Electrochemical reduction of cinnamonitrile in the presence of carbon dioxide: Synthesis of cyano- and phenyl-substituted propionic acids

Article abstract of DOI:10.1071/CH08092

Cyano- and phenyl-substituted propionic acids were synthesized simply and efficiently by electrocarboxylation of cinnamonitrile in undivided cells using the non-noble metal nickel as cathode and magnesium as the anode. The radical anion generated by the e

Full text of DOI:10.1071/CH08092

Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2008, 61, 526–530
www.publish.csiro.au/journals/ajc
Electrochemical Reduction of Cinnamonitrile in the
Presence of Carbon Dioxide: Synthesis of Cyano-
and Phenyl-Substituted Propionic Acids
Huan Wang,A Mei-Yu Lin,A Kai Zhang,A Su-Jiao Li,A and Jia-Xing LuA,B
AShanghai Key Laboratory of Green Chemistry and Chemical Processes, Department of Chemistry,
East China Normal University, Shanghai 200062, China.
BCorresponding author. Email: jxlu@chem.ecnu.edu.cn
Cyano-andphenyl-substitutedpropionicacidsweresynthesizedsimplyandefficientlybyelectrocarboxylationofcinnamo-
nitrile in undivided cells using the non-noble metal nickel as cathode and magnesium as the anode. The radical anion
generated by the electroreduction of cinnamonitrile in the absence of CO2 is involved in several competitive reactions that
lead to the formation of linear hydrodimers, cyclic hydrodimers, saturated dihydro products, and glutaronitrile derivatives.
While under 101.325 kPa of CO2, the electrocarboxylation could easily be carried out in the absence of additional catalysts,
and with good yield (84.8%).The influence of various synthetic parameters, such as the nature of the electrode, the working
potential, the concentration, and the temperature, on the electrocarboxylation reaction was investigated.
Manuscript received: 4 March 2008.
Final version: 16 April 2008.
400
Introduction
V [V sϪ1
a – 0.1
b – 0.2
c – 0.5
d – 1.0
e – 2.0
]
e
The syntheses of cyano- and phenyl-substituted propionic acids
and their esters are of interest owing to their biochemical prop-
erties as inhibitors of bovine carboxypeptidase A.[1] They are
also key molecules in the synthesis of γ-aminobutyric acid
(GABA),[2,3] an important inhibitory neurotransmitter in the
central nervous system.[4] The conventional synthesis of cyano-
and phenyl-substituted propionic acids and their esters, based
on the condensation reaction of ethyl cyanoacetate with alde-
hydes or ketones combined with the simultaneous reduction
of the C=C bond,[5–7] suffers from several drawbacks such
as expensive and complicated catalysts, and high pressure of
hydrogen.
300
200
100
0
d
c
b
a
It is well known, however, that electrocarboxylation is a
powerful pathway for the production of carboxylic acids.[8–12]
This methodology has been extensively used for obtaining alter-
native synthetic routes to the manufacture of pharmaceutical
industrial products.[13,14] Gennaro[15] and Filardo[16] et al. also
have reported the electrocarboxylation of chloroacetonitrile to
cyanoacetic acid. To achieve high yield of the electrocarboxyla-
tion, sacrificial anodes were normally used.[8–16]
Ingeneral, platinumisthemostcommonlyusedworkingelec-
trodeintheelectrocarboxylationreaction, butitisveryexpensive
for industrial production. Thus we are interested in using a non-
noble metal or other cheap material as a cathode. Recently, we
reported the electrochemical carboxylation of activated olefins
in ionic liquid,[17] cinnamate esters,[18] and styrene[19] in MeCN,
using titanium, nickel, etc., as a cathode electrode. As one of our
continuing studies on electrochemical carboxylation, we suc-
ceeded in an efficient electrochemical fixation of CO2 molecules
to cinnamonitrile to give cyano- and phenyl-substituted propi-
onic acids in moderate to high yield with a Ni electrode. In
Ϫ1.0
Ϫ1.2
Ϫ1.4
Ϫ1.6
Ϫ1.8
Ϫ2.0
Ϫ2.2
E [V] v. Ag/AgI/IϪ
Fig. 1. Cyclic voltammograms obtained at a glass carbon electrode in
MeCN–0.1 M Et4NBF4 for 5 mM cinnamonitrile with scan rates of 0.1–
2.0V s−1
.
the present paper, electroreduction of cinnamonitrile was stud-
ied by cyclic voltammetry and controlled-potential electrolysis
first. Then electrocarboxylation was carried out in an undivided
cell and synthesis conditions were optimized.
Results and Discussion
Electrochemical Reduction of Cinnamonitrile
The cyclic voltammetric behaviour for reduction of cinnamoni-
trile at a glass carbon (GC) electrode in MeCN is shown in
Fig. 1. A one-electron reduction peak at −1.46V (v. Ag/AgI/I−)
and a second chemically irreversible reduction at −2.0V
© CSIRO 2008
10.1071/CH08092
0004-9425/08/070526

Electrochemical Reduction of Cinnamonitrile
527
Ph
Ph
Ph
Ph
CN
CN
CN
CN
CN
Ph
CN
ϩ
NH2 ϩ
ϩ
Ph
Ph
CN
1
2
3
4
5
–
Ϫ1.6 V
Ϫ2.1 V
48%
36%
42%
33%
10%
29%
2%
Scheme 1.
ϩ eϪ
ϩ 2Hϩ
ϩ eϪ
CN
CN
Ph
CN
Ph
Ph
Ϫ
1
6
2
Ϫ
Ph
Ph
ϩ 2Hϩ
Ph
Ph
CN
CN
CN
CN
Ϫ
7
3
ϩ Hϩ
Ph
Ph
Ph
Ph
N
ϩ Hϩ
NH
2
CN
Ϫ
CN
8
4
CN
CN
CN
ϩ Hϩ
CNCHϪ2
CN
CN
ϩ
Ph
Ph
Ϫ
Ph
1
5
Scheme 2.
(v.Ag/AgI/I−) are observed at a scan rate of 0.1V s−1. When the
scan rate is increased, the first peak becomes partially reversible,
whereas the second remains irreversible (Fig. 1). This behaviour
is indicative of a chemical reaction following the first electron
transfer reaction.
To understand the process, preparative-scale electrolysis of
cinnamonitrile in MeCN was carried out at the two reduction
peaks, respectively (Scheme 1). The electroreduction at −1.6V
(v. Ag/AgI/I−) resulted in two forms of hydrodimer, the linear
hydrodimer 3 (48%), and a cyclic hydrodimer 4 (42%), with a
small amount of glutaronitrile derivative 5 (10%). At −2.1V (v.
Ag/AgI/I−), the saturated dihydro product 2 (2%) was formed
by the electroreduction (together with 36% of 3, 33% of 4, and
29% of 5). This indicates that the electron transfer numbers for
these two reductions are both one.
700
600
500
400
300
200
100
0
Ϫ1.0
Ϫ1.2
Ϫ1.4
Ϫ1.6
Ϫ1.8
Ϫ2.0
Ϫ2.2
E [V] v. Ag/AgI/IϪ
According to the literature,[20–23] the experimental results are
consistent with the reaction mechanism depicted in Scheme 2.
Electroreduction of cinnamonitrile gives rise to the correspond-
ing radical anion 6.After another one-electron reduction of 6 and
further protonation, the saturated dihydro product 2 is obtained.
Meanwhile, alkene radical anion 6 could also go through a
dimerization process, with the formation of the dimerized dian-
ion 7, which could be involved in two kinds of protonation
processes. In the first route, 7 gets two protons to produce the
linear hydrodimer 3. The other involves a Dieckmann condensa-
tion process of anion intermediate 8, protonated from dimerized
dianion 7, to cyclic hydrodimer 4. However, because the acid-
ity of MeCN, the conjugate base of MeCN, CNCH−2 , can react
Fig. 2. Cyclic voltammograms obtained at a glass carbon electrode
in MeCN–0.1 M Et4NBF4 for 25 mM cinnamonitrile with scan rates of
0.1V s−1 in the absence (- - -) and presence (—) of saturated CO2.
with the unreacted substrate by Michael addition, followed by
proton transfer from another solvent molecule, to produce the
glutaronitrile derivative 5 via canomethylation.
Electrochemical Carboxylation of Cinnamonitrile
When the cyclic voltammograms were obtained in MeCN sat-
urated with CO2, different behaviours were observed (Fig. 2).

528
H. Wang et al.
Ϫ
Ϫ
COO
COO
ϩ eϪ ϩ CO2
CN
CN
CN
CN
ϩ
Ph
Ph
ϩ
Ph
10
Ph
COO
0.1M Et4NBF4–MeCN
COO
Ϫ
Ϫ
1
11
9
Scheme 3.
Table 1. Electrocarboxylation of cinnamonitrile under various conditions
Solution of cinnamonitrile in MeCN–Et4NBF4, undivided cell, CO2 atmosphere, 2 F mol−1 of
cinnamonitrile
Entry
Electrodes
E [V]
Concentration [mM]
T [K] Yield [%]A
Ratio 9:10:11
1
2
3
4
5
6
7
8
Stainless steel–Mg
Cu–Mg
C–Mg
Ti–Mg
Ni–Mg
Ni–Zn
−1.70
−1.70
−1.70
−1.70
−1.70
−1.70
−1.60
−1.65
−1.75
−1.80
−1.85
−1.90
−1.75
−1.75
−1.75
−1.75
−1.75
−1.75
−1.75
−1.75
−1.75
−1.75
−1.75
100
100
100
100
100
100
100
100
100
100
100
100
50
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
288
278
273
67.1
62.4
35.4
47.8
71.7
50.7
64.8
68.5
72.6
72.5
72.1
72.8
72.3
71.7
72.1
72.4
73.1
72.0
56.7
47.9
75.2
80.3
84.8
71:13:14
35:18:48
79:21:00
65:6:29
81:1:18
54:17:29
79:14:7
80:11:9
74:6:20
60:12:28
62:7:31
59:13:28
71:8:21
67:26:7
82:14:4
77:16:7
58:23:19
57:12:31
50:25:25
48:19:33
56:10:34
42:8:50
38:8:54
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
Ni–Mg
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
75
125
150
175
200
250
300
100
100
100
AGas chromatography yield.
The firstreduction peak current increased, whereas the second
peak disappeared. Hence, carboxylation between the electro-
generated intermediate and CO2 is presumably more rapid than
dimerization of the radical anion.
To support this assumption, the controlled-potential elec-
troreduction of cinnamonitrile in MeCN saturated with CO2 was
carried out using a stainless steel plate as the cathode and an
Mg rod as the anode. 2-Cyano-3-phenylpropionic acid 9 was
obtained as the principal product, accompanied by 3-cyano-2-
phenylpropionic acid 10 and 2-cyano-3-phenylsuccinic acid 11
(Scheme 3).
With the aim of optimizing the yields, we focussed our atten-
tion on the influence of various synthetic parameters on the
process, such as the nature of the electrode, the electrolysis
potential, the concentration and the temperature. The results of
the electrolyses are reported in Table 1.
materials (stainless steel, Cu, C, Ti, Ni) as the cathode and
the same Mg anode. The yield of carboxylic acids decreased
depending on the cathode materials employed in the following
order: Ni (71.7%) > stainless steel (67.1%) > Cu (62.4%) >Ti
(47.8%) > C (35.4%).As nickel complexes can catalyze electro-
chemical incorporation of CO2 into alkenes, the excellent result
from the nickel cathode may be associated with its catalytic acti-
vation for substrates. This will be pursued in more detail in the
future.
In addition, the nature of the anodic material strongly deter-
mined the yield or reactivity of this electrocarboxylation reac-
tion. The replacement of the Mg by a Zn anode (with a Ni
cathode, entry 6) made the global carboxylation yield drop from
71.7to50.7%.Thisisattributedtothedifferentreducibilityofthe
metal cation generated at the anode.The reduction formal poten-
tial of Zn2+ (EZon2+/0 = −0.7626V v. normal hydrogen electrode
(NHE)) is much more positive than Mg2+ (EMo g2+/0 = −2.356V
v. NHE).[24] It indicates that Zn2+ is reduced more easily on
the cathode, which will compete with the electrocarboxylation
reaction, causing the decrease of the electrocarboxylation yield.
Actually, Zn particles were found on the cathode surface after
electrolysis.
Influence of the Nature of the Electrode
The reaction yield and selectivity were shown to be dependent
on the reaction conditions, particularly on the nature of the elec-
trodes. Therefore, we first investigated the effect of different
electrode materials on the electrocarboxylation of cinnamoni-
trile keeping other reaction conditions the same. Table 1 (entries
1–5) presents the results obtained with the use of different
According to the above experimental results, the most effec-
tive electrode couple is Ni as cathode and Mg as anode. Hence,
this electrode couple was applied in the following investigations.

Electrochemical Reduction of Cinnamonitrile
529
Influence of the Electrolysis Potential
(Et4NBF4) was prepared according to the literature.[26] Other
reagents were used as received.
Electrocarboxylation of cinnamonitrile under various electro-
lysis potentials was examined. As shown in Table 1 (entries
5 and 7–9), the carboxylation yield increased from 64.8 to
72.6% when the electrolysis potential changed from −1.60 to
−1.75V, indicating that a more negative potential favours elec-
trocarboxylation of cinnamonitrile in this region. However, the
sensitivity of global yield becomes negligible at potentials more
negative than −1.80V (Table 1, entries 10–12). Considering
minimizing energy costs for this electrocarboxylation in practi-
cal applications, −1.75V is the best working potential for further
investigation.
Electroanalytical Experiments
The electroanalytical experiment was carried out in a dry MeCN
solution containing Et4NBF4 as a supporting electrolyte in a one-
compartment electrochemical cell equipped with a gas inlet, a
glassy carbon electrode as the working electrode, a platinum
spiral as the counter-electrode and an Ag|AgI|0.1 M n-Bu4NI in
DMF as the reference electrode. Potential scan was performed
by a CHI 650 instrument.
Electrochemical Reduction of Cinnamonitrile
Influence of Concentration
The controlled-potential electroreduction was carried out in
a mixture of cinnamonitrile (0.1 M) and Et4NBF4 (0.1 M)
in 10 mL dry MeCN under a slow stream of N2 in a one-
compartment electrochemical cell equipped with a Mg sacri-
ficial anode and a stainless steel cathode (8 cm2) at 298 K, until
2 F mol−1 of charge was passed. The identification and quantifi-
cation of the products was done by gas chromatography–mass
spectrometry (GC–MS).
Electrocarboxylation of cinnamonitrile with different initial
concentrations was performed to investigate the influence of
concentration. As shown in Table 1 (entries 9 and 13–18), the
global electrocarboxylation yield was almost the same with low
initial amounts of substrate (50–200 mM). When the electro-
lysis was carried out in a more concentrated solution (more than
200 mM of cinnamonitrile), the carboxylation yield decreased
dramatically with increasing the concentration (Table 1, entries
19–20).According to our experimental results, the concentration
of cinnamonitrile that favoured the electrocarboxylation in good
yield is within the range 50–200 mM.
Electrochemical Carboxylation of Cinnamonitrile
(Typical Procedure)
Cinnamonitrile (0.1 M) was electroreduced as described above
under a slow stream of CO2.
Influence of Temperature
The reaction mixture was distilled under reduced pressure.
The residue was esterified in 10 mL DMF by adding anhydrous
K2CO3 (1 mmol) and MeI (3 mmol) and stirring the mixture at
323 K for 5 h. The solution was hydrolyzed and extracted with
Et2O, and the organic layers were washed with H2O, dried over
MgSO4, and evaporated. The methyl esters were isolated by
column chromatography with petroleum spirit/ethyl acetate as
eluent.After isolation and identification of the products, working
curves were used with biphenyl as internal standard for analysis
of the electrochemical carboxylation.
It is well known that CO2 solubility depends on tempera-
ture, which is larger at lower temperatures.[25] The solubility
may also affect the electrocarboxylation reaction. So to inves-
tigate the effect of temperature on the electrocarboxylation of
cinnamonitrile, a set of electrolyses was performed at differ-
ent temperatures. A strong increase in the global yield was
observed at lower temperatures (Table 1, entries 9 and 21–23),
in agreement with the CO2 solubility.
Conclusions
In some reactions, nickel (Ni), copper (Cu), titanium (Ti),
and graphite (C) were also used as cathode, and zinc (Zn) was
used as anode.
The electroreduction of cinnamonitrile was carried out in MeCN,
giving a linear hydrodimer and cyclic hydrodimer as the major
products, with small amount of glutaronitrile derivative and sat-
urated dihydro product. The reaction mechanism has also been
proposed.
Cyano- and phenyl-substituted propionic acids and their
esters (valuable inhibitors of bovine carboxypeptidase A) have
been obtained, under mild conditions, in good to high yields
by electrochemical reduction (under a CO2 atmosphere) of cin-
namonitrile. The influence of the electrolysis conditions on the
yields has been studied. Lower temperature, lower substrate con-
centrations, and a suitable potential were effective in increasing
dramatically the global yields (84.8%).
Acknowledgements
The present work was supported by the National Natural Science Foundation
of China (no. 20573037), the Natural Science Foundation of Shanghai (no.
05JC1470), a Shanghai Leading Academic Discipline Project (B409), and
PhD Program Scholarship Fund of ECNU 2007.
References
[1] L. D. Byers, R. Wolfenden, Biochemistry 1973, 12, 2070.
doi:10.1021/BI00735A008
[2] M. Calmes, F. Escale, J. Martinez, Tetrahedron Asymmetry 2002, 13,
293. doi:10.1016/S0957-4166(02)00075-7
[3] P. Camps, D. Munoz-Torrero, L. Sanchez, Tetrahedron Asymmetry
2004, 15, 311. doi:10.1016/J.TETASY.2003.10.035
Experimental
General
1H NMR spectra were recorded on an AVANCE 500 (500 MHz,
Bruker, Germany) spectrometer in CDCl3 with Me4Si as an
internal standard. Mass spectra were obtained on a 5973N
spectrometer connected with an HP 6890 gas chromatograph
(Agilent, USA). Electrochemicalmeasurementswerecarriedout
using a CHI 650 instrument (CHI 650 Instruments, USA). Ace-
tonitrile (MeCN) and dimethylformamide (DMF) were kept over
4-Å molecular sieves. Tetraethylammonium tetrafluoroborate
[4] P. Krogsgaard-Larsen, Compr. Med. Chem. 1990, 3, 493.
[5] K. Motokura, N. Fujita, K. Mori, T. Mizugaki, K. Ebitani, K. Kaneda,
TetrahedronLett. 2005, 46, 5507. doi:10.1016/J.TETLET.2005.06.053
[6] M. Baidossi, A. V. Joshi, S. Mukhopadhyay, Y. Sasson, Tetrahedron
Lett. 2005, 46, 1885. doi:10.1016/J.TETLET.2005.01.092
[7] F. Goettmann, D. Grosso, F. Mercier, F. Mathey, C. Sanchez, Chem.
Commun. 2004, 1240. doi:10.1039/B403491K
[8] M. Tokuda, A.Yoshikawa, H. Suginome, H. Senboku, Synthesis 1997,
1143. doi:10.1055/S-1997-1327

530
H. Wang et al.
[9] H. Senboku, H. Komatsu,Y. Fujimura, M. Tokuda, Synlett 2001, 418.
[18] H. Wang, K. Zhang,Y.-Z. Liu, M.-Y. Lin, J.-X. Lu, Tetrahedron 2008,
64, 314. doi:10.1016/J.TET.2007.10.104
[19] H. Wang, M.-Y. Lin, F.-J. Fang, T.-T. Chen, J.-X. Lu, Chin. J. Chem.
2007, 25, 913. doi:10.1002/CJOC.200790177
[10] F. Koster, E. Dinjus, E. Dunach, Eur. J. Org. Chem. 2001, 2507.
doi:10.1002/1099-0690(200107)2001:13<2507::AID-EJOC2507>
3.0.CO;2-P
[11] H. Senboku, H. Kanaya, M. Tokuda, Synlett 2002, 140. doi:10.1055/
S-2002-19350
[20] J. H. P. Utley, M. Gullu, M. Motevalli, J. Chem. Soc., Perkin Trans. 1
1995, 1961. doi:10.1039/P19950001961
[12] O. Scialdone, M. A. Sabatino, C. Belfiore, A. Galia, M. P. Pater-
nostro, G. Filardo, Electrochim. Acta 2006, 51, 3500. doi:10.1016/
J.ELECTACTA.2005.09.044
[21] I. Fussing, M. Gullu, O. Hammerich, A. Hussain, M. F. Nielsen,
J. H. P. Utley, J. Chem. Soc., Perkin Trans. 2 1996, 649. doi:10.1039/
P29960000649
[13] J. Damodar, S. R. K. Moha, S. R. J. Reddy, Electrochem. Commun.
2001, 3, 762. doi:10.1016/S1388-2481(01)00263-6
[22] I. Fussing, O. Hammerich, A. Hussain, M. Nielsen, J. H. P. Utley, Acta
Chem. Scand. 1998, 52, 328.
[14] A. K. Datta, P. A. Marron, C. J. H. King, J. H. Wagenknecht, J. Appl.
Electrochem. 1998, 28, 569. doi:10.1023/A:1003289800341
[15] A. A. Isse, A. Gennaro, J. Electrochem. Soc. 2002, 149, D113.
doi:10.1149/1.1490358
[16] O. Scialdone, A. Galia, C. Belfiore, G. Filardo, G. Silvestri, Ind. Eng.
Chem. Res. 2004, 43, 5006. doi:10.1021/IE034275+
[17] H. Wang, G.-R. Zhang, Y.-Z. Liu, Y.-W. Luo, J.-X. Lu, Electrochem.
Commun. 2007, 9, 2235. doi:10.1016/J.ELECOM.2007.06.031
[23] N. Kise, S. Iitaka, K. Iwasaki, N. Ueda, J. Org. Chem. 2002, 67, 8305.
doi:10.1021/JO026183K
[24] A. J. Bard, L. R. Faulkner, Electrochemical Methods: Fundamentals
andApplications, 2nd edn 2001 (John Wiley & Sons: NewYork, NY).
[25] A. Gennaro, A. A. Isse, E. Vianello, J. Electroanal. Chem. 1990, 289,
203. doi:10.1016/0022-0728(90)87217-8
[26] D. T. Sawyer, J. L. Robert, Jr, Experimental Electrochemistry for
Chemists 1974 (Wiley-Interscience, New York, NY).