Talanta 78 (2009) 171–177
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Pore-functionalized polymer membranes for preconcentration of heavy metal ions T. Vasudevan a , Sadananda Das b , Suparna Sodaye b , A.K. Pandey b,∗ , A.V.R. Reddy b a b
Research Reactor Services Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India Radiochemistry Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India
a r t i c l e
i n f o
Article history: Received 21 July 2008 Received in revised form 30 October 2008 Accepted 30 October 2008 Available online 7 November 2008 Keywords: Functionalized membrane Acidic monomer Ethylene glycol methacrylate phosphate UV-grafting Metal ions Radiotracers Selectivity
a b s t r a c t Functionalized membranes containing carboxylate, phosphate and sulfonate groups were prepared by UV-initiator induced graft polymerization of the functional monomer (acrylic acid, ethylene glycol methacrylate phosphate (EGMP) and 2-acrylamido-2-methyl-1-propane sulfonic acid) with a crosslinker (methylenebisacrylamide) in the pores of poly(propylene) host membranes. The functionalized membranes thus obtained were characterized by gravimetry, FTIR spectroscopy, radiotracers and scanning electron microscopy for the degree of grafting and water uptake, presence of functional groups, ionexchange capacity, and physical structure of the membranes, respectively. The uptakes of Cs+ , Ag+ , Sr2+ , Cd2+ , Hg2+ , Zn2+ , Eu3+ , Am3+ , Hf4+ and Pu4+ ions in the functionalized membranes were studied as a function of acidity of the equilibrating aqueous solution. Among the functionalized membranes prepared in the present work, the EGMP-grafted membrane (with phosphate groups) showed acid concentration dependent selectivity towards multivalent metal ions like Eu3+ , Am3+ , Hf4+ and Pu4+ . The solvent extraction studies of EGMP monomer in methyl isobutyl ketone (MIBK) solvent indicated that divalent and trivalent metal ions form complexes with EGMP in 1:2 proportion, but the distribution coefﬁcients of trivalent metal ions were signiﬁcantly higher that for the divalent ions. The uptakes of Eu3+ ions in monomeric EGMP (dissolved in MIBK) and polymeric EGMP (in the forms of crosslinked gel and membrane) were studied as a function of concentration of H+ ions in the equilibrating solution. This study indicated that polymeric EGMP has better binding ability towards Eu3+ as compared to monomeric EGMP. The variation of distribution coefﬁcients of Eu3+ /Am3+ in gel and membrane as a function of H+ ion concentration in the equilibrating aqueous solution indicated that ionic species held in the membrane and gel were not same. These results indicated that proximity of functional groups (phosphate) plays an important role in metal ion binding with polymeric EGMP. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Membranes with functional groups have applications in the chemical sensors, fuel cell, separation science, electrodialysis, and as a host for nonoparticles [1–5]. Most of the commercially available membranes are ion-exchange membranes. Therefore, these membranes cannot be used for the applications that require selective recognition of target ions. The target ion speciﬁc membrane is a key issue not only in development of the chemical sensors but also in developing separation schemes for various applications, e.g. the radioactive waste reprocessing. At present, target speciﬁc membranes are being developed by physically immobilizing speciﬁc ligands in a polymer matrix. These membranes are known as supported liquid membranes and polymer inclusion
∗ Corresponding author. Tel.: +91 22 25590641; fax: +91 22 25505150/25505151. E-mail address: [email protected]
(A.K. Pandey). 0039-9140/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.10.053
membranes. The supported liquid membranes (SLMs) are formed by the physical immobilization of the organic phase containing a mobile carrier (extractant) into the pores of a microporous host membrane. As the organic phase is held in the pores of the host membrane by capillary force, the long-term stability of the SLM is a major concern for its wide scale industrial applications . In order to overcome the stability problems associated with the SLM, a new class of membranes called polymer inclusion membranes (PIMs) have been developed . The PIMs are prepared by physical immobilization of a selective extractant into a plasticized polymer matrix. The PIMs have been reported to have better stability than SLMs. These membranes ﬁnd extensive applications to prepare the electrochemical and optical chemical sensors . However, the transport of ions across PIM has been found to be slower than SLM and ﬁxed-site carrier membrane. This is due to fact that viscosity of the liquid fraction and hydrophobicity of PIM are considerably higher than those for other two classes of membrane .
T. Vasudevan et al. / Talanta 78 (2009) 171–177
The membranes can be made quite stable by linking functional groups covalently with the polymer chains in the membrane. Such membranes can be synthesized by grafting a monomer on polymer chains and subsequently generating required functional group by chemical modiﬁcation of the precursor chemical groups on grafted polymer chains. This class of membranes is termed as ﬁxed-site membranes, functionalized membranes, or adsorptive membranes. The membranes used for the metal ions separation have been prepared by anchoring monomers 4-vinyl benzyl chloride (converted to phosphonate ester or triethyl ammonium) , 4-vinylpyridine [10–11], acrylonitrile (converted to amidoxime) [12–14], acrylic/methacrylic acid  and monomer containing benzo-18-crown-6 crown ether functional groups  in the host microporous membrane. Ethylene glycol methacrylate phosphate (EGMP) and 2-acrylamido-2-methyl-1-propane sulfonic acid (AMPS) grafted membranes have not been studied for metal ions separations. However, these monomers have been used for preparing the proton conducting composite materials [17,18]. The objective of the present study is to develop a functionalized ﬂat sheet membrane that can be used as a sorbent for the preconcentration of heavy metal ions. In the present work, the functionalized membranes have been prepared by anchoring carboxylic, phosphate and sulfonate groups in the pores of a microporous host substrate by UV photoinitiator induced in situ polymerization of the monomer (acrylic acid (AA)/EGMP/2acrylamido-2-methyl-1-propane sulfonic acid (AMPS)) along with a crosslinker methylenebisacrylamide. The membranes are characterized in terms of degree of grafting, ion-exchange capacity, water uptake capacity, and physical and chemical structures. Studies on the uptake of Cs+ , Ag+ , Sr2+ , Cd2+ , Hg2+ , Zn2+ , Eu3+ , Am3+ , Hf4+ and Pu4+ in the membranes were carried out as a function of acidity of the equilibrating solution. The extraction of multivalent ions in organic phase consisting of monomer EGMP dissolved in methylisobutyl ketone solvent (MIBK), EGMP gel, and EGMP membrane have also been studied to understand the mechanism of sorption of ions in the EGMP sorbents.
were dissolved in tetrahydrofuran (THF) and dimethylformamide (DMF) mixture. AMPS was ﬁrst dissolved in water–methanol (1:1) mixed solvent, and then added to the grafting solution. The amounts of crosslinker and monomer were adjusted in the polymerization solution to get 4–5 mol% of crosslinking. The amount of UV-initiator was taken as 1 wt.% as this was found to be minimum quantity required to initiate the graft-polymerization. For grafting, the microporous host poly(propylene) membrane (6 × 6 cm2 area, 60% porosity, 0.2 m pore diameter and 59 m thickness) was soaked in the polymerizing solution, and sandwiched the grafting solution ﬁlled membrane between two transparent polyester sheets to prevent any possible loss of grafting solution ﬁlled in the pores. Care was taken to remove excess grafting solution and air bubbles trapped between the membrane and polyester sheets covering the membrane surface. Finally, the sandwiched membrane was exposed to 365 nm UV light in a multilamp photoreactor (Heber Scientiﬁc, model No. HML-SW-MW-LW-888) for a period of 20 min. After irradiation in the photoreactor, the membrane was washed thoroughly with THF, methanol and distilled water to remove the un-grafted components, and then conditioned with a aqueous solution having 0.25 mol L−1 NaCl. The membrane sample was dried for 8–10 h under vacuum and weighed to obtain the weight of grafted membrane. The crosslinked EGMP gels were prepared by using same polymerizing solution in a 15 mL beaker, which was exposed to 365 nm UV light for 20 min in the photoreactor. After gelation, the corsslinked EGMP gels were washed to remove un-polymerized components and conditioned with 0.25 mol L−1 NaCl.
where Wwet is the weight of the wet membrane sample and Wdry is the weight of the same membrane sample dried under vacuum. The membrane samples were powdered in liquid nitrogen and mixed with KBr for FTIR measurements. FTIR spectra of the samples were recorded using the spectrophotometer (8400model) procured from Shimadzu, Japan. The KBr powder was used as an internal standard in the measurements. The ratio of the KBr to membrane samples was maintained to be 100:1. Diffused reﬂectance assembly was used for recording the spectra. The ion exchange capacities of the functionalized membranes were measured by acid-base titration. The weighed membrane sample (2 × 2 cm2 ) in H+ form was equilibrated with 10 mL of 0.5 mol L−1 NaCl for 5–6 h with a constant stirring at 25–27 ◦ C. After equilibration, the membrane sample was taken out from the solution and droplets adhering on the surface of membrane were washed with de-ionized water. The amount of H+ ions (moles) liberated in the solution (ﬁnal volume of 20 mL) was measured by potentiometric titration with standard NaOH solution. The ion-exchange capacity of the membrane was computed with the experimentally measured moles of H+ ions librated in the equilibrating solution and weight of the dried membrane sample.
2.1. Reagents and apparatus Analytical reagent grade chemicals and de-ionized water (18 M/cm) puriﬁed by model QuantumTM from Millipore (Mumbai, India) were used throughout the present studies. EGMP, AMPS, AA, N-N -methylenebisacrylamide (MBA), and ␣,␣ -dimethoxy-␣phenyl acetophenone (DMPA) were obtained from Sigma–Aldrich (Steinheim, Switzerland). Tetrahydrofuran (THF), MIBK, and N-N dimethylformamide (DMF) were obtained from Merck (Mumbai, India). Radiotracers 22 Na, 110m Ag, 137 Cs, 85,89 Sr, 115m Cd, 203 Hg, 65 Zn, 154 Eu, 241 Am, 239 Pu and 181 Hf dissolved in aqueous solutions were obtained either from the Board of Radiation and Isotope Technology, Mumbai, India or produced by reactor neutron irradiation of appropriate target samples in our laboratory. The ␥-activities of the radiotracers in the membrane and equilibrating solution were monitored by using either a high purity germanium (HPGe) detector or a well-type NaI(Tl) detector based gamma ray spectrometer. All the samples and standard containing radioactivity were counted in identical sample-detector geometry. The thickness of the membrane samples was measured using a digital micrometer (Mitutoyo, Japan) with a precision of ±0.001 mm. 2.2. Preparation of functionalized membranes and gel The monomer (AA, EGMP, or AMPS) along with a crosslinker, MBA and an UV initiator (␣,␣-dimethoxy-␣-phenyl acetophenone)
2.3. Characterization of membranes The degree of grafting in the membrane sample was calculated using weights of membrane sample before and after grafting. The water uptake capacity of the functionalized membranes was also determined by gravimetry using the following equation: water uptake capacity(%) =
(Wwet − Wdry ) (Wdry )
2.4. Uptake of metal ions For uptake studies, 2 × 2 cm2 pieces of membranes were equilibrated with 5 mL solutions containing desirable metal ions in different acidities for 24 h without stirring or 3 h with stirring at room temperature 25 ◦ C. Buffer was not used in the uptake
T. Vasudevan et al. / Talanta 78 (2009) 171–177
experiments. The equilibration time was based on the experiments that indicated the minimum equilibration time for quantitative uptake of metal ions in the membrane samples. The radiotracers of 22 Na, 110m Ag, 137 Cs, 85,89 Sr, 115m Cd, 203 Hg, 65 Zn, 154 Eu, 241 Am, 239 Pu and 181 Hf were used for the uptake studies. The concentrations of these radiotracers in the equilibrating solution were sub-ppm level depending upon the radioactivity of the radiotracer. The uptake of metal ions (except for 239 Pu) in the membrane was monitored by ␥-activity of the solutions (5 mL) before and after equilibration of membrane samples, using a NaI(Tl) scintillation detector. The ␥-activity of radiotracer used in these experiments was 40,000–50,000 counts/min in 5 mL solution. The uptake of 239 Pu in the membrane was obtained by ␣-scintillation counting of the samples taken before and after equilibration from the feed solution. The ␣-activity of the equilibrating solution (5 mL) was adjusted to obtain 10,000 counts/min in ␣-scintillation counting of its 50 L sample. A 50 L sample of the equilibrating solution was added to 5 mL of standard organic liquid scintillation cocktail and assayed by scintillation counting. The uptake of radiotracers in the membrane was obtained from Eq. (2). ions uptake(%) =
(Cbefore − Cafter ) × 100 (Cbefore )
where Cbefore and Cafter are the ␣-scintillations/␥-activity (counts/s) of the radiotracer in the samples taken from feed solution before and after equilibrating with the membrane sample, respectively. 2.5. Extraction study Monomer EGMP was dissolved in solvent isobutylmethyl ketone (MIBK) to prepare organic phase having EGMP concentration in the range of 0.50–3.75 mol L−1 . The equal volumes (5 mL) of organic phase and aqueous phase containing a radiotracer (65 Zn/154 Eu/241 Am) were mixed in an equilibration tube, and shaken for 2 h in a constant temperature bath kept at 25 ◦ C. The acidity and ionic strength of the aqueous phase were adjusted with HNO3 and NaNO3 , respectively. After equilibration, organic and aqueous phases were separated by ultracentrifugation. Finally, aliquots of 0.5 mL from each phase were taken for the radioactivity counting. The distribution coefﬁcient (Kd ) as a function of EGMP monomer concentration in MIBK was obtained from the ratio of measured radioactivity in organic to aqueous phase after equilibration. The Kd values of EGMP membrane and gel were also measured by equilibrating a known weight of solid sample (0.05 g) in 5 mL of aqueous phase for overnight with constant stirring at room temperature (25–27 ◦ C). The aqueous phase was prepared by adding a known radioactivity of radiotracer (154 Eu/241 Am) in aqueous solution whose ionic strength and acidity were adjusted with NaNO3 and HNO3 , respectively. After equilibration, the 0.5 mL sample was taken from the aqueous phase for monitoring the radioactivity. The Kd values of Eu3+ and Am3+ ions between gel/membrane and aque-
ous phase were calculated from following equation: Kd =
(Co − C)V CW
where Co , C, V and W denote initial and equilibrium radioactivity of the radiotracer in aqueous phase, volume of the equilibrating aqueous phase, and weight of dried gel/membrane, respectively. 3. Results and discussion The functionalized membranes were prepared by in situ graft polymerization of monomer (acrylic acid/2-acrylamido2-methyl-1-propane sulfonic acid/ethylene glycol methacrylate phosphate) with a 5 mol% crosslinker into the pores of microporous poly(propylene) host membrane. Un-grafted monomers from the pores of the membrane were removed completely by repeated washing with DMF and methanol. These cycles were continued until weight of the membrane sample remained constant. The grafting of monomer along with crosslinker in poly(propylene) microporous membranes samples was conﬁrmed gravimetrically after removal of the un-grafted materials. It was observed that the amount of grafting in the host poly(propylene) (PP) membrane could be controlled by adjusting the quantity of monomer and relative amounts of other components in the polymerization solution ﬁlled in the pores of the host membrane. As given in the Table 1, the functionalized membranes with a varying degree of grafting could be obtained in the present work. The grafting of monomers up to 180 wt.% did not affect thickness of the host poly(propylene) membrane (59 m) signiﬁcantly as thickness of the membranes samples after grafting was found to be within ±2 m of the nascent host membrane. Higher degree of grafting was found to increase the thickness of a membrane. For example, the thickness of the membrane was increased from 59 to 70 m after grafting to the extent of 290 wt.%. This indicated that higher degree of grafting leads to formation of the surface layer on host poly(propylene) membrane. The surface morphology of the membranes was studied by the scanning electron microscope, and images thus obtained are shown in Fig. 1. It can be seen from the comparison of the SEM micrographs of the host and grafted membranes that pores of the membrane are completely ﬁlled with the grafted polymer irrespective of their chemical composition. The comparison of FM-3 (290 wt.% grafted EGMP) and FM-6 (120 wt.% grafted AMPS) shows that higher degree of grafting leads to the ﬂat surface without any microstructure. These indicate that higher degree of grafting leads to surface deposition, and the grafted polymer is conﬁned to the pores only when grafting is below 180 wt.%. 3.1. Characterization of membranes All the functionalized membranes synthesized in the present work were found to take up water readily. These membranes were found to be dimensionally stable, and did not change their
Table 1 Degree of monomer grafting in poly(propylene) microporous membranes and the water uptake capacities of the functionalized membranes. The membranes with higher degree of grafting were obtained by increasing the concentration of monomer in the grafting solutions. Id.
FM-1 FM-2 FM-3 FM-4 FM-5 FM-6
Composition of grafting solution Monomer
EGMP EGMP EGMP AA AA AMPS
4 5 4 4 5 5
Degree of grafting (wt.%)
120 170 290 80 174 120
± ± ± ± ± ±
10 12 10 6 10 10
Water uptake (wt.%)
Ion-exchange capacity (meq/g)
– 52 ± 5 – – 84 ± 5 138 ± 8
– 1.0 ± 0.1 – – 0.3 ± 0.1 1.8 ± 0.1
T. Vasudevan et al. / Talanta 78 (2009) 171–177
obtained in the membrane having –SO3 H group which is stronger than –COOH and –PO(OH)2 acid groups. The chemical structures of functionalized membranes were studied by FTIR spectroscopy. The FTIR spectrum of un-grafted PP membrane shows the absorption bands at 2930, 1450 and 1370 cm−1 corresponding to –CH2 and –CH3 vibrations, respectively. The FTIR spectrum of AMPS-grafted poly(propylene) membrane showed the additional vibration bands of sulfonyl group at 1035 cm−1 (S–O), 1210 and 1370 cm−1 (S O), 1450 cm−1 (–SO2 –), and amide carbonyl group at 1645 cm−1 (C O). The FTIR spectra of EGMP monomer and EGMP-grafted membrane showed similar bands corresponding to free P O (1330 and 1350 cm−1 ), associated P O (1171 and 1170 cm−1 ), C O (1715 and 1720 cm−1 ), and P–OH vibrations at ≈1000, 2400–2200 cm−1 , and a broad band in the region of 3400–3000 cm−1 . The FTIR spectrum of EGMP-grafted membrane showed a shift in the free P O and C O vibration bands to a higher frequency with respect to EGMP monomer. Also, the C C stretching frequency observed at 1640 cm−1 in the FTIR spectrum of EGMP monomer has disappeared after grafting it in the poly(propylene) host membrane. The frequencies discussed were assigned based on the comparison of FTIR spectra with the characteristic frequencies reported in the literature for these groups . The comparison of the host and grafted membranes described above conﬁrmed the presence of the –SO3 H, –PO(OH)2 , and C O groups in the functionalized membranes. The possible chemical structures of the functionalized polymer anchored in the pores of membranes are shown in Scheme 1. As can be seen from the chemical structures, there would be considerable extent of the intra and inter-chain hydrogen bonding that are not shown in chemical structures of functionalized polymers given in Scheme 1. The ionexchange capacities of the AA, EGMP and AMPS membranes were found to be 0.3, 1.0 and 1.8 meq/g, respectively. This trend in the ionexchange capacity was expected based on the acid groups present in these membranes (see Scheme 1). 3.2. Uptake studies
Fig. 1. Scanning electron microscopic images of (a) poly(propylene) host membrane (b) grafted AMPS membrane (FM-6), and (c) grafted EGMP membrane (FM-3).
dimensions when swollen in the water. The two factors responsible for the dimension stability of the functionalized membrane are: (i) poly(propylene) acts as a mechanical containment and (ii) crosslinking of the grafted polymer would prevent the swelling of the polymer chains. The latter is responsible for preventing the swelling of the membranes with higher degree of grafting. As shown in Table 1, the water uptake capacity of the functionalized membrane was dependent on the nature of the functional groups anchored in the membrane. The highest water content was
In order to characterize the functionalized membranes in terms of their selectivity towards different valence metal ions, the membrane samples were equilibrated with the solutions containing a radiotracers of the ions like Cs+ , Ag+ , Sr2+ , Cd2+ , Hg2+ , Zn2+ , Eu3+ , Am3+ , Hf4+ and Pu4+ at different acidities. The comparisons of uptake of the monovalent ions (Cs+ ) and divalent ions (Cd2+ ) in the AA, AMPS and EGMP membranes from aqueous solutions having pH ranging from 1–6 are shown in Figs. 2 and 3, respectively. It can be seen from these ﬁgures that the uptake proﬁles of monovalent and divalent in AMPS and EGMP membranes are similar. The uptake of Cs+ and Cd2+ are quantitative (>80%) below pH 3 and 2, respectively, and thereafter falls steeply on decreasing the pH of the equilibrating solution. The uptakes of Cs+ and Cd2+ in AA membrane were considerably lower than AMPS and EGMP membranes, and quantitative only for Cd2+ ions above pH 5. This may be attributed to the fact that AA contains –COOH, which is a weak acid group. The uptake of trivalent ions (Eu3+ /Am3+ ) and tetravalent ions (Pu4+ ) in AMPS and EGMP membranes were studied as these membranes take up monovalent and divalent ions quantitatively at acidic pH range as described above. The uptake of Eu3+ and Am3+ ions in both these membranes is shown in Fig. 4. The uptake proﬁles of trivalent ions as a function of feed acidity was found to be similar in both the membranes. These membranes sorb trivalent ions quantitatively (>90%) from solution having acidity up to 0.2 mol L−1 , and sharply decrease on increasing the acidity above 0.2 mol L−1 . It is also seen from Fig. 4 that the AMPS and EGMP membranes are slightly more selective towards Eu3+ ions as compared to Am3+ ions. Fig. 5 shows the uptake proﬁles of Pu4+ ions in AMPS and EGMP
T. Vasudevan et al. / Talanta 78 (2009) 171–177
Scheme 1. Chemical structures of the functionalized polymer grafted in the pores of poly(propylene) host membranes.
Fig. 2. Cs+ uptake in AA (), AMPS (x), and EGMP () membranes as a function of pH of equilibrating solution.
brane by its electrostatic interactions with sulfonic acid groups. In EGMP-membrane, Pu4+ must be held by electrostatic as well as covalent interactions with phosphate groups, which gives rise to its quantitative uptake even at very high acidity. The sulfonic acid functionalized membrane (AMPS-membrane) is nonselective due to a narrow range of reaction free energy values for the exchange with various metal ions. The carboxylic acid membrane (AA-membrane) is less acidic, thus limiting its usefulness in the low pH environments. Of intermediate strength is phosphate functionalized membrane (EGMP-membrane). This is clearly reﬂected in their ion-exchange capacities (Table 1). The sorption of different valence ions like Ag+ , Sr2+ , Zn2+ , Hg2+ , Eu3+ , Am3+ and Hf4+ in EGMP-membrane as a function of log10 of HNO3 concentration in the equilibrating aqueous feed is shown in Fig. 6. It can be seen from this ﬁgure that the uptake proﬁles are dependent on oxidation states of the ions, and there is no appreciable difference in the uptake behavior of different ions having same valency. This gives rise to a possibility of selectivity uptake of higher valence ions in the EGMP at higher acidity. All the metal ions could be desorbed quantitatively from AMPSmembranes in 2 mol L−1 HNO3 . In the case of EGMP-membrane, the desorption of Pu(IV) and Hf(IV) was not found to be possible with
membranes. As can be seen from Fig. 5, the trends in the uptake of Pu4+ ions in EGMP and AMPS membranes as a function of acidity are quite different. Uptake of Pu4+ in AMPS membrane is negligible from solution having H+ ions concentration above 1 mol L−1 , whereas EGMP membrane takes up Pu4+ at acid concentration as high as 4 mol L−1 . This indicates that Pu4+ is held in AMPS mem-
Fig. 3. Cd2+ uptake in AA (), AMPS (x), and EGMP () membranes as a function of pH of equilibrating solution.
Fig. 4. Sorption of Eu3+ () and Am3+ (×) in AMPS and EGMP membranes as a function of feed acidity. Solid and broken lines represent AMPS and EGMP membranes, respectively.
T. Vasudevan et al. / Talanta 78 (2009) 171–177
Fig. 5. Comparison of Pu4+ sorption in EGMP and AMPS membranes as a function of acid concentration in the aqueous feed.
Fig. 7. Eu3+ -uptakes in solution of monomeric EGMP, EGMP-gel, and EGMPmembrane as a function of log10 of acid concentration in the aqueous feed.
3.4. Extraction studies the acid up to 4 mol L−1 . It was observed that 30–40% of Pu4+ could be desorbed by equilibration of EGMP-membrane with aqueous solution of 0.2 mol L−1 disodium salt of EDTA for overnight without stirring. 3.3. Comparison of uptake in monomer and polymeric EGMP The Eu3+ uptake in EGMP monomer (dissolved in MIBK), EGMPgel and EGMP-membrane were studied as a function of HNO3 concentration in the equilibrating feed solution, and shown in Fig. 7. It can be seen from this ﬁgure that uptake of Eu3+ in monomeric EGMP increases with increase in the concentration of EGMP, and get saturated at EGMP concentration of 3.5 mol L−1 in MIBK solvent. However, the Eu3+ uptake proﬁles in polymeric EGMP (gel and membrane) were considerably higher than that in monomeric EGMP solution. This seems to suggest that the binding of Eu3+ ions is much stronger in polymeric EGMP than its monomeric form. Similar observations were reported for other functionalized polymers [20,21]. It is also seen in Fig. 7 that the trends of Eu3+ uptake proﬁles of EGMP solution, gel and membrane are signiﬁcantly different.
Fig. 6. Sorption of different valence ions in EGMP membrane as a function of log10 of HNO3 concentration in the equilibrating aqueous feed.
In order to understand the formation of metal ion complexes in EGMP-membrane, the solvent extraction studies of Zn2+ , Eu3+ and Am3+ ions in organic phase containing EGMP-monomer in MIBK were carried out. The variation of distribution coefﬁcients (Kd ) of these ions as a function of EGMP concentration [LH2 ] in organic phase is shown in Fig. 8. It is seen from this ﬁgure that Kd values increase linearly as a function of EGMP concentration in all the cases. Kd values of Eu3+ and Am3+ were found to be similar, but signiﬁcantly higher the Kd values of Zn2+ at a given concentration of the ligand. Kd values for Eu3+ and Am3+ gets saturated at 2.5 mol L−1 concentration of EGMP. This may be due to the formation of hydrogen bonded micelle like species of EGMP monomer at higher concentration, which would inhibit the accessibility of some of the binding sites to the metal ions. Though the Kd values of Zn2+ were signiﬁcantly lower than that for Eu3+ and Am3+ , their constant
Fig. 8. Variation of logarithm of distribution coefﬁcients of Am3+ (), Eu3+ (), and Zn2+ () as a function of logarithm of the concentration of EGMP monomer (LH2 ) dissolved in the isobutylmethyl ketone.
T. Vasudevan et al. / Talanta 78 (2009) 171–177
cal containment that mitigates the impact of osmotic pressure. The swelling would lead to increase in the effective distance between ﬁxed-sites (phosphate moiety of the EGMP) of the sorbent. This seems to suggest that the proximity of the functional groups plays an important role in binding of the multivalent ions in solid sorbent. 4. Conclusions
Fig. 9. Slope analysis for the extraction of Am3+ (×) and Eu3+ () ions. Solid and broken lines represent EGMP-membrane and EGMP-gel, respectively.
slope value 2 in the plot of ln Kd vs. ln[LH2 ] (Fig. 8) indicates that two EGMP units are involved in binding with Zn2+ , Eu3+ , and Am3+ ions. The extraction of ions in EGMP-sorbent involves ion-exchange process as the uptake of ions decreases with increase in the concentration of H+ ions. This means Eu3+ and Am3+ ions are extracted in EGMP-sorbent by an ion-exchange reaction between these trivalent ions and hydroxyl groups of phosphate moiety in EGMP-sorbent. The stoichiometric equation involved in the uptake of Eu3+ /Am3+ in gel and membrane can be represented based on solvent extraction studies as follow: + M3+ aq + 2(−LH2 )S M(−L2 H4−n )S + nH
where M3+ and –LH2 denote Eu3+ /Am3+ ions and phosphate groups in the EGMP-sorbents, respectively. The subscripts aq and s denote aqueous and sorbent phases,respectively. Based on this equation, the extraction constant can be expressed as: ln Kex = ln
[M(L2 H4−n )s ] [M3+ ]
− 2 ln[(LH2 )] + n ln[H+ ]
The functionalized membranes with three different acidic groups viz. –SO3 H, –PO(OH)2 , and–COOH were prepared by UV initiated grafting of functional monomers in the pores of poly(propylene) host membrane. The presence of required functional groups in the membrane was conﬁrmed by FTIR spectroscopy. It was observed that the grafting yield could be controlled by adjusting the concentration of the monomer in grafting solution. The functionalized membranes developed in the present work were found to be hydrophilic. It was observed that the spacing of functional groups in the EGMP-sorbent inﬂuence binding of the species of the multivalent ions with functional groups. The uptake of different valence metal ions in the membranes as a function of the feed acidity indicated that the EGMP-grafted membrane (with phosphate groups) can be made selective towards higher valence metal ions by adjusting the feed acidity. The selectivity of the other functionalized membranes towards different valence ions was not as promising as in the case of EGMP-grafted membrane. Thus, EGMP-membrane reported in the present work can be used as the adsorptive membrane for selective preconcentration of the actinides and lanthanides. Acknowledgments Authors are thankful to Dr. V.K. Manchanda, Head, Radiochemistry Division, BARC, Mumbai for his keen interest in the present work, and Dr. P.K. Pujari for his valuable suggestions during the course of this work. Authors also thank J. K. Banerjee, Radiometallurgy Division, BARC, for SEM of the membranes. References
    
where [LH2 ] and [H+ ] denote the concentrations of unreacted –LH2
ln Kex = ln Kd + ln(1 + ˇ1 [NO3 ] + ˇ2 [NO3 ] ) −2 ln[(LH2 )] + n ln[H+ ]
in the sorbent and hydrogen ion in the aqueous solution, respectively. ˇ1 and ˇ2 denote the stability constants for the nitrato complexes M(NO3 )2+ and M(NO3 )+ , respectively. Since the ionic strength was kept constant, the second term in Eq. (6) can be considered to be almost constant. Therefore, Eq. (6) can be simpliﬁed as: ln Kex = ln Kd − 2 ln[(LH2 )] + n ln[H+ ]
     
The slope analysis was performed using experimentally measured Kd as a function of H+ ion concentration in the equilibrating solution, keeping constant concentration of radiotracer and weight of EGMP-sorbent. Fig. 9 shows the variation of (ln Kd − 2 ln[(LH2 )]) vs. ln[H+ ] for Eu3+ and Am3+ uptake in the gel or membrane. It is seen from Fig. 9 that the slope is 2 for Eu3+ and Am3+ uptake in the gel, and 3 for Eu3+ uptake in the membrane. As slope represents n in Eq. (7), it appears that Eu(NO3)2+ and Eu3+ are complexed with phosphate moiety of EGMP in the gel and membrane, respectively. The EGMP-gel can swell in water, but EGMP-anchored in the pores of membrane cannot swell as membrane acts as mechani-
        
K.A. Mauritz, R.B. Moore, Chem. Rev. 104 (2004) 4535. E. Bakker, P. Bühlmann, E. Pretsch, Chem. Rev. 97 (1997) 3083. A. Sachdeva, S. Sodaye, A.K. Pandey, A. Goswami, Anal. Chem. 78 (2006) 7169. M. Ulbricht, Polymer 47 (2006) 2217. C.R. Martin, M. Nishizawa, K. Jirage, M. Kang, S.B. Lee, Adv. Mater. 13 (2001) 1352. A.J.B. Kemperman, D. Bargeman, Th.V.D. Boomgaard, H. Strathmann, Sep. Sci. Technol. 31 (1996) 2733. L.D. Nghiem, P. Mornane, I.D. Potter, J.M. Perera, R.W. Cattrall, S.D. Kolev, J. Membr. Sci. 281 (2006) 7. Y.M. Scindia, A.K. Pandey, A.V.R. Reddy, J. Membr. Sci. 249 (2005) 143. K.L. Thunhorst, R.D. Noble, C.N. Bowman, J. Membr. Sci. 128 (1997) 183. A.M. Mika, R.F. Childs, M. West, J.N.A. Lott, J. Membr. Sci. 136 (1997) 221. R.F. Childs, A.M. Mika, A.K. Pandey, C. McCrory, S. Mouton, J.M. Dickson, Sep. Purif. Technol. 22–23 (2001) 507. Y. Kobuke, I. Tabushi, T. Aoki, T. Kamaishi, I. Hagiwara, Ind. Eng. Chem. Res. 27 (1988) 1461. T. Kawai, K. Saito, K. Sugita, A. Katakai, N. Seko, T. Sugo, J.-I. Kanno, T. Kawakami, Ind. Eng. Chem. Res. 39 (2000) 2910. S. Das, A.K. Pandey, A. Athawale, V. Kumar, Y.K. Bhardwaj, S. Sabharwal, V.K. Manchanda, Desalination 232 (2008) 243. E.A. Hegazy, B. El-Gammal, F.H. Khalil, T.M. Mabrouk, J. Appl. Polym. Sci. 102 (2006) 320. K.L. Thunhorst, R.D. Noble, C.N. Bowman, J. Membr. Sci. 156 (1999) 293. G. Zukowska, J. Williams, J.R. Stevens, K.R. Jeffrey, A. Lewera, P.J. Kulesza, Solid State Ionics 167 (2004) 123. M. Kufaci, A. Bozkurt, M. Tülü, Solid State Ionics 177 (2006) 1003. J. Coates, in: R.A. Meyers (Ed.), Encyclopedia of Analytical Chemistry, John Wiley & Sons Ltd., Chichester, 2000, pp. 10815–10837. T. Hirotsu, S. Katoh, K. Sugasaka, M. Seno, T. Itagaki, J. Chem. Soc. Dalton Trans. (1986) 1983. H. Nishikawa, E. Tsuchida, J. Phys. Chem. 79 (1975) 2072.