Highly regioselective and diastereoselective epoxidation of allylic amines with Oxone

Article abstract of DOI:10.1039/b503516c

Allylic amines (as their protonated ammonium salts) can be epoxidised with high syn diastereoselectivity and regioselectivity at the proximal alkene in substrates with several double bonds using Oxone. The protonated ammonium cation activates the Oxone by

Full text of DOI:10.1039/b503516c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Highly regioselective and diastereoselective epoxidation of allylic
amines with Oxone{
Varinder K. Aggarwal* and Guang Yu Fang
Received (in Cambridge, UK) 10th March 2005, Accepted 4th May 2005
First published as an Advance Article on the web 9th June 2005
DOI: 10.1039/b503516c
manner akin to the classical Henbest hydroxy-directed epoxidation
(Scheme 2).7
Allylic amines (as their protonated ammonium salts) can be
epoxidised with high syn diastereoselectivity and regioselectivity
at the proximal alkene in substrates with several double bonds
using Oxone. The protonated ammonium cation activates the
Oxone by hydrogen bonding, thus promoting the oxidation of
groups within the vicinity of the complex.
We recently reported a novel amine-catalysed epoxidation of
alkenes using Oxone.1,2 Using chiral amines, up to 66% ee was
achieved, demonstrating that the amine was intimately involved in
the epoxidation process. We proposed that the ammonium salt
(the active catalyst rather than the amine) had two distinct roles: (i)
it acted as a phase transfer catalyst bringing the Oxone into
solution and (ii) it activated Oxone as an electrophilic oxidant by
hydrogen bonding (Scheme 1).2 Our mechanism has now been
generally accepted by the scientific community.3
Scheme 2 Model for epoxidation of a protonated allylic ammonium salt
with m-CPBA.
We believed that if our proposed mechanism is correct, Oxone
should also be directed and activated by a protonated ammonium
ion, leading to high diastereoselectivity. We therefore tested the
ammonium salt 18 in our epoxidation process using Oxone and
observed complete diastereoselectivity (.95 : 5) in favour of the
syn isomer 2. The stereochemistry was proved by X-ray analysis{
of the corresponding tosylamide 3 (Scheme 3).
It occurred to us that this process could be applied to the
controlled epoxidation of protonated allylic ammonium salts
where issues of regioselectivity and stereoselectivity remain.
Protonated allylic ammonium salts are the simplest protected
forms of allylic amines that are compatible with strong oxidative
reagents. Unfortunately the ammonium salt strongly deactivates
the proximal alkene towards electrophilic attack and reaction times
can be very long.4a However, oxidants capable of hydrogen
bonding to the protonated ammonium salt like meta-chloroper-
oxybenzoic acid (m-CPBA) or even dimethyldioxirane (DMDO)
are rendered more electrophilic and so are activated, resulting in
diastereoselective epoxidation syn to the ammonium ion4–6 in a
Scheme 3 Epoxidation of ammonium salt 1 with Oxone. Reagents and
conditions: (i) 2 equiv. Oxone, 10 equiv. NaHCO3, 0.5 equiv. pyridine,
MeCN/H2O (95:5), 28 uC, 73%, .95:5 dr; (ii) TsCl/TEA, DMAP, 87%.
We also examined the quaternary ammonium salt 4,8 but this
failed to give any epoxide. This clearly demonstrates that the ionic
interaction between the peroxymonosulfate anion and the
ammonium cation is not sufficient alone to promote epoxidation;
epoxidation clearly requiring a protonated ammonium ion. This is
Scheme 1 The ammonium salt catalysed epoxidation of 1-phenylcyclo-
hexene and proposed working model.
{ Electronic Supplementary Information (ESI) available: Experimental
details and spectral data for synthesis and characterisation of compounds
1–8. See http://www.rsc.org/suppdata/cc/b5/b503516c/
*V.Aggarwal@bristol.ac.uk
Fig. 1 Factors affecting regioselective epoxidation of geranylammonium
tosylate 5 with Oxone.
3448 | Chem. Commun., 2005, 3448–3450
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Scheme 4 Epoxidation of geranylammonium tosylate 5 with m-CPBA and Oxone.
in keeping with our mechanism in which a protonated amine
activates by hydrogen bonding the peroxymonosulfate anion (it is
now a better electrophile) towards nucleophilic attack by the
alkene.
We then tested a more challenging substrate, geranylammonium
tosylate 5.9 This substrate has two alkenes: a remote trisubstituted,
electron rich alkene, and an alkene that is strongly deactivated
by the ammonium cation. We questioned whether the directing/
activating effects of the protonated ammonium ion with
Oxone could overcome the inherent reduced nucleophilicity of
the alkene, and so control the regiochemistry of epoxidation
(Fig. 1).
We tested both m-CPBA10 and Oxone in the epoxidation
process. Epoxidation of the geranylammonium salt with m-CPBA
gave a mixture of monoepoxides 6 and 7, diepoxide 8, and starting
material. This indicates that the directing effect of the ammonium
cation, which promotes epoxidation of the proximal alkene, was
effectively cancelled out by the deactivation caused by the strong
electron withdrawing group, resulting in similar rates of epoxida-
tion for the two alkenes (Scheme 4). In contrast using Oxone, a
much cleaner reaction was observed, and the only monoepoxida-
tion product isolated was the epoxyamine 6. This clearly
demonstrates that the protonated ammonium ion both
activated and directed Oxone epoxidation to take place at the
alkene proximal to it. The degree of activation of the oxidant
by the ammonium ion (relative to the power of the free
oxidant) is clearly greater for Oxone than m-CPBA as it
completely outweighed the reduced nucleophilicity of the alkene
(Scheme 4).
Scheme 5 Synthesis of ammonium salts
¯
conditions: (i) MeLi/CeCl3, THF, 278 oˆC to r.t. 78%; (ii) KH, Et2O,
¯
NCCCl3, 25 oˆC to r.t, 83%; (iii) 6 N NaOH, EtOH, 81%; (iv)
1
and 4. Reagents and
HCHO, NaBH3CN, 95%; (v) MeI/Et2O, 98%; (vi) TsOH, THF, 89%;
(vii) AgBF4/H2O, 89%.
Notes and references
{ Crystal data for compound 3: C14H19NO3S, M 5 281.1, triclinic, space
¯
group P1,
a
5 7.1014(6), b 5 8.6018(7),
c
5
12.6772(11) s,
a
5
101.5860(10),
b
5
98.6520(10),
c
5
109.5870(10)u, V 5
694.53(10) s3, Z 5 2, T 5 173(2), Dc 5 1.345 Mg m23, crystal dimensions
0.25 6 0.10 6 0.10 mm, Mo-Ka radiation, l 5 0.71073 s. Data were
collected on a Bruker Smart CCD area-detector diffractometer and a total
of 3146 of the 7277 reflections were unique (Rint 5 0.0247). Refinement on
F2, wR2 5 0.1051 (observed reflections), R1 5 0.0496 [I . 2s(I)]. CCDC
266382. See http://www.rsc.org/suppdata/cc/b5/b503516c/ for crystallo-
graphic data in CIF or other electronic format.
Oxone and m-CPBA show similarly high levels of diastereos-
electivity in the epoxidation of cyclic, allylic ammonium salts. As
Oxone is the safer and less hazardous reagent of the two, especially
on a large scale, we would expect it to be the reagent of choice for
such substrates. In reactions of allylic ammonium salts
containing multiple double bonds it is far superior, delivering
high levels of regiocontrol due to its strong activation by hydrogen
bonding.
1 M. F. A. Adamo, V. K. Aggarwal and M. A. Sage, J. Am. Chem. Soc.,
2000, 122, 8317; M. F. A. Adamo, V. K. Aggarwal and M. A. Sage,
J. Am. Chem. Soc., 2002, 124, 11223.
2 V. K. Aggarwal, C. Lopin and F. Sandrinelli, J. Am. Chem. Soc., 2003,
125, 7596.
3 A. Armstrong, Angew. Chem., Int. Ed., 2004, 43, 1460; C.-Y. Ho,
Y.-C. Chen, M.-K. Wong and D. Yang, J. Org. Chem., 2005, 70,
898.
4 (a) G. Asensio, R. Mello, C. Boix-Bernardini, M. E. Gonza´lez-Nu´n˜ez
and G. Castellano, J. Org. Chem., 1995, 60, 3692; (b) G. Asensio,
R. Mello, C. Boix-Bernardini, C. Andreu, M. E. Gonza´lez-Nu´n˜ez,
R. Mello, J. O. Edwards and G. B. Carpenter, J. Org. Chem., 1999, 64,
4705.
This work reinforces the proposed mechanism of the ammo-
nium salts in that they not only act as directing groups, but also as
activators of Oxone by hydrogen bonding. In addition, the work
fills a methodology gap by providing a method for controlling the
regiochemistry of epoxidation in allylic amines.
5 A. S. Edward, R. A. J. Wybrow, C. Johnstone, H. Adams and J. P.
R. Harrity, Chem. Commun., 2002, 1542.
We thank the EPSRC for their support of this work and
Chrystel Lopin for developing the synthesis of 1.
6 For a review of substrate directable reactions see: A. H. Hoveyda,
D. A. Evans and G. C. Fu, Chem. Rev., 1993, 93, 1307. For a recent
example of diastereoselective, amide-directed epoxidation see: P. O’Brien,
A. C. Childs, G. J. Ensor, C. L. Hill, J. P. Kirby, M. J. Dearden,
S. J. Oxenford and C. M. Rosser, Org. Lett., 2003, 5, 4955 and
references therein. For examples of ketone directed epoxidation see:
Varinder K. Aggarwal* and Guang Yu Fang
Department of Chemistry, University of Bristol, Cantock’s Close,
Bristol, UK BS8 1TS. E-mail: V.Aggarwal@bristol.ac.uk;
Fax: +44 (0)117 929 8611; Tel: +44 (0)117 9546319
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 3448–3450 | 3449

View Article Online
A. Armstrong, P. A. Barsanti, P. A. Clarke and A. Wood, J. Chem.
Soc., Perkin Trans. 1, 1996, 1373. For hydroxy-directed epoxidation
using sulfonic peracids see: R. Kluge, M. Schulz and S. Liebsch,
Tetrahedron, 1996, 52, 2957.
7 H. B. Henbest and R. A. L. Wilson, J. Chem. Soc., 1959, 1958;
H. B. Henbest, Proc. Chem. Soc., 1963, 159.
8 The ammonium salts 1 and 4 were made from 2-cyclohexen-1-one
(Scheme 5). Experimental details and related analytical data are included
in the Electronic Supplementary Information.
9 Geraniol has often been employed to examine the power of hydroxy-
promoted epoxidation. The remote 6,7-alkene is more electron rich than
the 2,3-alkene and the ratio of regioisomeric products gives a measure of
the relative rates of epoxidation. See: W. Adam and A. K. Smerz, Bull.
Soc. Chim. Belg., 1996, 105, 581.
10 According to the proposed mechanism for the epoxidation of
protonated allylic amines with m-CPBA (see ref. 4 and 5), the hydrogen
bonding interaction between m-CPBA and protonated amino group
should also enhance its reactivity.
3450 | Chem. Commun., 2005, 3448–3450
This journal is ß The Royal Society of Chemistry 2005