Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Engineering thiol-ene click chemistry for the fabrication of novel structurally well-defined multifunctional cyclodextrin separation materials for enhanced enantioseparation Xiaobin Yao, Hao Zheng, Yang Zhang, Xiaofei Ma, Yin Xiao, and Yong Wang Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b00897 • Publication Date (Web): 08 Apr 2016 Downloaded from http://pubs.acs.org on April 9, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 27
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Detailed Response to Referee’s Comments to Manuscript No. ac-2016-008973 Response to reviewer 3 Comments: Novel more powerful chiral stationary phases (CSP) are always welcome since the number of single enantiomers that are needed in e.g. pharmaceutical research, development, and production is growing. This manuscript describes performance of two CSPs based on functionalized cyclodextrin (CD) attached to silica beads via thiol-ene click reaction. The approach the authors used is interesting although use of click reactions for the preparation of various chromatographic media is known for some time now. The performance of the new CSPs are well desaturated with the separation of a large number of racemates differing in their structures. This work merits after some revisions are carried out. Point-to-point response is listed below: 1. A number of compounds were separated. Their structures should be presented. (Editor- I believe you included this information in the supporting information). Answer: The structures of the analytes have been included in the supporting information (Figure S-1). 2. I am missing explanation why the authors functionalized CD with allylimidazole. Is the imidazole functionality important? Can allylamine be used? Answer: Thanks for the reviewer’s suggestion. Allylamine should also work in anchoring CDs onto thiol silica to form cationic linkage, however, the linkage is not that stable to tolerate various separation conditions (especially buffer solutions) compared to imidazolium where a quaternary ammonium is formed. In addition, the imidazole ring can provide more H-bonding sites which may do favor to the CSP’s separation profile or selectivity. 3. SEM images of the CSPs say nothing. This part of fig. 1 can be removed. Answer: According to the reviewer’s suggestions, we have removed the SEM images into the supporting information. 4. P. 8, l. 56: 5,000 um/ml looks better as 5 mg/mL or 5 g/L. Answer: 5000 µg/mL has been changed to 5 mg/mL accordingly in the revised manuscript and highlighted in red colour. 5. Seeing just two peaks in all chromatograms of Fig. 2 is somewhat boring. Why not to show a single one and all other present in the Supporting Info. Answer: According to the reviewer’s suggestion, we only keep the chromatograms of 3ClPh-OPr (on CSP1) and 6-Hydroxyflavanone (on CSP2) in the main article, the other chromatograms have been removed to the supporting information (Fig. S-2) in the revised manuscript. 6. The text is full of abbreviations. I suggest to add a list of abbreviations. Answer: According to the reviewer’s suggestion, we have added the list of abbreviations and put in in the end of the main text in the revised manuscript. 7. The English of this manuscript may need some editing. Answer: According to the reviewer’s advice, we have carefully check through the manuscript and made necessary editing accordingly to make it sound native.
Thanks again for the reviewer’s valuable comments.
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 27
Engineering thiol-ene click chemistry for the fabrication of novel structurally well-defined multifunctional cyclodextrin separation materials for enhanced enantioseparation Xiaobin Yao1,2, Hao Zheng3, Yang Zhang4, Xiaofei Ma1,2, Yin Xiao*2,3, Yong Wang*1,2 1
Department of Chemistry, School of Science, Tianjin University, Tianjin 300072, China.
2
Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, China.
3
School of Chemical Engineering, Tianjin University, Tianjin 300072, China
4
Weifang Teda Environmental Protection Equipment Co., Ltd, Weifang 262100, China.
E-mail addresses:
[email protected],
[email protected].
Abstract The preparation of two novel multifunctional cyclodextrin (CD) separation materials and their ultimate enantioseparation performances in high performance liquid chromatography are reported.
A
mild
thiol-ene
click
reaction
was
used
1-allylimidazolium-per(p-methyl)phenylcarbamoylated-β-CD
to
anchor and
1-allylimidazolium-per(p-chloride)phenylcarbamoylated-β-CD onto thiol-modified porous silica giving structurally well-defined stable cationic multifunctional CD chiral stationary phases (CSP1 and CSP2 respectively). These covalently-bonded CD phases have typical interaction modes such as H-bonding, π−π effect, electrostatic and dipole-dipole interactions as well as steric effects which result in superior chiral resolution for a variety of chiral compounds in different separation modes. In a reverse-phase mode, both CSPs exhibited excellent separation abilities for isoxazolines, flavonoids, β-blockers and some other neutral and basic racemates. In a polar-organic mode, isoxazolines and flavonoids were well resolved. CSP1 with an electron-rich phenyl substitution on the CD rims gave a better resolution for isoxazolines whereas CSP2 with an electron-deficient phenyl substitution on the CD rims gave better resolution for flavonoids. Among isoxazolines, 4ClPh-OPr gained a high selectivity and resolution up to 18.6 and 38.7 respectively, which is an amazing result for CD enantioseparation materials.
ACS Paragon Plus Environment
Page 3 of 27
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Introduction The importance of enantioseparations continues to increase due to their broad range of applications to industries such as pharmaceuticals,1-4 pesticides,5 food additives,6 and many others.7,8 Chromatographic techniques, especially high performance liquid chromatography (HPLC) with chiral stationary phases (CSPs) have grown into one of the most important techniques for both the determination of enantiomeric purity and the quick preparation of pure enantiomers.9-11 Cyclodextrin (CD) based separation materials, especially CD-CSPs are extensively used in enantioseparations since the shape of the truncated hollow CD cone can form strong inclusion complexes with guest molecules.12 In addition to inclusion complexation, CD-derived CSPs have other advantages such as being amenable to chemical modification so that π-π, dipole-dipole, ion-pairing, hydrogen bonding, electrostatic and steric repulsion interactions can be constructed to promote chiral recognition processes.13 Multifarious CD derivatives have been synthesized and immobilized onto silica surfaces via different covalent bonding strategies.14 Although, these different synthesis and immobilization strategies have produced materials with enhanced lead to exert profound influence in their enantioseparation abilities,15 the fabrication of structurally well-defined multifunctional CD-CSPs is still a great challenge in this research area. Most reported CD-CSPs have been prepared using immobilizations with common linkers such as ethers,16-20 amino groups,21-26 or urea27-32 although more recently a triazole linkage has also been used.33 Click chemistry, i.e. the Cu(I) catalytic 1,3-dipolar cycloaddition reaction, has great advantages over other CD immobilization methods because of its tolerance to water, oxygen and various functional moieties. In addition, it can be performed under mild conditions and forms a stable covalent bond which is can withstand harsh conditions.33-35 Therefore, a series of native and functionalized CD-CSPs have been developed via click chemistry. Wang et al. first reported the development of triazole linked native and functionalized CD-CSPs.36 The native β-CD clicked CSP (CCN-CSP) afforded excellent enantioseparation towards dansyl amino acids and flavonoids.37 Designed perphenylcarbamoylated β-CD CSP (CCP-CSP) and permethylated β-CD CSP (CCM-CSP) afforded good enantioselectivities for aryl alcohols, flavonoids, β-blockers (CCP-CSP) and non-aromatic ionone derivatives (CCM-CSP).38 The enantiorecognition abilities of triazole linked CD-CSPs can be finely tuned by changing the substitutions on the phenyl rings on the CD rims to achieve good discriminations against a wide range of chiral analytes. Tang and coworkers investigated the effects of substitution on triazole linked phenylcarbamoylated CD-CSPs, where the phenyl rings were decorated with electron-donating methyl or electron-withdrawing chloro groups.39,40 Although this provided some insights about the impact of the phenyl ring substitution position on the CSPs’ enantioselectivity,40
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
the actual effect of the individual substituents was not discussed. In recent years, another important branch of click chemistry, the thiol-ene reaction has attracted more and more attention.41 This reaction has almost all the advantages of the Cu(I) catalytic 1,3-dipolar cycloaddition reaction but it avoids the use of a metal catalyst. So it has increasingly been employed in developing separation materials.41-46 Our group has previously reported the fabrication of stable and structurally well-defined cationic native CD-CSPs which have higher enantioseparation abilities for dansyl (Dns) amino acids and carboxylic aryl acids than those for triazol linked native CD-CSP in chiral HPLC.47,48 This result demonstrates that thiol-ene click chemistry provides an excellent method for preparing cationic CD-CSP. In order to more fully explore the potential of thiol-ene reactions and to develop a novel structurally well-defined CD-CSP which incorporates multiple interactions such as H-bonding, π−π, electrostatic, dipole-dipole and inclusion interactions, the preparation of two novel CD-CSPs is reported, herein. Thiol-ene click chemistry was used to immobilize two multifunctional cationic compounds onto silica. The compounds were per(p-methyl)phenylcarbamoylated (CSP1) and per(p-chloro)phenylcarbamoylated-β-CDs (CSP2) where CSP1 bears an electron-donating –CH3 group and CSP2 bears an electron-withdrawing -Cl group on the p-positions of the phenyl moieties. The enantioselective abilities of both CSPs were comprehensively evaluated using HPLC with nearly fifty model racemates including isoxazolines, flavonoids，β-blockers and some other neutral and basic analytes in both reversed-phase (RP) and polar-organic (PO) modes. A comparative study on the CSP1 and CSP2 separations was also conducted to determine the effect of phenyl substituents on the CD-CPS’s enantioseparation performance.
ACS Paragon Plus Environment
Page 4 of 27
Page 5 of 27
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Experimental Section Materials and general procedures For the CSP preparation, azoisobutyronitrile (AIBN) and pyridine were purchased from Tianjin
Chemical
Reagents
(Tianjin,
China);
1-allylimidazole
was
purchased
from
Energy-Chemical (Shanghai, China); anhydrous N,N-dimethylformamide (DMF) was provided by Heowns (Tianjin, China); and p-methyl phenylisocyanate and p-chloro phenylisocyanate were provided by Ouhe Technology (Beijing, China). Mono-6A-deoxy-(p-tolylsulfonyl)-β-cyclodextrin (TsO-CD) and thiol functionalized silica were synthesized according to our previously reported procedure.47 Kromasil spherical silica gel (5 µm, 100 Å) was purchased from Eka Chemicals (Bohus, Sweden). For the chromatographic experiments, HPLC-grade methanol (MeOH), acetonitrile (ACN), triethylamine (TEA), n-hexane (Hex), isopropanol (IPA) and acetic acid were purchased by Concord Technology Co. Ltd (Tianjin, China). Ultra-pure water was prepared using a Milli-Q water purification system (Billerica, MA, USA). Isoxazoline racemates were provided by Tang’s group at Tianjin University, Tianjin, China.49 The acidic racemates were purchased from Sigma-Aldrich (Shanghai, China) and the other racemates were purchased from Energy-Chemical (Shanghai, China). The analyte structures are shown in supplementary Figure S-1. The 1H NMR spectra were recorded on a Bruker AVANCE III (400 MHz), Bruker Biospin (Fällanden, Switzerland). Solid state 13C NMR was performed on a Varian Infinityplus 300 NMR spectrometer (300 MHz, 7.0 T) (USA). Fourier-transform infrared (FTIR) spectra were collected on an AVATR360 supplied by Thermo Nicolet (USA). Thermal gravimetric analysis (TGA) was conducted on a NET-ZSCH STA409PC analyzer (Bavaria, Germany) in an air atmosphere at a heating rate of 10°C min−1. Elemental analysis was performed on a Var-ioMICRO CHNOS elemental analyzer (Elementar Analysensysteme, Hanau, Germany). Mass spectra were collected on LCQ Deca XP MAX system (Thermo Fischer, USA). Chromatographic analyses were performed on an Agilent 1100 HPLC system with a diode array detection (DAD) system (State college, PA, USA).
Synthesis of mono-6A-deoxy-6-(1-allylimidazolium)-per(p-methyl)phenylcarbamoylated-β-CD and CSP1 1-allylimidazole (1.1 mL, 10.2 mmol) was added to a solution of dry TsO-CD (1) (5.2 g, 4.1 mmol) in anhydrous DMF (10 mL). The reaction solution was stirred at 85 oC for 12 h under N2. The reaction solution was then poured into acetone (30 mL) and stirred for one hour followed by filtration. The obtained solid product was washed with acetone (2 × 20 mL) to afford mono-6A-deoxy-6-(1-allylimidazolium)-β-cyclodextrin tosylate (2) (4.7 g, yield 94%). The tosyl
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 27
group was thereafter replaced with a chloride ion through ion exchange (Amberlite IRA-900) to give mono-6A-deoxy-6-(1-allylimidazolium)-β-cyclodextrin chloride (3) (3.2 g). Under the protection of nitrogen, p-methylphenyl isocyanate (12.0 mL, 89.3 mmol) was added to the freshly prepared mono-6A-deoxy-6-(1-allylimidazolium)-β-cyclodextrin chloride (3) (3.2 g, 2.6 mmol) solution in dry pyridine (50 mL). The mixture was stirred at 85 oC for 18 h. After cooling to room temperature, the pyridine and unreacted p-methylphenyl isocyanate were removed via vacuum distillation. The residue was dissolved in ethyl acetate (270 mL) and then washed with saturated sodium chloride solution (70 mL). After the organic solvent was evaporated, the crude product was subjected to silica column chromatography to afford the target mono-6A-deoxy-6-(1-allylimidazolium)-per(p-methyl)phenylcarbamoylated-β-CD (4a) as a white solid (3.7 g). 1H NMR (400 MHz DMSO-d6) δ: 9.71-9.22 (m, NH, 20H), 7.37-7.01 (m, 2H-Ar and 6H-Ar, 40H), 6.88-6.72 (m, 3H-Ar and 5H-Ar, 40H), 5.48-4.08 (m, H-1, H-6, H-2, H-3, H-4 and H-5, 49H), 2.33-1.92 (m, C6-Ar-CH3,C2-Ar-CH3 and C3-Ar-CH3, 60H). Mono-6A-6-(1-allylimidazolium)-per(p-methyl)phenylcarbamoylated-β-CD (4a) and AIBN (21 mg) were added to a suspension of thiol functionalized silica (5) (2.5 g) in MeOH (30 mL). The reaction mixture was stirred at 40 oC for 24 h under N2. After filtration, the obtained material was washed with EtOH (3 × 20 mL) and then acetone (3 × 20 mL) followed by Soxhlet extraction with acetone for 24 h to afford CSP1 (2.4 g).
13
C NMR (300 MHz, 7.0 T) δ: 153.48, 135.03,
128.55, 119.48, 100.51, 98.38, 80.18, 74.09, 49.87, 18.34, 9.60.
Synthesis
of
mono-6A-deoxy-6-(1-allylimidazolium)-per(p-chloro)phenylcarbamoylated-β-CD and CSP2 Using a synthetic procedure similar to that for 4a, p-chlorophenyl isocyanate (12 mL, 106.8 mmol) was reacted with compound (3) to afford mono-6A-deoxy-6-(1-allylimidazolium)per(p-chloro)phenylcarbamoylated-β-CD (4b) (3.5 g) as a white solid. 1H NMR (400 MHz DMSO-d6) δ: 9.79-9.63 (m, NH, 20H), 7.68-7.31 (m, 2H-Ar and 6H-Ar, 40H), 7.02-6.68 (m, 3H-Ar and 5H-Ar, 40H), 5.47-4.09 (m, H-1, H-6, H-2, H-3, H-4 and H-5, 49H). Using a procedure similar to that for CSP1, the click chemistry reaction catalyzed by AIBN between 4b and the thiol functionalized silica afforded the target CSP2. 13C NMR (300 MHz, 7.0 T) δ: 155.62, 153.87, 138.07, 136.13, 129.00, 123.50, 120.59, 113.33, 80.83, 73.84, 49.36.
Column packing and HPLC experiments The prepared CSPs were packed into stainless-steel columns (150 mm × 4.6 mm I.D.) using a typical slurry-packing technique with MeOH as the solvent. The packing pressure was maintained for at least 30 min. The column was then rinsed and equilibrated with mobile phase before use.
ACS Paragon Plus Environment
Page 7 of 27
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
The column efficiencies of the CSPs were determined to be 5000−20000 plates·m-1 under reversed phase conditions (MeOH/H2O = 70/30, 1.0 mL min−1) using racemates as the separated analytes. Samples for reversed-phase chromatography were dissolved in MeOH/H2O (v/v = 1:1) at a concentration of 1 mg mL−1, whereas samples for the polar-organic mode were dissolved in isopropanol at the same concentration. The injection volume was 10 µL. Each solution was injected in triplicate and the average values are reported. Triethylammonium acetate buffer (TEAA) was prepared by dissolving 1% (v/v) TEA in ultra-pure water and then adjusting to the desired pH with acetate acid. All the buffers and samples were filtered through 0.22 µm membranes before use. Detection was performed at 220-300 nm. Calculations of capacity factor, k; selectivity, α; and resolution, Rs were according to USP standards.
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 27
Results and Discussion Synthesis and characterization of the CD clicked CSPs The
CD-CSPs
(CSP1
and
CSP2)
were
prepared
from
the
synthesized
A
mono-6 -deoxy-6-(1-allylimidazolium)-β-CD (3), which had been functionalized to afford mono-6A-deoxy-6-(1-allylimidazolium)-per(p-methyl)phenylcarbamoylated-β-CD
(4a)
and
A
mono-6 -deoxy-6-(1-allylimidazolium)-per(p-chloro)phenylcarbamoylated-β-CD (4b). CSP1 and CSP2 were readily achieved by anchoring 4a and 4b onto thiol functionalized silica (5) via a final thiol-ene click step (Scheme 1).
Scheme 1 Click synthesis of CSP1 and CSP2
CSP1 and CSP2 were characterized by FTIR, solid state 13C NMR spectroscopy, SEM, TGA and elemental analysis. In the FTIR spectra (Figure 1a), the peaks at around 1739 cm-1 are due to the presence of carbonyl
groups,
which
contribute
to
improved
chiral
recognition
abilities
perphenylcarbamoylated CD-CSPs compared to those for native CD-CSPs. The solid state
of 13
C
NMR spectrum (Figure 1b) of the unfunctionalized thiol silica has peaks at 9, 27 and 49 ppm which can be assigned to the three linkage carbons. After the click reaction, new peaks appear in the CSP spectra. The peaks between 60 and 110 ppm can be assigned to the carbon atoms on the CD glucose units. The peaks between 110 and 170 ppm are due to the carbons on the phenyl isocyanate and the peaks around 150 ppm belong to the carbonyl carbon. In the CSP1 spectrum, the peak at 19 ppm is due to the methyl substituent carbon. The other peaks between 0 and 50 ppm are from the aliphatic linkage carbon atoms. The comparable spherical morphologies of CSP1, CSP2 and the thiol silica (Figure S-2) indicate that the thiol-ene click chemistry provided a very moderate reaction environment for the immobilization of the CDs. The FTIR, 13C NMR, SEM and TGA analysis (Figure S-3) all show that the desired CD-CSPs were successfully prepared by the thiol-ene reaction. The surface concentration of the CD selectors was determined by elemental analysis
ACS Paragon Plus Environment
Page 9 of 27
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
(supplementary Table S-1). The amount of carbon, nitrogen and hydrogen in the CSPs all increased significantly compared to the amounts in the thiol silica. The surface loadings were calculated to be 0.11 µmol·m-2 in CSP1 and 0.13 µmol·m-2 in CSP2 according to the equation:
% .. .
% .
(1)
where c% is carbon percentage which was determined from the elemental analysis of carbon, nc is the number of carbon atoms per CD molecule (here 208 for CSP1 and 188 for CSP2), Mr is the molecular weight of the CD molecules anchored onto the thiol silica, and S.A. is the surface area of the silica gel (300 g·cm-2).
Figure 1. FTIR spectra and solid state 13C NMR spectra of CSP1 and CSP2
Calibration curves for the CSPs for the enantioseparation of 4MOPh-OPr were prepared (Figure S-4). For both CSP1 and CSP2, linear relationships were observed for the 4MOPh-OPr peak areas over a concentration range of 1–5000 µg·mL-1 with R2 = 0.99 for concentrations below 1000 µg·mL-1. The retention factor, selectivity and resolution were also analyzed as a function of concentration (supplementary Table S-2 and Table S-3). As the analyte concentration increased, both the retention factor and selectivity remained almost constant. However the resolution decreased slightly due to peak broadening. Distortion of the enantiomer peaks occurred at sample concentrations over 5 mg·mL-1. The ability of the columns to separate such high sample loads indicates a good capacity for the clicked CD columns.
Influence of HPLC parameters on the enantioseparation It is known that HPLC parameters such as flow rate and temperature play important roles in separation processes. Therefore the effects of these parameters on the CSPs’ enantioseparation abilities were evaluated. Again 4MOPh-OPr was chosen as the model analyte. Separations were performed with column temperatures ranging from 20 to 45 oC and flow rates from 0.2 to 1.2 mL·min-1 (Figure S-5). As expected, with increasing temperature, the retention time, selectivity
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 27
and resolution all decreased due to the enhanced solute diffusion. Increasing the flow rate from 0.2 to 1.2 mL·min-1, resulted in large decreases in the retention time and resolution but the selectivity remained constant for CSP1 and fluctuated slightly for CSP2. These results can be explained by the chromatography van Deemter equation. Taking the efficiency, resolution and analysis time into consideration, 30
o
C and 1.0 mL·min-1 were chosen to carry out the subsequent
enantioseparation experiments. In order to explore the versatility of the CSPs for enantioseparations in different HPLC modes,
the
functionality-tuned
enantioselectivities
of
the
click
prepared
phenylcarbamoylated-β-CD-CSPs were investigated using 48 recemates including isoxazoline derivatives, flavonoids, β-blockers and some other neutral and basic analytes in both RP and PO modes. Representative enantioseparation results for the two CSPs are shown in Figure 2 and Figure S-6.
Figure 2. Representative enantioseparation chromatograms of selected analytes o
Conditions: flow rate 1 mL·min-1; 30 C.
Enantionseparation performance of CSP1 in the RP mode In RP-HPLC, the organic solvent can greatly influence the enantioseparation. Two commonly used solvents MeOH and ACN were used for enantioseparations on CSP1 using several representative analytes. The results are shown in supplementary Table S-4. The isoxazoline derivatives were better separated with a mobile phase of MeOH/H2O than with ACN/H2O. This is in contrast to our previous work on a native β-CD-CSP, where isoxazoline derivatives were better separated using ACN as the organic modifier.48 MeOH favors the separation of the more polar Ph-OPr whereas ACN is better at resolving the more hydrophobic Ph-Ph and Ph-Py. Enantionseparation on phenylcarbamoylated-β-CD-CSPs is different from the interaction mechanism on native β-CD-CSPs where the guest molecules can enter the CD cavities to form inclusion complexes. Separation by phenylcarbamoylated-β-CD-CSPs mostly relies on hydrogen bonding, π-π interactions, steric effects and electrostatic interactions. In this situation, MeOH acts as a protic solvent and it can form H-bonds which plays an important role in the steric interactions
ACS Paragon Plus Environment
Page 11 of 27
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
between the analytes and the selectors. The enantioseparation of thirty two isoxazolines were performed on CSP1 using MeOH/H2O as the mobile phase (which gave the best Rs for most analytes with fair retention factors) and the separation results are given in Table 1. Most of the analytes were baseline resolved (Rs > 1.5) under these conditions and the resolution of some analytes almost reached 40. These encouraging results demonstrate that CSP1 exhibits powerful resolving ability for isoxazoline derivatives. The Ar-OPr category were much better separated than the Ar-Ph and Ar-Py categories. This indicates that the pyrrolidinone ring on the 5-position of Ar-OPr, which is different from the phenyl on Ar-Ph and the pyridine on Ar-Py, plays a more important role in the enantionseparation process. The nitrogen and oxygen atoms on the pyrrolidinone ring can form strong H-bonds and dipole-dipole interactions with the CSP carbamoyl groups. In addition the MeOH in the mobile phase may also be responsible for the better separation. Further the non-planar structure of the pyrrolidinone ring may also produce a stronger steric effect which enhances the enantiorecognition ability of CSP1 for Ar-OPr analytes. All the Ar-OPr analytes had baseline separations in short analysis times and almost half of the analytes had resolutions over 15. The resolution of 2ClPh-OPr is much lower than those for 4ClPh-OPr, 3ClPh-OPr and Ph-OPr which may be attributed to the formation of intramolecular H-bonding between the Cl and N on the isoxazoline ring. This is further supported by the relatively low resolutions for 2,4ClPh-OPr and 2,6ClPh-OPr. The same phenomena was also found for 3NPh-OPr and 2NPh-OPr (Rs = 0). Due to CSP1’s amazing recognition ability for 4NPh-OPr, the second enantiomer was not found even after 2 h using 70% MeOH. When the MeOH content was increased to 90%, the resolution of 4NPh-OPr was 24.89. These results indicate that the nitrogen oxygen double bond in 4NPh-OPr undergoes intense dipole-dipole interactions with the carbamoyl groups and the resulting π-conjugated -4NPh moiety may then form strong π-π interactions with the phenyl substitutions on the CD rims, which accentuates the enantiorecognition. Generally, the 3- and 4-positions on the phenyl ring are better substitution sites for the enantioseparation of the Ar-OPr category. Most of the analytes in the Ar-Ph and Ar-Py categories were also baseline separated on CSP1. However, the selectivities and resolutions are much lower than those for the Ar-OPr category. This verifies the importance of the pyrrolidinone substitutions on the analyte skeleton. Making a comparison of 4MetPh-Py and 4MetPh-2Py under the same conditions, the former has a much better separation which may be because the position of the nitrogen atom on 4MetPh-2Py is too close to the chiral center to provide an effective interaction site for enantioseparation. This suggests that a tiny difference in analyte structure may result in a large difference in enantioseparation.
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 27
Table 1 Optimal separation results of isoxazolines on CSP1 k1
k2
α
Rs
3ClPh-OPr
0.62
11.56
18.61
24.55
4ClPh-OPr
0.80
12.64
15.76
38.74
3NPh-OPr
1.21
18.11
14.89
38.39
MDOPh-OPr
0.93
6.94
7.45
23.00
Ph-OPr
0.63
3.48
5.50
19.62
4MOPh-OPr
0.79
2.68
3.36
15.19
Py-OPr
0.65
1.24
1.91
6.90
4MetPh-OPr
0.71
1.35
1.89
7.17
3FPh-OPr
0.73
0.95
1.31
2.90
2ClPh-OPr
0.79
0.95
1.21
2.19
2,6ClPh-OPr
1.09
1.18
1.08
0.98
2,4ClPh-OPr
0.86
0.89
1.03
