Technique of supported liquid membranes (SLMs) for the facilitated transport of vanadium ions (VO
2+)
Parameters and mechanism on the transport process
M. Hor
a, A. Riad
a, A. Benjjar
a, L. Lebrun
b, M. Hlaïbi
a,b,⁎
aLaboratoire Interface Matériaux- chimie de l'Environnement (LIME), Université Hassan II, Faculté des Sciences Aïn Chock, B.P. 5366, Maârif, Casablanca, Morocco
bLaboratoire Polymères, Biopolymères et Surfaces, UMR 6270 CNRS, Université de Rouen, Faculté des Sciences, F-76821 Mont-Saint-Aingnan, France
a b s t r a c t a r t i c l e i n f o
Article history:
Received 23 July 2009
Received in revised form 18 December 2009 Accepted 30 December 2009
Available online 6 February 2010 Keywords:
Facilitated transport Supported liquid membrane Permeability
Flux
Apparent diffusion coefficient Association constant
Two supported liquid membranes (SLMs) containing two different carriers but formed by the same polymer support polyvinylidene difluoride (PVDF) and the same organic phase (xylene), were used to realize the facilitated transport of vanadium ions (VO2+) from the concentrated acid solutions. The SLM support was a micro porousPVDFpolymerfilm of thickness 100 µm with pore size 0.45 µm which has been impregnated with a solution of xylene containing 0.01 M of one of the carriers, the Di-(2-ethylhexyl) phosphoric acid (D2EHPA) or the Trioctylphosphine oxide (TOPO). The permeabilitiesPand the initialfluxesJ0for the transport of the VO2+ions were calculated from the proposed kinetic model for the two SLMs used. These two macroscopic parameters depend on both the concentration of the carrier (T) and that of the substrate (S) to be transported. The proposed mechanism indicates the formation of a complex (1/1) carrier–substrate (TS), and the migration of this complex through the organic phase of the SLM, is the rate-determining step in the transport mechanism. The initialfluxJ0is related to the initial substrate concentrationC0in the feed phase by a saturation law, which allowed the determination of the apparent diffusion coefficientsD*and the association constantsKassof the complexes (TS), formed in the organic phase of the two studied SLMs. These results clearly indicate that the values of these two microscopic parameters (D*andKass), depend certainly on the acidity of the medium and probably the concentration of co-ions NO3−.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
Currently, the membranes are used in many industrial applica- tions, either to recover or separate the components of a mixture or to control the selectively exchange of material between different environments. The use of membrane technology in recent years experienced rapid growth, particularly given the increasing appli- cation areas. This development should increase because of the emergence of the needs of environmental protection and through
energy efficiency and techno-economic increasingly competitive offered by membrane process. On the other hand, more advanced searches are conducted and are designed to better understand the functioning of membranes, to prepare and characterize membranes more efficiently and specifically and also to develop processes to access new applications[1,2].
Today, to advance in this area, it has become necessary and required to develop highly selective systems that are essential to achieve some separation and recovery of metal ions highly polluting to the environment (especially for radioactive species) from complex aqueous mixtures. For this purpose, the technique of liquid–liquid extraction was thefirst separation method used, with agents more or less suitable for the recovery of metal ions, from complex and loaded industrial solutions. First, this technique involves the use of complex- ing agents and large amounts of organic solvents which are often expensive and toxic. On the other hand, it includes two steps, an extraction with phase transfer, followed by a reextraction, these two steps can be quite consuming of organic solvents, especially in volatile solvents. An elegant alternative to liquid–liquid extraction is the development of artificial membrane systems, which reproduce the process of facilitated transport across bio-membranes, carried out by mobile carriers.
Abbreviations: a, Slope of the plot−ln (C0−2Cr) =f(t);C0, Initial concentration of VO2+
ions in the feed phase (mol L−1);Cr, Concentration of VO2+
ions in the receiving phase (mol L−1);Cs, Concentration of VO2+
ions in the source phase (mol L−1);P, The permeability of VO2+
ions (cm2s−1);J0, Initialflux of VO2+
ions (mmol cm−2s−1);D*, Apparent diffusion coefficient of the complex (TS) (cm2s−1);Kass, Association constant of the complex (TS) (L mol−1); l, The membrane thickness (mm or µm); S, The membrane area (cm2); [T]0, Concentration of carrier in the membrane (mol L−1); [TS], Concentration of the complex in the organic phase (mol L−1);T, Temperature (K or °C);
t, Time (s);V, Volume of the receiving compartment (cm3).
⁎ Corresponding author. Laboratoire Interface Matériaux-chimie de l'Environnement (LIME), Université Hassan II, Faculté des Sciences Aïn Chock, B.P. 5366, Maârif, Casablanca, Morocco.Tel.: + 212 5 22 23 06 80; fax: + 212 5 22 23 06 74.
E-mail address:miloudi58@hotmail.com(M. Hlaïbi).
0011-9164/$–see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.desal.2009.12.023
Contents lists available atScienceDirect
Desalination
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / d e s a l
Liquid membranes incorporating specific complexing agents are artificial systems well suited for the treatment of liquid medium loaded with metal ions. Of these membranes, supported liquid mem- branes (SLM) are the most commonly used systems for these ap- plications. These systems can be developed in the presence of an inert polymer support. Indeed, the organic solution containing a specific molecule (carrier) is incorporated most frequently by impregnating this polymer support. The polypropylene polymer support is the most used because of its high porosity, which generates the bestfluxes of metal ions through the SLM. The processes supported liquid membrane, have several advantages compared to liquid–liquid extraction. Indeed, this type of liquid membranes contains small amounts of organic solvents, which is an important criterion with respect to the constraints of environmental protection and limiting toxic emissions, and allows continuous operation, since the two steps, extraction and reextraction are so coupled to the two interfaces. However, a supported liquid membrane (SLM) is in the form of a thin barrier that separates two fluids, an essential property of these membranes is their permeability, their ability to allow selective passage of some species, from one medium to another, acting as a barrier to other species.
For this work, we have developed a simple and effective technique for the extraction of vanadium ions (VO2+
) from the concentrated acid environments. This technique is based on work related to membrane transport phenomena, especially the facilitated transport through the SLM. On the one hand, our objective is the development of specific liquid membranes, capable of extracting metal ions from industrial concentrated solutions of phosphoric acid, to purify these solutions obtained by hydrolysis of phosphate ore. On the other hand, the extraction of the oxygen ions (VO2+) through the liquid membranes and the determination of parameters relating to transport, to under- standing the mechanism of facilitated transport of oxygen ions. Our work will be limited to the supported liquid membrane (SLM), con- sisting of an inert micro porous polymer, polyvinylidene difluoride (PVDF), containing one of the following amphiphile carriers, the Di-(2-ethylhexyl) phosphoric acid (D2EHPA) or trioctylphosphine oxide (TOPO) (Fig. 1), soluble in xylene phase. These two complexing agents were often used for the extraction of some metal ions[3–6].
In addition, a kinetic model as well as a transport mechanism have been developed and tested for transport of VO2+ions from different solutions. The macroscopic parameters, permeabilities (P) and initial fluxes (J0), were determined and related to the microscopic charac- teristics, apparent diffusion coefficients (D*) and association constants (Kass), on the complex (Carrier–Substrate) formed in organic phase.
2. Description and operating principle of the SLM
A supported liquid membrane (SLM) consists of an organic solvent, immobilized in the pores of a polymer support by capillary forces, this membrane separates two compartments, the feed compartment
contained the substrate solution and the receiving compartment contained pure water. The support of these membranes is in general an inert hydrophobic micro porous polymer, which is characterized by a low thickness of around 25 to 100 µm, and pore diameters of 0.12 to 1 µm[7,8]. The passage of chemical species through these membranes, is an interfacial phenomenon, so the use of a support with a large porosity is very important and necessary, to increase the contact area and ensure the best conditions for transport, separation and selectivity for the transported species, across the supported liquid membrane (SLM)[8]. In order to improve the process of the separation and the transport through these membranes, researchers have added to the organic phase of the SLM, mobile carriers to accelerate and facilitate the transport of species, while increasing the selectivity of these prepared membranes[7–12]. The MLS technique is an approach widely used for the extraction and enrichment of metal ions and organic compounds [13–15]. Indeed, this technique has been used to study transport, selective extraction and enrichment of organic compounds such as amino acids[16,17], aromatic amino phosphates[18], sugars[19–21], herbicides[22,23]and some organic acids[24,25].
This process known facilitated transport through supported liquid membranes is based on the recognition of a substrate (S) by an amphiphilic molecule acting as a carrier (T). The process involves the mobility of a complex (TS) within the membrane, resulting in a reversible reaction between the carrier (T) and substrate (S), taking place at the interfaces of the membrane with the feed and receiving phases. This phenomenon of facilitated transport through the SLM is a cyclical process that takes place infive consecutive steps:
1- Diffusion of the substrate (S) in the source phase to the interface of the membrane.
2- Training of substrate-carrier complex (TS) at the membrane interface with the source phase.
3- Diffusion of the complex (TS) through the membrane organic phase to the interface of the membrane with the receiving phase.
4- Dissociation of the complex (TS) at the membrane interface with the receiving phase.
5- Diffusion of free substrate (S) in the receiving phase and the carrier (T) in the membrane organic phase to participate in the following cyclic process (complexation/diffusion/dissociation)
The diagram inFig. 2represents thefive steps on the facilitated transport mechanism through an SLM, with step three as a kinetically determining step.
3. Experimental procedure and theoretical model 3.1. Conditioning of the SLM
The transport of a model solution of vanadium ions (VO2+) (0.10 M in feed phase), was achieved through a SLM in which the membrane phase was a 0.01 M solution of Di-(2-ethylhexyl) phosphoric acid (D2EHPA) in pure xylene solvent, supported by a micro porous polyvinylidene difluoride (PDVF)film.
Fig. 1.Structure of tow carriers: (a)D2EHPAand (b)TOPO. Fig. 2.Facilitated transport mechanism through SLM.
When the SLM was used for transport immediately after its preparation, theflux of vanadium ions VO2+remained small for a long initial period, typically 8 to 10 h. After this time, the rate of transport increased and the concentration of vanadium ions (VO2+) in the receiving phaseCr, increased rapidly over several hours. This behavior has been observed previously with a SLM containing some lipophilic carriers[16,17], and was attributed to the slow incorporation of water into the membrane to form presumably a carrier–water complex.
After this initial period (induction period), the transport of vanadium ions (VO2+) increases, probably because the substrate rapidly exchanges with water at the interfaces, to form the carrier–substrate complex, or possibly a ternary carrier–substrate–water complex that is the active species for the transport process. The presence of water in the SLM was reported to be important phenomenon for the transport of carbohydrates through plasticized cellulose triacetate membranes containing ion-pair carriers[26,27].
The existence of this induction period complicates the analysis of the kinetic studies, because it makes difficult to approximate the exact value of the slope of the plots drawn for afirst-order reaction[21]. In this study, in order to suppress this period of slow transport, the prepared SLM was conditioned for 16 h in the cell between two phases of pure water. After this time, the induction period was not observed and the transport of vanadium ions (VO2+) began just after its introduction in the feed phase.
3.2. Transport cell
The transport experiments were performed in the cell represented by the diagram in Fig. 3. This cell has two compartments with a volume of each is 175 mL, separated by the micro porous membrane (M). This transport unit is immersed in a thermostated bath (TB) and a multi-magnetic stirrer; allow stirring the solutions in both compartments.
3.3. Determination of permeability andflux
The SLM is placed between two compartments (Fig. 3). Into the feed compartment, a known volume of a substrate solution (S) with a concentrationC0 is introduced, and we place the same volume of water in the receiving compartment. Small aliquots (v= 1.0 mL) of the receiving phase were withdrawn at known intervals. IfCris the concentration of substrate in the receiving phase at a time t, the substrate concentration in the source phaseCsat this time t is:Cs= C0−Cr. The evolution of the concentrationCrof substrateSin the receiving phase vs. time is related to theflux J by the following equation:
dCr=dt=J×S=V ð1Þ
with,Sas the diffusion surface of the membrane andVthe volume of receiving phase.
For a quasi-stationary state, theflux is related to the difference of the concentrations of two compartments,ΔC=Cs−Cr, by the Fick's first law:
J=P×ΔC=l ð2Þ
Pis the permeability of the membrane andlits thickness.
AsCs=C0−Cr;thereforeΔC=Cs−Cr=C0−2Cr ð3Þ by combining Eqs. (1)–(3), wefind the following equation:
Pdt=ðl×V=SÞdCr=ðC0−2CrÞ ð4Þ After integration:
Pðt−tLÞ=ðl×V=2SÞln½C0=ðC0−2CrÞ ð5Þ This equation shows that after an induction period (tL), which can last several hours, if the term−ln(C0−2Cr) of Eq. (5), follows a linear evolution as a function of the time, then, the diffusion rate of the substrate (S) through the SLM, is carried out by a kinetic law offirst order. The permeability is calculated from the slope“a”of the right (−ln (C0−2Cr) =f(t)), using the following equation:
P=a×Vl=2S ð6Þ
and that the initial flux J0, which can be calculated from the permeabilityPaccording to Eq. (7):
J0=P×C0=l ð7Þ
While, equality Cs=Cr=C0/2 of the concentrations in both compartments, reflects a dynamic equilibrium state, with equal diffusion rates of the substrate (S) through the SLM in the two opposite directions, and indicates the end of the transport process.
3.4. Model and theoretical calculations
The phenomenon of facilitated transport depends on the forma- tion and dissociation of the complex carrier–substrate (TS) at the solution–membrane interfaces. The carrier (T) is insoluble in the aqueous phase, while the substrate (S) is insoluble in the organic phase of the membrane. The complexation equilibrium is:
Torg + Saq⇔TSorg
org and aq represent respectively, the organic phase of the membrane and the feed aqueous phase. The concentration [TS]i of carrier– substrate complex (TS) at the membrane-source phase interface, is expressed from the mass action law according to the equation:
½TSi=Kass½Ti½Si ð8Þ
[T]iand [S]iare respectively the concentrations of the carrier and the substrate at the source phase-membrane interface.Kass is the association constant of the complex (TS), is also the constant of the heterogeneous equilibrium established at this interface. In kinetically determining step, thefluxJis determined by Eq. (9), derived from the Fick'sfirst law, which implies that the concentration of the complex is negligible at the receiving phase-membrane interface (dissociation of complex).
J=ðD*=lÞ×½TS ð9Þ
D*is the apparent diffusion coefficient andlis the thickness of membrane.
Fig. 3.Scheme of the transport cell.
However, at the source phase-membrane interface, [TS]i≪[S]i
(excess of substrate relative to the carrier) and at any moment, the concentration [S]tof the substrate in the feed phase is equal to the concentration [S]iat the interface of the membrane ([S]i= [S]t). The total carrier concentration [T]0 immobilized in the membrane is constant, equal to the sum of the concentrations [T]iand [TS]i.
½T0=½Ti+½TSi=½TSi×½ð1 +Kass×½SiÞ=ðKass×½SiÞ ð10Þ
½TSi=½T0×Kass×½Si=ð1 +Kass×½SiÞ ð11Þ But in the initial conditions, the substrate is in excess of the carrier and at the source phase-membrane interface, we can write [S]i= [S]0=C0and [TS]i≈[T]0.
Using Eqs. (9) and (11), the expression of initialfluxJ0= (D*/l) × [TS]iis given by the following relationship:
J0=ðD*=lÞ×ð½T0×Kass×½SiÞ=ð1 +Kass×½SiÞ
Hence thefinal expression ofJ0from the initial concentration, [T]0andC0with an excess of substrate (C0>> [T]0):
J0=ðD*=lÞ×ð½T0×Kass×C0Þ=ð1 +Kass×C0Þ ð12Þ This expression allows to calculate the permeabilityPas a function of [T]0,C0andKassusing the following equation:
P=J0×l=C0=ðD*Þ×ð½T0×KassÞ=ð1 +Kass×C0Þ ð13Þ We note that the evolution ofJ0andPis proportional to the initial substrate concentrationC0, and it is a Michaelis–Menten type, since for high concentrations of substrate, these two macroscopic para- meters reaching the limit values. To determine the microscopic parametersD*andKass, the relationship given by Eq. (12) is linearized according to the Lineweaver–Burk plot 1/J0=f(1 /C0), Eq. (14):
1=J0=ðl=D*Þ×½ð1=½T0×KassÞ×ð1=C0Þ+ð1=½T0Þ ð14Þ
and values ofKassandD*are calculated from the relationships:
Kass= InterceptðOOÞ=slopeðpÞandD* =ðl=OOÞ×ð1=½T0Þ ð15Þ
4. Experimental section
All chemicals, reagents and solvents were pure commercial products (Aldrich or Fluka) of analytical grade, used as received. The acid hydrolysis of pentaoxide vanadium V2O5in concentrated nitric acid solutions, allows obtaining studied solutions of vanadium ions (VO2+), with various concentrations of nitrate ions NO3−(1.0, 1.5 and 2.0 M). The SLM support was a micro porous PVDFfilm (Millipore) of thickness 100 µm. Characteristic values are porosity 60% and pore size 0.45 µm. The membrane area available for diffusion was 20 cm2 (diameter 5.0 cm).
The transport cell (Fig. 3) is made of two compartments of equal volumes (175 mL) separated by the SLM, prepared by soaking a square portion of the polymerfilm into a 0.01 M organic solution of one carrier D2EHPA or TOPO in pure xylene, during 10 h. The cell is immersed into a thermostated bath (T, 298 K). Initially, the feed phase is a solution of vanadium ions (VO2+
), with an initial concentrationC0
and afixed value of pH, while the receiving phase is a solution of pure water at the same pH. The solutions in both compartments are stirred with magnetic bars, using a Multimatic-9 S apparatus. Two different techniques are used to study the facilitated transport of substrate (vanadium ions (VO2+)) through the SLM.
− In the first one (mode a), for the study of the lifetime of the membrane and the reproducibility of results, we used for each of the experiments a new substrate solution with the samefixed concentration. With this procedure, we have verified that the results are reproducible with the same membrane was used during 10 days without showing any sign of failure.
− In the second one (mode b), to study the effect of the concentration of the substrate on the parameters of transport, the same membrane and the same solution have been used for all runs. At equilibrium after each run, the contents of both compartments were withdrawn and mixed together, then 175 mL of the resulting solution were introduced in the feed compartment and 175 mL of pure water at the same pH in the receiving compartment and the following run was started. This economic procedure for vanadium ions solutions (VO2+), was typically repeated four orfive times.
In these techniques, small aliquots (v= 0.50 mL) of the receiv- ing phase were withdrawn at known intervals. These samples were analyzed by standard HeliosγUV–visible spectrometer, the Table 1
Facilitated transport of substrate (VO2+) through an SLM.
T(min) Absorbance Cr(M) −ln(C0−2Cr)
90 0.130 0.014 2.118
180 0.210 0.024 2.283
270 0.310 0.035 2.538
360 0.391 0.044 2.805
420 0.410 0.046 2.880
C0= [VO2+
]0= 0.15 M, [HNO3] = 1.5 M, [D2EHPA] = 0.01 M,T= 298 K.
Fig. 4.Plots of−ln(C0−2Cr) vs. timetfor the transport of VO2+
ions through the SLM.
concentrationsCrof vanadium ions in the receiving phase are calcu- lated and the evolution of the function−ln(C0−2Cr) is determined as a function of time. The following table lists an example of results on the facilitated transport of vanadium ions (VO2+) from acidic medium [H3O+] = [NO3−] = 1.5 M (Table 1).
Some experiments were continued for 24 h, when samples of both aqueous phases were withdrawn and analyzed to ensure that equal concentrations were present, indicating that equilibrium was reached.
5. Results and discussion
Under the same experimental conditions, using the SLM with 0.01 M of the carrier (D2EHPA) dissolved in the organic phase, facilitated transport of substrate (VO2+) was carried out from different acid solutions with the successive acidity, [HNO3] = 1.0, 1.5 and 2.0 M.
The kinetic model proposed for this type of transport, indicates that the evolution of the term−ln(C0−2Cr) should be linear with time, which is verified by the straight lines represented by the graph in Fig. 4. The slopes determined from these straight lines, allow calculating the permeabilities P of this SLM for vanadium ions (VO2+) of different studied solutions (Eq. (6)), while the initialfluxes J0through the SLM for these ions, are calculated from the Eq. (7). The results relating to these parameters obtained for the different studied solutions are summarized inTable 2.
These results show that the permeability P of the SLM varies inversely with the initial concentrationC0of vanadium ions (VO2+) and an increase inC0results in a decrease ofP. In contrast, the initial fluxJ0of these ions (VO2+) through the SLM increases with the initial concentrationC0of substrate in the feed phase[17,18]. The study of
the Lineweaver–Burk plots (1/J0=f(1 /C0)) provided by Eq. (14), allows to verify the proposed mechanism for the facilitated transport of vanadium ions (VO2+) by the studied SLM and to determine microscopic parametersD*andKass, for the three studied acidity. The obtained straight lines are represented by the graph ofFig. 5.
The results show that the proposed mechanism is verified, the carrier–substrate complex (TS) formed in the organic phase of the SLM is the composition (1/1) and the migration of this complex through the organic phase is the kinetically determining step for the mechanism on the transport of vanadium ions (VO2+). The slopes (p) and the intercepts (OO) were determined from straight lines inFig. 5 and using the expressions given by Eq. (15), the apparent diffusion coefficients D* and the association constantsKass were calculated.
However, these results summarized inTable 3show that these two parameters (D*andKass) vary slightly with the acidity of the medium.
The formation of complex and its diffusion through the organic phase of the SLM are little affected by the acidity of the source and receiving phases.
According to these results, wefind that the values of apparent diffusion coefficients D* obtained (about 10−4cm2s−1) for the transport of the (VO2+) ions, are higher compared to literature values (about 10−6cm2s−1) [28,29]. The same remark was made for the transport of some alditols and sugars [20,21]. To explain this phenomenon, we suggested in the transport of alditols[21], that the diffusion process is perhaps accompanied by a convection movement, which would be responsible for accelerating the diffusion of complexes in the membrane. While the hypothesis of “fixed-site jumping”proposed by Smith et al. in the transport of sugars through a plasticized membrane of triacetate cellulose [26,30] is another suggestion, more likely to clarify this point. This theory assumes that the substrate moves by jumping from site to site inside the membrane. In this case, the transport mechanism has the same kinetic profile as the facilitated transport; nevertheless, its characteristic is to be faster than the diffusion[26].
Several studies show that the nature and structure of the carrier are two important and essential factors for the facilitated transport of metal ions or organic molecules through the SLMs[21,22,31,32]. To complement our results and investigate the influence of the nature of the carrier and the stripping agent in the feed phase, we performed the same experiments on the transport of vanadium ions (VO2+), in the same conditions, with a new SLM containing the trioctylpho- sphine oxide (TOPO) as carrier dissolved in xylene and with two stripping agents, respectively HNO3and H2SO4in the source phase.
For this new SLM, the kinetic model proposed for the evolution of the concentrationCrof vanadium ions (VO2+
), carried in the receiving phase, as a function of time was verified (Fig. 4). All determined Table 2
Influence of the medium acidity on the parameters of the transport.
[HNO3] M C0= [VO2+] M P× 107(cm2s−1) J0× 106(mmol cm−2s−1)
1.00 0.100 16.77 16.77
0.750 17.50 13.12
0.025 19.68 4.92
1.50 0.150 17.50 26.25
0.075 18.95 14.21
0.018 20.41 3.82
2.00 0.200 16,77 33.54
0.100 19.69 19.68
0.025 21.15 5.28
[D2EHPA] = 0.01 M, ±5% is the confidence interval for values, PermeabilityP is calculated from Eq.(6).
InitialfluxJ0is calculated from Eq.(7). For thePVDFfilm, the areaSis 20.0 cm2, the thicknesslis 0.01 cm, pore size 0.45 µm, xylene phase andT= 298 K.
Fig. 5.Plots of 1/J0vs. 1/C0for the transport of VO2+
ions across the SLM.
permeabilities and initial fluxes for the facilitated transport of substrate through this new SLM with the stripping agent HNO3, are summarized inTable 4.
In the same way as for the previous SLM, the study of the Lineweaver–Burk plots (1/J0=f(1 /C0)) provided by Eq. (14), allows to verify the proposed mechanism for the facilitated transport of substrate by this new SLM and to determine microscopic parameters D*andKass, for each studied agent, HNO3and H2SO4. The obtained straight lines are represented by the graph of the followingfigure:
From straight lines inFig. 6, the intercepts (OO) and the slopes (p) were determined and using the expressions given by Eq. (15), the apparent diffusion coefficients D*and the association constantsKass were calculated. For the three studied acidity, all results are summarized inTable 5. These results show that the constantKassincreases with the acidity of the medium, while the apparent coefficientD*decreases.
The values of these two parameters indicate that for this new SLM, the stability of the complex (TS) formed in the organic phase is lower,
while its diffusion through this SLM is more important. Moreover, the presence of the stripping agent HNO3 in the feed phase is more effective for facilitated transport of VO2+ions from the concentrated acid solutions, through supported liquid membranes, and the extrac- tion of this substrate (VO2+) from these solutions is more important in the presence of the stripping agent HNO3, as its counterpart H2SO4, (D*HNO3/D*H2SO4= 3 to 4). The obtained results for both prepared SLMs clearly show that facilitated transport of vanadium ions (VO2+) in this type of membranes is very effective and certainly more ap- plications can be envisaged. As the previous membrane, the high values of apparent diffusion coefficientsD*indicate that the migration of the complex (TS) through the organic phase is not a pure diffusion and the hypothesis of“fixed-site jumping”proposed by Smith et al.[26,30], is likely a suggestion to explain this phenomenon of high diffusion of some compounds, observed for the facilitated transport through these SLMs.
According to this hypothesis, the substrate moves by jumping from site to site inside the membrane and in this case the transport mechanism is characterized by high values of apparent diffusion coefficients.
5.1. Relations between the microscopic parameters Kassand D*
The comparison of the values of these microscopic factors (Kass
andD*), shows that the less stable complexes migrate rapidly through the membrane. As a result, the association constants vary inversely in relation to the apparent diffusion coefficients (Fig. 7).
This result can be explained by the hypothesis of “fixed-site jumping”proposed by Smith [26], or the transport of substrate is by skipping one carrier to another. Indeed, when the stability of complexes is low, their dissociation constants are large; therefore, the passage of substrate from one carrier to another is easy. The apparent diffusion coefficients in this case are larger than those observed in the case of a simple diffusion. The study of all results shows that the migration of complexes (Carrier–Substrate) through the organic phase of the membrane depends on the stability of these complexes, and low stability (smallKass), results in a large diffusion (highD*) through the membrane and a high permeability of the used SLM for transported vanadium ions (VO2+). However, we can propose three types of move- ments associated with the migration of these complexes through the organic phases of the SLMs.
1) Movement of pure diffusion (D*order 10−6) for a very stable complexes in organic phase.
Table 3
Evolution of the parametersD*andKassdepending on the acidity.
[HNO3] M Kass(L mol−1) D*. 105(cm2s−1)
1.00 1.97 10.61
1.50 2.14 9.70
2.00 2.33 9.30
[D2EHPA] = 0.01 M;Kass± 0.01 and (D*± 0.01) * 10−5;Kassis the association constant for the heterogeneous equilibrium:Torg+Saq⇔TSorg.
Table 4
Evolution of the transport parameters depending on the acidity.
[HNO3] M C0= [VO2+
] M P× 107(cm2s−1) J0× 106(mmol cm−2s−1)
1.00 0.100 23.33 23.33
0.025 24.06 6.02
0.0125 26.25 3.28
1.50 0.075 18.96 14.22
0.0375 22.61 8.48
0.01875 23.33 4.38
2.00 0.200 16.12 32.24
0.100 18.34 18.34
0.050 19.72 9.86
0.025 20.46 5.12
[TOPO] = 0.01 M, ±5% is the confidence interval for values, similar experimental conditions toTable 2.
Fig. 6.Plots of 1/J0vs. 1/C0for the transport of VO2+
ions across the new SLM, for each stripping agent HNO3and H2SO4.
2) Composed movement (diffusion + jump from site to site) for unstable complexes in organic phase of the SLM (D*order 10−5) 3) Movement characterized by the displacement of the substrate by
jumping from site to site (with highD*, at around 10−4).
This hypothesis of“fixed-site jumping”advanced by Smith for the transport of sugars[22], explains the rapid migration through the organic phase of the SLM for these vanadium ions (VO2+) complexes with low stability.
6. Conclusions
Two supported liquid membranes (SLMs) were prepared with the same polymer support (PVDF), the same organic phase (xylene) and two different carriers (D2EHPA and TOPO). These two SLMs have been used for the facilitated transport of vanadium ions (VO2+) from the concentrated acid mediums. On the one hand, these experimental results confirm the recent work on the transport of Dy3+and Tm3+
ions[33,34], which clearly indicate that the permeabilityP of the membrane increases when the concentration of each substrate Dy3+
or Tm3+decreases, the acidity factor is important for the source phase and the transport parameters depend on the nature of the used stripping agent. Moreover, these results verify the proposed kinetic model, allowing to determine the permeabilityPand initialfluxJ0for both adopted SLMs. The obtained results show that the used SLMs are very permeable for VO2+ions, which resulted in significantfluxes of transported substrate. To understand this facilitated transport phenomenon through this type of liquid membranes, we developed a mechanism based on the formation of a complex carrier–substrate at the interface of the membrane and the migration of this complex through the organic phase of the SLM, to explain the migration of VO2+ions from the source phase to receiving phase. This mechanism has been verified, thus determining the microscopic parametersKass
and D*, for the reaction of complex formation and its diffusion through the organic phase of the SLM. Low values of the constants Kassand high values of the coefficientsD*, can fully explain the large permeabilities andfluxes, obtained for facilitated transport of VO2+
ions in this type of SLMs. Indeed, the microscopic parametersKass
andD*vary inversely and low values of association constantsKass, correspond to high values of apparent coefficientsD*. This important result shows that these high values of the apparent coefficients do not reflect a pure movement of diffusion of the complex (TS) through the organic phase of the SLM. Recent studies[31,32]on the facilitated transport of metal ions by the same type of membranes, confirm the high values of these apparent diffusion coefficientsD*. They explain this result by the nature of the movement of the substrateSin the organic phase of the SLM during its migration from the source phase to receiving phase.
Various studies [35–38], show that some supported liquid membranes containing specific carriers, are also very effective for the facilitated transport of some organic compounds (sugars, organic acids), and can be fully operational for the separation of mixtures of these compounds. All these results clearly indicate that the SLMs may be potential tools for several applications, particularly the specific
extraction of compounds from mixtures, or the general extraction of toxic heavy metals from industrial waste.
Acknowledgement
All authors wish to thank Professors Jean-François Verchère from the University of Rouen (France) for his advice, fruitful discussions, strong encouragement and exemplary cooperation.
References
[1] Luis J. Lozano, Carlos Godý´nez, F.J. Alguacil, Facilitated transport of vanadium (V) by supported liquid membranes, Hydrometallurgy 80 (2005) 196–202.
[2] F.J. Alguacil, M. Alonso, Transport of Cd(II) from a mixture of HCl and H3PO4, using phosphine oxides (Cyanex 921 and Cyanex 923) as carriers, Hydrometallurgy 74 (2004) 195–202.
[3] R.K. Biswas, M.A. Hayat, Solvent extraction of Zirconium (IV) from chloride medium, by D2EHPA in kerosene, Hydrometallurgy 63 (2002) 149–158.
[4] R.K. Biswas, M.A. Hayat, Kinetics of solvent extraction of Zirconium (IV) from chloride medium by D2EHPA in kerosene using the single drop technique, Hydrometallurgy 65 (2002) 205–216.
[5] J.-H. Chen, Y.-Y. Kao, C.-H. Lin, Selective separation of vanadium from Molybdenum using D2EHPA-Immobilized Amberlite XAD-4 resin, Sep. Sci.
Technol. 38 (2003) 3827.
[6] G. Jia, G. Torri, M. Ptruzzi, Distribution coefficients of polonium between 5% TOPO in toluene and aqueous hydrochloric and nitric acids, Appl. Radiat. Isotopes 61 (2004) 279–282.
[7] L. Boyadzhiev, G. Kyuchukov, Short communication: further development of carrier-mediated extraction, J. Membr. Sci. 6 (1980) 107–202.
[8] W.F. van Straate-Nijenhuis, F. de Jong, D.N. Reinhoudt, Macrocyclic carriers in supported liquid membranes, Recl. Trav. Chim. Pays-Bas 112 (1993) 317–324.
[9] H.C. Visser, D.N. Reinhoudt, F. de Jong, Carrier-mediated transport through liquid membranes, Chem. Soc. Rev. 23 (1994) 75–81.
[10] M. Teresa, A. Reis, Jorge M.R. Carvalho, Modelling of zinc extraction from sulphate solutions with bis(2-ethylhexyl)thiophosphoric acid by emulsion liquid mem- branes, J. Membr. Sci. 237 (2004) 97–107.
[11] R.M. Izatt, J.D. Lamb, R.L. Brueining, Comparison of bulk, emulsion, thin sheet supported, and Hollofiber SLM, in macrocycle-mediated cation separations, Sep.
Sci. Technol. 23 (1988) 1645–1658.
[12] A.J.B. Kempermen, D. Bageman, T.V.D. Boomgard, H. Stratmen, Stability of supported liquid membranes: state of the art, Sep. Sci. Technol. 31 (1996) 2733–2762.
[13] M. Mulder, Basic principles in membrane technology, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1991.
[14] J.A. Jönsson, L. Mathiasson Liquid, membrane extraction in analytical sample preparation, Trends Anal. Chem. 18 (1999) 318–325.
[15] H. Singh, S.L. Mishra, R. Vijayalakshmi, Uranium recovery from phosphoric acid by solvent extraction using a synergistic mixture of di-nonyl phenyl phosphoric acid and tri-n-butyl phosphate, Hydrometallurgy 73 (2004) 63–70.
[16] T. Yamaguchi, K. Nishimura, T. Shinbo, M. Sugiura, Amino acid transport through supported liquid membrane: mechanism and its application to enantiomeric resolution, Bioelectrochem. Bioenerg. 20 (1988) 109–123.
[17] P. Dzygiel, P. Wieczorek, J.A. Jönsson, L. Mathiasson, Enrichment of amino acids by supported liquid membrane extraction using aliquat 336 as a carrier, Anal. Lett. 31 (1998) 1261–1274.
[18] M. Rak, P. Dzygiel, P. Wieczorek, Supported liquid membrane extraction of aromatic aminophosphonates, Anal. Chim. Acta 433 (2001) 227–236.
[19] M. Di Luccio, B.D. Smith, T. Kida, T.L.M. Alves, C.P. Borges, Evaluation offlat sheet and hollowfiber supported liquid membranes for fructose pertraction from a mixtures of sugars, Desalination 148 (2002) 213–320.
Table 5
Microscopic parametersKassandD*according to the acidity with the stripping agents:
HNO3and H2SO4.
(a) [NHO3] M (b) [H2SO4] M Kass(L mol−1) D*. 105(cm2s−1)
1.00 1.00 1.56 (a)/7.71 (b) 16.90 (a)/5.92 (b)
1.50 1.50 1.57 (a)/8.75 (b) 15.10 (a)/4.86 (b)
2.00 2.00 1.60 (a)/9.99 (b) 13.33 (a)/3.28 (b)
[TOPO] = 0.01 M;Kass± 0.01; (D*± 0.01) * 10−5;Kassis the association constant for the heterogeneous equilibrium:Torg+Saq⇔TSorg.
Stripping agent in the feed phase: (a) NHO3, (b) H2SO4.
Fig. 7.ParametersD*andKasson theTOPO-VO2+complex, depending on the acidity.
[20] N. Tbeur, T. Rhlalou, M. Hlaïbi, D. Langevin, M. Métayer, J.F. Verchère, Selective transport of alditols through a supported liquid membrane, Carbohydr. Res. 329 (2000) 409–422.
[21] T. Rhlalou, M. Ferhat, M.A. Frouji, D. Langevin, M. Métayer, J.F. Verchère, Facilitated transport of sugars by a resorcinarene through a supported liquid membrane, J. Membr. Sci. 168 (2000) 63–73.
[22] P. Dzygiel, P. Wieczorek, Extraction of glyphosate by a supported liquid membrane technique, J. Chromatogr. A 889 (2000) 93–98.
[23] M. Mulugeta, N. Megersa, Carrier-mediated extraction of bipyridilium herbicides across the hydrophobic liquid membrane, Talanta 64 (2004) 101–108.
[24] R.-S. Juang, L.-J. Chen, Transport of citric acid across a SLM containing various salts of a tertiary amine, J. Membr. Sci. 123 (1997) 81–87.
[25] E. Miyako, T. Maruyama, N. Kamiya, M. Goto, Transport of organic acids through a supported liquid membrane driven by lipase-catalyzed reactions, J. Biosci. Bioeng.
96 (2003) 370–374.
[26] J.A. Riggs, B.D. Smith, Facilitated transport of small carbohydrates through plasticized cellulose triacetate membranes. evidence for fixed-site jumping transport mechanism, J. Am. Chem. Soc. 119 (1997) 2765–2766.
[27] K.M. White, B.D. Smith, P.J. Duggan, S.L. Sheahan, E.M. Tyndall, Mechanism of facilitated saccharide transport through plasticized cellulose triacetate mem- branes, J. Membr. Sci. 194 (2001) 165–175.
[28] S. Ouahid, M. Métayer, D. Langevin, M. Labbé, Sorption of glucose by an anion- exchange membrane in the borate form. Stability of the complexes and the facilitated transport of glucose, J. Membr. Sci. 114 (1996) 13–25.
[29] M.F. Paugam, G.T. Morin, B.D. Smith, Metal cation: glucopyranoside co-transport through a liquid organic membrane, Tetrahedron Lett. 34 (1993) 7841–7844.
[30] B.D. Smith, Liquid membrane transport using boronic acid carriers, Supramol.
Chem. 7 (1996) 55–60.
[31] F.J. Alguacil, M. Alonso, Transport of Au(CN)2- across SLM using mixtures of amine and phosphine oxide, Hydrometallurgy 74 (2004) 157–163.
[32] F.J. Alguacil, H. Tayibi, Carrier-facilitated transport of Cd(II) from a high-salinity chloride medium across a supported liquid membrane containing Cyanex 923 in Solvesso 100, Hydrometallurgy 180 (2005) 181–187.
[33] L. Pei, B. Yao, X. Fu, Study on transport of Dy(III) by dispersion supported liquid membrane, J. Rare Earths 27 (3) (2009) 447.
[34] L. Pei, B. Yao, C. Zhang, Transport of Tm(III) through dispersion supported liquid membrane containing PC-88A in kerosene as carrier, Sep. Sci. Technol. 65 (2009) 220–227.
[35] R.S. Juang, R.H. Huang, R.T. Wu, Separation of citric and lactic acids in aqueous solutions by solvent extraction and liquid membrane processes, J. Membr. Sci. 136 (1997) 89–99.
[36] G.O. Yahaya, B.J. Brisdon, R. England, Facilitated transport of lactic acid and its ethyl ester by SLMs containing functionalized polyorganosiloxanes as carriers, J.
Membr. Sci. 168 (2000) 187–201.
[37] H. Hassoune, T. Rhlalou, M. Métayer, J.-F. Verchère, Facilitated transport of aldoses by methyl cholate through SLMs impregnated with various solvents, J. Membr. Sci.
248 (2005) 89–98.
[38] H. Hassoune, T. Rhlalou, M.A. Frouji, C. Chappey, J.-F. Verchère, Application of SLMs containing methyl cholate in cyclohexane for the carrier-mediated transport of sugars, Desalination 189 (2006) 31–42.
