Enantioselective fluoride ring opening of aziridines enabled by cooperative Lewis acid catalysis

Article abstract of DOI:10.1016/j.tet.2013.01.062

The enantioselective ring opening of aziridines using a latent source of HF is described. A combination of two Lewis acids, (salen)Co and an achiral Ti(IV) cocatalyst, provided optimal reactivity and enantioselectivity for the trans β-fluoroamine product.

Full text of DOI:10.1016/j.tet.2013.01.062

Tetrahedron xxx (2013) 1e8
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Enantioselective fluoride ring opening of aziridines enabled by
cooperative Lewis acid catalysis
*
Julia A. Kalow, Abigail G. Doyle
Department of Chemistry, Princeton University, Princeton, NJ 08544, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The enantioselective ring opening of aziridines using a latent source of HF is described. A combination of
two Lewis acids, (salen)Co and an achiral Ti(IV) cocatalyst, provided optimal reactivity and enantiose-
Received 10 December 2012
Received in revised form 17 January 2013
Accepted 22 January 2013
Available online xxx
lectivity for the trans
b-fluoroamine product. The use of a chelating aziridine protecting group was
crucial. Acyclic and cyclic meso N-picolinamide aziridines underwent fluoride ring opening in up to 84%
ee, and the kinetic resolution of a piperidine-derived aziridine was performed with krel¼6.6. The pico-
linamide group may be readily removed without epimerization of the fluoroamine. Preliminary studies
revealed a bimetallic mechanism wherein the chiral (salen)Co catalyst delivers the nucleophile and the
Ti(IV) cocatalyst activates the aziridine.
This manuscript is dedicated to Professor
Melanie S. Sanford on the occasion of her
receiving the Tetrahedron Young In-
vestigator Award
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Asymmetric catalysis
Fluorination
Aziridine opening
Cooperative catalysis
b
-Fluoroamine
1. Introduction
amine catalysts (<15% ee); instead, a chiral auxiliary was employed
for the asymmetric synthesis of b-fluoroamines. The lack of asym-
The asymmetric ring opening of aziridines is a powerful strategy
metric induction is attributed to the variable, non-rigid structure of
the active fluorinating reagent, an amine poly(hydrogen fluoride)
species.
to access valuable enantioenriched b
-substituted amines.1 However,
the use of halides as nucleophiles in this transformation has only
recently been achieved. Jacobsen and Mita reported the desym-
metrization of meso aziridines with HCl catalyzed by a bifunctional
thioureaephosphine,2 and Ooi and co-workers accomplished both
the desymmetrization and kinetic resolution of aziridines by TMSCl
and TMSBr in combination with a chiral triazolium catalyst.3
We previously reported the enantioselective ring opening of
epoxides by fluoride catalyzed by the chiral Lewis acid (salen)Co.7
Use of the same latent HF source, PhCOF/HFIP, enabled these re-
actions to occur under mild conditions and with high enantioin-
duction for the desymmetrization of meso epoxides and kinetic
resolution of terminal epoxides. On the basis of detailed mecha-
nistic investigations, we proposed that the active nucleophilic
fluorine source is a cobalt bifluoride, which delivers fluoride to an
epoxide substrate that is activated by another (salen)Co species.8
The interaction between two chiral (salen)Co species in the tran-
sition state has been proposed to be key to the high enantioin-
duction observed for epoxide-opening reactions.9
In developing an asymmetric catalytic fluoride ring opening
of aziridines, we envisioned taking advantage of the chiral metal
fluoride generated from (salen)Co, PhCOF, and HFIP. Although
(salen)Co(III) can act as a Lewis acid to activate epoxides, previous
work has shown that (salen)Co(III) Lewis acids are not effective for
the activation of protected aziridines, likely due to their increased
From a synthetic standpoint, enantioenriched
represent a desirable target.4 By lowering the pKa of neighboring
amines, -fluoro substitution affects a variety of properties relevant
b-fluoroamines
b
to medicinal chemistry, including metabolic stability, binding af-
finity, and bioavailability.5 Work in our laboratory has revealed that
the combination of benzoyl fluoride and 1,1,1,3,3,3-hexafluoroiso-
propanol (HFIP) serves as a latent source of HF for ring-opening
reactions. The addition of a Lewis base catalyst to PhCOF/HFIP
promotes the in situ generation of an amineeHF reagent that is
effective in the ring opening of aziridines.6 However, enantiose-
lective fluoride ring opening could not be achieved with chiral
* Corresponding author. E-mail address: agdoyle@princeton.edu (A.G. Doyle).
0040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.01.062
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062

2
J.A. Kalow, A.G. Doyle / Tetrahedron xxx (2013) 1e8
steric hindrance.10 The Jacobsen group has approached this chal-
lenge by designing tridentate Schiff base ligands; the resulting
Cr(III) complexes effect enantioselective opening of N-dinitro-
benzyl aziridines by TMSN3. Unfortunately, the relatively basic
N-alkyl aziridines compatible with this catalyst undergo facile au-
tocatalytic background reaction in the presence of PhCOF/HFIP.
We therefore considered whether a distinct Lewis acid could be
used to activate the aziridine toward fluoride ring opening by
(salen)CoF(HF). Shibasaki,11 Kobayashi,12 and Parquette and Rajan-
Babu13 have taken advantage of multimetallic chiral catalyst sys-
tems to achieve the highly asymmetric ring opening of aziridines by
nucleophiles including azide, cyanide, anilines, and malonates.
These systems rely on multidentate ligands that can coordinate to
multiple metal centers; for aziridine opening, homopolymetallic
systems are most often employed.14 We anticipated that catalyst
design could be simplified if a separate, readily available Lewis acid
could be used in combination with our existing (salen)Co/PhCOF/
HFIP system. In order to favor coordination of this Lewis acid to the
substrate, we decided to install a chelating protecting group on the
aziridine (Scheme 1). The proposed system represents an example
of synergistic catalysis, in which both the nucleophile and the
electrophile are activated by distinct catalysts.15
our designplan (Scheme 1), we tested aziridines functionalized with
chelating protecting groups. Intriguingly, N-acyl aziridines bearing
2-azaarenes proved uniquely effective (2-pyridyl, 2-thiazolyl, 1-
isoquinolinyl). All other aziridine protecting groups provided no
observable b-fluoroamine or racemic product in the presence of
a range of Lewis acid cocatalysts (Fig.1). The picolinamide-protected
aziridine 1a was selected for further optimization. This auxiliary was
introduced as a directing group for Pd-catalyzed C(sp3)eH activa-
tion16 and is proposed to act as a chelating protecting group in the
cinchona alkaloid/Zn(II)-catalyzed desymmetrization of aziridines
with phosphites.17
As the sole catalyst in the presence of PhCOFand HFIP in tert-butyl
methyl ether (TBME), (R,R)-(salen)Co provided 79% yield of the de-
siredb-fluoroamine2a; however, thisproduct wasgenerated in<10%
ee (Table 1, entry 1). The addition of Lewis acid cocatalysts typically
employed in aziridine-opening protocols (Y, Yb, Sc, Zn, Cu) to the
(salen)Co-catalyzed protocol did not improve the enantioselectivity
(entries 2e6). Interestingly, several oxophilic Lewis acids, including
Al(Oi-Pr)3, FeCl3, Zr(Oi-Pr)4, and Ti(Oi-Pr)4, provided significant en-
hancements in enantiomeric excess, up to 62% ee (entries 7e10).
Table 1
Optimization of Lewis acid cocatalysta
Entry
Lewis acid
Yieldb (%)
eec (%)
1
2
3
4
5
6
7
8
None
79
71
60
0
59
25
40
58
61
66
79
79
7 (S,S)
8 (S,S)
3 (S,S)
n/a
7 (S,S)
0
13 (R,R)
31 (R,R)
41 (R,R)
62 (R,R)
83 (R,R)
56 (R,R)
Y(Oi-Pr)3
Yb(Oi-Pr)3
Sc(OTf)3
Zn(OAc)2
Cu(TFA)2
Al(Oi-Pr)3
FeCl3
Scheme 1. Proposed catalytic cycle for aziridine ring opening.
2. Results and discussion
2.1. Optimization
9
Zr(Oi-Pr)4
Ti(Oi-Pr)4
Ti(NMe2)4
10
11
12
d
TiF4
a
Reaction conditions: 0.1 mmol 1a, 0.005 mmol (R,R)-(salen)Co, 0.005 mmol
TBHP (5 M solution in decane), 0.01 mmol Lewis acid, 0.2 mmol PhCOF, 0.4 mmol
HFIP, 0.5 mL TBME, 23 ꢀC, 18 h. All reactions proceeded to >95% conversion.
We chose to initiate reaction development using the chiral Lewis
acid catalyst and latent HF source that provided optimal reactivity
and enantioselectivity in the epoxide ring-opening reaction.7 Since
we reasoned that (salen)Co(III) would be able to activate fluoride
toward asymmetric nucleophilic delivery, but would not serve as
a competent Lewis acid for the N-substituted aziridine, we evaluated
a wide variety of achiral Lewis acid cocatalysts. In accordance with
b
Determined by HPLC in the presence of diphenyl ether as a quantitative internal
standard.
c
Determined by chiral HPLC analysis.
The cocatalyst was added as a 0.5 M solution in MeCN.
d
It is particularly remarkable that the addition of an achiral Lewis
acid cocatalyst enables asymmetric induction by (R,R)-(salen)Co
despite a highly efficient unselective background reaction by the chiral
catalyst. Accordingly, we found that optimal enantioselectivity was
achieved when a 2:1 molar ratio of Lewis acid relative to (salen)Co
was employed. Ti(IV) species proved to be the most selective co-
catalysts, and 2a could be obtained in 83% ee in the presence of
Ti(NMe2)4 (entry 11).
In previous studies, we observed that aerobic oxidation of the
commercially available (salen)Co(II) complex to the catalytically
active (salen)Co(III) species occurs in the presence of HFIP; however,
this slow in situ oxidation gives rise to an induction period.8 For
aziridine opening by PhCOF/HFIP, we found that pre-oxidized cat-
alysts provided superior performance. The (salen)Co(II) precatalyst
can be rapidly oxidized in situ prior to the reaction by anhydrous
tert-butyl hydroperoxide (TBHP). For substrates requiring extended
reaction times, (salen)Co(III)OCH(CF3)2$HFIP, readily prepared from
(salen)Co(II) and HFIP,8 provides optimal results.
Fig. 1. Aziridines tested in the fluoride ring opening reaction.
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062

J.A. Kalow, A.G. Doyle / Tetrahedron xxx (2013) 1e8
3
Further optimization was attempted by modifying the steric and
electronic nature of the picolinamide protecting group. Sub-
stitution at the 6-position of the picolinamide was not tolerated,
presumably because this hinders N,O-chelation by the Lewis acid.
Electron-withdrawing and -donating substituents elsewhere on
the pyridine ring had little effect on the enantioselectivity. Addi-
tionally, Ti(IV) cocatalysts bearing chiral ligands were evaluated;
while these chiral catalysts did not provide superior results to
Ti(NMe2)4, these data offered mechanistic insight into the syner-
gistic interaction between the Lewis acid cocatalysts (vide infra).
Dihydronaphthalene-derived aziridine 1b also performed well
in the reaction, delivering 2b in 66% yield and 80% ee (97% ee after
recrystallization, entry 2). Acyclic aliphatic aziridines such as 1c are
also competent substrates, providing the b-fluoroamine as a single
diastereomer in 63% ee (entry 3), but cis-diphenyl aziridine 1d
suffered from a number of side reactions leading to depressed yield
and enantioselectivity (entry 4). The major side products for the
reaction of 1d arose from cationic rearrangements leading to trans
aziridine18 as well as cis and trans oxazolines.
Other ring sizes performed modestly in the asymmetric fluori-
nation reaction. Five-membered cyclic aziridines delivered fair
yields, and fluoroamines 2e and 2f with cyclopentane and pyrro-
lidine backbones were obtained with 56 and 75% ee, respectively
(entries 5 and 6). A dihydrofuran-derived aziridine (1g), however,
underwent sluggish ring opening with low enantioselectivity (en-
try 7). Cycloheptane substrate 1h also performed poorly, providing
only 12% ee (entry 8). Significant amounts of oxazoline were also
observed in these reactions and the enantioselectivities of the re-
actions decreased over time, indicative of catalyst inhibition or
degradation. The oxazoline side product arises from the Lewis acid-
catalyzed GabrieleHeine rearrangement, in which acylaziridines
are isomerized to the corresponding oxazolines.19 Interestingly, the
oxazoline byproducts are generated with measurable ee, suggest-
ing that (salen)Co is involved in the isomerization.20
2.2. Substrate scope
With this protocol in hand, we sought to explore the substrate
scope of the dual Lewis acid-catalyzed aziridine ring opening. A
variety of meso aziridines 1aeh were synthesized by amide cou-
pling of the corresponding NeH aziridine and picolinic acid. On
preparative scale, 2a was isolated in 93% yield and 84% ee (Table 2,
entry 1). Promisingly, 2a may be recrystallized from i-PrOH by slow
evaporation to provide the trans
b
-fluoroamine in 95% ee.
Table 2
Substrate scope
b-Fluoroamines derived from unsymmetrical aziridines repre-
sent valuable building blocks for medicinal chemistry. For example,
the trans 3-amino-4-fluoropiperidine core has appeared in nu-
merous patents.21 Previous work has shown that nucleophilic at-
tack on 3,7-diazabicyclo[4.1.0]heptane occurs preferentially at the
position distal to the inductively electron-withdrawing hetero-
atom;22 however, asymmetric reactions of this substrate to arrive at
valuable enantioenriched products have not been reported. Grati-
fyingly, the kinetic resolution of aziridine 3 occurred with a selec-
tivity factor (s)23 of 6.6, providing 3-amino-4-fluoropiperidine 4 as
a single regioisomer in 29% yield and 68% ee after 4 h, and 41% yield
and 62% ee after 7 h (Scheme 2).
Entry
1
Conditionsa
A
Product
Yieldb (%)
eec (%)
84 (95)
2a
2b
2c
2d
2e
93 (51)
66 (56)
75
2
3
4
5
A
B
B
A
80 (97)
63
51
48
Scheme 2. Kinetic resolution of piperidine 3 by regioselective fluoride ring opening.
46
56
Importantly, the picolinamide group can be readily removed
under mild conditions without epimerization of the b-fluoroamine.
Product 2a (95% ee) was subjected to Boc protection and the amide
was subsequently cleaved by NaBH4 to furnish (R,R)-5 in 91% yield
6d
B
B
2f
48
27
75
32
and 95% ee (Scheme 3). The absolute configuration of the b-fluo-
roamine products was determined by comparison to the literature
data (5) and X-ray crystallography (2b24): notably, (R,R)-(salen)Co
provides the opposite absolute configuration for the aziridine ring
opening compared to epoxide ring opening, which generates (S,S)-
fluorohydrins.7
7d,e
2g
8e
B
2h
33
12
a
A: 5 mol % (salen)Co(II), 5 mol % TBHP (5 M in decane), 24 h; B: 5 mol % (salen)
Co(III)OCH(CF3)2$HOCH(CF3)2, 48 h. Data are the average of two experiments. Yields
based on recovered 2 and ees in parentheses are for products recrystallized from
i-PrOH.
b
Isolated yield on 0.2e0.5 mmol scale.
Enantiomeric excesses determined by chiral HPCL.
Reaction time: 96 h.
One run.
c
d
Scheme 3. Deprotection of 2a.
e
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062

4
J.A. Kalow, A.G. Doyle / Tetrahedron xxx (2013) 1e8
2.3. Mechanistic studies
which the Co center both activates the epoxide and delivers the
nucleophile.9a However, 8 provided no rate acceleration and de-
pressed ee relative to monomeric (salen)Co for the aziridine ring-
opening reactions. Therefore, the linear relationship between %ee
of 2a and %ee of (salen)Co is best ascribed to a mechanism in which
this catalyst plays only one role in the ring opening. This scenario
also provides a rationale for the observation that the absolute ste-
reochemistry of the aziridine ring opening is the opposite of that for
the epoxide opening.
2.3.1. Control experiments: protecting group. To confirm that the
unique reactivity observed for the picolinamide-protected azir-
idines 1 arises from chelation to the Lewis acid, rather than simply
an electronic effect, we synthesized benzamide- and isonicotina-
mide-protected aziridines 6 and 7 (Scheme 4). Under the cocata-
lytic reaction conditions, both aziridines displayed poor reactivity,
and benzamide-protected 6 delivered only 11% yield of the fluo-
roamine in racemic form. No fluorinated product was detected in
the reaction of isonicotinamide-protected aziridine 7, which un-
derwent only 4% conversion.
2.3.4. Role of the Ti(IV) cocatalyst. In the aziridine ring opening
reactions, we consistently observed a small amount of N,N-dime-
thylbenzamide as a byproduct, which is produced by the reaction of
benzoyl fluoride with the Ti(NMe2)4 cocatalyst. Indeed, when
Ti(NMe2)4 and 4 equiv PhCOF were combined in CDCl3 in an NMR
tube, the reaction mixture turned orange and became warm.
A yellow precipitate was formed upon standing for several
minutes. Addition of MeCN slightly solubilized this precipitate and
the resulting suspension displayed a 19F NMR resonance consis-
tent with a terminal TieF (þ135.0 ppm relative to CFCl3). Dia-
lkylaminotitanium fluorides, TiFn(NR2)4ꢁn, have been previously
prepared by alternative methods; these species exhibit insolubility
and complex concentration- and temperature-dependent equilibria
between oligomeric species via bridging fluorides.26 The species
obtained from Ti(NMe2)4 and PhCOF is a competent cocatalyst,
providing 67% yield and 67% ee for the ring opening of 1a under the
standard conditions; TiF4 provides 56% ee, despite its poor solubility
in organic solvents (Table 1, entry 12). This observation raises the
possibility that the nucleophilic fluorine source in the ring-opening
reaction is in fact a titanium fluoride complex. However, in the ab-
sence of PhCOF, 2a is not observed in the reaction of 1a with pre-
generated dimethylaminotitanium fluoride (2 equiv), (salen)Co, and
HFIP. Therefore, ligand exchange of Ti(NMe2)4 with PhCOF may in-
crease the Lewis acidity of this catalyst, but the resulting titanium
fluoride is not the source of fluoride for aziridine opening.27
Scheme 4. Control experiments with aziridines 6 and 7. Conversions, yields, and ees
determined by chiral HPLC.
2.3.2. Nonlinear effects with scalemic (salen)Co. The fluorination of
1a was performed with 0e100% ee (R,R)-(salen)Co, and a linear re-
lationship between %ee of 2a and %ee of the catalyst was observed
(Chart 1). The lack of nonlinear effects on enantioselectivity is con-
sistent with either a mechanism in which the (salen)Co species do
not interact, or one in which the (salen)Co species interact but the
homo- and heterochiral pathways occur with equal rates.25
2.3.5. Matchedemismatched effects with chiral Ti(IV) cocatalysts. In
light of our data, we propose a mechanism in which fluoride is
delivered by (salen)Co, as in the epoxide-opening reaction, and the
Ti(IV) cocatalyst activates the aziridine by chelating to the picoli-
namide protecting group (Fig. 3). Based on our previous studies, we
propose that the nucleophilic fluorine species is a cobalt(III)
bifluoride.
Chart 1. %ee of 2a versus %ee of (salen)Co for the ring opening of 1a (red squares) and
linear fit.
2.3.3. Reactions with a linked (salen)Co catalyst. Linked (salen)Co
catalyst 8 (Fig. 2) has been shown to accelerate the fluoride ring
opening of epoxides and provide improved ees for the fluorohydrin
products; together with nonlinear effects, substituent effects, and
kinetic data, this effect is consistent with a ring-opening reaction in
Fig. 3. Proposed mechanism for ring opening.
The proposed synergistic interaction between the cocatalysts is
further supported by experiments with chiral Ti(IV) cocatalysts.28
When TiF2X2 catalysts generated from chiral diols are used in the
reaction of 1a, modest matchedemismatched effects on enantio-
selectivity are observed for the diastereomeric combinations of the
catalysts (Scheme 5). These matchedemismatched effects are
consistent with a cooperative interaction between the two chiral
catalysts.
Fig. 2. Linked (salen)Co catalyst (R,R,R,R)-8.
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062

J.A. Kalow, A.G. Doyle / Tetrahedron xxx (2013) 1e8
5
(CDCl3¼
d
77.16 ppm, MeOH¼49.00 ppm). 19F spectra were recorded
on a Bruker NB 300 spectrometer (282 MHz); chemical shifts
are reported in parts per million (ppm) and are referenced to CFCl3
(
d
0 ppm). NMR data are represented as follows: chemical shift
ppm), multiplicity, coupling constant in hertz (Hz), integra-
(d
tion. Reversed-phase liquid chromatography/mass spectrometry
(LC/MS) was performed on an Agilent 1260 Infinity analytical LC
and Agilent 6120 Quadrupole LCMS system, using electrospray
ionization/atmospheric-pressure chemical ionization (ESI/APCI),
and UV detection at 254 and 280 nm. High-resolution mass
spectra were obtained on an Agilent 6220 using electrospray ioni-
zation time-of-flight (ESI-TOF). FTIR spectra were recorded on
a PerkineElmer Spectrum 100 and are reported in terms of fre-
quency of absorption (cmꢁ1). High-pressure liquid chromatography
(HPLC) was performed on an Agilent 1200 series instrument using
Chiracel OD-H (25 cmꢂ0.46 cm), Chiralpak AS-H (25 cmꢂ0.46 cm),
and Chiralpak AD-H (25 cmꢂ0.46 cm) columns equipped with the
corresponding guard columns (1 cmꢂ0.4 cm). Gas chromatography
(GC) was performed on an Agilent 7890A series instrument equip-
ped with a J&W Scientific Cyclodex-B column (30 mꢂ0.25 mm).
Optical rotations were recorded on a Jasco P-1010 polarimeter using
a 1-mL cell with a 0.5 dm path length; concentration (c) is in
Scheme 5. Matchedemismatched effects for the reaction of 1a with (salen)Co and
chiral Ti(IV) catalysts.
3. Conclusion
We have described the first asymmetric catalytic fluoride ring
opening of aziridines. This protocol generates valuable trans
b-
fluoroamine products through the desymmetrization of meso
aziridines in up to 84% ee. Additionally, the catalytic fluorination
may be applied to the kinetic resolution of a racemic substrate. Keys
to the success of this reaction are the use of PhCOF and HFIP as
a latent source of HF and the application of the picolinamide pro-
tecting group to enable Lewis acid activation of the aziridine. In-
triguingly, an achiral Lewis acid is required in addition to (salen)Co
to achieve enantioselectivity for the ring opening. Preliminary
mechanistic studies reveal a synergistic interaction between the
catalysts, in which (salen)Co serves to deliver fluoride and Ti(IV)
activates the aziridine. The discovery of new applications of benzoyl
fluoride as a latent fluoride source for asymmetric catalytic trans-
formations is ongoing in our laboratory.
g/100 mL and [a]D values are in degrees.
4.2. Representative procedure for synthesis of substrates
4.2.1. 7-Azabicyclo[4.1.0]heptan-7-yl(pyridin-2-yl)methanone (1a). To
a solution of 7-azabicyclo[4.1.0]heptane29 (0.55 mL, 94 wt %,
5.0 mmol) in anhydrous CH2Cl2 (25 mL) at 23 ꢀC were added picolinic
acid (0.62 g, 5.0 mmol), 4-dimethylaminopyridine (0.67 g, 5.5 mmol),
and N,N-diisopropylcarbodiimide (0.86 mL, 5.5 mmol). The reaction
mixture was stirred for 16 h, and the suspension was diluted with
hexanes, filtered, and concentrated. The crude reaction mixture was
purified by column chromatography (5e40% EtOAc/hexanes with
0.5% Et3N) to provide the title compound as a white solid (0.83 g, 82%
yield); mp 65.0e66.0 ꢀC (lit.17a 65.0e67.0 ꢀC). The product was stored
atꢁ10 ꢀC. Spectraldatawere inagreement withthe literature values.6
4. Experimental
4.1. General information
Unless otherwise noted, reactions were carried out in oven-dried
glassware under a N2 atmosphere. Commercial reagents were used
as received with the following exceptions. Dichloromethane
(CH2Cl2), tetrahydrofuran (THF), and toluene were dried by passing
through activated alumina columns; acetonitrile (CH3CN) was dried
by passing through a column of activated molecular sieves. tert-Butyl
4.2.2. Pyridin-2-yl(1a,2,7,7a-tetrahydro-1H-naphtho[2,3-b]azirin-1-
yl)methanone (1b). Prepared from 1a,2,7,7a-tetrahydro-1H-naph-
tho[2,3-b]azirine.30 FTIR (thin film, cmꢁ1) 3050, 1668, 1434, 1316;
1H NMR (500 MHz, CDCl3)
d
8.69 (d, J¼3.9 Hz, 1H), 7.92 (d, J¼7.8 Hz,
1H), 7.75 (td, J¼7.7, 1.6 Hz, 1H), 7.39 (dd, J¼7.4, 4.8 Hz, 1H), 7.16e7.10
ꢀ
methyl ether (TBME) was distilled from CaH2 and stored over 3 A
(m, 2H), 7.10e7.01 (m, 2H), 3.53 (d, J¼16.6 Hz, 2H), 3.29 (s, 2H), 3.21
molecular sieves (8e12 mesh), which had been flame-dried under
reduced pressure. 1,1,1,3,3,3-Hexafluoroisopropanol (HFIP) was dis-
(d, J¼16.8 Hz, 2H); 13C NMR (125 MHz, CDCl3)
d 176.85, 150.99,
149.30, 136.74, 132.51 (2C), 129.46 (2C), 126.80 (2C), 126.41, 123.87,
37.49 (2C), 29.49 (2C); HRMS (ESIþ) calculated for C16H15N2O
([MþH]þ): 251.1179; found: 251.1178.
ꢀ
tilled from powdered activated 3 A molecular sieves and stored over
ꢀ
activated 3 A molecular sieves (8e12 mesh). Commercial benzoyl
fluoride (PhCOF) was filtered through SiO2, eluting with pentanes,
and solvent was removed in vacuo; it was stored in a plastic vial in
a desiccator containing CaSO4. Ti(NMe2)4 (>99%) was purchased in
ampules, transferred to oven-dried vials fitted with a Mininert valve
cap, and stored in a dessicator. Reactions were monitored by LC/MS
or thin-layer chromatography (TLC) on EMD Silica Gel 60 F254 plates,
visualizing with fluorescence quenching, KMnO4, ninhydrin, or ceric
4.2.3. (cis-2,3-Dibutylaziridin-1-yl)(pyridin-2-yl)methanone
(1c). Prepared from cis-2,3-dibutylaziridine.31 FTIR (thin film, cmꢁ1
)
2957, 1673, 1585, 1439, 1332, 1147, 996; 1H NMR (500 MHz, CDCl3)
d
8.70 (d, J¼3.7 Hz, 1H), 8.09 (d, J¼7.8 Hz, 1H), 7.82 (t, J¼7.0 Hz, 1H),
7.44 (dd, J¼6.4, 4.8 Hz, 1H), 2.63 (d, J¼4.3 Hz, 2H), 2.00e1.77 (m, 2H),
1.80e1.28 (m, 10H), 0.94 (t, J¼7.2 Hz, 6H); 13C NMR (125 MHz, CDCl3)
ammonium molybdate (CAM). Automated column chromatography
d 178.41, 151.08,149.28, 136.85,126.50,124.05, 42.90 (2C), 29.65 (2C),
was performed using SiliCycle SiliaFlash F60 (40e53
m
m, 60 A) in
27.59 (2C), 22.72 (2C), 14.22 (2C); HRMS (ESIþ) calculated for
ꢀ
SNAP cartridges on a Biotage Isolera 4.
C
16H25N2O ([MþH]þ): 261.1961; found: 261.1962.
Proton and carbon nuclear magnetic resonance (1H and 13C NMR)
spectra were recorded on a Bruker 500 AVANCE spectrometer (500
and 125 MHz). Chemical shifts for protons are reported in parts per
million (ppm) downfield from tetramethylsilane and are referenced
4.2.4. (cis-2,3-Diphenylaziridin-1-yl)(pyridin-2-yl)methanone
(1d). Prepared from cis-2,3-diphenylaziridine.32 FTIR (thin film,
cmꢁ1) 3050, 1679, 1584, 1438, 1413, 1328, 1243, 1092, 911; 1H NMR
to residual protium in the NMR solvent (CHCl3¼
d
7.26 ppm,
(500 MHz, CDCl3)
d
8.63 (d, J¼4.6 Hz,1H), 8.16 (d, J¼7.8 Hz,1H), 7.81
MeOH¼3.31 ppm). Chemical shifts for carbon are reported in parts
per million (ppm) downfield from tetramethylsilane and are refer-
enced to the central carbon resonance of the solvent residual peak
(td, J¼7.7, 1.4 Hz, 1H), 7.43 (dd, J¼7.5, 4.8 Hz, 1H), 7.30 (d, J¼6.9 Hz,
4H), 7.23e7.10 (m, 6H), 4.21 (s, 2H); 13C NMR (125 MHz, CDCl3)
d
177.69, 150.35,149.68,136.98,134.23 (2C),128.16 (4C), 127.90 (4C),
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062

6
J.A. Kalow, A.G. Doyle / Tetrahedron xxx (2013) 1e8
127.39 (2C), 126.85, 124.30, 46.75 (2C); HRMS (ESIþ) calculated for
C
0.010 mmol) in TBME (1 mL) were added TBHP (2
and Ti(NMe2)4 (4.7 L, 0.020 mmol), followed by HFIP (84
0.80 mmol), 1b (50 mg, 0.20 mmol), and PhCOF (44
mL, 0.01 mmol)
20H17N2O ([MþH]þ): 301.1335; found: 301.1337.
m
mL,
L,
m
4.2.5. 6-Azabicyclo[3.1.0]hexan-6-yl(pyridin-2-yl)methanone
(1e). Prepared from 6-azabicyclo[3.1.0]hexane.33 FTIR (thin film,
cmꢁ1) 2960, 1655, 1520, 1436, 1393, 1183, 1140, 997; 1H NMR
0.40 mmol). The reaction mixture was sealed and stirred at 23 ꢀC
for 24 h then poured onto SiO2 and purified by column chroma-
tography (5e40% EtOAc/hexanes) to obtain the title compound as
a white solid: 63% yield (34 mg), 78% ee; 70% yield (38 mg), 82%
ee. Compound 2b (64 mg) was recrystallized from IPA to obtain
the product in 97% ee and 56% yield (36 mg). The absolute con-
figuration was assigned by X-ray crystallography (see
Supplementary data). FTIR (thin film, cmꢁ1) 3363, 3062, 3023,
2932, 1670, 1520, 1466, 1435, 1020, 998; 1H NMR (500 MHz, CDCl3)
(500 MHz, CDCl3)
d
8.70 (s, 1H), 8.09 (d, J¼7.8 Hz, 1H), 7.81 (t,
J¼7.7 Hz, 1H), 7.42 (dd, J¼7.4, 4.9 Hz, 1H), 3.31 (s, 2H), 2.22e2.14 (m,
2H), 1.74e1.61 (m, 3H), 1.30e1.19 (m, 1H); 13C NMR (125 MHz,
CDCl3)
d 175.55, 151.17, 149.44, 136.88, 126.36, 123.85, 43.85 (2C),
27.03 (2C), 19.40; HRMS (ESIþ) calculated for C11H13N2O ([MþH]þ):
189.1022; found: 189.1023.
d
8.54e8.48 (m, 1H), 8.21 (d, J¼7.8 Hz, 1H), 8.15 (d, J¼7.1 Hz, 1H),
4.2.6. Benzyl 6-picolinoyl-3,6-diazabicyclo[3.1.0]hexane-3-
carboxylate (1f). Prepared from benzyl 3,6-diazabicyclo[3.1.0]hex-
ane-3-carboxylate.32 FTIR (thin film, cmꢁ1) 2877, 1699, 1668, 1585,
1423, 1390, 1348, 1313, 1211, 1126, 1089, 1040, 995, 914; 1H NMR
7.86 (td, J¼7.7, 1.7 Hz, 1H), 7.43 (ddd, J¼7.6, 4.8, 1.1 Hz, 1H),
7.23e7.18 (m, 2H), 7.17e7.11 (m, 2H), 5.07 (dddd, J¼12.7, 6.6, 6.6,
5.3 Hz, 1H), 4.72e4.64 (m, 1H), 3.47 (dt, J¼16.8, 5.0 Hz, 1H), 3.31
(ddd, J¼22.3, 17.2, 5.1 Hz, 1H), 3.17 (td, J¼17.5, 6.5 Hz, 1H), 2.89 (dd,
(500 MHz, CDCl3)
d
8.67 (d, J¼4.6 Hz, 1H), 7.97 (d, J¼7.8 Hz, 1H), 7.76
J¼16.9, 7.3 Hz, 1H); 13C NMR (126 MHz, CDCl3)
d 164.61, 149.58,
(t, J¼7.7 Hz, 1H), 7.43 (dd, J¼7.3, 4.9 Hz, 1H), 7.37e7.25 (m, 5H), 5.03
(d, J¼3.1 Hz, 2H), 4.11 (d, J¼12.1 Hz, 1H), 4.03 (d, J¼12.3 Hz, 1H),
3.55e3.52 (m,1H), 3.50e3.47 (m,1H), 3.39 (d, J¼1.7 Hz,1H), 3.37 (d,
148.20, 137.57, 132.81 (d, 4J¼1.4 Hz), 132.18 (d, 3J¼6.8 Hz), 129.35,
129.20, 126.82, 126.78, 126.52, 122.42, 89.10 (d, 1J¼177.4 Hz), 48.44
(d, 2J¼22.6 Hz), 33.46 (d, 2J¼21.6 Hz), 32.75 (d, 3J¼4.3 Hz); 19F
J¼1.5 Hz, 1H); 13C NMR (125 MHz, CDCl3)
d
174.31, 155.01, 150.32,
NMR (282 MHz, CDCl3)
d
ꢁ182.77 (dddt, J¼50.1, 22.5, 17.9, 5.1 Hz);
149.32, 137.09, 136.60, 128.56 (2C), 128.12, 128.06 (2C), 126.90,
HRMS (ESIþ) calculated for C16H16FN2O ([MþH]þ): 271.1241;
21
123.94, 67.08, 46.61, 46.05, 41.14, 40.98; HRMS (ESIþ) calculated for
found: 271.1244; [
a]
ꢁ41.5 (c 1.0, CHCl3), 79% ee; HPLC (Chiracel
D
C
18H18N3O3 ([MþH]þ): 324.1343; found: 324.1341.
OD-H, 10% IPA/hexane, 1 mL/min): tR(major)¼17.3, tR(minor)¼
13.0 min.
4.2.7. 3-Oxa-6-azabicyclo[3.1.0]hexan-6-yl(pyridin-2-yl)methanone
(1g). Prepared from 3-oxa-6-azabicyclo[3.1.0]hexane.32 FTIR (thin
film, cmꢁ1) 2861, 1667, 1585, 1439, 1387, 1360, 1318, 1077, 906, 767;
Method B: N-((5R,6R)-6-fluorodecan-5-yl)picolinamide (2c). To
a
solution of (R,R)-(salen)Co(III)OCH(CF3)2$HFIP (9.4 mg,
0.010 mmol) in TBME (0.5 mL) was added Ti(NMe2)4 (4.7 L,
0.020 mmol), followed by HFIP (84 L, 0.80 mmol) and PhCOF
(44 L, 0.40 mmol). Compound 1c (52 mg, 0.20 mmol) was charged
m
1H NMR (500 MHz, CDCl3)
d
8.68 (d, J¼4.5 Hz, 1H), 8.14 (d, J¼7.8 Hz,
m
1H), 7.83 (td, J¼7.8, 1.6 Hz, 1H), 7.45 (dd, J¼7.0, 5.3 Hz, 1H), 4.05 (d,
m
J¼10.0 Hz, 2H), 3.66e3.49 (m, 4H); 13C NMR (125 MHz, CDCl3)
as a solution in TBME (0.5 mL) and the reaction mixture was stirred
under N2 at 23 ꢀC for 48 h. The reaction mixture was poured onto
SiO2 and purified by column chromatography (5e40% Et2O/hex-
anes) to obtain the title compound as a clear oil: 80% yield (45 mg),
67% ee; 70% yield (39 mg), 58% ee. FTIR (thin film, cmꢁ1) 3385,
2956, 2932, 2861, 1680, 1518, 1466, 1434, 998; 1H NMR (500 MHz,
d
174.21, 150.84, 149.07, 137.08, 126.69, 124.02, 66.42 (2C), 41.24
(2C); HRMS (ESIþ) calculated for C10H11N2O2 ([MþH]þ): 191.0815;
found: 191.0814.
4.2.8. 8-Azabicyclo[5.1.0]octan-8-yl(pyridin-2-yl)methanone
(1h). Prepared from 8-azabicyclo[5.1.0]octane.34 FTIR (thin film,
cmꢁ1) 2927, 1663, 1519, 1464, 1435, 1326, 1178, 1155, 997, 820; 1H
CDCl3)
d
8.59e8.56 (m,1H), 8.20 (d, J¼7.8 Hz,1H), 8.10 (d, J¼10.0 Hz,
1H), 7.86 (td, J¼7.7, 1.6 Hz, 1H), 7.44 (ddd, J¼7.5, 4.8, 1.0 Hz, 1H), 4.60
(dddd, J¼47.4, 8.8, 4.2, 1.4 Hz, 1H), 4.22 (dddd, J¼29.0, 17.2, 7.5,
1.4 Hz, 1H), 1.77e1.64 (m, 3H), 1.56e1.23 (m, 9H), 0.90e0.85 (m,
NMR (500 MHz, CDCl3)
d
8.70 (d, J¼3.9 Hz, 1H), 8.07 (d, J¼7.8 Hz,
1H), 7.81 (td, J¼7.7, 1.5 Hz, 1H), 7.42 (dd, J¼6.5, 4.7 Hz, 1H), 2.82 (d,
J¼2.8 Hz, 2H), 2.14e1.96 (m, 4H), 1.74e1.48 (m, 5H), 1.38e1.28 (m,
6H); 13C NMR (125 MHz, CDCl3)
d 164.45, 149.74, 148.29, 137.47,
1H); 13C NMR (125 MHz, CDCl3)
d
177.82, 151.10, 149.54, 136.79,
126.35, 122.52, 95.50 (d, 1J¼172.9 Hz), 51.69 (d, 2J¼18.9 Hz), 32.23
126.35,124.01, 42.22 (2C), 31.60, 29.19 (2C), 25.64 (2C); HRMS (ESIþ)
(d, 3J¼3.0 Hz), 31.77 (d, 2J¼20.8 Hz), 28.38, 27.55 (3d, J¼4.8 Hz),
calculated for C13H17N2O ([MþH]þ), 217.1335; found: 217.1333.
22.71, 22.58, 14.14, 14.12; 19F NMR (282 MHz, CDCl3)
d
ꢁ196.06
(dddd, J¼46.6, 31.4, 29.4, 14.5 Hz); HRMS (ESIþ) calculated for
16H26FN2O ([MþH]þ): 281.2024; found: 281.2025; [
a]
þ14.5 (c
21
4.2.9. Benzyl 7-picolinoyl-3,7-diazabicyclo[4.1.0]heptane-3-
carboxylate (3). Prepared from benzyl 3,7-diazabicyclo[4.1.0]hep-
tane-3-carboxylate.35 FTIR (thin film, cmꢁ1) 2929, 1680, 1673, 1585,
1415, 1319, 1242, 1216, 1137, 1107, 1046, 995, 906, 828; 1H NMR
C
D
1.4, CHCl3), 67% ee; HPLC (Chiralpak AD-H, 1% IPA/hexane, 1 mL/
min): tR(major)¼26.9, tR(minor)¼29.0 min.
(500 MHz, CDCl3)
d
8.73e8.61 (m, 1H), 8.15e8.02 (m, 1H), 7.87e7.74
4.3.1. N-((1R,2R)-2-Fluorocyclohexyl)picolinamide (2a). Method A,
1a (40 mg, 0.20 mmol), 5e40% EtOAc/hexanes with 0.5% Et3N: 94%
yield (42 mg), 83% ee; 92% yield (41 mg), 85% ee. Compound 2a
(350 mg) was recrystallized by from IPA to obtain the product in
(m, 1H), 7.51e7.40 (m, 1H), 7.38e7.27 (m, 5H), 5.13 (s, 2H),
4.20e3.93 (m, 2H), 3.70e3.53 (m, 1H), 3.47e3.22 (m, 1H), 3.13, 2.85
(m, 2H), 2.46e2.18 (m, 1H), 2.11e1.87 (m, 1H); 13C NMR (125 MHz,
CDCl3, peaks corresponding to rotamers indicated by *)
177.10 , 155.61, 155.54 , 150.44, 149.51, 149.45 , 136.97, 136.92
136.76, 128.57 (2C), 128.10, 127.98 (2C), 126.76, 124.19, 124.13
d
177.23,
95% ee and 51% yield (180 mg). White solid. Spectral data were in
were in agreement with the literature values;6
[
a
]
ꢁ36.5 (c 0.7,
21
*
*
*
*
*
*
,
,
,
D
CHCl3), 84% ee; HPLC (Chiralpak AS-H, 10% IPA/hexane, 1 mL/min):
67.27, 42.52, 42.43
*
, 38.48, 38.30
*
, 36.03, 35.79
*
, 35.75, 35.35
tR(major)¼13.1, tR(minor)¼14.8 min.
23.95, 23.50
*
; HRMS (ESIþ) calculated for C19H20N3O3 ([MþH]þ):
338.1499; found: 338.1497.
4.3.2. N-((1R,2R)-2-Fluoro-1,2-diphenylethyl)picolinamide
(2d). Method B, 1d (60 mg, 0.20 mmol), 5e15% acetone/hexanes:
42% yield (27 mg), 48% ee; 59% yield (38 mg), 48% ee. White solid.
FTIR (thin film, cmꢁ1) 3383, 3063, 3034, 2932, 1676, 1512, 1465,
4.3. Representative procedures for the desymmetrization of
meso aziridines
1455, 1435, 998, 697; 1H NMR (500 MHz, CDCl3)
d
8.88 (d, J¼8.9 Hz,
Method A: N-((2R,3R)-3-fluoro-1,2,3,4-tetrahydronaphthalen-2-
yl)picolinamide (2b). To a suspension of (R,R)-(salen)Co (6.0 mg,
1H), 8.59 (d, J¼4.7 Hz, 1H), 8.10 (d, J¼7.8 Hz, 1H), 7.81 (td, J¼7.7,
1.2 Hz,1H), 7.43 (dd, J¼6.9, 4.5 Hz,1H), 7.38e7.27 (m,10H), 5.84 (dd,
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062

J.A. Kalow, A.G. Doyle / Tetrahedron xxx (2013) 1e8
7
J¼45.9, 4.4 Hz, 1H), 5.61 (ddd, J¼21.5, 9.1, 4.4 Hz, 1H); 13C NMR
1.6 Hz, 1H), 7.42 (ddd, J¼7.5, 4.8, 1.0 Hz, 1H), 4.63 (dm,
JHF¼48.2 Hz, 1H), 4.28 (ddddd, J¼17.3, 8.7, 8.7, 8.6, 3.2 Hz, 1H),
2.03e1.93 (m, 3H), 1.82e1.59 (m, 5H), 1.59e1.44 (m, 2H); 13C NMR
(126 MHz, CDCl3)
d
163.95, 149.50, 148.32, 138.32 (d, 3J¼1.8 Hz),
137.45, 136.79 (d, 2J¼20.6 Hz), 128.73 (d, 3J¼1.5 Hz), 128.72 (2C),
128.42 (2C), 128.05, 127.53 (d, 3J¼0.5 Hz, 2C), 126.49, 126.11, 126.05,
122.47, 95.60 (d, 1J¼180.9 Hz), 57.64 (d, 2J¼21.6 Hz); 19F NMR
(125 MHz, CDCl3)
d 163.93, 149.97, 148.11, 137.51, 126.31, 122.39,
96.46 (d, 1J¼173.8 Hz), 55.24 (d, 2J¼21.8 Hz), 31.89 (d, 2J¼21.2 Hz),
(282 MHz, CDCl3)
d
ꢁ188.37 (dd, J¼46.0, 21.5 Hz); HRMS (ESIþ)
30.29 (d, 3J¼7.6 Hz), 27.95, 24.45, 21.34 (d, 3J¼8.2 Hz); 19F NMR
calculated for C20H18FN2O ([MþH]þ), 321.1398; found: 321.1394;
(282 MHz, CDCl3)
d
ꢁ169.35 (m); HRMS (ESIþ) calculated for
21
[a]D ꢁ4.5 (c 1.0, CHCl3), 48% ee; HPLC (Chiralpak AS-H, 5% IPA/
C
13H18FN2O ([MþH]þ): 237.1398; found: 237.1401; HPLC (Chir-
ꢁ
hexane, 1 mL/min): tR(major)¼22.9, tR(minor)¼16.8 min.
alpak AS-H, 10% IPA/hexane,
1
mL/min): tR(major)¼11.2,
tR(minor)¼12.1 min.
4.3.3. N-((1R,2R)-2-Fluorocyclopentyl)picolinamide (2e). Method A,
1e (75 mg, 0.40 mmol), 5e40% EtOAc/hexanes: 43% yield (36 mg),
4.4. Procedure for the kinetic resolution of 3
50% ee; 48% yield (40 mg), 61% ee. White solid. FTIR (thin film, cmꢁ1
)
3378, 3059, 2964, 2877, 1662, 1516, 1465, 1434, 998, 957, 820; 1H
NMR (500 MHz, CDCl3)
4.4.1. (3R,4R)-Benzyl 4-fluoro-3-(picolinamido)piperidine-1-
carboxylate (4). (R,R)-(salen)Co (12 mg, 0.020 mmol) was sus-
d
8.53 (d, J¼4.7 Hz, 1H), 8.20 (br d, J¼7.8 Hz,
1H), 7.94 (d, J¼3.0 Hz, 1H), 7.86 (td, J¼7.7, 1.1 Hz, 1H), 7.44 (dd, J¼7.1,
5.2 Hz,1H), 5.06 (ddd, J¼9.6, 6.8, 3.5 Hz,1H), 4.54e4.45 (m,1H), 2.30
(ddd, J¼13.5, 13.2, 7.6 Hz, 1H), 2.05e1.82 (m, 4H), 1.71e1.61 (m, 1H);
pended in TBME (1 mL) and TBHP (4
(9.5 L, 0.040 mmol) were added via syringe. Compound 3 (135 mg,
0.400 mmol) and HFIP (84 L, 0.80 mmol) were added as a solution
in TBME (1 mL), followed by PhCOF (35 L, 0.32 mmol). The reaction
mL, 0.02 mmol) and Ti(NMe2)4
m
m
13C NMR (125 MHz, CDCl3)
d
164.15, 149.67, 148.12, 137.60, 126.46,
m
122.37, 98.72 (d, 1J¼179.5 Hz), 56.41 (d, 2J¼28.2 Hz), 31.11 (d,
mixture was sealed and stirred at 23 ꢀC for 4 or 7 h, then quenched
with 7 M methanolic ammonia (4 mL) and stirred vigorously for
15 min. The red solution was poured into 1 M NaOH (10 mL) and
extracted with DCM (3ꢂ5 mL). The combined organic layers were
washed with saturated brine (5 mL), dried over MgSO4, filtered, and
concentrated in vacuo. The crude material was purified by column
chromatography on SiO2 (7e60% EtOAc/hexanes) to obtain the title
product as a yellow oil. The 4-h reaction provided 4 in 29% yield
(42 mg) and 68% ee (s¼6.9), with 65% yield recovered 3 (87 mg) in
29% ee. The 7-h reaction provided 4 in 41% yield (59 mg) and 62% ee
(s¼6.4). FTIR (thin film, cmꢁ1) 3362, 3062, 2932, 2858, 1697, 1675,
1519, 1466, 1433, 1314, 1269, 1229, 1139, 998; 1H NMR (500 MHz,
2J¼21.9 Hz), 30.27 (d, 3J¼1.8 Hz), 21.63 (d, 3J¼1.4 Hz); 19F NMR
(282 MHz, CDCl3)
d
ꢁ175.65 (ddddd, J¼46.0, 29.3, 28.4,18.2, 0.9 Hz);
HRMS (ESIþ) calculated for C11H14FN2O ([MþH]þ), 209.1085; found:
209.1088; [
a]
22 ꢁ11.8 (c 1.2, CHCl3), 61% ee; HPLC (Chiralpak AS-H,
D
5% IPA/hexane, 1 mL/min): tR(major)¼24.4, tR(minor)¼23.4 min.
4.3.4. (3R,4R)-Benzyl 3-fluoro-4-(picolinamido)pyrrolidine-1-
carboxylate (2f). Method B, 96 h, 1f (162 mg, 0.400 mmol), 7e60%
EtOAc/hexanes: 47% yield (80 mg), 78% ee; 49% yield (84 mg), 72%
ee. Yellow oil. FTIR (thin film, cmꢁ1) 3290, 3062, 2953, 2890, 1697,
1667, 1515, 1418, 1357, 1338, 1212, 1185, 1104, 997, 955, 766, 696; 1H
NMR (500 MHz, MeOD)
d
8.61 (d, J¼4.3 Hz, 1H), 8.08 (d, J¼7.8 Hz,
MeOD)
d
8.62 (d, J¼4.4 Hz, 1H), 8.10 (d, J¼7.8 Hz, 1H), 7.97 (td, J¼7.7,
1H), 7.95 (td, J¼7.7, 1.5 Hz,1H), 7.57e7.53 (m,1H), 7.41e7.27 (m, 5H),
5.18 (dm, JHF¼50.5 Hz, 1H), 5.15 (s, 2H), 4.71e4.62 (m, 2H),
3.93e3.63 (m, 4H); 13C NMR (125 MHz, MeOD, mixture of rotamers,
1.6 Hz, 1H), 7.56 (ddd, J¼7.5, 4.8, 1.0 Hz, 1H), 7.34 (d, J¼25.0 Hz, 5H),
5.15 (s, 2H), 4.79 (ddd, J¼48.8, 12.0, 8.2 Hz, 1H), 4.16 (ddd, J¼17.1,
8.6, 4.7 Hz, 1H), 4.12e4.06 (m, 1H), 3.94 (d, J¼1.2 Hz, 1H), 3.29e3.16
(m, 2H), 2.20e2.08 (m, 1H), 1.84e1.72 (m, 1H); 13C NMR (125 MHz,
indicated by *)
d
166.97, 156.66, 156.60 , 150.51, 149.81, 138.86,
*
137.94, 129.55, 129.17, 129.16
*
, 129.00, 128.99
*
, 128.05, 123.37, 95.31
MeOD) d 166.74, 156.90, 150.54, 149.79, 138.88, 137.92, 129.56,
(d, J¼182.1 Hz), 94.53
54.72
50.08, 49.85
dicated by *)
*
(d, 1J¼183.2 Hz), 68.35, 55.57 (d, 2J¼29.9 Hz),
129.16, 128.90, 128.01, 123.26, 90.89 (d, J¼179.8 Hz), 68.63, 51.34 (d,
*
(d, J¼29.2 Hz), 51.52 (d, J¼26.7 Hz), 51.34
*
(d, J¼26.7 Hz),
J¼20.71 Hz), 46.11 (d, J¼5.3 Hz), 41.93, 30.66; 19F NMR (282 MHz,
*
;
d
19F NMR (282 MHz, MeOD, mixture of rotamers, in-
MeOD)
C
d
ꢁ185.20 (dm, JHF¼49.4 Hz); HRMS (ESIþ) calculated for
22
ꢁ181.46 (dddd, J¼49.1, 36.1, 25.3, 12.8 Hz), ꢁ181.84
*
19H21FN3O3 ([MþH]þ): 358.1561; found: 358.1558; [
a]
þ17.8 (c
D
(dddd, J¼48.8, 37.6, 25.2, 12.7 Hz); HRMS (ESIþ) calculated for
1.1, CHCl3), 68% ee; HPLC (Chiralpak AD-H, 15% IPA/hexane, 0.9 mL/
min): tR(major)¼19.6, tR(minor)¼24.3 min. The absolute stereo-
chemistry was assigned in analogy to 2a,b.
22
C
18H19FN3O3 ([MþH]þ): 344.1405; found: 344.1403; [
a
]
þ30.7 (c
D
1.3, CHCl3), 72% ee; HPLC (Chiralpak AD-H, 17.5% IPA/hexane,
0.9 mL/min): tR(major)¼15.6, tR(minor)¼17.3 min.
4.5. Procedure for the deprotection of 2a
4.3.5. N-((3R,4S)-4-Fluorotetrahydrofuran-3-yl)picolinamide
(2g). Method B, 96 h, 1g (95 mg, 0.50 mmol), 12e100% EtOAc/
hexanes: 27% yield (28 mg, 85% purity), 32% ee. Off-white solid.
FTIR (thin film, cmꢁ1) 3369, 3061, 2970, 2869, 1717, 1665, 1514,
1466, 1383, 1351, 1279, 1075, 998, 904, 769, 713; 1H NMR (500 MHz,
4.5.1. tert-Butyl
((1R,2R)-2-fluorocyclohexyl)carbamate
(5). To
a refluxing solution of 2a (181 mg, 0.814 mmol, 95% ee) and 4-
dimethylaminopyridine (10 mg, 0.081 mmol) in THF (3.3 mL) was
added di-tert-butyl dicarbonate portionwise (8ꢂ0.19 mL, 6.51 mmol
total) over 24 h. Upon completion, the reaction mixture was
concentrated in vacuo and purified by column chromatography
on SiO2 (5e40% EtOAc/hexanes) to obtain tert-butyl ((1R,2R)-2-
fluorocyclohexyl)(picolinoyl)-carbamate as a white solid (249 mg,
95% yield). N-Boc-2a (247 mg) was dissolved in EtOH (6 mL) and
NaBH4 (116 mg, 3.06 mmol) was charged. The reaction mixture was
stirred at 23 ꢀC for 12 h, then quenched with 1 M NaOH (10 mL) and
extracted withEt2O (3ꢂ10 mL). Thecombined organics werewashed
with saturated brine (5 mL), dried over MgSO4, filtered, and con-
centrated in vacuo. The crude material was purified by column
chromatography on SiO2 (5e40% Et2O/hexanes) to obtain 5 as
a white solid in 96% yield (159 mg, 91% yield over two steps). The
enantiomeric excess was determined to be 95% bychiral GC. Spectral
CDCl3)
d
8.54 (d, J¼4.6 Hz, 1H), 8.19 (d, J¼7.8 Hz, 1H), 8.12 (br s, 1H),
7.87 (td, J¼7.7, 1.4 Hz, 1H), 7.46 (dd, J¼7.5, 4.8 Hz, 1H), 5.21 (dd,
J¼52.2, 3.6 Hz, 1H), 4.71 (dd, J¼13.1, 6.2 Hz, 1H), 4.26e3.84 (m, 4H);
13C NMR (126 MHz, CDCl3)
d 164.40, 149.11, 148.30, 137.67, 126.78,
122.40, 95.91 (d, 1J¼184.3 Hz), 72.29 (d, 2J¼24.0 Hz), 71.18, 55.68 (d,
2J¼28.8 Hz); 19F NMR (282 MHz, CDCl3)
d
ꢁ176.09 (dddd, J¼51.6,
37.1, 29.0, 14.1 Hz); HRMS (ESIþ) calculated for C10H12FN2O2
([MþH]þ), 211.0877; found: 211.0880; HPLC (Chiracel OD-H, 10%
IPA/hexane, 1 mL/min): tR(major)¼14.7, tR(minor)¼12.6 min.
4.3.6. N-((1R,2R)-2-Fluorocycloheptyl)picolinamide (2h). Method B,
1h (87 mg, 0.40 mmol), 2e20% EtOAc/hexanes: 33% yield (31 mg),
12% ee. Clear oil. FTIR (thin film, cmꢁ1) 3371, 2932, 3862, 1666,
1520, 1465, 1435, 998; 1H NMR (500 MHz, CDCl3)
d
8.56e8.53 (m,
1H), 8.24e8.21 (br m, 1H), 8.20 (d, J¼7.8 Hz, 1H), 7.84 (td, J¼7.7,
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062
data were in agreement with the literature values; [
a]
22 ꢁ34.7 (c 1.1,
D
CHCl3), 95% ee; GC (Cyclodex-B, 130 ꢀC isotherm, 1 mL/min):

8
J.A. Kalow, A.G. Doyle / Tetrahedron xxx (2013) 1e8
14. For a review, see: Park, J.; Hong, S. Chem. Soc. Rev. 2012, 41, 6931e6943.
15. Allen, A. E.; MacMillan, D. W. C. Chem. Sci. 2012, 3, 633e658.
16. For key references, see: (a) Zaitsev, V. G.; Shabashov, D.; Daugulis, O. J. Am.
Chem. Soc. 2005, 127, 13154e13155; (b) He, G.; Chen, G. Angew. Chem., Int. Ed.
2011, 50, 5192e5196.
tR(major)¼33.3, tR(minor)¼34.0 min. The absolute configuration of
the product was assigned as (R,R) on the basis of comparison to the
literature data for (S,S)-5 ([
a]
22 þ27.5 (c 1.01, CHCl3), 99% ee).6
D
17. (a) Hayashi, M.; Shiomi, N.; Funahashi, Y.; Nakamura, S. J. Am. Chem. Soc. 2012,
134, 19366e19369; (b) Nakamura, S.; Hayashi, M.; Kamada, Y.; Sasaki, R.; Hir-
amatsu, Y.; Shibata, N.; Toru, T. Tetrahedron Lett. 2010, 51, 3820e3823.
18. Fluoride ring opening of aziridine trans-1d was not observed in this reaction.
Based on literature precedent, this is expected to be a challenging substrate
for ring opening by (salen)M: Brandes, B. D.; Jacobsen, E. N. Synlett 2001,
1013e1015.
Acknowledgements
We thank Dana Schmitt for studies of the deprotection of 2a and
Phil Jeffrey for X-ray crystallographic structure determination of 2b.
Financial support was provided by Princeton University (fellowship
to J.A.K.), the NSF (CAREER-1148750) and the ACS Division of
Organic Chemistry (fellowship to J.A.K. sponsored by Pfizer).
19. Ferraris, D.; Drury, W. J.; Cox, C.; Lectka, T. J. Org. Chem. 1998, 63, 4568e4569.
20. Under the standard conditions using 1,4-dioxane as solvent in the presence of
ꢀ
activated powdered 4 A molecular sieves, reaction of 1a provided 2a in 68%
yield and 80% ee, accompanied by oxazoline in 20% yield and 80% ee. Control
experiments confirmed that the oxazoline did not arise from product 2a. In
TBME, the yield and ee of the oxazoline side product were depressed. The
asymmetric synthesis of oxazolines from alkenes in up to 92% ee using a stoi-
chiometric chiral nitridomanganese has been described; however, this is be-
lieved to proceed through rearrangement of the enantioenriched aziridine,
which does not apply to meso N-acyl aziridines: Nishimura, M.; Minakata, S.;
Takahashi, T.; Oderaotoshi, Y.; Komatsu, M. J. Org. Chem. 2002, 67, 2101e2110.
21. (a) Burger, M.; Lan, J.; Lindvall, M.; Nishigughi, G.; Tetalman, M. PCT Int. Patent.
Appl. EP2009/052506, March 3, 2009; (b) le Huerou, Y.; Blake, J. F.; Gunwar-
dana, I. W.; Mohr, P.; Wallace, E. M.; Wang, B.; Chicarelli, M.; Lyon, M. PCT Int.
Patent Appl. US2009/043,691, May 13, 2009; (c) Ninkovic, S.; Braganza, J. F.;
Collins, M. R.; Kath, J. C.; Li, H.; Richter, D. T. PCT Int. Patent Appl. IB2009/
053389, August 4, 2009.
Supplementary data
Supplementary data associated with this article can be found in
the online version, at http://dx.doi.org/10.1016/j.tet.2013.01.062.
References and notes
1. For reviews, see: (a) Hu, X. E. Tetrahedron 2004, 60, 2701e2743; (b) Pineschi, M.
Eur. J. Org. Chem. 2006, 4979e4988; (c) Schneider, C. Angew. Chem., Int. Ed. 2009,
48, 2082e2084.
2. Mita, T.; Jacobsen, E. Synlett 2009, 1680e1684.
22. (a) Hu, X. E.; Kim, N. K.; Ledoussal, B.; Colson, A.-O. Tetrahedron Lett. 2002, 43,
4289e4293; (b) Hu, X. E. Tetrahedron Lett. 2002, 43, 5315e5318.
3. Ohmatsu, K.; Hamajima, Y.; Ooi, T. J. Am. Chem. Soc. 2012, 134, 8794e8797.
4. For alternative approaches that employ asymmetric catalytic fluorination, see:
(a) Fadeyi, O. O.; Lindsley, C. W. Org. Lett. 2009, 11, 943e946; (b) Appayee, C.;
Brenner-Moyer, S. E. Org.Lett. 2010,12, 3356e3359; (c)Schulte, M.L.; Lindsley, C.W.
Org. Lett. 2011, 13, 5684e5687; (d) Phipps, R. J.; Hiramatsu, K.; Toste, F. D. J. Am.
Chem. Soc. 2012, 134, 8376e8379.
23. The value s represents a lower bound for the selectivity factor calculated from
the isolated yield (C) and ee of product (ee): s¼ln[1ꢁC(1þee)]/ln[1ꢁC(1ꢁee)].
24. CCDC 919698 contains the supplementary crystallographic data for this paper.
These data can be obtained free of charge from the Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
5. (a) O’Hagan, D. Chem. Soc. Rev. 2008, 37, 308e319; (b) Morgenthaler, M.;
€
25. Fora review, see:Girard, C.; Kagan, H. B. Angew. Chem., Int. Ed.1998, 37, 2922e2959.
Schweizer, E.; Hoffmann-Roder, A.; Benini, F.; Martin, R. E.; Jaeschke, G.;
€
26. Burger, V. H.; Wiegel, K. Z. Anorg. Allg. Chem. 1973, 398, 257e272.
Wagner, B.; Fischer, H.; Bendels, S.; Zimmerli, D.; Schneider, J.; Diederich, F.;
€
27. Pagenkopf, B. L.; Carreira, E. M. Chem.dEur. J. 1999, 5, 3437e3442.
28. (a) Pagenkopf, B.; Carreira, E. Tetrahedron Lett. 1998, 39, 9593e9596; (b)
Fleischer, R.; Wunderlich, H.; Braun, M. Eur. J. Org. Chem. 1998, 1063e1070.
29. Watson, I. D. G.; Afagh, N.; Yudin, A. K. Org. Synth. 2010, 87, 161e169.
30. Yano, T.; Kobayashi, H.; Ueno, K. Bull. Chem. Soc. Jpn. 1973, 46, 985e990.
31. The aziridine was prepared according to a method for the corresponding cis-
dipropylaziridine: (a) Ref. 13a. Characterization data are provided in: (b) Sen-
boku, H.; Nakahara, K.; Fukuhara, T.; Hara, S. Tetrahedron Lett. 2010, 51,
435e438.
32. Rowland, E. B.; Rowland, G. B.; Rivera-Otero, E.; Antilla, J. C. J. Am. Chem. Soc.
2007, 129, 12084e12085.
33. da Zhang, Z.; Scheffold, R. Helv. Chim. Acta 1993, 76, 2602e2615.
34. Hassner, A.; Matthews, G. J.; Fowler, F. W. J. Am. Chem. Soc. 1969, 91, 5046e5054.
35. Schramm, H.; Pavlova, M.; Hoenke, C.; Christoffers, J. Synthesis 2009,
1659e1662.
Kansy, M.; Muller, K. ChemMedChem 2007, 2, 1100e1115.
6. Kalow, J. A.; Schmitt, D. E.; Doyle, A. G. J. Org. Chem. 2012, 77, 4177e4183.
7. Kalow, J. A.; Doyle, A. G. J. Am. Chem. Soc. 2010, 132, 3268e3269.
8. Kalow, J. A.; Doyle, A. G. J. Am. Chem. Soc. 2011, 133, 16001e16012.
ꢁ
9. (a) Jacobsen, E. N. Acc. Chem. Res. 2000, 33, 421e431; (b) Kemper, S.; Hrobarik,
€
P.; Kaupp, M.; Schlorer, N. E. J. Am. Chem. Soc. 2009, 131, 4172e4173.
ꢁ
10. Li, Z.; Fernandez, M.; Jacobsen, E. N. Org. Lett. 1999, 1, 1611e1613.
11. (a) Fujimori, I.; Mita, T.; Maki, K.; Shiro, M.; Sato, A.; Furusho, S.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2006, 128, 16438e16439; (b) Xu, Y.; Lin, L.; Kanai,
M.; Matsunaga, S.; Shibasaki, M. J. Am. Chem. Soc. 2011, 133, 5791e5793.
12. (a) Arai, K.; Lucarini, S.; Salter, M. M.; Ohta, K.; Yamashita, Y.; Kobayashi, S. J. Am.
Chem. Soc. 2007, 129, 8103e8111; (b) Yu, R.; Yamashita, Y.; Kobayashi, S. Adv.
Synth. Catal. 2009, 351, 147e152.
13. (a) Wu, B.; Gallucci, J. C.; Parquette, J. R.; RajanBabu, T. V.Angew. Chem., Int. Ed. 2009,
48,1126e1129; (b) Wu, B.; Parquette, J. R.; RajanBabu, T. V. Science 2009, 326,1662.
Please cite this article in press as: Kalow, J. A.; Doyle, A. G., Tetrahedron (2013), http://dx.doi.org/10.1016/j.tet.2013.01.062