Electrochemical, Spectroscopic, and Quantum Chemical Study of Electrocatalytic Hydrogen Evolution in the Presence of N-Methyl-9-phenylacridinium Iodide

Article abstract of DOI:10.1134/S1070428019070030

Main intermediates in the electrocatalytic hydrogen evolution promoted by N-methyl-9-phenyl-acridinium iodide have been identified and characterized using cyclic voltammetry (CV) and NMR and ESR spectroscopy. A probable mechanism of the process has been p

Full text of DOI:10.1134/S1070428019070030

ISSN 1070-4280, Russian Journal of Organic Chemistry, 2019, Vol. 55, No. 7, pp. 938–943. © Pleiades Publishing, Ltd., 2019.
Russian Text © The Author(s), 2019, published in Zhurnal Organicheskoi Khimii, 2019, Vol. 55, No. 7, pp. 1030–1037.
Electrochemical, Spectroscopic, and Quantum Chemical Study
of Electrocatalytic Hydrogen Evolution in the Presence
of N-Methyl-9-phenylacridinium Iodide
A. V. Dolganova,* O. V. Tarasovaa, A. V. Balandinaa, O. Yu. Chernyaevaa, V. Yu. Yurovaa,
Yu. M. Selivanovaa, and A. D. Yudinaa
a Ogarev Mordovian National Research State University, Saransk, Russia
*e-mail: dolganov_sasha@mail.ru
Received September 13, 2018; revised April 1, 2019; accepted April 12, 2019
Abstract—Main intermediates in the electrocatalytic hydrogen evolution promoted by N-methyl-9-phenyl-
acridinium iodide have been identified and characterized using cyclic voltammetry (CV) and NMR and ESR
spectroscopy. A probable mechanism of the process has been proposed on the basis of DFT quantum chemical
calculations. Analysis of structural and energetic parameters of intermediate products has shown that the
electrocatalytic process involves formation of N-protonated radical cation, followed by reductive elimination of
molecular hydrogen in the bimolecular reaction of two radical cations.
Keywords: electrocatalysis, generation of hydrogen, reaction mechanism, electrochemical properties,
10-methyl-9-phenylacridinium iodide.
DOI: 10.1134/S1070428019070030
Electrolysis of water is an environmentally benign
method of generation of molecular hydrogen. Water is
the most accessible and almost inexhaustible source of
hydrogen, and transformation of the latter back into
water and electricity, which is utilized in hydrogen fuel
cells, ensures environmental safety and renewability of
the process [1]. Platinum is the most efficient catalyst
of the hydrogen evolution reaction (HER) in acid
medium, but due to high cost of platinum and its
sensitivity to traces of various impurities, there is
a need for catalysts based on accessible and inexpen-
sive metals [2–13]. Most problems related to the high
cost and low stability can be solved by development of
metal-free molecular catalysts consisting of elements
present in the atmosphere in large amounts (C, H,
N, O). N-Methyl-9-phenylacridinium iodide (PhAcrI)
is a promising candidate for the design of a family of
efficient metal-free electrocatalytic systems for genera-
tion of molecular hydrogen. As shown previously,
PhAcrI-based electrocatalytic hydrogen evolution
systems are efficient in both aprotic solvents and in
aqueous media at low pH values [14–15]. Figure 1
shows a catalytic scheme for generation of molecular
hydrogen, which was proposed on the basis of electro-
chemical data. In order to get a deeper insight into the
catalytic process and optimize the catalytic system, in
this work we studied in detail the mechanism of
electrocatalytic hydrogen evolution reaction and deter-
mined main intermediates using cyclic voltammetry,
NMR and ESR spectroscopy, and quantum chemical
calculations.
The mechanism outlined in Fig. 1 is essentially
similar to the classical homolytic and heterolytic pro-
ton reduction mechanisms in the presence of transition
metal-based electrocatalysts [16]. The heterolytic
mechanism implies two successive electrochemical
reductions of the same PhAcrI molecule with protona-
tion of the anion (EEC path) or protonation of inter-
mediate product (ECEC path); in both cases, the final
product would be 10-methyl-9-phenyl-9,10-dihydro-
acridine or its protonated form.
The homolytic mechanism involves one-electron
reduction of PhAcr+ and subsequent protonation and
bimolecular reaction between the two protonated
species, which is accompanied by evolution of
hydrogen.
As follows from Fig. 1, the first stage in the
proposed mechanism is electrochemical reduction of
acridinium ion to PhAcr· radical which was unam-
938

ELECTROCHEMICAL, SPECTROSCOPIC, AND QUANTUM CHEMICAL STUDY
939
–
Ph
Ph
C
e–
H+
·
e–
Epeak = –1.73 V
N
N
Epeak = –0.86 V
Me
Me
Ph
Ph
EEC
H+
N
N
Me
Me
ECEC
+·
Ph
C
Ph
H
H
e–, H+
1/2H2
H+
Epeak = –1.36 V
N
N
H
Me
Me
Fig. 1. Possible electrocatalytic mechanism of formation of molecular hydrogen in the presence of PhAcrI and competing EEC and
ECEC mechanisms. The potentials are given relative to Fc/Fc+ for acetonitrile solution.
biguously identified by the characteristic half-wave
reduction potential (–0.86 V vs Fc/Fc+). The potential
–0.850 V corresponds to one-electron reduction of the
cation to radical. Further reduction of PhAcr · to
PhAcr– is impossible at that potential, since it requires
a significantly more negative potential (–1.73 V vs
Fc/Fc+). It should be noted that this process is
irreversible even at high potential sweep rates (up to
10 V/s), which is likely to be related to the high
reactivity of PhAcr–. The anion is capable of readily
abstracting a proton from the solvent (acetonitrile)
molecules according to the PCET (proton-coupled
electron transfer) mechanism. Taking into account that
the electrochemical reaction produces a radical
species, it was reasonable to consider its stability
toward disproportionation leading to the initial cation
and anion. If the disproportionation reaction PhAcr· →
PhAcr+ + PhAcr– had taken place, the CV curve would
display increase of the current of the forward reduction
process and decrease of the current of the backward
reaction. We found that repeated potential scan in the
range from 0 to −1 V in a degassed solvent did not
change the CV pattern, and the currents of both pro-
cesses remained unchanged. These findings unambigu-
ously indicate that the radical is stable in solution and
that no its disproportionation occurs.
low acid concentration, the potential shifts to the
anodic region, and increase of the acid concentration
leads to cathodic shift of the potential [14].
The observed pattern corresponds to electro-
catalytic reactions limited by substrate (acid) diffusion
[14, 15, 17]. In the absence of PhAcrI, no reduction
processes were detected in the above potential range;
furthermore, no catalytic activity was observed when
the spent electrode (after electrochemical studies) was
placed in a fresh electrolyte solution containing no
PhAcrI. This means that the electrocatalytic process is
strictly homogeneous and that there is no effect of
[H+], mM
E, V
Fig. 2. (a) Cyclic voltammogram of a 1.0 mM solution of
PhAcrI in acetonitrile containing (1) 0, (2) 6.05, (3) 12.1,
(4) 18.15, (5) 24.2, (6) 30.25, (7) 48.4, (8) 60.5, and
(9) 84.7 mM of HCLO4 and (10) in the absence of PhAcrI.
(b) Plot of Icat/Ip versus HClO4 concentration (mM) at
a potential sweep rate of 200 mV/s; 1.0 mM of PhAcrI in
acetonitrile.
Addition of perchloric acid to a solution containing
PhAcrI resulted in a considerable increase of the
forward reduction current and complete disappearance
of the reverse process (Fig. 2); i.e., the CV curve is
typical of a homogeneous electrocatalytic process: at
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 55 No. 7 2019

DOLGANOV et al.
940
Scheme 1.
Ph
Ph
Ph
H
·
·
H+
H+
I
II
·
N
N
N
H
Me
Me
Me
I
II
adsorption of the catalyst or its decomposition prod-
ucts on the electrode surface (or this effect is negli-
gible). It should be noted that pure 10-methyl-9-
phenyl-9,10-dihydroacridine showed no electrocata-
lytic activity at a potential of –0.850 V in the presence
of 0.1 M perchloric acid; thus, EEC and ECEC mech-
anisms can be completely ruled out. In addition, all CV
curves displayed no characteristic plateau even at
a potential sweep rate of 10 V/s and high concentra-
tions of both perchloric acid (up to 1 M) and PhAcrI
(12.5 mM), indicating fast electrocatalytic process
limited by substrate diffusion to the electrode surface.
We can conclude that, in keeping with the proposed
mechanism, the only possible second stage of the cata-
lytic process under the given conditions is protonation
of PhAcr· at the most basic nitrogen atom with the
formation of N-substituted radical cation (Scheme 1,
path I). Interestingly, a proton could also add to the C9
carbon atom possessing the highest electron density to
form an isomeric radical cation (Scheme 1, path II).
However, the energy of activation along path II is
much higher than for path I.
In order to confirm the catalytic nature of radical
species PhAcr· and exclude all possible stages requir-
ing the second electrochemical reduction, PhAcr+ was
subjected to chemical reduction with metallic silver
[15, 18]. The addition of silver powder to a solution of
PhAcrI in acetonitrile led to the formation of PhAcr·
radicals which were detected by ESR, NMR, and UV
spectroscopy. The radicals generated in the absence of
oxygen were surprisingly stable. However, the ESR
spectrum of the solution (Fig. 3) was considerably
broadened due to formation of noncovalent radical–
radical or radical–cation dimers stabilized by π-stack-
ing [15]. The occurrence of dynamic exchange be-
tween the radical–cation pairs also followed from the
NMR spectra (Fig. 4) of a solution containing stable
PhAcr· radicals. The NMR spectra showed appreciable
broadening of signals belonging to the initial cation,
and these signals became narrower as the concentra-
tion of PhAcr· decreased. After addition of an acid
(a source of protons), signals belonging to the radical
immediately disappeared, whereas those of initial
PhAcr+ appeared again (Figs. 5, 6). Furthermore,
Fig. 3. ESR spectrum (X band) of a solution of PhAcr· in
benzene at room temperature.
δ, ppm
1
1
a new signal appeared in the H NMR spectrum at
Fig. 4. H NMR spectra (600 MHz) of a 3 mM solution of
δ 4.57 ppm (s), which was assigned to molecular
hydrogen; this signal disappeared from the spectrum
after purging with argon (Fig. 6). It should be noted
that no chemical reaction (including formation of
hydrogen) was observed when silver powder was
added to an acidic solution containing no PhAcrI.
PhAcrI in acetonitrile-d3 (1) before and (2) after addition of
fine silver powder and (3) after subsequent regeneration of
the initial cation by adding perchloric acid. Spectrum 2 is
magnified 8 times to demonstrate broadened signals assigned
to radical species. Narrow signals marked with an asterisk
were assigned to the products of reversible radical dimeriza-
tion on the basis of DOSY data.
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 55 No. 7 2019

ELECTROCHEMICAL, SPECTROSCOPIC, AND QUANTUM CHEMICAL STUDY
941
Although other catalytic paths may be proposed on
the basis of the obtained data, they were excluded by
the following considerations:
(1) The EEC path, which is also general for tran-
sition-metal based electrocatalysts, was ruled out
because of the very negative reduction potential of
PhAcr·/PhAcr– (–1.73 V vs Fc/Fc+); the absence of
signals assignable to PhAcr– in the NMR spectra of
3
a solution of chemically generated PhAcr· excluded
2
formation of PhAcr– as a result of disproportionation
of two PhAcr· radicals.
λ, nm
(2) The ECEC path, including formation of
PhAcrH+· and its subsequent reduction to PhAcrH, was
ruled out since pure 10-methyl-9-phenyl-9,10-dihydro-
acridine (PhAcrH) showed no catalytic activity in the
hydrogen evolution reaction.
Fig. 5. UV-vis spectra of a 0.1 mM solution of PhAcrI in
acetonitrile (1) before and (2) after addition of excess silver
powder and (3) after subsequent regeneration of the initial
cation by adding perchloric acid.
The formation of radical cation I was detected by
CV by scanning the potential in the cathodic region to
–1.8 V (Fig. 7). As seen from Fig. 7, the addition of
perchloric leads to the appearance of a new wave at
–1.36 V (vs Fc/Fc+). Most probably, the redox process
PhAcrH+· → PhAcrH occurs at that potential value,
i.e., catalytically inactive species is formed. It is im-
portant that the PhAcrH+ ·/PhAcrH reduction wave
disappears as high acid concentrations, which is very
consistent with the proposed mechanism. Increase of
the acidity accelerates protonation of PhAcrH·, so that
the concentration of PhAcrH+· and hence the rate of
reductive elimination of hydrogen in the bimolecular
reaction of two radical cations PhAcrH+· also increase.
The formation of catalytically inactive 10-methyl-9-
phenyl-9,10-dihydroacridine (PhAcrH) was confirmed
by preparative electrolysis of an acetonitrile solution
containing PhAcrI (0.1 mM) and HClO4 (0.01 M) at
a redox potential of –1.36 V vs Fc/Fc+. When the
δ, ppm
1
Fig. 6. H NMR spectra (600 MHz) of a 3 mM solution of
PhAcrI in acetonitrile-d3 after (1) successive addition of fine
silver powder and perchloric acid solution and (2) subse-
quent removal of molecular hydrogen (δ 4.565 ppm) by
purging the mixture with argon.
1
electrolysis was complete, the H NMR spectrum
contained signals belonging exclusively to PhAcrH.
Interestingly, the amount of electricity consumed for
preparative electrolysis at an acid concentration of
1 mM was larger by 28% than at an acid concentration
of 0.01 M, 0.31 and 0.25 C, respectively. Thus, the
preparative electrolysis data are very consistent with
the data obtained by cyclic voltammetry.
E, V
The proposed mechanism was also confirmed by
DFT quantum chemical calculations using B3LYP
functional and 6-31+G basis set. Analysis of the
vibrational frequencies showed that all assumed inter-
mediate structures correspond to energy minima on the
potential energy surface. Figure 8 shows the energy
profile for the electrocatalytic formation of molecular
Fig. 7. Cyclic voltammograms of a 1.0 mM solution of
PhAcrI in acetonitrile containing (1) 0, (2) 1.21, (3) 2.42,
(4) 3.63, (5) 4.84, (6) 6.05, (7) 7.26, and (8) 8.47 mM of
perchloric acid and (9) of a 20 mM solution of perchloric
acid in acetonitrile without PhAcrI. Potential sweep rate
200 mV s–1, 25°C, 0.1 M [Bu4N][BF4] in acetonitrile, glassy
carbon working electrode. The potentials are given relative
to Fc/Fc+.
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 55 No. 7 2019

DOLGANOV et al.
942
E, kJ/mol
165.58
90
0
85.42
–1.75
–10.55
+ H2
Reaction coordinate
Fig. 8. Calculated energy profile of the hydrogen evolution reaction in the presence of PhAcrI.
hydrogen in the presence of PhAcrI. According to the
results of calculations, the first step of the electro-
catalytic process is the reduction of PhAcr+ to PhAcr·.
The next step is protonation of PhAcr· at the nitrogen
atom with the formation of radical cation. This step
requires an energy of 87.17 kJ/mol to be overcome.
This value is consistent with the energies of activation
of many catalytic hydrogen evolution reactions in the
presence of electrocatalytc systems based on various
metal complexes and enzymes (~90 kJ/mol) [2]. Our
results are very interesting. Though the energy barriers
are fairly similar, catalysts of different natures (metal-
free and organic metal complexes) are characterized
by strongly different pK values. For instance, the
experimental pK value of PhAcrH+· in water is 4, and
the complex [(DPA-bipy)Co(OH2)]n+ [DPA-bipy is
N,N-bis(pyridin-2-ylmethyl)-2,2′-bipyridin-6-ylmethan-
amine which catalyzes the process with an energy of
about 60 kJ/mol] has pK 13.9 [19]. The most probable
reason is that radical cation PhAcrH+· is unstable due
to low degree of charge stabilization; stabilization of
charge in metal complexes is much more effective.
Therefore, we presume that hydride metal complexes
formed after protonation and generally having low pK
values are involved in the next stage (formation of
molecular hydrogen) according to the heterolytic
mechanism. In contrast, hydrogen elimination from
radical cation PhAcrH+· with a fairly high pK value is
energetically more favorable according to the homo-
lytic path since the N–H bond has a very low basicity.
This assumption is well supported by the obtained
data; homolytic bimolecular elimination of hydrogen
requires almost three times lower energy than the
heterolytic path. It should be noted that the formation
of isomeric radical cation II (protonated at the carbon
atom) requires almost twice as high energy. Our results
allow us to unambiguously affirm that protonation of
PhAcr+· involves the nitrogen atom and that this reac-
tion is the rate-limiting step. The results of calculations
are consistent with the experimental data [14], accord-
ing to which the rate of the catalytic process is limited
by the radical protonation step.
Thus, we can convincingly conclude that the
catalytic process follows the homolytic mechanism
which is determined by the low basicity of radical
cation formed as a result of protonation. Presumably,
rational design of the molecular skeleton of acridinium
cation could increase the basicity of the nitrogen atom,
which should reduce the energy of formation of the
radical cation and hence the rate of formation of
molecular hydrogen.
EXPERIMENTAL
1
The H and 13C NMR spectra were recorded on
a Bruker Avance-600 spectrometer (600.13 MHz for
1H) relative to the residual proton and carbon signals
of the deuterated solvent. The data were acquired and
processed using Topspin 2.1 and MestReNova 9.0.0
software, respectively.
The ESR spectra were recorded on a Bruker
Elexsys E500 spectrometer with the following param-
eters: SHF frequency 9.8 GHz, SHF power 200 mW,
band width 40 G, modulation frequency 100 kHz,
modulation amplitude 0.3 G, response time 200 ms,
resolution 2048 points.
The electrochemical data were obtained by cyclic
voltammetry in acetonitrile containing 0.1 M of tetra-
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 55 No. 7 2019

ELECTROCHEMICAL, SPECTROSCOPIC, AND QUANTUM CHEMICAL STUDY
943
5. Solis, B.H. and Hammes-Schiffer, S., Inorg. Chem.,
2011, vol. 50, p. 11252. doi 10.1021/ic201842v
6. Solis, B.H. and Hammes-Schiffer, S., Inorg. Chem.,
butylammonium tetrafluoroborate as supporting elec-
trolyte using a Gamry potentiostat and a 5-mL cell.
The working electrode was a glassy carbon electrode
with a surface area of 0.125 cm2. The electrode was
thoroughly polished and washed before measurements.
A platinum electrode was used as auxiliary, and
a silver chloride electrode (E0 = 0.33 V in MeCN vs
Fc/Fc+), as reference one. All solutions were prelimi-
narily degassed by purging argon.
2014, vol. 53, p. 6427. doi 10.1021/ic5002896
7. Zarkadoulas, A., Field, M.J., Papatriantafyllopoulou, C.,
Fize, J., and Artero, V., Inorg. Chem., 2016, vol. 55,
p. 432. doi 10.1021/acs.inorgchem.5b02000
8. Fogeron, T., Porcher, J.P., Gomez-Mingot, M., Todoro-
va, T., Chamoreau, L.-M., Mellot-Draznieks, C., Li, M.,
and Fontecave, Y., Dalton Trans., 2016, vol. 45,
p. 14754. doi 10.1039/C6DT01824F
9. Liao, R.Z., Wang, M., Sun, L.C., and Siegbahn, P.E.M.,
Dalton Trans., 2015, vol. 44, p. 9736. doi 10.1039/
C5DT01008J
10. Dolganov, A.V., Tarasova, O.V., Moiseeva, D.N., and
Muryumin, E.E., Int. J. Hydrogen Energy, 2016, vol. 41,
p. 9312. doi 10.1016/j.ijhydene.2016.03.131
11. Dolganov, A.V., Belov, A.S., Novikov, V.V., Vologzha-
nina, A.V., Romanenko, G.V., Budnikova, Y.G., Zelin-
skii, G.E., Buzin, M.I., and Voloshin, Y.Z., Dalton
Trans., 2015, vol. 44, p. 2476. doi 10.1039/
C4DT03082F
12. Dolganov, A.V., Belaya, I.G., and Voloshin, Y.Z.,
Electrochim. Acta, 2014, vol. 125, p. 302. doi 10.1016/
j.electacta.2014.01.060
10-Methyl-9-phenylacridinium iodide (PhAcrI)
was synthesized by quaternization of 9-phenylacridine
with methyl iodide in a hermetically closed test tube
for 10 h [19]. According to the NMR data, the product
contained 99.6% of the main substance. 1H NMR spec-
trum (CD3CN), δ, ppm: 8.69 d (2H, 4-H, J = 9.2 Hz),
8.42 d.d (2H, 3-H, J = 9.2, 6.8 Hz), 8.04 d (2H, 1-H,
J = 8.5 Hz), 7.94–7.83 d.d (2H, 2-H, J = 8.5, 6.8 Hz),
7.83–7.73 m (3H, m-H, p-H), 7.56 d (2H, o-H, J =
7.2 Hz), 4.90 s (3H, Me). 13C NMR spectrum
(CD3CN), δC, ppm: 161.65 (C9), 141.70 (C12), 138.74
(C3), 133.32 (Ci), 130.25 (Cp), 130.18 (C1), 129.92
(Co), 128.89 (Cm), 127.82 (C2), 126.21 (C11), 118.68
(C4), 39.15 (Me).
13. Dolganov, A.V., Belov, A.S., Novikov, V.V., Vologzha-
nina, A.V., Mokhir, A., Bubnov, Y.N., and Volo-
shin, Y.Z., Dalton Trans., 2013, vol. 42, p. 4373. doi
10.1039/C3DT33073G
14. Dolganov, A., Tanaseichuk, B., Moiseeva, D.,
Yurova, V., Sakanyan, J., Shmelkova, N., and Loba-
nov, V., Electrochem. Commun., 2016, vol. 68, p. 59.
doi 10.1016/j.elecom.2016.04.015
15. Dolganov, A.V., Tanaseichuk, B.S., Ivantsova, P.M.,
Tsebulaeva, Y.V., Kostrukov, S.G., Moiseeva, D.N.,
Shmelkova, N.M., and Yurova, V.Y., Int. J. Electrochem.
Sci., 2016, vol. 11, p. 9559. doi 10.20964/2016.11.24
ACKNOWLEDGMENTS
The authors thank Prof. V.V. Novikov for recording
the ESR and NMR spectra.
CONFLICT OF INTERESTS
No conflict of interests is declared by the authors.
REFERENCES
1. Panetier, J.A., Letko, C.S., Tilley, T.D., and Head-
Gordon, M., J. Chem. Theory Comput., 2016, vol. 12,
p. 223. doi 10.1021/acs.jctc.5b00968
16. Artero, V. and Montecave, M., Coord. Chem. Rev.,
2005, vol. 249, p. 1518. doi 10.1016/j.ccr.2005.01.014
2. Kaeffer, N., Chavarot-Kerlidou, M., and Artero, V., Acc.
Chem. Res., 2015, vol. 48, p. 1286. doi 10.1021/
acs.accounts.5b00058
17. Artero, V., Chavarot-Kerlidou, M., and Fontecave, M.,
Angew. Chem., Int. Ed., 2011, vol. 50, p. 7238. doi
10.1002/anie.201007987
3. Harshan, A.K., Solis, B.H., Winkler, J.R., Gray, H.B.,
and Hammes-Schiffer, S., Inorg. Chem., 2016, vol. 55,
p. 2934. doi 10.1021/acs.inorgchem.5b02857
18. Koper, N.W., Jonker, S.A., and Verhoeven, J.W., Recl.
Trav. Chim. Pays–Bas, 1985, vol. 104, p. 296. doi
10.1002/recl.19851041106
4. Anjali, B.A., Sayyed, F.B., and Suresh, C.H., J. Phys.
Chem. A, 2016, vol. 120, p. 1112. doi 10.1021/
acs.jpca.5b11543
19. Lewandowska-Andralojc, A., Baine, T., Zhao, X.,
Muckerman, J.T., Fujita, E., and Polyansky, D., Inorg.
Chem., 2015, vol. 54, p. 4310. doi 10.1021/ic5031137
RUSSIAN JOURNAL OF ORGANIC CHEMISTRY Vol. 55 No. 7 2019