Changed reactivity of the 1-bromo-4-nitrobenzene radical anion in a room temperature ionic liquid

Article abstract of DOI:10.1039/c3cp51004b

Radical anions of 1-bromo-4-nitrobenzene (p-BrC₆H ₄NO₂) are shown to be reactive in the room temperature ionic liquid N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide, ([C₄mPyrr][NTf₂]), by means of voltammetric measurements. In particular, they are shown to react via a DISP type mechanism such that the electrolysis of p-BrC₆H₄NO₂ occurs consuming between one and two electrons per reactant molecule, leading to the formation of the nitrobenzene radical anion and bromide ions. This behaviour is a stark contrast to that in conventional non-aqueous solvents such as acetonitrile, dimethyl sulfoxide or N,N-dimethylformamide, which suggests that the ionic solvent promotes the reactivity of the radical anion, probably via stabilisation of the charged products.

Full text of DOI:10.1039/c3cp51004b

PCCP
View Article Online
View Journal | View Issue
PAPER
Changed reactivity of the 1-bromo-4-nitrobenzene
radical anion in a room temperature ionic liquid
Sven Ernst,a Kristopher R. Ward,a Sarah E. Norman,b Christopher Hardacreb and
Richard G. Compton*a
Cite this: Phys. Chem. Chem. Phys., 2013,
15, 6382
Radical anions of 1-bromo-4-nitrobenzene (p-BrC6H4NO2) are shown to be reactive in the room
temperature ionic liquid N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide, ([C4mPyrr][NTf2]),
by means of voltammetric measurements. In particular, they are shown to react via a DISP type
mechanism such that the electrolysis of p-BrC6H4NO2 occurs consuming between one and two electrons
per reactant molecule, leading to the formation of the nitrobenzene radical anion and bromide ions. This
behaviour is a stark contrast to that in conventional non-aqueous solvents such as acetonitrile, dimethyl
sulfoxide or N,N-dimethylformamide, which suggests that the ionic solvent promotes the reactivity of the
radical anion, probably via stabilisation of the charged products.
Received 6th March 2013,
Accepted 14th March 2013
DOI: 10.1039/c3cp51004b
www.rsc.org/pccp
The area of radical ion (cation and anion) chemistry is well-
explored voltammetrically in conventional solvents.8–17 Such
Introduction
Room temperature ionic liquids show significantly changed
properties in comparison to conventional, molecular, non-
aqueous solvents in terms of physical properties1 tuneability,2,3
greenness,4 recyclability5 and negligible room temperature
volatility.6 Accordingly, they have been increasingly adopted
in synthetic procedures ranging from laboratory benchtops to
scaled up industrial reactions.7 On the other hand, the
measurement of physiochemical reaction parameters, both
kinetic and thermodynamic, has lagged significantly.8
In a previous paper we showed the use of voltammetry as a
ready means of obtaining quantitative data on a variety of
processes in RTILs.8 In the present paper we develop this
approach to probe changed solute reactivity between RTIL
solvents and conventional, molecular, aprotic solvents such
as acetonitrile, dimethyl sulfoxide, N,N-dimethylformamide,
etc. In particular, cyclic voltammetry can easily fingerprint
a stable species generated electrochemically so that this
approach is well-suited for identifying changed reactivity
between these different solvent systems. Specifically, we seek
to identify species which are reactive in RTILs but are (at least
kinetically) stable in conventional media.
measurements show that, for example, the radical anion of
nitrobenzene (C6H5NO2 ꢁ) is stable both in ionic liquid media18
ꢀ
and in aprotic molecular solvents.19–21 Complementary to the
electrochemical data, electron spin resonance (ESR) confirms
the formation of a persistent radical anion species.22,23 Similarly,
nitrotoluene radical anions are also seen to be generally stable.24
On the other hand, the introduction of halogen substituents
onto the nitrobenzene ring can have different effects. Introduc-
tion of an iodine substituent leads to radical ions which are
usually reactive, whereas chlorinated species are much more
stable.25,26 In the case of the bromo-nitrobenzenes, the radical
anion stability depends on the site of the substituent with
respect to the ortho-, meta- or para-positions.25–27 However,
1-bromo-4-nitrobenzene (p-BrC6H4NO2) radical anions, shown
in Fig. 1, are stable in acetonitrile, dimethyl sulfoxide and
N,N-dimethylformamide.19–21 In the following, we investigate the
voltammetry of p-BrC6H4NO2 in the RTIL N-butyl-N-methylpyrrol-
idinium bis(trifluoromethylsulfonyl)imide ([C4mPyrr][NTf2]) and
a Department of Chemistry, Physical and Theoretical Chemistry Laboratory,
Oxford University, South Parks Road, Oxford OX1 3QZ, UK.
E-mail: richard.compton@chem.ox.ac.uk; Fax: +44 (0)1865 275410;
Tel: +44 (0)1865 275 (413 or 957)
b School of Chemistry and Chemical Engineering/QUILL, Queen’s University Belfast,
Belfast, Northern Ireland BT9 5AG, UK
Fig. 1 Molecular structure of 1-bromo-4-nitrobenzene (p-BrC6H4NO2).
c
6382 Phys. Chem. Chem. Phys., 2013, 15, 6382--6389
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
show that, in contrast, it undergoes reduction leading to T-shaped cell was then connected to a rotary vacuum pump and
debromination and the loss of Brꢁ with the formation of the the sample was placed under vacuum (P = 4.0 ꢂ 0.1 mbar) for at
radical anion of nitrobenzene via a DISP-type mechanism, with least 2 hours before measurement and remained in vacuo for
the reaction occurring on the timescale of seconds.
the duration of the experiment. The system was studied
in vacuo to reduce the water33 and oxygen34 concentrations in
the RTIL to acceptably low levels. Cyclic voltammograms (CVs)
were background-subtracted (using a linear baseline from the
Experimental
N-Butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide non-Faradaic region) and filtered using an FFT function, within
([C4mPyrr][NTf2], electrochemically pure) was synthesised, as the GPES software, to remove high frequency electrical noise
reported previously,28 from the corresponding bromide salt and Modelling of voltammetric data was carried out using software
analysed by 1H-NMR and 13C-NMR. Nitrobenzene (C6H5NO2, described elsewhere.8
>99%), bromobenzene (C6H5Br, 99%), 1-bromo-4-nitrobenzene
(p-BrC6H4NO2, 99%), silver trifluoromethanesulfonate (Ag[OTf],
>99.95%) and Ferrocene (Fc, 98%) were obtained from Aldrich.
Acetonitrile (MeCN, Z99.9%) was obtained from Sigma-Aldrich.
Results & discussion
All reagents were used as received without further purification.
Electrochemical measurements were performed in a thermo-
stated (25.0 ꢂ 0.1 1C) Faraday cage, using a mAutolab Type III
computer-controlled potentiostat (Ecochemie). All measurements
were obtained using a standard three electrode configuration,
utilising platinum (Pt, r E 5 mm, BASi) working electrodes, a
platinum wire (99.9%, Goodfellow) counter electrode and a Ag/Ag+
reference electrode (see below). Prior to use the working
electrode was polished using 1, 0.3 and 0.05 mm alumina
lapping compounds (Buehler) on soft lapping pads (Buehler)
before being placed in deionised water (resistivity B 18.2 MO cm at
25 1C, Millipore water systems) in an ultrasonic bath for B3 minutes
to remove any adsorbed material. The working electrode was
then rinsed thoroughly with MeCN and dried using compressed
nitrogen. Before use, the counter electrode was cleaned in a hot
Bunsen flame before being rinsed thoroughly with MeCN.
Electrode radii were calibrated by measuring chronoampero-
metric transients in a 2 mM solution of ferrocene in MeCN and
then fitting the transients using the Shoup–Szabo approxi-
mation29 and the literature value of 2.3 ꢃ 10ꢁ9 m2 sꢁ1 for the
diffusion coefficient of Fc.30 Chronoamperometric transients
measured in [C4mPyrr][NTf2] were obtained by first holding the
potential at ꢁ1.0 V for 5 s and then stepping to a potential of
ꢁ2.0 V and recording the current response for 5 s, using a
sample time of 0.01 s. The Ag/Ag+ reference electrode was
prepared following a modified version of the method of Meng
et al.31 which involved attaching a 5 Å, 1.6 mm diameter
molecular sieve (Aldrich) to the end of a cut-down glass Pasteur
pipette (VWR International Ltd) using PTFE heatshrink
(TE Connectivity). The Ag/Ag+ reference electrode was then filled
with a 20 mM solution of Ag[OTf] in [C4mPyrr][NTf2] and a silver
wire was immersed in the solution. This electrode then yielded a
reproducible formal potential for the ferrocene|ferrocenium
couple of ꢁ363 ꢂ 4 mV at the Pt electrode in [C4mPyrr][NTf2].
Experiments were prepared by placing 50 mL of the desired
sample solution into a sample chamber on top of the electrode,
constructed by attaching a section of disposable micropipette
tip (VWR International Ltd) over the electrode tip. The working
electrode and sample were then placed into a glass T-shaped
cell, described previously,32 and the counter and reference
electrodes fed into the sample from above. One arm of the
In order to better understand and simulate the reduction of
p-BrC6H4NO2 in the [C4mPyrr][NTf2] ionic liquid, we first
investigate the electrochemical reduction C6H5NO2 and
C6H5Br in [C4mPyrr][NTf2].
Electrochemical reduction of nitrobenzene in [C4mPyrr][NTf2]
A 15.9 mM solution of C6H5NO2 in [C4mPyrr][NTf2] was placed
under vacuum to remove oxygen and water, which interfere
with electrochemical experiments in RTILs.33,34 Cyclic voltammo-
grams were then recorded for the reduction of nitrobenzene
in the [C4mPyrr][NTf2] ionic liquid. Fig. 2 shows a CV of
15.9 mM nitrobenzene in [C4mPyrr][NTf2], recorded at a scan rate
of 100 mV sꢁ1, where the potential was scanned between ꢁ1 and
ꢁ2.05 V. The CV shows single reduction and oxidation peaks at
ꢁ1.91 and ꢁ1.78 V vs. the Ag/Ag+ reference, respectively.
Chronoamperometric transients were then recorded for the
reduction of nitrobenzene and the experimental data was fitted
using the Shoup–Szabo approximation29 to simultaneously
determine the diffusion coefficient, D, and the number of
electrons, n, transferred in the reduction. Fig. 2 (inset) shows
Fig. 2 Cyclic voltammogram of 15.9 mM C6H5NO2 in [C4mPyrr][NTf2] on a Pt
microdisk (radius = 5.1 mm) at a scan rate of 100 mV sꢁ1. Inset: Chronoampero-
metric transients for the reduction of 15.9 mM C6H5NO2 in [C4mPyrr][NTf2] on a
Pt microdisk (radius = 5.4 mm), showing the experimental data (circles) and
Shoup–Szabo fit (line).
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 6382--6389 6383

View Article Online
PCCP
Paper
a typical chronoamperogram for the reduction of the 15.9 mM
C6H5NO2 solution.
As shown in Fig. 3, the experimental data is well simulated as a
one-electron reduction using the values obtained via chrono-
As the radius of the electrode and the initial concentration amperometry and Tafel analysis as a starting point. The reduction
of the solution were known, fitting via the Shoup–Szabo expres- was thus assigned as the one electron reduction of nitrobenzene
sion allows for the simultaneous determination of D and nc. to its radical anion, as has been previously shown for nitro-
Chronoamperometric transients measured at B10 minutes benzene in a range of aprotic solvents19,37,38 and in the ionic liquid
intervals showed that nc was found to decrease with time under 1-butyl-2,3-dimethylimidazolium bis(trifluoromethylsulfonyl)imide,
vacuum as the compound was slowly removed from solution, [C4dmim][NTf2].18 The fitting parameters used in the simula-
however, a diffusion coefficient of (3.4 ꢂ 0.1) ꢃ 10ꢁ11 m2 sꢁ1 was tion of C6H5NO2 are used below for the interpretation of the
reproducibly obtained, showing that D is independent of concen- voltammetry of p-BrC6H4NO2.
tration. Plotting nc vs. t then allowed for the extrapolation of nc to
t = 0, where c = c0, and thus n was determined as approximately
equal to one, which suggests that the electrode process is;
Electrochemical reduction of bromobenzene in [C4mPyrr][NTf2]
A 12.8 mM solution of C6H5Br in [C4mPyrr][NTf2] was then
ꢁ
C6H5NO2 + eꢁ " C6H5NO2
(1.1)
ꢀ
placed under vacuum and investigated via cyclic voltammetry.
However, no Faradaic features were visible in the resulting
where the stability of the radical anion is suggested by the voltammetry. This was rationalised as the result of fast loss of
pronounced backpeak in Fig. 2. material from the solution phase into the gas phase, given the
In order to confirm that n = 1, CVs were simulated as a single larger vapour pressure and lower boiling point reported for
electron transfer, as suggested by the chronoamperometric C6H5Br than for C6H5NO2. The vapour pressure at 25 1C and
data. The value for the electron transfer coefficient, a, was the boiling point at atmospheric pressure of nitrobenzene are
calculated, via Tafel analysis35,36 of the voltammetry, to be 0.274 Torr (0.365 mbar) and 210.8 ꢂ 0.0 1C, respectively, while
0.63 ꢂ 0.03 and this value was incorporated in the model, for bromobenzene these values are 4.12 Torr (5.49 mbar)
along with the value for D(C6H5NO2) of 3.4 ꢃ 10ꢁ11 m2 sꢁ1 and 154.2 ꢂ 9.0 1C, respectively (Calculated using Advanced
determined via chronoamperometry. The concentration of the Chemistry Development (ACD/Labs) Software V11.02 (r 1994–2013
solution at the time that the voltammetry was measured could ACD/Labs), via Scifinder). Therefore, in order to measure the
be estimated from the chronoamperometric data and thus qualitative voltammetry of C6H5Br, solutions of much higher
these values were also incorporated into the simulation. The concentration were investigated so that a small amount of material
voltammetry was initially simulated with D(C6H5NO2 ꢁ) = remained in solution for voltammetric study, albeit of unknown
ꢀ
D(C6H5NO2), however the simulated backpeaks did not match concentration. Fig. 4 shows the electrochemical reduction of 0.95 M
the experimental voltammetry. Given that the diffusion coefficients C6H5Br in the [C4mPyrr][NTf2] ionic liquid.
of charged species have been shown to significantly differ from
Apparent in Fig. 4 is a single reduction peak at ꢁ2.07 V with
1
ꢁ
ꢀ
their neutral counterparts in RTILs in many cases, D(C6H5NO2
)
no accompanying backpeak, as well as a crossover in the forward
was then considered as a fitting parameter. The fitting parameters and reverse sweep currents indicative of a change of the electrode
were thus the anion radical diffusion coefficient, D(C6H5NO2 ꢁ) = surface. Furthermore, the reductive wave diminishes and shifts to
ꢀ
(1.1 ꢂ 0.1) ꢃ 10ꢁ11 m2 sꢁ1, the heterogeneous rate constant, k0 = more negative potentials on successive scans until the signal
(7 ꢂ 1) ꢃ 10ꢁ5 m sꢁ1, and the formal potential, E0 = ꢁ1.81 ꢂ 0.01 V. is no longer detectable. These observations together then imply
Fig. 3 shows a comparison between the experimental and a modification of the surface.39 Due to this surface fouling, no
simulated voltammetry at 10 and 100 mV sꢁ1 scan rates.
further experiments were performed using the C6H5Br compound.
Fig. 3 Comparison of experimental (circles) and simulated (lines) data for the reduction of nitrobenzene at scan rates of (a) 10 mV sꢁ1 and (b) 100 mV sꢁ1
.
c
6384 Phys. Chem. Chem. Phys., 2013, 15, 6382--6389
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
by an adsorption–desorption type process.36 CVs were then
recorded at a range of scan rates between 5 and 1000 mV sꢁ1
as shown in Fig. 6.
,
Fig. 6 shows the effects of scan rate, n, on the voltammetry of
p-BrC6H4NO2 in the [C4mPyrr][NTf2] ionic liquid. It is apparent
from the figure that as the scan rate is increased the voltam-
metry becomes increasingly well-defined. This indicates that
the reduction wave shoulder and faster than expected current
decay after the backpeak are likely caused by a combination of
adsorption/desorption, as at faster scan rates these processes
are relatively outrun.
Chronoamperometric transients were then measured for the
reduction of p-BrC6H4NO2 and the experimental data was fitted
using the Shoup–Szabo approximation to simultaneously determine
the diffusion coefficient and the number of electrons transferred in
the reduction. Fig. 5 (inset) shows a typical chronoamperogram
for the reduction of 5.4 mM p-BrC6H4NO2.
Fig. 4 Cyclic voltammetry of 0.95 M C6H5Br in [C4mPyrr][NTf2] on a Pt microdisk
(radius = 5.4 mm) at a scan rate of 10 mV sꢁ1
.
As described above and elsewhere,29,40 fitting the experi-
mental transient via the Shoup–Szabo approximation allows
for the simultaneous determination of D and nc. Analogous
to the experimental procedure for C6H5NO2, long timescale
chronoamperometric transients were measured at B10 minutes
intervals and each transient was independently fitted using the
Shoup–Szabo relation. The fitting of transients with short
measurement times did not yield reproducible results and hence
transients were measured for 5 s. The fitting of transients with
long measurement times then yielded a reproducible diffusion
Nevertheless, the potentials at which there is an onset of
C6H5Br reduction were noted as being significantly more
negative than for nitrobenzene. This is important in rationalising
the reduction of p-BrC6H4NO2, to which we next turn.
Electrochemical reduction of 1-bromo-4-nitrobenzene in
[C4mPyrr][NTf2]
After studying the electrochemistry of C6H5NO2 and C6H5Br in
the [C4mPyrr][NTf2] ionic liquid, the voltammetry of p-BrC6H4NO2
was investigated. Fig. 5 shows the cyclic voltammetry of a 5.4 mM
solution of p-BrC6H4NO2 in the [C4mPyrr][NTf2] ionic liquid at a
coefficient of (1.8 ꢂ 0.1) ꢃ 10ꢁ11 m2
s
ꢁ1, while nc was again
found to decrease with time under vacuum. Extrapolation of the
nc vs. t data to overcome issues relating to solute volatility, as for
C6H5NO2, allowed for the determination of nc at t = 0, showing
that, for p-BrC6H4NO2, n was between one and two.
In order to determine the mechanism for the reduction of
p-BrC6H4NO2, it was necessary to simulate the voltammetric
data. Initially, the voltammetry was simulated as a one-electron
reduction, as for nitrobenzene, however, the one-electron
scan rate of 100 mV sꢁ1
.
Fig. 5 shows that the reduction and oxidation peaks occur at
ꢁ1.8 and ꢁ1.7 V, respectively. Also evident in the voltammetry
is a shoulder on the reduction wave and a sharper than
expected current decrease after the oxidation peak. This beha-
viour at potentials positive of the redox couple is likely caused
Fig. 5 Cyclic voltammetry of 5.4 mM p-BrC6H4NO2 in [C4mPyrr][NTf2] on a Pt
microdisk (radius = 5.0 mm) at a scan rate of 100 mV sꢁ1. Inset: Chronoamperogram
for the reduction of 5.4 mM p-BrC6H4NO2 in [C4mPyrr][NTf2] on a Pt microdisk
(radius = 5.0 mm), showing the experimental (circles) and Shoup–Szabo fitted
(line) transients.
Fig. 6 Cyclic voltammetry of 5.4 mM p-BrC6H4NO2 in [C4mPyrr][NTf2] on a Pt
microdisk (radius = 5.0 mm) at scan rates of 5, 10, 25, 50, 75, 100, 250, 500, 750
and 1000 mV sꢁ1
.
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 6382--6389 6385

View Article Online
PCCP
Paper
model was found to be insufficient in fitting the experimental If (3.2) is rate determining the mechanism is DISP1, while if
data. The experimental data was then simulated assuming a (3.3) is the rate determining step, the mechanism is DISP2.
pure two-electron reduction, however, the simulated voltam- The DISP2 case can be distinguished voltammetrically as the
mograms were again unable to describe the data and it was forward peak potential shows a concentration dependence as
evident that a more complicated mechanism was involved. well as a scan rate dependence of B20 mV decades, compared
Plotting the log of the peak currents, log(Ip), vs. the log of the to the B30 mV decades observed for both ECE and DISP1
scan rate, log(n), for the experimental voltammetry and for processes.36 The ECE and DISP1 schemes are often difficult to
the one- and two-electron simulations allows comparison of differentiate between voltammetrically when the 2nd electron
the experimental peak currents with the working curves for transfer is highly driven, as both have a chemical step which is
both mechanisms. This is demonstrated in Fig. 7, which shows rate-determining and show the same dependence of peak
the experimentally obtained peak currents varying between the potential with scan rate.36
two limiting cases, of one-electron and two-electron reduction,
The experimental voltammetry was then simulated using
with scan rate. This indicates that a chemical step may be both ECE and DISP1 schemes. The mechanisms assumed for
involved in the reaction scheme and also justifies why the short the ECE and DISP1 reduction of p-BrC6H4NO2 in the
time chronoamperometric transients were unable to yield a [C4mPyrr][NTf2] ionic liquid are shown below.
reproducible diffusion coefficient when fitted with the Shoup–
Szabo relation.
ꢁ
p-BrC6H4NO2 + eꢁ " [p-BrC6H4NO2]ꢀ
(4.1)
slow
ƒ!
½p-BrC6H4NO2ꢄꢀꢁþHS
C H NO þ Sꢀ þ Brꢁ
ð4:2Þ
The variation of the experimental peak current between the
one electron and simple two-electron limits, shown in Fig. 7,
suggests that either an ECE or DISP type mechanism is occur-
ring. The generalised ECE mechanism for a reduction is
represented as follows;
6
5
2
ꢁ
C6H5NO2 + eꢁ " C6H5NO2
(4.3.1)
ꢀ
where HS refers to the solvent system. For the DISP1 mechanism,
step 4.3.1 is replaced with;
A + eꢁ " B
(2.1)
(2.2)
(2.3)
B - C
fast
½p-BrC6H4NO2ꢄꢀꢁþC6H5NO2
p-BrC H NO þ C H NOꢀꢁ
ƒ!
6
4
2
6
5
2
C + eꢁ " D
ð4:3:2Þ
Note that reaction 4.3.2 is fully driven in the pure DISP1
case, whereas in the present system under study the reaction
will be in equilibrium with both reactants and products present
because C6H5NO2 and p-BrC6H4NO2 have clearly similar
reduction potentials. The fitting parameters used to simulate
the experimental voltammetry via the ECE and DISP1 mechanisms
shown in reaction 4 are presented in Tables 1 and 2, respectively,
where the values D, cbulk, E0, k0, kf and a refer to the diffusion
coefficients, initial concentrations, formal potentials, hetero-
geneous rate constants, homogeneous rate constants and
electron transfer coefficients, respectively, for steps involving
the specꢀies p-BrC6H4NO2, [p-BrC6H4NO2]ꢀꢁ, C6H5NO2 and
C6H5NO2 ꢁ. The parameter kf(1) represents a pseudo-first order
rate constant for the chemical step, as the solvent is present
in large excess. Simulations for the ECE and DISP1 reduction
of p-BrC6H4NO2 were constructed such that the simulated
diffusion coefficients for p-BrC6H4NO2 and C6H5NO2 were
those obtained from the Shoup–Szabo fitting of long time
chronoamperometric transients and the parameters for the
2nd heterogeneous step of the ECE mechanism were the same
as those determined via the one-electron simulation of
C6H5NO2. As a complex adsorption–desorption mechanism
was clearly influencing the experimental voltammetry, simula-
tions were limited to the fitting of the forward scan. A compar-
ison of experimental peak currents to those simulated using the
ECE mechanism is presented in Fig. 8.
The two generalised DISP mechanisms for the case of a
reduction are then represented by;
A + eꢁ " B
(3.1)
(3.2)
(3.3)
B - C
B + C - A + D
where the distinction between DISP1 and DISP2 depends on
which of the two homogeneous steps are rate determining.
Fig. 7 Comparison of the experimentally determined peak currents (circles)
with those simulated using one-electron (lower squares) and two-electron (upper
squares) models, with scan rate. The scan rates shown are 5, 10, 25, 50, 75, 100,
Fig. 8 shows a comparison of the reduction peak currents of
the voltammetry simulated via the ECE mechanism with the
experimental peak currents. It is evident that the simulated
250, 500, 750 and 1000 mV sꢁ1
.
c
6386 Phys. Chem. Chem. Phys., 2013, 15, 6382--6389
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
Table 1 Parameters employed in the ECE simulation of p-BrC6H4NO2 voltammetry
ꢁ
ꢁ
ꢀ
C6H5NO2
Chemical species
p-BrC6H4NO2
[p-BrC6H4NO2]ꢀ
C6H5NO2
D/m2 sꢁ1
cbulk/mM
(1.8 ꢂ 0.1) ꢃ 10ꢁ11
(5.0 ꢂ 0.5) ꢃ 10ꢁ12
(3.4 ꢂ 0.1) ꢃ 10ꢁ11
(1.1 ꢂ 0.1) ꢃ 10ꢁ11
5.4 ꢂ 0.5
0
0
0
Homogeneous reactions
B + HS - C
kf(1)/sꢁ1
1.0 ꢂ 0.1
Heterogeneous reactions
A + eꢁ - B
E0/V
k0/m sꢁ1
a
ꢁ1.74 ꢂ 0.01
>1 ꢃ 10ꢁ3 a
0.5 ꢂ 0.1
C + eꢁ - D
ꢁ1.81 ꢂ 0.01
(7 ꢂ 1) ꢃ 10ꢁ5
0.63 ꢂ 0.03
a
Not rate determining.
Table 2 Parameters employed in the DISP1 simulation of p-BrC6H4NO2 voltammetry
ꢁ
ꢁ
ꢀ
C6H5NO2
Chemical species
p-BrC6H4NO2
[p-BrC6H4NO2]ꢀ
C6H5NO2
D/m2 sꢁ1
cbulk/mM
(1.8 ꢂ 0.1) ꢃ 10ꢁ11
(5.0 ꢂ 0.5) ꢃ 10ꢁ12
(3.4 ꢂ 0.1) ꢃ 10ꢁ11
(1.1 ꢂ 0.1) ꢃ 10ꢁ11
5.4 ꢂ 0.5
0
0
0
Homogeneous reactions
B + HS - C
B + C - A + D
kf(1)/sꢁ1
1.0 ꢂ 0.1
kf(2)/m3 molꢁ1 sꢁ1
>1 ꢃ 1010 a
Heterogeneous reactions
E0/V
ꢁ1.74 ꢂ 0.01
k0/m sꢁ1
a
A + eꢁ - B
>1 ꢃ 10ꢁ3 a
0.5 ꢂ 0.1
a
Not rate determining.
peak currents are not too dissimilar from those obtained from 50, 250, 75, 500, 25 and 1000 mV sꢁ1, respectively. The
experimental voltammetry, however the fact that the second experimental voltammetry was then simulated using the DISP1
electron transfer is not highly driven, leading to slightly split- mechanism and the comparisons between experimental and
wave behaviour, would seem to indicate that an ECE mecha- simulated peak currents are shown in Fig. 9.
nism is ill-suited to describe the electrochemical reduction of p-
As shown in Fig. 9, the data simulated via the DISP1
BrC6H4NO2 in [C4mPyrr][NTf2]. It is also important to note that mechanism describes the experimental peak currents to a larger
the experimental peak currents yield a less well-defined work- extent than the voltammetry simulated via the ECE mechanism.
ing curve due to a steady decrease in concentration over the As such, the electrochemical reduction of p-BrC6H4NO2 in
timescale of the scan rate dependence study. The order of the the [C4mPyrr][NTf2] ionic liquid appears to occur via a DISP1
scan rates studied was random, to avoid the concentration drop mechanism, with simulations describing the experimental data
adding to the scan rate dependency, in the order 10, 100, 5, 750, well given the large parameter-space inherent in the simulation
of multi-step mechanisms.
It is of interest to compare the voltammetry of p-BrC6H4NO2
in [C4mPyrr][NTf2] with that reported for the photo-electro-
chemical reduction of p-BrC6H4NO2 in MeCN.21–23,26 The studies
in MeCN showed that, in the absence of light, p-BrC6H4NO2 was
reduced via an electrochemically reversible one-electron
reduction and that the resulting anion radical had a lifetime
of at least 10 s.22 Furthermore, irradiation of the system with
light corresponding to absorption bands in the radical anion,
led to the observation of significant photocurrents. These
photocurrents were shown to be due to the reduction of
nitrobenzene after photo-induced dissociation of Brꢁ from
the anion radical followed by hydrogen abstraction from the
solvent.23 In light of these results, the present voltammetry,
which was recorded in the absence of light inside a Faraday
cage, is all the more interesting as it shows that the
p-BrC6H4NO2 molecule has an increased reactivity in the
Fig. 8 Comparison of reduction peak currents obtained from experiment
[C4mPyrr][NTf2] ionic liquid, compared to MeCN. The most
obvious explanation for this enhanced reactivity is that the
(circles) and from ECE simulations (triangles) as well as for the two limiting cases
of one-electron (lower squares) and two-electron (upper squares) reductions.
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 6382--6389 6387

View Article Online
PCCP
Paper
light. Thus, it was shown that the p-BrC6H4NO2 radical
anion exhibits a changed reactivity in the [C4mPyrr][NTf2] ionic
liquid compared to molecular non-aqueous solvents, which was
rationalised in terms of much better solvation and stabilisation of
the bromide anion in the RTIL as coordination with the RTIL
cation is highly favoured.
Acknowledgements
S.E. thanks St John’s College Oxford and Syngenta for partial
funding. K.R.W. thanks Schlumberger Cambridge Research for
funding.
References
Fig. 9 Comparison of reduction peak currents obtained from experiment
(circles) and from DISP1 simulations (triangles) as well as for the two limiting
cases of one-electron (lower squares) and two-electron (upper squares)
reductions.
1 D. S. Silvester and R. G. Compton, Z. Phys. Chem., 2006, 220,
1247–1274.
2 J. D. Holbrey and K. R. Seddon, J. Chem. Soc., Dalton Trans.,
1999, 2133–2139.
ionic nature of the RTIL allows for much better solvation and
stabilisation of the bromide anion and C6H5NO2 radical anion as
compared to the solvation by MeCN because possible counter-ions
are present in much higher concentration than the supporting
electrolyte in MeCN. Furthermore, the bromide anion can be
envisaged to coordinate much more strongly with the pyrrolidinium
cation than the sterically hindered bis(trifluoromethylsulfonyl)imide
anion, as corroborated by examining the melting points of
[C4mPyrr][NTf2] and [C4mPyrr]Br, which are 186 K41 and
490 K,42 respectively. It has thus been demonstrated in this
work that the electrochemical reduction of p-BrC6H4NO2 in
[C4mPyrr][NTf2] occurs via a DISP1 mechanism in the absence
of light, which illustrates the possibility for enhanced reactivity
of the p-BrC6H4NO2 in the [C4mPyrr][NTf2] ionic liquid compared
to that in MeCN.
3 L. Xiong, A. M. Fletcher, S. G. Davies, S. E. Norman,
C. Hardacre and R. G. Compton, Chem. Commun., 2012,
48, 5784–5786.
4 M. J. Earle and K. R. Seddon, Pure Appl. Chem., 2000, 72,
1391–1398.
5 Z.-L. Shen, S.-J. Ji and T.-P. Loh, Tetrahedron Lett., 2005, 46,
3137–3139.
6 M. J. Earle, J. M. S. S. Esperança, M. A. Gilea, J. N. C. Lopes,
L. P. N. Rebelo, J. W. Magee, K. R. Seddon and
J. A. Widegren, Nature, 2006, 439, 831–834.
7 N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37,
123–150.
8 E. O. Barnes, A. M. O’Mahony, S. R. Belding and
R. G. Compton, J. Chem. Eng. Data, 2009, 55, 2219–2224.
´
9 J.-M. Saveant, Tetrahedron, 1994, 50, 10117–10165.
10 J. M. Saveant, Acc. Chem. Res., 1993, 26, 455–461.
11 L. Dunsch, Z. Chem., 1974, 14, 463–468.
12 J. Heinze, Angew. Chem., Int. Ed. Engl., 1984, 96, 823–840.
Conclusions
The reductions of C6H5NO2, C6H5Br and p-BrC6H4NO2 were 13 M. V. B. Zanoni, E. I. Rogers, C. Hardacre and
studied in the [C4mPyrr][NTf2] ionic liquid via cyclic voltammetry, R. G. Compton, Int. J. Electrochem. Sci., 2009, 4, 1607–1627.
chronoamperometry and voltammetric simulation. C6H5NO2 was 14 V. A. Nikitina, R. R. Nazmutdinov and G. A. Tsirlina, J. Phys.
found to be reduced via a one-electron mechanism to form the Chem. B, 2011, 115, 668–677.
stable radical anion, analogous to the behaviour in a range of 15 A. P. Doherty and C. A. Brooks, ACS Symp. Ser., 2003, 856,
aprotic solvents and the [C4dmim][NTf2] RTIL. The reduction 410–420.
of C6H5Br was found to occur at potentials negative to those of 16 J. Heinze, Angew. Chem., Int. Ed. Engl., 1993, 105, 1327–1349.
¨
C6H5NO2 and multiple scans were found to modify the electrode 17 G. Zeppenfeld, M. Stephan, L. Dunsch and H. G. Doge, Acta
surface. The reduction of p-BrC6H4NO2 was then examined Polym., 1979, 30, 664–669.
experimentally and via voltammetric simulations employing 18 D. S. Silvester, A. J. Wain, L. Aldous, C. Hardacre and
the analysis of the reductions of C6H5NO2 and C6H5Br. It was R. G. Compton, J. Electroanal. Chem., 2006, 596, 131–140.
shown that reduction proceeds via a DISP type mechanism in 19 W. Kemula and R. Sioda, Bull. Acad. Pol. Sci., Ser. Sci. Chim.,
which the parent molecule is reduced by one electron to form the 1962, 10, 507–512.
anion radical, at which point debromination occurs in a rate- 20 J. G. Lawless and M. D. Hawley, J. Electroanal. Chem.
determining chemical step involving the fast scavenging of a Interfacial Electrochem., 1969, 21, 365–375.
hydrogen atom from the solvent system to yield nitrobenzene 21 J. A. Cooper and R. G. Compton, Electroanalysis, 1998, 10,
which is then reduced to its radical anion. This shows significant 1182–1187.
contrast to the behaviour observed in conventional non-aqueous 22 R. G. Compton, R. A. W. Dryfe and A. C. Fisher,
solvents, as the debromination was observed in the absence of
J. Electroanal. Chem., 1993, 361, 275–278.
c
6388 Phys. Chem. Chem. Phys., 2013, 15, 6382--6389
This journal is the Owner Societies 2013

View Article Online
Paper
PCCP
23 R. G. Compton and R. A. W. Dryfe, J. Electroanal. Chem., 33 A. M. O’Mahony, D. S. Silvester, L. Aldous, C. Hardacre and
1994, 375, 247–255. R. G. Compton, J. Chem. Eng. Data, 2008, 53, 2884–2891.
24 L. Alvarez-Griera, I. Gallardo and G. Guirado, Electrochim. 34 C. Zhao, A. M. Bond, R. G. Compton, A. M. O’Mahony and
Acta, 2009, 54, 5098–5108. E. I. Rogers, Anal. Chem., 2010, 82, 3856–3861.
25 R. G. Compton, R. A. W. Dryfe, J. C. Eklund, S. D. Page, J. Hirst, 35 J. O. M. Bockris and A. K. N. Reddy, Modern Electrochemistry,
´
L. Nei, G. W. J. Fleet, K. Y. Hsia, D. Bethell and L. J. Martingale,
J. Chem. Soc., Perkin Trans. 2, 1995, 1673–1677.
26 R. G. Compton, R. A. W. Dryfe and A. C. Fisher, J. Chem. Soc.,
Perkin Trans. 2, 1994, 1581–1587.
Plenum Press, New York, 1998.
36 R. G. Compton and C. E. Banks, Understanding Voltammetry,
World Scientific, Singapore, London, 2nd edn, 2011.
37 P. Zuman and Z. Fijalek, J. Electroanal. Chem., 1990, 296,
583–588.
27 E. O. Barnes, Y. Wang, J. G. Limon-Petersen, S. R. Belding
and R. G. Compton, J. Electroanal. Chem., 2011, 659, 25–35. 38 A. Pelekourtsa, N. Missaelidis and D. Jannakoudakis, Chim.
ˆ
28 P. Bonhote, A.-P. Dias, N. Papageorgiou, K. Kalyanasundaram
Chron., 1997, 26, 39–48.
¨
and M. Gratzel, Inorg. Chem., 1996, 35, 1168–1178.
39 V. Jouikov and J. Simonet, Electrochem. Commun., 2010, 12,
331–334.
40 L. H. J. Xiong, L. Aldous, M. C. Henstridge and
R. G. Compton, Anal. Methods, 2012, 4, 371–376.
29 D. Shoup and A. Szabo, J. Electroanal. Chem. Interfacial
Electrochem., 1982, 140, 237–245.
30 M. Sharp, Electrochim. Acta, 1983, 28, 301–308.
31 Y. Meng, L. Aldous, S. R. Belding and R. G. Compton, Phys. 41 D. R. MacFarlane, P. Meakin, J. Sun, N. Amini and
Chem. Chem. Phys., 2012, 14, 5222–5228. M. Forsyth, J. Phys. Chem. B, 1999, 103, 4164–4170.
32 R. G. Evans, O. V. Klymenko, P. D. Price, S. G. Davies, 42 C. Jagadeeswara Rao, K. A. Venkatesan, K. Nagarajan,
C. Hardacre and R. G. Compton, ChemPhysChem, 2005, 6,
526–533.
T. G. Srinivasan and P. R. Vasudeva Rao, Electrochim. Acta,
2009, 54, 4718–4725.
c
This journal is the Owner Societies 2013
Phys. Chem. Chem. Phys., 2013, 15, 6382--6389 6389