Regioselective synthesis of substituted o-alkoxyphenol derivatives through thermal benzannulation of Fischer (alkenylcyclobutenyl)carbene complexes

Article abstract of DOI:10.1021/jo0264574

A convenient, regioselective, and general synthetic method for producing highly substituted o-phenol-containing polycycles from Fischer (alkenylcyclobutenyl)carbene complexes has been described. The starting complexes have been synthesized by means of the [2 + 2] cycloaddition reaction of (alkenylethynyl)carbene complexes and a range of enol ethers, and in most cases, they have proven to be stable at room temperature and therefore isolable. The key step of the synthesis consists of the thermal benzannulation reaction of these novel pentacarbonyl dienyl Fischer complexes, which is an unprecedented transformation for these kinds of complexes. The unexpected behavior of (alkenylcyclobutenyl)carbene complexes has been rationalized in terms of their geometries.

Full text of DOI:10.1021/jo0264574

Regioselective Syn th esis of Su bstitu ted o-Alk oxyp h en ol
Der iva tives th r ou gh Th er m a l Ben za n n u la tion of F isch er
(Alk en ylcyclobu ten yl)ca r ben e Com p lexes
J ose´ Barluenga,* Fernando Aznar, and M. AÄ ngel Palomero
Instituto Universitario de Quı´mica Organometa´lica “Enrique Moles”, Unidad Asociada al CSIC,
Universidad de Oviedo, J ulia´n Claverı´a 8, 33071 Oviedo, Spain
barluenga@sauron.quimica.uniovi.es
Received September 18, 2002
A convenient, regioselective, and general synthetic method for producing highly substituted o-phenol-
containing polycycles from Fischer (alkenylcyclobutenyl)carbene complexes has been described. The
starting complexes have been synthesized by means of the [2 + 2] cycloaddition reaction of
(alkenylethynyl)carbene complexes and a range of enol ethers, and in most cases, they have proven
to be stable at room temperature and therefore isolable. The key step of the synthesis consists of
the thermal benzannulation reaction of these novel pentacarbonyl dienyl Fischer complexes, which
is an unprecedented transformation for these kinds of complexes. The unexpected behavior of
(alkenylcyclobutenyl)carbene complexes has been rationalized in terms of their geometries.
In tr od u ction
The coupling reaction of Fischer R,â-unsaturated com-
plexes and alkynes (Do¨tz reaction) provides another
particularly noteworthy annulation reaction leading to
phenol derivatives.9 The process involves the initial
insertion of the alkyne molecule in the metal-carbon
double bond to produce unstable dienylcarbene complex
intermediates (also known as metallahexatrienes) that
finally afford p-alkoxyphenols by successive CO insertion
and electrocyclization.10 Given that the process is initi-
ated by the thermally or photochemically promoted
dissociation of a CO ligand in the metal coordination
sphere, the resulting dienyl complexes are coordinatively
unsaturated tetracarbonyl species (Scheme 1).
Pentacarbonyl analogues to the above-mentioned in-
termediates of the Do¨tz reaction have also been described.
These systems are accessible by several synthetic routes
and have proven to be potentially interesting substrates
for organic synthesis since they undergo selective and
high-yielding transformations in mild conditions.11 The
photochemically driven benzannulation reaction of chro-
mium dienylcarbene complexes is one of the most distinc-
tive processes of pentacarbonylmetallahexatrienes and
Substituted and fused phenols are commonly observed
structural units in many natural products,1 which has
resulted in the development of several synthetic ap-
proaches to these compounds over the past several
decades. The lack of regioselectivity shown by the clas-
sical methodologies involving the transformation of phe-
nolic precursors2 has been conveniently overcome by
novel procedures that are based on selective annulations
of acyclic starting materials.3 In this context, Diels-
Alder4 and several other transition-metal-catalyzed cy-
cloaddition reactions,5 Robinson annelations,6 electrocy-
clizations of dienylketene intermediates,7 or cycloaro-
matizations8 should be pointed out as convenient and
widely employed routes to substituted phenols.
* To whom correspondence should be addressed. Fax: + 34 98 510
34 46.
(1) (a) Ferruginol, Ohashi, M.; Maruishi, T.; Kakisama, H. Tetra-
hedron Lett. 1968, 719. (b) Hinokiol, Matsumoto, T.; Usui, S.; Ka-
mashima, H. Bull. Chem. Soc. J pn. 1981, 54, 581. (c) J uncusol: J acobi,
P. A.; Kravitz, J . I.; Zheng, W. J . Org. Chem. 1995, 60, 376.
(2) (a) Olah, G. A. Acc. Chem. Res. 1971, 4, 240. (b) Hartshorn, S.
R. Chem. Soc. Rev. 1974, 3, 167. (c) Bunnett, J . F. Acc. Chem. Res.
1978, 11, 413. (d) Chander, B.; Green, L.; Snieckus, V. Pure Appl.
Chem. 1999, 71, 1521.
(3) Reviews: (a) Bamfield, P.; Gordon, P. F. Chem. Soc. Rev. 1984,
13, 441. (b) Katritzky, A. R.; Li, J .; Xie, L. Tetrahedron 1999, 55, 8263.
(4) (a) Oppolzer, W. In Comprehensive Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 5, p 315.
(b) Andersen, N. G.; Maddaford, S. P.; Keay, B. A. J . Org. Chem. 1996,
61, 2885 and references therein.
(5) Review: Saito, S.; Yamamoto, Y. Chem. Rev. 2000, 100, 2901.
(6) (a) Gawley, R. E. Synthesis 1976, 777. (b) Banwell, M. G.;
Lambert, J . N.; Corbett, N.; Greenwood, R. J .; Gulbes, J . M.; Macky,
M. F. J . Chem. Soc., Perkin Trans. 1 1992, 1415. (c) Abad, A.; Agullo,
C.; Arno, M.; Cunat, A. C., Garc´ıa, T.; Zaragoza, R. J . J . Org. Chem.
1996, 61, 5916.
(7) (a) Danheiser, R. L.; Brisbois, R. G.; Kowalczyc, J . J .; Miller, R.
F. J . Am. Chem. Soc. 1990, 112, 3093. (b) Loebach, J . L.; Bennet, D.
M.; Danheiser, R. L. J . Org. Chem. 1998, 63, 8380.
(8) (a) Reviews: J unjappa, H.; Ila, H.; Asokan, C. V. Tetrahedron
1990, 46, 5423. J unjappa, H.; Ila, H.; Barun, O. J . Organomet. Chem.
2001, 64, 34. (b) Barun, O.; Nandi, S.; Panda, K.; Ila, H.; J unjappa, H.
J . Org. Chem. 2002, 67, 5398.
(9) (a) Do¨tz, K. H. Angew. Chem. 1975, 87, 674; Angew. Chem., Int.
Ed. Engl. 1975, 14, 644. (b) Recent review: Do¨tz, K. H.; Tomuschat,
P. Chem. Soc. Rev. 1999, 28, 187.
(10) For general reviews of the coupling reactions of Fischer carbene
complexes and alkynes, see: (a) Wulff, W. D. In Comprehensive
Organometallic Chemistry II; Abel, E. W., Stone, A. F. G., Wilkinson,
G., Eds.; Pergamon Press: New York, 1994; Vol. 12, p 469. (b) Harvey,
D. F.; Sigano, D. M. Chem. Rev. 1995, 96, 271. (c) Do¨rwald, F. Z. Metal
Carbenes in Organic Synthesis; Wiley-VCH: New York, 1999. (d) de
Meijere, A.; Schirmer, H.; Duetsch, M. Angew. Chem. 2000, 112, 4124;
Angew. Chem., Int. Ed. 2000, 39, 3964.
(11) For a recent review of the reactivity of dienylcarbene complexes,
see: Aumann R. Eur. J . Org. Chem. 2000, 17.
10.1021/jo0264574 CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/19/2002
J . Org. Chem. 2003, 68, 537-544
537

Barluenga et al.
SCHEME 1
SCHEME 2
TABLE 1. Gen er a liza tion of th e Ca r ben e Com p lex.
Syn th esis of 1-Meta lla -1,3,5-h exa tr ien es 3a -m
t
conversn yield
entry
1
M
W
R1
Me
R2
2
3
(h)
(%)
(%)
1
2
3
4
5
6
7
8
9
1a
Ph
Ph
2a 3a
2a 3b
2a 3c
2a 3d
72
24
12
4
12
5
100
100
100
100
100
100
100
77
71
100
85
69
52
64
83
85
93
60
79
87
52
64
58
88
50
79
1b Cr Me
1c
1d Cr
1e Cr -CH2CH2CH2CH2- 2a 3e
1f
W
-CH2CH2CH2-
-CH2CH2CH2-
provides another synthetic route to phenol derivatives
(o-alkoxy-substituted, in this case). However, complexes
in which one or both double bonds belong to aromatic
rings have mostly been employed as starting materials.12
cis-Metallahexatrienes in which none of the conjugated
double bonds are contained in an aromatic ring exhibit
much lower thermal stability and have to be isolated, in
those cases in which this is possible, at low temperature.
Cyclopentannulation is the most characteristic and com-
monly observed evolution path for this kind of metalla-
hexatriene (Scheme 1).13
W
W
Me
Me
H
Ph
Ph
Ph
2b 3f
2b 3g
1a
24
1b Cr Me
1g Cr
2b 3h 120
2b 3i 144
H
10 1c
11 1d Cr
12 1e Cr -CH2CH2CH2CH2- 2b 3l
13 1h Cr Me
W
-CH2CH2CH2-
-CH2CH2CH2-
2b 3j
2b 3k
48
72
72
-CH2OAlla
2b 3m 48
a
All ) CH2CHdCH2.
In the course of the investigation of the reactivity of
Fischer (alkenylethynyl)carbene complexes toward 2-
amino-1,3-butadienes, we discovered a cascade double or
triple tandem ([4 + 2] cycloaddition-cyclopentannula-
tion) process involving unstable metallahexatrienes de-
rived from the initial [4 + 2] cycloaddition and therefore
containing a cyclohexene moiety in their structure.14 We
decided to extend this methodology to the preparation of
polycycles starting from cyclobutene-containing dienyl
complexes, but [2 + 2] cycloadducts, arising from the
reaction of (alkenylethynyl)carbene complexes and cyclic
enol ethers, were unreactive at room temperature. In-
stead of cyclopentannulation, these adducts underwent
benzannulation when they were heated, thus giving rise
to o-alkoxyphenol derivatives in a completely regioselec-
tive fashion and with good yields.15
We envisioned the potential usefulness of this meth-
odology for the synthesis of fused substituted phenol
derivatives, and therefore, in this paper, we report the
comprehensive study carried out to establish the general-
ity and limitations of both the synthesis of stable cy-
clobutene-containing metallahexatrienes and their ther-
mal benzannulation. In addition, we have carried out the
transformation of the synthesized o-alkoxyphenol deriva-
tives into the corresponding catechols, and proposed a
rationalization for the odd behavior of these novel met-
allahexatrienes.
Resu lts a n d Discu ssion
P r ep a r a tion of Alk en ylcyclobu ten yl(a lk oxy)ca r -
ben e Com p lexes. The enol ethers initially chosen for
the synthesis of (alkenylcyclobutenyl)carbene complexes
were commercially available 2,3-dihydrofuran (2a ) and
3,4-dihydro-2H-pyran (2b), given that the latter was
known to undergo [2 + 2] cycloaddition to Fischer alkynyl
complexes.16 According to the general procedure described
in the literature for this type of process, reactions of
(alkenylethynyl)carbene complexes 1 and enol ethers 2a
and 2b were performed at room temperature, under a
nitrogen atmosphere, and using an excess of olefin as
solvent (Scheme 2, Table 1). Metallahexatrienes 3a -m
were obtained as single products in moderate to good
yields and purified by column chromatography. In all the
cases studied, compounds 3a -m were stable at room
temperature, which seemed surprising since dienylcar-
bene complexes arising from complexes 1, either by [4 +
(12) (a) Merlic, C. A.; Xu, D. J . Am. Chem. Soc. 1991, 113, 7418. (b)
Merlic, C. A.; Roberts, W. M. Tetrahedron Lett. 1993, 34, 7379. (c)
Merlic, C. A.; Xu, D.; Gladstone, B. G. J . Org. Chem. 1993, 58, 538.
(13) (a) Meyer, A. G.; Aumann, R. Synlett 1994, 1011. (b) Aumann,
R.; J asper, B.; Fro¨hlich, R. Organometallics 1994, 14, 231. (c) Aumann,
R.; Meyer, A. G.; Fro¨hlich, R. Organometallics 1995, 15, 5018. (d)
Barluenga, J .; Toma´s, M.; Lo´pez-Pelegr´ın, J . A.; Rubio, E. Tetrahedron
Lett. 1995, 38, 3981. (e) Aumann, R.; Go¨ttker-Schnetmann, I. J .;
Wibbeling, B.; Fro¨hlich, R. Tetrahedron Lett. 1998, 39, 795. (f)
Barluenga, J .; Trabanco, A. A.; Flo´rez, J .; Garc´ıa-Granda, S.; Llorca,
M. A. J . Am. Chem. Soc. 1998, 120, 12129. (g) Barluenga, J .; Lo´pez,
L. A.; Mart´ınez, S.; Toma´s, M. Synlett 1999, 219. (h) Aumann, R.;
Fro¨hlich, R.; Prigge, J .; Meyer, O. Organometallics 1999, 18, 1369. (i)
Wu, H. P.; Aumann, R.; Vennedunker, S.; Saarenketo, P. Eur. J . Org.
Chem. 2000, 3463. (j) Wu, H. P.; Aumann, R.; Fro¨hlich, R.; Wegelius,
E.; Saavenketo, P. Organometallics 2000, 19, 2373. (k) Wu, H. P.;
Aumann, R.; Fro¨hlich, R.; Wegelius, E. Organometallics 2001, 20, 2183.
(l) Aumann, R.; Go¨ttker-Schnetmann, I.; Fro¨hlich, R.; Saarenketo, P.;
Holst, C. Chem.sEur. J . 2001, 7, 711.
(14) (a) Barluenga, J .; Aznar, F.; Barluenga, S.; Garc´ıa-Granda, S.;
Alvarez-Ru´a, C. Synlett 1995, 1040. (b) Barluenga, J .; Aznar, F.;
Barluenga, S.; Ferna´ndez, M.; Martin, A.; Garc´ıa-Granda, S.; Pin˜era-
Nicola´s, A. Chem.sEur. J . 1998, 4, 2280. (c) Another example of a
cascade process involving (alkenylethynyl)complexes 1 and nucleo-
philes: Wu, H. P.; Aumann, R.; Fro¨hlich, R.; Saarenketo, P. Chem.s
Eur. J . 2001, 7, 700.
(15) Preliminary communication: Barluenga, J .; Aznar, F.; Palo-
mero, M. A.; Barluenga, S. Org. Lett. 1999, 1, 541
(16) Wulff, W. D.; Faron, K. L.; Su, J .; Springer, J . P.; Rheingold,
A. L. J . Chem. Soc., Perkin Trans. 1 1999, 197.
538 J . Org. Chem., Vol. 68, No. 2, 2003

Synthesis of Substituted o-Alkoxyphenol Derivatives
TABLE 2. Gen er a liza tion of th e En ol Eth er . Syn th esis
of 1-Meta lla -1,3,5-h exa tr ien es 3n -y
t
yielda
(%)
enttry
1
M
R1
R2
Ph
2
3
(h)
1
2
3
4
5
6
7
8
9
1b
1d
1c
1c
1d
1b
1d
1c
1d
1d
1b
1d
Cr Me
2c
2c
2d
2e
2e
2f
2f
2g
2h
2i
3n
3o
3p
3q
3r
48
48
48
12
6
68
80
66
82
77
67
78
58b
68
c
Cr
W
W
Cr
-CH2CH2CH2-
-CH2CH2CH2-
-CH2CH2CH2-
-CH2CH2CH2-
Cr Me
Ph
3s
3t
9
3
Cr
W
Cr
Cr
Cr Me
Cr
-CH2CH2CH2-
-CH2CH2CH2-
-CH2CH2CH2-
-CH2CH2CH2-
3u
3v
3w
3x
3y
72
6
12
12
3
F IGURE 1.
10
11
12
Ph
2j
2j
93
98
2] cycloaddition with 1,3-butadienes or from addition of
nucleophiles to their â-position, had never been isolated
at such a temperature due to their fast transformation
into cyclopentadiene derivatives under the reaction
conditions.13h-l,14
-CH2CH2CH2-
a
b
Conversion ) 100%. Conversion ) 50%. c Reaction carried
out in dichloromethane (concentration of 1d 0.1 M); the cycload-
duct could not be isolated and had to be directly used in the
following step.
Complexes 3a -m were obtained in typically long
reaction times; in some cases, it was not possible to reach
total conversion of the starting material even after the
reaction mixture was stirrred for several days (Table 1,
entries 8, 9, and 11-14). As indicated in Table 1,
cycloadditions were faster and proceeded to completion
when 2,3-dihydrofuran was employed (Table 1, entries
1-5). The use of tungsten starting complexes led to total
conversion in those cases in which the olefin was 3,4-
dihydro-2H-pyran (Table 1, entries 6, 7, and 10), which
was the only noticeable influence of the metal in the
cycloaddition process. The structures of compounds 3
SCHEME 3
Th er m a l Rea ctivity of Alk en ylcyclobu ten yl(a l-
k oxy)ca r ben e Com p lexes. Syn th esis of o-Meth oxy-
p h en ol Der iva tives. In a preliminary experiment,
solutions of compounds 3j and 3k in tetrahydrofuran
(THF) were refluxed until complete disappearance of the
starting complex, and unexpectedly, o-methoxyphenol
derivative 4g, the structure of which was elucidated on
the basis of its 1H NMR and 13C NMR analyses, was
obtained after conventional workup of the reactions
(Scheme 3, Table 3, entries 9 and 10). The process
involves a CO insertion followed by cyclization, and
therefore, it is very closely related to the known photo-
chemical benzannulation of chromium dienylcarbene
complexes that leads to the same aromatic compounds
(with regiochemistry complementary to that of the com-
pounds obtained by the Do¨tz reaction).12 However, this
result constitutes the first example of a thermal benzan-
nulation reaction of a pentacarbonyl(alkoxy)metallatriene
described in the literature.15,18
1
were assigned on the basis of their H NMR and 13C NMR
spectra, and the stereoselectivity in the formation of the
fused cycles was determined to be cis by NOE experi-
ments on compound 3j.
Once it had been established that this methodology
enabled the introduction of a wide range of substituents
in the terminal double bonds of metallahexatrienes 3, by
changing alkynyl complexes 1, we investigated the
generality of the process concerning the enol ether, by
testing the reaction of systems 2c-j depicted in Figure
1 with carbene complexes 1. Cycloadditions were carried
out in the same experimental conditions described above
with the exception of that involving 1,1-dimethoxyethene
(2i), which was performed in a solution of dichlo-
romethane because it was too exothermic to be carried
out neat (Scheme 2, Table 2, entry 10). When cyclic enol
ethers were employed, the process was again more
favored for five-membered systems than for their six-
membered analogues (Table 2, entries 2, 3, 7, and 8).
Actually, it was necessary to use tungsten alkynyl
complexes to accomplish the preparation of metalla-
trienes derived from alkenes 2d and 2g (Table 2, entries
3 and 8).
Complexes 3 were heated in the same experimental
conditions, furnishing polycycles 4 as major products in
most cases (Scheme 3, Table 3). A range of solvents could
be used in the benzannulation reaction, but the best
results were obtained in refluxing THF. As shown in
Table 4, when solutions of metallatriene 3j or 3k in THF
were heated at lower temperatures than the boiling point
All cycloaducts 3n -y, the same as previously described
compounds 3a -m , were isolated in moderate to high
yields by flash chromatography and were found to be
stable at room temperature, excluding 3w , resulting from
the reaction of 1d and 2i, which decomposed under the
standard purification conditions and had to be directly
used in the next reaction step (Table 1, entry 10). The
best results were observed when 1,1,2,2-tetramethoxy-
ethene (2j) was used, given that very good yields of stable
metallahexatrienes 3x,y were achieved in short reaction
times (Table 2, entries 11 and 12).17
(17) The use of 2j in the [2 + 2] cycloaddition reaction of Fischer
carbene complexes has been previously reported: Camps, F.; Llebar´ıa,
A.; Moreto´, J . M.; Ricart, S.; Vin˜as, J . M. Tetrahedron Lett. 1990, 31,
2479.
(18) Thermal benzannulation of a tetracarbonyl-1,4,5-η3-dienylcar-
bene complex has been previously observed: Barluenga, J .; Aznar, F.;
Gutie´rrez, I.; Mart´ın, A.; Garc´ıa-Granda, S.; Llorca-Baragan˜o, M. A.
J . Am. Chem. Soc. 2000, 122, 1314. Other examples of thermal
benzannulation of pentacarbonyldienylcarbene complexes were re-
ported after the preliminary communication of this work: (a) Wu, H.
P.; Aumann, R.; Fro¨hlich, R.; Wibbeling, B. Eur. J . Org. Chem. 2000,
1183. (b) Mathur, P.; Ghosh, S.; Sarkar, A.; Rheingold, A. L.; Guzei, I.
Tetrahedron 2000, 56, 4995.
J . Org. Chem, Vol. 68, No. 2, 2003 539

Barluenga et al.
TABLE 3. Th er m a l Ben za n n u la tion of
Meta lla h exa tr ien es 3. Syn th esis of P h en ol Der iva tives 4
SCHEME 4
ta
(h)
yielda
(%)
entry
complex
4
1
2
3
4
5
6
7
8
8
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
3p
3q
3r
4a
4a
4b
4b
4c
4d
4e
4e
4f
4g
4g
4h
4i
4j
4k
4l
4m
4m
4n
4o
4p
4q
12
7
9
4
6
9
12
6
7 (12)
9 (20)
8 (12)
9 (12)
3
2
3
9
89
72
77
56
46
79
60
55
56 (32)
59 (58)
72 (26)
69 (44)
77
90
95
70
83
SCHEME 5
9
10
11
12
13
14
15
16
17
18
19
20
21
12
acetals 2i and 2j, decomposed when treated in the same
conditions. Although both chromium and tungsten com-
plexes provided entry to the same compounds 4, the latter
required longer reaction times to be totally consumed,
probably due to their higher stability (Table 3, entries
1-4, 9, and 10).
Complexes 3s and 3t bearing a dioxol moiety in their
structure were particularly interesting substrates to
perform the benzannulation, given that the deprotection
of the acetal functionality in the final aromatic com-
pounds 4n and 4o gave entry to 1,2-diols, which could
be further modified. The hydrolysis of the acetal group
was carried out with trifluoroacetic acid (TFA) in THF,
and compounds 5 were obtained with high yields (Scheme
4).
3
6
4
6
8
90
72
67
65
3s
3t
3u
3v
72
a
Data in parentheses refer to the reactions carried out without
nitrogen purge.
TABLE 4. Effect of th e Solven t a n d Tem p er a tu r ea
3j 3k
reaction
conditions
yield
(%)
t
yield
(%)
t
(h)
(h)
THF, reflux
THF, 40 °C
hexane, reflux
MeCN, 65 °C
58
15
19
32
20
48
48
10
26
20
25
25
12
28
24
8
We finally focused our attention on the preparation of
the final products 4 in a one-pot process from alkynyl
complexes 1 and enol ethers 2. The refluxing of a solution
of 1 and different proportions of 2 in THF under a purge
of nitrogen only afforded complex mixtures of products
resulting from undesirable lateral thermal processes.
However, the purification of metallahexatrienes 3 could
be achieved by simple removal of the excess enol ether,
once the cycloadduct was formed, followed by addition
of THF and heating in the standard conditions. Polycycle
4g was synthesized by this procedure in a 48% or 51%
overall yield, starting from chromium or tungsten com-
plexes, respectively.
a
Yields of compound 4g isolated after total conversion of the
starting complexes. The reaction times also refer to total conver-
sion of the starting materials.
of THF, a noticeable increase in the reaction times, which
implied partial decomposition of the starting complex,
was observed. The use of a nonpolar solvent, such as
hexane, led to the same results. However, in the more
coordinating solvent acetonitrile, even though the disap-
pearance of the starting material was slightly acceler-
ated, a larger amount of byproducts was obtained, thus
diminishing the efficiency of the process.
It has already been remarked that tetracarbonyl- and
pentacarbonylmetallahexatrienes usually show different
reactivity, and consequently, the behavior of dienylcar-
bene complexes is expected to be strongly influenced by
the presence or absence of CO in the reacting mixture.
With the aim of determining the effect of CO on these
cyclization reactions, we heated complex 3k under a CO
atmosphere, finding that most of the initial carbene
complex was recovered unaltered after 9 days at 60 °C.
On the other hand, when reactions were carried under a
purge of nitrogen, shorter reaction times were needed,
providing better yields of phenols 4 (Scheme 3, Table 3,
entries 8-11). According to data displayed in Table 3,
corresponding to reactions performed in the optimized
conditions, metallahexatrienes arising from the cycload-
dition of complexes 1 and enol ethers 2a -h underwent
the benzannulation process, giving rise to phenol deriva-
tives 4. However, complexes 3w -y, derived from ketene
Syn th esis of Ca tech ol Der iva tives. P r ep a r a tion
a n d Rea ctivity of Allyloxy- a n d Ben zyloxy(a lk yn yl)-
ca r ben e Com p lexes. We envisioned that the synthetic
potential of this methodology leading to phenols could
be enhanced by removal of the methyl moiety in the final
products, given that highly substituted catechols would
be obtained in a regioselective manner. With that pur-
pose, some compounds 4 were treated with boron tribro-
mide according to the methodology described in the
literature,19 and the desired diols 6 were isolated as single
products in high yields (Scheme 5, Table 5). In those cases
in which additional methyl ethers were present in the
substrates, the treatment with boron tribromide led to
the deprotection of all of them, furnishing triol systems
(Table 5, entries 5 and 6). The structures of all catechol
(19) Demuynck, M.; Clercq, P.; Vandewalle, M. J . Org. Chem. 1979,
26, 4863.
540 J . Org. Chem., Vol. 68, No. 2, 2003

Synthesis of Substituted o-Alkoxyphenol Derivatives
TABLE 5. Syn th esis of Ca tech ol Der iva tives 6
ity: allyl-substituted complexes 7a ,b, although not stable
at room temperature for long times, could be isolated in
moderate to good yields. However, the analogous system
7c bearing a benzyl moiety quickly decomposed when
chromatographic purification was attempted and could
not be properly characterized.22 [2 + 2] cycloaddition of
compounds 7 to 2,3-dihydrofuran afforded the desired
(alkenylcyclobutenyl)carbene complexes 8, which not only
showed differential stability, the same as their precursors
7, but also evolved to different products when they were
heated in refluxing THF. Thus, allyloxy complexes 8a ,b
underwent intramolecular diastereoselective [4 + 2]
cycloaddition, giving rise to compound 9 after oxidative
workup (Scheme 6, Table 6, entries 1 and 2).23 Although
it was possible to detect the carbene complexes initially
formed in the [4 + 2] cycloaddition process, their isolation
and chromatographic purification implied a great loss in
the final yield of compound 9, and oxidation by means of
exposure to oxygen and sunlight was carried out in the
crude mixture. The yield of the process was improved
when the reaction was performed in an atmosphere of
carbon monoxide (Table 6, entry 2).
yield
(%)
entry
4
R3
R4
R5
6
R
1
2
3
4
5
4a
4b
4e
4f
4j
H
H
H
H
O(CH2)2
6a
6b
6c
6d
6e
H
H
H
H
93
93
94
88
95
O(CH2)2
O(CH2)3
O(CH2)3
(CH2)3
OMe
OH
SCHEME 6
On the other hand, benzyloxy complex 8c furnished
the expected o-benzyloxyphenol derivative 10 when it was
thermally treated in THF under a purge of nitrogen
(Scheme 6, Table 6, entry 3). Given the instability of both
intermediates involved in this route, neither of them was
isolated: the crude complex 7c was filtered through a
pad of Celite and treated with 2,3-dihydrofuran until
disappearance of the starting material. Then, the excess
olefin was removed in a vacuum, and crude 8c was
directly benzannulated in the usual conditions to yield
o-benzyloxyphenol 10. This compound was quantitatively
transformed into catechol 6b by hydrogenolysis; the
overall yield of catechol derivative accomplished by this
procedure was lower than that of the sequential benzan-
nulation-methyl moiety removal, and the latter syn-
thetic route was preferred for the preparation of the rest
of compounds 6 (Table 5).
TABLE 6. Syn th esis a n d Rea ctivity of Allyloxy- a n d
Ben zyloxyca r ben e Com p lexes 7
yield
(%)
yield
(%)
yield
(%)
entry
7
M
W
P
8
product
1
2
3
7a
allyl
59
77
b
8a
8b
8c
97
79
b
9
9
49a
55a
37d
7b Cr allyl
7c Cr benzyl
10c
a
Reactions performed in a CO atmosphere. When carbene
complex 8b was heated under nitrogen purge, the yield of 9 was
b
41%. Not isolable. c 10 was transformed into 6b by hydrogenoly-
d
sis (98%). Yield of the overall process from crude complex 7c.
Mech a n istic P r op osa l a n d Ra tion a liza tion of th e
Resu lts. Given the similarity of this benzannulation with
both the Do¨tz reaction and the photochemical process
that converts dienylcarbene complexes into o-alkoxyphe-
nol derivatives, it seemed reasonable to propose for it an
analogous mechanism, which is depicted in Scheme 7.
Hence, the reaction would begin with the dissociation of
a CO ligand to afford a tetracarbonylmetallahexatriene
intermediate, A, that would give rise to a cyclohexadi-
enone derivative, C, by successive CO insertion and
electrocyclic ring closure. A keto-enol tautomerism and
the oxidative cleavage of the tricarbonylmetal moiety
attached to the aromatic ring would furnish the final
phenol derivatives 4. It is commonly accepted that the
rate-determining step of this kind of reaction is the
release of the CO ligand in the starting complex, which
in this case would provide a rationalization for the
experimental results. Thus, the nitrogen purge would
facilitate the formation of the tetracarbonyl intermediate
1
derivatives 6 were determined by analysis of their H and
13C NMR spectra.
The employment of 2-haloalkoxy Fischer carbene com-
plexes as hydroxycarbene complex equivalents for the
synthesis of deprotected hydroxyl compounds has been
recently described by our research group,20 so we decided
to apply the same strategy to the direct preparation of
catechol derivatives 6. However, all the attempts directed
toward the synthesis of 2-iodoalkoxy(alkynyl)carbene
complexes were unsuccessful, and we resolved to position
a different protective group in the starting carbene
complex that similarly enabled its removal in the last
step of the synthesis. Chromium and tungsten alkynyl-
carbene complexes 7, bearing an allyl or benzyl moiety,
were synthesized by standard methodology from M(CO)6,
the corresponding alkynyllithium, and allyl or benzyl
triflate (Scheme 6, Table 6).21 The nature of the protective
group attached to the oxygen atom in the carbene
complex seemed to have a great influence on its stabil-
(22) The slow decomposition of alkynyl(allyloxy)carbene complexes
has been reported. See ref 21.
(20) Barluenga, J .; Lo´pez, S.; Trabanco, A. A.; Flo´rez, J . Chem.s
Eur. J . 2001, 7, 4723.
(21) Christoffers, J .; Do¨tz, K. H. Organometallics 1994, 13, 4189.
(23) For an analogous intramolecular [4 + 2] cycoaddition reaction
of Fischer carbene complexes, see: Do¨tz, K. H.; Noack, R.; Harms, K.
Tetrahedron 1990, 46, 1235.
J . Org. Chem, Vol. 68, No. 2, 2003 541

Barluenga et al.
SCHEME 7
Con clu sion s
In summary, we have applied the [2 + 2] cycloaddition
reaction of Fischer carbenes and enol ethers to alkenyl-
ethynyl complexes to prepare cyclobutenyl-containing
1-metalla-1,3,5-hexatrienes. These complexes have shown
unexpected stability at room temperature, in contrast to
previously reported dienylcarbene complexes, and un-
usual reactivity. Upon thermal treatment, they have
furnished o-alkoxyphenol derivatives through a process
that involves CO insertion and subsequent cyclization
and that did not have any precedent in the literature.
We believe that the geometrical restraints introduced in
the structure of the starting complexes by the cyclobutene
ring are responsible for both the stability of the starting
systems and the unusual reactivity they have shown.
The regioselectivity with which the reaction proceeds,
as well as the broad range of possibilities of substitution
in the starting substrates, convert this process into a good
synthetic method of o-methoxyphenol derivatives. These
compounds can be easily transformed into the corre-
sponding catechols by reaction with boron tribromide,
therefore increasing the potential of the methodology.
SCHEME 8
Exp er im en ta l Section
Gen er a l Meth od s. For all reactions run under
a N2
atmosphere, THF, diethyl ether, dichloromethane, toluene, and
methanol were dried and distilled by standard procedures
before use. Solvents used in column chromatography were
distilled prior to use. All other reagents used in the reactions
were of the best commercial grade available. Column chroma-
tography was carried out on silica gel 60 (230-400 mesh). Rf
data are referred to the mixture of solvents in which column
chromatography was carried out. All melting points are
uncorrected. NMR spectra were recorded at 300 or 200 MHz
for 1H and 75 or 50.3 MHz for 13C, with tetramethylsilane as
internal standard for 1H and the residual solvent signals as
standard for 13C. Chemical shifts are given in parts per million.
The multiplicity of the signals is indicated as follows: s )
singlet, d ) doblet, t ) triplet, q ) quadruplet, quint )
quintuplet, m ) multiplet. Mass spectra were obtained by EI
(70 eV), FAB+, or MALDI-TOF. IR spectra are given in inverse
centimeters.
P r ep a r a tion of (Alk en yleth yn yl)- a n d (Ar yleth yn yl)-
ca r ben e Com p lexes 1. Fischer carbene complexes 1 have
already been synthesized in our research group by standard
methodology from the corresponding acetylenes. Their analyti-
cal and spectroscopic data have been reported.14a,25
P r ep a r a tion of En ol Eth er s a n d Keten e Aceta ls 2. 2,3-
Dihydrofuran (2a ), 3,4-dihydro-2H-pyran (2b ), and 2-meth-
oxypropene (2h ) are commercially available reagents. The rest
of the enol ethers and ketene acetals were prepared according
to methods described in the literature: 1-methoxycyclopentene
(2c), 1-methoxycyclohexenene (2d ), and 1-methoxycyclohep-
tene (2e),26 2,2-dimethyl-1,3-dioxol (2f),27 2H-1,3-dioxine (2g),28
1,1-dimethoxyethene (2i),29 and 1,1,2,2-tetramethoxyethene
(2j).30
by the removal of CO and, consequently, accelerate the
reaction. On the other hand, the presence of CO would
make the key intermediate revert to the starting pen-
tacarbonyl species, avoiding the whole process.
A possible explanation for the different behavior
observed for metallatrienes 3 and those generated by
[4 + 2] cycloaddition of 2-amino-1,3-butadienes and
complexes 1 could be based on their geometries. Our
proposal is depicted in Scheme 8: we can assume that
tetracarbonyl intermediate A is in equilibrium with
metallacyclohexadiene E, resulting from its electrocy-
clization. Given that reductive elimination of E to afford
F is very disfavored due to the large angles existing
between the substituents of the cyclobutene moiety, the
equilibrium is displaced to the starting complex A, which
evolves through an alternative path leading to phenol
derivative 4. An identical explanation can be applied to
pentacarbonyl complexes 3. However, it is commonly
accepted that the cyclopentannulation of such compounds
takes place through an initial nucleophilic attack of the
terminal double bond to the carbene carbon atom to
afford intermediate G.13a,14a In the case of metallatrienes
3, this interaction is not likely to occur since the distance
between carbons C1 and C5 is considerably longer than
that of their cyclohexene-containing analogues (Scheme
8).24 Moreover, that fact would justify the stability of the
(alkenylcyclobutenyl)carbene complexes at room temper-
ature.
(24) Simple molecular mechanics modeling showed that the differ-
ence between the distances C1-C5 in metallatrienes 3 and 5 is about
0.7 Å (molecular mechanics calculations were carried out using the
MM2 force field; atom distance data for the pentacarbonylchromium
carbene moiety were taken from an X-ray structure of a similar
compound and kept fixed along the minimization).
(25) Barluenga, J .; Aznar, F.; Palomero, M. A. Chem.sEur. J . 2001,
7, 5318.
(26) (a) Rhoads, S. J .; Chattopadhyay, J . K.; Waali, E. E. J . Org.
Chem. 1950, 35, 3355. (b) Gassman, P. G.; Burns, S. J .; Pfister, K. B.
J . Org. Chem. 1993, 58, 1449.
(27) Posner, G. H.; Nelson, T. D. Tetrahedron 1990, 46, 4573.
(28) Groth, U.; Scho¨llkopf, U.; Tiller, T. Tetrahedron 1991, 47, 2835.
(29) Zheng, Q. H.; Su, J . Synth. Commun. 1999, 29, 3467.
542 J . Org. Chem., Vol. 68, No. 2, 2003

Synthesis of Substituted o-Alkoxyphenol Derivatives
Gen er a l P r oced u r e for th e P r ep a r a tion of F isch er
Dien yl Com p lexes. All dienylcarbene complexes 3 were
synthesized by the same experimental procedure previously
described by Wulff:16 1 mmol of carbene neat with 10 mmol of
the corresponding enol ether were stirred at room temperature,
under a nitrogen atmosphere, until TLC analyses revealed that
the starting complex had been consumed. Removal of the
volatiles by vacuum (10 mmHg) and flash chromatography
afforded the title compounds. Exceptionally, when 1,1-
dimethoxyethene was used, 3 mmol of the cited olefin and 1
mmol of complex 1d were dissolved in 10 mL of dry dichlo-
romethane, and the mixture was stirred under a nitrogen
atmosphere at room temperature until total consumption of
the starting carbene complex was observed. The resulting
adduct 3w could not be isolated by column chromatography,
so it was directly used in the next step after removal of solvents
in a vacuum. Data for compounds 3a -d and 3t,v,y have
already been reported.15,31
formation of the corresponding cycloadduct. At that point, the
remaining olefin was removed in a vacuum (10 mmHg), dry
THF was added (10 mL), and the resulting solution was heated
under a nitrogen purge. Phenol derivatives obtained by this
procedure were purified in the same manner previously
described.
Da ta for 2,3,5b,7,8,8a -h exa h yd r o-5-m eth oxy-1H-in d en -
[4′,5′:3,4]cyclobu t a [1,2-b]fu r a n -4-ol (4b): white solid; mp
155-157 °C; yield 56% or 77%; Rf ) 0.18 (hexane:diethyl ether:
dichloromethane ) 6:1:1); 1H NMR (400 MHz, CDCl3) δ 1.70-
1.88 (m, 2H), 2.09 (quint, J ) 7.2 Hz, 2H), 2.71 (t, J ) 7.2 Hz,
2H), 2.83 (t, J ) 7.2 Hz, 2H), 3.63-3.70 (m, 1H), 3.93 (dd, J )
7.8, 3.6 Hz, 1H), 4.01 (s, 3H), 4.10 (t, J ) 8.0 Hz, 1H), 5.44 (s,
1H), 5.58 (d, J ) 3.6 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ
26.0 (CH2), 28.8 (CH2), 28.9 (CH2), 29.4 (CH2), 47.7 (CH), 58.1
(CH3), 66.8 (CH2), 80.5 (CH), 124.2 (C), 130.3 (C), 132.0 (C),
132.9 (C), 140.0 (C), 140.7 (C); HRMS m/z calcd for C14H16O3
232.1099, found 232.1096; LRMS (EI) m/z 232 (75), 217 (100),
189 (27), 171 (7), 128 (8), 91 (5). Anal. Calcd for C14H16O3: C,
72.38; H, 6.95. Found: C, 72.32; H, 6.94.
Da ta for 2,3,5b,8a -tetr a h yd r o-5-m eth oxy-7,7-d im eth yl-
1H-in d en o[4′,5′:3,4]cyclobu ta [1,2-d ][1,3]d ioxol-4-ol (4o):
light yellow solid; mp 142-143 °C; yield 67%; Rf ) 0.32
(hexane:ethyl acetate ) 5:1); 1H NMR (300 MHz, CDCl3) δ 1.11
(s, 3H), 1.47 (s, 3H), 2.08-2.19 (m, 2H), 2.77-2.89 (m, 4H),
4.06 (s, 3H), 5.57 (s, 3H), 5.69 (d, J ) 3.6 Hz, 1H), 5.85 (d,
J ) 3.6 Hz, 1H); 13C NMR (75 MHz, CDCl3) δ 26.0 (CH2), 28.0
(CH3), 28.8 (CH3), 29.0 (CH2), 29.6 (CH2), 58.2 (CH3), 80.8 (CH),
81.0 (CH), 115.4 (C), 127.8 (C), 132.4 (C), 133.6 (C), 133.7 (C),
139.4 (C), 142.2 (C); HRMS m/z calcd for C15H18O4 262.1205,
found 262.1208; LRMS (EI) m/z 262 (12), 204 (38), 189 (100),
187 (9), 97 (25). Anal. Calcd for C15H18O4: C, 68.68; H, 6.92.
Found: C, 68.47; H, 7.03.
Gen er a l P r oced u r e for th e P r ep a r a tion of Der iva tives
5. Compounds 4n ,o (0.3 mmol) dissolved in a mixture THF/
H2O (10:1) were treated with an excess of trifluoroacetic acid
(2 mL) at room temperature. After 12 h, volatiles were removed
in high vacuum (0.1 mmHg), and the residue was redissolved
in dichloromethane, washed with 0.5 M NaOH and saturated
NaCl solutions, dried over anhydrous Na2SO4, and chromato-
graphed in a 1:1 mixture of hexane and ethyl acetate to yield
diols 5.
Da t a for p en t a ca r b on yl{[2,3,3a ,5a -t et r a h yd r o-4-(1-
m et h yl-2-p h en ylet h en yl)cyclob u t a [b]fu r a n -5-yl]m et h -
oxym eth ylen e}tu n gsten (0) (3a ): dark orange solid; mp 92-
94 °C; yield 64%; Rf ) 0.44 (hexane:diethyl ether:dichloro-
methane ) 4:1:1); 1H NMR (200 MHz, CDCl3) δ 1.82-2.07 (m,
2H), 1.93 (s, 3H), 3.54 (dd, J ) 7.7, 3.8 Hz, 1H), 3.90 (td, J )
8.9, 5.6 Hz, 1H), 4.17 (t, J ) 7.9 Hz, 1H), 4.62 (s, 3H), 5.57 (d,
J ) 3.8 Hz, 1H), 6.92 (br s, 1H), 7.32-7.41 (m, 5H); 13C NMR
(50.3 MHz, CDCl3) δ 16.6 (CH3), 27.4 (CH2), 43.9 (CH), 66.6
(CH2), 68.2 (CH3), 79.1 (CH), 127.9 (CH), 128.3 (2CH), 129.4
(2CH), 131.2 (C), 136.2 (CH), 136.7 (C), 143.7 (C), 151.2 (C),
196.9 (4C), 203.4 (C), 310.7 (C); IR (CH2Cl2) ν 2066, 1937 cm-1
.
Anal. Calcd for C22H18O7W: C, 45.67; H, 3.14. Found: C, 45.42;
H, 3.28.
Da ta for p en ta ca r bon yl{[6-(cyclop en ten yl)-1-m eth oxy-
b icyclo[3.2.0]h ep t a -6-en -7-yl]m et h oxym et h ylen e}ch r o-
m iu m (0) (3o): orange oil; yield 80%; Rf ) 0.61 (hexane:diethyl
ether:dichloromethane ) 4:1:1); 1H NMR (300 MHz, CDCl3) δ
1.30-2.70 (m, 12H), 2.90-3.40 (m, 1H), 3.23 (s, 3H), 4.66 (s,
3H), 6.12 (br s, 1H); 13C NMR (75 MHz, CDCl3) δ 23.6 (CH2),
23.6 (CH2), 24.7 (CH2), 31.2 (CH2), 33.1 (CH2), 33.2 (CH2), 47.8
(CH), 53.0 (CH3), 65.7 (CH3), 94.5 (CH), 136.9 (C), 137.3 (C),
139.0 (CH), 149.6 (C), 216.2 (4C), 223.8 (C), 340.4 (C); IR (CH2-
Cl2) ν 2058, 1942 cm-1. Anal. Calcd for C20H20CrO7: C, 56.61;
H, 4.75. Found: C, 56.83; H, 5.03.
Data for cis-2-m eth oxy-5-m eth yl-4-ph en ylbicyclo[4.2.0]-
octa -1,3,5-tr ien e-3,7,8-tr iol (5a ): white solid; mp 142-144
°C; yield 91%; Rf ) 0.18 (hexane:ethyl acetate ) 1:1); 1H NMR
(300 MHz, CDCl3) δ 1.97 (s, 3H), 4.08 (s, 3H), 5.06 (d, J ) 3.1
Hz, 1H), 5.13 (d, J ) 3.1 Hz, 1H), 5.49 (s, 1H), 7.20-7.26 (m,
2H), 7.34-7.46 (m, 3H); 13C NMR (75 MHz, CDCl3) δ 14.6
(CH3), 58.0 (CH3), 72.5 (CH), 72.8 (CH), 125.9 (C), 127.3 (CH),
128.2 (C), 128.4 (2CH), 129.7 (2CH), 131.2 (C), 136.0 (C), 136.1
Da ta for p en ta ca r bon yl{[4-(1-cyclop en ten yl)-3a ,5a -d i-
h ydr o-2,2-dim eth ylcyclobu ta[d]dioxol-5-yl]m eth oxym eth -
ylen e}ch r om iu m (0) (3t): dark red solid; mp 93-95 °C; yield
78%; Rf ) 0.30 (hexane:dichloromethane ) 5:1); 1H NMR (300
MHz, CDCl3) δ 1.43 (s, 3H), 1.86-2.03 (m, 2H), 2.15-2.26 (m,
1H), 2.33-2.39 (m, 1H), 2.39-2.51 (m, 2H), 4.75 (s, 3H), 5.08
(d, J ) 3.7 Hz, 1H), 5.84 (d, J ) 3.7 Hz, 1H), 6.47-6.48 (m,
1H); 13C NMR (75 MHz, CDCl3) δ 23.5 (CH2), 28.4 (CH3), 29.0
(CH3), 33.0 (CH2), 33.4 (CH2), 66.0 (CH3), 75.9 (CH), 81.0 (CH),
137.0 (C), 137.4 (C), 143.3 (CH), 150.3 (C), 215.9 (4C), 223.6
(C), 336.5 (C); IR (CH2Cl2) ν 2057, 1938 cm-1. Anal. Calcd for
(C), 140.2 (C), 143.4 (C); HRMS m/z calcd for
C16H16O4
272.1049, found 272.1053; LRMS (EI) m/z 272 (23), 270 (54),
257 (68), 241 (11). Anal. Calcd for C16H16O4: C, 70.57; H, 5.92.
Found: C, 70.68; H, 6.11.
C
19H18CrO8: C, 53.53; H, 4.26. Found: C, 53.81; H, 4.50.
Gen er a l P r oced u r e for th e P r ep a r a tion of Ca tech ol
Der iva tives 6. A solution of compound 4 (0.2 mmol) in dry
dichloromethane (5 mL) was treated with an excess of boron
tribromide (1.75 mmol, 1 mL of a 1.75 M solution in hexane)
at -78 °C. The reaction mixture was allowed to reach room
temperature over 6-7 h and then quenched with aqueous
saturated NaHCO3. The organic product was extracted twice
with dichloromethane and ethyl acetate and chromatographed
in a 1:1 mixture of hexane and ethyl acetate.
Da ta for 2,3,5b,7,8,8a -h exa h yd r o-1H-in d en [4′,5′:3,4]cy-
clobu ta [1,2-b]fu r a n -4,5-d iol (6b): white solid; mp 155-157
°C; yield 93%; Rf ) 0.19 (hexane:ethyl acetate ) 1:1); 1H NMR
(300 MHz, DMSO-d6) δ 1.48-1.61 (m, 1H), 1.74 (dd, J ) 12.2,
4.8 Hz, 1H), 1.94 (quint, J ) 7.4 Hz, 2H), 2.48-2.68 (m, 4H),
3.35-3.43 (m, 1H), 3.73 (dd, J ) 7.7, 3.4 Hz, 1H), 3.93 (t, J )
8.5 Hz, 1H), 5.35 (d, J ) 3.4 Hz, 1H), 8.01 (br s, 1H), 8.98 (br
s, 1H); 13C NMR (75 MHz, DMSO-d6) δ 25.6 (CH2), 28.4 (CH2),
29.1 (CH2), 29.4 (CH2), 46.4 (CH), 66.0 (CH2), 79.1 (CH), 125.9
Gen er a l P r oced u r e for th e Ben za n n u la tion P r ocess.
F r om Meta lla h exa tr ien es 3. In an optimized procedure, a
solution of complex 3 (0.5 mmol) in dry THF (25 mL) was
refluxed under a nitrogen purge until TLC analyses revealed
total comsumption of the starting material. For reactions
arising from chromium complexes, addition of hexane (50 mL)
and exposure to sunlight and air for 12 h, followed by flash
cromatoghaphy, afforded the title compounds. For tungsten
complexes, the residue was loaded directly onto a silica gel
column without exposure to light and air.
F r om Alk yn yl Com p lexes 1. Complex 1 was dissolved in
an excess of enol ether 2 (0.5 mL), and the mixture was stirred,
under a nitrogen atmosphere, until TLC analysis revealed the
(30) Scheeren, J . W.; Staps, R. J . F. M.; Nivard, F. R. J . Recl. Trav.
Chim. Pays-Bas 1953, 92, 11.
(31) Barluenga, J .; Aznar, F.; Palomero, M. A. Angew. Chem. 2000,
112, 4514; Angew. Chem., Int. Ed. 2000, 39, 4346.
J . Org. Chem, Vol. 68, No. 2, 2003 543

Barluenga et al.
(C), 129.2 (C), 129.5 (C), 133.8 (C), 138.0 (C), 141.1 (C); HRMS
m/z calcd for C13H14O3 218.0943, found 218.0944; LRMS (EI)
m/z 218 (100), 203 (33), 201 (17), 190 (49), 171 (16). Anal. Calcd
for C13H14O3: C, 71.54; H, 6.47. Found: C, 71.91; H, 6.22.
P r ep a r a tion of Com p lexes 7. Butyllithium (1.6 M in
hexanes, 3.3 mmol) was added dropwise to a solution of
chromium or tungsten hexacarbonyl (3 mmol) and acetylene
(3.3 mmol) in 10 mL of dry THF at -80 °C. The mixture was
allowed to reach room temperature, and the solvent was
removed in a vacuum and replaced by dry dichloromethane
(10 mL). Allyl or benzyl triflate was prepared by adding the
corresponding alcohol (4 mmol) to a solution of pyridine (4.2
mmol) and trifluoromethanesulfonic anhydride (4.2 mmol) in
dry dichloromethane (20 mL) at -25 °C. The mixture was
stirred at that temperature for 10 min and then poured via
cannula over the solution containing the lithium acylmetalate
at -10 °C. Despite the fact that complexes 7a ,b slowly
decomposed both neat and in solution, they could be purified
by flash chromatography in hexane and stored at -5 °C for
short times. However, complex 8c was filtrated through a thin
pad of Celite and directly used in the subsequent cycloaddition
reaction.
13.8, 6.0 Hz, 1H), 5.36 (dd, J ) 13.8, 6.0 Hz, 1H), 5.44 (d, J )
11.4 Hz, 1H), 5.54 (d, J ) 12.5 Hz, 1H), 6.13-6.26 (m, 1H),
6.46 (br s, 1H); 13C NMR (75 MHz, CDCl3) δ 23.6 (CH2), 27.0
(CH2), 33.4 (CH2), 33.6 (CH2), 44.5 (CH), 66.3 (CH2), 80.2 (CH),
83.2 (CH2), 120.5 (CH2), 131.3 (CH), 139.1 (C), 140.6 (C), 143.7
(CH), 148.6 (C), 197.1 (4C), 203.2 (C), 305.0 (C); IR (CH2Cl2) ν
2057, 1941 cm-1; LRMS (MALDI-TOF) m/z calcd for C20H18O7W
554, found 554. Anal. Calcd for C20H18O7W: C, 43.34; H, 3.27.
Found: C, 43.65; H, 2.97.
Da ta for (()-(1S,2R,6S,12S,13S)-3,15-d ioxa bicyclo[11.-
3.1.01,7.02,6.08,12]h ep ta d eca n -7-en -16-on e (9): colorless oil;
1
yield 49% or 55%; Rf ) 0.23 (hexane:ethyl acetate ) 3:1); H
NMR (300 MHz, CDCl3) δ 1.06-1.25 (m, 3H), 1.60-1.88 (m,
4H), 1.89-2.09 (m, 2H), 2.26-2.39 (m, 2H), 2.85-2.93 (m, 1H),
3.74 (td, J ) 5.4, 2.6 Hz, 1H), 3.91 (dd, J ) 9.4, 6.0 Hz, 1H),
4.07 (t, J ) 8.0 Hz, 1H), 4.13-4.22 (m, 1H), 4.47 (t, J ) 9.4
Hz, 1H), 4.84 (d, J ) 4.7 Hz, 1H); 13C NMR (75 MHz, CDCl3)
δ 25.4 (CH2), 27.9 (CH2), 28.9 (CH2), 31.6 (CH), 32.2 (CH2),
36.2 (CH2), 36.4 (CH), 49.6 (CH), 56.0 (C), 70.5 (CH2), 71.5
(CH2), 81.7 (CH), 126.0 (C), 142.5 (C), 178.6 (C); IR (CH2Cl2)
ν 1752 (s) cm-1; LRMS (MALDI-TOF) m/z calcd for C15H18O3
246, found 246. Anal. Calcd for C15H18O3: C, 73.15; H, 7.37.
Found: C, 73.02; H, 7.22.
Da t a for {a llyloxy[(1-cyclop en t en yl)et h yn yl]m et h yl-
en e}p en ta ca r bon ylch r om iu m (0) (7b): red oil; yield 77%;
Da ta for 5-ben zyloxy-2,3,5b,7,8,8a -h exa h yd r o-1H-in -
d en [4′,5′:3,4]cyclobu ta [1,2-b]fu r a n -4-ol (10): yellow oil;
yield 37%; Rf ) 0.36 (hexane:ethyl acetate ) 5:1); 1H NMR
(300 MHz, CDCl3) δ 1.71-1.82 (m, 1H), 1.87 (dd, J ) 12.2, 5.1
Hz, 1H), 2.11 (quint, J ) 7.4 Hz, 2H), 2.73 (t, J ) 7.4 Hz, 2H),
2.86 (t, J ) 7.4 Hz, 2H), 3.65-3.74 (m, 1H), 3.90 (dd, J ) 7.7,
3.4 Hz, 1H), 4.10 (dd, J ) 8.8, 7.7 Hz, 1H), 5.27 (d, J ) 12.0
Hz, 1H), 5.34 (d, J ) 12.0 Hz, 1H), 5.40 (d, J ) 3.4 Hz, 1H),
5.52 (br s, 1H), 7.35-7.45 (m, 5H); 13C NMR (75 MHz, CDCl3)
δ 26.2 (CH2), 28.8 (CH2), 29.0 (CH2), 29.4 (CH2), 47.5 (CH),
66.9 (CH2), 72.4 (CH2), 80.5 (CH), 124.7 (C), 127.4 (2CH), 128.0
(CH), 128.5 (2CH), 130.2 (C); 132.4 (C), 133.1 (C), 137.2 (C),
139.1 (C), 140.9 (C); LRMS (MALDI-TOF) m/z calcd for
1
Rf ) 0.27 (hexane); H NMR (300 MHz, CDCl3) δ 1.87-2.06
(m, 2H), 2.42-2.44 (m, 2H), 2.58-2.61 (m, 2H), 5.09 (d, J )
3.6 Hz, 2H), 5.33 (d, J ) 19.4 Hz, 1H), 5.42 (d, J ) 10.8 Hz,
1H), 5.50-6.07 (m, 1H), 6.50-6.52 (m, 1H); 13C NMR (75 MHz,
CDCl3) δ 23.2 (CH2), 34.5 (CH2), 35.6 (CH2), 79.5 (CH2), 93.5
(C), 119.5 (CH2), 124.1 (C), 131.1 (CH), 134.3 (C), 147.7 (CH),
216.1 (4C), 225.7 (C), 313.4 (C); IR (CH2Cl2) ν 2061, 1939 cm-1
.
No satisfactory elemental analysis could be achieved for
compound 7b due to its slow decomposition.
P r ep a r a tion of Com p lexes 8 a n d Com p ou n d s 9 a n d
10. Alkynyl complexes 7 were dissolved in 2,3-dihydrofuran
and stirred at room temperature, under a nitrogen atmo-
sphere, until TLC analyses revealed that the starting complex
had been consumed (1-2 h), and then the volatiles were
removed in a vacuum (10 mmHg). Flash chromatography in a
1:1 mixture of hexane and dichloromethane enabled the
purification of compounds 8a ,b, but 8c was directly used in
the next reaction step. For the preparation of 9, a solution of
complexes 8a ,b (0.5 mmol) in THF (5 mL) was degassed by
means of three freeze-pump-thaw cycles, then the atmo-
sphere was saturated with CO, and the mixture was heated
in an oil bath at 65 °C for 6 h. The crude mixture was diluted
2-fold with hexane, exposed to air and sunlight, filtered, and
chromatographed. Thermal benzannulation of cycloadduct 8c
to afford 10 was carried out in the standard conditions.
Da ta for {a llyloxy[4-(1-cyclop en ten yl)-2,3,3a ,5a -tetr a -
h yd r ocyclobu ta [b]fu r a n -5-yl]m eth ylen e}p en ta ca r bon yl-
tu n gsten (0) (8a ): orange oil; yield 97%; Rf ) 0.45 (hexane:
C
20H20O3 308, found 308. Anal. Calcd for C20H20O3: C, 77.90;
H, 6.54. Found: C, 78.02; H, 6.88.
Hyd r ogen olysis of 10. A mixture of compound 10 (0.2
mmol), ethanol (10 mL), and catalytical amounts of Pd/C (10%)
was stirred under 1 atm of hydrogen for 24 h. Filtration
through a pad of Celite and flash chromatography afforded
6b in 98% yield.
Ack n ow led gm en t. This research was supported by
the DGICYT (Grant PB97-1271). M.A.P. gratefully
acknowledges the MEC and University of Oviedo for
predoctoral fellowships.
Su p p or tin g In for m a tion Ava ila ble: Analytical and spec-
troscopic data for all new compounds prepared and 13C NMR
spectra of compounds 7a and 7b. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
dichloromethane ) 1:1); H NMR (300 MHz, CDCl3) δ 1.71-
1.98 (m, 5H), 2.30-2.43 (m, 4H), 3.31 (dd, J ) 7.7, 1.1 Hz,
1H), 3.79-3.87 (m, 1H), 4.11 (t, J ) 8.1 Hz, 1H), 5.30 (dd, J )
J O0264574
544 J . Org. Chem., Vol. 68, No. 2, 2003