Comparison of Chiral Separation on Amylose and Cellulose

NotesBull. Korean Chem. Soc. 2003, Vol. 24, No. 2 239

Comparison of Chiral Separation on Amylose and CelluloseTris(3,5-dimethylphenylcarbamate)-Coated Zirconia in HPLC

In Whan Kim,a Jong Kwon Ryu,b Sung Duck Ahn,b Jung Hag Park,b,* Kwang-Pill Lee,c Jae Jeong Ryoo,c

Myung Ho Hyun,d Yoshio Okamoto,e Chiyo Yamamoto,e and Peter W. CarrfDept. of Chemical Education, Taegu University, Gyeongsan 712-714, Koreab

Dept. of Chemistry, Yeungnam University, Gyeongsan 712-749, Korea

c

Dept. of Chemical Education, Kyungpook National University, Daegu 702-701, Korea

d

Dept. of Chemistry, Pusan National University, Busan 609-735, Korea

e

Dept. of Applied Chemistry, Graduate School of Engineering, Nagoya University, Nagoya 464-8603, Japan

f

Dept. of Chemistry, University of Minnesota, MN 55455, USA

Received August 2, 2002Key Words : Chiral stationary phase, Amylose and cellulose tris(3,5-dimethylphenylcarbamate), Zirconia,HPLC

a

HPLC separation method based on chiral stationary phases(CSPs) has become one of the most attractive approaches tochiral separations, due to their simplicity for determiningoptical purity and easy extension to the semipreparative andpreparative scales.1 One of the major problems in usingmany CSPs is their narrow range of analyte applicability;they can only discriminate a limited number of specific typesof chemical entities, and it is frequently necessary toderivatize the compounds of interest to achieve separation.2On the other hand, the polysaccharide derivative-based CSPsdeveloped by Okamoto and co-workers3-6 have proven to behighly versatile and rugged. Okamoto reported the resolutionof 64% of 483 racemic mixtures on cellulose tris(3,5-di-methylphenyl carbamate) (CDMPC) and 80% were success-fully resolved on either the cellulose or the correspondingamylose carbamate (ADMPC).7

Fast method development, high efficiency, rapid resolutionof enantiomers, and robustness are the main criteria forchiral separation methods, especially in the pharmaceuticalindustry. These priorities require stable CSPs capable ofachieving baseline separations in the minimum time, whichultimately means high selectivity and efficiency. Silica is themost popular choice for support for HPLC stationary phaseligands due to the mechanical strength, wide range of particleand pore dimensions, pore structure and well-establishedsilane chemistry. However, silica and bonded phase ligandshave stability problems. Silica dissolves in mobile phasebuffered at or above pH 8 with loss of bonded phase ligandand column packing.8 Loss of organosilanes from the silicasurface via hydrolysis proceeds rapidly at low pH (<3) andat higher temperature (40oC). These deficiencies of thecolumn packing create problems of poor injection reproduci-bility, poor peak shape, and high backpressure, thus makingmethod development tasks difficult. Over the last decade,zirconia has received considerable attention as a stationaryphase support for HPLC.9,10 Zirconia particles are very

*

robust material; they show no detectable signs of dissolutionover the pH range from 1 to 14 and have been used forprolonged periods at temperatures up to 200oC in chromato-graphic separations. We recently reported preparation ofzirconia based CSPs with cellulose, bovine serum albuminand β-cyclodextrin for use in either normal or reversed-phase LC separation of chiral compounds.11-14

In this work we compared chromatographic performancesof chiral separation for ADMPC and CDMPC coated on 3-µm zirconia particles by measuring retention of a set ofracemic compounds on them. We used narrow-bore (1-mmID) columns that lead to many advantages such as lowconsumption of both mobile and stationary phases etc.15-17

Experimental Section

Reagents and materials. All reagents used for the prepara-tion of the stationary phase were reagent grade or better.Microcrystalline cellulose and amylose were purchased fromNakarai Chemicals (Japan). 3,5-Dimethylphenyl isocyanate,N,N-dimethylacetamide and pyridine were obtained fromAldrich (Milwaukee, USA). Zirconia, having a mean poresize of 30 nm and a mean particle diameter of 3 µm, wasobtained from ZirChrom Separations (Anoka, USA). Acetone,and 2-propanol were HPLC grade (J.T. Baker, Phillipsburg,USA). n-Hexane and tetrahydrofuran (THF) were purchasedfrom EM Sciences (Gibbstown, USA). The racemic com-pounds studied are shown in Figure 1. All are commerciallyavailable. Solutions at a concentration of 0.1 mg/mL wereprepared by dissolving the compounds in the mobile phase.Preparation of CDMPC and ADMPC. CDMPC andADMPC were synthesized as previously reported5,18 andwere characterized by elemental analyses, IR and NMRspectroscopy. The data indicated that hydroxyl groups ofcellulose and amylose were almost completely converted tothe corresponding carbamate groups.

Preparation of CDMPC and ADMPC coated zirconia.To dehydroxylate zirconia's surface, the particles were

Corresponding author. E-mail: jhpark@yu.ac.kr

240 Bull. Korean Chem. Soc. 2003, Vol. 24, No. 2Notes

Figure 1

. Structures of racemic compounds.

heated at 750oC for 5 h and cooled over phosphorus pent-oxide before use. Typically, 1.0 g of particles was suspendedin 10 mL of THF and sonicated under vacuum for 15 min toeliminate the air from the pores. Polymer loading of 4% byweight was chosen since this loading has been shown tooffer excellent chiral recognition ability and column effici-ency.11,13 The corresponding amount of CDMPC or ADMPCwas dissolved in 10 mL of THF and the solution was addedto the slurry of zirconia in THF using a syringe pump at arate of 0.04 mL/min (~4 h). The suspension was stirredovernight and then the solvent was slowly removed by rotaryevaporation at room temperature. Finally, the particles weredried in vacuum at 50oC.

Chromatography. Packing materials were suspended in a(1:1) hexane/2-propanol mixture and packed into 25 cm×1 mm (ID) columns using the downward slurry method atca. 7000 psi. 2-Propanol was employed as the displacing

solvent. A chromatographic system consisting of a Model7520 injector with a 0.5-µL internal loop (Rheodyne, CA,USA), a Model 530 column oven (Alltech, IL, USA) set at30oC and a Linear Model 200 UV/VIS detector (Alltech, IL,USA) with a 0.25-µL flowcell set at 254 nm was used. AHewlett-Packard (Avondale, CA, USA) Series 3365 integrat-ing recorder was used to record chromatograms. The mobilephases were mixtures of 2-propanol and hexane (2/98 or10/90 v/v%). They were filtered through a membrane filterof 0.5-µm pore size and degassed prior to use. The flow ratewas 200 µL/min. The dead time was estimated by using1,3,5-tri-tert-butylbenzene as unretained compound.19

Results and Discussion

The performance of a column packed with ADMPC and

CDMPC-zirconia is shown for the resolution of trifluoro-

Figure 2. Chromatograms for the separation of racemic trifluoroanthryl ethanol on (a) ADMPC- and (b) CDMPC-zirconia. Columndimension; 25×0.1 cm I.D. Mobile phase; 90:10 (v/v %) n-hexane: 2-propanol. Flow rate; 0.2 mL/min. Column temperature; 25 oC.

Notes

Table 1. Chromatographic Data on ADMPC- and CDMPC-Zirconiain Hexane/2-propanolCompound Mobile PhaseADMPCCDMPCNo.(v/v%)

k1aαbk1aαb190:100.861.550.892.75298:29.551.009.241.10398:28.971.080.871.86498:23.201.062.801.18598:24.341.002.591.15698:210.021.076.771.1690:100.581.090.691.00798:23.171.072.871.08898:21.822.411.771.1190:100.471.870.581.17990:107.821.307.821.301098:20.461.340.733.0190:100.251.200.352.231198:20.861.500.762.6290:100.371.350.252.121298:215.061.226.011.001390:103.781.362.081.0014

98:21.681.271.181.0090:10

0.50

1.04

0.30

1.00

a

Retention factor for the first eluting enantiomer. b

Selectivity factor.

anthryl ethanol in 90:10(v/v) hexane/2-propanol (Fig. 2).Retention factors (k) for this analyte under the conditionsused are small but its enantiomers are baseline resolved withseparation factors of 1.55 and 2.75 on ADMPC- and CDMPC-zirconia, respectively. Separation data of twelve racemiccompounds are listed in Table 1. Most of the racemiccompounds studied were well resolved on the two CSPs.Retention and chiral selectivities of ADMPC- and CDMPC-zirconia vary extensively with the type of chiral compoundsas can be seen in Figure 3. For seven alcohols (1-7) investi-gated selectivity factors are in general greater on CDMPCthan ADMPC while retention is always longer on ADMPCthan CDMPC. For two bases (8, 9) retention values aresimilar on the two columns but chiral selectivity is better onADMPC-zirconia than on CDMPC-zirconia. For two cyclicethers (10-11) retention is very short on both columns butselectivity of CDMPC is much greater than that for ADMPC.For two lactones and cyanide (12-14) both retention andselectivity are greater on ADMPC than on CDMPC. TheCDMPC- and ADMPC-coated zirconia CSPs show comple-mentary chiral recognition capability for types of theracemates studied.

The stability of the polysaccharide-zirconia columns werechecked by measuring retention factor of the first elutingenantiomer of Tröger's base after passage of every 500column volume of the eluent through the columns. Therewas only less than 2% decrease in retention factor of the testsolute for the both columns after 6,000 column volume. Thehigh enantioselectivity of the zirconia CSPs may allow forthe use of a shorter column for reduced analysis time andsolvent consumption.

Bull. Korean Chem. Soc. 2003, Vol. 24, No. 2 241

Figure 3. Comparison of retention and chiral selectivity ofADMPC- and CDMPC-zirconia. When separations were carriedout at two different mobile phase compositions results for 98:2 n-hexane: 2-propanol were plotted. Solid bar, ADMPC; Open bar,CDMPC. Solutes: 1, trifluoroanthryl ethanol; 2, α-trifluoromethyl-benzyl alcohol; 3, α-methyl-1-naphthalene methanol; 4, 1-phenyl-1-propanol; 5, 1-phenyl-2-propanol; 6, 3-phenyl-1-butanol; 7, 1-phenyl-1-butanol; 8, Tröger's base; 9, 3,5-dinitrobenzoyl-α-methyl-benzylamine; 10, trans-stilbene oxide; 11, 4-phenyl-1,3-dioxane;12, γ-phenyl-γ-butyrolactone; 13, γ-(2-naphthyl)-γ-butyrolactone;14, α-methylbenzyl cyanide.

Acknowledgment. This work was supported by the KoreaResearch Foundation grant (2001-015-DP0288). IWKacknowledges financial support by the Taegu Universityresearch grant (2002).

References

1.Francotte, E. In Chiral Separations: Applications and Technology;

Ahuja, S., Ed.; American Chemical Society: Washington, 1997;Chapter 5.

2.Dingenen, J. In A Practical Approach to Chiral Separations by

Liquid Chromatography; Subramanian, G., Ed.; VCH: New York,1994; Chapter 6.

3.Okamoto, Y.; Kawashima, M.; Yamamoto, K.; Hatada, K. Chem.

Lett. 1984, 739. 4.Okamoto, Y.; Kawashima, M.; Hatada, K. J. Am. Chem. Soc.

1984, 106, 5357.

5.Okamoto, Y.; Kawashima, M.; Hatada, K. J. Chromatogr. 1986,

363, 173.

6.Okamoto, Y.; Yashima, E. Angew. Chem. Int. Ed. 1998, 37, 1020.

242 Bull. Korean Chem. Soc. 2003, Vol. 24, No. 2

7.Okamoto, Y.; Kaida, Y.; Aburatani, R.; Hatada, K. In Chiral

Separations by Liquid Chromatography; Ahuja, S., Ed.; ACSSymposium Series 471, American Chemical Society: Washington,DC, 1991; pp 101-113. 8.Kirkland, J. J.; van Straten, M. A.; Claessans, H. A. J.

Chromatogr. 1995, 691, 3. 9.Nawrocki, J.; Dunlap, C. J.; Carr, P. W.; Blackwell, J. A.

Biotechnol. Prog. 1994, 10, 561.

10.Jackson, P. T.; Carr, P. W. Chemtech. Oct. 29, 1998.11.Castells, C. B.; Carr, P. W. Anal. Chem. 1999, 71, 3013.12.Castells, C. B.; Carr, P. W. J. Chromatogr. 2000, 904, 17.

13.Park, J. H.; Ryu, J. K.; Park, J. K.; McNeff, C. V.; Carr, P. W.

Chromatographia 2001, 53, 405.

Notes

14.Park, S. Y.; Park, J. K.; Park, J. H.; McNeff, C. V.; Carr, P. W.

Microchem. J. 2001, 70, 179.

15.Hôrvath, C. G.; Preiss, B. A.; Lipsky, S. R. Anal. Chem. 1967, 39,

1422.

16.Scott, R. P. W.; Kucera, P. J. Chromatogr. 1976, 125, 251.

17.Ishii, D. In Introduction to Microscale High-Performance

Liquid Chromatography; Ishii, D., Ed.; VCH: Weinheim, 1988;Chapter 1.

18.Okamoto, Y.; Aburatani, R.; Fukumoto, T.; Hatada, K. Chem. Lett.

1987, 1857.

19.Koller, H.; Rimbock, K. H.; Mannschreck, A. J. Chromatogr.

1983, 282, 89.