A novel transition metal-catalyzed route to functionalized dihydropyrans and tetrahydrooxepines

Article abstract of DOI:10.1055/s-1998-1598

A straightforward route for the synthesis of α,α′-disubstituted dihydropyrans and tetrahydrooxepines has been developed involving Pd- and Ru-catalyzed reactions.

Full text of DOI:10.1055/s-1998-1598

192
LETTERS
SYNLETT
A Novel Transition Metal-Catalyzed Route to Functionalized Dihydropyrans and
Tetrahydrooxepines
a
Floris P. J. T. Rutjes,* T. Martijn Kooistra, Henk Hiemstra and Hans E. Schoemaker
Laboratory of Organic Chemistry, Amsterdam Institute of Molecular Studies, University of Amsterdam, Nieuwe Achtergracht 129, 1018 WS Amster-
dam, The Netherlands
Fax +31 20 525 5670; e-mail: florisr@org.chem.uva.nl
a
DSM Research, P.O. Box 18, 6160 MD Geleen, The Netherlands
Received 3 November 1997
Abstract
: A straightforward route for the synthesis of α,α'-disubstituted
dihydropyrans and tetrahydrooxepines has been developed involving
Pd- and Ru-catalyzed reactions.
1
2
2,6-Disubstituted dihydropyrans and (n = 1) occur frequently in
1
nature as a structural unit in a wide range of natural products. In
addition, these compounds could serve as useful building blocks for the
synthesis of biologically active saturated oxygen heterocycles via
functionalization of the double bond. Examples include (poly)cyclic
2
ethers, but also modified carbohydrates and 2-carboxyl-substituted
3
tetrahydropyrans with antibacterial activity such as KDO. More
1
2
specifically, dihydropyrans (R = OMe) have recently been used as
4,5
intermediates in the synthesis of mevinolin and compactin.
In
conjunction with previous work in our group on vinylsilane-terminated
6
formation of dihydropyrans via oxycarbenium ions, we set out to
explore a novel entry into this compound class, which would enable us
to vary the side chains and would also give facile access to enantiopure
heterocycles. Moreover, the methodology outlined in this paper is not
restricted to six-membered ring formation, but can also be applied to
form the corresponding seven-membered rings. Especially the latter
aspect provides a significant benefit compared to previously established
5,7
methods for forming such rings via hetero Diels-Alder reactions.
A
Scheme 2
short retrosynthetic outline of our route is shown in Scheme 1.
further optimization (Pd(OAc) (cat.), Ph P(CH ) PPh (dppp), excess
2
2
2 3
2
of 1-methoxy-1,2-propadiene, Et N, MeCN, 80 °C in a sealed tube) led
3
13
7
to a virtually quantitative yield of the desired acetal
.
Not
7
surprisingly, was obtained as a ca. 1:1 mixture of diastereoisomers,
which could not be separated by column chromatography. The diolefin
7
then cyclized smoothly under standard ring-closing metathesis
conditions (5 mol% of Ru-benzylidene catalyst, 0.1 M solution in
Scheme 1
8
CH Cl , rt), providing the dihydropyran in a satisfactory yield of 74%
2
2
1
2
(cis/trans 1:1). This mixture of isomers was then treated with a Lewis
The target compounds and should be accessible via the allylic acetal
9
acid (BF ·OEt ) to generate the allylic oxycarbenium intermediate ,
3 2
which in the presence of different nucleophiles (allyltrimethylsilane,
3
, which contains (i) a terminal olefin function that can be used as a
8,9
handle for ring-closing metathesis and (ii) an acetal function to enable
trimethylsilyl cyanide and α-(trimethylsilyloxy)styrene reacted to the
later modification of the system via oxycarbenium ion chemistry.
10–12
corresponding 1,2-adducts
in reasonable to good yields and
The first sequence (Scheme 2) commenced with anhydrous methyl
10
excellent trans-selectivity. This selectivity probably arises from
stereoelectronically preferred pseudo-axial attack of the incoming
nucleophile and is more generally found in similar unsaturated
5
glyoxylate ( ),
which upon treatment with BF ·OEt and
3 2
6
allyltrimethylsilane led to methyl 2-hydroxy-4-pentenoate ( ). Initially,
14
6
attempts were made to convert into the desired allylic acetal under
15,16
systems.
acid-catalyzed conditions with acrolein derivatives, but without much
An analogous sequence was developed to arrive at enantiopure six- and
seven-membered ring ethers as depicted in Scheme 3. Starting material
success. Therefore, we turned our attention to other strategies, involving
11
the use of 1-methoxy-1,2-propadiene,
which upon electrophilic
13
was the readily available benzyl-protected (R)-glycidol , which was
activation and subsequent reaction with the secondary alcohol might
give rise to the desired product. Several methods have been published
ring-opened with complete regioselectivity using vinyl- or
allylmagnesium bromide in the presence of Me S and a catalytic
describing activation of allenes towards coupling with oxygen
2
11
14
15
,
amount of CuI to give the corresponding secondary alcohols
and
nucleophiles (including N-bromosuccinimide, HgCl ),
but these
2
respectively. A similar strategy as in the previous scheme, namely
oxypalladation of 1-methoxy-1,2-propadiene (in both cases ca. 1:1
mixture of isomers were formed) and subsequent ring-closing
routes are not attractive in view of the stoichiometric use of the reagents.
At this point, a publication of Alper and coworkers appeared, who
obtained a similar acetal as a byproduct in Pd(II)-catalyzed annulation
reactions with substituted allenes. Application of their conditions and
12
18
19
metathesis provided the oxygen heterocycles and (cis/trans 1:1) in

February 1998
SYNLETT
193
References
1)
2)
3)
4)
For some examples, see e.g.: (a) Norcross, R.D.; Paterson, I.
Chem. Rev. 1995, 95, 2041; (b) Faulkner, D.J. Nat. Prod. Rep.
1997, 14, 259.
(a) Gleason, M.M.; McDonald, F.E. J. Org. Chem. 1997, 62,
6432; (b) Rychnovsky, S.D.; Dahanukar, V.H. J. Org. Chem.
1996, 61, 7648.
(a) Gao, J.; Härter, R.; Gordon, D.M.; Whitesides, G.M. J. Org.
Chem. 1994, 59, 3714; (b) Kragl, U.; Gödde, A.; Wandrey, C.;
Lubin, N.; Augé, C. J. Chem. Soc., Perkin Trans. 1 1994, 119.
(a) Bauer, T.; Chapuis, C.; Jezewski, A.; Kozak, J.; Jurczak, J.
Tetrahedron: Asymm. 1996, 7, 1391; (b) Bauer, T.; Jezewski, A.;
Jurczak, J. Tetrahedron: Asymm. 1996, 7, 1405.
5)
6)
7)
Terada, M.; Mikami, K.; Nakai, T. Tetrahedron Lett. 1991, 32,
935.
Semeyn, C.; Blaauw, R.H.; Hiemstra, H.; Speckamp, W.N. J. Org.
Chem. 1997, 62, 3426.
For a review article, see: Bednarski, M.D.; Lyssikatos, J.P. In
Comprehensive Organic Synthesis; Trost, B.M.; Fleming, I., Eds.;
Pergamon: Oxford, 1991; Vol. 2, 661 and references cited therein.
8)
9)
For reviews, see: (a) Grubbs, R.H.; Miller, S.J.; Fu, G.C. Acc.
Chem. Res. 1995, 28, 446; (b) Schmalz, H.-G. Angew. Chem. Int.
Ed. Engl. 1995, 34, 1833; (c) Mol, J.C.; Ivin K.J. Olefin
Metathesis and Metathesis Polymerization; Academic Press:
London, 1997.
Scheme 3
satisfactory yields. In the six-membered ring series, the oxycarbenium
ion-mediated coupling reactions with allyltrimethylsilane and
trimethylsilyl cyanide proceeded considerably better than in the ester-
substituted cases leading to the desired dihydropyrans 20 and 21 in good
yields. Again, coupling took place selectively at the 2-position, with
complete trans-selectivity.
membered ring oxycarbenium ion did not proceed in high yield, a single
diastereoisomer (22) was obtained whose stereochemistry in analogy
with the six-membered rings was tentatively assigned as trans.
For a selection of recent examples, see: (a) Rutjes, F.P.J.T.;
Schoemaker, H.E. Tetrahedron Lett. 1997, 38, 677; (b) Yang, Z.;
He, Y.; Vourloumis, D.; Vallberg, H.; Nicolaou, K.C. Angew.
Chem., Int. Ed. Engl. 1997, 36, 166; (c) Hammer, K.; Undheim,
K. Tetrahedron 1997, 53, 5925; (d) Fürstner, A.; Langemann, K.
J. Am. Chem. Soc. 1997, 119, 9130 and references cited therein.
15,16
Although the addition to the seven-
16
10) Hook, J.M. Synth. Commun. 1984, 14, 83.
11) For a review article about 1-alkoxy-1,2-propadienes, see: Zimmer,
R. Synthesis, 1993, 167.
12) Okuro, K.; Alper, H. J. Org. Chem. 1997, 62, 1566.
13) All new compounds were obtained in analytically pure form and
1
13
appropriately characterized by IR, H and C NMR, high
resolution mass data and rotational values.
Scheme 4
14) For a discussion of the stereoelectronic aspects of the addition
reaction, see: Deslongchamps, P. Stereoelectronic Effects in
Organic Chemistry; Pergamon: New York, 1983; pp 209-221.
The proposed mechanism of the Pd-catalyzed oxypalladation of
methoxyallene is shown in Scheme 4, where coordination of the Pd(II)-
species with the more electron-rich oxygen-substituted double bond
(viz. 23) renders the allene sufficiently electrophilic to be attacked by
the secondary alcohol. Protonolysis of the resulting vinyl-palladium
species 24 then leads to the desired acetal 25 and regeneration of the
1
15) The trans-relationship was established by comparison of H NMR
NOE data (obtained via irradiation of the α-oxygen hydrogen
atoms) with those of related structures (e.g. ref. 6). For related
coupling reactions with cyclic allylic oxycarbenium ions
selectively giving trans-disubstituted products, see: Postema,
M.H.D. Tetrahedron 1992, 48, 8572-8576 and references cited
therein.
17
Pd(II)-catalyst.
In conclusion, we have developed an efficient route to various 2,6-
trans-disubstituted oxygen heterocycles, which can also be appplied to
form the corresponding seven-membered rings. The pathway involves
(i) a novel Pd(II)-mediated coupling of 1-methoxy-1,2-propadiene with
secondary alcohols and (ii) a Ru-catalyzed ring-closing process as the
key transformations. One could envision that the use of similar allylic
acetals is not only restricted to ring-closing metathesis, but that they
could also be applied in many other types of ring-closing processes. At
present, we are exploring such possibilities including applications in
natural product synthesis, which will be presented in the near future.
16) Data for selected compounds:
11: colorless oil; R = 0.46 (silica, 70% ether in petroleum ether);
f
-1
1
IR (film) ν
2956, 2240, 1745, 1186, 1096 cm ; H NMR (400
max
MHz, CDCl ) δ 6.10-6.15 (m, 1H), 5.74-5.79 (m, 1H), 5.18 (dd, J
3
= 1.7, 3.5 Hz, 1H), 4.50 (dd, J = 5.5, 8.8 Hz, 1H), 3.81 (s, 3H),
13
2.42-2.46 (m, 2H); C NMR (100 MHz, CDCl ) δ 173.0, 127.7,
3
121.0, 116.1, 69.8, 62.6, 52.4, 26.7; HRMS (EI), calcd for
+
C H NO (M ): 167.0582, found 167.0594.
8
9
3
21: colorless oil; R = 0.65 (silica, 70% ether in petroleum ether);
f
22
[α]
-57.4 (c 0.5, CH Cl ); IR (film) ν
1104, 904, 739, 698 cm ; H NMR (400 MHz, CDCl ) δ 7.26-
3031, 2868, 2234,
max
D
2
2
-1
1
3

194
LETTERS
SYNLETT
7.38 (m, 5H), 6.08-6.13 (m, 1H), 5.70-5.74 (m, 1H), 5.04-5.06 (m,
Hz, 1H), 5.04-5.12 (m, 2H), 4.56 (s, 2H), 4.44-4.49 (m, 1H), 4.07
1H), 4.60 (s, 2H), 4.07-4.13 (m, 1H), 3.59 (d, J = 4.5 Hz, 1H),
2.26-2.34 (m, 1H), 2.01-2.08 (m, 2H); C NMR (100 MHz,
(dt, J = 5.1, 10.7 Hz, 1H), 3.57 (dd, J = 6.1, 9.9 Hz, 1H), 3.45 (dd,
13
13
J = 5.0, 9.9 Hz, 1H), 2.21-2.40 (m, 4H), 1.81-1.96 (m, 2H);
C
CDCl ) δ 137.8, 128.9, 128.3, 127.7, 127.6, 120.9, 116.9, 73.3,
NMR (100 MHz, CDCl ) δ 138.3, 135.3, 132.5, 130.7, 128.2,
3
3
+
71.5, 70.7, 62.9, 26.2; HRMS (EI), calcd for C
229.1103, found 229.1107.
H
NO (M ):
127.5, 127.4, 116.6, 75.4, 73.2, 72.4, 71.0, 40.3, 29.9, 26.4;
14 15
2
+
HRMS (FAB), calcd for C
259.1682.
H
O
(M+H ): 259.1698, found
17 23
2
22: colorless oil; R = 0.74 (silica, 70% ether in petroleum ether);
f
22
[α]
-0.5 (c 1, CH Cl ); IR (film) ν
743, 697 cm ; H NMR (400 MHz, CDCl ) δ 7.26-7.35 (m, 5H),
3020, 2927, 2859, 1093,
max
D
2
2
17) For an excellent summary of Pd-chemistry, see: Tsuji, J.
-1
1
3
Palladium Reagents and Catalysts; Wiley: New York, 1995.
5.83-5.93 (m, 1H), 5.70-5.76 (m, 1H), 5.49 (ddd, J = 1.8, 3.8, 11.1