High regio- and stereoselective Barbier reaction of carbonyl compounds mediated by NaBF4/Zn (Sn) in water

Article abstract of DOI:10.1039/b303187j

Studies with NaBF₄/M (M = Zn or Sn) showed that this novel mediator facilitated allylation of a variety of carbonyl compounds in water and had a great influence on the diastereoselectivity of the addition. More importantly, α- and γ-addition pr

Full text of DOI:10.1039/b303187j

View Article Online / Journal Homepage / Table of Contents for this issue
High regio- and stereoselective Barbier reaction of carbonyl
compounds mediated by NaBF4/Zn (Sn) in watery
Zhenggen Zha, Zhen Xie, Cunliu Zhou, Mingxin Chang and Zhiyong Wang*
Department of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026,
China. E-mail: zwang3@ustc.edu.cn; Fax: +86 551 363 1760; Tel: +86 551 360 3185
L e t t e r
Received (in Montpellier, France) 20th March 2003, Accepted 6th June 2003
First published as an Advance Article on the web 24th July 2003
Table 1 Effects of NaBF4 on the allylation of benzaldehyde
Studies with NaBF4/M (M ¼ Zn or Sn) showed that this novel
mediator facilitated allylation of a variety of carbonyl com-
pounds in water and had a great influence on the diastereosel-
ectivity of the addition. More importantly, a- and c-addition
products of crotylations can be alternatively obtained under the
control of this novel mediator. A reaction mechanism is proposed
based on our quantum calculation and experimental results.
Entry
Metal
NaBF4/mol Lꢀ1
Time/h
% Yielda
1
2
3
4
5
6
Zn
Zn
Zn
Sn
Sn
Sn
—
24
24
24
2
19
76
86
—
88
98
0.25
0.50
—
0.25
0.25
2
16
Allylation of carbonyl compounds with allyl halides is a very
important reaction, and numerous reagents and methods have
been developed to accomplish this transformation.1 Parti-
cularly, over the last decade, the Barbier-type reaction in
aqueous media2 has drawn great attention. Numerous metals
have been shown to mediate the addition of allyl halides to
carbonyl compounds in aqueous media to give rise to the cor-
responding homoallylic alcohols.3 All the reported metals are
commercially available metal powders. Several studies have
indicated that zinc- or tin-mediated allylation can only take
place in the presence of ammonium chloride, in organic solvent
or over long reaction times.4 In some cases, the commercial
metal is not active enough to mediate the allylation under
benign environmental conditions.5,6b,c For example, commer-
cial zinc was apparently not successful in mediating the allyla-
tion reaction in distilled water; then certain salts and reducing
reagents such as fluoride salt, cupric chloride, aluminum, and
NaBH4 were employed to activate the metals.6 In our studies,
we focus on the development of reaction conditions in which
commercial zinc or tin can be directly used in allylation reac-
tions without requiring an acidic environment or organic
solvent. In addition to nanotin,7 we recently have found that
sodium tetrafluoroborate is also effective in ‘‘activating’’ zinc
and tin in aqueous media to facilitate addition of allyl bromide
to carbonyl compounds. More interestingly, high regio- and
stereoselective crotylation can be realized under these condi-
tions while it was reported that this reaction can be carried
out only in organic solvent and specifically metallorganics.8
In our preliminary experiments, the reaction was carried out
in distilled water with the molar ratio of benzaldehyde : allyl
bromide : Zn at 1 : 2 : 2 and resulted in a low amount of the
corresponding homoallylic alcohol even after vigorous stirring
of the reaction mixture for 24 h (entry 1, Table 1). When Zn
was replaced with Sn under the same conditions, no allylation
product was observed after 2 h. Only by prolonging the reac-
tion time to more than 10 h can we get the corresponding alco-
hol in a modest yield. These results suggest that the metal
needs to be activated in order to facilitate this reaction. We
then examined a number of phase transfer catalysts9 and salts
a
All yields were determined by 1H NMR.
as additives in the aqueous media. Out of a large number of
conditions tested, we found that 0.12–0.50 mol Lꢀ1 of sodium
tetrafluoroborate was quite efficient in activating zinc to give
a good yield of the corresponding homoallylic alcohol (entries
2 and 3, Table 1). Moreover, it was shown that tin was more
effective than zinc and the tested allylation reaction proceeded
quantitatively (entries 5 and 6, Table 1). Taken together, we
conclude that 0.25 mol Lꢀ1 of NaBF4 is the optimal reaction
media to activate commercial zinc and tin metal in allylation
reactions.
We examined allylations of a variety of carbonyl com-
pounds mediated by tin or zinc in the presence of NaBF4
(Scheme 1). The results are summarized in Table 2. The experi-
mental results show that Sn-mediated carbonyl allylations gave
rise to excellent yields and can be accomplished in a relatively
short time (entries 1–6, entries 8–10, Table 2), whereas Zn
mediated allylations exhibited only relatively good or modest
yields. NaBF4 appeared to facilitate the allylation reactions
in both the reaction rate and the efficiency of the conversion
regardless of the type of aldehyde as allylations for both aro-
matic and aliphatic aldehydes proceeded well (entries 1–10,
Table 2).
To the best of our knowledge, ketones, either aromatic or
aliphatic, hardly undergo allylation, probably due to steric
hindrance. However, in the presence of NaBF4 , allylations of
ketones proceed smoothly (entries 7 and 10, Table 2). Further-
more, allylations of aldehydes that did not occur with Zn alone
in distilled water also proceeded with modest yields in the
presence of Zn and NaBF4 together (Table 2). As compared
to zinc, tin appeared more effective and comprehensive as a
mediator in various types of allylation reactions and produced
the corresponding homoallylic alcohols in high yields.
y Electronic supplementary information (ESI) available: spectral data
of Barbier-type reaction products and quantum calculation results. See
http://www.rsc.org/suppdata/nj/b3/b303187j/
Scheme 1
DOI: 10.1039/b303187j
New J. Chem., 2003, 27, 1297–1300
1297
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
Table 2 Allylation of carbonyl compounds mediated by NaBF4/Zn (Sn) in aqueous medium
Entry
1
Carbonyl compound
Metal
Time/h
Product
% Yielda
Zn
Sn
18
16
76
98
2
3
Zn
Sn
18
42
2.5
100
Zn
Sn
Zn
20
2
76
100
52
4
5
20
Sn
2
100
50
Zn
20
Sn
Zn
Sn
5
20
3
100
47
6
7
100
Zn
Sn
18
16
25
60
8
9
n-C6H13CHO
Zn
Sn
24
1
95
100
Zn
24
70
Sn
Zn
Sn
4
24
4
100
–
10
95
a
All yields were determined by 1H NMR.
In an effort to determine whether NaBF4 can control reac-
tion selectivity in allylations, a property that may be important
for the development of new organic synthetic methods, we stu-
died the regioselectivity and diastereoselectivity of allylations
mediated by NaBF4 (Scheme 2). The investigation was focused
on allylations of benzaldehyde with ethyl 4-bromobutenoate
or crotyl bromide, and of 3-hydroxybutan-2-one with allyl
bromide in water (entries 11–18, Table 3). First of all, NaBF4
had an important influence on the distribution of products.
When 1.5 mmol of tin was used without NaBF4 for allylation
of benzaldehyde with ethyl 4-bromobutenoate, the reaction
gave rise to the corresponding g-addition products in a yield
of 58% (entry 11, Table 3), whereas the allylation afforded
a-addition products in a yield of 74% in the presence of 0.25
mol L12 of NaBF4 (entry 12, Table 3). 1H and 13C NMR
spectral analysis indicated that the g-products were diastereo-
isomers and the syn isomer was predominant because it is a
chelate-controlled product, probably due to the hydrogen
bond between the proton in the hydroxyl group and the oxy-
gen in the ester group (entry 11, Table 3). Also, the reaction
of benzaldehyde with crotyl bromide generated exclusively
the a-addition product in the presence of NaBF4 (entries 15
and 16, Table 3), whereas the same reaction gave a mixture
of a- and g-addition products without NaBF4 (entry 14, Table
3). Moreover, NaBF4 had an influence on the stereoselectivity
of allylations. When benzaldehyde was reacted with crotyl bro-
mide, the amount of E a-addition product increased from 47%
to 59–62% with varying NaBF4 concentration, from 0 to 0.12–
0.25 mol Lꢀ1, respectively (entries 14–16, Table 3). The diaste-
reoselectivity for the allylation of 3-hydroxybutan-2-one with
allyl bromide was also studied under the same conditions. This
reaction gave rise to the corresponding addition products and
NMR spectra (1H and 13C) indicated that the syn isomer was
the dominant product because of the hydrogen bond between
the two hydroxyl groups in the product (entry 17, Table 3).
Scheme 2
1298
New J. Chem., 2003, 27, 1297–1300

View Article Online
Table 3 Regio- and diastereoselectivity for the allylation and crotylation of 1 with 2 under different conditions
% Yielda
Entry
R1
R2
R3
Sn/mmol
NaBF4/mol Lꢀ1
Time/h
3a (Z:E )
3g (syn:anti )b
11
12
13
14
15
16
17
Ph
Ph
Ph
Ph
Ph
Ph
H
COOC2H5
COOC2H5
COOC2H5
CH3
1.5
1.5
1.2
1.5
1.5
1.2
1.5
0
24
15
10
24
15
10
8
—
58 (74:26)
—
H
0.25
0.50
0
74 (2:98)
62 (1:99)
54 (53:47)
75 (41:59)
73 (38:62)
75 (63:37)c
H
—
22 (37:63)
H
H
CH3
CH3
0.12
0.25
0
—
—
H
CH3
H
18
CH3
H
1.5
0.12
6
90 (66:34)c
a
1
Isolated yield. b The ratio of syn isomer to anti isomer was determined by H NMR, 13C NMR and quantum calculations. c The two products
are diastereoisomers of each other.
When NaBF4 was added to this reaction, the yield was
enhanced from 75% to 90% but the ratio of syn to anti pro-
ducts was almost the same (entry 18, Table 3).
show that the electron density of the a-carbon atom that bonds
to tin in intermediate (2) is the richest (R3 ¼ COOC2H5 :
r ¼ ꢀ1.013; R3 ¼ CH3 : r ¼ ꢀ0.959) and thus it becomes
more nucleophilic. Then intermediate (2) coordinates with
the carbonyl group of the carbonyl compound to give rise to
the corresponding six-membered ring transition structure. As
a result, the a-addition product is the product of the allylation
after addition of sodium fluoroborate (entries 12 and 13, Table
3). In entry 14 of Table 3, quantum calculations10 show that
the a-carbon has a higher electron density (ꢀ0.695) than the
g-carbon (ꢀ0.421) in transition state (1). This result reveals
that the a-carbon adduct is formed preferentially, accompa-
nied by the g-addition product. When NaBF4 is added to the
reaction, intermediate (2) (the electron density of the a-carbon
is ꢀ0.959 compared with an electron density of ꢀ0.118 on
the g-carbon ) is generated and coordinates with the carbonyl
compound to afford transition state (3) as shown in Scheme 3,
resulting in the generation of the a-addition product (entries 15
and 16, Table 3). As for entries 17 and 18, two diastereo-
isomers are obtained. Quantum calculations indicate that the
syn isomer is more stable (the energy difference between syn
and anti is 9.32 kJ molꢀ1 under B3LYP/6-31G).10
In conclusion, NaBF4 in distilled water can facilitate tin-
or zinc-mediated allylations to give excellent yields as well as
to shorten the reaction time. More importantly for crotylation,
a- and g-addition products can be alternatively obtained
whether NaBF4 in water is used or not. Quantum calculations
further demonstrated that the dispersion and stereochemistry
of the corresponding products are consistent with the calcu-
lated results and the proposed mechanism in which NaBF4
promotes the reaction. Further studies of the effects of NaBF4
on the Barbier-type reaction of carbonyl compounds in water
are currently in progress.
We hypothesize that NaBF4 affects the stereoselectivity
according to the mechanism shown in Scheme 3. Firstly, tin
reacts with allyl bromide to afford a tin compound (1). In
the presence of NaBF4 , fluoroborate can coordinate with (1)
to give rise to the corresponding intermediate (2). Then this
nucleophile attacks carbonyl compounds through a six-mem-
bered ring transition state (3), which results in the a-adduct.
Without NaBF4 , quantum calculations10 indicate that the
electron density of the g-carbon is the richest when R3 is an
ester group. When R3 is a methyl group, the electron density
of the a-carbon is slightly higher than that of the g-carbon,
leading to the mixture of a- and g-adducts.
This proposed mechanism as shown in Scheme 3 can pro-
vide an explanation for the stereoselectivity of the allylations
in Table 3. As for entry 11 in Table 3, the rich electron density
of the g-carbon (ꢀ0.875) promotes the attack on the carbonyl
group of the aldehyde. In addition, the COOC2H5 group
can facilitate the formation of a hydrogen bond between the
hydroxyl group and the ester group in transition state (4)
and promote the formation of the g-addition product. Also,
quantum calculations show that the total energy of syn isomers
is the lowest among the isomers (see Electronic supplementary
information). Thus, the formation of the syn product is fav-
ored. After addition of sodium fluoroborate, intermediate
(2)11 is formed as shown in Scheme 3. Quantum calculations
Experimental
Spectral data—IR (Perkin–Elmer, 2000 FTIR), 1H NMR
(CD3Cl, 500 MHz or 400 MHz), 13C NMR (CDCl3 , 125.7
MHz or 100 MHz) and MS-GC (HP5890 (II)/HP5972, EI)—
were acquired on the indicated instrumentation. All reagents
were of commercial quality from Aldrich Chemical Company
and J&K Chemical Ltd. All organic solvents were purchased
from the First Factory of Chemical Reagents in Shanghai
(China).
General experimental procedure
To a mixture of aldehyde (1 mmol) in 0.25 mol Lꢀ1 aqueous
sodium tetrafluoroborate solution (4 mL) and allyl bromide
(2.0 mmol), metal powder (2 mmol) was added in one portion
and the mixture was vigorously stirred until the metal powder
Scheme 3 Proposed mechanism for allylations in the presence of
NaBF4 .
New J. Chem., 2003, 27, 1297–1300
1299

View Article Online
Chem., 1998, 63, 7498; ( f ) C. J. Li, Y. Meng, X. H. Yi, J. H. Ma
had reacted. Ethyl acetate was added to the reaction mixture
and the organic layer was separated. The aqueous phase was
extracted with ethyl acetate. The organic extracts were com-
bined and dried over anhydrous MgSO4 , then was filtered
and evaporated. The residue, for most of the aldehydes,
afforded the corresponding alcohols with high purity. If nec-
essary, purification was performed by flash chromatography
over silica gel using ethyl acetate–petroleum ether as eluent
(Rf ¼ 0.2).
and T. H. Chan, J. Org. Chem., 1997, 62, 8632; (g) M. Wada,
H. Ohki and K. Y. Akiba, Bull. Chem. Soc. Jpn., 1990, 63,
2751; (h) M. Wada, H. Ohki and K. Y. Akiba, J. Chem. Soc.,
Chem. Commun., 1987, 708; (i) K. Uneyama, N. Kamaki, A.
Moriya and S. Torii, J. Org. Chem., 1985, 50, 5396; ( j) J. Nokami,
S. Wakabayashi and R. Okawara, Chem. Lett., 1984, 869; (k) C.
Petrier and J. L. Luche, J. Org. Chem., 1983, 50, 910; (l) J.
Nokami, J. Otera, T. Sudo and R. Okawara, Organometallics,
1983, 2, 191.
(a) C. L. Zhou, Z. G. Zha and Z. Y. Wang, Chin. J. Chem., 2002,
20, 718; (b) D. Marton, D. Stivanello and G. Tagliavini, J. Org.
Chem., 1996, 61, 2731; (c) S. R. Wilson and M. E. Guazzaroni,
J. Org. Chem., 1989, 54, 3087; (d ) C. Einhorn and J. L. Luche,
J. Organomet. Chem., 1987, 322, 177.
4
Spectral data are given for one product below as an example
and full spectral data for all products in Tables 2 and 3 are
given in the ESI.
`
5
6
(a) B. Alcaide, P. Almendros, C. Aragoncillo and R. Rodrıguez-
syn-2-(Hydroxyphenylmethyl)-but-3-enoic acid ethyl ester.
(entry 1, Table 3) IR (NaCl, cmꢀ1): 3483, 1728, 1638, 1318,
1177, 765, 701; H NMR (CDCl3 , ppm): 7.26–7.23 (m, 5H),
5.88 (m, 1H), 5.18 (m, 1H), 5.09 (m, 1H), 4.91 (d, J ¼ 6.4
Hz, 1H), 3.95 (q, J ¼ 7.2 Hz, 2H), 3.25 (q, J ¼ 6.4 Hz,
J ¼ 8.8 Hz, 1H), 1.02 (t, J ¼ 7.2 Hz, 3H); 13C NMR (CDCl3 ,
ppm): 172.5, 140.7, 132.0, 128.2, 127.9, 126.5, 120.5, 74.0, 60.9,
58.4, 14.0; MS: m/z 220 (M+).
Acebes, J. Org. Chem., 2001, 66, 5208; (b) T. H. Chen, Y. Yang
and C. J. Li, J. Org. Chem., 1999, 64, 4452.
1
(a) X. H. Tan, B. Shen, L. Liu and Q. X. Guo, Tetrahedron Lett.,
2002, 43, 9373; (b) L. H. Li and T. H. Chen, Tetrahedron Lett.,
2000, 41, 5009; (c) L. H. Li and T. H. Chan, Org. Lett., 2000, 2,
1129; (d ) P. D. Ren, Q. H. Jin and Z. P. Yao, Synth. Commun.,
1997, 27, 2761; (e) A. Kundu, S. Prabhakar, M. Vairamani and
S. Roy, Organometallics, 1997, 16, 4796; ( f ) W. Wang, L. Shi
and Y. Huang, Tetrahedron, 1990, 46, 3315.
7
8
Z. Y. Wang, Z. G. Zha and C. L. Zhou, Org. Lett., 2002, 4, 1683.
(a) T. P. Loh, K. T. Tan, J. Y. Yang and C. L. Xiang, Tetrahedron
Lett., 2001, 42, 8701; (b) R. Gewald, M. Kira and H. Sakurai,
Synthesis, 1996, 111; (c) K. Mikami, Y. Kimura, N. Kishi and
T. Nakai, J. Org. Chem., 1983, 48, 281.
Acknowledgements
This research was supported by NSFC (No. 50073021), Nature
Science Foundation of Anhui Province (No. 01046301)
and Education Department of Anhui Province (No.
2002kj330ZD).
9
(a) Z. G. Zha, Z. Xie, C. L. Zhou, Z. Y. Wang and Y. S. Wang,
Chin. J. Chem., 2002, 20, 1477; (b) Z. G. Zha, Y. S. Wang, G.
Yang, L. Zhang and Z. Y. Wang, Green Chem., 2002, 4, 578.
10 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery,
R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D.
Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala,
Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L.
Andres, M. Head-Gordon, E. S. Replogle and J. A. Pople,
Gaussian 98, Revision A.11, Gaussian Inc., Pittsburgh, PA, 1998.
11 J. M. Blackwell, W. E. Piers and R. McDonald, J. Am. Chem.
Soc., 2002, 124, 1295.
References
1
2
Y. Yamamoto and N. Asao, Chem. Rev., 1993, 93, 2207.
(a) P. Sinha and S. Roy, Chem. Commun., 2001, 1798; (b) C. J. Li
and T. H. Chan, Organic Reactions in Aqueous Media, John Wiley
& Sons, New York, 1997; (c) C. J. Li, Tetrahedron, 1996, 52, 5643;
´
(d ) A. Lubineau, J. Auge and Y. Queneau, Synthesis, 1994, 741;
(e) C. J. Li, Chem. Rev., 1993, 93, 2023.
3
(a) Z. Y. Wang, S. Z. Yuan and C. J. Li, Tetrahedron Lett., 2002,
43, 5097; (b) W. C. Zhang and C. J. Li, J. Org. Chem., 1999, 64,
3230; (c) C. J. Li and T. H. Chan, Tetrahedron, 1999, 55, 11 149;
(d ) C. J. Li and W. C. Zhang, J. Am. Chem. Soc., 1998, 120, 9102;
(e) C. J. Li, Y. Meng, X. H. Yi, J. H. Ma and T. H. Chan, J. Org.
1300
New J. Chem., 2003, 27, 1297–1300