CHIRALITY 19:693–700 (2007)
Strategic Use of Preparative Chiral Chromatography for the Synthesis of a Preclinical Pharmaceutical Candidate WILLIAM R. LEONARD JR.,1 * DEREK W. HENDERSON,1 ROSS A. MILLER,1 GLENN A. SPENCER,2 OSAMA S. SUDAH,2 MIRLINDA BIBA,1 AND CHRISTOPHER J. WELCH1* 1 Separation and Analysis Technologies, Department of Process Research, Merck Research Laboratories, Rahway, New Jersey 2 Department of API Technology, Merck Manufacturing Division, Rahway, New Jersey
ABSTRACT The modern use of preparative chromatography in pharmaceutical development is illustrated by the case of a recent preclinical candidate from these laboratories. The synthesis of the candidate employed a coupling of two enantiopure intermediates, each of which could be resolved using preparative chiral chromatography. SFC screening was employed to identify the enantioselective stationary phases, and semipreparative SFC methods derived from this screening were used to produce gram amounts of enantiopure intermediate for initial studies. However, initial larger scale resolution required the translation of the SFC methods to HPLC conditions. Preparative chiral HPLC on a 30-cm i.d. column was then used to produce enantiopure intermediates which were coupled to give 170 g of the preclinical candidate. Subsequent preparation of the candidate at larger scale for later-stage clinical evaluation employed an improved synthesis in which one component was constructed by asymmetric synthesis. Resolution of the other component, now a more advanced intermediate, was carried out using newly obtained large-scale SFC equipment. Some discussion is presented on the varying strategies whereby preparative chiral chromatography can be used to support either short-term or long-term synthetic goals in preclinical pharmaceutical development. C 2007 Wiley-Liss, Inc. Chirality 19:693–700, 2007. V KEY WORDS: preparative HPLC; preparative SFC; chiral chromatography; synthetic strategy
INTRODUCTION
In recent years, preparative chromatography has emerged as a valuable tool for supporting synthesis of preclinical drug candidates.1–3 Long a staple technique of biomolecule production, preparative chromatography has gained favor in small molecule synthesis, especially for the chromatographic resolution of enantiomers.4,5 Several commercial molecules are now being produced at multiton scale using preparative chiral chromatography,6–8 and the technique has become widely accepted for the kilogram scale preparation of enantiopure new drug candidates needed for preclinical and early clinical studies. We herein present a case study featuring the intensive use of preparative chiral chromatography in the synthesis of the early development drug candidate 1 (Scheme 1). A semipreparative SFC resolution was used to produce gram amounts of an enantiopure intermediate for initial synthetic route investigations. Subsequently, two different relatively unproductive HPLC resolutions were used in the initial delivery of 170 g of the preclinical candidate for initial safety testing. Subsequent preparation of the candidate at larger scale employed a combination of asymmetric synthesis and preparative chiral SFC to establish product stereochemistry. C 2007 Wiley-Liss, Inc. V
EXPERIMENTAL Materials
Methanol, HPLC-grade solvent, was purchased from EMD Chemicals (Gibbstown, NJ). Carbon dioxide (SFC grade) was purchased from Scott Specialty Gases (Plumsteadville, PA). Chiralpak and Chiralcel columns were purchased from Chiral Technologies (Exton, PA), Whelk-O columns were purchased from Regis Technologies (Morton Grove, IL), Chiris AX-QD columns were purchased from Iris Technologies (Roswell, GA), Kromasil columns were purchased from Eka Chemicals (Dobbs Ferry, NY), and Chirobiotic columns were purchased from Astec (Whippany, NJ).
*Correspondence to: William R. Leonard Jr. or Christopher J. Welch, Separation and Analysis Technologies, Department of Process Research, Merck Research Laboratories, Rahway, NJ 07065. E-mail:
[email protected] or
[email protected] Received for publication 5 October 2006; Accepted 20 December 2006 DOI: 10.1002/chir.20378 Published online 12 March 2007 in Wiley InterScience (www.interscience.wiley.com).
694
LEONARD ET AL.
Scheme 1. Synthetic strategies for preparation of preclinical candidate 1. (a) Initial synthesis carried out for the production of 170 g relying on preparative HPLC resolution of two different enantiopure advanced intermediates 2 and 3. (b) Improved synthesis utilized for subsequent larger 1.6 kg delivery based on a highly productive preparative SFC resolution of intermediate 4.
Chiral SFC Method Development Screening
Chiral SFC method development screening was carried out using the Mettler Toledo Berger analytical supercritical fluid chromatograph (Mettler Toledo Autochem, Newark, DE) fitted with a six-position column selection valve with 4.6 3 250 mm columns and an Agilent model 1100 diode array UV-Visible detector (Agilent Technologies, Palo Alto, CA). Column screening was carried out using a standard gradient approach described previously.9 Semipreparative SFC
Semipreparative separations were carried out using the Mettler Toledo Berger multigram semipreparative SFC system (Mettler-Toledo Autochem) with a Knauer K-2600 UV detector (Knauer, Berlin) and a Chiralcel OJ or Chiralpak AD 2 3 25 cm column (Chiral Technologies) with a MeOH/CO2 eluent at a flow rate of 70 ml/min. Isolated fractions were rotary evaporated at room temperature. Preparative HPLC Loading Studies
Preparative loading studies were conducted on an Agilent 1100 system with a G1311A quaternary pump, G2260 Prep ALS autosampler, and a G1315B diode array UV-Vis detector using 4.6 3 250 mm Chiralpak AD or Chiralcel OJ 20-lm columns at 1.5 ml/min. Large Scale Preparative HPLC
Large-scale preparative HPLC was conducted using a Biotage KP3000 system (Biotage, Charlottesville, VA) and a 30-cm i.d. Prochrom LC3000 column (Novasep, Boothwyn, PA) containing 10.6 kg (25 cm bed depth) Chiralpak AD 20 lm. A flow rate of 6 liter/min was used with the eluent feed temperature controlled at 228C. MeOH, i-PrOH, and heptane were purchased from Pride Solvents and Chemicals (Avenel, NJ). Fraction collection was based on total volume and UV detection. Fractions were collected in metal drums, assayed for purity, combined Chirality DOI 10.1002/chir
and concentrated by distillation at reduced pressure. For the purification of 2, the maximum injection size was 820 ml of a 101 mg/ml solution of rac-2 in MeOH using detection at 259 nm. The flow rate was 6 liter/min with a cycle time of 15 min using MeOH as eluent. For the purification of 3, the maximum injection size was 610 ml of a 71 mg/ml solution of rac-3 in eluent using detection at 310 nm. The flow rate was 4.8 liter/min with a cycle time of 5 min using 1% i-PrOH/heptane (v/v) as eluent. This eluent was premixed and recirculated through a bed of ˚ , to a residual water level of 30 ppm molecular sieves, 3 A before use.
RESULTS AND DISCUSSION
Our strategy for using preparative chromatography in the initial 170 g synthesis of the preclinical development candidate 1 relied heavily on the original medicinal chemistry approach in which intermediates 2 and 3 were each prepared in enantiopure form by preparative chiral chromatography and then coupled in a subsequent step. The use of two different preparative separations within a single synthesis, while acceptable in medicinal chemistry, is less often utilized in the larger scale syntheses carried out to support preclinical development. A strategy based on coupling of the two racemates, followed by purification of the desired product from the statistical mixture of four stereoisomers would be cumbersome, affording a maximum of 25% yield (assuming no diastereoselectivity in the reaction) and clearly not competitive with the potential for 50% yield afforded by the approach in which the two enantiopure intermediates are first prepared and then coupled. For the case at hand, there was no reasonable opportunity to use one fixed stereocenter in the molecule to control the introduction of the second stereocenter. In addition, nonchromatographic options for quickly accessing enantiopure 2 or 3 at reasonable scale were quite limited, with
PREPARATIVE CHIRAL CHROMATOGRAPHY IN PHARMACEUTICS
695
Fig. 1. Chiral SFC screening of aminopiperidine intermediate rac-2 shows good resolution using either Chiralcel OJ-H or Chiralpak AD-H CSPs. All stationary phases are 10 lm unless noted, and the Kromasil silica column is presented as an achiral control. Conditions: 4% MeOH in CO2 for 4 min, then ramp at 2%/min to 40% MeOH, 25 min run time, 4.6 3 250 mm columns, 1.5 ml/min, 200 bar, 358C, UV 215 nm.
initial screening of classical resolution and enzymatic approaches affording no suitable leads. Chromatographic method development in these laboratories typically begins with analytical chiral SFC using a screening approach that we have described previously.9 Subsequent use of semipreparative SFC provides rapid access to the enantiopure intermediates that are essential for carrying out a rapid survey of potential synthetic routes. Chiral SFC screening quickly identifies chiral stationary phases (CSPs) that are suitable for optimization for analytical assays as well as for semipreparative and preparative
purification. Chiral SFC screening of Boc-aminopiperidine rac-2 afforded the results pictured in Figure 1, from which it can be seen that both Chiralpak AD (desired enantiomer eluted first) and Chiralcel OJ (desired enantiomer eluted second) CSPs afford quite good resolution of the enantiomers on the analytical scale. Chiral SFC screening results can be translated to semipreparative scale SFC to provide rapid access to gram amounts of enantiopure compounds for synthetic route investigation, crystallization studies, or other needs. Using this approach it is possible to access gram amounts of Chirality DOI 10.1002/chir
696
LEONARD ET AL.
Fig. 2. SFC resolution of the enantiomers of aminopiperidine intermediate 2. A. Optimized isocratic method using analytical loading amounts on a Chiralcel OJ-H (21 3 250 mm, 40% MeOH/CO2, 70 ml/min, 100 bar, 358C). B. Semipreparative method on the same column utilizing overlapping injections (0.5 ml of a 230 mg/ml solution in MeOH, injected every 1 min), affording a productivity of 1.4 kkd (kilograms purified enantiomer per kilogram of stationary phase per 24 h day). Note that the first injection is not a full load because of priming of the injection system. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]
enantiopure material from newly prepared racemates within a single day. Thus, a rapid (2-min) method was developed for resolution of the enantiomers of aminopiperidine 2 using a 21 3 250 mm Chiralcel OJ semipreparative column with an eluent of 40% MeOH/CO2 (Fig. 2A). The small impurity on the tail of the second peak was not of concern because the impurity can be removed by a downstream process in the synthesis. Following initial injection at analytical scale, the injection volume can be increased until the enantiomer peaks reach a ‘‘touching-band’’ point, which can provide the basis for fast chromatographic purification without generation of fractions of mixed purity. In cases such as this one, where the sample is relatively clean, injections can be overlapped to improve productivity. In the present example, 115 mg (0.5 ml of a 230 mg/ ml solution in MeOH) of rac-2 was injected on to the column every 1 min with automated sample injection and fraction collection (Fig. 2B). Using this approach, a total of 4 g racemate was resolved in about 1 h, providing the individual enantiomers in 98% ee (first eluted) and 95% ee (second eluted, desired enantiomer). While such small-scale separations are quite useful for providing limited quantities of material to carry out process research investigations, we were unable to carry out largescale preparative chiral SFC separations during the period of the initial portion of this development project owing to the lack of larger scale SFC equipment. We have subsequently added large-scale preparative SFC equipment which is now routinely used for carrying out kilogram-scale resolutions.10 Our strategy during the period of these separations was to use preparative HPLC at the kilogram-scale to provide the desired amount of each compound in a reasonable time. In our laboratories, we often use the initial SFC screening results as a starting point for developing suitable preparative HPLC methods. The two chiral SFC Chirality DOI 10.1002/chir
screening ‘‘hits’’ for rac-2 (Chiralpak AD and Chiracel OJ) were each further studied in order to find the optimal eluent for HPLC separation. The HPLC separation on stationary phase was examined using mixtures of polar modifiers (1:1, v/v, MeOH/EtOH, EtOH, or iPrOH) in heptane. The two columns were also examined in the ‘‘strong polar mode’’ using 100% polar solvents.11 Using these approaches, we identified 100% MeOH on Chiralpak AD and 1:1 (v/v) EtOH/heptane on Chiralcel OJ as suitable preparative methods for carrying out the separation of rac-2. Loading studies were next carried out to maximize the productivity of the separation. The studies were conducted on 20-lm CSPs that are available in bulk. Figure 3 illustrates the study of 100% MeOH on an analytical 4.6 3 250 mm column containing Chiralpak AD. Injection of 19 mg of racemate affords near-baseline resolution of enantiomers with a cycle time of about 15 min. Consequently, a productivity of 0.33 kkd (kilograms purified enantiomer per kilogram of stationary phase per 24-h day) was estimated (productivity calculations include a correction for racemate purity and an estimate of compound recovery.) A similar study on Chiralcel OJ using 1:1 EtOH/heptane provided a slightly lower calculated productivity. The greater productivity of the Chiralpak AD method, combined with the advantage of eluting the desired enantiomer first, led to the selection of this method for carrying out kiloscale preparative separation to support the first delivery of preclinical material. The ability to carry out small-scale modeling studies using an analytical column to reliably estimate requirements for larger scale preparative separation campaigns is one of the greatest advantages of the preparative chromatographic approach in kilogram-scale synthesis of pharmaceutical candidates. In contrast to organic reactions, where the influence of stirring rate, heat transfer, and
PREPARATIVE CHIRAL CHROMATOGRAPHY IN PHARMACEUTICS
697
Fig. 3. HPLC loading study for the separation of the enantiomers of aminopiperidine intermediate rac-2. Progressively larger injections of concentrated (100 mg/ml in MeOH) rac-2 feed lead to the identification of the optimal injection amount (19 mg racemate) to maintain ‘‘touching-band’’ separation of enantiomers, in which the two peaks are still near-baseline resolved. Conditions: Chiralpak AD 20 lm, 0.46 3 250 mm, 100% MeOH, 1.5 ml/min, 228C, 254 nm. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]
issues relating to reaction exotherms or induction periods are notoriously difficult to estimate, large-scale HPLC performance criteria such as production rate, amount of solvent used, and time required for separation can be confidently predicted based on model studies carried out on a standard analytical column. Based on the loading studies illustrated in Figure 3, we estimated that in order to resolve the 1.7 kg of racemic 2 required for the first delivery of 1 using our 30-cm i.d. column, *21 injections taking 5.3 h and 2000 l of MeOH solvent would be needed. The actual number of injections, time, and solvent usage are typically somewhat higher owing to column packing and equilibration, as well as chromatographic loading verification. Nevertheless, these requirements were judged to be in scope for carrying out the resolution, and we proceeded to further study a few critical parameters for this separation to ensure successful results under pilot-plant operating conditions. We first tested the influence of temperature on the preparative separation to investigate whether this parameter would significantly influence the broad and tailing nature
of the second eluting enantiomer. In addition, this information would indicate whether special attention needed to be paid to temperature (e.g., controlling column jacket temperature) other than controlling inlet eluent temperature. As is generally the case, we found a temperature dependence showing that decreasing temperature leads to increased retention and enantioselectivity without much effect on peak broadness as illustrated in Figure 4. We next investigated the influences of several variables in operation parameters upon the separation. Residual water content in eluents can vary greatly, especially in hygroscopic solvents like methanol or from adventitious water remaining from clean outs of multiuse pilot-plant vessels. Furthermore, residual water can sometimes have a profound influence on chromatographic performance, although in the present case, we found the separation to be relatively insensitive to variations in water content (Fig. 5A). We next studied the influence of the additive diethylamine (DEA) on the separation. Amine additives are often required for optimal performance in the separation of basic or active compounds on the Chiralpak AD
Fig. 4. Influence of temperature on the preparative chromatographic separation of the enantiomers of Boc-aminopiperidine rac-2 on Chiralpak AD. Conditions: Chiralpak AD 20 lm, 0.46 3 250 mm, 100% MeOH, 1.5 ml/min, 228C, 265 nm, 0.4-ml injection of 49 mg/ml in MeOH. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] Chirality DOI 10.1002/chir
698
LEONARD ET AL.
Fig. 5. Investigation of critical parameters on preparative chromatographic separation of the enantiomers of 2. A. Influence of water content. B. Influence of DEA on the preparative chromatographic separation of the enantiomers of Boc-aminopiperidine rac-2 on Chiralpak AD. Conditions: Chiralpak AD 20 lm, 0.46 3 250 mm, 100% MeOH, 1.5 ml/min, 228C, 254 nm, 0.15-ml injection of 100 mg/ml in MeOH. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]
column. In general, this requirement is more pronounced for relatively nonpolar alcohol/hydrocarbon eluents than for polar eluents such as methanol or acetonitrile. In the present case, we found the DEA additive to have a slight influence on retention time, but not to substantially alter peak shape or enantioselectivity (Fig. 5B). Consequently, we opted to proceed with the separation using neat methanol as eluent, with no amine modifier. With these studies in hand, we proceeded to carry out the separation using a 30-cm i.d. HPLC column pilot-plant installation. A total of 1.7 kg of racemate was resolved in 32 injections to afford 710 g (83% recovery) of the desired enantiomer in 98% ee. The maximum productivity obtained during the separation was 0.34 kkd, compared with an estimated value of 0.33 kkd. The specific solvent consumption obtained was 2400 liter/kg resolved enantiomer, compared with the modeled value of 2630 liter/kg. The chromatography, illustrated in Figure 6, went largely as expected. Although this method is fairly unproductive, our ability to access 710 g of enantiopure material quickly using the chromatographic approach greatly escalated the speed of this development project and compares favorably in speed and cost with any other possible solution.
Concurrent with the development and execution of the enantioseparation of aminopiperidine intermediate 2, we were investigating approaches to the chromatographic resolution of the enantiomers of the second key intermediate, the dihydropyrrole 3 required for the synthesis of preclinical candidate 1. Following a similar approach to that described previously, we developed a fairly productive 0.73 kkd preparative HPLC method for this resolution on Chiralpak AD using 1% i-PrOH/heptane as illustrated in Figure 7, with the desired enantiomer eluting first. Critical parameter studies revealed the importance of using an amine modifier such as DEA, especially for a ‘‘virgin’’ CSP being used for the first time. So-called memory effects with the use of amine modifiers with Chiralpak AD and similar stationary phases are well precedented.12 In addition, as is often the case for enantioseparations carried out with very low modifier concentrations, small changes in modifier concentration were found to result in profound changes in chromatographic performance (Fig. 8A).13 The sensitivity toward solvent composition for this separation is strikingly evident, making the careful control of solvent composition a key requirement for carrying out this separation in our pilot-plant facility. In addi-
Fig. 6. Representative preparative chromatogram of the separation of the enantiomers of rac-2 to support first developmental scale preparation of material for preclinical evaluation. Conditions: Chiralpak AD 20 lm, 30 i.d. 3 25 cm, 100% MeOH, 6 liter/min, 228C, 259 nm, 820-ml injection of 101 mg/ml in MeOH.
Fig. 7. Loading study chromatogram for the preparative separation of the enantiomers of dihydropyrrole intermediate rac-3 on an analytical 0.46 3 25 cm column to support first delivery of material for preclinical evaluation. Conditions: Chiralpak AD 20 lm, 0.46 3 250 mm, 1% (v/v) i-PrOH/heptane þ 0.1% Et2NH, 1.5 ml/min, 228C, 306 nm, 0.06-ml injection of 280 mg/ml in 1% (v/v) i-PrOH/heptane.
Chirality DOI 10.1002/chir
PREPARATIVE CHIRAL CHROMATOGRAPHY IN PHARMACEUTICS
699
Fig. 8. Influence of critical parameters on preparative separation of the enantiomers of dihydropyrrole intermediate 3. A. Influence of modifier concentration. B. Influence of water content. Conditions: Chiralpak AD 20 lm, 0.46 3 250 mm, 1% (v/v) i-PrOH/heptane þ 0.1% Et2NH (unless otherwise noted), 1.5 ml/min, 306 nm, 0.06-ml injection of 280 mg/ml in 1% (v/v) i-PrOH/heptane. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]
tion, a profound influence of water on the separation was also observed (Fig. 8B), suggesting that careful water control would be essential predictable chromatography. Given the somewhat delicate nature of this separation, we opted to premix the eluent so as to not be vulnerable to changes in composition owing to variability in pump metering rates. In addition, we instituted an analytical specification for water content. When the actual solvent batch was prepared in the pilot plant prior to the largescale separation, the water content was found to be higher than desired at 180 ppm, as measured by Karl Fischer titration. A front run separation carried out on a 5-cm column using this eluent confirmed poor chromatographic performance, which was returned to the expected performance upon drying the eluent with molecular sieves. We therefore implemented a plan to dry the mobile phase in the pilot plant by recirculation through a bed of molecular ˚ , whereupon the water content was reduced to sieves, 3 A an acceptable level of 30 ppm. We next proceeded with the pilot-plant separation campaign, where 470 g of racemic rac-3 was resolved in 16 injections to afford 205 g (97% recovery) of the desired enantiomer in 98% ee. The predicted productivity was 0.73 kkd compared with the maximum obtained productivity of 0.53 kkd. A representative chromatogram from the preparative campaign is shown in Figure 9. With the two enantiopure fragments in hand, the final coupling and deprotection was carried out to afford 170 g of the preclinical candidate 1. While certainly not elegant, it should again be emphasized that the chromatographic approach was expeditious, offering the fastest, cheapest, and most labor-efficient option for preparation of 1. Given that most drug candidates fail in preclinical development and that most initial developmental syntheses are carried out only once, the use of preparative chromatography to support the quantities of preclinical candidates for their first preparation is enjoying increasing favor. This approach allows a switch to a more attractive synthesis at a later stage, should the candidate fare well during preclinical studies.
In the present case, the use of two different preparative separations to construct a compound with two stereocenters was deemed less than optimal for the long term, and studies were undertaken to develop an alternative synthesis. A number of approaches were evaluated for formation of the two stereocenters, including the use of asymmetric catalysis or resolution by enzymatic or crystallization techniques. Ultimately, an efficient enantioselective synthesis of the dihydropyrrole intermediate 3 was developed for the second preparation of 1, which will be described elsewhere. While other attempts at enantioselective preparation of intermediates corresponding to aminopiperidine 2 failed, SFC screening investigations of a number of analogs of rac-2 led to the identification of a preparative SFC separation of an advanced intermediate (4) with a more simplified protecting group scheme than that utilized in the initial campaign (Scheme 1). Preparative SFC studies using acetamide 4 indicated a productivity of 2.5 kkd (a 750% improvement over the previous similar HPLC separation) using Chiralcel OF with 40% i-PrOH/CO2. Only 300 liters of i-PrOH/kg resolved enantiomer was expected to be consumed. This method was the basis for the preparative separation of 850 g of 4 used for the second preparation of 1.6 kg of the preclinical
Fig. 9. Representative chromatogram from pilot-plant separation of the enantiomers of dihyropyrrole intermediate 3. Conditions: Chiralpak AD 20 lm, 30 i.d. 3 25 cm, 1% (v/v) i-PrOH/heptane þ 0.1% Et2NH, 4.8 liter/min, 228C, 310 nm, 610-ml injection of 71 mg/ml in 1% (v/v) i-PrOH/heptane. Chirality DOI 10.1002/chir
700
LEONARD ET AL.
candidate 1 and may be economically feasible for its preparation on commercial scale. CONCLUSIONS
Preparative chiral chromatography is a valuable tool for supporting the rapid development of compounds for preclinical investigations. Different strategies for the use of preparative chromatography can be used at different stages of development, with productivity and efficiency being less of a concern at early stages of development, and becoming increasingly important as the project moves closer to commercialization. Use of chromatographic resolution to support first delivery of preclinical candidates can buy time, allowing the development of more cost-efficient approaches at a later stage, if warranted. Alternatively, highly productive chromatographic separations may sometimes afford the most economical process for industrial scale manufacturing. ACKNOWLEDGMENTS We are grateful to J. Albaneze-Walker and J. DaSilva for assistance with SFC purifications. LITERATURE CITED 1. Cox GB. Chiral chromatography: Given the new techniques currently in development, the use of chiral chromatography in the large-scale separation of racemic compounds will eventually become commonplace. Innovat Pharm Technol 2001;1:131–137. 2. Francotte ER. Enantioselective chromatography as a powerful alternative for the preparation of drug enantiomers. J Chromatogr 2001; 906:379–397.
Chirality DOI 10.1002/chir
3. Welch CJ, Fleitz F, Antia F, Yehl P, Waters B, Ikemoto N, Armstrong JD, Mathre DJ. Chromatography as an enabling technology in pharmaceutical process development: expedited multi-kilogram preparation of a candidate HIV protease inhibitor. Org Process Res Dev 2004; 8:186–191. 4. Welch CJ. Chiral chromatography in support of pharmaceutical process research. In: Cox GB, editor. Preparative enantioselective chromatography. Oxford: Blackwell; 2005. p 1–18. 5. Nelson TD, Welch CJ, Rosen JD, Smitrovich JH, Huffman MA, McNamara JM, Mathre DJ. Effective use of preparative chiral HPLC in a preclinical drug synthesis. Chirality 2004;16:609–613. 6. Quallich GJ. Development of the commercial process for Zoloft/ sertraline. Chirality 2005;17(Suppl):S120–S126. 7. Cavoy E, Hamende M, Deleers M, Canvat J-P, Zimmermann V. Process for preparing (S)- and (R)-a-ethyl-2-oxo-1-pyrrolidineacetamide using a simulated mobile bed chromatography with chiral phase of silica gel supported polymer. US Patent 6124473, 2000. 8. Blehaut J, Ludemann-Hombourger O, Perrin SR. Industrial chiral chromatography: the keys to rapid process development and successful implementation. Chim Oggi 2001; 19(Sept):24–28. 9. Biba M, Welch CJ, DaSilva JO, Mathre DJ. Impurity isolation in support of pharmaceutical process research. Am Pharm Rev 2005;8:68–72,74. 10. Welch CJ, Leonard WR Jr, DaSilva JO, Biba M, Albaneze-Walker J, Henderson DW, Laing B, Mathre DJ. Preparative chiral SFC as a green technology for rapid access to enantiopurity in pharmaceutical process research. LCGC N Am 2005;2:16–29. 11. Miller L, Orihuela C, Fronek R, Murphy J. Preparative chromatographic resolution of enantiomers using polar organic solvents with polysaccharide chiral stationary phases. J Chromatogr A 1999;865: 211–226. 12. Stringham RW, Lynam KG, Lord BS. Memory effect of diethylamine mobile phase additive on chiral separations on polysaccharide stationary phases. Chirality 2004;16:493–498. 13. Pirkle WH. Unusual effect of temperature on the retention of enantiomers on a chiral column. J Chromatogr 1991;558:1–6.
