Jacobsen protocols for large-scale epoxidation of cyclic dienyl sulfones: Application to the (+)-pretazettine core

Article abstract of DOI:10.1021/ol300996m

A Jacobsen epoxidation protocol using H₂O₂ as oxidant was designed for the large-scale preparation of various epoxy vinyl sulfones. A number of cocatalysts were screened, and pH control led to increased reaction rate, higher turnover

Full text of DOI:10.1021/ol300996m

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Jacobsen Protocols for Large-Scale
Epoxidation of Cyclic Dienyl Sulfones:
Application to the (þ)-Pretazettine Core
G. Reza Ebrahimian, Xavier Mollat du Jourdin, and Philip L. Fuchs*
Department of Chemistry, Purdue University, West Lafayette, Indiana 47906,
United States
pfuchs@purdue.edu
Received April 17, 2012
ABSTRACT
A Jacobsen epoxidation protocol using H2O2 as oxidant was designed for the large-scale preparation of various epoxy vinyl sulfones. A number of
cocatalysts were screened, and pH control led to increased reaction rate, higher turnover number, and improved reliability.
The formation of the CꢀO bond via epoxidation is one
of the most widely used reactions in the toolbox of the
synthetic organic chemist. Jacobsen and Katsuki further
expanded the synthetic utility of this transformation in the
Scheme 1. Hentemann and Myers/Torres epoxidations of
Dienyl Sulfones 1 and 2
early 1990s with the design of salen Mn(III) catalysts
offering access to a wide range of enantiomerically pure
epoxides.1 The nature of the oxidant (H2O2, NaOCl,
m-CPBA, urea hydrogen peroxide, PhIO)2 as well as
the presence of additives and axial ligands (carboxylates,
N-oxides)3 and the structure of the catalyst itself has been
studied, thus leading to numerous variants of the Jacobsen
protocol. In the last 15 years, Fuchs’ group further
extended the applicability of Jacobsen’s protocol to epox-
idation of dienyl sulfones in the course of the synthesis
of naturally occurring products such as apoptolidin,
aplyronine A, and (þ)-discodermolide. In the late
1990s, Hentemman demonstrated the synthetic utility of
Jacobsen’s asymmetric epoxidation (JAE) to obtain good
yield and high ee of enantiopure epoxy vinylsulfones 3
(1) (a) Jacobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J. R.; Deng,
L. J. Am. Chem. Soc. 1991, 113, 7063–7064. (b) Irie, R.; Noda, K.; Ito,
Y.; Matsumoto, N.; Katsuki, T. Tetrahedron Lett. 1990, 31, 7345–7348.
´
(2) (a) Jacobsen, E. N.; Deng, L.; Furukawa, Y.; Martınez, L. E.
and 5 (Scheme 1).4 Despite its high efficiency on small
scale, this method was not transferable to large scale.
Tetrahedron 1994, 50, 4323–4334. (b) Chang, S.; Galvin, J. M.; Jacobsen,
E. N. J. Am. Chem. Soc. 1994, 116, 6937–6938.
(3) Brandes, D. B.; Jacobsen, E. N. J. Org. Chem. 1994, 59, 4378.
(4) Hentemann, M.; Fuchs, P. L. Org. Lett. 1999, 1, 355.
r
10.1021/ol300996m
XXXX American Chemical Society

Table 1. Optimization of the Large-Scale Epoxidation of Dienyl Phenyl Sulfone 4
entry
scale (g)/cat. load (mol %)
additive (equiv)
n-Bu4NF (1)
oxidant (equiv)
time (h)
conversion (yield (%), ee (%))
1
2
0.2/1
0.2/1
6
6
6
6
6
6
6
4
4
4
4
4
4
17
17
17
2
traces
NH4F (1)
25
3
0.2/1
DMAP (1) or NH2C(S)NH2 (1)
Me3NO (1)
traces
4
0.2/1
50ꢀ60
traces
5
0.2/1
Ph3PO (1)
17
2
6
0.2/1
HMPA (1)
50ꢀ60
7
0.2 /1
0.2/1
Na3PO4 (1)
17
4.5
17
4.5
4.5
4.5
3d
10
8
Na3PO4 (0.4)/NH4BF4 (1.2)
Na2HPO4 (0.4)/NH4BF4 (1.2)
Na3PO4 (0.4)/NH4BF4 (1.2)
Na3PO4 (0.4)/NH4BF4 (1.2)
Na3PO4 (0.4)/NH4BF4 (1.2)
Na3PO4 (0.4)/NH4BF4 (1.2)
>97 (72,a 99b)c
NR
9
0.2/1
10
11
12
13
1/0.9
>95c
30/0.8
120/0.6
350/0.6
>95 (99)c
>95 (99)c
>95 (66e; 99)c
a The product was purified by flash chromatography. b The ee of the reaction was determined via chrial HPLC using a chiralpack AD column. c The
addition of H2O2 was performed dropwise and portionwise (1 equiv/30ꢀ40 min). d The reaction was run in MeOH (0.5 M), the concentration increase
afforded faster reaction. e the product was isolated by crystallization from MeOH (1 M) according to Torres’ procedure.
The reaction was usually carried out in methylene choride
with 5ꢀ15% of (salen)Mn(III) catalyst, 10ꢀ40 mol % of
4-(3-phenylpropyl)pyridine N-oxide (P3NO) as cocatalyst,
and buffered bleach as oxidant. The estimated first gen-
eration cost of of preparation of 1 mol of epoxide 3 was
>$6000 without factoring waste processing. Recently,
Torres used Pietikainen’s conditions5 and reported an
improved environmentally friendly large-scale preparation
of epoxy dienyl sulfone 5.6 The reaction was run for 12ꢀ
24 h in methanol with 1 mol % of catalyst, 1 equiv of
ammonium acetate, and 6 equiv of hydrogen peroxide.
Along withthe substantiallydecreasedcostassociatedwith
using 1% catalyst, the switch to MeOH made this new
protocol more environmentally friendly. Using this meth-
od, epoxy vinylsulfone 5 was obtained in 72% yield and
99% ee. Further optimization of this transformation
through screening of new additives led to a significant rate
increase at further decreased catalyst load. Using TBAF or
NH4F as sources of fluoride for chlorideꢀfluoride ex-
changemet with little success. The size ofthe countercation
was only slightly advantageous in the case of NH4F.
Switching to 4-(dimethylamino)pyridine (DMAP) or
thiourea did not accelerate the reaction under similar
conditions which employed Me3NO (Table 1, entry 4).
Phosphorus-centered species were screened next. How-
ever, Ph3P(O) had little effectpresumably for steric reasons
(Table 1, entry 5). Interestingly, HMPA dramatically
increased the rate of the reaction (Table 1, entry 6) but
did not meet environmental requirements. Significantly,
the use of Na3PO4 accelerated the reaction when NH4BF4
was also included to increase the solubility of the additive.
Itisbelieved thatthe accelerationof thereaction wasdue to
the basicity of the additive whichformedHOOꢀ in situand
accelerated the formation of the MndO complex. Filtra-
tion of the salts also improved purification of the product
on a large scale. Base-catalyzed epoxide opening being one
of the major pathways of decomposition of the product, it
was believed that using Na2HPO4 would afford higher
yield. However, no reaction and complete decomposition
of the catalyst were observed (Table 1, entry 9). Attempts
at decreasing the amount of Na3PO4 beyond 40 mol %
were unsuccessful, as lower reaction rate and decreased
conversions resulted. Finally, at higher scales (Table 1,
entries 10ꢀ12), the catalyst load was gradually decreased
to 0.6 mol % without a detrimental effect on the ee and the
reaction rate. However, portionwise (epoxidation on scale
larger than 10 g) and dropwise (epoxidation on 100 g scale)
additions were critical for the success of the reaction
(Table 1, entries 8ꢀ12). Finally, for concerns of scalability
(concentration scale larger than 100 g), addition of oxidant
to 0.5 M was found to have no visible effect on the outcome
of the reaction despite a noticeable increase in the
exotherm of the reaction. Direct crystallization of the
product from the methanolic reaction media according
to Torres’ protocol provided two crops of the desired
epoxide in 60ꢀ70% overall yield and 99% ee. The scope
of this new protocol was extended to various dienyl
sulfones such as the analogous 6-membered dienyl sulfone
2. Satisfactorily, this protocol was extended to this sub-
strate, although partial aromatization of the product to
diphenyl sulfone could not be avoided under these condi-
tions. Extension to dienyl triflate and its more stable
(5) Pietikainen, P. Tetrahedron 1998, 54, 4319.
(6) Park, T.; Torres, E.; Fuchs, P. L. Synthesis 2004, 1895.
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 2. Extension of the Na3PO4/NH4BF4 Protocol to Various Dienyl Sulfones and Dienyl Triflates
a For entries 1ꢀ8, (S,S)-1 was used, and for entries 9ꢀ12, (R,R)-1 was used. b Low yield is due to aromatization and formation of diphenyl sulfone.
nonaflate analogue was also successful and gave a good
yield and satisfactory ee. Finally, extension of this protocol
to more sterically hindered substrates was also attempted.
Although epoxidation leading to 11 under substrate-directed
conditions provided satisfactory results, it was studied
under Jacobsen conditions to test this new protocol.
Fortunately, epoxidation of 10 on the side opposite to
the neighboring ꢀMe and ꢀOTBS subsituents proved
successful. In practice, the desired epoxide was obtained
in 62ꢀ63% yield after 1.5 h at 0 °C with 1ꢀ2 mol % of
catalyst. Increased reaction rate as well as lower catalyst
load could be achieved under these new conditions. Con-
trary to epoxide 11, epoxidation to epoxide 13 constitutes a
more challenging transformation since epoxidation occurs
on the same side of the ring as the neighboring ꢀMe
substituent. Switching to our new modified protocol in-
creased the yield to75% and shortened the time of reaction
Although fast in the early stages of the reaction, epoxida-
tion of dienyl sulfone 12 could not be driven to com-
pletion and 75% yield based upon recovered starting
material resulted. This protocol was successfully used in
the synthesis of an advanced intermediate leading to
(þ)-pretazettine. Amaryllidaceae alkaloids havestimulated
the attention to the organic synthetic community due
to their complex structure and wide range of biological
activities.
Over 100 alkaloids of this family have been isolated
including (þ)-pretazettine (14).7 There have been only two
enantioselective total syntheses of (þ)-pretazettine to date:
one starts from chiral R-methyl-D-mannopyranoside,7g
and the other suffers from harsh reaction conditions and
moderate yield.7h,i It was envisioned that the new epoxida-
tion might enable a concise approach to the core skeleton
of (þ)- pretazettine 14 (Scheme 2). Using the above pro-
tocol (Table 2, entry 6), epoxide 7 was obtained with good
stereocontrol, high yield (86%), and low catalyst loading.
On a 10 g scale, this reaction furnished epoxide 7 in 70%
isolated yield. It is noteworthy that 7 is unstable in its pure
form and aromatizes within 30 min. Having the epoxide in
hand, the nitrogen stereocenter was installed after epoxide
(7) Previous syntheses: (a) Tsuda, Y.; et al. Tetrahedron Lett. 1972,
3153. (b) Danishefsky, S.; Morris, J.; Mullen, G.; Gammil, R. J. Am.
Chem. Soc. 1982, 104, 7591. (c) Martin, S. F.; Davidson, S. K. J. Am.
Chem. Soc. 1984, 106, 6431. (d) Martin, S. F.; Davidson, S. K.; Puckette,
T. A. J. Org. Chem. 1987, 52, 1962. (e) Abelman, M. M.; Overman, L. E.;
Tran, V. D. J. Am. Chem. Soc. 1990, 112, 6959. (f) White, J. D.; Chong,
W. K.-M.; Thirring, K. J. Org. Chem. 1983, 48, 2300. (g) Baldwin, S. W.;
Debenham, J. S. Org. Lett. 2000, 2, 99. (h) Nishimata, T.; Mori, M.
J. Org. Chem. 1998, 63, 7586. (i) Nishimata, T.; Sato, Y.; Mori, M. J. Org.
Chem. 2004, 69, 1837. (j) Rigby, J. H.; Cavezzo, A.; Hegg, M. J. J. Am.
Chem. Soc. 1998, 120, 3664.
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 2. Synthesis of (þ)-Pretazettine Core
opening with LiHMDS, ꢀOH-directed epoxidation of the
resulting dienol, TBS protection, and stereo- and regio-
controlled opening of the epoxide using tetramethylgua-
nidinium azide (TMGA), which afforded azide 20 in good
yield (>19:1 stereoselectivity at the N center). Reduction
of the azide followed by protection of the amine afforded
amine 21 in 82% yield. Finally, Suzuki coupling was
smoothly run with boronic acid 22 and provided advanced
intermediate 23 in excellent yield (90% as single dia-
stereomer). It is noteworthy to mention that Suzuki cou-
pling prior to azide reduction and nitrogen protection
diminished the overal yield of the reaction by 20% and
generated unidentifiable side products.
synthesis of apoptolidin, aplyronine A, (þ)-discodermo-
lide, and (þ)-pretazettine. A Na3PO4/NH4BF4 combina-
tion is used for the routine epoxidation of sterically less
demanding substrates on a 1ꢀ350 g scale. The advan-
tages of these protocols combine increased reaction rate,
higher turn over number, and significantly increased
reproducibility.
Acknowledgment. We thank Arlene Rothwell and
Dr. Karl Wood from Purdue University for MS data
and Purdue University for support of this research.
Supporting Information Available. Procedures, spectro-
scopic data, and spectra.This material is available free of
charge via the Internet at http://pubs.acs.org.
In conclusion, a new Jacobsen protocol has been
designed and optimized for the large scale (up to
300ꢀ400 g) asymmetric preparation of epoxy vinyl sul-
fones, key starting materials, and intermediates for the
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX