- Email: [email protected]
Molecular orientation of gel forming compounds and their effect on molecular-shape selectivity in liquid chromatography
Journal of Chromatography A, 1324 (2014) 149–154
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Molecular orientation of gel forming compounds and their effect on molecular-shape selectivity in liquid chromatography Abul K. Mallik a , Sudhina Guragain a , Hiroshi Hachisako a , Mohammed Mizanur Rahman c , Makoto Takafuji a,b , Hirotaka Ihara a,b,∗ a
Department of Applied Chemistry and Biochemistry, Faculty of Engineering, Kumamoto University, 2-39-1 Kurokami, Kumamoto, Japan Kumamoto Institute for Photo-Electro Organics (Phoenics), Kumamoto 862-0901, Japan c Department of Applied Chemistry and Chemical Engineering, University of Dhaka, Dhaka 1000, Bangladesh b
a r t i c l e
i n f o
Article history: Received 9 September 2013 Received in revised form 13 November 2013 Accepted 14 November 2013 Available online 22 November 2013 Keywords: Reversed-phase chromatography Molecular gel Molecular-shape selectivity Carbonyl− interaction Polycyclic aromatic hydrocarbons Tocopherol
a b s t r a c t Double-alkylated glutamic acid and aminoadipic acid derived gel systems (G1 and G2) have been prepared and then grafted onto silica particles. Gel-forming compounds-grafted silica particles (Sil-G1 and Sil-G2) were then applied for the liquid chromatographic separation of shape-constrained isomers of polycyclic aromatic hydrocarbons (PAHs) and tocopherols as stationary phases. G1 and G2 were analyzed before and after immobilization onto silica. Elemental analysis and thermogravimentric analysis (TGA) results showed almost the same amounts of organic phases were grafted in Sil-G1 and Sil-G2 phases. However, chromatographic results showed large differences of molecular-shape selectivity between the stationary phases. Little change in the chemical structures of the gel-forming compounds can change the molecular orientation as well as drastic change in molecular recognition in liquid chromatography. Molecular orientation of the functional groups affected the interaction mechanism for the separation of shape-constrained isomers. Comparison of the shape selectivity of Sil-G1 and Sil-G2 phases with other commercial columns is also described. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Due to the differences in stationary-phase chemistry affect the selectivity or column performance, a variety of synthetic approaches are utilized by column manufacturers to address separation needs. A considerable research has been performed on synthesis and characterization of stationary phases based on alkyl modified silica, since the development of reversed-phase liquid chromatography (RPLC) in the early 1970s. Among the alkyl bonded phases, octadecylsilica phases (ODSs or C18 ) are the most commonly used, and those phases can be basically divided into two types, depending on the bonding chemistry. One type is polymeric, and other is monomeric. The selectivity and the ordering differences between polymeric and monomeric ODS phases have already been reported [1–7]. In general, better molecular-shape selectivity for the separations of polycyclic aromatic hydrocarbons (PAHs) isomers can usually be achieved with polymeric stationary phases
∗ Corresponding author at: Department of Applied Chemistry and Biochemistry, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan. Tel.: +81 96 342 3661; fax: +81 96 342 3661. E-mail address: [email protected] (H. Ihara). 0021-9673/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chroma.2013.11.032
compared with monomeric one [1,6,8]. The term “shape selectivity” is used to describe the chromatographic quality of certain stationary phases to recognize and separate certain geometric isomers based on their molecular-shape [9]. There are some other factors, which influences molecular-shape recognition in LC, including stationary phase bonding density [9–12], alkyl-phase chain length [3,13], and column temperature [14]. In a recent paper, Sander et al. reported that the use of spacer between C18 chains could improve the shape selectivity [15]. Longer alkyl chain phases (C30 and C34 ) were also developed for the separation of shape-constrained larger bio-molecules, such as isomers of carotenoids and tocopherols [16,17]. On the other hand, molecular-shape selectivity can be achieved by orientation of weak interaction sites in molecular gel-forming compound-grafted silica stationary phases [18–21]. Through intermolecular interactions, molecular gel-forming compounds (consisting of self-assembling small molecules) form highly oriented structures [22–24]. They form analogous structures of the lipid membrane of a biological cell. The ordering of small molecules in such self-assemblies is effective to recognize guest molecules through efficient intermolecular interactions. Selfassembling integrated materials have been investigated with the viewpoint of developing molecular recognition systems [25,26].
150
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
Fig. 1. Schematic illustration of the possibility of intermolecular hydrogen bonding for both of the organic phases. Intermolecular hydrogen bonding in Sil-G2 is supposed to be stronger than Sil-G1 due to closer interaction sites.
Our previous analyses showed that dialkyl l-glutamide derivatives form fibrous supramolecular assemblies and demonstrate unique properties based on their highly oriented structures in aqueous and organic media [27]. When we applied molecular-gel forming compound-grafted silica as a stationary phase for molecular-recognition in high-performance liquid chromatography (HPLC), we observed very high molecular-shape selectivity. Dialkyl l-glutamide derivative-grafted silica stationary phases showed much better selectivities than commonly used commercial phases like C18 (monomeric and polymeric) and C30 phases [18–21]. In this work, we newly introduced dialkyl l-2-aminoadipic acidderived gel-forming compound-grafted silica (Sil-G2) phase and compared the properties and chromatographic results with dialkyl l-glutamide-derived organic phase (Sil-G1). The idea is based on the expectations that highly chirally ordered structure of aminoadipic acid derivative through intermolecular H-bonding interaction will be available than glutamic acid derivative as illustrated in Fig. 1. As a result, due to more oriented functional groups (multiple interaction sites) and interaction aspect ratio with solutes, higher retention for linear and planar molecules were expected on Sil-G2 phase than Sil-G1 phase. Finally, the effect of orientation of the organic phase for the separation of shape-constrained isomers of PAHs and tocophers is described with the view point of interaction mechanism. 2. Experimental 2.1. Materials The tocopherol isomers were obtained from CalBiochem, USA. l-Glutamic acid, stearylamine, diethylphosphorocyanidate (DPEC, peptide synthesis reagent), triethylamine (TEA), and l-2aminoadipic acid were purchased from Wako (Japan) and used without further purification. PAHs samples were purchased from TCI (Japan). The gel-forming compound-grafted silica (Sil-G1 and Sil-G2) stationary phases were synthesized, characterized, and packed into stainless steel columns (150 mm × 4.6 mm i.d.). YMC silica (YMC SIL-120-S5 having a 5 m diameter, a 12 nm pore size, and surface coverage 300 m2 g−1 ) was used. HPLC-grade solvents were used in chromatographic separations.
2.2. Synthesis of the molecular gel-forming compounds (N ,N -dioctadecyl-N˛ -[(4-carboxybutanoyl)]-l-glutamide (G1, m = 1) and N ,N -dioctadecyl-l-2-N˛ -[(4-carboxy) butanoyl]aminoadipamide (G2, m = 2))-grafted silica phases (Sil-G1 and Sil-G2) The compound G1 was synthesized from N-benzyloxycarbonyll-glutamic acid through alkylation, debenzyloxycarbonylation, and finally ring-opening reaction with glutaric anhydride to obtain G1. The compound G1 was then immobilized onto silica (SilG1) according to previously reported methods [18] as shown in Fig. 2. Initially, the surface of silica was modified with 3aminopropyltrimethoxysilane (APS) to obtain Sil-APS. G1 was then coupled with Sil-APS through amide coupling reaction to get Sil-G1. The compound G2 was synthesized from N-benzyloxycarbonyl-l2-aminoadipic acid through alkylation, debenzyloxycarbonylation, and finally ring-opening reaction with glutaric anhydride to obtain G2 according to previously reported methods [27] as shown in Fig. 2. Similar procedure was also applied for the immobilization of G2 onto silica to get Sil-G2. 2.3. Proton NMR 1H
NMR were recorded with a JEOL JNM-LA400 (Japan) instrument at 400 MHz in CDCl3 solutions at 25 ◦ C. Chemical shifts (␣) of 1 H, expressed in parts per million (ppm) with use of the internal standards Me4 Si (ı = 0.00 ppm). 2.4. FT-IR, DRIFT and elemental analysis FT-IR measurements were conducted with JASCO FT/IR-4100 (Japan). For DRIFT measurement accessory DR PRO410-M (JASCO, Japan) was used. Elemental analyses were carried out on a Yanaco CHN Corder MT-6 Apparatus (Japan). 2.5. Thermogravimetric analysis (TGA) TGA were performed on a Seiko EXSTAR 6000 TG/DTA 6300 thermobalance in static air from 30 to 900 ◦ C at a heating rate of 10 ◦ C/min.
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
151
Fig. 2. Synthesis of N ,N -dioctadecyl-N˛ -[(4-carboxybutanoyl)]-l-glutamide and N ,N -dioctadecyl-l-2-N˛ -[(4-carboxy)butanoyl]aminoadipamide-grafted silica phases (SilG1 and Sil-G2).
2.6. Liquid chromatography Liquid chromatography was carried out with methanol or a mixture of methanol/water as a mobile phase at a flow rate of 1.0 ml min−1 under isocratic condition. The chromatographic system included a JASCO 1580 pump and a JASCO MD-2010 plus UV–vis photodiode array detector. As the sensitivity of UV detector is high, 5 l sample was injected through a Reodyne Model 7725 injector having 20 l loop. The column temperature was maintained by using a column jacket having heating and cooling system. A personal computer connected to the detector and pump with ChromNAV (Ver 1.17 or later) software was used for system control and data analysis. Chromatographic grade solvent was used to prepare mobile phase solutions. The separation factor (˛) was given by the ratio of retention factors (k). The retention time of D2 O was used as the void volume (t0 ) marker (the absorption for D2 O was measured at 400 nm).
grafted amide bonded lipids on the silica surface for both of the phases. Equally important is the appearance of N H stretching (3284 cm−1 ) in the spectra for Sil-G1 and Sil-G2, providing further evidence that both of G1 and G2 was successfully grafted onto the silica surface. 3.3. Immobilization amount The TGA illustrates the distinct effect of silane functionality on bonded phases surface coverage to a greater extent than values from elemental analysis. The significant weight loss step from 300 to 900 ◦ C correlates to the decomposition of the chemically bonded organic phase onto silica surface. Typical TGA curves for the bare silica, Sil-APS, Sil-G1, and Sil-G2 are depicted in Fig. 4. The weight retention profile of silica and Sil-APS reached a plateau at about 110 ◦ C (drying period), indicating the removal of surface
3. Results and discussion 3.1. Synthesis of G1 and G2 and immobilization onto silica Molecular-gel forming compounds (G1 and G2) were designed, synthesized, immobilized onto the silica surface (Fig. 2), and used as packed column for liquid chromatography. The functional groups were oriented in such a way that l-2-aminoadipic acid derivative (G2) believed to be more oriented than in l-glutamic acid derivative (G1) due to stronger intermolecular H-bonding was anticipated for G2 compared to G1 (Fig. 1). 3.2. DRIFT analysis The grafting of organic molecules onto silica can be confirmed by DRIFT, elemental analysis, and thermogravimentric analysis (TGA). As shown in Fig. 3, a group of peaks at 2925 and 2853 cm−1 , respectively, were attributed to the C H bond stretching of the long alkyl chain for both of the stationary phases (Sil-G1 and Sil-G2). The intense bands at 1644 and 1549 cm−1 , indicating the presence of
Fig. 3. DRIFT spectra of bare silica, Sil-APS, Sil-G1, and Sil-G2.
152
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
analysis data strongly corresponds to the results obtained by TGA measurements. 3.4. Chromatographic evaluation of Sil-G1 and Sil-G2
Fig. 4. TGA spectra of bare silica, Sil-APS, Sil-G1, and Sil-G2.
water. After the thermal degradation of the APS the weight of the sample was constant from 600 to 900 ◦ C. A plateau in the weight retention curve of Sil-G1 and Sil-G2 was also observed as the temperature reached 600 ◦ C, confirming that there is no organic content remaining on the silica at 900 ◦ C. Using the TGA curve of silica as reference, the weight of the immobilized APS can be calculated as 8.49 wt.% of the total mass. Similarly, TGA revealed that 14.65 wt.% G1 and 13.82 wt.% G2 were grafted on the silica surface, using the weight retention of Sil-APS as a reference at 900 ◦ C. The immobilization amount in Sil-G1 was slightly higher than in Sil-G2. The grafted amount of organic phase onto silica can also be determined from the %C obtained from elemental analysis data. We found 8.56% C, 2.48% H, and 2.78% N for Sil-APS, 20.66% C, 3.88% H, and 3.24% N for Sil-G1, and 20.22% C, 3.61% H, and 3.12% N for Sil-G2, respectively from elemental analysis. The surface coverage of silica particles were determined to be 4.82, 0.88, and 0.82 mol m−2 , respectively according to previously reported method [28,29]. The percentage of carbon of Sil-APS was deducted from the values of Sil-G1 and Sil-G2 during the calculation of surface coverage to get the actual coverage by G1 and G2. The grafting amount from the elemental
To assess the selectivity differences between Sil-G1 and Sil-G2, evaluation was carried out with the separation of shapeconstrained isomers of PAHs and tocopherols. In this work, chromatographic results were analyzed and compared between only our synthesized two slightly different organic phases (SilG1 and Sil-G2). Because, previously we reported that dialkyl l-glutamide-derived stationary phases showed much better selectivity than commonly used important commercial columns (monomeric and polymeric ODSs and C30 ) [18–21]. The selectivity differences of the two columns were carried out with the separation of PAHs and the selectivity for two-dimensional shape was studied with a molecular-shape descriptor, such as molecular length-to-breadth (L/B) ratio. The chromatograms for the fourring PAHs by the Sil-G1 and Sil-G2 phases are shown in Fig. 5. If we analyze the separation of two nonlinear and linear PAHs benzo[a]anthracene (2) and naphthacene (4), which have the same number of carbon atoms and -electrons but differ only in their molecular-shape such as the length and aspect ratio (L/B = 1.60 and L/B = 1.90, respectively), Sil-G2 showed much better linearity selectivity (˛4/2 = 4.50) than Sil-G1 (˛4/2 = 2.09) with the same chromatographic conditions. At the same time, better separation was also observed with Sil-G2 than Sil-G1 for benzo[a]anthracene and chrysene (Fig. 5). To evaluate the shape recognition and interaction mechanism of Sil-G1 and Sil-G2, other chromatographic tests have been established. For example, the selectivity for oterphenyl (L/B = 1.11) and triphenylene (L/B = 1.12) probes with similar L/B values, number of carbon atoms, and -electrons but different molecular planarity, was introduced for the evaluation of the planarity recognition capability of C18 phases by Tanaka et al. [30] and Jinno et al. [31]. Here, we have used another mixture of o-(nonplanar), m-(nonplanar), and p-terphenyl (almost planar) isomers and triphenylene (planar) to investigate both planarity and linearity selectivity with Sil-G1 and Sil-G2 (Fig. 6). Dramatic differences in selectivity were observed for the Sil-G2 and Sil-G1 phases. In Sil-G1, the mixture was separated successfully in the order of o-, m-, and p-terphenyls and triphenylene; however, in
Fig. 5. Separation of four ring PAHs isomers (1, triphenylene; 2, benzo[a]anthracene; 3, chrycene; 4, naphthacene) on Sil-G1 and Sil-G2 phases. Mobile phase: methanol–water (90:10), column temperature 30 ◦ C, flow rate: 1.0 ml min−1 . UV detection: 275 nm.
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
153
Table 1 Separation factors (␣) of PAHs and their isomers for Sil-G1 and G2 phases. Organic phases
Dibenz[a,h]anthracene/ trans-/cis-Stilbene dibenz[a,c]anthracene
Coronene/ hexahelicene
Naphthacene/ Chrycene/ Triphenylene/ benzo[a]anthracene benzo[a]anthracene o-terphenyl
p-/-m-Terphenyl
Sil-G1 Sil-G2
1.79 1.46
16.3 19.6
2.09 4.50
2.03 4.31
2.92 2.06
Sil-G2, the retention order was changed so that p-terphenyl was eluted after triphenylene, which again suggesting very high linearity/slenderness selectivity of Sil-G2 compared to Sil-G1 phase. However, planarity selectivity (␣triphenylene/o-terphenyl ) was slightly higher for Sil-G1 compared to Sil-G2 (Fig. 6). For better assessment of linearity and planarity selectivity of Sil-G1 and Sil-G2, more results have been included in Table 1. Therefore, the orientation of the functional groups or interaction sites of Sil-G1 and Sil-G2 has much effect on linearity and planarity selectivity toward PAHs. From the above findings of molecular planarity and linearity selectivity of Sil-G1 and Sil-G2, we tried to separate a shapeconstrained vitamin E isomers (␣-, -, ␥-, and ␦-isomers). Among the ␣-,  -, ␥-, and ␦-isomers of tocopherol (Vitamin E), the separation of - and ␥-tocopherol has long presented a special challenge. The baseline separation of these two homologues especially for and ␥-isomers could not be achieved in conventional RP-HPLC and is completely impossible on ODS phases [32–34]. Previously, we reported the separation of tocopherol isomers by poly(octadecyl acrylate-alt-N-octadecylmaleimide)-grafted silica (Sil-poly(ODAalt-OMI)) and poly(octadecyl acrylate-alt-N-octadecyl-ˇ-alanyl maleimide)-silica hybrid (Sil-alt-P) [35,36]. Fig. 7 shows the separation of tocopherol isomers with Sil-G1 and Sil-G2. Slightly better separation of - and ␥-isomers was observed with Sil-G1 compared to Sil-G2, which may be related to more planarity selectivity of the former phase than the later. 3.5. Molecular-recognition mechanism We have reported about the interaction mechanism of the gel-forming compound-grafted silica phase for the separation of
Fig. 6. Separation of o-, m-, p-terphenyl, and triphenylene on Sil-G1 and Sil-G2 phases. Chromatographic conditions are the same as Fig. 7.
1.11 1.48
4.77 3.98
shape-constrained isomers [18,20,21]. In all the cases, our phases showed better selectivities and different interaction mechanism than commercial alkyl phases (ODSs and C30 ). Orientation of the functional groups (interaction sites) are the driving forces for multiple interactions with solutes and eventually obtained high selectivities on gel-forming compound-grafted silica phases [18,20,21]. Herein, we must need to discuss about the tunability of the selectivity due to slight changes in the chemical structure of the organic phase. Generally, the molecular-shape selectivity in the ODSs and C30 phases increases with increasing carbon loading, alkyl chain length, decreasing column temperature as we mentioned earlier [9–17]. All of these have been attributed by slight increase of alkyl chain ordering as well as hydrophobic interactions. Unlike ODS phases, the higher selectivity of Sil-G1 and Sil-G2 cannot be explained by the hydrophobic interactions with the solute molecules. To explain the difference of interaction mechanisms between Sil-G1 and Sil-G2, we propose the direct interaction of solutes with ordered functional groups (carbonyl groups). The gel-forming ability or chirally ordered functional groups of amino adipic acid derived lipid (G2) was expected to be higher than glutamic acid derived lipid (G1) as illustrated in Fig. 1. Therefore, very high linearity selectivity of Sil-G2 can be explained by multiple carbonyl– interactions with higher interaction aspect ratio with the solutes due to more oriented carbonyl groups compared to Sil-G1. As the effect of temperature has great effect on molecularrecognition [14,37–39], we have also observed the temperature dependencies of m- and p-terphenyl on Sil-G1 and Sil-G2 phase and found much higher selectivity with the later phase than the former (Fig. 8). Phase transition was not observed with increasing temperature. This result may be due to reason of more ordering carbonyl groups could be maintained at variable temperature in SilG2 compared to Sil-G1. On the other hand, slight higher planarity selectivity of Sil-G1 compared to Sil-G2 may be due to less accessibility of nonplanar compounds into the interaction sites on Sil-G1.
Fig. 7. Separation of tocopherols on Sil-G1 and Sil-G2 phases. Mobile phase: methanol–water (85:15), column temperature 40 ◦ C, flow rate: 1.0 ml min−1 . UV detection: 285 nm.
154
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
the other hand, Sil-G1 showed slightly higher molecular-planarity selectivity compared to Sil-G2. Therefore, the orientation of the functional groups can change the selectivity of the organic phases. Additionally, both of the phases showed very high molecular-shape selectivity compared to the commercially available alkyl phases. Acknowledgement This research was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. References [1] [2] [3] [4] [5] [6] Fig. 8. Temperature dependencies of the separation factors between m- and pterphenyls on Sil-G1 and Sil-G2 phases. Mobile phase: methanol (100%), flow rate: 1.0 ml min−1 . UV detection: 275 nm.
[7] [8]
In the interaction sites of both of the phases carbonyl groups are supposed to be the driving force for the high shape selectivity by multiple carbonyl– interactions with the solutes. A carbonyl– interaction has been discussed in our previous calculation works [40]. This interaction in a model complex of HCHO-benzene is much stronger (1.87 kcal mol−1 ) than a CH4 –benzene interaction (0.53 kcal mol−1 ) and a plane-to-plane interaction between two benzenes (0.49 kcal mol−1 ) [40]. On the basis of this calculation, when acetone with a carbonyl group was added to the mobile phase, both the retention time and selectivity decreased remarkably. This indicates that acetone functions as an inhibitor for a carbonyl– interaction [18]. Selectivity enhancement through a carbonyl– interaction has been also discussed in relation to homopolymers and copolymers from octadecyl acrylate (ODAn ). When ODAn was grafted onto silica and then evaluated by the retention time of PAHs, the resultant selectivity was higher, especially at crystalline temperatures of ODAn than for ODS columns [41–45]. This is attributed to the fact that ODAn has a crystal-to-isotropic phase transition, so that a multiple carbonyl– interaction with PAHs becomes possible through the ordering of carbonyl groups at lower temperatures (<30 ◦ C), where the polymers are in crystalline state. High shape selectivity was also observed for the alternating copolymer-grafted silica containing ODA and octadecyl maleimide due to ordered carbonyl groups along the polymer main chain, suitable for multiple interactions with the solutes [45]. Therefore, slight changes in the orientation of the functional groups can cause a large effect on the shape selectivity for the gel-forming compound-grafted phases. 4. Conclusion Here we newly introduced dialkyl l-2-aminoadipic acid-derived gel-forming compound-grafted silica (Sil-G2) phase as a promising organic phase for the separation of shape-constrained isomers especially for the separation of molecules with slight molecular linearity difference. The properties of the phase and chromatographic results were compared with our previously reported dialkyl l-glutamide-derived organic phase (Sil-G1). Although the chemical structure of both of the phases are almost similar comparatively highly chirally ordered structure of aminoadipic acid derived organic phase showed more molecular-linearity selectivity. On
[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]
[25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45]
L.C. Sander, S.A. Wise, Anal. Chem. 56 (1984) 504. L.C. Sander, S.A. Wise, J. Chromatogr. 316 (1984) 163. L.C. Sander, S.A. Wise, Anal. Chem. 59 (1987) 2309. J.G. Dorsey, W.T. Cooper, B.A. Siles, J.P. Foley, H.G. Barth, Anal. Chem. 70 (1998) 591R. K. Jinno, T. Ibuki, N. Tanaka, M. Okamoto, J.C. Fetzer, W.R. Biggs, P.R. Griffiths, J.M. Olinger, J. Chromatogr. 461 (1989) 209. K. Jinno, T. Nagoshi, N. Tanaka, M. Okamoto, J.C. Fetzer, W.R. Biggs, J. Chromatogr. 392 (1987) 75. K. Jinno, S. Shimura, N. Tanaka, K. Kimata, J.C. Fetzer, W.R. Biggs, Chromatographia 27 (1989) 285. L.C. Sander, S.A. Wise, in: R.M. Smith (Ed.), Retention and Selectivity Studies in HPLC, Elsevier, Amsterdam, 1994, p. 337. L.C. Sander, M. Pursch, S.A. Wise, Anal. Chem. 71 (1999) 4821. K.B. Sentell, J.G. Dorsey, Anal. Chem. 61 (1989) 930. A.K. Mallik, M.M. Rahman, M. Czaun, M. Takafuji, H. Ihara, J. Chromatogr. A 1187 (2008) 119. A.K. Mallik, M.M. Rahman, M. Czaun, M. Takafuji, H. Ihara, Chem. Lett. 36 (2007) 1460. N. Tanaka, K. Sakagami, M. Araki, J. Chromatogr. 199 (1980) 327. Y. Saito, K. Jinno, J.J. Pesek, Y.L. Chen, G. Luehr, J. Archer, J.C. Fetzer, W.R. Biggs, Chromatographia 38 (1994) 295. M. Kühnle, V. Friebolin, K. Albert, C.A. Rimmer, K.A. Lippa, L.C. Sander, Anal. Chem. 81 (2009) 10136. L.C. Sander, K.E. Sharpless, N.E. Craft, S.A. Wise, Anal. Chem. 66 (1994) 1667. L.C. Sander, K.E. Sharpless, M. Pursch, J. Chromatogr. A 880 (2000) 189. M.M. Rahman, M. Takafuji, H.R. Ansarian, H. Ihara, Anal. Chem. 77 (2005) 6671. M.M. Rahman, M. Takafuji, H. Ihara, J. Chromatogr. A 1119 (2006) 105. A.K. Mallik, H. Qiu, T. Sawada, M. Takafuji, H. Ihara, Chem. Commun. 47 (2011) 10341. A.K. Mallik, H. Qiu, T. Sawada, M. Takafuji, H. Ihara, Anal. Chem. 84 (2012) 6577. P. Terech, R.G. Weiss, Chem. Rev. 7 (1997) 3133. J.H. van Esch, B.L. Feringa, Angew. Chem. Int. Ed. 39 (2000) 2263. H. Ihara, M. Takafuji, T. Sakurai, in: H.S. Nalwa (Ed.), Encyclopedia of Nanoscience and Nanotechnology, vol. 9, American Scientific Publishers, Stevenson Ranch, CA, 2004, p. 473. S. Ong, H. Liu, X. Qiu, G. Bhat, C. Pidgeon, Anal. Chem. 67 (1995) 755. H. Liu, S. Ong, L. Glunz, C. Pidgeon, Anal. Chem. 67 (1995) 3550. H. Hachisako, N. Ryu, R. Murakami, Org. Biomol. Chem. 7 (2009) 2327. G.F. Berendsen, K.A. Pikaart, L. de Galan, J. Chromatogr. 3 (1980) 1437. A.K. Mallik, H. Qiu, M. Takafuji, H. Ihara, Anal. Bioanal. Chem. 404 (2012) 229. N. Tanaka, Y. Tokuda, K. Iwaguchi, J. Araki, J. Chromatogr. 239 (1982) 761. K. Jinno, K. Yamamoto, H. Nagashima, T. Ueda, K. Itoh, J. Chromatogr. 517 (1990) 193. S. Strohschein, M. Pursch, D. Lubda, K. Albert, Anal. Chem. 70 (1998) 13. C.W. Henry, C.A. Fortier, I.M. Warner, Anal. Chem. 73 (2001) 6077. S.L. Abidi, J. Chromatogr. A 881 (2000) 197. A.K. Mallik, T. Sawada, M. Takafuji, H. Ihara, Anal. Chem. 82 (2010) 3320. A.K. Mallik, H. Qiu, M. Takafuji, H. Ihara, Anal. Methods 3 (2011) 1277. M. Pursch, L.C. Sander, H.-J. Egelhaaf, M. Raitza, S.A. Wise, D. Oelkrug, K. Albert, J. Am. Chem. Soc. 121 (1999) 3201. A.K. Mallik, M. Takafuji, H. Ihara, Eur. Polym. J. 45 (2009) 1811. A. Shundo, A.K. Mallik, H. Nakashima, M. Takafuji, S. Nagaoka, M.A.J. Chowdhury, H. Ihara, Chem. Lett. 39 (2010) 844. S. Sakaki, K. Kato, T. Miyazaki, K. Ohkubo, H. Ihara, C. Hirayama, J. Chem. Soc. Faraday Trans. 89 (1993) 659. M.A.J. Chowdhury, R.I. Boyson, H. Ihara, M.T.W. Hearn, J. Phys. Chem. B 106 (2002) 11936. H. Ihara, Y. Goto, T. Sakurai, M. Takafuji, T. Sagawa, S. Nagaoka, Chem. Lett. (2001) 1252. H. Ihara, T. Sagawa, Y. Goto, S. Nagaoka, Polymer 40 (1999) 2555. A. Shundo, M. Fukui, M. Takafuji, K. Akasaka, H. Ohrui, D. Berek, H. Ihara, Anal. Sci. 23 (2007) 1. A.K. Mallik, H. Qiu, T. Sawada, M. Takafuji, H. Ihara, J. Chromatogr. A 1232 (2012) 183.
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
Molecular orientation of gel forming compounds and their effect on molecular-shape selectivity in liquid chromatography Abul K. Mallik a , Sudhina Guragain a , Hiroshi Hachisako a , Mohammed Mizanur Rahman c , Makoto Takafuji a,b , Hirotaka Ihara a,b,∗ a
Department of Applied Chemistry and Biochemistry, Faculty of Engineering, Kumamoto University, 2-39-1 Kurokami, Kumamoto, Japan Kumamoto Institute for Photo-Electro Organics (Phoenics), Kumamoto 862-0901, Japan c Department of Applied Chemistry and Chemical Engineering, University of Dhaka, Dhaka 1000, Bangladesh b
a r t i c l e
i n f o
Article history: Received 9 September 2013 Received in revised form 13 November 2013 Accepted 14 November 2013 Available online 22 November 2013 Keywords: Reversed-phase chromatography Molecular gel Molecular-shape selectivity Carbonyl− interaction Polycyclic aromatic hydrocarbons Tocopherol
a b s t r a c t Double-alkylated glutamic acid and aminoadipic acid derived gel systems (G1 and G2) have been prepared and then grafted onto silica particles. Gel-forming compounds-grafted silica particles (Sil-G1 and Sil-G2) were then applied for the liquid chromatographic separation of shape-constrained isomers of polycyclic aromatic hydrocarbons (PAHs) and tocopherols as stationary phases. G1 and G2 were analyzed before and after immobilization onto silica. Elemental analysis and thermogravimentric analysis (TGA) results showed almost the same amounts of organic phases were grafted in Sil-G1 and Sil-G2 phases. However, chromatographic results showed large differences of molecular-shape selectivity between the stationary phases. Little change in the chemical structures of the gel-forming compounds can change the molecular orientation as well as drastic change in molecular recognition in liquid chromatography. Molecular orientation of the functional groups affected the interaction mechanism for the separation of shape-constrained isomers. Comparison of the shape selectivity of Sil-G1 and Sil-G2 phases with other commercial columns is also described. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Due to the differences in stationary-phase chemistry affect the selectivity or column performance, a variety of synthetic approaches are utilized by column manufacturers to address separation needs. A considerable research has been performed on synthesis and characterization of stationary phases based on alkyl modified silica, since the development of reversed-phase liquid chromatography (RPLC) in the early 1970s. Among the alkyl bonded phases, octadecylsilica phases (ODSs or C18 ) are the most commonly used, and those phases can be basically divided into two types, depending on the bonding chemistry. One type is polymeric, and other is monomeric. The selectivity and the ordering differences between polymeric and monomeric ODS phases have already been reported [1–7]. In general, better molecular-shape selectivity for the separations of polycyclic aromatic hydrocarbons (PAHs) isomers can usually be achieved with polymeric stationary phases
∗ Corresponding author at: Department of Applied Chemistry and Biochemistry, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan. Tel.: +81 96 342 3661; fax: +81 96 342 3661. E-mail address: [email protected] (H. Ihara). 0021-9673/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chroma.2013.11.032
compared with monomeric one [1,6,8]. The term “shape selectivity” is used to describe the chromatographic quality of certain stationary phases to recognize and separate certain geometric isomers based on their molecular-shape [9]. There are some other factors, which influences molecular-shape recognition in LC, including stationary phase bonding density [9–12], alkyl-phase chain length [3,13], and column temperature [14]. In a recent paper, Sander et al. reported that the use of spacer between C18 chains could improve the shape selectivity [15]. Longer alkyl chain phases (C30 and C34 ) were also developed for the separation of shape-constrained larger bio-molecules, such as isomers of carotenoids and tocopherols [16,17]. On the other hand, molecular-shape selectivity can be achieved by orientation of weak interaction sites in molecular gel-forming compound-grafted silica stationary phases [18–21]. Through intermolecular interactions, molecular gel-forming compounds (consisting of self-assembling small molecules) form highly oriented structures [22–24]. They form analogous structures of the lipid membrane of a biological cell. The ordering of small molecules in such self-assemblies is effective to recognize guest molecules through efficient intermolecular interactions. Selfassembling integrated materials have been investigated with the viewpoint of developing molecular recognition systems [25,26].
150
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
Fig. 1. Schematic illustration of the possibility of intermolecular hydrogen bonding for both of the organic phases. Intermolecular hydrogen bonding in Sil-G2 is supposed to be stronger than Sil-G1 due to closer interaction sites.
Our previous analyses showed that dialkyl l-glutamide derivatives form fibrous supramolecular assemblies and demonstrate unique properties based on their highly oriented structures in aqueous and organic media [27]. When we applied molecular-gel forming compound-grafted silica as a stationary phase for molecular-recognition in high-performance liquid chromatography (HPLC), we observed very high molecular-shape selectivity. Dialkyl l-glutamide derivative-grafted silica stationary phases showed much better selectivities than commonly used commercial phases like C18 (monomeric and polymeric) and C30 phases [18–21]. In this work, we newly introduced dialkyl l-2-aminoadipic acidderived gel-forming compound-grafted silica (Sil-G2) phase and compared the properties and chromatographic results with dialkyl l-glutamide-derived organic phase (Sil-G1). The idea is based on the expectations that highly chirally ordered structure of aminoadipic acid derivative through intermolecular H-bonding interaction will be available than glutamic acid derivative as illustrated in Fig. 1. As a result, due to more oriented functional groups (multiple interaction sites) and interaction aspect ratio with solutes, higher retention for linear and planar molecules were expected on Sil-G2 phase than Sil-G1 phase. Finally, the effect of orientation of the organic phase for the separation of shape-constrained isomers of PAHs and tocophers is described with the view point of interaction mechanism. 2. Experimental 2.1. Materials The tocopherol isomers were obtained from CalBiochem, USA. l-Glutamic acid, stearylamine, diethylphosphorocyanidate (DPEC, peptide synthesis reagent), triethylamine (TEA), and l-2aminoadipic acid were purchased from Wako (Japan) and used without further purification. PAHs samples were purchased from TCI (Japan). The gel-forming compound-grafted silica (Sil-G1 and Sil-G2) stationary phases were synthesized, characterized, and packed into stainless steel columns (150 mm × 4.6 mm i.d.). YMC silica (YMC SIL-120-S5 having a 5 m diameter, a 12 nm pore size, and surface coverage 300 m2 g−1 ) was used. HPLC-grade solvents were used in chromatographic separations.
2.2. Synthesis of the molecular gel-forming compounds (N ,N -dioctadecyl-N˛ -[(4-carboxybutanoyl)]-l-glutamide (G1, m = 1) and N ,N -dioctadecyl-l-2-N˛ -[(4-carboxy) butanoyl]aminoadipamide (G2, m = 2))-grafted silica phases (Sil-G1 and Sil-G2) The compound G1 was synthesized from N-benzyloxycarbonyll-glutamic acid through alkylation, debenzyloxycarbonylation, and finally ring-opening reaction with glutaric anhydride to obtain G1. The compound G1 was then immobilized onto silica (SilG1) according to previously reported methods [18] as shown in Fig. 2. Initially, the surface of silica was modified with 3aminopropyltrimethoxysilane (APS) to obtain Sil-APS. G1 was then coupled with Sil-APS through amide coupling reaction to get Sil-G1. The compound G2 was synthesized from N-benzyloxycarbonyl-l2-aminoadipic acid through alkylation, debenzyloxycarbonylation, and finally ring-opening reaction with glutaric anhydride to obtain G2 according to previously reported methods [27] as shown in Fig. 2. Similar procedure was also applied for the immobilization of G2 onto silica to get Sil-G2. 2.3. Proton NMR 1H
NMR were recorded with a JEOL JNM-LA400 (Japan) instrument at 400 MHz in CDCl3 solutions at 25 ◦ C. Chemical shifts (␣) of 1 H, expressed in parts per million (ppm) with use of the internal standards Me4 Si (ı = 0.00 ppm). 2.4. FT-IR, DRIFT and elemental analysis FT-IR measurements were conducted with JASCO FT/IR-4100 (Japan). For DRIFT measurement accessory DR PRO410-M (JASCO, Japan) was used. Elemental analyses were carried out on a Yanaco CHN Corder MT-6 Apparatus (Japan). 2.5. Thermogravimetric analysis (TGA) TGA were performed on a Seiko EXSTAR 6000 TG/DTA 6300 thermobalance in static air from 30 to 900 ◦ C at a heating rate of 10 ◦ C/min.
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
151
Fig. 2. Synthesis of N ,N -dioctadecyl-N˛ -[(4-carboxybutanoyl)]-l-glutamide and N ,N -dioctadecyl-l-2-N˛ -[(4-carboxy)butanoyl]aminoadipamide-grafted silica phases (SilG1 and Sil-G2).
2.6. Liquid chromatography Liquid chromatography was carried out with methanol or a mixture of methanol/water as a mobile phase at a flow rate of 1.0 ml min−1 under isocratic condition. The chromatographic system included a JASCO 1580 pump and a JASCO MD-2010 plus UV–vis photodiode array detector. As the sensitivity of UV detector is high, 5 l sample was injected through a Reodyne Model 7725 injector having 20 l loop. The column temperature was maintained by using a column jacket having heating and cooling system. A personal computer connected to the detector and pump with ChromNAV (Ver 1.17 or later) software was used for system control and data analysis. Chromatographic grade solvent was used to prepare mobile phase solutions. The separation factor (˛) was given by the ratio of retention factors (k). The retention time of D2 O was used as the void volume (t0 ) marker (the absorption for D2 O was measured at 400 nm).
grafted amide bonded lipids on the silica surface for both of the phases. Equally important is the appearance of N H stretching (3284 cm−1 ) in the spectra for Sil-G1 and Sil-G2, providing further evidence that both of G1 and G2 was successfully grafted onto the silica surface. 3.3. Immobilization amount The TGA illustrates the distinct effect of silane functionality on bonded phases surface coverage to a greater extent than values from elemental analysis. The significant weight loss step from 300 to 900 ◦ C correlates to the decomposition of the chemically bonded organic phase onto silica surface. Typical TGA curves for the bare silica, Sil-APS, Sil-G1, and Sil-G2 are depicted in Fig. 4. The weight retention profile of silica and Sil-APS reached a plateau at about 110 ◦ C (drying period), indicating the removal of surface
3. Results and discussion 3.1. Synthesis of G1 and G2 and immobilization onto silica Molecular-gel forming compounds (G1 and G2) were designed, synthesized, immobilized onto the silica surface (Fig. 2), and used as packed column for liquid chromatography. The functional groups were oriented in such a way that l-2-aminoadipic acid derivative (G2) believed to be more oriented than in l-glutamic acid derivative (G1) due to stronger intermolecular H-bonding was anticipated for G2 compared to G1 (Fig. 1). 3.2. DRIFT analysis The grafting of organic molecules onto silica can be confirmed by DRIFT, elemental analysis, and thermogravimentric analysis (TGA). As shown in Fig. 3, a group of peaks at 2925 and 2853 cm−1 , respectively, were attributed to the C H bond stretching of the long alkyl chain for both of the stationary phases (Sil-G1 and Sil-G2). The intense bands at 1644 and 1549 cm−1 , indicating the presence of
Fig. 3. DRIFT spectra of bare silica, Sil-APS, Sil-G1, and Sil-G2.
152
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
analysis data strongly corresponds to the results obtained by TGA measurements. 3.4. Chromatographic evaluation of Sil-G1 and Sil-G2
Fig. 4. TGA spectra of bare silica, Sil-APS, Sil-G1, and Sil-G2.
water. After the thermal degradation of the APS the weight of the sample was constant from 600 to 900 ◦ C. A plateau in the weight retention curve of Sil-G1 and Sil-G2 was also observed as the temperature reached 600 ◦ C, confirming that there is no organic content remaining on the silica at 900 ◦ C. Using the TGA curve of silica as reference, the weight of the immobilized APS can be calculated as 8.49 wt.% of the total mass. Similarly, TGA revealed that 14.65 wt.% G1 and 13.82 wt.% G2 were grafted on the silica surface, using the weight retention of Sil-APS as a reference at 900 ◦ C. The immobilization amount in Sil-G1 was slightly higher than in Sil-G2. The grafted amount of organic phase onto silica can also be determined from the %C obtained from elemental analysis data. We found 8.56% C, 2.48% H, and 2.78% N for Sil-APS, 20.66% C, 3.88% H, and 3.24% N for Sil-G1, and 20.22% C, 3.61% H, and 3.12% N for Sil-G2, respectively from elemental analysis. The surface coverage of silica particles were determined to be 4.82, 0.88, and 0.82 mol m−2 , respectively according to previously reported method [28,29]. The percentage of carbon of Sil-APS was deducted from the values of Sil-G1 and Sil-G2 during the calculation of surface coverage to get the actual coverage by G1 and G2. The grafting amount from the elemental
To assess the selectivity differences between Sil-G1 and Sil-G2, evaluation was carried out with the separation of shapeconstrained isomers of PAHs and tocopherols. In this work, chromatographic results were analyzed and compared between only our synthesized two slightly different organic phases (SilG1 and Sil-G2). Because, previously we reported that dialkyl l-glutamide-derived stationary phases showed much better selectivity than commonly used important commercial columns (monomeric and polymeric ODSs and C30 ) [18–21]. The selectivity differences of the two columns were carried out with the separation of PAHs and the selectivity for two-dimensional shape was studied with a molecular-shape descriptor, such as molecular length-to-breadth (L/B) ratio. The chromatograms for the fourring PAHs by the Sil-G1 and Sil-G2 phases are shown in Fig. 5. If we analyze the separation of two nonlinear and linear PAHs benzo[a]anthracene (2) and naphthacene (4), which have the same number of carbon atoms and -electrons but differ only in their molecular-shape such as the length and aspect ratio (L/B = 1.60 and L/B = 1.90, respectively), Sil-G2 showed much better linearity selectivity (˛4/2 = 4.50) than Sil-G1 (˛4/2 = 2.09) with the same chromatographic conditions. At the same time, better separation was also observed with Sil-G2 than Sil-G1 for benzo[a]anthracene and chrysene (Fig. 5). To evaluate the shape recognition and interaction mechanism of Sil-G1 and Sil-G2, other chromatographic tests have been established. For example, the selectivity for oterphenyl (L/B = 1.11) and triphenylene (L/B = 1.12) probes with similar L/B values, number of carbon atoms, and -electrons but different molecular planarity, was introduced for the evaluation of the planarity recognition capability of C18 phases by Tanaka et al. [30] and Jinno et al. [31]. Here, we have used another mixture of o-(nonplanar), m-(nonplanar), and p-terphenyl (almost planar) isomers and triphenylene (planar) to investigate both planarity and linearity selectivity with Sil-G1 and Sil-G2 (Fig. 6). Dramatic differences in selectivity were observed for the Sil-G2 and Sil-G1 phases. In Sil-G1, the mixture was separated successfully in the order of o-, m-, and p-terphenyls and triphenylene; however, in
Fig. 5. Separation of four ring PAHs isomers (1, triphenylene; 2, benzo[a]anthracene; 3, chrycene; 4, naphthacene) on Sil-G1 and Sil-G2 phases. Mobile phase: methanol–water (90:10), column temperature 30 ◦ C, flow rate: 1.0 ml min−1 . UV detection: 275 nm.
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
153
Table 1 Separation factors (␣) of PAHs and their isomers for Sil-G1 and G2 phases. Organic phases
Dibenz[a,h]anthracene/ trans-/cis-Stilbene dibenz[a,c]anthracene
Coronene/ hexahelicene
Naphthacene/ Chrycene/ Triphenylene/ benzo[a]anthracene benzo[a]anthracene o-terphenyl
p-/-m-Terphenyl
Sil-G1 Sil-G2
1.79 1.46
16.3 19.6
2.09 4.50
2.03 4.31
2.92 2.06
Sil-G2, the retention order was changed so that p-terphenyl was eluted after triphenylene, which again suggesting very high linearity/slenderness selectivity of Sil-G2 compared to Sil-G1 phase. However, planarity selectivity (␣triphenylene/o-terphenyl ) was slightly higher for Sil-G1 compared to Sil-G2 (Fig. 6). For better assessment of linearity and planarity selectivity of Sil-G1 and Sil-G2, more results have been included in Table 1. Therefore, the orientation of the functional groups or interaction sites of Sil-G1 and Sil-G2 has much effect on linearity and planarity selectivity toward PAHs. From the above findings of molecular planarity and linearity selectivity of Sil-G1 and Sil-G2, we tried to separate a shapeconstrained vitamin E isomers (␣-, -, ␥-, and ␦-isomers). Among the ␣-,  -, ␥-, and ␦-isomers of tocopherol (Vitamin E), the separation of - and ␥-tocopherol has long presented a special challenge. The baseline separation of these two homologues especially for and ␥-isomers could not be achieved in conventional RP-HPLC and is completely impossible on ODS phases [32–34]. Previously, we reported the separation of tocopherol isomers by poly(octadecyl acrylate-alt-N-octadecylmaleimide)-grafted silica (Sil-poly(ODAalt-OMI)) and poly(octadecyl acrylate-alt-N-octadecyl-ˇ-alanyl maleimide)-silica hybrid (Sil-alt-P) [35,36]. Fig. 7 shows the separation of tocopherol isomers with Sil-G1 and Sil-G2. Slightly better separation of - and ␥-isomers was observed with Sil-G1 compared to Sil-G2, which may be related to more planarity selectivity of the former phase than the later. 3.5. Molecular-recognition mechanism We have reported about the interaction mechanism of the gel-forming compound-grafted silica phase for the separation of
Fig. 6. Separation of o-, m-, p-terphenyl, and triphenylene on Sil-G1 and Sil-G2 phases. Chromatographic conditions are the same as Fig. 7.
1.11 1.48
4.77 3.98
shape-constrained isomers [18,20,21]. In all the cases, our phases showed better selectivities and different interaction mechanism than commercial alkyl phases (ODSs and C30 ). Orientation of the functional groups (interaction sites) are the driving forces for multiple interactions with solutes and eventually obtained high selectivities on gel-forming compound-grafted silica phases [18,20,21]. Herein, we must need to discuss about the tunability of the selectivity due to slight changes in the chemical structure of the organic phase. Generally, the molecular-shape selectivity in the ODSs and C30 phases increases with increasing carbon loading, alkyl chain length, decreasing column temperature as we mentioned earlier [9–17]. All of these have been attributed by slight increase of alkyl chain ordering as well as hydrophobic interactions. Unlike ODS phases, the higher selectivity of Sil-G1 and Sil-G2 cannot be explained by the hydrophobic interactions with the solute molecules. To explain the difference of interaction mechanisms between Sil-G1 and Sil-G2, we propose the direct interaction of solutes with ordered functional groups (carbonyl groups). The gel-forming ability or chirally ordered functional groups of amino adipic acid derived lipid (G2) was expected to be higher than glutamic acid derived lipid (G1) as illustrated in Fig. 1. Therefore, very high linearity selectivity of Sil-G2 can be explained by multiple carbonyl– interactions with higher interaction aspect ratio with the solutes due to more oriented carbonyl groups compared to Sil-G1. As the effect of temperature has great effect on molecularrecognition [14,37–39], we have also observed the temperature dependencies of m- and p-terphenyl on Sil-G1 and Sil-G2 phase and found much higher selectivity with the later phase than the former (Fig. 8). Phase transition was not observed with increasing temperature. This result may be due to reason of more ordering carbonyl groups could be maintained at variable temperature in SilG2 compared to Sil-G1. On the other hand, slight higher planarity selectivity of Sil-G1 compared to Sil-G2 may be due to less accessibility of nonplanar compounds into the interaction sites on Sil-G1.
Fig. 7. Separation of tocopherols on Sil-G1 and Sil-G2 phases. Mobile phase: methanol–water (85:15), column temperature 40 ◦ C, flow rate: 1.0 ml min−1 . UV detection: 285 nm.
154
A.K. Mallik et al. / J. Chromatogr. A 1324 (2014) 149–154
the other hand, Sil-G1 showed slightly higher molecular-planarity selectivity compared to Sil-G2. Therefore, the orientation of the functional groups can change the selectivity of the organic phases. Additionally, both of the phases showed very high molecular-shape selectivity compared to the commercially available alkyl phases. Acknowledgement This research was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. References [1] [2] [3] [4] [5] [6] Fig. 8. Temperature dependencies of the separation factors between m- and pterphenyls on Sil-G1 and Sil-G2 phases. Mobile phase: methanol (100%), flow rate: 1.0 ml min−1 . UV detection: 275 nm.
[7] [8]
In the interaction sites of both of the phases carbonyl groups are supposed to be the driving force for the high shape selectivity by multiple carbonyl– interactions with the solutes. A carbonyl– interaction has been discussed in our previous calculation works [40]. This interaction in a model complex of HCHO-benzene is much stronger (1.87 kcal mol−1 ) than a CH4 –benzene interaction (0.53 kcal mol−1 ) and a plane-to-plane interaction between two benzenes (0.49 kcal mol−1 ) [40]. On the basis of this calculation, when acetone with a carbonyl group was added to the mobile phase, both the retention time and selectivity decreased remarkably. This indicates that acetone functions as an inhibitor for a carbonyl– interaction [18]. Selectivity enhancement through a carbonyl– interaction has been also discussed in relation to homopolymers and copolymers from octadecyl acrylate (ODAn ). When ODAn was grafted onto silica and then evaluated by the retention time of PAHs, the resultant selectivity was higher, especially at crystalline temperatures of ODAn than for ODS columns [41–45]. This is attributed to the fact that ODAn has a crystal-to-isotropic phase transition, so that a multiple carbonyl– interaction with PAHs becomes possible through the ordering of carbonyl groups at lower temperatures (<30 ◦ C), where the polymers are in crystalline state. High shape selectivity was also observed for the alternating copolymer-grafted silica containing ODA and octadecyl maleimide due to ordered carbonyl groups along the polymer main chain, suitable for multiple interactions with the solutes [45]. Therefore, slight changes in the orientation of the functional groups can cause a large effect on the shape selectivity for the gel-forming compound-grafted phases. 4. Conclusion Here we newly introduced dialkyl l-2-aminoadipic acid-derived gel-forming compound-grafted silica (Sil-G2) phase as a promising organic phase for the separation of shape-constrained isomers especially for the separation of molecules with slight molecular linearity difference. The properties of the phase and chromatographic results were compared with our previously reported dialkyl l-glutamide-derived organic phase (Sil-G1). Although the chemical structure of both of the phases are almost similar comparatively highly chirally ordered structure of aminoadipic acid derived organic phase showed more molecular-linearity selectivity. On
[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]
[25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45]
L.C. Sander, S.A. Wise, Anal. Chem. 56 (1984) 504. L.C. Sander, S.A. Wise, J. Chromatogr. 316 (1984) 163. L.C. Sander, S.A. Wise, Anal. Chem. 59 (1987) 2309. J.G. Dorsey, W.T. Cooper, B.A. Siles, J.P. Foley, H.G. Barth, Anal. Chem. 70 (1998) 591R. K. Jinno, T. Ibuki, N. Tanaka, M. Okamoto, J.C. Fetzer, W.R. Biggs, P.R. Griffiths, J.M. Olinger, J. Chromatogr. 461 (1989) 209. K. Jinno, T. Nagoshi, N. Tanaka, M. Okamoto, J.C. Fetzer, W.R. Biggs, J. Chromatogr. 392 (1987) 75. K. Jinno, S. Shimura, N. Tanaka, K. Kimata, J.C. Fetzer, W.R. Biggs, Chromatographia 27 (1989) 285. L.C. Sander, S.A. Wise, in: R.M. Smith (Ed.), Retention and Selectivity Studies in HPLC, Elsevier, Amsterdam, 1994, p. 337. L.C. Sander, M. Pursch, S.A. Wise, Anal. Chem. 71 (1999) 4821. K.B. Sentell, J.G. Dorsey, Anal. Chem. 61 (1989) 930. A.K. Mallik, M.M. Rahman, M. Czaun, M. Takafuji, H. Ihara, J. Chromatogr. A 1187 (2008) 119. A.K. Mallik, M.M. Rahman, M. Czaun, M. Takafuji, H. Ihara, Chem. Lett. 36 (2007) 1460. N. Tanaka, K. Sakagami, M. Araki, J. Chromatogr. 199 (1980) 327. Y. Saito, K. Jinno, J.J. Pesek, Y.L. Chen, G. Luehr, J. Archer, J.C. Fetzer, W.R. Biggs, Chromatographia 38 (1994) 295. M. Kühnle, V. Friebolin, K. Albert, C.A. Rimmer, K.A. Lippa, L.C. Sander, Anal. Chem. 81 (2009) 10136. L.C. Sander, K.E. Sharpless, N.E. Craft, S.A. Wise, Anal. Chem. 66 (1994) 1667. L.C. Sander, K.E. Sharpless, M. Pursch, J. Chromatogr. A 880 (2000) 189. M.M. Rahman, M. Takafuji, H.R. Ansarian, H. Ihara, Anal. Chem. 77 (2005) 6671. M.M. Rahman, M. Takafuji, H. Ihara, J. Chromatogr. A 1119 (2006) 105. A.K. Mallik, H. Qiu, T. Sawada, M. Takafuji, H. Ihara, Chem. Commun. 47 (2011) 10341. A.K. Mallik, H. Qiu, T. Sawada, M. Takafuji, H. Ihara, Anal. Chem. 84 (2012) 6577. P. Terech, R.G. Weiss, Chem. Rev. 7 (1997) 3133. J.H. van Esch, B.L. Feringa, Angew. Chem. Int. Ed. 39 (2000) 2263. H. Ihara, M. Takafuji, T. Sakurai, in: H.S. Nalwa (Ed.), Encyclopedia of Nanoscience and Nanotechnology, vol. 9, American Scientific Publishers, Stevenson Ranch, CA, 2004, p. 473. S. Ong, H. Liu, X. Qiu, G. Bhat, C. Pidgeon, Anal. Chem. 67 (1995) 755. H. Liu, S. Ong, L. Glunz, C. Pidgeon, Anal. Chem. 67 (1995) 3550. H. Hachisako, N. Ryu, R. Murakami, Org. Biomol. Chem. 7 (2009) 2327. G.F. Berendsen, K.A. Pikaart, L. de Galan, J. Chromatogr. 3 (1980) 1437. A.K. Mallik, H. Qiu, M. Takafuji, H. Ihara, Anal. Bioanal. Chem. 404 (2012) 229. N. Tanaka, Y. Tokuda, K. Iwaguchi, J. Araki, J. Chromatogr. 239 (1982) 761. K. Jinno, K. Yamamoto, H. Nagashima, T. Ueda, K. Itoh, J. Chromatogr. 517 (1990) 193. S. Strohschein, M. Pursch, D. Lubda, K. Albert, Anal. Chem. 70 (1998) 13. C.W. Henry, C.A. Fortier, I.M. Warner, Anal. Chem. 73 (2001) 6077. S.L. Abidi, J. Chromatogr. A 881 (2000) 197. A.K. Mallik, T. Sawada, M. Takafuji, H. Ihara, Anal. Chem. 82 (2010) 3320. A.K. Mallik, H. Qiu, M. Takafuji, H. Ihara, Anal. Methods 3 (2011) 1277. M. Pursch, L.C. Sander, H.-J. Egelhaaf, M. Raitza, S.A. Wise, D. Oelkrug, K. Albert, J. Am. Chem. Soc. 121 (1999) 3201. A.K. Mallik, M. Takafuji, H. Ihara, Eur. Polym. J. 45 (2009) 1811. A. Shundo, A.K. Mallik, H. Nakashima, M. Takafuji, S. Nagaoka, M.A.J. Chowdhury, H. Ihara, Chem. Lett. 39 (2010) 844. S. Sakaki, K. Kato, T. Miyazaki, K. Ohkubo, H. Ihara, C. Hirayama, J. Chem. Soc. Faraday Trans. 89 (1993) 659. M.A.J. Chowdhury, R.I. Boyson, H. Ihara, M.T.W. Hearn, J. Phys. Chem. B 106 (2002) 11936. H. Ihara, Y. Goto, T. Sakurai, M. Takafuji, T. Sagawa, S. Nagaoka, Chem. Lett. (2001) 1252. H. Ihara, T. Sagawa, Y. Goto, S. Nagaoka, Polymer 40 (1999) 2555. A. Shundo, M. Fukui, M. Takafuji, K. Akasaka, H. Ohrui, D. Berek, H. Ihara, Anal. Sci. 23 (2007) 1. A.K. Mallik, H. Qiu, T. Sawada, M. Takafuji, H. Ihara, J. Chromatogr. A 1232 (2012) 183.