Noticeable facilitation of the bismuth-mediated Barbier-type allylation of aromatic carbonyl compounds under solvent-free conditions

Article abstract of DOI:10.1039/b303783e

When milled together with bismuth shot in the presence of allyl halide, aromatic aldehydes readily underwent a Barbier-type allylation to afford the corresponding homoallyl alcohols in good yield. In contrast to the failure in solution reaction, aromatic

Full text of DOI:10.1039/b303783e

View Article Online / Journal Homepage / Table of Contents for this issue
Noticeable facilitation of the bismuth-mediated Barbier-type
allylation of aromatic carbonyl compounds under solvent-free
conditions
Shinobu Wada, Nobuyuki Hayashi and Hitomi Suzuki*
Department of Chemistry, School of Science and Technology, Kwansei Gakuin University,
Gakuen 2-1, Sanda 669-1337, Japan. E-mail: hsuzuki@ksc.kwansei.ac.jp;
Fax: ϩ81-79-565-9077
Received 9th April 2003, Accepted 2nd May 2003
First published as an Advance Article on the web 16th May 2003
When milled together with bismuth shot in the presence of allyl halide, aromatic aldehydes readily underwent a
Barbier-type allylation to afford the corresponding homoallyl alcohols in good yield. In contrast to the failure in
solution reaction, aromatic ketones also underwent allylic carbonyl addition under solvent-free conditions to give the
expected tertiary homoallyl alcohols in moderate to good yield.
been prepared in a dry powdery state from halonaphthalenes
Introduction
and magnesium using the ball milling technique and allowed to
The conversion of carbonyl compounds to homoallylic alco-
react with aromatic ketones.24 Aryl–aryl coupling is the major
hols is an important process in organic synthesis.1 This type of
competitive process observed therein.
functional group transformation has long been carried out
The literature to date contains few examples of mechano-
using organometallic compounds such as Grignard reagents,2
chemical organic synthesis. The ball milling of aromatic
organolithiums,3 organosilanes,4 and organostannanes.5 In
ketones and magnesium metal led to the McMurry and pinacol
recent years, however, an alternative approach involving a
coupling products along with the expected alcohol and its
reductive allylation process has become of increasing import-
parent hydrocarbon.24 A redox-type ligand exchange between
ance, where a combination of allyl halide and a metal powder
or a low valent metal halide is used for the in situ generation of
thallium() cyclopentadienide and rare earth metals under ball
milling conditions has been employed for the preparation of
allylmetal species. This type of allylic carbonyl addition has
rare earth cyclopentadienyls such as YbCp2 and EuCp3.25 Aro-
been called the Barbier-type allylation,6 and the metals
matic hydrocarbons are found to undergo disproportionation
employed therein include magnesium,7 tin,8 indium,9 zinc,10
by ball-milling in the presence of silica gel and alumina.26
cadmiun,11 amalgamated cerium,12 manganese,13 antimony,14
Chlorinated hydrocarbons can be destroyed by milling with
and bismuth.15 The low valent metal halides used include
magnesium, calcium and calcium oxide.27
chromium() chloride,16 samarium() iodide,17 and ytter-
bium() iodide.17 Bimetallic reducing systems consisting of a
Results and discussion
metal and a metal halide have been successfully employed for
the catalytic reductive allylation of carbonyl compounds. They
include Fe–SbCl3,18 Al–SbCl3,18 Mg–BiCl3,19 Al–BiCl3,20 Zn–
BiCl3,20 Zn–InCl3,21 and Al–InCl3 21 systems. This methodology
has attracted considerable attention, since the reaction can be
carried out in aqueous media.22 The reductive allylation based
on an electrochemically generated allylmetal species has also
been reported.23
To our knowledge, almost all known Barbier-type allylations
have been carried out in the presence of a solvent, stirring a
heterogeneous mixture for a specified period of time ranging
from several hours to a few days at room temperature or under
gentle heating. In the majority of reported cases, the procedure
has proved to be ineffective for the allylation of ketones. The
dimerization of carbonyl compounds to pinacols, Wurtz-type
coupling of allyl halides, isomerization of a branched allylic
framework, and Tishchenko reaction leading to esters are the
major side reactions which often accompany the reductive
allylation of carbonyl compounds in solution.
The bismuth-mediated allylation of aldehydes with allyl
bromide was first reported by Wada in 1985.15 As part of our
ongoing program on bismuth as a reagent in organic synthesis,
we have reinvestigated this pioneering work under ball milling
conditions, i.e. solvent-free environments. Bismuth is attractive
as a reagent for the reductive allylation of aromatic carbonyl
compounds, since it is non-toxic and low-priced.28 We found
that the Barbier-type allylation of aromatic aldehydes took
place smoothly under such completely dry conditions, giving
the expected homoallylic alcohols in moderate to good yield
(Scheme 1).
In this paper, we wish to report a remarkable facilitation of
the bismuth-mediated Barbier-type allylation observed under
ball milling conditions. Under such completely dry environ-
ments, the reaction time is considerably shortened, and both
aldehydes and ketones can be allylated with modest ease to
afford the expected homoallylic alcohols in moderate to good
yield. It has generally been recognized that the metal-promoted
allylation of carbonyl compounds is highly influenced by the
nature of the solvent employed. Under the present solvent-free
conditions, however, there is no need to be concerned about the
choice of solvent. Recently, aromatic Grignard reagents have
Scheme 1
Thus, benzaldehyde (1; R = H), allyl bromide 2b, bismuth
shot, and small stainless steel balls were placed in a stainless
steel vial with a screw cap. The reaction vessel was shaken using
a laboratory ball mill apparatus at a rate of 30 Hz for 0.5–1.5 h.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 6 0 – 2 1 6 3
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
2160

View Article Online
Table 1 Bi-mediated allylation of aromatic aldehydes 1 under ball mill conditions
Entry
R
X
Aldehyde 1/halide 2/bismuth ratioa
Milling time (h)
Yieldb, c (%)
1
2
3
4
5
6
7
8
9
H
H
H
Me
CN
CN
NO2
NO2
NO2
Br
Br
Br
Br
Br
Br
Br
Br
Cl
1/2/8
1/2/8
0.5
1.5
1.5
0.5
0.5
1.0
0.5
1.0
1.5
81
88
94
69
80
91
84
95
34
1/1.5/8
1/1.5/8
1/1.5/8
1/1.5/8
1/1.5/8
1/1.5/8
1/1.5/8
a Molar ratio. b Yield refers to the isolated compound. All products are known. To promote efficiency of milling bismuth shot was used in a large
excess. No effort was made to optimize milling conditions against product yield. c Extraction of the product was made with 0.1 M HCl and EtOAc.
The resulting brown to dark grey pasty mass was extracted with
EtOAc by trituration and the extract was evaporated to leave
the expected homoallylic alcohol (3; R = H). The results are
summarized in Table 1. Less reactive allyl chloride 2a was also
usable, but the reaction required a longer milling time to com-
plete (Table 1, entry 9). The chloride usually fails to react in
solution.15 Prolonged ball milling led to a gradual decline of
product yield, probably due to the decomposition of the initial
product to a diene and its descendents. Loss of allyl halides by
the Wurtz-type coupling was not observed.
The bismuth-mediated reaction of benzaldehyde and allyl
bromide can be carried out in protic solvents such as alcohol
and water.15 Thus, anticipating that an extent of asymmetric
induction might arise during the allylic carbonyl addition, the
bismuth-mediated Barbier-type reaction of 4-trifluoromethyl-
benzaldehyde (1; R = CF3) and allyl bromide 2b was performed
in the presence of a chiral powdery medium such as cane sugar
or a potato starch. However, the enantiomer ratio of the result-
ing homoallyl alcohol (3; R = CF3), determined by gas chrom-
atography using a chiral column, was close to the unity, and
therefore, we could not draw any definite conclusion as to pos-
sible asymmetric induction under the present ball milling condi-
tions. The enantiomers appeared as a pair of peaks of different
height on the gas chromatogram, but the ratio of integrated
peak areas was near to unity.
Generally speaking, aromatic ketones are quite sluggish or
fail to undergo the Barbier-type allylation in solution.15 How-
ever, they were able to undergo the reaction with moderate ease
under the solvent-free conditions (Scheme 2). In this case,
attempted extraction of the resulting pasty solid with EtOAc
alone led to a low yield of the expected product. However, when
the solid was first triturated with 0.1 M HCl and then extracted
with EtOAc, the product yield was dramatically improved.
Seemingly, the tertiary homoallyl alcohols 5 formed as the bis-
muth alkoxide in the pasty solid. Pinacols that often accom-
pany the Barbier-type reaction of ketones in solution were not
detected. It is of note that sensitive functional moieties such as
nitro and cyano groups remained intact throughout the reac-
tion (Table 2). Both commercial bismuth shot and powder
(Ishizu, ∼100 mesh) were used as received and provided almost
identical results. The physical form of the metallic bismuth does
not seem to have much influence on the product yield.
For comparison, the present methodology was extended to
the indium- and tin-mediated allylation of aromatic aldehydes.
Both metals proved to work similarly, though they somewhat
appeared to favor the formation of byproducts. The advan-
tageous use of bismuth over indium and tin in the chemo-
selective Barbier-type allylation was well manifested in the
allylation of 4-nitrobenzaldehyde (1; R = NO2) and 4-nitro-
acetophenone (4; R = NO2). Bismuth afforded the expected
alcohols (3 and 5; R = NO2) in good to moderate yields (Table
1, entries 7 and 8 and Table 2, entries 4–6), but indium and tin
gave a complex mixture containing the azo and azoxy com-
pounds. This may be attributed to the higher reducing ability of
tin (Sn to Sn2ϩ, Ϫ0.137 V) and indium (In to Inϩ3, Ϫ0.338 V) as
compared with bismuth (Bi to Bi3ϩ, ϩ0.317 V).29
With an intention to see the effect of steric factors on the sol-
vent-free carbonyl allylation, two types of ketones with differ-
ent steric requirements around the carbonyl function, 6 and 8,
have been reacted with allyl bromide 2b under the same condi-
tions. Cyclohexanone 6 reacted with moderate ease, but steric-
ally more crowded benzophenone 8 reacted with less ease
(Scheme 3). The reactivity of these ketones in the known
Barbier-type allylation considerably differs in solution reaction.
The former ketone can undergo reductive allylation under cer-
tain circumstances, whereas the latter often fails to react or
suffers reductive coupling to give a pinacol. However, the dif-
ference in reactivity between them is not so significant under the
solvent-free environments. Thus, the steric factor seems to be of
less influence in the solvent-free reaction than the solution reac-
tion. The ease with which carbonyl compounds undergo reduc-
tive allylation can be ranked in the order, aromatic aldehydes >
aryl methyl ketones > cycloalkanones > diaryl ketones. Aro-
matic carbonyl compounds bearing an electron-withdrawing
group afforded a higher yield relative to other ones, while those
bearing an electron-donating group led to a lower yield.
Scheme 2
Scheme 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 6 0 – 2 1 6 3
2161

View Article Online
Table 2 Bi-mediated allylation of aromatic ketones 4 under ball mill conditions
Entry
R
Bi form
Ketone 4/halide 2b/bismuth ratioa
Yieldb (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
H
H
H
NO2
NO2
NO2
CN
Br
Shot
Shot
Powder
Shot
Shot
Powder
Shot
Powder
Powder
Powder
Shot
1/1.5/4
1/1.5/8
1/1.5/8
1/1.5/4
1/1.5/8
1/1.5/8
1/1.5/4
1/1.5/8
1/1.5/8
1/1.5/8
1/1.5/4
1/1.5/4
1/1.5/8
44
46
56
36
70
51
40
69
53
50
20d
31d
36d
Cl
Me
OMe
OMe
OMe
Powder
Powder
a See footnotes in Table 1. For all reactions, milling time was fixed at 0.5 h. b Extraction of the product was made with 0.1 M HCl and EtOAc.
c Determined by 1H NMR.
Depending on the carbonyl compounds employed, the reac-
tion mixture sometimes became very sticky during the course of
ball milling, where the stainless steel balls and bismuth shot
clung together to form a doughy lump in the reaction vessel.
Under such a circumstance, the allylation did not proceed to
completion. This problem could partly be solved by adding
some non-reactive co-milling agent such as calcite grains to the
reaction mixture.
The mechanism of the reductive allylation of carbonyl com-
pounds under solvent-free conditions is not clear. However, it is
most likely that the reaction proceeds via an electron-transfer
process involving an allylbismuth species generated on the nas-
cent metal surface, which is continuously supplied during the
milling process of the bismuth metal. The alkylation of carb-
onyl compounds with Grignard and organolithium reagents
has been shown to occur via an electron-transfer mechanism.30
In summary, the bismuth-mediated Barbier-type allylation of
aromatic aldehydes with allyl halides has been carried out
smoothly under solvent-free conditions using the ball milling
technique. In contrast to the failure in solution reaction, aro-
matic ketones were also allylated with moderate ease under dry
conditions, giving the expected tertiary homoallylic alcohols in
modest yield. Although more work is needed before the method
becomes practically attractive, our study has shown the poten-
tial usefulness of carrying out organic reactions under solvent-
free environments.
0.37 mmol), bismuth shot (44.9 mg, Wako 99.999%), and two
stainless steel ball bearings of 7 mm diameter were placed in a
stainless steel vial (1.2 × 4 cm; 5 ml vol.) with a screw cap. The
reaction vessel was shaken using a ball mill apparatus at a rate
of 30 Hz at room temperature. The vessel warmed somewhat by
mechanical friction. After 1 h, milling was stopped and the
resulting gray pasty mass was extracted with EtOAc by tritur-
ation. The organic phase was filtered on a thin Celite bed and
evaporated to leave a pale yellow oil, which was chromato-
graphed on silica gel using hexane–EtOAc as the solvent to give
1-(4-nitrophenyl)but-3-en-1-ol31 (46.2 mg, 95%) as an oil. IR
1
(liquid film): 3410 (OH), 1520 (NO2), 1346 (NO2) cmϪ1. H
NMR: δ 8.20 (d, 2H, J = 8.6 Hz, ArH), 7.53 (d, 2H, J = 8.6 Hz,
ArH), 5.72–5.86 (m, 1H, CH), 5.20 (d, 1H, J = 10.5 Hz,
cis-vinylic CH), 5.18 (d, 1H, J = 17.5 Hz, trans-vinylic CH),
4.87 (dd, 1H, J = 5.2, 7.2 Hz, CH), 2.52–2.61 (m, 1H, CH2),
2.41–2.50 (m, 1H, CH2), 2.32 (s, 1H, OH).
Allylation of aromatic ketones. Typical procedure. 4-Nitro-
acetophenone (43.1 mg, 0.26 mmol), allyl bromide (45.3 mg,
0.37 mmol), bismuth shot (203 mg), and two stainless steel balls
were placed in a stainless steel vial with a screw cap. The reac-
tion vessel was shaken using a ball mill apparatus at a rate of
30 Hz. After 0.5 h, milling was stopped and the resulting pasty
solid was first triturated with 0.1 M HCl and then extracted
with EtOAc. Attempted extraction with EtOAc alone led to a
diminished yield of the product. The organic phase was separ-
ated, filtered on a thin Celite bed, and evaporated to leave a pale
yellow oil, which was chromatographed on silica gel using
hexane–EtOAc as the eluent to give 2-(4-nitrophenyl)pent-
4-en-2-ol32 (37.7 mg, 70%) as an oil. IR (liquid film): 3540
Experimental
General
All reagents and metals were reagent-grade commercial prod-
ucts and used as received. 1H-NMR spectra were determined in
CDCl3 on a JEOL ECA-300 or a Varian Unity Plus 300 spec-
trometer using tetramethylsilane as an internal standard. IR
spectra were measured on a JEOL FTIR-5300 spectrophoto-
meter against liquid film on NaCl plate and only characteristic
peaks were recorded. The optical purity of homoallyl alcohol 3
(R = CF3) was estimated on a gas chromatograph HP 6890 GC
using a chiral column CP-chiral-Dex CB. Milling was per-
formed using a ball mill apparatus (Retsch mixer mill, MM200;
Retsch GmbH, Haan, Germany) at a rate of 30 Hz. Since the
mechanical performance of our apparatus was fixed, no effort
was made to optimize the milling conditions nor product yield.
All products are known and identified by direct comparison of
their spectral data with those of authentic specimens obtained
by other routes.
1
(OH), 1518 (NO2), 1348 (NO2) cmϪ1. H NMR: δ 8.20 (d, 2H,
J = 8.9 Hz, ArH), 7.61 (d, 2H, J = 8.9 Hz, ArH), 5.17 (d, 1H,
J = 10.6 Hz, cis-vinylic CH), 5.16 (d, 1H, J = 16.1 Hz, trans-
vinylic CH), 2.69 (dd, 1H, J = 6.5, 13.7 Hz, CH2), 2.54 (dd, 1H,
J = 8.3, 13.7 Hz, CH2), 2.17 (s, 1H, OH), 1.58 (s, 3H, CH3).
Acknowledgements
The authors thank Professor Toyoshi Shimada of Kyoto
University for helpful discussions. We also thank Dr. Asato
Kina of the same university for estimating the optical purity of
a chiral homoallyl alcohol by gas chromatography.
References
1 For a review, see: (a) Y. Yamamoto and N. Asao, Chem. Rev., 1993,
93, 2207; (b) R. W. Hoffman, Angew. Chem., Int. Ed. Engl., 1982, 21,
555.
2 (a) M. S. Kharasch and C. F. Fuchs, J. Org. Chem., 1944, 9, 359;
(b) H. Gilman and J. H. McGlumphy, Bull. Soc. Chem. Fr., 1928, 43,
1322.
Bismuth-mediated allylation of aromatic carbonyl compounds
with allyl halides
Allylation of aromatic aldehydes. Typical procedure. 4-Nitro-
benzaldehyde (38.4 mg, 0.25 mmol), allyl bromide (45.3 mg,
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 6 0 – 2 1 6 3
2162

View Article Online
3 (a) D. Seyferth, G. J. Murphy and B. Mauze, J. Am. Chem. Soc.,
1977, 99, 5317; (b) D. Seyferth and M. A. Weiner, J. Org. Chem.,
1961, 26, 4797.
4 (a) A. P. Davis and M. Jaspars, Angew. Chem., Int. Ed. Engl., 1992,
31, 470; (b) I. Fleming, J. Dunogues and R. Smithers, Org. React.
(N. Y.), 1989, 37, 57; (c) A. Hosomi, A. Shirahata and H. Sakurai,
Tetrahedron Lett., 1978, 3043.
5 (a) Y. Yamamoto, H. Yatagai, Y. Ishihara, N. Maeda and
K. Maruyama, Tetrahedron, 1984, 40, 2239; (b) Y. Naruta, S. Ushida
and K. Maruyama, Chem. Lett., 1979, 919.
18 W.-B. Wang, L.-L. Shi and Y.-Z. Huang, Tetrahedron, 1990, 46,
3315.
19 M. Wada, H. Ohki and K.-y. Akiba, Tetrahedron Lett., 1986, 27,
4771.
20 (a) M. Wada, T. Fukuma, M. Morioka, T. Takahashi and
N. Miyoshi, Tetrahedron Lett., 1997, 38, 8045; (b) M. Wada, H. Ohki
and K.-y. Akiba, Bull. Chem. Soc. Jpn., 1990, 63, 1738; (c) M. Wada,
H. Ohki and K.-y. Akiba, J. Chem. Soc., Chem. Commun., 1987, 708.
21 S. Araki, S.-J. Jin, Y. Idou and Y. Butsugan, Bull. Chem. Soc. Jpn.,
1992, 65, 1736.
6 For a survey of the Barbier reaction see: C. Blomberg and
F. A. Hartog, Synthesis, 1977, 18.
7 R. T. Arnold and E. C. Coyner, J. Am. Chem. Soc., 1944, 66, 1542.
8 (a) T. Mandai, J. Nokami, T. Yano, Y. Yoshinaga and J. Otera,
J. Org. Chem., 1984, 49, 172; (b) T. Mukaiyama and T. Harada,
Chem. Lett., 1981, 1527.
9 (a) C.-J. Li and T. H. Chan, Tetrahedron Lett., 1991, 32, 7017;
(b) S. Araki, H. Ito and Y. Butsugan, J. Org. Chem., 1988, 53,
1831.
22 (a) For a survey, see M. Wada and N. Miyoshi, J. Synth. Org. Chem.,
1999, 57, 689; (b) C.-J. Li, Tetrahedron, 1996, 52, 5643; (c) C.-J. Li,
Chem. Rev., 1993, 93, 2023.
23 (a) M. Minato and J. Tsuji, Chem. Lett., 1988, 2049; (b) H. Tanaka,
S. Yamashita, T. Hamatani, Y. Ikemoto and S. Torii, Chem. Lett.,
1986, 1611; (c) K. Uneyama, H. Matsuda and S. Torii, Tetrahedron
Lett., 1984, 25, 6017.
24 J. M. Harrowfield, R. J. Hart and C. R. Whitaker, Aust. J. Chem.,
2001, 54, 423.
10 (a) K. Tanaka, S. Kishigami and F. Toda, J. Org. Chem., 1991, 56,
4333; (b) T. Shono, M. Ishifune and S. Kashimura, Chem. Lett.,
1990, 449; (c) C. Petrier and J.-L. Luche, J. Org. Chem., 1985, 50,
910.
11 S. Araki, H. Ito and Y. Butsugan, J. Organomet. Chem., 1988, 347, 5.
12 (a) T. Imamoto, T. Kusumoto, Y. Tawarayama, Y. Sugiura, T. Mita,
Y. Hatanaka and M. Yokoyama, J. Org. Chem., 1984, 49, 3904;
(b) T. Imamoto, Y. Hatanaka, Y. Tawarayama and M. Yokoyama,
Tetrahedron Lett., 1981, 22, 4987.
25 M. G. Aylmore, F. J. Lincoln, J. E. Cosgriff, G. B. Deacon,
B. M. Gatehouse, C. A. Sandoval and L. Spiccia, Eur. J. Solid State
Inorg. Chem., 1996, 33, 109.
26 L. D. Field, S. Sternhell and H. V. Wilton, Tetrahedron, 1997, 53,
4051.
27 S. A. Rowlands, A. K. Hall, P. G. McCormick, R. Street, R. J. Hart,
G. F. Ebell and P. Donecker, Nature, 1994, 367, 223.
28 Organobismuth Chemistry, H. Suzuki and Y. Matano, Eds. Elsevier,
Amsterdam, 2001, Ch. 1.
13 (a) T. Hiyama, M. Sawahata and M. Obayashi, Chem. Lett., 1983,
1237; (b) T. Hiyama, M. Obayashi and A. Nakamura, Organo-
metallics, 1982, 1, 1249.
14 Y. Butsugan, H. Ito and S. Araki, Tetrahedron Lett., 1987, 28, 3707.
15 M. Wada and K.-y. Akiba, Tetrahedron Lett., 1985, 26, 4211.
16 T. Hiyama, Y. Okude, K. Kimura and H. Nozaki, Bull. Chem. Soc.
Jpn., 1982, 55, 561.
17 (a) J. Souppe, J. L. Namy and H. B. Kagan, Tetrahedron Lett., 1982,
23, 3497; (b) P. Girard, J. L. Namy and H. B. Kagan, J. Am. Chem.
Soc., 1980, 102, 2693.
29 J. Emsley, Elements, 3rd Edn., Oxford University Press, Oxford,
1998.
30 (a) H. Yamataka, T. Matsuyama and T. Hanafusa, J. Am. Chem.
Soc., 1989, 111, 4912 and references cited therein; (b) H. Yamataka,
Y. Kawafuji, K. Nagareda, N. Miyano and T. Hanafusa, J. Org.
Chem., 1989, 54, 4706.
31 R. A. Batey, A. N. Thadani, D. V. Smil and A. J. Lough, Synlett,
2001, 1802.
32 R. Hamasaki, Y. Chounan, H. Horino and Y. Yamamoto,
Tetrahedron Lett., 2000, 41, 9883.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 2 1 6 0 – 2 1 6 3
2163