20

Description of Transport Mechanism During the Elimination of Copper (II) from Wastewaters Using Supported Liquid Membranes and ACORGA M5640 as Carrier

F.J.ALGUACIL* AND M.ALONSO

Centro Nacional de Investigaciones Metalúrgicas (CSIC), Avda Gregorio del Amo 8, Ciudad Universitaria, 28040 Madrid, Spain. E-mail:

______

The kinetics of copper facilitated transport through a flat-sheet supported liquid membrane is investigated, using the commercially available oxime Acorga M5640 as ionophore, as a function of hydrodynamic conditions, stripping phase composition, concentration of copper (II) and pH of the source phase and carrier concentration in the membrane phase. The performance of the system is also compared using organic diluents of different nature (aliphatic or aromatic) and against other available oxime extractants (LIX 860, LIX 622, LIX 973N and LIX 84-I). A model is presented that describes the transport mechanism, consisting of diffusion through source side aqueous diffusion layer, a fast interfacial chemical reaction, and diffusion carrier and its metal complex through the organic membrane. The organic membrane diffusional resistance (Δ org) and aqueous diffusional resistances (Δaq) were calculated from the proposed model, and their values were 9.3x10-7 and 46565 s/cm, respectively. It was observed that the copper flux across the membrane tends to reach a plateau at high concentration of copper or low concentration of H+ owing to carrier saturation within the membrane, and leads to a diffusion-controlled process. The values of the apparent diffusion coefficient (Dorga) and limiting metal flux (Jlim) were calculated from the limiting conditions and found to be 5.2x10-11 and 1.9x10-9 mmol/cm2 s, respectively. The values of the bulk diffusion coefficient (Db,org) and diffusion coefficient (Dorg) calculated from the model were 4.8x10-10 and 1.3x10-10 cm2/s, respectively. The method had proven its feasibility to recover copper (II) from wastewaters.

Introduction

The liquid-liquid extraction of metals by liquid ion exchangers had gained a prominent and key position in various processing schemes, being this particularly sound in the recovery of copper from the mining industry. The use of supported liquid membranes techniques in flat-sheet, hollow fiber and spiral wound can improve the liquid-liquid extraction procedure and extend its application field to the recovery of metal ions (or complexes) from wastewaters. In supported liquid membrane technology, the extraction, stripping and regeneration of the organic phase are combined in a single stage, furthermore, from the engineering and practical stand point, supported liquid membranes are of a particular interest because of its stability and simplicity.

During last years several studies deal about the permeation of copper (II) through supported liquid membranes using different extractants (1-14).

However, between first studies and a running process, a gap has to be filled up with concrete and specialised know-how, thus, before, scaling up the flat-sheet supported liquid membrane, a theoretical model of the liquid membrane system is needed in order to design an efficient recovery process in terms of better stability and performance.

In order to contribute to this goal, this work presents the kinetic modelling of active transport of copper (II) using ACORGA M5640 (oxime derivative) immobilized on a microporous hydrophobic support. The carrier is a chelating agent, highly selective for copper and used in industrial liquid-liquid extraction (15). The membrane and aqueous diffusional resistances were calculated from the proposed model. The influence of hydrodynamic conditions and chemical parameters were investigated in order to obtain efficient flat-sheet supported liquid membrane.

Experimental Section

The oxime ACORGA M5640 was obtained from Avecia (England) and was used without further purification; the active substance of the reagent is 5-nonylsalicylaldoxime which also contained a fatty ester as modified. According to its composition, the reagent a salicylaldehide alkyl derivative, belongs to the fourth group of hydroxyoxime extractants, which formed the second generation of hydroxyoxime reagents (16). This ester-modified oxime reagent offers the ultimate in selectivity (e.g. copper (II) over iron (III), performance over a wide range of pH, and hydrolytic stability). The concentration of ACORGA M5640 in the working organic solutions was determined by the ultimate loading (17).

Iberfluid (CS, Spain) is a kerosene type diluent and was used as the organic diluent for ACORGA M5640, its main characteristics are: boiling range (210-284ºC), flash point (96ºC), aromatics (<2%), density (785 kg/m3). Unless otherwise stated, all the other chemicals used in this work were of AR grade (Fluka).

The flat-sheet supported liquid membrane was impregnated with the carrier solution by immersing it for 24 h and then leaving it to drip for 10 s before being placed in the cell. Previous experiments had shown that prolonged immersion times (e.g. 48 h) of the membrane in the carrier solution had not influenced the permeation coefficient values obtained using 24 h. The flat-sheet membrane used was Millipore Durapore GVHP 4700 of 125x10-4 cm thick microporous polyvinylidenedifluoride film with nominal porosity of 75%, effective pore size of 2.2x10-5 cm, and tortuosity 1.67.

Batch transport experiments were performed in a permeation cell consisting of two cubic compartments made of methacrylate and separated by the microporous membrane. The membrane effective area was 11.3 cm2 and the volume of the feed and stripping solutions was 200 mL (each). The experiments were carried out at 20ºC at a mechanical stirring speed of 1300 rpm in the source and stripping phases, except in the experiments where the stirring speed was varied. Agitation was performed in both compartments by using cylindrical Teflon impellers having a diameter of 2.4 cm.

An Oakton (USA) combined electrode and pH-meter were used in all the experiments for measuring the pH of the source phase and, if needed, to correct it to a constant pH value by the addition of NaOH or H2SO4 solutions. A Perkin Elmer 1100B Atomic Absorption Spectrophotometer (England) was used to measured the copper (II) concentrations in the source and stripping phases.

The metal flux was evaluated from the ln [Cu]t/[Cu]0 versus t curves by using the equation:

(1)

where [Cu]t and [Cu]0 are the copper concentrations in the source phase at a given time and time zero, respectively, and P is the permeability coefficient which was estimated as described elsewhere (18).

Results and Discussion

Influence of the stirring speed. The effect of the stirring speed was examined in order to optimise uniform mixing of the solution and to minimise thickness of aqueous boundary layer with source and strip conditions being maintained as 0.16 mM Cu(II) at pH 2.0±0.05 and 1.8 M H2SO4, respectively. The extractant concentration was 0.36 M in Iberfluid immobilised on a Durapore Support. The metal flux becomes virtually independent of the stirring speed above 1100 rpm, indicating a decrease in the aqueous boundary layer thickness, and then a minimum value of the thickness is reached at 1300 rpm, thus, a stirring speed of 1300 rpm was kept constant throughout the experiments conducted.

Influence of stripping phase composition on the transport of copper (II). Since the extraction and stripping processes in flat-sheet supported liquid membrane systems are carried out simultaneously for continuous transport of metal ion, it is important to investigate the effect of the stripping phase composition in order to enhance effective ion transport by making the stripping process efficient at the interface of the membrane and stripping solutions. Of two strippant tested, such as 1.8 M H2SO4 or 0.16 mM Cu(II) and 1.8 M H2SO4, both proved to be equally efficient for copper transport (J= 6.6x10-10 mmol/cm2 s). In the above experiments, the source phase was of 0.08 mM Cu(II) at pH 2.0±0.05, whereas the membrane solution was of 0.36 M ACORGA M5640 in Iberfluid. Moreover, concentration profiles in the three bulk phases during the permeation experiment, using the stripping phase of 0.16 mM Cu(II) and 1.8 M H2SO4, are shown in Fig.1. From the initial stages of the experiment, the decrease in copper concentration in the source phase was accompanied by a corresponding increase in the stripping phase concentration. In the case of the membrane behaviour, and after an initial increase in the metal concentration, this reaches a plateau, indicating that copper was immediately transferred from the organic phase to the stripping phase. It is also noted, that from the first minute, copper (II) was being transported uphill against its concentration gradient, driven by the counter-current coupled transport of protons from the stripping to source solution.

1

FIGURE 1. Copper (II) concentration profiles in transport experiments. See text for experimental conditions.

Effect of source phase pH on the transport of copper (II). In the present flat-sheet supported liquid membrane system, the pH gradient between source and stripping solutions is the driving force for the transport of metal ion. In order to asses the role of source phase pH, pH variation studies in the range 1.0 to 3.0 were carried out. The receiving phase consisted of 1.8 M H2SO4, whereas the concentration of membrane carrier was 0.36 M. It is evident from Table 1 that the metal flux increases with an increase in pH from 1.0 to 2.0, though at higher pH it remained unaffected, thus it is considered that the pH value of 2.0 may be optimum for copper permeation; in addition at this pH value copper is transported selectively in the presence of other metals presented in the feed phase (see below). Moreover, at high pH values the diffusion of carrier through the membrane becomes rate determining. From eq.(11) (see further), two limiting cases can be derived:

i) at low pH and high carrier concentration, the equation reduces to:

(2)

ii) at high pH and high carrier concentration, the equation is:

(3)

and the flux is independent of pH (source phase).

Table 1. Influence of Initial pH on Metal Flux
pH J(mmol/cm2s)
1.0 4.1x10-11
1.5 2.5x10-10
2.0 1.5x10-9
3.0 1.5x10-9
Source phase: 0.16 mM Cu (II).

Effect of carrier concentration on the transport of copper (II). A supported liquid membrane having no carrier immobilized on the support results in no transport of copper ions. The results concerning transport of copper (II) from the source phase containing 0.16 mM Cu(II) at pH 2.0±0.05 and the stripping phase 1.8 M H2SO4 and varying concentrations of ACORGA M5640 in the range 0.09 to 0.72 M dissolved in Iberfluid are shown in Fig.2. As can be seen from the figure, the metal flux increases with carrier concentration until a maximum is reached and levels off. This is due to the fact that at low carrier concentrations, diffusion of the metal complex across the liquid membrane is the rate-determining step, while at higher carrier concentrations diffusion of the cation across the aqueous phase boundary layer is the rate-determining step.

1

FIGURE 2. The influence of carrier concentration on metal flux.

Assuming that the carrier concentration in the membrane phase is constant, the next equation can be used to determine the apparent diffusion coefficient of copper species in the organic phase (19):

(4)

where dorg is the thickness of the support. The value of the apparent diffusion coefficient was estimated to be 5.2x10-11 cm2/s using a carrier concentration of 0.36 M.

Effect of carrier diluent on the transport of copper (II). The effect of the diluent of the membrane phase on the transport of various metals had been reported in the literature (20-24). The diluent employed as a membrane should retain the carrier to the maximum extent and yet at the same time retain a relatively high concentration of water to aid the transfer of hydrated species without loss of extractant to the aqueous phase. Polymer-immobilised liquid membranes performance is primarily dependent on intrinsic membrane diluent properties such as viscosity, volatility, surface tension, water solubility, etc. In the present work, the effect of the membrane phase diluent on the transport of copper (II) had been investigated with source and strip conditions being maintained as 0.08 mM Cu(II) at pH 2.0±0.05 and 1.8 M sulphuric acid, respectively. The ACORGA M5640 concentration was 0.36 M, in each diluent, immobilised on a Durapore microporous support. The transport behaviour of copper, using Iberfluid or cumene as diluents, shows that in the present system the type of diluent (e.g. aliphatic versus aromatic) had not and apparent effect on the transport of the metal, since the fluxes obtained were the same (6.6x10-10 mmol/cm2 s) for each diluent. It is reported in the literature (25), that in acidic/chelating extractants-based systems the influence of the diluent, on the liquid-liquid extraction of metals, is less than in other reagent-based systems, e.g. amines and solvation extractants.

Effect of metal concentration on the transport of copper (II). Studying the effect of the initial concentration of copper (II) (0.08 to 2.5 mM) in the source solution, when the stripping side contain no copper, it was revealed that the metal flux initially increased from 0.08 to 0.125 mM and beyond this became independent of the initial concentration. The results of copper flux through the membrane as a function of initial metal concentration are shown in Fig.3. The initial increase in copper flux can be attributed to that the permeation process is controlled by diffusion of metal species in the lower range of copper concentrations. At higher metal concentrations, the observed behaviour is probably due to membrane saturation and a lower effective membrane area in the polymer-immobilised liquid membrane and also due to maximisation as a result of saturation of membrane pores with metal-carrier species and build-up of the carrier layer on the membrane interface, enhancing the retention of the separating constituent on the entry side and thus causing the flux to be constant (26). Under the limiting condition the total concentration of ACORGA M5640 [HR] becomes equivalent to [CuR2]org (see eq.(6) for details), and on the basis of the next equation (27):

(5)

where n is the stoichiometric coefficient of the extraction reaction (2 in the present case), the value Jlim is estimated to be 1.9x10-10 mmol/cm2 s.

1

FIGURE 3. Influence of initial concentration of Cu(II) on metal flux. Source phase: copper (II) at pH 2.0±0.05. Membrane phase: 0.36 M ACORGA M5640 in Iberfluid. Stripping phase: 1.8 M sulphuric acid.