A quantitative structure-reactivity relationship in decarboxylative Claisen rearrangement reactions of allylic tosylmalonate esters

Article abstract of DOI:10.1039/b812306c

First-order rate constants have been determined for the decarboxylative Claisen rearrangement reactions at 293 K of substituted methyl (E)-3-phenyl-2-propenyl 2-tosylmalonate esters, which show a linear free-energy relationship indicative of the development of positive charge on the benzylic position in the sigmatropic rearrangement transition-state. The Royal Society of Chemistry.

Full text of DOI:10.1039/b812306c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A quantitative structure–reactivity relationship in decarboxylative
Claisen rearrangement reactions of allylic tosylmalonate estersw
Donald Craig* and Nikolay K. Slavovz
Received (in Cambridge, UK) 18th July 2008, Accepted 29th September 2008
First published as an Advance Article on the web 14th October 2008
DOI: 10.1039/b812306c
First-order rate constants have been determined for the decar-
boxylative Claisen rearrangement reactions at 293 K of sub-
stituted methyl (E)-3-phenyl-2-propenyl 2-tosylmalonate esters,
which show a linear free-energy relationship indicative of the
development of positive charge on the benzylic position in the
sigmatropic rearrangement transition-state.
but was hampered in the case of 1f (isolated in 10% yield) by
the formation of a significant quantity (17%) of the g-lactone
4. This likely arises through cyclisation of the malonate
precursor of 1f via intramolecular nucleophilic attack of the
derived enolate on the styryl double bond, which is rendered
electrophilic by the electron-withdrawing 4-cyanophenyl
group. We have observed similar behaviour previously in an
analogous substrate possessing a 4-nitrophenyl residue,9 and
the 4-cyanostyryl moiety was shown to be susceptible to
intramolecular nucleophilic attack during ortho-Claisen
rearrangement of 4-cyanocinnamyl p-tolyl ether.10
The Ireland–Claisen rearrangement1 continues to be used
widely in synthesis as a strategy-level method for the regio-
specific and stereoselective formation of carbon–carbon
bonds.2 During the course of our studies of thermal decarboxy-
lative Claisen rearrangement (dCr) reactions of allylic
a-tosylesters,3–9 we have developed dCr reactions of allylic
tosylmalonate esters 1. These substrates may be converted into
the decarboxylated rearrangement products at room tempera-
ture using both the KOAc–N,O-bis(trimethylsilyl)acetamide
(BSA) reagent system developed initially,3,7 and modified
conditions involving treatment of 1 with DBU and tert-
butyldimethylsilyl triflate (TBDMSOTf).4 Further experimen-
tation showed that malonate substrates 2 possessing two
allylic ester groups underwent highly regioselective, room-
temperature mono-dCr reactions, in which the allylic group
possessing the more electron-rich R substituent rearranged
preferentially9 (Scheme 1). This communication presents 1H
nmr-based kinetic measurements of BSA–KOAc-mediated
dCr reactions of 1. The results provide further evidence10 for
significant positive charge build-up on the distal position of
the allylic moiety during sigmatropic rearrangement.
Reference samples of dCr products 3a–f were prepared
from 1 in 66–81% isolated yield using the DBU–TBDMSOTf
reagent system (CH2Cl2, rt, 40 min); all were formed as
diastereoisomeric mixtures in varying ratios (Fig. 1). A further
substrate 1 with S ¼ OMe was synthesised also, but its dCr
reaction was too rapid to provide useful kinetic data under the
conditions studied.
The dCr reactions of 1 to give 3 proceed via a stepwise
mechanism, in which silyl ketene acetal formation is
followed by [3,3]-sigmatropic rearrangement and subsequent
decarboxylation. In order to simplify reaction monitoring, we
required reaction conditions in which ketene acetal formation
was relatively rapid, followed by slower rearrangement
and decarboxylation. This was achieved by adding BSA
(2.3 equiv.) to CD2Cl2 solutions (ca. 0.08 M) of 1 containing
KOAc (0.1 equiv.) under nitrogen at 293 K, followed by brief
agitation and transfer to the nmr probe. Shimming typically
took 3–5 min, after which data acquisition began.
Our principal aim in this study was to evaluate the effect of
variations in electron density within the R group in tosylma-
lonates 1 without changing the associated steric demand.
Therefore, a series of substrates 1a–f was synthesised in which
R was a 4-substituted or an unsubstituted phenyl group. The
syntheses were carried out as previously described:4 coupling
of the appropriate cinnamyl alcohol11 with methyl malonate
(DMAP–DCC) gave the mixed malonate esters, which were
C-tosylated using tBuOK–TsF in DMSO.12 The C-tosylation
step provided acceptable yields (62–69%) of substrates 1a–e,
Under the conditions described above, mechanistic path-
ways may be envisaged as shown in Scheme 2. Initially,
substrates 1 undergo rapid formation of regioisomeric mix-
tures of geometric isomeric ketene acetals 5. Unlike 5y,
isomers 5x are inert with respect to [3,3]-sigmatropic
Department of Chemistry, Imperial College London, South Kensington
Campus, London, UK SW7 2AZ. E-mail: d.craig@imperial.ac.uk;
Fax: þ44 (0)20 7594 5868; Tel: þ44 (0)20 7594 5771
w Electronic supplementary information (ESI) available: Experimental
procedures, analytical data and nmr spectra (1H, 13C and where
appropriate 19F) for the synthesis and characterisation of 1, 3, and
synthetic intermediates; summarised kinetic data and first-order plots;
tabulated integral data vs. time for rearrangement reactions of 1a–f.
See DOI: 10.1039/b812306c
z Present address: Institut fur Organische Chemie, Universitat zu
¨
Koln, Greinstrasse 4, 50939 Koln, Germany.
¨
¨
¨
Scheme 1 dCr reactions of allylic 2-tosylmalonates.
ꢀc
This journal is The Royal Society of Chemistry 2008
6054 | Chem. Commun., 2008, 6054–6056

View Article Online
Fig. 2 1H nmr spectra of dCr reaction of 1b recorded at 60 s intervals.
Fig. 1 Structures of 1, 3 and 4.
and 5y on the nmr time-scale. Typically, 1H nmr analysis
showed depletion of 5, accumulation, and then depletion of 6,
and accumulation of 3 over the time-periods during which
measurements were taken. Fig. 2 shows 1H nmr traces ob-
tained at 60 s intervals for the reaction of 1b. Full details
appear in the ESI.w
Reactions were monitored over at least two half-lives for all
substrates apart from 1c, for which data were gathered over
ca. 1.5 half-lives. Plots of [A]/[A]0 versus t were obtained for
1a–f, where [A] is the integral for the observed protons in 5
(methylene or methyl ester, as described above), normalised by
taking into account the number of protons giving rise to the
signal being integrated, and [A]0 is the sum of the normalised
integrals for 5, 6 and 3. First-order rate constants kS were
obtained by fitting the data according to the equation
[A] ¼ [A]0eꢁkt þ B using nonlinear least squares,14 where B
is non-zero because of the formation of an unreactive species
alluded to above. The rate constant data collected in Table 1
were further analysed by plotting lg[kS/kH] versus s1
,
Spara
Scheme 2 Mechanistic pathways for the formation of 3.
which gave a straight-line plot and a r-value of ca. ꢁ2.3
(Fig. 3). The kS values for the reactions of 5 derived from
1a–f, and the corresponding s1
values15 are shown in
Spara
rearrangement; E-5x and E-5y may interconvert by
1,5-silatropic rearrangement.13 The lack of complete reaction
observed for all the substrates may indicate the formation of a
non-reactive species such as Z-5x. Following [3,3]-sigmatropic
rearrangement to silyl esters 6, further alternatives may be
considered. In one scenario, 6 may undergo desilylation by
acetate-mediated transesterification–decarboxylation, giving
the conjugate bases of 3 which then abstract a proton from
1. Alternatively, 6 may suffer loss of CO2 by a formal retro-ene
reaction to give ketene acetal derivatives of 3, which undergo
acetate-mediated desilylation and protonation by 1.
Table 1; full details are provided in the ESI.w
Several features of the data presented above are worthy of
note. Firstly, the first-order rate constants measured at 293 K
are all significantly higher than those reported for the ortho-
Claisen rearrangement reactions of 4-substituted cinnamyl
p-tolyl ethers carried out under significantly more forcing
conditions.10 This is well established semi-quantitatively,2
and may be rationalised in terms of the driving-force intrinsic
in the carbonyl-forming Ireland–Claisen process, in contrast
to the dearomatising nature of the [3,3]-sigmatropic rearran-
gement step in the ortho-Claisen variant. Secondly, the varia-
tion of rate constant as a function of the cinnamyl substituent
observed in the present work is significantly larger than that
reported previously, as evidenced by a r-value over five times
greater than that found for the ortho-Claisen rearrangements.
The dCr reactions of 1 were monitored by measuring the
changes in integrals of proton signals with respect to time. For
substrates 1a, 1b, 1d and 1e, signal resolution was such that it
was most convenient to measure integrals for the methylene
protons in 5 (which appeared as broad doublets), the C3
methines in 6 (pairs of broad doublets due to the diastereo-
isomeric mixture) and the C2 methines in 3 (pairs of sharp
doublets due to the diastereoisomeric mixture). For substrates
1c and 1f the integrals of the methyl ester CH3 groups in 5, 6
and 3 were measured. Care was taken to include signals
corresponding to both diastereoisomers in the integrated
ranges. The methylene and methoxy CH3 proton signals in 5
all showed significant broadening, indicative of the inter-
conversion through [1,5]-silatropic shift of regioisomers 5x
Table 1 Measured rate constants for dCr reactions of 1 at 293 K
Substrate
S
s1
105 kS/sꢁ1
lg[kS/kH]
Spara
1a
1b
1c
1d
1e
1f
H
Me
F
Cl
Br
CN
0
41.4
250.8
110.6
12.5
9.8
0
0.78
0.43
ꢁ0.52
–0.63
ꢁ1.34
ꢁ0.311
ꢁ0.073
0.114
0.15
0.659
1.9
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 6054–6056 | 6055

View Article Online
Notes and references
1 R. E. Ireland and R. H. Mueller, J. Am. Chem. Soc., 1972, 94,
5897. For reviews of the Ireland–Claisen rearrangement, see:
(a) S. Pereira and M. Srebnik, Aldrichimica Acta, 1993, 26, 17;
(b) Y. Chai, S. Hong, H. A. Lindsay, C. McFarland and
M. McIntosh, Tetrahedron, 2002, 58, 2905.
2 For a recent review of the Claisen rearrangement, see: A. M. Castro,
Chem. Rev., 2004, 104, 2939.
3 D. Bourgeois, D. Craig, N. P. King and D. M. Mountford, Angew.
Chem., Int. Ed., 2005, 44, 618.
4 D. Craig and F. Grellepois, Org. Lett., 2005, 7, 463.
5 D. Craig, F. Grellepois and A. J. P. White, J. Org. Chem., 2005, 70,
6827.
6 D. Craig, N. P. King, J. T. Kley and D. M. Mountford, Synthesis,
2005, 3279.
7 D. Bourgeois, D. Craig, F. Grellepois, D. M. Mountford and
A. J. W. Stewart, Tetrahedron, 2006, 62, 483.
Fig. 3 Plot of lg[kS/kH] vs. s1
for dCr reactions of 1a–f.
Spara
8 For the first report of Ireland–Claisen rearrangement of an allylic
a-phenylsulfonylacetate followed by decarboxylation, which took
place in a separate step, see: A. H. Davidson, N. Eggleton and
I. H. Wallace, J. Chem. Soc., Chem. Commun., 1991, 378.
9 D. Craig, M. I. Lansdell and S. E. Lewis, Tetrahedron Lett., 2007,
48, 7861.
10 For an early study of substituent effects on the rate of ortho-Claisen
rearrangement of substituted cinnamyl phenyl ethers, see: W. N.
White and W. K. Fife, J. Am. Chem. Soc., 1961, 83, 3846.
11 For substrates 1 having R ¼ 4-MeC6H4, 4-FC6H4, C6H5, 4-ClC6H4
and 4-BrC6H4, the cinnamyl alcohols were made by reduction of
the appropriate unsaturated ethyl esters using DIBAL-H; for
R ¼ 4-NCC6H4, NaBH4–CeCl3 was used to reduce the correspond-
ing aldehyde. The carbonyl reduction substrates were made by
Wittig reaction of EtO2CCHQPPh3 or OHCCHQPPh3 with the
corresponding benzaldehydes. Full experimental details are
provided in the ESIw.
12 D. Craig, F. Paina and S. C. Smith, Chem. Commun., 2008, 3408.
13 For high-temperature thermal decarboxylative Claisen rearrange-
ment reactions of b-keto ester-derived silyl enol ethers involving
silatropic rearrangement, see: R. M. Coates, L. O. Sandefur and
R. D Smillie, J. Am. Chem. Soc., 1975, 97, 1619.
14 We thank a referee for pointing out the importance of fitting the
data using nonlinear rather than linear least squares.
15 C. D. Ritchie and W. F. Sager, Prog. Phys. Org. Chem., 1964, 2,
323.
Indeed, the value for r determined in the present work is
similar in magnitude to that found for the hydrolysis of
substituted benzyl chlorides in aqueous ethanol.16
Acceleration of the Ireland–Claisen rearrangement of sub-
strates possessing an electron-donating substituent on the
distal position of the allylic moiety has been documented.
Curran et al. have shown17 that the presence of an oxygen
atom in this position leads to an increase in rate of one or more
orders of magnitude. This effect has been studied computa-
tionally,18 and has been described in terms of ‘‘vinylogous
anomeric’’ (p - s*) stabilisation17d,19 of an early transition-
state.20,21 In this model, there is significant weakening of the
allylic C–O bond, such that its scission may be significantly
more advanced than formation of the new C–C bond. This
induces positive charge character on the distal allylic carbon
atom, which is stabilised by more electron-donating substitu-
ents attached at that position.22 In the present work, the
rate enhancement observed for ketene acetals 5 possessing
electron-rich Ar is consistent with this analysis. The notion of
a dipolar transition-state is further supported by published
observations that an oxygen substituent at the allylic position
leads to reaction acceleration in the Claisen rearrangement,17e
and that solvent17e,21c and hydrogen-bonding additives17f
significantly affect the rate.
16 R. W. Alder, R. Baker and J. M. Brown, Mechanism in Organic
Chemistry, Wiley-Interscience, London, 1971.
17 (a) D. P. Curran, Tetrahedron Lett., 1982, 23, 4309; (b) D. P. Curran
and Y. Suh, Tetrahedron Lett., 1984, 25, 4179; (c) D. P. Curran and
Y. Suh, J. Am. Chem. Soc., 1984, 106, 5002; (d) D. P. Curran and
Y. Suh, Carbohydr. Res., 1987, 171, 161; (e) R. M. Coates,
B. D. Rogers, S. J. Hobbs, D. P. Curran and D. R. Peck, J. Am.
Chem. Soc., 1987, 109, 1160; (f) D. P. Curran and H. K. Lung,
Tetrahedron Lett., 1995, 36, 6647.
18 (a) H. Y. Yoo and K. N. Houk, J. Am. Chem. Soc., 1997, 119,
2877; (b) V. Aviyente and K. N. Houk, J. Phys. Chem. A, 2001,
105, 383; (c) V. Aviyente, H. Y. Yoo and K. N. Houk, J. Org.
Chem., 1997, 62, 6121.
19 S. E. Denmark and M. S. Dappen, J. Org. Chem., 1984, 49, 798,
and references therein.
20 K. D. McMichael and G. L. Korver, J. Am. Chem. Soc., 1979, 101,
2746.
21 (a) J. J. Gajewski and N. D. Conrad, J. Am. Chem. Soc., 1979, 101,
2747; (b) J. J. Gajewski and J. Emrani, J. Am. Chem. Soc., 1984,
106, 5733; (c) J. J. Gajewski, Acc. Chem. Res., 1997, 30, 219.
22 M. P. Meyer, A. J. DelMonte and D. A. Singleton, J. Am. Chem.
Soc., 1999, 121, 10865.
In summary, we have shown that the first-order rate con-
stants for dCr reactions of 1 show a pronounced dependency
on s1 for the para-aryl substituent S, and therefore on the
electron-density of the aryl group. This is indicative of sig-
nificant positive charge development at the benzylic position
during rearrangement, and the magnitude of the effect is
substantially greater than that found in previous investigations
of structure–rate relationships in [3,3]-sigmatropic processes.
This research was supported by Evangelisches Studienwerk
e.V. Villigst (scholarship to N.K.S.). We thank Dr John C. de
Mello for valuable assistance with data analysis, and Mr
Richard N. Sheppard and Mr Peter R. Haycock for nmr
measurements.
ꢀc
This journal is The Royal Society of Chemistry 2008
6056 | Chem. Commun., 2008, 6054–6056