Epoxidation of alkenes by amine catalyst precursors: Implication of aminium ion and radical cation intermediates [10]

Full text of DOI:10.1021/ja0004433

J. Am. Chem. Soc. 2000, 122, 8317-8318
8317
Scheme 1. Epoxidation of 1-Phenylcyclohexene by Oxone/
NaHCO3
Epoxidation of Alkenes by Amine Catalyst
Precursors: Implication of Aminium Ion and Radical
Cation Intermediates
Mauro F. A. Adamo, Varinder K. Aggarwal,* and
Matthew A. Sage
Table 1. Epoxidation of 1-Phenylcyclohexenea
Department of Chemistry, UniVersity of Sheffield
conversionb,c epoxidec
diolc
Brook Hill, Sheffield, S3 7HF, UK
entry
amine
(%)
yield (%) yield (%)
1
2
3
4
5
6
7
8
9
10
11
12d
13d
none
2
0
40
16
2
100
90
60
65
0
0
0
36
13
0
90
87
58
62
0
2
0
4
2
2
10
3
2
3
0
ReceiVed February 7, 2000
EtNH2
Et2NH
Et3N
NEt4OAc
pyrrolidine
piperidine
morpholine
N-methylpyrrolidine
EDTA
pyrrolidinone
pyrrolidine 10%
pyrrolidine 10% +
Py (0.5eq)
pyrrolidine 5%
pyrrolidine 5% +
Py (0.5 equiv)
pyridine
We report herein the discovery of a novel process for
epoxidation of alkenes, using Oxone, which, remarkably, is
catalyzed by simple amines. This process, which does not rely
on transition metal catalysts, utilizes Oxone/NaHCO3 as the
oxidant and simple, cheap, and readily available amines (see Table
1) as catalyst precursors for alkene epoxidation. The standard
reaction procedure is illustrated in Scheme 1.
Normally, Oxone buffered with NaHCO3 in MeCN:H2O
epoxidizes unfunctionalized alkenes only when oxygen-transfer
reagents such as ketones,1 imines,2 or iminium salts,3 are present.
Aqueous solutions of Oxone at approximately neutral pH are also
known to oxidize alkenes to give epoxides but in variable yield.4,5
At lower pH, mixtures of epoxides and diols are obtained.4
During our studies on iminium salt-catalyzed epoxidations of
alkenes3b we discovered that simple amines were also capable of
epoxidizing our test substrate, 1-phenylcyclohexene and so we
tested a broad range of amines (Table 1). In a control experiment,
alkene and oxidant were combined in the absence of amine, and
essentially no epoxide was obtained (entry 1). Primary amines
were not effective (entry 2), but in the presence of secondary
and tertiary amines (entries 3, 4, 6-9) rapid epoxidation ensued.
Secondary amines gave the highest yields, and within this class,
pyrrolidine (entry 6) was optimum. 1,2-Diamines (entry 10) and
amides (entry 11) were not effective catalysts. We therefore tested
pyrrolidine at loadings of 10 and 5 mol % and discovered that
good levels of turnover could be achieved (entries 12-15).
Unfortunately, with lower amine loadings (entries 12,14) a
significant amount of hydrolysis of the epoxide occurred despite
the fact that an excess of NaHCO3 was present. Although less
hydrolysis could be achieved by increasing the ratio of water
present, this also resulted in reduced conversion.6 Attempts to
increase the pH of the media with other inorganic bases were
unsuccessful but the use of 0.5 equiv of pyridine did largely
suppress epoxide hydrolysis7 (entries 13 and 15). A control
experiment showed that pyridine itself was not able to catalyze
the epoxidation process (entry 16). Pyrrolidine and a chiral
derivative [(S)-2-(diphenylmethyl)pyrrolidine-18] were tested as
catalysts at just 5 mol % loading9 with a range of alkenes (Table
2
100
100
2
60
95
0
40
4
14d
15d
65
59
35
56
30
3
16d
1
1
0
a Unless otherwise stated all reactions conducted with 1-phenylcy-
clohexene (0.125 mmol), 1 equiv amine, 2 equiv Oxone, 10 equiv
NaHCO3, 0.5 mL d-MeCN:D2O (95:5), 2 h. b Remainder is alkene.
c Yields determined by 1H NMR relative to an internal standard
(nitrobenzene). d Concentration of 1-phenylcyclohexene increased to
0.85 mol/L.
2) and it was found that good yields of epoxides could be obtained
in most cases although the reaction was sensitive to both the
structure of the alkene and the amine. The substituted pyrrolidine
1 was a more effective catalyst for a broader range of alkenes
than pyrrolidine itself (compare entries 5-7, and 9) and gave up
to 57% enantioselectivity with 1-phenylcyclohexene (entry 5).
Amine 1 was an effective catalyst for most alkenes; only stilbenes,
aliphatic disubstituted and terminal alkenes gave low yields.
We have carried out competition experiments with structurally
similar alkenes, and found that the amine-catalyzed epoxidation
reactions showed much greater selectivity compared to MTO-
catalyzed epoxidations10 (Table 3).
The mechanism of the reaction is intriguing. We found that in
the presence of Oxone (1 equiv) pyrrolidine was oxidized11 to
the corresponding hydroxylamine (10%), nitrone (60%), and
N-hydroxylactam (2-3%), but none of these oxidation products
either transferred their oxygen or acted as catalysts for epoxida-
tion. Tertiary amine N-oxides were similarly inactive catalysts.
However, the fact that asymmetric induction is observed means
that the amine is intimately involved in the oxygen transfer
process, but not simply as phase-transfer catalyst as quaternary
ammonium salts were also inactive (Table 1, entry 5). We have
considered the possibility that a single electron-transfer process
may be involved (Scheme 2). It is possible that the amine is
oxidized to its radical cation12,13 which in turn oxidizes the alkene
(1) (a) Wang, Z. X.; Tu, Y.; Frohn, M.; Zhang, J. R.; Shi, Y. J. Am. Chem.
Soc. 1997, 119, 11224-11235. (b) For a review, see: Denmark, S. E.; Wu,
Z. Synlett 1999, 847-849.
(2) Davis, F. A.; Reddy, R. T.; Han, W.; Reddy, R. E. Pure Appl. Chem.
1993, 65, 633-640.
(3) (a) Lusinchi, X.; Hanquet, G. Tetrahedron 1997, 53, 13727-13738. (b)
Aggarwal, V. K.; Wang, M. J. Chem. Soc., Chem. Commun. 1996, 191-192.
(c) Page, P. C. B.; Rassias, G. A.; Bethell, D.; Schilling, M. B. J. Org. Chem.
1998, 63, 2774-2777. (d) Armstrong, A.; Ahmed, G.; Garnett, I.; Goacolou,
K.; Wailes, J. S. Tetrahedron, 1999, 55, 2341-2352.
(4) Zhu, W. M.; Ford, W. T. J. Org. Chem. 1991, 56, 7022-7026.
(5) Zheng, T. C.; Richardson, D. E. Tetrahedron Lett. 1995, 36, 833-
836.
(9) We have not been able to reisolate the amine; it is ultimately oxidised
by the excess Oxone. Tertiary amines are oxidized more rapidly than secondary
amines which may account for their poor catalytic activity.
(10) (a) Ruldolph, J.; Reddy, L.; Chiang, J. P.; Sharpless, K. B. J. Am.
Chem. Soc. 1997, 119, 6189-6190 and references therein.
(11) Details are presented in the Supporting Information. Oxone is known
to oxidise amines in the presence of acetone (Neset, S. M.; Benneche, T.;
Undheim, K. Acta. Chem. Scan. 1993, 47, 1141-1143). However, we are
not aware of any report on the direct oxidation of amines with Oxone.
(6) For details see Supporting Information. It should be noted that small
variation in results have been experienced with different batches of Oxone,
as noted by others.1a
(7) Pyridine has also been used to suppress hydrolysis in MTO-catalysed
epoxidation.10
(8) Bailey, D. J.; O’Hagan, D.; Tavasli, M. Tetrahedron: Asymmetry 1997,
8, 149-153.
10.1021/ja0004433 CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/11/2000

8318 J. Am. Chem. Soc., Vol. 122, No. 34, 2000
Communications to the Editor
Table 2. Alkene Epoxidation Using 5 Mol % of Either Pyrrolidine or Amine 1a
pyrrolidine catalyst
(S)-2-diphenylmethylpyrrolidine-1 catalyst
entry
alkene
conv.c %
epoxided
diold
conv.c %
epoxided (eee%)
diold
1
2
3
4
5
6
7
8
9
10
11
12
13
methylenecyclohexene
1-methylcyclohexeneb
octahydronaphthalene
indeneb
1-phenylcyclohexene
R-methylstyreneb
â-methylstyrene
norbornylene
33
100f
100f
29
59f
77f
15
77
30
0
29
90
92
19
56
38
15
74
27
0
4
10
8
10
3
38
0
2
2
0
0
0
33
100f
100f
31
100f
88f
21
80
100
0
29
4
10
7
11
4
23
0
6
6
0
0
0
90 (15 R,S)
93
20 (25)
96 (57 S,S)
65 (15 S)
21 (13 S,S)
74
93 (9 S)
styrene
octa-4-ene
0
trans-stilbene
2
6
0
2
6
0
6
9
0
6
cis-stilbeneg
9h
0
dec-1-ene
0
0
a Alkene (0.424 mmol), 5 mol % amine, 2 equiv Oxone, 10 equiv NaHCO3, 0.5 mL d-MeCN:D2O (95:5), 0.5 equiv pyridine, 4 h, rt. b 1 equiv
1
pyridine. c Remainder is alkene. d Yields determined by H NMR relative to an internal standard (nitrobenzene). See Supporting Information for
isolated yields. e Enantioselectivities determined by chiral HPLC using a Chiracel OD column and correlated with literature data.1a f Conversion
after 2 h. g The same results are obtained in the absence of amine. h 6:1 ratio of cis:trans epoxides.
Table 3. Competition Experiments with Structurally Similar
Scheme 3. Epoxidation of Triene 2
Alkenes and Comparison of Different Oxidation Proceduresa
We have circumstantial evidence for the intermediacy of radical
cations. (i) In competition experiments similar selectivity is
observed to Bauld’s aminium-catalyzed reactions16 which have
been shown to occur via radical cations (Table 3). (ii) Triene 2
gave a single mono epoxide, uncontaminated by other regioiso-
mers (Scheme 3). The same selectivity has been observed in
aminium ion-catalyzed epoxidation17 and cyclopropanation of the
same substrate.18 (iii) In competition experiments trans-stilbene
reacted ∼2.5 times faster than cis-stilbene19 (similar observations
for radical cations20), whereas in electrophilic oxidations the cis
isomer is more reactive. (iv) Most substrates that are inert to
Bauld’s radical cation-mediated oxidations17 are also inert in our
system (Table 2, entries 10, 13).21
a 1 equiv of each alkene is used in all experiments. 1H NMR
conversion of the two alkenes A:B are given relative to an internal
standard (nitrobenzene). b Alkene (each 0.424 mmol); amine (0.04
mmol), Oxone (0.85 mmol), NaHCO3 (4.24 mmol) 0.5 mL of d-MeCN:
D2O (95:5), pyridine (0.424 mmol), 4 h, rt. c 20 mol % Ar3N+•SbCl6
,
-
3 equiv SeO2, d-CH2Cl2, rt, 5 min. d 0.5 mol % MTO (methyltrioxo-
rhenium), 12 mol % pyridine, 1.5 equiv H2O2, d-CH2Cl2, 6 h.
Scheme 2. Proposed Catalytic Cycle for Alkene Epoxidation
In summary, we have discovered a novel process for oxidation
of alkenes which is catalyzed by simple amines. It is believed
that the reactions are mediated by chiral amine radical cations,
and this is the first report of such a species. The plethora of readily
available chiral amines makes the likelihood of finding highly
enantioselective catalysts a real possibility. Efforts in this area
are in progress.
Acknowledgment. We thank EPSRC, DTI, and the LINK scheme in
Asymmetric Synthesis [contributions from Pfizer (P. Spargo), SB (D.
Bell), GlaxoWellcome (A. Walker), and Chiroscience (M. Burk)] for
support.
to the radical cation.14 In the presence of suitable oxidants alkene
radical cations are converted to epoxides,15-17 in what is often a
diffusion-controlled process. In our case the amine must still be
associated (complexed) with the alkene radical cation at the time
of the epoxidation process for asymmetric induction to be
observed.
Supporting Information Available: General procedures for alkene
epoxidation, the effect of water, pyridine, and base on yields of epoxide
and diol; tables of isolated yields of epoxides and results obtained using
1 mol % of amine 1 (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
JA0004433
(12) 1O2 oxidises amines to the corresponding radical cation (Ferroud, C.;
Rool, P.; Santamaria, J. Tetrahedron Lett. 1998, 39, 9423-9426 and references
therein).
(17) Kim, T.; Mirafzal, G. A.; Liu, J.; Bauld, N. L. J. Am. Chem. Soc.
1993, 115, 7653-7664.
(18) Mirafzal, G. A.; Lozeva, A. M.; Oslon, J. A. Tetrahedron Lett. 1998,
39, 9323-9326.
(13) 1O2 is produced during auto-decomposition of Oxone (Evans, D. F.;
Upton, M. W. J. Chem. Soc., Dalton Trans. 1985, 6, 1151-1153). We believe
that only a small amount of 1O2 is generated from auto-decomposition of Oxone
which leads to a small amount of the amine radical cation. Ample Oxone
remains for alkene epoxidation.
(19) These experiments were done at 10 mol % loading of amine 1 with
limited amount of oxidant (0.5 equiv) and ratios determined by 1H NMR after
1 h at 0.21 M in each alkene (12% conversion for trans-stilbene). It is important
to ensure that both alkenes are completely dissolved at the start.
(20) Bauld, N. L.; Yeuh, W. J. Am. Chem. Soc. 1994, 116, 8845-8846.
(21) A few substrates do not show the same reactivity as Bauld’s: stilbene
was expected to be a good substrate but is rather poor, and norbornene was
expected to be a poor substrate but is in fact good. We do not have a good
explanation for this at present.
(14) Aminium radical cations readily react with alkenes to give alkene
radical cations (Mirafzal, G. A.; Liu, J.; Bauld, N. L. J. Am. Chem. Soc. 1993,
115, 6072-6077 and references therein).
(15) Bosch, E.; Kochi, J. K. J. Am. Chem. Soc. 1996, 118, 1319-1329.
(16) Bauld, N.; Mirafzal, G. A. J. Am. Chem. Soc. 1991, 113, 3613-3614.