Desalination 190 (2006) 7988

Recovery of metals from complexed solutions by electrodeposition

Y. Oztekin*, Z. Yazicigil
Department of Chemistry, Selcuk University, Campus, Konya 42075, Turkey Tel. +90 (332) 223-1263; Fax: +90 (332) 241-0106; email:yoztekin@selcuk.edu.tr Received 9 May 2005; accepted 20 July 2005

Abstract Electrolytic recovery of metals from aqueous solutions containing complexing chelating agents such as EDTA, NTA, and citrate was studied in a two-chamber cell separating with a commercial cation-exchange membrane (CEM). Equimolar solutions of metal and a chelating agent as a catholyte and NaNO3 as an anolyte were used; the effect of current densities, initial catholyte and anolyte pH, metal concentration and the type of the CEM, chelating agent and metal on the recovery of metals was determined. The recovery of metal increased with higher initial anolyte pH, concentration and current density, whereas it decreased with lower initial catholyte pH. The results show that electrodeposition seems to be an applicable method for the recovery of metals under appropriate conditions. Keywords: Electrodeposition; Recovery of metals; Chelating agent; Cation-exchange membrane

1. Introduction Chelating agents (citrate, nitrilotriacetic acid and ethylenediaminetetraacetic acid) have been used in many industrial applications such as the metal plating and finishing industries, water softening, photography, textile processing and industrial cleaning. They bind with metal cations, preventing them from forming insoluble salts or other process problems [1,2], but they complicate the treatment of industrial polluted effluents by
*Corresponding author.

reducing the efficiency of metal removal using traditional chemical precipitation ion exchange and other processes [1]. Citrate, nitrilotriacetic acid (NTA) and ethylenediaminetetraacetic acid (EDTA) may make the chemical precipitation methods involving hydroxides and sulfides less effective, even with relatively high levels of metals. Moreover, the high buffer capacity provided by the complexing agents requires excessive amounts of chemicals to neutralize the alkalinity. These problems and restrictions warrant investigations to other recovery alternatives [3]. The recovery methods, which depend greatly

doi:10.1016/j.desal.2005.07.017

0011-9164/06/$ See front matter 2006 Elsevier B.V. All rights reserved

80

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

on the particular chelating agent and metals, as well as their concentrations, are generally grouped into three categories: chemical, physical and electrochemical. Among them, the electrochemical method seems to be particularly advantageous for concentrated solutions. The main concern becomes the recovery and reuse of metals and complexing agents [4,5]. Regardless of which concentration method is used, however, the chelated metals cannot be destroyed, and the final step should be aided via electrolysis operations [6,7]. For diluted solutions, ion-exchange resins [8,9], adsorption onto activated carbon [10,11], inorganic materials [12] and functional polymers [13] have been tried. However, it was seen that EDTA was oxidized into many compouds, including CO2, formaldehyde and ethylendiamine, during metal deposition on a Pt anode in acidic sulfate solutions [14]. Thus, simultaneous recovery of metals and chelating agents is not possible using common electrolytic methods. One of the most important techniques in these electrolytic methods, electrodeposition, has been usually applied for the recovery of metals from wastewater or electroplating [15]. Copper electrodeposition is a process of great practical importance, for example, in the electronics industry in which hard, uniform and bright deposits are desired and the formation of loose deposits has to be prevented [16]. In this work, attention was paid to examining the recovery of binary complexed solutions, Cu(II), Co(II) and Ni(II) EDTA, NTA and citrate by using commercial cation-exchange membranes (CEM) with electrodeposition. These metals were selected because the literature results of each single metal system were available for comparison. Experiments were carried out by taking at equimolar chelating agents metals as a function of current density, initial catholyte and anolyte pH, types of membranes, types of metals, and types of chelating agents, metal concentration and their concentration ratio.

2. Experimental 2.1. Materials All chemicals were purchased from Merck and were reagent grade. The catholyte solution was prepared by dissolving metal from the sulfate form and chelating agent in deionized water at equimolar. The pH of the resulting solution was between 2.80 to 3.30. When the effect of the initial catholyte pH was examined, the solution pH was adjusted to be in the range of 1.0 to 8.0 by adding a small amount of HCl or NaOH. The anolyte solution was 0.1 M NaNO3 and its initial pH was around 5.98. When the effect of the initial anolyte pH was examined, the solution pH was adjusted to a range of 1.08.0 by adding a small amount of HCl or NaOH. The CEM selected were Neosepta CMB, CMS, and CMX (kindly provided by Eurodia). They are homogeneous films comprising cross-linked sulfonated copolymers of vinyl compounds on synthetic reinforcing fabric (Table 1) [17]. 2.2. Preparation of membranes The following steps were conducted for all membrane samples (7.065 cm2): Prior to use, all membranes were treated with distilled water at 70EC for 1 h, 0.1 M HCl and 0.1 M NaOH at 50EC for 1 h, respectively, to remove inorganic impurities. The resulting membranes were finally immersed in 1 M HCl solutions to transform
Table 1 Properties of the CEM (functional group: polysulphone) CMB Ion-exchange capacity, meq/g Thickness, m Water content, % Burst strength, -cm2 2.72.8 220260 3341 3.05.0 CMS 2.02.5 140170 3545 1.53.5 CMX 1.51.8 170190 2530 2.03.5

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

81

membranes to their hydrogen form for 24 h and were used for the electrodeposition experiments. 2.3. Diagram for electrodeposition experiments The apparatus for the electrodeposition experiments, made from polyamide, is illustrated in Fig. 1. Two chambers were attached by means of inserting two screws on their flanges with a CEM in between. The membrane was cut as a circle with a diameter of 30 mm and glued onto the inner mounting wall of the two chambers with silicone rubber sealant to prevent any leakage between the chambers. The electrodes were fitted in the inner wall of the apparatus and were symmetrically separated about 65 mm from the membrane surface. Stainless steel and Pt were used as the cathode and anode, respectively, and their active dimensions were 30 mm. The flow of solutions was performed by a peristaltic pump (913 Mity Flex, double-way). The power supply used was capable of supplying a potential of up to 32 V and DC up to 2 Amp.

2.4. Experimental procedures and chemical analysis An equal volume (250 mL) of the anolyte and catholyte solution was filled in the respective chambers. The experiments were carried out at the constant current by changing the cell voltage according to the variation of the current. The pH values of the anolyte and the catholyte solutions were measured with an Orion EA 940 pH meter. As expected, vigorous mixing reduced the thickness of the double layer adjacent to the electrodes, thereby enhancing mass transfer of the chelates toward the cathode [1]. For this reason, a stirring speed of 500 rpm was selected to prevent the formation of a serious vortex in the electrolytic cell. Samples were taken at time intervals (of 15 min) from the cathode chamber and also from the anode chamber at the beginning and end of the experiments to observe whether or not metal ions were transported to the anode chamber. The concentration of Cu(II), Ni(II) and Co(II) ions was analyzed by an atomic absorption spectrophotometer (Unicam 929). The experiments were performed at 251EC. Each experiment was replicated under identical conditions.

3. Results and discussion 3.1. Reactions in the electrolysis cell Citric acid, NTA and EDTA are all polyprotic acids (HxL) which can exist in many protonated forms in aqeous solutions. They form stable chelates with most divalent metals in a 1:1 molar ratio over a rather wide pH range [18]. For an equimolar diluted solution of Cu(II) and EDTA, for example, it was found that the divalent chelated anions CuL!2 absolutely dominate at pH 3 to 12, but the univalent chelated anions CuHL! at pH<3 [8]. In the case of Cu(II) and NTA, the univalent chelated anions CuL! dominate at pH 1.5 to 11 [1].

Fig. 1. Diagram used for electrodeposition experiments.

82

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

The catholyte pH increased during electrolysis due to the production of OH! at the cathode in reaction 9; in addition to this, H+ is generated at the anode such as in reaction 1. As a result of this, the pH of the anolyte solution decreased during the experiment. Hence, when a current is applied, the following reactions may occur at the anode: 2H2O O2 + 4H+ + 4e! H2O H+ + OH! E0 = !1.229 V E0 = !0.828 V (1) (2) (3) (4) (5)

complex is different from one metal to another or from one chelate to another. If these values are examined, it appears that the reduction tendency of metal ions plays an important role; that is, the reaction of Eq. (13) occurs in two steps: ML!2 M+2 + L!4 (by virtue of applied cell voltage) ML!2 + 2e! M(s) M+2 + 2e! M(s) + L!4 (M=Co, Ni, Cu) Step 1 Step 2 (13)

4OH! O2 + 2H2O + 4e! E0 = !0.401 V 2OH! H2O2 + 2e! H2O2 O2 + 2H+ + 2e! whereas at the cathode: CuL2!x + 2e! Cu(s) + L!x O2 + 2H2O + 4e! 4OH! E0 = 0.401 V 2H+ + 2e! H2 E0 = 0 V E0 = !0.547 V E0 = !0.682 V

The standard reduction potential in step 1 can change due to the metal. The present results imply that the chelated anions are destroyed (step 1), possibly on the outside of an electric double layer of the cathode, before the metals are deposited onto the cathode [19]. 3.2. Effect of current density on electrolysis The effect of current density on Cu recovery, RCu, from aqeous EDTA solutions for cpH = 3 and apH = 6 is shown in Fig. 2 and as a slope of the graphs for different pH values in Table 2. It was found that RCu increases with increasing electrolysis time and current density. Here the recovery value of the metal, R, was calculated from the following equation: R (%) =100 (actual amount of metal deposited onto the cathode) divided by the (initial amount of metal in the catholyte) In practice, the differences of RCu and current efficiency () among the three types of complexes Cu(II)-EDTA, Cu(II)-NTA and Cu(II)citrate were found to be insignificant, although Kf differs in each by 67 orders of magnitude. Here, is commonly defined as [20]: (14)

(6) (7) (8) (9) (10)

2H2O + 2e! H2 + 2OH! E0 = !0.828 V Cu+2 + 2e! Cu0 Co+2 + 2e! Co0 Ni+2 + 2e! Ni0 E0 = 0,337 V

E0 = !0.277 V (11) E0 = !0.250 V (12)

Evidently the evolutions of O2 and H2 gases are the main side reactions at the anode and cathode, respectively, which would reduce the current efficiency [1]. In addition to this, the binding affinity of the metalchelating agent complexes cannot fully explain the percent of recovery values because the overall formation constant (log Kf) of the 1:1

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

83

Fig. 2. Effect of current density on RCu from EDTA for cpH = 3 and apH = 6 with the CMB membrane. Table 2 Effect of current density on RCu from EDTA for different pH values with the CMB membrane Current density, mA Slope cpH: 3, apH: 6 15 25 35 0.0471 0.1066 0.1282 cpH: 3 apH: 3 0.0565 0.0701 0.0790 cpH: 6 apH: 6 0.0812 0.0984 0.1115

is more ordered. Sometimes the deposition is more disordered and heterogeneous. The first one is crystalline; the second one is non-crystalline. As seen from Table 2, although there are many different pH experiments for three values of current density, the recovery value of Cu increased in the order: 35 mA > 25 mA > 15 mA; but the assay of this augmentation varied with pH of catholyte and anolyte solutions. For example, when the pH of the catholyte was 3 and the anolyte was 6, the slope difference between 15 mA and 35 mA is 0.0811. However, when the pH of catholyte was 3 and the anolyte was 3, the slope difference between 15 mA and 35 mA was 0.0225. The effect of the stirring speed was not examined because a more vigorous mixing reduced the thickness of the double layer adjacent to the electrodes, enhancing mass transfer of the chelates and the deposition of metals onto the cathode by way of metal recovery increase with increasing the stirring speed [6]. 3.3. Effect of membrane type on electrolysis The CEM selected in this study were CMB, CMS, CMX; their properties are the same except for their capacity for ion exchange, thickness and capacity of water retention. The effect of membrane types on recovery of Cu from EDTA is shown as a graph in Fig. 3 and as a slope of the other metals (nickel and cobalt) in Table 3. Similar cases were obtained when other chelating agents such as NTA and citrate were used. The differences in results of these experiments are explained by the structure of the membranes (Table 1). As from Table 3, the amount of copper deposited onto the cathode increased in the order: CMB > CMX > CMS. Although this membrane order is the same for the both nickel and cobalt, the value of their slopes varied with the type of metal. For example, the variations of slopes between the CMB and CMS membranes were 0.0289, 0.0211 and 0.0596 for copper, nickel and cobalt, respectively.

(%) =100 (actual amount of metal deposited onto the cathode) divided by (15) (amount of metals deposited based on the Faradays law of electrolysis) The structure of metal deposition mainly depends on current density during electrolysis [21]. At low current density, surface diffusion is faster than electron transfer, so that the deposition

84

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

Fig. 4. Effect of the initial catholyte pH on RCu from EDTA for apH 6 at 35 mA with the CMB membrane. Fig. 3. Effect of the membrane type on RCu from EDTA for cpH = 3, apH = 6 at 35 mA. Table 3 Effect of the membrane type on recovery of metals from EDTA for cpH = 3, apH = 6 at 35 mA Membrane type Slope Cu CMB CMX CMS 0.1282 0.1012 0.0993 Ni 0.1373 0.1213 0.1162 Co 0.1514 0.1288 0.0918

3.4. Effect of catholyte pH on electrolysis The experiments were performed using a catholyte solution having different pH values, and the effect of the initial catholyte pH on the recovery of metals was examined. Fig. 4 illustrates the effect of the initial catholyte pH on RCu. Although it was expected from Eqs. (7) and (9) that the catholyte pH would increase after electrolysis, it became nearly constant at pH 11.6 if the initial catholyte pH was about 3. Catholyte pH values increased not only for this pH value but also for the other initial catholyte pH values at the end of

the electrolysis. In the electrodeposition experiments, the optimum pH for recovery was 3 and 6 for catholyte and anolyte solutions, respectively. For comparing of recovering values, these optimum conditions were used. RCu had a maximum at pH 3 and the relatively low RCu pH<3 was related to the amount of H+ discharging at the cathode and hence producing H2 gas bubbles. As the initial catholyte pH was beyond 3, the greater amount of Cu was deposited on the cathode; and in this case, a greater amount of Na+ was transferred from the anolyte through the membrane into the catholyte to maintain electroneutrality in the catholyte. Thus, a certain amount of current was consumed to preferentially reduce water [Eq. (9)], rather than Na+ at the cathode, which leads to a quick rise of catholyte pH. This can account for the drop of low RCu at pH >3. 3.5. Effect of anolyte pH on electrolysis The experiments were performed using anolyte solutions with different pH values. Figs. 5 and 6 illustrate the effect of the initial anolyte pH on RCu. Opposite to the increase of catholyte pH

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

85

Fig. 5. Effect of initial anolyte pH on RCu from EDTA for cpH 1 at 35 mA with the CMB membrane.

Fig. 6. Effect of initial anolyte pH on RCu from EDTA for cpH 3 at 35 mA with the CMB membrane.

from the beginning, initial anolyte pH decreases at the end of the electrolysis [Eq. (1)]. The best deposition values were obtained for copper when the initial anolyte pH was 6. 3.6. Effect of catholyte concentration on electrolysis The experiments were performed using catholyte solutions with different concentrations (0.001, 0.003, 0.005 M), but the metalchelate concentration ratio was 1:1 at pH 3 and 6 values for catholyte and anolyte solutions, respectively. Fig. 7 shows the effect of CuEDTA catholyte concentration on electrolysis. In addition to this, Table 4 demonstrates the effect of the catholyte concentration of the other metals (nickel and cobalt) as a slope of their graphs. In the experiments for the 1:1 metalchelate concentration ratio but in different metalchelate concentrations, it was clear that the amount of metal rises in unit volume because of the recovery of metal value increased. When the slopes of the graphs were estimated (Table 4), the differences between

Fig. 7. Effect of catholyte concentration on RCu from EDTA for cpH 3, apH 6 at 35 mA with CMB membrane.

the concentration values of 1.10!3 and 5.10!3 changed according to the types of the metals. For

86

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

Table 4 Effect of catholyte concentration on recovery of metals from EDTA for cpH = 3, apH = 6 at 35 mA with the CMB membrane Catholyte conc. Slope Cu 1.10 3.10!3 5.10!3
!3

Ni 0.1373 0.1531 0.1739

Co 0.1514 0.1633 0.1825

0.1282 0.1346 0.165

Table 5 Effect of the type of the chelating agent on recovery of metals for cpH = 3, apH = 6 at 35 mA with the CMB membrane Chelate types Slope Cu EDTA NTA Citrate 0.1282 0.1523 0.1647 Ni 0.1373 0.1765 0.1876 Co 0.1514 0.1793 0.1970 Fig. 8. Effect of the type of chelating agent on RCu for cpH = 3, apH = 6 at 35 mA with the CMB membrane.

example; the distinctions among these values are 0.0368, 0.0366 and 0.0311 for copper, nickel and cobalt, respectively. 3.7. Effect of types chelates on electrolysis The effects of complexing agents on the recovery of metals were found to be different according to their complex formation constant with metals. The recovery value of copper in the presence of different chelating agents is presented in Fig. 8, and the slopes of nickel and cobalt graphs are given to compare with copper in Table 5. According to the results on types of chelating agents, it can be said that their effect on the recovery of metals is varied, depending on their complex formation constants with metals. The recovery of copper value is ordered as: from EDTA chelate < NTA chelate < citrate chelate; and because of the complex formation constant order: Kf CuEDTA > Kf CuNTA > Kf Cucitrate [22]. As seen in Table 5, the results obtained from the experiments made with copper are the same for the other metals and chelating agents.

3.8. Effect of concentration ratio of metal chelate on electrolysis Different concentration ratios of metalchelate were also examined in order to compare the effect of the chelate concentration to the deposition of metals onto the electrode. The recovery values of copper experiments as a graph and of nickel as a slope are demonstrated in Fig. 9 and Table 6, respectively. In general, an increase of the Cu EDTA, NiEDTA or CoEDTA chelate concentration caused an increase of the driving force for mass transfer, thus facilitating the deposition of metal onto the cathode. This reduces the occurrence of side reactions and increases the total amount of species recovered rather than recovery of metals. Different metalchelate concentration ratios occurred, and different results were obtained due to the effect of the amount of chelates on the unit volume of metal. Increasing with the concentration of chelating agent or decreasing with the ratio of metalchelate concentration, amount of the metal, recovered on the cathode, was decreased due to the permission of the

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

87

Fig. 9. Effect of the metalchelate concentration ratio on RCu from EDTA for cpH 3, apH 6 at 35 mA with the CMB membrane. Table 6 Effect of the metalchelate concentration ratio on recovery of metals from EDTA for cpH 3, apH 6 at 35 mA with the CMB membrane Metalchelate ratio Slope Cu 1 1:2 Only metal 0.1282 0.1138 0.4488 Ni 0.1373 0.1206 0.3492

Fig. 10. Effect of type of metal on Rmetal without chelating agents for cpH 3, apH 6 at 35 mA with the CMB membrane.

optimum conditions. The results obtained from these experiments are shown in Fig. 10. In the electrodeposition experiments using metals without chelating agents, the results show that the recovery of copper was greater not only of cobalt but also of nickel because of its standard reaction potential [Eqs. (10)(12)]. 4. Conclusions The factors affecting the recovery of metals from complexed solutions were systematically examined. Regarding the mechanism of the growth of metal electrodeposition onto cathodes, it is basically a net result of the following three processes: (1) transfer of metal ions (including hydrated and chelated) toward the cathode surface due to diffusion, convection and migration; (2) transfer of electrons toward the cathode surface to form an adatom; and (3) diffusion of the adatom into the crystal lattice after sliding along the cathode surface.

chelate to metal to transport toward the surface of the cathode. The distinction between the different ratio of M:C concentration changed according to the type of metal as it was seen in Table 6. For example, while the difference of slopes between M:C = 1 and M:C = 1/2 is 0.0144 for copper, this difference was 0.0167 for nickel. 3.9. Effect of type of metal on electrolysis Experiments were performed to compare the effect of type of metal without chelating agents to the deposition of metal onto the cathode under the

88

Y. Oztekin, Z. Yazicigil / Desalination 190 (2006) 7988

The drawback of using the special twocompartment cell for all three types of metal was demonstrated: recovery increased with increase of current density, concentration of catholyte solution, concentration ratio of metalchelate, changed for the type of membrane in CMB > CMS > CMX order, for the type of the chelating agent in citrate > NTA > EDTA order, for the type of the metal in the Cu > Ni > Co order. The least value of recovery of metal was approximately 40%. This value increased due the type of the experiments up to 90% for copper. Therefore, this membrane technique is applicable for the recovery of metals.

References
[1] R.-S. Juang and L.-C. Lin, Sep. Sci. Technol., 35(7) (2000) 10871098. [2] R.-S. Juang and L.-C. Lin, Sep. Purif. Technol., 2223 (2001) 627635. [3] R.-S. Juang and I. Pyng Huang, Sep. Sci. Technol., 35(6) (2000) 869881. [4] R.-S. Juang and S.-W. Wang, Water Res., 34(12) (2000) 31793185. [5] R.-S. Juang and M.-N. Chen, Ind. Eng. Chem. Res., 36 (1997) 179186. [6] R.-S. Juang and L.-C. Lin, J. Membr. Sci., 171 (2000) 1929. [7] R.-S. Juang, S.-W. Wang and L.-C. Lin, J. Membr. Sci., 160 (1999) 225233. [8] R.S. Juang and L.D. Shiau, Ind. Eng. Chem. Res., 37 (1998) 555560. [9] C.N. Haas and V. Tare, Reactive Polym., 2 (1984) 61. [10] C. Chang and Y. Ku, Sep. Sci. Technol., 33 (1998) 483. [11] B.E. Reed, S. Arunachalam and B. Thomas, Environ. Prog., 13 (1994) 60. [12] B. Nowack and L. Sigg, J. Colloid Interf. Sci., 177 (1996) 106. [13] Z. Matejka and Z. Zitkova, Reactive Funct. Polym., 35 (1997) 81. [14] J.W. Johnson, H.W. Jiang, S.B. Hana and W.J. James, J. Electrochem. Soc., 119 (1972) 574. [15] L. Doulakas, K. Novy, S. Stucki and C. Comninellis, Electrochimica Acta, 46 (2000) 349356. [16] A.L.Portela, G.I. Lacconi and M. Lopez Teijelo, J. Electroanal. Chem., 495 (2001) 169 172. [17] Catalog of Neosepta Products, 2001. [18] F.M.M. Morel and J.G. Hering, Principles and Applications of Aquatic Chemistry, Wiley, New York, 1997, pp. 159165. [19] K.B. Oldham and J.C. Myland, Fundamentals of Electrochemical Science, Academic Press, San Diego, 1994, Chap. 9. [20] R.S. Juang and L.C. Lin, Water Res., 34 (2000) 4350. [21] Y. Oztekin, Recovery of metals from aqueous solutions in the presence of chelating agents, M.Sc. Thesis, Selcuk University, 2003. [22] J. Kragten, Atlas of metal-ligand equilibria in aqueous solution. Thesis, University of Amsterdam, 1978.

5. Symbols apH cpH E0 H4L I RCu RCo RNi MC pH of anolyte pH of catholyte Standard reaction or oxidation potential, V Ethylenediaminetetraacetic acid Current density Recovery of copper Recovery of cobalt Recovery of nickel Concentration ratio between metal and chelate

Acknowledgements The authors wish to thank Eurodia Co. for support with their materials, and are also grateful for the financial support provided by the Selcuk University Research Foundation (SUAF) under project number FEF 2003/039.