The True Fate of Pyridinium in the Reportedly Pyridinium-Catalyzed Carbon Dioxide Electroreduction on Platinum

Article abstract of DOI:10.1002/anie.201808122

Protonated pyridine (PyH⁺) has been reported to act as a peculiar and promising catalyst for the direct electroreduction of CO₂ to methanol and/or formate. Because of recent strong incentives to turn CO₂ into valuable products, this claim triggered great interest, prompting many experiments and DFT simulations. However, when performing the electrolysis in near-neutral pH electrolyte, the local pH around the platinum electrode can easily increase, leading to Py and HCO₃^? being the predominant species next to the Pt electrode instead of PyH⁺ and CO₂. Using a carefully designed electrolysis setup which overcomes the local pH shift issue, we demonstrate that protonated pyridine undergoes a complete hydrogenation into piperidine upon mild reductive conditions (near 0 V vs. RHE). The reduction of the PyH⁺ ring occurs with and without the presence of CO₂ in the electrolyte, and no sign of CO₂ electroreduction products was observed, strongly questioning that PyH⁺ acts as a catalyst for CO₂ electroreduction.

Full text of DOI:10.1002/anie.201808122

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: The True Fate of Pyridinium in the Reportedly Pyridinium-
Catalyzed Carbon Dioxide Electroreduction on Platinum
Authors: Pierre-Yves Olu, Qi Li, and Katharina Krischer
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201808122
Angew. Chem. 10.1002/ange.201808122
Link to VoR: http://dx.doi.org/10.1002/anie.201808122
http://dx.doi.org/10.1002/ange.201808122

10.1002/anie.201808122
Angewandte Chemie International Edition
COMMUNICATION
The True Fate of Pyridinium in the Reportedly Pyridinium-
Catalyzed Carbon Dioxide Electroreduction on Platinum
Pierre-Yves Olu[a], Qi Li[a] and Katharina Krischer*[a]
Abstract: Protonated pyridine (PyH+) has been reported to act as a
peculiar and promising catalyst for the direct electroreduction of CO2
to methanol and/or formate. Because of recent strong incentives to
valorise CO2 into valuable products, this claim triggered
considerable interest, prompting many experiments and numerous
DFT simulations. However, when performing the electrolysis in near-
neutral pH electrolyte, the local pH around the platinum electrode
can easily increase, leading to Py and HCO3- being the predominant
species next to the Pt electrode instead of PyH+ and CO2. Using a
carefully designed electrolysis setup which overcomes the local pH
shift issue, in this communication we demonstrate that protonated
pyridine undergoes a complete hydrogenation into piperidine upon
mild reductive conditions (near 0 V vs. RHE). The reduction of the
PyH+ ring occurs with and without the presence of CO2 in the
electrolyte, and no sign of CO2 electroreduction products was
observed, strongly questioning that protonated pyridine acts as a
suitable catalyst for CO2 electroreduction reaction.
Recently, Costentin et al. expressed strong doubts that PyH+
20]
could catalyse the CO2 electroreduction reaction at all, as they
claimed that the pyridine ring could not be reduced under the
above-mentioned experimental conditions.[21] They also
reminded that an increase of the reduction current peak below 0
V vs. RHE on the cyclic voltammetry of Pt when PyH+ was
added to the CO2-saturated electrolyte was by no means a proof
of enhanced CO2 electroreduction.[21] Indeed, this higher
reduction current is most likely to arise from a stronger local pH
buffering after PyH+ addition, just as other kinds of Brønsted
acids would do.[22] The slightly stronger pH buffering capabilities
of the electrolyte after PyH+ addition can then better resist local
pH shift to alkaline values under reductive conditions and thus
increase the reduction current coming from gaseous hydrogen
evolution.[22] As for the cyclic voltammetry measurements, the
local pH shift is also an issue to consider in electrolysis
measurements. Our motivation for this work was therefore to
perform electrolysis measurements of PyH+ and CO2 on Pt with
minimized local pH shift.
The reduction of anthropogenic CO2 in the atmosphere is one of
the main challenges mankind is facing. An attractive strategy is
to valorise CO2 into small organic molecules such as formic acid
or methanol, thus closing the carbon cycle. Currently, different
routes are being explored. Besides the catalytic[1] and
photocatalytic[2] CO2 hydrogenation reactions, the direct electro-
catalytic reduction of CO2 on the active electrode surface is
considered a promising approach.[3,4]
In this context, a peculiar experimental result attracted a lot
of interest: protonated pyridine (PyH+, aka pyridinium) was
reported to be an efficient catalyst for CO2 electroreduction
reaction on platinum and palladium, achieving Faradaic yields of
30% methanol for Pd,[5] and 10% formate + 20% methanol for
Pt.[6] The formation of methanol and/or formate from CO2
catalysed by PyH+ on Pt has been further reported by other
groups,[7–11] although it is no easy task to draw a logical trend
from these dispersed experimental results. Yet, these
experimental results seemed promising and were tentatively
rationalized by numerous DFT studies, where protonated
pyridine PyH+ was the necessary initial species for catalysing the
CO2 electroreduction via various mechanistic pathways involving
different types of intermediate states of a partially-reduced PyH+
(such as the pyridyl radical or dihydropyridine).[12–17]
First,
suitable
parameters
for
the
electrolysis
measurements need to be evaluated. The working electrode, a
polycrystalline Pt mesh of 10 cm2 electrochemically active
surface area (EASA, see Figure S1 for EASA determination), is
characterized in a big volume (100 mL) of CO2-saturated 1 M
KCl + 10 mM Py electrolyte, which was also used in most of the
above-mentioned experimental studies. The measured pH of the
electrolyte is 5.3, which is near the pKa value of PyH+ (ca. 5.2),
implying that both PyH+ and Py coexist in the electrolyte. The
cyclic voltammogram (Figure 1A) is alike the ones previously
reported under comparable conditions[9,18] and shows a first
reduction current peak at -0.15 V vs. RHE coming mainly from
the reduction of hydrated protons into gaseous hydrogen. Note
that as weak Brønsted acids, PyH+ and carbonic acid are the
main sources of hydrated protons. Once the concentration of
protons near the Pt surface is depleted, the hydrogen evolution
proceeds through the reduction of water below ca. -0.25 V vs.
RHE. The gaseous hydrogen generated is then subsequently
electrooxidized in the following positive-going scan (oxidation
peak around 0.1 V vs. RHE in Figure 1A). The reduction current
peak of -1.5 mA cm-2 at -0.15 V vs. RHE is characteristic of local
pH shift: The concentration of equivalent hydrated protons near
the Pt surface is depleted as the hydrogen evolution reaction
proceeds.[18,21,22] The cyclic voltammogram thus gives a first
indication of the range of current that can be drawn in an
electrolysis measurement without experiencing the problematic
local pH shift.
However, other groups failed to detect any methanol or
formate under the same experimental conditions as in Ref.[6].[18–
[a]
Dr. P.-Y. Olu, Q. Li, Prof. Dr. K. Krischer
Physics Department
A more precise evaluation is done on the same system
using galvanostatic control, as shown in Figure 1B. When a
constant current density of -0.1 mA cm-2 is applied, a stable
electrode potential of ca. -30 mV vs. RHE is measured, in
agreement with the voltammogram of Figure 1A. However, for a
constant current density of -1 mA cm-2, the potential of the
Technical University of Munich (TUM)
James-Franck-Straße 1, 85748 Garching, Germany
E-mail: krischer@tum.de
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201808122
Angewandte Chemie International Edition
COMMUNICATION
electrode quickly decreases, indicating an alkaline local pH shift
near the Pt surface as the current applied is too high for protons
to be replenished under stagnant condition. Convection inside
the cell (using a magnetic stirrer) is needed in order to resist the
local pH shift and to obtain a stable electrode potential of ca. -60
mV vs. RHE (Figure 1B). The difference of potential for the
same constant current with and without stirring is ca. 0.25 V,
which translates into a local pH shift of 4.2 units. Consequently,
if we were to perform the electrolysis measurement (where no
stirring is applied) at -1 mA cm-2, the local pH of the working
After performing an electrolysis of -15 C at -0.1 mA cm-2 in
1 mL CO2-saturated 1 M KCl + 10 mM Py electrolyte, the only
product identified by water-suppressed 1H-NMR is piperidine
(Figure 2). The identification of piperidine is certified by
comparison with commercial pure piperidine solution (see Figure
S2). Apart from gaseous hydrogen evolved during the
electrolysis (see the gas chromatography analysis in Figure S3),
no other compounds such as methanol or formate are detected.
The detection limits of 1H-NMR for both formate and methanol is
ca. 10 µM (see Figure S4), corresponding to Faradaic
efficiencies below 0.01% and 0.03% for formate and methanol
formation reactions, respectively. This means that under our
experimental conditions, PyH+/Py undergo a full hydrogenation
into piperidine, without reacting with CO2. The concentration of
piperidine after electrolysis (Figure 2B) is estimated to be 0.53
-
electrode would be above 9. In this case Py and HCO3 , rather
than PyH+ and CO2, would be the predominant species near the
Pt electrode. Thus for the following, a constant current density of
-0.1 mA cm-2 was selected as a suitable value for electrolysis
measurements without local pH shift issues. In order to
maximize the concentration of possible products forming under
this relatively small current applied, the following electrolysis
measurements were performed in a small volume of electrolyte
(1 mL).
1
mM using the H-NMR calibration curve (see Figure S5). This
final concentration corresponds to a Faradaic efficiency of ca.
2.1% for the formation of piperidine. The rest of the electric
charge is attributed to the evolution of gaseous hydrogen. Note
the change of the 1H-NMR chemical shift for pyridine before
(Figure 2A) and after (Figure 2B) the electrolysis, due to an
increase of the bulk pH of the electrolyte.[23] Although our setup
is designed to overcome the issue of local pH shift against bulk
pH, it cannot prevent the bulk pH itself to increase when protons
react to evolve gaseous hydrogen if the dissolution of gaseous
CO2 does not buffer the system fast enough (assuming that
every electron injected during the -15 C electrolysis in the 1 mL
electrolysis cell is related to the consumption of one proton, it is
then equivalent to the addition of 0.15 M OH- to the electrolyte
with consequent increase of the bulk pH). As the initial pH of the
electrolyte is 5.3, both PyH+ and Py coexist in solution at the
start of the electrolysis. In order to disentangle which species
can undergo hydrogenation into piperidine, electrolysis
measurements were performed using more extreme pH values.
Figure 1. Characterization of the Pt mesh working electrode in CO2-saturated
1 M KCl + 10 mM Py electrolyte (pH = 5.3). A: stagnant cyclic voltammetry at
50 mV s-1. B: galvanostatic measurements at -0.1 and -1 mA cm-2, with or
without stirring (magnetic stirrer) inside the cell.
Figure 2. Water-suppressed 1H-NMR of CO2-saturated 1 M KCl + 10 mM Py
electrolyte (initial pH = 5.3) before (A) and after (B) -15 C electrolysis on Pt
mesh at -0.1 mA cm-2. Piperidine yield and Faradaic efficiency are estimated
to be 5.2% and 2.1%, respectively.
This article is protected by copyright. All rights reserved.

10.1002/anie.201808122
Angewandte Chemie International Edition
COMMUNICATION
The two next electrolysis runs were performed with only
one of the PyH+/Py acid/base species present in solution. In both
cases the electrolytes were saturated with N2 gas in order to
remove any CO2 interference. For 1 M KCl + 10 mM Py
electrolyte (initial pH = 7.5), no formation of piperidine occurs
after electrolysis (Figure 3A and 3B). The only modification
coming from to the electrolysis is the expected increase of the
electrolyte bulk pH, accounting for the change in pyridine 1H-
NMR chemical shift values.
interesting process for chemical synthesis applications, but it
does not seem suitable to catalyse the CO2 reduction reaction.
For 1 M KCl + 0.2 M HCl + 10 mM PyH+ electrolyte (initial
pH = 0.7), ca. 6 mM of PyH+ is reduced to piperidine after
transferring -5 C, corresponding to a Faradaic efficiency of ca.
72%. No PyH+ at all is left after -15 C electrolysis: PyH+ is
completely hydrogenated into piperidine (Figure 3C, 3D and 3E).
More precisely, in this case PyH+ is completely transformed into
protonated piperidine as the electrolyte is still acidic after
electrolysis due to the excess of initial HCl outweighing the
proton consumption inside the working electrode compartment.
The protonated nitrogen atom of piperidinium accounts for the
1:1:1 triplet (1JN-H = 52 Hz) around 7.9 ppm chemical shift
(Figure 3D and 3E) which is not observed for piperidine in
alkaline solution (Figure 2B). This was further confirmed by
reference spectra of commercial piperidine in acidic and alkaline
solutions in water (see Figure S2) and in heavy water (see
Figure S6).
The full hydrogenation of PyH+ into piperidinium in acidic
conditions, as shown in Figure 3E, falls in line with the recently
reported observations of (i) PyH+ hydrogenation into
“piperidinium-like species” on Pt electrodes upon reductive
conditions using operando reflectance FTIR in strongly acidic
(0.5 M H2SO4) electrolyte[24] and (ii) benzannulated pyridine
hydrogenation in H2O/CH3CN mixture with sufficient equivalent
protons source at Pt and glassy carbon electrodes surfaces.[25]
Considering these previous works[24,25] and the present results, it
is clear that near complete conversion of pyridine-based species
into their hydrogenated counterparts is feasible in mild reductive
conditions as long as the equivalent proton concentration near
the electrode surface is sufficient for the hydrogenation reaction
to occur. In addition, our results reveal that although complete
and selective hydrogenation of pyridinium into piperidinium
occurs in 0.2 M HCl electrolyte (Figure 3), the use of strongly
acidic electrolyte is not necessary. Indeed, piperidinium can also
be formed in CO2 electrolysis conditions (pH ~ 5) if the design of
the experiment accounts for the issue of local pH shift. The
relatively low 5.2% piperidinium conversion in CO2-saturated 1
M KCl electrolyte (Figure 2) is attributed to the progressive
increase of the electrolyte bulk pH during electrolysis, leaving
the stable non-protonated pyridine as the predominant species
of the PyH+/Py equilibrium (pKa ca. 5.2) and thus stopping the
hydrogenation reaction. At the beginning of the electrolysis
measurement, the conditions should be ideal for CO2 reduction
considering the previously proposed “CO2 reduction catalysed
by PyH+” pathway, as CO2 is present around PyH+ undergoing
hydrogenation. However neither formate nor methanol (detection
limits of ca. 10 µM, see Figure S4) were detected after the
electrolysis in CO2-saturated electrolyte although 0.52 mM
piperidine was produced (Figure 2). The electrochemical
hydrogenation of pyridine-based species could be a very
Figure 3. Water-suppressed 1H-NMR of N2-saturated 1 M KCl + 10 mM Py
electrolyte (initial pH = 7.5) before (A) and after (B) -15 C electrolysis on Pt
mesh at -0.1 mA cm-2. Water-suppressed 1H-NMR of N2-saturated 1 M KCl +
0.2 M HCl + 10 mM PyH+ electrolyte (initial pH = 0.7) before (C) and after -5 C
(D) and -15 C (E) electrolysis on Pt mesh at -0.1 mA cm-2.
We demonstrated that with or without the presence of CO2,
PyH+ is completely hydrogenated into piperidine/piperidinium on
This article is protected by copyright. All rights reserved.

10.1002/anie.201808122
Angewandte Chemie International Edition
COMMUNICATION
the Pt surface under mild reductive conditions (i.e. near 0 V vs.
RHE), while Py is electrochemically stable. The fact that we
detect piperidine as the one and only “product” of the CO2
electrolysis (apart from H2) suggests that most of the previous
works likely did not perform electrolysis with PyH+ and CO2 near
The authors are thankful to Prof. Dr. Wolfgang Eisenreich for
fruitful discussions about NMR data and Simon Filser for his
assistance on GC measurements. The project was funded by
the Deutsche Forschungsgemeinschaft (DFG, German
Research Foundation) project KR1189/17. The authors also
acknowledge financial support of TUM.solar in the frame of the
Bavarian Collaborative Research Project Solar technologies go
hybrid (SolTech).
-
the electrode surface but rather with Py and HCO3 , due to the
increase of local pH. Thus, although protonated pyridine was
previously presented as a peculiar and promising catalyst for
CO2 electroreduction on Pt by numerous tentative reaction
mechanisms and DFT investigations, the fact that PyH+ ring
hydrogenation was observed here in CO2 reduction conditions
without any formation of formate or methanol rules out
Keywords: carbon dioxide • platinum • electrolysis • pyridine •
piperidine
protonated pyridine as
electroreduction reaction.
a
suitable catalyst for CO2
[1]
W. Wang, S. Wang, X. Ma, J. Gong, Chem. Soc. Rev. 2011, 40,
3703–3727.
[2]
[3]
H. Takeda, O. Ishitani, Coord. Chem. Rev. 2010, 254, 346–354.
Experimental Section
C. Costentin, M. Robert, J.-M. Savéant, Chem. Soc. Rev. 2013, 42,
2423–2436.
Chemicals: H2SO4 (Suprapur, Merck), KCl (Suprapur, Merck), HCl
(Suprapur, Merck), Pyridine (≥99.7%, Normapur, VWR), Piperidine
(≥99.5%, Aldrich), D2O (99.9%, Aldrich), ultra-pure water (Elga Purelab
Ultra, 18.2 MΩ cm, TOC < 3 ppb), N2 gas (5.0, Linde), CO2 gas (4.5,
Westfalen).
[4]
J. Durst, A. Rudnev, A. Dutta, Y. Fu, J. Herranz, V. Kaliginedi, A.
Kuzume, A. A. Permyakova, Y. Paratcha, P. Broekmann, et al.,
Chim. Int. J. Chem. 2015, 69, 769–776.
Electrochemical measurements were performed using
a
SP-300
[5]
[6]
[7]
G. Seshadri, C. Lin, A. B. Bocarsly, J. Electroanal. Chem. 1994, 372,
(bi)potentiostat (Bio-Logic, France) in a two compartments glass cell,
cleaned by overnight soaking in a mixture of Suprapur H2SO4 + H2O2,
followed by thorough rinsing under UPW. The two compartments of the
cell were separated by a ca. 2 mm thick porous Vycor glass (average
pore size below 5 nm diameter). The platinum mesh working electrode
and platinum plate counter electrode were cleaned by annealing over a
butane flame. For the electrochemical characterization measurements
(Figure 1), the Pt mesh WE was placed along with a K2SO4-saturated
mercury/mercury(I) sulfate reference electrode (MSE) in the “big volume”
compartment (100 mL electrolyte). Measured potential vs. MSE were
translated with respect to the reversible hydrogen electrode (RHE)
potential using the following relation: E(V) vs. MSE = E(V) vs. RHE - 0.65
- 0.059 pH. For the electrolysis experiments (Figures 2 and 3), the Pt
mesh working electrode was placed into the “small volume” compartment
(1 mL electrolyte) for galvanostatic measurements (see the sketch of the
cell in Figure S7). CO2 or N2 gas was bubbled through the electrolyte for
at least 45 min before insertion of the Pt WE. During electrolysis
measurements, gas flow was kept over the electrolyte surface. All
experiments were performed at room temperature.
145–150.
E. B. Cole, P. S. Lakkaraju, D. M. Rampulla, A. J. Morris, E. Abelev,
A. B. Bocarsly, J. Am. Chem. Soc. 2010, 132, 11539–11551.
E. Portenkirchner, C. Enengl, S. Enengl, G. Hinterberger, S.
Schlager, D. Apaydin, H. Neugebauer, G. Knör, N. S. Sariciftci,
ChemElectroChem 2014, 1, 1543–1548.
[8]
S. I. Rybchenko, D. Touhami, J. D. Wadhawan, S. K. Haywood,
ChemSusChem 2016, 9, 1660–1669.
[9]
P. K. Giesbrecht, D. E. Herbert, ACS Energy Lett. 2017, 2, 549–555.
M. Dunwell, Y. Yan, B. Xu, ACS Catal. 2017, 7, 5410–5419.
[10]
[11]
N. R. de Tacconi, W. Chanmanee, B. H. Dennis, F. M. MacDonnell,
D. J. Boston, K. Rajeshwar, Electrochem. Solid-State Lett. 2012, 15,
B5–B8.
1H-NMR spectroscopy was performed at 300 K using an Avance III 500
System (Bruker Biospin, Germany) with an UltraShield 500 MHz magnet
and a SEI 500 S2 probe head (5 mm, inverse 1H/13C with Z-gradient). For
all samples, the 1H-NMR spectra were acquired using the one-
dimensional NOESY sequence “noesygppr1d.comp” with presaturation of
the residual water signal during the relaxation delay (5 s) and the mixing
time (10 ms) using spoil gradients. The acquisition time was 4.1 s.
Spectra were the result of 512 scans, with data collected into 64 k data
points. The analysed solutions (540 µL) were taken directly from the
electrochemical cell after electrolysis and mixed with 60 µL D2O.
[12]
[13]
[14]
[15]
C. H. Lim, A. M. Holder, C. B. Musgrave, J. Am. Chem. Soc. 2013,
135, 142–154.
C. H. Lim, A. M. Holder, J. T. Hynes, C. B. Musgrave, J. Am. Chem.
Soc. 2014, 136, 16081–16095.
C. H. Lim, A. M. Holder, J. T. Hynes, C. B. Musgrave, J. Phys.
Chem. Lett. 2015, 6, 5078–5092.
Acknowledgements
M. Z. Ertem, S. J. Konezny, C. M. Araujo, V. S. Batista, J. Phys.
This article is protected by copyright. All rights reserved.

10.1002/anie.201808122
Angewandte Chemie International Edition
COMMUNICATION
Chem. Lett. 2013, 4, 745–748.
[16]
[17]
[18]
J. A. Keith, E. A. Carter, Chem. Sci. 2013, 4, 1490–1496.
A. Marjolin, J. A. Keith, ACS Catal. 2015, 5, 1123–1130.
H. Dridi, C. Comminges, C. Morais, J. C. Meledje, K. B. Kokoh, C.
Costentin, J. M. Savéant, J. Am. Chem. Soc. 2017, 139, 13922–
13928.
[19]
[20]
[21]
[22]
C. Costentin, J. C. Canales, B. Haddou, J. M. Savéant, J. Am.
Chem. Soc. 2013, 135, 17671–17674.
E. Lebègue, J. Agullo, D. Bélanger, ChemSusChem 2018, 11, 219–
228.
C. Costentin, J. M. Savéant, C. Tard, ACS Energy Lett. 2018, 3,
695–703.
M. Auinger, I. Katsounaros, J. C. Meier, S. O. Klemm, P. U.
Biedermann, A. A. Topalov, M. Rohwerder, K. J. J. Mayrhofer, Phys.
Chem. Chem. Phys. 2011, 13, 16384–16394.
[23]
[24]
[25]
A. D. Gift, S. M. Stewart, P. Kwete Bokashanga, J. Chem. Educ.
2012, 89, 1458–1460.
C. X. Kronawitter, Z. Chen, P. Zhao, X. Yang, B. E. Koel, Catal. Sci.
Technol. 2017, 7, 831–837.
P. K. Giesbrecht, D. B. Nemez, D. E. Herbert, Chem. Commun.
2018, 54, 338–341.
This article is protected by copyright. All rights reserved.

10.1002/anie.201808122
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents
Layout 2:
COMMUNICATION
Dr. Pierre-Yves Olu, Qi Li, Prof. Dr.
Katharina Krischer*
Page No. – Page No.
The True Fate of Pyridinium in the
Reportedly Pyridinium-Catalyzed
Carbon Dioxide Electroreduction on
Platinum
Hey Pyridinium, you are fired! The protonated pyridine cation was believed to be a
promising although peculiar catalyst for CO2 electroreduction reaction on platinum
surfaces. In this work, using an electrolysis setup - carefully designed to overcome
the misleading issue of local pH shift - we put an end to the protonated pyridine
success story.
This article is protected by copyright. All rights reserved.