0.50	0.010	0.480	1.428	0.370	0.0070	0.0028	29.75	7.70	0.15	0.06
DI-4	0.67	0 013	0.639	1 573	0.467	0.0089	0.0046	24.61	7.30	0.14	0 07
DI-5	1.00	0.020	0.935	2.458	0.528	0.0099	0.0096	26.29	5.65	0.11	0.10
DI-6	1.25	0.025	1.005	4.059	0.442	0.0075	0.0046	40.38	4.39	0.07	0.05
DI-7	2.50	0.050	1.009	3.386	0 253	0.0043	0.0040	33 56	2.51	0 04	0.04
DI-8	3.33	0.067	1.003	3.410	0.200	0.0034	0.0026	33.99	2.00	0 .03	0.03
DI-9	5.00	0.100	1.002	2.956	0.130	0.0022	0.0018	29.50	1.30	0.02	0 02
DI-10	6.25	0.125	1.002	3.173	0.097	0.0016	0.0018	31.72	isMTs	0.02	0.02
DI-11	8.33	0.167	1.004	3.194	0.077	0.0013	0.0025	31.81	0.77	0.01	0.02
iBliill	12 50	0.250	1.006	iliB//	0 049	0 0008	0.0010	27 21	0.49	i/p/QI/i/	0.01

Table 12 % Meta I loading on air dry resin and %mass balance (out/ln)

/ID/////	OS/O0	Loading (%) based on both resin and Aq assays	/MO0
		Ni	Co	Cu	Fe	Ni	Co	Cu	Fe
DI-1	0.002	0.01	8.96	11.20	41.8	101.2	104.8	51.54
DI-2	0.005	0.05	22.67	14.92	2.51	100.4	100.0	87.91	79.57
Di-3	0.010	0.07	51.75	42.80	5.09	96 93	96.56	85.76	78 36
DI-4	0.013	0.06	65.36	56.73	8.77	92.34	96.33	83.19	75.80
DI-5	0.020	CIO	76 26	///:010///:	14.21	102 1	96 05	92.03	100 9
DI-6	0.025	0 35	78 96	58.67	9.13	96.70	90 21	79.17	84.71
DI-7	0.050	0.48	88.71	70.31	16.57	96.69	91.56	73.60	80.22
DI-8	0.067	0.64	91.97	75.95	15 53	99 49	93.79	71.43	73 75
DI-9	0.100	0.71	93.67	73.21	15.42	97.77	89.80	68.61	73.78
DI-10	0.125	1.02	94.58	70.01	15.65	101.6	83.03	63.35	92.68
DI-11	0.167	1.37	95 65	72.47	30.77	101 5	86.30	66.41	87.67
DI-12	0.250	1.44	97.14	73.88	21.20	97.58	80.76	49.80	68.49

Results: Co distribution isotherm shows a generally normal curve, but it appears not very efficient for low raffinate Co (Figure 17D). At R/S ratios below 1:75, Co loading

WO 2019/090389

PCT/AU2018/051203 approached maximum at about 7.5 mg/g. The raffinate Co decreased consistency with higher R/S ratios down to below 4 mg/L at a R/S ratio of 0.25 (Figure 17C). The raffinate Co concentrations assayed by ICP-AES decreased smoothly and consistently with increasing R/S ratio. However, this was a possible effect of the high concentration of nickel sulfate matrix on the assay of low Co concentrations by the ICP-AES, compared to ICP-MS. Ni loading remained high at 31 ±4 mg/g over the whole R/S ranges examined (Figure 17B). As a result, the selectivity, measured by separation factor (SF) of Co over Ni, peaked at a lower range of R/S ratio (Figure 17A), but the Ni/Co loading ratio on the resin increased linearly with increase in the R/S ratio (Figure 17B). This resulted from more than thousand times higher Ni 10 concentration than Co concentration in the solution phase. Raffinate Cu decreased to below 1 mg/L with increased R/S ratios, while raffinate Fe unexpectedly remained high in the range of 7-12 mg/L. Assay of Fe by ICP-AES appeared to be less affected by the high nickel sulfate matrix. However, assay errors by ICP-AES could be larger than ICP-MS.

Estimation of metal loadings on dry basis

Loading based on dry resin mass which was estimated according to the averaged

31.5% water content (28 - 35 wt%) provided by LANXESS supplier. The data based on the estimated dry resin mass are given in Table 13 and the Co distribution isotherms on air dry basis and dry basis are compared in Figure 17E. The mass capacity (mg/g) can be converted to bulk resin volume capacity (g/L) based on the bulk density of the resin (530 g/L) (Table

13). This gives about 6 g Co / L resin, compared to a minimum 12 g/L Zn provided by

LANXESS supplier. However, the systems, metal concentrations and test conditions were all different. In fact, the average total loading of Ni + Co was 26.8 ± 3.5 g/L (Figure 17E).

Table 13 Loading based on dry resin estimated based on average 31.5% (28 - 35 Wt%) ________________________water content by LANXESS________________________

lllii	IT·!				IIIBiill 1011111		(Ni+Co)
	(g)		ig)	Ni	Co	Cu	Fe	Ni/Co	Ni	Co	Cu	Fe	(g/L)
Feed	0.00	0.000	0.702	25 9	0.00	0.00	0.00		13.7	0.00	0.00	0.00
Di-1	0.07	0.001	0.055	39.0	10.6	0.17	0.21	4	20.6	5.60	0.09	0.11	26.3
Di-2	0 17	0 003	0.182	48.4	10.2	0.15	0.08	5	25.7	5 41	0.08	0.04	31 1
Di-3	0.34	0.007	0.329	43.4	11.2	6.21	0.08	4	23.0	5.96	0.11	0 04	29.0
Di-4	0.46	0.009	0 438	35.9	10 7	0 20	0 10	1113111	19.0	5.65	0.11	0 06	24.7
Di-5	0.69	0.014	0.640	38.4	8.24	0 15	0.15	5	20.3	4.37	0.08	0.08	24.7
ictili	0.86	0.017	0.688	59.0	6.42	0.11	0.07	111	31.2	3.40	0.06	0.04	34.6
Di-7	1 71	0.034	0.691	49.0	3.66	0.06	0.06	13	26.0	1 94	0.03	0.03	27 9
Di-8	2.28	0.046	0.687	49 6	2.92	0 05	0.04	17	26.3	1.55	0.03	0.02	27.8
Di-9	3.43	0.069	0 686	43.1	1.90	0.03	0 03	23	22.8	1.00	0.02	0.01	23.8
DM0	4.28	0.086	0 686	46.3	1.42	0.02	0.03	33	24.5	0.75	0.01	0 01	25.3
Di-11	5.7 Ϊ	0.114	0.688	46.4	1.12	0.02	0.04	42	24.6	0.59	0.01	0.02	25.2
Di-12	8 56	0 171	0.689	39 7	0.71	0.01	0.01	56	21.1	0 38	0.01	10)041	21 4
AVE													26 8
SD													3.5

WO 2019/090389

PCT/AU2018/051203

Langmuir and Freundlich isotherms

The Langmuir adsorption model assumes that molecules are adsorbed at a fixed number of well-defined sites, each of which can only hold one molecule and no transmigration of adsorbate in the plane of the surface. These sites are also assumed to be energetically equivalent and distant to each other so that there are no interactions between molecules adsorbed to adjacent sites. The linear form of the Langmuir isotherm is represented by the following equation:

Ce/Qe = 1/Qs-K + Ce/Qs (4) where Qe is the amount adsorbed (mg/g), Ce is the equilibrium concentration of the adsorbate ions (mg/L), and Qs and K are Langmuir constants related to maximum adsorption capacity (monolayer capacity) (mg/g) and energy of adsorption (L/mg), respectively. A plot of Ce/Qs versus Ce should indicate a straight line of slope 1/Qs and an intercept of 1/Qs K. Freundlich isotherm is an empirical equation that encompasses the heterogeneity of sites and the exponential distribution of sites and their energies.

InQe = (1/n) InCe + lnK_f (5) where K_F and n are Freundlich constants related to adsorption capacity and adsorption intensity, respectively. Data for Langmuir and Freundlich isotherms are given in Table 14 and shown in Figure 17F - 17H respectively. The monolayer loading capacity on dry resin mass was derived through fitting Langmuir isotherm model at about 11.5 mg/g by taking 6 pointe as the linear part, and about 13.2 mg/g by taking 8 pointe as the linear part (Table 14 and Figure 17G).

Table 14 Data for Langmuir and Freundlich isotherms on estimated dry resin basis

ID	Total resin	R/A ratio	Langmuir X	-Y data series	I Freundlich X-Y data I series
	II··			Ce/Qe
D!1	U.U /	i 0.001	147.2 I	13.92	4.99	2 36
DI-2	0 17	i 0.003	1193 ί	11.68	II 1/4)71111	2.32
DI-3	0.34	i 0 007	71.8	6.38	\| 4.27 j	2.42
DI-4	0.46	i O.OOS	...............................OU, ..................................:	4,76	\| 3.93 j	2.37
DI-5	0.69	i 0.014	35.2 i	4.26	3.56	2 11
DI-6	0.86	Ό: Ό: 1: : -Z: : : : : : : :	29.2 I	4.56	1 3.38 ΐ	1.86
DI-7		0.034	15.9 I	4.35	I 2.77 1	1.30
DI-8	2.28	: 0 046	11.6 1	3.98	\| 2 45 1	1.07
DI-9	3.43	0 069	8 77 ΐ	4.63	\| 2.17 \|	0.64
DI-10	4.28	i 0.086		4.SO		0.35
DI-11	S 71	1 0.114	5.78 1	5.17	1 7K	0 11
DI-12	8.56	I 0.171	3.56 ΐ	5.02	I 1.27 1	-0.34
Slope			1/Qs - 0.0585 (6	points) R^z = 0.97	\| 1/n ~	1.044
			1/Qs --- 0.057 (8	points) R^{2 ::} 0.94	R² =	0.99
ilhtetoeiat/					-1.642
					0.99

WO 2019/090389

PCT/AU2018/051203

Variation of solution pH with metal loadings

The solution pH rose sharply from initial 4.6 to above 5.3 when contacted with the preloaded resin (Figure 171). It varied at lower resin / solution ratios, mainly mirrored variations of equilibrium Ni loading, and then steadily decreased to about 5.1 with increased resin / solution ratios. The results again demonstrated that the Ni pre-loaded resin well stabilised or buffered the solution pH over a wide range of R/S examined. This is critical to maintain or buffer the solution pH in a continuous column running mode for practical applications. Example 13 - TP 272 Resin stability tests

Batch test work was conducted to assess the degradation of the resin. The vendor recommended maximum operating temperature for the TP 272 resin is 60^cC compared to the leach PLS temperature of 80°C. Degradation of the resin can either be mechanical degradation of the resin or loss in total ion exchange capacity. The vendor advised that the elevated temperature, the primary issue is the loss of the Cyanex 272 extractant. The extractant can dissolve in the PLS and be removed from the resin.

Test work was conducted in batch with pre-loading, loading, stripping and washing cycles. It was found that degradation of the resin occurred and after 12 cycles, 50% of the total ion exchange capacity was lost. The loss appeared to stabilize after 12 cycles and remained constant. It was found that the pH of the demineralized water used to wash the resin was pH 7-8 which could have cause a large proportion of the Cyanex to dissolve. Column Tests

Column tests were conducted to determine the breakthrough point of the TP 272 resin. The column used was a 150 mL jacketed column operating at 2 BVs/hr and 80°C with synthetic PLS. The composition of the feed and raffinate were analysed by ICP-OES. The results from the column trail are shown in Figure X. The cobalt concentration in the raffinate was much higher than expected, up to 20 mg/L. Samples were sent to Bureau Veritas Ultra Trace for comparison using ICP-MS, It was found that, the ICP-OES gave erroneous results at low cobalt levels due spectral interference from the high nickel concentration (Figure 57A). The breakthrough point of the column was set at 30 BVs/hr which corresponded to a raffinate concentration of 20 mg/L Co. The solution was re-passed through a polishing column which produced a raffinate with 1 - 2 mg/L Co, suitable for crystallisation.

A: Summary of test conditions

Batch loading, stripping and pre-loading tests with TP 272 resin were conducted for 4 weeks (20 days or cycles) at each cycle of 24 hours (Table 15) with a total of 20 cycles over 4 weeks (excluding weekends). Both aqueous and resin samples were taken from each cycle to determine metal loadings and the exchange capacity of the resin overtime. Two independent sets of fresh resin samples and the stability cycled resin samples were reacted with standard NaOH solution for 24 hours, followed by titration procedures above. Conditions

WO 2019/090389

PCT/AU2018/051203 for exchange capacity and typical titration data of fresh resin reference samples are given in Tables 15 and 16 respectively.

Table 15 Summary of stability test scheme and conditions

Cycle steps	Materials	Conditions	Time (h\|
Ni Preloading	Synthetic Ni sulphate solution, eluted / washed Lewatit TP 272	80 °C, pH 5- 5.5 by Ni(OH)₂	1
Washing		80 C, washing 3 times	0.5
Co-loading Elution	Ni loaded Lewatit TP 272, synthetic PLS Co loaded Lewatit TP 272, deionised water Sulfuric acid, Co loaded Lewatit TP 272	80 °C, -pH 5 /80/*CiBdShih§/8ifito 80 °C, 2 M H2SO4	20 0.5 1
	Eiuted Lewatit TP 272, Deionised water	iiSfriO/iOOihisSTlO	0.5
S/L operations	-		0.5

Table 16 Conditions for determination of resin exchange capacity

Conditions	Set 1	Set 2
Resin
Feed NaOH	196.1 mN	19.296 mN
Resin / Solution
Temp	25 °C	25 °C
Titer	H_;;SO₄ (104.3 mN)

B: Total exchange capacity

Two fresh TP 272 resin samples were treated by the standard NaOH solution and duplicate aqueous samples were titrated with standard H₂SO₄ solution. The total exchange capacity of fresh TP 272 was determined to be 1.21 + 0.02 (Taiale 17), and used as reference for calculation of the loss of exchange capacity of the used resin samples in the stability test cycles.

Table 17 Titration data of total exchange capacity using the fresh resin samples

ID	Titer - H₂SO₄	Sample	Resin	H₂SO₄ usage	[NaOH]	A[NaOH]	NaOH
	liWiiiii/i	IBlIiii	(9).........	/(toL)iiiiii///	/(iiN)ilil	/BN)\|i///\|	(rneq/g)
NaOH Feed					19.296	-
Fresh 1-1	21.681 21.681	10 10	0.251	6.20 6.10	13 442 13.225	5 854 6.071	1 17 1.21
Fresh 1-2	21 681 21.681	20 10	0.502	6 55 3.20	/7/1O1\|\|\| 6.938	12.195 12.358	1.21 1.23
Average							1 21
SD							0.02
%Rd							1.99

C: Variation of exchange capacity

Both Set 1 and Set 2 data showed gradual decrease or increased loss (%) in the exchange capacity with a tendency of slow change after a certain cycles (around 15 cycles) (Figure 18). Set 1 data tended to become constant about 42% loss of the total exchange

WO 2019/090389

PCT/AU2018/051203 capacity. In comparison, Set 2 data showed relatively larger fluctuations along the cycles with about 56% for the last cycle. The larger scattered range could be caused by the difference in the detection and sensitivity of end points, where more diluted NaOH feed solution and H₂SO₄titers were used in Set 2.

D: Variation of metal loadings

Variations of metal loadings with the stability test cycles are shown in Figure 19A, No significant variation of Co loading was observed The slightly increasing loading could result from the decreasing resin mass due to resin sampling at the end of each cycle. Ni loading decreased sharply from Cycle 2 and stabilised at about 9 mg/g up to Cycle 18, and then a significant decrease at the last Cycle 20. Fe and Cu loading maintained at a lower levels without significant variations in most of cycles. The variation of total loading, expressed by milimole / g resin (mM/g), mainly followed the trend of Ni loading as the dominant loaded metal.

E: Cobalt behaviour

Except for the initial large changes, the variations of the raffinate Co were small, though a slight increasing trend can be observed (Figure 19B). As samples were taken for each cycle, the total resin mass (initial 50 g air dried) in reaction was decreasing. This may contribute to the variation of Co loading based on the resin mass remaining. Loss of resins in the cycle operations may also introduce errors. The %loading based on resin stripping analysis remained nearly constant with most data >90% except for above 18 Cycles data dropping to --80% (Figure 19B). Behaviours of Ni, Fe and Cu

The initial Ni loading at 44 mg/g sharply decreased to below 10 mg/g after Cycle 2, and then nearly constant at ~9 mg/g, until above 18 Cycles at which a significant drop observed (Figure 19C). This drop was consistent to that observed for Co (Figure 19B), with uncertainty whether this drop suggest the loading capacity be affected by the total exchange capacity, Fe and Cu behaviour shown are in Figure 19D and Figure 19E, Both Fe and Cu were loaded consistently over the most of cycles, leaving the raffinate Fe and Cu below detection limits in most of cycles.

G: Maximum loading tests

In order to verify the loss of exchange capacity as measured by the titration method, maximum loadings of the fresh and the used resin samples (from the last cycle of the stability study) were carried out. Each resin sample was loaded three times with 20 g/L Co at 80 °C. The resin samples were, after wash, and stripped with 2 M H₂SO₄ to regenerate strip liquor samples. Both raffinate and strip liquor samples were assayed by ICP-AES. The test conditions are given in Table 18. The maximum loading of Co using the last cycle (Cycle 20)

WO 2019/090389

PCT/AU2018/051203 was decreased by 54.71%. This agreed well with the loss of total exchange capacity loss in the range previously determined.

Table 18 Test conditions for the maximum loading of cobalt

Resin| Lewatit TP 272 Fresh (F) and Used (U) from the last cycle of the stability study

S/L ratio I 2 g : 50 mL (loading three times with the fresh aqueous solution ___ >initial solution pH i 5.5

Sample stripping I x g resin with 20 mL of 2 M H₂SO₄ at 80 'C

Table 19 Assay data of the resin samples

Resin ID	Resin sample	Strip solution	Strip liquor Co	Co loading	Average	Loss	MB
)))))))))))))))))))))))))))))))))))))/	////(ILF///:	iiiilii		:)/%))/	))Hi))	ISSI)/	)))))«)))))
F1: Fresh resin 1		1.660	20	2.259	27.2	1.82			90.60
lF21FfBhO§iri;2/		1 316	1111201111	1.693	))))261:1	)1)72/	:))))26:::47//:		93 84
Ul: Used resin 1 cycle)	(last	2.038	20	1.211	Ϊ 1.9	0.79			91.57
Idl; Iblelltell 11 IMiillllllllll/	iliii	1.935	1110/11)	11)1/1011	))))1)1)1))))	s	)))))1)1)0)))))	54 71	94.91

H: Summary and recommendations re resin stability

The stability measurement indicated gradual loss of exchange capacity with cycles and significant loss after 20 cycles (42% by Set 1 measurement, and 56% by Set 2 measurement). This agreed well with 54% loss of maximum Co loading. No significant decrease in Co loading was observed. Ni loading was dominant and stabilised at a certain level (-9 mg/g) after the big drop from Cycle 2, until above Cycle 18 after which a significant drop in Co and Ni loading observed.

Example 13 - Tests using Lewatit TP 207

Tests using Lewatit TP 207 was mainly aimed at removal of Fe and Cu impurities, and behaviours of Co and Ni. The test results are described below.

A: Loading pH isotherms

Loading pH isotherms with consecutive loading and sampling

Initial loading pH isotherms with Lewatit® MonoPlus TP 207 and the synthetic PLS was carried out in a mixer at 80 °C and 1:2 (g/L) resin/PLS ratio. Solution pH was adjusted with the synthesised Ni(OH)₂. The aqueous samples was taken and analysed for metal concentrations by ICP-AES. The bulk resin at the last equilibrium pH was washed and stripped with 2 M H₂SO₄ and the strip liquor samples were assayed by ICP-AES for metal concentrations. The variation of raffinate metal concentrations and percent loadings with pH based on aqueous analysis results were shown in Figure 13J and Figure 20B, respectively, and metal loadings based on assay of final resin strip liquor in Table 20. The observations

WO 2019/090389

PCT/AU2018/051203 are outlined as follows: Raffinate Fe significantly decreased above pH 3 and lowered to below 1 mg/L Fe (> 90%) above pH 4; Raffinate Cu decreased slowly with increasing pH, reaching 55% at pH 4.6; Co loading on the resin based on aqueous analysis results showed a significant increase at pH above 4.6 (>30 mg Co / g resin); Ni loading was dominant at 114 mg/g at pH 4.6, together with minor loading of Fe (0.44 mg/g) and Cu (0.47 mg/g), and little loading of Co (0.02 mg/g), based on the assay of final resin strip liquor; Mass balance (Out/ln) for Ni (92%), Co (89%), Fe (12%), and Cu (51%), respectively, suggesting significant loss of Fe and Cu, through hydroxide precipitations during the loading tests, which could be promoted by addition of Ni(OH)₂ crystals via creating local alkaline conditions.

Table 20 Metal loadings at last equilibrium pH 4.6 based on resin stripping

Elements	Fe	Cu	Ni	Co
Loading on resin (mg/g)	0.44	9.47	114.4	0.92
%Mass balance (Out/ln)	12	51	92	89

Loading pH isotherms with separate loading and sampling TP 207 resin

To confirm the results in terms of possibility for extraction and removal of cobalt with Lewatit® MonoPlus TP 207 resin, the pH isotherm tests were carried out in a different way, i.e. loading doses of resin in separate closed contactors for respective target pH to minimise error and to enable resins to be stripped for each pH loading. Sodium hydroxide was used for adjustment of initial pH to compare the effect of Ni(OH)₂ on enhancement of Fe and Cu hydrolysis to low levels.

Table 21 Test conditions for loading pH isotherms in separate contactors

Resin	Lewatit® MonoPlus TP 207
Aqueous solution	Synthetic PLS
RiS ratio (g/L)	5:1
T (°C)	80
pH adjustment	H₂SO₄ and NaOH solutions
Sample stripping	x g resin with 2 M H₂SO₄ at 89°C

Metal loading efficiency (%) based on resin stripping assays and feed PLS, raffinate metal concentrations, and metal loadings on resin versus pH are shown in Figure 20C, Figure 20D and Figure 20E, respectively. The behaviours of Ni, Co and Cu in these tests were essentially similar to those observed for the consecutive loading and sampling above: dominant Ni loading (-150 mg/g Ni) with low loading of Fe and Cu, and little co-loading of Co. However, the behaviour of Fe was different. The level of Fe remained at about 10 mg/L in the raffinate. This may suggest that neutralisation with nickel hydroxide promote the precipitation of Fe. The metal accountabilities were reasonably good (Table 22).

WO 2019/090389

PCT/AU2018/051203

Table 22 Metal accountability data for the loading pH isotherms

pH	Mass balance (Out/ln %)
5.30	Ni 99	Co 96	Cu 79	Fe 117
4.80	99	98	80	85
4.10	102	99	86	101
3.65	101	99	82	102
3.15	96	94	83	88
2.47	98	98	88	liliTiiii
1.90	102	100	92	91

B: Loading kinetics with TP 207

Conditions of loading kinetics with TP 207 are given in Table 23, and result and findings are shown in Figure 21A and Figure 22B. The result are summarised as follows: Initial Fe at pH 4 was only 3 mg/L, but increased to ~10 mg/L within 5 min and then constant at -9 mg/L. This could result from possible Fe precipitated during pH adjustment and redissolution back Into solution as pH decreased from 4 to ~3, but little loading of Fe on the resin within 60 min; Cu slightly decreased with <15% loading within 60 min; Final resin stripping and analysis showed loading (mg/g): Ni 95.34, Co 0.03, Fe 0.16, Cu 0.18.

The results generally suggest that the TP 207 system was not efficient and selective for Cu and Fe, and Co, most likely caused by very high dominant Ni loading at this pH range.

Table 23 Test conditions for loading kinetics with TP 207

Resin	Lewatit® MonoPlus TP 207
Aqueous solution	Synthetic PLS
R/S ratio (q/L)	3:1
T (Ό)	80
pH adjustment	H₂SO₄ and NaOH solutions
Sample stripping	x g resin with 2 M H₂SO₄ at 80“C

C: Loading distribution isotherms

Conditions of loading distribution isotherms with TP 207 are given in Table 24, and results are shown in Figure 22A to 22D respectively. The results and finding are summarised as follows: Feed Cu concentration was low, attributable to its hydrolysis at relatively high pH, and remained low at < 1 mg/L; Raffinate Fe was low (~ 1 mg/L) only observed at low S/R ratio (<200 mL/g), and remained high at > 200 mL/g; Both loading of Fe and Cu on resin were low (<0.5 mg/g); Ni loading dominantly high at -150 mg/g; Co loading is very low < 0.5 mg/g; Fe distribution isotherm curve with selected consistent data points suggests unfavourable distribution shape for removal of Fe with the TP 207 resin system from the PLS.

Table 24 Test conditions for loading distribution isotherms with TP 207

Resin___________ Aqueous solution	Lewatit® MonoPius TP 207 50 mL Synthetic PLS (initial pH 4.5)
R/S ratio (g/L)	Various

WO 2019/090389

PCT/AU2018/051203

: (°C)	80
Mixing time (h)
Sample stripping	x g resin with 2 M H₂SO₄ at 80°C

D: TP 207 - Elution

Conditions of loading distribution isotherms with TP 207 are given in Table 25 and results are shown in Figure 23. The results and finding are summarised as follows: 1 -1.5 M H₂SO₄ for nearly complete stripping of Fe and Cu; 0.5 M H₂SO₄ for > 95% Ni stripping; M H₂SO₄ for -80% Co stripping, but it was uncertain if the remaining Co required 4 M H₂SO₄ or analysis error due to very low concentration of Co loading (< 1 mg/g); Potential selective scrubbing and stripping to separate metals.

Table 25 Test conditions for elution isotherms with TP 207

Resin	Lewatit® MonoPlus TP 207
Aqueous solution	Various concentrations of H₂SO₄ solutions
R/S ratio (q/mL)	1:30
T(^!C).............................	80
Mixing time (h)	'!
Sample stripping	x g resin with 2 M H₂SO₄ at 80°C

Example 14 - Other resins / SX systems

A: IX - Amberlite IRC 748

Amberlite IRC 748 has the same function group as Lewatit MonoPlus TP 207. pH isotherms were conducted to provide a reference. Conditions of loading distribution isotherms with TP 207 are given in Table 26, and results are shown in Figures 24A to 24C. The results and finding are summarised as follows: Loading behaviour with IRC 748 essentially similar to TP 207, with dominant Ni loading and minor Co loading. Fe decreased from feed -12 mg/L to about 2 mg/L in the raffinate, but the loading tended to decrease at pH > 4.0. Cu decreased from feed about 4 mg/L to below 1 mg/L and remained low in raffinate, corresponding to low and constant loading over the pH range.

Table 26 Test conditions for elution isotherms with Amberlite IRC 748

Resin	Amberlite IRC 748
Aqueous solution	Synthetic PLS
R/S ratio (g/mL)	25:90
TCC)	80
Mixing time (h)	1
Sample stripping	x g resin with 2 M H₂SO₄ at 80°C

B: IX - Purolite S910

Purolite S910 is amidoxime chelating resin. Metal pH isotherms were conducted for comparison with TP 272. Conditions of loading distribution isotherms with TP 207 are given in

WO 2019/090389

PCT/AU2018/051203

Table 27, and results are shown in Figure 25. The resin showed more selective and efficient for loading Cu over other metals.

Table 27 Test conditions for elution isotherms with Purolite S910

Resin	Purolite S910
Aqueous solution	Synthetic PLS
R/S ratio (a/rnL)	25:90
T (Ό)	80
Mixing time (h)	1
Sample stripping	x g resin with 2 M H₂SO₄ at 80^;;C

C: SX - Cyanex 272 system

A reference SX system of 10% Cyanex 272 with the same function group as TP 272 resin was proposed and metal extraction pH isotherms carried out using the same synthetic PLS. Solution pH was adjusted with H₂SO₄ and NaOH solutions. The organic samples were stripped with 2 M H₂SO₄ and the strip liquor and aqueous samples were assayed by ICPAES. Metal extraction efficiency and metal concentrations in the organic and raffinate are shown in Figure 26A to 26C, respectively.

The efficiency of metal extraction (%) was in the order of Fe > Cu > Co > Ni (Figure 26A). Selectivity of Co over Ni increased and reached the maxima at pH about 4.5 and then a sharp decrease at above pH 5 where the extraction of Co nearly completed, but the extraction of Ni continuously increased with increased pH (Figure 26B), In fact, the extraction of Ni commenced from a lower pH 2 and the concentration of extracted Ni was comparable to that of Co in the organic phase. Raffinate Co was lowered to below 3 mg/L at > pH 5 (Figure 26C).

Example 15 - mini column tests

A primary (lead) column (Column 1) was constructed and went through all the running steps, including Ni pre-loading, preload-wash, Co loading and load-wash, elution and elutionwash. Other two columns were similarly constructed and used for simulating polishing (lag) column, and fortesting the effect of pH on loading efficiency and selectivity. The results are briefly summarised in the following sections.

A: Ni pre-loading

Columns were pre-loaded with synthetic Ni sulfate solution under the conditions given in Table 28. Typical loading data for Column 1 were presented in Table 29. The solution pH was stable between 5 and 5.5 with slightly decreasing trend with increased bed volume (BV). The Ni loadings with 4 and 10 BV for different columns were all similar at about 13-14 mg/g (Table 30).

Table 28

Conditions for Ni preloading

WO 2019/090389

PCT/AU2018/051203

Column	Column 1 0 21.5 x 400 H
Volume	-Ί45 mL
Resin	64.9 g Lewatit TP 272
Feed	Ni sulfate solution
Temperature	80 °C
Flow rate	5 / BV/hr
	12 5 mL/min

Table 29 Column 1 Ni preloading data

ID	BV	Time	pH	Ni Concentration
		(min)		(g/L)
CPLT1-0	Feed	0	5.20	183.8
CPLT1-1	1	24	5.50	187.1
CPLT1-2	2	36	5.30	187.5
CPLT1-3	3	48	5.30	202.3
CPLT1-4	ill»	60	5.20	186.8
CPLT1-5	5	72	5.20	187.2
CPLT1-6	6	84	5.23	191.2
CPLT1-7	7	96	5.10	188.4
CPLT1-8	TfBlf	108	5.15	190.7
CPLT1-9	9	120	5.04	191.2

Table 30

Ni loading (sampling on top of the columns)

Column	lllllllll	BV	(min)	Illi		g resin)
Column 1	CPLT1-10	10	120	5.1	183.8	0.014
Column 2	CPLT3-4	4	122	5.1	138.9	0.013
Column 3	CPLT5-4	4	122	5.2	166.8	0.014

S: Column test conditions

Test conditions for Column 1 are summarised in Table 31, involving all running steps.

WO 2019/090389

PCT/AU2018/051203

Table 31 Conditions for Column 1 Co loading / washing and elution / washing

	Load	Load-Wash	Elution 1
Solid	Ni Preioaded TP 272	Co Loaded TP 272	Washed TP 272
Aqueous	Ni leach (batch 1 and	Di Water	0.1 M H;SO₄
Column	145 mL/64.9 g air dry resin	145 mL/64.9 g air dry resin	145 mL/64.9 g air dry resin
T (°C)	80	65 - 70	50
Flow rate (BV/hr)	2	5	2
Flow rate (mL/min)	lilllllllllllllll	lilillllllllllllli	lilllllllllllllll

	Elution 2	Elution 3	Elution Wash
Solid	Elution 1 TP 272	Elution 2 TP 272	Elution 3 TP 272
Aqueous	0.5 M H₂SO₄	i .5 M H_;;SC’4	DI Water
Column	145 mL/64.9 g air dry resin	145 mL/64.9 g air dry resin	145 mL/64.9 g air dry resin
T (°C)	50		50
Flow rate (BV/hr)	2	2	5
Flow rate (mL/min)	lilllllllllllllll	lilllllllllllllll	12.5

C: Column 1 preload washing

Washing of Ni from the pre-loaded resin was very efficient, requiring a couple of BV to obtain > 99% washing efficiency.

Table 32 Ni washing efficiency from the pre-loaded resins

ID	BV	COifrOWO (min)	Iilllllilllli	Ni	Column 1 iiililiil		lllllllll
CPLT1-0	0	0	5.2	184
CPLT2-1	1	24	6.30	4.503	97.55	99.64	99.86
CPLT2-2	2	36	6.50	0.281	99.85	99.86	99.89
CPLT2-3	113(1	48	6.55	0.120	99.93	99.91	99.94
CPLT2-4	4	60	6.60	0.113	99 94
CPLT2-5	5	72	6.50	0.096	99.95
CPLT2-6	6	84	6.40	0.093	99.95
CPLT2-7	117(1	96	6.40	0.102	99.94
CPLT2-8	8	108	6.40	0.083	99.96
CPLT2-9	9	120	6.38	0.080	99.96

D: Column 1 - loading

Column 1 loading using the pregnant leach solution (PLS) are shown in Figure 27A.

The aqueous samples were initially assayed by ICP-AES, and later some selected samples were verified by ICP-MS. Both sets of data were compared in Figure 27A.

WO 2019/090389

PCT/AU2018/051203

The initial two consecutive runs (up to 26 BV) and the two batch samples assayed by ICP-AES gave significant positive errors on the Co concentrations, compared to those by ICP-MS. The assay data above 26 BV agreed reasonably well between the ICP-AES and ICP-MS. One possible cause for the errors could be the high Ni sulfate matrix concentration. Rapid assays by ICP-AES as requested for obtaining data might also contribute to large errors. Based on the ICP-MS assays, raffinate Co was about 1% at beginning of 1 BV, and consistently increased almost linearly, if some new consecutive running points around 28 BV was excluded. This agreed with the observations from batch Co distribution isotherms regarding lower efficiency of the system with TP 272 and the PLS containing high Ni sulfate for low raffinate Co. About 10 mg/L Co (-7% breakthrough of the feed 136 mg/L Co) occurred at about 30 BV. The raffinate below 10% breakthrough were used as the feed for polishing running. The cobalt loading approached saturation at about 62 - 72 BV with loading about 11 mg/g Co on the air dry resin basis, while Ni loading is 27 - 35 mg/g (Figure 27B). E: Column 3 - polishing loading

Polishing loading was carried out using the raffinate of Column 1 (< 6 % breakthrough) as the feed. ICP-MS assay results are given in Table 33 and shown in Figure 28. With 3 different feed solutions (raffinate from different runs of Column 1) and 3 different flow rates (5, 2, 0,8 BV/h), the polishing resulted in the raffinate Co levels remaining in the range of 1 - 2 mg/L Co, The effect of flow rate or the extended residence time on the polishing was not very significant. The rate of mass transfer and diffusion may become slow at low flow rate, which needs further optimisation in large scale and is expected to be improved at a large scale.

WO 2019/090389

PCT/AU2018/051203

Table 33 Assay results (Co and Cu by ICP-MS) by Bureau Veritas for samples from ___________ Column 3 - Polishing Runs with Column 1 raffinate as feed_____________

Run		Sample	Assay by Bureau Veritas	Assay data x 4
		))(\|)i)ii)\|)		Mill	7C0))))))))^ liill	iiiii	liiiiii	liill	liill	till	IIIII	lillll
		UNITS	mg/L	mg/L	mg/L	mg/L	mg/L	..9/4........	mg/L	mg/L	mg/L	.9/1........
5 BV/h	T11-0	182	49200	1.94	0.1	1	27800	196.8	7.76	0.4	4	m2
T11-4	186	45300	0.38	0 09	-1	25300	181 2	1 52	0.36	“4	101.2
T11-7	189	44100	0 45	0.12	1	24600	176.4	1.8	0 48	4	96.4
T11-8	190	45100	0.29	0.08	-1	25000	180 4	1 16	0.32	“4	100
))11)77))))) ))7)7))))))))))	iilillli	44900	t)iiii	)Bii))))	))i)))))))))	25300	iiiii	))ili)))	ttii	114111	))71))1))
2 BV/h	T17-0	196	44100	0.67	0.15	-1	24700	176.4	2.68	0.6	-4	98.8
T17-0	leii	44600	1)7111	))0))1)))))	))7))))))))))	24300	178.4	))2)14)))	))0)6))))))	))4))))))))))	99.2
T17-3	199	44700	0.35	0.08	2	25000	178.8	1.4	0.32	8	100
T17-7	203	43900	))0)41)))))	))0)02)))))	))2))))))))))	27400	195.6	1.64	3.68	))8))))))))))	109.6
T17- 10	206	46900	0.45	0.11	-1	26200	187.6	1.8	0.44	~4	104.8
))11)77))))) iiiili	till!	51500	)Bii))))	)i)i))))))	))·)))))))	28500	titii	)i))iii	iiiii	1141111	))i)i))))))
T20-1	213	45200	0.42	0.12	-1	25000	180.8	1.68	0.48	~4	100
Τ2Ω-5	lITliil	43300	))0)73)))))	))0))1))))))	))2))))))))))	24200	173.2	2.92	0.4	))1))))))))))	96.8
T20- Raf	219	47000	0.51	0.18	-1	26600	188	2.04	0.72	-4	106.4
0.8 BV/h	T21-0	221	36900	))0)11))))	0.11	1	20500	147.6	1.48	0.44	Ti4iiili	82
T21-2	223	43300	0.41	0.11	-1	23700	173.2	1.64	0.44	4	94.8
T21-7	227	35500	))0)11)))))	0.09	))i)))))))))	19600	142	))7)11))	0 36	Ti4iiili	76.4
T21-7	227 Rpt	36600	0.31	0.1	1	20800	146 4	1.24	0.4	4	83.2

F: Column 1 Load washing

See Table 34 and Figure 29.

Table 34 Column 1 Co load washing.

						Concentration
Tests	ID	BV	Time	pH	Temp	Ni	Co	Cu	Fe
			(min)		(°C)	(g/L.)	(mg/L)	(mg/L)	(mg/L.)
Load Wash	T22-12	1.0	12	6.30	70	16.021	9.660	0.0002	0.0002
Load Wash	T22-13	)ii))))	24	6.57	65	3.691	2.944	0.0002	0.0002
Load Wash	T22-14	3.0	36	6.60	65	2.678	1.790	0.0002	0 0002
Load Wash	T22-15	4.0	48	6.55	65	2.201	1.519	0.0002	0.0002
Load Wash	T22-16	5.0	60	6.65	65	1.892	1.080	0.0002	0.0002
Load Wash	T22-17	6.0	))71))))))))	6.50	65	1.677	0.871	0.0002	0.0002
Load Wash	T22-18	7.0	84	6.50	65	1.453	1.935	0.0002	0.0002
Load Wash	T22-19	8.0	96	6.50	65	1.308	1.494	0.0002	0.0002
Load Wash	T22-20	9.0	108	6.50	'65	1.185	1.898	0.0002	0.0002
Load Wash	T22-21	10.0	120	6.50	Olli	1.323	2 068	0.0002	0.0002

WO 2019/090389

PCT/AU2018/051203

Table 35 Column 1 elution conditions

Column 1	Elution Run 1	Elution Run 2	Elution Run 3
Test ID	T24	T25	T26
Resin TP 272	Loaded resin	Eluted in Run 1	Eluted in Run 2
Aqueous	0.1 M H₂SO,_;	0.5 M H₂SO₄	1.5 M H₂SO₄
Resin bed volume (BV)	-145 mL	-145 mL	-145 mL
Resin mass (air dried)	64.9 g	64 9 g	64.9 g
T(°C)	50	50	50
Flow rate (BV/hr)	2	2	2
Flow rate (mL/min)	5	5	5
Resin sample stripping	10 mL 2 M H₂SO₄	10 mL 2 M H₂SO₄	10 mL 2 M H₂SO₄

Table 36 Eluate assay data by ICP-AES

Elution	lllll	lltl	Time	lliili (g/L)	Co (g/L)	liillil (mg/L)	lets (mg/L)
Elution Run 1	T24-1	1.0	30	7.219	2.652	0.200	0.200
T24-2	2.0	60	7.589	3.646	90.63	0.200
T24-3	3 2	95	2.137	1 232	37.49	17.02
T24-4	4.0	120	0.839	0.559	16.28	22.28
T24-5	5.0	150	0.027	0.011	0.200	22.21
T24-6	6 0	180	0.042	0 014	0.200	19.42
Elution Run 2	T25-1	6.4	193	0.028	0.007	0.200	6.991
T25-2	7.0	210	0.013	0.002	26.21	12.42
T25-3 T25-4	8 0 9.0	240 270	0.005 0.007	0 000 0.001	0.200 0.200	13.00 161.1
T25-5	10.0	300	0.003	0.000	0.200	13.08
T25-6	11.0	330	0.004	0.000	0.200	4.364
T25-7	12.0	360	0.004	0.000	0.200	2.249
Elution Run 3	T26-1	12.5	376	0.014	0.001	0.583	6.869
T26-2	13.0	390	0.005	0.000	0.429	5.623
T26-3	14.0	420	0.002	0.000	0.389	1.154
T26-4	15 0	450	0.001	0 000	0.390	0.708
T26-5	16.0	480	0.001	0.000	0.379	0.253
T26-6	17.0	510	0.002	0.000	0.374	0.200
T26-7	18 0	540	0.001	0 000	0.359	0.200

WO 2019/090389

PCT/AU2018/051203

Table 37 Resin sample stripping data

	Resin samples	Mass	Strip Sol Volume	Strip liquor fg/L} by ICP-AES	Loading (mg/g air dried resin}
		(9’	(mL\|	Ni	Co	Cu	Fe	Ni	Co	Cu	Fe
	Loaded ItaSihsIlIll	0.217	lilllill	0 583	lilBl	0.005	0.007		11 43	liiii	jiiij
Column 1	Loaded resin 2	0.268	10	0.740	0.360	0.007	0.011	27.61	13.42	0.24	0.42
Averaged loading	27.23	12.42	0.23	0.37
T24	Eluted resin	0.321	10	0.003	0.001	0.000	0.013	0.09	0.03	0.01	0.42
T25	lOOOiii Hll	0.196		0.001	liBi	0.000	0.001	liiii		li	0.07
T26	Eluted resin	0.196	10	0.001	0.000	0.000	0.000	0.06	0.01	0.02	0.01

Table 38 Elution efficiency based on resin strip liquor assay (%)

Test ID	Elution efficiency based on resin assays (%)
Ni	I Co	Cu	i Fe
T24~Run 1 (0.1 M H₂SO₄)	99 69	i 99.75	..i -I	i -12.32
T25~Run 2 (0 5 M H₂SO₄)	99.90	I 99.92	92 01	i 82 39
T26 -Run 3 (1.5 M H_?SO<)	99.80	ϊ 99.92	91.71	i 97.25

G: Column 1 Elution

See Tables 32 to 35 and Figure 30.

Table 39 Column 1 elution washing data

Tests	ID	BV	Time	pH	Temp	Concentration (g/L)
			(min)		(°C)	Ni	Co	Cu	Fe
Strip Wash	T27-1	0.6	11111	0.80	50	0.0050	0.0002	0.0004	0.0010
Strip Wash	T27-2	1.0	12	1.10	50	0.0034	0.0002	0.0004	0.0002
Strip Wash	T27-3	2.0	Illi	2.30	10111	0.0006	0.0002	0.0004	0.0002
Strip Wash	T27-4	3.2	38	2.70	50	0.0016	0 0002	0.0004	0.0002
Strip Wash	T27-5	4.0	48	2.70	50	0.0025	0.0002	0.0003	0.0002
Strip Wash	T27-6	5.0	60	3.22	50	0.0014	0.0002	0.0003	0.0002
Strip Wash	T27-7	6.0	1BIIII	2.90	llllll	0.0119	0.0002	0.0003	0.0002
Strip Wash Strip Wash	T27-8 T27-9	7.0 7.9	84 195111	2.80 2.90	50 50	0.0146 0.0006	0.0002 0.0002	0.0004 0.0004	0.0002 0.0002
Strip Wash	T27-10	9.0	108	3.00	50	0.0003	0.0002	0.0003	0.0002
Strip Wash	T27-11	10 0	111·	3.05	50	0.0006	0.0002	0.0003	0 0002

H: Column 1 Elution washing

See Table 36 and Figure 31.

Claims

Claims

1. A process for producing high purity nickel sulfate, preferably having a purity of >99.8%, more preferably >99.98%, suitable for use in battery manufacture or nickel plating, comprising the step of:

selectively removing non-nickel metal impurities from a nickel sulfate solution, preferably a sub-saturated nickel sulfate solution obtained from nickel powder, by ion exchange using a nickel pre-loaded ion exchange (IX) resin which adsorbs non-nickel metal impurities from the solution to form a substantially non-nickel metal impurities free nickel sulfate solution from which the high purity nickel sulfate can be recovered.
2. The process of claim 1, wherein the step of selectively removing the non-nickel metal impurities involves buffering the nickel sulfate solution prior to ion exchange (IX) with one or more basic nickel compounds to a pH optimised for selectivity and stability of non-nickel metal impurity loading onto the nickel pre-loaded ion exchange resin.
3. The process of claim 1 or claim 2, further comprising the step of recovering the high purity nickel sulfate from the substantially non-nickel metal impurities free nickel sulfate solution, preferably in the form of crystalline alpha nickel sulfate hexahydrate.
4. The process of any one of claims 1 to 3, wherein the nickel sulfate solution is an acidic sub-saturated nickel sulfate solution, and wherein the method comprises prior to the selective removal step, the additional steps of:

(i) generating the acidic sub-saturated nickel sulfate solution which comprises the one or more non-nickel metal impurities; and (ii) removing bulk non-nickel metal impurities from the acidic sub-saturated nickel sulfate solution to form a non-nickel metal depleted nickel sulfate solution comprising a subsaturated concentration of nickel sulfate and trace amounts of one or more non-nickel metal impurities.
5. A process, preferably a batch process, for leaching nickel sulfate from nickel powder comprising the steps of:

(i) leaching a stoichiometric excess of the nickel powder with sulfuric acid to form an acidic sub-saturated solution, preferably a pregnant leach solution, of dissolved nickel sulfate and one or more non-nickel metal impurities, together with unleached nickel powder;

(ii) separating the acidic sub-saturated nickel sulfate solution from the unleached nickel powder to provide a discharge solution which is a substantially solid-free acidic subsaturated nickel sulfate solution; and optionally repeating steps (I) and (ii) one or more times, wherein the one or more additional leaching steps (I) are carried out with sulfuric acid, preferably using the unleached nickel solid separated in step (ii).
6. The process of claim 5, further comprising the steps of:

WO 2019/090389

PCT/AU2018/051203 (I) precipitating substantially all of the non-nickel metal impurities as an insoluble non-nickel metal impurity precipitate; and (ii) removing the precipitate from the discharge solution to form a non-nickel metal impurity-depleted discharge solution comprising a sub-saturated solution of nickel sulfate and trace amounts of the one or more non-nickel metal impurities.
7. The process of claim 6, further comprising the step of oxidising the non-nickel metal impurities in the discharge solution, preferably prior to precipitation.
8. The process of any one of claims 5 to 7, wherein the precipitated non-nickel metal impurities comprise non-nickel metal hydroxides.
9. The process of any one of claims 5 to 8 wherein the precipitation step involves increasing the pH of the discharge solution to a pH of about 3 to about 7, preferably to a pH of about 4.5 to about 5.5, more preferably to a pH of about 5.
10. The process of claim 9 wherein the pH of the discharge solution is increased by the addition of one or more basic compounds, preferably one or more basic nickel compounds, most preferably nickel hydroxide.
11. The process of any one of claims 5 to 10, further comprising the step of removing hydrogen gas evolved during the process by flushing with steam and/or by operating the process at a positive gauge pressure to prevent air ingress, and/or by carrying out the process under a nitrogen blanket to prevent the hydrogen mixing with air to form an explosive environment.
12. The process of any one of claims 5 to 11, wherein the nickel powder has an average particle size of from about 10 microns to about 500 microns, and preferably has a purity of >99.8%, more preferably >99.98% nickel, for example, which is obtainable from solution by hydrogen pressure reduction.
13. The process of any one of the preceding claims, wherein the one or more non-nickel metal impurities include one or more of Ca, Al, Na, P, Si, K, Mg, Mn, Se, Cr, Co, Fe, Cu, Zn, As, Ru, Pb, Hr, Pd, Ag, Cd, Sb, Ir, Pt, Au, and Bi, more preferably Co, Cu, and/or Fe.
14. The process of any one of the preceding claims, wherein the process is carried out at a process temperature of from about 50 °C to about 100 °C, most preferably from about 70 °C to about 95 °C, most preferably still at about 80 °C.
15. The process of any one of claims 5 to 14, wherein leaching step (i) proceeds for a period of from about 2 to about. 30 hours, more preferably from about. 8 to about 24 hours, more preferably still from about 9 to about 12 hours, most preferably for about 10 hours, and/or wherein the step (i) proceeds until the pregnant leach solution has a pH of from about 0 to about 4, more preferably a pH of about 1 to about 3,5, most preferably a pH of about 3.

WO 2019/090389

PCT/AU2018/051203
16. The process of any one of the preceding claims, wherein the nickel sulfate solution is a sub-saturated nickel sulfate solution comprises dissolved nickel in a concentration that is about 70% to about 99%, more preferably from about 92% to about 97%, most preferably about 95% of the theoretical nickel sulfate saturation concentration.
17. The process of any one of claims 5 to 16, wherein the concentration of sulfuric acid used in leaching step (i) is initially from about 150 g/L and about 350 g/L, more preferably from about 200 g/L and about 300g/L, more preferably still from about 250 g/L to about 290g/L, most preferably about. 280 g/L, and/or wherein the nickel powder is provided/maintained at a loading of between about 850 g/L and 1200 g/L, more preferably, from about 950 g/L to about 1100 g/L, most preferably about 1000 g/L (mass of nickel per litre of sulfuric acid); and/or wherein the nickel in the sulfuric acid has a pulp density in the range of from about 500 g/L to about 1500 g/L, preferably from about 700 g/L to about 1300 g/L, more preferably from about 750 g/L to about 850 g/L, most preferably about 800 g/L.
18. The process of any one of the preceding claims, wherein the process occurs at about atmospheric pressure or at a positive gauge pressure.
19. The process of any one of claims 2 to 14, wherein after non-metal impurity precipitation, the discharge solution has a pH of from about 3 to about 7, more preferably about. 4 to 6, most preferably a pH of about 5; and/or wherein the discharge solution comprises from about 130 g/L to about 210 g/L nickel, more preferably from about 140 g/L to about 200 g/L nickel, more preferably still from about 150 g/L to about 195 g/L nickel, and most preferably about 192 g/L.
20. A process for producing a high purity nickel sulfate solution suitable for crystallisation of high purity, preferably >99.8%, more preferably >99.98% purity nickel sulfate hexahydrate, comprising the steps of:

(iv) providing a nickel sulfate solution, preferably being a substantially solid-free acidic sub-saturated nickel sulfate solution, comprising dissolved nickel sulfate and one or more non-nickel metal impurities, preferably a pregnant leach solution, more preferably obtainable by a process as defined in any one of claims 5 to 19;

(v) selectively removing non-nickel metal impurities from a nickel sulfate solution by ion exchange (IX) by subjecting a solution of nickel sulfate, comprising dissolved nickel sulfate and one or more non-nickel metal impurities to a first round of ion exchange using a resin pre-loaded with nickel, (vi) whereby non-nickel metal impurities in the nickel solution are selectively retained by the resin to generate a first IX discharge solution which is a clean nickel sulfate solution substantially free of non-nickel metal.

WO 2019/090389

PCT/AU2018/051203
21. The process of claim 20, wherein the nickel sulfate solution has an Initial pH of pH < 6, preferably a pH of from about 4.5 to about 6, more preferably, a pH of about 5 to about 5.5, most preferably, a pH of about 5 prior to the first round of ion exchange.
22. The process of claim 20 or claim 21, wherein the subjecting step includes the step of buffering the nickel sulfate solution prior to ion exchange (IX) with one or more basic nickel compounds to a pH optimised for selectivity and stability of non-nickel metal impurity loading onto the nickel pre-loaded ion exchange resin, for example, where the amount of nickel hydroxide provided is sufficient to neutralise acid released by the resin in exchange for nickel during nickel preloading,
23. The process of any one of claims 20 to 22, wherein the resin of the first round of ion exchange is preloaded with nickel using a nickel pre-loading feed solution which is preferably a portion of the first IX discharge solution, or an alternative source of filtered clean nickel sulfate solution, preferably in the presence of nickel hydroxide for pH adjustment/buffering.
24. The process of claim 23, wherein the pre-loading feed solution has a nickel concentration in the range of from about 130 g/L to about 210 g/L nickel, more preferably from about 150 g/L to about 200 g/L nickel, more preferably still from about 175 g/L to about 195 g/L nickel, and most preferably about 190 g/L.
25. The process of any one of claims 20 to 24, further comprising the step of generating the nickel hydroxide for pH adjustment/buffering from clean nickel sulfate solution, preferably which is a portion of the first IX discharge solution or an alternative source of filtered clean nickel sulfate solution.
26. The process of any one of claims 20 to 25, further comprising subjecting the first IX discharge solution to a second round of ion exchange using nickel preloaded resin to provide a second IX discharge solution being a polished IX discharge solution.
27. The process of claim 26, wherein the second round of ion exchange is used where impurity break through is observed in the first IX discharge solution.
28. The process of any one of claims 20 to 27, further comprising the step of regenerating the resin used in the first round of ion exchange, preferably on observation that (i) loading of substantially all of the non-nickel metal impurities onto the resin has been achieved or (ii) breakthrough of the non-nickel metal impurity in the first IX discharge solution.
29. The process of claim 28, wherein the step of regenerating the resin used in ion exchange involves the steps of:

(i) washing the resin free of entrained nickel with an aqueous solution of acid, preferably sulfuric acid, preferably having a pH of about 4 to about 5;

(ii) stripping retained non-nickel metal impurities from the resin using an aqueous acid solution, preferably sulfuric acid, preferably at a concentration of from about 0.25 M to

WO 2019/090389

PCT/AU2018/051203 about 2.5 M acid, more preferably 1.0 M to about 2.0 M acid, wherein the stripping step exchanges the retained non-metal impurities for hydrogen ions thereby loading hydrogen ions onto the resin to provide resin in the hydrogen form;

(iii) removing excess acid from the resin by washing with waler, preferably demineralised water;

(iv) pre-loading the resin with nickel by flushing the resin with a solution comprising a clean nickel sulfate solution preferably being a portion of the IX discharge solution, preferably together with nickel hydroxide.
30. The process of any one of claims 20 to 29, wherein the resin is provided in two or more columns arranged in a lead/lag configuration.
31. The process of any one of claims 20 to 30, wherein the resin is a cation exchange resin, preferably a macroporous crosslinked polystyrene based resin, which is preferably acidic or otherwise capable of chelating metal ions, for example, a phosphoric, phosphonic or phosphlnic acid based resin.
32. The process of any one of claims 20 to 31, wherein the first and one or more additional rounds of ion exchange are carried out at a temperature of from about 50 °C to about 95 °C, more preferably from about 70 °C to about 90 °C, most preferably at about 80°C.
33. The process of any one of the preceding claims, further comprising the step of recovering nickel sulfate from the clean nickel sulfate solution, preferably by crystallisation.
34. A process for producing high purity nickel suitable for use in battery manufacture or nickel plating, preferably having a purity of >99.8%, more preferably >99.98% nickel sulfate, comprising the steps of:

(i) providing a solution sub-saturated with nickel sulfate and comprising one or more non-nickel metal impurities, preferably a pregnant leach solution, more preferably obtainable by a process as defined in any one of claims 5 to 19;

(ii) removing the bulk of the one or more non-nickel metal impurities by raising the pH of the sub-saturated nickel sulfate solution by adding a basic nickel compound, preferably nickel hydroxide, to the sub-saturated nickel sulfate solution and separating the non-nickel metal impurities from the solution in the form of precipitated insoluble non-nickel metal hydroxides;

(iii) selectively removing remaining trace non-nickel metal impurities from the subsaturated nickel sulfate solution using one or more rounds of ion exchange purification involving at least a first round of ion exchange using a resin pre-loaded with nickel to form a first IX discharge solution comprising nickel sulfate and a reduced amount of non-nickel metal

WO 2019/090389

PCT/AU2018/051203 impurities which is a clean nickel sulfate solution being a substantially non-nickel metal impurities free sub-saturated nickel sulfate solution;

(iv) recovering high purity nickel sulfate from the substantially non-nickel metal impurities free sub-saturated nickel sulfate solution, preferably in the form of alpha nickel sulfate hexahydrate.
35. The process of claim 34, further comprising the step of oxidising non-nickel metal impurities in the sub-saturated nickel sulfate solution, preferably prior to raising the pH of the sub-saturated nickel sulfate solution, preferably a pH of about 3 to about 7, preferably to a pH of about 4.5 to about 5.5, more preferably to a pH of about >5, and separating the precipitate to form a non-nickel metal hydroxide-depleted sub-saturated nickel sulfate solution.
36. The process of claim 34 or 35, wherein the step of selectively removing the trace amounts of non-nickel metal impurities involves buffering the sub-s-aturated nickel sulfate solution prior to ion exchange (IX) with one or more basic nickel compounds to a pH optimised for selectivity and stability of non-nickel metal impurity loading onto a nickel preloaded ion exchange resin.
37. The process of any one of claims 34 to 36, wherein the recovering step involves crystallising the non-nickel metal hydroxide-depleted sub-saturated nickel sulfate solution to form high purity nickel sulfate crystals, preferably wherein the crystallising step is carried out at a temperature of from about 45 °C to about 65 °C, preferably from about 50 °C to about 60 °C, most preferably at about 53°C and/or under a pressure of about 10 kPa absolute, preferably forming crystalline alpha nickel sulfate hexahydrate.
38. The process of any one of claims 34 to 37, wherein the sub-saturated nickel sulfate solution is an acidic pregnant leach solution, for example, obtainable from a nickel sulfate leaching process as defined in any one of claims 5 to 19.
39. The process of any one of claims 34 to 38, wherein the selectively removal step (iii) is an ion exchange process as defined in any one of claims 19 to 34.
40. The process of claim 37, further comprising the step of drying the crystals to provide dry high purity alpha nickel sulfate hexahydrate, for example, in a fluid bed dryer.
41. A plant for producing high purity nickel sulfate, preferably having a purify of >99.8%, more preferably >99.98% nickel sulfate, from a nickel sulfate solution comprising:

(i) an acidic nickel leach module for generating sub-saturated pregnant leach solution comprising nickel sulfate and non-nickel metal impurities, preferably configured for batch leaching;

(ii) downstream of the acidic nickel leach module, a bulk non-nickel metal impurity removal module for precipitating the bulk of the non-nickel metal impurities, preferably configured to oxidise oxidisable non-nickel metal impurities in the pregnant leach solution;

WO 2019/090389

PCT/AU2018/051203 (iii) downstream of the bulk non-nickel metal impurity removal module, a trace non-nickel metal impurity ion exchange removal module for removing trace amounts of nonnickel metal impurity to form a purified sub-saturated nickel sulfate solution substantially free of non-nickel metal impurity, preferably configured for one or more rounds of ion exchange; and optionally, (iv) downstream of the trace non-nickel metal impurity ion exchange removal module, a nickel sulfate crystallisation module for crystallisation of high purity nickel sulfate crystals from the purified sub-saturated nickel sulfate solution substantially free of non-nickel metal impurity, preferably configured to crystallise alpha-nickel sulfate hexahydrate,
42. The plant of claim 41, further comprising a nickel hydroxide formation module for preparation of nickel hydroxide for use as an acid neutraliser and/or as a pH buffer in the production of the high purity nickel sulfate, wherein the nickel hydroxide formation module is in communication with the bulk non-nickel metal impurity removal module and/or the trace non-nickel metal impurity ion exchange removal module for providing nickel hydroxide solution thereto.
43. The plant of claim 41 or claim 42, wherein the nickel hydroxide formation module comprises a closed circuit for forming a nickel hydroxide solution such that ammonia and/or NaOH used for neutralisation and pH adjustment during nickel hydroxide preparation is isolated from the bulk non-nickel metal impurity removal module and/or the trace non-nickel metal impurity ion exchange removal module thereby avoiding contamination of nickel sulfate solutions.
44. The plant of any one of claims 41 to 43, wherein the bulk non-nickel metal impurity oxidiser module is in communication with a source of oxidant, for example, air, and a source of nickel hydroxide acid neutraliser and/or as a pH buffer preferably from the nickel hydroxide formation module of claim 46 or claim 47.
45. The plant of any one of claims 41 to 44, further comprising a hydrogen gas mitigation module associated with at least module (i) and/or module (ii) for management of hydrogen gas evolved during processing.
46. The plant of claim 45, wherein the hydrogen gas mitigation module comprises: a source of a steam flush for management of hydrogen evolved during the process; and/or a means for processing nickel leaching under a positive gauge pressure to prevent air ingress; and/or a means for providing a blanket of nitrogen gas over at least module (i) and/or module (ii).
47. The plant of any one of claims 41 to 46, configured (I) to provide sulfuric acid from a refinery to the plant, wherein the plant is configured to provide sulfuric acid to the acidic nickel leach module for leaching nickel sulfate from nickel powder and/or sulfuric acid ion exchange removal module for ion-exchange stripping during regeneration, and/or (ii) to recycle

WO 2019/090389

PCT/AU2018/051203 condensates from one or more plant modules, for example, heating coils in the leaching module, the product dryer module, and/or the crystallisation module for reuse.
48. The plant of any one of claims 41 to 47, further comprising: a nickel sulfate dewatering module for removing the bulk of solute from nickel sulfate crystals downstream of the nickel sulfate crystallisation module; and/or a nickel sulfate crystal dryer module to dry the crystals to a high purity nickel sulfate hexahydrate product; and/or a high purity nickel sulfate product packaging module; and/or one or more off-gas scrubbers, for example, which are associated with the acidic nickel leach module, the bulk non-nickel metal impurity oxidiser module and/or the bulk non-nickel metal impurity removal module for capture of gases and the nickel sulfate crystallisation module to capture nickel sulfate dust.
49. Use of nickel hydroxide as an acid neutraliser and/or a pH buffer in a process for preparing high purity nickel sulfate, preferably having a purity of >99.8%, more preferably >99.98% nickel sulfate suitable for use in battery manufacture or nickel plating including electroplating and electroless plating.
50. Use of nickel hydroxide as an acid neutraliser and/or a pH buffer in an ion exchange purification process for selective removal of non-nickel metal impurities from a nickel sulfate solution, preferable an acidic sub-saturated nickel sulfate solution, more preferably, an acidic sub-saturated nickel sulfate solution from which bulk non-nickel metal impurities have been removed.
51. Use of nickel hydroxide as an acid neutraliser and/or a pH buffer in nickel sulfate solution, preferable an acidic sub-saturated nickel sulfate solution to precipitate non-nickel metal impurities in as insoluble non-nickel metal impurities, for example, non-nickel metal hydroxides.
52. Use of a nickel pre-loaded ion exchange resin to selectively remover non-nickel metal impurities from a nickel sulfate solution, preferable an acidic sub-saturated nickel sulfate solution, more preferably, an acidic sub-saturated nickel sulfate solution from which bulk nonnickel metal impurities have been removed.
53. Use according to claim 52, wherein the resin is a phosphoric acid, a phosphonic acid or phosphinic acid resin.
54. Use of a nitrogen gas blanket to prevent formation of explosive air and hydrogen mixtures over an acidic nickel leach evolving hydrogen.
55. Use of a sub-saturated nickel sulfate solution in a process for the preparation of high purity nickel sulfate, preferably having a purity of >99.8%, more preferably >99.98% nickel sulfate, and preferably wherein the sub-saturated nickel sulfate solution is a pregnant leach solution, for example, obtainable by a process as defined in any one of claims 1 to 17.
56. High purity nickel sulfate obtainable by the process of any one of claims 1 to 40.

WO 2019/090389

PCT/AU2018/051203
57. Use of high purity nickel sulfate as defined in claim 56 in the manufacture of an energy storage device, for example, a battery or capacitor and/or in a nickel plating process including electroplating and electroless plating.
58. An energy storage device, for example, a battery or capacitor, comprising nickel

5 sulfate as defined in claim 56.
59. A product comprising plated nickel derived from nickel sulfate as defined in claim 56,
60. The process of any one of claims 1 to 40, wherein the subsaturated nickel sulfate solution is obtained from nickel powder, preferably having a purity of from 98% to 100%.