Highly efficient and selective electrophilic and free radical catalytic bromination reactions of simple aromatic compounds in the presence of reusable zeolites

Article abstract of DOI:10.1039/b002157l

Reactions of mono-substituted aromatics of moderate activity with bromine in the presence of stoichiometric amounts of zeolite NaY proceed in high yield and with high selectivity to the corresponding para-bromo products. The zeolites can easily be regenerated by heating and reused. Similar para-selectivity can be achieved in the case of toluene by use of tert-butyl hypobromite as reagent, zeolite HX as catalyst, and a solvent comprising a mixture of tetrachloromethane and diethyl ether. Radical bromination of ethyl 4-methylbenzoate using bromine in the presence of light is catalysed by various zeolites and affords a high yield of ethyl 4-(bromomethyl)benzoate but with no great improvement in selectivity for monobromination.

Full text of DOI:10.1039/b002157l

View Article Online / Journal Homepage / Table of Contents for this issue
Highly efficient and selective electrophilic and free radical catalytic
bromination reactions of simple aromatic compounds in the presence
of reusable zeolites
1
Keith Smith,* Gamal A. El-Hiti,† Mark E. W. Hammond, Dawoud Bahzad, Zhaoqiang Li and
Christophe Siquet
Centre for Clean Chemistry, Department of Chemistry, University of Wales Swansea,‡
Singleton Park, Swansea, UK SA2 8PP
Received (in Cambridge, UK) 16th March 2000, Accepted 13th June 2000
Published on the Web 19th July 2000
Reactions of mono-substituted aromatics of moderate activity with bromine in the presence of stoichiometric
amounts of zeolite NaY proceed in high yield and with high selectivity to the corresponding para-bromo products.
The zeolites can easily be regenerated by heating and reused. Similar para-selectivity can be achieved in the case of
toluene by use of tert-butyl hypobromite as reagent, zeolite HX as catalyst, and a solvent comprising a mixture of
tetrachloromethane and diethyl ether. Radical bromination of ethyl 4-methylbenzoate using bromine in the presence
of light is catalysed by various zeolites and affords a high yield of ethyl 4-(bromomethyl)benzoate but with no great
improvement in selectivity for monobromination.
significantly poorer selectivity. These results encouraged us to
Introduction
undertake a detailed study of zeolite catalysed bromination
In recent years considerable emphasis has been placed on
reactions in an attempt to develop highly selective bromination
improvement in the environmental impact of industrial
reactions.
chemical processes.1 It is well recognised that solids can play a
In a preliminary communication we reported that bromi-
significant role in the development of cleaner technologies
nation of aromatic compounds was highly efficient and para-
through their abilities to act as catalysts, support reagents,
selective when bromine was used as a brominating reagent
entrain by-products, avoid aqueous work-ups and influence
in the presence of a zeolite.12 We now also report the full details
product selectivities, and several books on the applications of
of this work, as well as results of free radical bromination
solids in organic synthesis have appeared.2 Advances are
reactions of ethyl 4-methylbenzoate using bromine in the
particularly needed in the area of electrophilic aromatic sub-
presence of zeolites and light.
stitutions,3 where traditional Lewis acid catalysts are a cause
of considerable concern and where reactions are frequently
unselective. Bromination, which is of considerable industrial
importance, is a case in point.
Results and discussion
para-Selective bromination using tert-butyl hypobromite
We have previously developed a highly para-selective method
for chlorination of toluene and other simple aromatic com-
pounds based on tert-butyl hypochlorite and a shape-selective
zeolite catalyst.4,5 Here we report the attempted adaptation
of the method, involving tert-butyl hypobromite.6 Although the
method is successful for toluene, the reagent is not very stable
and the reaction is not suitable for large scale application or
application to other substrates.
We have also previously reported a selective bromination
method for activated heterocycles based on N-bromo-
succinimide and silica.7 However, this method is not successful
for bromination of simple aromatic compounds of moderate
activity. We have therefore sought to develop an alternative
method for para-selective bromination of such compounds.
Early attempts to use zeolites to gain regioselective control
over bromination of toluene had some success.8 Limited
improvements in selectivity of bromination reactions have also
been obtained using clay supported reagents.9 Recently, how-
ever, Sasson and co-workers achieved significant improvements
by use of zeolite Y as a catalyst and an epoxide as a hydrogen
bromide scavenger.10,11 They were able to obtain excellent para-
selectivity (98%) in low yield (ca. 10%) or higher yields with
Following our previous success in developing a para-selective
chlorination reaction for aromatic substrates of moderate
reactivity by use of tert-butyl hypochlorite as reagent
and zeolite HX as catalyst,4,5 we first attempted the analogous
reaction with tert-butyl hypobromite as reagent and toluene
as substrate. The reagent was prepared as a solution in tetra-
chloromethane, dichloromethane or petroleum spirit by a
slightly modified version of a literature procedure.13 These
solutions, sometimes with diethyl ether added because of the
improved para-selectivity found with ether as co-solvent in
chlorination reactions,4,5 were allowed to react [eqn. (1)] with
(1)
an equimolar quantity of toluene in the presence of partially
proton exchanged zeolite NaX (13X molecular sieve). The
results are shown in Table 1.
From the results in Table 1, it is clear that the reaction is
successful in achieving para-selective bromination of toluene.
The yields were high, but not quantitative. The reactions would
† Permanent address; Department of Chemistry, Faculty of Science,
Tanta University, Tanta, Egypt.
‡ Formerly known as the University College of Swansea.
DOI: 10.1039/b002157l
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752
This journal is © The Royal Society of Chemistry 2000
2745

View Article Online
Table 1 Yields and selectivities in HNaX-catalysed reactions accord-
Table 3 Yields and selectivities according to eqn. (1) with different
ing to eqn. (1)a
catalystsa
Yield
(%)b
Isomer proportions
Yield
(%)b
Isomer proportions
Solvent
(2:3)c
Catalyst
(2:3)c
CH2Cl2
20% Et2O, 80% CH2Cl2
CCl4
20% Et2O, 80% CCl4
Petroleum spirit (30–40 ЊC)
20% Et2O, 80% petroleum spirit
a ButOBr (2.6 mmol), toluene (2.5 mmol), solvent (10 ml), zeolite
HNaX (1.5 g), 20 ЊC, 1 h. b Absolute yield of monobromotoluenes by
quantitative GC. No other toluene-derived products were observed
except unreacted toluene. c By GC.
85
77
92
78
81
73
29:71
6:94
24:76
2:98
19:81
3:97
None
Synclyst (13% Al2O3)
K10
MeSO3H
HNaX
HNaY
0
83
93
92
49
77
62
4
—
54:46
56:44
61:39
0:100
36:64
54:46
30:70
HMordenite
H-ZSM-5
a Catalyst (1.5 g), ButOBr (2.6 mmol), toluene (2.5 mmol), Et2O (7.5
ml), CCl4 (2.5 ml), 20 ЊC, 1 h. b,c As footnotes to Table 1.
Table 2 Yields and selectivities in HNaX-catalysed reactions accord-
to allow the shape-selective constraints of the zeolite pores to
exert maximum effect.
ing to eqn. (1)a
The low yield over HNaX probably results because the
rate of reaction is slow enough to allow effective competition
from destruction of the hypobromite in ether. Indeed, when
the reaction over HNaX was tried with less active substrates
(benzene, ethylbenzene, cumene, tert-butylbenzene and bromo-
benzene), destruction of the hypobromite seemed to become
the principle reaction, so that yields of product were very
low (0–36%). Consequently, although somewhat better yields
could probably be achieved by use of excess reagent, it seemed
unlikely that the reaction would be useful as a general method.
Therefore, our attention turned to the use of bromine as the
reagent.
% Ether in Et2O–CCl4
solvent
Yield
(%)b
Isomer proportions
(2:3)c
0
1
2.5
5
10
15
20
25
50
75
92
63
72
75
76
71
70
64
59
49
24:76
14:86
8:92
5:95
4:96
3:97
3:97
2:98
1:99
0:100
a ButOBr (2.6 mmol), toluene (2.5 mmol), solvent (10 ml), zeolite
HNaX (1.5 g), 20 ЊC, 1 h. b Absolute yield of monobromotoluenes by
quantitative GC. No other toluene-derived products were observed
except unreacted toluene. c By GC.
para-Selective bromination using bromine
Initial reactions were conducted on toluene as substrate and
with zeolite HX present as catalyst, but with bromine as reagent
[eqn. (2)]. The reaction vessel and the sample collected for
analysis were protected from light, to inhibit radical side-chain
bromination. Initially, several solvents were tested in order to
allow choice of the best solvent for the reaction. The results
obtained suggested that dichloromethane was the best solvent
of those tried for selective production of p-bromotoluene (3), at
least for reaction in the presence of HX zeolite as catalyst.
require either more time or excess tert-butyl hypobromite in
order to go to completion. For all solvents tried, selectivities
improved when diethyl ether was used as co-solvent. How-
ever, invariably the yield also suffered. In an attempt to
find the optimum proportion of ether in the solvent mixture, a
series of reactions was run in diethyl ether–tetrachloromethane
mixtures of different proportions. The results are shown in
Table 2.
(2)
The results in Table 2 demonstrate that the reaction can be
totally para-selective, though in modest yield, by use of a
solvent mixture consisting of a 3:1 volume ratio of diethyl
ether and tetrachloromethane. Leaving this reaction mixture for
a longer time did not produce a dramatic increase in yield. We
surmise that the low yield is a result of competitive destruction
of tert-butyl hypobromite in the ether-containing reaction
mixture.
For comparison, reactions were also conducted in the
absence of any catalyst, with an amorphous solid catalyst
(synclyst), with an acid-treated clay (K10), with a solution
phase acid (methanesulfonic acid) and with other zeolites
(Y, Mordenite and ZSM-5) instead of HNaX. The results are
shown in Table 3.
We next investigated the use of various large-pore, proton-
form zeolites as catalysts in the reaction of bromine with
toluene. Reactions were conducted with a small excess of
bromine (1.1 equiv.) at room temperature in dichloromethane
as solvent and in the presence of various zeolites. The reaction
mixture was stirred and monitored after 1–5 h. The yields
of o-bromotoluene (2) and p-bromotoluene (3) obtained are
shown in Table 4.
The results in Table 3 demonstrate that an acidic catalyst is
required, and that high yields are achievable with catalysts
where diffusion is not a problem (synclyst, K10, MeSO3H),
but that with such catalysts there is no para-selectivity. When
diffusion is a significant problem, as for the medium pore
zeolite H-ZSM-5, the reaction is slow and the yield is corre-
spondingly low. Mordenite and zeolite Y, which both have large
pores, give reasonable yields but with low para-selectivity. No
other catalyst matches the para-selectivity of HNaX, which
has large pores to allow reasonable diffusion but weakly acidic
sites. This appears to favour a reaction that is sufficiently mild
The results showed that all zeolites tried were capable of
catalysing the reaction to give high yields with fair para-
selectivities, typically in the region of 80%. The reaction was
fastest when HY was used but the para-selectivity was some-
what higher when Hβ was the catalyst. The reaction over
HY gave 68% yield of p-bromotoluene (3) within 1 h, which
increased to 79% after 3 h. Zeolite Hβ and H-Mordenite gave 3
in yields of 79% and 83%, respectively, after 5 h. This indicated
that bromination reactions of toluene over these catalysts were
slower but still effective. Clearly, if a proton-form catalyst were
to be used for bromination of toluene, then it would be best to
2746
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752

View Article Online
Table 4 Yields according to eqn. (2), over proton-form zeolitesa
Table 6 Yields according to eqn. (2) over different amounts of NaY
zeolitea
Yields (%)b
Reaction
Yields (%)b
Catalyst
time/h
1
2
3
Quantity of
NaY/g
Reaction
time/h
1
2
3
HY
HY
HY
HY
HX
HX
HX
HX
Hβ
Hβ
Hβ
Hβ
HMord
HMord
HMord
HMord
1
2
3
5
1
2
3
5
1
2
3
5
1
2
3
5
16
6
4
13
14
15
15
8
16
20
20
6
7
8
9
9
68
75
79
82
30
47
57
61
50
64
71
79
35
60
72
83
0.05
0.05
0.05
0.05
0.05
0.10
0.10
0.10
0.10
0.10
0.25
0.25
0.25
0.25
0.25
0.50
0.50
0.50
0.50
0.55
0.55
0.55
0.55
0.55
1
2
3
5
24
1
2
3
5
24
1
2
3
5
24
1
2
3
5
1
75
73
72
70
50
44
38
32
22
—
14
11
11
11
10
5
5
5
5
—
—
—
—
—
1
1
1
1
4
4
6
8
10
15
2
2
2
2
3
1
1
1
1
1
1
20
21
22
24
40
51
55
60
67
84
82
82
82
84
85
93
93
93
94
98
98
98
98
98
2
60
35
21
19
43
29
21
12
51
24
12
1
12
13
14
a Toluene (0.84 mmol), bromine (0.93 mmol), zeolite (0.5 g), in CH2Cl2
(6.6 ml), 20 ЊC. b Absolute yields determined by quantitative GC.
Table 5 Yields according to eqn. (2), over Na-form zeolitesa
2
3
5
24
1
1
1
Yields (%)b
Reaction
a Toluene (0.84 mmol), bromine (0.93 mmol), zeolite NaY, in CH2Cl2,
Catalyst
time/h
1
2
3
20 ЊC. b Absolute yields determined by quantitative GC.
NaY
NaY
NaY
NaY
NaX
NaX
NaX
NaX
Naβ
Naβ
Naβ
Naβ
NaMord
NaMord
NaMord
NaMord
NaMord
NaA
1
2
3
5
1
2
3
5
1
2
3
5
1
2
3
5
24
1
2
3
5
1
2
3
5
5
5
5
5
8
7
6
6
1
1
1
1
1
1
1
1
5
9
10
10
3
6
7
93
93
93
94
91
92
93
93
42
77
83
86
20
39
51
71
86
1
a small pore and a medium pore zeolite. The results obtained
are recorded in Table 5.
It is clear from Table 5 that the yields of brominated toluenes
2 and 3 formed were broadly a function of the size of the pores.
Thus, almost no reaction took place when the small pore zeolite
A was used. Reaction took place to a modest extent with the
medium pore Na-ZSM-5 and also with the larger pore Morden-
ite, which has linear channels. The other large pore zeolites,
which have three dimensional pore systems, catalysed the reac-
tion to a greater extent. This was particularly true of the X and
Y zeolites, where reaction was almost complete, even after 1 h.
Zeolite Naβ gave a better yield of p-bromotoluene (3), with
higher selectivity, than zeolite Hβ. To our surprise, NaY and
NaX produced high yields of almost entirely 4-bromotoluene
(3), in contrast to the results of Sasson.10,11 Furthermore, the
reactions using NaY and NaX were fast, giving high yields of
bromotoluenes by the time the reactions were first monitored
after 1 h at 20 ЊC. The rates of reactions with the other zeolites
were lower, giving lower yields of brominated products even
after 5 h. Nevertheless, the yield with Naβ was 86% after 5 h
and that with Na-Mordenite crept up to 86% after 24 h.
The relative reaction rates can again be understood in terms
of the pore structures of the zeolites. The large cavities created
by the interconnecting channels of NaY and NaX, and to a
lesser extent Naβ, allow readier diffusion of the reactants to the
active sites. By contrast, NaA fails to react because the pores
are too small for toluene to enter. However, it was not immedi-
ately obvious why NaY and NaX should induce reactions at
faster rates than their proton counterparts, nor why the yields
and selectivities should be so much higher than those reported
by Sasson.10,11
52
13
7
4
75
54
40
18
3
98
95
94
93
75
60
58
56
9
11
—
—
—
—
4
5
6
6
NaA
NaA
NaA
NaZSM5
NaZSM5
NaZSM5
NaZSM5
1
2
2
20
32
32
35
a Toluene (0.84 mmol), bromine (0.93 mmol), zeolite (0.5 g), in CH2Cl2
(6.6 ml), 20 ЊC. b Absolute yields determined by quantitative GC.
leave the reaction longer than one hour in order to obtain a
high yield.
The results with HY and HX, which have the same gross
crystalline structure, showed that zeolite Y gave a higher yield
and better selectivity than zeolite X. This is presumably due to
differences in the number and strengths of the acid sites and
therefore in the overall acidity/basicity of the solid. Therefore,
further experiments were carried out using Na-exchanged
zeolites in order to investigate further the effect of the basicity
of the zeolite used.
Some of Sasson’s best selectivities were achieved using NaY
zeolite10,11 so we next investigated the sodium forms of a range
of zeolites. Reactions were conducted in the same way as those
using proton-form zeolites. The sodium forms of zeolite A and
ZSM-5 were also included, in order to provide comparison with
The major difference between our experiments and corres-
ponding ones carried out by Sasson was in the quantity of
zeolite used. We therefore investigated the effects of variation in
the amount of zeolite NaY used. The results obtained are
recorded in Table 6.
As the results in Table 6 show, there is a strong dependency of
the yield on the quantity of NaY used and the reaction became
quantitative when 0.55 g was employed. The reaction was
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752
2747

View Article Online
Table 7 Yields according to eqn. (2) over different amounts of NaX
slower and produced a low yield when the quantity of the
catalyst was only 0.05 g, broadly in line with the observations of
Sasson.10,11 These results are consistent with the occurrence of a
rapid and stoichiometric reaction involving the NaY zeolite
[eqn. (3)] competing favourably with a slower and less selective
zeolitea
Yields (%)b
Quantity of
NaX/g
Reaction
time/h
1
2
3
(3)
0.10
0.10
0.10
0.10
0.10
0.20
0.20
0.20
0.20
0.20
0.50
0.50
0.50
0.50
0.50
0.55
0.55
0.55
0.55
0.55
1
2
3
5
24
1
2
3
5
24
1
2
3
5
24
1
2
3
5
24
68
64
63
61
47
41
36
32
26
12
8
7
6
6
6
4
1
1
1
1
2
2
2
4
2
4
5
6
8
1
1
1
1
1
1
1
1
1
1
30
33
33
36
49
54
59
62
65
79
91
92
93
93
93
93
96
96
96
97
catalytic reaction involving the HY zeolite (compare Table 4).
When such trends were seen before, the results were inter-
preted in terms of a catalytic process that was damaged by the
presence of HBr.8,10,11
Zeolite X has the same crystalline structure as zeolite Y but
the Si:Al ratio is lower (i.e. it contains a higher proportion of
Al). The number of anionic sites on the lattice is related to the
number of Al atoms and the number of exchangeable sodium
atoms in NaX or NaY is related to the number of anionic sites.
Therefore, NaX was used instead of NaY to see the effect of
changing the quantity of this catalyst. The results obtained are
recorded in Table 7.
—
a Toluene (0.84 mmol), bromine (0.93 mmol), zeolite NaX, in CH2Cl2
(6.6 ml), 20 ЊC. b Absolute yields determined by quantitative GC.
As can be seen from Table 7, the same effect was observed
when NaX was used as the catalyst as when NaY was used.
The yield of 3 depended on the quantity of NaX used in the
bromination reaction. The para-regioselectivity was high in all
cases. The only differences between NaX and NaY were in the
reaction rate and in the amount of product formed by the initial
reaction. Surprisingly, it was NaY rather than NaX that
resulted in the more rapid and higher yielding reaction, despite
the greater quantity of exchangeable sodium ions in the latter.
The explanation for this is not yet clear, but may be connected
with constraints to diffusion and with accessibility of sites
in the zeolite with the greater initial concentration of sodium
cations, especially as the quantity of NaBr deposited in the
pores builds up.
Although the best conditions allow the quantitative forma-
tion of 4-bromotoluene (3) with very high selectivity, it is
unattractive for large-scale use to require 3.8 times as much
NaY as the mass of product formed. This would be less of a
problem if the solid could be easily regenerated following
removal of the bromotoluene. It appears from eqn. (3) that the
spent solid should have finely dispersed sodium bromide in the
pores of HY zeolite. Simple heating of the solid might therefore
cause expulsion of gaseous hydrogen bromide and regeneration
of NaY [eqn. (4)]. Alternatively, a standard ion-exchange
Table 8 Yields according to eqn. (2), with different quantities of
solventa
Yields (%)b
CH2Cl2/ml
1
2
3
0c
2
3
—
—
—
5
20
1
1
79
95
97
93
5
1
a Toluene (0.84 mmol), bromine (0.93 mmol), zeolite NaX (0.5 g), in
CH2Cl2, 20 ЊC. b Absolute yields determined by quantitative GC.
c Toluene (5 ml) was used as its own solvent and calculation done
with respect to the amount of bromine added.
still converted with high selectivity into 4-bromotoluene (3)
when the amount of solvent was reduced to 2 ml. The para-
regioselectivity was much lower when toluene was used as its
own solvent, while with no solvent at all the mixture could not
be stirred. The solvent probably aids diffusion into the zeolite
pores by increasing the mobility of the system. However, in
an industrial situation it might be important economically to
avoid a solvent separation step. This would be possible if
4-bromotoluene could be utilised as the solvent. Therefore,
an experiment was conducted in 4-bromotoluene as solvent.
Indeed, 4-bromotoluene could be used successfully as a solvent
instead of dichloromethane, with no diminution in yield or
(4)
process involving reflux of the spent zeolite with aqueous NaBr
should effect its conversion back into its sodium form.
Calcination of the solid at 400 ЊC did indeed completely
regenerate its reactivity. The regenerated material gave the same
yield and selectivity as the original NaY. Therefore, in practice
it should become possible to produce much larger quantities of
4-bromotoluene from a particular batch of solid by repeated
regeneration and reuse.
The amount of solvent might also affect the yield and para-
selectivity of the reaction. Therefore, a series of experiments
was conducted in which different amounts of solvent were used.
The results obtained are recorded in Table 8.
selectivity. Consequently, this approach now becomes
a
commercially realistic possibility for the synthesis of 4-
bromotoluene (3).
In order to demonstrate that the reaction was applicable
on a preparative scale, a single reaction was conducted on a
76 mmol scale (using 7.0 g of toluene). The reaction produced
a product (12.5 g, corresponding to a 96% yield if all 3),
1
the H NMR spectrum of which indicated that it was about
96% pure. A single distillation without fractionation produced
a liquid forerun (0.27 g), which was mainly 3 together with
small quantities of solvent and other impurities, a solid main
fraction (10.35 g, 80% yield) that was essentially pure 3 (mp
26–28 ЊC), and a dark residue (1.35 g), which still contained
substantial quantities of 3. It is clear that with optimisation
of the reaction procedure for larger scale work and a more
The results showed that the quantity of dichloromethane
could be reduced without detriment until there was insufficient
to keep the reaction mobile. Decreasing the amount of solvent
made the reaction more concentrated but the toluene was
2748
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752

View Article Online
Table 9 Yields according to eqn. (5)a
sodium cation, and assistance for removal of the proton from
the aromatic ring coming from the zeolite framework. This
would lead directly to NaBr and HY as well as to the organic
product.
The alternative would involve initial reaction between the
zeolite and bromine to give a surface hypobromite species,
which would become the active brominating agent. An
aluminium site with Lewis-acid properties would be generated
simultaneously [eqn. (6)]. The surface hypobromite site, with
Yields (%)b
R of substituted
benzenes 4
4
5
6
H
Me
Et
i-Pr
t-Bu
F
Cl
Br
MeO
6
—
1
1
—
—
—
—
—
—
91c
98
97
—
—
—
2
98
97d
8 (6)e
92 (93)e
78 (82)e
56 (66)e
93
(6)
22 (17)e
44 (34)e
—
a Substrate (0.84 mmol), bromine (0.93 mmol), zeolite NaY (0.55 g), in
CH2Cl2 (ca. 8 ml), 20 ЊC, 5 h. b Absolute yields determined by quantita-
tive GC. c 2% of dibromo product also obtained. d ca. 1% of a side
product also obtained. e Figures in parentheses are for similar reactions,
but with 0.8 g of NaY.
NaBr deposited alongside it, might be rather inaccessible to the
more hindered ortho-position of the substrate. This would
enhance the tendency, already seen with tert-butyl hypobromite,
for para-preference in these reactions, thereby allowing the
extreme selectivity observed. This latter mechanism bears some
similarity to one proposed previously by Russian workers.14
sophisticated distillation process the reaction would produce a
very high isolated yield of extremely pure material, which is
impossible by any previous procedure.
Side-chain bromination
In order to explore the scope of the reaction, similar con-
ditions (small scale) were applied to a range of substituted
benzenes 4 according to eqn. (5). The yields of brominated
products 5 and 6 obtained are recorded in Table 9.
Having achieved selective nuclear bromination of substituted
benzenes, we next turned our attention to the possibility of
selective side-chain bromination. We chose ethyl 4-methyl-
benzoate (7) as substrate because the ester group would lower
reactivity towards nuclear substitution and because the
product, ethyl 4-(bromomethyl)benzoate (8), could be of com-
mercial interest.
(5)
Ethyl 4-(bromomethyl)benzoate can be obtained from the
reaction of 4-bromomethylbenzoyl chloride or bromide with
ethanol.15 Whilst such a method can be quite efficient, it suffers
from the poor availability and stability of acid halides, which
considerably limits its scope, while direct reaction between the
carboxylic acid and ethanol is prone to competitive displace-
ment of bromide by ethanol. An alternative approach involves
bromination of ethyl 4-methylbenzoate (7) using N-bromo-
succinimide (NBS),16,17 but NBS as a brominating reagent is
relatively expensive compared to bromine. The NBS reaction
also produces significant quantities of ethyl 4-(dibromomethyl)-
benzoate (9) as a by-product and requires separation of
succinimide from the product. It was therefore of interest to see
if zeolites could be used to catalyse and to control the direct
reaction of 7 with bromine. This should offer the advantages of
a cheap reagent, easy separation of the zeolite from the reaction
mixture by filtration, possible reuse of the catalyst, and an
environmentally friendly process.
As Table 9 shows, the reaction gives high yields and para-
selectivities for a range of substituted benzenes 4 of moderate
activity, including alkyl and halogenobenzenes.
Several features of these reactions are noteworthy. First, the
sodium forms of the zeolites generate much faster reactions
than the proton forms. This has been noted before,8,10,11 but
despite this fact, the common view has been that increasing the
Brønsted acidity of the active sites enhances the activity. Such
an inference appears to be unsustainable, and the somewhat
lesser activity of NaX than NaY may be due to greater dif-
fusion constraints or some other factor, rather than to the lower
acidity of the corresponding proton-form zeolite.
Second, the selectivity is remarkable, almost exclusive forma-
tion of the para-isomer being achieved in all cases. In the many
prior studies of zeolite-induced aromatic bromination reactions
(see refs. 8, 10, 11 and citations therein) selectivities at total
conversion were significantly less, especially for toluene. Also,
conventional wisdom would maintain that space within the
pores of zeolite Y is so abundant that there is little opportunity
for significant shape-selectivity. Yet the results defy this argu-
ment. It appears to be important that the reaction requires the
substrate, the bromine and the sodium form of the zeolite all to
be involved in the process. Perhaps even the presence of solvent
molecules plays a role. In this case, the space within the pores
could become much more congested than might be envisaged in
a cursory consideration, and this might lead to the observed
selectivity.
As a test case, the reaction of ethyl 4-methylbenzoate (7)
was conducted with excess bromine (1.33 mole equivalents) in
dry dichloromethane without zeolite [eqn. (7)].
(7)
A solution of bromine in dichloromethane was added in a
dropwise manner to a stirred solution of 7 in dichloromethane
over 20 min. The reaction mixture was then stirred at room
temperature for 2 h. After quenching and work-up, ethyl 4-
(bromomethyl)benzoate (8) was obtained in 42% yield as
determined by the use of 1H NMR spectroscopy. Ethyl 4-
(dibromomethyl)benzoate (9), presumably formed by reaction
of 8 with further bromine, was also obtained, in 0.5% yield.
Moreover, analysis of the reaction mixture showed that 57%
of the starting material remained unreacted.
We consider two potential mechanisms that might be
involved in the process occurring within the pores. One would
require a three-body process involving adsorption of both the
substrate and the bromine on the internal surface of the zeolite
and in close proximity to each other. The reaction could then
take place between bromine and the substrate, with assistance
to polarisation of the bromine molecule coming from the
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752
2749

View Article Online
Table 10 Yields according to eqn. (7)
Yields (%)a
Reaction
time/h
Reaction
temperature/ЊC
Br2 (equiv.)
Solvent
7
8
9
1.33b
1.33c
1.33
1.33
1.33
1.33
1.43
1.02d
1.02
1.23
1.07
1.02
1.02
1.02
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
MeOH
CH3CN
CHCl3
2
2
2
4
6
7
4
3
1
1.5
2
3
3
20
20
20
20
20
20
20
20
57
57
57
42
32
33
42
96
8
5
8
97
96
17
42
42
42
57
65
65
57
4
82
82
82
3
0.5
0.5
0.5
1
2
2
1
—
10
13
9
—
—
6
42 (reflux)
42 (reflux)
42 (reflux)
20
20
20
14
77
3
1
a Yields were determined by H NMR spectroscopy. b Dry dichloromethane (DCM) was used and bromine was added dropwise. c Dry DCM was
used. d Reaction carried out in the dark.
A series of experiments was conducted in which different
Conclusion
solvents were used and the conditions of the reaction were
The reaction of bromine with substituted benzenes of moderate
varied in an attempt to improve the yield of the desired product
activity in the presence of sufficient NaY zeolite at room
8 and to eliminate the formation of the side product 9. The
temperature provides a convenient, high yielding and highly
results obtained are recorded in Table 10.
regioselective method for synthesis of para-brominated
products. Furthermore, heating easily regenerates the zeolite
and the reaction works almost as well on a vastly larger scale
even without any modification to the conditions. Consequently,
this approach becomes a commercially realistic possibility for
the synthesis of para-bromo substituted benzenes.
High para-selectivity can also be achieved in bromination of
toluene by use of tert-butyl hypobromite as reagent, HNaX
It was found that dropwise addition of bromine and dry
conditions were not necessary. Not surprisingly, the yield of 8
was improved (to 82%) when a mixture of 7 and bromine
(1.07 equiv.) in dichloromethane was held under reflux for 2 h.
However, the side product 9 was obtained in a yield of about
9% under such conditions. It was also found that dichloro-
methane and chloroform are the best solvents of those tried
in which to conduct the bromination reaction. The room
zeolite as catalyst, and a solvent mixture comprising diethyl
temperature reaction took place faster in chloroform, in which
ether and tetrachloromethane, but the reaction is not general.
The use of zeolites also enhances the rate of light-induced
side-chain bromination of ethyl 4-methylbenzoate, but does not
substantially diminish the tendency to over-bromination.
the monobromo derivative 8 was obtained in 77% yield,
but there was a relatively high yield (ca. 6%) of the dibromo
derivative 9. Very low yields of the desired product 8 were
obtained when methanol or acetonitrile was used as solvent.
The yield of 8 was only ca. 4% when the bromination
reaction was carried out in dichloromethane in the dark,
verifying the radical nature of the process. As can be seen
from Table 10, the yield of the side product 9 could be as high
as 13%. This shows that the reaction between 8 and further
Experimental
A PU 4400 Gas Chromatograph (Philips) fitted with a capillary
column RTX-1 (100% polydimethylsiloxane, 0.32 mm ID)
was used to analyse reaction mixtures in the case of the nuclear
bromination reactions. Temperature programmed heating
between 70 ЊC and 200 ЊC was employed and decane or unde-
cane was added as an internal standard to allow quantification.
NMR was used for quantification in the case of side-chain
bromine is quite rapid under the reaction conditions tried.
Our attention was turned next to the use of various types
of zeolite (5A, 4A, 3A, 13X, NaM, HY and Hβ) to catalyse
and/or control the bromination process in different solvents.
When a mixture of 7 (6.10 mmol), bromine (6.22 mmol) and
zeolite 5A (2.0 g) in dichloromethane was stirred for 3 h at room
temperature, the yield of 8 was 83% and the side product 9 was
obtained in a yield of ca. 7%. Similar results were obtained
when a smaller amount of 5A (1.0 g or even 0.1 g) was used as
a catalyst. It had been hoped that slow release of bromine from
within the zeolite pores might help enhance the selectivity, but
the effect was not large if it was valid at all.
The results with other zeolites are shown in Table 11. Again,
when the reaction was carried out in methanol or acetonitrile
as solvent, or in the dark, very low yields of 8 were obtained.
The zeolites and solvent were varied in order to try to eliminate
the side product, but none of the conditions were completely
successful in this respect (Table 11).
1
bromination reactions. H NMR spectra were recorded on a
Bruker, AC 400 spectrometer operating at 400 MHz. Assign-
ments of signals are based on coupling patterns and expected
chemical shift values.
Zeolites
Commercial zeolites were purchased from Aldrich Chemical
Company (3A, 4A, 5A, 13X, NaY catalyst support) or provided
as gifts by Zeolyst International. Ion-exchanged forms were
obtained by heating the commercial sodium or proton forms
at reflux with a 1 M solution of the appropriate halide salt,
followed by washing with water until the washings were halide-
free and then calcination at 450 ЊC. Proton forms were obtained
by initial formation of the corresponding ammonium forms,
which break down to the proton forms during calcination.
As can be seen from Table 11, the yield of 8 could be
improved up to 84%, but the yield of 9 was always significant
also. The use of zeolites enhanced the rate of the bromination
process of ethyl 4-methylbenzoate (7) and improved the
yield of the desired product, but none of the zeolites brought
about substantial improvement over the selectivity of the
reaction.
Bromination reactions
Bromination of toluene with tert-butyl hypobromite. In an
oven-dried, 50 ml round-bottomed flask were placed zeolite
2750
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752

View Article Online
Table 11 Yields according to eqn. (7) in the presence of various zeolitesa
Yields (%)b
Zeolite
Mass used/g
Solvent
7
8
9
5A
5A
5A
5A
5A
5A
5A
5A
5A
4A
4A
3A
3A
3A
13X
13X
13X
NaMord
NaMord
NaMord
HY
HY
HY
Hβ
0.1
0.5
1.0
1.0d
1.0e
1.0
1.0
1.0
2.0
0.1
1.0
0.1
0.5
1.0
0.1
0.5
1.0
0.1
0.5
1.0
0.1
0.5
1.0
0.1
0.5
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
MeOH
MeCN
CHCl3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
10 (17)c
10
5
10
96
98
87
20
10
7
6
9
2
11
9
82 (77)c
82
83
82
4
8 (6)c
8
10
7
—
—
—
19
7
12
12
13
14
9
2
13
69
83
81
81
78
84
80
82
81
61
80
82
83
83
83
83
51
80
9
8
11
24
6
14
8
7
7
12
1
5
15
14
3
8
9
9
4
47
15
Hβ
a Compound 7 (6.1 mmol), bromine (6.22 mmol), zeolite, solvent, 20 ЊC. b Yields were determined by 1H NMR spectroscopy. c Figures in parentheses
are for a similar reaction but under reflux for 1 h. d Bromine was loaded into the zeolite. e Reaction carried out in the dark.
(1.50 g) and a magnetic follower. A standard solution contain-
ing toluene (2.50 mmol) and undecane (1.25 mmol, internal
standard), in the appropriate solvent (7.4 ml), was added,
followed by a solution of tert-butyl hypobromite in the same
solvent (1.00 M as estimated by iodometric titration, 2.60 ml,
2.6 mmol). The flask was stoppered and the mixture was stirred
for 1 h at ambient temperature. The mixture (or a sample of
it) was filtered through a glass sinter (P3) and the zeolite was
washed with a little extra solvent. The solution was analysed by
GC (see Tables 1–3).
product were combined and distilled at atmospheric pressure. A
forerun was collected up to 140 ЊC (uncalibrated thermometer)
and the product was then collected until the temperature
dropped again to 130 ЊC. The product (10.35 g, 80%) solidified
on cooling to yield an off-white solid, mp 26–28 ЊC (lit.18
26–29 ЊC). A brown residue (1.35 g) remained in the distillation
flask. Both the forerun and the residue contained a substantial
proportion of 4-bromotoluene, as shown by NMR, while the
main fraction was essentially the pure product.
Bromination of ethyl 4-methylbenzoate (7). A mixture of ethyl
4-methylbenzoate (1.0 g, 6.10 mmol), zeolite (0.1–2.0 g) and
bromine (0.32 ml, 6.22 mmol) in dichloromethane (20 ml)
was stirred at room temperature for a period of time. The
reaction was stopped by the addition of a dilute aqueous
sodium sulfite solution (5 ml). The reaction mixture was filtered
and the zeolite was washed with dichloromethane (2 × 5 ml).
The organic layer was separated, washed with water (2 ×
10 ml), dried (MgSO4), and evaporated to dryness under
Bromination of substituted benzenes 4 with bromine. A sample
of a calcined catalyst (0.05–0.55 g) was placed in a 25 ml round-
bottom flask with dichloromethane (5 ml). A solution of 4
(~0.70 ml of 10% solution in dichloromethane, 0.84 mmol)
and decane as a standard (0.43 ml of 10% solution in dichloro-
methane, 0.039 g) was then added. The mixture was protected
totally from light, and stirred for a few minutes. Bromine (0.48
ml of 10% solution in dichloromethane, 0.93 mmol) was then
added dropwise with rapid stirring in the dark. At various
times samples were taken (also protected from light) from the
reaction mixture and were analysed by GC.
1
reduced pressure. The residue obtained was submitted for H
NMR spectroscopy.
Acknowledgements
Preparative scale bromination of toluene. Catalyst NaY
(55.112 g, Aldrich 33,444-8, calcined at 550 ЊC for 24 h) was
added to dichloromethane (400 ml) in a 1 l round-bottomed
flask. To the mixture, a solution of toluene (7.011 g, 76.2 mmol)
in dichloromethane (100 ml) was added. The mixture was
stirred in the dark for 15 min and bromine (13.783 g, 86.1
mmol) in dichloromethane (100 ml) was then added dropwise
with rapid stirring over 45 min in the dark. The mixture was
stirred at room temperature for 5 h in the dark and then filtered.
The solid was washed with dichloromethane (3 × 100 ml) and
the washings were combined with the filtrate. The combined
solution was evaporated on a rotary evaporator at 20 ЊC to give
a pale yellow liquid (12.09 g, 93%). The zeolite was further
washed with acetone (2 × 100 ml) and the washings were
evaporated to give a brown liquid (0.40 g). The two parts of the
G. A. E.-H. thanks the University of Wales Swansea for
financial support and the Royal Society of Chemistry for an
international author grant. We thank the EPSRC and the
University of Wales for grants that enabled the purchase of
NMR equipment used in the course of this work. We also
thank Zeolyst International for gifts of zeolites.
References
1 Chemistry of Waste Minimization, ed. J. H. Clark, Chapman and
Hall, London, 1995.
2 Solids Supports and Catalysts in Organic Synthesis, ed. K. Smith,
Ellis Horwood, Chichester, 1992; Preparative Chemistry Using
Supported Reagents, ed. P. Laszlo, Academic Press, London, 1987;
H. Van Bekkum, E. M. Flanigen and J. C. Jansen, Stud. Surf. Sci.
Catal., 1991, 58.
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752
2751

View Article Online
3 R. Taylor, Electrophilic Aromatic Substitution, John Wiley and Sons,
Chichester, 1990.
4 L. Delaude, P. Laszlo and K. Smith, Acc. Chem. Res., 1993, 26, 607.
5 K. Smith, M. Butters, W. E. Paget and B. Nay, Synthesis, 1985,
1155; K. Smith, M. Butters and B. Nay, Synthesis, 1985, 1157;
K. Smith, M. Butters, W. E. Paget, D. Goubet, E. Fromentin and
B. Nay, Green Chem., 1999, 83.
9 J. H. Clark, J. C. Ross, D. J. Macquarrie, S. J. Barlow and
T. W. Bastock, Chem. Commun., 1997, 1203.
10 F. De La Vega and Y. Sasson, Zeolites, 1989, 9, 418; F. De La Vega
and Y. Sasson, J. Chem. Soc., Chem. Commun., 1989, 653.
11 F. De La Vega, Y. Sasson and K. Huddersman, Zeolites, 1991, 11,
617; F. De La Vega, Y. Sasson and K. Huddersman, Zeolites, 1993,
13, 341.
6 M. E. W. Hammond, PhD Thesis, University of Wales, Swansea,
1999.
7 A. G. Mistry, K. Smith and M. R. Bye, Tetrahedron Lett., 1986,
27, 1051; K. Smith, D. M. James, A. G. Mistry, M. R. Bye and
D. J. Faulkner, Tetrahedron, 1992, 48, 7479; K. Smith, D. M. James,
I. Matthews and M. R. Bye, J. Chem. Soc., Perkin Trans. 1, 1992,
1877.
12 K. Smith and D. Bahzad, Chem. Commun., 1996, 467.
13 C. Walling and A. Padwa, J. Org. Chem., 1962, 27, 2967.
14 S. M. Nagy, E. A. Zubkov, V. G. Shubin, A. G. Pelmenshchikov,
L. A. Vostrikova and K. G. Ione, Mendeleev Commun., 1991, 94.
15 A. F. Titley, J. Chem. Soc., 1928, 2581.
16 H. R. Snyder, E. P. Merica, C. G. Force and E. G. White, J. Am.
Chem. Soc., 1958, 80, 4622.
8 T. M. Wortel, D. Oudijn, C. J. Vleugel, D. P. Roelofsen and H. Van
Bekkum, J. Catal., 1979, 60, 110.
17 S. Goswami and A. K. Mahaptra, Tetrahedron Lett., 1998, 39, 1981.
18 Aldrich catalogue, Aldrich Chemical Company, UK, 1999–2000.
2752
J. Chem. Soc., Perkin Trans. 1, 2000, 2745–2752