Revisiting the Maitland-Japp reaction. Concise construction of highly functionalised tetrahydropyran-4-ones

Article abstract of DOI:10.1039/b416247a

Application of modern synthetic methods to the Maitland-Japp reaction has provided a one pot, one step procedure for the efficient construction of highly substituted tetrahydropyran-4-ones. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b416247a

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Revisiting the Maitland–Japp reaction. Concise construction of highly
functionalised tetrahydropyran-4-ones{
Paul A. Clarke,* William H. C. Martin, Jason M. Hargreaves, Claire Wilson and Alexander J. Blake
Received (in Cambridge, UK) 21st October 2004, Accepted 24th November 2004
First published as an Advance Article on the web 6th January 2005
DOI: 10.1039/b416247a
Application of modern synthetic methods to the
Maitland–Japp reaction has provided a one pot, one step
procedure for the efficient construction of highly substituted
tetrahydropyran-4-ones.
Tetrahydropyran (THP) rings are ubiquitous in the natural
product arena, and over the years much effort has been directed
towards the development of new strategies for their synthesis.1
Fig. 1 The Maitland–Japp reaction.
Continued interest in the synthesis of THP containing natural
Lewis acid in the presence of an aldehyde should generate aldol
products provided the impetus to initiate a project designed to
product 2, which could then undergo a Lewis acid catalysed
develop a new and more expedient route to the formation of THP
Knoevenagel condensation with a second equivalent of a different
rings. Our attention was attracted to the report published by
aldehyde or ketone to furnish 3. Finally, a Lewis acid catalysed
Maitland and Japp in 1904, which showed that a molecule of
intramolecular oxy-Michael reaction would deliver tetrahydro-
pentan-3-one and 2 molecules of benzaldehyde could be condensed
pyran-4-ones of type 4 (Scheme 1). As there was precedent in the
ring2 (Fig. 1). Much later this reaction was found to generate a
literature for the Lewis acid catalysed versions of all of these
in a low yielding process to generate a highly substituted THP
single diastereomer product.3 While the original reaction had
processes, it was envisaged that the three individual reactions
which make up the synthesis would all follow on from each other
several drawbacks such as long reaction times, use of excess
in one pot.
aqueous reagents, low yields and lack of generality, we were
The first conditions examined used TiCl4 as the Lewis acid as
this has been shown to promote both Mukaiyama aldol reactions6
and the fact that multiple carbon–carbon bond forming reactions
encouraged by its multi-component nature, its diastereoselectivity
and Knoevenagel condensations.7 When isobutyraldehyde was
were occurring in one pot. We realised that with current synthetic
treated with TiCl4 and 1 at 278 uC, smooth conversion to the
aldol product was observed. Addition of n-butanal to the reaction
mixture at this time and warming the reaction to room
temperature led to the very slow formation of tetrahydropyrans
5a and 6a. It was found that the rate of the tandem Knoevenagel–
Michael reaction could be increased dramatically if TFA was
added to the reaction before the addition of n-butanal. This
increase in rate presumably arose from the acid catalysed
removal of the silyl ethers initially formed in the Mukaiyama
aldol reaction. In this manner tetrahydropyrans 5a and 6a were
generated in a 1 : 1 ratio in an excellent 98% yield. However, when
Yb(OTf)3 was used as the Lewis acid it was discovered that the
diastereoselectivity of the reaction favored the formation of 5a
over 6a, in an excellent 91% yield (Table 1).8 Encouraged by this
success a range of different aldehyde partners were investigated
with each Lewis acid (Table 1).
technology it may prove fruitful to revisit this forgotten reaction
and to investigate the potential for a version of the Maitland–Japp
reaction to be a useful tool in the armory of the synthetic chemist.
As it is desirable to develop a procedure for the synthesis of non-
symmetrical tetrahydropyrans, it was decided to move away from
a symmetrical ketone based reaction and instead to concentrate on
the use of a b-ketoester derivative.4 In the case of b-ketoester
derivatives the difference in reactivity of the a- and c-positions
should allow for the reaction of the b-ketoester derivative with
different aldehydes or ketones at the a- and c-positions. The initial
b-ketoester derivative chosen was the bis-trimethylsilyl enol ether
of methyl acetoacetate (Chan’s diene, 1).5 Treatment of 1 with a
{ Electronic supplementary information (ESI) available: experimental
section. See http://www.rsc.org/suppdata/cc/b4/b416247a/
*paul.clarke@nottingham.ac.uk
Scheme 1 Strategy for the synthesis of highly functionalised tetrahydropyran-4-ones.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 1061–1063 | 1061

View Online
Table 1 One pot formation of tetrahydropyran-4-ones
TiCl4
Yield (%)
TiCl4
Ratio 5 : 6
Yb(OTf)3
Yield (%)
Yb(OTf)3
Ratio 5 : 6
R
R1
a
b
c
d
e
f
iPr
Ph
Pr
Ph
98
98
88
88
93
80
74
82
1 : 1
1 : 3
1 : 4
1 : 1
1 : 3
1 : 0
1 : 2
1 : 3
91
98
93
81
91
75
86
80
10 : 1
2.5 : 1
2 : 1
6 : 1
2 : 1
1 : 0
11 : 1
2.5 : 1
iPr
Pr
(CH2)2CHLCH2
CH2OBn
Ph
Ph
Cyhx
iPr
4-MeOC6H4
iPr
Ph
Ph
g
h
As can be seen from Table 1, a wide range of structurally
different aldehydes can be incorporated in either the 2 or 6 position
of the products, and the yields of the tetrahydropyran-4-ones range
from good to excellent. Of particular interest is the significant
change in the diastereoselectivity of the reaction when the Lewis
acid was changed from TiCl4 to Yb(OTf)3. In general, use of TiCl4
tended to favor the formation of the 2,6-trans diastereomer 6,
while use of Yb(OTf)3 favored the formation of the 2,6-cis
diastereomer 5. It is rationalised that the 2,6-cis-diastereomer exists
as the keto-tautomer as this obviates a large eclipsing interaction
between the C5-ester group and the C6-substituent which would
be present in the enol-tautomer. The 2,6-trans-diastereomer exists
as the enol-tautomer as the hydrogen bond formed between the
enol proton and the carbonyl of the adjacent ester compensates for
placing the C6-substituent in a pseudo-axial position. A possible
explanation for this change in diastereoselectivity with different
Lewis acids is that the stoichiometric amount of TiCl4 in solution is
chelated by the enol form of the trans-diastereomer. However,
Yb(OTf)3 is much less soluble in CH2Cl2 and therefore present in
solution in much smaller amounts. This has the effect that much
less of the product is chelated with the Lewis acid and so exists as
the cis-diastereomer in the keto form, where all the substituents are
equatorial.
Scheme 2 The decarboxylative Maitland–Japp reaction.
As the 2,6-cis and 2,6-trans diastereomers were in equilibrium
with one another diene 1 was changed for the bis-trimethylsilyl
enol ether of tert-butyl acetoacetate 7, in an attempt to initiate
in situ Lewis acid catalysed decarboxylation of the 2,6-cis-
tetrahydropyran-4-one 8 which existed as the keto tautomer, and
thus drive the equilibrium over to the 2,6-cis compound 9. It was
rationalised that as the 2,6-trans diastereomer 10 existed as the enol
tautomer decarboxylation would not be possible (Scheme 2).
As can be seen from Table 2, in situ decarboxylation did
not occur under the reaction conditions which instead
furnished excellent yields of mixtures of 8 and 10 with 8, the 2,6-
cis-tetrahydropyran-4-one, predominating. This cis selectivity is
the complete reverse of the selectivity obtained when the
reaction was run under identical conditions using Chan’s diene 1
(Table 1).
Exploration into the nature of the diastereoselection showed
that in the case of both Lewis acids the reaction was under
thermodynamic control. When separate samples of 5b and 6b were
resubmitted to the reaction conditions in the presence of TiCl4 for
18 h an identical 1 : 3 mixture of 5b and 6b resulted in both cases.
However, when this equilibration was carried out at 278 uC for
18 h, rather than at room temperature, the ratio of 5b to 6b was
found to be reversed at 3 : 1. Interestingly, when the formation
reaction was repeated with TiCl4 and held at 0 uC, then worked-up
within 2 minutes of the addition of the second aldehyde, 5b was
formed exclusively, implying that, under these conditions, 5b is the
kinetic product and 6b is the thermodynamic one. In the case of
the Yb(OTf)3 promoted reaction, separate samples of 5b and 6b
were resubmitted to the reaction conditions and after 18 h were
found to have equilibrated to identical 2.5 : 1 mixtures of 5b to 6b.
This shows that the Yb(OTf)3 promoted reaction is also under
thermodynamic control. In all cases there was no loss of material
from these equilibration studies.
It is also possible to construct enantiomerically pure tetrahy-
dropyran-4-ones as there is no erosion of the enantiomeric excess
of the d-hydroxy b-ketoesters 12 under the cyclisation reaction
conditions (Scheme 3). Enantiomerically pure b-hydroxy esters
11a/b are available commercially and were homologated via a
standard Claisen condensation reaction9 to yield 12a/b. d-Hydroxy
b-ketoesters 12a/b were subjected to our TiCl4 promoted tandem
Knoevenagel/oxy-Michael reaction and yielded 8i and 8f respec-
tively; pyranone 8f was then decarboxylated in situ by further
Table 2 Use of bis-trimethylsilyl enol ether of tert-butyl acetoacetate
in the Maitland–Japp reaction
R
R1
Yield (%)
Ratio 8 : 10
a
b
c
iPr
Ph
iPr
Pr
Ph
92
98
91
14 : 1
4 : 1
2 : 1
(CH2)2CLCH2
1062 | Chem. Commun., 2005, 1061–1063
This journal is ß The Royal Society of Chemistry 2005

View Online
Notes and references
{ Diffraction data were acquired on a Bruker SMART1000 (5b) or a
Bruker SMART APEX (6b) CCD area detector diffractometer equipped
with an Oxford Cryosystems open-flow cryostat operating at 150 K. The
structures were solved by direct methods and refined by full-matrix least-
squares on F2. Crystal data for 5b. C19H18O4, M 5 310.33, monoclinic,
˚
a 5 13.0132(10), b 5 8.6004(7), c 5 14.1620(11) A, b 5 91.558(2)u,
V 5 1584.4(2) A , T 5 150(2) K, Z 5 4, Dx 5 1.301 g cm23. Final R1 [2746
3
˚
F . 4s(F)] 5 0.0375, wR2 [all 3656 F2] 5 0.105. Crystal data for 6b.
C19H18O4, M 5 310.33, triclinic, a 5 5.5429(5), b 5 9.5351(8),
˚
c 5 15.1955(13) A, a 5 82.937(2), b 5 85.273(2), c 5 77.190(2)u,
V 5 775.9(2) A , T 5 150(2) K, Z 5 2, Dx 5 1.328 g cm23. Final R1 [2913
3
˚
F . 4s(F)] 5 0.0401, wR2 [all 3507 F2] 5 0.111. CCDC 250776–250777.
See http://www.rsc.org/suppdata/cc/b4/b416247a/ for crystallographic data
in .cif or other electronic format.
1 (a) For the manipulation of carbohydrates see: Total Synthesis of
Natural Products: The ‘Chiron’ Approach,S.Hanessian,Ed.J.E.Baldwin,
Pergamon (Oxford), 1983; C. Esteveza, A. J. Fairbanks and
G. W. J. Fleet, Tetrahedron, 1998, 54, 13591; (b) For a treatise on the
Hetero Diels–Alder reaction see: D. L. Boger and S. M. Weinreb,
Hetero Diels–Alder Methodology in Organic Synthesis, Academic Press
(San Diego), 1987; (c) For recent examples of the Prins reaction see:
C. St. J Barry, S. R. Crosby, J. R. Harding, R. A. Hughes, C. D. King,
G. D. Parker and C. L. Willis, Org. Lett., 2003, 5, 2429; S. R. Crosby,
J. R. Harding, C. D. King, G. D. Parker and C. L. Willis, Org. Lett.,
2002, 4, 3407; J. J. Jaber, K. Mitsui and S. D. Rychnovsky, J. Org.
Chem., 2001, 66, 4679; M. J. Cloninger and L. E. Overman, J. Am.
Chem. Soc., 1999, 121, 1092; S. A. Kozmin, Org. Lett., 2001, 3, 755. For
a review of the Prins reaction see: D. R. Adams and S. P. Bhaynagar,
Synthesis, 1977, 661; (d) For a review of the Michael reaction see:
R. D. Little, M. R. Masjedizadeh, O. Wallquist and J. I. McLoughlin,
Org. React., 1995, 47, 315.
Scheme 3 Construction of single enantiomers.
addition of 10 equiv. of TFA and heating to provide 9b. Analysis
of the enantiomeric excesses by chiral 1H NMR shift reagent
showed that both 8i and 9b were single enantiomers, proving that
the enantiomeric integrity of the d-hydroxy b-ketoesters 12 or the
pyran products 8 and 9 are not eroded by the Maitland–Japp
reaction conditions.10
In summary, we have developed an efficient one pot, multi-
component and diastereoselective synthesis of highly functionalised
tetrahydropyran-4-ones and have shown that it can be used to
prepare 2,6-cis-disubstituted tetrahydropyran-4-ones in enantio-
merically pure form. We are now investigating the possibility of
installing the hydroxyl stereocentre via a catalytic asymmetric
aldol reaction in the same pot as the subsequent Knoevenagel/
oxy-Michael reactions.{
2 F. R. Japp and W. Maitland, J. Chem. Soc., 1904, 85, 1473.
3 R. Sivakumar, N. Satyamurthy, K. Ramalingam, D. J. O’Donnell,
K. Ramarajan and K. D. Berlin, J. Org. Chem., 1979, 44, 1559.
4 For preliminary investigations see P. A. Clarke and W. H. C. Martin,
Org. Lett., 2002, 4, 4527.
5 T. H. Chan and P. Brownbridge, J. Am. Chem. Soc., 1980, 102, 3534.
6 T. Mukaiyama, Org. React., 1982, 28, 203.
7 W. Lehnert, Tetrahedron Lett., 1970, 11, 4723.
1
8 On the basis of H NMR studies 6 was initially assigned as the enol-
form of 5. This mis-assignment was corrected when crystals of 5b and 6b
suitable for X-ray analysis were obtained. Single crystal X-ray analysis
also confirmed the structures of the other tetrahydropyran-4-ones
synthesized.
Paul A. Clarke,* William H. C. Martin, Jason M. Hargreaves,
Claire Wilson and Alexander J. Blake
School of Chemistry, University of Nottingham, University Park,
Nottingham, Notts, UK NG7 2RD.
E-mail: paul.clarke@nottingham.ac.uk; Fax: +44 115 9513564;
Tel: 44 115 9513566
9 L. Shao, H. Kawano, M. Saburi and Y. Uchida, Tetrahedron, 1993, 49,
1997.
10 See supporting information for details.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 1061–1063 | 1063