Preparation of cobalt oxide from concentrated cathode material of spent lithium ion batteries by hydrometallurgical method

Advanced Powder Technology 21 (2010) 175–179

Contents lists available at ScienceDirect

Advanced Powder Technology journal homepage: www.elsevier.com/locate/apt

Original Research Paper

Preparation of cobalt oxide from concentrated cathode material of spent lithium ion batteries by hydrometallurgical method Jingu Kang a,b, Jeongsoo Sohn b, Hankwon Chang b, Gamini Senanayake c, Shun Myung Shin b,* a

Department of Resource Recycling, Korea University of Science and Technology (UST), 52 Eoeun-dong, Yuseong-gu, Daejeon, Republic of Korea Metal Recovery Group, Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources (KIGAM), 92 Kwahak-no, Yuseong-gu, Daejeon, Republic of Korea c Parker Centre, Faculty of Minerals and Energy, Murdoch University, Perth, WA 6150, Australia b

a r t i c l e

i n f o

Article history: Received 6 October 2009 Received in revised form 28 October 2009 Accepted 29 October 2009

Keywords: Spent Li-ion battery Reductive leaching Oxalate precipitation Cobalt oxide

a b s t r a c t Cobalt oxide was prepared from spent lithium ion batteries (LIBs) by reductive leaching, copper sulfide precipitation, cobalt oxalate precipitation and thermal decomposition. The cobalt rich non-magnetic 16 mesh fraction obtained from spent LIBs by mechanical separation was leached using 2 M H2SO4, 6 vol% H2O2, reaction temperature 60 °C, agitation speed 300 rpm, pulp density 100 g/L, reaction time 1 h. The leaching efficiency of cobalt was more than 99% and its concentration was 27.4 g/L. Copper was removed (99.9%) as CuS by precipitating with Na2S. The crystalline solid CoC2O42H2O selectively precipitated by treating the copper-free liquor with oxalic acid was calcined to produce crystalline Co3O4, of which primary average particle size was 340 nm. Ó 2009 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.

1. Introduction The usage of lithium ion batteries (LIBs) has rapidly increased as they are widely used as electrochemical power sources in mobile telephones, personal computers, video cameras and other modern-life appliances [1,2]. Due to their characteristics of light weight, high energy and good performance, LIBs are increasingly substituting for other batteries [3,4]. They not only dominate the cellular phone and laptop computer markets at present, but also will be the first category of dynamic batteries to be chosen to provide power for electronic automobiles in the future. The increasing public concern about the environment in the last decades has resulted in stricter regulations worldwide on the disposal of hazardous waste containing heavy metals, such as spent portable batteries. LIBs consist of organic chemicals (15%), plastics (7%), lithium (5–7%), cobalt (5–20%) and nickel (5–10%) with the composition varying slightly with different manufacturers [5]. Recycling practices are highly desirable at present and in the future from the viewpoints of environmental conservation, and the recovery of metal values from spent LIBs and their utilization as raw materials. Despite the extensive research activities for recycling spent LIBs in many countries there is no commercialized recycling plant as yet [6]. The recycling processes under development stages are mainly composed of acid leaching, solvent extraction, precipitation * Corresponding author. Tel.: +82 42 868 3622; fax: +82 42 868 3415. E-mail address: [email protected] (S.M. Shin).

and electro-winning. Zhang et al. [7] reported a hydrometallurgical process consisting of HCl leaching, solvent extraction of Co(II) with PC-88A, crystallization of CoSO4 and precipitation of Li2CO3. Contestabile et al. [8] studied a recycling process of spent LIBs on a laboratory scale which comprised of the steps: (i) sorting, (ii) selective separation of LiCoO2 powder using N-methylpyrrolidone, (iii) HCl leaching of LiCoO2, (iv) cobalt hydroxide precipitation and thermal treatment to produce Co3O4, and (v) thermal treatment of Co3O4 with Li2CO3 to produce LiCoO2 which was suitable for the production of new batteries. These studies, however, focused mainly synthetic solution or spent cathodic materials generated during the manufacturing process to investigate recycling processes. It is also required that the complex process to recover valuable metals as well as a high capital cost and the solvent cost to establish proper recycling processes. Comparing with previous studies, the present proposed method is a simple process to prepare cobalt compounds from spent LIBs using single leaching/two steps precipitation method. In this study, laboratory tests were conducted for (i) the reductive leaching of LIBs with sulfuric acid and hydrogen peroxide, (ii) the removal of copper by sulfide precipitation with sodium sulfide, (iii) the selective precipitation of cobalt oxalate from the purified leach liquor, and (iv) the preparation of cobalt oxide by calcination. 2. Experimental procedure The physical treatment of spent LIBs included discharging, dehydration, drying, and crushing. The square type spent LIBs were

0921-8831/$ - see front matter Ó 2009 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved. doi:10.1016/j.apt.2009.10.015

176

J. Kang et al. / Advanced Powder Technology 21 (2010) 175–179

range of 0.03–0.3. Desorption isotherm was used to determine the pore size distribution using the Barrett–Joyner–Halenda (BJH) method [10]. The Particle size of the produced powder was analyzed by a particle size analyzer (PAS; Malvern Instruments, Mastersizer 2000).

Li-ion battery powder H2SO4 H2O2

Leaching

Cu waste

Na2S solution

Cu precipitation

Co oxalate precipitation

3. Results and discussion 3.1. Physical treatment and leaching of spent LIBs

Oxalic acid solution

The mass and metal composition of different size fractions of spent LIBs after the physical treatment is listed in Table 1 [9]. The optimum leaching conditions for the 16 mesh fraction have been reported in a previous study [5]: 60 min at 100 g L1 pulp density, 2 M H2SO4, 6 vol% H2O2, 250 rpm, and 60 °C. The concentration of valuable metals in the leach liquor is shown in Table 2.

Calcination

Co3O4

Fig. 1. Flow sheet of cobalt oxide production from Li-ion batteries (conditions in Table 2).

3.2. Removal of Cu(II) from the leach liquor treated with a roll presser in order to short-circuit between the cathode and the anode. They directly fell into distilled water for discharging for 1 day. After dehydration and drying, discharged LIBs were subjected to crushing, magnetic separation, and screening to obtain three size fractions: +8 mesh, 8 + 16 mesh, and 16 mesh [9]. The 16 mesh fraction was selected for leaching. As shown in the flow sheet in Fig. 1, the leaching was conducted with 2 M sulfuric acid solution containing 6 vol% hydrogen peroxide. The addition of Na2S at the molar ratios of Na2S: Cu(II) in the ratio 3/1 to the leach liquor over a period of 30 min at 25 °C precipitated CuS. The residue was filtered and a solution of 1.5 M H2C2O4 was added to the filtrate to precipitate CoC2O42H2O over a period of 2 h. The composition of the precipitate was analyzed by AAS (Perkin Elmer, AAnalyst 400) and its crystal phases were analyzed by XRD (Rigaku, RU-200, Cu K-a). The oxalate product was heated for 2 h at 150–600 °C using an electric furnace to calcine to Co3O4 which was also analyzed by AAS and XRD. Morphology of the produced cobalt compounds was investigated by SEM (JEOL, JSM6380LA). Brunauer–Emmett–Teller (BET; Quantachrome Instruments, Quadrasorb SI) analysis was performed to determine the specific surface area (m2/g) and pore volume of the powders. The specific surface area (SBET) was determined by a multipoint BET method using the adsorption data in the relative pressure (P/P0)

Copper(II) must be removed from the solution before oxalic acid precipitation for recovery of cobalt to avoid the co-precipitation of CuC2O4. Copper(II) can be selectively separated from other metal ions in a wide range of solution pH. The sulfide precipitation method is used more widely than hydroxide and carbonate precipitation because of the low solubility of CuS in acid solutions. This is demonstrated in Fig. 2 which represents the solubility diagram for Cu(II), Ni(II), and Co(II) at 25 °C and 1 atm H2S based on reported thermodynamic data [11]. Lines for H+ and Cu2+ in Fig. 2 represent the equilibrium for Eqs. (1a) and (1b), respectively. Although Na2S was added as the precipitating agent in the present study, the solubility of metal ions as sulfides in acidic solutions is governed by the equilibrium in Eq. (1c) which represents the combined form of Eqs. (1a) and (1b):

H2 S ¼ 2Hþ þ S2

ð1aÞ

CuSðsÞ ¼ Cu2þ þ S2

ð1bÞ

Cu2þ þ H2 S ¼ CuSðsÞ þ 2Hþ

ð1cÞ 2

As predicted in Fig. 2 the free S ion concentration at pH 1 is 1017 M which corresponds to 1015 M Cu(II) compared to the concentration of Co(II) and Ni(II) which are greater than 104 M. However, the application of such predictions for quantitative treatment

Table 1 Mass and metal composition in different fractions of spent LIBs.

a

Fraction

Mass%

Co%

Li%

Cu%

Fe%

Mn%

Ni%

Al%

Magnetic material +8 mesh 8 + 16 mesh 16 mesh Total

3.5 49.4 7.6 36.5 97a

<0.01 9.3 2.1 11.8 23.3

<0.01 1.1 0.3 1.3 2.7

0.1 7.4 0.1 4.6 12.2

1.6 0.1 0.01 0.2 1.8

0.01 0.1 <0.01 <0.01 0.11

0.9 0.4 0.01 0.1 1.41

0.1 7.8 0.09 5.1 13.1

3% lost in the process (Shin et al. [9]).

Table 2 Metal composition of starting material and other products in the flow sheet.

a b c

Metal

Co

Fe

Cu

Li

Mn

Ni

Al

Mass% in feed (16 mesh) mg/L in leach liquor (pH 0.3)a Mass% in Co-oxalate productb Mass% in Co-oxide productc

11.8 27,400 35 79

1.3 88.6 0.03 0.1

4.6 962.4 0.01 0.01

0.2 4870 0.1 0.08

<0.01 15.7 0.3 0.6

0.1 71.1 0.2 0.4

– 1800 0.1 0.2

Leaching conditions for 16 mesh fraction: 60 min at 100 g L1 pulp density, 2 M H2SO4, 6 vol% H2O2, 250 rpm, and 60 °C. CoC2O42H2O obtained by adding oxalic acid to copper-free leach liquor. Co3O4 obtained by heating CoC2O42H2O for 2 h at 150–600 °C using an electric furnace.

177

J. Kang et al. / Advanced Powder Technology 21 (2010) 175–179 7

Co2+

-2

H

-4

5

Cu2+

-6

4

-8 -10

3

-12

2

-14

1

-16 0

-18 -20 -30

-1

-25

-20

-15

Before Cu removal After Cu removal

6

+

pH

2+

log {[M ]/ mol/l}

29000

Ni2+

-10

-5

log{[S2-] / mol/l}

Cocentration of valuable metal (mg/L)

0

28000 27000 26000 25000

3.3. Precipitation of cobalt from the leach liquor with oxalic acid Oxalic acid precipitation has been used for recovering metals [12,13]. Anion of the dibasic oxalic acid (pKa1 = 1.27 and pKa2 = 4.28) is an excellent ligand for metal ions. It usually binds to divalent metal ions as a bi-dentate ligand forming a five-membered MO2C2 ring. However, cobalt oxalate has a low solubility. Cobalt oxalate was precipitated (Eq. (2)) during the addition of oxalic acid when the molar ratio of H2C2O4 to Co(II) in the solution was 1.0

0.8

1000

500

Fig. 2. Metal sulfide precipitation diagram (25 °C, 1 atm H2S).

in actual precipitation processes is limited due to the fact that they do not take into account the association of cations with sulfate or bisulfate ions and the effect of dissolved salts on the saturated solubility of H2S. Fig. 3 shows the actual precipitation efficiency of Cu(II) as a function of molar ratio of Na2S to Cu(II) under conditions of pH 1, 25 °C and 30 min. As expected, the Cu(II) concentration rapidly decreased with increasing Na2S concentration. The need for a molar ratio of Na2S/Cu (II) = 3/1 for quantitative precipitation of CuS is due to the existence of equilibrium in Eq. (1c). Fig. 4 shows the composition of valuable metals in the solution before and after the precipitation of CuS. Copper was completely removed (99.9%) from the solution with Na2S. The concentration of other metals remained constant, whereas 11% Al was also removed during the precipitation of CuS.

1500

0 Co

Fe

Cu

Li

Mn

Ni

Al

Fig. 4. Metal ion concentration before and after copper removal by Na2S.

3:1. At this point, the volume ratio of the copper-free solution to the oxalic acid solution was 1:1.

Co2þ þ H2 C2 O4 ¼ CoC2 O4 ðsÞ þ 2Hþ

ð2Þ

The precipitate was separated by filtration. The oxalic acid remained in the solid residue was washed at 50 °C with distilled water and the residue was dried for 24 h in an oven. The chemical analysis of the final product was investigated by AAS. The residue contained 35% cobalt with only 0.2% Ni and 0.3% Mn as impurities (Table 2). As shown by the XRD analysis in Fig. 5, the product was confirmed to be crystalline CoC2O42H2O. Fig. 6 confirms the change of crystalline form to CoC2O4 at 150 °C, and also to Co3O4 at higher temperatures. The final product was crystalline Co3O4 (Fig. 6) which contained 79% of cobalt with other impurities: 0.1%, Fe, 0.4% Ni and 0.6% Mn. 3.4. Characterization of produced cobalt compounds The powder characteristics of the produced cobalt oxalate and cobalt oxide were investigated by SEM, BET, and PSA such as morphology, specific surface area, and particle size. Fig. 7 and 8 show SEM images of the cobalt oxalate and the cobalt oxide, respectively. The morphology of cobalt oxalate was columnar structure (Fig. 7(a), (b)). The SBET value of the produced cobalt oxalate was 32.6 m2/g and the pore volume was 0.076 cc/g between 1 nm and 300 nm. The average particle diameter (d0.5) measured by PSA was 16.4 lm.

CoC 2O 4 2H 2O

0.4

Intensity

C/Co ratio

0.6

0.2

0.0 0

1

2

3

Na2S/Cu molar ratio

0

10

20

30

40

50

60

70

2θ(degree) Fig. 3. Effect of Na2S on removal of Cu from leach liquor (Co: initial Cu concentration in feed, C: residual Cu concentration in solution).

Fig. 5. XRD analysis precipitated Co-oxalate.

80

90

178

J. Kang et al. / Advanced Powder Technology 21 (2010) 175–179

0.0014 Accumulated pore volume (cc/g)

0.007

o

600 C

Co3O4

0.0012 Co3O4

o

350 C

o

300 C

Co3O4

dV(d) (cc/nm/g)

Intensity

0.0010

o

150 C

CoC2O4

Precipitate

CoC2O4·· 2H2O

0

10

20

30

40

50

60

70

0.004 0.003 0.002 0.001 0.000

0.0008

1

10

100

Diameter (nm)

0.0006

80 0.0002

Fig. 6. XRD analysis showing formation of Co-oxide from Co-oxalate.

0.0000

The morphology of produced cobalt oxide calcined kept in the columnar structure of cobalt oxalate (Fig. 8(a)) but polyhedral single crystal was recrystallized (Fig. 8 (b)). The primary particle size obtained from SEM images ranged approximately 200–700 nm. The SBET value of produced cobalt oxide was 2.9 m2/g. Fig. 9 shows pore size distribution of the cobalt oxide obtained by BJH method from desorption isotherm. Since the pore volume was very low, 0.006 cc/g, average primary particle diameter can be calculated by spherical approximation (Eq. (3)).

6 So q

0.005

0.0004

2θ(degree)

d¼

0.006

ð3Þ

Where, d is an average particle diameter and So is the specific surface area. q is the density of Co3O4 (6.11 g/cm3). Average primary particle diameter obtained by Eq. (3) was 340 nm which well corresponded with the primary particle size observed by SEM image (Fig. 8(b)).

1

10

100

Diameter (nm) Fig. 9. Pore size distribution of Co-oxide.

4. Summary and conclusions Spent LIBs were physically treated for the separation of magnetic material, crushed and classified into magnetic materials (3.5%) and three size fractions: +8 mesh (49.4%), 8 + 16 (7.6%) mesh and 16 mesh (36.5%). The 16 mesh fraction of the LIBs was leached with sulfuric acid, Cu was eliminated with Na2S, and resulting solution was treated with oxalic acid for the recovery of cobalt. The leach liquor contained 27.4 g/L Co, 4.87 g/ L Li, 88.6 mg/L Fe and 962.4 mg/L Cu as impurities. More than 99.9% Cu was precipitated and removed by the addition of Na2S to the leach liquor at a Na2S to Cu molar ratio of 3 to 1.

Fig. 7. SEM images of produced Co-oxalate ((a):850; (b):3000).

Fig. 8. SEM images of prepared Co-oxide ((a):3000; (b):13,000).

J. Kang et al. / Advanced Powder Technology 21 (2010) 175–179

The cobalt oxalate precipitate was calcined to oxide as the final product. The copper-free liquor was treated with oxalic acid to precipitate the crystalline form of CoC2O42H2O. Cobalt oxide in the crystalline form of Co3O4 was successfully obtained by calcining CoC2O42H2O at 150–300 °C for 2 h. Morphology of cobalt oxalate was columnar structure. In the case of cobalt oxide, recrystallized polyhedral single crystal was observed. A primary average particle size of prepared cobalt oxide was 340 nm. Overall 98% cobalt was recovered through this process from spent LIBs. Acknowledgement This study was supported by a grant operated by Korea Environmental Industry & Technology Institute (KEITI) of the Ministry of Environment of Korea. The authors would like to thank KEITI for the financial support. Support from the Parker CRC for Integrated Hydrometallurgy Solutions is also gratefully acknowledged. References [1] J. Nan, D. Han, X. Zuo, Recovery of metal values from spent lithium-ion batteries with chemical deposition and solvent extraction, J. Power Sources 152 (2005) 278–284. [2] D.I. Ra, K.S. Han, Used lithium ion rechargeable battery recycling using EtoileRebatt technology, J. Power Sources 163 (2006) 284–288.

179

[3] A. Lundblad, B. Bergman, Synthesis of LiCoO2 starting from carbonate precursors 2. Influence of calcination conditions and leaching, Solid State Ionics 96 (1997) 183–193. [4] C.M. Sabin, Battery paste recycling process, US Patent 5690718, 1997. [5] S.M. Shin, N.H. Kim, J.S. Sohn, D.H. Yang, Y.H. Kim, Development of a metal recovery process from Li-ion battery wastes, Hydrometallurgy 79 (2005) 172– 181. [6] J.S. Sohn, S.M. Shin, D.H. Yang, S.K. Kim, C.K. Lee, Comparison of two acidic leaching processes for selecting the effective recycle process of spent lithium ion battery, Geosystem Eng. 9 (2006) 1–6. [7] P. Zhang, T. Yokoyama, O. Itabashi, T.M. Suzuki, K. Inoue, Hydrometallurgical process for recovery of metal values from spent lithium-ion secondary batteries, Hydrometallurgy 47 (1998) 259–271. [8] M. Contestabile, S. Panero, B. Scrosati, A laboratory-scale lithium-ion battery recycling process, J. Power Sources 92 (2001) 65–69. [9] S. M. Shin, J. S. Sohn, D. H. Yang, T. H. Kim, J. G. Kang, J. H. Park, Safe method for dismantling spent lithium-ion secondary batteries, KR Patent 10-0860972, 2008. [10] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T. Siemieniewska, Reporting, physisorption data for gas/solid systems with special reference to the determination of surface area and porosity, Pure Appl. Chem. 57 (1985) 603–619. [11] E. Jackson, Hydrometallurgical extraction and reclamation, John Wiley & Sons, New York, USA, 1986. pp. 147–153. [12] T.T. Fang, H.B. Lin, Factors affecting the preparation of barium titanyl oxalate tetrahydrate, J. Am. Ceram. Soc. 72 (1989) 1899–1906. [13] I.I. Tarasova, A.W.L. Dudeney, S. Pilurzu, Glass sand processing by oxalic acid leaching and photocatalytic effluent treatment, Miner. Eng. 14 (2001) 639– 646.