Investigation of Cyclam Based Re-Complexes as Potential Electrocatalysts for the CO2 Reduction Reaction

Article abstract of DOI:10.1002/zaac.202000450

Among the various homogenous electrocatalysts, especially Re(bpy)(CO)₃Cl and [Ni(cyclam)]²⁺ were shown to be highly efficient for the selective conversion of CO₂ to CO at moderate potentials. However, a purposeful combination of a Re^I tricarbonyl unit with a cyclam ligand hitherto received no attention. Herein, we report on a series of cyclam based Re complexes comprising the original {N₄} as well as heteroatom-altered ligand frameworks, describe their synthesis, reveal their coordination behavior and furthermore investigate their performance towards the electrochemical CO₂ reduction.

Full text of DOI:10.1002/zaac.202000450

Journal of Inorganic and General Chemistry
DOI: 10.1002/zaac.202000450
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
Investigation of Cyclam Based Re-Complexes as Potential
Electrocatalysts for the CO2 Reduction Reaction
Philipp Gerschel,[a] Anna L. Cordes,[a] Sarah Bimmermann,[a] Daniel Siegmund,[b]
Nils Metzler-Nolte,[a] and Ulf-Peter Apfel*[a, b]
Dedicated to Professor Dr. Peter Klüfers on the Occasion of his 70th Birthday
Among the various homogenous electrocatalysts, especially
Re(bpy)(CO)3Cl and [Ni(cyclam)]2+ were shown to be highly
efficient for the selective conversion of CO2 to CO at moderate
potentials. However,
tricarbonyl unit with a cyclam ligand hitherto received no
attention. Herein, we report on a series of cyclam based Re
complexes comprising the original {N4} as well as heteroatom-
altered ligand frameworks, describe their synthesis, reveal their
coordination behavior and furthermore investigate their per-
formance towards the electrochemical CO2 reduction.
a purposeful combination of a
ReI
Introduction
demands, suitable strategies are required to achieve a selective
and low-energy conversion.[1,2,10,11]
The earths major greenhouse gas, CO2, contributes massively to
the current global warming with a global radiative forcing of at
least 2.076 WmÀ 2, which accounts to 66% of the total
anthropogenic radiative forcing.[1–5] Since the ever-increasing
global warming triggers an unstoppable climate change, which
in turn affects the environment and its ecological systems
adversely, a decrease of further CO2 emissions and a reversal of
the unnaturally high concentrations of CO2 already accumulated
in the atmosphere is of crucial importance.[1,5–7] While the
avoidance of CO2 emissions can be achieved by increasing the
efficiency of existing energy conversion processes based on
fossil fuels and by establishing sustainable energy resources,
the reduction of the existing high CO2 concentrations in the
atmosphere can be achieved through carbon capture and
utilization (CCU), which deals with the chemical conversion and
utilization of CO2.[1,6,8,9] Since the activation and conversion of
CO2 into energy-rich products is associated with high energy
Among the variety of CO2 reduction approaches, electro-
catalysis was suggested to be the most promising and
sustainable, as it can be run directly with renewable
power.[2,12–14] However, finding suitable electrocatalysts that
operate at low overpotentials and with a high selectivity for the
CO2 reduction is challenging.[15–17] Although heterogeneous
electrocatalysts are preferred for industrial applications due to
their easy handling, high robustness and low price, homoge-
neous electrocatalysts are the key towards understanding and
identification of essential features for an efficient electro-
catalytic CO2 reduction reaction (CO2RR).[14,18,19] The enormous
bandwidth of homogeneous CO2RR electrocatalysts can be
divided into noble metal and non-noble metal containing
catalysts.
One of the most prominent noble metal containing electro-
catalyst is Re(bpy)(CO)3Cl (1, bpy=2,2’-bipyridine), that was first
described by Lehn et al. in 1984 (Scheme 1).[20] At a moderate
potential of À 1.25 V vs. NHE in DMF/water (9:1) in the presence
of a glassy carbon working electrode, complex 1 catalyzes the
selective reduction of CO2 to CO with a Faraday Efficiency (FE)
of up to 98%.[20,21] The catalytically active species was identified
as [Re(bpy)(CO)3]À , which is generated by two successive
electron uptakes, with the initial being located at the bpy
ligand, leaving a vacant coordination site at the Re0 center
upon loss of the chloride ligand.[22–24] During catalysis, CO2 is
initially activated and bound to the vacant site of the rhenium
center as η1-CO2H. A subsequent electron and proton uptake,
[a] P. Gerschel, Dr. A. L. Cordes, S. Bimmermann, Prof. Dr. N. Metzler-
Nolte, Prof. Dr. U.-P. Apfel
Inorganic Chemistry I
Ruhr-Universität Bochum
Universitätsstraße 150
44801 Bochum, Germany
E-mail: ulf.apfel@rub.de
[b] Dr. D. Siegmund, Prof. Dr. U.-P. Apfel
Department of Electrosynthesis
Fraunhofer UMSICHT
Osterfelder Straße 3
46047 Oberhausen, Germany
Supporting information for this article is available on the WWW
under https://doi.org/10.1002/zaac.202000450
© 2021 The Authors. Zeitschrift für anorganische und allgemeine
Chemie published by Wiley-VCH GmbH. This is an open access
article under the terms of the Creative Commons Attribution Non-
Commercial NoDerivs License, which permits use and distribution
in any medium, provided the original work is properly cited, the use
is non-commercial and no modifications or adaptations are made.
Scheme 1. Schematic overview of complexes Re(bpy)(CO)3Cl 1, [Ni
{N4}]2+ 2 and [Re(tacn)(CO)3]+ 3.
Z. Anorg. Allg. Chem. 2021, 647, 968–977
968 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
accompanied by water release, generates the intermediary Results and Discussion
tetracarbonyl ReI complex [Re(bpy)(CO)4]0. In a last step, CO is
released after additional electron uptake and the catalytically
active species is recovered.[23,24]
Complex Syntheses and Characterization
In order to further understand the electrocatalytic CO2RR
performance of complex 1 a broad variety of modulations of
the bpy ligand and furthermore substitutions of bpy for
comparable ligand platforms (e.g. pyNHC, pypz, pyiz) were
investigated.[23–37] Among the large number of modified
rhenium complexes and within the most studied class, the 4,4’-
substituted bpy rhenium complexes, the tBu-derivative Re-
(bpy-tBu)(CO)3Cl (bpy-tBu=4,4’-di-tert-butyl-2,2’-bipyridine) is
particularly noteworthy.
While the {N4} ligand can be purchased, the cyclam based
ligand {N2S2} as well as the isocyclam based ligands {i-SN3}, {i-
ON3} and {i-N4} itself were synthesized according to literature-
known procedures.[44,48] For the synthesis of the corresponding
rhenium complexes, the macrocyclic ligands {N4}, {N2S2}, {i-N4},
{i-SN3} and {i-ON3} were reacted with equimolar amounts of
Re(CO)5Br in acetone under reflux conditions. After purification
via recrystallization, the desired rhenium complexes Re{N4} (=
[Re(k3-N,N,N-{N4}(CO)3]Br, 87%), Re{N2S2} (=[Re(k3-S,S,N-
{N2S2}(CO)3]Br, 90%), Re{i-N4} (=[Re(k3-N,N,N-{i-N4}(CO)3]Br,
74%), Re{i-SN3} (=[Re(k3-S,N,N-{i-SN3}(CO)3]Br, 92%) and Re{i-
ON3} (=[Re(k3-N,N,N-{i-ON3}(CO)3]Br, 98%) were obtained as
colorless solids in good yields (Scheme 2). The successful
complexation and furthermore the exact composition of the
rhenium complexes was supported by electrospray-ionization
mass spectrometry (ESI-MS), revealing characteristic peaks at m/
z=470.9 (Re{N4}), 504.8 (Re{N2S2}), 470.9 (Re{i-N4}), 487.9 (Re{i-
SN3}) and 471.9 (Re{i-ON3}) for the corresponding [Re(L)(CO)3]+
fragments. Since the observed complex ions do not contain a
bromine, it can be concluded that the rhenium complexes are
cationic and bromide acts as non-coordinating counter ion.
Single crystals of complexes Re{N4}, Re{i-N4}, Re{i-SN3} and
Re{i-ON3} were obtained either from acetone or methanol
solutions and provided further information about the coordina-
tion environment of the rhenium centers and binding modes of
the corresponding macrocycles.
Although operating at a ~0.1 V higher overpotential, Re-
t
(bpy- Bu)(CO)3Cl reveals a ten times higher catalytic activity
compared to complex 1.[23,25] Likewise, a series of rhenium
tricarbonyl complexes containing aniline-substituted bpy deriv-
atives with the amine function at ortho-, meta- and para-
position was investigated. While the meta-version reveals the
best activity within the series with a more than three times
higher turnover frequency (TOF) compared to complex 1, the
para- version shows an inhibition, likely due to coordination of
the aniline-amine function to the reduced rhenium center.[31]
Among the non-noble metal containing electrocatalysts,
[Ni(cyclam)]2+ (2, cyclam={N4}=1,4,8,11-tetraazacyclotetra-dec-
ane) is one of the most efficient systems with a high FE of 96%
and a reaction rate of 32 hÀ 1 for the selective reduction of CO2
to CO at a potential of À 1.05 V vs. NHE at pH 4.1 in water on a
mercury working electrode (Scheme 1).[38–42]
Along this line, we recently investigated the effects
regarding electrocatalytic CO2RR upon heteroatom replacement
and geometric changes caused by the cyclam ligand.[43,44]
Although N/S replacement in case of [Ni{N2S2}]2+ ({N2S2}=1,8-
All rhenium centers are coordinated octahedrally as fac-
isomers by three carbonyl ligands and additional three donor
atoms of the macrocycles in a k3-manner, leaving one donor
atom of the macrocycles uncoordinated (Figure 1). Due to the
symmetry of ligand {N4} there is only one k3-isomer possible in
the case of complex Re{N4}. The averaged ReÀ N distance of Re
{N4} is 2.235 Å and the averaged ReÀ C and CÀ O bond lengths
are 1.913 Å and 1.153 Å, respectively. Analysis of the bond
angles furthermore reveals that the octahedral coordination
sphere of complex Re{N4} is highly distorted with the LÀ ReÀ L
dithia-4,12-diaazacyclotetradecane) resulted in
a 0.2 V de-
creased overpotential for the electrocatalytic CO2 reduction, the
CO2RR selectivity was severely mindered.[43] Similar results were
obtained for the Ni-isocyclam derivative (isocyclam={i-N4}=
1,4,7,11-tetraazacyclotetradecane), [Ni{i-SN3}]2+ ({i-SN3}=1-thia-
2+
4,8,12-triazacyclotetradecane) and [Ni{i-ON3}]
({i-ON3}=1-
oxa-4,8,12-triazacyclotetradecane).[44]
°
°
We herein set out to combine both mentioned approaches
and investigated rhenium complexes bearing macrocyclic
cyclam and isocyclam based ligands for the CO2RR. Notably,
besides the synthesis and characterization of [Re(tacn)(CO)3]+
(3, tacn=1,4,7-triaazacyclo-nonane) and its N/S or N/O heter-
oatom replaced derivatives, there are no studies on rhenium
tricarbonyl complexes comprising azamacrocyclic ligands
(Scheme 1).[45–47] We thus herein present the synthesis and
characterization of a novel series of cyclam based rhenium
tricarbonyl complexes and their electrochemical properties.
binding angles ranging from 102.9 to 77.8 (Table S1).
Scheme 2. Syntheses of the rhenium complexes Re{N4}, Re{N2S2}, Re
{i-N4}, Re{i-SN3} and Re{i-ON3}.
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
969 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
octahedral geometry in Re{i-SN3} is less distorted compared to
°
°
Re{i-N4} with LÀ ReÀ L binding angles ranging from 98.6 to 78.7
(Table S1).
Unlike Re{i-SN3}, where the non-nitrogen donor atom of the
macrocycle binds to the rhenium center, in Re{i-ON3} the
oxygen atom of the {i-ON3} macrocycle remains uncoordinated.
The formation of this specific k3-isomer of Re{i-ON3} can be
justified by the HSAB concept. Since ReI is a soft transition
metal, the coordination of the comparatively softer nitrogen
donor atoms is favored over the relatively hard oxygen donor
atom. Although the k3-N coordination mode herein is geometri-
cally unfavored, Re{i-ON3} is adopting this binding mode to
avoid the coordination of oxygen. With an averaged distance of
2.253 Å the ReÀ N bond length in Re{i-ON3} is significantly
longer compared to Re{i-N4} and slightly longer compared to
Re{N4}. The averaged ReÀ C and CÀ O bond lengths of Re{i-ON3}
are with 1.917 Å and 1.154 Å in a similar range as those of Re{i-
N4} and Re{N4}. However, complex Re{i-ON3} is compared to Re
{i-N4} overall less distorted with LÀ ReÀ L binding angles ranging
Figure 1. Molecular structures of the complexes Re{N4} (A), Re{i-N4}
°
°
from 97.6 to 79.4 (Table S1).
(B), Re{i-SN3} (C) and Re{i-ON3} (D) shown with thermal ellipsoid
drawn at the 50% probability level. Hydrogen atoms and bromide
counterions are omitted for clarity. Gray: carbon, blue: nitrogen,
yellow: sulfur, red: oxygen and dark blue: rhenium.
Especially for carbonyl complexes, IR spectroscopy is a
valuable tool in order to gain information about the geometry
and symmetry of the complexes and to determine the effects of
the remaining ligands on the electronic properties of the metal
center. The IR spectra of complexes Re{N4}, Re{N2S2}, Re{i-N4},
Re{i-SN3} and Re{i-ON3} reveal a similar band structure in the
carbonyl region, with a sharp band over 2000 cmÀ 1 and a broad
band around 1900 cmÀ 1, which is characteristic for [Re(L)(CO)3]+
fac-isomers (Figure 2).[46] However, within the different rhenium
complexes there are small alterations in the general band
structure, that can be explained by the different donor atom
sets and the differently distorted octahedron geometries. While
for all k3-N complexes Re{N4}, Re{i-N4} and Re{i-ON3} the sharp
In contrast to {N4}, three k3-isomers are possible for the
simplest structurally related ligand {i-N4}. However, only one
isomer is obtained in case of Re{i-N4} with the nitrogen atom
between the propyl units uncoordinated. With an averaged
ReÀ N distance of 2.214 Å the nitrogen atoms in Re{i-N4} are
slightly stronger bonded to the rhenium center compared to Re
{N4}. However, the averaged ReÀ C and CÀ O bond lengths of Re
{i-N4} remain comparatively equal with 1.912 Å and 1.150 Å.
Furthermore, bond angle analysis shows that the octahe-
dron of Re{i-N4} is overall less distorted compared to Re{N4}
(Table S1).
Similar to ligand {i-N4}, three k3-isomers for the correspond-
ing [Re(k3-{L})(CO)3]+ complexes are possible for the ligands {i-
SN3} and {i-ON3}. While the formation of the k3-isomer of Re{i-
N4} seems to be exclusively dependent on geometrical factors,
the k3-isomer choices of Re{i-SN3} and Re{i-ON3} are additionally
dependent on the nature of the donor atoms due to the mixed
donor sets provided.
In case of Re{i-SN3} the sulfur donor atom of the macrocycle
binds to the rhenium center and similar to Re{i-N4} the nitrogen
atom between the propyl units remains uncoordinated. The
formation of this specific k3-isomer is favored by two factors.
On the one hand the complex occupies a preferred thermody-
namic geometry and on the other hand sulfur is coordinated as
preferred donor atom according to the HSAB concept. Complex
Re{i-SN3} has a ReÀ S distance of 2.456 Å and an averaged ReÀ N
bond length of 2.237 Å. Moreover, due to the influence of the
sulfur donor atom the ReÀ C bond lengths is altered significantly
compared to Re{i-N4}. While the averaged CÀ O distance of Re{i-
SN3} is maintained with 1.150 Å, the averaged ReÀ C bond
length is comparatively elongated with 1.928 Å, indicating a
weaker π-backbonding in case of Re{i-SN3}. Moreover, the
Figure 2. Solid ATR-IR spectra of the rhenium complexes Re{N4}
(red), Re{N2S2} (blue), Re{i-N4} (yellow), Re{i-SN3} (green) and Re{i-
ON3} (violet). For comparison, the band positions of Re{N4} are
shown as grey dotted lines.
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
970 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
CO band is observed at ~2014 cmÀ 1, for the k3-S/N complexes
complex Re{N2S2} twenty signals are observed in the corre-
Re{i-SN3} (2019 cmÀ 1) and Re{N2S2} (2027 cmÀ 1) the same CO
band is observed at higher wavenumbers with increasing sulfur
content. This finding indicates a weaker π-backbonding of the
carbonyl ligands upon sulfur coordination, which is in line with
the ReÀ C and CÀ O bond length analysis of complex Re{i-SN3}.
While the CO band at higher wavenumbers seems to depend
mainly on the donor atoms nature of the macrocycles, the
broad CO band at lower wavenumbers depends additionally on
the geometry of the complexes. Here again the sulfur
coordination causes a shift of the broad band to higher
wavenumbers for Re{N2S2} and Re{i-SN3}, while a less distorted
octahedron geometry, like in cases of Re{i-N4}, Re{i-SN3} and
especially Re{i-ON3}, causes a shift to lower wavenumbers
(Figure 1 and Table 1).
sponding 13C{1H} NMR spectrum, indicating either an equili-
brium between the two possible k3-isomers or the presence of
two different conformational isomers in solution (Figure S2). In
case of the, according to their crystal structures, C2v symmetric
isocyclam based complexes Re{i-N4}, Re{i-SN3} and Re{i-ON3},
five different signals are expected in their 13C{1H} NMR spectra
for the corresponding macrocyclic ligand. However, all 13C NMR
spectra of the complexes Re{i-N4}, Re{i-SN3} and Re{i-ON3}
reveal additional signals besides the expected dominant five
signals, also indicating the presence of different k3-or conforma-
tional isomers in solution (Figures S3–S5).
Further investigations of the rhenium complexes using UV/
vis/NIR spectroscopy show in all cases a similar double
absorption band structure in the UV region around 320 nm,
whereby the exact band positions of the rhenium complexes
differ only by 1–2 nm (Table 1 and Figure S6). Furthermore,
while in case of the k3-N complexes Re{N4}, Re{i-N4} and Re{i-
ON3} the band at lower wavelengths can be seen as shoulder of
the absorption at higher wavelengths, the same band is more
pronounced in case of the k3-S/N complexes Re{N2S2} and Re{i-
SN3} and can be seen as discrete band.
Although no molecular structure of complex Re{N2S2} was
obtained, we can now predict the k3-binding mode of the
{N2S2} ligand based on the IR data of complexes Re{N2S2} and
Re{i-SN3} and the molecular structure of Re{i-SN3}. Since the CO
bands of Re{N2S2} are observed at significantly higher wave-
numbers compared to the bands of complex Re{i-SN3}, where
one sulfur atom is coordinated to the rhenium center, and the
coordination of sulfur atoms is known to result in a weaker π-
backbonding, complex Re{N2S2} is most likely the k3-S,S,N
isomer.
Electrochemical Characterization
While the thermodynamically preferred coordination modes
of the rhenium complexes Re{N4}, Re{i-N4}, Re{i-SN3} and Re{i-
ON3} in the solid state were revealed by crystal structure
analyses, no information is provided about the structures in
solution. In order to shed light on the structures of the rhenium
complexes in solution 13C{1H} NMR spectroscopic experiments in
deuterated methanol were carried out. Since for the C1
symmetric complex Re{N4} only one k3-isomer is thinkable, ten
different signals for the macrocyclic ligand are expected and
found in the corresponding 13C{1H} NMR spectrum (Figure S1).
Contrary to complex Re{N4}, in case of the C1 symmetric
All rhenium complexes Re{N4}, Re{N2S2}, Re{i-N4}, Re{i-SN3} and
Re{i-ON3} were characterized by cyclic voltammetry (CV) in dry
acetonitrile with 0.1 M [nBu4N]PF6 as supporting electrolyte in
an argon atmosphere at a scan rate of 100 mVsÀ 1 (Figure 3).
Comparison of the CVs of the different complexes reveals
that all rhenium complexes have the same shape and features
in their voltammograms. At potentials more negative than
À 2.30 V vs. Fc/Fc+ they show an irreversible ReI/0 redox couple
and at potentials more positive than 0.35 V vs. Fc/Fc+ they
show an irreversible ReII/I redox couple. In all cases, the
mentioned ReI/0 and ReII/I couples stay fully irreversible, when
altering the scan rate (Figures S7–S11).
Although the CVs of all rhenium complexes have the same
shape, the individual complexes differ significantly in their
cathodic peak potentials Epc for the ReI/0 couple (Table 1).
Compared to Re{N4}, which reveals an Epc of À 2.68 V vs. Fc/Fc+
for the ReI/0 couple, the exchange of two nitrogen donor atoms
against sulfur while maintaining the ligand geometry leads to
an anodic shift of the Epc by 0.37 V to À 2.31 V vs. Fc/Fc+, as in
the case of Re{N2S2}. Changing the geometry compared to Re
{N4} while maintaining the donor atoms, as in the case of Re{i-
N4}, also results in an anodic but comparatively smaller shift of
the Epc by 0.1 V to À 2.57 V vs. Fc/Fc+. When combining the
effects of geometric change and donor atom exchange, which
individually cause a shift of the Epc to more anodic potentials, a
more anodic Epc would be expected in the case of Re{i-SN3}
compared to Re{i-N4}. However, with À 2.56 V vs. Fc/Fc+ the Epc
of Re{i-SN3} remains unaltered compared to Re{i-N4}. Further-
more, complex Re{i-ON3}, with the actual unfavored k3- N
geometry shows an expected and compared to Re{i-N4} more
cathodic Epc of À 2.69 V vs. Fc/Fc+ for the ReI/0 redox couple.
Table 1. Characteristic IR and UV spectroscopic features and
electrochemical data of complexes Re{N4}, Re{N2S2}, Re{i-N4}, Re{i-
SN3} and Re{i-ON3}.
complex
UV MeCN
λ (nm)
Epc[a] MeCN
(V)
Epc[a] MeCN/
H2O (V)
IR solid
v (cm
À 1
~
)
Re{N4}
2014
1898
1880
2027
1906
2014
1873
2019
1886
2015
1894
1875
1853
317
322 (sh)
À 2.68
À 2.55
Re{N2S2}
Re{i-N4}
318
324
318
323 (sh)
318
324
318
323 (sh)
À 2.31
À 2.57
À 2.56
À 2.69
À 2.23
À 2.47
À 2.31
À 2.48
Re{i-SN3}
Re{i-ON3}
[a] Cathodic peak potential vs. Fc/Fc+ at 100 mVsÀ 1
.
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
971 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
The anodic peak potentials Epa for the ReI/II couple also differ
for the individual rhenium complexes. However, the difference
between all complexes is much less pronounced and within a
range of 80 mV. Moreover, the Epc for the ReI/0 couple is more
relevant compared to the Epa for the ReII/I couple, since the Re0
species is expected to be the potentially catalytic active species
for the CO2 reduction reaction.
Finally, the electrochemical properties of complexes Re{N4},
Re{N2S2}, Re{i-N4}, Re{i-SN3} and Re{i-ON3} were analyzed in an
acetonitrile/water mixture (4:1), since water acts as proton
source for the later examined electroreduction of CO2 (Figure 4,
dashed lines).
The CVs of the rhenium complexes in the presence of water
reveal the same features as the measurements in pure
acetonitrile. All complexes show a fully irreversible ReI/0 couple
and a compared to the measurements in pure acetonitrile
slightly more reversible ReII/I couple. While the ReI/0 couples of
all rhenium complexes and the ReII/I couples of Re{N2S2}, Re{i-
N4} and Re{i-SN3} stay irreversible when increasing the scan
rate, the nature of the ReII/I couples of Re{N4} and Re{i-ON3} can
be identified as quasi-reversible based on the measurements
with increased scan rates (Figures S12–S16).
Although in presence of water the cathodic peak potentials
of the ReI/0 couples are generally shifted to more anodic
potentials compared to the measurements in acetonitrile, the
individual shifts of the rhenium complexes are different
(Table 1). While for complexes Re{N4}, Re{N2S2} and Re{i-N4} the
potential shift is around 0.1 V, the shift for complexes Re{i-SN3}
and Re{i-ON3} is with 0.2 V significantly higher.
Figure 3. Cyclic voltammograms of 1 mM solutions of complexes
Re{N4}, Re{N2S2}, Re{i-N4}, Re{i-SN3} and Re{i-ON3} in acetonitrile
with 0.1 M [nBu4N]PF6 at 100 mVsÀ 1
.
Figure 4. Cyclic voltammograms of 1 mM solutions of complexes Re{N4}, Re{N2S2}, Re{i-N4}, Re{i-SN3} and Re{i-ON3} in acetonitrile/water
(4:1) under argon (dashed lines) and saturated with CO2 (solid lines) with 0.1 M [nBu4N]PF6 at 100 mVsÀ 1
.
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
972 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
After saturation of the acetonitrile/water (4:1) solutions of
Table 2. Quantification of the gaseous electrolysis products by
the rhenium complexes with CO2, the shape of the CVs changed
drastically in all cases compared to the measurements in an
argon atmosphere (Figure 4, solid lines). In the presence of CO2
a significant reductive current increase is observed for com-
plexes Re{N4}, Re{N2S2}, Re{i-N4} and Re{i-ON3} after generation
of the Re0 species, indicating a potential interaction with CO2.
For complex Re{i-SN3}, the same kind of current increase is
observed, however, much less pronounced. Moreover, a much
higher reductive current maximum ipc around 100 μA is
observed for all k3-N complexes Re{N4}, Re{i-N4} and Re{i-ON3},
compared to the ipc around 40 μA observed for the k3-S,N
complexes Re{N2S2} and Re{i-SN3}. Furthermore, in the presence
of CO2 all complexes reveal an additional oxidative signal prior
the wave associated with the ReII/I redox couple. This oxidation
peak only appears when generating the Re0 species beforehand.
If the CV scan is performed solely in oxidative direction, the
wave is missing in all cases (Figure S17). This finding further
suggests that the electrochemically generated Re0 species
interact with CO2 and potentially catalyze its reduction.
Analysis of the icat/ipc ratios of the complexes Re{N4} (1.63),
Re{N2S2} (1.45), Re{i-N4} (1.66), Re{i-SN3} (1.14) and Re{i-ON3}
(2.15) in acetonitrile/water (4:1) shows that the catalytic
activities of the rhenium complexes studied here are signifi-
cantly lower compared to the benchmark electrocatalysts
Re(bpy)(CO)3Cl (4.4 in anhydrous acetonitrile, 13.4 in
acetonitrile+4% TFE) and [Ni(cyclam)]2+ (~4 in acetonitrile/
water (4:1)).[31,49]
GC-MS.
[a]
complex
ECPC
(V)
Time
(h)
FEH2
(%)
FECO
(%)
Q
(C)
TON[b]
blank
À 2.68
À 2.68
À 2.47
À 2.59
À 2.35
À 2.61
2
4
6
8
24
2
4
6
8
24
2
4
6
8
24
2
4
6
8
24
2
4
6
8
24
2
4
6
78
69
89
95
96
53
67
63
69
57
46
56
44
53
41
44
56
69
70
75
89
83
76
72
69
52
60
45
44
49
2.1
0.2
0.5
0.3
0.7
4.8
7.4
9.3
11
0.12
0.35
1.5
-
3.7
22.3
0.17
0.56
1.0
1.4
3.9
0.12
0.22
0.31
0.43
1.56
0.47
1.3
2.3
3.6
22.4
0.10
0.40
0.75
1.0
Re{N4}
0.90
0.08
4.10
0.62
0.93
11
Re{N2S2}
Re{i-N4}
Re{i-SN3}
Re{i-ON3}
4.5
4.5
4.7
4.9
4.9
13
15
18
17
6.9
2.9
7.0
7.3
7.8
6.5
10
7.7
7.9
11
4.0
0.14
0.33
0.62
0.90
3.1
8
24
21
Electrocatalytic CO2 Reduction
[a] Potential of the controlled potential coulometry vs. Fc/Fc+. [b]
Turnover number (TON) of all CO2 reduction based processes.
In order to determine the formed reduction products and
furthermore the amount as well as selectivity they are formed
within the catalytic regions of the different rhenium complexes,
additionally controlled potential colometry (CPC) experiments
were carried out (Figure S18). These long-term electrolyses were
performed at defined potentials for 24 h in a one-compartment
cell using a three-electrode setup with glassy carbon as working
electrode, a silver wire as pseudo-reference electrode and a
platinum wire as counter electrode. After saturation of the
1 mM rhenium complex and 0.1 M [nBu4N]PF6 containing
acetonitrile/water (4:1) solution with CO2, the cell was sealed
tightly for the time of the experiment. The CPCs with the
different rhenium complexes were carried out at the corre-
sponding peak potentials of the catalytic currents observed in
the presence of CO2 (Figure 4 and Table 2). During electrolysis,
the headspace composition of the cell was analyzed every 2 h
within the first 8 h via GC-MS (Table 2) and after 24 h
electrolysis, the composition of the headspace of the cell and
additionally the composition of the liquid phase of the cell was
analyzed via GC-MS (Figure 5). The selectivity of the product
formation utilizing the different rhenium complexes is reflected
by the Faradaic Efficiencies (FE) and in combination with the
consumed charge the maximum amounts of formed products
can be determined along with a statement about the activity.
In general, during all experiments utilizing the different
rhenium complexes only H2 and CO were obtained as gaseous
products. The complexes generally herein suppress the H2
formation which is the main reaction in the absence of any ReI
complex. In addition, after 24 h of electrolysis, none of the
experiments showed the formation of a deposited species
during rinse test analysis, which would be an indication for
catalyst decomposition or at least render the catalytically active
species to be a heterogeneous one.
Within the first 8 h of CPC with Re{N4} as catalyst, the
cumulated FEs for the production of H2 and CO increased from
58% to 80% during electrolysis, whereby the CO/H2 ratio
shifted from 1:11 to 1:6 in favor of CO, resulting in a FE of 11%
for CO after 8 h.
After 24 h of electrolysis, the FE for the generation of H2
decreased to 57%, while the FE for the generation of CO was
maintained at 11%, further improving the CO/H2 ratio. Addi-
tionally, liquid phase analysis revealed formic acid with a FE of
7.2% as well as methanol with a notably FE of 12% as further
CO2RR products. Although the use of Re{N4} as electrocatalyst
reveals a low selectivity for the CO2RR with a total FE of ~30%
for the sum of all CO2 reduction products, the selectivity is
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
973 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
FE of 2.3% were identified as further reduction products.
Although Re{i-N4} with a summed FE of ~19% shows a lower
selectivity for the CO2RR compared to Re{N4}, the absolute
amounts of CO2 reduction products are significantly higher for
Re{i-N4}, since the total consumed charge and thus the activity
is almost six times higher compared to Re{N4}.
The averaged CO/H2 ratio of Re{i-SN3} is with 1:10 similar to
that of Re{N2S2} during the first 8 h, except for the first 2 h.
However, in contrast to Re{N2S2} the absolute FEs for Re{i-SN3}
are higher with an averaged FE of 77% for H2 and 7.4% for CO.
After 24 h electrolysis with Re{i-SN3} as catalyst, the FEs for the
gaseous products dropped to ~76% in total, while the previous
CO/H2 ratio of 1:10 was maintained. Again, the liquid phase
analysis revealed formic acid and methanol as further CO2RR
products with FEs of 7.1% and 3.9% respectively. Compared to
Re{N2S2} the CO2RR selectivity of Re{i-SN3} is significantly higher
and comparable with that of Re{i-N4}. However, the consumed
charge and consequently the catalytic activity of Re{i-SN3} is
significantly lower compared to Re{i-N4} and comparable with
that of Re{N4}.
The FEs of Re{i-ON3} for the formation of the gaseous
products within the first 8 h of electrolysis are relatively
constant with an averaged FE of 50% for H2 and 9% for CO,
resulting in an averaged CO/H2 ratio of ~1:5. However, product
analysis after 24 h of electrolysis with Re{i-ON3} as catalyst
revealed a significant shift of the CO/H2 ratio in favor of CO to
almost 1:2, resulting in FEs for H2 and CO of 49% and 21%
respectively. Furthermore, liquid phase analysis identified
methanol as further reduction product with a remarkable FE of
24%. While generation of formate/formic acid during electro-
catalytic CO2 reduction is rather common,[21,51,52] the formation
of methanol is particularly noteworthy. Although experiments
performed in CD3CN show no incorporation of deuterium into
methanol or formic acid, final proof for the formation of
methanol from CO2 must be obtained by utilization of 13CO2.
Consequently, an electrolysis experiment was carried out using
13CO2 and Re{i-ON3} as a proxy for all. Unfortunately, due to the
lack of a suitable analytical method for these small amounts of
methanol produced, we could not unequivocally identify CO2 as
the origin of methanol.
Figure 5. Faradaic Efficiencies (FE, stacked bars) and total consumed
charge (red lines) after 24 h long-term electrolysis of 1 mM rhenium
complex solutions in CO2 saturated acetonitrile/water (4:1) with
0.1 M [nBu4N]PF6 at a defined potential. The potential during
electrolysis was held at À 2.68 V vs Fc/Fc+ for blank, at À 2.68 V vs
Fc/Fc+ for Re{N4}, at À 2.47 V vs Fc/Fc+ for Re{N2S2}, at À 2.59 V vs
Fc/Fc+ for Re{i-N4}, at À 2.35 V vs Fc/Fc+ for Re{i-SN3} and at
À 2.61 V vs Fc/Fc+ for Re{i-ON3}. Gray: hydrogen (H2), blue: carbon
monoxide (CO), green: formic acid (FA) yellow: methanol (MeOH)
and red: acetaldehyde (CÀ CHO).
drastically increased compared to the blank measurement in
absence of any catalyst (FECO 0.7%). However, it must be taken
into account that the charge consumed during electrolysis in
presence of Re{N4} is almost six times lower compared to the
measurement without catalyst, indicating a lower total catalytic
activity for Re{N4}.
Compared to Re{N4}, the FEs of Re{N2S2} for the formation
of the gaseous products within the first 8 h are lower and
relatively constant with an averaged FE of 50% for H2 and 4.7%
for CO, resulting in an averaged CO/H2 ratio of 1:11. Product
quantification after 24 h electrolysis using Re{N2S2} as catalyst
revealed a decrease in the FE for the H2 production to 41%,
while the FE for the CO production was maintained at 4.9%.
Furthermore, liquid phase analysis showed the formation of
acetaldehyde with
a
FE of 18%. However, this product
While the catalytic activity of Re{i-ON3} based on the
consumed charge is comparable to that of Re{N4} and Re{i-SN3},
the CO2RR selectivity of Re{i-ON3} is highest within the series of
rhenium complexes investigated herein with a total FE of 45%.
One explanation for the above average CO2RR selectivity of
complex Re{i-ON3} compared to all other rhenium complexes
investigated herein is the nature of the uncoordinated donor
atom of the macrocyclic ligand. During the catalytic cycle a free
binding site at the rhenium center of the [Re(k3-{L})(CO)3]+
complexes is expected, as in case of Re(bpy)(CO)3Cl.[23,24] While
Re(bpy)(CO)3Cl only binds CO2 at this position, the [Re(k3-
{L})(CO)3]+ complexes can also bind the uncoordinated donor
atom of the macrocycle at this position and thus intrinsically
inhibit catalysis.[31] Since nitrogen donor atoms reveal a higher
binding affinity to low-valent rhenium centers compared to
oxygen donor atoms, the intrinsic inhibition in case of Re{N4},
originates from the hydrogenation of acetonitrile, rather than
CO2 reduction, as already observed in previous studies on
similar nickel- and cobalt-containing complexes.[44,50] Moreover,
besides the drastically lower CO2RR selectivity of Re{N2S2}
compared to Re{N4}, the total consumed charge and the total
FE of the electrolysis with Re{N2S2} are significantly lower
compared to that with Re{N4}, indicating a comparatively lower
catalytic activity for Re{N2S2}.
In case of Re{i-N4} the FEs for the generation of both H2 and
CO increases gradually during the first 8 h of electrolysis,
reaching a plateau with a FE of 70% for H2 and 17% for CO.
Although Re{i-N4} revealed a higher FE for the production of CO
after 8 h compared to Re{N4}, the FE of Re{i-N4} for CO
formation dropped drastically to ~7% after 24 h electrolysis,
while the FE for the generation of H2 increased slightly to 75%.
Additionally, formic acid with a FE of 10% and methanol with a
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
974 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
°
were recorded with a JASCO V-670 instrument at 25 C and are
reported in nm.
Re{N2S2}, Re{i-N4} and Re{i-SN3} is higher compared to Re{i-
ON3}.
General Procedure for the Syntheses of the Complexes [Re(k3-
{L})(CO)3]Br: Ligand {N4}, {N2S2}, {i-N4}, {i-SN3} or {i-ON3} (1 equiv.)
was dissolved in degassed acetone (48 mLmmolÀ 1) and Re(CO)5Br
(1 equiv.) was added to the macrocycle solution. The reaction
mixtures were heated under reflux conditions for 12 h. In all cases a
color change to light yellow/brown was observed. Subsequently,
the reaction mixtures were cooled to room temperature, filtered,
and concentrated. After recrystallization pure complexes Re{N4}, Re
{N2S2}, Re{i-N4}, Re{i-SN3} and Re{i-ON3} were obtained.
Conclusions
In summary, we herein describe the synthesis and character-
ization of hitherto unprecedented cyclam based rhenium
complexes Re{N4}, Re{N2S2}, Re{i-N4}, Re{i-SN3} and Re{i-ON3}
along with their performances in the electrocatalytic reduction
of CO2.
Within the series of the cyclam based rhenium complexes
interesting trends are observed. Since the rhenium center
selects three of the four donor atoms of the macrocyclic ligand
for coordination, a ranking of the binding affinity of ReI to
different donor atoms in the order of S>N>O was revealed.
Moreover, it was shown that in case of the isocyclam derivatives
the k3-coordination mode, which leaves the donor atom
between the propyl units uncoordinated, is preferred due to
the minimized ring strains.
Regarding the electrocatalytic CO2 reduction, all rhenium
complexes reveal a significant increase in CO2RR selectivity
compared to the measurement in absence of any catalyst.
However, H2 remains the major reduction product in all cases.
While Re{i-N4} shows the highest catalytic activity within the
series, Re{i-ON3} reveals the highest CO2RR selectivity after 24 h
with a total FE of 45%, which in turn is split into 21% CO and
24% methanol. The remaining complexes can be ranked
according to their CO2RR selectivity from Re{N4}>Re{i-N4}>Re
{i-SN3}>Re{N2S2}, where Re{N2S2} has only a FE of 4.9% for the
production of CO.
[Re(k3-N,N,N-{N4})(CO)3]Br (Re{N4}): The yellow solid obtained from
reaction of {N4} (50 mg, 0.25 mmol) with Re(CO)5Br (101 mg,
0.25 mmol) was recrystallized in hot acetone, yielding Re{N4} in
87% (120 mg, 0.22 mmol) as colorless solid. 13C{1H} NMR (75 MHz,
CD3OD): δ=24.6 (CH2CH2CH2), 43.9, 47.6, 48.5, 51.7, 57.7, 58.3, 59.7,
60.7 (CH2NH) ppm. ESI-MS: calcd. for [C13H24N4O3Re]+: m/z=471.1;
found: 470.9. EA: C13H24BrN4O3Re: calcd. C, 28.37; H, 4.39; N, 10.18%;
~
found: C, 28.47; H, 4.59; N, 10.35%. IR (ATR): v=3395, 3059, 2907,
2872, 2826, 2014, 1898, 1880, 1468, 1055, 1024 cmÀ 1. UV/vis/NIR
(CH3CN): 317, 322 (sh) nm.
[Re(k3-S,S/N,N-{N2S2})(CO)3]Br (Re{N2S2}): The yellow/brown solid
obtained from reaction of {N2S2} (50 mg, 0.21 mmol) with Re(CO)5Br
(87 mg, 0.21 mmol) was recrystallized using a mixture of acetone
and methanol as solvent and diethyl ether as precipitant. Complex
Re{N2S2} was obtained as colorless solid in 90% (110 mg,
0.19 mmol) yield. 13C{1H} NMR (75 MHz, CD3OD): δ=24.5, 27.2, 27.5,
28.5 (CH2CH2CH2), 28.8, 29.4, 29.7, 30.2, 30.7, 31.9, 32.5, 37.5 (CH2S),
39.1, 47.4, 47.7, 51.0, 53.7, 54.4, 57.4, 62.3 (CH2NH) ppm. ESI-MS:
calcd. for [C13H22N2O3ReS2]+: m/z=505.1; found: 504.8. EA:
C13H22BrN2O3ReS2 +0.33 CH3OH+0.67 H2O: calcd. C, 26.37; H, 4.09;
~
N, 4.61%; found: C, 26.25; H, 4.00; N, 4.72%. IR (ATR): v=3400,
2937, 2814, 2735, 2027, 1906, 1448, 1045, 1018 cmÀ 1. UV/vis/NIR
(CH3CN): 318, 324 nm.
[Re(k3-N,N,N-{i-N4})(CO)3]Br (Re{i-N4}): The yellow solid obtained from
reaction of {i-N4} (58 mg, 0.29 mmol) with Re(CO)5Br (117 mg,
0.29 mmol) was recrystallized using a mixture of acetone and
methanol as solvent and diethyl ether as precipitant. Complex Re{i-
N4} was obtained as colorless solid in 74% (118 mg, 0.21 mmol)
yield. 13C{1H} NMR (75 MHz, CD3OD): δ=28.2 (CH2CH2CH2), 50.4,
52.7, 56.5, 61.6 (CH2NH) ppm. ESI-MS: calcd. for [C13H24N4O3Re]+: m/
z=471.1; found: 470.9. EA: C13H24BrN4O3Re+0.33 CH3OH+0.33
H2O: calcd. C, 28.24; H, 4.62; N, 9.88%; found: C, 28.27; H, 4.53; N,
Notably, in comparison to the literature known electro-
catalyst Re(bpy)(CO)3Cl, the reduction of CO2 proceeds for all
herein investigated rhenium complexes at comparatively higher
overpotentials together with a lower CO2RR selectivity.
Experimental Section
All reactions were performed under a dry Ar or N2 atmosphere
using standard Schlenk techniques or by working in a glovebox.
Starting materials and chemicals were obtained from commercial
suppliers and used without further purification. All solvents were
dried and degassed according to standard methods.[53] 1,8-Dithia-
4,11-diazacyclotetradecane {N2S2},[48] 1,4,7,11-tetraazacyclotetrade-
cane {i-N4},[44] 1-thia-4,8,12-triazacyclotetradecane {i-SN3}[44] and 1-
oxa-4,8,12-triazacyclotetradecane {i-ON3}[44] were synthesized ac-
cording to literature-known procedures. NMR spectra were re-
corded at room temperature from the dissolved crystalline samples
with a Bruker AVIII-300 NMR spectrometer and are reported in parts
per million (ppm). Unfortunately, the recorded 13C{1H} NMR spectra
did not show the quaternary carbon signals associated with the
carbonyl ligands, and furthermore, assignment of the 1H signals
was not possible due to the high flexibility and anisotropy of the
complexes (Figure S19–S23). Mass spectra were obtained with a
Bruker Daltonics Esquire 6000 instrument. CHN analyses were
measured with an Elementar vario MICRO cube. IR spectra were
recorded with a Bruker Tensor 27 FT-IR instrument attached with a
Pike Miracle ATR unit and are reported in cmÀ 1. UV/vis/NIR spectra
~
9.77%. IR (ATR): v=3069, 2926, 2858, 2014, 1873, 1460, 1074, 1053,
1028 cm-1. UV/vis/NIR (CH3CN): 318, 323 (sh) nm.
[Re(k3-S,N,N-{i-SN3})(CO)3]Br (Re{i-SN3}): The yellow/brown solid ob-
tained from reaction of {i-SN3} (50 mg, 0.23 mmol) with Re(CO)5Br
(93 mg, 0.23 mmol) was recrystallized using a mixture of acetone
and methanol as solvent and diethyl ether as precipitant. Complex
Re{i-SN3} was obtained as light brown solid in 92% (120 mg,
0.21 mmol) yield. 13C{1H} NMR (75 MHz, CD3OD): δ=28.3
(CH2CH2CH2), 39.7 (CH2S), 50.3, 59.6, 62.8 (CH2NH) ppm. ESI-MS:
calcd. for [C13H23N3O3ReS1]+: m/z=488.1; found: 487.9. EA:
C13H23BrN3O3ReS+0.25 CH3OH+0.5 H2O: calcd. C, 27.23; H, 4.31; N,
~
7.19%; found: C, 27.20; H, 4.09; N, 6.89%. IR (ATR): v=3425, 3055,
2924, 2849, 2019, 1886, 1460, 1061, 1028 cm-1. UV/vis/NIR
(CH3CÀ N): 318, 324 nm.
[Re(k3-N,N,N-{i-ON3})(CO)3]Br (Re{i-ON3}): The yellow solid obtained
from reaction of {i-ON3} (50 mg, 0.25 mmol) with Re(CO)5Br
(101 mg, 0.25 mmol) was recrystallized using a mixture of acetone
and methanol as solvent and methanol as precipitant. Complex Re
{i-ON3} was obtained as colorless solid in 98% (135 mg, 0.24 mmol)
yield. 13C{1H} NMR (75 MHz, CD3OD): δ=26.6 (CH2CH2CH2), 50.4,
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
975 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
56.4, 62.6 (CH2NH), 71.0 (CH2O) ppm. ESI-MS: calcd. for
voltammograms, controlled potential coulometries, as well as
crystal structure refinements.
[C13H23N3O4Re]+: m/z=472.1; found: 471.9. EA: C13H23BrN3O4Re+
0.45 C3H6O: calcd. C, 29.84; H, 4.49; N, 7.28%; found: C, 29.99; H,
~
4.59; N, 7.57%. IR (ATR): v=3117, 3074, 2958, 2932, 2874, 2015,
1894, 1875, 1853, 1468, 1130, 1057 cm-1. UV/vis/NIR (CH3CN): 318,
323 (sh) nm.
Acknowledgements
Electrochemistry: Electrochemical studies were performed using a
PalmSens4 potentiostat in a standard three-electrode setup in a
one-compartment cell. A glassy-carbon electrode was used as
working electrode (WE), a Ag wire in a Luggin capillary as the
pseudo-reference electrode (PRE), and a Pt wire as the counter
electrode (CE). The working electrode was prepared by successive
polishing with 1.0, 0.3 and 0.1 μm sandpaper and subsequent
sonication in acetonitrile for 10 min. Tetrabutylammonium hexa-
fluorophosphate ([nBu4N]PF6, 0.1 M) was used as the electrolyte in
all electrochemical measurements either in anhydrous acetonitrile
or an acetonitrile/water mixture (4:1). Prior to each experiment, the
electrochemical cell was degassed with Ar for 10 min. The electro-
lyte solution was purged for 10 min either with Ar or CO2 and an Ar
or CO2 atmosphere was maintained throughout the measurement.
This work was supported by the Fonds of the Chemical Industry
(Liebig Grant to U.-P.A. and Kekulé fellowship to P.G.), the
Deutsche Forschungsgemeinschaft (Emmy Noether Grant to U.-
P.A., AP242/2-1 and AP242/5-1) and the Fraunhofer Internal
Programs (Attract, Grant No. 097-602175) as well as the
Fraunhofer Cluster of Excellence CINES. Furthermore, the work
was funded by the Deutsche Forschungsgemeinschaft under
Germany’s
Excellence
Strategy,
EXC-2033
(Projektnr.
390677874). Open access funding enabled and organized by
Projekt DEAL.
Keywords: Cyclam
· Rhenium · Electrocatalysis · Carbon
All cyclic voltammograms were recorded at
a scan rate of
100 mVsÀ 1 if not stated otherwise, and after every experiment all
pseudo-referenced potentials were referenced against the ferro-
cene/ferrocenium couple (Fc/Fc+). Controlled-potential coulometry
(CPC) was performed for 24 h at defined potentials in a sealed one-
compartment cell under conditions otherwise identical with those
reported above. The electrochemical cell for long-term electrolysis
contained 5 mL of electrolyte and a headspace of 57 mL.
dioxide · Carbonyl ligands
[1] M. Aresta, Ed., Carbon Dioxide as Chemical Feedstock, Wiley-
VCH Verlag GmbH&Co. KGaA, Weinheim, Germany, 2010.
[2] J. Qiao, Y. Liu, J. Zhang, Electrochemical Reduction of Carbon
Dioxide: Fundamentals and Technologies, CRC Press, Boca
Raton, 2016.
Quantification of Electrolysis Products: Quantification of the
headspace gas composition and liquid phase composition of the
electrochemical cell was performed using a Shimadzu GCMS-
QP2020 system equipped with two capillary columns and a MS
detector. Gas separation was performed via hand injection using a
Carboxen 1010 PLOT fused silica capillary GC column (L X I.D.
30 m×0.32 mm, average thickness 15 μm) and liquid phase separa-
tion was performed via headspace analysis using a SH-Rtx-200 ms
fused silica capillary GC column (L X I.D. 30 m×0.25 mm, average
thickness 1 μm). Helium was used as carrier gas and the product
amount was determined by calibration curves. The following
gaseous and liquid products/components were assayed via the
GCMS system: H2, O2, N2, CO, CH4, C2H4, C2H6, methanol, ethanol,
propanol, formic acid (formate), acetic acid (acetate), propionic acid
(propionate), acetaldehyde and propionaldehyde. Liquid samples
were acidified with concentrated sulfuric acid (10:1) prior injection.
[3] T. Stocker, The Physical Science Basis: Working Group
I
Contribution to the Fifth Assessment Report of the Inter-
governmental Panel on Climate Change, Climate Change 2013:
Cambridge University Press, New York, 2014.
[4] J. H. Butler, S. A. Montzka, “The NOAA Annual Greenhouse Gas
Index (AGGI),” can be found under https://www.esrl.noaa.gov/
gmd/aggi/aggi.html, 2019.
[5] R. A. Eggleton, A Short Introduction to Climate Change, Cam-
bridge University Press, Cambridge; Port Melbourne, Vic, 2012.
[6] M. Aresta, F. Nocito, A. Dibenedetto, Adv. Catal. 2018, pp. 49–
111.
[7] T. R. Karl, Science 2003, 302, 1719–1723.
[8] A. M. Appel, J. E. Bercaw, A. B. Bocarsly, H. Dobbek, D. L.
DuBois, M. Dupuis, J. G. Ferry, E. Fujita, R. Hille, P. J. A. Kenis,
C. A. Kerfeld, R. H. Morris, C. H. F. Peden, A. R. Portis, S. W.
Ragsdale, T. B. Rauchfuss, J. N. H. Reek, L. C. Seefeldt, R. K.
Thauer, G. L. Waldrop, Chem. Rev. 2013, 113, 6621–6658.
[9] R. M. Cuéllar-Franca, A. Azapagic, J. CO2 Util. 2015, 9, 82–102.
[10] C. Panzone, R. Philippe, A. Chappaz, P. Fongarland, A.
Bengaouer, J. CO2 Util. 2020, 38, 314–347.
[11] E. E. Benson, C. P. Kubiak, A. J. Sathrum, J. M. Smieja, Chem.
Soc. Rev. 2009, 38, 89–99.
[12] D. R. Kauffman, J. Thakkar, R. Siva, C. Matranga, P. R. Ohodnicki,
C. Zeng, R. Jin, ACS Appl. Mater. Interfaces 2015, 7, 15626–
15632.
X-ray Data Collection and Structure Solution Refinement: Single
crystals suitable for X-ray structure analysis were handled in
perfluorinated oil, mounted on a fiber loop, and placed in a cold N2
stream (Oxford Cryostream) throughout the measurement. X-ray
diffraction data were collected using either an Oxford Xcalibur-2
equipped with a Sapphire 2 CCD detector (Mo-Kα, λ=0.71073 Å), a
Rigaku Oxford diffraction XtaLAB SuperNova equipped with an
Atlas CCD detector (Cu-Kα, λ=1.54184 Å) or a Rigaku Oxford
diffraction XtaLAB Synergy-S with a HyPix-6000HE detector (Cu-Kα,
λ=1.54184 Å). The obtained data were analyzed with the program
packages CrysAlisPro and Platon.[54] The structures were solved by
intrinsic phasing using ShelXT[55] and refined against F2 (ShelXL[56]).
Hydrogen atoms were added in calculated positions and refined
using a rigid model. The program package OLEX2[57] served as a
graphical user interface. Crystallographic data and refinement
parameters are presented in Table S2 in the Supporting Informa-
tion.
[13] J. Qiao, Y. Liu, F. Hong, J. Zhang, Chem. Soc. Rev. 2014, 43, 631–
675.
[14] C. Long, X. Li, J. Guo, Y. Shi, S. Liu, Z. Tang, Small Methods
2018, 1800369.
[15] M. Rakowski Dubois, D. L. Dubois, Acc. Chem. Res. 2009, 42,
1974–1982.
[16] H. A. Hansen, J. B. Varley, A. A. Peterson, J. K. Nørskov, J. Phys.
Chem. Lett. 2013, 4, 388–392.
[17] A. A. Peterson, ECS Trans. 2015, 66, 41–52.
[18] F. Möller, S. Piontek, R. G. Miller, U.-P. Apfel, Chem. Eur. J. 2018,
24, 1471–1493.
Supporting Information (see footnote on the first page of this
article): The Supporting Information contains bond length and
bond angle analysis, NMR spectra, UV/vis/NIR spectra, cyclic
[19] D.-M. Feng, Y.-P. Zhu, P. Chen, T.-Y. Ma, Catalysts 2017, 7, 373.
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
976 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.

Journal of Inorganic and General Chemistry
ARTICLE
Zeitschrift für anorganische und allgemeine Chemie
[20] J. Hawecker, J.-M. Lehn, R. Ziessel, J. Chem. Soc. Chem.
[40] J. Song, E. L. Klein, F. Neese, S. Ye, Inorg. Chem. 2014, 53, 7500–
Commun. 1984, 328–330.
7507.
[21] J. Hawecker, J.-M. Lehn, R. Ziessel, Helv. Chim. Acta 1986, 69,
[41] G. B. Balazs, F. C. Anson, J. Electroanal. Chem. 1993, 361, 149–
1990–2012.
157.
[22] B. P. Sullivan, C. M. Bolinger, D. Conrad, W. J. Vining, T. J. Meyer,
J. Chem. Soc. Chem. Commun. 1985, 1414–1416.
[23] K. A. Grice, C. P. Kubiak, Adv. Inorg. Chem., Elsevier, 2014, pp.
163–188.
[24] A. W. Nichols, C. W. Machan, Front. Chem. 2019, 7, 397.
[25] J. M. Smieja, C. P. Kubiak, Inorg. Chem. 2010, 49, 9283–9289.
[26] S. Sung, D. Kumar, M. Gil-Sepulcre, M. Nippe, J. Am. Chem. Soc.
2017, 139, 13993–13996.
[42] J. D. Froehlich, C. P. Kubiak, Inorg. Chem. 2012, 51, 3932–3934.
[43] P. Gerschel, K. Warm, E. R. Farquhar, U. Englert, M. L. Reback, D.
Siegmund, K. Ray, U.-P. Apfel, Dalton Trans. 2019, 48, 5923–
5932.
[44] P. Gerschel, B. Battistella, D. Siegmund, K. Ray, U.-P. Apfel,
Organometallics 2020, 39, 1497–1510.
[45] K. Wieghardt, C. Pomp, B. Nuber, J. Weiss, Inorg. Chem. 1986,
25, 1659–1661.
[27] A. Nakada, O. Ishitani, ACS Catal. 2018, 8, 354–363.
[28] M. L. Clark, P. L. Cheung, M. Lessio, E. A. Carter, C. P. Kubiak,
ACS Catal. 2018, 8, 2021–2029.
[46] C. Pomp, S. Drüeke, H.-J. Küppers, K. Wieghardt, Z. Naturforsch.
B 1988, 43, 299–305.
[47] P. Hoffmann, A. Steinhoff, R. Mattes, Z. Naturforsch. B 1987, 42,
867–873.
[29] A. N. Hellman, R. Haiges, S. C. Marinescu, Dalton Trans. 2019,
48, 14251–14255.
[48] T. L. Walker, W. Malasi, S. Bhide, T. Parker, D. Zhang, A.
Freedman, J. M. Modarelli, J. T. Engle, C. J. Ziegler, P. Custer,
W. J. Youngs, M. J. Taschner, Tetrahedron Lett. 2012, 53, 6548–
6551.
[30] M. R. Madsen, J. B. Jakobsen, M. H. Rønne, H. Liang, H. C. D.
Hammershøj, P. Nørby, S. U. Pedersen, T. Skrydstrup, K.
Daasbjerg, Organometallics 2020, 39, 1480–1490.
[31] K. Talukdar, S. Sinha Roy, E. Amatya, E. A. Sleeper, P. Le Ma-
gueres, J. W. Jurss, Inorg. Chem. 2020, 59, 6087–6099.
[32] J. Mukherjee, I. Siewert, Eur. J. Inorg. Chem. 2020, 2020, 4319–
4333.
[49] J. D. Froehlich, C. P. Kubiak, J. Am. Chem. Soc. 2015, 137, 3565–
3573.
[50] L. Iffland, D. Siegmund, U. Apfel, Z. Anorg. Allg. Chem. 2020,
646, 746–753.
[33] H. Shirley, T. M. Sexton, N. P. Liyanage, M. A. Perkins, S. A.
Autry, L. E. McNamara, N. I. Hammer, S. R. Parkin, G. S.
[51] F. P. A. Johnson, M. W. George, F. Hartl, J. J. Turner, Organo-
metallics 1996, 15, 3374–3387.
[52] D. R. Case, A. Spear, A. F. Henwood, M. Nanao, S. Dampf, T. M.
Korter, T. Gunnlaugsson, J. Zubieta, R. P. Doyle, Dalton Trans.
2021, 50, 3479–3486.
[53] D. D. Perrin, W. L. F. Armarego, Purification of Laboratory
Chemicals, Pergamon Press, Oxford; New York, 1988.
[54] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7–13.
[55] G. M. Sheldrick, Acta Crystallogr. Sect. A 2015, 71, 3–8.
[56] G. M. Sheldrick, Acta Crystallogr. Sect. C 2015, 71, 3–8.
[57] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H.
Puschmann, J. Appl. Crystallogr. 2009, 42, 339–341.
Tschumper,
J. H.
Delcamp,
ChemPhotoChem
2021,
cptc.202000296.
[34] B. Merillas, E. Cuéllar, A. Diez-Varga, T. Torroba, G. García-
Herbosa, S. Fernández, J. Lloret-Fillol, J. M. Martín-Alvarez, D.
Miguel, F. Villafañe, Inorg. Chem. 2020, 59, 11152–11165.
[35] L. Suntrup, F. Stein, J. Klein, A. Wilting, F. G. L. Parlane, C. M.
Brown, J. Fiedler, C. P. Berlinguette, I. Siewert, B. Sarkar, Inorg.
Chem. 2020, 59, 4215–4227.
[36] J. Du, A. Wilting, I. Siewert, Chem. Eur. J. 2019, 25, 5555–5564.
[37] J. K. Nganga, L. M. Wolf, K. Mullick, E. Reinheimer, C. Saucedo,
M. E. Wilson, K. A. Grice, M. Z. Ertem, A. M. Angeles-Boza, Inorg.
Chem. 2021, 60, 3572–3584.
[38] M. Beley, J.-P. Collin, R. Ruppert, J.-P. Sauvage, J. Chem. Soc.
Chem. Commun. 1984, 1315.
[39] Marc. Beley, J. Paul Collin, Romain. Ruppert, J. Pierre Sauvage,
J. Am. Chem. Soc. 1986, 108, 7461–7467.
Manuscript received: December 8, 2020
Revised manuscript received: March 24, 2021
Accepted manuscript online: March 24, 2021
Z. Anorg. Allg. Chem. 2021, 968–977
www.zaac.wiley-vch.de
977 © 2021 The Authors. Zeitschrift für anorganische und allgemeine Chemie
published by Wiley-VCH GmbH.