Electroactive imidazolium salts based on 1,4-dimethoxybenzene redox groups: Synthesis and electrochemical characterisation

Article abstract of DOI:10.1039/c3ra41345d

Three new redox-active ionic salts have been synthesised based on 1,4-dimethoxybenzene (1) and 2,5-di-tert-butyl-1,4-dimethoxybenzene (2 and 3). These imidazolium salts are the first organic redox-active groups to be incorporated into an ionic liquid stru

Full text of DOI:10.1039/c3ra41345d

RSC Advances
View Article Online
View Journal | View Issue
COMMUNICATION
Electroactive imidazolium salts based on 1,4-
dimethoxybenzene redox groups: synthesis and
Cite this: RSC Advances, 2013, 3, 12035
electrochemical characterisation3
Received 19th March 2013,
Accepted 30th May 2013
John C. Forgie and Dominic Rochefort*
DOI: 10.1039/c3ra41345d
www.rsc.org/advances
Three new redox-active ionic salts have been synthesised based
on 1,4-dimethoxybenzene (1) and 2,5-di-tert-butyl-1,4-dimethox-
ybenzene (2 and 3). These imidazolium salts are the first organic
redox-active groups to be incorporated into an ionic liquid
structure. There are no negative effects on the transport proper-
ties by incorporating 2,5-di-tert-butyl-1,4-dimethoxybenzene into
an electroactive imidazolium salt. The thermal stability and
oxidation potential have been increased.
imidazolium cation, it makes a well-known and characterised
ionic liquid, EMI-TFSI. Compounds 2 and 3 are both imidazolium
salts with the only difference between them being the anion, TFSI,
and hexafluorophosphate (PF6), respectively.
Ionic liquids have attracted a great deal of attention in recent years
for use as solvents or additives to many applications such as
capacitors, batteries, and fuel cells.1–3 The reason can be ascribed
to their numerous desirable properties such as high electrical
conductivity and dielectric constant; excellent thermal and
chemical stability; and a non-existent vapour pressure.4 A redox-
active ionic liquid, the incorporation of a redox-active group
directly in an ionic liquid structure, is an interesting sub-section of
this field allowing studies on electron transport,5 and electron
transfer,6 to be conducted in higher concentrations. In a recent
publication, we have demonstrated the ability of redox-active ionic
liquids to be used as a redox shuttle for lithium-ion batteries.7
Redox-active ionic liquids are commonly organometallic, primarily
based on ferrocene and cobaltocene,5,8 and to the best of the
authors’ knowledge, no redox shuttle ionic liquids have been
synthesised that contain organic redox-active groups. Herein, we
report the first synthesis of three new redox-active imidazolium
salts, compounds 1–3, based on the redox-active groups 1,4-
dimethoxybenzene and 2,5-di-tert-butyl-1,4-dimethoxybenzene (4),
respectively.
The cation in both examples is propyl-methyl-imidazolium,
this time linked to a 2,5-di-tert-butyl-dimethoxyphenoxy group.
The molecule 2,5-di-tert-butyl-1,4-dimethoxybenzene was proposed
as a stable, high potential redox shuttle for an application in
lithium-ion batteries in 2005.9 It was shown to be fully reversible
when oxidised and can cycle more than 200 times,10 due to the
two bulky tert-butyl arms protecting the radical cation and prevent
it from dimerising and subsequently polymerising when oxidised,
as sometimes observed with 1,4-dimethoxybenzene. The same
strategy was employed here in order to obtain an electrochemically
reversible redox ionic liquid.
Ionic liquid 1 was synthesised in four steps, as depicted in
Scheme 1. Compound 6 was synthesised from the etherification of
4-methoxyphenol (5) with 2-chloroethanol in 80% yield.11 The
corresponding alcohol was brominated12 to give compound 7 in
78% yield. Reaction of 7 with 1-methylimidazole affords the
imidazolium bromide salt, 8 in 82% yield.13 Conversion of the
bromide counter ion to TFSI was achieved from a metathesis
reaction of 8 with lithium bis(trifluromethanesulfonyl)amide in an
aqueous solution to give the desired product 1.
Compound 1 is an ionic liquid in which an ethyl-methyl-
imidazolium (EMI) cation is directly connected to a methoxyphe-
noxy group. Bis(trifluoromethanesulfonyl)amide (TFSI) was chosen
as the anion for our ionic liquid as when combined with an
´
´
´
Departement de chimie, Universite de Montreal, CP6128 Succ. Centre-Ville,
´
Montreal, QC, Canada H3C 3J7. E-mail: dominic.rochefort@umontreal.ca;
Fax: +1-514-34-37585; Tel: +1-514-343-6733
Redox-active imidazolium salts 2 and 3 were synthesised in a
similar fashion using 3-bromo-propanol in the etherification step
to provide 10 in 56% yield as shown in Scheme 2.14 As before,
Electronic supplementary information (ESI) available: Experimental section,
3
differential scanning calorimetry, thermogravimetric analysis and additional cyclic
voltammograms. See DOI: 10.1039/c3ra41345d
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 12035–12038 | 12035

View Article Online
Communication
RSC Advances
butyl groups clearly has a large effect on the melting point, the
aliphatic groups possibly providing additional symmetry or
helping facilitate p–p stacking of the aromatic cores. The TFSI
derivative has a lower melting point than the PF6 version, as the
anion in this case is unable to hydrogen-bond and has a more
delocalised charge.17 Thermogravimetric analysis (TGA) was
performed on compounds 1–4 (Fig. S4–S7, ESI ) from room
3
temperature to 600 uC. Compound 1 is the most stable with an
onset of degradation at 320 uC, the addition of the aliphatic tert-
butyl groups lowers the onset to 250 uC for compound 2. The
change of anion from TFSI to PF6 lowers the onset further; this
time to 190 uC for 3, while 4 is the least stable, with an onset of
degradation occurring at 60 uC.
Scheme 1 Synthesis of compound 1.
These TGA results show that the incorporation of 2,5-di-tert-
butyl-1,4-dimethoxybenzene into an imidazolium salt increases
the thermal stability by at least 130 uC. In comparison, the
unmodified ionic liquid EMI-TFSI is stable up to over 400 uC.15
The electrochemical properties of the newly synthesised
compounds 1–3 were assessed by cyclic voltammetry (CV) using
Pt working and counter electrodes and a pseudo Ag reference
electrode in ethylene carbonate-diethyl carbonate (1 : 2 v/v) with
lithium TFSI as the supporting electrolyte (1.5 M). This solution
was selected to evaluate the electrochemistry of the redox ionic
liquids in an electrolyte usable in Li-ion batteries. Fig. 1 shows the
oxidation of all three compounds in 1 mM solutions, all three
compounds oxidise to give a radical cation. All data are
summarised in Table 1. Compound 1 shows a quasi-reversible
wave at +1.00 V, full reduction of the oxidised species in this case
to the neutral compound is not possible as the radical cation
formed from the oxidation reacts with another radical cation to
give a dimer. At this concentration, dimerisation is not seen when
multiple CV cycles are performed but at higher concentrations (10
mM) this phenomena can be observed upon the first ten cycles
(see Fig. 2). Indeed, Fig. 2 shows a decrease of the peaks at +0.98
and +1.13 V (compound 1), with new peaks appearing at +0.56 and
+0.69 V (dimers of 1). Compounds 2 and 3 however do show full
reversibility as the large tert-butyl groups protect the radical and
prevent dimerization (inset of Fig. 2). Consequently though, the
addition of the tert-butyl groups through a large inductive effect
lowers the oxidation potential greatly, the anodic peaks of 2 and 3
bromination of 10 to give 11 was performed with carbon
tetrabromide (82% yield) which was then reacted with 1-methy-
limidazole to give imidazolium bromide salt, 12 in 73% yield. The
TFSI salt (2) was synthesised from a metathesis reaction of 12 with
lithium bis(trifluoromethanesulfonyl)amide in methanol in 67%
yield. Methanol was used as the solvent rather than water as
compound 11 is insoluble in aqueous media. The PF6 salt (3) was
obtained from the reaction of 12 with silver hexaflurorophosphate
in acetonitrile, the resulting silver bromide precipitate was filtered
and the desired salt 3 was collected under reduced pressure in
75% yield. Compounds 1–3 were vacuum-dried at 80 uC for 24 h.
The melting points and crystallisation temperatures of 2 and 3
(Fig. S2 and S3 in ESI ) are considerably high for imidazolium TFSI
3
and PF6 based compounds (Tm/Tc: 94.35/51.30 uC and 149.37/
119.78 uC, respectively, measured by differential scanning
calorimetry). In comparison, EMI-TFSI has a melting point of
around 221 uC,15 and EMI-PF6 at 60 uC.16 Compound 1 is an ionic
liquid at room temperature with a glass transition temperature
(T ) at 262.4 uC (Fig. S1 in ESI ). The addition of the bulky tert-
3
g
Fig. 1 Cyclic voltammogram of compounds 1–3 (1 mM in EC:DEC + 1.5 M LiTFSI,
100 mV s21).
Scheme 2 Synthesis of compounds 2 and 3.
12036 | RSC Adv., 2013, 3, 12035–12038
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Communication
than the other imidazolium salts, which will encounter an
association of multiple TFSI anions with a lithium cation.19 The
ratio of DO/DR is 0.97 and 0.98 for 2 and 3, respectively; these
values are really close to 1 thus showing that the oxidation process
is fully reversible and the diffusion expectedly follows Arrhenius-
type behaviour. The heterogeneous rate transfer constant (ks) was
determined using peak-to-peak separation (Dp) according to
Nicholson’s method.20 The scan rate experiment was carried out
using iR compensation with positive feedback. At low scan rates,
the electron transfer is Nernstian and not controlled by the
electrode reaction kinetics. At scan rates of 2 V s21 and higher, the
peak separations from these scan rates were used to calculate an
average ks. All data are summarised in Table 1. As discussed above,
compound 3, with less interaction to the electrolyte than 2,
Table 1 Electrochemical data of compounds 1–4
ox
E1/2 (V)
DR (cm2 s21
3.72 6 1027
)
DO (cm2 s21
)
ks (cm s21
)
1
2
3
4
+1.00a
—
—
+0.51 (+0.56/+0.45) 3.64 6 1027
+0.51 (+0.56/+0.46) 5.21 6 1027
+0.44 (+0.50/+0.37) 3.24 6 1027
3.52 6 1027
5.08 6 1027
3.21 6 1027
0.0049
0.0090
0.0049
a
Quasi-reversible wave.
are 0.34 V lower than 1. For comparison, we also measured the
oxidation potential of 2,5-di-tert-butyl-1,4-dimethoxybenzene (4) in
the same conditions. This compound also shows a reversible
oxidation wave with E1/2ox = +0.44 V which is 0.07 V lower than the
ox
predictably has a higher ks value (0.0090 and 0.0049 cm s21
respectively).
In conclusion, three new imidazolium salts, 1–3 which are
,
E1/2 of 2 and 3 (+0.51 V for both). The difference in potential
could be attributed to the imidazolium cation on the chain pulling
electron density away from the core thus requiring more energy to
remove an electron or from possible p-stacking interactions
between the aromatic core and the imidazolium ring, allowed by
the flexibility of the propyl linker chain. The evaluation of the
electrochemistry of an analogue of compound 1 with a shorter
linker could provide an explanation on the shift in potential. To
examine the effect the addition of the tert-butyl groups and the
incorporation of the redox shuttle into an ionic liquid has on the
transport properties, the diffusion coefficients (reduced, DR and
oxidised, DO forms) of compounds 1–4 and the heterogeneous rate
transfer (ks) of compounds 2–4 were measured. Compound 1 does
not show a reversible oxidation and therefore DO and ks could not
be calculated. The diffusion coefficient, an important parameter
for redox shuttles as it is a factor in the determination of the
maximum current that the redox shuttle can carry,18 was
calculated from a series of oxidations at different scan rates
based on 1,4-dimethoxybenzene and 2,5-di-tert-butyl-1,4-dimethox-
ybenzene redox-active groups linked to imidazolium cations to
form salts with TFSI and PF6 anions were reported. Compound 1
has the highest oxidation potential but is irreversible; compounds
2 and 3 are fully reversible, the large tert-butyl groups prevent
dimerisation but also lower oxidation potential through a large
inductive effect. The PF6 salt, 2 shows the highest diffusion
coefficient and heterogeneous rate constant among the three
compounds. In comparison to 4, the incorporation of 2,5-di-tert-
butyl-1,4-dimethoxybenzene onto a redox-active imidazolium salt
improves the thermal stability, has no negative impact on the
transport properties and even actually increases the oxidation
potential. In a future paper, we will report on the use of these
redox-active imidazolium salts as redox shuttles for cathode
overcharge protection in Li-ion batteries.
(ESI, Fig. S8–S12). The diffusion coefficients of the reduced forms
3
´ ´
We acknowledge the financial support from Fonds Quebecois
pour la Recherche en Nature et Technologie (FQRNT).
are all in the same order of magnitude (61027 cm2 s21), for the
four compounds with the only difference being the PF6 salt (3)
which has a higher diffusion coefficient in both the reduced and
oxidised forms. The reason for the small increase can be explained
by the PF6 salt experiencing less interaction with the electrolyte
Notes and references
1 A. B. McEwen, H. L. Ngo, K. LeCompte and J. L. Goldman, J.
Electrochem. Soc., 1999, 146, 1687–1695.
2 A. Lewandowski and A. Swiderska-Mocek, J. Power Sources,
2009, 194, 601–609.
3 M. Armand, F. Endres, D. R. MacFarlane, H. Ohno and
B. Scrosati, Nat. Mater., 2009, 8, 621–629.
4 M. Galinski, A. Lewandowski and I. Stepniak, Electrochim. Acta,
2006, 51, 5567–5580.
5 W. Wang, R. Balasubramanian and R. W. Murray, J. Phys.
Chem. C, 2008, 112, 18207–18216.
6 O. Fontaine, C. Lagrost, J. Ghilane, P. Martin, G. Trippe,
C. Fave, J. C. Lacroix, P. Hapiot and H. N. Randriamahazaka, J.
Electroanal. Chem., 2009, 632, 88–96.
7 J. C. Forgie, S. El Khakani, D. D. MacNeil and D. Rochefort,
Phys. Chem. Chem. Phys, 2013, 15, 7713–7721.
8 Y. Gao, B. Twamley and J. M. Shreeve, Inorg. Chem., 2004, 43,
3406–3412.
9 J. Chen, C. Buhrmester and J. R. Dahn, Electrochem. Solid-State
Lett., 2005, 8, A59–A62.
Fig. 2 Cyclic voltammogram of compound 1 showing the formation of a new
compound upon cycling. Inset shows the stable behaviour of compound 2.
Conditions: 10 mM in EC:DEC + 1.5 M LiTFSI, 100 mV s21
.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 12035–12038 | 12037

View Article Online
Communication
RSC Advances
´
10 J. R. Dahn, J. W. Jiang, L. M. Moshurchak, M. D. Fleischauer,
C. Buhrmester and L. J. Krause, J. Electrochem. Soc., 2005, 152,
A1283–A1289.
11 M. Dong, Y. Q. Si, S. Y. Sun, X. P. Pu, Z. J. Yang, L. R. Zhang, L.
H. Zhang, F. P. Leung, C. M. C. Lam, A. K. Y. Kwong, J. B. Yue,
Y. Y. Zhou, I. A. Kriksunov, Q. Hao and H. C. Lee, Org. Biomol.
Chem., 2011, 9, 3246–3257.
12 Y. J. Ma, X. F. Ji, F. Xiang, X. D. Chi, C. Y. Han, J. M. He, Z. Abliz,
W. X. Chen and F. H. Huang, Chem. Commun., 2011, 47,
12340–12342.
13 J. Li, F. Hu, X.-K. Xie, F. Liu and Z.-Z. Huang, Catal. Commun.,
2009, 11, 276–279.
14 P. Nesvadba, L. B. Folger, P. Maire and P. Novak, Synth. Met.,
2011, 161, 259–262.
15 A. Noda, K. Hayamizu and M. Watanabe, J. Phys. Chem. B, 2001,
105, 4603–4610.
16 J. Fuller, R. T. Carlin, H. C. De Long and D. Haworth, J. Chem.
Soc., Chem. Commun., 1994, 299–300.
ˆ
17 P. Bonhote, A.-P. Dias, N. Papageorgiou, K. Kalyanasundaram
¨
and M. Gratzel, Inorg. Chem., 1996, 35, 1168–1178.
18 S. R. Narayanan, S. Surampudi, A. I. Attia and C. P. Bankston, J.
Electrochem. Soc., 1991, 138, 2224–2229.
19 A. Andriola, K. Singh, J. Lewis and L. Yu, J. Phys. Chem. B, 2010,
114, 11709–11714.
20 R. S. Nicholson, Anal. Chem., 1965, 37, 1351.
12038 | RSC Adv., 2013, 3, 12035–12038
This journal is ß The Royal Society of Chemistry 2013