Nickel catalyzed stereoselective conjugate addition of dimethylzinc upon aldimines across 1,3-dien-8-ynes and 1,3-dien-9-ynes

Article abstract of DOI:10.1039/b605728d

Ni(acac)₂ catalyzes the five-component connection reaction of Me₂Zn, alkynes, diene (of 1,3-dien-8-ynes and 1,3-dien-9-ynes), aldehydes and anisidine to furnish cyclic dienyl amines anti-2 with high remote 1,5-diastereoselectivity. T

Full text of DOI:10.1039/b605728d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Nickel catalyzed stereoselective conjugate addition of dimethylzinc
upon aldimines across 1,3-dien-8-ynes and 1,3-dien-9-ynes{
Masanari Kimura,b Masahiko Mori,b Nahoko Mukai,b Keisuke Kojimab and Yoshinao Tamaru*a
Received (in Cambridge, UK) 21st April 2006, Accepted 17th May 2006
First published as an Advance Article on the web 31st May 2006
DOI: 10.1039/b605728d
reaction shows wider flexibility regarding the type of aldehydes (see
below).
Ni(acac)2 catalyzes the five-component connection reaction of
Me2Zn, alkynes, diene (of 1,3-dien-8-ynes and 1,3-dien-9-ynes),
aldehydes and anisidine to furnish cyclic dienyl amines anti-2
with high remote 1,5-diastereoselectivity.
The reaction can be undertaken in one flask with great ease (run
1, Table 1): a mixture of benzaldehyde (1 mmol) and p-anisidine
(2 mmol) in dry THF (1.5 mL) was stirred at room temperature
overnight. This mixture, without removing water produced, was
treated with Ni(acac)2 (0.05 mmol) dissolved in THF (1.5 mL), 1a
(0.5 mmol) and Me2Zn (3.6 mmol, 1 M hexane) at room
temperature. The reaction was complete within 1 h and provided
2a in an excellent yield (90–100%) after usual workup and
purification by column chromatography over silica gel.{
Interestingly, both an excess amount of p-anisidine (AN) and
water produced seem to be essential, and cooperate to promote the
reaction (Table 2).6 Isolated aldimine itself did not undergo the
expected addition reaction; after a long reaction, an intractable
mixture of products containing unreacted 1a resulted (run 1). In
the presence of either water (1 mmol) or AN (1 mmol), the
reaction was complete within 1 h, and 2a was isolated in 19 or 71%
yield, respectively. The yield of 2a was greatly improved in the
presence of both water and AN (run 4). At the moment, the roles
of these additives are not clear, but we speculate that AN might
activate Me2Zn by coordination to it, dissociating Me2Zn-
oligomers to the lower oligomers or monomer, and water produces
some zinc oxide species, which might serve as a Lewis acid to
activate an aldimine.
Aldimines are among the least reactive electrophiles. Accordingly,
the C–C bond formation reactions of aldimines with organome-
tallic reagents have been achieved mostly by introduction of
electron-withdrawing activating groups either on the amine1 or on
the aldehyde moiety.2 This statement does not necessarily hold for
the reactions of organic transition metals. In fact, a few precedents
indicate that aldimines are more reactive than3 or show
comparable reactivity to aldehydes.4 In transition metal catalysis,
it is believed that the strong coordination of nitrogen to transition
metals activate aldimines as electrophiles.
Recently, we disclosed that Ni(acac)2 (acac = acetylacetonato)
served as a catalyst for the conjugate addition reaction of Me2Zn
upon aldehydes across 1,3-dien-8-ynes and 1,3-dien-9-ynes 1 (n =
1, 2) and furnished cyclic dienyl alcohols 29 with high 1,5-syn
stereoselectivity and in good yields (Scheme 1).5 Here we report
that the reaction can be successfully extended to aldimines;
aldimines showed comparable reactivity to aldehydes under the
Ni-catalyzed conditions and provided the nitrogen analogues,
cyclic dienyl amines 2. The reactions were complete at room
temperature within a few hours in most cases and showed better
performance than aldehydes regarding the isolated yields and
stereoselectivity. Notably, aldimines and aldehydes showed an
opposite stereoselectivity; aldehydes provided 29 in ratios of anti/
syn = 1 : 7 to 1 : >30, while aldimines exclusively provided 1,5-anti-
2 as single diastereomers in all cases examined. Furthermore, the
Table 1 summarizes the reactions of in-situ generated benzalde-
hyde-AN imine with a wide structural variety of 1, encompassing
three- and four-carbon tethers containing oxygen and nitrogen
atoms in their skeleton. As compared with the reactions with
benzaldehyde, the yields of 2a–2h are either comparable or better
(e.g., the alcohol analogues of 2a, 2b and 2c have been isolated in
63, 70 and 61% yields, respectively).5 A phenyl-substituted 1,3-
dien-8-yne 1c was unreactive and required heating at 50 uC for an
exceptionally long period of time.
The structure of 2g was determined unequivocally by X-ray
crystallographic analysis.7 The structures of other isomers were
tentatively assigned by analogy.8
Furthermore, the reaction showed wide flexibility for a variety
of aromatic aldehydes and aliphatic aldehydes. The results are
summarized in Fig. 1. Aromatic aldehydes bearing either electron-
donating (2i) or withdrawing group (2j), as well as a heteroaro-
matic aldehyde (2k) worked similarly well. Interestingly, judging
from the reaction time, aliphatic aldehyde imines, 2l and 2m, seem
to be more reactive than aromatic aldehyde imines.
Scheme 1 Opposite 1,5-diastereoselection between aldehydes and
aldimines.
aDepartment of Applied Chemistry, Faculty of Engineering, Nagasaki
University, 1-14 Bunkyo-Machi, Nagasaki, 852-8521, Japan.
E-mail: tamaru@net.nagasaki-u.ac.jp
bGraduate School of Science and Technology, Nagasaki University, 1-14
Bunkyo-Machi, Nagasaki, 852-8521
The compatibility of the reaction with water encouraged us to
examine the reaction with lactols. Curiously, 2-hydroxy-1-oxacy-
clohexane was unreactive under the usual conditions and no
{ Electronic supplementary information (ESI) available: Analytical and
spectral data of 2a–o. See DOI: 10.1039/b605728d
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2813–2815 | 2813

View Article Online
Table 1 Ni-catalyzed coupling of Me2Zn and benzaldehyde-p-anisi-
dine imine across 1,3-dien-8-ynes or 1,3-dien-9-ynes 1a
Run
1
1
t/h
Isolated yield of product 2 (%)
1
2
3
1
24b
Fig. 1 Reaction of 1a with aromatic and aliphatic aldehyde-AN imines.
4
2
1
expected product was obtained [eqn (1)], while pre-treatment of a
lactol with AN overnight and application of the usual reaction
conditions promoted the reaction as usual. The reaction was
complete within 1–3 h at room temperature and furnished the
expected cyclic dienyl amino alcohols 2n and 2o in modest to good
yields as single diastereomers [eqn (2)].
5
6
(1)
0.5
7
8
a
3
(2)
0.5
A working hypothesis that accommodates 1,5-anti selectivity of
2, being opposite to that of aldehydes, is outlined in Scheme 2. The
scenario is almost the same as that described for the reaction of 1
with aldehydes.5 Although, an intermediate I0, favoured for the
reaction with an aldehyde, exposing the substituent R1 on an
quasi-equatorial position and hence with an anti-conformation
between R1 and C4–C5, is not responsible any longer for the
reaction with an aldimine, since in a square planer NiII geometry
of I9, an alkyne and the PMP substituent on N might experience a
severe gauche repulsion. In an intermediate I, on the other hand, a
gauche repulsion between R1 and C4–C5 arises, but this repulsion
is expected to be much smaller than that between alkyne and PMP
in I9. Methyl transfer from Me2Zn to NiII might form II, which
would undergo alkyne insertion in a syn fashion to give III and
hence 1,5-anti-2 via reductive elimination.
Reaction conditions: benzaldehyde (1 mmol) and p-anisidine
(2 mmol) in THF (1.5 mL) at room temperature overnight, then
Ni(acac)2 (0.05 mmol) dissolved in THF (1.5 mL), a 1,v-dienyne 1
(0.5 mmol), and Me2Zn (3.6 mmol, 1 M hexane) at room
temperature under N2 for the period of time indicated. PMP =
p-methoxyphenyl, M = CO2Me. At 50 uC.
b
Table 2 Effect of additives for the reaction of 1a (0.5 mmol), isolated
aldimine (benzaldehyde-AN, 1 mmol) and Me2Zn
Run Additive (mmol)
Me2Zn/mmol t/h Yield of 2a (%)
1
2
3
4
a
None
H2O (1)
p-Anisidine (1)
1.2a
2.4
2.4
19
1
0.5 71
0.5 85
0
19
Although 2n and 2o were obtained as single diastereomers, their
stereochemistry is not known at the moment. The similarly large
steric environment (PMP and ZnMe) around N in the most
probable intermediate IV (Scheme 2) prevents from structure
determination by analogy with the mechanistic scenario. All
p-Anisidine (1), H2O (1) 3.6b
The amount of Me2Zn used is the same as that optimized for the
reactions with aldehydes.5 The amount of Me2Zn is the same as
that applied to the reactions shown in Table 1.
b
2814 | Chem. Commun., 2006, 2813–2815
This journal is ß The Royal Society of Chemistry 2006

View Article Online
13.2 Hz, 1 H), 2.18–2.22 (m, 2 H), 2.57 (dd, J = 8.5, 13.2 Hz, 1 H), 2.83–
2.92 (m, 2 H), 3.28 (ddm, J = 5.9, 8.5 Hz, 1 H), 3.30 (m, 1 H), 3.63 (t, J =
6.3 Hz, 2 H), 3.70 (s, 3 H), 3.71 (s, 3 H), 3.74 (s, 3 H), 5.31 (dm, J = 15.1 Hz,
1 H), 5.36 (dm, J = 15.1 Hz, 1 H), 6.55 (br d, J = 8.5 Hz, 2 H), 6.75 (d, J =
8.5 Hz, 2 H); 1H NMR (400 MHz, THF-d8) d 1.32–1.56 (m, 6 H), 1.57 (s,
3 H), 1.63 (s, 3 H), 2.00 (dd, J = 6.2, 13.1 Hz, 1 H), 2.13 (dt, J = 13.7,
6.8 Hz, 1 H), 2.18 (dt, J = 13.7, 7.6 Hz, 1 H), 2.53 (dd, J = 8.4, 13.1 Hz,
1 H), 2.83 (br d, J = 16.6 Hz, 1 H), 2.89 (br d, J = 16.6 Hz, 1 H), 3.22 (br
dd, J = 6.2, 7.6 Hz, 1 H), 3.28 (br d, J = 8.4 Hz, 1 H), 3.40–3.50 (m, 2 H),
3.61 (s, 3 H), 3.62 (s, 3 H), 3.63 (s, 3 H), 3.93 (br s, 1 H), 5.28 (dd, J = 7.6,
14.9 Hz, 1 H), 5.41 (dt, J = 14.9, 6.8 Hz, 1 H), 6.47 (d, J = 9.0 Hz, 2 H),
6.64 (d, J = 9.0 Hz, 2 H); 13C NMR (100 MHz, CDCl3) d 20.8, 21.6, 22.2,
32.7, 33.9, 36.7, 38.7, 41.2, 44.2, 52.6, 54.1, 55.8, 59.1, 62.7, 114.9, 125.2,
125.9, 132.9, 135.8, 152.0, 172.1, 172.2; HRMS: calc. for C27H39NO6:
473.2777. Found m/z (relative intensity) 474 (M+ + 1, 32), 473.2770 (M+,
100), 472 (7), 456 (2), 442 (13).
1 (a) D. J. Weix, Y. Shi and J. A. Ellman, J. Am. Chem. Soc., 2005, 127,
1092–1093; (b) K. Matsui, S. Takizawa and H. Sasai, J. Am. Chem. Soc.,
2005, 127, 3680–3681; (c) S.-W. Li and R. A. Batey, Chem. Commun.,
2004, 1382–1383; (d) X.-F. Zhu, J. Lan and O. Kwan, J. Am. Chem. Soc.,
2003, 125, 4716–4717; (e) N. Hermanns, S. Dahmen, C. Bohm and
S. Bra¨se, Angew. Chem., Int. Ed., 2002, 41, 3692–3694; (f) H. Fujihara,
K. Magai and K. Tomioka, J. Am. Chem. Soc., 2000, 122, 12055–12056.
2 (a) C. Ogawa, M. Sugiura and S. Kobayashi, Angew. Chem., Int. Ed.,
2004, 43, 6491–6493; (b) M. Marigo, A. Kjærsgaard, K. Juhl,
N. Gathergood and K. A. Jørgensen, Chem. Eur. J., 2003, 9,
2359–2367; (c) Y. Niwa and M. Shimizu, J. Am. Chem. Soc., 2003,
125, 3720–3721; (d) A. Co´rdova, W. Notz, G. Zhong, J. M. Betancort
and C. F. Barbas, III, J. Am. Chem. Soc., 2002, 124, 1842–1843.
3 H. Nakamura, H. Iwama and Y. Yamamoto, J. Am. Chem. Soc., 1996,
118, 6641–6647, and references therein.
4 (a) M. Shimizu, M. Kimura, T. Watanabe and Y. Tamaru, Org. Lett.,
2005, 7, 637–640; (b) M. Kimura, A. Miyachi, K. Kojima, S. Tanaka and
Y. Tamaru, J. Am. Chem. Soc., 2005, 127, 10117; (c) K. Kojima,
M. Kimura and Y. Tamaru, Chem. Commun., 2005, 4717–4719; (d)
M. Kimura, M. A. Miyachi, K. Kojima, S. Tanaka and Y. Tamaru,
J. Am. Chem. Soc., 2004, 126, 14360–14361; (e) N. Solin, J. Kjellgren and
K. J. Szabo´, J. Am. Chem. Soc., 2004, 126, 7026–7033; (f) S. Sakaguchi,
T. Mizuta, M. Furuwan, T. Kubo and Y. Ishii, Chem. Commun., 2004,
1638–1639; (g) J. L. Davis, R. Dhawan and B. A. Arndtsen, Angew.
Chem., Int. Ed., 2004, 43, 590–595; (h) S. J. Patel and T. F. Jamison,
Angew. Chem., Int. Ed., 2004, 43, 3941–3944; (i) S. Sebelius, O. A. Wallner
and K. J. Szabo´, Org. Lett., 2003, 5, 3065–3068; (j) R. A. Fernandes,
A. Stimac and Y. Yamamoto, J. Am. Chem. Soc., 2003, 125,
14133–14139; (k) S. J. Patel and T. F. Jamison, Angew. Chem., Int.
Ed., 2003, 42, 1364–1367; (l) S.-K. Kang, K.-J. Kim and Y.-T. Hong,
Angew. Chem., Int. Ed., 2002, 41, 1584–1586.
Scheme 2 Rationale for the selective formation of 1,5-anti-2.
attempts have been unsuccessful as yet to obtain crystalline solids
of 2n and 2o and their derivatives suitable for X-ray crystal-
lographic analysis [e.g., N-PMP and N-H pyrrolidine derivatives
and their HClO4 salts formed by cyclization via intramolecular
amination of 2n (TsCl/pyridine) and PMP deprotection (CAN/
CH3CN–H2O)].9
In summary, we demonstrated that Ni(acac)2 catalytically
promoted the five-component connection reactions of Me2Zn,
alkynes, diene (of 1,3-dien-8-ynes and 1,3-dien-9-ynes), aldehydes
and anisidine to furnish cyclic dienyl amines 2. Despite the low
reactivity, aldimines generated in situ showed comparable
reactivity to aldehydes under the nickel catalysis and even showed
better performance than aldehydes regarding the yields (2 vs. 29)
and stereoselectivity, providing 1,5-anti-2 as single diastereomers.
Furthermore, although lactols failed, the corresponding lactamines
successfully underwent the five-component connection reaction to
give cyclic dienyl amino alcohols 2n,o in acceptable yields.
This work was supported by the Ministry of Education, Culture
Sports, Science and Technology, Japanese Government [Grant-in-
Aid for Scientific Research B (16350058) and Priority Areas
(17035065)].
5 (a) M. Kimura, A. Ezoe, M. Mori and Y. Tamaru, J. Am. Chem. Soc.,
2005, 127, 201–209; (b) A. Ezoe, M. Kimura, T. Inoue, M. Mori and
Y. Tamaru, Angew. Chem., Int. Ed., 2002, 41, 2784–2786.
6 Similar acceleration effect of an excess amine and water has been reported
in ref. 4c,d.
Notes and references
{ Reactions are typically performed as follows (see eqn (2)): A mixture of
p-anisidine (246 mg, 2 mmol) and 2-hydroxy-1-oxacyclohexane (103 mg,
1 mmol) in dry THF (2 mL) was stirred at room temperature overnight
under N2. Into a flask containing Ni(acac)2 (12.8 mg, 0.05 mmol) purged
with N2 were added successively THF (1 mL), the above-prepared solution
(via cannula), 1a (125 mg, 0.5 mmol) and Me2Zn (3.6 mL, 1 M hexane).
The homogeneous solution was stirred at room temperature for 1 h;
Rf (1a) = 0.7, Rf (2o) = 0.07 (hexane–EtOAc = 2 : 1 v/v). The mixture was
partitioned into EtOAc (20 mL)/H2O (20 mL). The water phase was
saturated with NaCl and washed with EtOAc (2 6 10 mL). The combined
organic phase was dried (K2CO3) and concentrated in vacuo. The residue
was purified by column chromatography over silica gel (hexane–EtOAc
gradient; 2 : 1 to 1 : 1 v/v) to give 4,4-di(methoxycarbonyl)-2-[4-(p-anisidyl)-
8-hydroxy-(1E)-octenyl]-1-isopropylidenecyclopentane (2o, 171.3 mg) in
73% yield as a colorless oil. 2o: IR (neat) 3395 (m), 2932 (w), 1736 (s), 1512
(s), 1443 (s), 1242 (s), 1042 (s), 818 (s), 733 (w) cm21; 1H NMR (400 MHz,
CDCl3) d 1.41–1.60 (m, 6 H), 1.57 (s, 3 H), 1.65 (s, 3 H), 2.06 (dd, J = 5.9,
¯
7 Crystal data for 2g: C25H31NO2, M = 377.53, triclinic, space group P1,
˚
a = 8.8531(8), b = 10.122(2), c = 12.351(2) A, a = 85.224(5), b = 85.589(3),
3
˚
c = 76.189(3)u, U = 1069.2(3) A , T = 297.2 K, Z = 2, m(Mo-Ka) =
0.7107 mm21, 4501 reflections measured, 2394 unique (Rint = 0.021),
wR(F2) = 0.1450 (all data). This compound is a racemate with two
¯
inversion-related molecules in the P1 triclinic unit cell, and the X-ray
structure of 2g determines the relative stereochemistry. CCDC 209610.
For crystallographic data in CIF or other electronic format see DOI:
10.1039/b605728d.
8 The Z structures of 2c and 2d were deduced on the basis of NOE
experiments. For example, irradiation of the methyl group of
a-phenylethylidene of 2c caused increments of the area intensities of
C4–H (3.2%) and C29–H (1.5%), while no increment for C2–H was
observed.
9 S. Fustero, J. G. Soler, A. Bartolome´ and M. S. Rosello´, Org. Lett., 2003,
5, 2707–2710.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2813–2815 | 2815