Alternative approach to the free radical bromination of oligopyridine benzylic-methyl group

Article abstract of DOI:10.1016/S0040-4039(02)00127-2

We have undertaken a study on the influence of three solvents on the photobromination of two picolines. It shows that dichloromethane and benzene are better solvents than the classical carbon tetrachloride. The obtention of a good free radical bromination is described using aqueous biphasic media.

Full text of DOI:10.1016/S0040-4039(02)00127-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 1697–1700
Alternative approach to the free radical bromination of
oligopyridine benzylic-methyl group
S e´ bastien Bedel, Gilles Ulrich* and Claude Picard
Laboratoire de Synth e` se et Physico-chimie de Mol e´ cules d’Int e´ r eˆ t Biologique, UMR 5068, Universit e´ Paul Sabatier,
18 rte de Narbonne, 31062 Cedex 4, Toulouse, France
1
Received 22 November 2001; revised 16 January 2002; accepted 17 January 2002
Abstract—We have undertaken a study on the influence of three solvents on the photobromination of two picolines. It shows that
dichloromethane and benzene are better solvents than the classical carbon tetrachloride. The obtention of a good free radical
bromination is described using aqueous biphasic media. © 2002 Elsevier Science Ltd. All rights reserved.
1
8,9
In our studies on oligopyridine based ligands suitable
V o¨ glte and co-worker in the early 1980s. Their results
for the encapsulation of lanthanide(III) salts, designed
for time resolved fluorescence, such as podands and
cryptands, we faced one major synthetic problem: the
obtention in fair yields of benzylic bromide derivatives.
These functions appeared to us as the most appropriate
for an easy grafting of these oligopyridines on various
preorganized molecules via simple nucleophilic substi-
tution. Indeed, most of the syntheses described to
obtain bipyridines based podands or cryptands used
on the bromination of toluene type model substrates
using N-bromosuccinimide (NBS) and several solvents,
pointed out the strong influence of the solvent on the
selectivity of the reaction. Moreover, they correlated
2,3
4
1
0
their results to the reactivity–selectivity principle.
A
good compromise has to be found to preserve a suffi-
cient reactivity in order to brominate the benzylic
methyl, and a high selectivity to avoid multiple bromi-
nation. They have shown that the classical condition
5
benzylic bromides precursors.
for free radical bromination (NBS/CCl ) of a benzylic
4
methyl is not optimum, due to a high reactivity, induc-
ing a low selectivity.
The popular methods to obtain benzylic bromides are
either a direct bromination of the benzylic methyl, or a
deoxybromination of benzylic alcohol. The latter route
6
To apply V o¨ glte results to oligopyridine derivatives, we
selected two models of picolines bearing, respectively, a
benzylic methyl on the meta and ortho position of the
nitrogen: 2-bromo-5-methylpyridine (1) and 2-bromo-6-
methylpyridine (3). We present free radical bromination
results of 1 and 3 under several experimental conditions
implies a multistep synthesis which is time consuming.
The two-step bromination method, via the generation7
of the methyl anion, recently developed by Fraser
appears attractive. However, this method is incompat-
ible with the presence of nucleophilic sensitive function
1
on the molecule, such as an ester group. The softer and
simplest route remains the free radical bromination.
The disadvantage of this direct bromination is con-
comitant formation of dibromomethyl derivatives,
which reduces the yield of the desired bromomethyl and
complicates the purification process. Moreover, it
becomes a real synthetic problem when more than one
benzylic methyl on the molecule has to be brominated.
Br
Br
[
Br ]
Br
Br
1
3
N
Br
N
N
Br
1
2a
2b
[
Br ]
To improve the free radical bromination of pyridine
benzylic groups, we were inspired by the works of
Br
N
Br
N
Br
N
Br
Br
Br
3
4a
4b
*
Corresponding author. Tel.: +33 561556288; fax: +33 561556011;
e-mail: ulrich@chimie.ups-tlse.fr
Scheme 1.
0
040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00127-2

1
698
S. Bedel et al. / Tetrahedron Letters 43 (2002) 1697–1700
(
Scheme 1). All the experiments were carried out using
solvents, but a large amount of dibromomethyl was
produced (Table 1). For the bromination of 1, the use
of benzene appears to be the most appropriate method
in order to obtain 2a with a minimum of 2b produced.
Due to the toxicity of carbon tetrachloride and ben-
zene, dichloromethane can probably be considered as
the best choice. The same results were observed in the
bromination of 3 (Table 2). The yields were slightly
lower than for 1, probably due to an interaction of the
nitrogen lone pair near the methyl groups with the free
radical bromination process.
the same simple procedure and using easy to handle
materials: a standard 150W halogen lamp. This lamp
was placed at 5 cm from the reactor and its heat was
used to reflux the solvent. In this study, we compare the
yields of compounds obtained by reacting one equiva-
lent of picoline and one equivalent of bromine donor.
The yields of the compounds produced after the reac-
1
†
tion were determined by H NMR.
(
(
1)
2)
Br
R
+
RH
R
+ HBr
The low yields of production of dibromomethyl com-
pounds in the case of benzene in opposition to CCl4,
both having similar boiling points, indicate different
reactivity of the hydrogen abstracting species. In pres-
ence of benzene, a bromine radical/arene p-complex is
probably formed, similar to the complex observed with
+ Br2
RBr + Br
O
O
N
Br + HBr
H
+ Br2
N
(3)
13,14
chlorine.
This complex is probably characterized by
O
O
a different selectivity for the abstraction of a benzylic
hydrogen (Eq. (1)). The good results obtained with
dichloromethane are probably due to the low tempera-
ture of the reaction.
(
4)
R
+
HBr
Br
+
RH
hν
Br Br
2 Br
(5)
Water was described as an excellent medium for free
15
radical reaction, due to its remarkable non-reactivity
towards radicals (OH bond resistancy to homolytic
breaking). Unfortunately, the solubility of most organic
compounds is limited in water. Very good results for
the photobromination of toluene derivatives with
We first carried out experiments varying the solvent
with the most used bromine donor: N-bromosuccin-
imide, using AIBN as initiator. The small amount of
Br then produced initiates the main propagation steps
Eqs. (1) and (2)). NBS has a double role: it provides a
source of Br , in a low, steady state concentration, and
consumes the liberated HBr by a ionic process (Eq.
’
(
1
5
bromine in water were observed. In fact, toluene acts
as its own solvent and consequently the reaction is
2
11
(
3)). The consumption of HBr avoids the inhibition of
12
bromination (Eq. (4)).
Table 1. Bromination of 1 with NBS
d
d
d
In order to choose the solvents for the bromination
studies we tested some candidates described in the
literature, and we selected two of them to be compared
with carbon tetrachloride: dichloromethane and ben-
zene. Unfortunately, bromination in methyl formate
described as the solvent inducing the highest ‘selectiv-
ity’, in the case of toluene family, was non-effective
with picolines, due to a probable too low reactivity.
Solvent
CCl
4
CH
2
Cl
2
C H
6 6
a
9
1 (%)
12
66
22
19
69
12
84
24
67
9
77
0.13
a
2
2
a (%)
b (%)
a
b
Br cons. (%)
100
c
Ratio
0.34
0.17
a
Relative yields calculated from 1H NMR spectra.
Consumption of bromine corrected by the starting quantities of
bromine donor.
b
‡
Under these conditions, CCl appeared again to be the
4
c
‘
highest reactivity’ solvent. After 3 h of irradiation, all
Ratio of 2b over 2a.
d
Method A: average value for a minimum of three experiments.
the bromine was consumed. This reactivity seemed to
be in opposition to ‘selectivity’. The yield of bro-
8
momethyl derivatives was similar to the concurrent
Table 2. Bromination of 3 with NBS
†
Product analysis. In the 1H NMR spectrum, the benzylic protons
CCl d
4
d
d
6
Solvent
CH Cl
C H
6
2
2
show up: for the methyl group at 2.28 for 1 and 2.52 ppm for 3; for
the bromomethyl group at 4.44 for 2a and 4.49 ppm for 4a; for the
dibromomethyl group at 6.56 for 2b and 6.57 ppm for 4b. The
values obtained in Table 1, were confirmed by HPLC analysis.
a
3 (%)
20
58
22
93
26
64
10
77
26
62
12
78
a
4a (%)
4b (%)
a
‡
b
Method A: A mixture of bromo-methyl-pyridine (0.1 g, 0.58 mmol),
Br cons. (%)
c
NBS (0.115 g, 0.64 mmol) and a catalytic amount of AIBN in a
pure organic solvent (10 mL) or in a biphasic medium (organic
solvent/water, 10/10 mL) was lightened and refluxed using a halogen
lamp (150W) for 3 h. The resulting organic layer was filtered
through a short alumina (Act IV) column, and evaporated under
reduced pressure. 1H NMR analysis of the crude product gave the
relative percentages.
Ratio
0.38
0.16
0.19
a
Relative yields calculated from 1H NMR spectra.
Consumption of bromine corrected by the starting quantities of
bromine donor.
Ratio of 4b over 4a.
Method A: average value for a minimum of three experiments.
b
c
d

S. Bedel et al. / Tetrahedron Letters 43 (2002) 1697–1700
1699
operated in a biphasic medium. We decided to study
the photobromination of our picolines in biphasic mix-
funnel with a diluted organic solution of bromine could
be a good alternative to NBS.
tures using NBS or Br as bromine donors and our
three reference solvents.
2
The study of the free radical bromination of 3 with
NBS in the biphasic mixture shows almost similar
yields as the monophasic case (Table 4). The yield
differences between the bromination of 1 and 3 are
enhanced with the use of bromine. Under these condi-
tions, the presence of the nitrogen lone pair seems to
have a significant action on the free radical process.
These differences in the photobromination of methyl
groups in the a or b position to the nitrogen was
noticed when we applied these results to a more compli-
The results obtained in biphasic media, in the case of 1,
using NBS as bromine donors gave almost similar
yields of 2a as in absence of water, but the production
of 2b was lowered (Table 3). When bromine is used, the
§
reaction is really convenient: completed in only 30 min,
no succinimide residues have to be eliminated and less
dibromomethyl derivative is produced. In this case, the
reaction is initiated by irradiation (Eq. (5)). The propa-
gation steps take place in the organic phase and water
seems to be a very efficient scavenger of the coproduct
HBr (Eq. (1)). The abstraction of HBr prevents the
reverse reaction (Eq. (4)) and thus accelerates the reac-
tion. In the absence of water, we have observed few
bromination compounds even after more than 2 days
irradiation. When NBS is used, the abstraction of HBr
produced by the aqueous phase seems not to interact
with the ionic processes of the generation of bromine
1
cated oligopyridine. The free radical bromination of
ethyl 5-methyl-2,2%-bipyrine-6-carboxylate to its mono-
bromoethyl derivatives could be raised from 56 (NBS,
benzene) to 73% (Br , Benzene/H O), but not with the
2
2
a-methyl i1somer ethyl 6-methyl-2,2%-bipyridine-6-
carboxylate.
We can conclude from this study that the photobromi-
nation of picoline benzylic-methyl to obtain the corre-
sponding bromomethyl derivative is more efficient in
dichloromethane or benzene than in carbone tetra-
chloride. For environmental and toxical reasons the
best solvent is probably dichloromethane. The use of an
(
Eq. (3)). This process probably takes place in water or
at the interface. In fact the only difference is the slow
liberation of Br in the case of NBS. If large scale
2
quantities have to be brominated, the use of a dropping
Table 3. Bromination of 1 in biphasic media
Solvent
CCl /H O
CH Cl /H O
C H /H O
6 6 2
4
2
2
2
2
Donor
Br2d
21
62
17
96
NBSe
13
69
18
96
Br2d
25
67
8
83
NBSe
13
72
15
92
Br2d
27
66
7
79
NBSe
29
66
5
79
1
2
2
(%)a
a (%)
b (%)
a
a
b
Br cons. (%)
Ratioc
0.28
0.26
0.12
0.20
0.10
0.08
a
Relative yields calculated from 1H NMR spectra.
Consumption of bromine corrected by the starting quantities of bromine donor.
Ratio of 2b over 2a.
Method B: average value for a minimum of three experiments.
Method A: average value for a minimum of three experiments.
b
c
d
e
Table 4. Bromination of 3 in biphasic media
Solvent
CCl /H O
CH Cl /H O
C H /H O
6 6 2
4
2
2
2
2
Donor
Br2c
30
50
20
90
NBSd
23
57
20
89
Br2c
34
58
7
72
0.12
NBSd
19
67
14
86
Br2c
37
54
9
72
NBSd
30
59
11
75
3
4
4
(%)a
a (%)
b (%)
a
a
a
Br cons. (%)
b
Ratio
0.40
0.36
0.21
0.16
0.19
a
Calculated on the basis of the starting quantities of bromine.
Ratio of 4b over 4a.
Method B: average value for a minimum of three experiments.
Method A: average value for a minimum of three experiments.
b
c
d
§
Method B: To a refluxing mixture of bromo-methyl-pyridine (0.1 g, 0.58 mmol) in a biphasic medium (organic solvent/water, 10/10 mL) was
added Br2 (30 mL, 0.58 mmol). The mixture was then lightened and refluxed using a halogen lamp (150W) during 30 min. The solution was
neutralized with a NaHCO3 aqueous solution. The aqueous layer was extracted with CH Cl , the combined organic extracts were dried with
2
2
MgSO4 and evaporated in vacuo. NMR analysis of the crude product gave the relative yields.

1
700
S. Bedel et al. / Tetrahedron Letters 43 (2002) 1697–1700
aqueous biphasic mixture of solvent and bromine has
been shown to be a convenient synthetic procedure for
the free-radical bromination of selected compounds.
100, 3553–3590.
6. Funeriu, D. P.; Lehn, J.-M.; Baum, G.; Fenske, D.
Chem. Eur. J. 1997, 3, 99–104.
. Savage, S. A.; Smith, A. P.; Fraser, C. L. J. Org.
7
Further applications of the results of this study on
various oligopyridines are in progress.
Chem. 1998, 63, 10048–10051.
8. Offermann, W.; V o¨ gtle, F. Synthesis 1977, 272–273.
. Offermann, W.; V o¨ gtle, F. Angew. Chem., Int. Ed.
9
Engl. 1980, 19, 464–465.
References
10. Giese, B. Angew. Chem., Int. Ed. Engl. 1977, 16, 125–
36.
1
1
2
3
. Ulrich, G.; Bedel, S.; Picard, C.; Tisn e` s, P. Tetrahedron
Lett. 2001, 42, 6113–6115.
. Charbonni e` re, L. J.; Weibel, N.; Ziessel, R. F. Tetra-
hedron Lett. 2001, 42, 659–662.
. Charbonni e` re, L. J.; Ziessel, R.; Guardigli, M.; Roda,
11. Adam, J.; Gosselain, P. A.; Goldfinger, P. Nature 1953,
171, 704–705.
12. Walling, C.; Rieger, A. L.; Tanner, D. D. J. Am.
Chem. Soc. 1963, 85, 3129–3136.
13. Raner, K. D.; Lusztyk, J.; Ingold, K. U. J. Am. Chem.
A.; Sabbatini, N.; Cesario, M. J. Am. Chem. Soc. 2001,
Soc. 1989, 111, 3652–3658.
1
23, 2436–2437.
14. Ingold, K. U.; Lusztyk, J.; Raner, K. D. Acc. Chem.
Res. 1990, 23, 219–225.
15. Shaw, H.; Perlmutter, H. D.; Gu, C. J. Org. Chem.
4
5
. Galaup, C.; Couchet, J. M.; Picard, C.; Tisn e` s, P. Tet-
rahedron Lett. 2001, 42, 6275–6278.
. Kaes, C.; Katz, A.; Hosseini, M. W. Chem. Rev. 2000,
1997, 62, 236–237.