Effect of Template on Chiral Separation of Phenylalanine using Molecularly Imprinted Membrane in Aqueous Medium.
Summary: Wet phase inversion method was used to prepare L-Phenylalanine (L-Phe) and D-Phenylalanine (D-Phe) imprinted poly [(acrylonitrile)-co-(acrylic acid)] membranes for chiral separation. Ultrafiltration experiments were conducted to evaluate the chiral separation ability of the prepared membrane towards racemate aqueous solution of Phenylalanine. The continuous permselectivity was observed by novel membrane. The chiral resolution ability of L-Phe imprinted membrane was much better than that of D-Phe. It was observed that both membranes simultaneously selectively reject selectively adsorbed and selectively permeate solute. The achieved adsorption selectivities of L-Phe imprinted membrane [aAds]L and D-Phe imprinted membrane [aAds]D were 2.6 and 2.40 respectively. Permselectivity of L-Phe imprinted membrane [aPerm]L was 2.56 while D-Phe imprinted membrane's permselectivity [aPerm]D was 2.03.
The rejection selectivities of L-Phe and D-Phe imprinted membranes were [aRej]L=0.32 and [aRej]D =0.28 respectively.
Keywords: Amino acid; Chiral resolution; Molecularly imprinted membrane; Molecular recognition; selectivity; Separation.
In chemical and biological processes the selective separation and recognition of specific target molecule is an important issue . The optical resolution of racemates has been essential in the perfume production pharmaceutical industry food preparation and so forth due to the harmful effect of one of the enantiomer of racemate mixture. The resolution of racemates is the primary method to obtain pure enantiomers in industry [2-4].
The selective sorption abilities of molecularly imprinted polymers (MIPs) are remarkable by the virtue of their procedure of synthesis [5 6]. Usually they are synthesized by very particular substrate that has the ability of selectively bound and separated. A heavily crosslinked rigid polymer is usually synthesized with template molecule. After completion of polymerization a suitable solvent is used to remove template by washing the polymer. When polymer is exposed to a racemate solution the polymer adsorbs one of the enantiomer with notable selectivity [7 8]. The covalent and noncovalent are the major techniques for imprinting [9-12].
The prepared polymer can be cast in the form of membranes known as Molecularly Imprinted Membranes (MIMs). MIMs are considered to be a promising material for selective separation with low energy consumption and can be scale up easily .
The appropriate polymer selection for the membrane is a major issue. Acrylic acid (AA) and acrylonitrile (AN) are most commonly available and have been considered as the most promising materials in a wet phase inversion method . AA contains only one hydroxyl group that is sufficient to make hydrogen with amino acid of template i.e. phenylalanine. The AN in the polymer forms solid matrix for the membrane. While AA forms non-covalent interaction with template which helps in the fixation of template molecule in the membrane matrix .
Many researchers have used poly(AA-co-AN) imprinted membrane for the separation of target molecule (template). Trotta et al. imprinted poly(AA-co-AN) membrane with tetracycline hydrochloride  and naringin  for the separation of tetracycline hydrochloride from chloramphenicol; and to separate naringin from orange juice respectively. Kobayashi et al.  employed Theophylline (THO) as template molecule to imprint poly(AA-co-AN) membrane for the separation of THO and Caffeine (CAF). Cristallini et al.  prepared Uric acid (UA) and THO imprinted poly(AA-co-AN) membranes for separation of UA and THO.
Several groups have devoted their efforts for the chiral separation of amino acids. But very few researchers have employed MIMs for the chiral separation of Phenylalanine. Takeda et al.  used
Nylon 6 nylon 66 and terephthalic phenylene polyamide (TPPP) membranes were imprinted by L-Phe for chiral resolution of Phe in batch binding using ultrafiltration cell. Nylon 6 nylon 66 and TPPP L-Phe imprinted membranes resolute 6.8 4.2 and 1.7 partition coefficients of L and D forms respectively. Takeda et al. might have got much better results if they would have considered the rejection of the solute by membranes. Jiang et al.  imprinted chitosan (CS)/ -glycidoxypropyl- trimethoxysilane (GPTMS) hybrid membrane with L-Phe for chiral resolution of Phe by diffusion cells improving significantly selectivity and achieved a separation factor of the order of 4.5 was achieved in 24 hrs.
First the Kobayashi group [19-22 26 27] had introduced imprinting via phase separation starting with a solution containing the copolymer and the template; same approach has also been used for instance by the Drioli group [24 25]. Later Kobayashi introduced the method via polymerization of single functional monomer to imprint template and cast subsequent film using non-solvent induced phase separation method In this study we have prepared poly(AA-co-AN) MIM by wet phase inversion method in which the in situ implantation of template (L-Phe or D-Phe) was done by non-covalent interactions for the optical resolution of Phenylalanine. Sorption and binding ability was studied along with the permeation selectivity of underivatized enantiomeric solution of Phe using Ultrafiltration experiments. Ultrafiltration technique showed significantly high selectivities in a short period of time. The selective rejection phenomenon of slote was also observed.
The morphology of membrane was characterized by SEM. FT-IR spectroscopy was used to study the chemical structure of membrane.
Results and Discussions
Structure Analysis and Morphology of Membranes
The imprinting phenomenon is the virtue of hydrogen bonding . It is appealing that one hydrogen bond is ample for the imprinting selective recognition of template in aqueous medium. It is well known that weakening effect of water creates hurdle in the formation of noncovalent interactions. In current study the AA serves as functional monomer having single carboxylic group that is enough for the selective recognition of target molecule in water. The chemical functionality shape and size of moieties and selective binding sites in the membrane structure were created after the removal of template.
The spectra of L-Phe imprinted poly(AA-co-AN) and D-Phe imprinted poly(AA-co-AN) membrane were analyzed by FT-IR. The interpretation of FT-IR spectra is summarized in Table-1. The OH dimmer and free OH stretching can be realized at 3466 cm-1 and 3242 cm-1 respectively in L-Phe imprinted membrane and in D-Phe imprinted membrane OH dimmer and free OH stretching appeared at 3461 cm-1 and 3243 cm-1 respectively (Fig. 1). These free OH groups are might be due to the presence of COOH in imprinted poly(AA-co-AN) membranes. It is assumed that due to these free OH group form hydrogen bond with the template. SEM studies revealed that the average thickness of membrane was 25 m and average thickness of dense top layer was 6 m. The measured pore sizes of membrane were less than 25 nm.
Table-1: Assignment of FT-IR spectra L-Phe and D-Phe imprinted P(AA/AN) membranes.
Free COOH Group
Free COOH group
Effect on Swelling Separation Ability of Membrane
It is reported that the functional groups those are responsible for the recognition of template may change their three-dimensional configuration due to swelling of polymer . The swelling rate of L-Phe imprinted membrane was 73% and that of D-Phe imprinted membrane was 75%. It was observed that elasticity and swelling ability of membrane increased in the aqueous medium. According to the theory of induced fit effect" Piletsky et. al.  concluded that solvation of the functional monomer binding ligands are the cause of swelling. Most of functional ligands (from the functional monomers) after the removal of template are probably produced inside the selective cavities. After selective rebinding the volume of the polymer reduced nearly to the original volume. While Ulbricht observed that the increase in permeability is due to the swelling of membrane caused by the binding of template to the imprinting sites .
Template Effect on Selective Solute Rejection
During ultrafiltration process it was observed that solute not only adsorbed on membrane but also rejected by membrane. Fig. 2 shows that in L-Phe imprinted membranes the rejection of D-Phe was higher than the rejection of L-Phe. In case of filtration from D-Phe membranes the rejection of L-Phe was higher than that of D-Phe. The substrates (L-Phe or D-Phe used as template during the synthesis of membrane) after removal left imprinting cavities and channels (corresponding to the size and shape of L-Phe or D-Phe). The recognition of template took place by imprinted cavities and channels with in the membrane matrix worked as gate between pores . When L-Phe imprinted membrane was used these gates allowed L-Phe pass through it and rejected D-Phe. Similarly when D-Phe imprinted membrane was used; L-Phe was rejected and D-Phe was allowed to pass through membrane.
When L-Phe imprinted membranes were used; the rejection of D-Phe after 16 ml of filtration increased by 3.17-folds than rejection of L-Phe and D-Phe imprinted membranes resulted rejection of L-Phe of the order of 3.53-times higher than the rejection of D-Phe. The rejection selectivity for L-Phe imprinted membrane was 0.32 and that of D-Phe imprinted membrane was 0.28. The nano pours and rough surface of membrane can also be the reason of rejection . T. Gotoh et al. have reported that when the amino acid concentration increases the rejection decreases . NTR-7450 nanofiltration membrane was used for the separation of glutathione and its related amino acids (L-glutamate L-cysteine glycine and L-glutamine). We used D-Phe and L-Phe imprinted membranes for chiral separation of Phe.
We observed that the concentration of solute in retentate increased gradually with filtration time while rejection decreased. So we can assume that decrease in rejection with filtration time is due to the increase in concentration of solute in retentate. From above results we can also conclude that selective rejection is the combine effect of selective adsorption and selective permeation.
Template Effect on Selective Solute Adsorption
The adsorbed amounts of D-Phe and L-Phe were 0.0647 mg/g of membrane and 0.1685 mg/g of membrane respectively and adsorption selectivity [aAds]L of 2.6 was achieved using L-Phe imprinted membranes. While D-Phe imprinted membrane showed adsorption selectivity [aAds]D of 2.40 and adsorbed amount of D-Phe was 0.1674 mg/g of membrane while that of L-Phe was 0.0698 mg/g of membrane. It may be concluded that the selective performance of membrane is due to the memory" (the recognition imprinted sites) created in the membrane matrix after removal of template . The high adsorption selectivity and strong binding affinity was observed in both L-Phe and D-Phe imprinted membranes (Table-2). Fig. 3 shows the chiral recognition ability of L-Phe and D-Phe imprinted membranes. There was preferential adsorption of L-Phe over D-Phe using L-Phe imprinted membrane and when D-Phe imprinted membrane was used the adsorbed amount of D-Phe was much more than the amount of L-Phe.
The FT-IR spectra confirmed that the selective performance of L-Phe imprinted membrane is better than that of D-Phe imprinted membrane While the adsorption capacity of D-Phe was more than that of L-Phe it is might be due to swelling effect [31 32] We can say that selective recognition of template by imprinted cavities in the membrane matrix directly effect on selective permeation and selective rejection.
Template Effect on Selective Permeation
Fig. 4 shows typical permeation curves for concentration and flux versus time for separation of Phe isomers mixture obtained by permeation experiments using imprinted poly(AA-co-AN) membranes fixed in ultrafiltration kit by applying a pressure of 1 atm. The concentration of permeate increased gradually with time. The fluxes of the isomers in permeate also increased with time increase in permeability is due to membrane swelling [31 32]. Both L-Phe and D-Phe showed similar trend. The concentration and flux of the two isomers were different and chiral resolution of D L-Phe was thus realized. The maximum separation factor (permselectivity) achieved in this study were about 2.56 and 2.03 for L-Phe imprinted poly(AA-co-AN) membranes and D-Phe imprinted poly(AA-co-AN) membranes respectively (Table-2) and the permeability coefficient P was in 9 x 10-9 m2/s. Fig. 4 illustrates that the permselectivity of L-Phe imprinted and D-Phe imprinted membrane increased with ultrafiltration.
The permselectivity of L-Phe was found to be better than that of D-Phe imprinted membrane. It is be concluded that the template plays an important role on the performance of imprinting membrane used for chiral resolution facilitated permeation through imprinted gates in the membrane and directly influence on selective rejection and selective adsorption.
Table-2: The selective separation profile of L-Phe and D-Phe imprinted membranes after 16 ml filtration of 100-ppm racemate mixture of Phenylalanine
###L-Phe imprinted###D-Phe imprinted
Template Effect on Solute Transport Mechanism
In MIMs transport of solute can be considered by (1) facilitated permeation and/or (2) retarded permeation . The affinity binding is responsible for the facilitated permeation along with the preferential sorption of the template. In facilitated permeation transport of solute depend on the structure of membrane concentration and distribution of MIP sites coupled with transport phenomenon . The transmembrane pores having relatively small diameter might be responsible for the selective separation of enantiomers. Mostly liquid membranes are facilitated permeation membranes. The barrier structure of liquid membranes is non-porous. In retarded permeation other solute transports faster due to affinity binding until a saturation of MIP sites with template is reached. MIP binding capacity helps to evaluate separation efficiency due to the saturation behavior. Those MIM can be solute adsorbers as selectivity is caused by specific adsorption .
Separation efficiency is determined by MIM binding capacity due to the saturation behavior. Based on the aSep data obtained by permeation of substrate and aAds data obtained by uptake values of membrane transport selectivity aTrans was calculated according to the solution-transport mechanism model using equation (8) and listed in Table-2. The transport selectivity of L-Phe imprinted membranes [aTrans]L was 0.98 and transport selectivity of L-Phe imprinted membranes [aTrans]D was 0.84. From these data we conclude that after 16 ml of permeation the permselectivity was higher than adsorption selectivity for both L-Phe and D-Phe imprinted membranes.
From these three selectivity factors it can be concluded that gate effect  plays an important role in selective adsorption of L-Phe and facilitated permeation of L-Phe. The D-Phe imprinted membrane showed similar behavior. D-Phe imprinted membrane rejected L-Phe and retarded transport of L-Phe while D-Phe was successfully recognized by membrane as adsorbed amount of D-Phe was much higher than that of L-Phe and permeation curves show facilitated permeation of L-Phe (template). Therefore the separation mechanism of L-Phe and D-Phe imprinted poly(AA-co-AN) membranes for DL-Phe isomer separation agreed well with the above mechanism (1). Thus we can conclude that the template recognition and increase in facilitated permeation are also functions of membrane swelling along with imprinted gates and cavities in the membrane matrix.
The chemicals purchased from Sigma-Aldrich (USA) were 22-Azobiisobutyronitrile (AIBN) D-Phenylalanine (D-Phe) L-Phenylalanine (L-Phe) underivatized mixture of DL-Phenylalanine (Phe) and Trifluoroacetic acid (TFA). Dimethyl sulfoxide (DMSO) was product of Kanto (Japan). Sulfate (CuSO4) and Acrylic acid (AA) were purchased from Junsei (Japan). Acrylonitrile (AN) was purchased from Yakuri (Japan). Scharlau (Spain) are the suppliers of the solvents i.e. acetonitrile and methanol used in HPLC. All reagents were of analytical grade and used without further purification.
Preparation of Molecularly Imprinted Membrane
To prepare molecularly imprinted poly(AA-co-AN) membranes by wet phase inversion method imprinted polymer was prepared by radical polymerization. In 50 ml DMSO 7.19 ml AA 0.5 g template (L-Phe or D-Phe) and 2 ml TFA were dissolved at 50 OC for 2 h in a polymerization reactor. To the above solution 37.72 g AN and a solution 50 ml DMSO and 0.22 g AIBN were added to above solution and nitrogen gas was purged for 5~10 minutes. The polymerization was done at 60 OC for 6 h under nitrogen atmosphere. The solution was stirred at uniform rotation speed of 200 rpm. 100 ml DMSO was added to the polymer and stirred for 20 h with a uniform rotation speed of 200 rpm at 25 OC. Then the polymer solution was placed in vacuum oven for 24 h at 0.8 atm and 25 OC. With the help of gardener knife polymer solution was cast on glass plate and coagulated in deionized water at 25 OC to get polymeric membrane. DMSO was removed from membrane by extensive washing.
5 % (V/V) acetic acid solution was used for the removal of template.
The L-Phe and D-Phe imprinted membranes prepared by AA and AN successfully recognize template facilitate permeation of template and reject other enantiomer. It was observed that one carboxylic molecule is sufficient for imprinting and recognition.
The interacting imprinting sites in membrane matrix successfully bind template resulting in significantly improved chiral separation followed by ultrafiltration. Both L- and D-Phe imprinted membranes show similar trends. The results of L-Phe imprinted membranes were found to be remarkable. The L-Phe imprinted membrane was much better than D-Phe imprinted membrane in terms of permselectivity adsorption selectivity rejection selectivity and transport selectivity.
1. O. Bruggemann K. Haupt L. Ye E. Yilmaz and K. Mosbach Journal of Chromatography A 889 15 (2000).
3. R Noyori Angewandte Chemie International Edition 41 2008 (2002).
4. K. B. Sharpless AngewandteChemie International Edition 41 2024 (2002).
5. Sellergren B (Ed.) In Molecularly Imprinted Polymers Man-Made Mimics of Antibodies and their Applications in Analytical Chemistry Elsevier Amsterdam (2001).
6. T. Pap and G. Horvai Journal of Chromatography B 804 167 (2004).
7. A. Conesa and C. Palet Desalination 200 110 (2006).8. M. Yoshikawa J. Izumi T. Kitao and S. Sakamoto Macromolecules 29 8197 (1996).
9. A. Kugimiya and T. Takeuchi Electroanalysis 11 1158 (1999).
10. D. A. Spivak and K. J Shea Macromolecules 31 2160 (1998).
11. B. R. Hart D. J. Rush and K. J. Shea Journal of the American Chemical Society 122 460 (2000).
12. M. J. Whitcombe M. E. Rodriguez P. Villar and E. N Vulfson Journal of the American Chemical Society 117 7105 (1995).
13. H. H. Yang S. Q Zhang W. Yang X. L Chen Z. X Zhuang J. G Xu X. R. Wang Journal of the American Chemical Society 126 4054 (2004).
14. R. H. Schmidt K. Mosbach and K. Haupt Advanced Materials 16 719 (2004).15. D. J. Duffy K. Das S. L. Hsu J. Penelle V. M. Rotello and H. D. Stidham Journal of the American Chemical Society 124 8290 (2002).
16. N. Hilal and V. Kochkodan Journal of Membrane Science 213 97 (2003).
17. J. K. Mathew and J. Shea Journal of the American Chemical Society 118 8154 (1996).
18. M. Yoshikawa T. Fujisawa J. I. Izumi T. Kitao and S. Sakamoto Analytica Chimica Acta 365 59 (1998).
19. H. Y. Wang T. Kobayashi and N. Fujii Langmuir 12 4850 (1996).
20. H. Y. Wang T. Kobayashi T. Fukaya and N. Fujii Langmuir 13 5396 (1997).
21. T. Kobayashi H. Y. Wang and N. Fujii Analytica Chimica Acta 365 81 (1998).22. K. Takeda and T. Kobayashi Journal of Membrane Science 275 61 (2006).
23. C. Cristallini G. Ciardelli N. Barbani and P. Giusti Macromolecular Bioscience 4 31 (2004).
24. F. Trotta C. Baggiani M. P. Luda E. Drioli and T. Massari Journal of Membrane Science 254 13 (2005).
25. F. Trotta E. Drioli C. Baggiani and D. Lacopo Journal of Membrane Science 201 77 (2002).26. T. Kobayashi Y. Murawaki P. S. Reddy M. Abe and N. Fujii Analytica Chimica Acta 435 141 (2001).
27. K. Takeda M. Abe and T. Kobayashi Journal of Applied Polymer Science 97 620 (2005).
28. Z. Jiang Y. Yu and H. Wu Journal of Membrane Science 280 876 (2006).
29. T. Zhang F. Liu W. Chen J. Wang and K. Li Analytica Chimica Acta 450 53 (2001).
30. T. A. Sergeyeva S. A. Piletsky A. A. Brovko E. A. Slinchenko L. M. Sergeeva and A. V. El'skaya Analytica Chimica Acta 392 105 (1999).
31. S. A. Piletsky E. V. Piletskaya T. L. Panasyuk A. V. El'skaya R. Levi I. Karube and G. Wulff Macromolecules 311 2137 (1998).
32. M. Ulbricht Journal of Chromatography B 804 113 (2004).
33. S. A. Piletsky T. L. Panasyuk E. V. Piletskaya L. A. Nicholls and M. Ulbricht Journal of Membrane Science 157 263 (1999).
34. M. D. Petala and A. I. Zouboulis Journal of Membrane Science 269 1 (2006).
35. T. Gotoh H. Iguchi and K. Kikuchi Biochemical Engineering Journal 19 165 (2004).
36. D. K. Roper and E. N. Lightfoot Journal of Chromatography A 702 3 (1995).
37. R. J. Ansell Advanced Drug Delivery Reviews 57 1809 (2005).
38. F. Schneider F. S. Piletsky E. Piletska A. Guerreiro and M. Ulbricht Journal of Applied Polymer Science 98 362 (2005).
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of the Chemical Society of Pakistan|
|Date:||Aug 31, 2014|
|Previous Article:||Crystal Structure and DNA-Binding Study of a Monoclinic Polymorph of N N'-bis(2-pyridyl)oxamide.|
|Next Article:||Clerodendrum splendens: A Potential Source of Antimicrobials.|