Electrochemically promoted nucleophilic aromatic substitution in room temperature ionic liquids - An environmentally benign way to functionalize nitroaromatic compounds

Article abstract of DOI:10.1039/c1gc15303j

The current manuscript shows the electrochemical studies performed to rationalize the mechanism and develop new green synthetic routes for the synthesis of substituted nitroaromatics based on the advantages of the electrochemical approach to the nucleophilic aromatic substitution reaction (such as (a) low cost and ready availability of reagents, (b) atom economy, (c) high yields, approaching 100%) and the use of Room Temperature Ionic Liquids (RTILs) as green alternative solvents to organic aprotic solvents. Four of the most popular RTILs (1-butyl-3-methylimidazolium tetrafluoroborate ([BMIM]BF ₄), 1-butyl-3-methylimidazolium hexafluorophosphate ([BMIM]PF ₆), 1-butyl-3-methyl-imidazolium bis(trifluoromethylsulfonyl)-imide ([BMIM]TFSI) and 1-butyl-3-methylimidazolium acetate ([BMIM]AcO) have been chosen since they have different properties in terms of solvation effects that can increase the regioselectivity of the reaction. The nucleophiles used to study the feasibility and viability of the reaction were the classical hydride, methoxide, ketones, cyanides and amines, whereas the nitroarenes selected were 4-nitrotoluene, 1,3-dinitrobenzene, 2,4-dinitroaniline, 1,3,5-trinitrobenzene, 1,3-dinitronaphthalene, 1-chloro-2,4,6-trinitrobenzene and 2,4,6- trinitroanisole. The electrocatalysis and regioselectivity effects of using RTILs are also investigated. The article concludes by analyzing the economic cost of performing this electrosynthesis in RTILs and organic solvent electrolyte systems, which contain 0.1 M of supporting electrolyte.

Full text of DOI:10.1039/c1gc15303j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 2531
www.rsc.org/greenchem
PAPER
Electrochemically promoted nucleophilic aromatic substitution in room
temperature ionic liquids—an environmentally benign way to functionalize
nitroaromatic compounds
Hugo Cruz, Iluminada Gallardo* and Gonzalo Guirado*
Received 19th March 2011, Accepted 17th June 2011
DOI: 10.1039/c1gc15303j
The current manuscript shows the electrochemical studies performed to rationalize the mechanism
and develop new green synthetic routes for the synthesis of substituted nitroaromatics based on
the advantages of the electrochemical approach to the nucleophilic aromatic substitution reaction
(such as (a) low cost and ready availability of reagents, (b) atom economy, (c) high yields,
approaching 100%) and the use of Room Temperature Ionic Liquids (RTILs) as green alternative
solvents to organic aprotic solvents. Four of the most popular RTILs (1-butyl-3-methylimida-
zolium tetrafluoroborate ([BMIM]BF4), 1-butyl-3-methylimidazolium hexafluorophosphate
([BMIM]PF6), 1-butyl-3-methyl-imidazolium bis(trifluoromethylsulfonyl)-imide ([BMIM]TFSI)
and 1-butyl-3-methylimidazolium acetate ([BMIM]AcO) have been chosen since they have
different properties in terms of solvation effects that can increase the regioselectivity of the
reaction. The nucleophiles used to study the feasibility and viability of the reaction were the
classical hydride, methoxide, ketones, cyanides and amines, whereas the nitroarenes selected were
4-nitrotoluene, 1,3-dinitrobenzene, 2,4-dinitroaniline, 1,3,5-trinitrobenzene, 1,3-dinitronaph-
thalene, 1-chloro-2,4,6-trinitrobenzene and 2,4,6-trinitroanisole. The electrocatalysis and
regioselectivity effects of using RTILs are also investigated. The article concludes by analyzing the
economic cost of performing this electrosynthesis in RTILs and organic solvent electrolyte
systems, which contain 0.1 M of supporting electrolyte.
1. Introduction
Nucleophilic aromatic substitution of halogen or other nucleo-
fugal groups in nitroarenes, according to the SNAr addition-
elimination mechanism (Scheme 1),1 is a classical reaction of
major practical value. Studies of the mechanisms of aromatic
substitution of electron-deficient aromatic and heteroaromatic
compounds, establishing the factors that affect the choice
Scheme 1 General overview of a SNAr reaction.
of mechanistic path and the regioselectivity within a given
mechanism, can have a major impact on such important
X
The conversion of intermediate s -complexes into products
areas as drug synthesis, polymer research, and environmental
through the promoted nucleophilic aromatic substitution of a
chemistry.2,3 So, a thorough understanding of these mechanisms
heteroatom via spontaneous departure of the leaving group, the
will be of high value in the practical choice of conditions,
NASX process, is strongly dependent on the nature of the leaving
solvents, and nucleophiles for the preparation of those com-
group, the nature of the nucleophilic reagent and the effects of
pounds and in the choice of procedures used in environmental
the medium, as has been previously described.4 The solvent is
amelioration.
one of the most important parameters in SNAr reactions since it
determines its thermodynamics. Changes in the polarity of the
solvent from protic to aprotic solvents affect the nature of the
Departament de Qu´ımica, Universitat Auto`noma de Barcelona, 08193,
substitution product as well as the required time to complete
Bellaterra, (Barcelona), Spain.
the reaction.4 Nucleophilic aromatic substitution of hydrogen
(NASH) reactions formally require the replacement of a hydride
E-mail: Iluminada.Gallardo@uab.es, Gonzalo.Guirado@uab.es; Fax: 00
34 935812920; Tel: 00 34 9358121 37
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2531–2542 | 2531
©

View Article Online
ion, and they either proceed “spontaneously” with consumption
of part of the starting material in the oxidation step, or are
promoted by the addition of external oxidants. In both cases,
low yields (with few exceptions)5,6 and a lack of generality are
the main drawbacks of these synthetic procedures. Moreover,
some of the chemical substances used as chemical oxidants are
inherently hazardous.
a heteroatom (NASX process)14,15 when a substituted position is
attacked by the nucleophile.
The main advantage of the electrochemical approaches to
the synthesis of functionalized nitroaromatic compounds is
that electrons flowing as current are regarded as one of the
reagents, thus the route is environmentally friendly. Moreover,
reactions take place in low-temperature conditions, reducing
the local consumption of energy and the risk of corrosion,
material failure, and accidental release. Finally, it is important
to highlight that electrodes can be regarded as heterogeneous
catalysts that are easily separated from the products. The low
volatility or non-volatility of reaction media are other factors
to be taken into account. In this sense, the main drawbacks are
the use of organic polar solvents and supporting electrolytes. An
interesting alternative would be to replace those solvents with
safer ones. Hence, the use of Room Temperature Ionic Liquids
(RTILs) would be very attractive not only due to their chemical
features (they are non-flammable, non-volatile and thermally
stable over a wide range of temperatures) but also to their
electrochemical ones.16–19 These solvents are a liquid consisting
of only ions, so from moderate to high conductivities are
guaranteed and there is no need to use a supporting electrolyte.
Moreover, their electrochemical stability (from 0 to 1.7 V vs.
SCE for oxidation processes) makes them an environmentally
benign replacement for classic electrolyte solutions.20 With the
exception of a single study, this approach has been almost
completely neglected in the literature.21 In our continuous
effort to rationalize the mechanism and develop new synthetic
routes, this manuscript is devoted to developing a “greener”
We were one of the first groups to focus our attention on
performing the NASH and NASX reactions by electrochemical
methods without the need to use halogenated starting materials
or intermediates (Scheme 2).7,8 The mechanism for the elec-
H
trochemical oxidation of the s -complex involves the loss of
two electrons and a proton, which corresponds to the formal
elimination of a hydride anion. The NASX electrochemical
mechanism also involves the elimination of a leaving group,
although its departure is as a radical rather than an anion, as
X
the rearomatization of the s or ipso complex involves the loss
of an electron.
The electrochemical approach to the NASH reaction over-
comes most of the previously mentioned drawbacks and in-
troduces numerous advantages in comparison with the chem-
ical route, such as low cost, ready availability of reagents,
atom economy and high yields (approaching 100% based on
unrecovered starting material). This environmentally friendly
electrochemical approach to this reaction has been successfully
applied7–13 to the cyanation,7 amination9 and alkylation12,13 of
nitroarenes. Moreover, it has been reported by some of us that
the use of electrochemical techniques also makes it possible to
perform and accelerate the nucleophilic aromatic substitution of
Scheme 2 Electrochemical approaches to NASH and NASX.
2532 | Green Chem., 2011, 13, 2531–2542 This journal is
The Royal Society of Chemistry 2011
©

View Article Online
route for the synthesis of nitroarene derivatives based on the
advantages of an electrochemical approach to the nucleophilic
aromatic substitution reaction and the current trends in Green
Chemistry of using environmentally friendly solvents, such as
RTILs. In the current manuscript, the electrochemical NASH
and NASX reactions were performed on four of the most
popular RTILs, such as 1-butyl-3-methylimidazolium tetraflu-
oroborate ([BMIM]BF4), 1-butyl-3-methylimidazolium hex-
afluorophosphate ([BMIM]PF6), 1-butyl-3-methylimidazolium
bis(trifluoromethylsulfonyl)-imide ([BMIM]TFSI) and 1-butyl-
3-methylimidazolium acetate ([BMIM]AcO) using several ni-
2. Experimental
2.1. Chemicals
N,N-dimethylformamide (DMF) was stored in an inert
atmosphere and molecular sieves were purchased from
Across. nBu4NBF4 (Fluka, puriss.) was used without fur-
ther purification. 4-Nitrotoluene (1), 1,3-dinitrobenzene (2),
2,4-dinitroaniline (5), 1,3-dinitronaphtalene (6) was pur-
chased from Aldrich. Whereas, 1,3,5-trinitrobenzene (3)
was provided by Supelco and 2,4,6-trinitroanisole (4) and
1-chloro-2,4,6-trinitrobenzene (7) were synthesized accord-
ing to methods previously described in the literature.22,23
All the commercially available reactants are of the
-
-
troarenes (1–7) and nucleophiles (BH4 , OCH3, CN-, ketones
and amines). Those starting materials, as well as the s-
complexes, are depicted in Scheme 3.
Scheme 3 The structures of the nitroarenes, s–complexes and RTILs.
The Royal Society of Chemistry 2011 Green Chem., 2011, 13, 2531–2542 | 2533
This journal is
©

View Article Online
highest purity available and are used without further
purification.
and DMF/[BMIM]AcO, the ratio being 8.1, 12.1, 4.1 and 12.7,
respectively. Note that the organic solvent electrolyte system,
which is labeled DMF, is a solution of DMF + 0.1 M of TBABF4.
RTILs ([BMIM]BF4, [BMIM]PF6 and [BMIM]AcO) were
first purchased from Solvionic, who guarantees that the amount
of water present is always less than 500 ppm. Later both
the RTILs [BMIM]BF4, [BMIM]PF6 and [BMIM]TFSI were
prepared from BMIMCl (Fluka) aqueous solutions according
to previously published procedures (Scheme 4).24–26 The solvents
are purified by repeated washing with water and purified by
column chromatography over silica under continuous nitrogen-
flow. The ionic liquid was conveniently dried until the theoreti-
cally expected electrochemical response was obtained for water
sensitive redox probes, such as 2,4-dinitroaniline, when either
the commercial or the prepared RTILs were used.
2.3. General procedure for NASH and NASX in nitroarenes
The corresponding anions ([5-H]-, [3-CN]-, [3-NH-
(CH2)3CH3]-, [2-CH2COCH3]-, [2-CH(CH3)COCH3]-, [3-
H
AcO]-, [4-CN-s ]-, [6-CH2COCH3]-, [6-CH(CH3)COCH3]-,
X
[4- OCH3]-, [7-Cl,NH- (CH2)3CH3]-, [4-CN,OCH3-s ]-,[4-
X
CN,NO2-s ]-) are either prepared “in situ” or transferred into
the RTILs and oxidized electrochemically (exhaustive controlled
potential electrolysis) using a carbon graphite as a working
electrode. When the reaction is complete, the mixture is extracted
in water/dichloromethane mixtures. The organic layer is dried
with Na2SO4 and evaporated affording a residue that is analyzed
by gas chromatography–mass spectrometry. The analysis shows
the presence of substituted nitrocompounds (2,4-dinitroaniline,
2,4,6-trinitrosanisole, 1-(2,4-dinitrophenylacetone), 3-(2,4-dini-
trophenyl)-2-butanone, 1-(2,4-dinitro-1-naphtyl)acetone, 3-(2,
4-dinitro-1-naphthyl)-2-butanone,
line, 2,4,6-trinitrobenzonitrile,
N-butyl-2,4,6-trinitroani-
2-methoxy-3,5-dinitroben-
zonitrile, 2,4,6-trinitrobenzonitrile and 3-methoxy-2,4,6-
trinitrobenzonitrile, respectively). Later, either the NASH
or NASX products are analyzed by 1H NMR and cyclic
voltammetry and identified by comparison of their spectroscopic
behaviour with that of pure samples available in our laboratory
in each case. At the end of the process the RTILs are suitably
recovered and recycled by water/dichloromethane washings
and are used again in a new set of experiments.
Scheme 4 The synthesis of the RTILs ([BMIM]BF4, [BMIM]PF6 and
[BMIM]TFSI).
2.2. Cyclic voltammetry experiments
2.4. Instrumentation
An electrochemical conical cell equipped with a methanol jacket,
which makes it possible to fix the temperature by means of a
thermostat, is used for the set-up of the three-electrode system.
For cyclic voltammetry experiments, the working electrode is,
in all cases, a glassy carbon disk of a diameter of 0.5 mm. It is
polished using a 1 mm diamond paste. The counter electrode is a
Pt disk of a diameter of 1 mm. All of the potentials are reported
versus an aqueous saturated calomel electrode (SCE) isolated
from the working electrode compartment by a salt bridge. The
salt solution of the reference calomel electrode is separated
from the electrochemical solution by a salt-bridge ended with
a frit, which is made of a ceramic material, allowing ionic
conduction between the two solutions and avoiding appreciable
contamination. Ideally, the electrolyte solution present in the
bridge (pure RTIL in the ionic liquid experiments) is the same
as that used for the electrochemical solution so as to minimize
junction potentials. So, the above-described electrochemical set-
up makes it possible to accurately determine the redox potential
at which the electrosynthetic processes should be performed.
Due to the low conductivity and larger viscosity of RTILs
and in order to facilitate direct comparison between the cyclic
voltammograms of redox probes in the different media, the
current values obtained are first normalized with respect to the
concentration and the scan rate (Ip/c.v1/2). The relationship be-
tween the four solvents can be expressed by a total normalization
coefficient. This coefficient can be expressed as a ratio between
DMF/[BMIM]BF4, DMF/[BMIM]PF6, DMF/[BMIM]TFSI
The cyclic voltammetry at low scan rates and controlled poten-
tial electrolysis are performed using a PAR 273A potentiostat
controlled by a computer. Powersuite software is used for data
acquisition and data treatment.
3. Results and discussion
The aim of this work is to establish the synthetic scope and
versatility of the electrochemical approach to the SNAr reaction
in RTILs. Thus, it should be mandatory to first establish
the electrochemical behaviour and the chemical stability of
the nitroaramotic compounds (main reactants in the SNAr
reactions) in RTILs. The next step will be to analyze the
electrochemical mechanism of the electrochemical oxidation
H
X
of the s -complexes and s -complexes in RTILs. Finally, we
will discuss whether the electrochemical oxidation of the s-
complexes leads to the substitution product, and to what extent,
by formally replacing a hydride (H-) or halide (X-) following a
NASH or NASX mechanism in RTILs.
3.1. Reactivity of neutral, anion radical and dianion of
nitroderivatives using RTILs
We have previously established27 the electrochemical behaviour
of a large family of nitroaromatic compounds, such as 4-
nitrotoluene (1), 1,3-dinitrobenzene (2), 1,3,5-trinitrobenzene
(3) and 2,4,6-trinitroanisole (4) in second generation RTILs
2534 | Green Chem., 2011, 13, 2531–2542
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Cathodic and anodic peak potential, standard potential, and DEp for 1, 2, 3 and 4 in different solvents at 20 ◦C (Scan rate 1.0 V s-1)
COMPOUND
SOLVENT
Epc (V)a
Epa (V)a
E0 (V)a
DEp (mV)b
1
DMFc
-1.267
-1.142
-1.367
-1.220
-1.036
-0.954
-0.845
-1.009
-0.950
-0.763
-0.581
-0.720
-0.763
-0.725
-1.100e
-0.649
-0.674
-0.887
-0.716
-1.075e
-1.176
-1.064
-1.239
-1.137
-0.923
-0.866
-0.785
-0.928
-0.866
-0.654
—
—
—
—
—
—
—
—
—
-1.222
-1.103
-1.303
-1.179
-0.978
-0.910
-0.815
-0.969
-0.907
-0.708
—
—
—
—
—
—
—
—
—
91
78
128
83
113
78
60
81
84
109
51
57
58
58
110
59
50
58
62
89
[BMIM]BF4
[BMIM]PF6
[BMIM]TSFI
[BMIM]AcO
DMFc
[BMIM]BF4
[BMIM]PF6
[BMIM]TSFI
[BMIM]AcO
DMFc
[BMIM]BF4
[BMIM]PF6
[BMIM]TSFI
[BMIM]AcOe
DMFb
[BMIM]BF4
[BMIM]PF6
[BMIM]TSFI
[BMIM]AcOe
2
3d
4d
—
—
a Potential values given vs. SCE b DEp = Epa - Epc; c Solution of DMF + 0.1 M TBABF4; d Trinitrobenzene derivates show an irreversible reduction
wave; e This value corresponds to the first reduction wave detected in [BMIM]AcO.
[BMIM]BF4 and [BMIM]PF6. This study is now extended to one
of the most popular second generation RTILs ([BMIM]TFSI)
and a third generation RTIL ([BMIM]AcO). The main differ-
ence between the second and the third generation of RTILs
is related to their reactivity in the presence of water and to the
nature of the counter anion. In the case of the second generation
RTILs the main feature is the hydrophobic character of the
reveal differences in the E◦ values (Table 1), which are strongly
related to ion-pairing effects.27,28 This effect on the shift of the
E◦ values is also observed when compound 2 is studied in those
solvents (Table 1). Compound 2 shows the same electrochemical
behaviour as 1 in terms of the E◦ value shift, so it can be used
as a model compound for dinitroderivatives, 2,4-dinitroaniline
(5) and 1,3-dintronaphthalene (6). In the case of trinitrobenzene
derivatives (compounds 3 and 4) the peak potential value (Epc)
is also strongly dependent on the solvent, being more easily
reduced in DMF + 0.1 M TBABF4 than in RTILs.
counter anion (BF4 , PF6 and TFSI-), whereas in the case of
the third generation, the RTILs contain a task specific counter
anion (AcO-).
-
-
Due to the low conductivity and high viscosity of RTILs
and in order to facilitate direct comparison between the cyclic
voltammograms, the current values obtained have been nor-
malised with respect to the concentration of the nitroaromatic
compound and the scan rate (Ip/cv1/2). Fig. 1 shows the
electrochemical behaviour of 1 in several solvents. A well-defined
one-electron reversible wave is obtained in DMF + 0.1 M
TBABF4 and in RTILs, so neither the nature of the reactant nor
the reactivity of the electrogenerated anion radical is affected
in those media. However, the cyclic voltammetry experiments
When [BMIM]AcO is used, a new effect that is mainly
negligibly for compound 1 and it is related to the stability of
the reactant, is observed. For compound 2, Fig. 2a shows that
the height of the two successive one-electron reversible reduction
peaks corresponding to the formation of the anion radical and
dianion, respectively, decreases over time. After 15 min, the peak
current value of the first peak decreases by 25% and, after 30 min,
it has decreased by 60% and the solution becomes purple.29 These
facts can be explained by the nucleophilic attack of the acetate
anion on the nitroaromatic compounds. This effect becomes very
important in the case of trinitrobenzene derivatives (3 and 4),
for which the reduction wave associated to 3 or 4 is not observed
after a few seconds (Fig. 2b). Note that the reduction wave at
ca. -1.1 V detected for compounds 3 and 4 corresponds, in both
H
cases, to the reduction of the s -complex, [3-AcO]-, formed in
solution (Table 1).
3.2. Electrochemical behaviour of rH-complexes and
rX-complexes in RTILs
H
X
In order to analyse the stability of the s -complexes and s -
complexes, the key intermediates, to establish the synthetic scope
of the methodology, and disclose the electrochemical behaviour
and chemical stability of those intermediates in RTILs; two
stable s -complexes and an s -complex were synthesised and
isolated in their solid form according to previously reported
Fig. 1 Cyclic voltammograms of 1 showing the reduction of a 5 mM
solution in DMF + 0.1 M TBABF4 and RTILs on a glassy carbon disk
(0.5 mm diameter), scan rate 1.0 V s-1.
H
X
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2531–2542 | 2535
©

View Article Online
Fig. 3 Cyclic voltammograms of [5-H]-, 5 mM in DMF + 0.1 M
TBABF4 and RTILs on a glassy carbon disk (0.5 mm diameter), scan
rate 1.0 V s-1 (Scan Range 0.0/0.5/-1.2/0.0 V).
forward scan, whereas a one electron reversible reduction wave
is shown in the cathodic back scan. It is important to remark
that when the CV is scanned first is in cathodic direction,
the reversible reduction wave does not exist while the two
electron irreversible wave appears unchanged on the reverse
scan; on the second cycle, the reduction wave is present in the
cathodic scan. This means that the reduction wave corresponds
to the electrochemical behavior of the product formed after the
Fig. 2 Cyclic voltammograms showing the reduction of a 5 mM
solution in BMIMAcO on a glassy carbon disk (0.5 mm diameter),
scan rate 1.0 V s-1 of a) 2 with time b) 4 initial.
H
oxidation of the s -complex [5-H]-. In order to characterize
the oxidation product a controlled potential electrolysis was
performed at a potential ca. 100 mV more positive than the
peak potential value determined for the s -complex. After the
H
procedures.30,31 The use of hydride (tetramethylammonium tetra-
borohydride) as a nucleophile and 2,4-dinitroaniline (5) as a
nitroaromatic reactant yields the tetramethylammonium salt of
passage of 2F and the chemical treatment of the sample, the
compound recovered is 2,4-dinitroaniline (5) in a 100% yield.
This experimental result shows that the electrochemically pro-
moted NASH process leads to the re-aromatizated compound
5 after the passage of 2 electrons and the loss of one proton,
which corresponds to the “formal” loss of a hydride in RTILs
(Scheme 5). It is important to highlight that the oxidation
peak is shifted depending on the RTIL used, which can be
explained in terms of solvation (Table 2). A stronger solvation
H
the s -complex [5-H]-, whereas the use of potassium methoxide
X
and 2,4,6-trinitroanisole (4) leads to the s -complex [4-OCH3]-.
From a qualitative point of view, the electrochemical be-
H
haviour of the s -complex [5-H]- in RTILs is very similar to
the one reported in organic solvent electrolyte systems (DMF
+ 0.1 M TBABF4) (Fig. 3).4 In all the RTILs investigated, an
irreversible two electron oxidation wave is observed in the anodic
Table 2 The anodic peak potential for s-complexes [5-H]- and [4-OCH3]-, standard potential and cathodic peak potential of the NASH and NASX
products obtained after the electrochemical oxidation of the sample in different solvents at 20 ◦
C
E◦(5) (V vs. SCE)b
H
s -complexes
SOLVENT
Epa (V vs. SCE)
NASH Product
[5-H]-
DMFa
0.210
0.326
0.365
0.208
0.369
5
5
5
5
5
-1.012
-0.903
-1.016
-1.009
-0.879
[BMIM]BF4
[BMIM]PF6
[BMIM]TSFI
[BMIM]AcOc
X
s -complexes
SOLVENT
E
pa(V vs. SCE)
NASX Product
Epc(6) (V vs. SCE)b
[4-OCH3]-
DMFa
1.217
1.170
0.968
1.045
0.947
4
4
4
4
4
-0.649
-0.674
-0.887
-0.716
-0.746
[BMIM]BF4
[BMIM]PF6
[BMIM]TSFI
[BMIM]AcOc
a Solution of DMF + 0.1 M TBABF4; b Standard Potential (E◦) or Cathodic Peak Potential (Epc) Values obtained after measuring a solution by cyclic
voltammetry, which has previously been oxidized by exhaustive controlled potential electrolysis at 1.0 V s-1; c After exhaustive controlled potential
electrolysis in [BMIM]AcO the acetate anion has been probably protonated and the produced acetic acid is a much weaker nucleophile. Moreover, it
could also have been oxidised so its concentration would be considerably decreased.
2536 | Green Chem., 2011, 13, 2531–2542
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 5 The electrochemically promoted NASH process in RTILs.
H
(a stronger ion pairing) of the s -complex by the cation of the
RTIL or of the supporting electrolyte would render its oxidation
H
more difficult (more positive potential). Thus the s -complex
would be more strongly solvated by the BMIM cation than by
H
tetrabutylammonium cation (in DMF). The s -complex would
be more strongly solvated in [BMIM]BF4 and [BMIM]AcO than
H
in [BMIM]TFSI because the s -complex would displace the
BF4- and AcO- anions more easily than the larger TFSI- anion.
At this point, it is possible to establish an electrocatalytic effect
when [BMIM]TFSI is used, since the electrochemical reaction
can be performed at a lower oxidation potential than in the other
RTILs, as well as in organic aprotic solvents. However, since
the [BMIM]TFSI is more expensive than the other two RTILs,
we decided to perform the electrochemical NASH approach,
which will be explained in the next sections, in [BMIM]BF4 or
[BMIM]PF6 in order to make scaling-up processes more feasible
in the future.
Analogous electrochemical studies have been performed for
X
the s -complex [4-OCH3]- so as to study the electrochemically
promoted NASX process in RTILs. The typical voltammogram
in DMF + 0.1 M TBABF4, at low scan rates, is shown in
Fig. 4a. No reduction waves are observed in a first reduction
scan, while an irreversible one electron wave appears in the
oxidation scan at 1.12 V vs. SCE. On the second cathodic
scan a reduction wave at -0.73 V is observed. This reduction
wave corresponds to the product formed in the first anodic scan
(Fig. 4a). The same general electrochemical behaviour described
in DMF + 0.1 M TBABF4 is observed when the experiments are
performed in all the RTILs (Fig. 4b–d). However, a closer look
at the cyclic voltammograms registered in RTILs reveals some
important differences. For instance, when the cyclic voltammetry
experiments are performed in [BMIM]PF6, the peak intensity
Fig. 4 Cyclic voltammograms of [4-OCH3]- 5 mM on a glassy carbon
disk (0.5 mm diameter), scan rate 1.0 V s-1 a) in DMF+0.1 M
TBABF4 (Scan Range 0.0/1.5/-1.0/0.0 V, 2 cycles) b) [BMIM]PF6
(Scan Range ca. 0.0/1.5/-1.0/0.0 V) c) [BMIM]PF6 (Scan Range ca.
0.0/-1.0/1.5/0.0 V) d) in [BMIM]BF4, [BMIM]TFSI and [BMIM]AcO.
X
for the oxidation wave corresponding to the s -complex [4-
OCH3]- corresponds to 0.55 electrons (Fig. 4b) instead of 1
electron.32 Moreover, the reduction wave corresponding to the
reduction of 4 at ca. -0.90 V is observed in the first cathodic scan
(Fig. 4c).33 It is therefore possible to conclude that methoxy is
a good leaving group in [BMIM]PF6, the extension of the SNAr
reaction following the classic addition–elimination mechanism
corresponds to a 45% yield (Scheme 6). The stability of the
intermediate can be increased using [BMIM]BF4, [BMIM]TFSI
or [BMIM]AcO, which means that the “chemical” route for
the SNAr reaction can be reduced to 20%, 18% or 15%
(Fig. 4d), respectively; in other words, the number of electrons
associated to the first oxidation wave is 0.80, 0.82 and 0.85,
respectively. In all cases, the SNAr can be electrochemically
completed by exhaustive controlled potential electrolysis at
ca. 1.30 V (a potential about 180 mV more positive than
the oxidation potential). The re-aromatizated product obtained
after the electrochemical oxidation is 2,4,6-trinitroanisole (4) in
X
a quantitative yield. So, the electrochemical oxidation of the s -
complex [4- OCH3]- leads to 4 after the loss of a methoxy radical.
Finally, it is also remarkable that a very efficient electrocatalytic
effect is found for the SNAr of heteroatom in [BMIM]PF6
or [BMIM]AcO since the oxidation potential is considerably
lower than in the other RTIL or aprotic solvents studied
(Table 2).
After disclosing that the electrochemical oxidation of the s-
complexes yields the substitution product by formally replacing
a hydride (H-) or halide (X-) following a NASH or NASX
mechanism in RTILs, it is time to establish the synthetic scope
and versatility of the synthetic methodology.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2531–2542 | 2537
©

View Article Online
Scheme 6 Comparison between the percentages of substitution product obtained either “chemically” or electrochemically promoted through
Nucleophilic Aromatic Substitution of Heteroatom Reaction (NASX process).
3.3. Electrochemically Promoted NASH Process
Fig. 5a and 5b show the cyclic voltammetry of a reaction
mixture of 2 (5 mM) and ketone (acetone or butanone, respec-
tively) in the presence of potassium tert-butoxide in a DMF-
RTIL mixture ([BMIM]BF4 and [BMIM]PF6, respectively)
prior to the electrochemical oxidation of the solutions. In
both cases, the nitroaromatic compound almost completely
reacted with the nucleophile since, in the first scan in the
cathodic direction, the intensity of the reduction wave corre-
sponding to 1,3-dinitrobenzene (2) is very low. Moreover, the
This section is devoted to reporting for the first time in terms of
cyclic voltammetry and preparative electrolysis on the scope and
limitations of electrochemical NASH in RTILs. These results
can be compared with those previously obtained by our group
using DMF + 0.1 M TBABF4 where the NASH product was
obtained when ketone enolates, cyanides and amines were used
as nucleophiles in the above-described conditions.7,9–10
H
quantitative formation of s -complexes, [2- CH2COCH3]- and
3.3.1. Electrochemical synthesis of nitroaromatic ketones and
nitriles. Formation of a C–C bond. The electrosynthesis of
nitroaromatic ketones in RTILs has been perfomed with com-
pounds 2 and 6 using acetone and 2-butanone as nucleophiles.
The base used to deprotonate the ketones to promote the nucle-
ophilic attack is potassium tert-butoxide in all cases. The order
of addition of the reactants is important in order to favour the
stoichiometric formation of the most stable carboanion present
in the reaction mixture and a fast quantitative nucleophilic
attack on it. The nitroaromatic compound is dissolved in a
large excess of ketone under nitrogen atmosphere. After that, the
potassium tert-butoxide is slowly and uniformly added. Then a
mixture of solvents DMF (without supporting electrolyte) and
RTILs (at a 1 : 1 ratio) is added. Finally, the sample is analyzed
and fully oxidized in terms of electrochemical techniques. The
addition of DMF makes the electrochemical oxidation process
faster to help with the conductivity of the media. The same
results are obtained when the reaction is performed in the
absence of DMF (Scheme 7).
Fig. 5 Cyclic voltammograms of a) 5.0 mM of “in situ” generated
[2- CH2COCH3]- and b) [2- CH(CH3)COCH3]- before (solid line)
and after (dotted line) exhaustive controlled potential electrolysis of
the sample in [BMIM]BF4 and [BMIM]PF6 RTILs, respectively, on a
glassy carbon disk (0.5 mm diameter), scan rate 1.0 Vs-1 (Scan Range
0.0/0.5/-1.2/0.0 V, 2 cycles).
Scheme 7 Electrochemically promoted NASH process to obtain sub-
stitution aromatic using ketones as a nucleophile in basic media.
2538 | Green Chem., 2011, 13, 2531–2542
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
H
Table 3 Exhaustive electrolysis of s -complex, NASH processes
H
Epa s -complex
NASH
Product (%)
H
Compound
Solvent
Nucleophile (number of eq.)
s -complex (%)
(V vs. SCE)
0.47
2
DMF + 0.1 M
TBABF4
tBuOK/CH2COCH3 (1 : 160 eq)
[2-CH2COCH3]- (100)a
90
2
2
6
6
3
BMIMBF4
BMIMBF4
BMIMPF6
BMIMPF6
DMF + 0.1 M
TMAPF6
BMIMBF4
DMF + 0.1 M
TBABF4
tBuOK/CH2COCH3 (1 : 160 eq)
tBuOK/CH3COCH2CH3 (1 : 18 eq)
tBuOK/CH2COCH3 (1 : 160 eq)
tBuOK/CH3COCH2CH3 (1 : 12 eq)
KCN (1 : 2.35 eq)
[2-CH2COCH3]- (93)a
[2-CH(CH3)COCH3]- (100)a
[6-CH2COCH3]- (95)a
[6-CH(CH3)COCH3]- (80)a
[3-CN]- (82)a
0.314
0.449
0.233
0.294
1.040
82
100
80
91
80
3
3
TEACN (1 : 1.5 eq)
n-BuNH2 (1 : 3 eq)
[3-CN]- (100)a
0.650
1.12
100
30
[3-NH-(CH2)3CH3]- (30)b
3
3
BMIMBF4
BMIMAcO
n-BuNH2 (1 : 12 eq)
n-BuNH2 (1 : 63 eq)
[3-NH-(CH2)3CH3]- (100)b
[3-NH-(CH2)3CH3]- (94)b
0.755
0.710
25
94
H
a Percentage s -complex calculated by cyclic voltammetry (oxidation peak potential and intensity of the remaining nitroarene reduction wave) b Since
H
the excess of amine can modify the intensity of the oxidation wave; the percentage s -complex was only calculated from the intensity of the remaining
nitroarene reduction in the cyclic voltammogram.
[2- CH(CH3)COCH3]-, can be deduced from the peak current
value of the oxidation wave associated to each intermediate
(0.45 and 0.35 V, respectively). In a second cathodic scan, the
reduction wave presumably corresponding to the substitution
product is observed at ca. -1.00 V vs. SCE. Finally, controlled-
tetramethylammonium hexafluorophosphate (TMAPF6) even
when reducing the number of nucleophile equivalents (Table 3).
3.3.2. Electrochemical synthesis of nitroaromatic amines.
Formation of a C–N bond. For the study of the electrochemical
synthesis of nitroanilines by means of a NASH the selected
nitroaromatic compound is 3 and the RTILs are [BMIM]BF4
and [BMIM]AcO. Previous studies performed in DMF + 0.1 M
TBABF4 demonstrated that the use of an excess of amine
leads to the substituted nitroaromatic products in moderate
yields (30–40%).6 The first RTIL studied is [BMIM]BF4. A
controlled excess of n-butylamine (12 equivalents) is added
to a solution of 3 in 200 ml of DMF (Table 3). When the
reaction is complete, which means that there is no unreacted
H
potential electrolysis are performed at ca. 0.6 V, after the s -
complex oxidation wave. The cyclic voltammograms of the
H
oxidation mixture show, in both cases, the oxidation of the s -
complex (no oxidation wave is detected) and the consequent
formation of the substitution product (reduction wave). Analysis
of the oxidised reaction mixture, after the chemical treatment
of the sample (see Experimental Section), makes it possible to
obtain the substitution product in high yields (80–100%).
It is important to remark that the reaction is totally selective,
only one C-arylation product is obtained when compounds 2
and 6 react either with acetone or with 2-butanone. In the case
of using 2-butanone, the retroanalysis of the product obtained, 3-
(2,4-dinitrophenyl)-2-butanone or 3-(2,4-dinitro-1-naphthyl)-2-
butanone, when 2 and 6 react with 2-butanone, indicates not only
that the reaction takes place with the most stable enolate but also
nitroaromatic material, 5 mL of [BMIM]BF4 is added.34
A
controlled potential electrolysis of the mixture leads to 25% of
N-butyl-2,4,6-trinitroaniline, the NASH substitution product.
In order to improve the reaction yield, a larger excess of amine
is added. However, this did not improve the yield. The fact that
yields are not ameliorated when a huge excess of amine is added,
and also that there is no presence of unreacted starting material,
seems to indicate that most of the nitroaromatic compound
is mainly in a zwitterionic form (Scheme 8).35 Therefore, to
increase the reaction yield it would be mandatory to shift the
H
on the most stable s -complex. This indicates that the reaction
is operating under thermodynamic conditions. The results of the
electrochemically promoted NASH process are summarized in
Table 3, excellent yields of the substitution product are obtained
in all cases. At this point, it is worth concluding that in spite of
the good results obtained in DMF + 0.1 M of TBABF4, it is
possible to obtain similar results using RTILs and avoid the use
of large quantities of supporting electrolyte and even reduce the
quantity of nucleophile, either in the presence or the absence of
DMF.
H
equilibrium from the zwitterionic complex to the s -complex.
At this point the use of [BMIM]AcO as solvent seems to be the
most appropriate RTIL due to the basicity of the acetate group.
Analysis of the reaction mixture after exhaustive controlled
potential electrolysis at 1.00 V replacing the [BMIM]BF4 for with
The NASH process in RTILs is also applied to the cyanation
of nitroarenes. Salts of either tetraethylammonium or potassium
cyanide are used as a nucleophile and 1,3,5-trinitrobenzene
(3) as a nitroarene. The best results are obtained for an
electrolyzed mixture of 3 in the presence of 1.5 equivalents
of tetraethylamonium cyanide using [BMIM]BF4 as a solvent.
It is important to highlight that the best yield obtained in
RTILs is 20% higher than the yield obtained in DMF + 0.1 M
Scheme 8 A NASH electrochemically promoted process for obtaining
substituted aromatic compounds using amines as nucleophiles in RTILs.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2531–2542 | 2539
©

View Article Online
[BMIM]AcO yields 94% of the expected substitution product.36
Moreover, when RTILs are used an important electrocatalytic
effect is observed, since the peak potential of the s -complex is
SNAr substitution of a heteroatom when cyanide is used as a
nucleophile.14,37
H
Scheme 9 shows that the cyanide can attack three different
positions of the aromatic ring, one non-substituted ([4-CN-rH]-)
and two substituted ([4-CN,OCH3-rX]- or [4-CN,NO2-rX]-). No-
tice that the cyanide mainly attacks a non-substituted position
when [BMIM]BF4 is used as a pure solvent, and the oxidation
of the corresponding [4-CN-rH]- leads to the NASH product.
When 5% of DMF is added to the [BMIM]BF4, the cyanide
exclusively attacks position 1 of the nitroaromatic ring leading
to [4-CN,OCH3-rX]-, so a controlled potential electrolysis of
the solution affords 2,4,6-trinitrobenzonitrile quantitatively. On
the other hand, if the [BMIM]BF4 is replaced with [BMIM]PF6
and the percentage of DMF presented in solvent mixture is
increased to 20%, the main intermediate formed is [4-CN,NO2-
rX]-, so the final product obtained after oxidation of the sample
mixture is 50% 2-methoxy-3,5-dinitrobenzonitrile. Finally, it is
important to highlight that, as previously demonstrated for the
anion radical and dianion of nitroaromatic compounds, the
solvation effects of the RTILs can modify the relative thermo-
dynamic stability of the s–complexes introducing an amazing
regioselective effect. The electrochemical oxidation can take
advantage of this effect by promoting the quick departure of the
leaving group. So, the environmental advantages of using RTILs
as green solvents and the electrochemical methods to promote
nucleophilic aromatic substitution as a green technology, open
an excellent route to overcoming the main drawbacks associated
with the SNAr reaction.
Finally, in the analysis of the feasibility and viability of
replacing organic aprotic solvents that contain 0.1 M tetra-
alkylammonium salts (supporting electrolyte) by RTILs, it is
important to take into account the following points: cost of each
experiment DMF + 0.1 M of TBABF4 and cost of the RTILs
either commercially available or synthesized in the laboratory.
So, the expense of one single experiment in a commercially
available RTIL, such as [BMIM]BF4, is almost twice that
of performing it in DMF + 0.1 M of TBABF4. However,
if the RTILs are synthesized in the laboratory, in all cases
([BMIM]BF4, [BMIM]PF6, [BMIM]TFSI and [BMIM]AcO),
the cost of the experiment is almost the same as in DMF +
0.1 M of TBABF4. However, since the RTILs are recyclable,
which means that at the end of the experiment 60–70% can be
recovered and reused again, the cost of the following experiments
are reduced by more than 60% in comparison with the initial
charges in DMF + 0.1 M TBABF4.
shifted to less positive potentials compared to those measured
in DMF (see Table 3).
3.4. Electrochemically promoted NASX process
The electrochemical NASX reaction is less attractive than
the NASH reaction in terms of designing an environmentally
friendly route, since the leaving group is different from hydrogen.
However, it can be a nice alternative to overcome the low yields
obtained when the electrochemical NASH approach is used.
This situation can be illustrated by, for instance, synthesizing N-
butyl-2,4,6-trinitroaniline in [BMIM]BF4. As described in the
previous section, the reaction of 3 with N-butylamine affords
25% substitution product through an electrochemically pro-
moted NASH process. However, it is possible to obtain 88% of
the N-butyl-2,4,6-trinitroaniline in [BMIM]BF4 using an excess
(21 equivalents) of N-butylamine (nucleophile) and 1-chloro-
2,4,6-trinitrobenzene (7), as nitroaromatic starting materials
X
after controlled potential electrolysis of the s -complex formed
in the reaction mixture ([7-Cl,NH-(CH2)2CH3]-). Better results
are obtained when [BMIM]PF6 is used as a solvent, where
it is possible to reach almost 100% yield by changing the
electrochemical solvent and increasing the number of amine
equivalents five times (Table 4). As in the previously described
case of intermediate [4- OCH3]-, it is possible that part of the
departure of chloride in the intermediate [7-Cl,NH-(CH2)2CH3]-
can take place spontaneously, this route being more important
in [BMIM]PF6 than in [BMIM]BF4. Note that the oxidation
X
potential of the s -complex is considerably lower than in DMF
or [BMIM]BF4, which makes this solvent extremely attractive
for performing NASX reactions due to its electrocatalytic effects.
The electrochemical synthesis of 2,4,6-trinitrobenzonitrile can
also be designed using a NASX process. It could be expected
that the reaction of 2,4,6-trinitroanisole (4) with cyanide would
lead to the 2,4,6-trinitrobenzonitrile. However, this reaction
is strongly dependent on the solvent. The use of different
solvents or mixtures of solvents modify the regioselectivity of
the products (Scheme 9, Table 4). This study demonstrates the
power of electrochemistry and RTILs in terms of selectivity,
good yields and environmentally friendly routes. It is important
to remark that as far as we are aware there is no spontaneous
X
Table 4 Exhaustive electrolysis of s -complex, NASX processes
H
Nucleophile
(number of eq.)
Epa s -complex
X
Compound
Solvent
s -complex (%)
(V vs. SCE)
1.350
NASX Product (%)
7
DMF + 0.1 M
TBABF4
BMIMBF4
BMIMPF6
BMIMBF4
BMIMBF4 + 5%
DMF
n-BuNH2 (2 eq)
[7-Cl,NH-(CH2)3CH3]- (70)
95
7
7
4
4
n-BuNH2 (21 eq)
n-BuNH2 (103 eq)
TEACN (2.35 eq)
TEACN (3.5eq)
[7-Cl,NH-(CH2)3CH3]- (40)
[7-Cl,NH-(CH2)3CH3]- (100)
[4-CN-rH]- (50)
1.267
0.686
0.600
0.75
88
91
96(NASH product)
100
[4-CN,OCH3-rX]- (100)
4
BMIMPF6 + 20%
DMF
TEACN (6.0eq)
[4-CN,NO2-rX]-
[4-CN,OCH3-rX]-
[4-CN-rH]-
0.45
0.75
0.60
50
16
34
2540 | Green Chem., 2011, 13, 2531–2542
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 9 NASH and NASX in different RTILs when cyanide is used as a nucleophile.
used. The use of controlled potential electrolysis makes it
4. Conclusions
possible to fast and selectively complete the desired oxidation
product. The results obtained are similar or even higher than
those obtained in organic solvent electrolyte systems (DMF
+ 0.1 M TBABF4), the experimental conditions used being
less polluting. In summary, the described methodology is a
friendly environmental route, which is economically feasible
and accomplishes most of the important principles of green
chemistry, such as safer solvents (non-flammable, non-volatile
and thermally stable over a wide range of temperatures), low
cost of the reactant, ready availability of the reagents, atom
The use of RTILs as a solvent for performing electrochem-
ically promoted nucleophilic aromatic substitution reactions
of hydrogen (NASH) or of a heteroatom (NASX) opens an
attractive “green route” for synthesizing substituted nitroaro-
matic compounds. Cyclic voltammetry makes it possible to
evaluate not only the nature and stability of the s-complex
but also the extension of the reaction in RTILs. Moreover,
it has been demonstrated that the stability and nature of
the s-complexes can be modulated depending on the RTIL
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2531–2542 | 2541
©

View Article Online
economy, high yields (approaching 100% based on unrecovered
starting material). We are currently working on the extension of
this methodology to new RTILs and substrates.
19 G. Angelini, C. Chiappe, P. De Maria, A. Fontana, F. Gasparini, D.
Pieraccini, M. Pierini and G. Siani, J. Org. Chem., 2005, 70, 8193.
20 M. C. Buzzeo, C. Hardacre and R. G. Compton, ChemPhysChem,
2006, 7, 176.
21 F. D’Anna, V. Frenna, R. Noto, V. Pace and D. Spinelli, J. Org.
Chem., 2006, 71, 5144.
22 C. B. Reese and Z. Pei-Zhuo, Z., J. Chem. Soc., Perkin Trans. 1, 1993,
2291.
Acknowledgements
23 K. B. Lam, J. Miller and P. J. Samenho, J. Chem. Soc., Perkin Trans.
We gratefully acknowledge the financial support of the Ministe-
rio de Educacio´n y Ciencia of Spain though projects CTQ2006-
01040 and CTQ 2009-07469. The authors greatly appreciate
an anonymous referee for his/her valuable comments and
corrections.
2, 1977, 457.
24 M. Trivedi, S. K. Singh, D. S. Pandey, M. C. Puerta and P. Valerga,
Transition Met. Chem., 2005, 30, 861.
25 S. T. Handy and X. Zing, Org. Lett., 2001, 3(2), 233.
26 P. Bonhoˆte, A. Dias, N. Kalyanasundaran and M. Gra¨tzel, Inorg.
Chem., 1996, 35, 1168.
27 H. Cruz, I. Gallardo and G. Guirado, Electrochim. Acta, 2008, 53,
5968.
28 A. J. Fry, J. Electroanal. Chem., 2003, 546, 35.
29 The color of the solution is characteristic of the formation in solution
References
H
1 J. March, “Advanced Organic Chemistry”, 4th. edn, Wiley, New York,
1992, Chapter 13, 641.
2 E. Buncel, J. M. Dust and F. Terrier, Chem. Rev., 1995, 95,
2261.
of s -complexes. For more information about the UV-Vis features
H
of s -complexes see references 1–4.
30 J. Atkins, V. Gold and W. N. Wassef, J. Chem. Soc., Perkin Trans. 2,
1983, 1197.
3 O. N. Chupakhin, V. N. Chupakhin, H. C. Van der Plas, “Nucleophilic
31 W. E. Byrne, E. J. Fendler, J. H. Fendler and C. E. Griffin, J. Org.
Aromatic Sustitution of Hydrogen”, Academic Press, 1994.
Chem., 1967, 32, 2506.
4 F. Terrier, “Nucleophilic Aromatic Displacement”, ed. H. Feuer, VCH,
New York, 1991.
32 The value of the number of electrons involved in each electron
transfer in the different media used are calculated by comparison
with the oxidation of redox probes, i.e. tris(4-bromophenyl)-amine,
for more information see reference 27.
33 The presence of a reduction pre-peak between -0.5 and 0.0 V,
depending on the RTILs, is related with reduction 4 in the presence
of potassium. See reference 14.
5 M. Hamana, G. Iwasaki and S. Saeki, Heterocycles, 1982, 17, 177.
6 G. Iwasaki, K. Wada, S. Saeki and M. Hamana, Heterocycles, 1984,
22, 1811.
7 I. Gallardo, G. Guirado and J. Marquet, Chem.–Eur. J., 2001, 7,
1759.
8 G. Moutiers, J. Pinson, F. Terrier and R. Goumont, Chem.–Eur. J.,
34 The absence of the reduction wave associated with the 1,3,5-
trinitrobenzene in the cyclic voltammogram indicates that there is
no “free” nitroaromatic compound in solution.
2001, 7, 1712.
9 I. Gallardo, G. Guirado and J. Marquet, Eur. J. Org. Chem., 2002, 2,
251.
35 The fact that it is not possible to shift the equilibrium from reactants
to products even when using an excess of amine, indicates that most
10 I. Gallardo, G. Guirado and J. Marquet, Eur. J. Org. Chem., 2002, 2,
H
261.
of the intermediate is in zwitterionic complex form rather than s -
11 I. Gallardo, G. Guirado, J. Marquet, Patent Pending, ES2000/489.
complex. The oxidation of the zwitterionic complex leads to the
H
12 I. Gallardo, G. Guirado and J. Marquet, J. Org. Chem., 2003, 68,
reactant, whereas the oxidation of the s -complex yields the NASH
631.
products. It is also possible that some important part of the amine in
excess could be protonated in RTILs and it would not contribute to
shifting the equilibrium and increasing the reaction yield.
13 I. Gallardo, G. Guirado and J. Marquet, J. Org. Chem., 2003, 63,
7334.
14 I. Gallardo, G. Guirado and J. Marquet, J. Org. Chem., 2002, 67,
36 Exhaustive controlled potential electrolysis is performed until there
is no presence of an oxidation wave. Since the reaction takes place in
[BMIM]AcO, it is possible that the anion attacks the NASH product.
This is not a problem in terms of NASH product yield. Although the
time of electrolysis is longer, the electrochemical oxidation of the
s -complex (NASH product-acetate) provokes the departure of the
acetate.
37 The absence of the reduction wave associated to the 2,4,6-
trinitroanisole after 1 h of preparation of the reaction mixture
indicates that there is no “free” nitroaromatic compound in solution.
2548.
15 I. Gallardo, G. Guirado and J. Marquet, J. Electroanal. Chem., 2000,
488, 64.
16 A. S. Matlack, Introduction to Green Chemistry, Marcel Dekker, New
York, 2001.
H
17 H. Ohno, “Electrochemical Aspects of Ionic Liquids”, Wiley-
Interscience, New Jersey, 2005.
18 C. Lagrost, L. Preda, E. Volanschi and P. Hapiot, J. Electroanal.
Chem., 2005, 585, 1.
2542 | Green Chem., 2011, 13, 2531–2542
This journal is
The Royal Society of Chemistry 2011
©