Further studies on nickel-promoted or -catalyzed cyclization of 1,3- diene and a tethered carbonyl group

Article abstract of DOI:10.1021/ja991241d

A nickel-promoted intramolecular cyclization of 1,3-diene with the tethered carbonyl group was developed using the catalyst generated by reduction of Ni(acac)₂ with DIBAL-H in the presence of PPh₃. It was found that the addition of 1

Full text of DOI:10.1021/ja991241d

1624
J. Am. Chem. Soc. 2000, 122, 1624-1634
Further Studies on Nickel-Promoted or -Catalyzed Cyclization of
1,3-Diene and a Tethered Carbonyl Group
Yoshihiro Sato, Masanori Takimoto, and Miwako Mori*
Contribution from the Graduate School of Pharmaceutical Sciences, Hokkaido UniVersity,
Sapporo 060-0812, Japan
ReceiVed April 19, 1999
Abstract: A nickel-promoted intramolecular cyclization of 1,3-diene with the tethered carbonyl group was
developed using the catalyst generated by reduction of Ni(acac)2 with DIBAL-H in the presence of PPh3. It
was found that the addition of 1,3-CHD to the reaction mixture affected the regiochemistry of olefin on the
side chain of the cyclized product. The reaction course of this cyclization can be accounted for by two possible
mechanisms. In one mechanism, a nickel hydride complex plays a key role and the cyclization proceeds via
π-allylnickel intermediate. In the other mechanism, a zero-valent nickel complex is the active species and the
cyclization proceeds via nickelacycle intermediates. These mechanistic considerations led to finding two nickel-
(0)-catalyzed cyclizations of 1,3-diene and the tethered aldehyde, in which the five- to seven-membered ring
products were produced in a regio- and stereoselective manner via π-allylnickel intermediate or via a
i
transmetalation process of nickelacycle intermediates with Bu2-ALAC.
Introduction
A cocyclization of bis-1,3-dienes in the presence of hydrosilane
and a cyclization of 1,3-diene and tethered alkyne with
isocyanide were reported by Tamao and Ito.6 The utility of these
processes prompted us to investigate the cyclization of 1,3-
dienes and other multiple bonds containing a heteroatom, and
we achieved a nickel-promoted or -catalyzed cyclization of 1,3-
dienes and tethered carbonyl groups to give five- to seven-
membered ring compounds in a stereoselective manner.7 We
describe herein our full results as well as new findings that have
arisen from our continuing studies on nickel-promoted or
-catalyzed stereoselective cyclization of 1,3-dienes and the
tethered carbonyl groups.8
Transition metal-catalyzed cycloaddition is a useful and
promising strategy for the stereoselective construction of various
cyclic compounds.1 The co-oligomerization of 1,3-dienes is one
of the most important processes in cycloaddition, and the
pioneering work of nickel-promoted co-oligomerization of 1,3-
dienes, in which 1,5-cyclooctadiene was produced by the
reaction of two molecules of butadiene with Ni(CO)4 in the
presence of P(OPh)3, was reported by Reed in 1954.2 Since then,
although there have been many efforts to investigate the nickel-
promoted or -catalyzed oligomerization of 1,3-dienes and
multiple bonds, synthetic utilization of these processes has been
restricted due to competition with linear polymerization as well
as the difficulty in controlling regio- and stereoselectivity.3 On
the other hand, the intramolecular versions of this process are
useful for regio- and stereoselective ring construction, and a
few excellent reactions have been reported. For example,
Wender reported a nickel-catalyzed [4+4] cycloaddition of bis-
1,3-dienes, in which eight-membered-ring compounds were
produced stereoselectively by a zero-valent nickel catalyst under
mild conditions.4 A [4+2] cycloaddition of 1,3-diene and
tethered alkyne or allene was also reported by the same group.5
Results and Discussion
Cyclization of 1,3-Diene with Tethered Aldehyde Using
the Catalyst Generated from Ni(acac)2 and DIBAL-H in the
Presence of PPh3. At first, we investigated the cyclization of
diene aldehydes 10-12, and these substrates were synthesized
as shown in Scheme 1. The reaction of the dianion, generated
from sorbic acid (1) and LDA,9 with the iodide 2 or 3 followed
(4) For [4+4] cycloadditions, see: (a) Wender, P. A.; Tebbe, M. J.
Synthesis 1991, 1089. (b) Wender, P. A.; Ihle, N. C.; Correia, C. R. D. J.
Am. Chem. Soc. 1988, 110, 5904. (c) Wender, P. A.; Ihle, N. C. Tetrahedron
Lett. 1987, 28, 2451. (d) Wender, P. A.; Snapper, M. L. Tetrahedron Lett.
1987, 28, 2221. (e) Wender, P. A.; Ihle, N. C. J. Am. Chem. Soc. 1986,
108, 4678.
(5) For [4+2] cycloadditions, see: (a) Wender, P. A.; Smith, T. E.
Tetrahedron 1998, 54, 1255. (b) Wender, P. A.; Smith, T. E. J. Org. Chem.
1996, 61, 824. (c) Wender, P. A.; Smith, T. E. J. Org. Chem. 1995, 60,
2962. (d) Wender, P. A.; Jenkins, T. E. J. Am. Chem. Soc. 1989, 111, 6432.
For a cyclization of 1,3-diene and allene, see: (e) Wender, P. A.; Jenkins,
T. E.; Suzuki, S. J. Am. Chem. Soc. 1995, 117, 1843.
(6) (a) Tamao, K.; Kobayashi, K.; Ito, Y. Synlett. 1992, 539. (b) Tamao,
K.; Kobayashi, K.; Ito, Y. J. Synth. Org. Chem. Jpn. 1990, 48, 381.
(7) Portions of this work have previously been communicated; see: (a)
Sato, Y.; Takimoto, M.; Hayashi, K.; Katsuhara, T.; Takagi, K.; Mori, M.
J. Am. Chem. Soc. 1994, 116, 9771. (b) Sato, Y.; Takimoto, M.; Mori, M.
Tetrahedron Lett. 1996, 37, 887. For extensions to a nitrogen-containing
system, see: (c) Sato, Y.; Saito, N.; Mori, M. Tetrahedron Lett. 1997, 38,
3931. (d) Sato, Y.; Saito, N.; Mori, M. Tetrahedron 1998, 54, 1153.
* To whom correspondence should be addressed: (tel, fax) +81-11-
706-4982; (e-mail) mori@pharm.hokudai.ac.jp.
(1) (a) Lautens, M.; Klute, W.; Tam, W. Chem. ReV. 1996, 96, 49-92.
(b) Ojima, I.; Tzamarioudaki, M.; Li, Z.; Donovan, R. J. Chem. ReV. 1996,
96, 635-662.
(2) Reed, H. W. B. J. Chem. Soc. 1954, 1931.
(3) For reviews, see: (a) Heimback, P.; Jolly, P. W.; Wilke, G. In
AdVances in Organometallic Chemistry; Stone, F. G. A., West, R., Eds.;
Academic: New York, 1970; Vol. 8, p 29. (b) Jolly, P. W.; Wilke, G. The
Organic Chemistry of Nickel; Academic: New York, 1975; Vol. 2. (c) Jolly,
P. W. In ComprehensiVe Organometallic Chemistry; Wilkinson, G., Stone,
F. G. A., Abel, E. W., Eds.; Pergamon: New York, 1982; Vol. 8, p 613.
(d) Keim, W.; Behr, A.; Roper, M. In ComprehensiVe Organometallic
Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon:
New York, 1982; Vol. 8, p 371. (e) Fischer, K.; Jonas, K.; Misbach, P.;
Stabba, R.; Wilke, G. Angew. Chem., Int. Ed. Engl. 1973, 12, 943. (f)
Heimback, P. Angew. Chem., Int. Ed. Engl. 1973, 12, 975. (g) Wilke, G.
Angew. Chem., Int. Ed. Engl. 1988, 27, 185.
10.1021/ja991241d CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/15/2000

Cyclization of 1,3-Diene and a Tethered Carbonyl Group
J. Am. Chem. Soc., Vol. 122, No. 8, 2000 1625
Scheme 1
Scheme 2
Table 1. Cyclization Using Various Amounts of Ni Complex 14
yield (%)
run Ni (mol %) temp (°C) time (hr) 15-I + 15-T 15-I 15-T
by treatment with diazomethane afforded the ester 4 or 5 in
good yield, respectively. After reduction of ester 4 or 5, the
resulting alcohol was treated with benzyl bromide to give 6 or
7, and deprotection of the TBDMS group in 6 or 7 followed by
oxidation of the alcohol produced substrate 10 or 11, respec-
tively. On the other hand, substrate 12 was obtained from 9 via
13 in three steps.
1
2
3
4
100
50
30
0
0
0
6
6
20
20
69
65
36
10
69a
6b
59
36
10
10
rt
a E:Z ) 8:1. b The ratio was not determined.
Scheme 3
To a toluene solution containing a stoichiometric amount of
the nickel complex 14, prepared in situ from Ni(acac)2 (100
mol %) and 2 equiv of DIBAL-H in the presence of PPh3 (200
mol %),6,10 was added a solution of 10 in toluene. The resulting
solution was stirred at 0 °C for 6 h. Hydrolysis of the reaction
mixture with 10% HCl at 0 °C afforded an alcohol 15-I in 69%
yield. Hydrogenation of 15-I with Pd/C gave the cyclopentanol
16 in 87% yield as a sole product, which indicates that three
stereocenters at C1, C2, and C3 on the cyclopentane ring in
15-I were produced in a stereoselective manner during the
cyclization. The stereochemistry of 15-I was unequivocally
determined by X-ray structural analysis of p-bromobenzoate 17
derived from (E)-15-I (Scheme 2). Cyclization of 10 using
various amounts of the nickel complex 14 was carried out, and
the results are summarized in Table 1. Although the yields in
cyclization using 50 and 30 mol % of 14 (runs 2 and 3) exceeded
the amounts of 14, reducing the amount of the catalyst tended
to decrease the yield of the cyclized product 15. It is interesting
that the cyclized product 15-T, having a terminal olefin on the
side chain, was produced in preference to 15-I, having an
internal olefin on the side chain, in cyclization using a catalytic
amount of 14 (runs 2-4). This phenomenon will be discussed
in a later section.
Encouraged by these results, the cyclization of various
substrates was investigated. When substrate 11 was treated with
a stoichiometric amount of 14, prepared by the above-mentioned
method, in toluene at 0 °C, cyclohexanols 18-I and 18-T (ratio
of 1.9:1) were obtained in 82% yield. Hydrogenation of a
mixture of 18-I and 18-T with 10% Pd/C gave cyclohexanol
19 in 92% yield as a sole product (Scheme 3), which indicates
that the three stereocenters at C1, C2, and C3 on the cyclohexane
ring in 18-I and 18-T were formed stereoselectively during
cyclization.11 It is noteworthy that the cyclization is applicable
(11) The stereochemistry of 18 (a mixture of 18-I and 18-T) was
determined by the coupling constants of Ha and Hb from the NMR spectrum
of II, which was obtained by acetylation of 18 followed by hydrogenation
with Pd(OH)2 and then PCC oxidation. Reduction of II afforded the same
alcohol I again, which indicates that no epimerization at the C3 position
occurred during conversion of I into II.
(8) A nickel-promoted or -catalyzed intermolecular coupling of 1,3-diene
and a carbonyl compound has been reported by Baker; see: (a) Baker, R.;
Cook, A. H.; Crimmin, M. J. J. Chem. Soc., Chem. Commun. 1975, 727.
(b) Baker, R.; Crimmin, M. J. J. Chem. Soc., Perkin Trans. 1 1979, 1264.
For recent reports on nickel-catalyzed intermolecular coupling of 1,3-diene
and a carbonyl compound, see: (c) Kimura, M.; Ezoe, A.; Shibata, K.;
Tamaru, Y. J. Am. Chem. Soc. 1998, 120, 4033. (d) Takimoto, M.; Hiraga,
Y.; Sato, Y.; Mori, M. Tetrahedron Lett. 1998, 39, 4543. For a cyclization
of 1,3-diene and a tethered R,â-unsaturated carbonyl group, see: Mont-
gomery, J.; Oblinger, E.; Savchenko, A. V. J. Am. Chem. Soc. 1997, 119,
4911.
(9) Ballester, P.; Costa, A.; Garcia-Raso, A.; Gomez-Solivellas, A.;
Mestres, R. Tetrahedron Lett. 1985, 26, 3625.
(10) Krysan, D. J.; Mackenzie, P. B. J. Org. Chem. 1990, 55, 4229.

1626 J. Am. Chem. Soc., Vol. 122, No. 8, 2000
Sato et al.
Table 2. Cyclizations of 23 and 28 Using Ni Complex 14
to the construction of a seven-membered ring. The reaction of
12 with a stoichiometric amount of 14 proceeded to give a
mixture of cycloheptanols 20, having a seven-membered ring,
in 43% yield. Hydrogenation of 20 with 10% Pd/C gave 21 as
two isomers (ratio of 3.5:1). To confirm that the cyclized product
had a seven-membered-ring framework, 21 was finally con-
verted into cycloheptanone 22 as a sole product by PCC
oxidation (Scheme 4). To examine the applicability of this
Scheme 4
Scheme 6
reaction to other carbonyl groups or internal dienes, the
cyclizations of 23 and 28 were planned, and these substrates
were prepared as shown in Scheme 5. Substrate 23, having a
Scheme 5
mixture of 30-I and 30-T by hydrogenation (Scheme 6).13 These
results indicate that three stereocenters at C1, C2, and C3 of
each cyclopentane ring in 29 and 30 are produced in a stereo-
selective manner in these cyclizations.
Regiocontrolled Effect of 1,3-Dienes as an Additive on
Nickel-Promoted Cyclization. As shown in Table 1, we found
that cyclized product 15-T having a terminal olefin on the side
chain was obtained in preference to 15-I having an internal one
when a catalytic amount of nickel complex 14 was used.
Although the reason is still not clear, it was speculated that an
excess of the diene part on substrate 10 coordinated to nickel
complex 14 acting as a ligand and affected the regiochemistry
of the cyclized product. That is, nickel complex 14′ coordinated
by 1,3-diene might possess a reactivity different from that of
(12) The stereochemistry of 29 was determined as follows; PCC oxidation
of 15-I followed by treatment with MeMgBr afforded a separable mixture
of cycloheptanols IIIA and IIIB. The stereochemistry of IIIA was
determined from NMR (COSY, NOESY) spectra. Hydrogenation of IIIA
afforded 31, whose spectral data were completely identical with that obtained
from a mixture of 29 by hydrogenation.
ketone moiety as a carbonyl group, was easily synthesized by
the nucleophilic addition of methylmagnesium bromide to
aldehyde 10 followed by oxidation of the resulting alcohol. For
the synthesis of 28, a coupling reaction of dimethyl malonate
(24) with iodide 2 followed by reduction of the ester group and
partial protection of the alcohol afforded 25 in good yield.
Oxidation of alcohol 25 followed by the Wadsworth-Horner-
Emmons reaction with 26 gave the corresponding dienoate,
which was converted to 27 by reduction with DIBAL-H
followed by methylation. Deprotection followed by oxidation
gave substrate 28, having an internal diene moiety.
The cyclization of ketone 23 proceeded smoothly to give a
mixture of cyclopentanols 29-I and 29-T (ratio of 1.8:1) in 56%
yield (Table 2). Hydrogenation of a mixture of 29-I and 29-T
with Pd/C gave the cyclopentanol 31 as a sole product.12 On
the other hand, cyclization of 28 afforded a mixture of
cyclopentanols 30-I and 30-T (ratio of 2.0:1) in 61% yield, and
cyclopentanol 32 was also obtained as a sole product from a
(13) The stereochemistry of 30 was determined by conversion of 30-T
into the aldehyde IV, which was obtained from 15-T as follows:

Cyclization of 1,3-Diene and a Tethered Carbonyl Group
J. Am. Chem. Soc., Vol. 122, No. 8, 2000 1627
Table 3. Cyclization of 10 in the Presence of Various Alkenes
Scheme 7
14 (Scheme 7). On the basis of this assumption, we examined
the effect of 1,3-diene on the cyclization of 10 using nickel
complex 14. To a stirred toluene solution of 14, generated in
situ by treatment of Ni(acac)2 (100 mol %) and PPh3 (200 mol
%) with DIBAL-H (200 mol %), was added 150 mol % of 1,3-
cyclohexadiene (1,3-CHD) at 0 °C, and the solution was stirred
for a few minutes. A toluene solution of substrate 10 was then
added to the resultant mixture, and the solution was stirred at 0
°C for 2 h. It was very surprising to find that hydrolysis of the
reaction mixture with 10% HCl at 0 °C preferentially produced
the cyclopentane derivative 15-T, having a terminal olefin on
the side chain, along with a small amount of 15-I (ratio of 95:
5) in 61% yield. As described above, under similar conditions
in the absence of 1,3-CHD, only 15-I was obtained in 69% yield.
This result prompted us to try cyclization of various substrates
under similar conditions in the presence of 1,3-CHD.
a All reactions were carried out in toluene at 0 °C. b A total of 150
mol % alkenes (runs 2-7) was used. c The ratio was determined by
1H NMR.
Scheme 10
The reaction of 23 proceeded smoothly at 25 °C, preferentially
giving 29-T along with 29-I in 70% yield (ratio of 86:14), while
the ratio of 29-T to 29-I was 36:64 in the above-mentioned
cyclization in the absence 1,3-CHD (Scheme 8). A remarkable
respect to the C1, C2, or C3 position of the cycloheptane ring
in 47% yield. It is notable that both cyclized products in a
mixture of 20-T had only terminal olefins on the side chain,
and no cyclized products having an internal olefin on the side
chain were formed (Scheme 11).
Scheme 8
Scheme 11
effect of the addition of 1,3-CHD was also observed in the
cyclization of 28, having an internal 1,3-diene moiety, and the
cyclized product 30-T was obtained in 73% yield as a sole
product (Scheme 9). In the cyclization of 11 in the presence of
To find the best additive to control the regioselectivity of
olefin on the side chain, reactions of 10 with nickel complex
14 in the presence of various alkenes were carried out. The
results are summarized in Table 3. A 1,4-diene (run 5) and a
monoolefin (run 6) were less effective for the formation of 15-T
than 1,3-dienes (runs 2, 3, 4, and 7), and 1,3-CHD (run 2)
showed higher selectivity than a linear 1,3-diene 33 or s-trans-
1,3-diene 34. These results indicated that a s-cis conformation
of 1,3-diene would be necessary for coordination to nickel
complex 14 in order to change the reactivity of the catalyst. It
was very interesting that 15-T was obtained in 70% yield with
the highest selectivity (15-T:15-I ) 98:2) when cyclopentadiene
3714 was used as an additive.15
Scheme 9
1,3-CHD, the cyclohexane derivative 18-T was obtained, in
preference to 18-I, in 86% yield in the ratio of 93:7, while the
above-mentioned cyclization in the absence of 1,3-CHD gave
preferentially 18-I (ratio of 31:69) in 82% yield (Scheme 10).
The reaction of 12 in the presence of 1,3-CHD produced the
cycloheptane derivatives 20-T as a mixture of two isomers with
(14) Halterman, R. L.; Vollhardt, K. P. C. Tetrahedron Lett. 1986, 27,
1461.
(15) In this reaction, unchanged 37 was recovered in 69% yield, which
suggests that 37 would not react with the complex 14 under the conditions
of this cyclization.

1628 J. Am. Chem. Soc., Vol. 122, No. 8, 2000
Sato et al.
It is possible that the formation of the cyclized product having
a terminal olefin was caused by the isomerization of that having
an internal olefin during the cyclization. To investigate this
possibility, silyl ether 38-I was treated with nickel catalyst 14,
generated from Ni(acac)2 (30 mol %), PPh3 (60 mol %), and
DIBAL-H (60 mol %), in the presence of 1,3-CHD (45 mol %)
in toluene at 0 °C for 2 h. As a result, unchanged 38-I was
recovered in 90%. Furthermore, when 38-I (1 equiv) was added
to the reaction solution of the cyclization of 10 (0.3 equiv) with
14 (0.3 equiv) in the presence of 1,3-CHD (0.45 equiv) in
toluene at 0 °C, unchanged 38-I was recovered again in 91%
along with the usual cyclized product 15-T in 58% yield based
on 10 (Scheme 12). These results suggest that the cyclized
Scheme 14. Possible Mechanisms for Cyclization Promoted
by a Ni(0) Complex
Scheme 12
mechanism, however, arose from further consideration of the
process of generation of zero-valent nickel complex 40 from
Ni(acac)2 (Scheme 15). Namely, nickel(II) hydride complex 46
Scheme 15
product having a terminal olefin was not produced from that
having an internal one via olefin isomerization during the
cyclization, although the role of 1,3-diene in regioselectivity
of the cyclized product remains unclear.
Possible Mechanism for Cyclization Using the Catalyst
Generated from Ni(acac)2-PPh3 and DIBAL-H. Having
established the nickel-promoted stereoselective cyclization of
1,3-dienes and the tethered carbonyl groups, we turned our
attention to the investigation of the mechanism of this reaction.
In this cyclization, a catalyst generated in situ by reduction of
Ni(acac)2 with DIBAL-H (2 equiv to Ni(acac)2) in the presence
of PPh3 was used. However, the structure of this catalyst was
not clear, although this catalytic system had been used in various
reactions.6,10,16 Among them, a modified method for preparation
of zero-valent nickel complex, Ni(cod)2, from Ni(acac)2 and
DIBAL-H in the presence of 1,5-cyclooctadiene (1,5-COD), was
reported by Krysan and Mackenzie.10 They assumed that zero-
valent nickel complex 40 was formed by the reduction of Ni-
(acac)2 with DIBAL-H (2 equiv to Ni(acac)2) followed by
reductive elimination from bis(hydride)nickel(II) intermediate
39 (Scheme 13). If the active species in our cyclization is a
would be formed along with diisobutylaluminum acetylacetonate
(iBu2-ALAC, 45) by reduction of Ni(acac)2 with DIBAL-H
during the generation of zero-valent nickel complex 40 via 39.17
Since complex 46 might be reactive to olefins, the reaction of
46 with the 1,3-diene moiety in 10 could produce π-allylnickel
intermediate 47.18 The π-allylnickel part in 47 could react with
tethered aldehyde giving 48, which is hydrolyzed with acidic
workup to afford cyclized product 15-I. However, the formation
of cyclized products having a terminal olefin cannot be explained
by this mechanism, because the isomerization of an internal
olefin into a terminal one in the cyclized products did not occur
during the cyclization, as described in the previous section.
Various experiments were carried out to investigate which of
the mechanisms, that shown in Scheme 14 or that shown in
Scheme 15, is more reasonable, and the results are described in
the following sections.
Scheme 13
Cyclization Using Zero-Valent Nickel Complexes. At first,
the reaction mixture of 10 with a stoichiometric amount of nickel
complex 14 was hydrolyzed with DCl-D2O, since it was
thought that a product containing deuterium on the carbon
framework should be obtained if this cyclization proceeded via
nickel complexes 41-44 in Scheme 14. As a result, treatment
of the reaction mixture with DCl-D2O provided no deuterated
product, only giving 15-I (not containing deuterium) in 69%
yield (E:Z ) 8:1). To examine the possibility that complexes
zero-valent nickel complex such as 40, the reaction mechanism
could assumed to be that shown in Scheme 14. In the cyclization
of 10, the substrate would oxidatively add to the zero-valent
nickel complex producing nickelacycle 41, which might be in
equilibration with π-allylnickel intermediates 42 and 44 and
another nickelacycle 43. Hydrolysis of nickel complexes 42,
43, and 44 by acidic workup would produce the cyclized product
15-I. It is also possible that the cyclized product having a
terminal olefin (i.e., 15-T, 18-T, 29-T, and 30-T) was produced
from a nickelacycle such as 41. The possibility of an alternative
(17) Formation of the nickel hydride complex 46 by the reaction of Ni-
(acac)2 with iBu3Al was reported by Wilke.3e In this paper, it was proposed
that iBu-Ni-acac complex was initially formed and successive â-elimination
from this complex produced 46.
(18) Wilke also reported that an olefin insertion into the hydrogen-
metal bond of complex 46 proceeded to give the corresponding alkyl-Ni-
acac complex.3e
(16) (a) Cuvigny, T.; Julia, M. J. Organomet. Chem. 1986, 317, 383.
(b) Hansen, R. T.; Carr, D. B.; Schwartz, J. J. Am. Chem. Soc. 1978, 100,
2244. (c) Sato, Y.; Nishimata, T.; Mori, M. J. Org. Chem. 1994, 59, 6133.
(d) Sato, Y.; Nishimata, T.; Mori, M. Heterocycles 1997, 44, 443. (e)
Suginome, M.; Matsuda, T.; Ito, Y. Organometallics 1998, 17, 5233.

Cyclization of 1,3-Diene and a Tethered Carbonyl Group
J. Am. Chem. Soc., Vol. 122, No. 8, 2000 1629
41-44 were decomposed in situ with a trace amount of H2O
contained in the reaction mixture or by a hydrogen abstraction
reaction from the solvent before acidic workup, the cyclization
of 10 was carried out in toluene containing D2O (10 equiv to
nickel complex 14) or in toluene-d8, and 15-I (not containing
deuterium) was obtained in 65 or 62% yield, respectively. Next,
we tried the cyclization of 10 using the zero-valent nickel
complex, Ni(cod)2, or the catalyst generated from NiCl2(PPh3)2
and n-BuLi.19 In both cyclizations, only a trace amount of 15-I
was obtained along with trace amounts of 49 and 50, and the
starting material 10 was recovered in both cases (Scheme 16).
Scheme 17
Scheme 16
yield as the sole product. It indicates that the nickel hydride
complex is more reactive to the 1,3-diene moiety of the substrate
than 1,3-CHD and that the effect of 1,3-CHD on the regiose-
lectivity of the olefin in the cyclized product could not be
reproduced in this reaction system.
Ni(0)-Catalyzed Cyclization Using a Zero-Valent Nickel
Complex in the Presence of Trialkylsilane. As shown in
Scheme 17, the cyclization of 10 using Ni(cod)2 and PPh3 in
the presence of Et3SiH produced the cyclized product 38-I as a
silyl ether. It indicates that reductive elimination of 38-I from
52 accompanying regeneration of the zero-valent nickel complex
occurred in this cyclization and that this cyclization would
proceed with a catalytic amount of Ni(0) complex (Scheme 18).
These results lead us to study the alternative mechanism depicted
in Scheme 15, in which the cyclization would proceed via
π-nickelacycle intermediate 47 generated from the 1,3-diene
moiety in 10 and a nickel(II) hydride complex such as 46.
Cyclization Using the Nickel(II) Hydride Complex Gener-
ated from a Zero-Valent Nickel Complex and Trialkylsilane
It was known that trialkylsilane oxidatively adds to a low-
valent transition metal complex to form a metal hydride complex
such as 46.20 In addition, it was also reported that a π-allylnickel
intermediate was formed by the reaction of 1,3-diene and the
nickel hydride complex generated from a zero-valent nickel
complex and trialkyl- or trialkoxysilane.21 To investigate the
mechanism concerned with nickel hydride complex 46, we tried
the cyclization of 10 using the nickel hydride complex generated
from Ni(cod)2 and PPh3 in the presence of triethylsilane (Et3-
SiH). When a toluene solution of 10, Ni(cod)2 (100 mol %),
PPh3 (200 mol %), and Et3SiH (1.5 equiv) was stirred at 0 °C
for 6.5 h, we were pleased to find that the highly regio- and
stereo-controlled cyclized product 38-I was obtained as the sole
product in 59% yield, in which three stereocenters at the C1,
C2, and C3 positions on the cyclopentane were produced with
the same selectivity as that of the cyclized product 15-I. The
reaction of 10 with Ni(cod)2 (100 mol %) and PPh3 (200 mol
%) in the presence of triethylsilyl deuteride (Et3SiD) instead of
Et3SiH also stereoselectively produced deuterated product 51
in 70% yield (D content >95%) (Scheme 17). These results
are consistent with the mechanism shown in Scheme 15,
although the formation of nickel hydride complex 46 from Ni-
(acac)2 and DIBAL-H was not directly proved. Thus, we
ultimately adopted the mechanism shown in Scheme 15 in
preliminary communications.7 It was interesting that the reaction
of 10 using a stoichiometric amount of the nickel hydride
complex formed from Ni(cod)2 and Et3SiH in the presence of
1,3-CHD (150 mol %) gave the cyclized product 38-I in 66%
Scheme 18
On the basis of this assumption, the reaction of 10 was carried
out with a catalytic amount of Ni(cod)2 (10 mol %) and PPh3
(20 mol %) in the presence of Et3SiH (5 equiv). As the result,
we succeeded in obtaining 38-I in 70% yield (Table 4, run 1).
Similarly, the cyclization of other substrates using a catalytic
amount of zero-valent nickel complex was investigated, and the
results are summarized in Table 4. The cyclization of 28, having
an internal diene moiety, also proceeded stereoselectively to
Table 4. Cyclization of Various Substrates Using a Catalytic
Amount of Ni(cod)2 and PPh3 in the Presence of Et3SiH
(19) Henningsen, M. C.; Jeropoulos, S.; Smith, E. H. J. Org. Chem. 1989,
54, 3015.
(20) Ojima, I. In The Chemistry of Organic Silicon Compounds; Patai,
S., Rappoport, Z., Eds.; John Wiley: Chichester, U.K., 1989; p 1479 and
references therein.
(21) Lappert, M. F.; Nile, T. A.; Takahashi, S. J. Organomet. Chem.
1974, 72, 425. Also, see: ref 6.
a All reactions were carried out in toluene in the presence of PPh3
(2 equiv to Ni(cod)2) and Et3SiH (5 equiv). b All cyclized products were
obtained as a sole product.

1630 J. Am. Chem. Soc., Vol. 122, No. 8, 2000
Sato et al.
give the cyclized product 53 in 67% yield as the sole product.
The cyclohexanol derivative 54 was obtained from 11 in a ste-
reoselective manner. The stereochemistry of these cyclized pro-
ducts 53 and 54 was determined by conversion into 30-I and
18-I, respectively. Surprisingly, cycloheptanol derivative 55 was
also obtained as the sole product in this reaction system.22
The probable catalytic cycle of this cyclization is shown in
Scheme 19. Nickel hydride complex 57 is formed by the oxida-
Scheme 21
Scheme 19
the reaction of 1,3-diene and nickel hydride complex 57, which
made it possible for the reductive elimination of 38-I from 52
to proceed and for a catalytic cycle of the cyclization to be
established.
tive addition of Et3SiH to the zero-valent nickel complex, and
the reaction of 10 with 57 produces π-allylnickel intermediate
47 via 56.21 The π-allylnickel moiety in 47 reacts with the
tethered carbonyl group to give 52, and the cyclized product
38-I is afforded by the reductive elimination from 52 accom-
panying the regeneration of the zero-valent nickel complex. It
is known that a π-allylnickel complex can react with a carbonyl
group to give a homoallyl alcohol derivative.23 In general,
π-allylnickel species 47 was prepared by the reaction of allyl
halide 58 with the zero-valent nickel complex in these studies
(Scheme 20). Since the reductive elimination of 61 from 60
Reinvestigation of Cyclization Using a Zero-Valent Nickel
Complex: Role of iBu2-ALAC (45). During the course of our
continuing investigations of this cyclization, we have very re-
cently found that a zero-valent nickel complex could promote
cyclization in the case of substrate 62.24 Namely, nickel complex
14, generated from Ni(acac)2 and DIBAL-H in the presence of
PPh3, was initially used in the cyclization of 62, in which it
was expected that product 65 or 66 would be obtained via
π-allylnickel intermediate 63 or 64 according to the above-men-
tioned mechanism (Scheme 21). However, the cyclized products
67, 68, and 65-T were obtained in a total 39% yield (ratio of
1.5:1:1.3) along with other undefined products. The formation
of 67, 68, and 65-T could not be explained by the mechanism
via π-allylnickel intermediate 63 or 64, and the result was quite
similar to that in the cyclization of 10 using the zero-valent
nickel complex shown in Scheme 16. In addition, the cyclization
of 62 using the nickel hydride complex generated from Ni(cod)2,
PPh3, and Et3SiH did not give the corresponding cyclized pro-
duct as a silyl ether but gave complex mixtures. These results
suggest that a nickel hydride complex cannot promote the cycli-
zation of 62. Thus, the cyclization of 62 using zero-valent nickel
complex, Ni(cod)2, was investigated. As a result, the cyclized
products 69 and 70 were obtained in good yields in this cycli-
zation. It was noteworthy that 69 and 70 could be produced by
â-hydride elimination from nickelacycle 71, which was gener-
ated from 62 and a zero-valent nickel complex (Scheme 22).
These results lead us to reconsider the mechanism depicted
in Scheme 14, in which the cyclization would proceed via
nickelacycle intermediates 41-44 generated from substrate 10
and a zero-valent nickel complex. As described in Scheme 16,
other zero-valent nickel complexes did not promote the cy-
clization of 10, while the catalyst 14 generated from Ni(acac)2
and DIBAL-H could promote the cyclization. If 14 is a zero-
valent nickel complex, we wondered what was the difference
between 14 and other zero-valent catalysts such as Ni(cod)2 or
Scheme 20
does not proceed, a stoichiometric amount of zero-valent nickel
complex is needed in this reaction. On the other hand, the
π-allylnickel intermediate in our cyclization was produced by
(22) The stereochemistry of 55 was determined by X-ray analysis of
p-nitrobenzoate V derived from 55.
(23) (a) Billington, D. C. Chem. Soc. ReV. 1985, 14, 93 and references
therein. (b) Hegedus, L. S.; Varaprath, S. Organometallics 1982, 1, 259.
(c) Hegedus, L. S.; Evans, B. R.; Korte, D. E.; Waterman, E. L.; Sjoberg,
K. J. Am. Chem. Soc. 1976, 98, 3901. (d) Hegedus, L. S.; Wagner, S. D.;
Waterman, E. L.; Siirala-Hansen, K. J. Org. Chem. 1975, 40, 593. (e)
Semmelhack, M. F.; Brichner, S. J. J. Am. Chem. Soc. 1981, 103, 3945. (f)
Semmelhack, M. F.; Yamashita, A.; Tomesch, J. C.; Hirotsu, K. J. Am.
Chem. Soc. 1978, 100, 5565. (g) Semmelhack, M. F.; Wu, E. C. S. J. Am.
Chem. Soc. J. Am. Chem. Soc. 1976, 98, 3384. (h) Semmelhack, M. F.
Org. React. 1972, 19, 115.
(24) Sato, Y.; Takanashi, T.; Hoshiba, M.; Mori, M. Tetrahedron Lett.
1998, 39, 5579.

Cyclization of 1,3-Diene and a Tethered Carbonyl Group
J. Am. Chem. Soc., Vol. 122, No. 8, 2000 1631
Scheme 22
of 10 under the same conditions in the presence of 1,3-CHD
gave 15-T exclusively in good yield (Scheme 24). These results
indicated that iBu2-ALAC plays an important role in producing
the cyclized product from nickelacycle intermediates 41-44,
and a possible mechanism is shown in Scheme 25. The
nickelacycle intermediates 41-44 would be generated by the
reaction of 10 and a zero-valent nickel complex.26 Among these
nickelacycles, it was thought that the nickelacycle 41 might be
i
most reactive to Bu2-ALAC (45) in the next transmetalation
step, since the nickelacycles 42 and 44 were stabilized by the
coordination of the η3-allyl group, and the abundance of 43 in
the equilibrium seemed to be less than that of other nickela-
cycles. The transmetalation of 41 with aluminum reagent 45
would proceed via 72 to give σ-allylnickel complex 73. The
π-allylnickel hydride complex 76 would be formed from 73
via π-allylnickel intermediate 74 or via σ-allylnickel hydride
intermediate 75. Reductive elimination of 77 from 76 followed
by hydrolysis afforded cyclized product 15. As mentioned above,
it was shown that 1,3-CHD controls the regioselectivity of the
cyclized product. It was thought that the additive 1,3-CHD might
prevent the formation of π-allylnickel complex 74 or 76 by
coordination to the η1-allyl complex 73 or 75. As a result,
spontaneous reductive elimination of 77-T from σ-allylnickel
hydride 75 would be the major pathway in the cyclization in
the presence of 1,3-CHD, producing 15-T predominantly.
a catalyst generated from NiCl2(PPh3)2 and n-BuLi. Thus, we
directed our attention to 45, which might be formed during the
conversion of Ni(acac)2 to a zero-valent nickel complex by
DIBAL-H. Diisobutylaluminum acetylacetonate (iBu2-ALAC,
45) was easily prepared from triisobutylaluminum and 2,4-
pentanedione according to the literature (Scheme 23).25 The
Scheme 23
Nickel(0)-Catalyzed Cyclization via Transmetalation Us-
ing Bu2-ALAC. As shown in Scheme 25, it was thought that
cyclization of 10 using Ni(cod)2 (100 mol %) and PPh3 (200
mol %) in the presence of Bu2-ALAC (2.2 equiv) in toluene
i
i
a zero-valent nickel species would be regenerated by reductive
elimination of 77-T or 77 from 75 or 76 and that the cyclization
would proceed catalytically with respect to the zero-valent nickel
complex. The cyclization of 10 using a catalytic amount of Ni-
(cod)2 (5 mol %) and PPh3 (10 mol %) in the presence of iBu2-
ALAC (45) (1.5 equiv) was carried out. As a result, we
succeeded in obtaining the cyclized product 15-T as the sole
product in 84% yield (Scheme 26). It was very interesting that
only 15-T having a terminal olefin on the side chain was
obtained in this catalytic cyclization, while the reaction using a
stoichiometric amount of Ni(cod)2 and PPh3 in the presence of
45 produced only 15-I having an internal olefin on the side
chain as previously shown in Scheme 24. In addition, the effect
of 1,3-diene was also observed in the cyclization using a
stoichiometric amount of zero-valent nickel complex in the
presence of 45, in which 15-T was preferentially produced in
was carried out, and we were very pleased to find that the
cyclized product 15-I was obtained in 82% yield. In addition,
Scheme 24
the effect of 1,3-CHD on the regioselectivity of the cyclized
product was also seen in this cyclization. Thus, the cyclization
i
Scheme 25. Possible Mechanism for the Cyclization Promoted by a Ni(0) Complex in the Presence of Bu2ALAC

1632 J. Am. Chem. Soc., Vol. 122, No. 8, 2000
Sato et al.
Table 5. Ni(0)-Catalyzed Cyclization of Various Substrates with
Scheme 27
iBu2-ALAC
i
a All reactions were carried out in toluene using Bu2-ALAC (1.5
equiv). b Ni(cod)2 (10 mol %), PPh3 (20 mol %). c Ni(cod)2 (5 mol %),
PPh3 (10 mol %).
the cyclization in the presence of 1,3-CHD. The fact that 15-T
was selectively produced in the catalytic reaction can be
1,3-diene with the tethered carbonyl group was developed using
catalyst 14 generated by reduction of Ni(acac)2 with DIBAL-H
in the presence of PPh3, and the cyclized product 79-I and 79-T
were obtained in a stereoselective manner with respect to the
stereochemistry of the substituents on the cycloalkane ring. It
was found that the addition of 1,3-CHD to the reaction mixture
under the same conditions affected the regiochemistry of olefin
on the side chain, and 79-T having a terminal olefin on the
side chain was exclusively produced. The reaction course of
this cyclization can be accounted for by two possible mecha-
nisms. In one mechanism, a nickel hydride complex plays a
key role and the cyclization proceeds via a π-allylnickel
intermediate. In the other mechanism, a zero-valent nickel
complex is the active species and the cyclization proceeds via
nickelacycle intermediates. These mechanistic considerations led
us to find two nickel(0)-catalyzed cyclizations of 1,3-diene and
the tethered aldehyde, in which the cyclized product 79′-I having
an internal olefin or the cyclized product 79-T having a terminal
olefin was produced respectively via π-allylnickel intermediate
8027 or via a transmetalation process of nickelacycle intermedi-
ates with iBu2-ALAC (45) such as 81. These cyclizations should
be quite interesting and synthetically useful since they are
complementary to each other to afford the cyclized product 79′-I
or 79-T in a stereoselective manner using the same catalyst
system, Ni(cod)2-PPh3, only depending upon the difference of
Scheme 26
explained by the effect of 1,3-diene on the regiochemistry of
the product. Namely, the 1,3-diene moiety in the case of an
excess amount of the substrate compared to the Ni(0) complex
would coordinate to the nickel complex and affect the regio-
chemistry of the product, in a manner similar to that shown in
Table 1.
Next, nickel(0)-catalyzed cyclization of other substrates was
carried out, and the results are summarized in Table 5. The
reaction of 28 using Ni(cod)2 (10 mol %), PPh3 (20 mol %),
and iBu2-ALAC (45) (1.5 equiv) afforded the cyclized product
30-T as the sole product in 75% yield. In the cyclization of 11,
the cyclized product 18-T was obtained as the sole product in
96% yield. In the case of a seven-membered-ring construction,
the cyclization of 12 using Ni(cod)2 (5 mol %), PPh3 (10 mol
%), and iBu2-ALAC (45) (1.5 equiv) produced the cycloheptane
derivatives 20-T in 57% yield as an inseparable mixture of two
isomers (ratio of 1:4.2) with respect to the C1, C2, or
C3-position on the cycloheptane ring. However, both products
in 20-T have only a terminal olefin at each side chain, indicating
that the regiochemistry of olefin on the side chain was
completely controlled in this system. Thus, we could extend
the zero-valent nickel-promoted cyclization of 1,3-diene and a
tethered carbonyl group via transmetalation of nickellacycles
i
additives, Et3SiH or Bu2-ALAC.
Experimental Section
General Information. All manipulations were performed under an
argon atmosphere unless otherwise mentioned. All solvents and reagents
were purified when necessary using standard procedures. Column
chromatography was performed on silica gel 60 (Merck, 70-230 mesh),
and flash chromatography was performed on silica gel 60 (Merck, 230-
400 mesh) using the indicated solvent. Melting points were determined
with a Yanagimoto meilting point microapparatus and are uncorrected.
NMR spectra were measured on JEOL GX 270 (1H at 270 MHz), JEOL
EX 270 (1H at 270 MHz, 13C at 67.5 MHz), JEOL EX 400 (1H at 400
i
and Bu2-ALAC to a catalytic reaction from which a cyclized
product having a terminal olefin on the side chain is obtained
in a stereoselective manner.
Conclusions
The main results described in this article are summarized in
Scheme 27. A nickel-promoted intramolecular cyclization of
(25) Kroll, W. R.; Naegele, W. J. Organomet. Chem. 1969, 19, 439.
(26) A similar mechanism via transmetalation of oxanickelacycles and
organometallic reagents was recently proposed by Montgomery in nickel-
(0)-catalyzed cyclization of alkyne and tethered aldehyde and by Tamaru
in nickel(0)-catalyzed coupling of 1,3-dienes and aldehydes; see: (a)
Oblinger, E.; Montgomery J. J. Am. Chem. Soc. 1997, 119, 9065. (b)
Kimura, M.; Fujimatsu, H.; Ezoe, A.; Shibata, K.; Shimizu, M.; Matsumoto,
S.; Tamaru, Y. Angew. Chem., Int. Ed. Engl. 1999, 38, 397.
(27) Very recently, a nickel-catalyzed cyclization of alkyne and tethered
aldehyde in the presence of triethylsilane was reported, in which the
mechanism of the cyclization was accounted for by σ bond metathesis of
triethylsilane and the nickel-oxygen bond of oxanickelacycle; see: Tang,
X.-Q.; Montgomery, J. J. Am. Chem. Soc. 1999, 121, 6098. The mechanism
of the cyclization using triethylsilane described in this article could be also
accounted for by the same mechanism based on the σ bond methathesis,
and its possibility could not be ruled out.

Cyclization of 1,3-Diene and a Tethered Carbonyl Group
J. Am. Chem. Soc., Vol. 122, No. 8, 2000 1633
MHz, 13C at 400 MHz), JEOL AL 400 (1H at 400 MHz, 13C at 100
MHz), or Bruker ARX-500 (1H at 500 MHz, 13C at 125 MHz) magnetic
resonance spectrometer. Infrared spectra were recorded on a Perkin-
Elmer FTIR 1605 spectrometer. Mass spectra were measured on JEOL
DX-303 and JEOL HX-110 mass spectrometers. Elemental analysis
were determined with Yanaco CHN CORDER MT-3.
was purified by silica gel column chromatography (hexane/EtOAc 5:1)
to afford 29 (28 mg, 70%, 29-T:29-I ) 86:14 based on H NMR).
1
Spectral data for 29-T: IR (neat) 3459, 3074, 3031, 2956, 2860, 1716,
1638, 1494, 1455 cm-1; 1H NMR (270 MHz, CDCl3) δ 1.31 (s, 3 H),
1.39-1.59 (m, 2 H), 1.60-1.76 (m, 3 H), 1.86-2.24 (m, 1 H), 2.05-
2.25 (m, 2 H), 2.26-2.42 (m, 1 H), 3.33 (dd, J ) 9.1, 7.1 Hz, 1 H),
3.50 (dd, J ) 9.1, 4.5 Hz, 1 H), 4.49 (d, J ) 12.2 Hz, 1 H), 4.53 (d,
J ) 12.2 Hz, 1 H), 5.06 (br d, J ) 10.0 Hz, 1 H), 5.09 (br d, J ) 17.2
Hz, 1 H), 5.87 (ddq, J ) 17.2, 10.0, 7.2 Hz, 1 H), 7.26-7.42 (m, 5 H);
MS m/z 260 (M+), 242, 169, 152, 91 (bp). Anal. Calcd for C17H24O2:
C, 74.42; H, 9.29. Found: C, 74.45; H, 9.36.
(1R*,2S*,3R*)-3-(Benzyloxymethyl)-2-[(2E)-4-methoxy-2-butenyl]-
cyclopentan-1-ol (30-T). Following the general procedure, a crude
material, which was obtained from the cyclization of 28 (55 mg, 0.19
mmol) using nickel complex 14 generated from Ni(acac)2 (49 mg, 0.19
mmol), PPh3 (100 mg, 0.38 mmol), and DIBAL-H (0.37 mL, 0.38
mmol) in the presence of 1,3-CHD (0.027 mL, 0.29 mmol) at 0 °C for
2 h, was purified by silica gel column chromatography (hexane/EtOAc
2:1) to afford 30-T (41 mg, 73%) as a single isomer: IR (neat) 3445,
3038, 2921, 2851, 1666, 1596, 1498, 1442 cm-1; 1H NMR (270 MHz,
CDCl3) δ 1.40-1.87 (m, 5 H), 1.93-2.13 (m, 2 H), 2.25 (dd, J ) 7.3,
7.3 Hz, 2 H), 3.30 (s, 3 H), 3.35 (dd, J ) 9.1, 6.3 Hz, 1 H), 3.47 (dd,
J ) 9.1, 6.4 Hz, 1 H), 3.86 (d, J ) 6.0 Hz, 2 H), 4.18-4.26 (m, 1 H),
4.48 (d, J ) 12.3 Hz, 1 H), 4.53 (d, J ) 12.3 Hz, 1 H), 5.62 (dt, J )
15.6, 6.0 Hz, 1 H), 5.75 (dt, J ) 15.6, 6.7 Hz, 1 H), 7.23-7.38 (m, 5
H); MS m/z 272 (M+ - H2O), 258, 240, 227, 91 (bp); HRMS (EI)
calcd for C18H24O2 (M+ - H2O) 272.1777, found 272.1768. Anal. Calcd
for C18H26O3: C, 74.45; H, 9.02. Found: C, 74.41; H, 9.05.
(1R*,2S*,3R*)-3-(Benzyloxymethyl)-2-(2-propenyl)cyclohexan-1-
ol (18-T). Following the general procedure, a crude material, which
was obtained from the cyclization of 11 (50 mg, 0.19 mmol) using
nickel complex 14 generated from Ni(acac)2 (50 mg, 0.19 mmol), PPh3
(101 mg, 0.39 mmol), and DIBAL-H (0.38 mL, 0.39 mmol) in the
presence of 1,3-CHD (0.028 mL, 0.29 mmol) at 0 °C for 2 h, was
purified by silica gel column chromatography (hexane/EtOAc 5:1) to
afford 18 (43 mg, 86%, 18-T:18-I ) 93:7 based on 1H NMR). Spectral
data for 18-T: IR (neat) 3392, 3056, 2934, 2858, 1641, 1495, 1452
cm-1; 1H NMR (270 MHz, CDCl3) δ 1.15-1.58 (m, 5 H), 1.63-1.87
(m, 3 H), 1.88-2.01 (m, 1 H), 2.15-2.29 (m, 1 H), 2.45-2.59 (m, 1
H), 3.37-3.50 (m, 2 H), 3.54 (dd, J ) 9.2, 3.1 Hz, 1 H), 4.47 (d, J )
12.1 Hz, 1 H), 4.52 (d, J ) 12.1 Hz, 1 H), 5.00-5.12 (m, 2 H), 5.88
(ddt, J ) 17.3, 10.2, 7.4 Hz, 1 H), 7.24-7.43 (m, 5 H); MS m/z 260
(M+), 242, 169, 91 (bp). Anal. Calcd for C17H24O2: C, 78.42; H, 9.29.
Found: C, 78.23; H, 9.35.
3-(Benzyloxymethyl)-2-(2-propenyl)cycloheptan-1-ol (20-T). Fol-
lowing the general procedure, a crude material, which was obtained
from the cyclization of 12 (50 mg, 0.18 mmol) using nickel complex
14 generated from Ni(acac)2 (47 mg, 0.18 mmol), PPh3 (96 mg, 0.37
mmol), and DIBAL-H (0.36 mL, 0.37 mmol) in the presence of 1,3-
CHD (0.026 mL, 0.28 mmol) at 23 °C for 2 h, was purified by silica
gel column chromatography (hexane/EtOAc 7:1) to afford 20-T (23
mg, 47% as an inseparable mixture of two isomers (ratio of 5:23 based
on 1H NMR): IR (neat) 3440, 3027, 2925, 2862, 1640, 1495 cm-1; 1H
NMR (270 MHz, CDCl3) δ 1.24-1.99 (m, 11 H), 1.93-2.53 (m, 2
H), 3.31-3.50 (m, 2 H), 3.75-3.89 (m, 23/28 H), 3.99-4.08 (m, 5/28
H), 4.45-4.56 (m, 2 H), 5.00-5.12 (m, 2 H), 5.70-5.94 (m, 1 H),
7.24-7.43 (m, 5 H); MS m/z 274 (M+), 256, 183, 91. Anal. Calcd for
C18H26O2: C, 78.79; H, 9.55. Found: C, 78.68; H, 9.60.
Typical Procedure for Cyclization Using a Stoichiometric Amount
of Nickel Complex 14 Generated from Ni(acac)2, PPh3, and
DIBAL-H (Cyclization of 10 in Scheme 2). To a cooled solution of
Ni(acac)2 (55 mg, 0.21 mmol) and PPh3 (112 mg, 0.42 mmol) in
degassed toluene (5.0 mL) was added dropwise a solution of DIBAL-H
(1.02 M in toluene, 0.42 mL, 0.42 mmol) at 0 °C, and the mixture was
stirred for 15 min at room temperature. To the mixture was added a
solution of 10 (55 mg, 0.21 mmol) in degassed toluene (5.0 mL) at 0
°C, and the mixture was stirred at 0 °C for 6 h. The reaction mixture
was hydrolyzed with 10% HCl aqueous solution, and the mixture was
extracted with ether. The organic layer was washed with saturated
NaHCO3 aqueous solution, and brine and dried over Na2SO4. After
removal of the solvent, the residue was purified by silica gel column
chromatography (hexane/EtOAc 5:1) to afford the cyclized product 15-I
(36 mg, 69%, E:Z ) 8:1). Spectral data for (E)-15-I: IR (neat) 3422,
3063, 3028, 2956, 2932, 1586, 1495, 1453, 1363 cm-1; 1H NMR (270
MHz, CDCl3) δ 1.42-1.60 (m, 3 H), 1.71 (d, J ) 5.5 Hz, 3 H), 1.80-
1.96 (m, 1 H), 1.96-2.12 (m, 1 H), 2.17-2.32 (m, 2 H), 3.30 (dd, J
) 9.2, 6.7 Hz, 1 H), 3.49 (dd, J ) 9.2, 3.7 Hz, 1 H), 4.10-4.17 (m,
1 H), 4.47 (d, J ) 12.1 Hz, 1 H), 4.53 (d, J ) 12.1 Hz, 1 H), 5.48 (dd,
J ) 15.4, 5.5 Hz, 1 H), 5.57 (dq, J ) 15.4, 5.5 Hz, 1 H), 7.28-7.37
(m, 5 H); MS m/z 246 (M+), 229, 169, 155, 91 (bp); HRMS m/z calcd
for C16H22O2 246.1621, found 246.1609. Anal. Calcd for C16H22O2:
C, 78.01; H, 9.00. Found: C, 77.98; H, 9.06.
General Procedure for Cyclization Using a Stoichiometric
Amount of Nickel Complex 14 in the Presence of 1,3-Cyclohexa-
diene. To a cooled suspension of Ni(acac)2 (100 mol % to substrate)
and PPh3 (200 mol % to substrate) in degassed toluene (0.04 M) was
added dropwise a solution of DIBAL-H (1.02 M in toluene, 200 mol
% to substrate) at 0 °C, and the mixture was stirred at room temperature
for 15 min. To the mixture was added 1,3-cyclohexadiene (150 mol %
to substrate) at 0 °C, and the solution was stirred for ∼3 min. Then a
solution of substrate in degassed toluene (0.04 M) was added to the
mixture, and the mixture was stirred. The reaction was monitored by
TLC. The reaction mixture was hydrolyzed with 10% aqueous HCl,
and the aqueous layer was extracted with ether. The organic layer was
washed with saturated aqueous NaHCO3 and brine, dried over Na2-
SO4, and concentrated. The residue was purified by silica gel column
chromatography to give the desired cyclized product.
(1R*,2S*,3R*)-3-Benzyloxymethyl-2-(2-propenyl)cyclopentan-1-
ol (15-T). Following the general procedure, a crude material, which
was obtained from the cyclization of 10 (49 mg, 0.20 mmol) using
nickel complex 14 generated from Ni(acac)2 (51 mg, 0.20 mmol), PPh3
(105 mg, 0.40 mmol), and DIBAL-H (0.41 mL, 0.41 mmol) in the
presence of 1,3-CHD (0.028 mL, 0.30 mmol) at 0 °C for 2 h, was
purified by silica gel column chromatography (hexane/EtOAc 5:1) to
afford 15 (30 mg, 61%, 15-T:15-I ) 95:5 based on 1H NMR). Spectral
data for 15-T: IR (neat) 3340, 3070, 3020, 2925, 1640, 1600, 1450,
1360 cm-1; 1H NMR (400 MHz, CDCl3) δ 1.41 (br s, 1 H), 1.43-1.54
(m, 1 H), 1.60-1.72 (m, 2 H), 1.77-1.86 (m, 1 H), 1.98-2.12 (m, 2
H), 2.19-2.32 (m, 2 H), 3.35 (dd, J ) 8.8, 6.3 Hz, 1 H), 3.37 (dd, J
) 8.8, 4.9 Hz, 1 H), 4.20-4.25 (m, 1 H), 4.50 (d, J ) 12.1 Hz, 1 H),
4.52 (d, J ) 12.1 Hz, 1 H), 5.00 (br.d, J ) 10.3 Hz, 1 H), 5.10 (ddd,
J ) 17.1, 2.0, 3.4 Hz, 1 H), 5.72 (dddd, J ) 17.1, 10.3, 7.8, 6.4 Hz,
1 H), 7.25-7.37 (m, 5 H); 13C NMR (100 MHz, CDCl3) δ 26.7 (CH2),
33.2 (CH2), 33.5 (CH2), 42.0 (CH), 47.5 (CH), 73.0 (CH2), 73.6 (CH2),
74.8 (CH), 115.4 (dCH2), 127.5 (Ar), 128.3 (Ar), 138.1 (CHd), 138.7
(Ar); MS m/z 247 (M+ + 1), 228, 169, 155, 91 (bp). Anal. Calcd for
C16H22O2: C, 78.01; H, 9.00. Found: C, 78.10; H, 8.89.
General Procedure for Catalytic Cyclization Using Ni(cod)2 and
PPh3 in the Presence of Et3SiH (Table 4). Ni(cod)2 and PPh3 were
dissolved in degassed toluene, and the mixture was stirred at 0 °C for
30 min. To the mixture were added Et3SiH (5 equiv) and a solution of
substrate in degassed toluene, and the mixture was stirred at the
temperature shown in Table 4. After usual workup, the residue was
purified by silica gel column chromatography to give the cyclized
product.
(1R*,2S*,3R*)-1-(Benzyloxymethyl)-2-[(E)-1-propenyl]-3-(trieth-
ylsilyloxy)cyclopentane (38-I). Following the general procedure, a
crude material, which was obtained from the cyclization of 10 (49 mg,
0.20 mmol) with Ni(cod)2 (5.5 mg, 0.02 mmol), PPh3 (10 mg, 0.04
mmol), and Et3SiH (0.16 mL, 1.0 mmol) at 23 °C for 21 h, was purified
(1R*,2S*,3R*)-3-(Benzyloxymethyl)-1-methyl-2-(2-propenyl)cy-
clopentan-1-ol (29-T). Following the general procedure, a crude
material, which was obtained from the cyclization of 23 (40 mg, 0.16
mmol) using nickel complex 14 generated from Ni(acac)2 (40 mg, 0.16
mmol), PPh3 (81 mg, 0.33 mmol), and DIBAL-H (0.32 mL, 0.32 mmol)
in the presence of 1,3-CHD (0.022 mL, 0.24 mmol) at 23 °C for 7 h,

1634 J. Am. Chem. Soc., Vol. 122, No. 8, 2000
Sato et al.
by silica gel column chromatography (hexane/ether 50:1) to afford 38-I
(50 mg, 70%): IR (neat) 3020, 2950, 1580, 1450, 1238 cm-1; 1H NMR
(270 MHz, CDCl3) δ 0.55 (q, J ) 7.7 Hz, 6 H), 0.94 (t, J ) 7.7 Hz,
9 H), 1.40-1.63 (m, 2 H), 1.65 (d, J ) 6.2 Hz, 3 H), 1.69-1.84 (m,
1 H), 1.95-2.09 (m, 2 H), 2.13-2.29 (m,1 H), 3.26 (dd, J ) 9.2, 7.3
Hz, 1 H), 3.49 (dd, J ) 9.2, 4.0 Hz, 1 H), 4.08-4.15 (m 1 H), 4.46 (d,
J ) 12.4 Hz, 1 H), 4.53 (d, J ) 12.4 Hz, 1 H), 5.37 (dq, J ) 15.5, 6.2
Hz, 1 H), 5.50 (dd, J ) 15.4, 8.8 Hz, 1 H), 7.27-7.33 (m, 5 H); MS
m/z 360 (M+), 331, 269, 87; HRMS calcd for C22H36O2Si 360.2486,
found 360.2463. Anal. Calcd for C22H36O2Si: C, 73.28; H, 10.06.
Found: C, 73.27; H, 9.96.
was stirred at the same temperature for 2 h. The reaction mixture was
hydrolyzed with 10% aqueous HCl, and the aqueous layer was extracted
with ether. After the usual workup, the cyclized product 15-I was
obtained in 82% yield.
i
Cyclization of 10 Using Ni(cod)2, PPh3, and Bu2-ALAC in the
Presence of 1,3-CHD (Scheme 24). In a similar manner, 15 was
1
obtained in 85% yield (48 mg, 15-T:15-I ) 98:2 based on H NMR)
from the reaction of 10 (56 mg, 0.23 mmol) with Ni(cod)2 (63 mg,
i
0.23 mmol), PPh3 (120 mg, 0.45 mmol), and Bu2-ALAC (0.5 M in
toluene, 1.0 mL, 0.5 mmol) in the presence of 1,3-CHD (0.033 mL,
0.34 mmol) in toluene at 0 °C for 2 h.
(1R*,2S*,3R*)-1-(Benzyloxymethyl)-2-[(E)-4-methoxy-1-butenyl]-
3-(triethyl-silyloxy)cyclopentane (53). Following the general proce-
dure, a crude material, which was obtained from the cyclization of 28
(45 mg, 0.16 mmol) with Ni(cod)2 (8.6 mg, 0.031 mmol), PPh3 (16
mg, 0.062 mmol), and Et3SiH (0.13 mL, 0.78 mmol) at 30 °C for 3.5
h, was purified by silica gel column chromatography (hexane/EtOAc
15:1) to afford 53 (43 mg, 67%): IR (neat) 3030, 2953, 2563, 2730,
General Procedure for Catalytic Cyclization Using Ni(cod)2,
PPh3, and iBu2-ALAC. Ni(cod)2 and PPh3 were dissolved in degassed
toluene (0.04 M), and the mixture was stirred at 0 °C for 20 min. To
i
the mixture was added Bu2-ALAC (1.0 M in toluene, 1.5 equiv to
substrate), and the mixture was stirred at room temperature for 15 min.
To the mixture was added a solution of substrate in degassed toluene
(0.04 M) at 0 °C, and the mixture was stirred at the temperature
indicated in Scheme 2 or Table 1. The reaction mixture was hydrolyzed
with saturated NH4Cl aqueous solution, and the aqueous layer was
extracted with ether. After the usual workup, the residue was purified
by silica gel column chromatography to give the cyclized product.
Cyclization of 10 (Scheme 26). Following the general procedure, a
crude material, which was obtained from 10 (57 mg, 0.23 mmol), Ni-
1
1724, 1604, 1587, 1490, 1450 cm-1; H NMR (270 MHz, CDCl3) δ
0.56 (q, J ) 7.8 Hz, 6 H), 0.95 (t, J ) 7.8 Hz, 9 H), 1.41-1.86 (m, 3
H), 1.94-2.12 (m, 1 H), 2.14-2.32 (m, 2 H), 2.29 (dd, J ) 7.0, 7.0
Hz, 2 H), 3.27 (dd, J ) 9.1, 7.1 Hz, 1 H), 3.33 (s, 3 H), 3.37 (t, J )
7.0 Hz, 2 H), 3.49 (dd, J ) 9.1, 4.9 Hz, 1 H), 4.11-4.17 (m, 1 H),
4.47 (d, J ) 12.2 Hz, 1 H), 4.53 (d, J ) 12.2 Hz, 1 H), 5.36 (dt, J )
15.6, 6.6 Hz, 1 H), 5.59 (dd, J ) 15.5, 8.9 Hz, 1 H), 7.22-7.38 (m, 5
H); MS m/z 404 (M+), 375, 283, 171, 91; HRMS (EI) calcd for
C24H40O3Si 404.2748, found 404.2762. Anal. Calcd for C24H40O3Si:
C, 71.24; H, 9.96. Found: C, 71.31; H, 10.01.
(1R*,2R*,3S*)-1-(Benzyloxymethyl)-2-[(E)-1-propenyl]-3-(trieth-
ylsilyloxy)cyclohexane (54). Following the general procedure, a crude
material, which was obtained from the cyclization of 11 (56 mg, 0.22
mmol) with Ni(cod)2 (12 mg, 0.044 mmol), PPh3 (23 mg, 0.087 mmol),
and Et3SiH (0.17 mL, 1.1 mmol) at 23 °C for 1.5 h, was purified by
silica gel column chromatography (hexane/ether 100:3) to afford 54
(58 mg, 71%): IR (neat) 2937, 2879, 1494, 1455 cm-1; 1H NMR (270
MHz, CDCl3) δ 0.56 (q, J ) 7.8 Hz, 6 H), 0.94 (t, J ) 7.8 Hz, 9 H),
1.00-1.50 (m, 4 H), 1.64 (dd, J ) 6.4, 1.6 Hz, 3 H), 1.69-1.84 (m,
2 H), 1.84-1.98 (m, 2 H), 3.21 (dd, J ) 9.1, 7.6 Hz, 1 H), 3.29 (ddd,
J ) 10.2, 9.5, 4.5 Hz, 1 H), 3.51 (dd, J ) 9.1, 3.3 Hz, 1 H), 4.41 (d,
J ) 12.0 Hz, 1 H), 4.49 (d, J ) 12.0 Hz, 1 H), 5.07 (ddq, J ) 15.2,
9.2, 1.6 Hz, 1 H), 5.39 (dq, J ) 15.2, 6.4 Hz, 1 H), 7.23-7.42 (m, 5
H); MS m/z 374 (M+), 346, 283, 91. Anal. Calcd for C23H38O2Si: C,
73.34; H, 10.22. Found: C, 73.55; H, 10.38.
i
(cod)2 (3.2 mg, 0.012 mmol), PPh3 (6.1 mg, 0.023 mmol), and Bu2-
ALAC (0.35 mL, 0.35 mmol), was purified by silica gel column
chromatography (hexane/AcOEt 6:1) to afford 15-T (48 mg, 84%),
whose spectral data were completely identical with those noted above.
Cyclization of 28 (Table 5, Run 1). Following the general
procedure, a crude material, which was obtained from 28 (61 mg, 0.21
mmol), Ni(cod)2 (5.8 mg, 0.021 mmol), PPh3 (11 mg, 0.042 mmol),
i
and Bu2-ALAC (0.32 mL, 0.32 mmol), was purified by silica gel
column chromatography (hexane/AcOEt 5:2) to afford 30-T (46 mg,
75%), whose spectral data were completely identical with those noted
above.
Cyclization of 11 (Table 5, Run 2). Following the general
procedure, a crude material, which was obtained from 11 (60 mg, 0.23
mmol), Ni(cod)2 (3.2 mg, 0.012 mmol), PPh3 (6.1 mg, 0.023 mmol),
i
and Bu2-ALAC (0.35 mL, 0.35 mmol), was purified by silica gel
column chromatography (hexane/AcOEt 6:1) to afford 18-T (58 mg,
96%), whose spectral data were completely identical with those noted
above.
Cyclization of 12 (Table 5, Run 3). Following the general
procedure, a crude material, which was obtained from 12 (62 mg, 0.23
mmol), Ni(cod)2 (3.1 mg, 0.011 mmol), PPh3 (6.0 mg, 0.023 mmol),
(1R*,2R*,3S*)-1-(Benzyloxymethyl)-2-[(E)-1-propenyl]-3-(trieth-
ylsilyloxy)cycloheptane (55). Following the general procedure, a crude
material, which was obtained from the cyclization of 12 (50 mg, 0.18
mmol) with Ni(cod)2 (10 mg, 0.037 mmol), PPh3 (19 mg, 0.073 mmol),
and Et3SiH (0.15 mL, 0.92 mmol) at 30 °C for 12 h, was purified by
silica gel column chromatography (hexane/ether 100:1) to afford 55
(48 mg, 66%): IR (neat) 3027, 2940, 1657, 1499, 1456 cm-1; 1H NMR
(270 MHz, CDCl3) δ 0.55 (q, J ) 8.0 Hz, 6 H), 0.94 (t, J ) 8.0 Hz,
9 H), 1.20-2.00 (m, 10 H), 1.62 (d, J ) 5.7 Hz, 3 H). 3.18 (dd, J )
9.1, 8.1 Hz, 1 H), 3.42 (dd, J ) 9.1, 3.6 Hz, 1 H), 3.70 (ddd, J ) 7.5,
5.5, 2.0 Hz, 1 H), 4.41 (d, J ) 12.0 Hz, 1 H), 4.49 (d, J ) 12.0 Hz, 1
H), 5.19 (dd, J ) 15.1, 8.8 Hz, 1 H), 5.31 (dq, J ) 15.1, 5.7 Hz, 1 H),
7.24-7.42 (m, 5 H); MS m/z 388 (M+), 359, 293, 91. Anal. Calcd for
C24H40O2Si: C, 74.17; H, 10.37. Found: C, 74.28; H, 10.45.
i
and Bu2-ALAC (0.34 mL, 0.34 mmol), was purified by silica gel
column chromatography (hexane/AcOEt 6:1) to afford 20-T (36 mg,
57%) as an inseparable mixture of two isomers (ratio of 1/4.2), whose
spectral data were completely identical with those noted above.
Acknowledgment. The authors are grateful to Drs. Koji
Hayashi, Takao Katsuhara, and Koji Takagi (Tsumura Central
Research Laboratories, Tsumura & Co.) for their help in X-ray
structural analysis of 17 and V. M.T. thanks the Japan Society
for the Promotion of Science (JSPS) for a Research Fellowship
for Young Scientists.
Cyclization of 10 Using Ni(cod)2, PPh3, and iBu2-ALAC (Scheme
24). Ni(cod)2 (62 mg, 0.23 mmol) and PPh3 (118 mg, 0.45 mmol) were
dissolved in degassed toluene (5.1 mL), and the mixture was stirred at
Supporting Information Available: Experimental proce-
dures and spectral data for the synthesis of substrates 10-12,
23, and 28; spectral data for 16, 17, 19, 31, 32, 51, I, II, IIIA,
IV, and V. This material is available free of charge via the
Internet at http://pubs.acs.org.
i
0 °C for 30 min. To the mixture was added Bu2-ALAC (0.5 M in
toluene, 1.0 mL, 0.5 mmol), and the mixture was stirred at room
temperature for 15 min. To the mixture was added a solution of 10 (55
mg, 0.23 mmol) in degassed toluene (5.1 mL) at 0 °C, and the mixture
JA991241D