Electrocatalytic debromination of open-chain and cyclic dibromides in ionic liquids with cobalt(II)salen complex as mediator

Article abstract of DOI:10.1007/s11164-012-0634-6

The electrocatalytic reduction of open-chain and cyclic dibromides in ionic liquids, mediated by cobalt(II)salen, was investigated. Macro-scale constant-potential electrolysis in an undivided cell gave the corresponding debrominated products in moderate to good yields. The workup process after electrolysis proved to be much simpler in the ionic liquid than that in organic solvents. The possibility of reuse of the ionic liquid was demonstrated.

Full text of DOI:10.1007/s11164-012-0634-6

Res Chem Intermed (2013) 39:89–99
DOI 10.1007/s11164-012-0634-6
Electrocatalytic debromination of open-chain and cyclic
dibromides in ionic liquids with cobalt(II)salen complex
as mediator
•
•
•
Yi Shen Shinsuke Inagi Mahito Atobe
Toshio Fuchigami
Received: 26 July 2011 / Accepted: 18 October 2011 / Published online: 16 June 2012
Ó Springer Science+Business Media B.V. 2012
Abstract The electrocatalytic reduction of open-chain and cyclic dibromides in
ionic liquids, mediated by cobalt(II)salen, was investigated. Macro-scale constant-
potential electrolysis in an undivided cell gave the corresponding debrominated
products in moderate to good yields. The workup process after electrolysis proved to
be much simpler in the ionic liquid than that in organic solvents. The possibility of
reuse of the ionic liquid was demonstrated.
Keywords Mediator ꢀ Ionic liquid ꢀ Controlled potential electrolysis ꢀ
Indirect electrolysis
Introduction
Recently, room temperature ionic liquids have attracted much attention as ‘‘green
solvents’’ because of their low melting point, high polarity, negligible vapor
pressure, non-flammability, ability to dissolve many materials, good conductivity,
and reusability [1–5]. Therefore, considering their non-volatile and good conductive
properties, ionic liquids seem to be ideal media for organic electrochemistry [6–14].
In fact, several papers dealing with the voltammetric study of metal complexes in
ionic liquids have been reported [7].
Gaillon and Bedioui [12], Sweeny and Peters [13], and Lagunas et al. [14]
reported cyclic voltammetric studies of electrocatalytic reduction in ionic liquids
using Co(II)salen (N,N0-bis(salicylidene)ethylenediaminocobalt(II); Fig. 1),
Ni(II)salen, and vitamin B12, respectively. In these studies, they showed that the
complexes operate as mediators for the cathodic reduction of various organic
Y. Shen ꢀ S. Inagi ꢀ M. Atobe ꢀ T. Fuchigami (&)
Department of Electronic Chemistry, Tokyo Institute of Technology, 4259 Nagatsuta,
Midori-ku, Yokohama 226-8502, Japan
e-mail: fuchi@echem.titech.ac.jp
123

90
Y. Shen et al.
Fig. 1 Structure of
cobalt(II)salen
O
N
O
N
Co
halides. However, they did not report results on the preparative electrolysis and the
reusability of ionic liquids, although reusability is a key advantage. Mellah
et al. [15], Barhdadi et al. [16], and Sawamura et al. [17] have reported the use of
redox mediators in ionic liquids for electrochemical organic synthesis. However,
a limited number of successful examples of organic electrosynthesis in ionic
liquids have yet been reported. With these facts in mind, we report herein the
electrochemistry of cobalt(II)salen in a variety of ionic liquids and describe use of
cobalt(II)salen as a mediator for preparative electrocatalytic reduction of dihalo
compounds. We also discuss the reusability of the ionic liquids.
Experimental
Materials
All reagents and dehydrated solvents were purchased from commercial sources and
used without further purification. The ionic liquids 1-ethyl-3-methylimidazolium
tetrafluoroborate (EMIMBF4), 1-n-butyl-3-methylimidazolium tetrafluoroborate
(BMIMBF4), and 1-n-butyl-3-methyl-imidazolium hexafluorophosphate (BMIMPF6)
were purchased from Kanto Chemical, and 1-ethyl-3-methylimidazolium trifluor-
omethanesulfonate (EMIMTfO) was obtained from Central Glass (Chart 1).
General
1
1
All products were analyzed by H NMR and 13C NMR spectroscopy and MS. H
NMR (270.05 MHz) and 13C NMR (67.80 MHz) spectra were acquired on a Jeol
1
EX270 NMR spectrometer in CDCl3 as solvent. The chemical shifts for the H
NMR and 13C NMR spectra are given in d (ppm) downfield from an internal TMS
standard. Mass spectra and high-resolution mass spectra were obtained with
Shimadzu GC-MSOP-200A and Jeol JMS-700 mass spectrometers, respectively.
Cyclic voltammetric measurements were obtained by use of an electrochemical
analyzer (model 600A, ALS/CH Instruments). Preparative electrolysis was
BF4
BF4
PF6
CF3SO3
C2H5
N
N
N
N
N
N
N
N
H3C
C2H5 H3C
C4H9 H3C
C4H9
H3C
EMIMBF4
BMIMBF4
BMIMPF6
EMIMTfO
Chart 1 Structures of the ionic liquids
123

Electrocatalytic debromination of open-chain and cyclic dibromides
91
performed with a potentiostat/galvanostat (Hokutodenko HA-501) and monitored
with a coulomb/amperehour meter (Hokutodenko HF-201).
Preparation of the starting materials: diethyl 2,7-dibromooctanedioate
and 2,3-dibromo-1,2,3,4-tetrahydronaphthalene
Diethyl 2,7-dibromooctanedioate (1)
In a three-necked flask equipped with a dropping-funnel, a double-surface
condenser, and a device for absorbing evolved sulfur dioxide was placed 37.56 g
(315 mmol) redistilled thionyl chloride, which was heated to boiling (76 °C). Pure
suberic acid (25 g, 144 mmol) was added at such a rate that the mixture continued
to boil gently under reflux for 1 h. The mixture was then heated under reflux for
0.5 h to expel any dissolved sulfur dioxide and left to cool to room temperature,
before addition of 0.1 g red phosphorus. Dry bromine (55.1 g, 344 mmol) was then
introduced to the gently boiling octanedioyl dichloride over a period of 2 h, and the
mixture was then heated under reflux for 7 h, by which time evolution of hydrogen
bromide had almost ceased. The crude 2,7-dibromooctanedionyl chloride was then
added to 40 mL ethanol in a three-necked flask equipped with a mechanical stirrer
and reflux condenser. The reaction was completed by heating for 4 h, during which
hydrogen chloride evolved slowly. The crude ester was poured into 43 g distilled
water and the oil was separated and washed successively with water and sodium
hydrogen carbonate solution, and then dried under vacuum. The resulting material
was subjected to silica gel column chromatography (hexane–EtOAc = 9:1) to give
41.6 g (75 % yield) pure diethyl 2,7-dibromooctanedioate (1).
1H NMR (CDCl3) d 1.30 (6H, t, J = 7.1 Hz), 1.42–1.54 (4H, m), 1.98–2.13 (4H,
m), 4.20 (2H, t, J = 6.8 Hz), 4.23 (4H, q, J = 7.1 Hz); 13C NMR (CDCl3) d 14.0,
26.5, 34.4, 45.7, 61.9, 169.4; MS (m/z) 388 (M?).
2,3-Dibromo-1,2,3,4-tetrahydronaphthalene (2)
To a solution of 2.5 g (19 mmol) 1,4-dihydronaphthalene in 15 mL of carbon
tetrachloride, 15 mL carbon tetrachloride containing 3.2 g (20 mmol) bromine was
added dropwise slowly with stirring at 0 °C. The solvent was removed under
vacuum, and the residue was recrystallized several times from ethanol to give 3.3 g
1
(60 % yield) 2. The structure of 2 was identified by comparison of H NMR, 13C
NMR, and MS with those of the authentic sample.
Cyclic voltammetry and electrolysis
Cyclic voltammetric measurements were obtained by use of a three-electrode setup
in a two-compartment electrochemical cell. A glassy carbon (GC) disk (d = 3 mm)
was used as the working electrode and a Pt plate (0.5 9 0.5 cm2) as the counter
electrode. The reference electrode was a saturated calomel electrode (SCE). The
two compartments of the electrochemical cell were connected via a glass tube filled
with the same ionic liquid as the cell medium.
123

92
Y. Shen et al.
Preparative electrolysis
Macro-electrolysis was performed with a potentiostat/galvanostat connected to a
coulomb/amperehour meter. A dimensionally stable anode (DSA) electrode (RuO2/
Ti) was obtained from Permelec Electrode. The electrolytic cell was a conventional
undivided cell equipped with a GC cathode (2 9 2 cm2), an appropriate anode, as
shown in Table 1, and an SCE reference electrode.
Measurements and electrolysis were performed under a nitrogen atmosphere.
After electrolysis, the product in the ionic liquid was extracted three times with
10 mL diethyl ether and the extracts evaporated under vacuum to provide the
product. The product was identified by comparison with an authentic sample
purchased from Tokyo Kasei Kogyo. The yield of the products was estimated by 1H
NMR using phenol as internal standard. The remaining ionic liquid was dried under
vacuum for more than 6 h and was reused for the next electrolysis.
Results and discussion
Cyclic voltammetry of cobalt(II)salen in different ionic liquids
Cyclic voltammetric measurements of the Co(II)salen complex were performed in
four ionic liquids, EMIMBF4, EMIMTfO, BMIMBF4, and BMIMPF6, using a two-
compartment electrochemical cell with a GC disk (d = 3 mm) as working electrode,
a Pt plate counter electrode, and an SCE reference electrode. A well-defined,
reversible Co(II)/Co(I) redox couple was observed at -1.23 and -1.15 V in
BMIMBF4 (Fig. 2) and similar redox peaks were also observed at -1.18 and
-1.10 V in EMIMBF4. The peak separation was 80 mV and the redox peak currents
were almost identical in both ionic liquids. In contrast, no clear redox peaks were
observed in EMIMTfO and BMIMPF6.
For a reversible system, the peak current ip can be expressed by Eq. 1.
ip ¼ ð2:69 ꢁ 105Þn3=2AD1=2Cꢂm1=2
ð1Þ
where n is the number of electrons involved in the electrode reaction (n = 1), A is the
area of the electrode (A = 0.07 cm2), D is the diffusion coefficient of the substrate in
the ionic liquid, C* is the concentration of Co(II)salen in the ionic liquid, and m is the
potential scan rate. Based on the plot of ip vs. m1/2 in Fig. 3, the Dco(II)salen value was
calculated to be 1.78 9 10-8 cm2 s-1. This value is similar to that observed for a
Ni(II)salen complex in the same ionic liquid (1.8 9 10-8 cm2 s-1)7 and is approxi-
mately a factor of 100 smaller than that observed for ordinary molecular solvents.
Electrocatalytic reduction of 1 using a Co(II)salen mediator in ionic liquids
On the basis of the results obtained above, we attempted to use Co(II)salen as
mediator in the electrocatalytic reduction of 1 in ionic liquids. First, the reduction
peak potential of 1 was measured by cyclic voltammetry in EMIMBF4 and an
123

Electrocatalytic debromination of open-chain and cyclic dibromides
93
123

94
Y. Shen et al.
Fig. 2 Cyclic voltammograms
obtained from Co(I/II)salen
(3 mM) at a GC electrode
(d = 3 mm) in BMIMBF4. Scan
rates: 10, 30, 50, and 100 mV/s.
Room temperature
Fig. 3 Plot of ip vs. m1/2. Data were obtained from Fig. 2
irreversible first reduction peak was observed at -2.10 V vs. SCE, which was more
negative than that of Co(II)salen (Erped = -1.23 V vs. SCE). The possibility of
utilizing Co(II)salen as a mediator for cathodic reduction of 1 was suggested by the
following cyclic voltammetric analysis (Fig. 4).
As shown in Fig. 4, the cathodic peak current increased substantially in the
presence of 1, and the re-oxidation peak disappeared completely. This type of
catalytic current suggests that reaction of electrogenerated Co(I)salen with 1 is
fast and irreversible. However, in the ionic liquid EMIMBF4, both the anodic
and cathodic peak currents decreased in the presence of 1, which seems to be
123

Electrocatalytic debromination of open-chain and cyclic dibromides
95
Fig. 4 Cyclic voltammograms
of (a) Co(II)salen (3 mM) and
(b) Co(II)salen (3 mM) in the
presence of 1 (5 mM) at a GC
electrode (d = 3 mm) in
BMIMBF4. Scan rate: 100 mV/s
attributable to the low solubility of 1 in EMIMBF4 and adsorption of insoluble 1 on
the surface of the GC electrode, suppressing the redox current.
Next, a macro-scale Co(II)salen-mediated dehalogenation reaction of 1 was
performed in an undivided cell using a GC cathode at 40 °C. As shown in Table 1,
cathodic reduction of 1 using the Co(II)salen mediator in BMIMBF4 proceeded to
give the corresponding debrominated product 4. Although it has been shown that
cathodic reduction of 1 in DMF gives the cyclic product 3 [18], this product was not
formed in our experiment. Constant-potential electrolysis of 1 using the Co(II)salen
mediator (Table 1, entry 3) gave a higher yield of 4 than constant current
electrolysis (entry 2). It is noted that use of a DSA anode gave a higher yield than
use of a Pt anode (entries 3 and 4). Although such an anode effect is interesting, its
cause is not clear. In the absence of Co(II)salen, neither compound 3 nor 4 was
detected even after a 4 F mol-1 charge was passed, and the starting material was
mostly recovered (entry 5).
From these results, the reaction mechanism depicted in Scheme 1 was proposed.
One-electron reduction of Co(II)salen generates Co(I)salen, which reacts with
starting material 1 to form the radical via elimination of bromide. Further one-
electron reduction of the radical produces anion A. According to a previous report
[18], anion A can afford the cyclic product 3. However, cyclization did not occur in
the viscous ionic liquid [19, 20]. Thus anion A is protonated and repeated
debromination occurs to provide product 4. The proton source has not been
identified but is likely to be a trace amount of water in the ionic liquid. Because a
substantially enhanced catalytic reduction current was observed after addition of
dibromide 1, chemical reaction followed by electron-transfer seems to be very rapid.
In general, a divided cell has to be used for mediator systems to prevent anodic
oxidation of the reduced species once generated at the cathode, which is rather
inconvenient. However, in such a viscous ionic liquid, the mediator can remain in
the vicinity of the cathode surface and will encounter the substrate before oxidation
123

96
Y. Shen et al.
e
e
O
OEt
Br
O
OEt
Co(I)salen Co(II)salen
Co(I)salen Co(II)salen
Br
Et
Br
Et
-Br-
-Br-
O
O
1
2e, -Br-, +H+
H
O
OEt
O
OEt
O
OEt
H
+H+
Br
Et
Br
Et
H
O
O
Et
O
A
4
O
OEt
OEt
O
3
Scheme 1 Proposed reaction mechanism for reduction of 1 using Co(II)salen as mediator
at the anode, enabling use of a simple undivided cell. This is a big advantage of
using such an ionic liquid system.
Electrocatalytic reduction of 2 using a Co(II)salen mediator in the ionic liquid
BMIMBF4
Although reductive cyclization cannot be achieved, as demonstrated above,
debromination proceeds well with Co(II)salen as mediator in the ionic liquid
BMIMBF4. With this in mind, macro-scale electrocatalytic reduction of 2 using the
Co(II)salen mediator was also investigated in BMIMBF4. First, cyclic voltammetry
measurements of 2 were carried out separately in DMF and BMIMBF4. The
reduction peaks were observed at -2.05 and -1.91 V vs. SCE, respectively. The
reduction peak potential in BMIMBF4 thus obtained was more negative than that of
Co(II)salen (Erped = -1.23 V vs. SCE).
Figure 5 shows the cyclic voltammograms of Co(II)salen with (curve b) and
without (curve a) 2 in the ionic liquid BMIMBF4. The cathodic peak current of
Co(II)salen increased markedly after addition of 2, whereas the re-oxidation peak
disappeared completely. Because the reduction potential (Erped) of 2 is -1.91 V vs.
SCE in BMIMBF4, the enhanced cathodic current of Co(II)salen seems to be a
typical catalytic current.
Hence, constant-potential electrolysis of 2 was carried out using the Co(II)salen
complex as mediator in BMIMBF4 at -1.30 V vs. SCE, at which potential 2 cannot
be reduced. Furthermore, the effects of anode materials on the yield of the cathodic
reduction product of 2 were investigated. The electrolytic results are shown in
Table 2. When a platinum plate was used as anode, the reduction product 1,4-
dihydronaphthalene (5), only, was formed in 38 % yield (Table 2, entry 1). Use of a
sacrificial aluminium anode resulted in yields comparable with those obtained with
a Pt anode (Table 2, entry 2). When an iron (Fe) plate was used as a sacrificial
anode, the yield of 5 increased to 53 % (entry 3). However, in this case, the
123

Electrocatalytic debromination of open-chain and cyclic dibromides
97
Fig. 5 Cyclic voltammograms
of (a) Co(II)salen (3 mM) and
(b) Co(II)salen (3 mM) in the
presence of 2 (5 mM) at a GC
electrode (d = 3 mm) in
BMIMBF4. Scan rate: 100 mV/s
Table 2 Effect of anode materials on electrocatalytic reduction of 2 in ionic liquid
Br
+ ne
GC cathode
in BMIMBF4
Br
5
2
Entry
Anode material
Yield of 5 (%)a
1
Pt
38
40
53
68
73
2
Al
3
Fe
4
5b
DSA
DSA
Electrolytic conditions: substrate 2, 0.5 mmol in 10 mL BMIMBF4; Co(II)salen, 0.05 eq; electrolysis
potential, -1.30 V vs. SCE; charge passed, 4 F mol-1
According to 1H NMR
a
b
Electrolysis in 0.1 M Bu4NBF4/DMF
Table 3 Reuse of Co(II)salen
and ionic liquid BMIMBF4 for
electrocatalytic reduction of 2
Cycle
Yield of 5 (%)
1st
68
52
37
69a
2nd
3rd
4th
4 F mol-1 charge was passed
a
0.05 eq fresh Co(II)salen was
added
electrolyte became very dark and viscous after electrolysis, and unidentified
precipitates were observed. When DSA was used as an anode, the yield of 5
increased to approximately 70 % (entry 4). Similar anode effects on product yield
123

98
Y. Shen et al.
e
Co(I)salen Co(II)salen
-Br-
e
Co(I)salen Co(II)salen
-Br-
Br
Br
Br
5
2
B
Scheme 2 Proposed reaction mechanism for the reduction of 2 using Co(II)salen mediator
were observed for 1. For comparison, electrolysis was also performed in DMF using
n-Bu4NBF4 as supporting electrolyte. Although the yield of 5 obtained in DMF
(Table 2, entry 5) was slightly higher than that in the ionic liquid (entry 4), the
workup process after electrolysis of 2 proved difficult. That is, both the product 5
and the Co(II)salen complex were extracted with a non-polar organic solvent from
the DMF solution. In contrast, with the ionic liquid, the product could be extracted
conveniently with diethyl ether and the Co(II)salen mediator remained in the ionic
liquid phase. The remaining mediator could be reused for subsequent electro-
catalytic debromination of 2.
Reuse of Co(II)salen and ionic liquid BMIMBF4 for electrocatalytic reduction
of 2
To demonstrate the advantages of using an ionic liquid, we examined the possibility
of reusing both the Co(II)salen mediator and the ionic liquid. After electrolysis of 2
in BMIMBF4, the ionic liquid was readily recovered simply by extracting 5 with
diethyl ether. The recovered ionic liquid was then dried under vacuum for 6 h to
remove any remaining ether or moisture and was reused for subsequent electrolysis.
The second electrolysis was performed in the recovered ionic liquid with addition of
2. The experimental results are listed in Table 3. The yield decreased from the first
cycle (68 %) to the third cycle (37 %). Because the amount of Co(II)salen lost
during workup can be disregarded, the sharp decrease in yield seems to be
attributable to deactivation and/or decomposition of Co(II)salen during electrolysis.
However, when 0.05 equiv fresh Co(II)salen was added with 2 before the fourth
cycle, the yield of 5 increased to 69 %. This clearly suggests that the ionic liquid is
reusable although maintaining the activity of the mediator remains an issue.
Taking into account the reported mechanism of electrocatalytic reduction of
organic halides with similar complexes [12] we propose the reaction mechanism
shown in Scheme 2. First, one-electron reduction of Co(II)salen generates
Co(I)salen, which reacts with the starting dibromide 2 to form radical B and a
bromide ion. Next, B reacts with electrogenerated Co(I)salen to provide 5.
Conclusions
In summary, we have successfully performed electroreductive debromination
in several ionic liquids, including BMIMBF4, using a Co(II)salen complex as
123

Electrocatalytic debromination of open-chain and cyclic dibromides
99
mediator. The catalytic effect of the Co(II)salen mediator on the reduction of
open-chain and cyclic dibromides in ionic liquids was investigated by cyclic voltam-
metry. Macro-scale controlled potential electrolysis gave the corresponding
debrominated product in moderate to good yields. The workup after electrolysis
proved to be much simpler for the ionic liquid than that for conventional organic
solvents. The possibility of the reuse of the ionic liquid was demonstrated to some
extent. Further efforts to enable reuse of the ionic liquid and mediator are now in
progress.
References
1. A.E. Visser, R.P. Swatloski, R.D. Rogers, Green Chem. 2, 1 (2001)
2. T. Welton, Chem. Rev. 99, 2071 (1999)
3. P. Wassersheid, W. Keim, Angew. Chem. Int. Ed. 39, 3772 (2000)
4. R.D. Rogers, K.R. Seddon, S. Volkov (eds.), Green Industrial Applications of Ionic liquids (Kluwer,
Dordrecht, 2002)
5. R.D. Rogers, K.R. Seddon (eds.), Ionic liquids as Green Solvents, Progress and Prospects (American
Chemical Society, Washington, DC, 2003)
6. T. Fuchigami, S. Inagi, in Electrochemical Aspects of Ionic Liquids, 2nd edn., ed. by H. Ohno (Wiley,
New Jersey, 2011), Chap. 8, pp. 101–127
7. P. Hapiot, C. Lagrost, Chem. Rev. 108, 2238 (2008)
8. J. Yoshida, K. Kataoka, R. Horcajada, A. Nagaki, Chem. Rev. 108, 2265 (2008)
9. M.C. Buzzeo, R.G. Evans, R.G. Compton, ChemPhysChem 5, 1106 (2004)
10. D.S. Silvester, R.G. Compton, Z. Phys. Chem. 220, 1247 (2006)
11. A.J. Fry, J. Electroanal. Chem. 546, 35 (2003)
12. L. Gaillon, F. Bedioui, J. Mol. Catal. A 214, 91 (2004)
13. B.K. Sweeny, D.G. Peters, Electrochem. Commun. 3, 712 (2001)
14. M.C. Lagunas, D.S. Silvester, L. Aldous, R.G. Compton, Electroanalysis 18, 2263 (2006)
15. M. Mellah, S. Gmouh, M. Vaultier, V. Jouikov, Electrochem. Commun. 5, 591 (2003)
´ ´
16. R. Barhdadi, C. Comminges, A.P. Doherty, J.Y. Nedelec, S. O’Toole, M. Troupel, J. Appl. Elect-
rochem. 37, 723 (2007)
17. T. Sawamura, S. Kuribayashi, S. Inagi, T. Fuchigami, Org. Lett. 12, 644 (2010)
18. M. Tokuda, A. Hayashi, H. Suginome, Bull. Chem. Soc. Jpn. 64, 2590 (1991)
19. P. Hapiot, D. Lorcy, A. Tallec, R. Carlier, A. Robert, J. Phys. Chem. 100, 14823 (1996)
20. C. Lagrost, D. Carrie, M. Vaultier, P. Hapiot, J. Phys. Chem. A 107, 745 (2003)
123