A R T I C L E S
Kimura et al.
v/v) to give 1d in 76% yield (189 mg) and 2b in 15% yield (25 mg).
Rf (1d) ) 0.46, Rf (2b) ) 0.39 (hexane/ethyl acetate ) 4:1, v/v). The
structure of 2b was confirmed by comparison of the spectral data (IR,
1H NMR) with those of authentic sample.8a
elimination and reductive elimination delivering a hydrogen to
the allylic terminus proximal to the OH-bearing carbon and
finally liberates a mixture of 8, an Ni0 species, and ethylene.
The lack of regioselectivity encountered in the intermolecular
reactions of unsymmetrical alkynes (Table 2) might be rational-
ized on the basis of the structure of an intermediate XIV, shown
in the red rectangle of Scheme 5 (an equivalent to VI for an
intramolecular reaction). The central nickelII of XIV is coor-
dinatively saturated and located in the middle between an allylic
terminal CH2 group and a methyl group of a similar steric size;
hence, an access of unsymmetric alkynes to the NiII atom is
expected to be free from any restrictions imposed by steric,
electronic, and coordination effects. On the other hand, large
steric and coordination effects are expected, if an intermediate
such as VII were involved (path b), where the alkyne carbon
bearing bulky substituents as well as the substituents of
coordination ability would monopolize the C2 position, since
the position is free from strain and located nicely for the
coordinating groups to interact with the NiII metal as L.
(3E,6E)-6-Ethyl-1-(2-furyl)-7-methyl-3,6-nonadien-1-ol (1d). IR
(neat) 3361 (s), 2963 (s), 1439 (m), 1373 (m), 1149 (m), 1010 (s), 968
(m), 734 (s) cm-1; 1H NMR (400 MHz, CDCl3) δ 0.92 (t, J ) 7.7 Hz,
3 H), 0.95 (t, J ) 7.7 Hz, 3 H), 1.60 (s, 3 H), 1.98 (q, J ) 7.7 Hz, 2
H), 2.02 (q, J ) 7.7 Hz, 2 H), 2.56 (m, 2 H), 2.72 (d, J ) 6.2 Hz, 2
H), 4.68 (t, J ) 6.2 Hz, 1 H), 5.36 (dt, J ) 15.4, 7.1 Hz, 1 H), 5.52
(dt, J ) 15.4, 6.2 Hz, 1 H), 6.22 (br d, J ) 3.3 Hz, 1 H), 6.31 (dd, J
) 1.8, 3.3 Hz, 1 H), 7.36 (br d, J ) 1.8 Hz, 1 H); HRMS calcd for
C16H24O2: 248.1776, found m/z (relative intensity) 249 (M+ + 1, 7),
248.1777 (M+, 41), 230 (18), 152 (100), 136 (22).
Ni-Catalyzed Reaction of Me2Zn, 1,ω-Dienyne 5h and Benzal-
dehyde (run 2, Table 4). To a homogeneous solution of Ni(acac)2
(25.6 mg, 0.1 mmol) and 5h (303 mg, 1 mmol) in dry THF (5 mL)
were successively added benzaldehyde (212 mg, 2.0 mmol) and
dimethylzinc (2.4 mL, 1 M in hexane). The mixture was stirred at room
temperature for 1 h under N2 and then poured into ice-cold water. After
addition of 2M HCl (5 mL), the mixture was extracted twice with ethyl
acetate. The combined organic extracts were washed with saturated
NaHCO3 and with brine, dried (MgSO4), and concentrated in vacuo.
The residue was purified by flash column chromatography over silica
gel (hexane/ethyl acetate ) 12:1, v/v) to give 6h in 96% yield (408
mg). Rf (6h) ) 0.20 (hexane/ethyl acetate ) 4:1, v/v).
An intervention of XIV might also be supported by the results
in runs 11-16, Table 1. The alkyne in XIV might effectively
prevent another butadiene molecule from approaching the NiII
and, hence, prohibit a 1:2:1 adduct formation of Me2Zn,
butadiene, and a bulky carbonyl (cf., eq 2).8a
(3S*,4′S*)-4-Isopropylidene-3-[4-hydroxy-4-phenyl-(1E)-butenyl]-
N-(p-toluenesulfonyl)piperidine (6h). mp 92.0-93.0 °C (hexane-
dichloromethane); IR (KBr disk) 3460 (m), 1340 (s), 1169 (s), 1103
(m), 932 (m) cm-1; 1H NMR (400 MHz, CDCl3) δ 1.61 (s, 6 H), 2.08-
2.24 (m, 2 H), 2.32 (dd, J ) 3.7, 11.4 Hz, 1 H), 2.40 (s, 3 H), 2.43-
2.51 (m, 3 H), 3.37 (m, 1 H), 3.76 (m, 1 H), 3.78 (dm, J ) 11.4 Hz,
1 H), 4.73 (br dd, J ) 5.5, 7.0 Hz, 1 H), 5.50 (ddt, J ) 1.5, 15.6, 7.1
Hz, 1 H), 5.67 (br dd, J ) 5.7, 15.6 Hz, 1 H), 7.24-7.36 (m, 5 H),
7.30 (d, J ) 8.2 Hz, 2 H), 7.62 (d, J ) 8.2 Hz, 2 H); 13C NMR (100
MHz, CDCl3) δ 19.9, 20.1, 21.5, 25.4, 39.8, 42.7, 47.1, 51.1, 73.7,
125.9, 126.8, 127.2, 127.4, 127.7, 128.4, 129.6, 133.4, 134.0, 143.3,
144.1; HRMS calcd for C25H31O3NS: 425.2025. Found m/z (relative
intensity): 425.2042 (M+, 1), 407 (2), 319 (100), 278 (17).
To detect intermediates, we examined the reaction of 5b in
the absence of an aldehyde under catalytic (conditions a) and
stoichiometric conditions using 1 equiv of Ni(cod)2. However,
the reactions proceeded slowly, taking almost 2-3 h for
complete disappearance of 5b. In all cases, the reaction mixtures
became very dirty, showing many tailing spots on TLC. No
fractions separated at random by means of flash column
1
chromatography showed characteristic absorptions in the H
NMR spectra ascribable to compounds expected from the
intermediates in Scheme 5 (including a [4 + 2]cycloaddition
product). These results indicate that a carbonyl compound is
essential to selectively promote the facile connection reaction
of R2Zn, an alkyne, and a diene.
Conclusion
We have demonstrated that Ni(acac)2 can be nicely utilized
as a catalyst for the four-component connection reaction of Me2-
Zn, alkynes, 1,3-butadiene, and carbonyl compounds in this
order in 1:1:1:1 ratio to furnish (3E,6Z)-octadien-1-ols 1 with
high stereoselectivity and in excellent yields. The same condi-
tions are applicable to the coupling reaction of Me2Zn, 1,ω-
dienynes 5, and carbonyls, furnishing 1-alkylidene-2-(4′-hydroxy-
(1′E)-alkenyl)cyclopentanes and -cyclohexanes 6 and their
oxygen and nitrogen heterocycles with an excellent level of 1,5-
diastereoselectivity. The reaction proceeds smoothly at room
temperature, tolerates an ester, a hydroxy, an allyl ether, a
propargyl ether, an allylamino, and pyridyl functionalities,
andaccommodates a variety of aromatic and aliphatic alkynes
and carbonyls (aromatic and aliphatic aldehydes and ketones).21
The reaction constitutes one of the few Ni-catalyzed nucleophilic
addition reactions of alkynes and dienes toward carbonyl
compounds.
After all, pathway a seems to be the most plausible, which
accounts for all the observations, regardless of inter- and
intramolecular four-component connection reactions: the high
yield formation of 1, with no regioselectively for unsymmetric
alkynes, the 1,5-diastereoselective formation of 6, the selective
formation of 2 (1:1:1 adducts and not 1:2:1 adducts of Me2Zn,
butadiene, and carbonyls). The mechanism is especially backed
by the fact that the reaction in eq 8 gives rise to a mixture of
6m and 8.
Typical Experimental Procedure
Ni-Catalyzed Four-Component Connection Reaction of Me2Zn,
3-Hexyne, 1,3-Butadiene, and 2-Furaldehyde (run 7, Table 1). Into
a flask containing Ni(acac)2 (25.7 mg, 0.1 mmol) purged with N2 were
successively added THF (5 mL), 3-hexyne (98.6 mg, 1.2 mmol), 1,3-
butadiene (0.11 mL, 1.2 mmol), 2-furaldehyde (96.1 mg, 1.0 mmol),
and dimethylzinc (2.4 mL, 1 M in hexane) via syringe at room
temperature. The homogeneous solution was stirred at room temperature
for 30 min under N2 and then quenched by adding 2 M HCl (25 mL).
The mixture was extracted twice with ethyl acetate, and the combined
extracts were washed with saturated NaHCO3 and with brine, dried
(MgSO4), and concentrated in vacuo. The residue was purified by flash
column chromatography over silica gel (hexane/ethyl acetate ) 16:1,
(21) The present catalytic system seems not to be applicable to thioether
compounds, e.g., phenyl 1-propynyl sulfide and 2-(2-butynyl)-2-(2,4-
pentadienyl)-1,3-dithiane. Under usual conditions, the reactions of these
sulfides proceeded slowly (at room temperature for 3-12 h) and provided
intractable mixtures of products. The reaction of bis(trimethylsilylmethyl)-
zinc with 5j and benzaldehyde was completed within 30 min; however, a
complex mixture of products resulted (7 to 8 spots on TLC).
9
208 J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005