Separation and recovery of cadmium(II), cobalt(II) and nickel(II) from sulphate leach liquors of spent Ni–Cd batteries using phosphorus based extractants

Separation and Purification Technology 50 (2006) 161–166

Separation and recovery of cadmium(II), cobalt(II) and nickel(II) from sulphate leach liquors of spent Ni–Cd batteries using phosphorus based extractants B. Ramachandra Reddy a,∗ , D. Neela Priya a , Kyung Ho Park b a

b

Inorganic Chemistry Division, Indian Institute of Chemical Technology (CSIR), Hyderabad 500007, India Minerals and Materials Processing Division, Korea Institute of Geoscience and Mineral Resources (KIGAM), Daejeon 305-350, Republic of Korea Received 26 May 2005; received in revised form 12 November 2005; accepted 16 November 2005

Abstract Solvent extraction separation and recovery of cadmium(II), cobalt(II) and nickel(II) from sulphate leach liquor of spent Ni–Cd batteries was investigated using phosphorus based extractants such as TOPS 99, Cyanex 923, Cyanex 272, Cyanex 302 and Cyanex 301 diluted in kerosene. The composition of the leach liquor used for the present study contains (g/L): Cd, 2.40; Ni, 5.94; Co, 0.05. Among the phosphorus based extractants screened at 0.1 M concentration as a function of aqueous phase equilibrium pH, Cyanex 301 showed selective separation of Cd(II) from Co(II) and Ni(II). On the other hand, stripping of metal from Cd–Cyanex 301 loaded organic (LO) phase needs high acid solution. Two-stage counter-current extraction with 0.06 M Cyanex 301 at unit phase ratio and three-stage stripping of the metal from LO with 6 M HCl at an aqueous to organic (A:O) phase ratio of 2 yielded >99.9% Cd(II) extraction and stripping efficiency. However, Cd(II) concentration in strip solution was reduced to half. Cobalt(II) extraction efficiency of ∼99% was achieved from cadmium raffinate at an equilibrium pH of 6.25 with 0.03 M Cyanex 272 in two counter-current stages at an A:O ratio of 2:1. Complete stripping of metal from LO containing 0.15 g/L Co(II) was carried out using a strip solution pH of 1.0 in two stages at an O:A ratio of 1.75. The enrichment of cobalt during extraction and stripping operations was ∼5.2 times. Complete process flowsheet for the separation and recovery of Cd(II), Co(II) and Ni(II) was proposed. © 2005 Elsevier B.V. All rights reserved. Keywords: Spent batteries; Sulphate leach liquor; Solvent extraction separation; Metal recovery; Phosphorus based extractants

1. Introduction Batteries are popular source of portable energy. The principal applications include illumination, communication and entertainment. Nickel–cadmium batteries represent the kind of secondary batteries that are used in many portable electronic devices, military and defense applications. Nickel is a useful metal whereas cadmium is toxic and poisoning occurs through inhalation and ingestion. Discarded waste Ni–Cd batteries into the environment results in pollution of soil and water. As a result, these are being replaced with nickel–metal hydride batteries to some extent. Recycling of spent batteries will be important, both from an environmental and economic point of view. In several European countries, the collection rates of spent batteries vary between 32 and 54% of battery sales. To the best of our knowledge, data on

∗

Corresponding author. Tel.: +91 40 27193510; fax: +91 40 27160921. E-mail address: brcreddy [email protected] (B.R. Reddy).

1383-5866/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.seppur.2005.11.020

disposal and recycling of spent Ni–Cd batteries is not available in India. Several methods to recover metal values from spent batteries are already reported in the literature but industrial routes generally employ either pyrometallurgical/hydrometallurgical processes such as the SNAM-SAVAN, SAB-NIFE and INMETCO [1–3], which employ basically selective volatilization of metals at elevated temperatures around 900 ◦ C followed by condensation to recover cadmium. High pure cadmium (>99%) is obtained, but most valuable metals such as nickel and cobalt are not usually recovered. Therefore, hydrometallurgical methods such as solvent extraction (SX) are found more economical and efficient which are characterized by low energy consumption, higher metal selectivity, high purity products and avoid effluent emissions. A number of hydrometallurgical methods involving acid leaching of battery active materials followed by recovery of Cd, Ni and Co by precipitation, ion exchange and solvent extraction methods were reported. Bartolozzi et al. [4] reported hydromet-

162

B.R. Reddy et al. / Separation and Purification Technology 50 (2006) 161–166

allurgical methodology for the recovery of cadmium and nickel from spent batteries by chemical and electrochemical methods. Xue et al. [5] proposed a method with initial acid leaching followed by precipitation of iron and manganese hydroxides; nickel was precipitated as a double salt at first and finally recovered as nickel hydroxide and cadmium was then precipitated as carbonate. The BATENUS [6] process provides a solution for the treatment of mixed battery waste, using sulphuric acid leaching and a combination of ion exchange and solvent extraction for metal recycling and purification. Solvent extraction flowsheet for the recovery of Cd, Co and Ni from synthetic sulphate solutions using D2EHPA, Ionquest 801 and Cyanex 272 was reported. Cd extraction with D2EHPA involves co-extraction of both Ni and Co, which needs scrubbing stage with pure Cd solution [7]. Solvent extraction studies of cadmium with possible stoichiometry of extracted species from acidic chloride media using Cyanex 302 [8], selective extraction of Zn from Co from sulphate solutions by organophosphorus acidic extractants [9] and Co–Ni separation by Cyanex 301 and Cyanex 302 was studied [10], but the stripping behavior of these metals has not been reported. Application of phosphorus based extractants such as D2EHPA, PC 88A, Cyanex 272 for the solvent extraction studies of Ni at macro-level concentrations from sulphate/chloride solutions [11], separation studies of Co(II)/Ni(II) from synthetic chloride solutions containing 0.5 g/L Co and 0.5/2 g/L Ni concentrations by the sodium salts of extractants and their mixtures [12], separation of Co and Ni from sulphate solutions of Indian Ocean nodules using Cyanex 272 [13], solvent extraction studies of Cd from sulphate solutions [14] and separation of Cd(II), Ni(II) and Co(II) from chloride leach liquor of spent batteries [15] were explored in our earlier studies. In this paper we report, studies carried out on the screening of suitable extractant and development of a hydrometallurgical route for the separation and recovery of Cd(II), Co(II) and Ni(II) from sulphate leach liquor of spent nickel–cadmium batteries using phosphorus based extractants, Cyanex 301, Cyanex 272 and TOPS 99 diluted in kerosene. The parameters studied for optimization of process are: leaching, effect of equilibrium pH, selection of extractant and its concentration, phase ratios, extraction and stripping isotherms, counter-current extraction and stripping simulations. 2. Experimental 2.1. Apparatus A Perkin-Elmer Model A300 Atomic Absorption Spectrophotometer (AAS) and a digital Digisun pH meter (model DI 707) were used for the measurement of metal concentration and pH in the aqueous phase. 2.2. Reagents The spent Ni–Cd batteries used in this work were prismatic shaped batteries supplied by HBL-NIFE Industries, Hyderabad, India. These are low rate cells with a capacity of 7 A and 1.2 V used in radio military communications. The dismantled elec-

trodes were washed thoroughly with distilled water and dried overnight around 100 ◦ C. The powdery materials from the wire mesh were scrapped and ground to −100 ␮m. Chemical analysis of the spent Ni–Cd battery powder indicated 61.3% Ni, 21.4% Cd and 0.5% Co. Cyanex 301, bis(2,4,4-trimethylpentyl)dithiophosphinic acid; Cyanex 302, bis(2,4,4-trimethylpentyl)monothiophosphinic acid; Cyanex 923, a mixture of four trialkylphosphine oxides (R3 P O, R2 R P O, R3 P O, R2 RP O, where R = hexyl and R = octyl); Cyanex 272, bis(2,4,4trimethylpentyl)phosphinic acid obtained from Cytec Canada Inc.; TOPS 99 (an equivalent of di-2-ethylhexyl phosphoric acid) from Heavy Water Plant, Talchar, India, were used as such with out purification. Distilled kerosene (b.p.: 160–200 ◦ C) mostly aliphatic (96.2%) was used as the diluent. All other chemicals used were analar grade. 2.3. Leaching studies Leaching experiments were conducted by taking 200 mL H2 SO4 (2.5 M) in a 500 mL round bottom flask fitted with water cooled condenser and stirrer, which is immersed in oil bath maintained at 85 ± 1 ◦ C. As soon as the temperature of flask reached 85 ± 1 ◦ C, weighed amounts (2 g) of sample was added and samples were with drawn periodically at different time intervals till the sample gets dissolved totally. It was observed that 8 h time is required for the total dissolution of the sample. Analysis of Cd, Ni and Co metal contents was determined after proper dilution by AAS to estimate the percentage leaching. The composition of leach liquor with H2 SO4 under optimum conditions (H2 SO4 : 2.5 M, temperature: 85 ◦ C, solid/liquid ratio: 1, time: 8 h) was (g/L): Cd, 2.40; Ni, 5.94; Co, 0.05. pH of the liquor was 0.3. 2.4. Solvent extraction procedure Suitable volumes of aqueous (leach liquor) and organic phases were contacted in separating funnels for 5 min (initial experiments showed that equilibrium was reached within 1 min), the phases were separated and the metal concentration in the aqueous phase (raffinate) was estimated directly or after suitable dilution. pH adjustment was performed by the addition of dilute NaOH/H2 SO4 . The loaded organic (LO) phases were stripped with respective HCl/H2 SO4 acid concentrations after which the strip solutions were diluted to the required concentration and analyzed for metal values by AAS. All the experiments were carried out at room temperature (30 ± 1 ◦ C). The distribution ratio, D, was calculated as the concentration of metal present in the organic phase to that part in the aqueous phase at equilibrium. 3. Results and discussion 3.1. Influence of equilibrium pH and extractant concentration on Cd(II) extraction Preliminary experiments on effect of equilibrium pH in the range from 1 to 7 on the extraction of metals with 0.1 M of TOPS 99 and Cyanex 923 showed low extraction efficiencies and non-

B.R. Reddy et al. / Separation and Purification Technology 50 (2006) 161–166

163

Fig. 2. McCabe–Thiele plot for cadmium extraction using 0.06 M Cyanex 301.

Fig. 1. (a and b) Effect of equilibrium pH on the percentage extraction of metals using 0.1 M extractants.

selectivity (Fig. 1a). Extraction of metals using 0.1 M Cyanex 302 showed systematic increase in the percentage extraction of Cd(II) from 75 to 99.5% with increase in pH of aqueous phase from 0.3 to 3.0. The co-extraction of Ni and Co varied from 4.4 to 31.8 and 0 to 15.4%, respectively (Fig. 1b). On the other hand, Cyanex 301 showed quantitative extraction of Cd(II) even at low pH of 0.3 (Fig. 1b) and nil extraction of Ni(II) and Co(II). Further, the variation of extractant concentration from 0.01 to 0.1 M at pH 0.3 (Table 1) indicated >99% extraction of Cd(II) with 0.06 M Cyanex 301. The co-extraction of Ni and Co was nil at this pH, indicating clear separation of Cd(II). Therefore, in the present study, Cyanex 301 was used for the extraction of Cd(II) from sulphate leach liquor. 3.1.1. Cadmium extraction To determine the number of stages required at a chosen volume phase ratio, the extraction data was obtained by contactTable 1 Effect of extractant concentration on percentage extraction of cadmium [Cyanex 301] (M) Cd extraction (%)

0.01 65.4

0.02 69.1

0.03 73.0

0.04 79.7

0.06 99.6

0.08 100.0

0.1 100.0

ing the leach liquor (pH 0.3) and 0.06 M Cyanex 301 (99.6% Cd(II) extraction efficiency at unit phase ratio) at different A:O phase ratios from 1 to 2 and O:A phase ratios from 1 to 6. The McCabe–Thiele plot (Fig. 2) indicates that complete cadmium extraction is possible in two counter-current stages at A:O phase ratio of unity. To confirm this, a two-stage countercurrent extraction simulation (CCES) test was carried out at the A:O phase ratio of unity, which resulted in a raffinate containing 1 mg/L of Cd(II) (99.96% extraction efficiency), 0.05 g/L of Co(II) and 5.94 g/L of Ni(II) and a loaded organic containing 2.399 g/L Cd(II). Co-extraction of Ni(II) and Co(II) was nil. Sufficient quantities of loaded organic phase and raffinates were generated by carrying out a two-stage CCES under the above conditions. The LO thus obtained was used for generating data on stripping and the raffinate was used to optimize conditions for Co(II) recovery. 3.1.2. Stripping of cadmium from loaded organic The generated LO contains 2.399 g/L Cd. Stripping of Cd(II) from Cyanex 301 loaded organic has not been reported elsewhere. However stripping studies with various reagents, viz. HCl, H2 SO4 , thiourea and oxalic acid of different concentrations was carried out to see the stripping behavior of cadmium from the Cd(II) LO (Table 3). It was observed that, stripping efficiency of Cd(II) from the LO was 50% even at higher concentrations of HCl (6–8 M) indicating strong Cd–Cyanex 301 complex formation and requires theoretically more than seven stages for quantitative recovery of metal if operated at unit phase ratio. The efficiency of the stripping reagents in the acid range 1–6 M follows the increasing order: oxalic acid ≈ thiourea < H2 SO4 < HCl. In order to increase the stripping efficiency and decrease the stages required for quantitative stripping of Cd(II) from LO, stripping was carried out further with 6 M HCl at different O:A phase ratios from 1:1 to 1:3, which resulted in increasing stripping efficiencies from 50 to 92.7%, respectively (Table 2). The results further indicate that stripping of Cd(II) from the loaded organic phase requires strong acidic conditions and high A:O phase

164

B.R. Reddy et al. / Separation and Purification Technology 50 (2006) 161–166

Table 2 Effect of various reagents and phase ratio on stripping of cadmium from LO Stripping agent

Phase ratio (A:O)

Concentration (M)

Stripping efficiency (%)

Thiourea Oxalic acid

1:1 1:1

1–2 1

<0.4 0.1

H2 SO4

1:1 1:1

1–6 18

<0.7 37.6

HCl

1:1 1:1 1:1 1:1 1:1 1:1 1.5:1 2.0:1 3.0:1

1 2 3 4 5 6–8 6 6 6

3.0 15.3 29.2 35.6 46.1 50 73.8 91.8 92.7

ratios to obtain quantitative efficiencies. Based on these results, an A:O ratio of 2 was selected, which can yield >99% stripping efficiency of Cd from LO, when operated in 2/3 stages. A three-stage counter-current stripping simulation (CCSS) test was carried out at A:O phase ratio of 2:1 with 6 M HCl to generate representative strip solution (SS) and spent organic (SO). The combined strip solution and spent organic were collected. SO phase was found to contain 1 mg/L Cd corresponding to a stripping efficiency of >99.9%. Strip solution contains 1.199 g/L Cd and strongly acidic. 3.2. Processing of cadmium-free raffinate for cobalt recovery using Cyanex 272 The raffinate after Cd(II) extraction, contains Co: 0.05 g/L and Ni: 5.94 g/L with a pH of 0.3, was used in optimizing conditions for cobalt recovery. It is already reported from the studies on selectivity of phosphorus based extractants, Cyanex 272 appears to be the best, further studies on separation of Co from Ni was carried out using this extractant [12,13,15].

Fig. 3. Effect of equilibrium pH on the percentage extraction of cobalt and nickel from cadmium raffinate using 0.01 and 0.03 M Cyanex 272.

phosphinic acid. But most of the studies are concerned with synthetic solutions of low and equal concentrations of Co and Ni, where as in the present study, cadmium raffinate contains Co to Ni in the ratio 1:119. To optimize the conditions with respect to phase ratio and number of stages required for the quantitative extraction of cobalt, extraction of metal was obtained at different A:O phase ratios from 1:3 and 7:1 using 0.03 M Cyanex 272 at an aqueous phase equilibrium pH of 6.25 (Fig. 4). The coextraction of Ni varied from 0.006 to 0.064 and 0.18 to 0.01 g/L (data presented in Fig. 4 at each phase ratio) corresponding to 0.1–3.1 and 0.43–0.1%, respectively. The results indicate that complete cobalt extraction is possible in two counter-current stages at the A:O phase ratio of 2:1. Considering the percent extraction, phase ratio, minimum stages required for complete extraction of Ni and at the same time possible enrichment of metal in the organic phase, A:O phase ratio of 2 was selected

3.2.1. Effect of equilibrium pH on extraction of cobalt From the cadmium raffinate, extraction of cobalt and nickel as a function of equilibrium pH in the range 5.4–7.0 was carried out at unit phase ratio using 0.01 and 0.03 M Cyanex 272 (Fig. 3). Percentage extraction of Co increased with increase in the pH of the aqueous phase. Comparison of the results indicate that an equilibrium pH of 7 was needed to achieve >99% Co efficiency using 0.01 M extractant concentration and the co-extraction of Ni was 0.72%. On the other hand, similar extraction efficiency of Co was achieved at about pH 6.4 using 0.03 M extractant and co-extraction of Ni was less (0.5%). Based on these results, an equilibrium pH of 6.25 and extractant concentration of 0.03 M was selected (cobalt extraction of >96% in single stage) for the cobalt extraction isotherm at different phase ratios in order to determine the number of stages required at a chosen ratio. 3.2.2. Cobalt separation It is well reported that the separation factor for Co–Ni separation increases in the order, phosphoric < phosphonic <

Fig. 4. McCabe–Thiele plot for cobalt extraction using 0.03 M Cyanex 272.

B.R. Reddy et al. / Separation and Purification Technology 50 (2006) 161–166

165

Table 3 Scrubbing of nickel from LO with acidified distilled water/cobalt sulphate Equilibrium pH

Co in scrub raffinate (mg/L)

Ni in scrub raffinate (mg/L)

Ni scrubbing efficiency (%)

2.2 3.1 4.1 4.8 5.2 5.80a 6.10a 6.25a 6.30a

99 85 7.0 0.4 0.5 3.8 2.7 1.1 Nil

50 50 41 23 18 46.2 47.3 48.9 50

100 100 82 46 36 92.4 94.6 97.8 100

a

Scrubb feed: 50 mg/L cobalt sulphate.

in order to extract Co in two counter-current stages. To confirm this, a two-stage CCES test was carried out at A:O phase ratio of 2:1, which resulted in raffinate containing 0.37 mg/L of cobalt corresponding to 99.3% extraction efficiency. The co-extraction of nickel into the loaded organic was 50 mg/L, corresponding to 0.42%. The loaded organic (Co: 0.099 g/L and Ni: 0.05 g/L) thus obtained was used for optimizing conditions for nickel scrubbing and cobalt stripping. Nickel scrubbing from LO was carried out with acidified distilled water in the pH range 2.2–5.2 (Table 3). It was observed that scrubbing of nickel alone was difficult, because cobalt was also stripped along with nickel. Therefore, CoSO4 scrub feed of 50 mg/L was chosen to scrub nickel in the equilibrium pH range from 5.8 to 6.3. The results clearly demonstrate quantitative removal of Ni from LO with a scrub feed equilibrium pH of 6.3. The scrubbed LO contains 0.149 g/L Co corresponding to quantitative scrubbing efficiency.

Fig. 5. McCabe–Thiele plot for cobalt stripping.

3.2.3. Stripping of cobalt from the loaded organic Stripping of cobalt from the scrubbed LO (SLO) at unit phase ratio with distilled water adjusted to pH values of 1, 2 and 2.5 resulted corresponding equilibrium pH of 1.1, 1.6 and 1.95 and stripping efficiencies of 100, 96.6 and 81%, respectively. In order to find out the extent of enrichment possible and also the theoretical number of stages required at a chosen A:O phase ratio, the stripping isotherm was obtained at different O:A ratios using the strip feed (distilled water adjusted to pH 1) and loaded organic. The McCabe–Thiele plot (Fig. 5) indicates that quantitative stripping of Co is achievable in two CCSS at A:O phase ratio of 1:1.75. To confirm the stripping isotherm prediction data,

Fig. 6. Flowsheet of the process for the recovery of Cd(II), Co(II) and Ni(II) from sulphate leach liquor of spent Ni–Cd batteries.

166

B.R. Reddy et al. / Separation and Purification Technology 50 (2006) 161–166

a two-stage CCSS test carried at A:O phase ratio of 1:1.75 gave spent organic containing 0.25 mg/L Co corresponding to 99.83% stripping efficiency. The enrichment of cobalt during extraction and stripping operations was ∼5.2 times. The raffinate from cobalt extraction circuit, contain 5.915 g/L of nickel with a pH of 6.25 for nickel recovery. Complete process flowsheet describing the phase ratio, number stages for the quantitative separation and recovery of Cd(II) and Co(II) from sulphate leach liquor of spent batteries is presented in Fig. 6. 4. Conclusions A complete hydrometallurgical process for the separation and recovery of cadmium(II), cobalt(II) and nickel(II) from sulphate leach liquor of spent batteries was developed on laboratory scale by solvent extraction route using phosphorus based extractants. Cyanex 301 and Cyanex 272 were selectively used for Cd(II) and Co(II) separations, respectively. Though, Cyanex 301 showed selective separation of Cd from Ni and Co, but the stripping of metal from loaded organic needs strong acid solutions. The generated Cd(II), Co(II) and Ni(II) salts are in pure form with a recovery efficiency >99.9%. The enrichment of Co(II) during extraction and stripping stages was about six times. Finally, the present approach demonstrates solving the environmental related issues by proper treatment methods of generated wastes and at the same time recovering the valuable metals. Acknowledgements Sincere thanks to Ministry of Environment & Forests (MOEF), Government of India, New Delhi, India, for the finan-

cial support under Project No. 19-125/99-RE. DNP thanks the Council of Scientific and Industrial Research (CSIR), India, for the award of Senior Research Fellowship, and BRR thanks the Korean Foundation of Technology and Science (KOFTS), Korea, for the award of Brain pool program. References [1] M.E. Schweers, J.C. Onuska, R.K. Hanewald, A Pyrometallurgical Process for Recycling Cadmium Containing Batteries, Proceedings of the HMC-South’92, New Orleans, 1992, p. 333. [2] T. Anulf, in: S.A. Hiscock, R.A. Volpe (Eds.), SAB-NIFE Recycling Concept for Nickel–Cadmium Batteries—An Industrialized and Environmentally Safe Process, Proceedings of the 6th International Cadmium Conference on Cadmium Association, London, UK, 1990, p. 161. [3] R.H. Hanewald, W.A. Munson, D.L. Shweyer, Miner. Metall. Process 9 (1992) 169. [4] M. Bartolozzi, G. Braccine, S. Bonvini, P.F. Marconi, J. Power Sources 55 (1995) 247. [5] Z. Xue, Z. Hua, N. Yao, S. Chen, Sep. Sci. Technol. 27 (1992) 213. [6] S. Frohlich, D. Sewing, J. Power Sources 57 (1995) 27. [7] C.A. Nogueira, F. Delmas, Hydrometallurgy 52 (1999) 267. [8] A. Almela, M.P. Elizalde, Hydrometallurgy 37 (1995) 47. [9] B. Gajda, W. Apostoluk, Ars Separatoria Acta 1 (2002) 55. [10] A.M.M. Daudinot, E.G. Liranza, in: K.C. Sole, P.M. Cole, J.S. Preston, D.J. Robinson (Eds.), Cobalt–Nickel Separation by Solvent Extraction in Cuba, Proceedings of the International Solvent Extraction Conference, ISEC, Johannesburg, South Africa, March 17–21, 2002, p. 964. [11] P.V.R. Bhaskara Sarma, B.R. Reddy, Miner. Eng. 15 (2002) 461. [12] K. Sarangi, B.R. Reddy, R.P. Das, Hydrometallurgy 52 (1999) 253. [13] B.R. Reddy, P.V.R. Bhaskara Sarma, Miner. Metall. Process 18 (2001) 172. [14] B. Ramachandra Reddy, D. Neela Priya, J. Rajesh Kumar, Hydrometallurgy 74 (2004) 277. [15] B. Ramachandra Reddy, D. Neela Priya, S. Venkateswara Rao, P. Radhika, Hydrometallurgy 77 (2005) 253.