Indium(III) chloride-catalyzed coupling reaction of acid chlorides and allylic tins

Article abstract of DOI:10.1055/s-2001-17482

Allylic tributyltins and acid chlorides are effectively coupled to yield allylic ketones in indium(III) chloride-catalyzed reactions. In the case of γ-unsubstituted allylic tins, the assistance of a phosphine is essential to prevent further allylation of the produced ketones.

Full text of DOI:10.1055/s-2001-17482

LETTER
1659
Indium(III) Chloride-Catalyzed Coupling Reaction of Acid Chlorides and
Allylic Tins
Indium(III)
C
hlor
i
a
de-Catalyze
t
d
R
eac
s
tion uyuki Inoue, Yasuhisa Shimizu, Ikuya Shibata, Akio Baba*
Department of Molecular Chemistry, Graduate School of Engineering, Osaka University, Yamada-Oka 2-1, Suita, Osaka 565-0871, Japan
Fax +81(6)67897387; E-mail: baba@ap.chem.eng.osaka-u.ac.jp
Received 10 July 2001
O
Bu3Sn
R2
Abstract: Allylic tributyltins and acid chlorides are effectively
coupled to yield allylic ketones in indium(III) chloride-catalyzed re-
actions. In the case of -unsubstituted allylic tins, the assistance of
a phosphine is essential to prevent further allylation of the produced
ketones.
+
R1
1a R1=H, R2=Me
1b R1=H,
2=Ph
1c R1=H, R2=H
1d R1=Me,
R
Cl
2e R=p-MeOC6H4
2f R=n-C5H11
2g R=c-C6H11
2h R=n-Bu(Et)CH
2a R=Ph
2b R=p-NO2C6H4
2c R=p-ClC6H4
2d R=p-MeC6H4
R
R2=H
Key words: indium, allylic tin, acid chloride, phosphine, coupling
OH R1
reaction
O
R1
R
+
R
R2
R2
R2
R1
The synthetic application of indium compounds has at-
tracted much attention in the past decade.1 The present au-
thors have already reported the characteristic reactivities
of organoindium compounds generated by the transmeta-
lation of organotin compounds with indium trichloride
(InCl3).1k-m In particular, an indium hydride was shown to
reduce acid chlorides to the corresponding aldehydes
without side reactions, where the addition of a phosphine
is essential.2 In our efforts to expand the scope of the cat-
alytic use of indium compounds, we focused on the cou-
pling reaction of acid chlorides with allyltin compounds.
Among reported attempts, Migita and Kosugi achieved
selective coupling between benzoyl chloride and allyltri-
n-butyltin using a rhodium catalyst at 80 °C.3 Unfortu-
nately, the Pd-catalyzed procedure, called the Migita-
Stille reaction, is not ideal due to undesirable side reac-
tions such as isomerization and further allylation of the
produced ketones.4 Here we report the InCl3-catalyzed se-
lective coupling of allylic tributyltins and acid chlorides,
for which in certain cases the coordination of phosphines
is indispensable for preventing side reactions.5
3aa - 3da
3ca - 3ch
4ca, 4da
Scheme 1
The use of allyltri-n-butyltin (1c) resulted in the further al-
lylation of the produced allylketone, yielding the diallylat-
ed product 4ca (entry 3) exclusively in only 29% yield
based on the acid chloride. None of desired allylketone
3ca was obtained under any of the conditions examined.
However, the catalytic effect of InCl3 in the coupling is
characteristic, because other group-13 elements such as
AlCl3 and BCl3 exhibit no catalytic effect despite their
stronger Lewis acidity (entries 4 and 5). In a similar way
to 1c, methallyltri-n-butyltin (1d) predominantly gave the
diallylated product 4da in 44% yield along with a minor
amount of 3da (entry 6). The lack of further allylation of
3aa and 3ab is thought to depend on the sluggish route to
steric demanding tertiary alcohols 4 (R2 H) from 3
(R2 H).
A variety of coordinative compounds were tested in the
reaction of 1c and benzoyl chloride (2a) as a means of pre-
venting diallylation, which occurs in the coupling of -un-
substituted allylic tins. Amines such as triethylamine,
tetramethyl ethylenediamine and pyridine, however, had
only a negative effect, and no reaction took place. The ad-
dition of HMPA also resulted in merely the recovery of
the starting substrates. It is expected that these additives
coordinate to indium species too strongly to direct any
subtler reactivity. Fortunately, phosphines were found to
have a remarkable effect, as shown in Table 1. While 10
mol% of Ph3P was sufficient to achieve coupling (91%),
the use of 20 mol% increased the yield up to 99% (entries
7 and 8).7 The moderate coordination ability of phos-
phines is thought to promote selective coupling, giving
monoallylated ketones in 73%–99% yield (entries 8-11).
A large difference between tri-n-butylphosphine and tri-
cyclohexylphosphine was observed, affording the adduct
Initially we examined the coupling reaction between ben-
zoyl chloride (2a) and crotyltri-n-butyltin (1a) or cin-
namyltri-n-butyltin (1b). When 10 mol% of InCl3 was
added under the conditions detailed in Table 1, the desired
adducts 3aa and 3ba were obtained exclusively, at ambi-
ent temperature without side reactions (entries 1 and 2).
No reaction occurred in the absence of InCl3. The perfect
-selectivity, giving -substituted allyl ketones 3aa and
3ba, is in sharp contrast to the low regioselectivity in the
Rh(III)-catalyzed reaction.4,6
Synlett 2001, No. 10, 28 09 2001. Article Identifier:
1437-2096,E;2001,0,10,1659,1661,ftx,en;Y14701st.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

1660
K. Inoue et al.
LETTER
Table 2 Phosphine-Controlled Monoallylation of Acid Chlorides 2
by 1ca
Table 1 Coupling Reaction of 2a with Allylic Tins 1, and Effect of
Additivesa
Entry R (2)
Conditions
r.t., 2.5 h
Yield of 3 (%)
Entry Allylic Additive
Tin
Yield (%)b
1
Ph
(2a)
(2a)
(2b)
(2c)
(2d)
(2e)
(2f)
3ca
3ca
3cb
3cc
3cd
3ce
3cf
99
98
99
99
91
86
64
60
77
17
83
3
4
2b
3
Ph
r.t., 2.5 h
1
2
1a
1b
1c
1c
1c
1d
1c
1c
1c
1c
1c
1c
1d
1a
1b
none
none
none
none
none
none
Ph3P
Ph3P
n-Bu3P
dppb
dppe
3aa
3ba
3ca
3ca
3ca
3da
3ca
3ca
3ca
3ca
3ca
70
89
0
4aa
4ba
4ca
4ca
4ca
4da
4ca
4ca
4ca
4ca
4ca
4ca
4da
4aa
4ba
0
0
p-NO2C6H4
p-ClC6H4
p-MeC6H4
p-MeOC6H4
n-C5H11
r.t., 0.5 h
4
r.t., 0.5 h
3
58
0
(29)f
(22)f
5
r.t., 1.0 h
4c
5d
6
0
6
r.t., 7.0 h
0
0
7c
8
r.t., 17.5 h
r.t., 22.0 h
60 °C, 1.0 h
r.t., 22.0 h
60 °C, 2.5 h
8
44
0
c-C6H11
(2g)
(2g)
(2h)
(2h)
3cg
3cg
3ch
3ch
7e
8
91
99
88
81
73
0
9
c-C6H11
0
10
11
n-Bu(Et)CH
n-Bu(Et)CH
9
0
10e
11e
12
13
14
15
0
a InCl3 0.1 mmol, Ph3P 0.2 mmol, 1c 1 mmol, 2 1 mmol, MeCN 1 mL.
b Galvinoxyl (0.01 mmol) was added.
0
c Toluene was used instead of MeCN.
(c-C6H11)3P 3ca
0
Ph3P
Ph3P
Ph3P
3da
3aa
3ba
99
0
0
Table 2 lists the results for the selective monoallylation of
various acid chlorides with 1c. The possibility of free rad-
ical processes can be discounted by the observation that
the inclusion of a radical scavenger, galvinoxyl, did not
affect the coupling at all (entry 2). Electron-withdrawing
substituents on the aromatic ring were found to accelerate
the allylation, giving 3cb and 3cc in quantitative yields
within 30 min (entries 3 and 4). Of worth noting is the
high selectivity with the use of p-nitrobenzoyl chloride
(2b), because Pd-catalyzed allylation has been reported to
result in the low selective formation of 3cb.4a The intro-
duction of an electron-releasing substituent lengthened
the reaction time yet afforded 3ce in high yield (entry 6).
Aliphatic acid chlorides, 2f and 2g, gave only moderate
yields at ambient temperature even after long periods (en-
tries 7 and 8). The reaction of secondary aliphatic acid
chloride 2h resulted in only 17% yield ever after 22 h (en-
try 10), however heating to 60 °C increased the yield of
3ch to 83% without side products (entry 11).
0
0
0
a InCl3 0.1 mmol, additive 0.2 mmol, 1 1 mmol, 2a 1 mmol, MeCN 1
mL, r.t., 2.5 h.
b Yields based on tin compounds.
c AlCl3 was used instead of InCl3.
d BCl3 was used instead of InCl3.
e 0.1 mmol of additive was used.
f Yields based on 2a.
3ca in 88% and 0% yields, respectively (entries 9 and 12).
This result indicates that increasing the steric demand of
phosphines primarily disturbs the interaction between in-
dium species and acid chloride or allylic tin,8 because both
phosphines have similar basicities. The choice of solvent
is also essential for efficient reaction; acetonitrile (99%) is
superior to CH2Cl2 (90%), THF (67%) and toluene (37%).
Among them, THF suppresses the coordination of triphe-
nylphosphine, giving the diallylated product 4ca in 30%
yield along with the desired adduct 3ca. Methallyltri-n-
butyltin (1d) also effected the selective monoallylation
with the aid of triphenylphosphine, giving the correspond-
ing monomethallylated ketone 3da in 99% yield (entry
13). In contrast, a surprising result was that the coupling
reaction using crotyltri-n-butyltin (1a) or cinnamyltri-n-
butyltin (1b) was completely inhibited by the addition of
triphenylphosphine (entries 14 and 15). Apparently, the
phosphine suppressed the allylation in contrast to the re-
ported reduction of acid chlorides.2
Scheme 2 illustrates the proposed catalytic cycle, consist-
ing of three steps; (1) the generation of allyl indium by
transmetalation, (2) the nucleophilic addition to an acid
chloride, and (3) the formation of a , -unsaturated ke-
tone and regeneration of InCl3.9 This cycle is quite similar
to that previously proposed for the reduction of acid chlo-
rides to aldehydes, where allyltri-n-butyltin has been only
replaced by tri-n-butyltin hydride.2 However, there is a
sharp contrast in the role of phosphine. The coupling reac-
tion is somewhat depressed by the coordination of phos-
phine as observed in the treatment of crotyl-n-tributyltin,
whereas indium hydride affected the reduction only in the
presence of phosphine. Although the origin of this differ-
ence has yet to be clarified explicitly, it appears that the
coordination reduces the Lewis acidity of allylindium and
Synlett 2001, No. 10, 1659–1661 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Indium(III) Chloride-Catalyzed Reaction
1661
Araki, S.; Kamei, T.; Hirashita, T.; Yamamura, H.; Kawai,
M. Org. Lett. 2000, 2, 847. (h) Pinacol coupling: Lim, H. J.;
Keum, G.; Kang, S. B.; Chung, B. Y.; Kim, Y. Tetrahedrom
Lett. 1998, 39, 4367. (i) Transmetallation: Marshall, J. A.;
Hinkle, K. W. J. Org. Chem. 1995, 60, 1920. (j) Marshall,
J. A.; Hincke, K. W. J. Org. Chem. 1996, 61, 105. (k)
Alkynylindium: Yasuda, M.; Miyai, T.; Shibata, I.; Nomura,
R.; Matsuda, H.; Baba, A. Tetrahedron Lett. 1995, 36, 9497.
(l) Allylindium: Miyai, T.; Inoue, K.; Yasuda, M.; Baba, A.
Synlett 1997, 699. (m) Indium hydride: Miyai, T.; Inoue,
K.; Yasuda, M.; Baba, A. Tetrahedron Lett. 1998, 39, 1929.
(n) As a review of catalytic use: Chauhan, K. K.; Frost, C.
G. J. Chem. Soc., Perkin Trans. 1. 2000, 3015.
Bu3SnCl
Bu3Sn
1a
PPh3
Cl2In
PPh3
InCl3
5
O
O
5
PPh3
R
Cl
R
3
2
InCl2
OInCl2
R
O
R
Cl
(2) Inoue, K.; Yasuda, M.; Shibata, I.; Baba, A. Tetrahedron
Lett. 2000, 41, 113.
(3) Kosugi, M.; Shimizu, Y.; Migita, T. J. Organomet. Chem.
1977, 129, C36.
6
OH
(H )
R
(4) (a) Labadie, J. W.; Tueting, D.; Stille, J. K. J. Org. Chem.
1983, 48, 4634. (b) Yamamoto, Y.; Hatsuya, S.; Yamada, J.
J. Chem. Soc., Chem. Commun. 1988, 86. (c) Yamamoto,
Y.; Hatsuya, S.; Yamada, J. J. Org. Chem. 1990, 55, 3119.
(d) Kuhn, H.; Neumann, W. P. Synlett 1994, 123.
(5) An encouraging result has been recently reported in which
an equimolar amount of metallic indium promoted the
coupling reaction between acid chlorides and allylic
bromides: Yadav, J. S.; Srinivas, D.; Reddy, G. S.; Bindu, K.
H. Tetrahedron Lett. 1997, 38, 8745.
4
Scheme 2
indium hydride, inhibiting extended reactions. If further
allylation did take place, the catalytic cycle would be in-
terrupted because indium becomes trapped as alkoxide 6,
which exhibits virtually no regeneration of active species
as reported previously.1l
(6) Andrianome, M.; Haberle, K.; Delmond, B. Tetrahedron
1989, 45, 1079.
Therefore, the coordination of phosphine is essential for
the completion of the catalytic cycle. Although a mecha-
nistic study is now under investigation, our indium-cata-
lyzed coupling of allylic tins with acid chlorides has some
synthetic advantages; (1) no expensive transition metal
catalyst is necessary, and (2) side reactions such as
isomerization of the double bond and further allylation are
not problems.
(7) A typical experimental procedure: A 10 mL flask charged
with InCl3 (0.1 mmol) was heated at 150 °C in vacuo for 1 h.
After nitrogen gas was filled, MeCN (1 mL), triphenyl-
phosphine (0.2 mmol), allyltri-n-butyltin (1c) (1.0 mmol)
and benzoyl chloride (2a) (1.0 mmol) were successively
added at ambient temperature and the resulting mixture was
stirred for 2.5 h. After quenching with 10 mL of aqueous
NH4F, the mixture was extracted with Et2O (30 mL 2),
dried over anhyd MgSO4, filtered and concentrated in vacuo.
NMR analysis revealed the formation of 3ca in 99% yield.
Purification was performed by column chromatography
(silica gel) and then Kügelrohr distillation.
Acknowledgement
This work was financially supported by a Grant-in-Aid for Scienti-
fic Research from the Ministry of Education, Science and Culture,
and JSPS Research Fellowships for Young Scientists.
(8) (a) The corn angles of Ph3P, n-Bu3P and (c-C6H11)3P are
145°, 136° and 179°, respectively: Tolman, C. A. Chem.
Rev. 1977, 77, 313. (b) The basicities (pKa values of the
corresponding conjugate acids) are 2.73, 8.43 and 9.70
respectively: Golovin, M. N.; Rahman, M. M.; Belmonte,
J. E.; Giering, W. P. Organometallics 1985, 4, 1981.
(9) Allylindium species generated by the transmetallation
between tributylallyltin and indium trichloride in THF-d8
solvent were characterized by 1H NMR: 1.97 (d, 2 H,
J = 8.3 Hz), 4.6-4.95 (m, 2 H), 5.9-6.1 (m, 1 H). However,
no spectral proof of coordination of Ph3P was available
because of the poor solubility of the indium-phosphine
complex.
References and Notes
(1) (a) Barbier-type allylation: Araki, S.; Ito, H.; Butsugan, Y.
J. Org. Chem. 1988, 53, 1831. (b) Li, C. J.; Chan, T. H.
Tetrahedron Lett. 1991, 32, 7017. (c) Li, C. J.; Chan, T. H.
Tetrahedron 1999, 55, 11149. (d) Li, C. J. Tetrahedron Lett.
1995, 36, 517. (e) Isaac, M. B.; Chan, T. H. Tetrahedron
Lett. 1995, 36, 8957. (f) Paquette, L. A.; Mitzel, T. M. J.
Org. Chem. 1996, 61, 8799. (g) Pd-catalyzed allylation:
Synlett 2001, No. 10, 1659–1661 ISSN 0936-5214 © Thieme Stuttgart · New York