Sultones and Sultines via a Julia-Kocienski Reaction of Epoxides

Article abstract of DOI:10.1002/anie.201508467

The development of the homologous Julia-Kocienski reaction has led to the discovery of two new reaction modes of epoxides with sulfones. These pathways allow rapid and direct access to a range of γ-sultones and γ-sultines.

Full text of DOI:10.1002/anie.201508467

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201508467
German Edition: DOI: 10.1002/ange.201508467
Synthetic Methods
Sultones and Sultines via a Julia–Kocienski Reaction of Epoxides
Geoffrey M. T. Smith, Paul M. Burton, and Christopher D. Bray*
Abstract: The development of the homologous Julia–Kocien-
ski reaction has led to the discovery of two new reaction modes
of epoxides with sulfones. These pathways allow rapid and
direct access to a range of g-sultones and g-sultines.
Despite their usefulness, synthesis of these fundamental
heterocycles is surprisingly difficult, especially given how long
they have been known. There are some notable routes to b-
and d-sultones,[14] but existing routes to g-sultones and to
sultines in general are invariably lengthy and low yielding,[12]
produce racemates,[15] utilize chiral auxiliaries[16] or are based
on chiral pool approaches.[13a,17] Herein, we present two new
reaction modes of epoxides, that leads either to g-sultones or
to g-sultines in one or two steps respectively (Figure 2).[18]
S
ulfur containing heterocycles play a major role in the
pharmaceutical, agrochemical, materials and petrochemical
industries. First made over 125 years ago,[1] sultones and
sultines (the thia-analogues of lactones) (Figure 1) are among
Figure 1. Reactivity profile and selected uses of sultones and sultines.
the oldest known sulfur heterocycles.[2] Sultines react akin to
their carbocyclic cousins and undergo nucleophilic substitu-
tion at sulfur. Sultines are used as lactone bioisosteres,[3] occur
as natural products,[4] find use in the perfume industry[5] and
are characterized as one of the distinctive olfactants of
Sauternes wines.[6] In contrast, ring-opening of sultones occurs
Figure 2. Current leading routes to sultones and sultines versus our
work.
The Julia–Kocienski reaction of aldehydes and ketones is
one of the preeminent methods for the stereocontrolled
synthesis of alkenes.[19] Related to our interest in three-
membered rings,[20] we wondered whether a homologous
variant of this reaction might be developed. Epoxides were
the first electrophiles considered since they are readily
available as single enantiomers via a multitude of methods.[21]
It was thought that a Julia–Kocienski sulfone for example, 1a
(Scheme 1) could ring-open an epoxide (2) to give a g-
alkoxysulfone 3. Anion-relay[22] (Smiles rearrangement)[23] of
À
À
with cleavage of the C O, rather than S O bond. Sultones act
as sulfoxylating agents[7] and are important in a wide variety
of fields including soap manufacture, drug discovery,[8]
polymer modification,[9] imaging[10] and energy storage.[11]
A
myriad of synthetic methodologies involve sultones and
sultines as key intermediates[12] and they have been used in
many total syntheses.[13]
[*] Dr. G. M. T. Smith, Dr. C. D. Bray
Department of Chemistry, Queen Mary University of London
Mile End Road, London, E1 4NS (UK)
E-mail: c.bray@qmul.ac.uk
Dr. P. M. Burton
Syngenta, Jealott’s Hill International Research Centre
Bracknell, Berkshire, RG42 6EY (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201508467.
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co.
KGaA. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly
cited.
Scheme 1. A new reaction mode of epoxides with sulfones to give g-
sultones.
15236
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 15236 –15240

Angewandte
Chemie
3 would give a sulfinate for example, 6, whose fate was unclear
and demonstrate the functional group tolerance of this
but by analogy with the known phosphorus chemistry[24] we
process. Curiously, styrene oxide simply returned the starting
materials. We next examined disubstituted epoxides; cis-1,2-
epoxybutane gave solely trans-sultone 4o whereas trans-1,2-
epoxybutane, gave solely the cis-sultone 4p, clear evidence
for the proposed reaction pathway involving a single inver-
sion of stereochemistry and revealing increased steric bulk on
the epoxide could be tolerated. In view of the widespread
interest from discovery chemists in spirocycles[29] we extended
this study to 1,1-dialkyl-substituted epoxides, and pleasingly,
we were able to synthesize the sultones 4q–v.
[25]
anticipated loss of SO2
and cyclization to create a new
general method for cyclopropane synthesis.
Evaluation of a series of epoxide, sulfone, base, temper-
ature, solvent and Lewis-acid combinations ultimately led us
to react TBTSO2Me[26] 1a and (R)-propylene oxide 2a (R =
Me) with LiN(SiMe3)2 as base at room temperature. How-
ever, rather than a cyclopropane, this gave the g-sultone (R)-
4a (R = Me) in > 99% yield and ee.[27] This indicates the
initially formed g-alkoxysulfone 3 (R = Me) did not undergo
the expected anion-relay, but instead directly cyclized on
sulfur with loss of the tetrazolide as a leaving group.[28] As
a result, the stereochemistry of the original epoxide is
completely retained in the g-sultone product. An operation-
ally simple one-step synthesis of g-sultones is therefore
achieved. A range of epoxides were then examined to
determine the scope of this new process (Table 1). A variety
of alkyl substituted (enantiopure)[21] terminal epoxides
reacted with 1a to give the g-sultones 4b–i in good to
excellent yields. Substrates bearing a protected alcohol,
amine and ketone as well as containing a halogen were
converted to the corresponding g-sultones 4j–m. A bis-
epoxide was examined as a substrate, but even with five equiv
of base/sulfone only the mono-sultonylated product 4n was
obtained, which suggests that formation of the first sultone
ring retards that of the second. The g-sultones 4 f–n are of
interest since they could allow for further functionalization
Our inability to form sultones from arylepoxides was
intriguing. We therefore synthesized the g-hydroxysulfone 5a
(Scheme 2) by BF3-mediated ring-opening of (S)-styrene
oxide with sulfone 1a with KN(SiMe3)2 as base (76%
Scheme 2. An alternative reaction pathway to give g-sultines.
yield).[27] This intermediate was treated with a variety of
bases. Upon treatment with DBU, the product was neither
a cyclopropane nor even a sultone, but instead the g-sultines
7a (trans:cis 76:24) in 76% yield and > 99% ee.[27] This
indicates a second reaction pathway where anion-relay had
occurred but, rather than loss of SO2, the sulfinate 6a had
directly cyclized through oxygen. This is unusual since
sulfinates primarily alkylate on sulfur and rarely on
oxygen.[30] No g-aryl-g-sultone products were ever detected
from this reaction but the g-sultines 7a could be oxidized (to
4w) or alternatively photolysed to give cyclopropylben-
zene,[31] one of the products originally mooted by us for this
process. In this instance, the stereocentre of the epoxide has
been inverted and this pathway is stereodivergent from the
one that produces g-sultones.[32]
Table 1: One-step synthesis of g-sultones from epoxides.
To uncover further examples of g-sultine formation using
the homologous Julia-Kocienski reaction we examined the
reactions of other g-hydroxysulfones 5 with DBU (Table 2). g-
Sultine products 7b were isolated from the 2-naphthyloxirane
derived substrate 5b. In a similar manner, the vinyl sub-
stituted g-sultines 7c could be observed through in situ NMR
monitoring, though they could not be isolated. The unstable
sulfinic acid 8 (entry 4) which results from 5d via anion-relay/
protonation could also be observed but even under forcing
conditions it did not cyclize to a g-sultine.
The two reaction pathways that produce g-sultones and g-
sultines proceed via similar g-alkoxysulfone intermediates
(3). Collectively, the results thus far suggest that when these
have been formed using a lithium amide base and are
[a] 1 Equiv of base/sulfone unless otherwise stated. [b] 3 Equiv of base/
sulfone.
Angew. Chem. Int. Ed. 2015, 54, 15236 –15240
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 15237

.
Angewandte
Communications
Table 2: Synthesis of g-sultines through anion-relay/cyclization.
situations. Treatment of this com-
pound with DBU in MeCN gave an
inseparable mixture of the sultines
7e alongside traces of the sultone
4a (Table 2, entry 5). In contrast
treatment of 5e with LiN(SiMe3)2
gave solely the sultone 4a (68%
isolated yield). This demonstrates
that the two reaction pathways can
compete, however use of Et-, iPr- or
tBu-substituted g-hydroxysulfones 5
led exclusively to the sultones
4b,d,e in > 90% yield in each case
irrespective of the base used. This
reveals how delicately balanced the
stereoelectronics are for the two
pathways. We next reasoned that
despite the Thorpe–Ingold effect,
increased substitution adjacent to
sulfur would disfavor direct ring-
closure as it would be pseudo neo-
pentyl and might lead exclusively to
g-sultine formation. This was
indeed found to be the case and
the g-alkoxysulfone 5 f derived
from ring-opening of propylene
Entry g-Hydroxysulfone
Product(s)
Yield
[%][a]
1
5a
7a (76:24)
76
2
3
5b
5c
7b (2:1)
7c (1:1)
51
>99[b]
4
5d
8
>99[b]
i
oxide with TBTSO2 Pr 1b gave
clean, albeit slow, conversion to g-
5
6
7
8
5e
5 f
7e (69:25:6)
7 f (1:1)
>99[b,c]
sultine products on treatment with
1
DBU (as judged by H NMR spec-
21
troscopy), though the labile sultines
7 f were only isolated in 21% yield.
Cyclization of the styrene oxide
derived g-alkoxysulfones 5g and
5h proceeded with greater ease
indicating that cyclization at a ben-
zylic position was more favourable.
Both diastereomers of 7g were
suitable for X-ray crystallographic
analysis. Of note was the fact that in
5g
5h
7g (17:83)
7h (23:77)
47
45
9
10
5i, R=Me
5j, R=Et
7i
7j
65
62
=
each case, the S O bonds were
oriented pseudo-axial,[33] forcing
one of the methyl substituents into
a seemingly unfavorable position.
This unusual observation can be
rationalized by the presence of an
anomeric-like effect, a phenomenon
which has been proposed previously
[a] Isolated yield. [b] Conversion as judged by 1H NMR spectroscopy. [c] Isolated as a complex mixture
alongside TBT-derived by-products.
substituted with an alkyl group (i.e. for all compounds in
Table 1), the oxy-anion is sufficiently nucleophilic to directly
displace a tetrazolide as leaving group and give g-sultones as
products. Conversely, when they are formed from g-hydroxy-
sulfones (5) with DBU and bear softer/less electron rich
(withdrawing) groups for example, Ph, vinyl (i.e. those in
Table 2), these are unable to directly attack on sulfur and
instead reaction is diverted along the pathway involving
anion-relay to give g-sultines. We hypothesized that use of the
propylene oxide derived substrate 5e would provide one of
the most evenly balanced scenarios between these two
for sultines.[34] Finally, the regioisomeric substrates 5i and 5j
underwent smooth cyclization to the sultines 7i and 7j
demonstrating that substitution in the b-position was possible
and that the need for an electron withdrawing substituent at
the g-position can be avoided if direct attack on sulfur can be
prevented. Sultines 7i and 7j being isolated solely as the cis-
diastereomers indicated high levels of 1,3-stereocontrol for
this substitution pathway.
In conclusion we present the first examples of a homolo-
gous Julia–Kocienski reaction which reveals two mechanisti-
cally novel reaction pathways of epoxides with sulfones.
15238 www.angewandte.org
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 15236 –15240

Angewandte
Chemie
M. Gruner, A. Jäger, O. Kataeva, P. Metz, Chem. Eur. J. 2011, 17,
13334.
[14] a) S. A. Wolckenhauer, A. S. Devlin, J. Du Bois, Org. Lett. 2007,
9, 4363; b) F. M. Koch, R. Peters, Angew. Chem. Int. Ed. 2007, 46,
2685; Angew. Chem. 2007, 119, 2739.
These two pathways provide access to g-sultones and g-
sultines in very rapid one and two-pot processes in a stereo-
controlled manner, something which existing methods fail to
achieve for either heterocycle. Sulfur containing heterocycles
are prevalent in drugs[35] and there is a current desire within
discovery chemistry to introduce saturated chiral scaffolds,[36]
as it is seen as a way to improve clinical success. Sultines and
sultones represent an interesting and largely unexplored area
of chemical space which is ripe for exploration given that
ready access to them is now possible.
[15] Sultones: a) Y. Morimoto, H. Kurihara, T. Kinoshita, Chem.
Commun. 2000, 189; Sultines: b) F. G. Bordwell, R. D. Chap-
man, C. E. Osborne, J. Am. Chem. Soc. 1959, 81, 2002; c) T.
Durst, T. K.-C. Tin, Can. J. Chem. 1970, 48, 845; d) O. B.
Bondarenko, T. I. Voevodskaya, L. G. Saginova, V. A. Tafeenko,
Y. S. Shabarov, J. Org. Chem. USSR 1987, 23, 1550; e) J.
Coulomb, V. Certal, L. Fensterbank, E. Lacôte, M. Malacria,
Angew. Chem. Int. Ed. 2006, 45, 633; Angew. Chem. 2006, 118,
649; f) J. Coulomb, V. Certal, M.-H. Larraufie, C. Ollivier, J.-P.
Corbet, G. Mignani, L. Fensterbank, E. Lacôte, M. Malacria,
Chem. Eur. J. 2009, 15, 10225; g) S. H. Kyne, H. M. Aitken, C. H.
Schiesser, E. Lacôte, M. Malacria, C. Ollivier, L. Fensterbank,
Org. Biomol. Chem. 2011, 9, 3331.
[16] a) D. Enders, W. Harnying, N. Vignola, Synlett 2002, 1727; b) D.
Enders, W. Harnying, N. Vignola, Eur. J. Org. Chem. 2003, 3939;
c) D. Enders, W. Harnying, Synthesis 2004, 2910.
[17] a) B. Fraser-Reid, K. M. Sun, R. Y. K. Tsang, P. Sinaþ, M.
Pietraszkiewicz, Can. J. Chem. 1981, 59, 260; b) P. A. Crooks,
R. C. Reynolds, J. A. Maddry, A. Rathore, M. S. Akhtar, J. A.
Montgomery, J. A. Secrist, J. Org. Chem. 1992, 57, 2830.
[18] For the analogous conversion of terminal epoxides to g-lactones,
see: M. Movassaghi, E. N. Jacobsen, J. Am. Chem. Soc. 2002, 124,
2456.
[19] a) M. Julia, J.-M. Paris, Tetrahedron Lett. 1973, 14, 4833; b) P. R.
Blakemore, W. J. Cole, P. J. Kocienski, A. Morley, Synlett 1998,
26; c) P. R. Blakemore, J. Chem. Soc. Perkin Trans. 1 2002, 2563.
[20] a) C. D. Bray, F. Minicone, Chem. Commun. 2010, 46, 5867;
b) C. D. Bray, G. De Faveri, J. Org. Chem. 2010, 75, 4652; c) F.
Minicone, W. J. Rogers, J. F. J. Green, M. Khan, G. M. T. Smith,
C. D. Bray, Tetrahedron Lett. 2014, 55, 5890.
[21] a) S. E. Schaus, B. D. Brandes, J. F. Larrow, M. Tokunaga, K. B.
Hansen, A. E. Gould, M. E. Furrow, E. N. Jacobsen, J. Am.
Chem. Soc. 2002, 124, 1307; b) O. A. Wong, Y. Shi, Chem. Rev.
2008, 108, 3958; c) M. Amatore, T. D. Beeson, S. P. Brown,
D. W. C. MacMillan, Angew. Chem. Int. Ed. 2009, 48, 5121;
Angew. Chem. 2009, 121, 5223.
[22] A. B. Smith III, W. M. Wuest, Chem. Commun. 2008, 5883.
[23] T. J. Snape, Chem. Soc. Rev. 2008, 37, 2452.
[24] W. S. Wadsworth, W. D. Emmons, J. Am. Chem. Soc. 1961, 83,
1733.
[25] For a recent reaction that involve loss of SO2 as a key step, see:
R. Gianatassio, S. Kawamura, C. L. Eprile, K. Foo, J. Ge, A. C.
Burns, M. R. Collins, P. S. Baran, Angew. Chem. Int. Ed. 2014, 53,
9851; Angew. Chem. 2014, 126, 10009.
[26] P. J. Kocienski, A. Bell, P. R. Blakemore, Synlett 2000, 365.
[27] See the Supporting Information.
[28] Two isolated examples of similar cyclizations have been
reported, see: a) S. O. Nwaukwa, S. Lee, P. M. Keehn, Synth.
Commun. 1986, 16, 309; b) V. Chudasama, A. R. Akhbar, K. A.
Bahou, R. J. Fitzmaurice, S. Caddick, Org. Biomol. Chem. 2013,
11, 7301.
[29] E. M. Carreira, T. C. Fessard, Chem. Rev. 2014, 114, 8257.
[30] a) H. Mayr, M. Breugst, A. Ofial, Angew. Chem. Int. Ed. 2011,
50, 6470; Angew. Chem. 2011, 123, 6598; b) J. S. Meek, J. S.
Fowler, J. Org. Chem. 1968, 33, 3422; c) M. Kobayashi, Bull.
Chem. Soc. Jpn. 1966, 39, 1296. See also: d) P. Messinger, R.
Vietinghoff-Scheel, Arch. Pharm. 1985, 318, 813; e) P. Mes-
singer, Arch. Pharm. 1985, 318, 950.
Acknowledgements
We thank the EPSRC/Syngenta for a CASE Ph.D. Student-
ship (to G.M.T.S. under grant EP/J50029X/1), the EPSRC UK
National Mass Spectrometry Facility at Swansea University
and the EPSRC UK National Crystallography Service Facility
at Southampton University.[37] We are indebted to Andrew
Plant and Janice Black (Syngenta) for the initial suggestion of
using the combination of LiN(SiMe3)2/CH2Cl2 and Matthew
Reid (Syngenta) for NMR assistance.
Keywords: anion-relay · anions · spiro compounds ·
sulfur heterocycles · synthetic methods
How to cite: Angew. Chem. Int. Ed. 2015, 54, 15236–15240
Angew. Chem. 2015, 127, 15451–15455
[1] a) H. Erdmann, Ann. Chem. 1888, 247, 306; b) E. Baumann, G.
Walter, Chem. Ber. 1893, 26, 1124.
[2] a) V. Meyer, Ber. Dtsch. Chem. Ges. 1883, 16, 1465; b) A.
Hantzsch, J. H. Weber, Ber. Dtsch. Chem. Ges. 1887, 20, 3118.
[3] a) T. Ojasoo, J. C. Dore, J. Gilbert, J.-P. Raynaud, J. Med. Chem.
1988, 31, 1160; b) D. V. Smil, F. E. S. Souza, A. D. Fallis, Bioorg.
Med. Chem. Lett. 2005, 15, 2057.
[4] C. Starkenmann, Y. Niclass, I. Cayeux, R. Brauchli, A.-C.
Gagnon, Flavour Frag. J. 2015, 30, 91.
[5] S. Yolka, E. Dunach, M. Loiseau, L. Lizzani-Cuvelier, R. Fellous,
S. Rochard, C. Schippa, G. George, Flavour Frag. J. 2002, 17, 425.
[6] E. Sarrazin, S. Shinkaruk, M. Pons, C. Thibon, B. Bennetau, P.
Darriet, J. Agric. Food Chem. 2010, 58, 10606.
[7] D. Postel, A. N. Van Nhien, J. L. Marco, Eur. J. Org. Chem. 2003,
3713.
[8] For recent examples, see: a) S. De Castro, C. García-Aparicio, G.
Andrei, R. Snoeck, J. Balzarini, M. J. Camarasa, S. Velµzquez, J.
Med. Chem. 2009, 52, 1582; b) H.-W. Xu, et al., Bioorg. Med.
Chem. Lett. 2014, 24, 2388.
[9] A. Laschewsky, Polymers 2014, 6, 1544.
[10] S. Schmitt, C. Bouteiller, L. BarrØ, C. Perrio, Chem. Commun.
2011, 47, 11465.
[11] For a recent example, see: J. Xia, L. Ma, J. R. Dahn, J. Power
Sources 2015, 287, 377.
[12] For a comprehensive review of sultones, see a) S. Mondal, Chem.
Rev. 2012, 112, 5339; For reviews of sultines, see: b) D. C.
Dittmer, M. D. Hoey, The Chemistry of Sulphinic Acids, Esters
and their Derivatives (Ed.: S. Patai), Wiley, New York, 1990,
Chap. 9, p. 239; c) O. B. Bondarenko, L. G. Saginova, N. V. Zyk,
Russ. Chem. Rev. 1996, 65, 147.
[31] T. Durst, J. C. Huang, N. K. Sharma, D. J. H. Smith, Can. J.
Chem. 1978, 56, 512.
[32] For recent examples of base-dependent stereodivergent meth-
odology, see: a) M. Sasaki, T. Takegawa, K. Sakamoto, Y.
[13] For representative examples, see: a) R. M. J. Liskamp, H. J.
Zeegers, H. C. J. Ottenheijm, J. Org. Chem. 1981, 46, 5408; b) P.
Fischer, A. B. G. Segovia, M. Gruner, P. Metz, Angew. Chem. Int.
Ed. 2005, 44, 6231; Angew. Chem. 2005, 117, 6387; c) P. Fischer,
Angew. Chem. Int. Ed. 2015, 54, 15236 –15240
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 15239

.
Angewandte
Communications
Kotomori, Y. Otani, T. Ohwada, M. Kawahata, K. Yamaguchi,
K. Takeda, Angew. Chem. Int. Ed. 2013, 52, 12956; Angew.
Chem. 2013, 125, 13194; b) S. Fustero, L. Herrera, R. Lazaro, E.
Rodriguez, M. Maestro, N. Mateu, P. Barrio, Chem. Eur. J. 2013,
19, 11776.
[35] a) C. W. Lindsley, ACS Chem. Neurosci. 2013, 4, 905; b) L. M.
Jarvis, Chem. Eng. News 2014, 92, 10.
[36] a) F. Lovering, J. Bikker, C. Humblet, J. Med. Chem. 2009, 52,
6752; b) R. D. Taylor, M. MacCoss, A. D. G. Lawson, J. Med.
Chem. 2014, 57, 5845.
[37] S. J. Coles, P. A. Gale, Chem. Sci. 2012, 3, 683.
[33] All of the saturated g-sultines in the Cambridge Crystallographic
=
Database have pseudo-axial S O bonds.
[34] C. A. G. Haasnoot, R. M. J. Liskamp, P. A. W. van Dael, J. H.
Received: September 9, 2015
Noordik, H. C. J. Ottenheim, J. Am. Chem. Soc. 1983, 105, 5406.
Published online: October 27, 2015
15240 www.angewandte.org
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 15236 –15240