Synthesis of conjugated γ- and δ-lactones from aldehydes and ketones via a vinylation/allylation-ring closing metathesis-oxidation sequence

Article abstract of DOI:10.1055/s-1999-2897

Nucleophilic C-vinylation or C-allylation of aldehydes and ketones followed by O-allylation of the obtained carbinols gave the corresponding allyl ethers. Ring-closing metathesis of the latter afforded cyclic ethers (dihydrofurans and dihydropyrans, respectively) which were then subjected to allylic oxidation to yield conjugated γ- and δ-lactones.

Full text of DOI:10.1055/s-1999-2897

LETTER
1639
Synthesis of Conjugated g- and d-Lactones from Aldehydes and Ketones via a
Vinylation/Allylation-Ring Closing Metathesis-Oxidation Sequence
Miguel Carda,*a Encarnación Castillo,a Santiago Rodríguez,a Santiago Uriel,a J. Alberto Marco*b
aDepart. de Q. Inorgánica y Orgánica, Univ. Jaume I, Castellón, Spain
Fax (+34)-964-728214; E-mail: mcarda@qio.uji.es;
bDepart. de Q. Orgánica, Univ. de Valencia, c/D. Moliner, 50, 46100 Burjassot, Valencia, Spain
Fax (+34)-96-3864328; e-mail: alberto.marco@uv.es
Received 5 August 1999
Abstract: Nucleophilic C-vinylation or C-allylation of aldehydes
and ketones followed by O-allylation of the obtained carbinols gave
the corresponding allyl ethers. Ring-closing metathesis of the latter
afforded cyclic ethers (dihydrofurans and dihydropyrans, respec-
tively) which were then subjected to allylic oxidation to yield con-
jugated g- and d-lactones.
Key words: Barbier allylation, ring-closing metathesis, allylic oxi-
dation, conjugated lactones.
For some time, we have been interested in the synthesis of
chiral natural products using easily available chirons,
most particularly erythrulose, as starting materials.1
Among the different types of structures that have attracted
our attention are (–)-malyngolide 1,2 (Z)-topostines (e.g.
2)3 and other saturated and unsaturated d-lactones with
potential pharmacological application.
As a method to construct the lactone ring, we initially con-
sidered a ring closing metathesis (RCM)4 in the ester of an
allyl carbinol with an acrylic-type acid using the well es-
tablished catalysts 3 or 4 (see scheme below). This would
yield a conjugated d-lactone which, if desired, should pro-
vide a saturated lactone by subsequent hydrogenation,
Michael addition or similar functional manipulations. The
overall reaction sequence would represent a synthetic
We then acylated the alcohol group of the allylation prod-
equivalent of the general d 4 synthon depicted below. The
ucts 6 with acryloyl or methacryloyl chloride to yield the
feasibility of this sequence has been very recently estab-
lished for allyl and homoallyl acrylates to yield conjugat-
corresponding esters 7. However, and in contrast with that
observed with acrylates 7 (R3, R4 = H),5 compounds 7 (R3
ed g- and d-lactones, respectively, with a disubstituted
or R4 π H) did not undergo RCM with catalyst 3, neither
C=C bond.5 In view of our ongoing synthetic projects
after a prolonged reaction time nor after addition of Lewis
mentioned above, our purpose was to test the generality of
acids.5,8 In view of the failure to obtain lactones 8 (R3 or
this procedure for the preparation of conjugated lactones
with a higher degree of substitution at the C=C bond.
R4 π H) by this procedure,9 we considered an alternative
method based on O-allylation of allyl carbinols 6 and
A number of aldehydes and ketones 5a-f was thus allowed
to react with allyl bromide or methallyl chloride and zinc
in an aqueous medium.6 The expected allyl carbinols 6 (R3
= H or Me) were obtained in all cases with good yields.7
RCM of the resulting ethers 9 (R3, R4 = H or Me) to dihy-
dropyrans 10,10 followed by allylic oxidation of the latter
compounds (Scheme 1).
Synlett 1999, No. 10, 1639–1641 ISSN 0936-5214 © Thieme Stuttgart · New York

1640
M. Carda et al.
LETTER
other hand, the molybdenum catalyst 4 gave rise to RCM
on ether 12b (R3 = Me, R4 = H) to yield the trisubstituted
olefin 13b (R3 = Me, R4 = H), precursor in turn of b-me-
thyl-g-lactone 14b (R3 = Me, R4 = H). In contrast, 4 failed
to promote the formation of the trisubstituted olefin 13b
(R3 = H, R4 = Me). As regards the allylic oxidation step, it
is worth mentioning that, while it was successful with 13a
(R3 = R4 = H, see Table 2), this was not the case with 13a
(R3 = H, R4 = Me), where mixtures of oxidation products
at the methylene and/or at the tertiary benzylic position
were obtained. Most likely, the additional methyl group
sterically retards the oxidation of the methylene carbon to
such an extent that attack at the methine carbon becomes
competitive.
Scheme 1
As a matter of fact, O-allylation of carbinols 6 to allyl
ethers 9 transcurred uneventfully. RCM of the latter com-
pounds using 3 afforded cyclic ethers 10 in very good
yields. Trisubstituted C=C bonds (10, R3 or R4 = Me) were
also formed under these conditions.11,12 Unfortunately,
however, tetrasubstituted olefins (10, R3 = R4 = Me) were
not obtained, even when using the more reactive Schrock
molybdenum catalyst 4.4c Finally, oxidation of the allylic
methylene next to the oxygen atom in cyclic ethers 10 was
achieved using the chromium trioxide complexes with
pyridine13 or 3,5-dimethylpyrazole.14 While both reagents
gave similar yields, CrO3/3,5-dimethylpyrazole was reac-
tive at lower temperatures, a feature of interest in the case
of sensitive substrates (see yields in Table 1).15-17
Scheme 2
The synthetic sequence described above constitutes there-
fore a useful procedure for the preparation of conjugated
g- and d-lactones of various structural types, which may
then be converted into other types of lactones by suitable
functional manipulation. We are currently investigating
the synthesis of some naturally occurring lactones with
the aid of this protocol. Our results will be reported in due
course.
We also tested the feasibility of the preparation of conju-
gated five-membered lactones with the aforementioned
procedure (Scheme 2). C-alkenylation of compounds 5
with either vinyl or isopropenyl magnesium bromide af-
forded allyl alcohols 11, which were then subjected to the
same allylation/metathesis/oxidation sequence. Here
again, the yields of the metathesis steps (Table 2) were
very high in all cases where the reaction worked.15-17 As
expected,10 catalyst 3 was able to promote RCM in bisal-
lyl ethers 12 (R3 = R4 = H) to yield dihydrofurans 13,
which were then converted by allylic oxidation into con-
jugated g-lactones 14 (R3 = R4 = H). However, the same
catalyst was much less effective with ethers 12 when R3 or
R4 π H. Only with aldehyde 5a as the starting materials did
the procedure work in some cases (see Table 2). On the
Acknowledgement
This work has been financially supported by the DGICYT (Project
PB96-0760) and by the Conselleria de Cultura de la Generalitat Va-
lenciana (project GV97-CB-11-77). E.C. thanks the latter instituti-
on for a pre-doctoral fellowship.
Synlett 1999, No. 10, 1639–1641 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Synthesis of Conjugated Lactones via Ring-Closing Metathesis
1641
worth mentioning here that we have had good results working
at a concentration of about 0.1M.
References and Notes
(1) Marco, J.A.; Carda, M.; González, F.; Rodríguez, S.; Castillo,
E.; Murga, J. J.Org.Chem. 1998, 63, 698.
(2) Cardillina II, J.H.; Moore, R.E.; Arnold, E.V.; Clardy, J.
J.Org. Chem. 1979, 44, 4039.
(3) Fujiwara, H.; Egawa, S.; Terao, Y.; Aoyama, T.; Shioiri, T.
Tetrahedron 1998, 54, 565.
(4) For two recent reviews on olefin metathesis, see: (a)
Armstrong, S.K. J.Chem.Soc.Perkin Trans. I, 1998, 371. (b)
Grubbs, R.H.; Chang, S. Tetrahedron 1998, 54, 4413.
(c) Schrock, R. R. Tetrahedron 1999, 55, 8141.
(5) (a) Ghosh, A.K.; Cappiello, J.; Shin, D. Tetrahedron Lett.
1998, 4651. (b) Bassindale, M.J.; Hamley, P.; Leitner, A.;
Harrity, J.P.A. Tetrahedron Lett. 1999, 3247.
(6) Carda, M.; Castillo, E.; Rodríguez, S.; Murga, J.; Marco, J.A.
Tetrahedron: Asymmetry 1998, 9, 1117. Zinc worked well
here and was preferred to indium because of its much lower
price. These results are part of the Ph.D. Thesis of E.C. and
will be published with detail in due course.
(7) (+)-Camphor 5e did not react under aqueous Barbier-type
conditions. Reaction with allyl magnesium bromide in THF
gave 15 (∫ 6e, R3 = H) which was then converted into lactones
16 (∫ 8e, R3 = R4 = H) and 17 (∫ 8e, R3 = H, R4 = Me) by the
sequence described in this paper. Erythrulose derivative 5f
reacted under Barbier-type conditions to yield a ca. 3:1
mixture of allylation products, with 18 (∫ 6f, R3 = H) being the
major isomer. The depicted configuration of the quaternary
stereogenic centre has been recently confirmed by an X-ray
diffraction analysis of lactone 19 (∫ 8f, R3 = R4 = H)
(unpublished result).
(12) As indicated in Table 1, RCM of diolefin 9a (R3 = H, R4 =
Me), prepared from aldehyde 5a, was unsuccessful. Only
starting material was recovered together with a small amount
of a product which was identified as the dimer resulting from
a intermolecular metathesis on the monosubstituted double
bond. Similar results were observed with olefins 12a and 12c
(R3 = Me, R4 = H) (Table 2). There is precedent for this type
of behaviour.4
(13) Dauben, W.G.; Lorber, M.; Fullerton, D.S. J.Org.Chem. 1969,
34, 3857.
(14) Salmond, W.G.; Barta, M.A.; Havens, J.L. J.Org.Chem. 1978,
43, 2057.
(15) General conditions of metathesis. A) With catalyst 3: the
substrate (1 mmol) was dissolved in dry, degassed CH2Cl2 (10
ml) and treated under Ar with 3 (25 mg, ca. 0.03 mmol). After
stirring overnight at reflux, the reaction mixture was conc. in
a rotary evaporator and chromatographed on silica gel (elution
with hexane-EtOAc 9:1). B) With catalyst 4: 23 mg of the
catalyst (ca. 0.03 mmol) was weighed in a dry box and placed
in a flame-dried flask under argon. The substrate (0.5 mmol)
was dissolved in dry, degassed benzene (5 ml) and added via
syringe to the flask containing the catalyst. After stirring for
24 h at reflux, the reaction mixture was concentrated in a
rotary evaporator and chromatographed on silica gel (elution
with hexane-EtOAc 9:1).
(16) General conditions for allylic oxidations: chromium trioxide
(1.2 g, 12 mmol) was suspended under Ar in dry CH2Cl2 (10
ml), cooled to -20º and treated rapidly at this temp. with 3,5-
dimethylpyrazole (12 mmol). After stirring the mixture for 15
min., the substrate (1 mmol) dissolved in dry CH2Cl2 (1 ml)
was added. The reaction mixture was then stirred for 1 hr at
the same temperature, treated with 5M aqueous NaOH (5 ml)
and further stirred at 0º for 1 hr. The reaction mixture was then
poured into diluted HCl, and the organic layer was washed
with brine, dried on anhydrous Na2SO4, filtered, conc. in a
rotary evaporator and chromatographed on silica gel (elution
with hexane-EtOAc 9:1).
(17) Physical and spectral data of lactone 16: white solid, mp 95-
96º, [a]D22 -80 (CHCl3; c, 5.2); IR nmax 1710, 1381, 1054, 997
cm-1; UV l max 228 nm; HREIMS, m/z (rel. int.): 220.1465
(M+, 33), 205 (30), 177 (32), 159 (32), 137 (50), 111 (100),
108 (94), 95 (94). Calc. for C14H20O2, M = 220.1463; 1H NMR
(CDCl3, 400 MHz): d = 6.80 (1H, ddd, J = 9.5, 6.5, 2 Hz), 5.95
(1H, dd, J = 9.5, 2.8 Hz), 2.56 (1H, ddd, J = 15.5, 2.8, 2 Hz),
2.20 (2H, m), 1.75-1.65 (2H, m), 1.50-1.40 (2H, m), 1.34 (1H,
ddd, J = 14, 9, 3.5 Hz), 1.10 (3H, s), 0.95 (1H, ddd, J = 12, 8.8,
5.3 Hz), 0.91 (3H, s), 0.85 (3H, s); 13C NMR (CDCl3, 100
MHz): d = 164.2, 90.6, 52.4, 49.3 (C), 144.4, 121.6, 45.0
(CH), 45.5, 31.6, 30.0, 26.3 (CH2), 21.0, 20.9, 10.8 (CH3).
Physical and spectral data of lactone 17: white solid, mp 72-
73º, [a]D22 -74 (CHCl3; c, 6); IR nmax 1703, 1362, 1123, 1055
cm-1; UV lmax 232 nm; HREIMS, m/z (rel. int.): 234.1620
(M+, 34), 219 (9), 191 (8), 173 (22), 151 (28), 125 (60), 108
(34), 95 (100). Calc. for C15H22O2, M = 234.1620; 1H NMR
(CDCl3, 400 MHz): d = 6.51 (1H, br dd, J = 6.5, 2 Hz), 2.62
(1H, br dd, J = 17.5, 2 Hz), 2.20 (2H, m), 1.90 (3H, br s), 1.80-
1.70 (2H, m), 1.50-1.40 (2H, m), 1.36 (1H, ddd, J = 14, 9, 3.5
Hz), 1.16 (3H, s), 1.00 (1H, ddd, J = 12, 9, 5.5 Hz), 0.96 (3H,
s), 0.90 (3H, s); 13C NMR (CDCl3, 100 MHz): d = 165.7,
128.7, 90.6, 52.4, 49.4 (C), 138.1, 45.1 (CH), 45.6, 32.1, 30.0,
26.5 (CH2), 21.1, 21.0, 16.7, 10.9 (CH3).
(8) Fürstner, A.; Langemann, K. J.Am.Chem.Soc. 1997, 119,
9130.
(9) For a similar situation, see also: Cossy, J.; Bauer, D.; Bellosta,
V. Tetrahedron Lett. 1999, 4187. That the direct preparation
of lactones with a trisubstituted C=C bond via RCM of
acrylate-type esters is not feasible was further shown here in
the case of lactone 8c (R3 = Me, R4 = H), which could not be
obtained from the corresponding acrylate 7c but was easily
prepared, however, via the allylation/RCM/oxidation
sequence (Table 1):
(10) (a) Delgado, M.; Martín, J.D. Tetrahedron Lett. 1997, 6299.
(b) Maier, M.E.; Bugl, M. Synlett 1998, 1390.
(11) (a) Kirkland, T.A.; Grubbs, R.H. J.Org.Chem. 1997, 62, 7310.
(b) Chang, S.; Grubbs, R.H. J.Org.Chem. 1998, 63, 864. It is
Article Identifier:
1437-2096,E;1999,0,10,1639,1641,ftx,en;L09999ST.pdf
Synlett 1999, No. 10, 1639–1641 ISSN 0936-5214 © Thieme Stuttgart · New York