AU2018363879B2

AU2018363879B2 - Production of high purity nickel sulfate

Info

Publication number: AU2018363879B2
Application number: AU2018363879A
Authority: AU
Inventors: Richard Clout; John Stewart
Original assignee: BHP Billiton Nickel West Pty Ltd
Current assignee: BHP Nickel West Pty Ltd
Priority date: 2017-11-10
Filing date: 2018-11-08
Publication date: 2024-06-13
Anticipated expiration: 2038-11-08
Also published as: AU2018363879A1; 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

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 a first aspect, the invention provides a process for producing high purity 99.98% nickel sulfate, the process comprising the steps of:

selectively removing by ion exchange non-nickel metal impurities from an acidic sub saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities, which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100%, preferably 99.8%, using a nickel pre-loaded ion exchange resin which adsorbs non-nickel metal impurities from the solution to form a substantially non-nickel metal impurities free nickel sulfate solution; and buffering the nickel sulfate solution prior to ion exchange 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; and

recovering high purity nickel sulfate from the non-nickel metal impurities free nickel sulfate solution.

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

(i) leaching a stoichiometric excess of a nickel powder having a purity of from about 98% to about 100%, with sulfuric acid to form an acidic sub-saturated solution of dissolved nickel sulfate comprising from about 130 g/L to about 210 g/L nickel 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 sub saturated 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.

In a third aspect, the invention provides an ion exchange process for producing a high purity nickel sulfate solution suitable for crystallisation of 99.98% purity nickel sulfate hexahydrate, comprising the steps of:

providing an acidic sub-saturated solution of dissolved nickel sulfate comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel and one or more non-nickel metal impurities;

selectively removing the non-nickel metal impurities from the nickel sulfate solution by ion exchange by subjecting the nickel sulfate solution to an ion exchange step using a resin pre-loaded with nickel,

whereby non-nickel metal impurities in the nickel sulfate solution are selectively retained by the resin to generate a first ion exchange discharge solution which is a clean nickel sulfate solution substantially free of non-nickel metal,

wherein the selectively removing step includes the step of buffering the nickel sulfate solution prior to ion exchange 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 wherein the amount of basic nickel compounds provided is sufficient to neutralise acid released by the resin in exchange for nickel during nickel preloading.

In a fourth aspect, the invention provides a process for producing 99.98% purity nickel sulfate comprising the steps of:

providing an acidic sub-saturated solution of dissolved nickel sulfate comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel;

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 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; selectively removing remaining trace non-nickel metal impurities from the sub-saturated nickel sulfate solution by ion exchange purification involving an ion exchange step using a resin pre-loaded with nickel to form a first ion exchange 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, wherein the step of selectively removing the trace amounts of non-nickel metal impurities involves buffering the sub saturated nickel sulfate solution prior to ion exchange 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; recovering high purity nickel sulfate from the substantially non-nickel metal impurities free sub-saturated nickel sulfate solution by crystallisation. In a fifth aspect, the invention provides a plant for producing high purity 99.98% nickel sulfate comprising:

(i) an acidic nickel leach module for generating sub-saturated pregnant leach solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel ; (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 configured to oxidise oxidisable non-nickel metal impurities in the pregnant leach solution; (iii) downstream of the bulk non-nickel metal impurity removal module, a selective trace non-nickel metal impurity nickel-preloaded ion exchange removal module for removing trace amounts of non-nickel metal impurity to form a purified sub-saturated nickel sulfate solution substantially free of non-nickel metal impurity; and optionally, (iv) downstream of the selective 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; and

(v) 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, and 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. In a sixth aspect, the invention provides a use of one or more basic nickel compound solution as an acid neutralizer/pH buffer in a nickel pre-loaded ion exchange process to selectively remover non-nickel metal impurities from an acidic sub-saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities, which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel. In a seventh aspect, the invention provides a use of a nickel pre-loaded ion exchange resin with a basic nickel compound as an acid neutralizer/ pH buffer in a selective ion exchange process to selectively remove non-nickel metal impurities from an acidic sub-saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel in a process for preparing high purity 99.98% nickel sulfate. In an eighth aspect, the invention provides a use of a nitrogen gas blanket to prevent formation of explosive air and hydrogen mixtures over an acidic nickel leach evolving hydrogen, wherein the leach generates an acidic sub-saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel. In a ninth aspect, the invention provides a high purity 99.98% nickel sulfate obtained by the process of any one of the first to fourth aspects. In a tenth aspect, the invention provides a use of high purity 99.98% nickel sulfate obtained by the process of any one of the first to fourth aspects in the manufacture of an energy storage device and/or in a nickel plating process including electroplating and electroless plating. Described herein is 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 non-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 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. 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 sub saturated 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 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 sub-saturated 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-nickel 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; (iii) 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 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; (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 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

Also described herein is 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 sub saturated 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: Ni(s) + H2S0 4 aq) - H2(g) + NiSO4(aq)

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 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 production 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 be provided to the nickel sulfate processing plant directly from the discharge of wet metals dryers in the nickel powder production 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 embodiments, it is believed that coarser particles leach more slowly such that there is a preference for finer particles where faster leaching is desirable. The nickel powder raw material may have a purity of from about 98% to about 100%. Preferably the nickel powder 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 sub-saturated 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-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 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, 800C, 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 a solution 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 solution). Suitably, the nickel in the sulfuric acid solution has a pulp density, that is amount of solids in a pulp of nickel in acid, in the range of about500toabout 1500 g/L, more preferably from about 600 to about 1200 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 at 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 (ii), 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/2 L reactor volume, more preferably 1 - 5 L/min/2 L reactor volume, most preferably about 1 L/min/2 L reactor volume. Lower flow rates, for example, : 3.5 L/min/2 L reactor volume, 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 N 2 sparging. As hydrogen is evolved as a side product of the leaching step, 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 are complete, it may be beneficial to provide the steam flush and/or positive gauge pressure to further steps of the process, particularly, the filtering and bulk impurities removal steps also described herein. For these latter steps, 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 off gas 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 it conveniently leaves the unleached nickel in place for easy commencement of a subsequent leach batch where fresh acid is 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 leaching step has been allowed to progress for a 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 embodiments, 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. 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 190 g/ L or 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 plus/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 non-nickel 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 non-nickel 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 ferricwhereby 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. 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. 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 a pH of about 5 to about 5.5, with reasonable neutralisation kinetics and efficiency. This is achievable with sufficient neutralisation 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 lead 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 impurity-depleted discharge solution is filtered to remove any remaining solids prior to storage in the ion exchange feed tank and/or prior to ion exchange. Suitably, the non-nickel metal impurity-depleted discharge solution is held in the ion exchange feed tank prior to the ion exchange processing. Suitably, filter sludge recovered from filtering is preferably transferred to a refinery for further processing.

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 with 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 Described herein is 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 by subjecting a solution of nickel sulfate comprising dissolved nickel sulfate and one or more non-nickel metal impurities to an ion exchange step using a resin pre-loaded with nickel, whereby non-nickel metal impurities in the nickel solution are selectively retained by the resin to generate a first ion exchange 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 of the invention; (ii) selectively removing non-nickel metal impurities from a nickel sulfate solution by ion exchange by subjecting a solution of nickel sulfate, comprising dissolved nickel sulfate and one or more non-nickel metal impurities to an ion exchange step using a resin pre-loaded with nickel, (iii) whereby non-nickel metal impurities in the nickel solution are selectively retained by the resin to generate a first ion exchange 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 or about 192 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 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 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 pre-loaded 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-ion exchange and post-ion exchange 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 determinable as shown in the examples provided herein. Preferably, prior to ion exchange, the nickel sulfate solution has an initial pH of pH 5 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 5 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 ion exchange step is 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 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 operating 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 ion exchange 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 over time such that the equilibrium pH tends to be less than the initial pH. In some embodiments, 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 used in the ion exchange step 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 with time. Using nickel hydroxide as base/buffer advantageously avoids pH decrease during ion exchange, 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 (pH < 6); and providing for minimum washing requirements.

Preloading feed solution & pH control Suitably, the resin used in the ion exchange step is preloaded with nickel using a nickel pre-loading feed solution, which is preferably a portion of the first ion exchange 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 ion exchange 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 their presence may lead to contamination in the final nickel sulfate product. In a preferred embodiment, the pre-loading step involves introducing a clean nickel sulfate solution substantially free of non-nickel metal impurities. In other embodiments, the resin may be preloaded using an alternative source of filtered clean nickel sulfate solution. Preferably, the process involves the step of filtering the first ion exchange 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 pre loading feed, preferably a maximum strength clean 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 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 or about 192 g/L.

Nickel Hydroxide Desirably a portion of the first ion exchange discharge solution is used to generate nickel hydroxide and/or to preload the resin of the ion exchange step with nickel. In a preferred embodiment, the nickel hydroxide base is generated from a portion of the ion exchange purified clean nickel sulfate solution, which is the first ion exchange discharge solution. Suitably, the clean nickel sulfate solution, substantially free of non-nickel metal impurities may be a portion of the first ion exchange 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 an 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 all 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 ion exchange discharge solution. In a preferred embodiment, cobalt breakthrough in the ion exchange column discharge solution can signify that resin regeneration or replacement is required. Desirably, the method further comprises the step of replacing or regenerating the resin used in the ion exchange step. 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 ion exchange 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. 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 occurs 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 ion exchange discharge solution. It should be understood that the clean nickel sulfate solution and the first ion exchange discharge solution described herein is an acidic sub-saturated nickel sulfate solution, preferably obtained 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. 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.

Crvstallisation In particularly preferred embodiments, the clean nickel sulfate solution preferably being the first ion exchange 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 example, 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. Described herein is 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; (iii) selectively removing remaining trace non-nickel metal impurities from the sub-saturated nickel sulfate solution by ion exchange purification involving an ion exchange step using a resin pre-loaded with nickel to form a first ion exchange 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. 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; (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 sub-saturated nickel sulfate solution by ion exchange purification involving an ion exchange step involving a resin pre-loaded with nickel to form a first ion exchange 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. Preferably, the process further comprises the optional step of oxidising non-nickel metal impurities in the sub-saturated nickel sulfate solution during 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, recovery 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 electroplating. 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 (iii) 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 reduced pressure of about 10 kPa absolute, preferably forming alpha crystalline nickel sulfate hexahydrate. Suitably, the crystallising step is carried out in an evaporative crystalliser, for example, a mechanical vapour recompression type crystalliser, preferably a draft tube baffle crystalliser. 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 Described herein is 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; (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 non-nickel metal impurity to form a purified sub-saturated nickel sulfate solution substantially free of non-nickel metal impurity, preferably configured for an ion exchange step in lead/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 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 off gas 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. 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-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. 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 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. 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 for a 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 sub-saturated 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 for a 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-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 according to the invention. 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 purity 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 in 1 L solution in 2 L vessels; (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 tests were operated for a period of 10 hours; (B) Impact of Pulp density on leaching rate at pH 1, 80°C; (C) Comparison of Nickel concentration for tests operated under anaerobic conditions at 290 g/L sulfuric acid, 800C; (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 in 1 L solution in a 2 L vessel; (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 tests, 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 in 1 L solution in a 2 L vessel; (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 N 2 sparged test; (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 in 1 L solution in a 2 L vessel; (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 in 1 L solution in 2 L vessels; (A) Impact of temperature on nickel powder dissolution at pH 1.0, 1000 g/L nickel and 5 L/min aeration. Red data points represent tests undertaken using uncoated impellors; (B) Effect of Temperature on Leaching Rate; Figure 8(A) - (G) illustrate the results of bench scale experiments relating to batch acid leaching in 1 L solution in a 2 L vessel; (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 tests at 80 and 100 °C; (C) Nickel concentration over time 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 ion exchange process of the invention and results of bench scale experiments relating to the effect of preloading of resin on ion exchange; (A) schematic of the ion exchange 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 0C;

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 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 pre-loaded 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 (H 2 SO4 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; 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 A/ 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 ion exchange - 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 ion exchange ion exchange - 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 Cyanex 272; (A) Metal extraction pH isotherms with 10% Cyanex 272 in Shellsol D70 at 80 °C and 1:1 A/O ratio; (B) Metal extraction in organic phase with 10% Cyanex 272 in Shellsol D70 at 800C and 1:1 NO ratio; (C) Metals in raffinate vs. pH with 10% Cyanex 272 in Shellsol D70 at 80 °C and 1:1 NO 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 29 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 (H 2 SO4 concentrations).

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 Overalldescription 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 nickelleach module A. Nickelleach 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 refinery. The tanks are adapted to include a source of steam for heating, for example, indirect steam heating via heating coils or jackets 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 nickelleach 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 are 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, a solution filter sludge tank (not shown), and transfer pump (not shown). Module B further comprises an ion exchange feed tank (not shown) for holding filtered non-nickel metal impurity depleted discharge solution 250 before ion exchange.

Ion exchange module C

This module comprises one or more ion exchange columns 270 which ideally are arranged in a lead/lag configuration. A third column (not shown) may be provided for off-line regeneration, which includes washing, stripping, and pre-loading. A 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 ion exchange 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 ion exchange columns 270. A crystalliser plant feed tank (not shown) holds filtered clean nickel sulfate solution 420 and is in fluid communication with the crystalliser 450 of a crystalliser plant E. The fluid line between the one or more ion exchange discharge polishing filters 400 and the crystalliser 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 ion exchange column 270 regeneration is required. The ion exchange 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 crystalliser 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 leach 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 transferof nickel powder 100 tothetanks 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. Typicallyonly approximately 10% of the nickel powderin the tanks 110 reacts during the batch cycle. The reaction proceeds are per the equation below.

Ni(s) + H 2S0 4 (aq) - H2(g)+ NiSO4(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 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 80C minimises the evaporation duty of the crystalliser 450 in crystallisation module E, and eliminates the need for a 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 800C. 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 vol% steam atmosphere in the vapour space about the tanks 110. The resultant hydrogen containing off-gas 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-leached solids remain in the heel of the leach tanks 110 and are re-processed in the next batch of the leach cycle. An exemplary leach batch cycle is shown below in Table 1: Table I Step by Step Activity Value Units STEP 1 :Load Acid/Demin Mix into Tank 45 minutes STEP 2 :Load Nickel into Tank 20 minutes STEP 3:Leach 24 hours STEP 4: Settling 15 minutes STEP 5: Decant Liquor 45 minutes Technical time (hold time) 115 minutes Total Batch Leach Cycle Time 28.0 hours 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 ion exchange feed tank (not shown) to await ion exchange. 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 ion exchange feed tank (not shown) is fed to one or more ion exchange 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 ion exchange columns 270 can be arranged in a lead/lag configuration, with the filtered non-nickel metal impurity depleted discharge solution 250 entering an ion exchange column 270 with partially loaded resin first (lead) before proceeding to an ion exchange 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 ion exchange 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 ion exchange columns 270 which can be removed from the filters as sludge 440. The filtered clean nickel sulfate solution 420 is then held in a crystalliser plant feed tank (not shown) before being pumped to the crystalliser plant E. A small portion of the filtered clean nickel sulfate solution 420, diverted filtered clean nickel sulfate solution 430 is transferred to a nickelhydroxide preparation module B; or to an ion exchange nickel preload tank (not shown) to be held until ion exchange 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 ion exchange 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 ion exchange waste 330 from the washing and other regeneration reagents may be collected in an ion exchange 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 is prepared in a mixing tank (not shown) and pumped for several bed volumes through the ion exchange 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 ion exchange waste 330 is sent to the ion exchange waste collection tank (not shown). An additional wash stage of several bed volumes of demineralised water 280 is then used to wash excess sulfuric acid from the column and is also sent to the ion exchange 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 ion exchange 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 ion exchange 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 crystallisermodule 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 this 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 generated in the crystalliser 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 crystalliser 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 crystalliser 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 crystalliser 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 pre-load recirculation tanks 340.

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 ion exchange 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) + H2SO4(aq) - Ni2+(aq) + H 2 ( 9)+SO4 2 -(aq) (1) Ni(s) + H2SO4(aq) + 1/2 02- Ni2+(aq) + S042-(aq) + H 20 (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 H 2 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 impellers 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 Test Nickel Aeration T No. pH/Acid load (mL, g) powder rate (°C) mass (g) (L/min) 1 1.0 600 5 80 2 1.0 800 5 1 80 3 1.0 1000 5 80 4 1.0 1200 5 80 5 1.0 700 5 80 6 1.0 1000 5 180 7 0 800 5 80 8 0 1000 5 1 80 9 1.0 1000 2(N2gas) 80 10 2.0 1000 5 1 80 11 3.0 1000 5 80 12 248 mL (456.3 g) H2SO4 made up to 1 L with tap water 1000 5 80 13 1.0 1000 10 80 14 1.0 1000 2.5 1 80 15 1.0 1000 1.0 80 16 90 mL (165.6 g) H2SO4 made up to 1 L with tap water 1000 5 80 17 100 mL (184.0 g) H2SO4 made up to 1 L with tap water 1000 5 80 18 130 mL (239.2 g) H2SO4 made up to 1 L with tap water 1000 5 80 19 150 mL (276.0 g) H2SO4 made up to 1 L with tap water 1000 5 80 20 1.0 1000 5 15 21 1.0 1000 5 40 22 1.0 1000 5 1 60 23 1.0 1000 5 100 24 90 mL (165.6 g) H2SO4 made up to 1 L with tap water 1000 5 80 25 1.0 1000 5 80 26 1.0 1000 5 | 100 27 150 mL (276.0 g) H2SO4 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 1000 1 80 29 180 mL (331.2 g) plant H2SO4 made up to 1 L with plant water 1000 1 80 30 200 mL (368.0 g) plant H2SO4 made up to 1 L with plant water 1000 1 80 31 150 mL (276.0 g) plant H2SO4 made up to 1 L with plant water 1000 1 80 32 128.8 mL (237.0 g) of plant H2SO4 made up to 1 L with plant water 1000 1 80

33 139.1 mL (256.0 g) of plant H2SO4 made up to 1 L with plant water 1000 1 80 34 149.5 mL (275.0 g) of plant H2SO4 made up to 1 L with plant water 1000 1 80 35 159.2 mL (293.0 g) of plant H2SO4 made up to 1 L with plant water 1000 1 80

Materials and methods 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 I - impact of pulp density on nickelpowder dissolution Impact of pulp density under aerobic conditions Test work was conducted in 2 L glass agitated vessels which were maintained at 800 C for 10 hours and pH controlled using sulfuric acid (98%) under aeration. The pulp density 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 800-1000 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 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). Itisunclearwhythe1000 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 acid loading Acid 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; R 2 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 impellor 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. 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 air flow 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, 106.7and 90.3gof nickel dissolved, respectively (Figure5A). Test9, operatedwith 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 air flow 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 air flow 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 impellors 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 build-up 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. 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 (0.2-1.5 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 nickelpowder 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 7A 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 1000 C. It should be noted that at elevated temperatures the evaporation rate also increased which may have biased the 1000 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. It 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 NiFe 20 4 .Iron corrosion may be a source of this. Example 4 - batch acid loading: impact on nickel dissolutionand solution neutralisation 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 8B). 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 dissolutionand solutionneutralisation under near optimal conditions A set of three tests, viz. tests 28 (150 mls, 276 g acid), 29 (180 mls, 331.2 g acid) & 30 (200 mls, 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 mls (276 g acid), 180 mls (331.2 g acid) and 200 mls (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 period of 8 hours (Figure 8D). In general it 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. Cr and 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(II) and Fe(III) 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(II) to Fe(III). Fe(III) typically starts to form precipitates at pH above 3 in sulfuric acid. Fe(II) 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 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 impeller agitation rate under acid loaded anaerobic conditions The impact of impellor agitation rate was assessed (under anaerobic, acid loaded conditions (290 g/L). Impellor 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 11A) 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 impellor agitation speed within the range tested had minimal impact on nickel powder dissolution. Example 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 nickelpowder 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 80°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. Evaporation rates increased with respect to aeration rate and temperature. At a temperature of 80 0C, 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 purificationof 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 (Cyanex 272). 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. 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 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 Cobalt impurity loading distribution isotherms were prepared using air-dried Ni pre-loaded 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 tests were carried out with air-dried cobalt loaded resin and 1.5 M H 2 SO4 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. 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 over time. The tests were conducted in a 500 mL Schott bottle in an incubator shaker at 80 0C. 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). 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 water jacket condensers. Hot water from water 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 ion exchange 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 (-2BV 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 - nickelhydroxide 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.

Table 4 Conditions for preparation of nickel hydroxide. Feed material Nickel sulphate (ILR) 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 <0.2 <1 92.7 <0.02 -7.3 *Total 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 - nickelpre-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 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. A typical 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 preloading 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 at 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 over time 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 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 - nickelhydroxide 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 ion exchange 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 = MA 2 + 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). It is known that the affinity of Co complexation to the organic phosphinic acid extractant, e.g. Cyanex 272, 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 Neutralisation Ni (g/L) Resin sample Strip liquor Ni Ni loading Ni PLS at pH 5 240.2 (g) (g/L) (mg/g) Ni Raf at pH 5 Ni(OH)2 189.4 0.92 3.847 17.73 Ni PLS at pH 6 192.9 Ni Raf at pH 6 NH4 0H 122.1 1.27 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 for mass 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 Cyanex 272. 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 re-equilibrium 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 re-equilibrium 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 2+: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-loadedresin 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 Final equilibrium pH pH Ni pre-loaded resin at pH 5 with Ni pre-loaded resin at pH 6 with Ni(OH) 2 NH 40H 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 , Cu 2 + and C02+), by metal ion complexes (e.g. bisulphate), and by metal ions including Niz 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 in resin micro pore structure, which may contribute to the net change of solution pH, after re-equilibrium. 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(NH 3 )n2+] (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 NH 3 into solution to cause an increased pH through the equilibrium reaction: NH 3 + H 2 0 = NH 4 + + 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-equilibrium. 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 pre-loaded 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 pre-loading and high temperature associated variations. C: elution kinetics Metal elution kinetics studies were performed using 1.5 M H 2 SO4 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 H 2 SO4 and 1.5 M H 2 SO 4 , respectively. Near complete stripping of Cu occurred at 2M H 2 SO 4 .

As little Fe was loaded, the stripping of Fe could not assessed. From solvent extraction experience using Cyanex 272 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 internal 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 mL of 2 M H 2 SO 4 at 80 0C

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.

Table 10 Feed and raffinate data for the distribution isotherm tests. ID Total resin R/S Concentration (g/L) Loading based on Aq (%) (g) (g/mL) Ni Co Cu Fe Ni Co Cu Fe Feed 0.00 0.000 168.9 0.154 0.0039 0.0141 DI-1 0.10 0.002 170.9 0.147 0.0018 0.0004 -1.22 4.52 54.23 DI-2 0.25 0.005 169.6 0.119 0.0029 0.0109 -0.42 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 67.10 64.00 30.85 DI-5 1.00 0.020 172.4 0.035 0.0015 0.0122 -2.08 77.20 61.07 13.42 DI-6 1.25 0.025 162.7 0.029 0.0013 0.0108 3.63 81.04 67.28 23.03 DI-7 2.50 0.050 162.5 0.016 0.0009 0.0094 3.78 89.67 78.15 33.07 DI-8 3.33 0.067 166.9 0.012 0.0007 0.0088 1.16 92.47 82.83 37.70 DI-9 5.00 0.100 163.9 0.009 0.0007 0.0088 2.93 94.31 81.62 37.59 DI-10 6.25 0.125 169.9 0.007 0.0007 0.0110 -0.62 95.50 81.00 21.83 DI-11| 8.33 0.167 169.1 0.006 0.0007 0.0085 -0.17 96.25 81.72 39.30 DI-12 12.50 0.250 162.4 0.004 0.0005 0.0076 3.82 97.69 86.99 46.03

Table 11 Metal loading on air dry basis (original assay data in blue) ID Total R/S Resin Resin sample strip liquor (g/L) Loading (mg/g) on air dry resin samples in 10 mLof 2 M H2SO4 basis (g) (g/mL) (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 0.185 0.0028 0.0015 33.16 6.99 0.10 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 3.178 0.097 0.0016 0.0018 31.72 0.97 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 DI-12 12.50 0.250 1.006 2.737 0.049 0.0008 0.0010 27.21 0.49 0.01 0.01

Table 12 %Metal loading on air dry resin and %mass balance (out/In) ID R/S ratio|Loading (%) based on both resin and Aq assays Mass Balance (Out/In, %) 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|78.38 DI-4 0.013 0.06 65.86 56.73 8.77 92.34|96.38|83.19|75.80 DI-5 0.020 0.10 76.26 57.70 14.21 102.1|96.05|92.03|100.9 DI-6 0.025 0.35 78.98 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 approached maximum at about 7.5 mg/g. The raffinate Co decreased consistently 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 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 ID Total R/A Resin Loading (mg/g resin) Loading Loading (g/L resin) based on (Ni+Co) resin ratio samples on dry basis ratio Bulk Density (530 g/L) loading (w/w) (g) (g) 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 0.21 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 3 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 DI-6 0.86 0.017 0.688 59.0 6.42 0.11 0.07 9 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 Dl- 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 10 Dl- 5.71 0.114 0.688 46.4 1.12 0.02 0.04 42 24.6 0.59 0.01 0.02 25.2 11 Dl- 8.56 0.171 0.689 39.7 0.71 0.01 0.01 56 21.1 0.38 0.01 0.01 21.4 12 AVE 26.8 SID 3.5

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 + InKf (5) where KF 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 points as the linear part, and about 13.2 mg/g by taking 8 points as the linear part (Table 14 and Figure 17G).

Table 14 Data for Langmuir and Freundlich isotherms on estimated dry resin basis ID Total R/A ratio Langmuir X-Y data series Freundlich X-Y data resin series (g) (g/mL) Ce Ce/Qe Ln Ce In Qe (mg/L) (g/L) DI-1 0.07 0.001 147.2 13.92 4.99 2.36 DI-2 0.17 0.003 1193 11.68 4.78 2.32 DI-3 0.34 0.007 71.8 6.38 4.27 2.42 DI-4 0.46 0.009 50.7 4.76 3.93 2.37 DI-5 0.69 0.014 35.2 4.26 3.56 2.11 DI-6 0.86 0.017 29.2 4.56 3.38 1.86 DI-7 1.71 0.034 15.9 4.35 2.77 1.30 DI-8 2.28 0.046 11.6 3.98 2.45 1.07 DI-9 3.43 0.069 8.77 4.63 2.17 0.64 DI-10 4.28 0.086 6.94 4.90 1.94 0.35 DI-11 5.71 0.114 5.78 5.17 1.75 0.11 DI-12 8.56 0.171 3.56 5.02 1.27 -0.34 Slope 1/Qs = 0.0585 (6 points) R2 =0.97 1/n= 1.044 1/Qs = 0.057 (8 points) R2 = 0.94 R2 =0.99 Intercept 1/Qs-K = 1.113 (6 points) R2 = 0.97 LnKF =-1.642 1/Qs-K = 2.188 (8 points) R2 = 0.94 R 2 = 0.99

Variation of solution PH with metal loadings The solution pH rose sharply from initial 4.6 to above 5.3 when contacted with the pre loaded 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 stabilitytests Batch test work was conducted to assess the degradation of the resin. The vendor recommended maximum operating temperature for the TP 272 resin is 600C compared to the leach PLS temperature of 800C. 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 800C 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 over time. 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 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 Ni Pre- Synthetic Ni sulphate solution, eluted / washed 80 0C, pH 5 - 5.5 by 1 loading Lewatit TP 272 Ni(OH)2 Washing Ni loaded Lewatit TP 272, deionised water 80 °C, washing 3 times 0.5 Co-loading Ni loaded Lewatit TP 272, synthetic PLS 80 °C, -pH 5 20 Washing Co loaded Lewatit TP 272, deionised water 80 °C, washing 3 times 0.5 Elution Sulfuric acid, Co loaded Lewatit TP 272 80 °C, 2 M H2SO4 1 Washing Eluted Lewatit TP 272, Deionised water 80 °C, washing 3 times 0.5 S/L - 0.5 operations

Table 16 Conditions for determination of resin exchange capacity Conditions Set 1 Set 2 Resin Fresh resin samples as reference Fresh resin samples as reference Cycled samples from the stability tests Cycled samples from the stability tests Feed NaOH 196.1 mN 19.296 mN sin x g resin / 50 mL NaOH x g resin / 50 mL NaOH Temp 25 °C 25 0C Titer H2SO4 (104.3 mN) H2SO4(21.681 mN) Titer_ _ H2SO4 (5.677 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 H 2S0 4 solution. The total exchange capacity of fresh TP 272 was determined to be 1.21 ±0.02 (Table 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 - H 2 SO 4 Sample Resin H2 S0 4 [NaOH] A[NaOH] NaOH usage (mN) (mL) (g) (mL) (mN) (mN) (meq/g) NaOH Feed 19.296 Fresh 1-1 21.681 10 0.251 6.20 13.442 5.854 1.17 21.681 10 6.10 13.225 6.071 1.21 Fresh 1-2 21.681 20 0.502 6.55 7.101 12.195 1.21 21.681 10 3.20 6.938 12.358 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 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 H 2SO4titers 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 H 2 SO4 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) 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 Aqueous 20 g/L Co solution prepared from Co sulphate solution S/L ratio 2 g : 50 mL (loading three times with the fresh aqueous solution T(°C) 80 Initial solution pH 5.5 Sample stripping x g resin with 20 mL of 2 M H2 SO 4 at 800C

Table 19 Assay data of the resin samples Resin ID Resin Strip Strip Co loading Average Loss MB sample solution liquor Co (g) (mL) (g/L) (mg/g) (%) (mg/g) ( (

Fl: Fresh resin 1 1.660 20 2.259 27.2 1.82 90.60 F2: Fresh resin 2 1.316 20 1.693 25.7 1.72 26.47 93.84 U1: Used resin 1 2.038 20 1.211 11.9 0.79 91.57 (last cycle) U2: Used resin 2 1.935 20 1.170 12.1 0.81 11.99 54.71 94.91 (last cycle)

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 Loadinq 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 H 2 SO4 , 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 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/In) 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 0.47 114.4 0.02 %Mass balance (Out/In) 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 R/S ratio (g/L) 5:1 T (°C) 80 pH adjustment H 2SO 4 and NaOH solutions Sample stripping x g resin with 2 M H 2 SO 4 at 800C

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).

Table 22 Metal accountability data for the loading pH isotherms pH Mass balance (Out/In %) Ni Co Cu Fe 5.30 99 96 79 117 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 87 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 re-dissolution 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 (g/L) 3:1 T (°C) 80 pH adjustment H 2SO 4 and NaOH solutions Sample stripping x g resin with 2 M H 2 SO 4 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 Lewatit@ MonoPlus TP 207 Aqueous solution 50 mL Synthetic PLS (initial pH 4.5) R/S ratio (g/L) Various T (°C) 80 Mixing time (h) 1 Sample stripping x g resin with 2 M H 2 SO 4 at 800 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 H 2 SO4 for nearly complete stripping of Fe and Cu; 0.5 M H2 SO4 for > 95% Ni stripping; M H2 SO4 for -80% Co stripping, but it was uncertain if the remaining Co required 4 M H2 SO4or 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 H 2SO 4 solutions R/S ratio (g/mL) 1:30 T (°C) 80 Mixing time (h) 1 Sample stripping x g resin with 2 M H 2 SO 4 at 800C

Example 14 - Other resins / SX systems A: Ion Exchange - 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 T (°C) 80 Mixing time (h) 1 Sample stripping x g resin with 2 M H 2 SO 4 at 800C

B: Ion Exchange - 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 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 (g/mL) 25:90 T (°C) 80 Mixing time (h) 1 Sample stripping x g resin with 2 M H 2 SO 4 at 800C

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 H 2SO4and NaOH solutions. The organic samples were stripped with 2 M H 2 SO4 and the strip liquor and aqueous samples were assayed by ICP-AES. 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 elution-wash. Other two columns were similarly constructed and used for simulating polishing (lag) column, and for testing 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 Column Column 1 0 21.5 x 400 H Volume -145 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 I 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 4 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 8 108 5.15 190.7 CPLT1-9 9 120 5.04 191.2

Table 30 Ni loading (sampling on top of the columns)

Column ID BV TimepH Ni feed (g/L) Ni Loading (g/ g pHmifedn)L 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

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

Table 31 Conditions for Column I Co loading / washing and elution / washing Load Load-Wash Elution I Solid Ni Preloaded TP 272 Co Loaded TP 272 Washed TP 272 Aqueous Ni leach (batch 1 and 2) DI Water 0.1 M H2SO4 Column 145 mL/64.9 g air dry 145 mL/64.9 g air dry 145 mL/64.9 g air dry resin resin resin T(°C) 80 65-70 50 Flow rate (BV/hr) 2 5 2 Flow rate 12.5 5 (mL/min)

Elution 2 Elution 3 Elution Wash Solid Elution 1 TP 272 Elution 2 TP 272 Elution 3 TP 272 Aqueous 0.5 M H2SO4 1.5 M H2SO4 DI Water Column 145 mL/64.9 g air dry 145 mL/64.9 g air dry 145 mL/64.9 g air dry resin resin resin T(°C) 50 50 50 Flow rate (BV/hr) 2 2 5 Flow rate 12.5 (mL/min)

C: Column I 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 Time . Column 1 Column 2 Column 3 ID BV (min) pH Ni 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 3 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 7 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 I - 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. 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 theICP-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. Table 33 Assay results (Co and Cu by ICP-MS) by Bureau Veritas for samples from Column 3 - Polishing Runs with Column I raffinate as feed Run Sample Assay by Bureau Veritas Assay data x 4 (4 x dil.) Ni Co MS Cu MS Fe S Ni Co Cu Fe S UNITS mg/L mg/L mg/L mg/L mg/L g/L mg/L mg/L mg/L g/L T11-0 182 49200 1.94 0.1 1 27800 196.8 7.76 0.4 4 111.2 T11-4 186 45300 0.38 0.09 -1 25300 181.2 1.52 0.36 -4 101.2 BV/h T11-7 189 44100 0.45 0.12 1 24600 176.4 1.8 0.48 4 98.4 T11-8 190 45100 0.29 0.08 -1 25000 180.4 1.16 0.32 -4 100 T11-11 193 44900 0.58 0.15 -1 25300 179.6 2.32 0.6 -4 101.2 T17-0 196 44100 0.67 0.15 -1 24700 176.4 2.68 0.6 -4 98.8 T17-0 196 Rpt 44600 0.71 0.15 1 24800 178.4 2.84 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 48900 0.41 0.92 2 27400 195.6 1.64 3.68 8 109.6 2 T17-10 206 46900 0.45 0.11 -1 26200 187.6 1.8 0.44 -4 104.8 BV/h T1 7- 210 51500 0.43 0.91 -1 28500 206 1.72 3.64 -4 114 Raf T20-1 213 45200 0.42 0.12 -1 25000 180.8 1.68 0.48 -4 100 T20-5 217 43300 0.73 0.1 2 24200 173.2 2.92 0.4 8 96.8 T20- 219 47000 0.51 0.18 -1 26600 188 2.04 0.72 -4 106.4 Raf T21-0 221 36900 0.37 0.11 -1 20500 147.6 1.48 0.44 -4 82 0.8 T21-2 223 43300 0.41 0.11 -1 23700 173.2 1.64 0.44 -4 94.8 BV/h T21-7 227 35500 0.32 0.09 -1 19600 142 1.28 0.36 -4 78.4 T21-7 227 Rpt 36600 0.31 0.1 1 20800 146.4 1.24 0.4 4 83.2

F: Column I Load washing See Table 34 and Figure 29.

Table 34 Column I 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 2.0 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 72 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 65 1.323 2.068 0.0002 0.0002

Table 35 Column I 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 MH2SO4 0.5 MH2SO4 1.5 MH2SO4 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 byICP-AES Elution Elto(IID BV Time Ni Co Cu Fe V min) (g/L) (g/L) (mg/L) (mg/L) 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 Elution Run 1 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 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 8.0 240 0.005 0.000 0.200 13.00 Elution Run 2 T25-4 9.0 270 0.007 0.001 0.200 161.1 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 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 Elution Run 3 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

Table 37 Resin sample stripping data

Resin Strip Resmn Mas Sol Strip liquor (g/L) Loading ID sample s Volum by ICP-AES (mg/g air dried resin) e (g) (mL) Ni Co Cu Fe Ni Co Cu Fe Colum Loaded 0.21 0.58 0.24 0.00 0.00 26.8 11.4 0.2 0.3 n1 resin1 7 0 3 8 5 7 5 3 2 2 Colum Loaded 0.26 10 0.74 0.36 0.00 0.01 27.6 13.4 0.2 0.4 n1 resin2 8 0 0 7 1 1 2 4 2 27.2 12.4 0.2 0.3 Averaged loading 3 2 3 7 Eluted 0.32 10 0.00 0.00 0.00 0.01 0.09 0.03 0.0 0.4 T24 resin 1 3 1 0 3 1 2 Eluted 0.19 10 0.00 0.00 0.00 0.00 0.03 0.01 0.0 0.0 T25 resin 6 1 0 0 1 2 7 T26 Eluted 0.19 10 0.00 0.00 0.00 0.00 0.06 0.01 0.0 0.0 resin 6 1 0 0 0 2 1

Table 38 Elution efficiency based on resin strip liquor assay(%) Test ID Elution efficiency based on resin assays(%) Ni Co Cu Fe T24 -Run 1 (0.1 M H2SO4) 99.69 99.75 97.32 -12.32 T25 -Run 2 (0.5 M H2SO4) 99.90 99.92 92.01 82.39 T26 -Run 3 (1.5 M H2SO4) 99.80 99.92 91.71 97.25

G: Column I Elution See Tables 32 to 35 and Figure 30. Table 39 Column I elution washing data Tests ID BV Time pH Temp Concentration (g/L) (min) (°C) Ni Co Cu Fe Strip Wash T27-1 0.6 7 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 24 2.30 50 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 72 2.90 50 0.0119 0.0002 0.0003 0.0002 Strip Wash T27-8 7.0 84 2.80 50 0.0146 0.0002 0.0004 0.0002 Strip Wash T27-9 7.9 95 2.90 50 0.0006 0.0002 0.0004 0.0002 Strip Wash T27-10 9.0 108 3.00 50 0.0003 0.0002 0.0003 0.0002 Strip Wash T27-11 10.0 120 3.05 50 0.0006 0.0002 0.0003 0.0002

H: Column I Elution washing See Table 36 and Figure 31.

Claims

1. A process for producing high purity 99.98% nickel sulfate, the process comprising the steps of:

selectively removing by ion exchange non-nickel metal impurities from an acidic sub saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities, which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100%, using a nickel pre-loaded ion exchange resin which adsorbs non-nickel metal impurities from the solution to form a substantially non-nickel metal impurities free nickel sulfate solution; and buffering the nickel sulfate solution prior to ion exchange 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; and

2. The process of claim 1, further comprising the step of recovering the high purity nickel sulfate from the substantially non-nickel metal impurities free nickel sulfate solution in the form of crystalline alpha nickel sulfate hexahydrate.

3. The process of claim 1 or claim 2, 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 sulphate solution to form a non-nickel metal depleted nickel sulfate solution comprising a sub saturated concentration of nickel sulfate and trace amounts of one or more non-nickel metal impurities.

4. A process for leaching nickel sulfate from nickel powder comprising the steps of:

5. The process of claim 4, further comprising the steps of:

(i) precipitating substantially all of the non-nickel metal impurities as an insoluble non nickel metal impurity precipitate comprising non-nickel metal hydroxides; 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.

6. The process of claim 5, further comprising the step of oxidising the non-nickel metal impurities in the discharge solution prior to precipitation.

7. The process of claim 5 or 6, wherein the precipitation step involves increasing the pH of the discharge solution to a pH of about 3 to about 7.

8. The process of claim 7, wherein the pH of the discharge solution is increased by the addition of one or more basic compounds including nickel hydroxide.

9. The process of any one of claims 4 to 8, 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. 10. The process of any one of the preceding claims, wherein the nickel powder has an average particle size of from about 10 microns to about 500 microns. 11. 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. 12. The process of any one of claims 4 to 11, wherein leaching step (i) proceeds for a period of from about 2 to about 30 hours, and/or wherein the step (i) proceeds until the pregnant leach solution has a pH of from about 0 to about 4. 13. The process of any one of claims 4 to 12, wherein the concentration of sulfuric acid used in the solution of the leaching step (i) is initially from about 150 g/L and about 350 g/L, and/or wherein the nickel powder is provided/maintained at a loading of between about 850 g/L and 1200 g/L, (mass of nickel per litre of sulfuric acid solution); and/or wherein the nickel in the sulfuric acid solution has a pulp density in the range of from about 500 g/L to about 1500 g/L. 14. The process of any one of the preceding claims, wherein the process occurs at about atmospheric pressure or at a positive gauge pressure. 15. The process of any one of claims 5 to 14, wherein after non-nickel metal impurity precipitation, the discharge solution has a pH of from about 5; and/or wherein the discharge solution comprises about 192 g/L nickel. 16. An ion exchange process for producing a high purity nickel sulfate solution suitable for crystallisation of 99.98% nickel sulfate hexahydrate, comprising the steps of: providing an acidic sub-saturated solution of dissolved nickel sulfate comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel and one or more non-nickel metal impurities; selectively removing the non-nickel metal impurities from the nickel sulfate solution by ion exchange by subjecting the nickel sulfate solution to a first round of ion exchange using a resin pre-loaded with nickel, whereby non-nickel metal impurities in the nickel sulfate solution are selectively retained by the resin to generate a first ion exchange discharge solution which is a clean nickel sulfate solution substantially free of non-nickel metal, wherein the selectively removing step includes the step of buffering the nickel sulfate solution prior to ion exchange 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 wherein the amount of basic nickel compounds provided is sufficient to neutralise acid released by the resin in exchange for nickel during nickel preloading. 17. The process of claim 16, wherein the nickel sulfate solution has an initial pH of 5 6 prior to the first round of ion exchange. 18. The process of claim 16 or claim 17, wherein the resin of the first round of ion exchange is preloaded with nickel using a nickel pre-loading feed solution which is a portion of the first ion exchange discharge solution, or an alternative source of filtered clean nickel sulfate solution in the presence of nickel hydroxide for pH adjustment/buffering. 19. The process of claim 18, 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. 20. The process of any one of claims 16 to 19, further comprising the step of generating the nickel hydroxide for pH adjustment/buffering from a clean nickel sulfate solution which is a portion of the first ion exchange discharge solution or an alternative source of filtered clean nickel sulfate solution. 21. The process of any one of claims 16 to 20, further comprising the step of regenerating the resin used in the first round of ion exchange 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 ion exchange discharge solution. 22. The process of claim 21, 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 having a pH ofabout4 to about5;

(ii) stripping retained non-nickel metal impurities from the resin using an aqueous acid solution wherein the stripping step exchanges the retained non-nickel 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 water; (iv) pre-loading the resin with nickel by flushing the resin with a solution comprising a clean nickel sulfate solution together with nickel hydroxide. 23. The process of any one of claims 16 to 22, wherein the resin is provided in two or more operating columns arranged in a lead/lag configuration. 24. The process of any one of claims 16 to 23, wherein the resin is a cation exchange resin for chelating metal ions. 25. The process of any one of claims 16 to 24, wherein the ion exchange step is carried out at a temperature of from about 50 °C to about 95 °C.

26. The process of any one of the preceding claims, further comprising the step of recovering nickel sulfate by crystallisation. 27. A process for producing 99.98% purity nickel sulfate comprising the steps of:

(i) providing an acidic sub-saturated solution of dissolved nickel sulfate comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel; (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 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 sub saturated nickel sulfate solution by ion exchange purification involving an ion exchange step using a resin pre-loaded with nickel to form a first ion exchange 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, wherein the step of selectively removing the trace amounts of non-nickel metal impurities involves buffering the sub-saturated nickel sulfate solution prior to ion exchange 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; (iv) recovering high purity nickel sulfate from the substantially non-nickel metal impurities free sub-saturated nickel sulfate solution by crystallisation.

28. The process of claim 27, further comprising the step of oxidising non-nickel metal impurities in the sub-saturated nickel sulfate solution prior to raising the pH of the sub-saturated nickel sulfate solution to a pH of about 3 to about 7, and separating the precipitate to form a non-nickel metal hydroxide-depleted sub-saturated nickel sulfate solution. 29. The process of claim 27 or claim 28, wherein the recovering step involves crystallising the non-nickel metal hydroxide-depleted sub-saturated nickel sulfate solution to form alpha nickel sulfate hexahydrate. 30. The process of any one of claims 27 to 29, wherein the sub-saturated nickel sulfate solution is an acidic pregnant leach solution obtained from a nickel sulfate leaching process as defined in any one of claims 4 to 15. 31. The process of any one of claims 27 to 30, wherein the selectively removal step (iii) is an ion exchange process as defined in any one of claims 16 to 28. 32. The process of claim 31, further comprising the step of drying the crystals to provide dry high purity alpha nickel sulfate hexahydrate. 33. A plant for producing high purity 99.98% nickel sulfate comprising:

(i) an acidic nickel leach module for generating sub-saturated pregnant leach solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel; (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 configured to oxidise oxidisable non-nickel metal impurities in the pregnant leach solution; (iii) downstream of the bulk non-nickel metal impurity removal module, a selective trace non-nickel metal impurity nickel-preloaded ion exchange removal module for removing trace amounts of non-nickel metal impurity to form a purified sub-saturated nickel sulfate solution substantially free of non-nickel metal impurity; and optionally, (iv) downstream of the selective 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; and

(v) 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, and wherein the nickel hydroxide formation module comprises a closed circuit for forming a nickel hydroxide precipitate 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.

34. Use of one or more basic nickel compound solution as an acid neutralizer/pH buffer in a nickel pre-loaded ion exchange process to selectively remover non-nickel metal impurities from an acidic sub-saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities, which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel.

35. Use of a nickel pre-loaded ion exchange resin with a basic nickel compound as an acid neutralizer/ pH buffer in a selection ion exchange process to selectively remove non-nickel metal impurities from an acidic sub-saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non-nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel in a process for preparing high purity 99.98% nickel sulfate.

36. Use according to claim 35, wherein the nickel pre-loaded ion exchange resin is used in conjunction 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 wherein the amount of basic nickel compounds provided is sufficient to neutralise acid released by the resin in exchange for nickel during nickel preloading and wherein the basic nickel compound includes nickel hydroxide.

37. Use of a nitrogen gas blanket to prevent formation of explosive air and hydrogen mixtures over an acidic nickel leach evolving hydrogen, wherein the leach generates an acidic sub-saturated nickel sulfate solution comprising from about 130 g/L to about 210 g/L nickel and one or more non nickel metal impurities which is obtained from an acid leach of nickel powder having a purity of from about 98% to about 100% nickel.

38. High purity 99.98% nickel sulfate obtained by the process of any one of claims 1 to 32.

39. Use of high purity 99.98% nickel sulfate obtained by the process of any one of claims 1 to 32 in the manufacture of an energy storage device and/or in a nickel plating process including electroplating and electroless plating.

Drying Product Vent Gas Off and Bagging Dry Product

Off Gas System

560 540 530 520

500

Demineralised Centrifuging Water

Return Solids Recycle Centrate 480 490

470 510

Sulphate Nickel Demineralised Crystalliser

550 Water

450 460 Filter Polishing 420 Filter Discharge Sludge Demineralised Ion-Exchange Filter Sludge

Water

200 Air / Oxygen

260

400 440 430

220

lon Exchange

410 Off Gas 310 Waste Polishing Solution Vessels Aeration 210 Ion-Exchange

Filter

170 330 190 Hydrogen Vent 250 Solution Sulphate Nickel Scrubber

Leach

230 270 Adjustment pH Demineralised lon Exchange Preload Nickel 180 140 300 Demineralised Reagents

Water 320

Water

160

280 340 240 Hydroxide Nickel Leach Nickel 290

Preparation (Batch)

Waste Liquor Acid Sulphuric Nitrogen / Steam 390 380 Slurry Hydroxide Nickel 110

370 Powder Nickel 150 Demineralised Hydroxide Demineralised Sodium

130 Water

Water

360 Figure 1 120 100 350

0.01 0.1 1 10 100 1000 10000

Size (um)

Figure 2

160

140

120

100

80

60

40 pH 1.0

20 pH o

o 500 600 700 800 900 1000 1100 1200 1300 Pulp Density (g/L) (A)

Impact of Pulp density on leaching rate at pH 1, 80°C 140

120 600 g/L Ni 100 700 g/L Ni 80 1000 g/L Ni 60

40 1200 g/L Ni

20

0 0 2 4 6 8 10 12 Time (hrs) (B)

Substitute Sheet

(Rule26) RO/AU

800 g/L Ni 1000 g/L Ni 40 1300 g/L Ni 700 g/L Ni 20 900 g/L Ni 1000 g/L Ni (rpt) 0 0 5 10 15 20 25 30 Time (hours) (C)

7 800 g/L Ni 1000 g/L Ni 6 1300 g/L Ni 700 g/L Ni 900 g/L Ni 1000 g/L Ni (rpt) 5

4

3 H 2

1

o 0 5 10 15 20 25 30 Time (hours) (D)

100 90 80 70 60 Head 700 g/L 50 800 g/L 40 900 g/L 30 1000 g/L 20 1000 g/L (Rpt) 10 1300 g/L 0 0 50 100 150 200 250 300 Size (um) (E)

Substitute Sheet

(Rule26) RO/AU

Impact of Pulp density on Ni Extraction for Anerobic conditons with batch acid addition 180 160 140 120 100 80 60 40 20 0 700 800 900 1000 1100 1200 1300 Pulp Density (g Ni/L solution)

(F)

Impact of Pulp Density on Ni Extraction Rate for Anerobic conditions with batch acid addition 200 180 160 140 700 g/L Ni 120 100 900 g/L Ni

80 1000 g/L Ni 60 1000 g/L Ni rpt 40 20 1300 g/L Ni 0 0 5 10 15 20 25 30 Time (hr) (G)

Figure 3A - 3G

Response to pH 160.0

120.0

100.0

80.0

00.0

300 20.0

1.5.4 1 is 1.5

pH (A)

Substitute Sheet

(Rule26) RO/AU

Response to pH 160.0

100.0

120.0 y = 13.632In(x) + 142.34 100.0 R2 = 0.9855

AUU 60.0

500 200 use 0.00 1.00 1.20

conrenttation (H+) (B)

4.0

3.5 y = 0.9112x-0.0139 R2 = 0.9414

3.0

2.5

2.0

1.5

1.0

0.5

0.0 o 1 2 3 4 Total sulfuric acid added (mol) (C)

Impact of Leaching Time on pH 200 180 160 140 pH 0 120 100 pH 1

80 pH 2 60 pH 3 40 20 0 0 2 4 6 8 10 Time (hr)

(D)

Figure 4(a) - (d)

Substitute Sheet

(Rule26) RO/AU

O o 2 4 6 8 10 Aeration rate (L/min) (A)

35.0

30.0

25.0

20.0

15.0

10.0

5.0

0.0 o 2 4 6 8 10 12 Aeration rate (L/min compressed air) (B)

Effect of Aeration rate on Ni Extraction Rate 140

120

100

80

60

40

20

0 0 2 4 6 8 10 12

Time (hr)

2 L/min Nitrogen 1 L/min air 2.5 L/min air 10 L/min air (C)

Figure 5A - 5C

Substitute Sheet

(Rule26) RO/AU

237 g 256 g 6 275 g 293 g 5

4

H 3

2

1

o 0 5 10 15 20 25 30 Time (hours) (A)

350

300

250

200

150

100 237 g 256 g

50 275 g 293 g

o o 5 10 15 20 25 30 Time (hours) (B)

250 Ni 200

150

100

50 237 g H2SO4 256 g H2SO4 275 g H2SO4 293 g H2SO4 O O 5 10 15 20 25 30 Time (hours) (C)

Substitute Sheet

(Rule26) RO/AU

Co 150

100

50 237 g H2SO4 256 g H2SO4 275 g H2SO4 293 g H2S04 o o 5 10 15 20 25 30 Time (hours) (D)

2 237 g H2SO4 256 g H2S04 275 g H2SO4 293 g H2S04 Cu 1.5

1

0.5

o o 5 10 15 20 25 30 Time (hours) (E)

Figure 6A to 6E

140

120

100

80

60

40

20

o o 20 40 60 80 100 Temperature (C) (A)

Substitute Sheet

(Rule26) RO/AU

Effect of Temperature on Leaching Rate

180

160

140 15C 120

100 40C

80 60C

60 80C 40 100C 20

0 0 2 4 6 8 10 Time (hr)

(B)

Figure 7A & 7B

160

140

120

100

80

60

40 A 20

0 80 90 100 110 120 130 140 150 160

Total acid added (mL)

(A)

7

6 B

5

4 Test 24 -90 mL

3 Test 17 -100 ml

ID Test 18-130 ml 2

Test 19 -150 mL 1

Test 12 -248 mL 0 Test 27 -150 mL 100C -1

0 5 10 15 20 25

Time (hours) (B)

Substitute Sheet

(Rule26) RO/AU

A 150

100 Near optimal (150 mL acid)

Near optimal (180 mL acid) 50 Near optimal (200 mL acid)

0 0 5 10 15 20 25

Time (hours) (C)

7

6 0 Acidioad (150 mL) adidas Acid load (180 mL) 5

Acid load (200 mL) 4

3

2 1000,000 1

0

-1

0 5 10 15 20 25

Time (hours) (D)

Aerobic Batch Acid Addition Ni Extraction Rate 250

200

150 276.0 g acid

331.2 g acid 100 368.0 g acid

50

0 0 5 10 15 20 25

Time (hr)

(E)

Substitute Sheet

(Rule26) RO/AU

Impact of Acid loading on Nickel Extraction Rate 250 340 g acid

350 g acid 200 360 g acid

150 380 g acid

237 g acid 100 256 g acid

50 275 g acid

293 g acid 0 5 10 15 0 Time (hr)20 25 30 35 (F)

Batch Acid Addition pH Profile 7

6

5

4 14

3

2

1

0 0 5 10 15 20 25 30 35

Time (hr) 340 g acid 350 g acid 360 g acid 380 g acid

237 g acid 256 g acid 275 g acid 293 g acid

(G)

Figure 8A - 8G

90 70

Co Fe 80 60

70 50 60

50 40

40 30

30 150 ssit add load # 180 mi acid load 20 20 200 ssit acid load 150 ml acid lead 180 mL acid load # 10 10 200 mL acid load

0 0 0 10 20 30 0 10 20 30 Time (hours) (A) Time (hours) (B)

Substitute Sheet

(Rule26) RO/AU

30 25.00

Cr K 25 20.00

20 15.00

15 * 10.00 10

5 150 ml acid load 180 ml acid load 5.00 150 ml acid feed 130 ml acid lead

200 ml acid food 200 mLace load o 0.00 0 10 20 30 Time (hours) 0 10 20 30 (C) Time (hours) (D)

14

4.0 12 Na 3.5 Ca 10 3.0

2.5 8

2.0

$4 6 1.5

4 to# 1.0

180 mL acid load 150 ml acid load 180 ml acid feed 150 mLac load 0.5 2 200 ml JC load 200 mi acid load

0.0 0 D 10 20 30 0 10 20 30

Time (hours) (E) Time (hourc) (F)

90 450 150 ml acid load 180 not adid load @ 80 Cu 400 Zn 200 ml acid load

70 350

60 300 50 250 40 200 30 150 20 100 150 mi acid load 180 mi acid load

10 50 200 ml acid load

0 0 0 10 20 30 10 20 0 30 Time (hours) Time (hours) (G) (H)

Figure 9A - 9H

300

250

4 200

DR 1

150

2

100 1

50 o

-1

o o 1 2 3 4 5 7 8 9 10 Time Ihours

Figure 10

Substitute Sheet

(Rule26) RO/AU

50 % Agitation 75 % Agitation 40 20 100 % Agitation 150 % Agitation

0 o 5 10 15 20 25 30 Time (hours) (A)

7 50 % Agitation 6 75 % Agitation

5 100 % Agitation

150 % Agitation 4

3

2

1

0 0 5 10 15 20 25 30

Time (hours) (B)

Figure 11(A) & (B)

Base reagent Ni(OH)2 / NH 4 OH Co/Ni Effluent

Bleed Co/Ni Effluent Solution Preparation

Co/Ni Raffinate (Desired pHs)

Regenerated resin IX Ni Effluent IX Co (Pre-Loading) (NH1),SO Ni Effluent

To Crystallisation Ni-loaded resin

Washing Washing Effluent Washed Ni-Loaded Resin (A)

Substitute Sheet

(Rule26) RO/AU

5.5

pH isotherms with fresh TP 272 5.0 without pH control

4.5 Initial pH

Final pH 4.0

3.5

3.0

2.5 3.0 3.5 4.0 4.5 5.0

pH (B)

40

35 Ni pre-loading

30 Ni in solution

25

20

15

10

5

0 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0

pH (C)

70 Ni Pre-loading pH profile 60

22 g/L Ni 50 48 g/l Ni

100 g/L Ni 40 145 g/L Ni

30

20

10

0 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0

pH (D)

Substitute Sheet

(Rule26) RO/AU pH 5 with feed 220 g/L Ni pH 6 with feed 111 g/L Ni 5

0 0 2 4 6 810 12 14 16 18 20 22 2426 28 30

Time (min) (E)

Figure 12A to 12E

5.5

pH variation with fresh TP 272

5.0 without Ni pre-loading

4.5 Final pH

4.0

3.5

3.0

2.5 2.5 3.0 3.5 4.0 4.5 5.0 5.5

Initial pH (A)

100

90

80 Fresh TP 272 70 without Ni pre-loading

60

50 Ni 40

30 Co 20 Cu 10 Fe

0 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6

Final pH (B)

Substitute Sheet

(Rule26) RO/AU

8.0 2.0

Fresh TP 272 without 7.0 Ni 1.8 Ni pre-loading Co 6.0 1.6 Cu 5.0 Fe 1.4

Ni/Co 4.0 1.2

3.0 1.0

2.0 0.8

1.0 0,6

0.0 * 0.4 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6

Final pH (C)

40 1000 38 Metal loading using resin 36 Ni pre-loaded at pH 5 900 34 32 800 30 28 700 26 Ni 24 600 22 Co 20 500 18 Fe 16 400 14 Cu 12 Total 300 10 Initial Ni 8 200 6 SF(Co/Ni) 4 100 2 0 0 4.0 4.2 4.4 4.6 4.8 5.0 X 5.2 5.4 5.6 5.8

Final pH (D)

100 Metal loading using resin Ni pre-loaded at pH 5 90

80 Ni 70 Co 60 Fe 50 Cu

40

30

20

10

0 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 Final pH (E)

Substitute Sheet

(Rule26) RO/AU

100.000 Ni

Co Raffinate metals using resin Fe 10.000 Ni pre-loaded at pH 5 Cu

1.000

0.100

0.010

0.001 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 Final pH (F)

100.000

Metals in the neutralised synthetic PLS Ni 10.000 Co Fe 1.000 Cu

0.100

0.010

0.001 3.8 4.0 4.2 4.4 4.6 4.8 5,0 5.2 Initial pH (G) 36 900 34 Metal loading using resin 32 Ni pre-loaded at pH 6 800 30 28 700 26 Ni 24 600 22 Co 20 Fe 500 18 Cu 16 400 Total 14 Initial Ni 12 300 10 SF(Co/Ni) 8 200 6 4 100 2 0 o 4.0 4.2 4.4 4.6 4.8 5,0 5.2 Final pH (H)

Substitute Sheet

(Rule26) RO/AU

Metal loading using resin Ni pre-loaded at pH 6 90

80 Ni 70 Co 60 Fe

50 Cu

40

30

20

10

0 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6

Final pH (I)

100.000 Ni

Co 10.000 Fe Raffinate metals using resin Cu Ni pre-loaded at pH 6 1.000

0.100

0.010

0.001 4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 Final pH (J)

5.5 Metal loading with the syntetic PLS

using Ni preloaded resin without pH control 5.0

4.5

4.0 Using 18 mg Ni / g resin Pre-loaded at pH 5

Using 18 mg Ni / g resin Pre-loaded at pH 6

3.5 3.5 4.0 4.5 5.0 5.5 Initial pH (K)

Figure 13A to 13K

Substitute Sheet

(Rule26) RO/AU

8.0 100 20000 Initial 18 mg Ni / g resin loaded at pH 5

7.0 90 18000 Initial 20.5 mg Ni / g resin loaded at pH 6

80 16000 6.0

70 14000 5.0 Ni 60 12000

4.0 Co 50 10000 Cu 3.0 40 Fe 8000 CO SF(Co/Ni) 30 6000 2.0

Co loading pH profile 20 4000 1.0 EX pH isotherms 10 10% Cyanex 272 2000 0.0 3.5 4.0 4.5 5.0 5.5 0 0 1.5 2.0 2.5 3.0 3.5 4,0 4.5 5.0 5.5 6.0 Final synthetic PLS pH pH C D Figure 14A - 14D

0.06 6.0

0,05 5.0 Ni

0,04 Co 4.0 pH

0,03 3.0

0.02 Co loading kinetics 2.0 with initial pH 4.7 feed PLS

0,01 1.0

0,00 0.0 0 10 20 30 40 50 60

Time (min) (A)

100 6.0

90

80 5.0

Ni 70 Co 60 4,0 pH 50

40 3.0

30 Co loading kinetics with initial pH 4.7 feed PLS 20 2.0

10

0 1.0 0 10 20 30 40 50 60 Time (min) (B)

Substitute Sheet

(Rule26) RO/AU

70 Ni

60 Co

50

40 Ni and Co loading kinetics with initial feed pH 4.7 feed PLS 30

20

10

0 0 10 20 30 40 50 60 Time (min) (C)

100

90

80

70 Ni

60 Co Fe 50 Cu 40

Co strip kinetics 30 with 1.5 M H2SO4 20

10

0 o 10 20 30 40 50 60 Time (min) (D)

Figure 15A - 15D

100 Ni 90 Co 80 Cu 70 Fe 60

50

40

30 Stripping Profile 20 10 Y o 0.0 - 0.5 1.0 1.5 2.0

H2SO4 (M)

Figure 16

Substitute Sheet

(Rule26) RO/AU

Metal loading 80 based on both resin and Aq assays 4000

70 3500 Ni

60 Co 3000 Cu 50 Fe 2500 SF (Co/Ni) 40 2000

30 X 1500

20 1000

10 500

o o 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Resin / solution ratio (g/mL) (A)

100.0 80

70

10.0 60

50

1.0 40 Loading isotherms on air dry basis Ni 30 Co

0.1 Cu Fe 20 Ni/Co 10

0.0 o 0.00 0.05 0.10 0,15 0.20 0.25

Resin / Solution ratio (g/mL) (B)

1000000,0

100000.0

Ni 10000.0

Co 1000.0 Cu 100.0 Fe

10.0

1.0

0.1

0.00 0.05 0.10 0.15 0.20 0,25

Resin / Solution ratio (g/mL) (C)

Substitute Sheet

(Rule26) RO/AU

Co distribution isotherms

8

6

4 On air dry basis

2

0 0 20 40 60 80 100 120 140 160

Co in raffinate (mg/L) (D)

12

10 Co distribution isotherms

8

6

On air dry basis 4 On estimated dry basis

2

0 0 20 40 60 80 100 120 140 160

Co in raffinate (mg/L) (E)

14

y = 0.0858x + 1.1133 12 R2 0.9731 Fitting 1: Qs = 11.7 mg/g 10 (Linear part taking 6 points)

8

Langmuir Model: 6 Ce/Qe = 1/Qs.K Ce/Qs Where Qe is the amount adsorbed (mg/g), Ce is the equilibrium concentration of the 4 adsorbate ions (mg/L), Qs related to

maximum adsorption capacity (monolayer 2 capacity) (mg/g), K are Langmur constant

(L/mg).

0 0 20 40 60 80 100 120 140 160 180 200

Ce (mg/L) (F)

Substitute Sheet

(Rule26) RO/AU y = 0.0757x + 2.1882 12 R2 0.944 I

10 Fitting 2: Qs = 13.2 mg/g

(Linear part taking 8 points)

8

6 Langmuir Model: Ce/Qe = 1/Qs.K + Ce/Qs 4 Where Qe is the amount adsorbed (mg/g), Ce is the equilibrium

2 concentration of the adsorbate

0 0 20 40 60 80 100 120 140 160 180

Ce (mg/L)

G 4.0 y = 1.044x 1.642 3.5 R2 0.9938

3.0

2.5

2.0

1.5

Freundlich isotherm: 1.0 InQe = (1/n) InCe + Ink,

Where K, and n are Freundlich 0.5 constants related to adsorption

0.0 capacity and adsorption intensity, respectively.

-0.5

0 1 2 3 4 5 6 Ln Ce (mg/L) (H)

50 6.0

45 Loading isotherms 5.8

on air dry basis 40 5.6

35 5.4

30 5.2

25 5.0 Ni 20 4.8 Co 15 4.6 Cu 10 Fe 4.4

5 Equilibroum pH 4.2

0 4.0 0.00 0.05 0.10 0.15 0.20 0.25

Resin / Solution ratio (g/mL) (I)

Substitute Sheet

(Rule26) RO/AU

Figure 17A 171

5.0 100

4.5 EC-Set 1 (meq/g) 90 EC-Set 2 (meg/g) 4.0 80 Loss Set (%) Loss Set 2 (%) 3.5 70

3.0 60

2.5 50

2.0 40

1.5 30

1.0 20

0.5 10

0.0 0 CO C1 C203 C45 67891011 C12 C13 C14 C15 C16 C17 C18 C19 C20 Cycles

Figure 18

Ni Co Fe 100.00 Cu Total (mM/g) Resin (g)

10.00

Stability tests - metal loading

Lewatit TP 272 1.00

0.10 H *

0.01

C1 C2 C4 C6 C8 C10 C12 C14 C16 C18 C20 Cycles (A)

200 1.0

180 0.9

160 0.8

Feed Co (mg/L) 140 Raffinate Co (mg/L) 0.7

Resin mass (g) 120 0.6 Co loading (%) 100 Co loading (mg/L) 0.5

80 0.4 TP 272 Stability

60 Co behaviour 0.3

40 0.2

20 0.1

0.0 0 C1 C2 C4 C6 C8 C10 C12 C14 C16 C18 C20 Cycles (B)

Substitute Sheet

(Rule26) RO/AU

Feed Ni (g/L) 140 35 Raffinate Ni (g/L) 120 TP 272 Stability 30

Ni Behaviour Resin mass (g) 100 25 Ni loading (%) 80 20 Ni loading (mg/g) 60 15

40 10

20 5

0 0 C1 C2 C4 C6 C8 C10 C12 C14 C16 C18 C20 Cycles (C)

100 0.10

90 0.09 Feed Fe (mg/L) Raffinate Fe (mg/L) 0.08 80 Resin mass (g) 70 Fe loading (%) 0.07

Fe loading (mg/g) 60 0.06

50 0.05

40 0.04

TP 272 Stability 30 0.03 Fe behaviour

20 0.02

10 0.01

III 0 0.00 C1 C2 C4 C6 CB C10 C12 C14 C16 C18 C20 Cycles (D)

100 0.10

90 0.09

80 0.08

Feed Cu (mg/L) 70 0.07 TP 272 Stability Raffinate Cu (mg/L) Cu behaviour Resin mass (g) 60 0.06 Cu loading (%) 50 Cu loading (mg/g) 0.05

40 0.04

30 0.03

20 0.02

10 0.01

III 0 II 0.00 C1 # C2 C4 C6 C8 C10 C12 C14 = C16 C18 C20 Cycles (E)

Figure 19A to 19E

Substitute Sheet

(Rule26) RO/AU

Y Loading with Lewatit TP 207 Fe (Based on aqueous analysis) Cu 10 Ni

Co

1

0 1.5 1.8 2.0 2.3 2.5 2.8 3.0 3.3 3.5 3.8 4.0 4.3 4.5 4.8 5.0

pH (A) 100 Loading with Lewatit TP 207 90 (Based on aqueous analysis)

80 Fe 70 Cu 60 Ni

50 Co

40

30

20

10

0 1.5 1.8 2.0 2.3 2.5 2.8 3.0 3.3 3.5 3.8 4.0 4.3 4.5 4.8 5.0

pH (B)

100

90 Ni Loading pH isotherms 80 Lewatit TP 207 Co 70 Cu 60 Fe 50

40

30

20

10

0 1.0 1.5 2,0 2.5 3.0 3,5 4.0 4.5 5.0 5,5 6.0

Equilibrium pH (C)

Substitute Sheet

(Rule26) RO/AU

100.0

Loading with Lewatit TP 207

10.0

1.0 Fe

Cu Ni

Co 0.1

1.5 1.8 2.0 2.3 2.5 2.8 3.0 3.3 3.5 3.8 4.0 4.3 4.5 4.8 5.0 5.3 5.5

pH (D)

100.00

Ni

10.00 Loading pH isotherms Co Lewatit TP 207 Cu Fe 1.00

0.10

0.01 1.5 1.8 2.0 2.3 2.5 2.8 3.0 3.3 3.5 3.8 4.0 4.3 4.5 4.8 5.0 5.3 5.5

pH (E)

Figure 20A to 20E

1000 4.0

Kinetics with Lewatit TP 207

3.5

100 3.0

Ni Co Fe 2.5

Cu pH 10 2.0

1.5

1 1.0

o 10 20 30 40 50 60 Time (min) (A)

Substitute Sheet

(Rule26) RO/AU

100 5.0

90 4.5 Kinetics with Lewatit TP 207

80 4.0

70 3.5

60 3,0

50 2,5

Ni Fe H Co 2.0 40 Cu pH 30 1.5

20 1.0

10 0.5

0,0 0 0 10 20 30 40 50 60 Time (min) (B)

Figure 21A & 21B

100

90

80

70 Ni

60 Co Cu 50 Fe 40

30

20

10

0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Solution / resin ratio (A)

1000.00

100.00

10.00

1.00

Ni Co 0.10 Cu Fe

0.01 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Solution / resin ratio (mL/g) (B)

Substitute Sheet

(Rule26) RO/AU

1000.00

100.00

Ni

10.00 Co Cu Fe 1.00

0.10

0,01

0 200 400 600 800 1000 1200 1400 1600 1800 2000 Solution/resin ratio (mL/g) (C)

0,5

Lewatit TP 207 Fe 0.4 Loading distribution isotherms

0.3

0.2

0.1

0.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0

Fe in raffinate (mg/L) (D)

Figure 22A to 22D

100

90

80

70 Ni 60

50 Co Cu 40 Fe 30

20

10

o 0.0 1.0 2.0 3.0 4.0 5.0 6.0 H2SO4 (M)

Figure 23

Substitute Sheet

(Rule26) RO/AU

Ni 70 Loading pH isotherms Co with Amberlite IRC 748 60 Cu 50 Fe

40

30

20

10

0 1.5 2.0 2.5 3.0 3,5 4.0 4.5 5.0 5.5

pH (A)

100.000

Ni Loading pH isotherms 10.000 Co with Amberlite IRC 748

Cu 1.000 Fe

0.100

0.010

0.001

0.000 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5

pH (B)

0.10 130

0.09 120

0.08 110 Loading pH isotherms 0.07 with Amberlite IRC 748 100

0.06 90

0.05 80

0.04 70

0.03 Co 60 Cu 0.02 50 Fe 0,01 Ni 40

0.00 30 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5

pH (C)

Figure 24A to 24C

Substitute Sheet

(Rule26) RO/AU

Loading pH isotherms 90 Purolite 910

80

70

Ni 60

Co 50 Cu 40 Fe

30

20

10

0 2.5 3.0 3.5 4.0 4.5 5.0 5.5

pH

Figure 25

100 20000

90 18000

Extraction pH isotherms 80 10% Cyanex 272 in D70 16000 I

70 14000 Based on aqueous & 60 organic assays 12000

50 10000

40 Ni 8000

30 Co 6000

20 Cu 4000 Fe 10 2000 - SF(Co/Ni)

0 0 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5

pH (A)

10.000

EX pH isotherms 10% Cyanex 272

1.000

Ni

Co 0.100 Cu Fe

0.010

0.001 2.0 2.5 3,0 3.5 4.0 4.5 5.0 5.5

pH (B)

Substitute Sheet

(Rule26) RO/AU

100.000

EX pH isotherms Ni 10.000 10% Cyanex 272 Co 1.000 Cu Fe

0.100

0.010

0.001

0.000 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5

pH (C)

Figure 26A to 26C

1000.0

100.0 Feed Co Raffinate Co Raffinate Cu Raffinate Fe 10.0 Column 1 Loading Resin: Average 145 mL TP 272 Co ICP-MS Feed: Real PLS Flowrate: 2 BV/h

1.0

0.1

O 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72

BV (A) 36

34

32 Column 1 Loading 30 Resin BV: Average 145 mL TP 272 28 Feed: PLS 1&2 26 Flowrate: 2 BV/h 24 Ni 22

20 Co 18 Cu 16 Fe 14

12

10

8 6 4

2

0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72

BV (B)

Substitute Sheet

(Rule26) RO/AU

Figure 27

10 10 10

Column 3 Loading Column 3 Loading Column 3 Loading 9 9 9 Resin: 150 mL TP 272 Resin: 150 mL TP 272 Resin: 150 mL TP 272 8 Feed: C1 Raffinate 8 Feed: C1 Raffinate 8 Feed: C1 Raffinate Assay: ICP-MS Assay: ICP-MS Assay: ICP-MS 7 7 7

6 6 6 Co at 5 BV/h Co at 2 Bv/hr Co at 0.8 Bv/hr 5 5 5

4 4 4

3 3 3

2 2 2

1 1 1

0 0 o 1 0 2 4 6 8 10 0 2 4 6 8 10 12 14 16 18 0 2 3 4 5

BV BV BV

Figure 28(A) - (C)

18 8.0

16 7.0

14 6.0

12 Column 1 Load Wash Resin: 145 mL TP 272 Ni 5.0

Aq: DI water 10 Co Flowrate: 5 BV/h 4.0 Cu 8 Fe pH 3.0 6

2.0 4

1.0 2 1 0 0,0 o 2 4 6 8 10 BV

Figure 29 TU.U zuu Elution Run 1 Elution Run 2 Elution Run 3 9.0 180 0.1 M H2SO4 0.5 M H2SO4 1.5 M H2SO4 8.0 160

7.0 140 Ni 6.0 120 Co

5.0 Cu 100

Fe 4.0 80

3.0 60

2.0 40

1.0 20

0.0 A 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Bed Volume Figure 30

Substitute Sheet

(Rule26) RO/AU

0.016 8.0

Column 1 Elution Wash 0.014 7.0 Resin: 145 mL TP 272 Aq: DI water 0.012 6.0 Flowrate: 5 BV/h I

0.010 5.0 Ni

0.008 Co 4.0

Cu Fe 3.0 0.006 pH 0.004 2.0

0.002 1.0

0,000 0.0 1 0 2 3 4 5 6 7 8 9 10

BV

Figure 31

160

150

140

130 Feed Co 120 Column 1 Loading Raffinate Co 110 Resin: Average 145 mL TP 272 Raffinate Cu 100 Feed: Real PLS

Flowrate: 2 BV/h 90 Raffinate Fe

80

70

60

50

40

30

20

10 +

0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64

BV (A)

Cobalt breakthrough curve comparsion

100

120

10%

8.7

10 70 2 BVS Figure 32 (B)

Substitute Sheet

(Rule26) RO/AU