Silyl-modified Bellus-Claisen rearrangement

Article abstract of DOI:10.1039/b614535c

A silyl-modified variant of the Bellus-Claisen rearrangement is described; the generality of this rearrangement has been demonstrated with a range of allylic amines and ketenes. The Royal Society of Chemistry.

Full text of DOI:10.1039/b614535c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Silyl-modified Bellusˇ–Claisen rearrangement{
Donald Craig,*a N. Paul Kingb and David M. Mountforda
Received (in Cambridge, UK) 6th October 2006, Accepted 9th November 2006
First published as an Advance Article on the web 21st December 2006
DOI: 10.1039/b614535c
used. We ascribed this failure to competing interactions of
A silyl-modified variant of the Belluˇs–Claisen rearrangement is
described; the generality of this rearrangement has been
demonstrated with a range of allylic amines and ketenes.
the Lewis basic groups with the powerful Lewis acid, and an
alternative metal-free approach was sought, in which these
unwanted interactions might be avoided.
The Claisen rearrangement1 has found widespread use in synthetic
organic chemistry due to the stereospecific and highly stereo-
selective nature of the [3,3]-sigmatropic carbon–carbon bond-
forming process. Modern variants of this rearrangement2 have
ensured the continued prominence of this powerful organic
synthesis tool, which allows access to a range of carboxylic acid
derivatives, depending on the reaction mode employed.
We envisaged combining zwitterion-like motif 2 of the Lewis
acid-assisted Belluˇs–Claisen rearrangement with the silyl ketene
acetal moiety of the Ireland–Claisen intermediate 3 to generate a
hybrid array 4 (Fig. 1). It was anticipated that silylation of a ketene
generated in situ with trimethylsilyl triflate (TMSOTf) would
sufficiently activate the ketene towards interception by the allylic
amine, thereby enabling Claisen rearrangement to take place
without the need for a powerful Lewis acid such as Me3Al6c,6d
or TiCl4.7
Treatment of a mixture of amine 5,8 TMSOTf and Hu¨nig’s base
with phenylacetyl chloride, propionyl chloride or dichloroacetyl
chloride using the modified conditions gave, respectively, the
rearranged products 7–9 in good yields. It seems likely that the
ketenes generated in situ are activated by TMSOTf prior to
interception by the allylic amine to give intermediates 6. [3,3]-
Sigmatropic rearrangement then gives rise to the unstable iminium
species 10, which readily collapse with concomitant loss of the silyl
group to give 7–9 (Scheme 2).{
We have begun a new programme to investigate, in-depth, the
effect of stereocentres outside the six-membered pericyclic array
upon the stereochemical outcome of Claisen rearrangements.3 The
Belluˇs–Claisen rearrangement4,5 was selected as the basis of the
new study.6 In this version of the transformation, the highly
reactive electrophile dichloroketene is intercepted by an allylic
ether, thioether or tertiary amine, and the resulting zwitterionic
intermediate, 1, undergoes
a spontaneous [3,3]-sigmatropic
rearrangement to form the corresponding a,a-dichloro-substituted
ester, thiol ester or tertiary amide (Scheme 1).
Our initial studies focused on the Lewis acid-assisted version
of the Belluˇs–Claisen rearrangement,6c,6d,7 in which reactions of
less highly activated ketenes with allylic tertiary amines deliver
c,d-unsaturated tertiary amide products in excellent yields. Whilst
our initial findings confirmed that the use of TiCl4 yielded
rearranged products for a range of simple allylic tertiary amines
and in situ generated ketenes,7 reactions of more complex allylic
amines containing protected heteroatoms resulted in either
decomposition or recovery of starting material when TiCl4 was
For comparison purposes, all three transformations were
carried out with TMSOTf, TiCl4 and TiCl(OTf)39 as Lewis acidic
additives (Table 1). These results show that TMSOTf performs
Fig. 1 Claisen rearrangement intermediates.
Scheme 1 The Belluˇs–Claisen rearrangement.
aDepartment of Chemistry, Imperial College London, South Kensington
Campus, London, UK SW7 2AZ. E-mail: d.craig@imperial.ac.uk;
Fax: +44 (0)20 7594 5868; Tel: +44 (0)20 7594 5771
bNeurology and GI Centre of Excellence for Drug Discovery,
GlaxoSmithKline, New Frontiers Science Park, Third Avenue, Harlow,
Essex, UK CM19 5AW
{ Electronic supplementary information (ESI) available: Experimental
details and full spectroscopic data for all allylic amines and rearrangement
products. See DOI: 10.1039/b614535c
Scheme 2 Reagents and conditions: RR9CHCOCl (1.2 equiv.), TMSOTf
(1.0 equiv.), iPr2NEt (1.5 equiv.), CH2Cl2, rt, 4.5 h.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 1077–1079 | 1077

View Article Online
Table 1 Silyl-modified Bellusˇ–Claisen rearrangement of 5
Acid Chloride
Yield (%)a
TMSOTfb TiCl4 TiCl(OTf)3
c
d
Entry R
R9
Product
1
2
3
a
Ph
Me
Cl
H
74
83
57
79
58
44
84
96
64
H
Cl
Products 7 and 8 were obtained as single diastereoisomers. All
reactions were carried out at rt with 1.5 equiv. iPr2NEt and
1.2 equiv. RR9CHCOCl. 1.0 equiv. TMSOTf, 4.5 h. 0.1 equiv.
Scheme 3 Rationale of observed stereoselectivity.
b
c
d
TiCl4, 1 h. 0.1 equiv. TiCl(OTf)3, 7 h.
gave the corresponding anti-products. Product stereochemistry
was assigned indirectly in these two transformations by analysis of
the derived a,b-anti- and a,b-syn-c-lactones, respectively, made by
acid-catalysed cyclisation followed by base-mediated silylation.
We speculate that in Nubbemeyer’s work, the phenyl-bearing
stereocentre may have undergone epimerisation to the sterically
more favourable a,b-anti-isomer during the ring closure or silyl
protection steps.
similarly to TiCl(OTf)3, and that it gives better yields than TiCl4 in
two of the three cases studied.10 Interestingly, the lowest yields in
these reactions were obtained for dichloroketene, which is the least
nucleophilic of the three ketenes evaluated. Substitution of
TMSOTf with the bulkier TBDMSOTf in the reaction of 5 with
phenylacetyl chloride caused a reduction in yield of 7 to 30%, and
use of TIPSOTf and TMSCl resulted in the recovery only of 5.
Reducing the amount of TMSOTf used to 0.1 equiv. gave 44% 7
and 48% recovered 5, suggesting that whilst some turnover of
TMSOTf takes place, catalyst decomposition is a significant
process. We speculate that this arises through the interception of
TMSOTf by chloride ions, which generates rearrangement-inactive
TMSCl.
The new silyl-modified Belluˇs–Claisen rearrangement conditions
were next applied to a range of allylic tertiary amines. The amines
were prepared in excellent yields by treatment of the corresponding
allylic alcohols with PPh3–NBS–morpholine as before.7b The
exposure of mixtures of amine, Hu¨nig’s base and TMSOTf to
phenylacetyl chloride or propionyl chloride yielded the corre-
sponding rearranged amides in good-to-excellent yields with high
and sometimes complete stereoselectivity (Table 2).
In the case of amide 7, the assignment of anti-stereochemistry
followed from X-ray crystallographic analysis (Fig. 2).§11 The
stereochemistry of 8 was inferred from that of 7. We rationalise
the stereochemistry of formation of 7 and 8 in terms of the
intermediacy of a Z-intermediate 6, formed by attack of amine 5
on the silylated ketene in an anti-sense with respect to the R group.
Rearrangement then takes place via a chair-like six-membered
transition-state (Scheme 3).
Some trends emerge from the data collected in Table 2. The
stereochemistry of all of the amide products may be rationalised in
terms of the chair-like transition-state model depicted in Scheme 3.
The presence of exo-pericyclic stereocentres was investigated with
allylic amines 17 and 19 (Table 2, entries 4 and 5); both amines
successfully underwent rearrangement to yield the corresponding
amides 18 and 20 as single diastereomers, which corresponds to
delivery of the carboxamide-containing moiety to the less
sterically-crowded face of the allylic double bond. The successful
rearrangement of substrates 23 and 25 (Table 2, entries 7 and 8)
demonstrates that quaternary carbon atoms may be accessed using
this chemistry, though the relatively low yield of the conversion of
25 into 26 may reflect the reduced nucleophilicity of the amine on
account of steric crowding. In this context, subjection of Z-allylic
amine 27 to a reaction with phenylacetyl chloride (Table 2, entry 9)
gave the phenyl analogue of 28 in only 19% yield. The low yield of
the reaction of Z-substrate 29 (Table 2, entry 10), even with the
more reactive propionyl chloride-derived ketene, may be indicative
of the combined effects of the Z-double bond and the electron-
withdrawing benzyloxy substituent, both of which serve to
attenuate amine nucleophilicity. No rearrangement product was
obtained when 29 was exposed to phenylacetyl chloride under the
standard conditions.
The proven anti-stereochemistry of amide 7 is in contrast with
the findings of Nubbemeyer et al.,6d who reported that in
trimethylaluminium-assisted Belluˇs–Claisen rearrangements invol-
ving pyrrolidine-containing tertiary amines, the use of phenylacetyl
chloride gave syn-diastereoisomers, whereas propionyl chloride
In summary, our investigations have led to the development of
a metal-free variant of the Lewis acid-assisted Belluˇs–Claisen
Fig. 2 The molecular structure of 7.
1078 | Chem. Commun., 2007, 1077–1079
This journal is ß The Royal Society of Chemistry 2007

View Article Online
and ketenes. Ongoing work is concerned with improving the sub-
stoichiometric process and investigating the effect of exo-pericyclic
stereocentres in acyclic substrates, and the results of these
investigations will be reported in due course.
Table 2 Silyl-modified Bellusˇ–Claisen rearrangement of allylic
amines
Diaster-
eomer
Yield
Entry Substrate
1
Producta,b
ratio (%)c (%)
The authors thank EPSRC/GlaxoSmithKline (supported DTA
studentship to D. M. M.) for support.
100 : 0
79
Notes and references
{ Typical experimental procedure for the silyl-modified Belluˇs–Claisen
rearrangement of allylic amines: To a solution of the allylic amine
(1.18 mmol, 1.0 equiv.) in CH2Cl2 (10 mL) at rt was added TMSOTf
(1.18 mmol, 1.0 equiv.), followed by N,N-diisopropylethylamine
(1.77 mmol, 1.5 equiv.). The solution was stirred for 5 min before a
solution of phenylacetyl chloride (1.42 mmol, 1.2 equiv.) in CH2Cl2 (5 mL)
was added dropwise over 4 h. The resulting mixture was stirred for 30 min.
The reaction was then diluted with diethyl ether (10 mL), treated with
aqueous NaOH (1 M; 5 mL) and stirred for a further 10 min. The aqueous
layer was extracted with diethyl ether (3 6 20 mL), the combined organic
layers washed with brine (3 6 10 mL), and dried with Na2SO4.
Concentration under reduced pressure and chromatography (30% ethyl
acetate–petroleum ether) gave the corresponding amide.
2
3
4
5
88 : 12 90
83 : 17 56
100 : 0
100 : 0
76
73
§ Crystal data for 7: C18H25NO2, M = 287.39, orthorhombic, a = 6.122(3),
b = 15.768(2), c = 17.431(2) s, b = 90u, U = 1682.7(8) s3, T = 293(2) K,
P212121, Z = 4, m = 0.574 mm21, Rint = 0.0000, R1 = 0.0816, wR2 = 0.1679
(all data). CCDC 623206. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/b614535c
1 (a) L. Claisen, Chem. Ber., 1912, 45, 3157; (b) For a recent, comprehen-
sive review of the Claisen rearrangement, see: A. M. M. Castro, Chem.
Rev., 2004, 104, 2939(c) See also: The Claisen Rearrangement: Methods
and Applications, ed. M. Hiersemann and U. Nubbemeyer, Wiley-VCH,
Weinheim, 2006.
2 (a) A. E. Wick, D. Felix, K. Steen and A. Eschenmoser, Helv. Chim.
Acta, 1964, 47, 2425; (b) W. S. Johnson, L. Werthemann, W. R. Bartlett,
T. J. Brocksom, T.-T. Li, D. J. Faulkner and M. R. Petersen, J. Am.
Chem. Soc., 1970, 92, 741; (c) R. E. Ireland and R. H. Mueller, J. Am.
Chem. Soc., 1972, 94, 5897.
6
7
80 : 20 79
97 : 3
89
3 For examples, see: (a) J. K. Cha and S. C. Lewis, Tetrahedron Lett.,
1984, 25, 5263; (b) T. Suzuki, E. Sato, S. Kamada, H. Tada, K. Unno
and T. Kametani, J. Chem. Soc., Perkin Trans. 1, 1986, 387; (c)
J. R. Hauske and S. M. Julin, Tetrahedron Lett., 1993, 34, 4909.
4 (a) R. Malherbe and D. Belluˇs, Helv. Chim. Acta, 1978, 61, 3096; (b)
R. Malherbe, G. Rist and D. Belluˇs, J. Org. Chem., 1983, 48, 860.
5 For a review of the Belluˇs–Claisen rearrangement, see: J. Gonda,
Angew. Chem., Int. Ed., 2004, 43, 3516.
6 For previous examples of 1,2-exo-pericyclic stereoselectivity in Belluˇs–
¨
Claisen rearrangement reactions, see: (a) U. Nubbemeyer, R. Ohrlein,
J. Gonda, B. Ernst and D. Belluˇs, Angew. Chem., Int. Ed. Engl., 1991,
30, 1465; (b) U. Nubbemeyer, J. Org. Chem., 1995, 60, 3773; (c)
U. Nubbemeyer, J. Org. Chem., 1996, 61, 3677; (d) B. Ernst, J. Gonda,
8
9
100 : 0
100 : 0
100 : 0
49
79
39
¨
R. Jeschke, U. Nubbemeyer, R. Ohrlein and D. Belluˇs, Helv. Chim.
Acta, 1997, 80, 876; (e) J. Gonda, M. Martinkova, B. Ernst and
D. Belluˇs, Tetrahedron, 2001, 57, 5607.
10
7 (a) T. P. Yoon, V. M. Dong and D. W. C. MacMillan, J. Am. Chem.
Soc., 1999, 121, 9726; (b) T. P. Yoon and D. W. C. MacMillan, J. Am.
Chem. Soc., 2001, 123, 2911.
8 Amine 5 was synthesised from (E)-2-hexen-1-ol by treatment with
morpholine, PPh3 and N-bromosuccinimide. See ref. 7b.
9 S. Kobayashi, M. Moriwaki and I. Hachiya, Bull. Chem. Soc. Jpn.,
1997, 70, 267.
10 A recent report describes the failure of TMSOTf, in combination with
an unspecified base, to effect an aza-Claisen rearrangement of a
prolinol-derived allylic amine. See: Q. Xu and E. Rozners, Org. Lett.,
2005, 7, 2821.
a
All reactions were carried out at rt with 1.5 equiv. iPr2NEt,
1.2 equiv. PhCH2COCl or MeCH2COCl and 1.0 equiv. TMSOTf,
4.5 h. Major diastereoisomer shown. Determined by 1H NMR
analysis of the crude reaction mixture.
b
c
rearrangement. The generality of the rearrangement has been
demonstrated with a range of structurally diverse allylic amines
11 We thank Dr. Andrew J. P. White (Department of Chemistry, Imperial
College) for the X-ray determinations.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 1077–1079 | 1079