Copper cation transport and scaling of ionic exchange membranes using electrodialysis under electroconvection conditions

Jih-Hsing Chang a , *, Amanda V. Ellisb, Cheng-Hung Tung c, Wen-Chi Huang d

aDepartment of Environmental Engineering and Management, Chaoyang University of Technology, 168 JiFong E. Rd., WuFong Township, 41349 Taichung County, Taiwan

bCentre for Nanoscale Science and Technology, School of Chemistry, Physics and Earth Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia

cDepartment of Environmental Engineering, National Chung Hsing University, 250, Kuo Kuang Road, Taichung, 40277, Taiwan

dDepartment of Environmental Engineering and Management, Chaoyang University of Technology, 168 JiFong E. Rd., WuFong Township, 41349 Taichung County, Taiwan

Corresponding Author: Jih-Hsing Chang

Address: Department of Environmental Engineering and Management, Chaoyang University of Technology, 168 GiFeng East Road, WuFong Hsiang, 413 Taichung County, Taiwan

Email:

Telephone No: 886-4-23323000 Ext 4210

Fax No: 886-4-23742365

Abstract

Here we report on the removal efficiency of copper cations from model Cu2+ ion wastewater by electrodialysis (ED) under electroconvection conditions. Experiments were conducted on commercial ionic exchange membranes (IEMs). Results are presented on the copper cation transport properties through a cation exchange membrane (CEM) showing that Cu2+ ions can penetrate a copper-saturated CEM and still maintain a stable cation removal efficiency rate. We use field-emission scanning electron microscopy (FE-SEM) and X-ray diffraction (XRD) to analyze the scale formed on the IEMs after ED treatment. XRD results show that the scaling on the IEMs is Cu(OH)2 and/or CuO potentially formed from the reaction of Cu2+ ions with hydrolysis products. In particular, results show that the anion exchange membranes (AEMs) are the most prone to scaling which results in a decrease in the overall ion removal efficiency of the ED system. Under electroconvection conditions, hydrolysis of water causes the formation of H+ and OH- ions which penetrate the IEMs leading to a lowering of the pH in both the treated and concentrated model Cu2+ ion wastewater. Finally we show that a stronger field is necessary to drive ion transport through the CEM due to scaling and that transport is heterogeneous in comparison to that of a CEM with no scaling.

Keywords: Electrodialysis, electroconvection, ion exchange membrane, copper wastewater

1. Introduction

Over the past several decades ED membranes have been heavily employed by industries for the removal or isolation of chemicals such technologies include, brine water desalination, acid liquid concentration, mineral recovery, and metal ion wastewater purification [1]. The predominance of ED in these applications is attributed to its remarkable ion removal efficiency, particularly in wastewater treatments and affords many advantages such as, small installation sites, short treatment times, and minimal sludge production. The mechanisms involved in ion-removal from wastewater include electromigration and ion-exchange [2]. Electromigration is induced by an applied electric field which drives ions through the ED membranes while ion-exchange involves the passage of ions in the aqueous phase through the membrane either by attraction or repulsion, primarily due to co-ionic functional groups on the membrane [3] used ED to treat a solution of copper chloride, copper sulfate, nickel sulfate and zinc sulfate finding a current efficiency of more than 70 % [4] used ED to decrease the concentration of Zn2+ from 1.440 g L-1 to 0.032 g L-1 with a current efficiency of approximately 60-90 %. ED has also been used to recover nickel (90 %) [5], zinc (90 %) [5] and silver (~100 %) [6].

Under ED the current running through a CEM changes with the electrical voltage due to the different ion transport mechanisms and electrochemical reactions involved in the ED process. Fig. 1 shows a current density versus potential plot of a typical CEM. The plot can be divided into three clearly defined regions based on the slope of the curve, that is, the ohmic region, the limiting current region and the electroconvection region [7]. In the ohmic region the current possesses a linear relationship with voltage (V = I × R, Voltage = current × resistance), which implies the penetrating flux of ions through the CEM increases proportionally with the electric field. In the limiting current region, the concentration of exchange ions approaches zero at the CEM interface (i.e., polarization) and the resistance tends to infinity. This phenomenon can be attributed to the fact that the flow rate of ion migration though a CEM is faster than that of the molecular diffusion. Under such conditions the current density of ED is constantly maintained and the plot shows a pseudo-plateau (Fig. 1) [8, 9]. As the electric voltage is increased into the electroconvection region the interface between the electrolyte and the CEM is hydrolyzed and releases a large quantity of H+ and OH- ions into the solution [10]. This results impose in an increase in the current density (greater than the over-limiting current density), which may disrupt the the diffusion layer on the membrane surface.

Historically ED is operated within the ohmic region primarily because the current efficiency is reduced in the limiting current region. However, more recently interest is being paid to operating ED in the electroconvection region as there have been reports of improved ion transport efficiency [11,12], although this region might afford unstable transport [13]. The production of H+ and OH- ions from the hydrolysis of water in this region may be a major contributor to this instability. In particular these hydrolysis products can lead to solution acidification and alkalization in both the treated and concentrated wastewater. Of more concern is the reaction of these hydrolysis products with metal ions giving rise to scaling of the membrane surface and thus simultaneously reducing the current and ion removal efficiency [14]. Metal ion removal under electroconvention conditions and the ion transport properties of IEMs is far from understood. Additionally, to date little is known about how scale/fouling affects the ion transport.

In this study, we use ED through IEMs to treat model Cu2+ ion wastewater under electroconvection operating conditions. We discuss the results in terms of copper removal efficiency, electrochemical reactions taking place during ED, membrane scaling/fouling, and the transport profiles of Cu2+ ions through the membranes.

2. Experimental

Commercial CEMs (Serial No. CMI 7000) and AEMs (Serial No. AMI 7001) were purchased from Membranes International Incorporation, USA. Their physical and chemical characteristics are listed in Table 1. These membranes were specifically chosen as they afford a high ion exchange capacity, low water permeability, low electric resistance and high physical strength. The model Cu2+ ion (500 mg L-1) wastewater was made using copper sulphate and the electrolyte was 0.1 M NaCl. Sodium hydroxide (Mallinckrodt, 99.84 %) and sulfuric acid (Merck, 98 %) were used to adjust the pH where necessary.

2.1 ED system for treating model Cu2+ ion wastewater

A schematic of the continuous-mode ED system used is shown in Fig. 2. This system consisted of 4 reaction cells, a pair of graphite electrodes, a pair of Ag/AgCl reference electrodes (REs), 2 peristaltic pumps and 3 IEMs (2 AEMs (Fig. 2(A)) and 1 CEM (Fig. 2(C)). The dimensions of the ED system were 38 cm (long) × 14 cm (wide) × 5.5 cm (high). The ED system was divided into four 150 mL cells by the IEMs, giving a cathodic cell (Fig. 2(1), a concentrated Cu2+ ion wastewater cell (Fig. 2(2)), a model Cu2+ ion (500 mg L-1) wastewater cell (Fig. 2(3)), and an anodic cell (Fig. 2(4)). The system was operated continuously for 48 hrs in the electroconvection region under a constant applied voltage of 33 V cm-1. A feed supply of model Cu2+ ion wastewater (500 mg L-1) was feed into cell 3 (Fig. 2) at a flow rate of 50 mL min-1 (i.e., the hydraulic retention time was 3 min) and discharged after ED treatment. The total volume of model Cu2+ ion wastewater used was 144 L. In cell 2 (Fig. 2) the concentrated Cu2+ ion solution passed through the CEM was recirculated into a 7 L concentrated Cu2+ ion reservoir (Fig. 2(6)). A 7 L reservoir of 0.1 M NaCl electrolyte (Fig. 2(8)) was circulated into and out of the anodic cell (Fig. 2(4)) and then into and out of the cathodic cell (Fig. 2(1)) and finally back into the 7 L reservoir of 0.1 M NaCl electrolyte (Fig. 2(8)). The electrolyte solution was refreshed ever 24 hrs. Each of the 4 cells had its pH, conductivity, Cu2+ ion concentration and current monitored hourly as the ED progressed. The pH and conductivity were determined with a pH meter (Suntex SP-701) and a conductivity meter (Suntex SP-700), respectively. The Cu2+ ion concentration was determined using flame atomic absorption (AA) spectrometry (Varian 220 spectrometer) and the current was monitored using a power supply (IP 200-21 DS). In order to monitor the degree and type of scaling/fouling that was forming on the surface of the IEMs during ED treatment field-emission scanning electron microscopy (FESEM, JEOL-JSM-6700F) and X-ray diffraction (MAC-MXP18 diffractometer with Cu Kα radiation, λ = 1.54056 Å) were used.

2.2 Ion transport profiles of the CEM by electrochemical analysis

Electrochemical analysis was used to study the ion transport profiles of the CEM, both with and without scaling. Fig. 3 shows a schematic of the electrochemical analysis system relating to the CEM component, Fig. 2C. The system was composed of a reaction cell made of poly(vinyl chloride) (PVC) with dimensions 12 cm (long) × 5 cm (wide) × 4 cm (high), a graphite working electrode (WE) (3 cm2) and a graphite counter electrode (CE) (3 cm2) and two Ag/AgCl REs. The reaction cell was divided into two separate 150 mL cells by the CEM (4 cm × 3 cm). A working station (WS1) provided an electric voltage between the WE and the CE while a second working station (WS2) measured the voltage drop across the CEM between the 2 REs. The ion transport properties of the CEM were measured by the voltage drop and the scanning current [15].

In order to study the effect of Cu2+ ion concentration on ion transport through the CEM, with and without scaling, six different copper sulfate solutions were prepared with concentrations of 1.5 × 10-3 M, 2 × 10-3 M, 2.5 × 10-3 M, 3 × 10-3 M, 3.5 × 10-3 M and 4 × 10-3 M. For each ED experiment cell 3 (Fig. 2) was filled with a copper sulfate solution and both cells on either side of the CEM were stirred with a magnetic stirrer for 3 mins. The galvanodynamic process was measured at WS1 at a current scan rate of 0.05 mA s-1 and the voltage and current were recorded at 1 plot s-1 scan rate. At the same time WS2 measured the voltage drop across the CEM by an open circuit method. The current density versus voltage curve produced at each concentration was then analyzed in order to explain the ion transport properties of the CEM.

3. Results and discussion

3.1 Cu2+ ion removal efficiency under electroconvection conditions

Fig. 4a and 4b shows the change in the concentration of the Cu2+ ion wastewater concentration (mg L1) with time (min) in the concentrated cell (Fig. 2(2)) and the model Cu2+ ion (500 mg L-1) wastewater cell (Fig. 2(3)), respectively. Fig. 4a, (first) shows the first run of ED treatment through the CEM with the model Cu2+ ion wastewater into the concentrated cell. The concentration gradually increased from 500 mg L-1 to 3,000 mg L-1 after 24 hrs. The electrolyte solution was then replenished and ED treatment through the CEM was carried out for a further 24 hrs. Again the Cu2+ concentration gradually increased from now 3,000 mg L-1 to 7,000 mg L-1 (Fig. 4a, second). These results indicate that the Cu2+ ions can continually penetrate the CEM and enter the more concentrated cell with time. Given that the slopes of both the first and second ED treatment in Fig 4a are the same, this would imply that the replenishment of the Cu2+ ion solution has no influence on the ion transport in the CEM. Correspondingly, Fig. 4b, first) shows that the Cu2+ ion concentration in the model Cu2+ ion wastewater cell was reduced from 500 mg L-1 to 125 mg L-1 after the first ED treatment, then reaching a stable concentration of approximately 120 mg L-1after 120 mins. After replenishment and a further 24 hrs of ED the Cu2+ ion concentration increased from 120 mg L-1 to 160 mg and stabilized at this value after 120 mins (Fig. 4b, second). This slight increase of 40 mg L-1 in Cu2+ ion concentration may be due to scaling of the CEM, which is observed as a decrease in the Cu2+ ion transport rate in the CEM. Previous results not shown indicate that the sorption capacity of Cu2+ ions on the CEM is around 1.07 g/cm2. After calculating the copper mass balance of the ED system it was observed that the CEM had reached its copper saturation concentration (i.e., sorption capacity) in the first run. Since there is still ion transport after the second run this implies that the Cu2+ ions could still penetrate the CEM even under saturation conditions.

3.2 Change in pH under electroconvection conditions

Fig. 5a and 5b show the change in pH of the concentrated cell (Fig. 2(2)) and the model Cu2+ ion wastewater cell (Fig. 2(3)) with time (min), respectively. At the commencement of the first run, see Fig. 5a, (first), the pH of the solution in the concentrated cell was approximately 5 and rapidly dropped to 3 after 60 mins. Subsequently a stable pH of 2 was maintained after 360 mins. After the second run, (see Fig. 5a, second), the pH dropped further from 2 to approximately 0.8. Similar results were observed for the model Cu2+ ion wastewater cell with the pH plateauing at 2.8 and 1.7 after the first and second runs, respectively (see Fig. 5b, first and second). This rapid drop in pH of the ED treated wastewater is attributed to water hydrolysis on the surface of the membrane [7]. It is believed that under electroconvection conditions H+ and OH- ions produced can penetrate the CEM and be transported to the anode and cathode, respectively [10].