Facilitated transport of Cr(VI) through a novel activated composite membrane containing Cyanex 923 as a carrier

Journal of Membrane Science 337 (2009) 224–231

Contents lists available at ScienceDirect

Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci

Facilitated transport of Cr(VI) through a novel activated composite membrane containing Cyanex 923 as a carrier Gulsin Arslan a, Ali Tor b,∗, Harun Muslu a, Mustafa Ozmen a, Ilker Akin a, Yunus Cengeloglu a, Mustafa Ersoz a a b

Department of Chemistry, Selcuk University, 42031, Campus, Konya, Turkey Department of Environmental Engineering, Selcuk University, 42031, Campus, Konya, Turkey

a r t i c l e

i n f o

Article history: Received 2 January 2009 Received in revised form 27 March 2009 Accepted 27 March 2009 Available online 8 April 2009 Keywords: Activated composite membrane Facilitated transport Cr(VI) Cyanex 923 Supported liquid membrane

a b s t r a c t This paper describes the facilitated transport of Cr(VI) through a novel Activated Composite Membrane (ACM) containing Cyanex 923 as a carrier. The ACM was prepared by immobilization of the Cyanex 923 on a polysulfone support by means of interfacial polymerisation. The prepared ACM was characterized by using scanning electron microscopy (SEM) and atomic force microscopy (AFM) techniques and contact angle measurements. The effect of feed phase composition, carrier concentration of the casting solution and stripping phase composition on the transport of Cr(VI) was investigated. When the feed phase contained 1 × 10−3 M Cr(VI) at pH 1.0, 99% of Cr(VI) was transported through the ACM (prepared with 3% carrier solution) by using 1 M NaOH as a stripping phase. Furthermore, Cr(VI) was preferably transported in the presence of various metal ions (i.e., Cr(III), Ni(II), Cu(II), Zn(II), Cd(II), Co(II), etc.) and sulphate and nitrate ions had no negative effect on the transport of Cr(VI). The results also showed that transport efficiency of the ACM was reproducible and it could be efficiently used in the long-term separation processes instead of supported liquid membrane (SLM). © 2009 Elsevier B.V. All rights reserved.

1. Introduction The extensive use of chromium in many industries (i.e., leather tanning, metallurgy, electroplating, etc.) has resulted in the release of aqueous chromium to the environment. In aqueous solution, chromium frequently exists as Cr(VI) or Cr(III). These two oxidation states have different chemical, biological, and environmental properties. Cr(VI) exists as anionic species such as HCrO4 − , Cr2 O7 2− and CrO4 2− , which are highly mobile in soil and aquatic systems. Oxidizing potential of these Cr(VI) species make them highly toxic for biological systems. However, Cr(III) readily precipitates as Cr(OH)3 under alkaline or slightly acidic conditions [1]. The limit for the discharge of Cr(VI) into inland surface waters is 0.1 mg/L. Furthermore, potable water containing more than 0.05 mg/L of Cr(VI) is considered to be toxic for living beings [2]. Therefore, the removal of Cr(VI) from industrial effluents is important before discharging them into aquatic environments or onto land. The different methodologies for recovery or removal of Cr(VI) from aqueous solutions have been developed utilizing ion-exchange [3], solvent extraction [4,5], non-dispersive solvent extraction [6,7] and membrane-based technologies. The

∗ Corresponding author. Tel.: +90 332 223 1914; fax: +90 332 241 0635. E-mail address: [email protected] (A. Tor). 0376-7388/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2009.03.049

membrane-based methods (i.e., nanofiltration [8], micellarenhanced ultrafiltration [9,10], Donnan dialysis [11], electrodialysis [12,13] and facilitated transport through bulk liquid membrane (BLM) [14], supported liquid membrane (SLM) [15], polymer inclusion membrane (PIM) [16], emulsion liquid membrane (ELM) [17] and activated composite membrane (ACM) [18], etc.) have been used for the separation of Cr(VI) from aqueous phase. The ACMs have been recently designed and developed to obtain simple, practical, fast and effective separation devices, which are based on the operation principles of the SLM and composite membrane. In comparison to the SLMs, the main advantage of ACMs is their high stability in long-term separation processes [19,20]. Due to this advantage of the ACMs, they have been used for the separation of metal ions and some organic molecules particularly amino acids [21–25]. During the last decades, Cyanex 923 has been used as extractant because of its marked selectivity and hydrolytic stability [26,27]. According to the literature, Alguacil et al. [28,29] studied the transport of Cr(VI) through a SLM containing Cyanex 923 as a carrier. However, the facilitated transport of Cr(VI) through an ACM prepared with Cyanex 923 has not been reported. Therefore, the objective of the present study is to investigate the efficiency of a novel ACM containing Cyanex 923 for the facilitated transport of Cr(VI). For this purpose, the ACM was prepared by immobilization

G. Arslan et al. / Journal of Membrane Science 337 (2009) 224–231

of the Cyanex 923 on a polysulfone support by means of interfacial polymerisation and it was characterized by contact angle measurements, atomic force microscopy (AFM) and scanning electron microscopy (SEM). The effects of the parameters (i.e., initial pH and Cr(VI) concentration of the feed phase, carrier concentration of the casting solution, type and concentration of the stripping phase) and other ionic species on the transport of Cr(VI) were investigated. Finally, transport efficiency of the ACM was compared with that of the SLM. 2. Experimental 2.1. Chemicals The chemical reagents used for the ACM preparation and experiments, 1,3-phenylene diamine, 1,3,5-benzene-tricarbonyl chloride, n-hexane, N,N-dimethyl formamide (DMF), hydrochloric acid, potassium chromate, chromium(III) chloride, copper(II) chloride, nickel(II) chloride, zinc(II) chloride, cobalt(II) chloride, cadmium(II) chloride, sodium hydroxide, sodium chloride, were from Merck Co. (Darmstadt, Germany). Cyanex 923, the mixture of tertiary octyl and hexyl phosphine oxides, was obtained from CYTEC (Canada). Composition of the Cyanex 923 was as follows: dioctyl-monohexyl phosphine oxide (40–44%), mono-octyldihexyl phosphine oxide (28–32%) and tri-n-octylphosphine oxide (12–16%). Polysulfone and sodium dodecyl sulfate were from Aldrich (Steinheim, Germany) and J.T. Baker (Dewenter, Holland), respectively. Kerosene used for the preparation of the SLM was from Aldrich (Germany). 2.2. Preparation of ACM The schematic representation for the preparation of the ACM was given in Refs. [22] and [25]. The polysulfone casting solution was prepared by dissolving the polymer (15%, w/w) in DMF by vigorous stirring for 12 h. Deposition of a polysulfone layer onto the non-woven fabric support (Hollytex 3329) was performed by using a spin coater (Laurell Model WS400A-6NPP/LITE). Then, the impregnated support was immersed in a water bath at room temperature for 5 min to induce phase inversion polymerisation. A second polyamide layer containing the carrier was formed on the obtained polysulfone support by means of interfacial polymerisation as follows. The polysulfone support was impregnated with an aqueous diamine solution for 5 min, then a n-hexane organic solution containing both 1,3,5-benzene-tricarbonyl chloride and Cyanex 923 for 5 min. The excess of the solution was washed off the membrane surface with distilled water and finally the membrane was dried in an oven at 60 ◦ C for 60 min. The resulting ACM sheet was cut to produce a 3.0 cm diameter of circle that was placed into the transport cell. A blank membrane was also prepared by following the same procedure without including carrier (Cyanex 923). The diamine solution consisted of 3 g of 1,3-phenylene diamine in 50 mL of distilled water, containing 0.01 g of sodium dodecyl sulfate as an aqueous stabilizing agent. The organic solution contained 0.04 g of 1,3,5-benzene-tricarbonyl chloride and desired concentration of the Cyanex 923 in 50 mL of n-hexane.

225

2.4. Surface characterization 2.4.1. Scanning electron microscopy The surface of the blank membrane and ACM was observed by using SEM (JEOL 5600-LU). The SEM images were obtained after the membranes were fixed with conductive glue and covered with a thin Au layer (10–20 nm) [31]. 2.4.2. Atomic force microscopy AFM images, obtained from the tapping mode of a Veeco diCaliber instrument, were used to investigate the surface morphology of the membranes. The speed of scanning was 2 kHz and silicone nitride cantilevers (antimony, ndoptFi) were employed. The AFM images (of ca. 10 ␮m × 10 ␮m) of five different parts of each membrane were properly analyzed in order to obtain the mean roughness (Ra) parameter. 2.4.3. Contact angle measurement The sessile drop method was used to measure the contact angle of the prepared membranes [32]. A 4 ␮L droplet of distilled water was placed on the membrane surface by using a 0.10 mL capacity of syringe and the contact angle was measured by means of a horizontal beam comparator (KSV CAM 200). 2.5. Transport experiments The transport experiments were carried out by using a cell consisting of two detachable teflon chambers. The ACM was placed between the chambers and the chambers were tightened with screws. The silicone rubber seals were used to prevent any leakage from the chambers. Equal volumes (40 mL) of the feed and stripping phases were placed in the respective chambers of the transport cell. In each experiment, the stirring rates of both phases were equal and constant at 600 rpm throughout the experiment. All experiments were performed at 25 ± 1 ◦ C and effective membrane area was 7.07 cm2 . The studied experimental parameters were initial pH (0.3–6.0) and Cr(VI) concentration of the feed phase (1 × 10−4 –5 × 10−3 M), Cyanex 923 concentration of the casting solution (0.5–5.0%), type of stripping phase (NaOH, HCl and NaCl) and concentration (0.1–1.0 M) of the most suitable stripping phase (NaOH). In addition, the effects of various metal ions (i.e., Cr(III), Ni(II), Cu(II), Zn(II), Cd(II), Co(II), etc.) as well as sulphate and nitrate ions on the Cr(VI) transport were investigated. Transport efficiency of the ACM was also compared with that of the SLM. Ionic strengths of the feed and stripping phases were equal. NaCl was used to adjust the ionic strength of both phases. The influence of the studied parameters on the transport of Cr(VI) was evaluated by means of the permeability coefficient (P). P can be defined according to Eq. (1) [23]. P=

dC 1 V C dt A

(1)

By integration of Eq. (1), the following equation is obtained: A Co = Pt Ct V

(2)

2.3. Preparation of SLM

ln

The SLM was prepared by submerging the support (Celgard 2500, which possessed a porosity of 45%, a thickness of 0.025 mm, and effective pore size of 0.04 ␮m) in kerosene containing 1% Cyanex 923. The support was impregnated for 24 h, then it was removed from the organic solution. The impregnated support was allowed to drain off for 10 min before it was used [30].

where P is the permeability coefficient (m/s), Co is the initial concentration of Cr(VI) in the feed phase and Ct is the concentration of Cr(VI) in the feed phase at time t (s). A is the effective membrane area (cm2 ) and V is the volume of the feed phase (mL). Thus, permeability coefficient can be obtained from the slope of the linear plot of ln(Co /Ct ) vs. time. A sample for ln(Co /Ct ) vs. time is presented in Fig. 1.

226

G. Arslan et al. / Journal of Membrane Science 337 (2009) 224–231

Fig. 1. A sample for the relationship of ln(Co /Ct ) vs. time for the transport of Cr(VI) through the ACM (feed phase: 1 × 10−3 M Cr(VI), initial pH of the feed phase: 1.0 ± 0.1, stripping phase: 0.5 M NaOH, ACM prepared with 1% Cyanex 923 solution).

The recovery of Cr(VI) was also defined according to the following equation: Recovery (%) = 100

C  s

Co

(3)

where Cs is the concentration of Cr(VI) in the stripping phase at time t, and Co is the initial concentration of Cr(VI) in the feed phase. In order to determine the concentration of Cr(VI), samples of 1 mL were periodically withdrawn from the feed and stripping phases (every 1 h) over 6 h, and analysed by using Continuum Source Atomic Absorption Spectrometer (ContrAA 300, Analytik jena) in air-acetylene flame at wavelength of 357.9 nm. For the competitive transport experiment, the concentrations of Cu(II), Ni(II), Zn(II), Co(II) and Cd(II) were also determined by ContrAA 300 in air-acetylene flame at wavelength of 324.7, 232.0, 213.8, 240.7 and 228.8 nm, respectively. Cr(VI) concentration in the presence of Cr(III) was determined by UV spectrophotometry (Shimadzu UV-16A) at wavelength of 540 nm based on the complex formation between 1,5-diphenyl carbazide and Cr(VI). Hence, Cr(III) concentration was determined by subtracting of the Cr(VI) concentration from the total concentration of chromium. Each analytical value reported is the mean of two replicates. A pH-meter (Orion) equipped with a combined glass–Ag/AgCl electrode was used for pH measurements. 3. Results and discussion 3.1. Surface characterization 3.1.1. Scanning electron microscopy SEM images of blank membrane (containing no carrier) and ACM prepared with 3% Cyanex 923 solution are presented in Fig. 2(a) and (b), respectively. Despite some heterogeneity, surface of the ACM seems to be compact due to the incorporation of the Cyanex 923 on the blank membrane (Fig. 2 (a)). The heterogeneity may be resulted from the variation of the polymerisation procedure in the presence of the carrier [23]. 3.1.2. Contact angle measurements The results from the contact angle measurements are given in Table 1. The contact angles of the blank and activated membranes were less than 90◦ , which means that each membrane had a hydrophilic surface. The results also indicated that contact angle

Fig. 2. (a) SEM image of blank membrane, (b) SEM image of ACM prepared with 3% Cyanex 923 solution.

of ACMs was lower than that of the blank membrane and this value slightly decreased with increase in the carrier concentration. It is likely that the capillary force working from the rough surface of the membrane increases the resistance against the movement of the water droplet on the membrane surface, hence, the higher contact angle is obtained [33,34]. Similar findings were observed in our previous paper [25], which described the facilitated transport of Cr(III) through ACM containing di-(2-ethylhexyl)phosphoric acid (DEHPA) as carrier agent. 3.1.3. Atomic force microscopy The AFM images of the blank membrane and ACMs prepared with 1–5% Cyanex 923 solutions are presented in Fig. 3(a)–(d), respectively. The differences among the surfaces of the membranes clearly indicated that immobilization of the Cyanex 923 onto the blank membrane was successfully performed. The roughnesses (Ra) for blank membrane and ACM prepared with 1%, 3% and 5% of casting solutions were found to be (77 ± 11), (21 ± 8), (13 ± 5) and (12 ± 5) nm, respectively (n = 5). It was found that Ra value of the Table 1 Results of contact angle measurements for the surface of the ACM, [n = 5]. Membrane

Contact angle (◦ )

ACM—blank ACM prepared with 1% Cyanex 923 solution ACM prepared with 3% Cyanex 923 solution ACM prepared with 5% Cyanex 923 solution

73 69 61 56

± ± ± ±

2 2 1 2

G. Arslan et al. / Journal of Membrane Science 337 (2009) 224–231

227

Fig. 3. AFM images of blank membrane (a), ACM prepared with 1% Cyanex 923 solution (b), ACM prepared with 3% Cyanex 923 solution (c) and ACM prepared with 5% Cyanex 923 solution (d).

blank membrane was distinctively higher than those of the ACMs. This result may be due to the filling of surface pores with the carrier molecules (Fig. 3). This finding is in agreement with Ref. [35], which reports that the surface roughness is enhanced when polyamides are meta-oriented due to the presence of free amide groups that are unbound to other amide groups by hydrogen bonding as a consequence of a large intermolecular N–H···O distance.

for the species of Cr(VI) and related equilibrium constants (K) are presented in Eqs. (4)–(7) [36,37]. H2 CrO4 ⇔ HCrO4 − + H+ HCrO4 − ⇔ CrO4 − + H+

HCr2 O7 ⇔ Cr2 O7

3.2.1. Effect of initial pH of feed phase In order to investigate the effect of initial pH of the feed phase on the transport of Cr(VI), the experiments were performed at initial pH of 0.3, 1.0, 2.0, 4.0 and 6.0. The results are presented in Fig. 4. A pH deviation of ±0.1 was observed for each pH measurement. The chromate ions may exist in the aqueous solution in different ionic forms (HCrO4 − , CrO4 2− , Cr2 O7 2− , HCr2 O7 − ), which depend on the Cr(VI) concentration and pH of the solution. The equilibrium

(4)

K 2 = 3 × 10−7

(5)

2HCrO4 − ⇔ Cr2 O7 2− + H2 O −

3.2. Transport experiments

K 1 = 1.21

2−

+H

+

K 3 = 35.5 K 4 = 0.85

(6) (7)

According to both Eqs. (4)–(7) and their K values, when Cr(VI) concentration is equal or less than 1 × 10−3 M at pH 6, approximately 75% of Cr(VI) exists as HCrO4 − and 25% of Cr(VI) exists as CrO4 2− [37]. In acidic solution, when Cr(VI) concentration is less than 0.02 M, HCrO4 − is the dominant species and when Cr(VI) concentration is greater than 0.02 M then Cr2 O7 2− is the dominant species [38]. Furthermore, Cr2 O7 2− converts into HCrO4 − in acidic aqueous solution at a total Cr(VI) concentration less than (1.26–1.74) × 10−2 M [39]. Based on this explanation and Refs. [40–44], in the present study, Cr(VI) ions exist as HCrO4 − in the feed phase because its concentration ranged from 1 × 10−4 to 5 × 10−3 M.

228

G. Arslan et al. / Journal of Membrane Science 337 (2009) 224–231

Fig. 4. Effect of initial pH of the feed phase on the transport of Cr(VI) through the ACM (feed phase: 1 × 10−3 M Cr(VI), stripping phase: 0.5 M NaOH, ACM prepared with 1% Cyanex 923 solution).

It was found that the transport of Cr(VI) was maximum when initial pH of the feed phase was in the range of 0.3–1.0 and an increase in pH decreased the transport of Cr(VI) (Fig. 4). Similar result was reported by Agrawal et al. [38], who studied the extractive removal of Cr(VI) from industrial waste solution by using Cyanex 923. The obtained result can be attributed to that formation of the complex between Cr(VI) and Cyanex 923 mostly favours at solution pH ≤1.0. Agrawal et al. [38] also pointed out that Cr(VI) was extracted from the hydrochloric acid media by Cyanex 923 as follows: HCrO4 − (aq) + 2H+ (aq) + Cl− (aq) + 2·Cyanex-923(org) ⇔ HCrO3 Cl·2·Cyanex-923(org) + H2 O

(8)

According to this reaction, it can be stated that rate of the complex formation depends on pH of the feed solution. A decrease in pH increases the rate of the complex formation (HCrO3 Cl·2·Cyanex923) at the feed phase-ACM interface, hence, it increases the transport of Cr(VI). Therefore, initial pH of the feed phase should be equal or less than 1.0 for an efficient transport of Cr(VI) through the ACM containing Cyanex 923 (see Fig. 4). After 6 h of transport, the final pH values were 0.8, 1.8, 2.6, 4.4 and 6.5 for the initial pH of 0.3, 1.0, 2.0, 4.0 and 6.0, respectively. Apart from the HCl, the effect of other mineral acids (i.e., H2 SO4 and HNO3 ) on the extraction of Cr(VI) by Cyanex 923 was investigated by Agrawal et al. [38]. They reported that the maximum extraction efficiency was obtained when HCl was involved in the extraction. Therefore, in further experiments of this study, initial pH of the feed phase was adjusted to 1.0 ± 0.1 by using HCl. 3.2.2. Effect of Cr(VI) concentration of feed phase In order to investigate the effect of Cr(VI) concentration on the transport, the experiments were performed with Cr(VI) concentration in the range of 1 × 10−4 –5 × 10−3 M. The results are presented in Fig. 5. The transport percentages were determined to be 94, 69 and 45%, for Cr(VI) concentrations of 1 × 10−4 , 1 × 10−3 and 5 × 10−3 M, respectively. This finding is in agreement with result reported by Reyes-Aguilera et al. [45], who observed a decrease in transport percentage with increasing feed phase concentration for bismuth recovery by using SLM containing Cyanex 921. Similar result was also reported in Ref. [46]. The effect of the Cr(VI) concentration on the transport is also evaluated by using flux values, which are defined according to Eq. (9) [47]. J = PCo

(9)

Fig. 5. Effect of Cr(VI) concentration of the feed phase on the transport (initial pH of the feed phase: 1.0 ± 0.1, stripping phase: 0.5 M NaOH, ACM prepared with 1% Cyanex 923 solution).

where, J is the flux, P is the permeability coefficient and Co is the initial Cr(VI) concentration of the feed phase. The flux values were (0.62 ± 0.07), (3.03 ± 0.08) and (8.15 ± 0.15) ␮mol/(m2 s) for Cr(VI) concentrations of 1 × 10−4 , 1 × 10−3 and 5 × 10−3 M, respectively (see Fig. 5). The obtained result is in accordance with the expected trend because according to Eq. (9) higher concentration results in a higher flux [15,28,45]. The result also indicated that membrane did not become saturated. Similar explanation was presented by Reyes-Aguilera et al. [45]. In each time, amount of Cr(VI) inside the membrane was calculated from the mass balance. The amount of Cr(VI) in the feed and stripping phases was subtracted from the original amount, corresponding to feed phase at the beginning of the experiment. As a result, no retention of Cr(VI) in the membrane was determined and time flag did not occur during the transport. Similar observation was reported by Gumi et al. [23], who studied the facilitated transport of lead(II) and cadmium(II) through ACM containing di(2-ethylhexyl)phosphoric acid.

3.2.3. Effect of Cyanex 923 concentration of casting solution The variation of the Cr(VI) transport with the concentration of Cyanex 923 in the casting solution is presented in Fig. 6. The result showed the transport of Cr(VI) increased with increasing the amount of Cyanex 923 up to 3.0%. This finding can be explained when it is considered that the stiochiometry of the Cr(VI)–Cyanex 923 complex is 1:2 [38]. Hence, increasing the amount of Cyanex 923 in the membrane resulted in a positive impact on the transport. However, transport efficiency of the ACM decreased when Cyanex 923 amount in the casting solution was beyond 3%. This result may be because of a steric hindrance effect. In other words, all the reactive sites are not involved in the transport mechanism when carrier concentration of the casting solution is over 3%. According to Navarro et al. [48], high loading of the extractant inside the porous network may limit accessibility to reactive sites. Similar explanation was reported by Martinez et al. [49] for the liquid–liquid extraction of gold by Cyanex 921 in HCl solutions. Moreover, a blank membrane (containing no carrier) was also examined in the transport experiment. As a result, any movement of Cr(VI) through the blank membrane was not determined. This result corroborates that the transport of Cr(VI) is fulfilled by the carrier.

G. Arslan et al. / Journal of Membrane Science 337 (2009) 224–231

229

Table 2 Effect of other metal ions on the recovery of Cr(VI). Metal ion in feed phase

Recovery, %

Cr(III) Ni(II) Cu(II) Co(II) Zn(II) Cd(II) Cr(VI)

No recovery No recovery No recovery No recovery 8 6 53

Experimental conditions: feed phase contains a mixture of metal ions, each metal ion concentration: 1 × 10−3 M, initial pH of the feed phase:1.0 ± 0.1 and receiver phase: 0.5 M NaOH, ACM prepared with 1% Cyanex 923 solution.

Fig. 6. Effect of Cyanex 923 amount in the casting solution on the transport of Cr(VI) through the ACM (feed phase: 1 × 10−3 M Cr(VI), initial pH of the feed phase: 1.0 ± 0.1, stripping phase: 0.5 M NaOH).

3.2.4. Effect of type and concentration of stripping phase According to the literature, selection of a suitable stripping phase is considered as one of the key factors for an effective transport system. Other researchers [5,15,38,37,50] evaluated different types of solutions (i.e., NaOH, NaCl, NaNO3 , Na2 SO3 , (NH4 )2 CO3 , etc.) for stripping of Cr(VI) from various Cr(VI)-carrier complex. In this study, NaOH, NaCl and HCl solutions were examined for Cr(VI) stripping from the Cr(VI)–Cyanex 923 complex. At the end of 6 h, 99%, 20% and 9% of Cr(VI) were transported by using the NaOH, NaCl and HCl solutions, respectively (concentration of each solution: 0.5 M). Therefore, in further studies, NaOH solution was chosen as stripping phase and the effect of NaOH concentration on the transport was investigated. It was found that recovery of Cr(VI) increased with increase in the concentration of NaOH (see Fig. 7). This result is expected because increasing the concentration of a strippant increased the decomplexation rate at the membranestripping phase interface [14]. 3.2.5. Transport mechanism The transport mechanism can be explained as follows:

Fig. 7. Effect of stripping phase concentration on the recovery of Cr(VI) (feed phase: 1 × 10−3 M Cr(VI), initial pH of the feed phase: 1.0 ± 0.1, ACM prepared with 3% Cyanex 923 solution).

i. At the feed phase-membrane interface, Cr(VI) is extracted by Cyanex 923 and the complex of HCrO3 Cl·2·Cyanex-923 was formed as described in Section 3.2.1. ii. The resulting complex diffuses through the membrane towards the membrane-stripping phase interface. iii. According to Eq. (10), at the membrane-stripping phase interface, HCrO3 Cl·2·Cyanex-923 is decomplexed by OH− and CrO3 Cl− (trioxochlorochromate) is formed. HCrO3 Cl·2·Cyanex-923 + OH− ⇔ CrO3 Cl− + H2 O + 2Cyanex-923

(10)

iv. Then, CrO3 Cl− is converted into CrO4 2− in the presence of highly basic stripping phase as follows [38]. CrO3 Cl− + H2 O ⇔ CrO4 2− + Cl− + 2H+

(11)

v. After release of Cyanex 923 by decomplexation, the transport cycle starts again. 3.2.6. Effect of other ionic species on the Cr(VI) transport Since chromium is widely used in many industries, the corresponding wastewaters may contain Cr(III) as well as other metal ions, i.e. Cu(II), Zn(II), Ni(II), Co(II) and Cd(II), etc. Therefore, the effects of these ions on the transport of Cr(VI) were investigated. The results indicated that Cr(III), Ni(II), Cu(II) and Co(II) were not transported through the ACM (Table 2). However, at the end of 6 h, 8% of Zn(II) and 6% of Cd(II) were co-transported with Cr(VI) through the ACM. Therefore, it can be concluded that Cr(VI) was preferably transported against the examined metal ions. However, transport percentage of Cr(VI) in the presence of these ions was 53%. This value was lower than that (69%) of the presence of only-Cr(VI) in the feed phase. This observation may be explained by considering the decrease in nominal carrier concentration in the ACM because Zn(II) and Cd(II) compete with Cr(VI) for transport [51]. Similar results were reported by Alguacil et al. [28], who studied the facilitated transport of Cr(VI) through SLM containing Cyanex 921 and Cyanex 923. Agrawal et al. [38] investigated the effect of various metal ions (i.e., Cr(III), Ni(II), Ca(II), Mg(II), Fe(II), Zn(II), Cu(II), etc.) on the extraction of Cr(VI) by Cyanex 923 and they reported that only Zn(II) was found to be interfering species. The results also indicated that speciation of chromium could be efficiently carried out by using the prepared ACM. Moreover, influence of the anionic species (i.e. SO4 2− and NO3 − , etc.) on the transport of Cr(VI) was investigated under the following experimental conditions (feed phase: mixture of Cr(VI), SO4 2− and NO3 − (each concentration: 1 × 10−3 M), initial pH of feed phase: 1.0 ± 0.1, stripping phase: 0.5 M NaOH, ACM prepared with 1% Cyanex 923 solution). The result showed that transport of Cr(VI) was not influenced by SO4 2− and NO3 − ions. Similar result was reported by Agrawal et al. [38].

230

G. Arslan et al. / Journal of Membrane Science 337 (2009) 224–231

Table 3 Comparison of the transport efficiency of the ACM and SLM.a . Membrane

Thickness (␮m)

J (␮mol/(m2 s))

JN (␮mol/(m2 s))

ACM SLM

90 ± 5 25

3.03 11.94

10.91 11.94

a Both membranes were prepared with 1% Cyanex 923 solution, feed phase: 1 × 10−3 M Cr(VI), initial pH of the feed phase 1.0 ± 0.1, stripping phase: 0.5 M NaOH.

3.2.7. Reproducibility and stability of ACM The transport behavior of the ACMs prepared from four different batches of 1% Cyanex 923 solution was examined under the following experimental conditions (initial pH of feed phase: 1.0 ± 0.1, Cr(VI) concentration: 1 × 10−3 M, stripping phase: 0.5 M NaOH). The results indicated that transport efficiency of the ACM was reproducible (relative standard deviation ≤4%, mean transport percentage: 69%). The stability of the ACM was investigated by evaluation of the permeability coefficients from eight sequential experiments in which the same membrane was used under the conditions above. It was found that the permeability coefficient remained practically constant, (2.98 ± 0.09 to 3.05 ± 0.11) × 10−6 m/s, within all transport cycles (each cycle: 6 h). As a result, the prepared ACM can be used in the long-term separation processes. 3.2.8. Comparison of Cr(VI) transport efficiency of ACM and SLM Maintaining the same content of the feed and stripping phases, transport efficiency of the ACM was compared with that of the SLM by means of the flux values as defined according to Eq. (9). The results are presented in Table 3. The flux from the SLM (11.94 ␮mol/m2 s) was higher than that from the ACM (3.03 ␮mol/m2 s). The obtained result is because of the different thickness of the membranes. Therefore, to make a real comparison between the efficiencies of the membranes, experimental flux from the ACM should be normalized to the SLM flux according to Eq. (12) [52].



JN = JEXP ×

ıACM ıSLM



(12)

where JN is the normalized flux (␮mol/(m2 s)), JEXP is the experimental flux (␮mol/(m2 s)), ıACM and ıSLM are the thickness of the ACM and SLM, respectively. The results in Table 3 show that the normalized flux from the ACM is close to the experimental SLM flux. The obtained results are in agreement with the results reported by Kozlowski and Walkowiak [16], who studied the applicability of SLM and PIM on the transport of Cr(VI) with different amines as ion carriers. They observed that transport efficiency of SLM was higher than that of PIM. However, after normalization of the fluxes, they concluded that PIM could be used instead of SLM for practical applications because normalized flux for PIM were close to experimental SLM flux. When it is considered that one of the major problems concerning SLMs is their low stability due to loss of either a membrane solvent or a carrier [30], we can also conclude that ACM can be used instead of SLM for practical application. 4. Conclusion In this study, the facilitated transport of Cr(VI) through a novel ACM containing Cyanex 923 was investigated. First, the prepared ACM was characterized by using SEM and AFM techniques and contact angle measurements. Then, the transport experiments were carried out. The obtained results can be concluded as follows: i. The results from the characterization studies corroborated that Cyanex 923 was successfully immobilized onto the poly-

ii.

iii.

iv.

v.

sulfone support. Hence, the preparation of the ACM was achieved. The transport of Cr(VI) through the ACM was influenced by a number of variables, including initial pH and Cr(VI) concentration of the feed phase, Cyanex 923 concentration of the casting solution and type and concentration of the stripping phase. The maximum transport of Cr(VI) was obtained when the following experimental conditions were employed: initial pH of feed phase ≤1.0, ACM prepared with 3% Cyanex 923 solution and stripping phase of 1 M NaOH. In addition, increasing the concentration of Cr(VI) in the feed phase decreased the transport percentage. The Cr(VI) was preferably transported against the other metal ions (i.e., Cr(III), Ni(II), Cu(II), Zn(II), Cd(II) and Co(II), etc.). From this result, it can be also concluded that speciation of chromium can be efficiently carried out by using the prepared ACM. It was found that sulphate and nitrate ions had no negative effect on the transport of Cr(VI). The obtained results also showed that transport efficiency of the ACM was reproducible and it could be efficiently used in the longterm separation processes instead of the SLM.

Acknowledgments We thank the Scientific and Technical Research Council of Turkey (TUBITAK-Grant number 104Y394) and Selcuk University Research Foundation (SUAF) for the financial support of this work. References [1] ATSDR, Agency for Toxic Substances and Disease Registary, Division of Toxicology/Toxicology Information Branch (ATSDR), Toxicological profile for chromium, ATSDR, Atlanta, USA, 2000, pp. 1–157. [2] EPA, Environmental Protection Agency, Environmental Pollution Control Alternatives, EPA/625/5-90/025, EPA/625/4-89/023, Cincinnati, US, 1990. [3] A.K. Sengupta, S. Subramonian, D. Clifford, More on mechanism and some important properties of chromate ion exchange, J. Environ. Eng. 114 (1988) 137–153. [4] S.L. Lo, S.F. Shiue, Recovery of Cr(VI) by quaternary ammonium compounds, Water Res. 32 (1998) 174–178. [5] P. Venkateswaran, K. Palanivelu, Solvent extraction of hexavalent chromium with tetrabutyl ammonium bromide from aqueous solution, Sep. Purif. Technol. 40 (2004) 279–284. [6] I. Ortiz, B. Galon, A. Arabien, Kinetic analysis of the simultaneous nondispersive extraction and back extraction of Cr(VI), Ind. Eng. Chem. Res. 35 (1996) 1369–1377. [7] A.I. Alonso, B. Galon, M. Gonzalez, I. Ortiz, Experimental and theoretical analysis of a non-dispersive solvent extraction pilot plant for the removal of Cr(VI) from a galvanic process waste waters, Ind. Eng. Chem. Res. 38 (1999) 1666–1675. [8] A. Hafiane, D. Lemordant, M. Dhahbi, Removal of hexavalent chromium by nanofilteration, Desalination 130 (2000) 305–312. [9] L. Gzara, M. Dhahbi, Removal of chromate anions by micellar enhanced ultrafilteration using cationic surfactants, Desalination 137 (2000) 241–250. [10] E.K. Paleologos, C.D. Stalikas, S.M. Tzouware-Karayanni, M.I. Karayannis, Selective speciation of trace chromium through micelle-mediated preconcentration, coupled with micellar flow injection analysis spectrofluorimetry, Anal. Chim. Acta 436 (2001) 49–57. [11] Y. Cengeloglu, A. Tor, E. Kir, M. Ersoz, Transport of hexavalent chromium through anion exchange membranes, Desalination 154 (2003) 239–246. [12] A. Tor, T. Buyukerkek, Y. Cengeloglu, M. Ersoz, Simultaneous recovery of Cr(III) and Cr(VI) from the aqueous phase with ion-exchange membranes, Desalination 171 (2004) 233–241. [13] M.E. Valleyo, F. Persin, C. Innocent, P. Sistas, G. Pourcelly, Electrotransport of Cr(VI) through anion exchange membrane, Sep. Purif. Technol. 21 (2000) 61–69. [14] A. Yilmaz, A. Kaya, H.K. Alpoguz, M. Ersoz, M. Yilmaz, Kinetic analysis of chromium(VI) ions transport through a bulk liquid membrane containing p-tert-butylcalix[4]arene dioxaoctylamide derivative, Sep. Purif. Technol. 59 (2008) 1–8. [15] P. Venkateswaran, K. Palanivelu, Studies on recovery of hexavalent chromium from plating wastewater by supported liquid membrane using tri-n-butyl phosphate as carrier, Hydrometallurgy 78 (2005) 107–115. [16] C.A. Kozlowski, W. Walkowiak, Applicability of liquid membranes in chromium(VI) transport with amines as ion carriers, J. Membr. Sci. 266 (2005) 143–150. [17] E. Salazar, M.I. Ortiz, A.M. Urtiaga, J.A. Irabien, Kinetics of the separation concentration of Cr(VI) with emulsion liquid membranes, Ind. Eng. Chem. Res. 31 (1992) 1523–1529.

G. Arslan et al. / Journal of Membrane Science 337 (2009) 224–231 [18] L. Melita, M. Popescu, Removal of Cr (VI) from industrial water effluents and surface waters using activated composite membranes, J. Membr. Sci. 312 (2008) 157–162. [19] M. Ersoz, Transport of mercury through liquid membranes containing calixarenes, Adv. Colloid Interface Sci. 134–135 (2007) 96–104. [20] M. Mulder, Basic Principles of Membrane Technology, Kluwer Academic Publishers, Dordrecht, 1992. ˜ [21] M. Oleinikova, C. Gonzalez, M. Valiente, M. Munoz, Selective transport of zinc through activated composite membranes containing di(2ethylhexyl)dithiophosphoric acid as a carrier, Polyhedron 18 (1999) 3353–3359. [22] M.E. Paez-Hernandez, K. Aguilar-Arteaga, M. Valiente, M.T. Ramirez-Silva, M. Romero-Romo, M. Palmomar-Pardave, Facilitated transport of Hg(II) through novel activated composite membranes, Anal. Bioanal. Chem. 380 (2004) 690–697. ˜ [23] T. Gumi, M. Oleinikova, C. Palet, M. Valiente, M. Munoz, Facilitated transport of lead(II) and cadmium(II) through novel activated composite membranes containing di-(2-ethyl-hexyl)phosphoric acid as carrier, Anal. Chim. Acta 408 (2000) 65–74. [24] A. Calzado, C. Palet, M. Valiente, Facilitated transport and separation of aromatic amino acids through activated composite membranes, Anal. Chim. Acta 431 (2001) 59–67. [25] G. Arslan, A. Tor, Y. Cengeloglu, M. Ersoz, Facilitated transport of Cr(III) through activated composite membrane containing di-(2ethylhexyl)phosphoric acid (DEHPA) as carrier agent, J. Hazard. Mater. (2008) doi:10.1016/j.jhazmat.2008.10.050. [26] Y.A. El-Nadi, N.E. El-Hefny, J.A. Daoud, Extraction of lanthanum and samarium from nitrate medium by some commercial organophosphorus extractants, Solv. Extr. Ion Exch. 25 (2007) 225–240. [27] B.R. Reddy, P. Radhika, J.R. Kumar, D.N. Priya, K. Rajgopal, Extractive spectrometric determination of Co(II) in synthetic and pharmaceutical samples using CYANEX 923, Anal. Sci. 20 (2004) 345–349. [28] F.J. Alguacil, A. López-Delgado, M. Alonso, A.M. Sastre, The phosphine oxides Cyanex 921 and Cyanex 923 as carriers for facilitated transport of chromium(VI)-chloride aqueous solutions, Chemosphere 57 (2004) 813–819. [29] F.J. Alguacil, A.G. Coedo, M.T. Dorado, Transport of chromium (VI) through a Cyanex 923—xylene flat-sheet supported liquid membrane, Hydrometallurgy 57 (2000) 51–56. [30] A. Tor, G. Arslan, H. Muslu, A. Celiktas, Y. Cengeloglu, M. Ersoz, Facilitated transport of Cr(III) through polymer inclusion membrane with Di(2-ethylhexyl)phosphoric acid (DEHPA), J. Membr. Sci. 329 (2008) 169–174. [31] B. Sartowska, M. Buchzkowski, W. Starosta, SEM observations of particle track membrane surfaces modificated using plasma treatment, Mater. Chem. Phys. 81 (2003) 352–355. [32] F. Garbassi, M. Morra, E. Occhiello, Polymer Surfaces. From Physics to Technology, Wiley, New York, 1994. [33] T. Gumi, M. Valiente, K.C. Khulbe, C. Palet, T. Matsuura, Characterization of activated composite membranes by solute transport, contact angle measurement, AFM and ESR, J. Membr. Sci. 212 (2003) 123–134. [34] K.C. Khulbe, G. Chowdhury, B. Kruczek, R. Vujosevic, T. Matsuura, G. Lamarche, Characterization of the PPO dense membrane prepared at different temperatures by ESR, atomic force microscope and gas permeation, J. Membr. Sci. 126 (1997) 115–122.

231

[35] S.-Y. Kwak, S.G. Jung, Y.S. Yoon, D.W. Ihm, Details of surface features in aromatic polyamide reverse osmosis membranes characterized by scanning electron and atomic force microscopy, J. Polym. Sci. Part B: Polym. Phys. 37 (1999) 1429–1440. [36] J.K. Cabatingan, R.C. Agapay, J.L.L. Rakels, M. Ottens, L.A.M. Vander Wielen, Potential of biosorption for the recovery of chromate in industrial wastewaters, Ind. Eng. Chem. Res. 40 (2001) 2302–2309. [37] B. Saha, R.J. Gill, D.G. Bailey, N. Kabay, M. Arda, Sorption of Cr(VI) from aqueous solution by Amberlite XAD-7 resin impregnated with Aliquat 336, React. Func. Polym. 60 (2004) 223–244. [38] A. Agrawal, C. Pal, K.K. Sahu, Extractive removal of chromium (VI) from industrial waste solution, J. Hazard. Mater. 159 (2008) 458–464. [39] A. Bhowal, S. Datta, Studies on transport mechanism of Cr(VI) extraction from an acidic solutions using liquid surfactant membranes, J. Membr. Sci. 188 (2001) 1–8. [40] R.A. Kumbasar, Selective separation of chromium (VI) from acidic solutions containing various metal ions through emulsion liquid membrane using trioctylamine as extractant, Sep. Purif. Technol. 64 (2008) 56–62. [41] R.A. Kumbasar, Studies on extraction of chromium (VI) from acidic solutions containing various metal ions by emulsion liquid membrane using Alamine 336 as extractant, J. Membr. Sci. 325 (2008) 460–466. [42] W. Zhang, J. Liu, Z. Ren, S. Wang, C. Du, J. Ma, Kinetic study of chromium(VI) facilitated transport through a bulk liquid membrane using tri-n-butyl phosphate as carrier, Chem. Eng. J. (2008) doi:10.1016/j.cej.2008.12.009. [43] M. Chiha, M.H. Samar, O. Hamdaoui, Extraction of chromium(VI) from sulphuric acid aqueous solutions by a liquid surfactant membrane (LSM), Desalination 194 (2006) 69–80. [44] P.J. Harrington, G.W. Stevens, Steady-state mass transfer and modelling in hollow fiber liquid membranes, J. Membr. Sci. 192 (2001) 83–98. [45] J.A. Reyes-Aguilera, M.P. Gonzalez, R. Navarro, T.I. Saucedo, M. Avila-Rodriguez, Supported liquid membranes (SLM) for recovery of bismuth from aqueous solutions, J. Membr. Sci. 310 (2008) 13–19. [46] E. Ergun, A. Tor, Y. Cengeloglu, I. Kocak, Electrodialytic removal of fluoride from water: effects of process parameters and accompanying anions, Sep. Purif. Technol. 64 (2008) 147–153. [47] F.J. Alguacil, A.G. Coedo, M.T. Dorado, A.M. Sastre, Uphill permeation of chromium Cr(VI) using Cyanex 921 as ionophore across an immobilized liquid membrane, Chemosphere 61 (2001) 13–19. ˜ [48] R. Navarro, I. Saucedo, A. Núnez, M. Ávila, E. Guibal, Cadmium extraction from hydrochloric acid solutions using Amberlite XAD-7 impregnated with Cyanex 921 (tri-octyl phosphine oxide), React. Funct. Polym. 68 (2008) 557–571. [49] S. Martínez, A. Sastre, F.J. Alguacil, Gold extraction equilibrium in the system Cyanex 921-HCl-Au(III), Hydrometallurgy 46 (1997) 205–214. [50] L. Sadoun, F. Hassaine-Sadi, Purification–concentration process. Studies on the transport mechanism of a chromium (VI)-sulfuric acid tri-noctylamine (TOA)–ammonium carbonate system, Desalination 167 (2004) 159–163. [51] J. De Gyves, E. Rodriguez de San Miguel, Metal ion separations by supported liquid membranes, Ind. Eng. Chem. Res. 38 (1999) 2182–2202. ˜ [52] M. Resina, C. Fontás, M. Cristina Palet, Munoz, Comparative study of hybrid and activated composite membranes containing Aliquat 336 for the transport of Pt(IV), J. Membr. Sci. 311 (2008) 235–242.