Cationic Palladium Complex-Catalyzed Cyclization-Hydrosilylation of 1,6-Heptadiyne and its Homologs

Article abstract of DOI:10.1055/s-2003-41494

Palladium complex (A)-catalyzed dimerization-hydrosilylation of 1-alkynes, which has previously been reported, can not be applied effectively to the intramolecular version of α,ω-alkanediynes, while a cationic palladium complex (B) is a very effective cat

Full text of DOI:10.1055/s-2003-41494

LETTER
1997
Cationic Palladium Complex-Catalyzed Cyclization–Hydrosilylation of
1,6-Heptadiyne and its Homologs
Cationic
P
alladi
a
um Compl
k
ex-Catalyzed
a
Cyclizatio
n
n–hydrosilyl
o
ation ri Uno, Shigeru Wakayanagi, Yoshiya Sonoda, Keiji Yamamoto*
Department of Materials Science and Environmental Engineering, Tokyo University of Science, Yamaguchi, 1-1-1 Daigaku-dori,
Onoda, Yamaguchi 756-0884, Japan
Fax +81(836)883844; E-mail: yamamoto@ed.yama.tus.ac.jp
Received 17 July 2003
H
E
E
Abstract: Palladium complex (A)-catalyzed dimerization-hydro-
silylation of 1-alkynes, which has previously been reported, can
not be applied effectively to the intramolecular version of a,w-
alkanediynes, while a cationic palladium complex (B) is a very ef-
fective catalyst for cyclization–hydrosilylation of functionalized
a,w-diynes to give (Z)-1-methylene-2-silylmethylenecycloalkane
derivatives, with 5-, 6-, and 7-membered ring, in good to moderate
yields.
1/2[(η3-C3H5)PdCl]2
SiCl3
(2 mol%)
E
E
(36%)
+
+ HSiCl3
R'
2 P(OEt)3
CH2Cl2, r.t.
(E = CO2Me)
R'
(14%)
H
SiCl3
(R' = CH2CE2CH2C ßCH)
Key words: cyclization, hydrosilylation, cationic palladium com-
plex, a,w-diyne, 5-, 6-, and 7-membered carbocycle
Scheme 2
nickel(0)-catalyzed hydrosilylation-cyclization of 1,7-
octadiynes in this context, being unable to apply to the
reaction of 1,6-heptadiynes.3
We have reported a clean dimerization-hydrosilylation of
a variety of 1-alkynes, which is catalyzed by 1/2[(h3-
C3H5)PdCl]2 (A) with two equivalents of bulky phosphite,
typically tris(2,4-di-t-butylphenyl) phosphite, in dichloro-
methane solution (Scheme 1).1
Widenhoefer and coworkers have pointed out, in their
consecutive articles,4 that the cyclization/hydrosilylation
of a,w-dienes with trialkylsilanes catalyzed by cationic
palladium complexes is a useful method for the synthesis
of silylated cycloalkanes, whereas the cyclization/hydro-
silylation of a,w-diynes, particularly 1,6-heptadiynes, to
form 1,2-dialkylidenecycloalkanes remains problematic.
Recently, they solved these difficulties by using cationic
platinum complexes, but not palladium ones, as catalysts
that contain bidentate nitrogen ligands.5 On the other
hand, the rhodium-catalyzed silylative cyclization of 1,6-
heptadiyne derivatives with trialkylsilanes proceeds
smoothly to give 1,2-dialkylidenecyclopentane frame-
work, albeit concomitant formation of aromatized
byproducts seems to be inevitable.6
1/2[(η3-C3H5)PdCl]2
R
R
R
(0.5 mol%)
+ HSiCl3
+
R
R
2 P(OAr)3
H
SiCl3
H
SiCl3
CH2Cl2, 50 °C, 15 h
head-to-tail head-to-head
:
(85–97)
(15–3)
Scheme 1
Head-to-tail dimer became by far the major product, when
the bulkiest phosphite examined was used as a supporting
ligand. It was then found that the above catalyst system
was not satisfactorily effective to the reaction of a,w-
alkanediynes, which may well undergo cyclization–hy-
drosilylation as the dimerization to take place intramolec-
ularly.2 In fact, the attempted reaction of dimethyl
dipropargylmalonate (1) under similar conditions as
above resulted in giving both cyclized and dimerized hy-
drosilylation products (Scheme 2).
Now we are pleased to find that a cationic palladium
complex, [(h3-C3H5)Pd(cod)]+[PF6]– (B) (cod = 1,5-
cyclooctadiene), catalyzed the effective hydrosilylation of
1 with trichlorosilane at room temperature to afford a
cyclization–hydrosilylation product, (Z)-1-(trichlorosil-
yl)methylene-2-methylene-4,4-bis(methoxycarbonyl)cy-
clopentane (2a), which was isolated as a triethoxy
derivative (2b), the latter being contaminated with a small
amount of (E)-isomer (Scheme 3).7,8
It is apparent that the expected cyclization–hydrosilyla-
tion of a,w-alkanediynes would result from head-to-head
dimerization, whereas the previous intermolecular dimer-
ization was found to prefer the head-to-tail product.
Tamao and his coworkers have already reported the
No phosphorus ligand was required for the catalyst B,
while [BF4]– as a counter anion seemed to exhibit little ef-
fect. It was also found that HSiMenCl3–n (n = 0, 1, and 2)
can equally be employed for the cyclization–hydrosilyla-
tion to give the corresponding products in comparable
yields. In order to obtain certain mechanistic rationale, the
palladium complex B-catalyzed hydrosilylation of un-
symmetrical dimethyl (2-butynyl)-propargylmalonate (3)
SYNLETT 2003, No. 13, pp 1997–2000
1
6
.1
0
.2
0
0
3
Advanced online publication: 08.10.2003
DOI: 10.1055/s-2003-41494; Art ID: U13903ST
© Georg Thieme Verlag Stuttgart · New York

1998
T. Uno et al.
LETTER
[(η3-C3H5)Pd(cod)]+
Table 1 Carionic Palladium Complex-Catalyzed Cyclisation-
Hydrosilylation of Functionalized 1,6-Diynes
(0.5 mol%)
E
E
H
E
E
+ HSiCl3
R
SiX3
1,6-Diynes
Hydrosi- Product
lane
Yield (%)
CH2Cl2, r.t.
(Z/E)b
R
(E = CO2Me)
1: R = H
2a: X = Cl, 2b : X = OEt
4a: X = Cl, 4b : X = OEt
E
E
H
E
3: R = Me
HSiCl3
56 (92/8)
Si(OEt)3
E
1
1
2b
Scheme 3
H
E
HMeSiCl2
54 (96/4)
54 (93/7)
was carried out. Thus, cyclization–hydrosilylation of 3
took place cleanly with HSiCl3 to afford, as a single
product in moderate yield, (Z)-1-(1¢-trichlorosilyl)eth-
ylidene-2-methylene-4,4-bis-(methoxy-carbonyl)cyclo-
pentane (4a), whose structure was readily identified as its
triethoxy derivative (4b) by 1H NMR spectrum.9
SiMe(OEt)2
E
2c
2d
H
E
1
HMe2SiCl
SiMe2(OEt)
E
Also, dipropargyl ether (5), unsymmetrical 2-butynyl pro-
pargyl ether (5¢), and N-tosyldipropargylamine (7) under-
went the cationic palladium complex (B)-catalyzed
cyclization–hydrosilylation with either HSiCl3 or
HMeSiCl2 in the same manner as for 1 to give, after ethox-
ylation, the corresponding derivatives of 1,2-dimethyle-
necyclopentanes 6b, 6¢b, and 8c, respectively, with
predominant Z-geometry. All results are summarized in
Table 1.
H
E
E
E
HSiCl3
53 (93/7)
33 (93/7)
40 (–)c
Si(OEt)3
E
Me
4b
3
Me
H
O
O
HSiCl3
Si(OEt)3
5
6b
H
O
O
HSiCl3
Si(OEt)3
Especially, the structure of 6¢b was fully characterized by
the NOE measurements,10 reinforcing Z-geometry with
respect to the silyl substituent in overall cyclization–hy-
drosilylation pattern. The fact that the silyl substituent is
exclusively introduced into the internal alkyne site in sub-
strates 3 and 5¢ indicates that the hydropalladation takes
place first at the terminal alkyne site, giving rise to form
4b and 6¢b, respectively.
Me
5'
6'b
Me
H
SiMe(OEt)2
HMeSiCl2
53 (–)c
TsN
TsN
8c
7
a Isolated yield by bulb-to-bulb distillation or by column chromatog-
raphy.
The protocol of the cationic palladium complex B-cata-
lyzed reaction was found to apply to 1,7-octadiyne (9) and
even to 1,8-nonadiyne (11), though the latter appeared to
undergo sluggish cyclization. Thus, the cyclization–hy-
drosilylation of 1,7-octadiyne (9) with HSiCl3 proceeded
smoothly at 50 °C for 22 hours to afford (Z)-1-(trichloro-
silyl)methylene-2-methylenecyclohexane (10a), which
was isolated as its triethoxy derivative (10b) in 68% iso-
lated yield and with Z/E = 84/16 by glc analysis.11 Also,
the reaction of 1,8-nonadiyne (11) was carried out under
the same conditions as above, resulting in giving cyclo-
heptane analogs (12a) and (12b), respectively, but in an
inferior yield (<20%).12 Others were found to be recov-
ered 11 and largely intractable materials (Scheme 4). Im-
provement of the reaction conditions for 11 is underway.
b Peak area ratio by glc analysis of immediate reaction product(s).
c Exclusive one isomer as shown.
1) Pd+
HSiCl3
H
H
or
n
SiX3
SiX3
2) EtOH
Et3N
9 (n = 2)
10a : X = Cl
12a : X = Cl
11 (n = 3)
10b : X = OEt
12b : X = OEt
(68% isolated yield)
(~20% isolated yield)
Scheme 4
choice,1 and (3) in contrast to the Ni(0),3 cationic Pt(II),5
and either neutral or cationic Rh(I)-catalyzed cyclization–
hydrosilylation,6 where the silylmetallation of triple bond
was suggested as an initial step of the catalytic cycle, we
propose the hydrometallation as the initial step, on the ba-
sis of regioselectivity observed in the products 4b and 6¢b,
respectively: The silyl group was incorporated exclusive-
ly into the internal alkyne site.
The salient features of the present cyclization–hydrosily-
lation of a,w-diynes are three fold; (1) the ordinary cation-
ic palladium complex B without phosphorus supporting
ligands played a key role in the cyclization to form 5-, 6-
and even 7-membered rings with concomitant hydrosily-
lation, (2) HMenSiCl3–n (n = 0–2) could equally be em-
ployed for the reaction as far as the substrate 1 was
concerned, while this was not the case for the intermolec-
ular dimerization of 1-alkynes, HSiCl3 being of only
Synlett 2003, No. 13, 1997–2000 © Thieme Stuttgart · New York

LETTER
Cationic Palladium Complex-Catalyzed Cyclization–Hydrosilylation
Synlett 1992, 539. (c) For a related Ni(0)-catalyzed
1999
Cycloisomerization of linear enynes using Pd(0) and an
acid, [cationic palladium hydride],13 to give cyclic dienes
has been extensively exploited most notably by Trost.14
Similar to a possible catalytic cycle for this transforma-
tion,15 the cyclization–hydrosilylation of a,w-diynes
would proceed as follows (Scheme 5): Following (i) com-
plexation, (ii) hydropalladation, and (iii) intramolecular
carbopalladation, the cyclized (Z)-alkenylpalladium inter-
mediate is formed and finally undergoes (iv) a possible
s-metathesis with a hydrosilane16 to give the silylated
product, regenerating the cationic catalyst. Further exam-
ination of the mechanistic rationale must be necessary and
will be presented in due course.
dimerization of 1-alkynes, see: Lappert, M. F.; Nile, T. A.;
Takahashi, S. J. Organomet. Chem. 1974, 72, 425.
(4) (a) Widenhoefer, R. A.; DeCarli, M. A. J. Am. Chem Soc.
1998, 120, 3805. (b) Stengone, C. N.; Wiedenhoefer, R. A.
Tetrahedron Lett. 1999, 40, 1451. (c) Perch, N. S.; Pei, T.;
Widenhoefer, R. A. J. Am. Chem. Soc. 1999, 121, 6960.
(d) Widenhoeder, R. A.; Stengone, C. N. J. Org. Chem.
1999, 64, 8681. (e) Widenhoefer, R. A.; Vadehra, A.
Tetrahedron Lett. 1999, 40, 8499. (f) Pei, T.; Widenhoefer,
R. A. Org. Lett. 2000, 2, 1469. (g) Perch, N. S.; Pei, T.;
Widenhoefer, R. A. J. Org. Chem. 2000, 65, 3836. (h) Pei,
T.; Widenhoefer, R. A. Tetrahedron Lett. 2000, 41, 7597.
(5) (a) Madine, J. W.; Wang, X.; Widenhoefer, R. A. Org. Lett.
2001, 3, 385. (b) Wang, X.; Chakrapani, H.; Madine, J. W.;
Widenhoefer, R. A. J. Org. Chem. 2002, 67, 2778.
(6) (a) Maruoka, T.; Matsuda, I.; Itoh, K. Organometallics 2002,
21, 3650. (b) Muraoka, T.; Matsuda, I.; Itoh, K. Tetrahedron
Lett. 1998, 39, 7325. (c) Ojima, I.; Vu, A. T.; McCullagh, J.
V.; Kinoshita, A. J. Am. Chem. Soc. 1999, 121, 3230.
(d) Liu, C.; Widenhoefer, R. A. Organometallics 2002, 21,
5666. (e) For a similar rhodium-catalyzed silylcarbo-
cyclization, but restricted within 1,6-enynes, see: Ojima, I.;
Vu, A. T.; Lee, S.-Y.; McCullagh, J. V.; Moralee, A. C.;
Fujiwara, M.; Hoang, T. H. J. Am. Chem. Soc. 2002, 124,
9164.
[(η3-C3H5)Pd(cod)]+
A
SiCl3
A
R
R
+
HPd
HSiCl3
(iv)
Pd
(i)
+
(7) Typical procedure for the cyclization–hydrosilylation of 1
with HSiCl3 is as follows: In a 10 mL screw-capped glass
tube were placed, under an argon atmosphere, 1 (0.208 g, 1.0
mmol), the catalyst precursor B (2.0 mg, 5 × 10–3 mmol, 0.5
mol%), and HSiCl3 (2 M CH2Cl2 solution 0.55 mL, 1.1
mmol) diluted with dry CH2Cl2 (0.45 mL). The orange-
yellow clear solution was magnetically stirred for 24 h at r.t.
GLC (10% Silicone SE-30 on uniport B, 3 mm × 3 m
column, programmed 100~280 °C) analysis of the reaction
mixture revealed recovered 1 (ca.4%) and the product peaks
(96%) in a ratio 8:92 (TR = 16.8 and 17.7 min). The whole
mixture was treated with dry EtOH (0.20 mL, 3.3 mmol) and
Et3N (0.46 mL, 3.3 mmol) dissolved in CH2Cl2 (6 mL) for 2
h in an ice-water bath. The turbid solution formed was
filtered through a celite plug, the latter being rinsed with dry
hexane, and the combined filtrate was thoroughly
+
PdH
R
A
A
R
H
+
(iii)
(ii)
A
Pd
R
Scheme 5a (i) Complexation, (ii) hydropalladation, (iii) carbopallada-
tion, (iv) s-metathesis with HSiCl3. a Other ligands on palladium are
omitted for clarity.
In conclusion, we have found that a cationic palladium
complex (B) is very effective catalyst for domino cycliza-
tion–hydrosilylation of functionalized a,w-diynes to form
5-, 6- and even 7-membered (Z)-1-methylene-2-silylme-
thylenecycloalkane derivatives in good to moderate
yields. Some of the products are subjected to the Diels–
Alder reaction with maleic anhydride or N-methylmale-
imide.17
concentrated by rotary evaporation. The crude products
(0.332 g, ca. 89%), in a ratio 8:92 (TR = 18.2 and 19.3 min),
was subjected to a bulb-to-bulb distillation (150–170 °C/3
Torr) to give pure 2b as a colorless liquid (0.207 g, 56%
yield), the probable (E)-isomer being hard to be detected
within an accuracy of NMR spectra.
(8) Spectral data for 2b: 1H NMR (270 MHz, CDCl3):
d (ppm) = 1.21 (t, J = 6.9 Hz, 1 H), 3.08 (br t, J = 2.2 Hz,
2 H), 3.13 (br d, J = 1.7 Hz, 2 H), 3.72 (s, 6 H), 3.80 (q,
J = 6.9 Hz, 6 H), 5.17 (br t, J = 2.2 Hz, 1 H), 5.32 (br t,
J = 2.0 Hz, 1 H), 5.95 (t, J = 2.3 Hz, 1 H). 13C NMR (67.8
MHz): d (ppm) = 18.1 (× 3), 42.1, 45.1, 52.8, 56.9, 58.4
(× 3), 112.0, 112.5, 143.7, 158.3, 171.6.
Acknowledgment
We thank Dr. Yukio Kawanami for the discussion as well as his
experimental skill. Mr. Jun-ichi Uchiyama is thanked for his expe-
rimental assistance at the initial stage of this work.
(9) Spectral data for 4b: 1H NMR: d = 1.20 (t, J = 6.9 Hz, 9 H),
1.85 (s, 3 H), 3.00 (t, J = 1.7 Hz, 2 H), 3.07 (d, J = 1.3 Hz, 2
H), 3.75 (s, 6 H), 3.77 (q, J = 6.9 Hz, 6 H), 5.04 (s, 1 H), 5.73
(s, 1 H). 13C NMR: d = 18.1 (× 3), 20.0, 40.4, 42.7, 52.8,
56.4, 58.3 (× 3), 111.1, 122.0, 144.3, 151.8, and 171.9.
(10) Spectral data for (Z)-6¢b: 1H NMR: d = 1.23 (t, J = 6.9 Hz, 9
H), 1.82 (br s, 3 H), 3.82 (q, J = 6.9 Hz, 6 H), 4.47 (t, J = 2.0
Hz, 2 H), 4.55 (d, J = 1.3 Hz, 2 H), 5.09 (d, J = 0.66 Hz, 1
H), 5.96 (t, J = 2.0 Hz, 1 H). 13C NMR: d = 18.1 (× 3), 19.7,
58.5 (× 3), 73.4, 74.5, 108.0, 120.9, 143.1, 150.9. The
stereochemistry was determined by NOE experiments
(NOE, Figure 1).
References
(1) Kawanami, Y.; Yamamoto, K. Synlett 1995, 1232.
(2) Kawanami, Y.; Yamamoto, K. 43rd Symposium on
Organometallic Chemistry, Abstr. PB114; Osaka: Japan,
1996.
(3) (a) Tamao, K.; Kobayashi, K.; Ito, Y. J. Am. Chem. Soc.
1989, 111, 6478. (b) Tamao, K.; Kobayashi, K.; Ito, Y.
Synlett 2003, No. 13, 1997–2000 © Thieme Stuttgart · New York

2000
T. Uno et al.
LETTER
(12) Spectral data for 12b: 1H NMR: d = 1.21 (t, J = 6.9 Hz, 9 H),
1.59 (br s, 6 H), 2.34 (m, 2 H), 2.42 (m, 2 H), 3.78 (q, J = 6.9
Hz, 6 H), 4.87 (dt, J = 2.0, 1.3 Hz, 1 H), 5.07 (t, J = 1.3 Hz,
1 H), 5.18 (d, J = 2.0 Hz, 1 H). 13C NMR: d = 18.21 (× 3),
28.7, 29.4, 30.9, 35.5, 40.4, 49.7, 58.3 (× 3), 113.4, 114.6,
151.9, 167.7.
(13) Amatore, C.; Jutand, A.; Meyer, G.; Carelli, I.; Chiaretto, I.
Eur. J. Inorg. Chem. 2000, 1855.
(14) (a) Trost, B. M.; Romero, D. L.; Rise, F. J. Am. Chem. Soc.
1994, 116, 4268. (b) Trost, B. M. Chem.– Eur. J. 1998, 4,
2405.
(15) Aggarwal, V. K.; Butters, M.; Davies, P. W. Chem.
Commun. 2003, 1046.
(16) (a) Maruyama, Y.; Yamamura, K.; Sagawa, T.; Katayama,
H.; Ozawa, F. Organometallics 2002, 19, 1308. (b) For a
pertinent review, see: Ozawa, F. J. Organomet. Chem. 2000,
611, 332.
nOe
H
H
H
H
H
H
O
O
Si(OEt)3
Si(OEt)3
H
H
Me
Me
4.2%
3.4%
Figure 1
It was found that pure (Z)-6¢b, dissolved in degassed CDCl3
in a sealed NMR tube, isomerized significantly to (E)-6¢b in
months: Z/E = 30/70. (E)-6¢b: 1H NMR: d = 1.24 (t, J = 6.9
Hz, 9 H), 2.05 (t, J = 2.0 Hz, 3 H), 3.83 (q, J = 6.9 Hz, 6 H),
4.43 (t, J = 2.0 Hz, 2 H), 4.60 (q, J = 2.0 Hz, 2 H), 5.27 (t,
J = 2.0 Hz, 1 H), 5.45 (br s, 1 H).
(17) Spectral data for an adduct of 2c with N-methylmaleimide:
1H NMR: d = 0.13 (s, Me), 1.21 and 1.22 (t, J = 6.9 Hz,
diastereotopic Me), 2.37 (br s, 1 H), 2.46 (br s, 2 H), 2.84–
3.24 (m, 6 H), 2.95 (s, 3 H), 3.69 and 3.71 (s, 2 × Me), 3.78
(11) Spectral data for (Z)-10b (estimated 84%): 1H NMR: d =
1.22 (t, J = 6.9 Hz, 9 H), 1.67 (m, 4 H), 2.27 (m, 2 H), 2.33
(m, 2 H), 3.79 (q, 6.9 Hz, 6 H), 4.80 (d, J = 1.3 Hz, 1 H), 5.00
(d, J = 1.5 Hz, 1 H), 5.07 (d, J = 1.7 Hz, 1 H). 13C NMR: d =
18.2 (× 3), 27.5, 27.7, 36.6, 41.1, 58.3 (× 3), 110.5 112.1,
149.4, 164 8. (E)-10b (estimated 16%): 1H NMR: d = 1.19
(t, J = 6.9 Hz, 9 H), 3.78 (q, J = 6.9 Hz, 6 H), 4.78 (br s, 1
H), 5.02 (br s, 1 H), 5.11 (d, J = 2.3 Hz, 1 H). Other signals
are overlapped with those of Z-isomer. 13C NMR (diagnostic
signals): d = 110.7, 113.4, 149.3, 163.8.
and 3.79 (q, J = 6.9 Hz, diastereotopic methylene). 13
C
NMR: d = –5.80, 18.35, 24.31, 25.28, 26.00, 39.28, 39.55,
43.56, 44.01, 52.71, 52.76, 57.68, 58.58 and 58.65
(diastereotopic methylene), 128.25, 131.68, 172.20, 172.51,
180.59, 180.79.
Synlett 2003, No. 13, 1997–2000 © Thieme Stuttgart · New York