Electrocatalytic Reduction of CO2 to Acetic Acid by a Molecular Manganese Corrole Complex

Article abstract of DOI:10.1002/anie.202000601

The controlled electrochemical reduction of carbon dioxide to value added chemicals is an important strategy in terms of renewable energy technologies. Therefore, the development of efficient and stable catalysts in an aqueous environment is of great importance. In this context, we focused on synthesizing and studying a molecular Mn^III-corrole complex, which is modified on the three meso-positions with polyethylene glycol moieties for direct and selective production of acetic acid from CO₂. Electrochemical reduction of Mn^III leads to an electroactive Mn^II species, which binds CO₂ and stabilizes the reduced intermediates. This catalyst allows to electrochemically reduce CO₂ to acetic acid in a moderate acidic aqueous medium (pH 6) with a selectivity of 63 percent and a turn over frequency (TOF) of 8.25 h^?1, when immobilized on a carbon paper (CP) electrode. In terms of high selectivity towards acetate, we propose the formation and reduction of an oxalate type intermediate, stabilized at the Mn^III-corrole center.

Full text of DOI:10.1002/anie.202000601

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Electrocatalytic Reduction of CO2 to Acetic Acid by a Molecular
Manganese Corrole Complex
Authors: Wolfgang Schoefberger, Ratnadip De, Sabrina Gonglach,
Shounik Paul, Michael Haas, S. S. Sreejith, Philipp Gerschel,
Ulf-Peter Apfel, Huyen Vuong, Jabor Rabeah, and Soumyajit
Roy
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.202000601
Link to VoR: https://doi.org/10.1002/anie.202000601

10.1002/anie.202000601
Angewandte Chemie International Edition
Electrocatalytic Reduction of CO2 to Acetic Acid by a
Molecular Manganese Corrole Complex
Ratnadip De1, Sabrina Gonglach2, Shounik Paul1, Michael Haas2, S. S. Sreejith1, Philipp
Gerschel3, Ulf-Peter Apfel3,4, Huyen Vuong5, Jabor Rabeah5, Soumyajit Roy*1, and
Wolfgang Schöfberger*2
1Eco-Friendly Applied Materials Laboratory (EFAML), Materials Science Centre, Department of
Chemical Sciences, Mohanpur Campus, Indian Institute of Science Education and Research,
Kolkata,741246, West Bengal.
2Institute of Organic Chemistry, Johannes Kepler University Linz, Altenberger Straße 69, 4040
Linz, Austria.
3Inorganic Chemistry I, Ruhr-Universität Bochum, Universitätsstraße 150, 44801 Bochum,
Germany.
4Fraunhofer UMSICHT, Osterfelder Straße 3, 46047 Oberhausen, Germany.
5Leibniz-Institut für Katalyse e. V., Albert-Einstein-Straße 29a, 18059 Rostock, Germany.
 Soumyajit Roy
s.roy@iiserkol.ac.in
 Wolfgang Schöfberger
wolfgang.schoefberger@jku.at
Abstract
The controlled electrochemical reduction of carbon dioxide to value added chemicals is an
important strategy in terms of renewable energy technologies. Therefore, the development of
efficient and stable catalysts in an aqueous environment is of great importance. In this context, we
focused on synthesizing and studying a molecular Mn(III)-Corrole complex, which is modified on
the three meso-positions with polyethylene glycol moieties for direct production of acetic acid
from CO2 in a selective manner. Electrochemical reduction of Mn(III) to a Mn(II)-electroactive
species, which significantly binds CO2 and stabilizes the reduced intermediates. This catalyst
allows to electrochemically reduce CO2 to acetic acid in a moderate acidic aqueous medium (pH 6)
with a selectivity of 63 % and a turn over frequency (TOF) of 8.25 h-1, when immobilized on a
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
Carbon Paper (CP) electrode. In terms of high selectivity towards acetate we propose the formation
and reduction of an oxalate type intermediate, stabilized at the Mn(III)-Corrole center.
Introduction
The growing consumption of fuels led to a drastic increase of the CO2 level in the atmosphere
(33.1 Gt of CO2 in 2018[1]). Recent trend in research focuses primarily on remediation of the CO2
emission from non-renewable fossil fuels as well as providing a clean, environmentally benign
and cheaper alternative which can solve the current energy crisis.[2] Getting inspired by nature,
scientists found a way to mitigate this problem mainly through the use of catalysts which can
mimic the role of photosystem in green plants.[3] In the past few decades huge efforts have been
made to solve this problem and several classes of catalysts have been explored to reduce CO2 to
valuable chemicals like e.g. acetic acid. Many metallic catalysts for CO2 reduction reaction were
investigated, like Cu[4], which shows a higher selectivity for the key C-C bond formation during
the CO2 reduction, where acetate was obtained with faradaic efficiencies (FE) varying from 2.5 %
to 56 %[5]. Introduction of Ag into the system, like Cu-Ag bimetallic nanoparticles[6], marked the
enhanced formation of acetate with a FE of 21.2 %. An improved FE of 60.9 % could be achieved
using Fe(III)oxyhydroxide on N-doped carbon[7], while N-doped nanodiamond on Si rod arrays[8]
showed a higher FE of 77.6 %. Table S1 summarizes above stated electrocatalysts for reducing
CO2 to acetate. Aside from that, well-designed molecular metal complexes were developed to
reduce CO2. In this regard some molecular Mn-complexes like Mn(bipyridine)-pyrene,[9] N-
heterocyclic carbene Mn(I) complex[10] and other similar complexes immobilized over supports
like multi-walled-carbon-nanotubes (MWCNT)[11] or Nafion membrane[12] have been reported to
reduce CO2 to CO. Recently, we have observed that a cobalt(III) triphenylphosphine corrole
complex exhibits potential dependent CO2 reduction at pH 6 to produce methanol, ethanol, and
low amounts of acetic acid (FEmax. 13% at -0.95 V vs. RHE)[13]. In this study we focus on the
design of manganese corroles as potent electrocatalysts for CO2 reduction. Manganese(III)
corroles are square-planar complexes and initially possess no axial ligand coordination compared
to the Co(III)(-L) corroles (e.g. L = triphenylphosphine or pyridine), which renders the complex
freely accessible towards axial coordination by nucleophiles (e.g. H2O or OH-)[14–16], therefore, we
have immobilized a novel manganese corrole complex (“Mn-Corrole”) on a Carbon
Paperelectrode (Mn-Cor-CP) to conduct the heterogeneous electroreduction of CO2. We observed
acetate as the main product in a moderate acidic aqueous medium (pH 6, phosphate buffer), at a
potential of -1.25 V vs Ag/AgCl (1 M KCl) with a FE as high as 63 % and a TOF of 8.25 h-1.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
Results and discussion
Synthesis, fabrication and characterization of the electrode material
Mn-Corrole is a molecular catalyst (Figure 1A), which was synthesized over four steps (chemical
syntheses and characterization are stated in the Supplement, Figure S1-S8). The free base form is
the 5,10,15–tris(2,3,5,6-tetrafluoro-4-(MeO-PEG(7))thiophenyl)corrole which complexes with
Mn(III) to form the four-coordinated Mn(TpFPC)(-S-PEG(7)-OMe)3 (“Mn-Corrole”). Owing to
its solubility in acetonitrile and insolubility in water heterogenization over different platforms like
Carbon Paper (CP), Indium Doped Tin Oxide (ITO) and Glassy Carbon (GC) leads to facile
fabrication of electrodes via drop casting, which are stable in aqueous solutions. This kind of
functionalization of the Mn-Corrole with S-PEG(7)-OMe moieties leads to physisorption over a
conducting surface via - interactions and equal distribution of the macrocycle.
XPS analysis was conducted with the catalyst immobilized Carbon Paper to assess
the properties of the Mn-Corrole showing the characteristic binding energy regions corresponding
to Mn2p, F1s, C1s, N1s, and S2p (Figure S9A). The Mn2p scan showed two distinct peaks at
642 eV and 653 eV attributed to 2p3/2 and 2p respectively. The intensity ratio and the position of
½
these two peaks matches well with the Mn(III) center of previously reported Mn-Corrole[17]. The
position of the N1s band at 398.3 eV and the main peak for F1s confirms the integrity of the metal
bound corrole species over the Carbon Paper electrode[17,18] (Figure S9A).
The XPS measurements were also performed after the electrolysis where the peaks for Mn2p, C1s,
N1s and S2p were positioned at the same energy regions depicting the stability of the catalyst even
after the CO2 reduction reaction (Figure S9B). Absence of any peak at 401 eV omits the possibility
of demetallation and formation of free base corrole.[18]
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
Figure 1: (A) Chemical structure of the Mn-Corrole (B) cyclic voltammetry of Mn-Corrole dissolved in
ACN under Ar (black), CO2 (red), in ACN/H2O (4:1) under Ar (blue) and in ACN/H2O (4:1) under CO2
(green). (C) Heterogeneous electrocatalysis of 0.5 mg cm-2 Mn-corrole on carbon-paper electrode under Ar
(red) and CO2 (black) at pH 6.0 (Ag/AgCl/1 M KCl, Pt, 100 mV s−1). (D) Controlled potential electrolysis
of Mn-Corrole immobilized on a Carbon Paper electrode in 0.1 M phosphate buffer (pH 6) saturated with
CO2 at -1.25 V vs Ag/AgCl for 5 h. All homogeneous cyclic voltammetry measurements were performed
with 0.1 M TBAPF6 as supporting electrolyte using glassy carbon as working, platinum wire as counter and
a non-aqueous pseudo-Ag/AgCl reference electrode at a scan rate of 100 mV s-1.
Electrochemical characterization of the Mn-Corrole
The electrochemical properties of 0.8 mM Mn-Corrole in acetonitrile were investigated via cyclic
voltammetry measurements using glassy carbon as working electrode with 0.1 M TBAPF6 as
supporting electrolyte at a scan rate of 100 mV s-1. In the cathodic part, two one electron reductions
at E1/2 = -0.94 V vs NHE (Mn(III)/Mn(II)) and Epc = -1.60 V vs NHE are shown in Figure 1B and
Figure S10, where the first redox couple is reversible[14,19], whereas the second one is irreversible.
Corresponding electro paramagnetic resonance (EPR) measurements, suggest the formation of a
Mn(II) radical anion at -1.60 V (Figure S11). On sweeping in the anodic region there is a metal
centered and ligand centered oxidation reaction occurring. The first metal centered oxidation at
E1/2 = 0.68 V and E1/2 = 0.96 V vs NHE of Mn(III) to Mn(IV) is split into two waves, where the
current of the first redox couple is significantly smaller than of the second one. Similar redox
behavior for other triaryl-corroles was found by the group of Kadish and co-workers[14,19]. This
split redox-waves result, since two different Mn(IV) species can be formed, a ion pair with the
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
anion of the electrolyte salt (Mn(IV)PF6) and the Mn(IV)+ cation species[14,19]. The second
oxidation at Ep = 1.45 V vs NHE is ring-centered where Mn(IV) is oxidized to a Mn(IV) π radical
cation[14]. We further studied the influence of water under CO2 in the electrochemical system. In
Figure 1B, the comparison of Mn-Corrole dissolved in ACN (red curve) and in ACN/H2O = 4/1
(green curve), is shown. The second reduction wave is increasing upon CO2 saturation at -1.43 V
vs NHE, which clearly illustrates the influence of protons in the CO2 reduction reaction through
proton coupled electron transfer (PCET). In the cyclic voltammogram (Figure 1B, green line)
‘‘curve crossing’’ occurs where the current on the return scan is higher than the current on the
forward scan. Curve crossing stems from the increasing concentration of active catalyst (in situ
generation of the active catalyst) – and so higher catalytic current – as the CV progresses.
Heterogeneous Electrochemistry with Mn-Corrole towards CO2 reduction
Cyclic voltammetry measurements under heterogenized conditions were conducted, to study its
efficacy for CO2 reduction reaction in an aqueous electrolyte solution. Mn-Corrole was
physisorbed on the Carbon Paper with an effective loading of 0.5 mg cm-2. Therefore, a three
electrode electrochemical cell with Mn-Corrole-CP as working, Ag/AgCl/1 M KCl as reference
and platinum wire as counter electrode was chosen. On performing cyclic voltammetry in 0.1 M
pH 6 phosphate buffer, it was found that the redox behavior of the catalyst remains almost the
same like under homogeneous conditions. Cyclic voltammetry measurements were performed
under Ar (red) and CO2 (black) with a pH 6 saturated electrolyte solution (Figure 1C). A steep
increase in current density upon CO2 saturation occurred, and a broad wave appeared in the cyclic
voltammogram starting at -1.1 V vs Ag/AgCl. This suggests the CO2 reduction is catalyzed with
the immobilized Mn-Corrole on the carbon paper. Further we performed controlled potential
electrolysis (CPE) at varying potentials, starting from -1.1 to -1.4 V vs Ag/AgCl over 5 h, with
resulting current densities between 0.35 to 1.1 mA cm-2 (Figure S12 and S13). The product
analysis was conducted with the resulting proton coupled nuclear magnetic resonance
spectroscopy (1H-NMR) and gas chromatography mass-spectroscopy (GC-MS) measurements
(Table 1, Figure S14 and S15). Methanol and acetate are the only liquid products, CO and H2 are
the detected gaseous compounds. CPE over 5 h with Mn-Corrole-CP, formed acetate as main
product with a FE of 63 % at -1.25 V vs Ag/AgCl and a TOF of 8.25 h-1 (Figure 1D and Figure
S12-S15). Additionally, the FE was measured at different intervals of time, where Figure 1D shows
that the FE remains quite constant over 5 h CPE. Methanol exhibited faradaic efficiencies between
9 and 23 %, with decreasing product formation towards lower potentials (Table 1, Figure S14A).
At more negative potentials the formation of C2 product acetate is favored with a substantial
increase in its FE (40 to 63 %, Table 1, Figure S14A). CO production was found only at higher
potentials, -1.1 V and -1.2 V (Table 1).
Control measurements were performed with an unmodified CP electrode, a CP electrode with the
corresponding free-base corrole and a CP electrode with PEG7-OMe-SH (Figure S16A-C). No
product formation (methanol or acetate) was observed, which proves CO2 reduction ability of the
Mn-Corrole. XPS analysis of the modified CP before and after the electrocatalysis reaction
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
clarifies that the Mn-Corrole is stable through the CO2 reduction process (Figure S9). Recorded
UV-vis spectra after electrolysis of the dissolved catalyst in ACN verified that the macrocycle is
still intact (Figure S17A) as well as measured heterogeneous cyclic voltammogram of the Mn-
Corrole modified CP after CPE depicted no change in the electrochemical behavior of the
fabricated electrode after catalysis (Figure S17B).
Table 1: Comparative data for the Mn-Corrole-CP working electrode during a 5 h electrolysis as a function
of the potential in 0.1 M phosphate buffer (pH 6) saturated with CO2. Standard deviations were calculated
after three electrochemical measurements followed by two product analysis measurements.
Potential Potential
FE % of each CO2 reduced products at different Total Charge passed
vs
Ag/AgCl
(V)
vs RHE
potential with Mn-Cor-CP electrode after 5 h of
electrolysis
FE%
(Coulomb)
(V)
Acetate
Methanol
CO
H2
FE% SD FE% SD FE% SD FE% SD
average
SD
-1.1
-1.2
-1.25
-1.3
-1.4
-0.524
-0.624
-0.674
-0.724
-0.824
40
55
63
60
61
4.3
4
3.75
3.55
4
23
20
16
19
9
3
3
3.5
2.5
3
31
20
-
-
-
5
4
-
-
-
-
-
94
95
97
98
95
6.3
9
14.4
17.1
19.8
0.2
0.25
0.3
0.5
0.8
18
19
25
4.5
5
6
Partial current
Potential
vs RHE
(V)
-0.524
-0.624
-0.674
-0.724
-0.824
Overpotential Overpotential
density (mA cm-2)
TONAcOH TONMeOH
ηAcOH (mV)
ηMeOH (mV)
jAcOH
0.140
0.275
0.504
0.570
0.671
jMeOH
0.080
0.100
0.128
0.180
0.099
264
364
414
464
564
527
627
677
727
827
11.45
22.50
41.25
46.65
54.90
6.58
8.18
10.47
14.76
8.10
Origin of the products
To verify the source of protons and carbon in the products, we performed isotopic labelling
experiments by using D2O/H2O confirming the source of proton and 13CO2 reduction
measurements for determining the carbon source. Deuterium distribution during the electrolysis
was investigated via CPE at -1.25 V vs Ag/AgCl in D2O/H2O (1/5) at pH 6. The resulting
GC-MS spectrum, confirmed the presence of deuterium enriched acetic acid (e.g. m/z = 64), which
was confirmed from the appearance of partial and fully proton-deuterium exchanged fragments
(m/z = 43-46) and shows a shift of m/z = 1 for the respective molecular ion peak and fragments
(Figure 2A and 2B). From the resulting peaks, the one with m/z = 64 can be assigned to the tetra-
deuterated (CD3COOD) acetic acid (Figure 2B, S18). Among the molecular ion peak, there are
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
also other characteristic fragments of CD3COOD occurring, like m/z = 46 for ·COOD+ and
m/z = 44 for ·CH2DCO+ (Figure 2B, S18). Further deuterium labelling was also observed for
methanol (m/z = 36, Figure S19). This D2O/H2O experiments verify that protons are incorporated
via proton coupled electron transfer (PCET) in the CO2 reduction reaction and confirms water is
the proton source in the formed products. Moreover, CPE measurements with 13CO2 were
performed to confirm the carbon source. The resulting GC-MS spectrum for acetate showed a shift
13
of m/z = 2 of the molecular ion peak (CH3COOH+, m/z = 60 to 62) due to C incorporation in
both carbon atoms of acetic acid (Figure 2C). This trend is also reflected for different fragments,
e.g. where COOH+ m/z = 45 shifts to m/z = 46 due to 13C substitution. The fragment obtained at
12
m/z = 43 during the reduction of CO2 corresponds to ·CH3CO+ and shifted to m/z = 45 upon
replacement with 13CO2 (Figure 2C). In the case of methanol, the molecular ion peak indicates a
shift of m/z = 1 upon 13C enrichment (m/z = 32 to 33) indicating the formation of 13CH3OH (Figure
2D and 2E). This data were further substantiated by performing 1H-decoupled carbon NMR (13C-
NMR) and 1H-NMR of the 13CO2 reduced products (Figure 2F). The obtained 13C-NMR spectrum
reveals a quartet at 20 ppm (1JC,H = 129.7 Hz) and a doublet at 176.2 ppm (JC,C = 55.3 Hz), which
verifies the presence of fully 13C-labeled acetic acid. Methanol appears at 49.3 ppm. The 1H-NMR
spectrum illustrated in Figure S14 indicates resonances at 3.2 ppm (13CH3OH, d, JC,H = 134 Hz)
13
and at 1.8 ppm (13CH3 COO-, dd, 1JC,H = 163 Hz, 2JC,H = 6.5 Hz). These GC-MS, 1H-NMR and
13C-NMR experiments confirm that the only carbon source of methanol and acetic acid is CO2.
Figure 2: (A-E) GC-MS-spectra of the electrolyte solution after 5 h CPE at -1.25 V vs Ag/AgCl, pH 6,
13
12
12CO2 and CO2, respectively (A) of 12CH3 COOH and (B) deuterium substituted acetic acid, D2O/H2O
(1/5) (C) 13CH3 COOH(D) 12CH3OH (E)13CH3OH. (F)1H-coupled-13C-NMR-spectrum in 13CO2 saturated
phosphate buffer (pH 6) after 5 h CPE at -1.25 V vs Ag/AgCl under 13CO2 saturation.
13
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
From these experiments, it was confirmed that the reduced form of Mn-Corrole can catalytically
activate CO2. Substantial increase in current density under CO2 atmosphere could be attributed to
the adsorption and reduction of reactive CO2 species. The role of water was validated like
mentioned above via a homogenous cyclic voltammetry measurement in a CO2 saturated
acetonitrile solution with 20 % water content (Figure 1B, green curve). There was a characteristic
increase in the second Mn reduction current observed, which suggests the formation of the
catalytically active species with subsequent CO2 reduction reaction. This signifies the need for an
intervening pathway of PCET in the presence of a proton donor to facilitate CO2 electroreduction.
Plausible mechanistic pathway for CO2 reduction
To have a deeper insight into the electron transfer phenomena obtained from the cyclic
voltammetry investigations we performed chemical reduction of the Mn(III)-Corrole in
acetonitrile as solvent. To demonstrate the charge transfer occurring between Mn-Corrole and
CO2, we have reduced Mn(III)-Corrole typically by the addition of KC8 in dry acetonitrile solution.
The electronic absorbtion spectrum (EAS) encounters a prominent change to a single Soret-band
maximum at 470 nm with Q-bands at 561 nm, 610 nm, and 663 nm (Figure 3, red dashed line).
After the addition of water, the Soret band exhibits splitting with absorption maxima at 406 nm
and 471 nm, which indicates the formation of MnCor(OH)2 species (Figure 3, blue dashed line).
This species was first investigated by the group of Kadish and co-workers[14,15]. After exposing the
solution to CO2, we observe a distinct change in the EAS spectrum (Figure 3, green solid line).
The obtained UV-vis spectrum shows striking similarity to the one observed for Mn(III)-Corrole
(Figure 3, black line). The position and intensity of the band at 475 nm is slightly altered compared
to the initial spectrum for the Mn-Corrole, which is attributed to the axial ligation of CO2 at the
Mn site. During the subsequent PCET the formation of Mn(III)-CO2H is facilitated. At this step,
the reaction can take different routes one of which is another PCET on carbon site to form
HCOOH, which can eventually lead to formation of other C1 products like CH3OH (Figure S20).
On the other hand, second protonation on the hydroxyl oxygen followed by dehydration generates
CO. Another possibility is the C-C bond formation to generate an oxalate type intermediate. From
here successive reduction of the oxalate leads to the formation of acetate as the final product
(Figure S21).
Furthermore, the electrochemical redox behavior of Mn(III)-Corrole was characterized under
argon and CO2 atmosphere via spectroelectrochemical UV-vis (SEC-UV-vis) measurements in an
optical transmission thin layer electrode cell (OTTLE cell, see SI experimental section). Therefore,
a light transparent platinum mini-grid acts as working, as counter and Ag-microwire as pseudo-
reference electrode. Measurements were pretreated for 1 min at constant potential, afterwards the
UV-vis recording was started with additional electrolysis of 1 min at the same potential. The first
one electron reduction at -1.0 V vs NHE in ACN with 2 % water under Ar ascribes the
electrochemical reduction of initial Mn(III) to Mn(II) with subsequent chemical reaction to
[Mn(III)(OH)2]2- [14,15] (Scheme 1 and Figure 3B, red line).
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
Figure 3: (A) EAS of Mn-Corrole in acetonitrile under argon (black solid line), the chemically reduced
form in presence of KC8 (red dotted line), after the addition of water (blue, dashed line), and subsequent
dosage of CO2 (green solid line). (B)-(D) Potential dependent SEC-UV-vis of 0.7 mM Mn-Corrole in
acetonitrile with 2 % water and 0.2 M TBAPF6 as electrolyte after 2 min CPE (B) under Argon (C) under
CO2, and (D) comparison of UV-vis spectra observed at -1.3 V vs NHE under Argon (black) and CO2
dosage (red). SEC-UV-vis measurements were recorded with a light transparent platinum mini-grid as
working, as counter and an Ag-microwire as pseudo-reference electrode.
Scheme 1: Redox reactions of Mn-Corrole in ACN with 2 % water[14,15]
.
9
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
This electrogenerated species shows a splitted Soret band at 428 nm, a significant intense band at
470 nm and Q-bands at 563, 597 and 637 nm, which is similar to the formed bis-hydroxy species
after chemical reduction and water addition. When moving on to more negative potentials like
-1.3 V vs NHE, [Mn(III)(OH)2]2- is reduced to [Mn(II)(OH)2]3-, where one sharp visible band at
471 nm, bathochromic shifted Q-bands at 567 nm, 612 nm and one intense Q-band at 660 nm was
formed (Figure 3B green and orange line)[14,15]. On performing measurements in ACN with 2 %
water under CO2 at various negative potentials (-0.2 to -1.3 V), a decrease of the initial Soret-band
at 417 nm with a consequently increase of a hypsochromic shifted visible band from 470 nm to
461 nm and an intense Q-band at 643 nm were observed (Figure 3C). At the first reduction
potential (-1.0 V) of Mn(III)-Corrole, the visible band at 467 nm (Figure 3C, green line) was more
intense compared to under Ar (470 nm), which can be ascribed to the additional axial ligation of
CO2 to the Mn-site. On shifting the applied potential to -1.3 V vs NHE, one bathochromic shifted
sharp visible band was formed at 461 nm (Figure 3D, red line) compared to 471 nm under Ar
(Figure 3D, black line). This suggests, that [Mn(II)(OH)2]- is the possible catalytic active species,
where CO2 can coordinate to the metal site to be transformed via PCET processes into respective
reaction intermediates and products.
Figure 4: SEC- FTIR of 15 mM Mn-Corrole in ACN/H2O (4/1) and 0.1 M TBAPF6 under CO2 at various
potentials from -0.6 to -1.8 V vs NHE. Glassy carbon was used as working, platinium wire as counter and
silver wire as pseudo-reference electrode. (background = bg)
For further insight in possible reaction intermediates we performed spectroelectrochemical
attenuated total reflection Fourier-transform infrared spectroscopy measurements (SEC-ATR-
10
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
FTIR) at various potentials from -0.6 to -1.8 V vs NHE. Figure 4 illustrates increasing IR-bands
at 3440-3420 cm-1 (O-H vibrational stretching), 1654 cm-1 (C=O vibrational stretching) and 1420-
1400 cm-1 (C-H and C-OH stretching), which suggests the stepwise formation of an alcohol or
acid. The respective IR-band for CO2 at 2360-2340 cm-1 is decreasing towards lower potentials,
verifying that CO2 is converted. Further decreasing IR-bands at 1738 cm-1 (C=O vibrational
stretching), 1217 cm-1 (vibrational stretching C-OH) and 1116 cm-1 (vibrational stretching C-OH)
suggest the formation and consumption of the proposed Mn-C(H)OOH intermediate to further
products, like methanol or acetic acid.
Mechanistic investigations show that metal and ligand centered activity of porphyrins[20] or
phthalocyanines[21] are a determining factor for the selectivity of CO and HCOO-/HCOOH
formation. On similar trails, Mn-Corrole acting as metal active corrole species possesses the ability
to stabilize a metal bound carboxyl group. Due to this, CO2 reduction reaction with Mn-Corrole
preferably results in the formation of the 2e- reduced CO, 6e- reduced methanol or via 8e- reduced
acetate (SI p.29, Equation S1-S14). Presence of protons (pH 6) lowers the activation energy for
the first PCET towards the O-atom[22] of CO2 resulting in the formation of a carboxyhydroxyl
intermediate stabilized on Mn(III)-Corrole site. This stabilization results not only in achieving a
low energy cost but also plays decisive role in the selectivity of the products. From this step another
PCET on the C-atom leads to the formation of HCOOH type intermediate which on subsequent
reduction forms methanol. On the other hand a PCET over the O-atom of the carboxyhydroxyl
intermediate resulted in the formation of CO. Higher selectivity for acetate achieved in this work,
originated from the Lewis acidity of Mn(III)center, which has a higher tendency to bind with the
Lewis basic O-site of the carboxyl group, hence facilitating the C-C dimerization leading to an
oxalate type intermediate[8,23]. This proposition was further substantiated by DFT calculations
(Figure S22-S25), which showed a potential stabilization of -1.04 eV upon C-C dimerization.
From this step, consecutive reduction results in the formation of acetate (SI p.29, Equation S1-
S14). During CO2 electrolysis over Mn-Corrole-CP electrodes no trace of oxalate or formate was
detected. To get a better idea of the mechanistic pathways we performed electrochemical reduction
of possible reaction intermediates. The reduction of 10 mM formic acid by Mn-Corrole under
similar electrocatalytic conditions generates methanol as the only liquid product (Figure S20) with
no gaseous components. From these observations we come to the conclusion that the reduction of
formic acid only yields methanol, thus it serves as an important 2e- reduced intermediate in CO2
reduction to methanol. When reduction of 10 mM oxalate in pH 6 phosphate buffer was performed
with Mn-Corrole-CP as cathode, acetate (Figure S21) was obtained exclusively which supports
our proposition for the formation of an oxalic acid type intermediate which follows multiple
reduction steps on the pathway of acetate formation (SI p.29, Equation S1-S14).
11
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
Conclusion
To conclude, in this work we have explored a molecular Manganese-Corrole-CP catalyst for the
electrochemical CO2 reduction to achieve acetate at -1.25 V vs Ag/AgCl over 5 h electrolysis time
with a FE of 63 %. Herein the S-PEG(7)OMe modified A3-Corrole is designed in such a way that
it exhibited the prerequisite 8e- reduction of CO2 in a selective manner. We suggest a mechanistic
pathway which involves the formation and reduction of a Mn(III) bound oxalate intermediate to
acetate which was further substantiated by reduction studies of oxalic acid and DFT calculations.
This work opens up opportunities for direct production of acetic acid from CO2 with high
selectivity, which currently involves multi step process via production of methanol and its
carbonylation starting from the synthesis of syngas. In summary we presented an efficient
approach for direct acetic acid formation via electrocatalytic CO2 reduction using a Mn-Corrole
based molecular catalyst. Further studies on this kind of catalysts will help in designing more
superior molecular catalysts for the purpose of storing energy through chemical bond formation in
a selective manner.
Reference
[1] International Energy Agency (Ed.) Global Energy & CO2 Status Report 2019,
https://www.iea.org/reports/global-energy-and-co2-status-report-2019, 2019.
[2] a) M. Aresta, A. Dibenedetto, A. Angelini, Chem. Rev. 2014, 114, 1709; b) T. Sakakura, J.-
C. Choi, H. Yasuda, Chem. Rev. 2007, 107, 2365; c) B. Khezri, A. C. Fisher, M. Pumera, J.
Mater. Chem. A 2017, 5, 8230.
[3] a) L. Sun, H. Berglund, R. Davydov, T. Norrby, L. Hammarström, P. Korall, A. Börje, C.
Philouze, K. Berg, A. Tran et al., J. Am. Chem. Soc. 1997, 119, 6996; b) C. Liu, B. C. Colón,
M. Ziesack, P. A. Silver, D. G. Nocera, Science 2016, 352, 1210.
[4] a) T. Kim, A. Kargar, Y. Luo, R. Mohammed, E. Martinez-Loran, A. Ganapathi, P. Shah, D.
P. Fenning, ACS Appl. Energy Mater. 2018, 1, 1965; b) K. P. Kuhl, E. R. Cave, D. N.
Abram, T. F. Jaramillo, Energy Environ. Sci. 2012, 5, 7050.
[5] C. Genovese, C. Ampelli, S. Perathoner, G. Centi, Green Chem. 2017, 19, 2406.
[6] Y. Wang, D. Wang, C. J. Dares, S. L. Marquard, M. V. Sheridan, T. J. Meyer, Proc. Natl.
Acad. Sci. USA 2018, 115, 278.
[7] C. Genovese, M. E. Schuster, E. K. Gibson, D. Gianolio, V. Posligua, R. Grau-Crespo, G.
Cibin, P. P. Wells, D. Garai, V. Solokha et al., Nat. commun. 2018, 9, 935.
[8] Y. Liu, S. Chen, X. Quan, H. Yu, J. Am. Chem. Soc. 2015, 137, 11631.
12
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
[9] B. Reuillard, K. H. Ly, T. E. Rosser, M. F. Kuehnel, I. Zebger, E. Reisner, J. Am. Chem.
Soc. 2017, 139, 14425.
[10] F. Franco, M. F. Pinto, B. Royo, J. Lloret-Fillol, Angew. Chem. Int. Ed. 2018, 57, 4603.
[11] a) S. Sato, K. Saita, K. Sekizawa, S. Maeda, T. Morikawa, ACS Catal. 2018, 8, 4452; b) J. J.
Walsh, C. L. Smith, G. Neri, G. F. S. Whitehead, C. M. Robertson, A. J. Cowan, Faraday
Discuss. 2015, 183, 147.
[12] J. J. Walsh, G. Neri, C. L. Smith, A. J. Cowan, Chem. commun. 2014, 50, 12698.
[13] S. Gonglach, S. Paul, M. Haas, F. Pillwein, S. S. Sreejith, S. Barman, R. De, S. Müllegger,
P. Gerschel, U.-P. Apfel et al., H. Coskun, A. Aljabour, P. Stadler, W. Schöfberger, S. Roy,
Nat. commun. 2019, 10, 3864.
[14] J. Shen, M. El Ojaimi, M. Chkounda, C. P. Gros, J.-M. Barbe, J. Shao, R. Guilard, K. M.
Kadish, Inorg. Chem. 2008, 47, 7717.
[15] P. Chen, M. E. Ojaimi, C. P. Gros, J.-M. Barbe, R. Guilard, J. Shen, K. M. Kadish, J.
Porphyrins Phthalocyanines 2011, 15, 188.
[16] M. Tiffner, S. Gonglach, M. Haas, W. Schöfberger, M. Waser, Chem. Asian J. 2017, 12,
1048.
[17] M. Paszkiewicz, T. Biktagirov, H. Aldahhak, F. Allegretti, E. Rauls, W. Schöfberger, W. G.
Schmidt, J. V. Barth, U. Gerstmann, F. Klappenberger, J. Phys. Chem. Lett. 2018, 9, 6412.
[18] H. Marbach, Acc. Chem. Res. 2015, 48, 2649.
[19] B. Gao, Z. Ou, X. Chen, S. Huang, B. Li, Y. Fang, K. M. Kadish, J. Porphyrins
Phthalocyanines 2014, 18, 1131.
[20] A. J. Göttle, M. T. M. Koper, J. Am. Chem. Soc. 2018, 140, 4826.
[21] a) M. Wang, K. Torbensen, D. Salvatore, S. Ren, D. Joulié, F. Dumoulin, D. Mendoza, B.
Lassalle-Kaiser, U. Işci, C. P. Berlinguette and M. Robert, Nat. commun. 2019, 10, 3602; b)
E. Boutin, M. Wang, J. C. Lin, M. Mesnage, D. Mendoza, B. Lassalle-Kaiser, C. Hahn, T. F.
Jaramillo, M. Robert, Angew. Chem. Int. Ed. 2019, 58, 16172.
[22] A. Seifitokaldani, C. M. Gabardo, T. Burdyny, C.-T. Dinh, J. P. Edwards, M. G. Kibria, O.
S. Bushuyev, S. O. Kelley, D. Sinton, E. H. Sargent, J. Am. Chem. Soc. 2018, 140, 3833.
[23] S. Sun, M. Watanabe, J. Wu, Q. An, T. Ishihara, J. Am. Chem. Soc. 2018, 140, 6474.
13
This article is protected by copyright. All rights reserved.

10.1002/anie.202000601
Angewandte Chemie International Edition
Acknowledgements
W.S acknowledges the financial support of the Austrian Science Fund (FWF-P28167-N34 and
FWF-P32045-NBL). S.R. acknowledges the start-up and FIRE and PRIS grants from IISER
Kolkata, India. S.R. further thanks CCNU, P.R. China and NSFC (B050704), P.R. China and ‘111
project’ for financial support. R.D. acknowledges IISER-K for fellowship, S.P. acknowledges
UGC for fellowships, S.S.S. thanks SERB (PDF/2017/000676).
The NMR spectrometers were acquired in collaboration with the University of South Bohemia
(CZ) with financial support from the European Union through the EFRE INTERREG IV ETC-
AT-CZ program (project M00146, “RERI-uasb”). W.S and S.G thanks Dr. Markus Himmelsbach
and DI Thomas Bögl for the ESI HRMS measurements. P.G. thanks the Fonds of the Chemical
Industry for a PhD fellowship. U.-P.A. is grateful for support by the Deutsche
Forschungsgemeinschaft (Emmy Noether grant, AP242/2-1; AP242/6-1) and the Fraunhofer
Internal Programs under Grant no. Attract 097-602175. U.-P.A. and P.G. were funded by the
Deutsche Forschungsgemeinschaft (DFG, German Research Foundation—Germany's Excellence
Strategy, EXC-2033—Project number 390677874). We would like to thank Prof. Dr. Angelika
Brückner for fruitful discussions and interpretation of the ESR spectra.
Author information
These authors contributed equally: Sabrina Gonglach, Shounik Paul, Ratnadip De.
14
This article is protected by copyright. All rights reserved.