Asymmetric electrophilic amination of enolates by a chiral N-alkoxycarbonyloxaziridine

Article abstract of DOI:10.1016/S0957-4166(01)00104-5

A new chiral 3-aryl-N-alkoxycarbonyloxaziridine, derived from menthol, has been prepared and tested as a reagent for asymmetric electrophilic amination of enolates. The aminated products were obtained in low diastereoselectivities of up to 21% d.e.

Full text of DOI:10.1016/S0957-4166(01)00104-5

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 535–538
Asymmetric electrophilic amination of enolates by a chiral
N-alkoxycarbonyloxaziridine
Alan Armstrong,a,* Mark A. Atkina and Steven Swallowb
aDepartment of Chemistry, Imperial College of Science, Technology and Medicine, London SW7 2AY, UK
bRoche Discovery Welwyn, Broadwater Road, Welwyn Garden City, Herts. AL7 3AY, UK
Received 12 February 2001; accepted 1 March 2001
Abstract—A new chiral 3-aryl-N-alkoxycarbonyloxaziridine, derived from menthol, has been prepared and tested as a reagent for
asymmetric electrophilic amination of enolates. The aminated products were obtained in low diastereoselectivities of up to 21%
d.e. © 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
2. Results and discussion
Electrophilic amination provides a powerful method for
the introduction of nitrogen into organic molecules.1
An asymmetric variant of this methodology would be
especially useful.2 Work to date has largely involved the
use of chiral enolate derivatives with achiral aminating
agents, but the development of chiral sources of elec-
trophilic nitrogen would offer greater synthetic flexibil-
ity. We have been attracted3 to the pioneering work of
Vidal and Collet,4–8 who showed that N-alkoxycar-
bonyloxaziridines (e.g. 1) transfer nitrogen rather than
oxygen to a number of nucleophiles. When the amina-
tion of ester enolates is considered, this methodology
provides a conceptually simple synthesis of a-amino
acids (Scheme 1). In this paper, we report the first
examples of the use of chiral, non-racemic oxaziridines
in this process, leading to asymmetric electrophilic
amination.
The first class of chiral, non-racemic oxaziridines we
investigated are analogues of the Vidal and Collet
compounds 1 where the Bu-carbamate is replaced by
t
one derived from a chiral alcohol, specifically 4, derived
from (−)-menthol. The preparation of 4 is shown in
Scheme 2. Aza–Wittig reaction9 of p-cyanobenzalde-
hyde with the iminophosphorane10 derived from men-
thyl chloroformate 211 gave the imine 3; this sensitive
intermediate was used in the next step without purifica-
tion since analysis of the crude 1H NMR spectrum
showed a very clean conversion (>95%) from the start-
ing aldehyde. Oxidation to the oxaziridine was then
performed using m-CPBA/BuLi.12 Since the ring car-
bon and nitrogen in 4 are both stereocentres, this
oxidation step could potentially lead to the formation
of four diastereoisomers, two pairs of which could
undergo interconversion by inversion at nitrogen. We
O
O
O
OLi
N
OtBu
+
NHBoc
tBuO
tBuO
NC
1
35%
Scheme 1.
* Corresponding author. E-mail: a.armstrong@ic.ac.uk
0957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00104-5

536
A. Armstrong et al. / Tetrahedron: Asymmetry 12 (2001) 535–538
O
O
H
(i), (ii), (iii)
O
Cl
O
N
2
CN
O
3
(iv)
O
H
N
O
O
N
O
H
O
4a
4b
CN
CN
Scheme 2. Reagents: (i) NaN3, Bu4NBr, CH2Cl2, H2O; (ii) PPh3, CH2Cl2, 81% for two steps; (iii) (p-CN)C6H4CHO, PhMe, reflux,
48 h; (iv) m-CPBA, n-BuLi, CH2Cl2, −78°C to rt, 61% for two steps.
1
had originally assumed that facial selectivity on nucleo-
philic addition to the imine would be low, leading to a
mixture of diastereomers at the ring carbon. Providing
these were separable, testing of the individual isomers
in the amination reaction would then provide valuable
information on the mechanism of asymmetric induction
in the amination process. Interestingly, however, the
oxidation product was shown by HPLC analysis under
detect other diastereomers by H or 13C NMR, suggest-
ing that the compounds were single diastereomers at
the ring carbon. Thus, the oxidation appears to have
proceeded highly diastereoselectively with respect to
facial attack on the imine carbon. It is not clear at this
time whether this phenomenon represents kinetic or
thermodynamic control, since nucleophilic addition to
the imine is potentially reversible. An X-ray crystal
structure14 of 4 (Fig. 1) allowed assignment of configu-
ration to the ring carbon. Compound 4 therefore
appears to represent a rare example of a chiral, non-
racemic N-alkoxycarbonyloxaziridine.15
1
several sets of conditions to be a single isomer. The H
NMR spectrum of 4 indicated the presence of two
compounds in a ca. 9:1 ratio. Variable temperature
studies showed coalescence of the two sets of reso-
nances over a temperature range of 70–80°C. This
suggests that they are (E)- and (Z)-isomers 4a and 4b,
interrelated by inversion at nitrogen with a barrier of
ca. 16–17 kcal/mol.13 However, we were unable to
With the synthesis of the desired non-racemic
oxaziridine accomplished, attention was then turned to
its use in the amination of enolates derived from achiral
Figure 1. X-Ray structure of oxaziridine 4a.

A. Armstrong et al. / Tetrahedron: Asymmetry 12 (2001) 535–538
Table 1. Enolate amination with oxaziridine 4a
537
Since the nature of the ester substituent appeared to
have little effect on the diastereoselectivity (compare
entries 4–6), we next investigated an alternative R
group in 5. Using an iso-valerate (entry 8), an improved
diastereoselectivity, with product of 17% d.e., was
observed (entry 8). While the low levels of diastereose-
lectivity observed make construction of a working TS-
model premature, it is conceivable that coordination
between the enolate and the oxaziridine might lead to a
more highly organised TS. In a preliminary study along
these lines, we found that the addition of Ti(OiPr)4 to
the lithium enolate of ethyl iso-valerate (entry 9)
R
R
O
(i) LDA, THF
X
X
N
H
O
(ii) 4
O
O
5
6
Entry
R
X
Yield (%)c
D.e. (%)d
1
2
3
4
5
6
7
8
9b
H
H
H
Me
Me
Me
Me
iPr
iPr
OtBu
Ph
NMe2
OMe
OEt
OtBu
Ph
60
59
56
57
52
51
62
49
51
–
–
–
5
8
afforded
a
further small increase in amination
diastereoselectivity to give a d.e. of 21%.
7
5e
17
21
3. Conclusion
OEt
OEt
In conclusion, we have prepared the novel chiral, non-
racemic oxaziridine 4 and demonstrated for the first
time that such compounds can be used to effect the
asymmetric electrophilic amination of enolates, albeit
with low diastereoselectivity. Having demonstrated that
the concept is feasible, future studies will be aimed at
improving the diastereoselectivity, both by optimisation
of the reaction conditions and in particular by the
design and preparation of further structurally varied
novel chiral oxaziridines that provide a better defined
chiral environment.
a Using 1 equiv. of oxaziridine with respect to enolate (see Ref. 16).
b 1 Equiv. Ti(OiPr)4 added to the lithium enolate prior to the addition
of oxaziridine.
c Isolated yields of pure products.
d Determined by GC. Configuration assigned by comparison to
authentic samples prepared from enantiomerically pure amino acid
esters and (−)-menthyl chloroformate. Major isomer as depicted.
e Relative configuration not determined.
ketones and esters 5 (Table 1). In a typical amination
reaction,16 a THF solution of the oxaziridine was added
to the lithium enolate of the substrate in THF at
−78°C; reaction was completed by allowing it to warm
to room temperature after 2–3 h. Diastereoselectivity
was measured where appropriate by GC analysis, and
configuration assigned by comparison to authentic sam-
ples of the diastereomerically pure products prepared
by reaction of the commercially available enantiomeri-
cally pure amino acid esters with (−)-menthyl chlorofor-
mate. To confirm the reactivity of 4, a number of
investigations were performed initially using simple ter-
minal enolates derived from esters, ketones and amides
(entries 1–3). In all cases the amination reaction pro-
ceeded to afford a-amino compounds 6 in good yield.
Amination of a range of lithium enolates of propi-
onates (entries 4–7) revealed that there was low but
reproducible diastereoselectivity in the nitrogen transfer
process with 5–8% d.e. values, with the same relative
configuration being obtained at the new stereocentre in
each case.
Acknowledgements
We thank the EPSRC and Roche (CASE award to
MAA) for funding this work. We are grateful to Ian
Whitcombe (Roche) for performing the variable tem-
perature NMR studies.
References
1. Reviews: (a) Erdik, E.; Ay, M. Chem. Rev. 1989, 89,
1947–1980; (b) Greck, C.; Genet, J. P. Synlett 1997,
741–748; (c) Mulzer, J.; Altenbach, H.-J. In Organic
Synthesis Highlights; VCH: Weinheim, 1991; pp. 45–53;
(d) Dembach, P.; Seconi, G.; Ricci, A. Chem. Eur. J.
2000, 6, 1281; (e) Modern Amination Methods; Ricci, A.,
Ed.; Wiley-VCH: Weinheim, 2000.
2. Recent examples: (a) Evans, D. A.; Johnson, D. S. Org.
Lett. 1999, 1, 595–598; (b) Harris, J. M.; McDonald, R.;
Vederas, J. C. J. Chem. Soc., Perkin Trans. 1 1996,
2669–2674; (c) Svenstrup, N.; Bøgevig, A.; Hazell, R. G.;
Jørgensen, K. A. J. Chem. Soc., Perkin Trans. 1 1999,
1559–1565; (d) Oppolzer, W.; Tamuira, O.; Sunarababu,
G.; Signer, M. J. Am. Chem. Soc. 1992, 114, 5900.
3. Armstrong, A.; Atkin, M. A.; Swallow, S. Tetrahedron
Lett. 2000, 41, 2247–2251.
Importantly, it was established that the products were
not undergoing epimerisation under the reaction condi-
tions by submitting diastereomerically pure samples to
the basic reaction conditions; these were recovered
unchanged. The low degree of stereoselectivity in the
amination process could be related either to low facial
selectivity in the approach of the oxaziridine to the
enolate, or to the presence of a mixture of (E)- and (Z)-
enolates. Precedent for highly stereoselective enolate
formation under these conditions17 suggests that the
former explanation is more likely.
4. Vidal, J.; Drouin, J.; Collet, A. J. Chem. Soc., Chem.
Commun. 1991, 435–437.
5. Vidal, J.; Guy, L.; Sterin, S.; Collet, A. J. Org. Chem.
1993, 58, 4791–4793.

538
A. Armstrong et al. / Tetrahedron: Asymmetry 12 (2001) 535–538
6. Vidal, J.; Damestoy, S.; Collet, A. Tetrahedron. Lett.
J 8.1 Hz), 5.39 (1H, s, trans), 5.09 (1H, s, cis), 4.73 (1H,
td, J 10.9 and 4.5 Hz, trans), 4.40 (1H, td, J 10.9 and 4.5
Hz, cis), 2.07 (1H, m), 1.94 (1H, m), 1.72 (2H, m), 1.52
(1H, m), 1.09 (2H, m), 0.94 (3H, d, J 6.6 Hz), 0.91 (3H,
d, J 7.0 Hz), 0.86 (2H, m), 0.81 (3H, d, J 7.0); lC (67
MHz, CDCl3) 159.9 (C), 137.7 (C), 132.5 (2×CH), 128.7
(2×CH), 128.0 (C), 118.5 (C), 79.9 (CH), 77.2 (CH), 46.9
(CH), 40.4 (CH2), 33.9 (CH2), 31.4 (CH), 26.1 (CH), 23.3
(CH2), 21.9 (CH3), 20.6 (CH3), 16.2 (CH3); MS (FAB)
m/z 329 (M+H, 21%), 189 (M−menthyl, 85%); HRMS
observed 329.1870, C19H25N2O3 requires 329.1865. Calcd
for C19H25N2O3: C, 69.47; H, 7.37; N, 8.53. Found: C,
69.31; H, 7.36; N, 8.49%.
1995, 36, 1439–1442.
7. Vidal, J.; Damestoy, S.; Guy, L.; Hannachi, J.; Aubry,
A.; Collet, A. Chem. Eur. J. 1997, 3, 1691–1709.
8. Vidal, J.; Hannachi, J.-C.; Hourdin, G.; Mulatier, J.-C.;
Collet, A. Tetrahedron Lett. 1998, 39, 8845–8848.
9. Golobokov, Y. G.; Kasukhin, L. F. Tetrahedron 1992,
48, 1353.
10. CAUTION: to avoid the risk of explosion in large scale
preparations, it is advised NOT to concentrate the
CH2Cl2 solution of the intermediate azidoformate. On a
small scale, the azidoformate was concentrated in vacuo
to yield the crude product as an off-white solid, which
could be recrystallised from ethanol.
13. This barrier to ring inversion is consistent with those
reported previously (see Ref. 7).
14. We are grateful to Dr. A. J. Blake (School of Chemistry,
University of Nottingham) for this structure determina-
tion. Full details will be provided in a separate account of
this work.
15. The first example of a chiral, non-racemic N-alkoxycar-
bonyloxaziridine was reported recently: Page, P. C. B.;
Murrell, V. L.; Limousin, C.; Laffan, D. D. P.; Bethell,
D.; Slawin, A. M. Z.; Smith, T. A. D. J. Org. Chem.
2000, 65, 4204–4207.
16. Typical procedure for enolate aminations: as in Ref. 3.
17. Ireland, R. E.; Wipf, P.; Armstrong, III, J. D. J. Org.
Chem. 1991, 56, 650–657.
11. Purchased from Aldrich, or prepared from (−)-menthol
according to: Goren, Z.; Heeg, M. J.; Mobashery, S. J.
Org. Chem. 1991, 56, 7186–7188.
12. The procedure of Ref. 3 was employed using anhydrous
m-CPBA (1.6 equiv.) and n-BuLi (1.5 equiv.), except that
after addition of the imine, the reaction was stirred at
−78°C for 3 h, then brought to rt over 90 min before
quenching and work-up. Data for 4: formed as a white
solid (61% from aldehyde): [h]2D0 +39.5 (c 1.0, CHCl3); mp
55°C; wmax (CHCl3 solution)/cm−1 2960, 2931, 2872, 2233,
1772, 1754, 1494, 1387, 1319, 1081, 981; lH (400 MHz,
CDCl3), equilibrium mixture of trans and cis isomers in
90:10 ratio at 300 K: 7.73 (2H, d, J 8.1 Hz), 7.60 (2H, d,
.
.