Article pubs.acs.org/ac
Design of C18 Organic Phases with Multiple Embedded Polar Groups for Ultraversatile Applications with Ultrahigh Selectivity Abul K. Mallik,*,† Hongdeng Qiu,∥ Tomohiro Oishi,‡ Yutaka Kuwahara,†,§ Makoto Takafuji,†,§ and Hirotaka Ihara*,†,§ †
Department of Applied Chemistry and Biochemistry and ‡Technical Division, Faculty of Engineering, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan § Kumamoto Institute for Photo-Electro Organics (Phoenics), Kumamoto 862-0901, Japan ∥ Key Laboratory of Chemistry of Northwestern Plant Resources and Key Laboratory for Natural Medicine of Gansu Province, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China S Supporting Information *
ABSTRACT: For the first time, we synthesized multiple embedded polar groups (EPGs) containing linear C18 organic phases. The new materials were characterized by elemental analysis, IR spectroscopy, 1H NMR, diffuse reflectance infrared Fourier transform (DRIFT), solid-state 13C cross-polarization magic angle spinning (CP/MAS) NMR, suspended-state 1H NMR, and differential scanning calorimetry (DSC). 29Si CP/MAS NMR was carried out to investigate the degree of cross-linking of the silane and silane functionality of the modified silica. Solid-state 13C CP/MAS NMR and suspended-state 1H NMR spectroscopy indicated a higher alkyl chain order for the phase containing four EPGs than for the phase with three EPGs. To correlate the NMR results with temperaturedependent chromatographic studies, standard reference materials (SRM 869b and SRM 1647e), a column selectivity test mixture for liquid chromatography was employed. A single EPG containing the C18 phase was also prepared in a similar manner to be used as a reference column especially for the separation of basic and polar compounds in reversed-phase liquid chromatography (RPLC) and hydrophilic interaction liquid chromatography (HILIC), respectively. Detailed chromatographic characterization of the new phases was performed in terms of their surface coverage, hydrophobic selectivity, shape selectivity, hydrogen bonding capacity, and ion-exchange capacity at pH 2.7 and 7.6 for RPLC as well as their hydrophilicity, the selectivity for hydrophilic−hydrophobic substituents, the selectivity for the region and configurational differences in hydrophilic substituents, the evaluation of electrostatic interactions, and the evaluation of the acidic−basic nature for HILIC-mode separation. Furthermore, peak shapes for the basic analytes propranolol and amitriptyline were studied as a function of the number of EPGs on the C18 phases in the RPLC. The chromatographic performance of multiple EPGs containing C18 HILIC phases is illustrated by the separation of sulfa drugs, β-blockers, xanthines, nucleic acid bases, nucleosides, and water-soluble vitamins. Both of the phases showed the best performance for the separation of shape-constrained isomers, nonpolar, polar, and basic compounds in RPLC- and HILIC-mode separation of sulfa drugs, and other polar and basic analytes compared to the conventional alkyl phases with and without embedded polar groups and HILIC phases. Surprisingly, one phase would be able to serve the performance of three different types of phases with very high selectivity, and we named this phase the “smart phase”. Versatile applications with a single column will reduce the column purchasing cost for the analyst as well as achieve high separation, which is challenging with the commercially available columns.
H
which are important but challenging due to their similar molecular shape. 6−9 Stationary phases for RPLC, also developed on the basis of shape selectivity, depend on the stationary-phase bonding density,6,10−12 alkyl-phase chain length,13−16 column temperature,17−19 or architecture and dynamics of the organic phases.20−24 However, the separation of basic and highly polar analytes by RPLC using alkyl phases
igh-performance liquid chromatography (HPLC) as a separation technique has resulted in significant progress in the analytical sciences since the 1970s, and its advancement has relied mostly on the development of new stationary phases. For example, reversed-phase (RP) separations, using hydrophobic or n-alkyl-type stationary phases, have greatly increased the application of HPLC due to its versatility and to the constant development of new stationary phases and instrumentation.1−5 Generally, two types of n-alkyl chromatographic sorbent can be distinguished on the basis of bonding chemistry (monomeric and polymeric C18). Polymeric C18 phases are more effective for the separation of a certain class of isomers, © XXXX American Chemical Society
Received: February 17, 2015 Accepted: June 4, 2015
A
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
of Standards and Technology (NIST), Gaithersburg, MD, USA). PAHs were purchased from TCI, Tokyo, Japan. The nonpolar, polar, and basic compound set (uracil, propranolol, butyl paraben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene) was obtained from Wako and TCI. Sulfa drugs (sulfanilamide, sulfamethoxypyridazine, sulfadiazine, sulfamethoxazole, sulfactamide, sulfamonomethoxine, and sulfaquinoxaline) were obtained from TCI or from SigmaAldrich (St. Louis, MO, USA). β-Blockers (Carvedilol, Metoprolol, Pindolol, Acebutolol, Sotalol, Labetalol, Atenolol, and Nadolol), xanthines (caffeine, 7-β-hydroxypropyltheophylline, dyphylline, xanthine, and hyphoxynthine), bases and nucleosides (thymine, uracil, 4,6-diaminopyrimidine, uridine, adenosine, cytosine, and cytidine), and water-soluble vitamins (nicotinamide, pyridoxine, nicotinic acid, and riboflavin) were purchased either from Sigma-Aldrich, Wako, or from TCI. Other nucleosides including 5-methyluridine, sodium ptoluenesulfonate, theobromine, theophylline, 2′-deoxyuridine, N,N,N-trimethylphenylammonium chloride, and vidarabine were obtained from TCI. 2′-Deoxyguanosine, 3′-deoxyguanosine, 4-nitrophenyl α-D-glucopyranoside, and 4-nitrophenyl βD-glucopyranoside were purchased from Sigma-Aldrich. 17βEstradiaol was obtained from Sigma-Aldrich. 17α-Estradiol was purchased from Dr. Ehrennstorfer GmbH (Augsburg, Germany). The multiple EPGs-containing C18 phases (Sil-MEPG-C18-1 and Sil-MEPG-C18-2) and the single EPG-containing C18 phase (Sil-SEPG-C18) were synthesized, characterized, and packed into stainless steel columns (150 mm × 4.6 mm i.d.). YMC silica (YMC SIL-120-S5) with a 5 μm diameter and a 12 nm pore size was used. In contrast, we used a commercial amide group (single) embedded C18 column (Ascentis RP-amide) as a reference column from Sigma-Aldrich. Our synthesized SilSEPG-C18 column was also used as a reference column. Many other reference columns were used in this work for both RPand HILIC-mode separation. Commercial monomeric and polymeric C18 columns (Inertsil ODS-3 and Inertsil ODS-P, respectively) were obtained from GL Sciences, Tokyo, Japan. Additionally, another polymeric C18 column (Shodex C18 P4D), purchased from Shodex, Tokyo, Japan, was also used as a reference column for better comparison. Longer alkyl phase C30 (Develosil C30-UG-5) was purchased from Nomura Chemical Co., Ltd., Tokyo, Japan. Commercial HILIC columns, for example, silica (Inertsil SIL), diol (Inertsil HILIC-Diol), and amide (Inertsil HILIC-Amide), were obtained from GL Sciences. Properties of the commercial reference columns are given in the Supporting Information (Table S1). Multiple EPG-containing C18 (N′-octadecyl-Nα-[(4-carboxybutanoyl)-2β-alanineamide (6) and N′-octadecyl-Nα-[(4-carboxybutanoyl)-β-alanineamide (7)) were synthesized using βalanine through alkylation, debenzyloxycarbonylation, and ringopening reactions with glutaric anhydride and then immobilized onto silica from 3-aminopropyltrimethoxysilane (APS)modified silica (Sil-APS), named Sil-MEPG-C18-1 and SilMEPG-C18-2, respectively (Supporting Information, Scheme S1). To obtain single EPG-C18-grafted silica (Sil-SEPG-C18), nonadecanoic acid was similarly grafted onto silica (Supporting Information, Scheme S1). The detailed synthetic procedures for 6 and 7, immobilization onto silica, and other experimental details are given in the Supporting Information. Liquid Chromatography. The chromatographic system consisted of a Gulliver PU-1580 intelligent HPLC pump with a Rheodyne sample injector. A JASCO multiwavelength UV
continues to challenge chromatographers due to peak tailing, phase collapse in highly aqueous environments, and poor retention and selectivity.25−28 To solve these problems, analysts have developed many types of single EPG-containing alkyl phases.29−31 However, these single EPG-containing alkyl phases were not suitable for the separation of shape-constrained isomers of polycyclic aromatic hydrocarbons (PAHs) and highly polar compounds.32 Moreover, an alternative and rival technique to RPLC for separating polar compounds was first reported by Alpert in 1990 and is called hydrophilic interaction chromatography (HILIC).33 Different types of stationary phases have been developed, including bare silica, amine, amide, diol, and cyclodextrin phases, for HILIC separations of polar analytes.34−38 Of course, these phases are not able to separate nonpolar and shape-constrained isomers. Furthermore, mixed-mode RPLC/HILIC stationary phases have also been synthesized. However, these phases do not show very high selectivity either in RPLC- or in HILIC-mode when compared to conventional single RPLC and HILIC phases.39,40 In this work, we report the first instance of multiple EPGcontaining C18 organic phases for the separation of versatile analytes including shape-constrained isomers, nonpolar, polar, and basic compounds with ultrahigh selectivity both in RPLCand HILIC-mode separations in a single column. In short, a single column can do the job of three different types of columns without losing the level of performance. Our designed novel materials (g and h) produced by embedding multiple polar groups differ from typical previous C18 (a and b),6−9 C30 (c),13−16 HILIC (d),34−38 mixed-mode (e),39,40 and EPG C18 (f)29−31 phase materials, as shown in Scheme 1. Our successful Scheme 1. Schematic Illustrations of Different Stationary Phases: (a) Monomeric C18, (b) Polymeric C18), and (c) C30 RPLC; (d) HILIC; (e) Mixed-Mode; (f) EPG C18; (g, h) Novel Multiple EPG C18 (Sil-MEPG-C18-2 and Sil-MEPGC18-1, Respectively) Materials Designed and Synthesized in This Study
strategy utilizes two concepts: a balanced combination of a highly hydrophilic moiety and a highly hydrophobic moiety in a single compound to achieve a synergistic effect of both moieties as well as individual effects in chromatographic separation.
■
EXPERIMENTAL SECTION Materials and Reagents. Stearylamine, nonadecanoic acid, diethylphosphorocyanidate (DPEC, peptide synthesis reagent), triethylamine (TEA), and β-alanine were purchased from Wako (Tokyo, Japan) and used without further purification. Standard reference material SRM 869b (column selectivity test mixture for liquid chromatography) and SRM 1647e (priority pollutant polycyclic aromatic hydrocarbons (PAHs)) were obtained from the Standard Reference Materials Program (National Institute B
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry detector MD 1510 plus was used. The chromatography was performed under isocratic elution conditions with a flow rate of 1 mL min−1 for most of the analysis. The column temperature was maintained by using a column jacket having a 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. Chromatographicgrade solvent was used to prepare mobile-phase solutions. The retention factor (k) was determined by (te − t0)/t0, where te was the retention time of the samples. The first disturbance of the baseline on the injection of methanol was used as the dead time marker (t0).41 The separation factor (α) was given by the ratio of retention factors. In RPLC, methanol was used as a mobile phase for the separation of SRM 869b and other PAHs. Separation of SRM 1647e was carried out using a 95:5 (volume fraction) methanol/water mobile phase at 40 °C. The other chromatographic conditions for the Tanaka RPLC characterization of the phases were as reported previously.41,42 In this protocol, six variables reflecting different chromatographic ion properties such as the retention factor for pentylbenzene, kPB; hydrophobic selectivity, α(CH2); the retention factor ratio between pentylbenzene and butylbenzene; shape selectivity, αT/O; the retention factor ratio between triphenylene and oterphenyl; hydrogen bonding capacity, αC/P; the retention factor ratio between caffeine and phenol; the total ion-exchange capacity at pH 7.6, αA/P; the retention factor ratio between benzylamine and phenol; acidic ion-exchange capacity at pH 2.7, αA/P; and the retention factor ratio between benzylamine and phenol were used for the characterization. A test mixture of nonpolar, polar, and basic compounds (uracil, propranolol, butylparaben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene) was separated in 45:55, 65:35, 75:25, and 75:25, methanol/20 mM KH2PO4/K2HPO4 at pH 7.00 for SilMEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases at 25 °C, respectively. For the separation of polar and basic analytes in the HILIC-mode, an ammonium acetate (NH4Ac) buffer and acetonitrile (ACN) mixture were used. In HILIC, most of the experiments were carried out at room temperature. The other chromatographic conditions for the HILIC characterization of the phases (and variables reflecting different chromatographic properties) were used as described by Kawachi et al.43 The detailed chromatographic conditions are also given in the Supporting Information.
Scheme 2. Chemical Structures of the Multiple Embedded Polar Groups (EPGs) Containing C18 Phases, Indicating Structural Properties and Possible Interaction Sites
2).13−16,44−47 Another column was prepared in a similar manner with a single EPG-containing C18 (Sil-SEPG-C18) to be used as a reference column, especially for the comparison and interaction mechanism elucidations of polar compound separation both in RPLC- and HILIC-mode (Supporting Information, Scheme S1). The materials were completely characterized by elemental analysis, thermogravimetric analysis (TGA), diffuse reflectance infrared Fourier transform (DRIFT), differential scanning calorimetry (DSC), solid-state 13 C cross-polarization (CP)48,49 magic angle spinning (MAS),50 and suspended-state 1H NMR51,52 and 29Si CP/MAS NMR53 spectroscopy before packing into stainless steel columns for chromatographic evaluation (see the Supporting Information for details). From the elemental analysis results, the surface coverage of APS, MEPG-C18-1, MEPG-C18-2, and SEPG-C18 to the silica surface was calculated according to the previously reported method6,12 as 7.72, 1.31, 1.55, and 2.19 μmol m−2 for Sil-APS, Sil-MEPG-C18-1, Sil-MEPG-C18-2, and Sil-SEPG-C18, respectively (Supporting Information, Table S2). The organic load in the new phases was also confirmed by the TGA measurements and showed agreement with elemental analysis results (Supporting Information, Table S2 and Figure S3). Additionally, grafting of organic molecules onto a silica surface was confirmed by DRIFT measurement (Supporting Information, Figure S4). The solid-state 13C CP/MAS NMR and suspended-state 1H NMR measurements were carried out at different temperatures from 20 to 50 °C to investigate the
■
RESULTS AND DISCUSSION As was mentioned earlier, multiple EPG-containing C18 compounds, MEPG-C18-1 (6) and MEPG-C18-2 (7), were synthesized using β-alanine through alkylation, debenzyloxycarbonylation, and ring-opening reactions with glutaric anhydride and then immobilized onto silica from 3-aminopropyltrimethoxysilane (APS)-modified silica (Sil-APS), named Sil-MEPG-C18-1 and Sil-MEPG-C18-2, respectively (Supporting Information, Scheme S1). Four and three EPG (amide)containing C18 phases (Sil-MEPG-C18-1 and Sil-MEPG-C18-2, respectively) were synthesized. The material contains two important units with five functions (Scheme 2). Four amide groups provide hydrophilicity and wettability, multiple carbonyl−π interaction sources,22−24 multiple H-bonding sources,44−47 and increase the chain length together with C18. However, rigid C18 provides hydrophobicity to the organic phase. H-bonding between polar groups and increased chain length leads to a more ordered alkyl chain structure (Scheme C
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
phase collapse was observed for both of the phases, indicating ordered alkyl chains (Supporting Information, Figure S6).52 29 Si CP/MAS NMR was carried out for the Sil-MEPG-C18-1 and Sil-MEPG-C18-2 phases (Supporting Information, Figure S7) attributed to a very high degree of cross-linking, even higher than the best C22 or C18 bonded phases.20,21 The results suggested a smaller number of free OH groups on the surface, which leads to fewer silanophilic interactions in HPLC.20,53 The novel phases were very suitable for the separation of shape-constrained isomers20−24 and neutral, polar, and basic compounds29−31 in RPLC and polar compounds34−38 in HILIC (i.e., three in one) compared to the most commonly used commercial RP, EPG, and HILIC columns. First, analysis was carried out with the separation of standard reference materials (SRMs) SRM869b and SRM1647e, as a column (shape) selectivity test mixture for liquid chromatography.54 SRM869b consists of 1,2:3,4:5,6:7,8-tetrabenzonaphthalene (TBN), phenanthro[3,4-c]phenanthrene (PhPh), and benzo[a]pyrene (BaP) (Supporting Information, Figure S8) and was developed after evaluation of over 100 PAH solutes. The three solutes (PAHs) selected provided the most sensitive indication of changes in selectivity due to solute shape. The ratio (α) of retention factors (k) for TBN and BaP (i.e., αTBN/BaP = kTBN/ kBaP) provides a measure of shape selectivity that is useful for column intercomparisons, and columns exhibit a high degree of shape selectivity (a lower value indicates higher shape selectivity); values typically fall within the range of 0.3− 1.0.55−57 We observed selectivity coefficients for the Sil-MEPGC18-1 phase ranging from 0.16 (at 10 °C) to 0.56 (at 50 °C), which exhibited very high shape selectivity compared to the most commonly used commercial RP columns (monomeric and polymeric C18 (1.93−1.95 and 0.28−0.76, respectively), EPG C18 (1.29−1.54), and C30 (0.51−1.81)) (Figure 2).6−24,32
conformations and mobility of the long alkyl chains, respectively, of the Sil-MEPG-C18-1 and Sil-MEPG-C18-2 phases. 13C CP/MAS NMR spectroscopy revealed highly ordered (trans conformation) alkyl chains in the Sil-MEPGC18-1 phase (Figure 1a) compared to the Sil-MEPG-C18-2
Figure 1. Partial solid-state 13C CP/MAS NMR spectra of Sil-MEPGC18-1 (a) and Sil-MEPG-C18-2 (b) at variable temperatures. Figure 2. Phase selectivity (αTBN/BaP = kTBN/kBaP) plotted as a function of temperature for Sil-MEPG-C18-1, Sil-MEPG-C18-2, and most commonly used RP commercial columns.
phase (Figure 1b) irrespective of temperature,17−19 polymeric chains,1,8 alkyl chain length,13−16 and surface coverage (1.31 μmol m−2).6,10−12 The degree of alkyl chain ordering increases with increasing alkyl chain length, grafting density/surface coverage, and lower column temperature. Highly ordered alkyl chains on the new phases may be due to embedded multiple polar groups (amide) and H-bonding among them.31,44−47 Similar results were also obtained by suspended-state 1H NMR in methanol (Supporting Information, Figure S5a,b), indicating that the mobility of the alkyl chains is low even at higher temperature and the results agree with the results of 13C CP/ MAS NMR spectroscopy.51,52 Moreover, DSC of the new phases was conducted in methanol, and no phase transition or
Figure 3 shows the comparative chromatogram for the separation of SRM869b in the RP-mode. Sil-MEPG-C18-1 was also able to separate other important standard PAH mixtures (SRM1647e; 16 PAHs are listed as priority pollutants by the EPA) into 16 peaks (Supporting Information, Figures S9 and S10a) within 27 min in isocratic elution for the first time, which represented one of the most important challenges.5,22,24 However, polymeric C18 phases (Shodex C18P-4D and Inertsil ODS-P) were not able to separate the standard PAH solution D
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Furthermore, Sil-MEPG-C18-1 and Sil-MEPG-C18-2 showed ultrahigh selectivity for other PAHs and their isomers in comparison with commercial RP columns, especially αcoronene/hexahelicene = 28.4 and 21.0, respectively (Supporting Information, Table S3). Additionally, to have a fair comparison, all parameters, such as capacity, hydrophobicity, steric or shape selectivity, silanol capacity, and ion-exchange capacities, at high and low pH were compared according to the Tanaka characterization protocol (Table 1).60 The Tanaka characterization protocol is a wellestablished approach that has been favored by academic groups61−64 and many stationary-phase manufacturers such as ThermoHypersil-Keystone, Merck, and Phenomenex to assess their phases. EPG phases were originally developed to reduce the peak tailing of basic analytes.29−31,65 Figure 5 shows the separation of
Figure 3. Chromatograms for the separation of SRM869b at 35 °C on Sil-MEPG-C18-1, Sil-MEPG-C18-2, and most commonly used RP commercial columns.
into 16 peaks under the same chromatographic conditions (Supporting Information, Figure S10b). Although Inertsil ODS-P column (highest carbon loading, 29% C) showed better separation ability, compounds 3 and 4 coeluted. Previously, we also reported that the SRM1647e cannot be separated in isocratic elution on C18 and C30 phases.22 The highest shape selectivity value was also observed for another shape selective probe (o-terphenyl and triphenylene, αtriphenylene/o‑terphenyl = 10.6 and 7.90, respectively, at 20 °C) defined by Jinno and Tanaka as shown in Figure 4.20,58,59
Figure 5. Chromatograms for the separation of uracil, propranolol, butylparaben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene on Sil-MEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases, respectively.
Figure 4. Phase selectivity (αtriphenylene/o‑terphenyl = ktriphenylene/ko‑terphenyl) plotted as a function of temperature for Sil-MEPG-C18-1, Sil-MEPGC18-2, and most commonly used RP commercial columns.
Table 1. Chromatographic Characterization (Tanaka Column Characterization Protocol) of the Newly Synthesized and Commercial Reference Columns in This Study (in RPLC)
Inertsil ODS-3 Ascentis RP-amide Shodex C18P-4D Develosil C30-UG-5 Inertsil ODS-P Sil-SEPG-C18 Sil-MEPG-C18-2 Sil-MEPG-C18-1
kPB
α(CH2)
αT/O
αC/P
αA/P at pH 7.5
αA/P at pH 2.7
6.85 4.82 5.10 5.42 6.21 2.84 1.26 0.86
1.46 1.42 1.45 1.45 1.47 1.43 1.43 1.34
1.30 1.65 1.98 1.58 2.47 2.48 3.65 5.91
0.47 0.20 0.40 0.59 0.76 0.15 0.23 0.37
0.12 0.09 0.26 0.16 0.46 0.09 0.08 0.23
0.47 0.92 0.44 0.54 0.40 0.99 0.99 1.00
E
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
sulfamethizole (SMT))66−68 and the most retained phase was Sil-SEPG-C18 with poor peak shapes (Supporting Information, Figure S12), possibly due to strong ion-exchange interactions with the easily accessed (due to shorter alkyl chains) residual aminopropyl groups on Sil-SEPG-C18. However, almost no separation was obtained with Ascentis RP-amide and most commonly used HILIC columns. The detailed chromatographic parameters for the separation of sulfa drugs on SilMEPG-C18-1 are given in Table S4 (Supporting Information). Again, eight β-blockers were selected and used as probe analytes to evaluate the stationary phases in HILIC-mode separation. Figure S13 (Supporting Information) gives the structures of these compounds and shows that they are hydroxylamine-containing compounds and contain at least one aromatic ring. Figure S13 (Supporting Information) also shows the chromatograms of β-blockers on Sil-MEPG-C18-1, SilMEPG-C 18-2, Sil-SEPG-C18 , and conventional EPG C 18 (Ascentis RP-amide). Conventional EPG C18 (Ascentis RPamide) was not able to function in HILIC-mode separation. Here again, the best separation was obtained with Sil-MEPGC18-1 phase. Furthermore, the new columns separated polar and hydrophilic compounds successfully, including xanthines and water-soluble vitamins (Supporting Information, Figures S14 and S15) To obtain more insight about the chromatographic characterization and interaction mechanism in HILIC-mode separation, similar to RPLC phases (e.g., by the Tanaka protocol), the new phases were characterized by the Tanaka protocol (that is supposed to be) for HILIC phases.43 In the work of Tanaka et al., characterization was carried out in terms of the degree of hydrophilicity, the selectivity for hydrophilic−hydrophobic substituents, the selectivity for the regio and configurational differences in hydrophilic substituents, the selectivity for molecular shapes, the evaluation of electrostatic interactions, and the evaluation of the acidic−basic nature of the stationary phases using nucleoside derivatives, phenyl glucoside derivatives, xanthine derivatives, sodium p-toluenesulfonate, and trimethylphenylammonium chloride as a set of test samples. The column efficiency and the peak asymmetry factors were also considered. Figure S16 (Supporting Information) shows the structure of the test samples. Detailed chromatographic data for the HILIC-mode separation are given in the Supporting Information (Tables S5, S6, S7, S8, S9, and S10). Properties of the new phases are obviously comparable with the typical HILIC phases. Among the prepared phases, Sil-MEPGC18-1 (maximum EPG-containing phase) showed the best column efficiency and peak asymmetry factors. The pH on the surface of the phases represented basic characteristics, and SilSEPG-C18 is the most basic phase. The reason for ultradiverse and ultrahigh selectivity of the MEPG C18 phases can be understood from the unique chemical structure and properties, which can work both in RPLC-mode (Supporting Information, Figure S17) and HILIC-mode (Supporting Information, Figure S18) separation. 13C CP/ MAS NMR spectroscopy revealed highly ordered alkyl chains in the Sil-MEPG-C18-1 phase (Figure 1a) compared to the SilMEPG-C18-2 phase (Figure 1b), although the phases were not polymeric type and high surface concentration (only 1.31 and 1.55 μmol/m2, respectively). In RPLC, the contribution of bonded phase rigidity and ordering to shape selectivity have been confirmed on alkyl stationary phases.6−24 However, extremely high shape selectivity (αcoronene/hexahelicene = 28.4) of our phases cannot be explained only by the ordering of the
a test sample containing a mixture of nonpolar, polar, and basic compounds (uracil, propranolol, butyl paraben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene) on Sil-MEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases, respectively. Our designed multiple EPG phases (Sil-MEPG-C18-1 and Sil-MEPG-C18-2) showed very good peak asymmetry (for basic amitriptyline and propranolol, especially for amitriptyline) compared to the reference single EPG C18 phases (Sil-SEPG-C18 and Ascentis RP-amide; Supporting Information, Figure S11). Finally, the new phases were also evaluated in HILIC-mode separation. Surprisingly, good separation of neutral polar molecules (nucleic acid bases and nucleosides) was also achieved on Sil-MEPG-C18-1 and Sil-MEPG-C18-2 phases compared to single EPG-containing C18 phases (Sil-SEPG-C18 and Ascentis RP-amide), as shown in Figure 6. The best
Figure 6. Separation of nucleic acid bases and nucleosides on SilMEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases: mobile phase, ACN−ammonium acetate buffer (10 mM in the aqueous portion, pH 6.7; 95:5, v/v); column temperature, 25 °C; flow rate, 1 mL/min.
separation was observed with the maximum EPG-containing phase (Sil-MEPG-C18-1). Very nice separation of polar ionic molecules was observed with the prepared phases compared to conventional commercial HILIC columns (silica, amide, and diol).34−38 For example, the separation of seven sulfa drugs (sulfanilamide (SNA), sulfamethoxypyridazine (SMP), sulfadiazine (SD), sulfamethoxazole (SMX), sulfectamide (STA), sulfamonomethoxine (SMM), sulfaquinoxaline (SQ), and F
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
the HILIC interaction mechanism. To prove the ion-exchange interaction mechanism, we investigated the impact of increased buffer concentration on retention factors for the neutral and ionic molecules. Figures S21 and S22 (Supporting Information) show that the retention decreases nonlinearly and linearly for the neutral and ionic molecules, respectively, on Sil-MEPG-C181. In short, the mixed-interaction mechanism has been investigated. Among the prepared phases, the maximum EPG-containing phase proved to be the best column in RPLC- and HILIC-mode separation. Therefore, multiple EPGs are important not only for RPLC but also for HILIC.
alkyl chains. From careful investigations on the retention behavior of shape-constrained isomers, we have concluded that the polymeric phases form a rigid ligand structure on the support, explained as a type of “slot-like” structure.57 Figure 7
■
CONCLUSIONS We have demonstrated the synthesis and applications of multiple EPG-containing C18 phases both in RPLC- and HILIC-mode separation. The unique stationary phases were completely characterized from a physiochemical standpoint as well as a chromatographic viewpoint. These phases showed ultrahigh selectivity and separation ability in both retention modes, compared to the most widely used conventional RP and HILIC columns. This improved performance is based on the presence of highly hydrophobic moieties and highly hydrophilic moieties in the same phase, induced by hydrogen bonding between amide groups and ordering of the alkyl chains. In RPLC, multiple EPG-containing phases are believed to form not only a slot-like structure but also to facilitate multiple carbonyl−π interactions leading to ultrahigh shape selectivity. In HILIC, multiple interactions were investigated, and the residual aminopropyl functionalities are supposed to influence ion-exchange interactions. The present results illustrate the separation of shape-constrained isomers, steroids, neutral, polar, and basic analytes with utilization of the multiple EPG-C18 phases. The concept, for which we propose the name “three-inone” or “smart phase”, has the potential to reduce column purchasing costs as well as yielding good selectivity, leading to a breakthrough in the field of analytical chemistry and materials science dealing with separation.
Figure 7. In RPLC, slot model explaining the selectivity for planar/ nonplanar solutes, including carbonyl−π interactions.
illustrates the retention model on the Sil-MEPG-C18-1 phase with a slot-like phase structure on silica gel, where a planar solute can interact (including multiple carbonyl−π interactions)5 to a greater extent than a nonplanar solute. Therefore, Sil-MEPG-C18-1 not only forms a slot-like structure but also facilitates multiple carbonyl−π interactions leading to ultrahigh shape selectivity. Additionally, new phases not only showed higher selectivity for the separation of shape-constrained isomers but also for other positional isomers (e.g., steroids)69 than any other reference columns (Supporting Information, Figure S19). Multiple H-bonding interactions are also possible due to the presence of multiple H-bonding (amide bonds or carbonyl groups) sources.31,44−47 In HILIC, a stagnant water layer forms on the surface of the hydrophilic stationary phase. Analytes partition between the stagnant water-rich solvent layer and the moving organic-rich eluent.33 The primary mechanism of retention is postulated to be partitioning of the analyte into this water-rich layer. However, adsorption, ion-exchange, dipole−dipole interaction, hydrogen bonding,70−73 π−π, and n−π interactions74 also contribute to retention on HILIC stationary phases. In this work, for the separation of polar and basic analytes in HILIC-mode, NH4Ac buffer and ACN mixture have been used. We found that, in those mobile-phase compositions, the alkyl chains of Sil-MEPG-C18-1 and SilMEPG-C18-2 were rigid and ordered by suspended-state 1H NMR spectroscopy (Supporting Information Figure S20). Polar analytes are believed to have penetrated into hydrophilic interaction sites and to have been retained. However, the established retention mechanism in HILIC-mode separation is not yet developed. From the analysis of HILIC results for multiple and single EPG-C18 phases, if we consider separation of neutral polar molecules, partitioning is supposed to be dominating (Figure 6). In the case of ionic polar molecules (sulfa drugs, Supporting Information Figure S12), ion-exchange interactions are believed to be dominating, and Sil-SEPG-C18 showed maximum ion-exchange properties, possibly due to easy access of ionic molecules to the residual aminopropyl groups. Perhaps the residual aminopropyl functionalities are at play in
■
ASSOCIATED CONTENT
S Supporting Information *
Synthetic procedures, experimental details, additional characterization data, and additional results. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b00663.
■
AUTHOR INFORMATION
Corresponding Authors
*(A.K.M.) E-mail: [email protected]. Tel. and Fax: +8196-342-3661. *(H.I.) E-mail: [email protected]. Tel. and Fax: +81-96342-3661. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Dr. Abul K. Mallik (ID No. PU13012) gratefully acknowledges the Japan Society for the Promotion of Science (JSPS) for providing financial support to carry out this research (Grant No. 26.03912).
■
REFERENCES
(1) Sander, L. C.; Wise, S. A. Anal. Chem. 1984, 56, 504−510.
G
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry (2) Sentell, K. B.; Dorsey, J. G. Anal. Chem. 1989, 61, 930−934. (3) Kimata, K.; Hirose, T.; Morichi, K.; Hosoya, K.; Araki, T.; Tanaka, N. Anal. Chem. 1995, 67, 2556−2561. (4) Jinno, K.; Tanabe, K.; Saito, Y.; Nagashima, H. Analyst 1997, 122, 787−791. (5) Mallik, A. K.; Sawada, T.; Takafuji, M.; Ihara, H. Anal. Chem. 2010, 82, 3320−3328. (6) Sander, L. C.; Pursch, M.; Wise, S. A. Anal. Chem. 1999, 71, 4821−4830. (7) Wise, S. A.; Sander, L. C. Chromatographic Separations Based on Molecular Recognition; Jinno, K., Ed.; Wiley-VCH: New York, 1997; Chapter 1. (8) Jinno, K.; Nagoshi, T.; Tanaka, N.; Okamoto, M.; Fetzer, J. C.; Biggs, W. R. J. Chromatogr. 1987, 392, 75−82. (9) Sander, L. C.; Wise, S. A. In Retention and Selectivity Studies in HPLC; Smith, R. M., Ed.; Elsevier: Amsterdam, 1994; pp 337−369. (10) Dorsey, J. G.; Dill, K. A. Chem. Rev. 1989, 89, 331−346. (11) Mallik, A. K.; Rahman, M. M.; Czaun, M.; Takafuji, M.; Ihara, H. J. Chromatogr. A 2008, 1187, 119−127. (12) Wise, S. A.; May, W. E. Anal. Chem. 1983, 55, 1479−1485. (13) Tanaka, N.; Sakagami, K.; Araki, M. J. Chromatogr. 1980, 199, 327−337. (14) Sander, L. C.; Wise, S. A. Anal. Chem. 1987, 59, 2309−2313. (15) Bell, C. M.; Sander, L. C.; Wise, S. A. J. Chromatogr. 1997, 757, 29−39. (16) Sander, L. C.; Sharpless, K. E.; Craft, N. E.; Wise, S. A. Anal. Chem. 1994, 66, 1667−1674. (17) Saito, Y.; Jinno, K.; Pesek, J. J.; Chen, Y. L.; Luehr, G.; Archer, J.; Fetzer, J. C.; Biggs, W. R. Chromatographia 1994, 38, 295−303. (18) Sander, L. C.; Wise, S. A. Anal. Chem. 1989, 61, 1749−1754. (19) Sander, L. C.; Lippa, K. A.; Wise, S. A. Anal. Bioanal. Chem. 2005, 382, 646−668. (20) Pursch, M.; Sander, L. C.; Egelhaaf, H.-J.; Raitza, M.; Wise, S. A.; Oelkrug, D.; Albert, K. J. Am. Chem. Soc. 1999, 121, 3201−3213. (21) Kühnle, M.; Friebolin, V.; Albert, K.; Rimmer, C. A.; Lippa, K. A.; Sander, L. C. Anal. Chem. 2009, 81, 10136−10142. (22) Mallik, A. K.; Qiu, H.; Sawada, T.; Takafuji, M.; Ihara, H. Anal. Chem. 2012, 84, 6577−6585. (23) Mallik, A. K.; Qiu, H.; Sawada, T.; Takafuji, M.; Ihara, H. Chem. Commun. (Cambridge, U. K.) 2011, 47, 10341−10343. (24) Pursch, M.; Vanderhart, D. L.; Sander, L. C.; Gu, X.; Nguyen, T.; Wise, S. A.; Gajewski, D. A. J. Am. Chem. Soc. 2000, 122, 6997− 7011. (25) McCalley, D. V. J. Chromatogr. 1993, 636, 213−220. (26) Nolan, J. W. LC-GC 1998, 16, 350−354. (27) Przybyciel, M.; Majors, R. E. LCGC North Am. 2002, 20, 516− 523. (28) Wang, C.; Jiang, C.; Armstrong, D. W. J. Sep. Sci. 2008, 31, 1980−1990. (29) O’Gara, J. E.; Alden, B. A.; Walter, T. H.; Petersen, J.; Niederlander, C. L.; Neue, U. D. Anal. Chem. 1995, 67, 3809−3813. (30) O’Gara, J. E.; Walsh, D. P.; Alden, B. A.; Casellini, P.; Walter, T. H. Anal. Chem. 1999, 71, 2992−2997. (31) O’Sullivan, G. P.; Scully, N. M.; Glennon, J. D. Anal. Lett. 2010, 43, 1609−1629. (32) Rimmer, C. A.; Sander, L. C. Anal. Bioanal. Chem. 2009, 394, 285−291. (33) Alpert, A. J. J. Chromatogr. 1990, 499, 177−196. (34) Strege, M. A. Anal. Chem. 1998, 70, 2439−2445. (35) Weng, N.; Shou, W.; Chen, Y. L.; Jiang, X. J. Chromatogr. B 2001, 754, 387−399. (36) Guo, Y.; Huang, A. J. Pharm. Biomed. Anal. 2003, 31, 1191− 1201. (37) Yoshida, T. Anal. Chem. 1997, 69, 3038−3043. (38) Berthod, A.; Chang, S. S.; Kullman, J. P.; Armstrong, D. W. Talanta 1998, 47, 1001−1012. (39) Liu, X.; Pohl, C. J. Chromatogr. A 2008, 1191, 83−89. (40) Qiu, H.; Zhang, M.; Gu, T.; Takafuji, M.; Ihara, H. Chem.Eur. J. 2013, 19, 18004−18010.
(41) Euerby, M. R.; Petersson, P. J. Chromatogr. A 2003, 994, 13−36. (42) Euerby, M. R.; McKeown, A. P.; Petersson, P. J. Sep. Sci. 2003, 26, 295−306. (43) Kawachi, Y.; Ikegami, T.; Takubo, H.; Ikegami, Y.; Miyamoto, M.; Tanaka, N. J. Chromatogr. A 2011, 1218, 5903−5919. (44) Dudones, L. P.; Gilpin, R. K. Langmuir 1999, 15, 6901−6907. (45) Rafferty, J. L.; Siepmann, J. I.; Schure, M. R. Anal. Chem. 2008, 80, 6214−6221. (46) Johansson, A.; Kollman, P.; Rothenberg, S.; McKelvey, J. J. Am. Chem. Soc. 1974, 96, 3794−3800. (47) Shundo, A.; Hori, K.; Ikeda, T.; Kimizuka, N.; Tanaka, K. J. Am. Chem. Soc. 2013, 135, 10282−10285. (48) Hartmann, S. R.; Hahn, E. L. Phys. Rev. 1962, 128, 2042. (49) Pines, A.; Gibby, M. G.; Waugh, J. S. J. Chem. Phys. 1972, 56, 1776−1777. (50) Andrew, E. R.; Bradbury, A.; Eades, R. G. Nature 1959, 183, 1802−1803. (51) Brindle, R.; Pursch, M.; Albert, K. Solid State Nucl. Magn. Reson. 1996, 6, 251−265. (52) Ansarian, H. R.; Derakhsan, M.; Rahman, M. M.; Sakurai, T.; Takafuji, M.; Taniguchi, I.; Ihara, H. Anal. Chim. Acta 2005, 547, 179− 187. (53) Albert, K.; Bayer, E. J. Chromatogr. 1991, 544, 345−370. (54) Sander, L. C.; Wise, S. A. SRM 869b, Column Selectivity Test Mixture for Liquid Chromatography (Polycyclic Aromatic Hydrocarbons), Certificate of Analysis; National Institute of Standards and Technology: Gaithersburg, MD, USA, 2008. (55) Sander, L. C.; Wise, S. A. Anal. Chem. 1995, 67, 3284−3292. (56) Sander, L. C.; Wise, S. A. J. High Resolut. Chromatogr. Chromatogr. Commun. 1988, 11, 383−387. (57) Wise, S. A.; Sander, L. C. J. High Resolut. Chromatogr. Chromatogr. Commun. 1985, 8, 248−255. (58) Tanaka, N.; Tokuda, Y.; Iwaguchi, K.; Araki, J. J. Chromatogr. 1982, 239, 761−772. (59) Jinno, K.; Yamamoto, K.; Nagashima, H.; Ueda, T.; Itoh, K. J. Chromatogr. 1990, 517, 193−207. (60) Kimata, K.; Iwaguchi, K.; Onishi, S.; Jinno, K.; Eksteen, R.; Hosoya, K.; Arki, M.; Tanaka, N. J. Chromatogr. Sci. 1989, 27, 721− 728. (61) Petersson, P.; Euerby, M. R. J. Sep. Sci. 2005, 28, 2120−2129. (62) Iványi, T.; Heyden, Y. V.; Visky, D.; Baten, P.; Beer, J. D.; Lázár, I.; Massart, D. L.; Roets, E.; Hoogmartens, J. J. Chromatogr. A 2002, 954, 99−114. (63) Cruz, E.; Euerby, M. R.; Johnson, C. M.; Hackett, C. A. Chromatographia 1997, 44, 151−161. (64) Euerby, M. R.; Petersson, P. J. Chromatogr. A 2005, 1088, 1−15. (65) Ascah, T. L.; Feibush, B. J. Chromatogr. A 1990, 506, 357−369. (66) Zheng, M.-M.; Zhang, M.-Y.; Peng, G.-Y.; Feng, Y.-Q. Anal. Chim. Acta 2008, 625, 160−172. (67) Peng, X.-T.; Feng, Y.-Q.; Hu, X.-Z.; Hu, D.-J. Chromatographia 2013, 76, 459−465. (68) Xu, L.; Peng, R.; Guan, X.; Tang, W.; Liu, X.; Zhang, H. Anal. Bioanal. Chem. 2013, 405, 8311−8318. (69) Mallik, A. K.; Qiu, H.; Takafuji, M.; Ihara, H. Anal. Bioanal. Chem. 2012, 404, 229−238. (70) Armstrong, D. W.; Jin, H. L. J. Chromatogr. A 1989, 462, 219− 232. (71) Dinh, N. P.; Jonsson, T.; Irgum, K. J. Chromatogr. A 2011, 1218, 5880−5891. (72) Mallik, A. K.; Cheah, W. K.; Shingo, K.; Ezaki, A.; Takafuji, M.; Ihara, H. Anal. Bioanal. Chem. 2014, 406, 4585−4593. (73) Hemström, P.; Irgum, K. J. Sep. Sci. 2006, 29, 1784−1821. (74) Qiu, H.; Wanigasekara, E.; Zhang, Y.; Tran, T.; Armstrong, D. W. J. Chromatogr. A 2011, 1218, 8075−8082.
H
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Design of C18 Organic Phases with Multiple Embedded Polar Groups for Ultraversatile Applications with Ultrahigh Selectivity Abul K. Mallik,*,† Hongdeng Qiu,∥ Tomohiro Oishi,‡ Yutaka Kuwahara,†,§ Makoto Takafuji,†,§ and Hirotaka Ihara*,†,§ †
Department of Applied Chemistry and Biochemistry and ‡Technical Division, Faculty of Engineering, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan § Kumamoto Institute for Photo-Electro Organics (Phoenics), Kumamoto 862-0901, Japan ∥ Key Laboratory of Chemistry of Northwestern Plant Resources and Key Laboratory for Natural Medicine of Gansu Province, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China S Supporting Information *
ABSTRACT: For the first time, we synthesized multiple embedded polar groups (EPGs) containing linear C18 organic phases. The new materials were characterized by elemental analysis, IR spectroscopy, 1H NMR, diffuse reflectance infrared Fourier transform (DRIFT), solid-state 13C cross-polarization magic angle spinning (CP/MAS) NMR, suspended-state 1H NMR, and differential scanning calorimetry (DSC). 29Si CP/MAS NMR was carried out to investigate the degree of cross-linking of the silane and silane functionality of the modified silica. Solid-state 13C CP/MAS NMR and suspended-state 1H NMR spectroscopy indicated a higher alkyl chain order for the phase containing four EPGs than for the phase with three EPGs. To correlate the NMR results with temperaturedependent chromatographic studies, standard reference materials (SRM 869b and SRM 1647e), a column selectivity test mixture for liquid chromatography was employed. A single EPG containing the C18 phase was also prepared in a similar manner to be used as a reference column especially for the separation of basic and polar compounds in reversed-phase liquid chromatography (RPLC) and hydrophilic interaction liquid chromatography (HILIC), respectively. Detailed chromatographic characterization of the new phases was performed in terms of their surface coverage, hydrophobic selectivity, shape selectivity, hydrogen bonding capacity, and ion-exchange capacity at pH 2.7 and 7.6 for RPLC as well as their hydrophilicity, the selectivity for hydrophilic−hydrophobic substituents, the selectivity for the region and configurational differences in hydrophilic substituents, the evaluation of electrostatic interactions, and the evaluation of the acidic−basic nature for HILIC-mode separation. Furthermore, peak shapes for the basic analytes propranolol and amitriptyline were studied as a function of the number of EPGs on the C18 phases in the RPLC. The chromatographic performance of multiple EPGs containing C18 HILIC phases is illustrated by the separation of sulfa drugs, β-blockers, xanthines, nucleic acid bases, nucleosides, and water-soluble vitamins. Both of the phases showed the best performance for the separation of shape-constrained isomers, nonpolar, polar, and basic compounds in RPLC- and HILIC-mode separation of sulfa drugs, and other polar and basic analytes compared to the conventional alkyl phases with and without embedded polar groups and HILIC phases. Surprisingly, one phase would be able to serve the performance of three different types of phases with very high selectivity, and we named this phase the “smart phase”. Versatile applications with a single column will reduce the column purchasing cost for the analyst as well as achieve high separation, which is challenging with the commercially available columns.
H
which are important but challenging due to their similar molecular shape. 6−9 Stationary phases for RPLC, also developed on the basis of shape selectivity, depend on the stationary-phase bonding density,6,10−12 alkyl-phase chain length,13−16 column temperature,17−19 or architecture and dynamics of the organic phases.20−24 However, the separation of basic and highly polar analytes by RPLC using alkyl phases
igh-performance liquid chromatography (HPLC) as a separation technique has resulted in significant progress in the analytical sciences since the 1970s, and its advancement has relied mostly on the development of new stationary phases. For example, reversed-phase (RP) separations, using hydrophobic or n-alkyl-type stationary phases, have greatly increased the application of HPLC due to its versatility and to the constant development of new stationary phases and instrumentation.1−5 Generally, two types of n-alkyl chromatographic sorbent can be distinguished on the basis of bonding chemistry (monomeric and polymeric C18). Polymeric C18 phases are more effective for the separation of a certain class of isomers, © XXXX American Chemical Society
Received: February 17, 2015 Accepted: June 4, 2015
A
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
of Standards and Technology (NIST), Gaithersburg, MD, USA). PAHs were purchased from TCI, Tokyo, Japan. The nonpolar, polar, and basic compound set (uracil, propranolol, butyl paraben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene) was obtained from Wako and TCI. Sulfa drugs (sulfanilamide, sulfamethoxypyridazine, sulfadiazine, sulfamethoxazole, sulfactamide, sulfamonomethoxine, and sulfaquinoxaline) were obtained from TCI or from SigmaAldrich (St. Louis, MO, USA). β-Blockers (Carvedilol, Metoprolol, Pindolol, Acebutolol, Sotalol, Labetalol, Atenolol, and Nadolol), xanthines (caffeine, 7-β-hydroxypropyltheophylline, dyphylline, xanthine, and hyphoxynthine), bases and nucleosides (thymine, uracil, 4,6-diaminopyrimidine, uridine, adenosine, cytosine, and cytidine), and water-soluble vitamins (nicotinamide, pyridoxine, nicotinic acid, and riboflavin) were purchased either from Sigma-Aldrich, Wako, or from TCI. Other nucleosides including 5-methyluridine, sodium ptoluenesulfonate, theobromine, theophylline, 2′-deoxyuridine, N,N,N-trimethylphenylammonium chloride, and vidarabine were obtained from TCI. 2′-Deoxyguanosine, 3′-deoxyguanosine, 4-nitrophenyl α-D-glucopyranoside, and 4-nitrophenyl βD-glucopyranoside were purchased from Sigma-Aldrich. 17βEstradiaol was obtained from Sigma-Aldrich. 17α-Estradiol was purchased from Dr. Ehrennstorfer GmbH (Augsburg, Germany). The multiple EPGs-containing C18 phases (Sil-MEPG-C18-1 and Sil-MEPG-C18-2) and the single EPG-containing C18 phase (Sil-SEPG-C18) were synthesized, characterized, and packed into stainless steel columns (150 mm × 4.6 mm i.d.). YMC silica (YMC SIL-120-S5) with a 5 μm diameter and a 12 nm pore size was used. In contrast, we used a commercial amide group (single) embedded C18 column (Ascentis RP-amide) as a reference column from Sigma-Aldrich. Our synthesized SilSEPG-C18 column was also used as a reference column. Many other reference columns were used in this work for both RPand HILIC-mode separation. Commercial monomeric and polymeric C18 columns (Inertsil ODS-3 and Inertsil ODS-P, respectively) were obtained from GL Sciences, Tokyo, Japan. Additionally, another polymeric C18 column (Shodex C18 P4D), purchased from Shodex, Tokyo, Japan, was also used as a reference column for better comparison. Longer alkyl phase C30 (Develosil C30-UG-5) was purchased from Nomura Chemical Co., Ltd., Tokyo, Japan. Commercial HILIC columns, for example, silica (Inertsil SIL), diol (Inertsil HILIC-Diol), and amide (Inertsil HILIC-Amide), were obtained from GL Sciences. Properties of the commercial reference columns are given in the Supporting Information (Table S1). Multiple EPG-containing C18 (N′-octadecyl-Nα-[(4-carboxybutanoyl)-2β-alanineamide (6) and N′-octadecyl-Nα-[(4-carboxybutanoyl)-β-alanineamide (7)) were synthesized using βalanine through alkylation, debenzyloxycarbonylation, and ringopening reactions with glutaric anhydride and then immobilized onto silica from 3-aminopropyltrimethoxysilane (APS)modified silica (Sil-APS), named Sil-MEPG-C18-1 and SilMEPG-C18-2, respectively (Supporting Information, Scheme S1). To obtain single EPG-C18-grafted silica (Sil-SEPG-C18), nonadecanoic acid was similarly grafted onto silica (Supporting Information, Scheme S1). The detailed synthetic procedures for 6 and 7, immobilization onto silica, and other experimental details are given in the Supporting Information. Liquid Chromatography. The chromatographic system consisted of a Gulliver PU-1580 intelligent HPLC pump with a Rheodyne sample injector. A JASCO multiwavelength UV
continues to challenge chromatographers due to peak tailing, phase collapse in highly aqueous environments, and poor retention and selectivity.25−28 To solve these problems, analysts have developed many types of single EPG-containing alkyl phases.29−31 However, these single EPG-containing alkyl phases were not suitable for the separation of shape-constrained isomers of polycyclic aromatic hydrocarbons (PAHs) and highly polar compounds.32 Moreover, an alternative and rival technique to RPLC for separating polar compounds was first reported by Alpert in 1990 and is called hydrophilic interaction chromatography (HILIC).33 Different types of stationary phases have been developed, including bare silica, amine, amide, diol, and cyclodextrin phases, for HILIC separations of polar analytes.34−38 Of course, these phases are not able to separate nonpolar and shape-constrained isomers. Furthermore, mixed-mode RPLC/HILIC stationary phases have also been synthesized. However, these phases do not show very high selectivity either in RPLC- or in HILIC-mode when compared to conventional single RPLC and HILIC phases.39,40 In this work, we report the first instance of multiple EPGcontaining C18 organic phases for the separation of versatile analytes including shape-constrained isomers, nonpolar, polar, and basic compounds with ultrahigh selectivity both in RPLCand HILIC-mode separations in a single column. In short, a single column can do the job of three different types of columns without losing the level of performance. Our designed novel materials (g and h) produced by embedding multiple polar groups differ from typical previous C18 (a and b),6−9 C30 (c),13−16 HILIC (d),34−38 mixed-mode (e),39,40 and EPG C18 (f)29−31 phase materials, as shown in Scheme 1. Our successful Scheme 1. Schematic Illustrations of Different Stationary Phases: (a) Monomeric C18, (b) Polymeric C18), and (c) C30 RPLC; (d) HILIC; (e) Mixed-Mode; (f) EPG C18; (g, h) Novel Multiple EPG C18 (Sil-MEPG-C18-2 and Sil-MEPGC18-1, Respectively) Materials Designed and Synthesized in This Study
strategy utilizes two concepts: a balanced combination of a highly hydrophilic moiety and a highly hydrophobic moiety in a single compound to achieve a synergistic effect of both moieties as well as individual effects in chromatographic separation.
■
EXPERIMENTAL SECTION Materials and Reagents. Stearylamine, nonadecanoic acid, diethylphosphorocyanidate (DPEC, peptide synthesis reagent), triethylamine (TEA), and β-alanine were purchased from Wako (Tokyo, Japan) and used without further purification. Standard reference material SRM 869b (column selectivity test mixture for liquid chromatography) and SRM 1647e (priority pollutant polycyclic aromatic hydrocarbons (PAHs)) were obtained from the Standard Reference Materials Program (National Institute B
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry detector MD 1510 plus was used. The chromatography was performed under isocratic elution conditions with a flow rate of 1 mL min−1 for most of the analysis. The column temperature was maintained by using a column jacket having a 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. Chromatographicgrade solvent was used to prepare mobile-phase solutions. The retention factor (k) was determined by (te − t0)/t0, where te was the retention time of the samples. The first disturbance of the baseline on the injection of methanol was used as the dead time marker (t0).41 The separation factor (α) was given by the ratio of retention factors. In RPLC, methanol was used as a mobile phase for the separation of SRM 869b and other PAHs. Separation of SRM 1647e was carried out using a 95:5 (volume fraction) methanol/water mobile phase at 40 °C. The other chromatographic conditions for the Tanaka RPLC characterization of the phases were as reported previously.41,42 In this protocol, six variables reflecting different chromatographic ion properties such as the retention factor for pentylbenzene, kPB; hydrophobic selectivity, α(CH2); the retention factor ratio between pentylbenzene and butylbenzene; shape selectivity, αT/O; the retention factor ratio between triphenylene and oterphenyl; hydrogen bonding capacity, αC/P; the retention factor ratio between caffeine and phenol; the total ion-exchange capacity at pH 7.6, αA/P; the retention factor ratio between benzylamine and phenol; acidic ion-exchange capacity at pH 2.7, αA/P; and the retention factor ratio between benzylamine and phenol were used for the characterization. A test mixture of nonpolar, polar, and basic compounds (uracil, propranolol, butylparaben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene) was separated in 45:55, 65:35, 75:25, and 75:25, methanol/20 mM KH2PO4/K2HPO4 at pH 7.00 for SilMEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases at 25 °C, respectively. For the separation of polar and basic analytes in the HILIC-mode, an ammonium acetate (NH4Ac) buffer and acetonitrile (ACN) mixture were used. In HILIC, most of the experiments were carried out at room temperature. The other chromatographic conditions for the HILIC characterization of the phases (and variables reflecting different chromatographic properties) were used as described by Kawachi et al.43 The detailed chromatographic conditions are also given in the Supporting Information.
Scheme 2. Chemical Structures of the Multiple Embedded Polar Groups (EPGs) Containing C18 Phases, Indicating Structural Properties and Possible Interaction Sites
2).13−16,44−47 Another column was prepared in a similar manner with a single EPG-containing C18 (Sil-SEPG-C18) to be used as a reference column, especially for the comparison and interaction mechanism elucidations of polar compound separation both in RPLC- and HILIC-mode (Supporting Information, Scheme S1). The materials were completely characterized by elemental analysis, thermogravimetric analysis (TGA), diffuse reflectance infrared Fourier transform (DRIFT), differential scanning calorimetry (DSC), solid-state 13 C cross-polarization (CP)48,49 magic angle spinning (MAS),50 and suspended-state 1H NMR51,52 and 29Si CP/MAS NMR53 spectroscopy before packing into stainless steel columns for chromatographic evaluation (see the Supporting Information for details). From the elemental analysis results, the surface coverage of APS, MEPG-C18-1, MEPG-C18-2, and SEPG-C18 to the silica surface was calculated according to the previously reported method6,12 as 7.72, 1.31, 1.55, and 2.19 μmol m−2 for Sil-APS, Sil-MEPG-C18-1, Sil-MEPG-C18-2, and Sil-SEPG-C18, respectively (Supporting Information, Table S2). The organic load in the new phases was also confirmed by the TGA measurements and showed agreement with elemental analysis results (Supporting Information, Table S2 and Figure S3). Additionally, grafting of organic molecules onto a silica surface was confirmed by DRIFT measurement (Supporting Information, Figure S4). The solid-state 13C CP/MAS NMR and suspended-state 1H NMR measurements were carried out at different temperatures from 20 to 50 °C to investigate the
■
RESULTS AND DISCUSSION As was mentioned earlier, multiple EPG-containing C18 compounds, MEPG-C18-1 (6) and MEPG-C18-2 (7), were synthesized using β-alanine through alkylation, debenzyloxycarbonylation, and ring-opening reactions with glutaric anhydride and then immobilized onto silica from 3-aminopropyltrimethoxysilane (APS)-modified silica (Sil-APS), named Sil-MEPG-C18-1 and Sil-MEPG-C18-2, respectively (Supporting Information, Scheme S1). Four and three EPG (amide)containing C18 phases (Sil-MEPG-C18-1 and Sil-MEPG-C18-2, respectively) were synthesized. The material contains two important units with five functions (Scheme 2). Four amide groups provide hydrophilicity and wettability, multiple carbonyl−π interaction sources,22−24 multiple H-bonding sources,44−47 and increase the chain length together with C18. However, rigid C18 provides hydrophobicity to the organic phase. H-bonding between polar groups and increased chain length leads to a more ordered alkyl chain structure (Scheme C
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
phase collapse was observed for both of the phases, indicating ordered alkyl chains (Supporting Information, Figure S6).52 29 Si CP/MAS NMR was carried out for the Sil-MEPG-C18-1 and Sil-MEPG-C18-2 phases (Supporting Information, Figure S7) attributed to a very high degree of cross-linking, even higher than the best C22 or C18 bonded phases.20,21 The results suggested a smaller number of free OH groups on the surface, which leads to fewer silanophilic interactions in HPLC.20,53 The novel phases were very suitable for the separation of shape-constrained isomers20−24 and neutral, polar, and basic compounds29−31 in RPLC and polar compounds34−38 in HILIC (i.e., three in one) compared to the most commonly used commercial RP, EPG, and HILIC columns. First, analysis was carried out with the separation of standard reference materials (SRMs) SRM869b and SRM1647e, as a column (shape) selectivity test mixture for liquid chromatography.54 SRM869b consists of 1,2:3,4:5,6:7,8-tetrabenzonaphthalene (TBN), phenanthro[3,4-c]phenanthrene (PhPh), and benzo[a]pyrene (BaP) (Supporting Information, Figure S8) and was developed after evaluation of over 100 PAH solutes. The three solutes (PAHs) selected provided the most sensitive indication of changes in selectivity due to solute shape. The ratio (α) of retention factors (k) for TBN and BaP (i.e., αTBN/BaP = kTBN/ kBaP) provides a measure of shape selectivity that is useful for column intercomparisons, and columns exhibit a high degree of shape selectivity (a lower value indicates higher shape selectivity); values typically fall within the range of 0.3− 1.0.55−57 We observed selectivity coefficients for the Sil-MEPGC18-1 phase ranging from 0.16 (at 10 °C) to 0.56 (at 50 °C), which exhibited very high shape selectivity compared to the most commonly used commercial RP columns (monomeric and polymeric C18 (1.93−1.95 and 0.28−0.76, respectively), EPG C18 (1.29−1.54), and C30 (0.51−1.81)) (Figure 2).6−24,32
conformations and mobility of the long alkyl chains, respectively, of the Sil-MEPG-C18-1 and Sil-MEPG-C18-2 phases. 13C CP/MAS NMR spectroscopy revealed highly ordered (trans conformation) alkyl chains in the Sil-MEPGC18-1 phase (Figure 1a) compared to the Sil-MEPG-C18-2
Figure 1. Partial solid-state 13C CP/MAS NMR spectra of Sil-MEPGC18-1 (a) and Sil-MEPG-C18-2 (b) at variable temperatures. Figure 2. Phase selectivity (αTBN/BaP = kTBN/kBaP) plotted as a function of temperature for Sil-MEPG-C18-1, Sil-MEPG-C18-2, and most commonly used RP commercial columns.
phase (Figure 1b) irrespective of temperature,17−19 polymeric chains,1,8 alkyl chain length,13−16 and surface coverage (1.31 μmol m−2).6,10−12 The degree of alkyl chain ordering increases with increasing alkyl chain length, grafting density/surface coverage, and lower column temperature. Highly ordered alkyl chains on the new phases may be due to embedded multiple polar groups (amide) and H-bonding among them.31,44−47 Similar results were also obtained by suspended-state 1H NMR in methanol (Supporting Information, Figure S5a,b), indicating that the mobility of the alkyl chains is low even at higher temperature and the results agree with the results of 13C CP/ MAS NMR spectroscopy.51,52 Moreover, DSC of the new phases was conducted in methanol, and no phase transition or
Figure 3 shows the comparative chromatogram for the separation of SRM869b in the RP-mode. Sil-MEPG-C18-1 was also able to separate other important standard PAH mixtures (SRM1647e; 16 PAHs are listed as priority pollutants by the EPA) into 16 peaks (Supporting Information, Figures S9 and S10a) within 27 min in isocratic elution for the first time, which represented one of the most important challenges.5,22,24 However, polymeric C18 phases (Shodex C18P-4D and Inertsil ODS-P) were not able to separate the standard PAH solution D
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Furthermore, Sil-MEPG-C18-1 and Sil-MEPG-C18-2 showed ultrahigh selectivity for other PAHs and their isomers in comparison with commercial RP columns, especially αcoronene/hexahelicene = 28.4 and 21.0, respectively (Supporting Information, Table S3). Additionally, to have a fair comparison, all parameters, such as capacity, hydrophobicity, steric or shape selectivity, silanol capacity, and ion-exchange capacities, at high and low pH were compared according to the Tanaka characterization protocol (Table 1).60 The Tanaka characterization protocol is a wellestablished approach that has been favored by academic groups61−64 and many stationary-phase manufacturers such as ThermoHypersil-Keystone, Merck, and Phenomenex to assess their phases. EPG phases were originally developed to reduce the peak tailing of basic analytes.29−31,65 Figure 5 shows the separation of
Figure 3. Chromatograms for the separation of SRM869b at 35 °C on Sil-MEPG-C18-1, Sil-MEPG-C18-2, and most commonly used RP commercial columns.
into 16 peaks under the same chromatographic conditions (Supporting Information, Figure S10b). Although Inertsil ODS-P column (highest carbon loading, 29% C) showed better separation ability, compounds 3 and 4 coeluted. Previously, we also reported that the SRM1647e cannot be separated in isocratic elution on C18 and C30 phases.22 The highest shape selectivity value was also observed for another shape selective probe (o-terphenyl and triphenylene, αtriphenylene/o‑terphenyl = 10.6 and 7.90, respectively, at 20 °C) defined by Jinno and Tanaka as shown in Figure 4.20,58,59
Figure 5. Chromatograms for the separation of uracil, propranolol, butylparaben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene on Sil-MEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases, respectively.
Figure 4. Phase selectivity (αtriphenylene/o‑terphenyl = ktriphenylene/ko‑terphenyl) plotted as a function of temperature for Sil-MEPG-C18-1, Sil-MEPGC18-2, and most commonly used RP commercial columns.
Table 1. Chromatographic Characterization (Tanaka Column Characterization Protocol) of the Newly Synthesized and Commercial Reference Columns in This Study (in RPLC)
Inertsil ODS-3 Ascentis RP-amide Shodex C18P-4D Develosil C30-UG-5 Inertsil ODS-P Sil-SEPG-C18 Sil-MEPG-C18-2 Sil-MEPG-C18-1
kPB
α(CH2)
αT/O
αC/P
αA/P at pH 7.5
αA/P at pH 2.7
6.85 4.82 5.10 5.42 6.21 2.84 1.26 0.86
1.46 1.42 1.45 1.45 1.47 1.43 1.43 1.34
1.30 1.65 1.98 1.58 2.47 2.48 3.65 5.91
0.47 0.20 0.40 0.59 0.76 0.15 0.23 0.37
0.12 0.09 0.26 0.16 0.46 0.09 0.08 0.23
0.47 0.92 0.44 0.54 0.40 0.99 0.99 1.00
E
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
sulfamethizole (SMT))66−68 and the most retained phase was Sil-SEPG-C18 with poor peak shapes (Supporting Information, Figure S12), possibly due to strong ion-exchange interactions with the easily accessed (due to shorter alkyl chains) residual aminopropyl groups on Sil-SEPG-C18. However, almost no separation was obtained with Ascentis RP-amide and most commonly used HILIC columns. The detailed chromatographic parameters for the separation of sulfa drugs on SilMEPG-C18-1 are given in Table S4 (Supporting Information). Again, eight β-blockers were selected and used as probe analytes to evaluate the stationary phases in HILIC-mode separation. Figure S13 (Supporting Information) gives the structures of these compounds and shows that they are hydroxylamine-containing compounds and contain at least one aromatic ring. Figure S13 (Supporting Information) also shows the chromatograms of β-blockers on Sil-MEPG-C18-1, SilMEPG-C 18-2, Sil-SEPG-C18 , and conventional EPG C 18 (Ascentis RP-amide). Conventional EPG C18 (Ascentis RPamide) was not able to function in HILIC-mode separation. Here again, the best separation was obtained with Sil-MEPGC18-1 phase. Furthermore, the new columns separated polar and hydrophilic compounds successfully, including xanthines and water-soluble vitamins (Supporting Information, Figures S14 and S15) To obtain more insight about the chromatographic characterization and interaction mechanism in HILIC-mode separation, similar to RPLC phases (e.g., by the Tanaka protocol), the new phases were characterized by the Tanaka protocol (that is supposed to be) for HILIC phases.43 In the work of Tanaka et al., characterization was carried out in terms of the degree of hydrophilicity, the selectivity for hydrophilic−hydrophobic substituents, the selectivity for the regio and configurational differences in hydrophilic substituents, the selectivity for molecular shapes, the evaluation of electrostatic interactions, and the evaluation of the acidic−basic nature of the stationary phases using nucleoside derivatives, phenyl glucoside derivatives, xanthine derivatives, sodium p-toluenesulfonate, and trimethylphenylammonium chloride as a set of test samples. The column efficiency and the peak asymmetry factors were also considered. Figure S16 (Supporting Information) shows the structure of the test samples. Detailed chromatographic data for the HILIC-mode separation are given in the Supporting Information (Tables S5, S6, S7, S8, S9, and S10). Properties of the new phases are obviously comparable with the typical HILIC phases. Among the prepared phases, Sil-MEPGC18-1 (maximum EPG-containing phase) showed the best column efficiency and peak asymmetry factors. The pH on the surface of the phases represented basic characteristics, and SilSEPG-C18 is the most basic phase. The reason for ultradiverse and ultrahigh selectivity of the MEPG C18 phases can be understood from the unique chemical structure and properties, which can work both in RPLC-mode (Supporting Information, Figure S17) and HILIC-mode (Supporting Information, Figure S18) separation. 13C CP/ MAS NMR spectroscopy revealed highly ordered alkyl chains in the Sil-MEPG-C18-1 phase (Figure 1a) compared to the SilMEPG-C18-2 phase (Figure 1b), although the phases were not polymeric type and high surface concentration (only 1.31 and 1.55 μmol/m2, respectively). In RPLC, the contribution of bonded phase rigidity and ordering to shape selectivity have been confirmed on alkyl stationary phases.6−24 However, extremely high shape selectivity (αcoronene/hexahelicene = 28.4) of our phases cannot be explained only by the ordering of the
a test sample containing a mixture of nonpolar, polar, and basic compounds (uracil, propranolol, butyl paraben, dipropyl phthalate, naphthalene, amitriptyline, and acenaphthene) on Sil-MEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases, respectively. Our designed multiple EPG phases (Sil-MEPG-C18-1 and Sil-MEPG-C18-2) showed very good peak asymmetry (for basic amitriptyline and propranolol, especially for amitriptyline) compared to the reference single EPG C18 phases (Sil-SEPG-C18 and Ascentis RP-amide; Supporting Information, Figure S11). Finally, the new phases were also evaluated in HILIC-mode separation. Surprisingly, good separation of neutral polar molecules (nucleic acid bases and nucleosides) was also achieved on Sil-MEPG-C18-1 and Sil-MEPG-C18-2 phases compared to single EPG-containing C18 phases (Sil-SEPG-C18 and Ascentis RP-amide), as shown in Figure 6. The best
Figure 6. Separation of nucleic acid bases and nucleosides on SilMEPG-C18-1, Sil-MEPG-C18-2, Sil-SEPG-C18, and EPG C18 (Ascentis RP-amide) phases: mobile phase, ACN−ammonium acetate buffer (10 mM in the aqueous portion, pH 6.7; 95:5, v/v); column temperature, 25 °C; flow rate, 1 mL/min.
separation was observed with the maximum EPG-containing phase (Sil-MEPG-C18-1). Very nice separation of polar ionic molecules was observed with the prepared phases compared to conventional commercial HILIC columns (silica, amide, and diol).34−38 For example, the separation of seven sulfa drugs (sulfanilamide (SNA), sulfamethoxypyridazine (SMP), sulfadiazine (SD), sulfamethoxazole (SMX), sulfectamide (STA), sulfamonomethoxine (SMM), sulfaquinoxaline (SQ), and F
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
the HILIC interaction mechanism. To prove the ion-exchange interaction mechanism, we investigated the impact of increased buffer concentration on retention factors for the neutral and ionic molecules. Figures S21 and S22 (Supporting Information) show that the retention decreases nonlinearly and linearly for the neutral and ionic molecules, respectively, on Sil-MEPG-C181. In short, the mixed-interaction mechanism has been investigated. Among the prepared phases, the maximum EPG-containing phase proved to be the best column in RPLC- and HILIC-mode separation. Therefore, multiple EPGs are important not only for RPLC but also for HILIC.
alkyl chains. From careful investigations on the retention behavior of shape-constrained isomers, we have concluded that the polymeric phases form a rigid ligand structure on the support, explained as a type of “slot-like” structure.57 Figure 7
■
CONCLUSIONS We have demonstrated the synthesis and applications of multiple EPG-containing C18 phases both in RPLC- and HILIC-mode separation. The unique stationary phases were completely characterized from a physiochemical standpoint as well as a chromatographic viewpoint. These phases showed ultrahigh selectivity and separation ability in both retention modes, compared to the most widely used conventional RP and HILIC columns. This improved performance is based on the presence of highly hydrophobic moieties and highly hydrophilic moieties in the same phase, induced by hydrogen bonding between amide groups and ordering of the alkyl chains. In RPLC, multiple EPG-containing phases are believed to form not only a slot-like structure but also to facilitate multiple carbonyl−π interactions leading to ultrahigh shape selectivity. In HILIC, multiple interactions were investigated, and the residual aminopropyl functionalities are supposed to influence ion-exchange interactions. The present results illustrate the separation of shape-constrained isomers, steroids, neutral, polar, and basic analytes with utilization of the multiple EPG-C18 phases. The concept, for which we propose the name “three-inone” or “smart phase”, has the potential to reduce column purchasing costs as well as yielding good selectivity, leading to a breakthrough in the field of analytical chemistry and materials science dealing with separation.
Figure 7. In RPLC, slot model explaining the selectivity for planar/ nonplanar solutes, including carbonyl−π interactions.
illustrates the retention model on the Sil-MEPG-C18-1 phase with a slot-like phase structure on silica gel, where a planar solute can interact (including multiple carbonyl−π interactions)5 to a greater extent than a nonplanar solute. Therefore, Sil-MEPG-C18-1 not only forms a slot-like structure but also facilitates multiple carbonyl−π interactions leading to ultrahigh shape selectivity. Additionally, new phases not only showed higher selectivity for the separation of shape-constrained isomers but also for other positional isomers (e.g., steroids)69 than any other reference columns (Supporting Information, Figure S19). Multiple H-bonding interactions are also possible due to the presence of multiple H-bonding (amide bonds or carbonyl groups) sources.31,44−47 In HILIC, a stagnant water layer forms on the surface of the hydrophilic stationary phase. Analytes partition between the stagnant water-rich solvent layer and the moving organic-rich eluent.33 The primary mechanism of retention is postulated to be partitioning of the analyte into this water-rich layer. However, adsorption, ion-exchange, dipole−dipole interaction, hydrogen bonding,70−73 π−π, and n−π interactions74 also contribute to retention on HILIC stationary phases. In this work, for the separation of polar and basic analytes in HILIC-mode, NH4Ac buffer and ACN mixture have been used. We found that, in those mobile-phase compositions, the alkyl chains of Sil-MEPG-C18-1 and SilMEPG-C18-2 were rigid and ordered by suspended-state 1H NMR spectroscopy (Supporting Information Figure S20). Polar analytes are believed to have penetrated into hydrophilic interaction sites and to have been retained. However, the established retention mechanism in HILIC-mode separation is not yet developed. From the analysis of HILIC results for multiple and single EPG-C18 phases, if we consider separation of neutral polar molecules, partitioning is supposed to be dominating (Figure 6). In the case of ionic polar molecules (sulfa drugs, Supporting Information Figure S12), ion-exchange interactions are believed to be dominating, and Sil-SEPG-C18 showed maximum ion-exchange properties, possibly due to easy access of ionic molecules to the residual aminopropyl groups. Perhaps the residual aminopropyl functionalities are at play in
■
ASSOCIATED CONTENT
S Supporting Information *
Synthetic procedures, experimental details, additional characterization data, and additional results. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b00663.
■
AUTHOR INFORMATION
Corresponding Authors
*(A.K.M.) E-mail: [email protected]. Tel. and Fax: +8196-342-3661. *(H.I.) E-mail: [email protected]. Tel. and Fax: +81-96342-3661. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Dr. Abul K. Mallik (ID No. PU13012) gratefully acknowledges the Japan Society for the Promotion of Science (JSPS) for providing financial support to carry out this research (Grant No. 26.03912).
■
REFERENCES
(1) Sander, L. C.; Wise, S. A. Anal. Chem. 1984, 56, 504−510.
G
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry (2) Sentell, K. B.; Dorsey, J. G. Anal. Chem. 1989, 61, 930−934. (3) Kimata, K.; Hirose, T.; Morichi, K.; Hosoya, K.; Araki, T.; Tanaka, N. Anal. Chem. 1995, 67, 2556−2561. (4) Jinno, K.; Tanabe, K.; Saito, Y.; Nagashima, H. Analyst 1997, 122, 787−791. (5) Mallik, A. K.; Sawada, T.; Takafuji, M.; Ihara, H. Anal. Chem. 2010, 82, 3320−3328. (6) Sander, L. C.; Pursch, M.; Wise, S. A. Anal. Chem. 1999, 71, 4821−4830. (7) Wise, S. A.; Sander, L. C. Chromatographic Separations Based on Molecular Recognition; Jinno, K., Ed.; Wiley-VCH: New York, 1997; Chapter 1. (8) Jinno, K.; Nagoshi, T.; Tanaka, N.; Okamoto, M.; Fetzer, J. C.; Biggs, W. R. J. Chromatogr. 1987, 392, 75−82. (9) Sander, L. C.; Wise, S. A. In Retention and Selectivity Studies in HPLC; Smith, R. M., Ed.; Elsevier: Amsterdam, 1994; pp 337−369. (10) Dorsey, J. G.; Dill, K. A. Chem. Rev. 1989, 89, 331−346. (11) Mallik, A. K.; Rahman, M. M.; Czaun, M.; Takafuji, M.; Ihara, H. J. Chromatogr. A 2008, 1187, 119−127. (12) Wise, S. A.; May, W. E. Anal. Chem. 1983, 55, 1479−1485. (13) Tanaka, N.; Sakagami, K.; Araki, M. J. Chromatogr. 1980, 199, 327−337. (14) Sander, L. C.; Wise, S. A. Anal. Chem. 1987, 59, 2309−2313. (15) Bell, C. M.; Sander, L. C.; Wise, S. A. J. Chromatogr. 1997, 757, 29−39. (16) Sander, L. C.; Sharpless, K. E.; Craft, N. E.; Wise, S. A. Anal. Chem. 1994, 66, 1667−1674. (17) Saito, Y.; Jinno, K.; Pesek, J. J.; Chen, Y. L.; Luehr, G.; Archer, J.; Fetzer, J. C.; Biggs, W. R. Chromatographia 1994, 38, 295−303. (18) Sander, L. C.; Wise, S. A. Anal. Chem. 1989, 61, 1749−1754. (19) Sander, L. C.; Lippa, K. A.; Wise, S. A. Anal. Bioanal. Chem. 2005, 382, 646−668. (20) Pursch, M.; Sander, L. C.; Egelhaaf, H.-J.; Raitza, M.; Wise, S. A.; Oelkrug, D.; Albert, K. J. Am. Chem. Soc. 1999, 121, 3201−3213. (21) Kühnle, M.; Friebolin, V.; Albert, K.; Rimmer, C. A.; Lippa, K. A.; Sander, L. C. Anal. Chem. 2009, 81, 10136−10142. (22) Mallik, A. K.; Qiu, H.; Sawada, T.; Takafuji, M.; Ihara, H. Anal. Chem. 2012, 84, 6577−6585. (23) Mallik, A. K.; Qiu, H.; Sawada, T.; Takafuji, M.; Ihara, H. Chem. Commun. (Cambridge, U. K.) 2011, 47, 10341−10343. (24) Pursch, M.; Vanderhart, D. L.; Sander, L. C.; Gu, X.; Nguyen, T.; Wise, S. A.; Gajewski, D. A. J. Am. Chem. Soc. 2000, 122, 6997− 7011. (25) McCalley, D. V. J. Chromatogr. 1993, 636, 213−220. (26) Nolan, J. W. LC-GC 1998, 16, 350−354. (27) Przybyciel, M.; Majors, R. E. LCGC North Am. 2002, 20, 516− 523. (28) Wang, C.; Jiang, C.; Armstrong, D. W. J. Sep. Sci. 2008, 31, 1980−1990. (29) O’Gara, J. E.; Alden, B. A.; Walter, T. H.; Petersen, J.; Niederlander, C. L.; Neue, U. D. Anal. Chem. 1995, 67, 3809−3813. (30) O’Gara, J. E.; Walsh, D. P.; Alden, B. A.; Casellini, P.; Walter, T. H. Anal. Chem. 1999, 71, 2992−2997. (31) O’Sullivan, G. P.; Scully, N. M.; Glennon, J. D. Anal. Lett. 2010, 43, 1609−1629. (32) Rimmer, C. A.; Sander, L. C. Anal. Bioanal. Chem. 2009, 394, 285−291. (33) Alpert, A. J. J. Chromatogr. 1990, 499, 177−196. (34) Strege, M. A. Anal. Chem. 1998, 70, 2439−2445. (35) Weng, N.; Shou, W.; Chen, Y. L.; Jiang, X. J. Chromatogr. B 2001, 754, 387−399. (36) Guo, Y.; Huang, A. J. Pharm. Biomed. Anal. 2003, 31, 1191− 1201. (37) Yoshida, T. Anal. Chem. 1997, 69, 3038−3043. (38) Berthod, A.; Chang, S. S.; Kullman, J. P.; Armstrong, D. W. Talanta 1998, 47, 1001−1012. (39) Liu, X.; Pohl, C. J. Chromatogr. A 2008, 1191, 83−89. (40) Qiu, H.; Zhang, M.; Gu, T.; Takafuji, M.; Ihara, H. Chem.Eur. J. 2013, 19, 18004−18010.
(41) Euerby, M. R.; Petersson, P. J. Chromatogr. A 2003, 994, 13−36. (42) Euerby, M. R.; McKeown, A. P.; Petersson, P. J. Sep. Sci. 2003, 26, 295−306. (43) Kawachi, Y.; Ikegami, T.; Takubo, H.; Ikegami, Y.; Miyamoto, M.; Tanaka, N. J. Chromatogr. A 2011, 1218, 5903−5919. (44) Dudones, L. P.; Gilpin, R. K. Langmuir 1999, 15, 6901−6907. (45) Rafferty, J. L.; Siepmann, J. I.; Schure, M. R. Anal. Chem. 2008, 80, 6214−6221. (46) Johansson, A.; Kollman, P.; Rothenberg, S.; McKelvey, J. J. Am. Chem. Soc. 1974, 96, 3794−3800. (47) Shundo, A.; Hori, K.; Ikeda, T.; Kimizuka, N.; Tanaka, K. J. Am. Chem. Soc. 2013, 135, 10282−10285. (48) Hartmann, S. R.; Hahn, E. L. Phys. Rev. 1962, 128, 2042. (49) Pines, A.; Gibby, M. G.; Waugh, J. S. J. Chem. Phys. 1972, 56, 1776−1777. (50) Andrew, E. R.; Bradbury, A.; Eades, R. G. Nature 1959, 183, 1802−1803. (51) Brindle, R.; Pursch, M.; Albert, K. Solid State Nucl. Magn. Reson. 1996, 6, 251−265. (52) Ansarian, H. R.; Derakhsan, M.; Rahman, M. M.; Sakurai, T.; Takafuji, M.; Taniguchi, I.; Ihara, H. Anal. Chim. Acta 2005, 547, 179− 187. (53) Albert, K.; Bayer, E. J. Chromatogr. 1991, 544, 345−370. (54) Sander, L. C.; Wise, S. A. SRM 869b, Column Selectivity Test Mixture for Liquid Chromatography (Polycyclic Aromatic Hydrocarbons), Certificate of Analysis; National Institute of Standards and Technology: Gaithersburg, MD, USA, 2008. (55) Sander, L. C.; Wise, S. A. Anal. Chem. 1995, 67, 3284−3292. (56) Sander, L. C.; Wise, S. A. J. High Resolut. Chromatogr. Chromatogr. Commun. 1988, 11, 383−387. (57) Wise, S. A.; Sander, L. C. J. High Resolut. Chromatogr. Chromatogr. Commun. 1985, 8, 248−255. (58) Tanaka, N.; Tokuda, Y.; Iwaguchi, K.; Araki, J. J. Chromatogr. 1982, 239, 761−772. (59) Jinno, K.; Yamamoto, K.; Nagashima, H.; Ueda, T.; Itoh, K. J. Chromatogr. 1990, 517, 193−207. (60) Kimata, K.; Iwaguchi, K.; Onishi, S.; Jinno, K.; Eksteen, R.; Hosoya, K.; Arki, M.; Tanaka, N. J. Chromatogr. Sci. 1989, 27, 721− 728. (61) Petersson, P.; Euerby, M. R. J. Sep. Sci. 2005, 28, 2120−2129. (62) Iványi, T.; Heyden, Y. V.; Visky, D.; Baten, P.; Beer, J. D.; Lázár, I.; Massart, D. L.; Roets, E.; Hoogmartens, J. J. Chromatogr. A 2002, 954, 99−114. (63) Cruz, E.; Euerby, M. R.; Johnson, C. M.; Hackett, C. A. Chromatographia 1997, 44, 151−161. (64) Euerby, M. R.; Petersson, P. J. Chromatogr. A 2005, 1088, 1−15. (65) Ascah, T. L.; Feibush, B. J. Chromatogr. A 1990, 506, 357−369. (66) Zheng, M.-M.; Zhang, M.-Y.; Peng, G.-Y.; Feng, Y.-Q. Anal. Chim. Acta 2008, 625, 160−172. (67) Peng, X.-T.; Feng, Y.-Q.; Hu, X.-Z.; Hu, D.-J. Chromatographia 2013, 76, 459−465. (68) Xu, L.; Peng, R.; Guan, X.; Tang, W.; Liu, X.; Zhang, H. Anal. Bioanal. Chem. 2013, 405, 8311−8318. (69) Mallik, A. K.; Qiu, H.; Takafuji, M.; Ihara, H. Anal. Bioanal. Chem. 2012, 404, 229−238. (70) Armstrong, D. W.; Jin, H. L. J. Chromatogr. A 1989, 462, 219− 232. (71) Dinh, N. P.; Jonsson, T.; Irgum, K. J. Chromatogr. A 2011, 1218, 5880−5891. (72) Mallik, A. K.; Cheah, W. K.; Shingo, K.; Ezaki, A.; Takafuji, M.; Ihara, H. Anal. Bioanal. Chem. 2014, 406, 4585−4593. (73) Hemström, P.; Irgum, K. J. Sep. Sci. 2006, 29, 1784−1821. (74) Qiu, H.; Wanigasekara, E.; Zhang, Y.; Tran, T.; Armstrong, D. W. J. Chromatogr. A 2011, 1218, 8075−8082.
H
DOI: 10.1021/acs.analchem.5b00663 Anal. Chem. XXXX, XXX, XXX−XXX
