Aromatic nitrations in ionic liquids: The importance of cation choice

Article abstract of DOI:10.1039/b310312a

Aromatic substrates can be nitrated in high yields and with efficient use of the nitrating agent in ionic liquids, although a suitably inert ionic liquid cation must be used.

Full text of DOI:10.1039/b310312a

Aromatic nitrations in ionic liquids: the importance of cation choice
N. Llewellyn Lancaster* and Verónica Llopis-Mestre
Chemistry Department, Imperial College London, South Kensington Campus, London, UK SW7 2AX.
E-mail: l.lancaster@imperial.ac.uk
Received (in Cambridge, UK) 27th August 2003, Accepted 28th September 2003
First published as an Advance Article on the web 11th October 2003
Aromatic substrates can be nitrated in high yields and with
efficient use of the nitrating agent in ionic liquids, although
a suitably inert ionic liquid cation must be used.
where the cation has been immobilised on clay or zeolite
supports. Preliminary studies showed that when the salts were
treated with acetic anhydride, no nitrotoluene was produced.
When the system was modified by use of trifluoroacetic acid
anhydride, nitrotolouenes were produced, but in low yield.
These reactions were more effective in [bmpy][N(Tf)2] than in
any of ther other ionic liquids studied. However, the HNO3–
TFAA system gave very poor yield in the other ionic liquids and
only a low yield in [bmpy][N(Tf)2].
The reactions of the HNO3–Ac2O system with toluene in the
ionic liquids generally gave low yields in the ionic liquids, and
within 1 h the reaction had stopped. Further reaction did not
occur if left, suggesting that the reagent had been entirely
depleted. This is in contrast to the reaction in dichloromethane,
in which the yield after 1 h was low, but which, if left, gave a
near quantitative yield of nitrotoluenes. However, it is clear that
the reaction in [bmpy][N(Tf)2] does not fit that trend. Instead
93% of the nitric acid used has been converted into ni-
trotoluenes within 1 h.
In both acyl nitrate systems, the regioselectivity of nitration is
fairly similar, although the reactions in [bmpy][N(Tf)2] are
apparently slightly more p- selective than the other systems.
Returning to the yields of nitrotoluenes obtained in the
various ionic liquids, there is an interesting pattern. The lowest
yield (i.e., the least efficient use of the nitrating agent) is
obtained in the [bmim]+ ionic liquids, with [bm2im]+ being
somewhat better and [bmpy]+ giving very high yields. This
observation can be turned around, so that instead we consider
the stability of the ionic liquid towards the nitrating agents.
Within the literature, the nitration of bases is well known and
Ridd has demonstrated the nitration of imidazolium salts.11
In order to study this, we put ionic liquid, nitric acid and
acetic anhdyride together in the same proportions as before, but
without the substrate. The reaction was quenched and the
product was analysed by FAB+ MS. Where the ionic liquid
cation was [bmim]+, peaks due to the addition of 45 or 90 mass
units (1 or 2 nitro groups) were observed. The same observation
was made where the cation was [bm2im]+. Thus the low
efficiency of use of the nitrating agent is explained.
Ionic liquids continue to excite interest as alternative solvents
for organic reactions.1 One reaction of academic, synthetic and
industrial interest is aromatic nitration.2 Strategies for improv-
ing this notoriously “ungreen” reaction have included the use of
clays3,4 or zeolites5–7 as solid supports, or perfluorocarbons8 as
solvents. Yet to date, ionic liquids have only been used in two
studies of aromatic nitration. Laali and Gettwert used a range of
nitrating agents in the nitration of mainly activated substrates9
whilst Srinivasan and Rajagopal used iron nitrate in the nitration
of phenols.10 In this work we have achieved the nitration of a
range of aromatic substrates including chlorobenzene and
bromobenzene by using HNO3–Ac2O in an ionic liquid.
Ionic liquids have many useful physical properties, not least
of which is their generally low miscibility with hydrocarbons. In
this study, we deliberately used an excess of substrate to
nitrating agent and then studied the extent of reaction under a
range of conditions and the efficiency with which the nitrating
agent is used. We used 4 different ionic liquids derived from 3
different cations, shown in Fig. 1 (where bmim = 1-butyl-
3-methylimidazolium, bm2im = 1-butyl-2,3-dimethylimidazo-
lium and bmpy = 1-butyl-1-methylpyrrolidinium).†
The nitrating agents used are all (fairly) mild, being acyl
nitrates formed in situ from the reaction of acetic (or
trifluoroacetic) anhydride with the nitrate salt or nitric acid. The
results of nitrations of toluene by these systems are shown in
Table 1.
A range of nitrate salts were used in all of the ionic liquids to
see whether the identity of the cation (even in the ionic liquid)
might influence reaction times or selectivities, as often observed
Having established that the best cation was [bmpy]+, we
started to study the range of substrates (activated and deacti-
vated) that could be nitrated using the nitric acid–Ac2O system.
These reactions became homogeneous except where mesitylene
was substrate. The results of these reactions are shown in Table
2. For comparison, the same reactions were performed in
dichloromethane, which is an excellent solvent for aromatic
nitrations.
Fig. 1
Table 1 The nitration of toluene by acyl nitrates at 25 °C in ionic liquids and
dichloromethane; yields after 1 h
Solvent
System
Yield (%)
Time/h
o-/p-
In the nitration of activated substrates there was little
difference between the ionic liquid and the molecular solvent.
For example, in the nitration of mesitylene, high yields of
nitroaromatic product were achieved in each system. In the
nitration of anisole, the yields were similar, although the
reaction was significantly more p- selective in [bmpy][N(Tf)2]
than in dichloromethane. The nitration of toluene showed near
identical regioselectivity in the two solvents, but the yield was
significantly higher in the ionic liquid.
CH2Cl2
[bmim][BF4]
HNO3–Ac2O
HNO3–Ac2O
HNO3–Ac2O
HNO3–Ac2O
HNO3–Ac2O
NH4NO3–TFAA
Cu(NO3)2–TFAA
Fe(NO3)3–TFAA
HNO3–TFAA
HNO3
35
35
42
63
93
4
18
25
32
< 1
1
1
24
1
1
1
1
1
1
1
1.4
1.5
1.5
1.4
1.3
1.1
1.3
1.4
1.0
—
[bmim][N(Tf)2]
[bm2im][N(Tf)2]
[bmpy][N(Tf)2]
[bmpy][N(Tf)2]
[bmpy][N(Tf)2]
[bmpy][N(Tf)2]
[bmpy][N(Tf)2]
[bmpy][N(Tf)2]
The nitration of deactivated substrates shows greater impact
of using the ionic liquid. As is shown by the table, under the
2812
CHEM. COMMUN., 2003, 2812–2813
This journal is © The Royal Society of Chemistry 2003

View Article Online
Table 2 Nitration of aromatic substrates at 25 °C; yields after 1 h
[bmpy][N(Tf)2]
Yield (%)
Dichloromethane
Fig. 2
Substrate
o-/p-
Yield (%)
o-/p-
Mesitylene
Toluene
Anisole
Bromobenzene
Chlorobenzene
Nitrobenzene
63
91a
89
63
50
< 1c
—
85
33b
96
0
0
< 1c
—
1.4
2.7
—
—
—
Clearly, these proposals will need to be tested further, and
this will best be achieved by a thorough kinetic and mechanistic
study. This work is ongoing. This work has shown that, by using
an ionic liquid, it is possible to achieve high yields of
nitroaromatic products using HNO3–Ac2O, and even to nitrate
compounds that might otherwise not react with this reagent. We
have shown that the regioselectivity is no lower than in a
molecular solvent, and that in fact we can achieve higher p-
selectivity in some cases. We have also shown that it is
necessary to chose a suitable ionic liquid, i.e., one which is
stable to nitration itself.
1.5
2.0
0.28
0.28
—
a Includes ca. 3% 3-nitrotoluene. b Includes ca. 1% 3-nitrotoluene. c 24 h,
m-dinitrobenzene only.
reaction conditions employed in dichloromethane, the yield of
bromonitrobenzenes is very low. Under the same conditions in
[bmpy][N(Tf)2], the yield is 63%. The result of chlorobenzene
nitration is similar, with again a low yield in the molecular
solvent and a yield of 50% in the ionic liquid.
We wish to acknowledge the Socrates program (V.Ll.M) for
support
The limit of the system appears to be reached in the nitration
of nitrobenzene. The reaction was sluggish even in the ionic
liquid, and a yield of less than 1 % was achieved, even after 24
h.
We must now address the solvent properties of the ionic
liquid. There appear to be two possible explanations for the
difference between the reaction in dichloromethane and in
[bmpy][N(Tf)2].
The first is that the reagent and the substrate are both charge
neutral but reacting to form a charged intermediate. The fact that
the outcomes of nucleophilic substitutions (using charged
reagents) can be predicted using the Hughes–Ingold rules for
ionic liquids has been discussed elsewhere.12 The same rules
show (in the case of neutrally charged nucleophiles), that the
rate of reaction would be increased by using an ionic liquid. We
could be observing a related effect in this work.
The HNO3–Ac2O system is generally thought to be a
molecular nitrating agent, although it is proposed that this
compound may be a nitronium ion carrier, dissociating before
reaction to give the free nitronium ion as shown (eqn. 1)
below.
Notes and references
† All reactions were carried out under a dry nitrogen atmosphere at 25 °C.
The ionic liquids were prepared as described elsewhere.12,13 Nitric acid
(0.13 cm3, 69% w/w, 2.0 mmol) was combined with the solvent (0.5 cm3)
and acetic anhydride (1 cm3, 10.6 mmol) or trifluroacetic acid anhydride
(1.2 cm3, 8.5 mmol). To this was added the substrate (2 cm3).
At known time, a sample was taken and quenched in saturated aqueous
sodium bicarbonate solution and extracted with dichloromethane, then
analysed by GC using an internal standard.
1 (a) Ionic Liquids in Synthesis, ed. P. Wasserscheid and T. Welton,
Wiley-VCH, Weinheim, 2002; (b) T. Welton, Chem. Rev., 1999, 99,
2071; (c) P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed. Engl.,
2000, 39, 3772; (d) R. Sheldon, Chem. Commun., 2001, 2399.
2 (a) K. Schofield, Aromatic Nitration, Cambridge University Press,
Cambridge, 1980; (b) G. A. Olah, R. Malhotra and S. C. Narang,
Nitration: Methods and Mechanisms, VCH, 1989; (c) N. Ono, The Nitro
group in Organic Synthesis, Wiley-VCH, 2001.
3 A. Cornelis, L. Delaude, A. Gerstmanns and P. Laszlo, Tetrahedron
Lett., 1988, 29, 5909.
4 B. M. Choudary, M. Ravichandra Sarma and K. Vijaya Kumar, J. Mol.
Catal., 1994, 87, 33.
5 (a) R. P. Claridge, N. L. Lancaster, R. W. Millar, R. B. Moodie and J.
P. B. Sandall, J. Chem. Soc., Perkin Trans. 2, 2001, 197; (b) R. P.
Claridge, N. L. Lancaster, R. W. Millar, R. B. Moodie and J. P. B.
Sandall, J. Chem. Soc., Perkin Trans. 2, 1999, 1815.
6 K. Smith, S. Almeer and C. Peters, Chem. Commun., 2001, 2748 and
references therein.
7 (a) D. Vassena, A. Kogelbauer and R. Prins, Catal. Today, 2000, 60,
275; (b) M. Haouas, S. Bernasconi, A. Kogelbauer and R. Prins, Phys.
Chem. Chem. Phys., 2001, 3, 5067.
8 (a) M. R. Crampton, L. M. Gibbons and R. Millar, J. Chem. Soc., Perkin
Trans. 2, 2001, 1662; (b) M. R. Crampton, E. L. Cropper, L. M. Gibbons
and R. W. Millar, Green Chem., 2002, 4, 275.
AcONO2 = AcO2 + NO2
+
(1)
This allows a second possible explanation. In the ionic liquid,
it would not be unreasonable to expect the complete dissocia-
tion of acetyl nitrate to nitronium acetate. This would be
expected to be a more potent nitrating agent and thus give the
higher yields or shorter reaction times that are generally
observed.
There is an additional consideration, based on the properties
of the ionic liquid cations and anions, and on the reagents and
intermediates of the nitration. If we consider aromatic nitration
by the nitronium ion, the rate determining step is the formation
of the Wheland intermediate which, once formed, undergoes
rapid deprotonation to give the nitroaromatic product. To
enhance the rate of the reaction, the rate of formation of the
Wheland intermediate must be increased. This is illustrated in
Fig. 2. If the Wheland intermediate can interact more strongly
with the ionic liquid than with the molecular solvent then we
would expect an increase in yield after unit time.
9 K. K. Laali and V. J. Gettwert, J. Org. Chem., 2001, 66, 35.
10 R. Rajagopal and K. V. Srinivasan, Synth. Commun., 2003, 33, 961.
11 M. W. Austin, J. R. Blackborrow, J. H. Ridd and B. V. Smith, J. Chem.
Soc., 1965, 1051.
12 N. L. Lancaster, P. A. Salter, T. Welton and G. B. Young, J. Org. Chem.,
2002, 67, 8855.
13 L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton, Phys. Chem.
Chem. Phys., 2001, 3, 5192.
CHEM. COMMUN., 2003, 2812–2813
2813