`
`Volume 19, Issue 3 (1988)
`
`175
`
`Authors:
`
`Daniel W. Armstrong
`Soon M. Han
`Department of Chemistry
`University of Missouri-Rolla
`Rolla, Missouri
`
`Referee:
`
`Willie L. Hinze
`Department of Chemistry
`Wake Forest University
`Winston-Salem, North Carolina
`
`I. INTRODUCTION
`
`The resolution of enantiomers (nonsuperimposable, mirror-image isomers) has tradition-
`ally been considered one of the more difficult problems in separation science. Enantiomers
`have identical physical and chemical properties in an isotropic environment except that they
`rotate the plane of polarized light in opposite directions. A mixture containing equal amounts
`of enantiomers is referred to as a racemic mixture. Neither racemic mixtures nor solutions
`of achiral compounds are able to rotate the plane of polarized light.
`In the mid-nineteenth century, the French physicist Biot discovered that certain substances
`had the unusual property of being able to rotate the plane of polarization of a linearly
`polarized incident light beam.1-2 These substances, which were said to be optically active
`or to possess optical rotatory power, were characterized by a lack of symmetry in their
`molecular or crystalline structure. In 1848, Pasteur reported the first deliberate separation
`of enantiomers from a racemic mixture.3 This separation was possible because saturated,
`racemic solutions of sodium ammonium tartrate form two types of morphologically distinct
`crystals, each containing a single enantiomer at temperatures below 27°C. In 1874, Van't
`Hoff and Le Bel independently deduced that the molecular basis of optical activity lay with
`the "asymmetric carbon atom".4-5
`Enantiomeric separations are very important in many fields. Some typical fields include
`chiral synthesis, mechanistic studies, catalysis, kinetics, geochronology, biology, biochem-
`istry, pharmacology, and medicine. There are many methods for the separation of enan-
`tiomers. Several nonchromatographic methods classically have been used to isolate optically
`pure compounds from racemic mixtures. The most generally useful of these methods involves
`conversion of the racemic mixture to a pair of diastereomers which have different chemical
`and physical properties and which may be separated by conventional techniques. Separation
`of the diastereomers can often be achieved by fractional recrystallization. Enantiomeric
`excesses have also been obtained via microbiological or enzymatic digestion. In this case,
`the enzyme must preferentially catalyze the reaction of one enantiomer relative to the other.
`Crystallizations followed by mechanical separation are useful for the few compounds that
`segregate into morphologically distinct crystals. Another type of crystallization technique
`involves seeding a supersaturated racemic solution with a small optically pure crystal. The
`resulting crystals often contain an enantiomeric excess. While all of these techniques have
`been utilized successfully, none can be considered generally useful, and all are relatively
`time consuming and tedious. In addition, these methods often fail to afford total separation
`of enantiomers. Recently there has been a dramatic increase in the number and type of
`racemic separations achieved by chromatographic methods. The popularity of the chromat-
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-1
`IPR2016-00379
`
`
`
`176
`
`CRC Critical Reviews in Analytical Chemistry
`
`ographic approach stems from its relative ease and efficiency. In this review, the enantiomeric
`separations with gas chromatography (GC), liquid chromatography (LC), and thin layer
`chromatography (TLC) are discussed. It appears that the most interesting research in this
`area involves the development of new, highly selective stationary phases. A large number
`of chiral stationary phases (CSPs) were synthesized, and it was found that minor structural
`modifications in a stationary phase may have a tremendous influence on an enantiomeric
`separation.
`
`II. GAS CHROMATOGRAPHY
`
`In GC, the separation of enantiomers is generally achieved in two ways. The first involves
`the use of CSPs.6"" The second involves derivatizing a racemate with a chiral group and
`making a pair of diastereomers, and then separating those on achiral or chiral stationary
`phases.6"" In 1966, Gil-Av and co-workers12 first made CSPs for GC and demonstrated their
`usefulness for separating enantiomers. This approach attracted great attention since it enabled
`one to precisely measure enantiomeric purities in complex mixtures while utilizing very
`small samples. The use of chiral derivatization methods has a few limitations: (1) active
`functional groups for forming diastereomeric derivatives are required, (2) it is not easy to
`get optically pure chiral reagents, (3) individual enantiomers have different reaction rates,
`and (4) the diastereomeric mixture must be chemically and stereochemically stable.
`
`A. CSPs
`/. Amino Acid Derivatives
`Gil-Av and co-workers12 first introduced an enantioselective GC stationary phase which
`consisted of N-trifluoroacetyl (N-TFA) L-isoleucine lauryl ester. This liquid phase was used
`for the separation of N-TFA-a-amino acid esters. It showed excellent resolution for volatile
`N-TFA-ot-amino acid esters (e.g., N-TFA-valine and leucine esters of 2-butanol), but was
`not suitable for N-TFA-alanine esters, 2-heptyl acetate, and a-acetoxypropionate esters of
`2-butanol.
`Feibush13 reported a highly efficient diamide-liquid stationary phase, n-dodecanoyl-L-
`valine-/m-butylamide. This phase was less polar than dipeptide phases but produced rela-
`tively high enantioselectivities. Retention times were generally shorter than on the more
`polar phases. However, this phase was limited to the separation of amino acid and amino
`alcohol enantiomers. Also, this phase had relatively high column bleeding at the optimum
`column operating temperature of 130 to 140°C. To reduce the column bleeding, AT-docos-
`anoyl-L-valine fm-butylamide and N-lauroyl-L-valine 2-methyl-2-heptadecylamide were used
`as the active coating.14 N-TFA isopropyl esters of 14 protein amino acids were studied, and
`12 of the 14 compounds were resolved on relatively short columns (2 to 4 m). These packed
`columns could be used at temperatures as high as 190°C (for 7V-docosanoyl-L-valine tert-
`butylamide) and 180°C (for N-lauroyl-L-valine 2-methyl 2-hepadecylamide) without losing
`their efficiency, even after prolonged use. The influence of structural factors on selectivity
`of CSPs (R,CONHCH[CH(CH3)2]CONHR2) was studied.15 Hobo and co-workers studied
`the stereoselectivity of the chiral diamide stationary phase, N-lauroyl-L-valine rerf-butylam-
`ide, diluted with diethylene glycol succinate polyester (DEGS) or squalane for the separation
`of several N-trifluoroacetyl isopropyl esters of amino acids.16 The addition of DEGS greatly
`reduced the resolution coefficient. It was thought that this resulted from the blocking of the
`chiral sites of the stationary phase by the polar solvent. When squalane was added, the
`resolution coefficient either remained the same or increased. Dilution of the chiral liquid
`stationary phase with achiral additives may lead to the dissolution of the hydrogen-bonded
`networks of these dipeptide-analog phases. It was concluded that the unblocked monomeric
`form of the diamide gave higher retention and stereoselectivity.
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-2
`IPR2016-00379
`
`
`
`Volume 19, Issue 3 (1988)
`
`177
`
`Table 1
`RESOLUTION FACTORS OF W-TFA-0-ACYL DERIVATIVES OF
`AMINOALKANOLS
`
`Propionyl
`
`Acyl group
`
`Isobutyryl
`
`Pivaloyl
`
`r(D/L)
`
`1.040
`
`1.076
`
`1.089
`
`1.087
`
`1.071
`
`1.071
`
`1.094
`
`1.057
`
`T(°C)C
`
`r*
`
`"(°C)C
`
`r*
`
`r
`
`b
`
`T{°CY
`
`120
`
`120
`
`120
`
`120
`
`140
`
`140
`
`120
`
`140
`
`1.051
`
`140
`
`1.089
`
`1.075
`
`1.071
`
`1.062
`
`1.067
`
`1.105
`
`1.085
`
`140
`
`160
`
`160
`
`170
`
`170
`
`140
`
`140
`
`9.40
`9.88
`14.56
`15.86
`10.94
`11.76
`16.86
`18.06
`19.86
`21.10
`31.98
`34.14
`16.90
`18.68
`29.84
`32.38
`
`1.075
`
`1.112
`
`1.097
`
`1.096
`
`1.082
`
`1.083
`
`1.125
`
`1.130
`
`12.00
`12.90
`15.86
`17.64
`11.48
`12.60
`18.14
`19.88
`20.88
`22.60
`33.14
`35.90
`18.66
`21.00
`30.04
`33.94
`
`140
`
`140
`
`160
`
`160
`
`170
`
`170
`
`140
`
`140
`
`Aminoalkanol
`
`Enantiomer
`
`r*
`
`14.58
`15.16
`25.00
`26.90
`42.28
`46.00
`69.88
`76.00
`51.88
`55.58
`88.80
`95.10
`28.44
`31.12
`24.10
`25.48
`
`L D L D L D L D L D L D L D L D
`
`2-Aminopropan-l-ol
`
`2-Aminobutan-l-ol
`
`2-Aminopentan-l-ol
`
`2-Aminohexan-I-ol
`
`2-Aminoheptan-l-ol
`
`2-Aminooctan-l-ol
`
`2-Amino-3-methyl-
`butan-1-ol
`2-Amino-4-methyl-
`pentan-1-ol
`
`'
`• Corrected retention time (minutes).
`b L/D, resolution factor, i.e., ratio of the corrected retention time of the L over that of the D enantiomer, calculated
`with r values expressed to the second decimal place.
`c Temperature at which good peak resolution and relatively short retention were observed.
`
`From Charles, R. and Gil-Av, E., J. Chromatogr., 195, 317, 1980. With permission.
`
`Koenig and co-workers prepared chiral liquid stationary phases by coupling S-a-phen-
`ylethylamine with S-2-hydroxyisopentanoic acid and S-2-hydroxyoctanoic acid.17 The sep-
`arations of the enantiomers of racemic amines, amino alcohols, and hydroxy acids were
`achieved using these stationary phases.
`Weinstein and co-workers18 reported that CSPs containing an amide group and an asym-
`metric carbon atom, attached to the nitrogen atom [RCONHCH(CH3)R']( often showed
`enantiomeric selectivity for amino acids containing N-trifluoroacetylamine, N-trifluoroace-
`tylamino acid esters, and a-methyl- or a-phenylcarboxylic acid amides. The best efficiency
`was obtained when R' was an aromatic group. Particularly good separations were obtained
`when R' was an a-naphthyl group, as in Af-lauroyl-S-a-(l-naphthyl)ethylamine. Also, the
`highest resolution factors were found for aromatic solutes such as iV-trifluoroacetyl-a-phen-
`ylethylamine and a-phenyl butyric acid amides.
`Other successful GC separations of enantiomers were reported on modified diamide
`phases.19-20 A/-Docosanoyl-L-valine-2-(2-methyl)-n-heptadecyl amide phases20 were used at
`column temperatures of up to 200°C. These phases showed excellent stereoselectivity for
`various compounds (aromatic N-TFA amines, TV-TFA-O-acetyl amino alcohol, N-TFA-a-
`methylvaline isopropyl ester, and N-TFA-esters of a- and -y-amino acids). Table 1 shows
`the resolution factors of Af-TFA-0-acetyl derivatives of amino alkanols. Lochmuller and co-
`workers reported ureide phases (i.e., carbonyl-bis [amino acid esters])2124 and examined the
`separation mechanisms of these phases. The general formula of the ureide phase is shown
`in Figure 1. Nuclear magnetic resonance (NMR) evidence was interpreted to indicate that
`only one significant portion of attachment is involved in the fonnation of diastereoisomeric
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-3
`IPR2016-00379
`
`
`
`178
`
`CRC Critical Reviews in Analytical Chemistry
`
`R'
`0
`R'
`II
`I
`I
`R-0—C— C—N—C—N-C—C—0—R
`II
`I
`I
`I
`II
`0
`0 H H
`H H
`FIGURE 1. General formula of the ureide phase. (From Lochmuller, C. H.
`and Souter, R. W., J. Chromatogr., 113, 283, 1975. With permission.)
`
`H
`H
`I
`I
`R4—0 — C —C — N — C — C — N —C —CF 3
`II
`I
`I
`II
`I
`I
`II
`0
`R, H
`0
`R2 H
`0
`*
`• £•
`^
`^"
`H H
`0
`I
`II
`p _Q Q Q N
`I
`
`*>•
`^
`0
`IB
`£
`
`£p
`
`L-Dlpeptlde ester phase
`
`D-Conflguration of solute
`
`H
`H
`I
`I
`R 4 -0 — C—C — N — C — C— N— C — CF3
`H
`I
`I
`II
`I
`I
`I
`0 R, H 0 R2 H
`0
`
`L-DIpeptlde ester phase
`
`0
`
`H H
`
`0
`
`c
`
`o
`
`R
`
`L-Conflguratlon of solute
`
`c
`
`I R
`
`Cp _ c
`
`N
`
`3
`FIGURE 2. Schematic showing the formation of H-bonded diastereomeric association complexes. (From Loch-
`muller, C. H. and Souter, R. W., J. Chromatogr., 113, 283, 1975. With permission.)
`
`association complexes. This, of course, is highly unlikely given the well-known geometrical
`requirements (i.e., three-point attachment) for chiral recognition. Also studied were the
`effects of stationary phase structure on selectivity.23
`
`2. Peptide Phases
`Gil-Av and co-workers also developed optically active stationary phases composed of N-
`TFA-a-amino acids (e.g., TV-L-valyl-L-valine isopropylester, N-TFA-L-valyl-L-valine cy-
`clohexyl ester,25 Af-acetyl-L-valyl-L-valine isopropyl ester,26 arid N-TFA-(L-valyI)2-L-valine
`isopropyl ester26). However, these phases had low efficiencies and long analysis times
`because of limited thermal stability. For example, a AT-TFA-L-valyl-L-valine cyclohexyl ester
`stationary phase was used at 110°C. Above 110°C, the column began to bleed, and at
`temperatures below 110°C, the column efficiency was very low and the peaks were broad.
`In peptide phases, the resolution of enantiomers is thought to be through the formation
`of hydrogen-bonded diastereomeric association complexes between the enantiomeric com-
`pounds and asymmetric chiral phases. Figure 2 shows this interaction on a dipeptide ester
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-4
`IPR2016-00379
`
`
`
`Volume 19, Issue 3 (1988)
`
`179
`
`phase: CF3CO-NH —CH —CO — NH —CH — C00C6H,,
`I
`I
`
`N-TFA-L-alanyl-L-alanlne cyclohexyl
`ester
`
`N-TFA-L-a-amlno-n-butyryl-L-a-amlno-n-
`butyric acid cyclohexyl ester
`
`N-TFA-L-norvaiyl-L-norvaltne cyclohexyl
`ester
`
`N-TFA-L-norleucyl-L-norleuclne
`cyclohexyl ester
`
`Rj, R2
`
`-CH3
`
`abbreviation
`
`ala-ala
`
`-CH2-CH3
`
`aba-aba
`
`-CH2-CH2-CH3
`
`nval-nval
`
`-CH2-CH2-CH2-CH3 nleu-nleu
`
`FIGURE 3. Homologous dipeptide stationary phases. (From Parr, W. and Howard P.
`1973. With permission.)
`
` Y.,Anal. Chem., 4 5, 711,
`
`phase.24 Parr and co-workers synthesized many dipeptide ester phases.2731 They also studied
`the structural effects on selected dipeptides as stationary phases for the enantiomeric sepa-
`ration of amino acids.32 Four optically-active dipeptide cyclohexyl esters (N-TFA-L-alanyl-
`L-alanine cyclohexyl ester, //-TFA-L-a-amino-w-butyryl-L-ot-amino-n-butyric acid cyclo-
`hexyl ester, N-TFA-L-norvalyl-L-norvaline cyclohexyl ester, and //-TFA-L-norleucyl-L-nor-
`leucine cyclohexyl ester) were used (see Figure 3). Separation factors and thermodynamic
`properties of liquid stationary phase-solute interactions were studied. Table 2 shows relative
`retention times and separation factors for Af-TFA-D,L-amino acid isopropyl esters. An increase
`in the size of the alkyl substituent on the asymmetric centers of the dipeptide solvent produced
`a greater liquid stationary phase-solute interaction. When the analogous modification was
`made with the side chain on the a carbon, the interaction decreased.
`In order to obtain a dipeptide ester phase of greater thermal stability and higher molecular
`weight, aromatic amino acids such as phenylalanine were employed (i.e., N-TFA-L-phen-
`ylalanyl-L-leucine cyclohexyl ester,29-31-33-34 N-TFA-L-phenylalanyl-L-phenylalanine cyclo-
`hexyl ester,35-36 and N-TFA-L-phenyl-L-aspartic acid bis [cyclohexyl] ester3537). These phases
`were able to be used at higher temperatures (130 to 165°C) than the original dipeptide phases
`and therefore were more effective for separating less volatile enantiomeric amino acid
`derivatives.
`Koenig and co-workers used glass capillaries because they found partial decomposition
`of some compounds during chromatography with steel capillaries. They reported that steel
`capillaries promoted the decomposition of solutes (cysteine, serine, and threonine derivatives)
`to a greater extent than glass capillaries did. The glass capillary columns also possessed
`higher efficiencies compared to analogous stainless steel capillaries. Some of the stationary
`phases synthesized and used with glass capillaries include /V-TFA-L-phenylalanyl-L-phen-
`ylalanine cyclohexyl ester, Af-TFA-L-phenylalanyl-L-leucine cyclohexyl ester, and N-TFA-
`L-phenylalanyl-L-aspartic acid bis (cyclohexyl)ester.35 Figure 4 shows the separation of
`racemic TFA-amino acid isopropyl esters on a glass capillary coated with N-TFA-L-valine-
`L-valine cyclohexyl ester.
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-5
`IPR2016-00379
`
`
`
`180
`
`CRC Critical Reviews in Analytical Chemistry
`
`Table 2
`RELATIVE RETENTION TIMES AND SEPARATION FACTORS FOR
`A/-TFA-D.L-AMINO ACID ISOPROPYL ESTERS ON FOUR DIFFERENT
`STATIONARY PHASES
`
`ala-ala
`110
`
`°C
`
`r
`
`a
`
`1.055
`
`1.064
`
`1.071
`
`1.059
`
`1.066
`
`1.078
`
`1.064
`
`1.050
`
`0.289
`0.305
`0.476
`0.507
`0.382
`0.409
`0.597
`0.631
`0.844
`0.900
`0.580
`0.625
`0.940
`1.000
`0.387
`0.407
`
`Amino
`acid
`
`D-ala
`L-ala
`D-aba
`L-aba
`D-val
`L-val
`D-nval
`L-nval
`D-leu
`L-leu
`D-ile
`L-ile
`D-n!eu
`L-nleu
`D-/-leu
`L-/-leu
`
`aba-aba
`°C
`100
`
`r
`
`a
`
`1.091
`
`1.091
`
`1.090
`
`1.086
`
`1.099
`
`1.105
`
`1.098
`
`1.052
`
`0.246
`0.268
`0.341
`0.372
`0.366
`0.399
`0.563
`0.611
`0.764
`0.839
`0.535
`0.591
`0.911
`1.000
`0.362
`0.381
`
`Amino
`acid
`
`D-ala
`L-ala
`D-aba
`L-aba
`D-val
`L-val
`D-nval
`L-nval
`D-leu
`L-leu
`D-ile
`L-ile
`D-nleu
`L-nleu
`D-/-leu
`L-f-leu
`
`nval-nval
`100
`°C
`
`nleu-nleu
`100
`°C
`
`Amino
`acid
`
`D-ala
`L-ala
`D-aba
`L-aba
`D-val
`L-val
`D-nval
`L-nval
`D-leu
`L-leu
`D-ile
`L-ile
`D-nleu
`L-nleu
`D-/-Ieu
`L-/-Ieu
`
`r
`
`a
`
`1.100
`
`1.097
`
`1.083
`
`1.092
`
`1.110
`
`1.108
`
`1.108
`
`1.044
`
`0.231
`0.254
`0.330
`0.362
`0.364
`0.397
`0.553
`0.604
`0.749
`0.832
`0.535
`0.592
`0.903
`1.000
`0.364
`0.381
`
`Amino
`acid
`
`D-ala
`L-ala
`D-aba
`L-aba
`D-val
`L-val
`D-nval
`L-nval
`D-leu
`L-leu
`D-ile
`L-ile
`D-nleu
`L-nleu
`D-f-leu
`L-/-Ieu
`
`r
`
`Of
`
`1.096
`
`1.091
`
`1.080
`
`1.090
`
`1.107
`
`1.101
`
`1.107
`
`1.033
`
`0.228
`0.249
`0.327
`0.356
`0.366
`0.396
`0.551
`0.600
`0.748
`0.828
`0.541
`0.595
`0.904
`1.000
`0.368
`0.380
`
`Note: See Figure 3 for the stationary phase structures and abbreviations, r. Relative retention time;
`reference compound in all cases is A'-TFA-L-norleucine isopropyl ester.
`
`From Parr, W. and Howard, P. Y., Anal. Chem.. 45, 711, 1973. With permission.
`
`u
`
`MMM
`
`M
`
`FIGURE 4. Gas chromatogram showing the separation of TFA-amino acid isopropyl esters.
`(From Koenig, W. A. and Nicholson, G. J., Anal. Chem., 47, 951, 1975. With permission.)
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-6
`IPR2016-00379
`
`
`
`Volume 19, Issue 3 (1988)
`
`181
`
`Carbin and co-workers synthesized several systematically substituted, optically active
`peptide stationary phases (N-TFA-L-valyl-L-leucine cyclohexyl ester, N-PFP-L-valyl-L-leu-
`cine cyclohexyl ester, N-TFA-L-leucyl-L-valine cyclohexyl ester, and JV-TFA-[L-leucyl]2-L-
`leucine cyclohexyl ester).38 They studied the effects of systematic changes in the side chains
`of the peptides and found that the separation factor (a) values were sensitive to changes in
`the structure of the side group at the ester end.
`Some of the dipeptide ester phases were converted to tripeptide phases.38 Af-Pentafluo-
`ropropionyl (N-PFP) groups were substituted for the Af-TFA group.29'3138 Many dipeptide
`and tripeptide stationary phases have been used in separating amino acids. The tripeptide
`phases have better thermal stability than dipeptide phases do, but their resolving powers
`were found to be very similar.3840 //-TFA-L-methionyl-L-methionine cyclohexyl ester and
`disulfide and disulfone derivatives of the TFA phases have been synthesized.41 Af-TFA-L-
`methionyl-L-methionine cyclohexyl ester stationary phases gave good separations of N-TFA-
`D,L-amino acid isopropyl esters. Although these phases were stable at temperatures as high
`as 150°C, increasing the temperature reduced the separation factors. N-PFP-derivatized
`stationary phases have been shown to be stable, giving larger a values and shorter retention
`times than N-TFA phases. In general, the peptide stationary phases that produced the best
`resolution of separated amino acid derivatives contained N-TFA groups or bulky ester groups
`(e.g., tert-bulyl, isopropyl, or cyclohexyl ester). Optimal performance of these columns was
`obtained at lower temperatures.
`Oi and co-workers synthesized the 5-triazine derivative of a tripeptide ester.42'44 This phase
`gave good separations of various a-alkyl phenylacetic acid and aryl alkylamine enantiomers.
`For example, Ar,N',N"-(2,4,6-[l,3,5-triazine]-triyl)-tris-L-valine isopropyl esters (OA-100)
`gave excellent separations of N-TFA-amino acid ester and N-TFA-amine enantiomers. How-
`ever, the enantioselectivity of these phases was insufficient for the resolution of some
`carboxylic acids and amines; therefore, the determination of the optical purity was impos-
`sible.42"44 OA-200 (A^,A^'-[2,4-(6-Ethoxy-l,3,5-triazine)diyl])-bis-(L-valyl-L-valine isopropyl
`ester), OA-300 (A^N'-[2,4-(6-ethoxy-l,3,5-triazine)diyl])-bis-(L-valyl-L-valyl-L-valine iso-
`propyl ester), and OA-400 (W,/V',AT'-[2,4,6-(l,3,5-triazine)triyl])tris-(Na-lauroyl-L-lysine-
`fe/7-butylamide) were used for separating enantiomers of a-hydroxycarboxylic acid esters
`and some alcohols.45-46 Racemic l-phenyl-2,2,2-trifluoroethanol was separated on an OA-
`400 column. The S isomers of the chiral alcohols were found to elute before the R isomers.
`N-Lauroyl-(S)-proline-(S)-l-(a-naphthyl)ethyIamide also was used to separate the R- and 5-
`isomers of l-phenyl-2-(4-tolyl)ethylamine.47
`
`3. Polymer Phases
`Frank and co-workers synthesized a CSP named Chirasil-val48"50 by coupling L-valine-
`te/7-butylamide to a copolymer of dimethylsiloxane and carboxy-alkyl-methyl-siloxane of
`appropriate viscosity and molecular weight. This polymeric GC stationary phase had a higher
`thermal stability and a lower volatility than the monomeric stationary phases. This polymeric
`phase was used to separate amino acid and some amino alcohol enantiomers,51"54 enantiomeric
`drugs, and metabolites.55 This phase allowed a greater range of temperatures to be used in
`the separation of enantiomers by GC. As a result, compounds of lower volatility could be
`analyzed for the first time. Mass spectrometric detection was also used with this polymeric
`stationary phase.56
`A thermally stable chiral phase was made by incorporating L-valine-/e/7-butylamide into
`the well-known GC stationary phase polycyanopropylmethyl phenylmethyl silicone.57 The
`cyano groups of commercially available GC phases (such as OV 225 and Silar IOC) were
`converted by acid hydrolysis into carboxylic groups, then into acid chlorides, and were
`finally coupled with L-valine-ter/-butylamide. This phase was similar to Chirasil-val, but
`had structural differences (e.g., propyl linkages instead of ethyl linkages to the chiral center
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-7
`IPR2016-00379
`
`
`
`182
`
`CRC Critical Reviews in Analytical Chemistry
`
`0
`0
`II
`II
`CH3-S1 —(CH 2)2 —C —NH —CH —C —NH —CH —
`I
`I
`I
`0
`™
`CH3
`•
`CH3 CH3
`C H 3 - Si — CH3
`
`I0 F
`
`IGURE 5. The molecular structure of XE-60-5-valine-5(or fl)-a-phenylethylamine.
`
`and the presence of phenyl groups on the silicone matrix). This phase was used for the
`separation of amino acid enantiomers at temperatures ranging from 60 to 230°C. The sep-
`aration of several protein amino acid enantiomers in one run using temperature programming
`was demonstrated. This phase also had good enantioselectivity for praline and other sec-
`ondary amino acid derivatives.58
`Koenig and co-workers59 separated chiral aliphatic, aromatic, and monoterpene alcohols
`as isopropyl urethane derivatives on glass capillary columns coated with XE-60-5-valine-5-
`a-phenylethylamide (Figure 5). By forming stable isopropyl urethanes in a facile derivati-
`zation step, the enantioselective intermolecular interactions between the racemic alcohols
`and the CSP were sufficiently enhanced to give enantiomeric resolution. This column was
`used for the separation of carbohydrate enantiomers as either TFA derivatives or TFA-methyl
`glycosides.17-60 A XE-60-/?-vaIine-^?-a-phenylethylamide column was used for the enan-
`tiomeric separation of phenolic a- and p-receptor-active drugs after derivatizing with di-
`azomethane and phosgene.61 Schomburg and co-workers reported modified GC phases for
`the separation of optically active compounds.62 Cross-linking experiments for the immobi-
`lization of XE-60-L-valine-(S)- and XE-60-L-valine-(ft)-a-phenylethylamine within capillary
`columns were successfully completed.62 Another approach to the modification of silicone
`polymers was reported by Koenig and Benecke.63 They bonded chiral constituents onto
`silicone OV-225 after reducing the cyano groups to amino groups with lithium aluminum
`hydride (LiAlH4), and then coupled benzyloxycarbonyl-L-valine and benzyloxycarbonyl-L-
`leucine to the amino groups (see Figure 6). These phases separated enantiomers of trifluo-
`roacetylated secondary amines and amino alcohols. It was found that the D amino alcohols
`had longer retention times than the L enantiomers, while the S enantiomers of amines had
`longer retention times than the R enantiomers. These columns were also used for resolving
`the racemates of arabinitol, fucitol, mannitol, and C5-8,2-aminoalkanes as the trifluoroacetyl
`derivatives.64
`
`4. Miscellaneous Phases
`Chiral liquid stationary phases composed of di-/-menthyl-( + )-tartrate and di-<//-methyI-
`(— )-malate were used in conjunction with glass capillary columns to separate enantiomers
`of amino acids, amines, and carboxylic acids.65 Figure 7A shows the GC separation of N-
`trifluoroacetyl-D,L-leucine isopropyl ester, and Figure 7B shows that of racemic N-penta-
`fluoro propyl-a-phenylethylamine. Oi and co-workers also developed some amide-type sta-
`tionary phases which contained two asymmetric carbon atoms. These were used to separate
`some carboxylic acid and amine enantiomers.66 Liquid stationary phases composed of N-
`(lfi,3/?)-rrarts-chrysanthemoyl-(/?)-Ha-naphthyl)ethylamine phase,66 which contains two
`asymmetric carbon atoms attached to both nitrogen and carbon atoms of the amide group,
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-8
`IPR2016-00379
`
`
`
`Volume 19, Issue 3 (1988)
`
`183
`
`CH3
`
`0
`
`b\
`
`1C
`
`6H 5
`
`0
`
`51
`
`1(CH2)3
`
`1C
`
`H2-NH2
`
`L1A1H4
`
`Ether
`
`CH3
`
`CH3
`
`—si —o —si—o
`I
`I
`(CH 2) 3
`C 6H S
`
`IC
`
`SN
`
`OV-225 amine
`
`CH3
`
`— Si —0 — S I — 0-
`1
`(CH2)3
`
`1C
`
`6H 5
`
`1C
`
`H2
`1
`NH-CO-CH-NH-Z
`
`1R
`
`Ri
`
`Z-NH-CH-COOH
`
`DCC/CHC13
`
`R:a)
`b
`
`CH(CH3)2
`> CH2-CH(CH3)2
`
`FIGURE 6. Schematic showing the synthetic pathway for the preparation of new chiral stationary phases by
`reduction of OV-225 polysiloxane. (From Koenig, W. A. and Benecke, I., J. Chromalogr., 209, 91, 1981. With
`permission.)
`
`showed better enantioselectivity than those composed of N-(l/?,3/?)-fra/i.r-chrysanthemoyl-
`lauryl-amine or N-lauroyl-(i?)-l-(a-naphthyl)ethylamine, both of which contain only one
`asymmetric center. An 0-(lZ?,3/?)-fran.s-chrysanthemoyl-(.S)-mandelic acid (/?)-l-(a-naph-
`thyl)-ethylamide stationary phase was synthesized and shown to have good enantioselectivity
`for chrysanthemic acid ester and 3-(2,2-dichlorovinyl)2,2-dimethylcyclopropanecarboxylic
`acid esters.67 Table 3 lists some of the enantiomers separated with this phase, as well as
`pertinent chromatographic data, a- and p-Cyclodextrin were used by Sybilska and co-workers
`as stationary phase coatings in the GC separation of enantiomers of a- and p-pinene.68-69
`
`B. Derivatization Methods
`/. Alcohols
`Diastereomers are often separated on conventional, achiral GC stationary phases. Con-
`sequently, there are many reports in which racemates have been resolved after derivatization
`with an optically pure reagent. This method is particularly important for difficult-to-resolve
`compounds such as branched aliphatic alcohols70 and 2,3-butanediol esters.71 The first GC
`resolution of racemic mixtures of secondary n-alkanols as the corresponding diastereoiso-
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-9
`IPR2016-00379
`
`
`
`184
`
`CRC Critical Reviews in Analytical Chemistry
`
`B
`
`50
`
`60
`
`70
`
`80
`
`RETENTION TIKE ( HIN )
`
`80
`RETEKTIOH TIME ( HIM )
`
`90
`
`FIGURE 7. Gas chromatogram showing the separation of the enantiomers of (A) AMrifluoroacetyl-D.L-leucine
`isopropyl ester and (B) yV-pentafluoropropyl-a-phenylethylamine. (From Oi, N., Kitahara, H., and Doi, T., J.
`Chromawgr.. 207, 252, 1981. With permission.)
`
`meric a-hydroxypropionates was reported in 1962.72-73 Resolution of enantiomorphs by
`conversion to volatile diastereoisomers followed by rectification and reconversion to the
`original compounds was done by Bailey and Hass.74 Partial resolution of racemic 2-butanol
`and 2-pentanol was done using levorotatory lactic acid. Lactic acid is commonly used as a
`derivatization reagent because it is inexpensive and easily obtained in pure form.
`Bulk dissymmetry at the alcoholic asymmetric carbon atom and the distance between
`optical centers and their effects on separation have been studied by Rose and co-workers.75
`They investigated the degree of separation of diastereomeric esters of acetylated lactic acid
`(a-acetoxypropionic acid) as a function of systematic variation in alcohol structure in order
`to gain insight into the factors responsible for the separation of these esters. Using a 1,2,3-
`tris(2-cyanoethoxy) propane column, the separation factors of the diastereoisomeric esters
`of the secondary alcohols (±)-2-n-butanol, (±)-2-«-pentanol, (±)-2-«-hexanol, (±)-2-n-
`heptanol, and (± )-2-/i-octanol) were found to change from 1.059 to 1.107 in order of
`increasing chain length. The separation factors of the esters of these alcohols also changed
`from 1.016 to 1.079 on a D.C. (Dow Corning Corp.) 710 silicone oil column. Gault and
`Felkin76 found that the diastereoisomers of unsaturated alcohols or diols are less likely to
`form an intramolecular hydrogen bond. Among the more effective diastereomeric derivative
`methods for resolving alcohols include the use of /V-TFA-(L)-( + )-alanyl ester,77 W-TFA-
`(5)-(-)-prolyl ester,47 (+ )-/ra«s-chrysanthemoyl ester, 78 (S)-acetoxypropionyl ester,79 and
`(/?)-( + )-l-phenylurethane.80 A comparison of the easy separation of three different deriv-
`atives of racemic alcohols as well as the determination of the configuration of myrmica ant
`3-octanol was done by Attygalle and co-workers.81 Three diastereomeric derivatives of 3-
`( ±) octanol were prepared. These were N-TFA-(5)-( + )-alanyl ester, the N-TFA-(S)-(-)-
`proyl ester, and the (+ )-rrans-chrysanthemoyl ester. They used OV-1 and DEGS columns,
`and the (+ Hra/w-chrysanthemoyl ester was decided to be the most useful derivatizing
`reagent for the study of the naturally occurring 3-octanol (Table 4).
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-10
`IPR2016-00379
`
`
`
`VI
`oo
`
`O
`
`oo
`oo
`
`From Oi, N., Kitahara, H., and Doi, T., J. Chromalogr., 254, 282, 1983. With permission.
`
`** Separation factor calculated from 2nd peak/1st peak retention time ratio.
`* Measured from solvent peak.
`
`themoyI-(5)-mande!ic acid (#)-l-(a-naphthyl)ethylamide. Carrier gas, helium at 0.7 to 0.8 m^/min.
`
`Note: Chromatography on 40 m x 0.25 mm I.D. glass capillary columns coated with O-(l/?,3K)-'rans-chrysan-
`
`1.013
`1.013
`1.023
`1.016
`1.020
`1.020
`1.014
`1.000
`1.024
`1.024
`1,020
`1.014
`1.025
`1.026
`1.024
`1.024
`1.025
`1.026
`
`102.0
`80.69
`233,3
`176.5
`158.4
`120.6
`58.88
`44.48
`177.5
`131.9
`60.20
`44.94
`101.4
`76.49
`60.54
`46.04
`42.10
`32.67
`
`100.7
`79.63
`22B.1
`173.8
`155.3
`118.2
`58.08
`44.48
`173.4
`128.8
`59.00
`44.30
`98.89
`74.56
`59.14
`44.96
`41.07
`31.83
`
`150
`
`120
`
`120
`
`100
`
`TOO
`
`100
`
`100
`
`2nd ptak
`
`1st ptak
`
`(min)
`Retention time*
`
`rc>
`temperature
`Column
`
`.C-CH-CH-CH-COOfl
`
`X
`
`.016
`.012
`.025
`.017
`.021
`.022
`.013
`.000
`1.022
`1.019
`.019
`.010
`.022
`
`.018
`.007
`.021
`
`142.5
`132.1
`226.9
`214.2
`153.3
`145.6
`14.37
`13.53
`44.92
`42.86
`15.55
`14.83
`25.97
`
`14.87
`14.41
`10.12
`
`140.2
`130.5
`221.3
`210.7
`15H.2
`I-J2.4
`14.19
`13.53
`43.97
`42.06
`15.26
`14.68
`25.42
`
`14.61
`14.31
`9.91
`
`2nd peak
`
`1st peak
`
`(min)
`Retention time*
`
`CM
`
`120
`
`10(1
`
`ion
`
`100
`
`100
`
`100
`
`100
`
`100
`
`100
`
`ro
`temperature
`Column
`
`C — CH-CH-COOR
`
`X
`
`CH
`
`Ch
`Trans
`Os
`Trans
`Ch
`Twin
`Ch
`Tram
`Ch
`Trans
`Ch
`Trans
`Cis
`Trans
`Cis
`Trans
`Cis
`
`n-C.H,,
`
`cyclo-C»H,,
`
`n-C H ,
`
`lert.-C,H,
`
`n-C4H,
`
`I'JO-CJH,
`
`n-CjH.
`
`C2»s
`
`CH5
`
`Enanllomtr
`
`GAS CHROMATOGRAPHIC SEPARATION OF ENANTIOMERS
`
`Table 3
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-11
`IPR2016-00379
`
`
`
`SQI
`
`to
`
`3 p
`
`From Attygalle, A. B., Morgan, E. D., Evershed, R. P., and Rowland, S. J., J. Chromatogr., 260, 411, 1983. With permission.
`
`Note: Analyses performed isothermally at 150°C (OV-1) and 16O°C (DEGS).
`
`S S R
`
`R R S
`
`isomer
`Second
`
`isomer
`First
`
`1.1
`
`1.2
`
`.2
`
`R 1
`
`1.04
`
`1.03
`
`1.03
`
`a
`
`2.5
`
`11.7
`
`3.2
`
`2.4
`
`11.4
`
`3.1
`
`S S R
`
`R R S
`
`isomer
`Second
`
`isomer
`First
`
`isomer
`Second
`
`isomer
`First
`
`1.9
`
`1.8
`
`.2
`
`R 1
`
`1.03
`
`1.04
`
`1.03
`
`a
`
`8.6
`
`8.9
`
`2.8
`
`8.3
`
`8.6
`
`2.7
`
`isomer
`Second
`
`isomer
`First
`
`order
`
`Elution
`
`time (min)
`
`Retention
`
`Elution order
`
`time (min)
`
`Retention
`
`DEGS
`
`OV-1
`
`santhematc ester
`(+)-trans-Giry-
`prolyl ester
`/V-TFA-(S)-( - )-
`alanyl ester
`JV-TFA(S)-( + )-
`
`Derivative
`
`RESOLVING AGENTS ON APOLAR (OV-1) AND POLAR (DEGS), WCOT CAPILLARY COLUMNS
`
`COMPARISON OF GC PROPERTIES OF 3-(±)-OCTANOL DERIVATIZED WITH DIFFERENT CHIRAL
`
`Table 4
`
`
`
`LOWER DRUG PRICES FOR CONSUMERS, LLC
`Exhibit 1009-12
`IPR2016-00379
`
`
`
`Volume 19, Issue 3 (1988)
`
`187
`
`Table 5
`SEPARATION FACTORS (a) AND OPERATING
`TEMPERATURES FOR THE SEPARATION OF 3-
`HYDROXY ACIDS AS N-tert BUTYL-CARBAMATE-ferf-
`BUTYLAMIDE DERIVATIVES (A) AND N-
`ISOPROPYLCARBAMATE-ISOPROPYLAMIDE
`DERIVATIVES (B)