Recovery of Rare Earth Oxide from Waste Nimh Batteries by Simple Wet Chemical Valorization Process

metals

Article Recovery of Rare Earth Oxide from Waste NiMH Batteries by Simple Wet Chemical Valorization Process

Nak-Kyoon Ahn 1 , Basudev Swain 1,* , Hyun-Woo Shim 1 and Dae-Weon Kim 2,*

1 Materials Science and Chemical Engineering Center, Institute for Advanced Engineering (IAE), Yongin 17180, Korea; [email protected] (N.-K.A.); [email protected] (H.-W.S.) 2 Advanced Materials and Processing Center, Institute for Advanced Engineering (IAE), Yongin 17180, Korea * Correspondence: [email protected] (B.S.); [email protected] (D.-W.K.)

Received: 6 September 2019; Accepted: 22 October 2019; Published: 27 October 2019

Abstract: Nickel metal hydride (NiMH) batteries contain a significant amount of rare earth metals (REMs) such as Ce, La, and Nd, which are critical to the supply chain. Recovery of these metals from waste NiMH batteries can be a potential secondary resource for REMs. In our current REM recovery process, REM oxide from waste NiMH batteries was recovered by a simple wet chemical valorization process. The process followed the chemical metallurgy process to recover REM oxides and included the following stages: (1) H2SO4 leaching; (2) selective separation of REM as sulfate salt from Ni/Co sulfate solution; (3) metathesis purification reaction process for the conversion REM sulfate to REM carbonate; and (4) recovery of REM oxide from REM carbonate by heat treatment. Through H2SO4 leaching optimization, almost all the metal from the electrode active material of waste NiMH batteries was leached out. From the filtered leach liquor managing pH (at pH 1.8) with 10 M NaOH, REM was precipitated as hydrated NaREE(SO ) H O, which was then further valorized 4 2· 2 through the metathesis reaction process. From NaREE(SO ) H O through carbocation, REM was 4 2· 2 purified as hydrated (REM) CO H O precipitate. From (REM) CO H O through calcination at 2 3· 2 2 3· 2 800 ◦C, (REM)2O3 could be recovered.

Keywords: spent NiMH batteries; circular economy; rare earths; valorization

1. Introduction Owing to the strategic, economic, and industrial importance of rare earth metals (REMs), the EU has classified them as critical raw materials (CRMs). A criticality assessment of metals based on economic importance and supply risk done in 2017 resulted in the EU considering heavy rare earth metals (HREMs), light rare earth metals (LREM), and Sc as CRMs [1,2]. The EU, along with the US Department of Energy (DOE) [3] and the American Physical Society (APS), [4] have reported REM as critical for energy and emerging technologies. The global supply chain of critical metals like HREMs, LREMs, and Sc is mainly dominated by China, which has a market share of up to 95% [2]. The global distribution of primary resources for these critical REMs is quite uneven and currently dominated by China. Hence, the recovery of these CRMs from secondary resources is very important in order to secure these metals for industrial economies like Korea, where natural resources are scarce. The nickel metal hydride (NiMH) battery could be a secondary resource for REMs like Ce, La, and Nd, as it contains a significant amount of these elements (which are all REMs). By efficiently recycling NiMH waste batteries, the REMs can be recovered and circulated in the industrial ecosystem. This could simultaneously address issues like urban mining, environmental directive, and CRM supply chain challenges, and could secure the supply of these metals for industrial development.

Metals 2019, 9, 1151; doi:10.3390/met9111151 www.mdpi.com/journal/metals Metals 2019, 9, 1151 2 of 13

With the emergence of more efficient batteries like the lithium-ion battery, the use of NiMH batteries has gradually decreased, which is leading to the NiMH batteries reaching end of life (EOL). Different industries around the world are recycling various batteries using either pyrometallurgy or hydrometallurgy processes, mainly targeting base metals like Co and Ni. Table1 summarizes battery industries worldwide who are treating or recycling NiMH batteries in particular, and all types of batteries in general [5,6]. Though an industrial recycling process is available for NiMH batteries, the industrial-scale recovery of REMs has hardly been available. Numerous authors have reported on NiMH battery recycling research by hydrometallurgy, pyrometallurgy, or a hybrid of both [7–16]. Xia et al. reported on REM recovery by leaching–solvent extraction route where battery powder was leached by H2SO4, and REMs were then purified by solvent extraction using Cyanex 923. REM oxide was recovered using oxalic acid [16]. Yang et al. reported a HCl leaching and oxalic acid precipitation route for the recovery of REM oxide by calcination [11]. Korkmaz et al. have reported H2SO4 + SO2 mixing, drying at 110 ◦C, roasting 850 ◦C and water leaching routes for the recovery of REMs from NiMH anode active materials [17]. Korkmaz et al. also reported a HCl acid leaching and oxalic acid precipitation route for the recovery of REM from waste NiMH batteries [18]. Innocenzi and Vegliò reported REM recovery by H2SO4 leaching followed by a (REM)2SO4 precipitation route using NaOH [19]. Though there have been several reports on the recovery of REMs by leaching precipitation, the REM oxide recovery leaching–precipitation–metathesis reaction route has hardly been discussed in the literature. As the industrial-scale recovery of REMs from waste NiMH is the primary challenge, in our current research we developed an industrially feasible recycling process where the circular economy goal can be achieved.

Table 1. Summary of battery recycling processes by various companies around the world (data from [5,6]).

Company Battery Type Process/Technology Used Location Pyrometallurgy Trail, BC, Canada; Baltimore, OH, Retriev Technologies All types Hydrometallurgy USA Salesco Systems All types Pyrometallurgy Phoenix, AZ, USA Allentown PA, USA; Hayward CA, AERC All types Pyrometallurgy USA; West Melbourne FL, USA Dowa All types Pyrometallurgy Japan Japan Recycle All types Pyrometallurgy Osaka, Japan Sony Corp. & Sumitomo All types Pyrometallurgy Japan Metals and Mining Co. Pyrometallurgy Horne Que, Nikkelverk Norway; XStrata All types Electrowinning Sudbury, ON., Canada Accurec All types Pyrometallurgy Mulhiem, Grenada DK All types Pyrometallurgy Duisburg, Greece Zurich, Switzerland; Rogerville, AFE Group (Valdi) All types Pyrometallurgy France Zurich, Switzerland; Rogerville, Citron All types Pyrometallurgy France Euro Dieuze/SARP All types Hydrometallurgy Lorraine, France SNAM Cd, Ni, MH, Li Pyrometallurgy Saint Quentin Fallavier, France IPGNA Ent. (Recupyl) All types Hydrometallurgy Grenoble, France Pyrometallurgy Umicore All types Hydrometallurgy Hooboken, Belgium Electrowinning

In our current investigation, we developed a simple yet industrially feasible and sustainable REM oxide recovery process. The complete process is depicted in a ﬂowsheet in Figure1. As shown in Metals 2019, 9, x FOR PEER REVIEW 3 of 13

Pyrometallurgy MetalsUmicore2019, 9, 1151 All types Hydrometallurgy Hooboken, Belgium 3 of 13 Electrowinning

the figure,In our thecurrent developed investigation, process we comprises developed sequential a simple H 2yetSO 4industriallyacid leaching feasible followed and bysustainable selective NaREE(SO ) H O precipitation and purification of (REM) CO H O by carbocation–mixed REM REM oxide4 recovery2· 2 process. The complete process is depicted2 in3 ·a flowsheet2 in Figure 1. As shown oxide isolation by calcination. As shown in Figure1, the process follows the integration of physical in the figure, the developed process comprises sequential H2SO4 acid leaching followed by selective dismantling and classification followed by the chemical metallurgy process. The physical separation NaREE(SO4)2 H2O precipitation and purification of (REM)2CO3.H2O by carbocation–mixed REM processoxide isolation comprises by calcin the followingation. As mechanical shown in Figure steps: (1)1, the residual process stored follows energy the discharge;integration (2) of grindingphysical anddismantling calcination; and andclassification (3) classification followed and by the access chemical of the metallurgy electrode material. process. The This physical is followed separation by the chemicalprocess comprises metallurgy the process, following which mechanical has the steps: following (1) residual stages: stored (1) H2SO energy4 leaching; discharge (2) separation; (2) grinding of theand REMscalcination as sulfate; and salt (3) fromclassification the Ni/Co and sulfate access solution; of the (3)electrode the metathesis material. synthesis This is f reactionollowed process by the for conversion of REM sulfate to REM carbonate; and (4) REM oxide recovery by heat treatment. The chemical metallurgy process, which has the following stages: (1) H2SO4 leaching; (2) separation of the noveltyREMs as of sulfate this process salt from is described the Ni/Co below. sulfate solution; (3) the metathesis synthesis reaction process for conversion of REM sulfate to REM carbonate; and (4) REM oxide recovery by heat treatment. The i. All battery recycling processes by various companies around the world that have been reviewed novelty of this process is described below. mostly follow pyrometallurgical or pyrometallurgical-dominated processes, in contrast to our i. All battery recycling processes by various companies around the world that have been reviewed developed process which is hydrometallurical. mostly follow pyrometallurgical or pyrometallurgical-dominated processes, in contrast to our ii. It is a simple acid leaching and precipitation process for the recovery of REMs. developed process which is hydrometallurical. iii. Recovered REM sulfate is value-added through a simple carbocation reaction. ii. It is a simple acid leaching and precipitation process for the recovery of REMs. iii. Recovered REM sulfate is value-added through a simple carbocation reaction.

Figure 1. 1. ExperimentalExperimental procedure procedure for for the the recovery recovery of rare of rareearth earth oxide oxide from a from waste a nickelwaste metal nickel hydride metal (NiMH)hydride battery(NiMH) by battery a simple by a wet simple chemical wet chemical valorization valorization process. REM:process. rare REM: earth rare metal. earth metal.

Metals 2019, 9, x FOR PEER REVIEW 4 of 13

Metals2. Material2019, 9, 1151s and Methods 4 of 13

2.1. Materials 2. Materials and Methods Waste NiMH battery modules (Model: HHR-33AH72W6) were imported from Japan by GM- 2.1.Tech Materials Co., KOREA. After dismantling, electrode active material was supplied for the current investigationWaste NiMH by GM battery-Tech modules Co., KOREA. (Model: All HHR-33AH72W6) chemicals, H2SO4 were(97%), imported NaOH (97%), from Japanand Na by2CO GM-Tech3 (97%), Co.,were KOREA. of analytical After dismantling, grade purchased electrode from active Daejung, material Korea was. Thesupplied composition for the current of recovered investigation NiMH battery powder was analyzed by XRD (X-ray diffraction spectroscopy, XRD-6100, Shimadzu, Kyoto, by GM-Tech Co., KOREA. All chemicals, H2SO4 (97%), NaOH (97%), and Na2CO3 (97%), were of analyticalJapan), XRF grade (X- purchasedray fluorescence from Daejung, spectroscopy, Korea. ZSX The compositionPrimus Ⅱ, Rigaku of recovered, Houston NiMH, TX, battery USA) and powder MP- wasAES analyzed (Microwave by XRD Plasma (X-ray-Atomic diffraction Emission spectroscopy, Spectroscopy XRD-6100, 4200, Agilent Shimadzu,, Santa Kyoto, Clara, Japan), CA, U XRFSA). (X-rayThermal fluorescence analysis was spectroscopy, carried out ZSX by Primus TGA (Thermo II, Rigaku, Gravimetric Houston, TX, Analyzer, USA) and TG MP-AES 8121, Houston (Microwave, TX, Plasma-AtomicUSA). Emission Spectroscopy 4200, Agilent, Santa Clara, CA, USA). Thermal analysis was carried out by TGA (Thermo Gravimetric Analyzer, TG 8121, Houston, TX, USA). 2.2. Leaching and REM Separation 2.2. LeachingThe leaching and REM reactor Separation used for leaching of the NiMH battery powder is shown in Figure 2. The reactorThe vessel leaching was reactor a three used-necked for round leaching-bottom of the flask NiMH of 1 battery L capacity. powder The isflask shown was injacketed Figure 2for. Theheating reactor and vessel controlling was a three-necked the reaction round-bottom temperature. The flask reactor of 1 L wascapacity. equipped The flask with wasan overheadjacketed foragitator heating, and and was controlling driven the by reaction a variable temperature.-speed motor, The areactor digital was thermocouple equipped with to an measure overhead the agitator,temperature and was during driven continuous by a variable-speed operation of motor, the reactor, a digital and thermocouple a condenser to for measure cooling the. Before temperature starting duringthe leaching, continuous the requisite operation volume of the reactor,of H2SO and4 solution a condenser was added for cooling. to the Beforereaction starting flask and the leaching,heated to thethe requisitetarget temperature volume of. HWaste2SO4 NiMHsolution battery was addedpowder to was the added reaction to flaskthe reaction and heated vessel to, thewhich target was temperature.then closed. After Waste the NiMH reaction battery was powdercomplete was, the added leaching to the liquor reaction was separated vessel, which from was residue then through closed. Afterfiltration. the reaction From the was leach complete, liquor, theNaREE(SO leaching4) liquor2 H2O was separated fromby precipitation residue through, controlling filtration. pH Fromusing the NaOH leach while liquor, stirring NaREE(SO at 400) rpmH O under was separated pH control. by precipitation,Isolated NaREE(SO controlling4)2 H2 pHO was using dried. NaOH For 4 2· 2 whilethe metathesis stirring at 400reaction rpm,under the NaREE(SO pH control.4)2 IsolatedH2O was NaREE(SO added to the) H NaO2CO was3 dried.solution For and the the metathesis reaction 4 2· 2 reaction,was allowed the NaREE(SO to complete.) HAfterO wasreaction added completion to the Na wasCO achieved,solution andthe (REM) the reaction2CO3 was was precipitated allowed to 4 2· 2 2 3 complete.out and Afterrecovered reaction through completion filtration. was achieved, After drying the (REM), the 2 (REM)CO3 was2CO precipitated3 was calcined out and for recovered (REM)2O3 throughrecovery. filtration. After drying, the (REM)2CO3 was calcined for (REM)2O3 recovery.

Figure 2. Schematic of the leaching reactor used for leaching waste electroactive powder material from Figure 2. Schematic of the leaching reactor used for leaching waste electroactive powder material a waste NiMH battery. from a waste NiMH battery.

Metals 2019, 9, x FOR PEER REVIEW 5 of 13

Metals2.3. Characterization2019, 9, 1151 5 of 13

Each of the isolated samples, including NaREE(SO4)2 H2O, (REM)2CO3, and(REM)2O3, were 2.3.analyzed Characterization by XRD (X-ray diffraction spectroscopy, XRD-6100, Shimadzu). The thermal decomposition behavior of the recovered sample was also analyzed by TGA (Thermo Gravimetric Analyzer, TG 8121, Each of the isolated samples, including NaREE(SO ) H O, (REM) CO , and(REM) O , were analyzed Rigaku). 4 2· 2 2 3 2 3 by XRD (X-ray diffraction spectroscopy, XRD-6100, Shimadzu). The thermal decomposition behavior of the3. Results recovered and sample Discussion was also analyzed by TGA (Thermo Gravimetric Analyzer, TG 8121, Rigaku). 3. Results and Discussion 3.1. Characterization of Waste NiMH Battery Material 3.1. CharacterizationPrior to characterizatio of Wasten, NiMH the waste Battery NiMH Material battery material module was discharged inside the brine.Prior Grinding to characterization, was followed theby classification waste NiMH, and battery electroactive material modulematerial was of 100 discharged mesh-size inside powder the brine.was recovered Grinding using was followeda pulverizer. by classification,XRD (Figure 3) and of electroactivethe waste NiMH material battery of 100powder mesh-size was analyzed. powder wasFrom recovered XRD analysis using, amajor pulverizer. phases XRDsuch (Figure as Ni(OH)3) of2, the NiO, waste Ni, NiMHNi7Ce2, battery Co5Nd, powder and La was2O3 and analyzed. minor 7 2 3 Fromphases XRD such analysis, as NiO, majorNiOOH, phases Ni Nd such, and as Ni(OH)Co Nd 2were, NiO, observed Ni, Ni7.Ce XRF2, Co analysis5Nd, and forLa the2O same3 and sample minor indicated Ni, Co, Ce, La, and Nd were the primary constituents in battery material with significant phases such as NiO, NiOOH, Ni7Nd2, and Co3Nd were observed. XRF analysis for the same sample indicatedNi content Ni, found Co, Ce, in La, the and waste Nd were NiMH the battery. primary The constituents composition in battery of the material recovered with NiMH significant battery Ni contentpowder found was also in the analyzed waste NiMH by MP battery.-AES The(Microwave composition Plasma of the Atomic recovered Emission NiMH batterySpectroscopy powder 4200, was alsoAgilent). analyzed It was by confirmed MP-AES (Microwavethat Ni, Co, Plasmaand REM Atomics (La, Ce, Emission Nd) comp Spectroscopyrised 45.8%, 4200, 8.5%, Agilent). and 20.2% It was of confirmedthe recovered that battery Ni, Co, pow andd REMser, respectively (La, Ce, Nd) (Table comprised 2). As Table 45.8%, 2 indica 8.5%,tes and, the 20.2% NiMH of thebattery recovered waste batterycontained powder, a significant respectively amount (Table of 2REMs). As Table(20.2%)2 indicates,, which would the NiMH be valuable battery waste CRM containeds for Korean a significantindustries. amount of REMs (20.2%), which would be valuable CRMs for Korean industries.

Figure 3. XRD pattern of ground waste NiMH battery powder. Figure 3. XRD pattern of ground waste NiMH battery powder. Table 2. The composition (%w/w) of metal elements contained in the spent NiMH batteries. Table 2. The composition (%w/w) of metal elements contained in the spent NiMH batteries. Rare Earth Elements (REMs) Total REMs Ni Co Others Rare Earth Elements (REMs) Ce La Nd Total REMs Ni Co Others Ce La Nd Weight (%) 10.4 6.7 3.1 20.2 45.8 8.5 25.5 Weight (%) 10.4 6.7 3.1 20.2 45.8 8.5 25.5

3.2. Leaching Optimization of Waste NiMH Battery Material

Metals 2019, 9, x FOR PEER REVIEW 6 of 13

For the development of a simple REM valorization process from the waste NiMH battery module, leaching parameters like temperature and reaction time were selected from the literature Metals 2019, 9, 1151 6 of 13 survey. Pietrelli et al. reported REM recovery from NiMH spent batteries using H2SO4 as a suitable lixiviant [20]. For the extraction of valuable REMs like Ce, La, and Nd from waste NiMH battery module3.2. Leachings, important Optimization leaching of Wasteparameters NiMH like Battery the Materiallixiviant (H2SO4) concentration and pulp densities were optimized, keeping the leaching temperature constant at 90 °C and leaching reaction time of 4 h. TemperatureFor the development and time of were a simple considered REM valorization from the reported process from studies the wasteby Pietrelli NiMH et battery al. [20 module,]. The leachingleaching optimization parameters like experiments temperature were and reactioncarried out time and were both selected parameters from the were literature varied survey. at the Pietrelli same et al. reported REM recovery from NiMH spent batteries using H SO as a suitable lixiviant [20]. For the time. The lixiviant concentration ranged from 1 to 4 M of H2SO42, while4 pulp density ranged from 25 toextraction 200 g/L. of The valuable reaction REMs temperature like Ce, La, was and 90 Nd °C from the waste reaction NiMH time battery was modules,4 h. Figure important 4a depicts leaching the leachingparameters behavior like the of lixiviantAl, Co, Fe, (H 2Mn,SO4 Ni,) concentration and Zn, and andFigure pulp 4b densities depicts the were leaching optimized, behavior keeping of Ce, the La,leaching and Nd temperature as a function constant of lixiviant at 90 concentration◦C and leaching and reaction pulp density. time of Figure 4 h. Temperature 4a shows that and Al, time Co, were Fe, Mn,considered Ni, and from Zn extraction the reported gradually studies increaseby Pietrellid as et a al. function [20]. The of leaching lixiviant optimization concentration experiments and pulp densitywere carried when outboth and were both abundant parameters in were the leaching varied at theprocess. same Hence, time. The higher lixiviant efficiency concentration, higher ranged pulp from 1 to 4 M of H SO , while pulp density ranged from 25 to 200 g/L. The reaction temperature was density, and higher2 H42SO4 acid concentration could be suitable for efficient leaching process 90 C the reaction time was 4 h. Figure4a depicts the leaching behavior of Al, Co, Fe, Mn, Ni, and Zn, development.◦ Similarly, as shown in Figure 4b, Ce, La, and Nd extraction increased up to 3 M H2SO4. and Figure4b depicts the leaching behavior of Ce, La, and Nd as a function of lixiviant concentration Extraction decreased gradually when 4 M H2SO4 was used as a lixiviant. The same figure also indicateand pulps a density. higher Figureamount4a of shows Ce, La,that and Al, Co, Nd Fe, were Mn, extracted Ni, and Znas extraction function of gradually pulp density increased. To understandas a function the of variable lixiviant leaching concentration phenomena and pulpobserved density in Figure when both4, thewere leaching abundant chemistry in the needed leaching to beprocess. understood. Hence, As higher the focus efficiency, of the higher investigation pulp density, is only and higherREMs, Hthe2SO possible4 acid concentration leaching reaction could chemistrybe suitable is forexplained efficient below. leaching As processXRD indicated development., La, Ce, Similarly, and Nd (REM as showns) present in Figureed mainly4b, Ce, as La, REM and oxideNd extraction and a Co/Ni increased alloy up compound to 3 M H. 2 ThereforeSO4. Extraction the leaching decreased chemistry gradually can when be explained 4 M H2SO using4 was Equationsused as a (1 lixiviant.)–(3) presented The same below. figure also indicates a higher amount of Ce, La, and Nd were extracted as function of pulp density. To understand the variable leaching phenomena observed in Figure4, + 3 → ( ) + 3 the leaching chemistry needed to be understood. As the ( focus) of the investigation is only REMs,(1 the) possible leaching() reaction + ( chemistry+ 2) is → explained()( below.)( As) + XRD indicated,() + ( La,+ 2 Ce,) and Nd (REMs)(2) presented mainly as REM oxide and a Co/Ni alloy compound. Therefore the leaching chemistry can be + ( + 3) → ()() + () + ( + 3) (3) explained using Equations (1)–(3) presented below.

Figure 4. Cont.

Metals 2019, 9, x FOR PEER REVIEW 7 of 13 Metals 2019, 9, 1151 7 of 13

Figure 4. Eﬀect of lixiviant concentration and pulp density on the extraction of metals (a) Al, Co, Fe, Mn,Figure Ni, and4. Effect Zn, ( bof) Ce,lixiviant La, and concentration Nd, from waste and pulp NiMH density battery on material. the extraction of metals (a) Al, Co, Fe, Mn, Ni, and Zn, (b) Ce, La, and Nd, from waste NiMH battery material.

3.3. Separation and Recovery Laof REMO + by3H MetathesisSO La Reaction(SO ) + 3H O (1) 2 3 2 4 → 2 4 3(aq) 2 Though base metal extraction for Ni and Co was more efficient with higher pulp density and + higher lixiviant(REM) concentrationNix + (x + 2)H, REMSO extraction(REE) ( SO was) not + muchxNiSO more efficient+ (x + 2 ) atH higherO lixiviant(2) 2 2 4 → 2 4 3(aq) 4(aq) 3 concentrations. As the main focus of the investigation was REM recovery, conditions such as using 1 M of H2SO4 acid as a lixiviant and a pulp density of 25 g/L were considered for+ further studies. CoxNd + (x + 3)H2SO4 Nd2(SO4)3(aq) + xCoSO4(aq) + (x + 3)H3O (3) Sufficient volume of leach liquor was→ generated using 1 M H2SO4 acid at a pulp density of 25 g/L, at a temperature of 90 °C with a reaction time of 4 h. Followed by solid–liquid separation, REMs were 3.3. Separation and Recovery of REM by Metathesis Reaction separated the leach liquor through precipitation using NaOH. The initial pH of the leaching liquor wasThough about 0.05 base after metal solid extraction–liquid for separation. Ni and Co The was pH more of eleachfficient liquor with higherwas adjusted pulp density by slowly and higheradding lixiviant10 M NaOH concentration, to the leach REM liquor extraction under was constant not much stirring. more eAsfficient the pH at higher of the lixiviant solution concentrations. was increased Asslowly the main to 1.8, focus REM ofs the were investigation precipitated was, leaving REM recovery, Co, Ni, and conditions other metals such as in using the leach 1 M liquor. of H2SO The4 acidprecipitated as a lixiviant REM ands w aere pulp separated density from of 25 the g/L solution were considered by simple for filtration further and studies. the sample Sufficient was volume washed ofseveral leach liquortimes to was remove generated soluble using components. 1 M H2SO After4 acid washing, at a pulp the density samples of 25 were g/L, dried at a temperature overnight in of an 90oven◦C with at 80 a °C reaction, and were time then of 4 h.characterized Followed by by solid–liquid XRD, ICP- separation,MS, and TGA. REMs were separated the leach liquorThe through recovered precipitation REM powder using NaOH.was analyzed The initial by XRD pH ofand the the leaching obtained liquor XRD was pattern about is 0.05 presented after solid–liquidin Figure 5 separation.. The powder The was pH identified of leach liquor as a mixture was adjusted of NaNd(SO by slowly4)2 addingH2O, NaCe(SO 10 M NaOH4)2 H2 toO, theand leachNaLa(SO liquor4)2 underH2O, which constant corresponded stirring. As to the JCPDS pH of # the40–1481, solution JCPDS was # increased86–0526, and slowly JCPDS to 1.8, # 82 REMs–1199, wererespectively. precipitated, From leaving now on Co,, the Ni, NaNd(SO and other4)2 metalsH2O, NaCe(SO in the leach4)2 H liquor.2O, and The NaLa(SO precipitated4)2 H2O REMs mixture were will separatedbe termed from (NaREE(SO the solution4)2 H by2O). simple The c filtrationontent of andCe, theLa, sampleand Nd was impurities washed including several times Ni and to remove Co was solubleanalyzed components. by MP-AES After by dissolving washing, thethe samplespowder. wereThis is dried presented overnight in Table in an 3. oven The atMP 80-AES◦C, andconfirmed were thenthat characterized Ce (17.2%), Laby XRD,(13.1% ICP-MS,), and Nd and (5.4% TGA.) were the main metal components in the precipitate, whereasThe recovered Co and Ni REM impurities powder only was made analyzed up 0.01%. by XRD Hence, and selective the obtained separation XRD pattern of NaREE(SO is presented4)2 H2O inusing Figure a 5precipitation. The powder technique was identified by adding as a NaOH mixture to ofthe NaNd(SO leachate is) anH efficientO, NaCe(SO method) H toO, recover and 4 2· 2 4 2· 2 NaLa(SOREMs from) H leachO, which liquor. corresponded to JCPDS # 40–1481, JCPDS # 86–0526, and JCPDS # 82–1199, 4 2· 2 respectively. From now on, the NaNd(SO ) H O, NaCe(SO ) H O, and NaLa(SO ) H O mixture 4 2· 2 4 2· 2 4 2· 2

Metals 2019, 9, 1151 8 of 13 will be termed (NaREE(SO ) H O). The content of Ce, La, and Nd impurities including Ni and Co was 4 2· 2 analyzed by MP-AES by dissolving the powder. This is presented in Table3. The MP-AES conﬁrmed that Ce (17.2%), La (13.1%), and Nd (5.4%) were the main metal components in the precipitate, whereas Co and Ni impurities only made up 0.01%. Hence, selective separation of NaREE(SO ) H O using a 4 2· 2 precipitation technique by adding NaOH to the leachate is an eﬃcient method to recover REMs from leach liquor. Metals 2019, 9, x FOR PEER REVIEW 8 of 13

Figure 5. XRD pattern of isolated REM (the NaNd(SO ) H O, NaCe(SO ) H O, and NaLa(SO ) H O 4 2· 2 4 2· 2 4 2· 2 mixtureFigure is referred 5. XRD to aspattern NaREE(SO of isolated) H REMO)) precipitates. (the NaNd(SO4)2 H2O, NaCe(SO4)2 H2O, and NaLa(SO4)2 H2O 4 2· 2 mixture is referred to as NaREE(SO4)2 H2O)) precipitates. Table 3. The composition (%w/w) of the metal elements contained in NaREE(SO ) H O. 4 2· 2 Table 3. The composition (%w/w) of the metal elements contained in NaREE(SO4)2 H2O. Rare Earth Elements # Ni + Co Ce LaRare Earth E Ndlements # Ni + Co Weight (%) 17.2Ce 13.1La 5.44Nd 0.01 Weight (%) 17.2 13.1 5.44 0.01 From the isolated NaREE(SO ) H O, a carbonation reaction using Na CO was completed, and From the isolated NaREE(SO4 2· 2 4)2 H2O, a carbonation reaction using2 Na3 2CO3 was completed, and the REMs were purified as anhydrous (REE)2(CO23)3 x3H32‧O.2 The carbonation reaction was carried the REMs were purified as anhydrous (REE) (CO· ) xH O. The carbonation reaction was carried out out by adding the isolated NaREE(SO4)422H‧ 22O powder to 200 mL of Na22CO3 3 solution in 500 mL of by adding the isolated NaREE(SO )· H O powder to 200 mL of Na CO solution in 500 mL of reactor, reactor,and and the the reaction reaction was was carried carried out out for for 5 h 5. The h. The stoichiometric stoichiometric ratio ratio of 1:1.1 of 1:1.1 for REM(III) for REM(III) with with CO32− was 2 CO3 − was maintained for the purification of anhydrous (REE)22(CO333)‧3 xH2 2O. The carbonation reaction maintained for the purification of anhydrous (REE) (CO ) xH· O. The carbonation reaction was was carriedcarried out out in in two two didifferentfferent temperaturestemperatures (i.e.,(i.e., r roomoom temperature temperature (RT) (RT) and and at at 70 70 °C◦)C).. After After the the reaction reactionwas was complete completed,d, the theisolated isolated sample sample was was dried dried and andanalyzed analyzed by XRD by XRD.. The XRD The XRDpattern pattern is depicted is in depictedFigure in Figure 6. The6 .XRD The pattern XRD pattern in Figure in Figure6 indicates6 indicates that there that was there almost was almost no difference no di ffinerence XRD pattern in s XRD patternsbetween between the two thedifferent two di samplesfferent samples isolated isolated after the after reaction the reaction at both atroo bothm temperature room temperature and at 70 °C. and at 70 ◦C. The XRD pattern was confirmed to be (CeLa)2(CO23)3 4H3 32‧O, (LaNd)2 2(CO32)3 8H3 23O,‧ and2 The XRD pattern was confirmed to be (CeLa) (CO· ) 4H O, (LaNd) (CO· ) 8H O, and NaNd(CO3)2 6H32O,2‧ which2 correspond to JCPDS #06–0076, #30–0678, and #30–1223, respectively. NaNd(CO· ) 6H O, which correspond to JCPDS #06–0076, #30–0678, and #30–1223, respectively.

MetalsMetals2019 2019, 9, 1151, 9, x FOR PEER REVIEW 9 of9 13 of 13

Figure 6. XRD pattern of purified (REM) CO H O powder, obtained through a metathesis reaction 2 3· 2 fromFigure NaREE(SO 6. XRD) patternH O precipitates. of purified (REM)2CO3‧H2O powder, obtained through a metathesis reaction 4 2· 2 from NaREE(SO4)2 H2O precipitates. The XRD pattern observed is a mixture of all three REM carbonates, so it will be presented as (REM)2COThe3. XRD The (REM)pattern2CO observed3 was analyzed is a mixture by TGA of all (Figure three7 ).REM The carbonate TGA figures, indicatesso it will twobe presented stages of as weight(REM) loss2CO as3. temperature The (REM)2CO increased3 was analyzed to 1000 ◦byC. TGA In the (Figure first step, 7). The the weightTGA figure loss wasindica 26.1%tes two between stages of roomweight temperature loss as temperature and 250 ◦C. increased The weight to 1000 loss ° isC. associatedIn the first withstep, lossthe weight of water loss and was conversion 26.1% between of hydratedroom temperature (REM) CO HandO 250 to anhydrous °C. The weight (REM) lossCO is. associated In the second with step, loss the of weightwater and loss conversion was 44.7% of 2 3· 2 2 3 fromhydrated room temperature (REM)2CO3 to H 8002O to◦C. anhydrous At 800 ◦C, (REM) (REM)22COCO3.3 Inwas the converted second step, to (REM) the weight2O3 by loss loss was of CO 44.7%2. To understandfrom room temperature the (REM) O toformation 800 °C. At through 800 °C,the (REM) metathesis2CO3 was reaction, converted the (REM)to (REM)CO2O3H byO loss sample of CO2. 2 3 2 3· 2 wasTo calcined understand at 800 the◦C (REM) for 1 h2O and3 formation characterized through by XRDthe metathesis analysis. Thereaction, XRD the pattern (REM) is represented2CO3.H2O sample in Figurewas8 .calc Theined XRD at peak800 °C in for the 1 XRD h and analysis characterized indicates by that XRD the analysis. sample wasThe XRD a mixture pattern of threeis represented types of in rareFigure earth oxides8. The XRD (i.e., Ndpeak0.5 Cein 0.5theO 1.75XRD, La analysis2O3, and indicates Nd6O11 ).that Hence, the sample pure (REM) was a2 Omixture3 can be of valorized three types throughof rare heat earth treatment oxides ( ofi.e., (REM) Nd0.52CeCO0.53Oat1.75 800, La◦2C.O3, and Nd6O11). Hence, pure (REM)2O3 can be valorized throughA complete heat treatment flowsheet of describing (REM)2CO the3 at recovery 800 °C. of (REM)2O3 from a waste NiMH battery via a simple wet chemical valorization process was developed and is shown in Figure9. The figure depicts the sequential process, i.e., H SO acid leaching, NaREE(SO ) H O precipitation, and metathesis 2 4 4 2· 2 reaction for the complete recycling of waste NiMH batteries. The process was completed with the integration of physical separation followed by the chemical metallurgy process. The physical separation process was followed by residual stored energy discharge, grinding and calcination, classification, and access to the electrode active material. In the chemical metallurgy process, the electrode active powder material was extracted by leaching, REMs were separated selectively by precipitation, and purified by carbocation and calcination processes. The chemical metallurgy process sequentially followed (i) 1 M H2SO4 leaching from a pulp density of 25 g/L at 90 ◦C, (ii) selective separation of REMs as NaREE(SO ) H O salt leaving Ni/Co sulfate solution, (iii) conversion of NaREE(SO ) H O 4 2· 2 4 2· 2 to (REM) CO H O through carbocation, and (iv) synthesis of REM oxide from (REM) CO H O by 2 3· 2 2 3· 2 calcination at 800 ◦C. The mixed REM oxide could be further purified to individual REMs by further

Metals 2019, 9, 1151 10 of 13 investigations,Metals 2019, 9, x FOR mainly PEER REVIEW by hydrometallurgy. As this is a complete systematic sequential process10 of for 13 the recovery of mixed critical REMs like Ce, La, and Nd, it could address the circular economy and recyclingMetals 2019 challenges, 9, x FOR PEER associated REVIEW with waste NiMH batteries. 10 of 13

Figure 7. Thermal characteristics of purified (REM)2CO3.H2O powder.

Figure 7. Thermal characteristics of puriﬁed (REM)2CO3 H2O powder. Figure 7. Thermal characteristics of purified (REM)2CO·3.H2O powder.

Figure 8. XRD pattern of pure (REM) O powder, obtained through the calcination of (REM) CO H O 2 3 2 3· 2 powder.Figure 8. XRD pattern of pure (REM)2O3 powder, obtained through the calcination of (REM)2CO3.H2O powder.

Figure 8. XRD pattern of pure (REM)2O3 powder, obtained through the calcination of (REM)2CO3.H2O A complete flowsheet describing the recovery of (REM)2O3 from a waste NiMH battery via a powder. simple wet chemical valorization process was developed and is shown in Figure 9. The figure depicts the sequential process, i.e., H2SO4 acid leaching, NaREE(SO4)2‧H2O precipitation, and metathesis A complete flowsheet describing the recovery of (REM)2O3 from a waste NiMH battery via a simple wet chemical valorization process was developed and is shown in Figure 9. The figure depicts

the sequential process, i.e., H2SO4 acid leaching, NaREE(SO4)2‧H2O precipitation, and metathesis

Metals 2019, 9, x FOR PEER REVIEW 11 of 13

reaction for the complete recycling of waste NiMH batteries. The process was completed with the integration of physical separation followed by the chemical metallurgy process. The physical separation process was followed by residual stored energy discharge, grinding and calcination, classification, and access to the electrode active material. In the chemical metallurgy process, the electrode active powder material was extracted by leaching, REMs were separated selectively by precipitation, and purified by carbocation and calcination processes. The chemical metallurgy process sequentially followed (i) 1 M H2SO4 leaching from a pulp density of 25 g/L at 90 °C, (ii) selective separation of REMs as NaREE(SO4)2 H2O salt leaving Ni/Co sulfate solution, (iii) conversion of NaREE(SO4)2 H2O to (REM)2CO3.H2O through carbocation, and (iv) synthesis of REM oxide from (REM)2CO3.H2O by calcination at 800 °C. The mixed REM oxide could be further purified to individual REMs by further investigations, mainly by hydrometallurgy. As this is a complete Metalssystematic2019, 9, sequential 1151 process for the recovery of mixed critical REMs like Ce, La, and Nd, it11 could of 13 address the circular economy and recycling challenges associated with waste NiMH batteries.

FigureFigure 9.9. ProcessProcess ﬂowflow sheetsheet forfor the the recovery recovery of of rare rare earth earth metal metal oxide oxide from from waste waste NiMH NiMH batteries batteries by by a simplea simple wet wet chemical chemical valorization valorization process. process.

4.4. ConclusionsConclusion

Through the H2SO4 leaching–selective precipitation–metathesis reaction route of waste NiMH batteries, REM oxide could be recovered quantitatively. Through H2SO4 leaching optimization, almost all metal from the electrode active material of waste NiMH batteries was leached out. From the filtered leach liquor with pH managed (pH 1.8) by 10 M NaOH, REMs were precipitated as hydrated NaREE(SO ) H O, which was further valorized through a metathesis reaction process. 4 2· 2 Isolated NaREE(SO ) H O was further purified through a carbocation reaction using Na CO . 4 2· 2 2 3 From NaREE(SO ) H O through carbocation, REMs were purified as hydrated (REM) CO H O 4 2· 2 2 3· 2 precipitate. From (REM) CO H O through the calcination process at 800 C, (REM) O could 2 3· 2 ◦ 2 3 be recovered. The mixed (REM)2O3 could be further purified through hydrometallurgy, such as dissolution followed by solvent extraction purification. The process is very simple, versatile, and easy to implement in the industry. Advantages of the process are that it is simple, versatile, and flexible, and industrial mass production is feasible. It is also sustainable and eco-efficient, and produces minimal Metals 2019, 9, 1151 12 of 13 waste emissions. The process addresses several issues concurrently, including waste valorization, environment management, critical REM circular economy, and urban mining.

Author Contributions: N.-K.A. and B.S. conceived and designed the experiments; H.-W.S. managed project and analyzed data, and D.-W.K. contributed for funding arrangement, writing and editing the manuscript. B.S. prepared draft and edited the manuscript. Funding: This work was supported by the Korea Evaluation Institute of Industrial Technology (KEIT), which is funded by the Ministry of Trade, Industry and Energy, Republic of Korea (Project No. 10077752). Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Critical Raw Materials. Available online: http://ec.europa.eu/growth/sectors/raw-materials/specific-interest/ critical_en (accessed on 12 September 2019). 2. E.U. Commission. Study on the review of the list of Critical Raw Materials Critical Raw Materials Factsheets Publications Office of the European Union; European Commission: Brussels, Belgium, 2017. 3. U.S. Department of Energy Critical Metal Strategy; U.S. Department of Energy: Washington, DC, USA, 2011. 4. Jaffe, R.; Ceder, G.; Eggert, R.; Graedel, T.; Gschneidner, K.; Hitzman, M.; Houle, F.; Hurd, A.; Kelley, R.; King, A.; et al. Energy Critical Elements. Available online: https://www.aps.org/policy/reports/popa-reports/ upload/elementsreport.pdf (accessed on 21 October 2019). 5. Ellis, T.W.; Mirza, A.H. Battery Recycling: Defining the Market and Identifying the Technology Required to Keep High Value Materials in the Economy and Out of the Waste Dump. Available online: http://www.nist.gov/ tip/wp/pswp/upload/245_battery_recycling_defining_the_market.pdf (accessed on 14 April 2016). 6. European Li-Ion Battery Advanced Manufacturing For Electric Vehicles. Available online: https://elibama. files.wordpress.com/2014/10/v-d-batteries-recycling1.pdf (accessed on 14 April 2019). 7. Müller, T.; Friedrich, B. Development of a recycling process for nickel-metal hydride batteries. J. Power Sources 2006, 158, 1498–1509. [CrossRef] 8. Tenório, J.A.S.; Espinosa, D.C.R. Recovery of Ni-based alloys from spent NiMH batteries. J. Power Sources 2002, 108, 70–73. [CrossRef] 9. Zhang, P.; Yokoyama, T.; Itabashi, O.; Wakui, Y.; Suzuki, T.M.; Inoue, K. Hydrometallurgical process for recovery of metal values from spent nickel-metal hydride secondary batteries. Hydrometallurgy 1998, 50, 61–75. [CrossRef] 10. Tzanetakis, N.; Scott, K. Recycling of nickel–metal hydride batteries. I: Dissolution and solvent extraction of metals. J. Chem. Technol. Biotechnol. 2004, 79, 919–926. [CrossRef] 11. Yang, X.; Zhang, J.; Fang, X. Rare earth element recycling from waste nickel-metal hydride batteries. J Hazard Mater 2014, 279, 384–388. [CrossRef][PubMed] 12. Larsson, K.; Binnemans, K. Selective extraction of metals using ionic liquids for nickel metal hydride battery recycling. Green Chem. 2014, 16, 4595–4603. [CrossRef] 13. Tzanetakis, N.; Scott, K. Recycling of nickel–metal hydride batteries. II: Electrochemical deposition of cobalt and nickel. J. Chem. Technol. Biotechnol. 2004, 79, 927–934. [CrossRef] 14. Valadares, A.; Valadares, C.F.; de Lemos, L.R.; Mageste, A.B.; Rodrigues, G.D. Separation of cobalt and nickel in leach solutions of spent nickel-metal hydride batteries using aqueous two-phase systems (ATPS). Hydrometallurgy 2018, 181, 180–188. [CrossRef] 15. Xi, G.; Xu, H.; Yao, L. Study on preparation of NiCo ferrite using spent lithium-ion and nickel–metal hydride batteries. Sep. Purif. Technol. 2015, 145, 50–55. [CrossRef] 16. Xia, Y.; Xiao, L.; Tian, J.; Li, Z.; Zeng, L. Recovery of rare earths from acid leach solutions of spent nickel-metal hydride batteries using solvent extraction. J. Rare Earths 2015, 33, 1348–1354. [CrossRef] 17. Korkmaz, K.; Alemrajabi, M.; Rasmuson, Å.; Forsberg, K. Recoveries of Valuable Metals from Spent Nickel Metal Hydride Vehicle Batteries via Sulfation, Selective Roasting, and Water Leaching. J. Sustain. Metall. 2018, 4, 313–325. [CrossRef] 18. Korkmaz, K.; Alemrajabi, M.; Rasmuson, Å.; Forsberg, K. Sustainable Hydrometallurgical Recovery of Valuable Elements from Spent Nickel–Metal Hydride HEV Batteries. Metals 2018, 8, 1062. [CrossRef] Metals 2019, 9, 1151 13 of 13

19. Innocenzi, V.; Vegliò, F. Recovery of rare earths and base metals from spent nickel-metal hydride batteries by sequential sulphuric acid leaching and selective precipitations. J. Power Sources 2012, 211, 184–191. [CrossRef] 20. Pietrelli, L.; Bellomo, B.; Fontana, D.; Montereali, M.R. Rare earths recovery from NiMH spent batteries. Hydrometallurgy 2002, 66, 135–139. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).