Kinetic resolution of protected α-amino acid derivatives by a chiral O-nucleophilic acyl transfer catalyst

Article abstract of DOI:10.1021/ja058010t

The kinetic resolution of α-trifluoroacetamido N-acyl oxazolidinethiones using 5-10% of the chiral, nonracemic O-nucleophilic acyl transfer catalyst 2 is described. A variety of substrates participate in this reaction in excellent yields, with s-factors ranging from 20 to 86. Copyright

Full text of DOI:10.1021/ja058010t

Page S1
Kinetic Resolution of Protected α-Amino Acid Derivatives
by a Chiral O-Nucleophilic Acyl Transfer Catalyst
Gregory T. Notte and Tarek Sammakia*
Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado
80309-0215
Supporting Information
Table of Contents:
S1-S2
S2-S6
S7
General Information
Synthesis and characterization of compounds 9 to 18.
Typical procedure for the kinetic resolutions.
Kinetic resolution of compounds 9 to 18.
S7-S10
S11-S12 Peptide couplings.
S12
S13
S14
Determination of absolute configuration.
Protecting group swap.
Reaction kinetics
S15-S38 1H and 13C NMR spectra for compounds 9 – 18 and 20 – 21 (separate PDF
file).
General Information. All air or moisture sensitive reactions were conducted in
oven dried glassware under a dry nitrogen atmosphere. Methanol, triethyl amine,
ethyltrifluoroacetate, and methylene chloride were distilled from CaH2. Toluene was
washed with concentrated H2SO4, H2O, 1M NaOH, and H2O, dried over MgSO4, then
filtered and distilled from CaH2. Except for the precursor to compound 14, all of the
racemic amino acids were available from commercial sources and most were purchased
from Fluka or Aldrich. Azetidine-2-carboxylic acid was purchased from Toronto
Research, and azepane-2-carboxylic acid was purchased from ASDI. The racemic amino
acid 1,2,3,4-tetrahyrdoisoquinoline-1-carboxylic acid was synthesized by reduction of 1-
isoquinolinecarboxylic acid following the procedure of Schuman.1
Flash
chromatography of substrate 16 (R = isobutyl) was performed on neutral silica gel
(Mallinckrodt Silicar Silica Gel 150, 60-200 mesh (75-250 micron)) following the
procedure of Still.2 Flash chromatography of this substrate on other “ordinary” silica gel
(i.e. silica gel with a pH other than 7.0) led to partial cleavage of the oxazolidinethione
auxiliary. All other flash chromatography was performed using Sorbent Technologies 60
1
2
Shuman R. T.; Rothenberger R. B.; Campbell C. S.; Smith G. F.; Gifford-Moore D. S.;
Gesellchen P. D.; J. Med. Chem. 1993, 36, 314–319.
Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923.

Page S2
1
Å silica gel (32-63 micron). H NMR spectra were recorded at 400 MHz in CDCl3 and
were referenced to residual CHCl3 (7.24 ppm). 13C NMR spectra were obtained at 100
MHz in CDCl3 and were referenced to the center line of the CDCl3 triplet (77.23 ppm).
Chiral HPLC analysis was performed using a semi-preparative Chiralcel OD column.
Compounds were detected by monitoring UV absorbance at 254 nm. Infrared spectra
were recorded as thin films on NaCl plates. Melting points were determined in open
capillaries and are uncorrected. Optical rotations were determined on a JASCO 1030
polarimeter at 25 °C. Exact mass was determined on M+H or M+Na using electrospray
ionization.
Synthesis of Substrates:
General Procedure for Preparation from α-Amino Acids
S
O
S
O
R
R
R
F3C OEt
R3N
O
HN
OH
OH
N
N
N
N
H
DCC, DMAP
TFA
TFA
O
O
O
This is a general procedure and some changes were made for specific substrates.
In general, tetramethylguanidine was used in the protection of the cyclic amino acids
while triethyl amine was used for acyclic amino acids (proline and azetidine-2-carbozylic
acid are exceptions). Excess base was also required if starting from the amino acid
hyrdochloride salt.
To a stirred solution of the α-amino acid (1.0 mmol) in MeOH (1.0 ml, 1.0M) at
room temperature was added amine base (1.0 mmol). Ethyltrifluoroacetate (1.25 mmol)
was then added to this suspension, and the reaction was stirred at room temperature for
24-48 hours. The reaction progresses from a suspension to a nearly clear solution upon
completion. MeOH was removed under reduced pressure, and the residue was acidified
by the addition of a slight excess of 2M HCl. The aqueous layer was extracted with
EtOAc (three times), and the combined organic extracts were washed once with a 10:1
1M HCl / brine solution, then dried over MgSO4, filtered, and concentrated at reduced
pressure to a solid. This material was of sufficient purity for use in the subsequent
coupling reactions.
To a stirred solution of the crude N-TFA protected amino acid prepared above
(1.0 mmol), oxazolidinethione (1.1 mmol) and DMAP (0.10 mmol) in CH2Cl2 (6 ml,
0.1M overall) at room temperature was cannulated dropwise a solution of DCC in CH2Cl2
(4 ml). The reaction was monitored by TLC, and was complete in less than 8 hrs. The
reaction mixture was then filtered to remove dicyclohexyl urea and concentrated at
reduced pressure. Purification by flash chromatography (3:1 hexanes-ethyl acetate)
affords the product as a white solid.

Page S3
S
O
N
N
F3C
O
O
9
Compound 9 was prepared in 77% overall yield from the amino acid.
Tetramethylguanidine (1.0 equiv) was used as the base in the TFA protection.
1H NMR revealed a 4:1 mixture of conformers (400 MHz, CDCl3): δ 6.61 (dd,
.20H, J=6, 10 Hz), 6.45 (dd, .80H, J= 4, 12 Hz), 4.51-4.61 (m, 2H), 4.12-4.34 (m, 2.20
H), 3.99 (dd, .80H, J=4, 16 Hz), 3.50 (dd, .80 H, J=11, 16 Hz), 3.24 (app t, .20H, J= 13
Hz), 2.88 (ddd, .20 H, J=6, 8, 14 Hz), 2.65 (ddd, .80 H, J=4, 8, 14 Hz), 1.79-2.06 (m,
3H), 1.23-1.69 (m, 4H). 13C NMR for the major rotamer (100 MHz, CDCl3): δ 184.99,
173.57, 158.39 (q, J=36 Hz), 116.68 (q, J=285 Hz), 66.92, 60.31, 47.26, 45.67 (q, J=3
Hz), 30.81, 28.63, 28.46, 26.32. mp = 157 °C. IR (cm-1) 3397, 1704, 1679, 1377.
HRMS m/z calcd for C12H15F3N2O3SNa+: 347.0647; found: 347.0653
S
O
N
N
O
O
CF3
10
Compound 10 was prepared in 88% overall yield from the amino acid.
Tetramethylguanidine (1.0 equiv) was used as the base in the TFA protection.
1H NMR revealed a 3.4:1 mixture of two conformers (400 MHz, CDCl3): δ 6.92
(d, .23H, J=7 Hz), 6.68 (dd, .77H, J=4, 7 Hz), 4.46-4.65 (m, 2H), 4.24-4.32 (m, 1H),
4.12-4.19 (m, 1H), 3.86-3.90 (m, 1H), 3.74-3.81 (m, .77H), 3.43-3.52 (m, .23H), 2.18-
2.34 (m, 1H), 1.31-2.04 (m, 5H). 13C NMR for the major rotamer (100 MHz, CDCl3): δ
184.80, 172.21, 157.80 (q, J=36 Hz), 116.54 (q, J=286 Hz), 66.93, 53.83, 47.42, 44.28 (q,
J=4 Hz), 26.58, 24.54, 19.90. mp = 132 °C. IR (cm-1) 2951, 1686, 1376, 1332. HRMS
m/z calcd for C11H13F3N2O3SH+: 311.0671; found: 311.0675
S
O
N
N
O
CF3
O
11
Compound 11 was prepared in 76% overall yield from the amino acid.
Triethylamine (1.0 equiv) was used as the base in the TFA protection.
1H NMR revealed a 7.3:1 mixture of two conformers (400 MHz, CDCl3): δ 6.71,

Page S4
(bm, .12H), 6.27 (dd, .88 H, J=4, 9 Hz), 4.54-4.62 (m, 2H), 4.13-4.32 (m, 2H), 3.68-3.94
(m, 2H), 2.42-2.57 (m, 1H), 2.03-2.16 (m, 3H). 13C NMR for the major rotamer (100
MHz, CDCl3): δ 185.31, 171.76, 155.73 (q, J=38 Hz), 116.28 (q, J=285 Hz), 67.35,
60.77, 48.14 (q, J=3 Hz), 47.83, 29.34, 24.86. mp = 124-126 °C. IR (cm-1) 2985, 1688,
1375, 1328. HRMS m/z calcd for C10H11F3N2O3SH+: 297.0515; found: 297.0523
S
O
N
N
F3C
O
O
12
Compound 12 was prepared in 87% overall yield from the amino acid.
Triethylamine (1.0 equiv) was used as the base in the TFA protection. This compound
was isolated as a white foam and then as a clear oil upon heating and subjection to
vacuum.
1H NMR revealed a 3.1:1 mixture of conformers (500 MHz, CDCl3): δ 6.49-6.52
(m, .24H), 6.28 (dd, .75H, J=6, 10 Hz), 4.58-4.68 (m, 2H), 4.00-4.44 (m, 4H), 2.95-3.10
(m, 1H), 2.23-2.35 (m, 1H). 13C NMR for the major rotamer (100 MHz, CDCl3): δ
185.02, 168.61, 155.96 (q, J=38 Hz), 116.05 (q, J=286 Hz), 68.00, 61.53, 49.40, 47.03.
IR (cm-1) 1700, 1471, 1388. HRMS m/z calcd for C9H9F3N2O3SNa+: 305.0178; found:
305.0181
S
O
N
N
O
O
CF3
13
Compound 13 was prepared in 87% overall yield from the amino acid
hydrochloride salt. Tetramethylguanidine (3.0 equiv) was used as the base in the TFA
protection, and 2.05 equiv of ethyltrifluoroacetate were used.
1H NMR revealed a 3:1 mixture of conformers (400 MHz, CDCl3): δ 7.18-7.35
(m, 4H), 7.10 (d, .25H, J=8 Hz), 6.55 (dd, .75H, J= 6, 8 Hz), 4.88 (s, .25 H), 4.80 (dd,
1.50H, J= 15, 26 Hz), 4.55-4.60 (m, 2H), 4.48 (dd, .25 H, J=9, 18 Hz), 4.23-4.30 (m, 1H),
4.08-4.17 (m, 1H), 3.87 (ddd, .25H, J=6, 9, 15 Hz), 3.58 (dd, .75H, J=6, 16 Hz), 3.42
(ddd, .50H, J=3, 17, 26 Hz), 3.13 (dd, .75H, J=8, 15 Hz). 13C NMR for the major rotamer
(100 MHz, CDCl3): δ 185.17, 171.78, 158.31 (q, J=36 Hz), 133.24, 132.76, 128.73,
128.00, 127.91, 126.00, 116.47 (q, J=285 Hz), 67.23, 55.24, 47.40, 46.87 (q, J=4 Hz),
30.64. mp = 74-76 °C. IR (cm-1) 1686, 1375, 1330. HRMS m/z calcd for
C15H13F3N2O3SH+: 359.0671; found: 359.0667

Page S5
N
O
CF3
S
O
N
O
14
Compound 14 was prepared in 63% overall yield from 1-isoquinolinecarboxylic
acid. In the reduction of 1-isoquinolinecarboxylic acid by the method of Shuman, it was
observed that increasing the pressure of H2 from 40 to 75 psi, and allowing the reaction to
stir for two days afforded nearly quantitative yield. Tetramethylguanidine (3.0 equiv)
was used as the base in the TFA protection, and 2.05 equiv of ethyltrifluoroacetate were
used.
1H NMR revealed a 30:1 mixture of conformers, for the major rotamer (400 MHz,
CDCl3): δ 8.35 (s, 1H), 6.64 (d, 1H, J=7 Hz), 7.19-7.30 (m, 3H), 4.56-4.63 (m, 1H), 4.50
(app q, 1H J=9 Hz), 4.30 (ddd, 1H, J=8, 10, 11 Hz), 4.15-4.21 (m, 1H), 4.07 (ddd, 1H,
J=6, 9, 11 Hz), 3.75 (ddd, 1H, J=4, 9, 16 Hz), 3.25 (ddd, 1H, 4, 9, 15 Hz), 2.90 (ddd, 1H,
J=4, 6, 16 Hz). 13C NMR for the major rotamer: (100 MHz, CDCl3): δ 185.61, 172.03,
157.50 (q, J=37 Hz), 135.55, 129.52, 128.95, 128.60, 127.60, 127.15, 116.43 (q, J=285
Hz), 66.57, 55.19, 47.63, 43.18 (q, J=4 Hz), 29.21. mp = 80-82 °C. IR (cm-1) 1688,
1375, 1333. HRMS m/z calcd for C15H13F3N2O3SH+: 359.0671; found: 359.0677
O
S
F3C
NH
O
N
O
15
Compound 15 was prepared in 74% overall yield from the amino acid.
Triethylamine (1.0 equiv) was used as the base in the TFA protection.
1H NMR (400 MHz, CDCl3): δ 6.86 (br d, 1H, J=8 Hz), 6.73 (dd, 1H, J=3, 9 Hz),
4.53-4.65 (m, 2H), 4.29 (ddd, 1H, J=9, 10, 11 Hz), 4.18 (ddd, 1H, J=6, 9, 11 Hz), 2.50
(hd, 1H, J=3, 7 Hz), 1.06 (d, 3H, J=7 Hz), 0.85 (d, 3H, J=7Hz). 13C NMR (100 MHz,
CDCl3): δ 184.70, 171.83, 157.19 (q, J=37 Hz), 115.88 (q, J=286 Hz), 67.06, 55.75,
47.30, 31.05, 19.60, 15.55. mp = 130 °C. IR (cm-1) 3343, 1721, 1689, 1368. HRMS
m/z calcd for C10H13F3N2O3SH+: 299.0671; found: 299.0678
O
S
F3C
NH
O
N
O
16
Compound 16 was prepared in 70% overall yield from the amino acid.
Triethylamine (1.0 equiv) was used as the base in the TFA protection. This compound
was flashed using neutal silica.
1H NMR (400 MHz, CDCl3): δ 6.90 (d, 1H, J=5 Hz), 6.67-6.73 (ddd, 1H, J=3, 8,

Page S6
11 Hz), 4.53-4.64 (m, 2H), 4.13-4.30 (m, 2H), 1.91 (ddd, 1H, J=3, 10, 14 Hz), 1.66-1.75
(m, 1H), 1.52 (ddd, 1H, J=4, 11, 14 Hz), 1.02 (d, 3H, J=7 Hz), 0.931 (d, 3H, J=7 Hz).
13C NMR (100 MHz, CDCl3): δ 184.39, 173.36, 156.99 (q, J=37 Hz), 115.89 (q, J=286
Hz), 67.07, 51.17, 47.47, 41.02, 25.47, 23.62, 21.45. mp = 146 °C. IR (cm-1) 3329,
1709, 1689, 1374. HRMS m/z calcd for C11H15F3N2O3SH+: 313.0828; found: 313.0830
O
S
F3C
NH
O
N
O
17
Compound 17 was prepared in 40% overall yield from the amino acid.
Triethylamine (1.0 equiv) was used as the base in the TFA protection. The DCC
coupling was performed at 0 °C.
1H NMR (400 MHz, CDCl3): δ 6.97 (d, 1H, J=6 Hz), 6.66 (ddd, 1H, J=4, 8, 8
Hz), 5.64-5.75 (m, 1H), 5.14-5.20 (m, 2H), 4.54-4.65 (m, 2H), 4.14-4.32 (m, 2H), 2.83-
2.89 (m, 1H), 2.51 (app p, 1H, J=8 Hz). 13C NMR (100 MHz, CDCl3): δ 184.61, 171.60,
156.73 (q, J=37 Hz), 131.13, 126.60, 115.78 (q, J=286 Hz), 67.19, 51.49, 47.38, 36.49.
mp = 132 °C.
IR (cm-1) 3313, 1712, 1695, 1373.
HRMS m/z calcd for
C10H11F3N2O3SH+: 297.0515; found: 297.0522
O
F3C
S
S
N
NH
O
O
18
Compound 18 was prepared in 88% overall yield from the amino acid.
Triethylamine (1.0 equiv) was used as the base in the TFA protection.
1H NMR (400 MHz, CDCl3): δ 7.32, (d, 1H), 6.62 (ddd, 1H, J=4, 9, 17 Hz), 4.57-
4.65 (m, 2H), 4.29 (ddd, 1H, J=8, 9, 12 Hz), 4.22 (ddd, 1H, J=8, 9, 12 Hz), 2.66 (ddd, 1H,
J=6, 9, 14 Hz), 2.51-2.56 (m, 1H), 2.10 (s, 3H), 1.96-2.03 (m, 1H). 13C NMR (100 MHz,
CDCl3): δ 184.58, 171.66, 157.09 (q, J=38 Hz), 115.79 (q, J=286 Hz), 67.22, 52.26,
47.49, 31.89, 30.02, 15.91. mp = 150-151 °C. IR (cm-1) 3293, 1716, 1698, 1378.
HRMS m/z calcd for C10H13F3N2O3S2Na+: 353.0211; found: 353.0216

Page S7
Typical Procedure for the Kinetic Resolutions
All entries in Table 1 were determined by averaging the s-factors from both runs,
however the enatiomeric excess and % conversion data reported are from individual runs.
S
Me
S
O
O
NMe2
OH
MeOH
Toluene
N
OMe
N
+
+
N
N
N
O
O
CF3
O
CF3
O
O
O
CF3
CF3
2
10
Table 1, Entry 2: A mixture of thioimide 10 (86.8 mg, 0.2797 mmol) and
benzo[k]fluoranthene (279 µL, 0.025 M solution in toluene, 0.000699 mmol, internal
standard) was dissolved in toluene (1.4 ml), and cooled to 0 °C. Methanol (340 µL, 8.40
mmol) was added and stirred until solution became homogenous, at which point the first
aliquot is taken. Catalyst 2 (6.92 mg, 0.02797 mmol) was then added as a solution in
toluene (70 µL) and the reaction stirred at 0 °C while monitoring by HPLC analysis as
follows. 2µL Aliquots of the reaction were taken and diluted with 200 µl of a 1:1
hexanes - ethanol solution, 50 µl of which were directly injected onto a chiralcel OD
column. For some compounds, the liberated oxazolidinethione would co-elute with one
or both enantiomers of starting material. In these cases a 10 µL aliquot was flashed using
a pipet column, and eluting 3 column volumes with 1:1 hexanes/EtOAc. These
conditions allow for complete separation of the starting material from oxazolidinethione
without the need for TLC of the eluting solvent. 50 µL of the eluent was then injected
directly onto the column. Elution with 3:1 hexanes - isopropanol provided separation of
all the reaction components, and the amount of starting material remaining was measured
against the internal standard. The reaction was allowed to proceed for 71 hours, then an
aliquot was taken to determine conversion, which was found to be 52.8%. The reaction
mixture was immediately purified by flash chromatography on silica (5:1 Æ 2:1 hexanes
- ethyl acetate) to provide recovered 10 (40.4 mg, 100% of theory based on 52.8%
conversion). The enantiomeric excess of the flashed recovered starting material was
determined to be 98.9% (Chiralcel OD semi-preperative column, 3:1 hexanes-
isopropanol, flow rate = 1.5 mL / min, retention times = 30.5 min, 40.4 min). An ee of
98.9% at 52.8% conversion corresponds to an s-factor of 83.
A duplicate run provided recovered starting material of 99.5% ee at 53.1%
conversion which corresponds to an s-factor of 89. [α]D = +23.5 (c = 1.50, EtOAc).
Me
S
S
NMe2
OH
MeOH
Toluene
O
O
+
N
N
N
+
N
OMe
N
F3C
F3C
F3C
O
O
O
O
O
O
CF3
2
9
Table 1, Entry 1: The typical procedure was followed using 20.2 mg (0.0623
mmol) of compound 9, 75 µL (1.87 mmol) of MeOH, and 1.54 mg (0.00623 mmol) of
catalyst 2. The reaction was stirred at room temperature for 54 hours (58% conversion),
and flash chromatography provided 8.2 mg (97% of theory) of recovered 9 of 95.8% ee
(Chiralcel OD semi-preperative column, 3:1 hexanes-isopropanol, flow rate = 1.5 mL /

Page S8
min, retention times = 27.6 min, 43.1 min), [α]D = +93.6 (c = 0.81, EtOAc). An ee of
95.8% at 58.0% conversion corresponds to an s-factor of 20.6.
A duplicate run provided recovered 9 in 85.8% ee at 53.0% conversion which
corresponds to an s-factor of 19.7.
S
S
Me
O
O
N
N
OMe
NMe2
OH
MeOH
Toluene
+
N
N
N
O
CF3
O
CF3
O
CF3
O
O
O
CF3
2
11
Table 1, Entry 3: The typical procedure was followed using 53.4 mg (0.1802
mmol) of compound 11, 219 µL (5.407 mmol) of MeOH, and 4.46 mg (0.01802 mmol)
of catalyst 2. The reaction was stirred at room temperature for 99 hours (58.2%
conversion), and flash chromatography provided 22.0 mg (99% of theory) of recovered
11 of 97.8% ee (Chiralcel OD semi-preperative column, 2:1 hexanes-isopropanol, flow
rate = 1.5 mL / min, retention times = 25.9 min, 98.9 min), [α]D = +146.7 (c = 2.2,
EtOAc). An ee of 97.8% at 58.2% conversion corresponds to an s-factor of 25.
A duplicate run provided recovered 11 in 93.9% ee at 57.5% conversion which
corresponds to an s-factor of 19.
Me
S
S
O
O
NMe2
OH
MeOH
Toluene
N
N
N
N
N
+
F3C
F3C
F3C
OMe
+
O
O
O
O
O
O
2
CF3
12
Table 1, Entry 4: The typical procedure was followed using 28.8 mg (0.1020
mmol) of compound 12, 124 µL (3.06 mmol) of MeOH, and 2.52 mg (0.01020 mmol) of
catalyst 2. The reaction was stirred at 0 °C for 30 hours (54% conversion), and HPLC
shows a nearly equal rate of reaction for each enantiomer (Chiralcel OD semi-preperative
column, 2:1 hexanes-isopropanol, flow rate = 1.5 mL / min, retention times = 36.6 min,
152.4 min), An ee of 3% at 54.0% conversion corresponds to an s-factor of 1.1.
Me
S
S
NMe2
OH
MeOH
Toluene
O
O
+
OMe
N
N
+
N
N
N
O
CF3
O
CF3
O
CF3
O
O
O
CF3
13
2
Table 1, Entry 5: The typical procedure was followed using 44.2 mg (0.1233
mmol) of compound 13, 150 µL (3.699 mmol) of MeOH, and 3.05 mg (0.01233 mmol)
of catalyst 2. The reaction was stirred at 0 °C for 72 hours (52.0% conversion), and flash
chromatography provided 20.5 mg (97% of theory) of recovered 13 of 92.1% ee
(Chiralcel OD semi-preperative column, 3:1 hexanes-isopropanol, flow rate = 1.5 mL /
min retention times = 43.5 min, 70.6 min). An ee of 92.1% at 52.0% conversion
corresponds to an s-factor of 40.

Page S9
A duplicate run provided recovered 13 in 92.9% ee at 52.2% conversion which
corresponds to an s-factor of 42. [α]D = +133.7 (c = 1.30, EtOAc)
Me
N
O
CF3
N
O
CF3
N
O
CF3
NMe2
OH
MeOH
Toluene
+
+
S
S
O
O
O
N
N
MeO
CF3
O
O
14
2
Table 1, Entry 6: The typical procedure was followed using 50.4 mg (0.14065
mmol) of compound 14, 171 µL (4.220 mmol) of MeOH, and 3.48 mg (0.014065 mmol)
of catalyst 2. The reaction was stirred at 0 °C for 55 hours (53.9% conversion), and flash
chromatography provided 23.0 mg (99% of theory) of recovered 14 of 96.2% ee
(Chiralcel OD semi-preperative column, 3:1 hexanes-isopropanol, flow rate = 1.5 mL /
min retention times = 35.8 min, 63.8 min), [α]D = +239.4 (c = 2.28, EtOAc). An ee of
96.2% at 53.9% conversion corresponds to an s-factor of 40.
A duplicate run provided recovered 14 in 95.5% ee at 53.8% conversion which
corresponds to an s-factor of 39.
O
O
O
Me
S
N
S
N
F3C
NH
F3C
NH
F3C
NH
O
NMe2
OH
O
MeOH
Toluene
+
OMe
+
O
O
O
CF3
15
2
Table 1, Entry 7: The typical procedure was followed using 46.3 mg (0.1552
mmol) of compound 15, 189 µL (4.656 mmol) of MeOH, and 3.84 mg (0.01552 mmol)
of catalyst 2. The reaction was stirred at 0 °C for 38 hours (54.9% conversion), and flash
chromatography provided 20.8 mg (99% of theory) of recovered 15 of 90.0% ee
(Chiralcel OD semi-preperative column, 4:1 hexanes-isopropanol, flow rate = 1.5 mL /
min, retention times = 37.7 min, 48.5 min), [α]D = - 26.4 (c = 2.08, EtOAc). An ee of
90.0% at 54.6% conversion corresponds to an s-factor of 20.
A duplicate run provided recovered 15 in 87.3% ee at 53.3% conversion which
corresponds to an s-factor of 21.
O
O
O
Me
S
N
S
N
F3C
NH
F3C
NH
F3C
NH
O
O
NMe2
OH
MeOH
Toluene
OMe
+
+
O
O
O
CF3
16
2
Table 1, Entry 8: The typical procedure was followed using 47.5mg (0.1521
mmol) of compound 16, 185 µL (4.56 mmol) of MeOH, and 3.76 mg (0.01521 mmol) of
catalyst 2. The reaction was stirred at -26 °C for 25 hours (54.5% conversion), and flash
chromatography on neutral silica provided 21.1 mg (98% of theory) of recovered 16 of
99.7% ee (Chiralcel OD semi-preperative column, 4:1 hexanes-isopropanol, flow rate =
1.5 mL / min, retention times = 25.8 min, 66.5 min), [α]D = +29.7 (c = 2.11, EtOAc). An
ee of 99.7% at 54.5% conversion corresponds to an s-factor of 69.
A duplicate run provided recovered 16 in 99.1% ee at 53.8% conversion which

Page S10
corresponds to an s-factor of 66.
Table 1, Entry 9: This run was conducted using 5% catalyst loading. The
typical procedure was followed using 89.9mg (0.2879 mmol) of compound 16, 350 µL
(8.64 mmol) of MeOH, and 3.56 mg (0.01439 mmol) of catalyst 2. The reaction was
stirred at -26 °C for 50 hours (56.8% conversion), and flash chromatography on neutral
silica provided 37.1 mg (96% of theory) of recovered 16 of 99.97% ee (Chiralcel OD
semi-preperative column, 4:1 hexanes-isopropanol, flow rate = 1.5 mL / min, retention
times = 25.8 min, 66.5 min), [α]D = +29.7 (c = 2.11, EtOAc). An ee of 99.97% at 56.8%
conversion corresponds to an s-factor of 61.
O
O
O
Me
S
N
S
N
F3C
NH
F3C
NH
F3C
NH
O
NMe2
OH
O
MeOH
Toluene
+
+
OMe
O
O
O
CF3
17
2
Table 1, Entry 10: The typical procedure was followed using 40.1mg (0.1353
mmol) of compound 17, 165 µL (4.06 mmol) of MeOH, and 3.35 mg (0.01353 mmol) of
catalyst 2. The reaction was stirred at -26 °C for 7 hours (53.9% conversion), and flash
chromatography on neutral silica provided 17.6 mg (96% of theory) of recovered 17 of
98.7% ee (Chiralcel OD semi-preperative column, 3:1 hexanes-isopropanol, flow rate =
1.5 mL / min, retention times = 30.1 min, 47.4 min), [α]D = +3.81 (c = 1.65, EtOAc). An
ee of 98.9% at 53.9% conversion corresponds to an s-factor of 61.
A duplicate run provided recovered 17 in 94.8% ee at 51.7% conversion which
corresponds to an s-factor of 60.
O
O
O
S
N
S
N
Me
F3C
S
NH
F3C
S
NH
F3C
S
NH
O
O
NMe2
OH
MeOH
CH2Cl2
OMe
+
+
O
O
O
CF3
2
18
Table 1, Entry 11: The typical procedure was followed (except the reaction was
conducted in methylene chloride at a concentration of 0.1 M due to the insolubility of
compound 18 using the typical conditions) using 73.2 mg (0.2216 mmol) of compound
18, 270 µL (6.650 mmol) of MeOH, and 5.48 mg (0.02216 mmol) of catalyst 2. The
reaction was stirred at -26 °C for 7 hours (56% conversion), and flash chromatography
provided 31.5 mg (98% of theory) of recovered 18 of 98.1% ee (Chiralcel OD semi-
preperative column, 3:1 hexanes-isopropanol, flow rate = 1.5 mL / min, retention times =
37.5 min, 82.3 min), [α]D = +43.2 (c = 0.64, EtOAc). An ee of 98.1% at 56.0%
conversion corresponds to an s-factor of 35.
A duplicate run provided recovered 18 in 99.4% ee at 57.0% conversion which
corresponds to an s-factor of 38.8.

Page S11
Peptide Coupling Reactions
S
Cl
O
O
H
N
O
N
H3N
Hunig's base
+
N
N
OEt
OEt
Ph
HOBt
92%
O
CF3
O
CF3
O
O
Ph
10
19
20
To a stirred solution of 36.0 mg thioimide 10 (0.1160 mmol) and 28.0 mg L-
phenylalanine ethyl ester hydrochloride (0.1160 mmol) in 580 µL THF (0.2 M) at room
temperature was added 101 µL Hunig’s base (0.5800 mmol). The reaction mixture
became homogenous, and 31.4 mg HOBt (0.2320 mmol) was added as a solid. After
stirring at room temperature for 36 hours, the reaction was concentrated, and purification
by flash chromatography (7:1 hexanes - ethyl acetate) provided the dipeptide 20 (42.7mg,
92%) as a white solid. The ee of the starting material was >99% and the d.r. of the
dipeptide was >100:1. This was determined by an independent synthesis of the peptide
N-TFA-DL-Pip-L-Phe-OEt and separation of the diastereomers using a Chiralcel OD
column.
1H NMR revealed a 4:1 mixture of conformers (400 MHz, CDCl3): δ 7.23-7.30
(m, 3H), 7.05-7.08 (m, 2H), 6.13 (d, 1H, J=8 Hz), 5.07 (br d, 0.8H, J=5 Hz), 4.81 (q, 1H,
J=8 Hz), 4.51 (br d, 0.2H, J=5 Hz), 4.39 (br d, 0.2H, J=14 Hz), 4.16 (q, 2H, J=7 Hz), 3.83
(br d, 0.8H, J=14 Hz), 3.20-3.28 (m, 0.8H), 3.09 (d, 1.6H, J=6 Hz), 3.04-3.17 (m, 0.2H),
2.35-2.38 (m, 0.2H), 2.21-2.27 (m, 0.8H), 1.39-1.70 (m, 5H), 1.26 (t, 0.6H, J=7 Hz), 1.23
(t, 2.4H, J=7 Hz). 13C NMR for the major rotamer (100 MHz, CDCl3): δ 171.29, 168.65,
157.46 (q, J=36 Hz), 135.52, 129.32, 128.94, 127.52, 116.55 (q, J=286 Hz), 61.85, 53.95,
53.04, 44.20 (q, J=3 Hz), 37.80, 25.78, 25.4, 20.26, 14.28. mp = 100-101. IR (cm-1)
3426, 1737, 1678, 1194. [α]D = +21.41 (c = 1.49, EtOAc). HRMS m/z calcd for
C19H23F3N2O4H+: 401.1682; found: 401.1672
O
O
Cl
O
S
N
F3C
F3C
NH
Hunig's base
94%
H3N
O
NH
O
H
N
OEt
+
OEt
Ph
O
16
19
O
21
Ph
To a stirred solution of 53.0 mg thioimide 16 (0.1697 mmol) and 39.0 mg L-
phenylalanine ethyl ester hydrochloride (0.1697 mmol) in 1.7 ml THF (0.1 M) at room
temperature was added 65.1 µL Hunig’s base (0.3733 mmol). The reaction mixture
became homogenous, and was refluxed for 36 hours. The reaction was then
concentrated, and purification by flash chromatography (7:1 hexanes - ethyl acetate)
provided the dipeptide 21 (64.2mg, 94%) as a white solid. The ee of the starting material
was 99% and the d.r. of the dipeptide was 98:2. This was determined by an independent
synthesis of the peptide N-TFA-DL-Leu-L-Phe-OEt and separation of the diastereomers
using a Chiralcel OD column.
1H NMR (400 MHz, CDCl3): δ 7.21-7.29 (m, 3H), 7.02-7.11 (m, 3H), 6.30-6.48
(m, 1H), 4.83(ddd, 2H, J=4, 6, 6 Hz), 4.41-4.48 (m, 1H), 4.17 (q, 2H, J=6 Hz), 3.17 (dd,
1H, 4 Hz), 3.00-3.04 (m, 1H), 1.47 (d, 2H, J=6 Hz), 1.41 (1H, hd, J=5 Hz) 1.24 (t, 3H,
J=8 Hz), 0.85 (app t, 6H, J=5 Hz). 13C NMR (100 MHz, CDCl3): δ 171.42, 170.35,

Page S12
157.07 (q, J=37 Hz), 135.79, 129.43, 128.83, 127.49, 115.89 (q, J=286 Hz), 62.07, 53.47,
52.17, 41.86, 38.07, 24.81, 22.87, 22.29, 14.25. mp = 89-91. IR (cm-1) 3270, 1703, 1655
1187. [α]D = +17.10 (c = 1.18, EtOAc). HRMS m/z calcd for C19H25F3N2O4H+:
403.1839; found: 403.1840
Determination of Absolute Configuration
The absolute configurations of the recovered cyclic substrates 11 and 13 were
determined by preparing authentic samples of the α-trifluoroacetamido oxazolidinethione
imides from the corresponding L-amino acids (proline and 1,2,3,4-
tetrahydroisoquinoline-3-carboxylic acid). The retention times of these materials were
compared with that of our recovered starting materials, which showed that the faster
reacting enantiomer is the L-amino acid-derived substrates. Thus, the recovered starting
materials correspond to the D- amino acid-derived substrates.
The absolute
configurations of compounds 9, 10 and 14 were assigned by analogy. Furthermore, we
observed that in all cases of the cyclic amino acid derivatives, using our HPLC method,
the slower reacting enantiomer is the slower eluting enantiomer.
In the case of the valine-derived substrate 15, the absolute configuration of the
methanolysis product (N-TFA-Val-OMe) was determined by comparing the optical
rotation of this material with that reported in the literature (Lit: N-TFA-L-Val-OMe
[α]24D = +33.4 (c = 0.72, CH2Cl2);3 N-TFA-Val-OMe from our kinetic resolution: [α]24
D
= +22.7 (c = 1.1, CH2Cl2, 74.5% ee)). Thus, the starting material recovered from
substrate 15 is of the D-configuration. To determine the absolute configuration of the
recovered leucine-derived substrate 16, N-TFA-L-Leu-L-Phe-OEt was prepared as was
N-TFA-DL-Leu-L-Phe-OEt. An HPLC method was then developed to separate the
diastereomers of the N-TFA-DL-Leu-L-Phe-OEt. Subjection of compound 21 to this
HPLC method allowed us to assign this compound as N-TFA-D-Leu-L-Phe-OEt. Thus,
the recovered starting material from compound 16 is also of the D configuration. The
absolute configuration of substrates 17 and 18 were assigned by analogy. In the case of
the acyclic amino acid substrates, using our HPLC method, the slow reacting enantiomer
is always the faster eluting enantiomer.
3
Denmark, S. E.; Stavenger, R. A.; Faucher, A.; Edwards, J. P. J. Org. Chem. 1997, 62, 3375.

Page S13
Protecting Group Swap
O
NHBoc
OMe
F3C
NH
1. Boc2O, DMAP, THF
OMe
2. hydrazine, MeOH
92%
O
O
Using a slightly modified version of an amide to carbamate transformation
reported by Burk,4 the trifluoracetamide can be converted to the t-butyl carbamate by the
following procedure. To a stirred solution of the trifluroacetamide (44.9 mg, 0.1861
mmol) and DMAP (4.5 mg, 0.03723 mmol) in THF (930 µL, 0.2 M) was added Boc2O
(85.5 µL, 0.3723 mmol), and the solution was heated to reflux for 4 hours. After cooling
to room temperature, MeOH (930 µL) and hydrazine (23.4 µL, 0.7440 mmol) were added
and the solution was stirred for an additional hour. The reaction mixture was diluted with
methylene chloride and the organic layer was washed with 1M HCl, 1M CuSO4, and sat.
NaHCO3. The organic layer was dried over MgSO4, filtered, and concentrated at reduced
pressure to a yellow oil. Purification by flash chromatography (10:1 hexanes:ethyl
acetate) afforded the product as a clear oil (41.9 mg, 92%). There was no detectable
1
13
racemization observed by HPLC, and H and C NMR spectral data were in agreement
with those reported in the literature.4
4
Burk, M. J.; Allen, J. G.; J. Org. Chem. 1997, 62, 7054.

Page S14
Reaction Profile: %Conversion vs. Time
Kinetic resolutions which display s-factors greater than 10 will slow drastically as
they approach 50% conversion. To illustrate this point, we have monitored the kinetic
resolution of compound 16 by NMR with an enhanced catalyst loading of 15%, as shown
in the figure below. Because of the large difference in reactivities of the two
enantiomeric substrates, there is a large window of time during which the reaction can be
stopped without compromising the yield of recovered starting material.
O
O
O
Me
S
N
S
N
F3C
NH
F3C
+
NH
F3C
NH
O
O
NMe2
OH
CD3OD
OCD3
+
Toluene - d8
O
O
O
CF3
16
2
Reaction Conversion vs Time Profile
1
0.9
0.8
0.7
0.6
0.5
0.4
0
20
40
60
80
100
120
140
time (min)