Synthesis of novel glucose-based polymers and their applications as chiral stationary phases for high performance liquid chromatography
2016-06-22TomoyukiIKAITakayukiYAMADA
Tomoyuki IKAI, Takayuki YAMADA
(Graduate School of Natural Science and Technology, Kanazawa University,Kakuma-machi, Kanazawa 920-1192, Japan)
Synthesis of novel glucose-based polymers and their applications as chiral stationary phases for high performance liquid chromatography
Tomoyuki IKAI*, Takayuki YAMADA
(Graduate School of Natural Science and Technology, Kanazawa University,Kakuma-machi, Kanazawa 920-1192, Japan)
Abstract:Two novel polymers containing glucose units as the main-chain that only differ in terms of their regioregularity were synthesized to evaluate their chiral recognition abilities as chiral stationary phases (CSPs) for high performance liquid chromatography (HPLC). The regioregular polymer (poly-5) shows clear resolution ability for the racemate of cobalt (Ⅲ) acetylacetonate (Co(acac)3), whereas the corresponding regioirregular polymer (poly-3) does not show any chiral recognition for Co(acac)3. The regioregular polymer main-chain seems to play an important role not only in providing an efficient interaction with the racemate but also in expressing the chiral recognition ability as a CSP for HPLC.
Key words:biomass; chiral discrimination; chirality; enantioseparation; glucose; regioregularity
Since Pasteur [1] first reported biological enantioselectivity in 1858, chirality has attracted much attention particularly in the pharmaceutical field, and drug companies have been requested to systematically evaluate the biological activity of individual enantiomers. Most of the world’s top selling drugs are now provided as single enantiomers with the desired medicinal effect. Recently, microanalysis of chiral compounds and large-scale preparation of pure enantiomers have become indispensable not only in the fields of pharmaceuticals, agrochemicals, foods, and fragrances but also in the field of functional materials, including nonlinear optical molecules and ferroelectric liquid crystals. As a consequence, enormous efforts have been undertaken to explore practical ways for chiral analysis and preparation of optically active compounds. Since Pasteur [2] found that crystals of racemic sodium ammonium tartrate consisted of two types of enantiomorphic shapes in 1848, which was the first breakthrough in the field of resolution, several resolution techniques have been proposed. Among these methods, direct enantioseparation using chiral stationary phases (CSPs) in high performance liquid chromatography (HPLC) is an effective and simple method capable of applying to both analytical and preparative purposes [3-17]. The recent progress in the chiral HPLC method is mainly based on the development of efficient CSPs. It is well known that polysaccharide-based CSPs, such as cellulose and amylose derivatives, show excellent resolution abilities for various chiral compounds [18-25]. These successful examples have inspired chemists to develop more practical CSPs using optically active polymers.
In this study, we synthesized two novel polymers consisting of glucose units as the main-chain, which only differ from each other in terms of the regioregularity, and investigated their chiral recognition abilities as CSPs for HPLC. Because the regioregularity has a potential to provide significant influences in the functionality of the materials [26,27], two kinds of polymers in this paper may show a different chiral recognition ability.
1Experimental
1.1Chemicals
The anhydrous solvents (dichloromethane and tetrahydrofuran (THF)), the common organic solvents, copper (Ⅰ) iodide (CuI), and triphenylphosphine were purchased from Kanto (Tokyo, Japan). Methyl-4,6-O-benzylidene-α-D-glucopyranoside and 4-iodobenzoic acid were bought from Tokyo Kasei (TCI, Tokyo, Japan). 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC-HCl), diethyl azodicarboxylate, diisopropylamine (DIPA), andN,N-dimethyl-4-aminopyridine (DMAP) were purchased from Wako (Osaka, Japan). Sodium hydrogen carbonate (NaHCO3) was obtained from Kishida (Osaka, Japan). Tetrakis(triphenylphosphine)palladium(0) (Pd(PPh3)4) was purchased from Nacalai (Kyoto, Japan). Tetrabutylammonium bromide (TBAB) and cobalt (Ⅲ) acetylacetonate (Co(acac)3) were from Aldrich (Milwaukee, WI, USA). 4-Ethynylbenzoic acid was prepared according to the procedure in the literature [28]. The wide-pore silica gel (Daiso gel SP-1000) with a mean particle size of 7 μm and a mean pore diameter of 100 nm was kindly supplied by OSAKA SODA Co., Ltd. (Osaka, Japan).
1.2Monomer synthesis
1.2.1Synthesis of compound 1
Methyl-4,6-O-benzylidene-α-D-glucopyranoside (1.00 g, 3.54 mmol) was added to a solution of 4-ethynylbenzoic acid (1.14 g, 7.79 mmol) and DMAP (0.95 g, 7.79 mmol) in dichloromethane (11 mL) and the mixture was cooled to 0 ℃ under a nitrogen atmosphere. To this mixture was added EDC-HCl (1.49 g, 7.79 mmol). After stirred at room temperature for 24 h, the reaction system was diluted with dichloromethane and the solution was washed with 1 mol/L HCl aqueous solution and saturated NaHCO3aqueous solution and then dried over Na2SO4. After filtration, the solvent was removed by evaporation and the crude product was purified by silica gel chromatography usingn-hexane-ethyl acetate (3∶1, v/v) as the eluent to give compound 1 as a white solid (1.48 g, 78% yield).1H NMR (500 MHz, CDCl3, room temperature (rt)):δ7.94-7.92 (m, 4H, Ph-H), 7.49-7.45 (m, 6H, Ph-H), 7.44-7.31 (m, 3H, Ph-H), 6.06 (t,J=10.0 Hz, 1H), 5.56 (s, 1H), 5.25 (dd,J=19.0 Hz, 3.5 Hz, 1H), 5.16 (d,J=4.0 Hz, 1H), 4.37 (dd,J=11.0 Hz, 5.5 Hz, 1H), 4.11-4.02 (m, 1H), 3.92-3.84 (m, 2H), 3.43 (s, 3H, OCH3), 3.22 (s, 1H), 3.20 (s, 1H).13C NMR (125 MHz, CDCl3, rt):δ165.37, 165.04, 136.89, 132.23, 132.10, 129.88, 129.67, 129.58, 129.17, 128.94, 128.31, 127.42, 127.05, 126.24, 101.73, 97.81, 82.78, 82.75, 80.60, 80.36, 79.32, 72.73, 69.88, 68.97, 62.61, and 55.62. Anal. Calcd for C32H26O8: C, 71.37; H, 4.87. Found: C, 71.42; H, 5.03.
1.2.2Synthesis of compound 2
The title compound 2 was prepared from methyl-4,6-O-benzylidene-α-D-glucopyranoside and 4-iodobenzoic acid in the same way for compound 1 and obtained in 90% yield as a white solid.1H NMR (500 MHz, CDCl3, rt):δ7.76-7.72 (m, 4H, Ph-H), 7.54-7.50 (dd,J=9.0 Hz, 3.0 Hz, 4H, Ph-H), 7.42-7.38 (m, 2H, Ph-H), 7.33-7.31 (m, 3H, Ph-H), 5.99 (t,J=10.0 Hz, 1H), 5.56 (s, 1H), 5.20 (dd,J=9.5 Hz, 4.0 Hz, 1H), 5.14 (d,J=4.0 Hz, 1H), 4.37 (dd,J=10.0 Hz, 5.0 Hz, 1H), 4.06 (m, 1H), 3.90-3.83 (m, 2H), 3.43 (s, 3H, OCH3).13C NMR (125 MHz, CDCl3, rt):δ165.62, 165.25, 137.98, 137.82, 136.89, 131.41, 131.24, 129.22, 129.07, 128.47, 128.34, 126.26, 101.77, 101.22, 97.79, 79.30, 72.72, 69.90, 69.01, 62.62, and 55.67. Anal. Calcd for C28H24I2O8: C, 45.31; H, 3.26. Found: C, 45.15; H, 3.27.
1.2.3Synthesis of compound 4
Methyl-4,6-O-benzylidene-α-D-glucopyranoside (1.00 g, 3.54 mmol) was added to a solution of 4-iodobenzoic acid (3.52 g, 14.2 mmol) and triphenylphosphine (3.71 g, 14.2 mmol) in THF (28 mL) and the mixture was cooled to 0 ℃ under a nitrogen atmosphere. To this mixture was added diethyl azodicarboxylate (1.49 g, 8.56 mmol). After stirred at 60 ℃ for 10 h, the reaction system was diluted with dichloromethane and the solution was washed with saturated NaHCO3aqueous solution and then dried over Na2SO4. After filtration, the solvent was removed by evaporation and the crude product was passed through a pad of silica gel usingn-hexane-dichloromethane-THF (7∶1∶1, v/v/v) as the eluent. The obtained white solid was added to a solution of 4-ethynylbenzoic acid (0.28 g, 1.9 mmol) and DMAP (0.23 g, 1.86 mmol) in dichloromethane (17 mL) and the mixture was cooled to 0 ℃ under a nitrogen atmosphere. To this mixture was added EDC-HCl (0.36 g, 1.86 mmol). After stirred at room temperature for 20 h, the reaction system was diluted with dichloromethane and the solution was washed with 1 mol/L HCl aqueous solution and saturated NaHCO3aqueous solution and then dried over Na2SO4. After filtration, the solvent was removed by evaporation and the crude product was purified by silica gel chromatography usingn-hexane-ethyl acetate (4/1, v/v) as the eluent to give compound 4 as a white solid (1.00 g, 44% yield).1H NMR (500 MHz, CDCl3, rt):δ7.92 (d,J=8.4 Hz, 2H, Ph-H), 7.75 (d,J=8.8 Hz, 2H, Ph-H), 7.67 (d,J=8.8 Hz, 2H, Ph-H), 7.48 (d,J=8.4 Hz, 2H, Ph-H), 7.43-7.41 (m, 2H, Ph-H), 7.33-7.31 (m, 3H, Ph-H), 6.01 (t,J=9.6 Hz, 1H), 5.56 (s, 1H), 5.21 (dd,J=9.6 Hz, 4.0 Hz, 1H), 5.15 (d,J=3.6 Hz, 1H), 4.37 (dd,J=10.4 Hz, 5.2 Hz, 1H), 4.06 (m, 1H), 3.91-3.83 (m, 2H), 3.44 (s, 3H, OCH3), 3.20 (s, 1H).13C NMR (125 MHz, CDCl3, rt):δ165.66, 165.07, 137.99, 136.92, 132.17, 131.43, 129.73, 129.61, 129.23, 128.51, 128.36, 127.10, 126.27, 101.78, 101.74, 97.81, 82.81, 80.35, 79.35, 72.78, 69.89, 69.03, 62.65, and 55.68. Anal. Calcd for C30H25IO850.3H2O: C, 55.79; H, 4.00. Found: C, 55.85; H, 3.95.
1.3Polymerization
1.3.1Synthesis of poly-3
To a solution of compound 1 (0.22 g, 0.40 mmol) and compound 2 (0.29 g, 0.40 mmol) in degassed toluene/THF/DIPA (6∶4∶5, v/v/v) (15 mL) were added Pd(PPh3)4(40 mg, 34 μmol), and CuI (31 mg, 0.16 mmol). The solution was stirred at 60 ℃ for 24 h. After cooling to room temperature, the reaction mixture was poured into a large amount of hexane and the resulting precipitate was collected by centrifugation to give the target polymer poly-3 as a pale yellow solid (0.30 g, 74% yield).1H NMR (500 MHz, CDCl3, 55 ℃):δ8.02-7.88 (br, 4H, Ph-H), 7.57-7.23 (br, 9H, Ph-H), 6.08-6.00 (br, 1H), 5.55 (s, 1H), 5.27-5.20 (br, 1H), 5.20-5.10 (br, 1H), 4.40-4.31 (br, 1H), 4.10-4.00 (br, 1H), 3.92-3.80 (m, 2H), 3.43 (s, 3H, OCH3). Anal. Calcd for (C30H24O850.1C30H24IO851.2H2O)n: C, 66.27; H, 4.85. Found: C, 66.31; H, 4.57.
1.3.2Synthesis of poly-5
The title compound poly-5 was prepared from compound 4 in the same way for poly-3 and obtained in 91% yield as a pale yellow solid.1H NMR (500 MHz, CDCl3, 55 ℃):δ7.93 (t,J=10.0 Hz, 4H, Ph-H), 7.50-7.37 (m, 6H, Ph-H), 7.32-7.25 (br, 3H, Ph-H), 6.00 (t,J=10.0 Hz, 1H), 5.54 (s, 1H), 5.24 (dd,J=8.5 Hz, 3.5 Hz, 1H), 5.13 (d,J=2.0 Hz, 1H), 4.35 (m, 1H), 4.05 (m, 1H), 3.86 (m, 2H), 3.42 (s, 3H, OCH3). Anal. Calcd for (C30H24O850.5H2O)n: C, 69.09; H, 4.83. Found: C, 69.12; H, 4.74.
1.4Preparation of HPLC columns
The CSPs for HPLC were prepared by coating polymers on silica gel (polymer/silica, 20∶80, mass ratio) according to the method reported previously [29], and the solvent was evaporated under reduced pressure. After fractionating with sieves, the obtained silica was packed into a column (25 cm×0.20 cm) by a slurry packing technique [30].
1.5Analytical methods
The1H and13C NMR spectra were measured in CDCl3at room temperature or 55 ℃ with a JEOL ECA-500 spectrometer (JEOL, Tokyo, Japan). The relative molecular masses and distributions of the polymers were estimated using size-exclusion chromatography (SEC) equipped with a TSKgel MultiporeHXL-M column (Tosoh, Tokyo, Japan), a JASCO PU-2080 Plus HPLC pump (JASCO, Tokyo, Japan) and a JASCO UV-970 UV/VIS detector at 254 nm, where CHCl3was used as the eluent. The molecular weight calibration curve was obtained with polystyrene standards (Tosoh). The UV-vis absorption and circular dichroism (CD) spectra were measured in CHCl3with 1.0 mm quartz cell using JASCO V-570 and JASCO J-720 spectrometers. The chromatographic experiments were performed using a JASCO PU-980 chromatograph equipped with UV (JASCO MD-2010) and polarimetric (JASCO OR-990, Hg-Xe without filter) detectors at room temperature. A solution of the racemate was injected into the chromatographic system using a Rheodyne Model 7125 injector.
Fig. 1 Synthesis of poly-3 and poly-5 by Sonogashira-Hagihara cross-coupling reaction
2Results and discussion
2.1Polymer synthesis
Palladium/copper-catalyzed copolymerization of the two optically active glucose-based monomers 1 and 2, which possess 4-ethynylbenzoate and 4-iodobenzoate units as polymerizable groups, respectively, resulted in the formation of a regioirregular polymer (poly-3 in Fig. 1). The regioregular polymer poly-5 was also synthesized by homopolymerization of compound 4 bearing 4-iodobenzoate at the 2-position of the glucose unit and 4-ethynylbenzoate at the 3-position. The relative molecular masses of the obtained polymers were estimated to exceed 2.0×104g/mol by SEC (Table 1). Poly-3 and poly-5 have the same primary structures but different regioregularities, and their solubilities in organic solvents are different. For example, poly-3 is soluble in CHCl3and THF, whereas poly-5 is soluble in CHCl3, but totally insoluble in THF. Such a difference in the solubility of the polymers seems to be attributed to some sort of the difference in their aggregate environments. Therefore, we used CHCl3as the solvent in the following spectroscopic analysis and the preparation of the CSPs.
Table 1 Polymerization results
Mn: determined by SEC in CHCl3at 40 ℃ (PSt standards).
2.2Chiroptical properties of poly-3 and poly-5
The CD and absorption spectra of poly-3 and poly-5 were measured in CHCl3(Fig. 2). Both polymers exhibited characteristic CD in the absorption region of the diphenylacetylene units in the backbone (250-350 nm), and their CD spectral patterns and intensities are almost identical to each other. This result indicates that the regioregularity in the main-chains hardly affects the chiroptical properties of the polymers. The clear split-type CD absorption, in which the first Cotton effect at around 320 nm is positive, can be attributed to chiral exciton coupling between the diphenylacetylene units in the backbone.
Fig. 2 CD and absorption spectra of poly-3 and poly-5 in CHCl3 at 25 ℃[polymer]=0.10 mmol/L.
2.3Application to chiral stationary phases
The packing materials were prepared by coating the polymers on macroporous silica gel. The plate numbers of the columns were approximately 2 000 for benzene using a hexane/2-propanol (90∶10, v/v) mixture as the eluent at a flow rate of 0.1 mL/min. Fig. 3a shows the chromatograms for the resolution of the racemic Co(acac)3on the poly-5-based CSP. The enantiomers were eluted at the retention times oft1andt2with a partial resolution, where the (-)-enantiomer eluted first followed by the (+)-isomer. The dead time (t0) was estimated to be 7.91 min with 1,3,5-tri-tert-butylbenzene as a non-retained compound under the same conditions [31]. The retention factors,k1(=(t1-t0)/t0) andk2(=(t2-t0)/t0), were determined to be 1.78 and 2.05, respectively, which gives a separation factorα(=k2/k1) of 1.15. On the other hand, the regioirregular poly-3-based CSP failed to resolve the racemate of Co(acac)3, as shown in Fig. 3b, where a single peak containing both components of the enantiomers was detected by a UV detector (254 nm) and a polarimetric detector did not show any detectable signal. In addition, itsk1value decreased to approximately one-fifth of the value of the poly-5-based CSP. These results indicate that the regioregular structure in the poly-5 backbone plays an important role in not only efficiently interacting with the racemate but also in the chiral recognition ability as a CSP for HPLC.
Fig. 3 Resolution results of Co(acac)3 on (a) poly-5 and (b) poly-3
3Conclusions
We synthesized two novel polymers containing glucose units as the main-chains that have the same primary structures except for a difference in the regioregularities, and their chiral recognition abilities were investigated as CSPs for HPLC. It was found that poly-5 with a regioregular structure exhibits resolution ability for the racemate of Co(acac)3, in which the regioregular backbone seems to provide the enantioselective interaction ability for the CSP. Detailed investigations of the resolution ability of the regioregular poly-5-based CSP are currently in progress and will be reported in due course.
References:
[1]Pasteur L. C R Acad Sci, 1858, 46: 615
[2]Pasteur L. C R Acad Sci, 1848, 34: 535
[3]Allenmark S G. Chromatographic Enantioseparation: Methods and Applications. Chichester: Ellis Horwood, 1988
[4]Pirkle W H, Pochapsky T C. Chem Rev, 1989, 89: 347
[5]Ahuja S. Chiral Separations by Chromatography. Washington, DC: American Chemical Society, 2000
[6]Francotte E R. J Chromatogr A, 2001, 906: 379
[7]Nakano T. J Chromatogr A, 2001, 906: 205
[8]Yamamoto C, Okamoto Y. Bull Chem Soc Jpn, 2004, 77: 227
[9]Thompson R. J Liq Chromatogr Relat Technol, 2005, 28: 1215
[10]Okamoto Y, Ikai T. Chem Soc Rev, 2008, 37: 2593
[11]Ward T J, Ward K D. Anal Chem, 2012, 84: 626
[12]Chankvetadze B. J Chromatogr A, 2012, 1269: 26
[13]Yashima E, Iida H, Okamoto Y//Schurig V, ed. Differentiation of Enantiomers I, Topics in Current Chemistry 340. Berlin Heidelberg: Springer-Verlag, 2013: 41
[14]Fernandes C, Tiritan M E, Pinto M. Chromatographia, 2013, 76: 871
[15]Sierra I, Pérez-Quintanilla D, Morante S, et al. J Chromatogr A, 2014, 1363: 27
[16]Al-Othman Z A, Al-Warthan A, Ali I. J Sep Sci, 2014, 37: 1033
[17]Hyun M H. Chirality, 2015, 27: 576
[18]Okamoto Y, Kaida Y. J Chromatogr A, 1994, 666: 403
[19]Yashima E, Yamamoto C, Okamoto Y. Synlett, 1998: 344
[20]Okamoto Y, Yashima E. Angew Chem Int Ed, 1998, 37: 1020
[21]Yashima E, Okamoto Y. Bull Chem Soc Jpn, 1998, 68: 3289
[22]Tachibana K, Ohnishi A. J Chromatogr A, 2001, 906: 127
[23]Stringham R W. Adv Chromatogr, 2006, 44: 257
[24]Ikai T, Okamoto Y. Chem Rev, 2009, 109: 6077
[25]Shen J, Ikai T, Okamoto Y. J Chromatogr A, 2014, 1363: 51
[26]Osaka I, Mccullough R D. Accounts Chem Res, 2008, 41: 1202
[27]Deckers S, Steverlynck J, Willot P, et al. J Phys Chem C, 2015, 119: 18513
[28]Melissaris A P, Litt M H. J Org Chem, 1992, 57: 6998
[29]Okamoto Y, Aburatani R, Hatada K. J Chromatogr A, 1987, 389: 95
[30]Okamoto Y, Kawashima M, Hatada K. J Chromatogr A, 1986, 363: 173
[31]Koller H, Rimbock K H, Mannschreck A. J Chromatogr, 1983, 282: 89
DOI:10.3724/SP.J.1123.2015.10037
*Received date:2015-10-26
*Corresponding author. Tel/Fax: +81 76 234 4781, E-mail: ikai@se.kanazawa-u.ac.jp.
Foundation item:This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grants-in-Aid for Scientific Research (C) (Grant No. 26410129).
CLC number:O658
Document code:AArticle IC:1000-8713(2016)01-0004-06
Special Issue on Chiral Separation by Chromatography·Communication