Intramolecular Ni-mediated cyclizations with α,ω-dienals

Article abstract of DOI:10.1002/1099-0690(200104)2001:7<1359::AID-EJOC1359>3.0.CO;2-7

Fifteen α,ω-dienals and one αen-ω-ynal were prepared. These substrates were treated with Ni(COD)₂/TMSC1 under carbonylative and noncarbonylative conditions, with subsequent methanol quenching of the resulting metal intermediates to afford different types of cycloadducts. From the compounds thus obtained, it is concluded that the reaction proceeds through an intermediate (π-ally1)Ni complex, which readily cyclises in a "metallo-ene"-type reaction. The different pathways were characterised by further carbonyl insertion or by protonolysis, resulting either in cyclopentanone formation (insertion followed by (β-elimination) when under carbon monoxide gas, or in single cyclization when not (or, even better, when in the presence of protic acids), Coordination of the distal olefin has been found essential for either of the two processes to proceed. Thus, coordinating heteroatoms (N,S), excessive tether length and particular substituents hampering satisfactory double bond chelation all disrupt the process, with the ketals of the original aldehyde (occasionally as Michael adducts) largely being recovered in such cases.

Full text of DOI:10.1002/1099-0690(200104)2001:7<1359::AID-EJOC1359>3.0.CO;2-7

FULL PAPER
Intramolecular Ni-Mediated Cyclizations with α,ω-Dienals
[a]
[a]
´
´
´
Gabriel Garcıa-Gomez and Josep M. Moreto*
Keywords: Allyl complexes / Carbonylations / Cyclizations / Metallo-ene / Nickel
Fifteen α,ω-dienals and one α-en-ω-ynal were prepared.
These substrates were treated with Ni(COD)2/TMSCl under
carbonylative and noncarbonylative conditions, with sub-
formation (insertion followed by β-elimination) when under
carbon monoxide gas, or in single cyclization when not (or,
even better, when in the presence of protic acids). Coordina-
sequent methanol quenching of the resulting metal interme- tion of the distal olefin has been found essential for either of
diates to afford different types of cycloadducts. From the
the two processes to proceed. Thus, coordinating het-
compounds thus obtained, it is concluded that the reaction eroatoms (N,S), excessive tether length and particular sub-
proceeds through an intermediate (π-allyl)Ni complex, which
readily cyclises in a ‘‘metallo-ene’’-type reaction. The differ-
ent pathways were characterised by further carbonyl inser-
stituents hampering satisfactory double bond chelation all
disrupt the process, with the ketals of the original aldehyde
(occasionally as Michael adducts) largely being recovered in
tion or by protonolysis, resulting either in cyclopentanone such cases.
Introduction
Under CO gas, these complexes react with either strained
or tethered olefins to give an enol ether 3 or a ketal 4 (or a
mixture of both, depending on reaction conditions). After
hydrolysis, a single oxocyclopentenylcarboxaldehyde is ob-
tained from these. We report here on the results of our stud-
ies concerning the carbonylative cycloaddition of olefin-
tethered acrylaldehydes (Scheme 1).
The use of transition metal complexes in organic syn-
thesis has provided not only shortened path sequences to
interesting target molecules, by altering the conventional re-
activity of the organic substrates, but also, often, the
achievement of higher selectivity in the preparation of the
intermediates involved in the development of the synthesis.
It should also be stressed that transition metals are able
to perform more than one synthetic step in a multicompon-
ent single operation (tandem, cascade or domino pro-
cesses), usually operate under very mild conditions and dis-
play catalytic character (atom economy) in many cases.[1]
Recently, one of the most actively studied processes in
this category is the PausonϪKhand reaction, which has
been applied to the synthesis of a variety of bioactive com-
pounds.[2]
Scheme 1. Ni-mediated carbonylative cycloaddition of α,ω-dienals
During the last decade we have been studying the scope
and applications of Ni-mediated carbonylative cycloaddi-
tion of allyl derivatives and alkynes.[3] Among our more re-
cent contributions in the field are the use of a ready-made
and low-risk Ni mediator, and also the replacement of allyl
halides by α,β unsaturated aldehydes, a system in which
acetylenes have also been found to be susceptible to replace-
ment by olefins in the intramolecular version of the pro-
cess.[4] In this version, the allyl derivative (prepared in situ
by addition of a Lewis acid, namely TMSCl, to the alde-
hyde) underwent an oxidative addition with Ni(COD)2 to
afford the corresponding π-allyl complex (Mackenzie com-
plexes).
Preparation of the Substrates and General
Procedure for the Cycloaddition
Sixteen different acrylaldehydes with a distal double
bond (one of them, exceptionally, with a triple bond) were
prepared through the corresponding intermediate allyl alco-
hol 1, according to two general methods. The first involved
treatment of an O-monoprotected (Z)-but-2-ene-1,4-diol
derivative (for 1f and 1g) or the unprotected compound (for
1aϪe, h, i, l, m) with a base (NaH) in THF, followed by
the addition of the corresponding allyl halide (and, when
applicable, deprotection), affording the corresponding alco-
hols 1. The second was based on treatment of the alkoxide
(for 1j and 1k), carbanion (1n), thiol or amine (1o and 1p,
respectively) with an O-protected (Z)-4-halobut-2-en-1-ol
derivative, followed by deprotection, also giving the alco-
hols 1 corresponding to this group of substrates. The re-
sulting alcohols were then oxidised to the corresponding
[a]
IIQAB ‘‘Josep Pascual Vila’’ CID, CSIC,
c/ J. Girona 18Ϫ26, 08034 Barcelona, Spain
Fax: (internat.) ϩ 34-3/204-5904
E-mail: jmmqob@cid.csic.es
Eur. J. Org. Chem. 2001, 1359Ϫ1369
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/0407Ϫ1359 $ 17.50ϩ.50/0
1359

´
´
´
G. Garcıa-Gomez, J. M. Moreto
FULL PAPER
aldehydes 2 by using pyridinium chlorochromate, with Finally, methanol was added to quench the organometallic
yields ranging from 40 to 90%. In the design of these, we intermediate and the contents of the flask were concen-
mainly paid attention to three aspects: the type of tether, trated to dryness. The residue was then worked up as indic-
the length of the tether, and the substitution of the distal ated for each of the cases. The results obtained are given
double bond. These are listed in Figure 1.
in Table 1.
Results and Discussion
The first substrate to be treated with Ni(COD)2 was en-
yne 2a. We had already successfully performed the intramo-
lecular cycloaddition with the corresponding halides[5] and
did not expect to meet any problem in this. However, we
could not isolate any defined product from the analogous
acryl derivatives. In this methodology, the triple bond was
allowed to interact with Ni0 [6] first, since the formation of
the π-allyl complex would be delayed until the introduction
of TMSCl, and reaction would take place without the pro-
tective presence of CO. We attributed our failure to alkyne
polymerization and reasoned that, since olefins were less
reactive towards Ni0 complexes, they would remain un-
altered until the π-allyl complex was formed. Accordingly,
we started out from a 1:1 mixture of 2b and Ni(COD)2 in
toluene, at which point a deep red, insoluble complex was
formed, turning orange after the addition of TMSCl and
Figure 1. Preparation of the substrates
The obtained dienal was added to an equimolar amount the introduction of CO by means of bubbling under atmo-
of Ni(COD)2 in toluene at low temperature. A red precipit- spheric pressure. Further addition of the quencher (meth-
ate (or solution) was thus formed. In this work, we changed anol) and workup of the crude product afforded a 41%
the original stoichiometry (2 mols of acryl derivative per yield of the cyclopentanone 3b. No ketal was produced un-
mol of Ni),[4b] since we supposed that the distal olefin could der these conditions. A single isomer of the enol ether 3b
replace in its ligand role the extra mol of acrylaldehyde re- was obtained, the stereochemistry of which, as depicted,
quired in the original methodology. After addition of had arisen from two stereodefined processes: syn-carbomet-
TMSCl, the red precipitate dissolved and CO was let in. allation and syn-periplanar β-H elimination.[7] In the pres-
Table 1. Reaction of substrates 2aϪ2p with Ni(COD)2, TMSCl and methanol
Entry
Dienal
Solvent
Reaction products
with CO bubbling
Reaction products
without CO
1
2
2a
2b
2b
2b
2b
2c[b]
2d
2e
2f
toluene
Ϫ[a]
Ϫ
Ϫ
Ϫ
Ϫ
toluene
3b (56%), 4b (4%)
3b (55%), 4b (18%)
7b (15%), 8b (25%)
Ϫ
3
acetonitrile
methanol ϩ TFA
toluene
4
5
9b (20%), 10b (40%)
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
2i (60%)
Ϫ
Ϫ
Ϫ
6
toluene
3c (52%)[b]
Ϫ
7
toluene
8
toluene
3e (60%)[c]
3f (69%)[d]
5g (40%), 12g (15%)
Ϫ
9
toluene
10
11
12
13
14
15
16
17
18
19
2g
2h
2i
toluene
toluene
toluene
Ϫ
2j
2k
2l
2m
2n
2o
2p
toluene
12j (37%)
Ϫ
3l (85%)
5m (42%)
3n (65%), 5n (8%)
decomposition
5p (10%)
toluene
toluene
toluene
Ϫ
acetonitrile
toluene
11n (52%)
Ϫ
Ϫ
toluene
[a]
Although only polymers could be obtained from the substrate 2a, the corresponding allyl halide (2E)-4-bromobut-2-enyl prop-2-ynyl
ether was prepared and the analogous carbonylated ene product, methyl (2E)-4-vinyldihydrofuran-3(2H)-ylideneacetate, was obtained
[b]
[c]
[d]
from this in 84% yield. Ϫ Isomeric cis/trans mixture (1:4) and diastereomeric ratio 75:25 for 3c. Ϫ Diastereomeric ratio 95:5. Ϫ
Diastereomeric ratio 75:25.
1360
Eur. J. Org. Chem. 2001, 1359Ϫ1369

Intramolecular Ni-Mediated Cyclizations with α,ω-Dienals
FULL PAPER
ence of 2 mol of TMSCl, however, the reaction gave a 56%
With the aim of making the reaction catalytic in nickel,
yield of cycloadduct 3b and 4% of the ketal 4b (Table 1, we then replaced the Lewis acid TMSCl with protic trifluo-
Entry 2). Both products converged after hydrolysis to the roacetic acid. No catalytic character resulted, but from the
common α-oxocarbaldehyde 6b.
reaction in toluene we were able, after quenching the inter-
mediates with methanol, to isolate the product formally
corresponding to the Michael adduct 7b,[8] with no traces
of cycloadducts.
However, when methanol in the presence of TFA was
used as the solvent we observed a rapid change as soon as
CO was allowed to enter the reaction flask: The original
deep red colour, typical of the generated allyl complexes,
rapidly turned apple green. After workup of the mixture
and purification of the products, we isolated, together with
the former product 7b, a 25% yield of a product 8b arising
from a formal ‘‘metallo-ene’’ reaction.
The ketal 4b was obtained as a diastereomeric mixture
(93:7). The formation of the major stereoisomer with the
dimethoxymethyl group in exo orientation, as deduced from
NOE experiments, most probably arises from the different
arrangement of the substituent bearing the TMSO moiety
in the two possible coordination faces of the double bond
in the (acyl)Ni intermediate. Obviously, the one disposing
this substituent in a pseudoequatorial arrangement will be
favoured over that positioning it with the pseudoaxial geo-
metry (Scheme 2). The reaction proved to proceed even
better (although with lower selectivity) in acetonitrile (re-
placing the original toluene), and a 55% yield of enol ether
3b and an 18% yield of the ketal 4b were obtained in this
solvent (Entry 3).
Interestingly, it consisted of only one diastereomer (the
cis isomer as deduced from the chemical shift of the annular
methyne protons).[9a] We interpret this result as a function
of a low degree of stabilisation of a putative (π-allyl)Ni cat-
ion, brought about by the enol proton. Thus, rapid coor-
dination of carbon monoxide would alter the π-allyl system
Scheme 2. Origin of the stereoselectivity found in products 3b and 4b
Eur. J. Org. Chem. 2001, 1359Ϫ1369
1361

´
´
´
G. Garcıa-Gomez, J. M. Moreto
FULL PAPER
to a σ-allyl state, promoting either a ‘‘metallo-ene’’ reac- substrate 2n, in the absence of CO and in the presence of
tion[9b] or alternatively a shift to a weakly coordinated π- TFA gave a 52% yield of the ‘‘ene-type’’ product 11n, the
oxaallyl complex (through insertion and proton capture), dimethylketal of 8n (Entry 17).
which would eventually be cleaved by methanol (Scheme 3).
Not only did thioether 2o not afford any defined product,
but it promoted decomposition of the original Ni(COD)2
complex, since the typical red colour of the allyl derivative
was not formed and, instead, a black deposition was ob-
served. From the diallylamine 2p, only a minor amount of
the original substrate (10%) could be recovered, as the cor-
responding ketal, from the reaction mixture and no cyclo-
adduct was detected. Therefore, both nitrogen and sulfur,
being strongly coordinating heteroatoms, were found to in-
teract in a detrimental way in the cycloaddition pathway.
In our next move, we investigated the effects of substitu-
Scheme 3. Two plausible paths for the formation of tetrahydrofu-
tion in the distal allylic region. With this aim, we prepared
ran 8b
the starting compound 2c, with a cis/trans ratio of 75:25
(as in the original crotyl chloride). Treatment of this with
Ni(COD)2, TMSCl and CO in toluene gave a 52% yield of
two isomeric cyclopentanones 3c in a molar ratio of 3:1,
similar to that in the starting dienal (Entry 6).
The formation of 8b represents a diversion of the process
leading to cyclopentanones in the presence of a protic acid.
It took place even in the presence of CO when TFA was
present. The question now arose of whether or not the first
CϪC insertion, producing the furan ring, actually required
CO. After treatment of 2b with Ni(COD)2, TMSCl and
methanol in toluene, we were able to isolate a 60% yield of
a 1:2 mixture of 9b and 10b (Entry 5).
We ruled out the possibility of the presence of a pair of
cis and trans isomers at the enol ether double bond because
1
of the very different H and 13C NMR chemical shifts for
the atoms in the neighbourhood of the saturated α-oxo site
compared with the similarity of those near the enol ether.
We thought that the transformation might have been ste-
reospecific and therefore decided to start from pure isomers.
We prepared the cis-dienal 2f, which under the former reac-
tion conditions gave a 69% yield of the oxadiquinane 3f
(Entry 9). This compound, however, consisted of two differ-
ent diastereomers in a ratio of 75:25. Surprisingly, the relat-
ive stereochemistry of the n-propyl moiety and the oxa ring
in the major isomer is trans (as deduced from NOE experi-
ments), exactly the opposite of that expected from the ori-
ginal substrate.
Since the product of β-elimination (10b) was the most
abundant, we concluded that the role of TFA in the pro-
tonation of the aldehyde was very efficient and completely
avoided a β-elimination. These last experiments told us that
while the first insertion would not require the presence of CO,
its presence and the stabilising role of the TMS group would
be essential for the carbonylation and second cyclization to
occur.
We then introduced changes in the tether, and the sub-
strates 2n, 2o and 2p were tested in the carbonylative cyclo-
addition. Treatment of 2n and 2 mol of TMSCl in acetonitr-
ile gave similar results to those for 2b, and a 65% yield of
3n was isolated (Entry 17). Together with 3n, an 8% yield
of the dimethyl ketal of the original dienal 5n was produced
and also a minor quantity of 4n (1:12 related to 3n) that
Since the trans geometry appeared the most favoured for
did not permit isolation and characterisation. The same both the starting dienal and the cyclopentanone, we set out
1362
Eur. J. Org. Chem. 2001, 1359Ϫ1369

Intramolecular Ni-Mediated Cyclizations with α,ω-Dienals
FULL PAPER
to discriminate between two different isomerization paths: dition, did indeed give good and excellent yields of the ex-
(a) prior to the closure of the cyclopentanone ring (either pected cycloadducts in highly stereoselective manner (60%
before coordination or during the process), or (b) with the and 85% respectively, diastereomeric ratio for 3e 95:5 and
oxadiquinane retaining its original geometry but the epi- one single diastereomer for 3l) (Scheme 5).
merizable α-oxo site suffering thermodynamic equilibration
after the process. Thus, we prepared the other (trans) isomer
2g, which was tested in the reaction under the same condi-
tions as for 2f. We were surprised to see that no adduct was
detected, but that the ketal derived from the substrate (5g)
was produced (Entry 10). We tried under more forcing con-
ditions (acetonitrile and 2 mol of TMSCl), but again the
cycloaddition failed and we were able only to isolate, to-
gether with 5g, a 15% yield of 12g.
It had become clear that the isomerization occurred when
Scheme 5. Allowed cycloadditions in unhindered dienals
the cyclopentanone ring had already been formed, and that
the geometry of the substituents at the double bond was of
The configuration of the three stereogenic tertiary fusion
centres in 3l was found to be cis-trans by NOE experiments
(cis for the cyclopentane ring; trans for the furan ring). This
stereochemistry would correspond to the most favoured ar-
rangement of the cyclooctene ring allowing coordination of
the metal ion by the re face. This coordination would dis-
pose the ring in a pseudoaxial configuration near the metal
centre and a pseudoequatorial one at the α-oxy tertiary
centre. The substantially higher degree of planarity in the
cyclohexene ring when 2m was used as the substrate would,
therefore, not be expected to allow a stable coordination of
the double bond with the metal ion and, again, 5m and 12m
were produced instead in variable amounts.
the utmost importance before ring closure. Therefore, if a
substrate does not undergo any cycloaddition (usually re-
verting to the ketal of the original aldehyde), it is because
the coordination of the distal olefin is somehow hampered.
Searching for the factor that might explain these observa-
tions, we came to the conclusion that, in addition to the
known fact that (Z)-olefins coordinate to transition metal
ions more easily than their (E) counterparts in this process,
those substituents liable to sterically interact with the sub-
stituents and ligands placed in the coordination hemisphere
opposed to the central (apical) π-allyl carbon atom were
completely preventing the chelation of the double bond,
and also its activation and final insertion. This conclusion
would fit with the known tilting ‘‘out-above, in-below’’ of
the π-allyl plane[10] and would very much point to a slanted
coordination of the olefin to the nickel centre in a square-
pyramidal environment (Scheme 4).
As expected, under noncarbonylative conditions, 2f
yielded 8f (among other products),[11] while neither 2g or 2i
(Entries 10, 12) proved able similarly to undergo the ‘‘ene-
type’’ reaction, giving, as usual, linear methanolysis prod-
ucts.
Finally, we also attempted the formation of a cyclopen-
tanone ring fused to a six-membered ring, using the sub-
strates 2j and 2k. No cycloaddition products were ever de-
tected, pointing to a high entropy barrier, probably adding
to the conformational restrictions, for these cases.
Conclusions
Ni(COD)2 efficiently promotes intramolecular carbonyl-
ative cycloaddition with 2,7-dienals to give diquinane-like
systems. In the absence of CO, or in the presence of protic
Scheme 4. Putative intermediates involved in the formation of 3f
Consistently with this, reactions with 2d, 2h and 2i also acids, ‘‘ene-type’’ reactions are also promoted, which can be
produced uncyclized products (starting products and/or the considered to constitute the initiation of the carbonylative
corresponding ketals). Again, these substrates were liable to cycloaddition. Both classes of reactions are stereoselective
bring with them the same kind of steric interactions and and are blocked by the presence of highly coordinating het-
we were not surprised by the lack of cyclization. Moreover, eroatoms and/or the presence of substituents that may in-
corresponding treatment of 2e and 2l (Entries 8 and 15, teract sterically with the initially formed π-allyl complex
respectively), both with substitution patterns close to that when the double bond approaches to coordinate the metal
of 2f and therefore prone to undergo carbonylative cycload- centre. Surprisingly, the pseudoaxial arrangement of these
Eur. J. Org. Chem. 2001, 1359Ϫ1369
1363

´
´
´
G. Garcıa-Gomez, J. M. Moreto
FULL PAPER
Syntheses of 1j, 1k, 1n: Under Ar, NaH (0.6 g, 25 mmol) and THF
(30 mL) were introduced into a three-necked, 100-mL flask. The
alcohol (1j, 1k) or ester (1n) (25 mmol) was added dropwise at 0 °C.
Once the H2 evolution had ceased, (2Z)-4-bromobut-2-enyl acetate
(4.82 g, 25 mmol) was added dropwise at room temperature. The
contents were then heated to reflux for 4 h, monitoring (by TLC)
the completion of the reaction by the disappearance of the allyl
halide. The crude product was extracted as above and the residue,
without purification, hydrolysed[12] for deprotection.
substituents in the putative intermediate [from (Z)-olefins]
would be more favoured than that in the pseudoequatorial
ones [from (E)-olefins]. A possible explanation for these re-
sults could be found in the known distortion of the π-allyl
plane from orthogonality.
Experimental Section
(2Z)-4-(But-3-enyloxy)but-2-en-1-ol (1j): 73%.
Ϫ
1H NMR
General Remarks: IR spectra were recorded with a Bomem MB-
120 instrument with Fourier transform capability and are reported
(300 MHz, CDCl3): δ ϭ 2.34 (tq, J ϭ 1.2, 6.6 Hz, 3 H), 3.49 (t,
J ϭ 6.6 Hz, 2 H), 4.05 (d, J ϭ 6.0 Hz, 2 H), 4.18 (d, J ϭ 6.3 Hz,
2 H), 5.06 (m, 2 H), 5.70 (m, 1 H), 5.80 (m, 2 H). Ϫ 13C NMR
(75 MHz): δ ϭ 34.1 (CH2), 58.7 (CH2), 66.4 (CH2), 69.8 (CH2),
116.6 (CH2), 128.3 (CH), 132.21 (CH), 134.9 (CH).
1
in cmϪ1. H and 13C NMR spectra were obtained in CDCl3 solu-
tions (unless otherwise indicated) and were recorded with Gemini
200 Varian and Unity 300 Varian machines, operating at 200 and
300 MHz for 1H and 50 and 75 MHz for 13C, respectively. Chemical
shifts are reported in δ units (ppm) downfield from Me4Si, or relat-
1
ive to the singlet at δ ϭ 7.26 for CDCl3 for H and relative to the
centre line of a triplet at δ ϭ 77.0 of CDCl3 for 13C. Splitting
patterns are designated as: s, singlet; d, doublet; t, triplet; q, quad-
ruplet; quint, quintuplet; *, apparent multiplicity; m, multiplet; br,
broad. Elemental analyses were performed with a Carlo Erba ap-
paratus (1106 and 1108 Models). GC-MS data were determined
with a Fisons MD800 mass spectrometer coupled to a gas chroma-
tograph equipped with an SPB-5 (30 m ϫ 0.32 mm i.d.) fused silica
capillary column. High resolution MS (EI) spectra (70 eV) were
obtained with an Auto Spec-Q instrument. TLC was performed on
Merck 60 F254 silica gel plates. Flash chromatography was per-
formed on 230Ϫ400 mesh Merck 60 silica gel. Commercial analyt-
ical grade reagents were obtained from commercial suppliers and
were used directly without further purification. Solvents were dis-
tilled under Ar prior to use and dried by standard methods.
(2Z)-4-(2-Allylphenoxy)but-2-en-1-ol (1k): 47%.
Ϫ
1H NMR
(200 MHz, CDCl3): δ ϭ 1.50 (br. s, 1 H), 3.39 (d, J ϭ 6.4 Hz, 2
H), 4.29 (br. s, 2 H), 4.63 (d, J ϭ 3.6 Hz, 2 H), 5.05 (dd, J ϭ 15.8,
4.4 Hz, 2 H), 5.85 (t, J ϭ 3.8 Hz, 2 H), 6.00 (m, 1 H), 6.84 (d, J ϭ
8.0 Hz, 1 H), 6.93 (d, J ϭ 7.4 Hz, 1 H), 7.17 (d, J ϭ 7.4 Hz, 2 H).
Ϫ
13C NMR (75 MHz): δ ϭ 34.3 (CH2), 59.0 (CH2), 64.1 (CH2),
111.7 (CH), 115.4 (CH2), 120.9 (CH), 127.3 (CH), 127.6 (CH),
129.9 (CH), 132.1 (CH), 136.9 (CH), 156.0 (C).
Diethyl 2-Allyl-2-[(2Z)-4-hydroxybut-2-enyl]malonate (1n): 82%. Ϫ
IR (film): ν˜ ϭ 3448, 1735, 1641 cmϪ1. Ϫ 1H NMR (200 MHz,
CDCl3): δ ϭ 1.23 (t, J ϭ 7.0 Hz, 6 H), 2.65 (d, J ϭ 7.6 Hz, 4 H),
4.15 (m, 4 H), 5.10 (m, 2 H), 5.40 (m, 1 H), 5.66 (m, 2 H). Ϫ 13C
NMR (50 MHz): δ ϭ 13.9 (CH3), 30.3 (CH2), 37.0 (CH2), 57.2 (C),
58.2 (CH2), 61.3 (CH2), 119.1 (CH2), 125.6 (CH), 132.1 (CH), 132.2
(CH), 170.7 (C).
Syntheses of 1a؊i, 1l, 1m: Under Ar, NaH (6 g, 0.25 mol) and
anhydrous THF (50 mL) were introduced into a three-necked, 250-
mL flask. (Z)-But-2-ene-1,4-diol (44 g, 500 mmol) was slowly ad-
ded, while keeping the temperature at 0 °C, at which point a white
slurry appeared. When the H2 evolution ceased, the mixture was
allowed to come to room temperature and the corresponding allyl
bromide (or allyl chloride for 1d, 1e and 1h) (500 mmol) was added
dropwise with vigorous stirring. The mixture warmed up while the
sodium halide separated. After the addition was complete, the con-
tents of the flask were refluxed for 4 h. The solvent was evaporated
to dryness and a saturated solution of NaCl was added. The or-
ganic material was extracted three times with diethyl ether. The
combined organic layers were dried with Na2SO4 and purified by
passing through a flash chromatography column (eluent: ethyl acet-
ate/hexane, 1:1).
Syntheses of 1o, 1p: Solid K2CO3 (4.5 g, 105 mmol) was added to
a solution of diallylamine (5.09 g, 52.5 mmol) (or prop-2-ene-1-
thiol) in 25 mL of THF. The mixture was vigorously stirred while
2-{[(2Z)-4-chlorobut-2-enyl]oxy}tetrahydro-2H-pyran
(10.00 g,
52.5 mmol) was added dropwise, and then left to stir for 24 h. The
precipitate was filtered off and the solution concentrated to dry-
ness. Saturated NaCl solution was then added to the residue, and
the mixture was extracted a few times with diethyl ether until a
drop did not show a residue after solvent evaporation. The com-
bined organic phases were concentrated and the product treated
with an excess of MeOH and p-toluenesulfonic acid for deprotec-
tion.[13] After concentration, the residue was found pure enough to
be used in the case of 1o, while 1p was purified by passing it
through a chromatography column (silica gel; EtOAc/hexane, 1:1).
(2Z)-4-(Prop-2-ynyloxy)but-2-en-1-ol (1a): 45%.
Ϫ
1H NMR
(300 MHz, CDCl3): δ ϭ 2.45 (t, J ϭ 2.4 Hz, 1 H), 2.71 (br. s, 1
H), 4.13 (m, 4 H), 4.21 (d, J ϭ 6.3 Hz, 2 H), 5.60 (m, 1 H), 5.80
(m, 1 H). Ϫ 13C NMR (75 MHz, CDCl3): δ ϭ 57.1 (CH2), 58.2
(CH2), 64.8 (CH2), 74.7 (CH), 79.3 (C), 126.9 (CH), 133.1 (CH).
(2Z)-4-(Allyloxy)but-2-en-1-ol (1b): 55%. Ϫ IR (film): ν˜ ϭ 3383,
1
3022, 1645 cmϪ1. Ϫ H NMR (200 MHz, CDCl3): δ ϭ 2.70 (br.
1H NMR
s, 1 H), 3.94 (d, J ϭ 8.7 Hz, 2 H), 4.03 (d, J ϭ 8.4 Hz, 2 H), 4.16
(br. s, 2 H), 5.20 (m, 2 H), 5.74 (m, 3 H). Ϫ 13C NMR (50 MHz,
CDCl3): δ ϭ 58.4 (CH2), 65.5 (CH2), 71.2 (CH2), 117.3 (CH),
127.8 (CH2), 132.2 (CH), 134.2 (CH).
(2Z)-4-(Allylsulfanyl)but-2-en-1-ol (1o): 77%.
Ϫ
(200 MHz, CDCl3): δ ϭ 1.43 (br. s, 1 H), 3.11 (d, J ϭ 5.8 Hz, 2
H), 3.14 (d, J ϭ 5.4 Hz, 2 H), 4.21 (d, J ϭ 6.0 Hz, 2 H), 5.11 (m,
2 H), 5.74 (m, 3 H). Ϫ 13C NMR (50 MHz): δ ϭ 27.0 (CH2),
34.0 (CH2), 58.2 (CH2), 117.1 (CH2), 128.2 (CH), 131.1 (CH),
134.2 (CH).
1364
Eur. J. Org. Chem. 2001, 1359Ϫ1369

Intramolecular Ni-Mediated Cyclizations with α,ω-Dienals
FULL PAPER
General Procedure for the Cyclization: Ni(COD)2 (1.12 g,
4.07 mmol) and anhydrous toluene (or acetonitrile) (20 mL) were
introduced under Ar into a 100-mL Schlenk flask and kept at Ϫ
20 °C while the dienal 2 (0.51 g, 4.07 mmol for 2b) was added drop-
wise. Soon a red precipitate or solution was formed. The temper-
ature was allowed to increase until it reached Ϫ10 °C. At this point,
Me3SiCl (0.88 g, 8.14 mmol) was added dropwise. After 1 h, the
oxidative addition was complete. The temperature was lowered
once more to Ϫ78 °C (Ϫ50 °C for acetonitrile) and the Ar was
replaced by CO, by bubbling under atmospheric pressure. The tem-
perature was again allowed to rise to room temperature. An excess
of methanol (2 mL) was added (quenching). The reaction was al-
lowed to proceed for a further 4 h. The solvent was completely
evaporated in vacuo. The residue was treated with water and ex-
tracted several times with diethyl ether. The combined organic
phases were dried with Na2SO4 and concentrated to dryness. The
crude oil was chromatographed on silica gel (eluent: hexane/
EtOAc, 6:1) and dried to obtain the expected adducts. For the reac-
tion in acetonitrile, a 55% yield of 3b and an 18% yield of 4b
were isolated.
(2Z)-4-(Diallylamino)but-2-en-1-ol (1p): 46%.
Ϫ
1H NMR
(300 MHz, CDCl3): δ ϭ 3.08 (dq, J ϭ 1.5, 6.0 Hz, 2 H), 3.13 (tt,
J ϭ 1.5, 6.6 Hz, 4 H), 4.15 (dq, J ϭ 1.5, 6.0 Hz, 2 H), 5.18 (m, 2
H), 5.22 (m, 2 H), 5.68 (dtt, J ϭ 1.5, 6.0, 11.7 Hz, 1 H), 5.80Ϫ5.96
(m, 3 H).
Syntheses of 2a؊o: A slight excess of pyridinium chlorochromate
(5.05 g, 23.4 mmol) in CH2Cl2 (1 mL) was added to a solution of 1b
(2 g, 15.6 mmol) (or 1aϪo) in CH2Cl2 (5 mL) at room temperature.
After stirring for 30Ϫ45 min, the reaction was complete (TLC
monitoring) and anhydrous diethyl ether (20 mL) was added. The
crude product was filtered through silica gel and the filtrate was
concentrated to dryness to afford the aldehydes 2aϪo.
(2E)-4-(Prop-2-ynyloxy)but-2-enal (2a): 45%.
Ϫ
1H NMR
(300 MHz, CDCl3): δ ϭ 2.47 (t, J ϭ 2.1 Hz, 1 H)4.19 (d, J ϭ
2.1 Hz, 2 H), 4.31 (dd, J ϭ 1.8, 4.2 Hz, 2 H), 6.30 (ddt, J ϭ 1.8,
8.1, 15.6 Hz, 1 H), 6.80 (dt, J ϭ 4.2, 15.6 Hz, 1 H), 9.60 (d, J ϭ
8.1 Hz, 1 H). Ϫ 13C NMR (75 MHz): δ ϭ 58.1 (CH2), 68.0 (CH2),
75.3 (CH), 78.8 (C), 132.0 (CH), 152.0 (CH), 193.1 (CH);
(4E)-4-(Methoxymethylene)tetrahydro-1H-cyclopenta[c]furan-
5(3H)-one (3b): Colourless solid, m.p 58 °C. Ϫ IR (film): ν˜ ϭ 1706,
1625 cmϪ1. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 2.25 (dd, J ϭ 4.8,
18.9 Hz, 1 H), 2.61 (dd, J ϭ 9.6, 18.9 Hz, 1 H), 2.88 (m, 1 H), 3.52
(m, 1 H), 3.53 (dd, J ϭ 4.8, 9.0 Hz, 1 H), 3.73 (dd, J ϭ 4.2, 9.0 Hz,
1 H), 3.86 (s, 3 H), 3.92 (dd, J ϭ 6.6, 9.0 Hz, 1 H), 4.01 (dd, J ϭ
7.8, 9.0 Hz, 1 H), 7.21 (d, J ϭ 2.1 Hz, 1 H). Ϫ 13C NMR (75 MHz):
δ ϭ 37.1 (CH2), 43.0 (CH), 43.4 (CH), 62.0 (CH3), 73.8 (CH2),
74.7 (CH2), 118.6 (C), 156.3 (CH), 206.5 (C). Ϫ MS (40 eV, EI);
m/z (%): 168 (76) [M]ϩ, 138 (29), 126 (33), 109 (100). Ϫ HR MS
for C9H12O3: calcd: 168.078644; found 168.078640.
(2E)-4-(Allyloxy)but-2-enal (2b): 67%. Ϫ 1H NMR (300 MHz,
CDCl3): δ ϭ 4.05 (dt, J ϭ 1.8, 5.7 Hz, 2 H), 4.26 (dd, J ϭ 1.8,
3.9 Hz, 2 H), 5.22 (dq, J ϭ 1.8, 10.5 Hz, 1 H), 5.30 (dq, J ϭ 1.8,
17.4 Hz, 1 H), 5.90 (ddt, J ϭ 5.7, 10.5, 17.4 Hz, 1 H), 6.38 (ddt,
J ϭ 1.8, 7.8, 15.6 Hz, 1 H), 6.84 (dt, J ϭ 3.9, 15.6 Hz, 1 H), 9.58
(d, J ϭ 7.8 Hz, 1 H). Ϫ 13C NMR (75 MHz): δ ϭ 65.6 (CH2), 71.1
(CH2), 117.5 (CH2), 130.9 (CH), 131.7 (CH), 153.1 (CH), 193.2
(CH);
4-(Dimethoxymethyl)tetrahydro-1H-cyclopenta[c]furan-5(3H)-one
(4b): Diastereomeric mixture (93:7), colourless oil. Ϫ IR (film): ν˜ ϭ
1741 cmϪ1. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 2.21 (dd, J ϭ 4.2,
18.9 Hz, 1 H), 2.45 (dd, J ϭ 3.3, 6.3 Hz, 1 H), 2.56 (dd, J ϭ 9.0,
18.9 Hz, 1 H), 2.92 (m, 1 H), 3.14 (m, 1 H), 3.40 (s, 3 H), 3.41 (s,
3 H), 3.53 (dd, J ϭ 5.1, 9.0 Hz, 1 H), 3.70 (dd, J ϭ 3.3, 9.0 Hz, 1
H), 3.92 (dd, J ϭ 3.3, 9.0 Hz, 1 H), 3.96 (dd, J ϭ 4.2, 9.0 Hz, 1
H), 4.56 (d, J ϭ 3.3 Hz, 1 H). Ϫ 13C NMR (75 MHz): δ ϭ 38.4
(CH), 40.8 (CH), 43.5 (CH2), 55.7 (CH3), 56.8 (CH), 57.1 (CH3),
74.3 (CH2), 74.6 (CH2), 105.5 (CH), 217.3 (C).
Synthesis of 2p: The oxidation with PCC gave a mixture of prod-
ucts. For this particular dienal, MnO2 was preferred, in a molar
excess of 40:1. The aldehyde was obtained as the crude product in
60% yield (corresponding by 1H NMR to an estimated 50% actual
yield) and used as such.
(2E)-4-(Diallylamino)but-2-enal (2p): 1H NMR (300 MHz, CDCl3):
δ ϭ 3.10 (dt, J ϭ 1.2, 6.3 Hz, 4 H), 3.33 (dd, J ϭ 1.6, 5.6 Hz, 2
H), 5.19 (m, 4 H), 5.80 (m, 2 H), 6.26 (ddt, J ϭ 1.6, 8.0, 15.6 Hz,
1 H), 6.83 (dt, J ϭ 5.6, 15.6 Hz, 1 H), 9.58 (d, J ϭ 8.0 Hz, 1 H).
Ϫ
13C NMR (75 MHz): δ ϭ 54.3 (CH2), 56.9 (CH2), 118.0 (CH2),
133.6 (CH), 134.9 (CH), 155.7 (CH), 193.6 (CH).
(2E)-4-Bromobut-2-enyl Prop-2-ynyl Ether): 1H NMR (300 MHz,
CDCl3): δ ϭ 2.48 (t, J ϭ 2.4 Hz, 1 H), 4.05 (d, J ϭ 8.4 Hz, 2 H),
(2E)-4-(Allyloxy)-1,1-dimethoxybut-2-ene (5b): As a result of apply-
4.18 (d, J ϭ 2.4 Hz, 2 H), 4.22 (dd, J ϭ 1.5, 6.6 Hz, 2 H), 5.71 (m, ing the general procedure, from 2b (0.28 g, 2.20 mmol), Ni(COD)2
1 H), 5.93 (m, 1 H). Ϫ 13C NMR (75 MHz, CDCl3): δ ϭ 26.2
(0.61 g, 2.20 mmol) and trimethylsilyl chloride (0.24 g, 2.20 mmol)
(CH2), 57.3 (CH2), 64.1 (CH2), 74.8 (CH), 79.3 (C), 129.3 (CH), in toluene (20 mL), 5b was obtained after flash chromatography
130.0 (CH).
(hexane/EtOAc, 9:1) as a yellow oil (0.152 g, 40%). Ϫ 1H NMR
Eur. J. Org. Chem. 2001, 1359Ϫ1369
1365

´
´
´
G. Garcıa-Gomez, J. M. Moreto
FULL PAPER
(300 MHz, CDCl3): δ ϭ 3.31 (s, 6 H), 3.98 (dt, J ϭ 1.5, 5.4 Hz, 2 (4E)-4-(Methoxymethylene)-6-methyltetrahydro-1H-cyclopenta[c]-
H), 4.01 (d, J ϭ 5.4 Hz, 2 H), 4.78 (d, J ϭ 4.8 Hz, 1 H), 5.17 (dd,
J ϭ 1.5, 10.2 Hz, 1 H), 5.27 (dq, J ϭ 1.5, 17.1 Hz, 1 H), 5.70 (ddt,
furan-5(3H)-one (3c): As a result of applying the general procedure,
from 2c (0.46 g, 3.23 mmol), Ni(COD)2 (0.89 g, 3.23 mmol) and
J ϭ 1.5, 4.8, 15.9 Hz, 1 H), 5.90 (m, 2 H). Ϫ 13C NMR (75 MHz): trimethylsilyl chloride (0.35 g, 3.23 mmol), 3c was obtained after
δ ϭ 52.7 (CH3), 69.5 (CH2), 71.2 (CH2), 102.4 (CH), 117.0 (CH2), flash chromatography (hexane/EtOAc, 6:1) as a 3:1 diastereomeric
128.6 (CH), 131.2 (CH), 134.5 (CH). Ϫ MS (40 eV, EI); m/z (%): mixture, yellow oil (0.306 g, 52%).
141 (30) [M Ϫ OCH3]ϩ, 115 (28), 101 (100), 81 (43), 75 (73), 71
(60).
1
Compound 3c (Major Isomer): H NMR (300 MHz, CDCl3): δ ϭ
1.15 (d, J ϭ 7.2 Hz, 3 H), 2.19 (q, J ϭ 7.2 Hz, 1 H), 2.43 (m, 1 H),
3.43 (m, 1 H), 3.69 (dd, J ϭ 3.0, 9.0 Hz, 1 H), 3.82 (dd, J ϭ 3.6,
7.2 Hz, 1 H), 3.84 (s, 3 H), 3.85 (dd, J ϭ 3.9, 9.0 Hz, 1 H), 3.95 (t,
J ϭ 8.4 Hz, 1 H), 7.22 (d, J ϭ 2.1 Hz, 1 H). Ϫ 13C NMR (75 MHz):
δ ϭ 15.7 (CH3), 41.0 (CH), 47.0 (CH), 48.8 (CH), 61.9 (CH3), 73.5
(CH2), 73.6 (CH2), 118.3 (C), 156.0 (CH), 208.7 (C). Ϫ For the
mixture of both isomers: HR MS for C10H14O3: calcd. 182.094294;
found 182.093965.
4-(Allyloxy)-3-methoxybutanal (7b) and (4-Methyltetrahydrofuran-
3-yl)acetaldehyde (8b): As a result of applying the general proced-
ure, from 2b (0.44 g, 3.53 mmol), Ni(COD)2 (0.97 g, 3.53 mmol)
and trifluoroacetic acid (0.40 g, 3.80 mmol) in MeOH (20 mL), 7b
(15%) and 8b (25%) were obtained after flash chromatography
(hexane/EtOAc, 9:1).
Compound 7b: Yellow oil (0.083 g, 15%). Ϫ 1H NMR (300 MHz,
CDCl3): δ ϭ 2.64 (dd, J ϭ 2.1, 6.3 Hz, 2 H), 3.41 (s, 3 H), 3.50
(dd, J ϭ 1.2, 4.8 Hz, 2 H), 3.86 (m, 1 H), 4.00 (dq, J ϭ 1.5, 5.7 Hz,
2 H), 5.17 (dd, J ϭ 1.2, 9.0 Hz, 1 H), 5.25 (dq, J ϭ 1.5, 17.1 Hz,
1 H), 5.78 (m, 1 H), 9.80 (t, J ϭ 1.8 Hz, 1 H). Ϫ 13C NMR
(75 MHz): δ ϭ 45.9 (CH2), 57.6 (CH3), 70.7 (CH2), 72.3 (CH2),
75.2 (CH), 117.2 (CH2), 134.2 (CH), 200.7 (CH).
Minor Isomer: Data deduced from the mixture by signal subtrac-
tion. Ϫ 1H NMR (300 MHz, CDCl3; main visible signals): δ ϭ
1.09 (d, J ϭ 7.2 Hz, 3 H), 2.57 (dq, J ϭ 7.2, 8.4 Hz, 1 H), 2.92 (q,
J ϭ 8.1 Hz, 1 H), 4.06 (dd, J ϭ 7.2, 9.0 Hz, 1 H), 7.25 (d, J ϭ
2.1 Hz, 1 H). Ϫ 13C NMR (75 MHz): δ ϭ 11.1 (CH3), 41.3 (CH),
42.8 (CH), 45.1 (CH), 62.0 (CH3), 69.1 (CH2), 73.4 (CH2), 117.6
(C), 156.5 (CH), 207.2 (C).
Compound 8b: Colourless oil (0.113 g, 25%).
Ϫ
1H NMR
(300 MHz, CDCl3): δ ϭ 0.91 (d, J ϭ 7.2 Hz, 3 H), 2.40 (m, 1 H),
2.45 (dd, J ϭ 1.5, 8.4 Hz, 1 H), 2.56 (dd, J ϭ 1.5, 6.0 Hz, 1 H),
2.67 (m, 1 H), 3.41 (dd, J ϭ 3.3, 8.4 Hz, 1 H), 3.44 (dd, J ϭ 6.6,
8.4 Hz, 1 H), 3.93 (dd, J ϭ 6.6, 8.4 Hz, 1 H), 3.99 (dd, J ϭ 7.2,
8.4 Hz, 1 H), 9.81 (t, J ϭ 1.5 Hz, 1 H). Ϫ 13C NMR (75 MHz):
δ ϭ 14.0 (CH3), 35.2 (CH), 36.15 (CH), 42.4 (CH2), 72.0 (CH2),
74.5 (CH2), 201.0 (CH).
(4E)-4-(Methoxymethylene)-1-methyltetrahydro-1H-cyclopenta[c]-
furan-5(3H)-one (3e): As a result of applying the general procedure,
from 2e (0.46 g, 3.23 mmol), Ni(COD)2 (0.89 g, 3.23 mmol) and
trimethylsilyl chloride (0.35 g, 3.23 mmol), 3e was obtained after
flash chromatography (hexane/EtOAc, 6:1) as a 95:5 diastereomeric
mixture, colourless oil (0.353 g, 65%). Ϫ IR (film): ν˜ ϭ 1708, 1631,
1456 cmϪ1. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 1.24 (d, J ϭ
6.0 Hz, 3 H), 2.10 (dd, J ϭ 3.0, 18.6 Hz, 1 H), 2.34 (m, 1 H), 2.54
(dd, J ϭ 9.3, 18.6 Hz, 1 H), 3.45 (dq, J ϭ 6.0, 7.8 Hz, 1 H), 3.51
(m, 2 H), 3.86 (s, 3 H), 4.30 (tt, J ϭ 3.3, 7.8 Hz, 1 H), 7.23 (d, J ϭ
2.1 Hz, 1 H). Ϫ 13C NMR (75 MHz, CDCl3): δ ϭ 18.7 (CH3), 40.9
(CH2), 43.4 (CH), 44.3 (CH), 61.9 (CH3), 73.1 (CH2), 81.2 (CH),
118.4 (C), 156.6 (CH), 206.2 (C). Ϫ MS (40 eV, EI); m/z (%): 182
(32) [M]ϩ, 137 (100), 123 (42), 109 (93), 93 (44). Ϫ Data for the
mixture of isomers: HR MS for C10H14O3: calcd. 182.094294;
found 182.094516.
3-[(E)-2-Trimethylsilyloxyethenyl]-4-methylenetetrahydrofuran
(10b): As a result of applying the general procedure, from 2b
(0.41 g, 3.23 mmol), Ni(COD)2 (0.89 g, 3.23 mmol) and trimethylsi-
lyl chloride (0.38 g, 3.56 mmol), 10b (0.26 g 40%) was obtained
after flash chromatography (hexane/EtOAc, 9:1). Ϫ 1H NMR
(300 MHz, CDCl3): δ ϭ 0.2 (s, 9 H), 3.20 (m, 1 H), 3.40 (dd, J ϭ
8.4, 9.6 Hz, 1 H), 4.06 (t, J ϭ 8.4 Hz, 1 H), 4.30 (dq, J ϭ 2.4,
13.2 Hz, 1 H), 4.45 (dq, J ϭ 2.4, 13.2 Hz, 1 H), 4.80 (dd, J ϭ 9.4,
12.0 Hz, 1 H), 4.92 (q, J ϭ 2.4 Hz, 1 H), 4.96 (q, J ϭ 2.4 Hz, 1 H),
6.31 (d, J ϭ 12.0 Hz, 1 H). Ϫ 13C NMR (75 MHz): δ ϭ Ϫ0.52
(CH3), 43.6 (CH), 71.3 (CH2), 74.1 (CH2), 104.7 (CH2), 109.4
(CH), 142.1 (CH), 151.7 (C). Ϫ MS (40 eV, EI); m/z (%): 198 (9)
[M]ϩ, 168 (70), 75 (79), 73 (100). Ϫ Product 9b could only be tent-
atively identified in the mixture since, being the minor product,
some signals were overlapped by those of the major. Any attempt
to separate them resulted in isolation of 8b.
(4E)-4-(Methoxymethylene)-6-propyltetrahydro-1H-cyclopenta[c]-
furan-5(3H)-one (3f): As a result of applying the general procedure,
from 2f (0.40 g, 2.40 mmol), Ni(COD)2 (0.66 g, 2.40 mmol) and tri-
methylsilyl chloride (0.26 g, 2.40 mmol), 3f was obtained after flash
chromatography (hexane/EtOAc, 9:1) as a 3:1 diastereomeric mix-
1366
Eur. J. Org. Chem. 2001, 1359Ϫ1369

Intramolecular Ni-Mediated Cyclizations with α,ω-Dienals
FULL PAPER
ture, colourless oil (0.35 g, 69%). Ϫ Data for the mixture: IR (film):
ν˜ ϭ 1706, 1633, 1463 cmϪ1
.
Major Isomer: 1H NMR (300 MHz, CDCl3): δ ϭ 0.91 (t, J ϭ
6.6 Hz, 3 H), 1.37 (m, 4 H), 2.17 (m, 1 H), 2.53 (m, 1 H), 3.44 (m,
1 H), 3.60 (dd, J ϭ 3.6, 9.0 Hz, 1 H), 3.68 (dd, J ϭ 4.5, 8.7 Hz, 1
H), 3.85 (s, 3 H), 3.89 (dd, J ϭ 6.6, 9.0 Hz, 1 H), 3.98 (t, J ϭ
8.4 Hz, 1 H), 7.20 (d, J ϭ 2.1 Hz, 1 H). Ϫ 13C NMR (75 MHz,
CDCl3): δ ϭ 14.0 (CH3), 20.3 (CH2), 33.6 (CH2), 41.3 (CH), 44.7
(CH), 53.9 (CH), 61.8 (CH3), 73.6 (CH2), 74.3 (CH2), 118.5 (C),
155.9 (CH), 208.4 (C). Ϫ MS (40 eV, EI); m/z (%): 181 (5), 168
(100), 137 (34), 106 (41). ϪData for the mixture: HR MS for
C12H18O3: calcd. 210.125595; found 210.123947.
4-(2,4,4-Trimethoxybutoxy)but-1-ene (12j): As a result of applying
the general procedure, from 2j (0.22 g, 1.53 mmol), Ni(COD)2
(0.42 g, 1.53 mmol) and trimethylsilyl chloride (0.33 g, 3.05 mmol),
12j was obtained after flash chromatography (hexane/EtOAc, 9:1)
1
as a colourless oil (0.123 g, 27%). Ϫ H NMR (300 MHz, CDCl3):
δ ϭ 1.79 (t, J ϭ 6.0 Hz, 2 H)2.43 (tq, J ϭ 1.2, 6.6 Hz, 2 H), 3.32
(s, 3 H), 3.33 (s, 3 H), 3.44 (s, 3 H), 3.50 (m, 5 H), 4.54 (t, J ϭ
6.0 Hz, 1 H), 5.06 (m, 2 H), 5.80 (m, 1 H).
(3E)-3-(Methoxymethylene)decahydro-1-oxacycloocta[cd]pentalen-
4(2H)-one (3l): As a result of applying the general procedure, from
2l (0.41 g, 2.11 mmol), Ni(COD)2 (0.58 g, 2.11 mmol) and trime-
thylsilyl chloride (0.23 g, 2.11 mmol), 3l was obtained after flash
chromatography (hexane/EtOAc, 9:1) as a single diastereomer, yel-
low oil (0.42 g, 85%), m.p. 107 ° C. Ϫ IR (film): ν˜ ϭ 1706, 1631
cmϪ1. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 1.2Ϫ1.6 (m, 6 H), 1.87
(m, 2 H), 2.11 (m, 1 H), 2.21 (m, 1 H), 2.36 (m, 1 H), 2.66 (q, J ϭ
9.9 Hz, 1 H), 3.39 (dd, J ϭ 6.3, 8.7 Hz, 1 H), 3.48 (dt, J ϭ 3.3,
10.2 Hz, 1 H), 3.58 (m, 1 H), 3.86(s, 3 H), 4.36 (t, J ϭ 8.4 Hz, 1
H), 7.26 (d, J ϭ 2.4 Hz, 1 H). Ϫ 13C NMR (75 MHz, CDCl3): δ ϭ
21.5 (CH2), 22.9 (CH2), 25.4 (CH2), 27.9 (CH2), 35.7 (CH2), 43.0
(CH), 47.9 (CH), 50.4 (CH), 61.9 (CH3), 72.4 (CH2), 81.0 (CH),
117.8 (C), 156.5 (CH), 207.5 (C). Ϫ MS (40 eV, EI); m/z (%): 236
(30) [M]ϩ, 208 (35), 178 (33), 137 (50), 84 (100). Ϫ C14H20O3
(236.31): calcd. C 71.18, H 8.47; found C 70.97, H 8.64
Minor Isomer: Data deduced from the mixture by signal subtrac-
tion. Ϫ 1H NMR (300 MHz, CDCl3; main visible signals): δ ϭ
1.17 (d, J ϭ 6.3 Hz, 3 H), 2.10 (dd, J ϭ 3.0, 18.6 Hz, 1 H), 2.29
(m, 1 H), 2.50 (dd, J ϭ 9.3, 18.6 Hz, 1 H), 3.39 (dq, J ϭ 6.3, 7.8 Hz,
1 H), 3.51 (m, 2 H), 3.81 (s, 3 H), 4.30 (tt, J ϭ 3.3, 7.8 Hz, 1 H),
7.13 (d, J ϭ 2.1 Hz, 1 H). Ϫ 13C NMR (75 MHz, CDCl3): δ ϭ
14.1 (CH3), 15.7, 20.9 (CH2), 38.3 (CH), 41.2 (CH), 60.3, 61.8
(CH3), 73.0 (CH2), 81.2 (CH), 118.4 (C), 155.2 (CH), 206.2 (C). Ϫ
MS (40 eV, EI); m/z (%): 181 (8), 168 (100), 137 (35), 106 (26).
(2E)-1-{[(2E)-4,4-Dimethoxybut-2-enyl]oxy}hex-2-ene (5g): As a re-
sult of applying the general procedure, from 2g (0.36 g, 2.14 mmol),
Ni(COD)2 (0.59 g, 2.14 mmol) and trimethylsilyl chloride (0.23 g,
2.14 mmol), 5g was obtained after flash chromatography (hexane/
EtOAc, 9:1) as a 3:1 diastereomeric mixture, colourless oil (0.183 g,
3-{[(2E)-4,4-Dimethoxybut-2-enyl]oxy}cyclohex-1-ene (5m): As a re-
sult of applying the general procedure, from 2m (0.25 g,
1.53 mmol), Ni(COD)2 (0.42 g, 1.53 mmol) and trimethylsilyl
chloride (0.16 g, 1.53 mmol), 5m was obtained after flash chroma-
1
40%). Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.90 (t, J ϭ 7.2 Hz, 3
H), 1.40 (quint*, J ϭ 7.2 Hz, 2 H), 2.04 (q, J ϭ 7.2 Hz, 2 H), 3.33
(s, 6 H), 3.93 (d, J ϭ 6.0 Hz, 2 H), 4.01 (d, J ϭ 6.3 Hz, 2 H), 4.80
(d, J ϭ 4.8 Hz, 1 H), 5.56 (m, 2 H), 5.70 (m, 2 H). Ϫ 13C NMR
(75 MHz): δ ϭ 13.6 (CH3), 22.1 (CH2), 34.3 (CH2), 52.7 (CH3),
69.2 (CH2), 71.0 (CH2), 102.4 (CH), 126.1 (CH), 128.5 (CH), 131.4
(CH), 134.8 (CH);
1
tography (hexane/EtOAc, 6:1) as a yellow oil (0.136 g, 42%). Ϫ H
NMR (300 MHz, CDCl3): δ ϭ 1.48Ϫ2.11 (m, 6 H), 3.32 (s, 6 H),
3.93 (m, 1 H), 4.05 (m, 2 H), 4.80 (d, J ϭ 4.5 Hz, 1 H), 5.78 (m,
4 H).
(2E)-1-(2,4,4-Trimethoxybutoxy)hex-2-ene (12g): As a result of ap-
plying the general procedure (without CO) for 2g (0.76 g,
4.54 mmol), Ni(COD)2 (1.25 g, 4.54 mmol) and trimethylsilyl
chloride (0.99 g, 9.09 mmol) in toluene (25 mL), 12g was obtained
after flash chromatography (hexane/EtOAc, 9:1) as a yellow oil
Diethyl
(4E)-4-(Methoxymethylene)-5-oxohexahydropentalene-
2,2(1H)-dicarboxylate (3n): As a result of applying the general pro-
cedure, from 2n (1.05 g, 3.93 mmol), Ni(COD)2 (1.08 g, 3.93 mmol)
and trimethylsilyl chloride (0.85 g, 7.85 mmol), 3n was obtained
after flash chromatography (hexane/EtOAc, 9:1) as a yellow oil
(0.791 g, 65%), together with 5n (8%).
1
(0.168 g, 15%). Ϫ H NMR (300 MHz, CDCl3): δ ϭ 0.88 (t, J ϭ
7.2 Hz, 3 H), 1.40 (m, 2 H), 1.79 (t, J ϭ 6.0 Hz, 2 H), 2.02 (q, J ϭ
7.2 Hz, 2 H), 3.31 (s, 3 H), 3.32 (s, 3 H), 3.39 (s, 3 H), 3.42 (m, 2
H), 3.93 (d, J ϭ 6.0 Hz, 2 H), 4.03 (m, 1 H), 4.54 (t, J ϭ 6.0 Hz,
1 H), 5.52 (m, 2 H). Ϫ 13C NMR (75 MHz): δ ϭ 13.6 (CH3), 22.1 Compound 3n: IR (film): ν˜ ϭ 1706, 1633, 1463 cmϪ1. Ϫ H NMR
1
(CH2), 34.2 (CH2), 34.9 (CH2), 52.7 (CH3), 52.9 (CH3), 57.6 (CH3), (300 MHz, CDCl3): δ ϭ 1.22 (t, J ϭ 7.2 Hz, 3 H), 1.26 (t, J ϭ
71.2 (CH2), 72.0 (CH2), 77.4 (CH), 102.0 (CH), 126.3 (CH),
134.5 (CH).
7.2 Hz, 3 H), 1.90 (dd, J ϭ 9.0, 13.5 Hz, 1 H), 2.12 (dd, J ϭ 6.6,
13.5 Hz, 1 H), 2.21 (dd, J ϭ 3.6, 18.9 Hz, 1 H), 2.55 (dd, J ϭ 9.0,
Eur. J. Org. Chem. 2001, 1359Ϫ1369
1367

´
´
´
G. Garcıa-Gomez, J. M. Moreto
FULL PAPER
18.9 Hz, 1 H), 2.58 (ddd, J ϭ 1.5, 7.8, 13.2 Hz, 1 H), 2.77 (m, 1
from
(2E)-4-bromobut-2-enyl prop-2-ynyl ether (0.41 g,
H), 2.81 (ddd, J ϭ 1.5, 8.4, 13.8 Hz, 1 H), 3.43 (m, 1 H), 3.86 (s, 2.18 mmol), Ni(COD)2 (0.60 g, 2.18 mmol) in toluene (20 mL), this
3 H), 4.15 (q, J ϭ 7.2 Hz, 2 H), 4.21 (q, J ϭ 7.2 Hz, 2 H), 7.20 (d,
compound was isolated after flash chromatography (hexane/
J ϭ 2.4 Hz, 1 H). Ϫ 13C NMR (75 MHz, CDCl3): δ ϭ 13.9 (CH3), EtOAc, 5:1) as a yellow oil (0.307 g, 84%). Ϫ IR (film): ν˜ ϭ 1720,
1
36.0 (CH2), 39.7 (CH2), 41.1 (CH), 41.6 (CH2), 44.4 (CH), 61.4
1672, 1434 cmϪ1. Ϫ H NMR (300 MHz, CDCl3): δ ϭ 3.71 (s, 3
(CH2), 61.9 (CH3), 62.0 (CH2), 119.7 (C), 156.5 (CH), 171.4 (C), H, CH3), 3.90 (dd, J ϭ 5.7, 9.0 Hz, 1 H, CH2O), 4.01 (dd, J ϭ 2.1
171.8 (C), 206.9 (C).
9.0 Hz, 1 H, CH2O), 4.18 (m, 1 H, CH), 4.32 (dt, J ϭ 1.5, 15.3 Hz,
1 H, CH2O), 4.55 (dt, J ϭ 1.5, 15.3 Hz, 1 H, CH2O), 5.09 (dt, J ϭ
1.5, 10.2 Hz, 1 H, ϭCH), 5.17 (dt, J ϭ 1.5, 17.1 Hz, 1 H, ϭCH),
5.83 (q, J ϭ 1.8 Hz, 1 H, ϭCH), 5.89 (ddd, J ϭ 6.9, 10.2, 17.1 Hz,
1 H, ϭCH). Ϫ 13C NMR (75 MHz, CDCl3): δ ϭ 46.2 (CH), 51.2
(CH3), 71.7 (CH2), 74.0 (CH2), 111.2 (CH2), 115.4 (CH2), 136.1
(CH), 162.7 (C), 165.8 (C).
Diethyl 2-Allyl-2-[(2E)-4,4-dimethoxybut-2-enyl]malonate (5n):
98 mg, 8%. Ϫ 1H NMR (300 MHz, CDCl3): δ ϭ 1.23 (t, J ϭ
6.8 Hz, 6 H), 2.64 (m, 4 H), 3.27 (s, 6 H), 4.21 (q, J ϭ 6.8 Hz, 4
H), 4.68 (d, J ϭ 4.8 Hz, 1 H), 5.13 (m, 2 H), 5.66 (m, 3 H). Ϫ 13C
NMR (75 MHz): δ ϭ 13.8 (CH3), 13.9 (CH3), 34.9 (CH2), 36.7
(CH2), 52.5 (CH3), 57.1 (C), 61.3 (CH2), 61.2 (CH2), 102.4 (CH),
119.2 (CH2), 128.6 (CH), 131.0 (CH), 131.9 (CH), 170.4 (C).
Acknowledgments
We thank DGESIC and Generalitat de Catalunya (Project
PB97Ϫ0407-C05Ϫ05 and Grant 5GR95Ϫ0439) for financial sup-
port. G. G.-G. also thanks CIRIT for a fellowship. We thank J.
Roget and BASF Espan˜ola S.A. for support.
Diethyl
3-(2,2-Dimethoxyethyl)-4-methylcyclopentane-1,1-dicarb-
oxylate (11n): As a result of applying the general procedure, from
2n (0.51 g, 1.89 mmol), Ni(COD)2 (0.52 g, 1.89 mmol) and trifluo-
roacetic acid (0.22 g, 1.89 mmol), 11n was obtained after flash
chromatography (hexane/EtOAc, 9:1) as a yellow oil (0.311 g, 52%).
[1]
[1a]
For leading references see:
B. M. Trost, I. Fleming, M. F.
Semmelhack, Comprehensive Organic Synthesis Pergamon
[1b]
Press, New York, 1991, vol. 4. Ϫ
L. S. Hegedus, Transition
Metals in the Synthesis of Complex Organic Molecules, Univer-
[1c]
sity Science Books, New York 1994. ؊
L. Haughton, J. M.
Ϫ
1H NMR (300 MHz, CDCl3): δ ϭ 0.98 (d, J ϭ 5.7 Hz, 3 H),
J. Williams, J. Chem. Soc., Perkin Trans.1 1999, 2645Ϫ2658. Ϫ
1.22 (t, J ϭ 7.2 Hz, 6 H), 1.53 (m, 2 H), 1.72 (d, J ϭ 13.2 Hz, 1
H), 1.80 (dd, J ϭ 4.2, 13.5 Hz, 1 H), 1.85 (m, 1 H), 1.90 (dd, J ϭ
3.0, 7.2 Hz, 1 H), 2.44 (dd, J ϭ 6.0, 13.2 Hz, 1 H), 2.55 (dd, J ϭ
6.9, 13.5 Hz, 1 H), 3.31 (s, 3 H), 3.33 (s, 3 H), 4.17 (q, J ϭ 7.2 Hz,
4 H), 4.40 (dd, J ϭ 4.5, 7.2 Hz, 1 H). Ϫ 13C NMR (75 MHz,
CDCl3): δ ϭ 14.0 (CH3), 17.5 (CH3), 36.3 (CH2), 40.2 (CH2), 40.5
(CH2), 42.2 (CH), 42.7 (CH), 53.0 (CH3), 56.9 (C), 61.2 (CH2),
61.3 (CH2), 103.7 (CH), 177.7 (C), 177.8 (C). Ϫ MS (40 eV, EI);
m/z (%): 285 (7) [M Ϫ OCH3]ϩ, 239 (26), 165 (12), 137 (15), 107
(14), 75 (100).
[1d]
A. C. Comely, S. E. Gibson, S. Sur, J. Chem. Soc., Perkin
Trans. 1 2000, 109Ϫ124.
[2] [2a]
O. Geis, H. G. Schmalz, Angew. Chem. Int. Engl. 1998, 37,
911Ϫ915. Ϫ [2b] N. E. Schore in Comprehensive Organometallic
Chemistry II, vol. 12 (Eds.: G. Wilkinson, F. G. A. Stone, E.
W. Abel), Pergamon, Oxford, 1995, p. 703Ϫ739.
[3] [3a]
´
F. Camps, J. Coll, J. M. Moreto, J. Torras, J. Org. Chem.
1989, 54, 1969Ϫ1978. Ϫ [3b] F. Camps, J. M. Moreto, Ll. Pages,
´
`
Tetrahedron 1992, 48, 3147Ϫ3162. Ϫ [3c] Ll. Pages, A. Llebaria,
`
´
F. Camps, E. Molins, C. Miravitlles, J. M. Moreto, J. Am.
[3d]
Chem. Soc. 1992, 114, 10449Ϫ10461. Ϫ
F. Vilaseca, Ll.
`
´
Pages, J. M. Villar, A. Llebaria, A. Delgado, J. M. Moreto, J.
Orgamomet. Chem. 1998, 551, 107Ϫ115.
[4]
[4a] G. Garcia-Gomez, X. Camps, A. Jauma Cayuela, J. M. Mo-
[4b]
´
reto, Inorg. Chim. Acta 1999, 296, 94Ϫ102. Ϫ
G. Garcia-
´
Gomez, J. M. Moreto, J. Am. Chem. Soc. 1999, 121, 878Ϫ879.
[5]
[6]
´
F. Camps, J. Coll, J. M. Moreto, J. Torras, Tetrahedron Lett.
1987, 28, 4745Ϫ4748
Ni0 complexes are active alkyne polymerisation catalysts; see:
P. W. Jolly, G. Wilke, The Organic Chemistry of Nickel Aca-
demic Press, New York, 1975, vol. II, chapter II.
Additional evidence for the stereochemistry of the enol ethers 3
reported here was provided by the X-ray crystal diffractometry
resolution of cycloadduct 3l.
We think that under the present conditions these kinds of prod-
ucts actually arise from a methoxy group attack onto the π-
allyl ligand.
(2E)-N,N-Diallyl-4,4-dimethoxybut-2-en-1-ylamine (5p): As a result
of applying the general procedure, from a 1:1 mixture of 1p/2p
(0.624 g, 1.89 mmol), Ni(COD)2 (0.52 g, 1.89 mmol) and trimethyl-
silyl chloride (0.23 g, 2.08 mmol), 5p was isolated after flash chro-
matography (hexane/EtOAc, 5:1) as a yellow oil (40 mg, 10%). Ϫ
1H NMR (300 MHz, CDCl3): δ ϭ 3.10 (m, 6 H), 3.32 (s, 3 H), 3.38
(s, 3 H), 4.16 (d, J ϭ 5.4 Hz, 1 H), 5.18 (m, 4 H), 5.80 (m, 4 H).
[7]
[8]
Ϫ
13C NMR (75 MHz): δ ϭ 52.8 (CH2), 56.2(CH2), 57.6(CH2),
[9] [9a]
W. Oppolzer, M. Bedoya-Zurita, C. Y. Switzer, Tetrahedron
[9b]
102.3 (CH), 118.7(CH2), 127.7(CH), 130.0 (CH), 134.2 (CH).
Lett. 1988, 29, 6433Ϫ6436. Ϫ
W. Oppolzer, Angew. Chem.
Int. Ed. Engl. 1989, 28, 38Ϫ52.
[10]
[11]
The dihedral angle formed between the π-allyl plane and the
equatorial is found always to be larger than 90° (usually around
106° and even attaining values of 120° in particular cases). See
ref.[6], vol. I, chapter VI.
Although 8f was produced as the major product in the crude
product, its isolation after purification turned out to be im-
Methyl (2E)-[4-Vinyldihydrofuran-3(2H)-ylidene]acetate: See foot-
note [a] in Table 1. As a result of applying the general procedure,
1368
Eur. J. Org. Chem. 2001, 1359Ϫ1369

Intramolecular Ni-Mediated Cyclizations with α,ω-Dienals
FULL PAPER
E. J. Corey, H. Niwa, J. Knolle, J. Am. Chem. Soc. 1978, 100,
possible and it could only be poorly identified in the 1H NMR
of the resulting mixture.
J. J. Plattner, R. D. Gless, H. Rapoport, J. Am. Chem. Soc.
[13]
1942 and references therein.
[12]
Received October 2, 2000
1972, 94, 8613 and references therein.
[O00504]
Eur. J. Org. Chem. 2001, 1359Ϫ1369
1369