BiBr3-catalyzed benzylation of alcohols. Stereochemistry and mechanistic investigations

Article abstract of DOI:10.1016/S0040-4020(01)00014-X

We have investigated the benzylation of optically active aliphatic alcohols (octan-2-ol and butan-2-ol) as well as cis and trans 2-methylcyclohexanol or (S)-(-)-menthol, catalyzed by bismuth (III) bromide in the presence of (-) or (±)-1-phenylethanol (PeOH). Under mild conditions, aliphatic alcohols provided an equimolar mixture of diastereomeric ethers while alicyclic alcohols gave rise to ethers with retention of configuration. To explain these results, we assumed that BiBr₃, acting as Lewis acid, would lead to the formation of a 6-fold coordinated bismuth intermediate involving two molecules of 1-phenylethanol and one molecule of aliphatic alcohol. We have proposed a mechanism to explain these BiBr₃-promoted etherification reactions.

Full text of DOI:10.1016/S0040-4020(01)00014-X

TETRAHEDRON
Pergamon
Tetrahedron 57 (2001) 1917±1921
BiBr -Catalyzed benzylation of alcohols. Stereochemistry
3
and mechanistic investigations
p
El Mehdi Keramane, Bernard Boyer and Jean-Pierre Roque
Laboratoire de Chimie Organique Physique, Universit e Montpellier II, Place E. Bataillon, 34095 Montpellier Cedex 5, France
Received 3 July 2000; revised 20 December 2000; accepted 21 December 2000
AbstractÐWe have investigated the benzylation of optically active aliphatic alcohols (octan-2-ol and butan-2-ol) as well as cis and trans
2-methylcyclohexanol or (S)-(2)-menthol, catalyzed by bismuth (III) bromide in the presence of (2) or (^)-1-phenylethanol (PeOH). Under
mild conditions, aliphatic alcohols provided an equimolar mixture of diastereomeric ethers while alicyclic alcohols gave rise to ethers with
retention of con®guration. To explain these results, we assumed that BiBr , acting as Lewis acid, would lead to the formation of a 6-fold
3
coordinated bismuth intermediate involving two molecules of 1-phenylethanol and one molecule of aliphatic alcohol. We have proposed a
3
mechanism to explain these BiBr -promoted etheri®cation reactions. q 2001 Elsevier Science Ltd. All rights reserved.
1
. Introduction
2. Results and discussion
In a series of preliminary studies, we have established the
general usefulness of bismuth (III) reagents for the halo-
genations of alcohols, the halogen exchange reactions of
The stereochemistry of the reaction and the mechanistic
examination were investigated under the same previous
experimental conditions using alternatively (2) or (^)-1-
1
2
3
halides with various BiX and more recently, the protection
3
of aliphatic alcohols as benzylic ethers.
phenylethanol (PeOH) as benzylic alcohol, and (1) or (^)-
octan-2-ol and butan-2-ol as aliphatic alcohols. Reactions
3
were run in the presence of BiBr . The general scheme was
3
as follows:
Therefore, we showed that BiBr was very ef®cient in
3
promoting the benzylation of aliphatic alcohols (primary,
secondary and tertiary) by various benzylic alcohols (benzyl
alcohol, 1-phenylethanol and 2-phenylpropan-2-ol). This
3
was achieved under good yield, under mild conditions.
In this paper, we extended our investigation, by studying the
reaction mechanism of the noteworthy catalytic activity of
BiBr for the benzylation of alcohols. In order to do so, we
extended the examples to a series of optically active alco-
holsÐboth benzylic and aliphatic. Concerning the benzylic
alcohols, we have chosen the 1-phenylethanol (PeOH) for
the following reasons:
For each reaction, optical rotations of isolated phenylethyl
ethers (PeOR) were measured. These results are reported in
Table 1.
3
In any case, the ether is obtained by the condensation of two
chiral alcohols. One alcohol molecule undergoes a C±O
bond breaking and can be considered as the substrate. The
other alcohol molecule retains its oxygen atom and plays the
role of a nucleophilic reactant. Only the alcohol substrate is
able to undergo a con®guration change (partial or complete
inversion), unless the reaction proceeds with retention of
con®guration.
(
i) it provides the possibility of introducing a chiral center
in the corresponding unsymmetrical ether;
ii) regardless of the aliphatic alcohol, its reactivity was
(
slightly lower than that observed with the 2-phenyl-
propan-2-ol, but signi®cantly higher than that of benzyl
alcohol;
3
The data reported in Table 1 calls for the following remarks:
(iii) ethers derived from 1-phenylethanol proved to be
stable under acidic conditions, but were easily cleaved
by hydrogenolysis.
3
(
i) entry 1: in this case, a symmetrical ether was formed as
a mixture of two diastereoisomers in a 50/50 ratio. This
result indicates that there is a total racemisation of the
alcohol substrate and therefore yields the two ethers (S,S)
and (S,R);
Keywords: reaction mechanism; bismuth bromide; benzylation; Lewis acid.
Corresponding author. Fax: 133-4-67-14-38-8;
e-mail: bboyer@univ-montp2.fr
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(01)00014-X

1
918
E. M. Keramane et al. / Tetrahedron 57 (2001) 1917±1921
3
3 4
Table 1. Optical rotations of phenylethylethers obtained using BiBr in CCl
a
a
b
D
c
Yield (%)
d
Entry
PeOH
ROH
Reaction times (min)
[a]
GC
e
1
2
3
4
5
6
7
(2)-S
(2)-S
(2)-S
(^)-
(2)-S
(^)-
(2)-S
±
20
45
45
45
20
20
20
236.28e
227.68
227.68
95
90
92
95
95
92
92
2 (50/50)
2 (50/50)
2 (50/50)
2 (50/50)
2 (55/45)
2 (55/45)
2 (55/45)
(^)-octan-2-ol
(1)-octan-2-ol
(1)-octan-2-ol
(1)-butan-2-ol
(1)-butan-2-ol
(^)-butan-2-ol
e
e
0.08
15.08
f
f
113.58
15.08
f
a
The optical rotations of pure alcohols are: (2)-PeOH ([a]
[a] OH), respectively.
ˆ1128; cˆ10, CH
Optical rotations were determined at 258C.
Isolated yields of ether, exhibiting H NMR data consistent with their assigned structures.
Signals detected by GC on the completion of the reaction. In all cases, two ethers are formed as observed by GC. Percentage of both ethers is given in
parentheses.
D
ˆ2458; cˆ5, CH
3
OH); (1)-octan-2-ol ([a]
D
ˆ19.8; cˆ5, CH
3
OH); (1)-butan-2-ol
(
D
3
b
c
d
1
e
f
cˆ5, CH
3
OH.
OH.
cˆ10, CH
3
(
ii) entries 2 and 3: in both cases, the same mixture of
Therefore, from these results, we can presume that only the
PeOH attack is responsible for the observed racemisation.
This observation led us to assume that the BiBr -promoted
diastereomeric ethers (50/50) is formed with the same
optical rotation. Clearly, the octan-2-ol plays the role of
the alcohol substrate and is completely racemised (entry
3
(
3
etheri®cation mechanism of alcohols was similar to that
previously suggested for the reaction of alcohols, in the
);
4
presence of BiBr . In that preceding study, we showed
iii) entry 4: the formation of an optically inactive ether's
3
mixture has been observed. This result con®rms the total
racemisation of the octan-2-ol;
that alcohols could yield ethers and alkenes, in appropriate
cases, involving a Bi(III) intermediate. The only role of
BiBr is, an ef®cient catalyst for both the etheri®cation
(
iv) entries 5±7: the butan-2-ol presents a similar beha-
3
vior to that of the octan-2-ol, except that the racemisation
is not complete.
and the elimination reactions.
It clearly appeared that the alcohol substrate undergoes
racemisation. Since this alcohol is incubated with BiBr3
before 1-phenylethanol is added, we presumed that the
alcohol racemisation would result from the reaction of the
water, which we thought might be present in BiBr with
3
the ROH/BiBr complex.
3
In a similar manner, we thought that this O-benzylation
reaction in the presence of BiBr would involve the forma-
tion of an intermediate complex, possessing several coordi-
nation-types around the bismuth atom. Thus, in such an
intermediate, the bismuth atom would interact with more
than one alcohol molecule.
3
In order to test this hypothesis, we carried out the following
experiment. We incubated (1)-octanol with BiBr3
Our reasons for postulating the existence of such an inter-
mediate were provided by the following observations:
(
1 equiv.) in CCl at room temperature for 45 min. Then,
4
we recovered octanol and measured its optical rotation. It
was the same as that of the pure starting alcohol. This leads
us to two possibilities: either no water is present in BiBr or
(i) the valence extension ability characteristic of heavy,
non-metallic elements confers to bismuth (III), the poten-
tial of being a soft Lewis acid. The introduction of elec-
5
3
water found in BiBr is in a too small quantity to have an
3
effect on racemisation.
tron withdrawing substituents (halogen atoms) onto the
bismuth atom enhances the Lewis acidity of the metal
center;
6
,13
However, if we take a closer look, we can notice that
the etheri®cation reaction leads to water formation. There-
fore, we foresaw that the water formed during the reaction
would promote the racemisation. To check this point, we
repeated the same previous experiment, except that, this
time one molar equivalent of water was added after
(ii) bismuth trihalides readily form adducts or coordina-
tion complexes with a large variety of Lewis bases.
Recently, an increasing number of complexes formed
by the reaction of organic substances with bismuth (III)
7
salts have been reported. There are many reports which
describe the coordination of a bismuth atom with
sili-
8
oxygen, nitrogen,
cium and sulfonyl groups. These results show that the
coordination number around the bismuth center increases
7d,9
10
sulfur, carbonyl groups,
7b,11
30 min of incubation. This mixture was maintained at
1
2
13
room temperature for 20 min, and then, the alcohol was
isolated.
7
a,b,10
from three to nine.
tion number
Furthermore, the high coordina-
makes the intermolecular interac-
7
a,b,e,8d,10
As in the previous experiment, the alcohol was optically
proving that the water formation cannot explain the racemi-
sation, occurring during the etheri®cation reaction.
tions possible around bismuth;
(iii) in addition, it has been shown
7
b,14
that BiX forms
3
donor±acceptor complexes with aromatic compounds.
These arene adducts consist of sheets and have a sand-
wich structure in which the benzene rings are bound
alternatively above and below the BiX3 sheets. So,
bismuth is centrally located above the aromatic ring by
4
Furthermore, we have previously suggested that the reac-
tion of alcohols with BiBr is inconsistent with a mechanism
3
involving carbocation, showing that racemisation is not due
to such an intermediate.

E. M. Keramane et al. / Tetrahedron 57 (2001) 1917±1921
1919
their electrophilic activity. Besides, the existence of such a
Bi±O bonded intermediate has been well demonstrated.
7
c,17
Finally, this activated oxygen atom, through coordination to
bismuth, turns the carbon atom bearing this hydroxyl group
into an electrophilic center. That would be likely to favor
the benzylic oxygen nucleophilic attack on either side of this
electron-de®cient carbon atom yielding the corresponding
mixture of diastereomeric ethers. Attack from side (a)
involves a con®guration inversion on the aliphatic alcohol,
while attack from side (b) implies a con®guration
retention. For some reason, both attacks are equivalent
with octan-2-ol, leading to an equimolar mixture of two
diastereoisomers. However, in the case of butan-2-ol, one
of the two attacks is slightly preponderant. So, the
mechanism for the benzylation of alcohols would involve
the activation of its hydroxyl function, through a complex
Scheme 1.
forming a p bond. Moreover, the interactions between
metal and arene are rather weak. Since the symmetrical
ether formation is speci®c to benzylic alcohols (i.e. no
reaction with aliphatic alcohols ), a particular interaction
with BiBr , followed by the nucleophilic attack of the
3
hydroxyl group of a benzylic alcohol bonded to the same
bismuth atom.
3
between BiBr and benzylic alcohols can be expected.
3
This interaction is most probably due to the p bond
formation between the bismuth atom and the benzene
ring.
Then, this study was extended to a series of cyclic alcohols
possessing a well-de®ned con®guration namely 2-methyl-
cyclohexanol (cis and trans) and the (S)-(2)-menthol.
Results are illustrated in the Scheme 2.
These results might indicate that BiBr can discriminate the
3
1
All the products gave satisfactory H NMR spectra. A
types of the hydroxyl functions, namely aliphatic and
benzylic alcohols. Besides, a similar observation
1
1a,15
marked contrast was observed between those cyclic alco-
hols and the aliphatic ones. Indeed, the spectroscopy data
have clearly demonstrated that corresponding ethers were
obtained speci®cally with con®guration retention of the
cyclic alcohols.
concerning an intramolecular complex, in which the
bismuth is coordinated to a carbonyl group, has been already
mentioned. The authors clearly showed that BiBr could
3
effectively discriminate between phenyl ketones and ali-
cyclic or aliphatic ketones.
1
6
However, in each case, an equimolar mixture of dia-
stereomers is formed (showed by GC analysis), since reac-
tions were performed from the racemic 1-phenylethanol.
A postulated reaction mechanism for the benzylation of
aliphatic alcohols is illustrated in Scheme 1.
According to experimental procedure, the initial stage of the
benzylation would be the formation of a key tetra-coordi-
nated complex between aliphatic alcohols and BiBr . Then,
3
the formation of bismuth±benzene ring interactions, with
two molecules of 1-phenylethanol, would lead to a 6-fold
coordinated bismuth atom.
We presumed that the reaction involved the same inter-
mediate-type as described above for aliphatic alcohols
(i.e. on the one hand a Bi±O interaction with cyclic alcohols
and on the other hand a bismuth±benzene ring interaction
with 1-phenylethanol (Scheme 3).
From these illustrations, we assumed that, whatever the
considered cyclic alcohol, the benzylic oxygen nucleophilic
would approach from the less-hindered side attack of cyclic
alcohols.
Thus, in this hexa-coordinated complex, the activation of
aliphatic alcohols is promoted by the Lewis acid coordi-
nation with the hydroxyl group, enhancing consequently,
Scheme 2.

1
920
E. M. Keramane et al. / Tetrahedron 57 (2001) 1917±1921
Scheme 3.
Scheme 4.
Only this nucleophilic attack by the hydroxyl group of the
racemic PeOH could, in such a case, explain the retention of
con®guration observed in the formed ethers.
using a Perkin Elmer 241 polarimeter with speci®c rotations
determined at 208C. All starting materials, optically active
alcohols and BiBr were commercially available (Aldrich).
3
Benzylation reactions were performed following the general
procedure previously described.
3
3
. Conclusion
4
.1.1. Compound I. d: 7.4 (m, 5H, C H ), 4.6 and 4.3 (q,
6 5
In summary, the catalytic activity of BiBr in the ®eld of
3
1H, C8-H)p, 3.4 (q, 1H, C1-H), 1.5 (d, 3H, C8-CH
1.6 (m, 10H, CH (2±6)), 1.1 (d, 3H, C1-CH ), 0.9 (t, 3H,
CH (7)). p represents diastereomeric pairs.
3
), 1.2±
O-benzylation reactions by PeOH has been extended to
optically active and cyclic alcohols. BiBr proved to be
2
3
3
3
capable of realizing an outstanding catalytic process. To
explain our results, we stated the hypothesis that benzyl-
ation reaction of alcohols in the presence of BiBr should
involve the formation of an intermediate, possessing a
6-fold coordinated bismuth atom. This intermediate
4.1.2. Compound II. d: 7.4 (m, 5H, C
1H, C4-H)p, 3.3 (q, 1H, C1-H), 1.5 (d, 3H, C8-CH
2H, CH (2)), 1.1 (d, 3H, C1-CH ), 0.8 (t, 3H, CH
p represents diastereomeric pairs.
H
), 4.6 and 4.3 (q,
), 1.4 (m,
(3)).
6
5
3
3
2
3
3
complex would contain a bismuth±oxygen bond (ROH)
and two bismuth±benzene ring interactions (PeOH). The
breakdown of this intermediate would be generated by the
benzylic oxygen nucleophilic attack, yielding the corre-
sponding ether. Inasmuch as the existence of a structure of
4.1.3. Compound III. d: 7.4 (m, 5H, C H ), 4.6 and 4.3 (q,
6 5
1H, C7-H), 3.3 and 3.2 (quint., 1H, Jˆ4, 8 Hz, C1-H)p, 1.5
(d, 3H, C7-CH
(2±6)). p represents diastereomeric pairs.
), 1 (d, 3H, C2-CH ), 0.9±2 (m, 10H, CH
3
3
2
1
8
[EX ] type has been already mentioned, our proposed
6
mechanism would be reasonable.
4.1.4. Compound IV. d: 7.4 (m, 5H, C
1
H
), 4.6 and 4.3 (q,
6
5
H, C7-H), 2.9 and 2.7 (dt, 1H, Jˆ4, 10 Hz, C1-H)p, 1.5 (d,
3
(2±6)). p represents diastereomeric pairs.
H, C7-CH ), 1 (d, 3H, C2-CH ), 0.9±1.8 (m, 10H, CH
3
3
2
4
. Experimental
4.1. General
4.1.5. Compound V. d: 7.4 (m, 5H, C
6
H
5
), 4.6 and 4.3 (q,
1
H, C8-H)p, 3.2 and 3.0 (dt, 1H, Jˆ5, 12 Hz, C1-H)p,
2.3 (m, 2H, CH (6)), 1.7 (m, 1H, C7-H), 1.5±1.6 (m,
4H, CH (3 and 4)), 1.5 (d, 3H, C8-CH ), 1 (m, 3H,
C5-CH ), 0.9 (m, 6H, C7-(CH ). p represents dia-
stereomeric pairs.
1
H NMR was recorded in a Brucker Ace-250 instrument at
2
2
50 MHz in CDCl solutions. We used a numbering system
3
2
3
as presented in Scheme 4. GC analysis was performed with
a Delsi 330 apparatus. Optical rotations were measured
3
)
3 2

E. M. Keramane et al. / Tetrahedron 57 (2001) 1917±1921
1921
References
Postel, M. New J. Chem. 1997, 21, 495. (c) Habata, Y.;
Fujishiro, F.; Akabori, S. J. Chem. Soc., Perkin Trans. 1
1
A. J. J. Am. Chem. Soc. 1985, 107, 3397.
996, 953. (d) Steward, C. A.; Calabrese, J. C.; Arduengo,
1
2
3
4
5
. Boyer, B.; Keramane, E. M.; Montero, J. L.; Roque, J. P.
Synth. Commun. 1998, 28, 1737.
9
. (a) Hancock, R. D.; Cukrowski, I.; Baloyi, J.; Mashishi, J.
J. Chem. Soc., Dalton Trans. 1993, 2895. (b) Cukrowski, I.;
Marsicano, F.; Hancock, R. D.; Tshetlo, T. T.; Van Otterlo,
W. A. L. Polyhedron 1995, 14, 1661.
. Boyer, B.; Keramane, E. M.; Arpin, S.; Montero, J. L.; Roque,
J. P. Tetrahedron 1999, 55, 1971.
. Boyer, B.; Keramane, E. M.; Roque, J. P.; Pavia, A. A.
Tetrahedron Lett. 2000, 41, 2891.
. Keramane, E. M.; Boyer, B.; Roque, J.-P. Tetrahedron 2001,
1
0. Genge, A. R. J.; Levason, W.; Reid, G. Chem. Commun. 1998,
159.
1. (a) Komatsu, N.; Uda, M.; Suzuki, H. Synlett 1995, 984.
b) Murafuji, T.; Mutoh, T.; Satoh, K.; Tsunenari, K.
2
5
7, 1909±1916.
. (a) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533.
b) Pearson, R. G. J. Chem. Educ. 1987, 64, 561. (c) Repichet,
1
(
(
J. Organomet. 1995, 14, 3848.
2. (a) Breunig, H. J.; Rosler, R.; Lork, E. Angew. Chem. Int. Ed.
S.; Leroux, C.; Hernandez, P.; Dubac, J.; Desmurs, J. R. J. Org.
Chem. 1999, 64, 6479 (and references cited therein).
1
1998, 37, 3175. (b) Althaus, H.; Breunig, H. J.; Rosler, R.;
Lork, E. J. Organomet. 1999, 18, 328.
6
. (a) Reese, A. L.; McMartin, K.; Miller, D.; Wickham, P. P.
J. Org. Chem. 1973, 38, 764. (b) Nakamura, E.; Shimada, J.;
Kuwajima, J. Organometallics 1985, 4, 641. (c) Garrigues, B.;
Gonzaga, F.; Robert, H.; Dubac, J. J. Org. Chem. 1997, 62,
1
1
3. Suzuki, H.; Murafuji, T.; Azuma, N. J. Chem. Soc., Perkin
Trans. 1 1993, 1169.
4. (a) Franck, W.; Schneider, J.; Muller-Becker, S. J. Chem. Soc.,
Chem. Commun. 1998, 799 (and references cited therein).
4
880 (and references cited therein).
7
. (a) Briand, G. G.; Burford, N. Chem. Rev. 1999, 99, 2601 (and
references cited therein). (b) Lange, K. C. H., Klapotke, T. M.
In The Chemistry of Organic Arsenic, Antimony and Bismuth
Compounds; Patai, S., Ed.; Wiley, Academic Press: 1994; pp.
(b) Schier, A.; Wallis, J. M.; Muller, G.; Schmidbaur, H.
Angew. Chem. Int. Ed. 1986, 25, 757.
1
1
1
5. Komatsu, N.; Ishida, J. Y.; Suzuki, H. Tetrahedron Lett. 1997,
3
8, 7219.
6. Komatsu, N.; Uda, M.; Suzuki, H.; Takahashi, T.; Domae, T.;
Wada, M. Tetrahedron Lett. 1997, 38, 7215.
315±366. (c) Guo, Z.; Sadler, P. J. Angew. Chem. Int. Ed.
1
999, 38, 1512 (d) Di Vaira, M.; Mani, F.; Stoppioni, P. Di
Vaira. Eur. J. Inorg. Chem. 1999, 833. (e) Carmalt, C. J.;
Clegg, W.; Elsegood, M. R. J.; Errington, R. J.; Havelock,
J.; Lightfoot, P.; Norman, N. C.; Scott, A. J. Inorg. Chem.
7. (a) Arnauld, T.; Barton, D. H. R.; Normant, J. F.; Doris, E.
J. Org. Chem. 1999, 64, 6915 (and references cited therein).
(b) Zevaco, T.; Postel, M.; Benali-Cherif, N. Main Group
Chem. 1992, 15, 217.
1996, 35, 3709 (and references cited therein). (f) Suzuki, H.;
Ikegami, T.; Matano, Y. Synthesis 1997, 249.
1
8. Fisher, G. A.; Norman, N. C. In Advances in Inorganic
Chemistry; Sykes, A. G., Ed.; Academic Press: New York,
8
. (a) Ogama, T.; Yoshikawa, A.; Wada, H.; Ogawa, C.; Uno, N.;
Suzuki, H. J. Chem. Soc., Chem. Commun. 1995, 1407 (and
references cited therein). (b) Kallal, K.; Coin, C.; Dunach, E.;
1
994; Vol. 41, pp 248.