Development of chiral stationary phases for high-performance liquid chromatographic separation

Develop l phases fo m chromat r Mengling Tang, Jin eip Chromatography technology b anti preparation of biochemicals ( In this review, we focus o nce l recognition mechanism, appli disco the rational design of future C Crown Copyright ª 2012 Pub Keywords: Biochemical; Chiral st al che chromatography (HPLC); Preparati Abbreviations: See, Appendix afte Mengling Tang, Jing Zhang, Shulin Zhuang, Weiping Liu* MOE Key Lab of Environmental Trends Trends in Analytical Chemistry, Vol. 39, 2012 Fax: +86 571 8898 2341.; preparation in chromatographic technol- ogy have developed rapidly. Direct reso- lution using CSPs in high-performance liquid chromatography (HPLC) is cur- rently the most important separation method, when compared with other methods [e.g., gas chromatography (GC), supercritical fluid chromatography (SFC), thin layer chromatography (TLC), capil- lary electrophoresis (CE) and capillary electrochromatography (CEC)], especially at the preparative scale. More than 110 CSPs developed for HPLC are now com- mercially available [1]. The general pre- parative method is simple (i.e. a single enantiomer of a chiral molecule is immo- bilized onto a solid support and, as ana- lytes having enantiomeric dissimilarities of stability and retentivity elute, the chiral analytes are separated) [2]. Chiral recognition and enantiomer dis- tinction are fundamental phenomena in natural and engineered systems. Recently, a number of researchers investigated the specific molecular recognition mecha- several modern CSPs, their chiral recogni- tion and their separation mechanisms, and summarized themost important theoretical principles using the ‘‘three-point attach- ment model’’ (Fig. 1). Cancelliere et al. [2] proposed some enlightening ideas: (1) CSPs with a receptor-like enantiose- lectivity seem to be powerful to study the interactions that drive and regu- late host-guest associations at molec- ular levels; (2) for molecular selectors coming from the natural pool, their modes of ac- tion can be investigated to provide structural and energetic information about chiral systems where associa- tion is mainly driven by hydrophobic forces. The present article focuses mainly on CSPs developed for HPLC and overviews the main structure, recognition mecha- nism, limitations and applications of commercially-available CSPs in HPLC. Classical CSPs are classified into ligand exchange, cyclodextrin (CD), Pirkle- Health, College of Environmental and Resource Sciences, Zhejiang University, Hangzhou 310058, China *Corresponding author. Tel.: +86 571 8898 2740; Remediation and Ecosystem E-mail: [email protected] 180 0165-9936/$ - see front m ment of chira r high-perfor ographic sepa g Zhang, Shulin Zhuang, W ased on chiral stationary phases (CSPs) for en e.g., drugs, foods, fragrances and pollutants). n the development of CSPs for high-performa cations and limitations of classical CSPs, newly SPs on the basis of computational chemistry. lished by Elsevier Ltd. All rights reserved. ationary phase (CSP); Chromatography; Computation on; Recognition; Rational design; Resolution r Reference section 1. Introduction In the past two decades, the chiral sepa- ration materials used for analysis and nisms of CSPs and corresponding selector atter Crown Copyright ª 2012 Published by Elsevier Ltd. All rights stationary ance liquid ation ing Liu oseparation is widely used for resolution and iquid chromatography (HPLC), including the vered types of CSP, and also the methods for mistry; Enantioseparation; High-performance liquid moieties with HPLC. However, the under- lying molecular mechanisms are still elusive. La¨mmerhofer [3] recently overviewed type, synthetic polymers, polysaccharide, reserved. doi:http://dx.doi.org/10.1016/j.trac.2012.07.006 There are currently a handful of classical CSPs that have oint in Trends in Analytical Chemistry, Vol. 39, 2012 Trends appeared to satisfy the majority of chiral separation needs in various chemical fields, including pharmaceu- tical discovery, preparation of food additives and fra- grances, and pollutant analysis using HPLC. The CSPs can be divided into naturally occurring, semi-synthetic and synthetic CSPs by the origins and sources of chiral selectors and they can also be classified by their chemical structure. 2.1. Ligand-exchange CSPs The first complete separation of a racemic analyte [amino acid (AA)] using chromatographic techniques was realized in the late 1960s by Davankov, who used ligand-exchange CSPs to immobilize proline onto a polystyrene support with Cu(II) ions in the eluent [8]. It was based on the diastereomeric ternary coordination protein, macrocyclic antibiotic and crown ether-based CSPs [4–7]. Many types of CSP or chiral selectors with potential applications for CSP preparation were devel- oped, but not commercialized. We discuss the modified classical CSPs and newly discovered cyclofructan, metal complex-bonded, boron-containing, ionic liquids, asym- metric organocatalyst and monolithic column CSPs. The rational design of CSPs suitable for chiral separation in chromatography technology warrants further study, and new techniques based on high-throughput screen- ing and computational chemistry [e.g., molecular mod- eling/simulation and quantitative structure-activity relationship/quantitative structure-enantiomers relation relationship (QSAR/QSEER) analysis] are proposed as the future direction of CSP design. 2. The classical commercial CSPs Figure 1. Three-p complexes formatted by chiral selector, metal ion and analyte, in which the chiral selector and the analyte are adsorbed in the coordination sphere of the central metal ions. After synthesis and test of the chiral ligand- exchanging polystyrene-type stationary phases in 1968, the analogous silica-bonded stationary phases emerged a decade later, and the other polymer-type CSPs were developed subsequently. The epoxy-activated and (L)- proline-incorporating silica was the first chiral-bonded stationary phase on the market (Chiral ProCu, Serva, Heidelberg, Germany). Of the many polymeric and silica- bonded CSPs for ligand-exchange chromatography, several CSPs became commercially available (Table 1). Ligand-exchange CSPs are available and used primar- ily for the enantiomeric separation of classes of biologi- cally-relevant compounds (e.g., AAs and hydroxyl acids). The enantiomers must be able to form bidentate chelates to be separated on this kind of CSP. The ligand- exchanging column 5 lm TSK gel Enantio L1 was tested on 18 essential AAs combined with Cu(II), Zn(II) and Ni(II) ions [9]. Acceptable results were achieved with aromatic AAs only, whereas Zn(II) ions performed well with hydroxyl-containing AAs, and the Cu(II) ions were the most universal [10]. The ChiralpakWH column could well resolve all enantiomeric pairs of the drug imidapril [11] and the enantiomers of imidapril antifungal agents (e.g., econazole, miconazole, and sulconazole), which contain N, O or S atoms in their molecules [12]. 2.2. Cyclodextrin CSPs CD has a long history of development dated back to the 1970s [1] and CD-bonded stationary phases developed by Armstrong�s group were the first commercially- available reversed-phase CSPs. CDs are a class of toroid- shaped cyclic oligosaccharides produced naturally from starch by the enzyme CD transglycosylase, containing six (a-CD), seven (b-CD) or eight (c-CD) a-(1,4)-linked D- glucopyranose units [13]. The interior of the cavity contains no hydroxyl groups that are hydrophobic, and the outer surface is hydrophilic. The chiral recognition mechanism is based on not only analyte interactions with hydrophilic CD surface, but also the inclusion of a bulky hydrophobic group of the analyte, preferably aromatic groups, into the hydrophobic cavity of the CD. Hydrogen bonding and dipole-dipole interaction with the hydroxyl groups at C-2 and C-3 positions at the mouth of teraction model. the cavity also contribute to the chiral recognition [14]. The enantioseparation abilities of CD can be enhanced by the derivatization of its hydroxyl groups. b-CD is the most widely-used chiral separation mate- rial in HPLC technology. Various types of CD-bonded phase columns designed for chiral separation were reviewed by Bressolle et al. in 1996 [15]. The commercialized CD CSPs are listed in Table 1. Most CD CSPs are available for only reversed-phase HPLC to separate polar-organic compounds. Differing from other http://www.elsevier.com/locate/trac 181 Table 1. Classical commercialized CSPs used in HPLC Chiral selector Trade name Manufacturer Ligand-exchange CSPs (LL)-Proline Chiralpak WH Daicel, Japan Chiralpak WH J.T. Baker, USA (LL)-Hydroxyproline Nucleosil Chiral-1 Macherey-Nagel, Germany (LL)-Hydroxy-proline Chiral-Si 100 L-HyProCu Serva, Germany N,S-Dioctyl-(D)-penicillamine Chirex 3126 Phenomenex, USA Sumichiral OA-5000 Sumika, Japan Chirex D-penicillamin Phenomenex, USA Binuclear Copper (II) of N-salicyliden-(R)-2-amino-1,1- bis(2-butoxy-5-tert-butylphenyl)-3-phenylpropanol Sumichiral OA-5500 Sumika, Japan (R,R)-tartaric acid-(R)-1-(a-naphthyl)ethylamide Sumichiral OA-6000 Sumika, Japan (R,R)-tartaric acid-(S)-valine-(S)-1-(a-naphthyl) ethylamide Sumichiral OA-6100 Sumika, Japan (D)-Pipecolinic Chirosolve D-Pipec. JPS Chimie, Switzerland N,N-Dioctyl-(LL)-alanine MCI CRS10 W Mitsubishi Kasei, Japan N,N-Dioctyl-(D)-alanine MCI CRS10WD Mitsubishi Kasei, Japan Chiralpak MA(+) Daicel, Japan Cyclodextrin CSPs b-CD Cyclobond I Astec, USA Cyclobond I Serva, Germany Cyclobond I Technicol, UK Cyclobond I Rainin, USA Cyclobond I ICT, Germany Cyclobond I 2000 Astec, USA Cyclobond I 2000 Technicol, UK ChiraDex Astec, USA Ultron ES-CD Shinwakako, Japan c-CD Cyclobond II Astec, USA a-CD Cyclobond III Astec, USA (R)-naphthylethylisocyanate derivatized b-CD Cyclobond I 2000 RN Astec, USA Acetylated b-CD Cyclobond I Ac Astec, USA Cyclobond I 2000 Ac Astec, USA Cyclobond I 2000 Ac Serva, Germany b-CD (S) and (R,S)-hydroxypropylderivative Cyclobond I RSP Astec, USA Beta-RSP-2000 Astec, USA Cyclobond I RSP Rainin, USA b-CD-(S)-naphthylethyl carbamate derivative Cyclobond I SN Astec, USA Cyclobond I 2000 SN Astec, USA b-CD-(S)-hydroxypropylether derivative Cyclobond I SP Astec, USA Cyclobond I SP Rainin, USA Carboxymethylated b-CD OR-Pak CDBS-453 Showa Denko, Japan b-CD phenylcarbamate Ultron ES-PhCD Shinwakako, Japan Perphenylcarbamate-heptakis(6-azido-6-deoxy) b-CD CHIDEX-MKP Chiral Sciences &Technologies Pte. Ltd, Singapore Perphenylcarbamate-b-CD CHIDEX-SKP Chiral Sciences &Technologies Pte. Ltd, Singapore Permethylated-b-CD Nucleodex b-PM Macherey Nagel, Germany Permethylated-c-CD Nucleodexc-PM Macherey Nagel, Germany Pirkle-type CSPs (R)-N-(3,5-Dinitrobenzoyl)phenylglycine (R)-DNBPG covalent Regis, USA Pirkle 1A covalent Regis, USA Sumichiral OA 2000 Sumika, Japan Pirkle 1-A Technicol, UK Chirex 3001 Phenomenex, USA Hi-Chrom Pirkle HiChrom Ltd, UK (continued on next page) Trends Trends in Analytical Chemistry, Vol. 39, 2012 182 http://www.elsevier.com/locate/trac Table 1 (continued) Chiral selector Trade name Manufacturer (S)-N-(3,5-Dinitrobenzoyl)phenylglycine DNBPG covalent Regis, USA Chirasep DNBPG Merck, Germany (R)-N-(3,5-Dinitrobenzoyl)phenylglycine Pirkle 1A ionic Regis, USA (S)-N-(3,5-Dinitroben zoyl)leucine Pirkle-Leu covalent Regis, USA Rexchrom Pirkle covalent Regis, USA L-leucine J.T. Baker, USA S-DNBL covalent Hi-Chrom Reversible-Leucine covalent L-Leucine covalent HiChrom Ltd, UK Interchim, France (R)-3,5-Dinitrobenzoyl-leucine Hi-Chrom Pirkle Covalent Leucine HiChrom Ltd, UK (S)-N-(3,5-Dinitroben-zoyl)leucine S-DNBL ionic Regis, USA (S)-N-(3,5-DNB)-phenylalanine Chirachrom A1 Interchim, France (S)-dinitrobenzoyltyrosineamide ChyRoSine-A Sedere, France (R)-DNB-Aminophosphonate derivative (R)-alpha-Burke 1 Regis, USA Interchim, France (R)-DNB-Aminophosphonate derivative (R)-alpha-Burke 2 Regis, USA (S)-DNB-Aminophosphonate derivative (S)-alpha-Burke 2 (S)-alpha-Burke 2 Regis, USA Phenomenex, USA Mono-3,5-dinitrobenzoyl-(R,R)-diphenylethanediamine (R,R) ULMO Regis, USA Mono-3,5-dinitrobenzoyl-(S,S)-diphenylethanediamine (S,S) ULMO Regis, USA 3-(3�,5�-Dinitrobenzoyl)amino-(R)-3-phenyl-(R)- 2-tBu-propanoic acid undecyl ester (R,R)-beta-GEM 1 Regis, USA (3R,4S)-4-(3,5-Dinitrobenzamido)-3-[3-(dimethylsilyloxy)propyl]-1, 2,3,4-tert-rahydrophenanthrene (S,S) Whelk-O1 Regis, USA (3R,4S)-4-(3,5-Dinitrobenzamido)-3-[3-(trioxysilyl)-propyl]-1, 2,3,4-tetrahydro-phenanthrene (R,R) Whelk-O2 Regis, USA (R)-a-(1-Naphthyl)ethylamine urea KK-Carnu YMC, USA (S)-Naphthylethylurea SNU5 (S)- Naphthylurea Supelco, USA Shandon, UK (LL)-Tartaric acid-(LL)-pheny-lethylamine Nucleosil chiral 2 Macherey-Nagel, Germany cis-N-3-(3,5-Dinitrobenzoyl)-amino-4-phenyl-b-lactam Pirkle 1-J Regis, USA (R,R)-N,N�-3,5-Dinitrobenzoyl-trans-1,2-diaminocyclohexane (R,R)-DACH-DNB Regis, USA N-Dinitrobenzoyl-(R)-1-Naphthylglycine Sumichiral OA-2500 Chirex 3005 Sumichiral OA-2500R Sumika, Japan Phenomenex, USA Sumika, Japan N-Dinitrobenzoyl-(S)-1-Naphthylglycine Sumichiral OA-2500S Sumika, Japan (R)-N-(3,5-Dinitrobenzoyl)-1-naphthylglycine Sumichiral OA-2500I Sumika, Japan 3,5-Dinitrophenylaminocarbonyl-(S)-valine Sumichiral OA-3100 Sumichiral OA-3100 Chirex 3010 Sumichiral OA-3100 Sumika, Japan Regis, USA Phenomenex, USA YMC, USA N-(3,5-Dinitrophenylcarbamoyl)-S-tertiobutylglycine Sumichiral OA-3200 Sumika, Japan N-(3,5-Dinitrophenylcarbamoyl)-(D)-phenylglycine Sumichiral OA-3300 Sumika, Japan N-(S)-1-(1-naphthyl)ethy-laminocarbonyl-(LL)-valine Sumichiral OA-4000 Sumika, Japan N-(R)-1-(a-Naphthy-l)ethylaminocarbonyl-(S)-valine Sumichiral OA-4100 Sumichiral OA-4100 Chirex 3014 Sumika, Japan Interchim, France Phenomenex, USA (S)-Benzoproline-(S)-1-naphthylethylurea derivative Sumichiral OA-4800 Sumika, Japan (S)-Benzoproline-(R)-1-naphthylethylurea derivative Sumichiral OA-4900 Chirex 3022 Sumichiral OA-4900 Sumika, Japan Phenomenex, USA Rainin, USA (S)-Proline-(S)-1-naphthylethylurea derivative Sumichiral OA-4400 Chirex 3017 Sumika, Japan Phenomenex, USA (R)-1-(1-Naphthyl)ethylaminocarbonyl-(LL)-proline Sumichiral OA-4500 Chirex 3018 Sumika, Japan Phenomenex, USA tert-Butylcarbamoyl-quinine Chiral AX QN-1 Bischoff, Germany (continued on next page) Trends in Analytical Chemistry, Vol. 39, 2012 Trends http://www.elsevier.com/locate/trac 183 Table 1 (continued) Chiral selector Trade name Manufacturer Synthetic polymers (+)-Poly(triphenylmethylmethacrylate) Chiralpak OT(+) Daicel, Japan J.T. Baker, USA Poly-[(S)-N-Acryloylphenylalanine-ethylester] Chiraspher Merck, Germany Bodman Chemical, USA Curtin-Matheson Scientific, USA O,O�-di(3,5-dimethyl-ben-zoyl)-(2R,3R)-diallyl-tartardiamide Kromasil CHI-I EKA Chemical, Sweden O,O�-di(4-tBu-phenyl)-(2R,3R)-diallyl-tartardiamide Kromasil CHI-DMB Kromasil CHI-II Kromasil CHI-TBB EKA Chemical, Sweden EKA Chemical, Sweden EKA Chemical, Sweden N-(2-acryloylamine-(1R,2R)-cyclohexyl)-acrylamide radical-polymerized with 4,4�-azo-bis-(4-cyano)-valeric acid (R,R)-P-CAP Astec, USA N-(2-acryloylamine-(1S,2S)-cyclohexyl)-acrylamide radical-polymerized with 4,4�-azo-bis-(4-cyano)-valeric acid (S,S)-P-CAP Astec, USA Polysaccharide CSPs Microcrystalline cellulose triacetate CTA-I Merck, Germany MCTA Merck, Germany CEL-AC-40-XF Chiralcel CA-1 CTA-I Conbrio-TAC Macherey-Nagel, Germany Daicel, Japan Fluka, Switzerland Perstop Biochem, Sweden Microcrystalline tribenzoylcellulose Chiral Tribencel Macherey-Nagel, Germany Cellulose triacetate Chiralcel OA Daicel, Japan Optipak-TA Waters, USA Chiralcel OA J.T. Baker, USA Cellulose tribenzoate Chiralcel OB Daicel, Japan Chiralcel OB-H Daicel, Japan Chiralcel OB J.T. Baker, USA Cellulose triphenylcarbamate Chiralcel OC Daicel, Japan Cellulose tris(3,5-dimethylphenylcarbamate) Chiralcel OD Daicel, Japan Chiralcel OD J.T. Baker, USA Chiralcel OD-H Daicel, Japan Chiralcel OD-R Daicel, Japan Chiralcel OD-RH Daicel, Japan Cellulose tris(4-Chloro phenylcarbamate) Chiralcel OF Daicel, Japan Chiralcel OF J.T. Baker, USA Cellulose tris(4-Methyl phenylcarbamate) Chiralcel OG Daicel, Japan Cellulose tris(4-methylbenzoate) Chiralcel OJ Daicel, Japan Chiralcel OJ-R J.T. Baker, USA Chiralcel OJ-RH Daicel, Japan Exp B101 Bio-Rad RSL, Belgium Cellulose tricinnamate Chiralcel OK Daicel, Japan Chiralcel OK J.T. Baker, USA Amylose tris(3,5-dimethyl phenylcarbamate) Chiralpak AD Daicel, Japan Chiralpak AD-RH J.T. Baker, USA Chiralpak AD-R Daicel, Japan Amylose tris[(S)-a-phenethyl]carbamate Chiralpak AS Daicel, Japan Chiralpak AS J.T. Baker, USA Chiralpak AS-H Daicel, Japan Amylose tris(3,5-dimethylphenylcarbamate) Chiralpak IA Daicel, Japan Cellulose tris(3,5-dimethylphenylcarbamate) Chiralpak IB Daicel, Japan Macrocyclic antibiotic CSPs Ristocetin A Chirobiotic R Astec, USA (continued on next page) Trends Trends in Analytical Chemistry, Vol. 39, 2012 184 http://www.elsevier.com/locate/trac Trends in Analytical Chemistry, Vol. 39, 2012 Trends CDs, Cyclobond I RN or SN perform well in normal-phase HPLC because of the ability to serve as a p-electron donor, so an analyte with a p-electron acceptor can be separated on this kind of CSP [16]. To date, a wide variety of compounds (e.g., b-blockers, alkaloids, carboxylic acids, and neutral molecules) have been successfully separated on these CD CSPs. b-blockers were observed to have an improved resolution using the organic polar mode on native b-CD or c-CD phases [17]. Sulfonated CD-derivative stationary phases can be used for the separation of neutral, cationic, and anionic substances based on their multiple interactions (e.g., hydrophobic and electrostatic). Sulfonyl-b-CD, the first charged CD CSP prepared by Stalcup and Gahm, was used successfully for the enantioseparation of 33 race- mates of biological or pharmaceutical interest [18]. A CD phase based on mixed functional silica, comprising both carbamate bonded b-CD and diol functions, was devel- oped for the separation of hexobarbital and chlorphe- niramine [19]. Table 1 (continued) Chiral selector Amphoteric Teicoplanin Vancomycin Teicoplanin aglycone Methylated Teicoplanin aglycone Vancomycin aglycone Crown ether-based CSPs (R)-18-Crown-6-ether (S)-18-Crown-6-ether (+)-(18-Crown-6)-2,3,11,12-tetracarboxylic acid (S,S)-Pseudo-18-crown-6-ether 2.3. Pirkle-type CSPs The Pirkle-type CSP was first developed by Pirkle�s group at the end of the 1970s [20]. A variety of Pirkle-type CSPs were subsequently prepared by this group and other groups. This type of CSP is formed as low molec- ular mass selectors bonded to a solid support (e.g., silica). They were named as Pirkle-type, brush-type, p-donor and p-acceptor CSPs. With the brush-type distribution of chiral molecules on the surface of the inert matrix, they are easily accessible to the analytes. The main interac- tion between the chiral selector and the analyte is p-p association, but other interactions can also be involved simultaneously (e.g., hydrogen bonding, dipole-dipole interaction and steric repulsion). Good loading capability makes Pirkle-type CSPs fully compatible for the prepa- ration of preclinical drugs with HPLC [21]. Welch [22] overviewed the contributions made by Pirkle�s group, including the development of Pirkle-type CSPs and the broad applications in chromatographic separation. Several Pirkle-type CSPs are commercially available, and Whelk-O1 (Regis Technologies, Morton Groove, IL), originally designed by Pirkle and Welch, appears to have the broadest applications nowadays in both industrial and academic laboratories. Other types of Pirkle-type CSP are listed in Table 1. Pirkle-type CSPs are most often operated in normal- phase HPLC because they are mostly activated in the interactions in normal phase and they are also limited to separate analytes with aromaticity, given the need for p-p interaction. Most of these CSPs provide good enantiose- lectivity for the separation of AAs, amino alcohols, amines, and other acidic racemates. Pirkle et al. [20] used 9-anthryltrifluoromethyl alcohol chemically-bonded CSP with p-donor group for the separation of 3,5- dinitrobenzoyl (DNB)-chloride derivatives (DNBPG) of p acid amines, AAs, and racemic sulfoxide compounds. Further, they prepared DNBPG CSP and (R)-DNB leucine derivatives of CSP (DNB Leucine CSP) as the second- generation brush-type CSP for the separation of many Trade name Manufacturer Chirobiotic T Astec, USA Chirobiotic V Astec, USA Chirobiotic V Technicol, UK Chirobiotic TAG Astec, USA Chirobiotic MTAG Astec, USA Chirobiotic VAG Astec, USA Crownpak CR(�) Daicel, Japan Crownpak CR(+) Daicel, Japan J.T. Baker, USA Crownpak CR(+) Opticrown RCA(+) US Mac Corporation, USA Sumichiral OA-8000 Sumika, Japan compounds with alkyl, ether, or amino-substituted aro- matic rings. Based on the second-generation CSP, the third-generation brush-type CSP for the resolution of acidic groups was developed [5]. 2.4. Synthetic polymers Naturally-generated polymers are widely used for chiral separation for their structural variabilities. Another kind of polymer-type CSP, the synthetic polymer, mimics recognition mechanisms of natural polymers by poly- merizing the chemical compounds artificially. Synthetic polymers have the advantages of high column capacity, high column efficiency, and broad adaptability to the mobile phase in chromatographic analysis. Three types of polymerization were proposed, including addition polymers, condensation polymers and cross-linked gels. Helical polymethacrylates, polyacrylamides and poly- methacrylamides, polyolefin, polystyrene derivatives, http://www.elsevier.com/locate/trac 185 Trends Trends in Analytical Chemistry, Vol. 39, 2012 poly(vinyl ether)s, polychloral, polyisocyanide, poly- acetylene and polyether are several effective types for chiral analysis. The cross-linked gels (template poly- merization) based on the molecular imprinting tech- nique have received great attention due to their advantages, including ease of preparation, scalability, low costs, stability, and eurytopicity with respect to solvents. Single-handed helical polymethacrylates and poly(meth)acrylamides with chiral groups are kinds of chiral linear polymer prepared by addition polymeriza- tion and several of them are commercialized (Table 1). The one-handed helical triphenylmethyl methacrylate (TrMA) was first synthesized in 1979 by Okamoto et al. [23]. It involves anionic polymerization at low temper- ature, and the helicity of TrMA is the reason for having chirality. Many chemicals (e.g., hydrocarbons, ethers, amides, halides, and phosphoric compounds) have been resolved on TrMA CSPs. Molecularly-imprinted CSPs, which are cross-linked polymer gels with enantioselective binding cavities. were prepared in 1972 [24]. The preparation of molecularly imprinted polymers (MIPs) may be divided into two sections (i.e. polymerization of the functional monomers and cross-linking monomers around the imprinted molecules, followed by grinding of particles and extrac- tion of the imprinted species). The most widely used functional monomer is methacrylic acid and ethylene glycol dimethacrylate (EDMA) is the most common cross-linking monomer [25]. The use of MIPs in CSPs is limited because of issues related to making consistent, homogeneous, and stable stationary-phase particles, and poor mass transfer leading to inefficient separations [26]. 2.5. Polysaccharide CSPs Polysaccharides are naturally-generated polymers and their derivatives. These CSPs can link various substitu- ents on the hydroxyl groups, showing enhanced capa- bility of chiral recognition in HPLC. The good feasibility of this type of CSP is attributed to its chiral individual carbohydrate monomers and a long-range helical sec- ondary structure that effects separations. These struc- tures constitute many possible interaction sites that can resolve many analytes. They were among the most widely used CSPs of all the existing chiral selectors and only surpassed by proteins. However, in contrast to proteins, they have higher loading capacity [27]. More than 85% (2003) and 90% (2005) of chiral HPLC was performed using polysaccharide CSPs [28]. The deriva- tives of cellulose and amylose usually exhibit higher recognition abilities, especially for phenylcarbamates and benzoate derivatives. About 200 different polysaccharide derivatives based on cellulose, amylase, chitin, chitosan, galactosamine, curdlan, dextran, xylan, and inulin have been prepared for chiral recognition in HPLC [28]. Microcrystalline 186 http://www.elsevier.com/locate/trac cellulose triacetate (MCTA, CTA-I) was the first poly- saccharide CSP used in practice, and was developed by Hesse and Hagel [29] in 1970. The mechanical stability problem of MCTA was overcome by Okamoto et al. in 1984 by coating cellulose derivatives onto the surface of macroporous silica beads [30]. After that, many commercialized polysaccharide CSPs were developed (Table 1). It was reported that 84% of the small-molecule race- mates could be resolved by Chiralcel OD, Chiralcel OJ, Chiralpak AD and Chiralpak AS [31]. Among some 500 racemates tested, �80% have been successfully resolved on only two kinds of polysaccharide derivative-based CSPs [28]. CTA-I was first used for the separation of many aromatic and aliphatic pharmaceuticals [32]. Some 14 heterocyclic oxiranes were successfully sepa- rated by five polysaccharide-based CSPs (i.e. Chiralpak AD and AS, and Chiralcel OD, OG and OJ) [33]. The pesticides of synthetic pyrethroids with one chiral center were successfully separated by Chiralcel OD CSP, and those having two pairs of enantiomers could normally be separated on Chiralpak AD [34]. 2.6. Protein CSPs Proteins employed in CSPs are typically naturally gen- erated . The recognition mechanism of enantiomers may be attributed to their formation of a three-dimen- sional structure with several interactive sites, based on the linking orders of AAs and the degree of glycosyla- tion. Stewart and Doherty first used protein CSP based on bovine serum albumin (BSA) for enantioseparation in LC [35]. Also, Allenmark [36] and Hermansson [37] were the pioneers who developed the covalently immobilized protein phases in LC using BSA and a1- acid glycoprotein (AGP), respectively. Although feasible to obtain, other than polysaccharide CSPs, proteins used in CSPs have significant limitations aside from their low capacities, in that they also have low effi- ciencies and the denaturing of proteins will limit their ranges in pH, ionic strength, and organic composition in the mobile phase. Historically, various protein-type CSPs have been commercialized, but several CSPs are no longer available today. CSPs based on BSA, the first reported protein-type CSPs, were developed and used for the separation of several racemates (e.g., N-derivatized AAs, aromatic AAs, uncharged solutes, sulfoxides and sulfoximine derivatives) [38]. AGP CSP was reported for both prep- aration and analysis of penicillamine enantiomers [39], and the separation of b2 agonist formoterol [40]. An- other widely used protein-type CSP was based on human serum albumin (HSA), which was used for the resolution of ibuprofen, ketoprofen, fenoprofen, reduced folates (e.g., leucovorin and 5-methyltetrahydrofolate) and benzodiazepines (e.g., oxazepam, lorazepam and temaz- epam) [38]. silica via hydrosilylation [52]. Specific p-donating analytes (e.g., aryloxypropionic Trends in Analytical Chemistry, Vol. 39, 2012 Trends 2.7. Macrocyclic antibiotic CSPs Macrocyclic antibiotic CSPs are among frequently used CSPs, which include two other CSPs of polysaccharide and CD CSPs in HPLC. In 1994, Armstrong�s group proposed the use of vancomycin, a glycopeptide antibi- otic, to separate AA enantiomers [41]. They showed that, unlike other types of chiral selector, macrocyclic antibiotics comprised a large variety of structural types (e.g., macrocyclic polyene-polyols, ansa compounds, macrocyclic glycopeptides, peptides, and peptide-hetero- cycle conjugates). The various structures of this type of macrocyclic CSP allow multiple interactions (e.g., hydrophobic, dipole-dipole, p-p interactions, hydrogen bonding, and steric repulsion). Macrocyclic antibiotics have a broad spectrum of applications that can work in all chromatographic modes. Currently, several macrocyclic antibiotics are used for chiral separation. Ansamycins and glycopep- tides are two kinds of commonly used chiral selectors (see details in Table 1). Chirobotic V column was used for the enantioseparation of bufaralol and duloxetine [42]. Chirobiotic T columns were used to separate seven aryl- substituted b-lactams, ifosfamide in human plasma and its N-dechloroethylated metabolites. Chiral chemicals of vigabatuin in pharmaceutical products [43], Molindone [44], sulfoxides and urinaty lactic acid [6] were analyzed using the Chirobiotic TAG column. 2.8. Crown-ether-based CSPs Crown ethers, first introduced in 1967, are macrocyclic polyethers, with a cavity of a specific size. The ether oxygens, electron-donor ligands, are distributed on the inside wall of the cavity, which allows access of metal or ammonium cations. In 1979, Cram et al. prepared the first crown ether-based CSP by immobilizing bis-(1,19- binaphthyl)-22-crown-6 on polystyrene or silica gel. Chiral crown ethers were prepared by incorporating various chiral units as chiral barriers into crown ethers (e.g., bulky chiral aromatic rings including binaphthyl, biphenanthryl units, chiral helicene derivatives, and naturally-occurring compounds, including tartaric acid, or carbohydrates) [45]. Crown-ether CSPs are mainly used for enantiosepa- ration of compounds with a primary amine chiral center (e.g., AAs and their derivatives). Two types of crown- ether CSP are widely used and also commercially avail- able (Table 1). One incorporates 3,3�-diphenyl-1, 1�-binaphthyl, and the other is combined with the tartaric acid group. The type based on 3,3�-diphenyl- 1,1�-binaphthyl was used for enantioseparation of a-AAs, 1-phenyl-ethylamine, 3-aminocaprolactam [46], b-AAs, aryl a-amino ketones, and fluoroquinolone antibacterials [47]. The type based on the tartaric acid group is widely used to separate a-AAs, alanine- b-naphthylamide, amino alcohols, 1-(1-naphthyl)ethyl- acids and profens) were better separated by a new Pirkle- anion exchange hybrid-type CSP, which immobilized amine, a-methyl- tryptamine, afloqualone (muscle relaxant) and primaquine (antimalarial) [48]. 3. Newly-developed CSPs The currently available classes of CSPs offer more ra- tional approaches for designs and actions. However, there still exist certain structural types that require new concepts in CSP development (e.g., compounds with stereogenic centers that are non-polar – offering no hydrogen bonding or interactive groups by HPLC, and compounds with multiple chiral centers, as well as a significant number of chiral primary amines) [1]. Expanding the menu of chiral selectors is a research priority. With the expanding needs for various chiral chemicals in pharmaceuticals and other uses, the development for new CSPs has attracted creativity and efforts of researchers all over the world. Newly-developed CSPs have recently been reported for the chiral separa- tion market. The developments focus on not only exploration of chiral selectors, but also the technologies for immobilizing chiral compounds on the support and material preparation. The majority of studies have been carried out to synthesize novel CSPs. The papers cited here represent some of the studies reported to date. 3.1. Modification of traditional CSPs Many derivatives of classical CSPs were developed as new CSPs for their high specificity and enantioselectivity. Cavazzini et al. reviewed the recent development of CSP applications in HPLC in detail from 2009 to 2011 [7]. Since the rapid evolution of various modified classical CSPs, we only overviewed the main types of CSPs developed from 2011 to date. A novel polysaccharide-based CSP, cellulose tris(3- chloro-4-methylphenylcarbamated) was evaluated by enantiomeric separation of 23 Fmoc-glycine derivatives of AAs and results showed that 19 of them were resolved [49]. A chitasan tris(3-chlorophenylcarbamated) deriv- ative-based CSP was used for the chiral recognition of racemic compounds in HPLC, and, of 9 analytes, 7 were separated more quickly on this type of CSP compared with Chiralcel OD-H and Chiralpak AD-H columns [50]. A mono-6-deoxy-(2,4-dihydroxybenzimide)-b-CD CSP was prepared and shown to perform excellently in the chiral separation of 1-phenyl-2-nitroethanol derivatives in reversed-phase mode HPLC [51]. Another CD-type CSP based on mono(6A-N-allylami- no-6A-deoxy)perphenyl -carbamoylated b-CD was re- ported using a novel approach to immobilize onto porous http://www.elsevier.com/locate/trac 187 n (CF Trends Trends in Analytical Chemistry, Vol. 39, 2012 10,11-didehydroquinine-3,5-dintrophenylcarbamate onto 3-azidopropyl silica gel using click chemistry [53]. A Pirkle-type CSP based on pracoterol was reported to have good enantioselectivity [54]. 3.2. Cyclofructans Cyclofructans (CFs), a class of derivatives of CF, were Figure 2. The structure of cyclofructan with six units of polymerizatio of CF6. Color code: carbon, gray; oxygen, red. proposed as a novel group of CSPs for their potential use in chiral separation, due to their broader applicability for superior separations of specific compounds, and ease of preparation, and they have filled an important gap. The first reported use in chiral separation was published by Armstrong and co-workers in 2009 [55]. This new type of chiral selector has a high loading potential, allowing it to purify various kinds of materials. Cyclofructans are cyclic oligosaccharides consisting of b-2,1 linked (D)- fructofuranose units. The ones with six to eight units of polymerization have been commercialized under the abbreviations CF6 (Fig. 2), CF7 or CF8. The mechanism of chiral recognition is attributed to the complex struc- ture, in that each unit of fructofuranose contains four stereogenic centers [7]. Derivatized CF selectors can effect enantiomeric sepa- ration, especially for molecules containing a primary amine functional group (e.g., AA amines, amino alco- hols, and diamines). The first application of cyclofructans in chiral recognition was the resolution of AA esters based on permethylated CF6 and CF7 [55]. Isopropyl carbamate-cyclofructan 6 (IP-CF6), (R)-naphthylethylc- arbamate cyclofructan 6 (RN-CF6), dimethylphenylcar- bamate cyclofructan 7 (DMP-CF7) are currently developed CSPs, and the dominant interactions in the 188 http://www.elsevier.com/locate/trac complex retention mechanism were studied by using the linear free-energy relationship model [56]. 3.3. Metal complex-bonded CSPs Metal complexes used for enantioseparation have not received much attention in HPLC, and only a few CSPs have been reported. 6): (a) 2D molecular structure of CF6; and, (b) 3D molecular structure In 1981, Yamagishi et al. prepared such CSPs by absorbing a metal-chelate complex on colloidal clay for use in enantioseparation [57]. With this approach, they separated the aromatic compounds of 2,3-diphenylpyr- azine and binaphthyl enantiomers by absorbing k-Ru(1,10-phenanthroline)3 2+ metal-chelate complex onto montmorillonite as a CSP [58]. Cu(II) complexes with AA amides as chiral selectors for CSPs were also explored by Chen et al. in 2003 [59]. Silica-based Ru complex-bonded CSPs were recently presented for the first time by Sun [60]. Enantiopure ruthenium(II) tris(diimine) complexes were chosen as the chiral selectors and three binding methods were tested by two Ru complexes, including [Ru(phen)2 aminophen](PF6)2 and [Ru(phen)2phendiamine](PF6)2 (Fig. 3). The results showed k-[Ru(phen)2(phendi- amine)]2+ provided higher enantioselectivity and reso- lution, and was more suitable for preparing CSPs. This new CSP provided good resolution for compounds of the binaphthyl type in normal-phase mode, as p-p interac- tions may play an important role in chiral recognition. In addition, the combination of hydrogen bonding, electrostatic, and dipolar interactions were considered dominant in the chiral recognition of acidic compounds in the polar organic mode. Trends in Analytical Chemistry, Vol. 39, 2012 Trends 3.4. Boron-containing CSPs A newly developed CSP based on boromycin (Fig. 4) containing boron was reported in 2007 as an effective chiral selector. As a macrocyclic antibiotic, boromycin consists of boric acid and a chiral polyhydroxy macro- cyclic ligand. It has strong enantioselectivity for a wide variety of primary amine-containing compounds. The enantiomeric recognition mechanism is attributed to the nature and the location of its enantioselective binding site. Boron is the stereogenic center and essential for chiral molecular recognition. ‘‘Steric bulk’’ on the ana- lytes, charge-charge interactions, hydrogen bonding with the cleft oxygens were also considered to be responsible for the resolution of chiral compounds [61]. 3.5. Ionic liquid CSPs Ionic liquids (ILs) are a class of semi-organic salts with melting points at or close to room temperature, includ- ing polyatomic inorganic anions [e.g., hexafluorophos- phate (PF6 -)] and cations (e.g., pyridinium). Ethyl ammonium nitrate, the first low melting organic salt Figure 3. The structures of two Ru complexes: (a) 2D molecular struct [Ru(phen)2phendiamine](PF6)2; (c) 3D molecular structure of [Ru(phen)2am diamine](PF6)2. Color code: carbon, gray; nitrogen, blue; Ru, crimson. which was liquid at room temperature was reported in 1914 [62]. ILs were recently considered as green mate- rials in chemical reactions and separation processes. These novel materials have many advantages (e.g., wide liquid range, low volatility, good thermal stability, elec- trolytic conductivity, wide range of viscosity, adjustable miscibility, reusability, and non-flammability). Com- pared to the research using ILs as mobile-phase modifiers to improve HPLC separations, methods using covalently- bonded IL CSPs are still limited in the literature. Zhou et al. [63] developed silica-bonded IL-function- alized b-CDs by treating 6-tosyl-b-CD with 1,2-dimeth- ylimidazole or 1-amino-1,2,3-triazole as new CSPs. 16 chiral aromatic alcohol derivatives and 2 racemic drugs were separated by those new CSPs. Wang et al. prepared an IL-modified b-CD CSP by graft polymerization of 6A-(3-vinylimidazolium)-6-deoxyper- phenylcarbamate-b-CD chloride or 6A-(N,N-allylmethyl - ammonium)-6-deoxyperphenylcarbamoyl-b-CD chloride onto silica gel. Compared to a b-CD CSP containing an ammonium moiety, the cationic moiety-containing CSP ure of [Ru(phen)2aminophen](PF6)2; (b) 2D molecular structure of inophen](PF6)2; and, (d) 3D molecular structure of [Ru(phen)2phen- http://www.elsevier.com/locate/trac 189 boro Trends Trends in Analytical Chemistry, Vol. 39, 2012 performed better in enantioseparation of 12 racemic pharmaceuticals and 6 carboxylic acids [64]. 3.6. Asymmetric organocatalysts Asymmetric organocatalysts are metal-free and low- molecular-weight organic molecules, some of which are naturally occurring (e.g., AAs) and considered low cost without the need for preparation compared with the Figure 4. The structure of boromycin: (a) 2D molecular structure of carbon, gray; oxygen, red; nitrogen, blue; boron, yellow. commonly used synthetic metal catalysts. Organocata- lysts are widely used for asymmetric synthesis, and whether the mechanism of chiral induction is due to support-bonded organocatalysts should be exploited in HPLC [7]. Adamantyl, neopentyl derivatives of quinine (AN-QN) and quinidine (AN-QD) were prepared for the separation of six racemic a-AA derivatives [65]. Kacprzak et al. immobilized quinine tert-butyl carbamate onto a silica surface as a CSP to enantioseparate man- delic acid and derivatives thereof. These CSPs had effi- cient resolution with high sample loading [66]. They further tested a representative set of structurally diverse racemic acids, including N-protected AAs, aromatic, aryloxycarboxylic acids, and binaphthol phosphate, using an anion exchange-type CSP, triazolo-linked cin- chona alkaloid carbamate. This type of CSP showed excellent stability in the enantioseparation process [67]. 3.7. Monolithic columns Monolithic columns are advantageous due to their net- work-type one-piece structures. They are prepared by polymerization of monomers in a column, usually in a capillary column. Consequently, these columns can have much higher external porosity than conventional 190 http://www.elsevier.com/locate/trac columns with densely packed particles, resulting in low backpressure, and hence high separation efficiency. Two categories are allotted by the constituent materials: or- ganic polymer and bonded silica-based, with the latter being used more widely. Previous studies on chiral monolithic columns reported the use of native silica monoliths, or silica monoliths with octadecyl or amino functionalities, but such columns for enantiomeric res- mycin; and, (b) 3D molecular structure of boromycin. Color code: olution were not commercialized [68]. Various monolithic columns were developed based on popular chiral selectors, including AA derivatives, CD derivatives, cellulose derivatives and proteins. Chen et al. prepared monolithic CSPs modified by (L)-phenylalan- inamide, (L)-alaninamide and (L)-prolinaminde, respec- tively, via a sol-gel process and successfully separated dansyl AAs and hydroxy acids by l-middle-pressure liquid chromatography (l-MPLC) [69]. Recently, Gotti et al. immobilized penicillin G acylase onto an epoxy-derivatized monolithic silica capillary column to determine (S)-ketoprofen in pharmaceutical samples [70], 13 racemates were enantioseparated by a perphenylcarbamoylated b-CD-silica hybrid monolithic column in capillary liquid chromatography [71], and chiral (R)-acryloyloxy-b-b-dimethyl-c-butyrolactone used as low-molecular-mass chiral selector in a mono- lithic capillary column successfully separated a set of racemic secondary alcohols [72]. 4. Design methods for prospective CSPs A proper choice or rational design of an enantioselective, high-efficiency CSP is one of the most critical steps for Trends in Analytical Chemistry, Vol. 39, 2012 Trends enantiomer separations. The rational design of CSPs is an emerging research field and various research groups have been engaged in the design of more efficient CSPs with pioneering works stemming from Davankov, Pirkle, Okamoto, Blaschke, Allenmark, Hermansson, Arm- strong, Gasparrini, and Lindner, amongst others. So far, reliable predictions of suitable columns and appropriate operational conditions for new selectands remain an unrealized scientific goal [3]. 4.1. Rational design rules Cancelliere et al. introduced some basic rules for rational design of a new CSP [2]. Common structural features for highly-efficient CSPs are as follows: (1) functional groups with high directionality, often hydrogen-bond donor and/or acceptor sites located in the proximity of stereogenic elements, partly leading to relatively high enantioselectivity; (2) conformational homogeneity favors enantioselec- tive recognition; (3) chiral selectors should be immobilized on the matrix to form a thermally, chemically, and stereochemi- cally stable linkage, which would ensure full mo- bile-phase compatibility and low column bleeding, with its active sites exposed to the flowing mobile phase; (4) the geometrical arrangement of the chiral fragment should be realized in a way that maximizes the retention of only one enantiomer; and, (5) the original chromatographic efficiency is preserved by removing strong achiral interacting sites on the matrix surface and avoiding the formation of thick polymeric layers during surface modification. 4.2. Applications of CSP-screening technologies Advances in screening technologies, complimented by the introduction of appropriate instrumentation, have offered better and faster outcomes. Screening and detection systems for chiral chromatography (e.g., opti- cal rotation, circular dichroism and mass spectrometry) are used commonly. Beesley [1] introduced the Express LC systems of Eksigent (Dublin, California, USA), Sepia- tec GmbH (Berlin, Germany), and PDR-Chiral (Lake Park, Florida, USA). All offer a multiple parallel-column- screening system to reduce development time. For bio- polymer-derived selectors, an in-vitro procedure called systematic evolution of ligands by exponential enrich- ment (SELEX) is applied. High-throughput synthesis and screening of newly- designed CSPs will no doubt be fruitful. But clearly, it is hard to generate all possibilities, so some methods of focusing or directing library design are urgently needed [73]. Furthermore, detailed understanding of chiral separation mechanisms in HPLC is essential to facilitate the rational design of novel stationary phases and to optimize separation processes [74]. Recently, combinatorial approaches together with high-throughput screening have been used to develop highly selective stationary phases for chiral recognition. La¨mmerhofer pointed out that databases (e.g., ChirBase) combined with extended automatic screening method- ologies and multiparallel microfluidic HPLC would pro- vide a real high-throughput option to identify the most promising CSPs quickly. Brahmachary et al. [75] iden- tified the best candidates from the library of phenyl amides of 2-oxo-azetidineacetic acid derivatives prepared by Ugi multi-component condensation reactions and screened using the reciprocal approach. Their findings revealed that the substituents of the phenyl ring adjacent to the chiral center of the selector candidates exhibited the most profound effect on the chiral recognition (Fig. 5). 4.3. Applications of computational chemistry With the development of computational chemistry, techniques like molecular modeling are frequently ap- plied (Fig. 6) in research into chiral recognition mechanisms. Hu et al. [74] reported a molecular sim- ulation study to investigate the chiral separation of racemic (D,L)-phenylglycines with thermolysin crystal and water acting as CSP and mobile phase, respec- tively. (D)-Phenylglycine was observed to transport more slowly than (L)-phenylglycine, and interact more strongly with thermolysin, in accordance with the experimentally-observed elution order. The chiral dis- crimination of (D,L)-phenylglycines was mainly caused by the collective contribution from the chiral centers of thermolysin residues. Hu et al. explained two factors attributed to chiral separation, namely the stationary phase (thermolysin crystal) in their simulations pos- sessed a significant number of chiral centers and the high flow rate of the mobile phase was responsible for an enhanced signal/noise ratio within a nanosecond time scale. Bu et al. [76] utilized an Obelisc R column, a novel mixed-mode stationary phase to separate triphenyl atropisomers successfully, and investigated the under- lying mechanism by dynamic chromatography, dynamic NMR and molecular modeling. From a thermodynamic perspective, separation of atropisomer was dominated by enthalpic interactions arising from a combination of both hydrophobic and charge interactions. Based on research into the chiral recognition mecha- nism, some researchers focused on the design of chiral selectands and selectors. Alcaro et al. [77] performed a rational design of novel asymmetric pyrazoles with im- proved enantioselective properties with a molecular docking technique using a model of Chiralcel OJ CSP. The theoretical predictions were confirmed by synthesis and HPLC experiments, and a detailed investigation of the chiral recognition process was carried out by molecular dynamic simulations. http://www.elsevier.com/locate/trac 191 estab Trends Trends in Analytical Chemistry, Vol. 39, 2012 Schefzick et al. [73] carried out a comparative Figure 5. The processes of molecular field analysis (CoMFA) on a set of quinine- based CSPs in an effort to obtain new information that may be useful for designing enhanced stereo-discrimi- nating CSPs. The 3D-QSAR results showed that steric fields alone described most of the variance, backed up by quantum mechanical calculations. Based on the CoMFA, a proposed set of compounds for synthesis and testing was finally given. Figure 6. The process of CSP 192 http://www.elsevier.com/locate/trac 4.4. Perspectives on novel CSP design lishing a prediction model. The contributions of scientific researchers on mecha- nism investigation and reliable prediction, partly men- tioned above, will certainly facilitate rational design of novel CSPs to a large extent. Based on these studies, we can extract some novel, practical ideas. Analysis of experimental data, combined with quantum chemical calculations and computational models, will explain the underlying mechanisms of enantiomer separation to a design using QSERR. comprehension of the underlying chiral recognition mechanisms, multi-disciplinary methods and coopera- s support China ( National Natural Scie d 21177 Appendix A. Abbrevi AGP CD CFs DNB GC HSA PFB SELEX Systematic evolution of ligands by exponential enrichment Trends in Analytical Chemistry, Vol. 39, 2012 Trends tion are required. 5. Conclusion This review addressed the development of CSPs for both analysis and preparation based on chromatographic technology, especially HPLC. Among the various kinds of CSPs, polysaccharide CSPs are currently the most useful for enantioseparation, followed by CD and macrocyclic antibiotics. The derivatives of cellulose and amylose, b- CD, and ansamycins and the glycopeptides are the main groups of chiral selectors in polysaccharide CSPs, CD CSPs and macrocyclic antibiotics, respectively. Recently, several newly-developed chemical com- pounds were found with potential capabilities for chiral selectors in CSPs and commercialization. In particular, in the coming years, monolithic CSPs and organicatalyst CSPs will probably attract increased interest because of their particular advantages in enantioseparation. With development in the applications of computational chemistry in biochemistry, design and screening of effi- cient and stable chiral selectors of CSPs may reach a higher level than conventional design methods. These developments of CSPs used for chromatographic sepa- ration will bring about extensive applications in bio- chemistry, pharmaceuticals, the environment and the food industry. large extent. This can be rather useful to establish solid interaction modes for molecular docking and dynamic simulation between similar CSPs changed certain moi- eties with target analytes, and further predict recogni- tion ability to determine whether the design is feasible or not. Supported by the contributions of studies on a range of chiral screening strategies, new insights into mecha- nisms, quantum chemical calculations, and molecular modeling, we propose that efforts should be made to investigate possible separation mechanisms and rela- tionships of popular classes of CSPs with corresponding separable categories of chiral compounds. Based on coherent, theoretical information, chiral recognition models can be built, and rational choice and design of CSPs may be achieved more easily and effectively. However, there are still some problems remaining to be explored {e.g., Berthod [78] indicated that failure to consider solvent effects critically is currently a limitation when evaluating the predictive value of establishing LC molecular modeling of chiral molecule-selector associa- tion}. To establish a reliable association between exper- imental data and computational calculation/simulation results, proper prediction models, and subsequent guid- ance/rules to optimize/design CSPs, and to improve SFC Supercritical fluid chromatography TLC Thin layer chromatography TrMA Triphenylmethyl methacrylate enantiomers relation relationship RN-CF6 (R) -naphthylethylcarbamate cyclofructan 6 QSEER relationship Quantitative structure- QSAR Pentafluorobenzoyl Quantitative structure-activity PF6 - Hexafluorophosphate chromatography MPLC Middle pressure liquid MIPs Molecular imprinting polymers triacetate MCTA/CTA-I cyclofructan 6 Microcrystalline cellulose IP-CF6 Isopropyl carbamate- ILs Human serum albumin Ionic liquids chromatography HPLC Gas chromatography High-performance liquid EDMA Ethylene glycol dimethacrylate dinitrobenzoyl)phenylglycine DNBPG 3,5-dinitrobenzoyl (3,5- cyclofructan 7 DMP-CF7 Dimethylphenylcarbamate CSP analysis Chiral stationary phase CoMFA Cyclofructans Comparative molecular field CEC Capillary electrochromatography CE Cyclodextrin Capillary electrophoresis BSA Bovine serum albumin of quinine AN-QN of quinidine Adamantyl, neopentyl derivative AN-QD a-acid glycoprotein Adamantyl, neopentyl derivative ations 20837002 an nce Foundations of China (Nos. 112). Program of No. 2009CB421603), and the This study wa ed by the National Basic Research Acknowledgement http://www.elsevier.com/locate/trac 193 References [1] T.E. Beesley, LC-GC Eur. 24 (2011) 270. [42] T.J. Ward, B.A. Baker, Anal. Chem. 80 (2008) 4363. [43] A.A. Al-Majed, J. Pharm. Biomed. Anal. 50 (2009) 96. Trends Trends in Analytical Chemistry, Vol. 39, 2012 [2] G. Cancelliere, I. D�Acquarica, F. Gasparrini, M. Maggini, D. Misiti, C. Villani, J. Sep. Sci. 29 (2006) 770. [3] M. Lammerhofer, J. Chromatogr. Sep. Technol. 1217 (2010) 814. [4] Y. Okamoto, T. Ikai, Chem. Soc. Rev. 37 (2008) 2593. [5] H.D. Qiu, X.J. Liang, M. Sun, S.X. Jiang, Anal. Bioanal. Chem. 399 (2011) 3307. [6] T.J. Ward, K.D. Ward, Anal. Chem. 84 (2012) 626. [7] A. Cavazzini, L. Pasti, A. Massi, N. Marchetti, F. Dondi, Anal. Chim. Acta 706 (2011) 205. [8] V.A. Davankov, Enantiomer 5 (2000) 209. [9] C. Cheng, F.Y. Lin, Chromatographia 39 (1994) 15. [10] V.A. Davankov, J. Chromatogr., A 1000 (2003) 891. [11] H. Nishi, K. Yamasaki, Y. Kokusenya, T. Sato, J. Chromatogr., A 672 (1994) 125. [12] H.Y. Aboul-Enein, I. Ali, Chromatographia 54 (2001) 200. [13] I.W. Muderawan, T.T. Ong, S.C. Ng, J. Sep. Sci. 29 (2006) 1849. [14] D.W. Armstrong, T.J. Ward, R.D. Armstrong, T.E. Beesley, Science (Washington, DC) 232 (1986) 1132. [15] F. Bressolle, M. Audran, T.N. Pham, J.J. Vallon, J. Chromatogr., B 687 (1996) 303. [16] Y. Liu, A.W. Lantz, D.W. Armstrong, J. Liq. Chromatogr. Relat. Technol. 27 (2004) 1121. [17] D.W. Armstrong, S. Chen, C. Chang, S. Chang, J. Liq. Chromatogr. 15 (1992) 545. [18] A.M. Stalcup, K.H. Gahm, Anal. Chem. 68 (1996) 1369. [19] J. Haginaka, J. Wakai, Anal. Chem. 62 (1990) 997. [20] W.H. Pirkle, D.W. House, J.M. Finn, J. Chromatogr. 192 (1980) 143. [21] C.M. Kraml, D.H. Zhou, N. Byrne, O. McConnell, J. Chromatogr., A 1100 (2005) 108. [22] C.J. Welch, J. Chromatogr., A 666 (1994) 3. [23] Y. Okamoto, K. Suzuki, K. Ohta, K. Hatada, H. Yuki, J. Am. Chem. Soc. 101 (1979) 4763. [24] G. Wulff, A. Sarhan, Angew. Chem., Int. Ed. Engl. 11 (1972) 341. [25] M. Kempe, K. Mosbach, J. Chromatogr., A 694 (1995) 3. [26] N.M. Maier, W. Lindner, Anal. Bioanal. Chem. 389 (2007) 377. [27] X.M. Chenl, C. Yamamoto, Y. Okamoto, Pure Appl. Chem. 79 (2007) 1561. [28] P. Franco, A. Senso, L. Oliveros, C. Minguillon, J. Chromatogr., A 906 (2001) 155. [29] G. Hesse, R. Hagel, Chromatographia 9 (1976) 62–68. [30] Y. Okamoto, M. Kawashima, K. Hatada, J. Am. Chem. Soc. 106 (1984) 5357. [31] H.Y. Aboul-Enein, J. Chromatogr., A 906 (2001) 185. [32] G. Blaschke, J. Liq. Chromatogr. 9 (1986) 341. [33] W. Zhou, B. Liu, G. Yang, B. Koppenhoefer, H.Y. Aboul-Enein, J. Liq. Chromatogr. Relat. Technol. 31 (2008) 2879. [34] J. Ye, J. Wu, W.P. Liu, Trends Anal. Chem. 28 (2009) 1148. [35] K.K. Stewart, R.F. Doherty, Proc. Natl. Acad. Sci. USA 70 (1973) 2850. [36] S. Allenmark, B. Bomgren, H. Boren, J. Chromatogr. 237 (1982) 473. [37] J. Hermansson, J. Chromatogr. 269 (1983) 71. [38] J. Haginaka, J. Chromatogr., A 906 (2001) 253. [39] R. Bhushan, R. Kumar, J. Chromatogr., A 1216 (2009) 3413. [40] S. Akapo, C. McCrea, J. Gupta, M. Roach, W. Skinner, J. Pharm. Biomed. Anal. 49 (2009) 632. [41] D.W. Armstrong, Y.B. Tang, S.S. Chen, Y.W. Zhou, C. Bagwill, J.R. Chen, Anal. Chem. 66 (1994) 1473. 194 http://www.elsevier.com/locate/trac [44] H.L. Jiang, Y.H. Li, M. Pelzer, M.J. Cannon, C. Randlett, H. Junga, X.Y. Jiang, Q.C. Ji, J. Chromatogr., A 1192 (2008) 230. [45] H.J. Choi, M.H. Hyun, J. Liq. Chromatogr. Relat. Technol. 30 (2007) 853. [46] T. Shinbo, T. Yamaguchi, H. Yanagishita, D. Kitamoto, K. Sakaki, M. Sugiura, J. Chromatogr. 625 (1992) 101. [47] T.J. Ward, K.D. Ward, Anal. Chem. 82 (2010) 4712. [48] Y. Machida, H. Nishi, K. Nakamura, H. Nakai, T. Sato, J. Chromatogr., A 805 (1998) 85. [49] E. Dominguez-Vega, A.L. Crego, K. Lomsadze, B. Chankvetadze, M.L. Marina, Electrophoresis 32 (2011) 2700. [50] X.Q. Han, L.L. An, H. Cui, H.Y. Li, W.L. Liu, Chromatographia 73 (2011) 1043. [51] X. Li, Z. Zhou, L. Dai, W. Zhou, J. Wang, Talanta 86 (2011) 452. [52] X.H. Lai, W.H. Tang, S.C. Ng, J. Chromatogr., A 1218 (2011) 5597. [53] K.M. Kacprzak, W. Lindner, J. Sep. Sci. 34 (2011) 2391. [54] Y.W. Zhang, Z. Hao, Y. Shun, S. Hang, Adv. Mater. Res. 396 (2011) 1513. [55] P. Sun, C.L. Wang, Z.S. Breitbach, Y. Zhang, D.W. Armstrong, Anal. Chem. 81 (2009) 10215. [56] L. Janeckova, K. Kalikova, J. Vozka, D.W. Armstrong, Z. Bosakova, E. Tesarova, J. Sep. Sci. 34 (2011) 2639. [57] A. Yamagishi, M. Soma, J. Am. Chem. Soc. 103 (1981) 4640. [58] A. Yamagishi, J. Chromatogr. 319 (1985) 299. [59] Z.L. Chen, M. Niitsuma, K. Uchiyama, T. Hobo, J. Chromatogr., A 990 (2003) 75. [60] P. Sun, Ph.D. Dissertation, University of Texas, Arlington, TX, USA, 2010. [61] C.L. Wang, D.W. Armstrong, D.S. Risley, Anal. Chem. 79 (2007) 8125. [62] K. Bica, P. Gaertner, Eur. J. Org. Chem. (2008) 3235. [63] Z. Zhou, X. Li, X. Chen, X. Hao, Anal. Chim. Acta 678 (2010) 208. [64] R.Q. Wang, T.T. Ong, W. Tang, S.C. Ng, Anal. Chim. Acta 718 (2012) 121. [65] P.A. Levkin, N.M. Maier, V. Schurig, W. Lindner, Angew. Chem., Int. Ed. Engl. 49 (2010) 7742. [66] K. Kacprzak, N. Maier, W. Lindner, J. Sep. Sci. 33 (2010) 2590. [67] K.M. Kacprzak, N.M. Maier, W. Lindner, J. Chromatogr., A 1218 (2011) 1452. [68] B. Chankvetadze, T. Ikai, C. Yamamoto, Y. Okamoto, J. Chroma- togr., A 1042 (2004) 55. [69] Z.L. Chen, K. Uchiyama, T. Hobo, J. Chromatogr., A 942 (2002) 83. [70] R. Gotti, J. Fiori, E. Calleri, C. Temporini, D. Lubda, G. Massolini, Chiral, J. Chromatogr., A 1234 (2011) 45. [71] Z. Zhang, M. Wu, R. Wu, J. Dong, J. Ou, H. Zou, Anal. Chem. 83 (2011) 3616. [72] A. Ghanem, T. Ikegami, N. Tanaka, Chirality 23 (2011) 887. [73] S. Schefzick, M. Lammerhofer, W. Lindner, K.B. Lipkowitz, M. Jalaie, Chirality 12 (2000) 742. [74] Z. Hu, J. Jiang, J. Phys. Chem. B 113 (2009) 15851. [75] E. Brahmachary, F.H. Ling, F. Svec, J. Frechet, J. Comb. Chem. 5 (2003) 441. [76] X. Bu, P.J. Skrdla, P.G. Dormer, J. Chromatogr., A 1217 (2010) 7255. [77] S. Alcaro, A. Bolasco, R. Cirilli, R. Ferretti, R. Fioravanti, F. Ortuso, J. Chem. Inf. Model. 52 (2012) 649. [78] A. Berthod, Anal. Chem. 78 (2006) 2093. Development of chiral stationary phases for high-performance liquid chromatographic separation 1 Introduction 2 The classical commercial CSPs 2.1 Ligand-exchange CSPs 2.2 Cyclodextrin CSPs 2.3 Pirkle-type CSPs 2.4 Synthetic polymers 2.5 Polysaccharide CSPs 2.6 Protein CSPs 2.7 Macrocyclic antibiotic CSPs 2.8 Crown-ether-based CSPs 3 Newly-developed CSPs 3.1 Modification of traditional CSPs 3.2 Cyclofructans 3.3 Metal complex-bonded CSPs 3.4 Boron-containing CSPs 3.5 Ionic liquid CSPs 3.6 Asymmetric organocatalysts 3.7 Monolithic columns 4 Design methods for prospective CSPs 4.1 Rational design rules 4.2 Applications of CSP-screening technologies 4.3 Applications of computational chemistry 4.4 Perspectives on novel CSP design 5 Conclusion Acknowledgement Appendix A Abbreviations References