Tuneable Redox Chemistry and Electrochromism of Persistent Symmetric and Asymmetric Azine Radical Cations

Article abstract of DOI:10.1002/chem.201902216

Molecular organic radicals have been intensively studied in the last decades, due to their interesting optical, magnetic and redox properties. Here we report the synthesis and characterisation of persistent organic radicals from one-electron oxidation of redox-active azines (RAAs), composed of two guanidinyl or related groups. By connecting two different groups together, asymmetric compounds result. In this way a series of compounds with varying redox potential is obtained that could be oxidised reversibly to the mono- and the dicationic charge states. The accessible redox states were fully determined by chemical redox reactions. The standard Gibbs free energy change for disproportionation of the radical monocation into the dication and the neutral molecule in solution, estimated from cyclovoltammetric measurements, varies between 43 and 71 kJ mol^?1. While the neutral RAAs absorb predominately UV light, the radical monocations display strong absorptions covering almost the entire visible region and extending for some compounds into the NIR region. A detailed analysis of this highly reversible electrochromism is presented, and the fast switching characteristics are demonstrated in an electrochromic test device.

Full text of DOI:10.1002/chem.201902216

A Journal of
Accepted Article
Title: Tuneable Redox-Chemistry and Electrochromism of Persistent
Symmetric and Asymmetric Azine Radical Cations
Authors: Marco Werr, Elisabeth Kaifer, Hubert Wadepohl, and Hans-
Jörg Himmel
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: Chem. Eur. J. 10.1002/chem.201902216
Link to VoR: http://dx.doi.org/10.1002/chem.201902216
Supported by

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
Tuneable Redox-Chemistry and Electrochromism of Persistent
Symmetric and Asymmetric Azine Radical Cations
Marco Werr, Elisabeth Kaifer, Hubert Wadepohl, Hans-Jörg Himmel*[a]
Abstract: Molecular organic radicals were intensively studied in the
last decades, due to their interesting optical, magnetic and redox
properties. Here we report the synthesis and characterization of
persistent organic radicals from one-electron oxidation of redox-active
azines (RAAs), composed of two guanidinyl or related groups. By
connecting two different groups together, asymmetric compounds
result. In this way a series of compounds with varying redox potential
is obtained, that could be oxidized reversibly to the mono- and the
dicationic charge states. The accessible redox states were fully
determined by chemical redox reactions. The standard Gibbs free
energy change for disproportionation of the radical monocation into
the dication and the neutral molecule in solution, estimated from
cyclovoltammetric measurements, varies between 43 and 71 kJ mol−1.
While the neutral RAAs absorb predominately UV light, the radical
monocations display strong absorptions covering almost the entire
visible region and extending for some compounds into the NIR region.
A detailed analysis of this highly reversible electrochromism is
presented, and the fast switching characteristics are demonstrated in
an electrochromic test device.
the oxidized states, was not undertaken. NMR data were also not
recorded. Most importantly, the redox reactivity of these
interesting species was not further assessed. The first urea azine
containing coordination compound was published 2007 by Villiers
et al. featuring a deprotonated hemi-alkylated urea azine forming
tetra- and hexanuclear uranium complexes.[5]
Introduction
The redox-active azine 2,2’-azino-bis(3-ethylbenzthiazoline-6-
sulphonic acid (ABTS, Figure 1) is a well-known compound that
found applications in food industry to measure the antioxidant
capacities, and in material science as a redox mediator in the
bioelectrocatalytic reduction of dioxygen to water.[1] Its redox
potential measures −0.035 V vs. Fc+/Fc at pH 4.0.[2]. Some
polymeric derivatives were also synthesized.[3] Besides, redox-
active azines (RAAs), that also comprise urea azines
Figure 1. Lewis representations of the redox-active azines (RAAs) 1 - 8 relevant
for this work as well as the derivative ABTS applied in biochemistry.
A few years ago, our group started to systematically analyze this
class of organic electron-donor compounds.[6-7] The new
compounds 1 - 3 (see Figure 1) were synthesized and their redox
potentials determined.[8] Compounds 1 and 2 were used as
ligands in zinc(II)[9], cobalt(II)[8] and copper(II)[10] complexes,
featuring a rather unusual coordination mode in which the metal
atom is coordinated by an imino nitrogen as well as an amino
nitrogen atom, resulting in a five-membered metallacycle. This
coordination mode is special since imine π-interaction in the
guanidine strongly reduces the Lewis basicity of the amino
nitrogen atoms. Single-crystal XRD and dynamic NMR studies[11]
of [1(ZnCl2)] showed that the amino nitrogen-metal bond is hemi-
labile, pointing to a special reactivity.
(compounds 1 - 5 in Figure 1), constitute
a very much
underdeveloped class of compounds. The first symmetric urea
azines were reported already more than 50 years ago by Hünig
et al.[4] The optical properties of a few compounds and their
Brønsted basicity were investigated. Furthermore, the redox
potentials and their pH dependency were determined by
polarography in an aqueous medium. On the other hand, no
structural characterization of the different oxidation states was
carried out, and a detailed analysis of the electronic structures in
[a]
M. Werr, Dr. E. Kaifer, Prof. H. Wadepohl, Prof. H.-J. Himmel
Anorganisch Chemisches Institut, Ruprecht-Karls-Universität
Heidelberg
Due to the low barrier for ligand-metal electron-transfer processes
and the resulting flexibility in the electronic structures of
guanidine-copper complexes,[12] CuII complexes of urea azines
were in the focus of recent work by our group.[10] Reaction of
copper halides CuX2 (X = Cl or Br) with 2 and 3 led not as
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
E-mail: hans-jorg.himmel@aci.uni-heidelberg.de
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
expected to complexation, but to ligand oxidation and subsequent
slow decomposition (Scheme 1a).
Scheme 1a. The radical 1-4 could be trapped by addition of
TEMPO.[10]
Furthermore, it was shown that one-electron oxidation of 2 with
AgBF4 is accompanied by dehydrogenation of an ethylene bridge
in the backbone to give (2-2H)+ (compound 2-2, Scheme 1b). The
isolation of 2·+ (compound 2-1) was finally possible by reaction
with TCNQ, yielding the donor-acceptor material 2(TCNQ)
(compound 1-3). On the other hand, oxidation of 2 with the oxidant
NOSbF6 led to formation of 22+ (compound 2-4, see Scheme 1b).
While the neutral compounds are colourless - pale yellow, the
oxidized compounds are intensively coloured.
The different redox states of redox-active organic compounds
often exhibit distinct electronic (UV-vis) absorption bands.[13] The
possibility to switch electrochemically between these redox-states
is then accompanied by electrochromism.[14] Such physical
responsiveness is widely investigated for switching and memory
devices,[15] e.g. reading a stored information by an optical
response (change in absorption spectra).[13a] Redox-active
molecules were also studied in the context of organic mixed
valence compounds.[16]
It should be added at this place that the oxidized RAAs are
(structurally) related to azoimidazolium dyes (such as Basic Red
51).[8] Moreover, N-heterocyclic carbene (NHC)-derived nitrogen-
centred radicals, formed by reduction of azoimidazolium dyes,
were recently reported by Severin and co-workers (see
Scheme 2).[17] Finally, azoimidazolium dyes were used as
precursors for “mesoionic carbene ligands” (Azo-MICs).[18]
Scheme 1. Hydrogen-atom transfer (HAT) (a) and dehydrogenation (b)
processes initiated by one-electron oxidation of the RAAs 1 and 2. Due to this
reactivity, radical monocations of 1 and 2 are not stable, despite of the reversible
redox events observed in the cyclovoltammetric (CV) curves.
On the other hand, the reaction of CuX2 with 1 (slightly higher
redox potential than 2 or 3) gave the corresponding complexes
[1(CuX2)] (X = Cl or Br, see Scheme 1a, compound 1-1). An
electron paramagnetic resonance (EPR) study for the compound
in the solid state showed that the spin-density is predominantly
located on the copper atom.[10] In solution, the copper complexes
slowly decomposed within minutes, a process that is initiated by
intramolecular electron transfer from the electron-rich urea azine
ligand to the CuII atom (intermediate 1-2). The electron transfer is
followed by decomplexation and formation of the free radical
cation 1·+ (species 1-3) together with a CuI counter-ion. Monitoring
the reaction via UV/Vis and EPR spectroscopy showed that the
decomposition kinetics depend on the solvent polarity and the
choice of halide co-ligands. The EPR signal of free 1·+
(species 1-3), with its spin density mainly located on the central
N-N unit, eventually vanishes. A hydrogen atom transfer (HAT)
takes place leading to a reactive carbon-centered radical
(intermediate 1-4), that subsequently takes up or loses a
hydrogen by reaction with itself (compounds 1-5 and 1-6).
Hydrogen loss combined with dimerization leads to formation of a
12-membered ring product (compound 1-7) as illustrated in
Scheme 2. Structural similarity between azoimidazolium dyes and aminyl
radicals with urea azine (RAA) derived radical cations. Recent complex
chemistry of azoimidazolium dyes shows their application as redox-active
mesoionic carbene ligands (Azo-MICs).
Furthermore, persistent organic radicals derived from NHCs
featuring sulfur atoms have been synthesized,[19] and cyclic
(alkyl)(amino)carbenes (CAACs) were used for the formation of
stable carbon centered radicals and diradicaloids.[20] Carbon-
centered classical NHC radicals have lately been reported by
Ghadwal et al., highlighting the key role of the substituent at the
C2 position of the 1,3-substituted imidazolium salt precursor in the
stabilization of the radical formed upon reduction.[21]. It was also
possible to prepare diradicaloids derived from classical NHCs.[22]
As part of our research in this area we seek to obtain persistent
radical monocations of RAAs by prohibiting HAT processes, and
to this end modified the urea azine backbone. In this work new
derivatives are presented and the characterization of others
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
completed. The redox properties of all known redox-active
symmetric and asymmetric azines 1 - 8 (see Figure 1) are
compared, and the distinct electrochromism studied.
compiled in Table 1. The central N-N bond and the two C=N imine
bond lengths fall in the ranges 1.405(3) - 1.429(2) Å and
1.294(2) - 1.306(2) Å, respectively. The introduction of sulphur
into the guanidino backbone (compounds 6 and 7) leads only to a
slight decrease in the C=N imine bond lengths due to the reduced
conjugation.
Results and Discussion
Synthesis and characterization. The Lewis representations of
all RAAs relevant for this work are displayed in Figure 1, while the
synthetic routes are illustrated in Scheme 3. Synthesis of the
symmetric azines 4 - 6 was carried out by reaction of the
corresponding chloro-formamidinium salt (Vilsmeier salt) with the
isothiouronium salt respectively with hydrazine hydrate. The
chloro-formamidinium precursors could readily be synthesized
from 2-chlorobenzimidazol by sequential reaction with
alkyliodides and a Meerwein reagent. The isothiouronium salt was
synthesized by reaction of a secondary amine with potassium
thiocyanate and cyclisation with elemental bromine. The
compounds were obtained after crystallization in isolated yields of
78% for 4, 33% for 5 and 28% for 6. The synthesis of the new
asymmetric RAAs 7 and 8 was carried out by reaction of 3-
4
6
5
methyl-2-benzothiazolinone-hydrazone
with
chloro-
5
7
formamidinium salts. The hydrazone was synthesized according
to a modified literature protocol from the isothiouronium salt and
hydrazine hydrate.[23] The compounds were isolated in pure form
in yields of 90% (7) and 81% (8) with respect to the hydazone
reactant.
(8+2H)Cl2
Figure 2. Molecular structures of 4 - 7 and (8+2H)Cl2 (only one of the two
independent molecules of 5 is shown). Hydrogen atoms omitted for clarity.
Displacement ellipsoids drawn at the 50% probability level. Selected bond
parameters are given in Table 1. In the case of 5, both independent molecules
in the unit cell are displayed.
In principal, isomerisation of the imine bond (E/Z isomer) might
result in the formation of different isomers in the case of the
asymmetric RAAs 7 and 8 (two possible isomers) as well as 6
(three possible isomers). Quantum-chemical calculations
(B3LYP/TZVP) for 6 and 7 showed that their energy differs at least
by 30 kJ mol−1 (see Figure 3).
Indeed, exclusively the lowest-energy isomer was present in the
solid state. The NMR spectra of 7 displayed three signals for three
non-equivalent methyl groups, arguing for the presence of only
Scheme 3. Reaction equations illustrating the synthetic routes to symmetric and
asymmetric RAAs. Conditions:
I (a) NaH, RI, DMF; (b) [R3O][BF4],
1,2-dichloroethane; (c) N2H5OH, NEt3, DCM; II (a) KSCN, HClaq; (b) Br2, CH3Cl;
(c) N2H5OH, EtOH; III (a) MeOTs, toluene; (b) N2H5OH, EtOH/H2O; IV (a) NEt3,
DCM. See SI for detailed information.
1
one isomer in solution. Furthermore, the H NMR spectra for 4
and 5 contain two equivalent signal sets for the methyl and ethyl
groups, respectively. The obvious inference is that 4 and 5 are
also present in a single isomeric form in solution. A variable-
temperature 1H NMR study of 4 in DMF gave no sign for
coalescence at higher temperature (see SI), indicating that
isomerisation of the urea azines does not occur in the temperature
range from r.t. to 100 °C. Un-catalysed imine isomerisation is
known to procced either via inversion or rotation and can be
stimulated thermally or by light irradiation.[24-28]
At standard conditions, all studied RAAs are colorless to beige
solids except of 8 which is a pale-yellow oil. Compounds 6 - 8 are
air-stable as solids. By contrast, compounds 4 and 5 are quickly
oxidized by air. Structural characterization of RAAs 1 - 7 and the
hydrochloride of 8 in the solid state was achieved by single-crystal
X-ray diffraction (XRD). Figure 2 illustrates the resulting
structures, and selected structural data of compounds 1 - 8 are
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
Table 1. Selected experimentally determined bond lengths (in Å) for all studied RAAs.
RAA
C=N
C=N
NN
C-N1
C-N2
C-N3
C-N4
1
1.306(2) 1.306(2) 1.415(2) 1.371(2) 1.403(2) 1.371(2) 1.403(2)
1.296(1) 1.296(1) 1.416(1) 1.394(1) 1.398(1) 1.394(1) 1.398(1)
1.295(1) 1.295(1) 1.427(2) 1.374(1) 1.403(1) 1.374(1) 1.403(1)
1.295(2) 1.295(2) 1.407(2) 1.400(2) 1.392(2) 1.400(2) 1.392(2)
1.301(2) 1.300(2) 1.407(2) 1.394(2) 1.386(2) 1.391(2) 1.393(1)
2
3
4
5a
RAA
6
C=N
C=N
NN
C-N
C-S
C-N
C-S
1.238(2) 1.238(2) 1.409(2) 1.381(2) 1.768(1) 1.381(2) 1.768(1)
C=NA C=NB NN C-NA1 C-NA2 C-NB C-S
7
1.296(2) 1.286(2) 1.413(2) 1.396(2) 1.391(2) 1.381(2) 1.767(2)
(8+2H)Cl2 1.364(2) 1.330(2) 1.389(2) 1.336(2) 1.331(2) 1.328(2) 1.729(2)
a For 5, the asymmetric unit contains two independent molecules with slightly different
bonding parameters, the values of one isomer are given.
Thermal isomerisation of hydrazine-based imines is generally
Redox properties.
associated with a high activation barrier, leading to coalescence
temperatures above 200 °C.[29] Isomerisation by light, for which
Cyclic voltammetry (CV). CV curves were recorded for
compounds 4 - 8 (see Figure 4 and Table 2) to compare the redox
interconversion occurs in the excited state, was also not observed. potentials of the RAAs. The redox potentials for compounds 1 - 3
TD-DFT calculations and the inspection of the frontier orbitals of
the neutral RAAs suggest that the HOMO-LUMO excitation might
favour imine isomerisation through the shift of electron-density
from the π-imine (C=N) orbitals into the π-orbital of the central
azine N-N unit (see SI). However, fast thermal relaxation might
immediately re-establish the initial configuration at ambient
temperature.[25] Since this issue is outside the scope of this work,
it is not further pursued herein.
were taken from the literature.[8] All compounds are oxidized in
two reversible, clearly separated one-electron redox steps,
indicating that the radical monocations are thermodynamically
stable towards disproportionation. Compound 4 exhibits the
lowest redox potential (E1/2 = −0.57 V vs. Fc+/Fc for the redox
couple 4·+/4), and compound 6 the highest one (E1/2 = +0.07 V vs.
Fc+/Fc for the redox couple 6·+/6). Moreover, the potential is lower
for aromatic RAAs (compounds 4 - 6) than for non-aromatic ones
(compounds 1 and 2), and significantly rises upon replacing an
NR group by sulfur. It can also be seen that structural
modifications in the backbone of the asymmetric compounds 7
and 8 slightly affect the redox potential of the first wave while the
second is almost unaffected. The redox potential variations are
generally larger for the first redox process(RAA·+/RAA) than for
the second (RAA2+/RAA·+).
70
60
50
40
30
20
10
0
Figure 3. Relative energies of the isomers of compounds 7 (left) and 6 (right)
from B3LYP/TZVP calculations.
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
Table 2. Redox potentials (vs. Fc+/Fc) for compounds 1 - 8 in CH2Cl2 solution
[Ag/AgCl reference electrode, scan speed 100 mV s−1, 0.1 M N(nBu)4(PF6) as
supporting electrolyte]. The DG0 and Kdisp values for disproportionation of the
radical monocation, as estimated from the potential differences, are also
a)
20
4
0
-10
10
given.All E values given in V, DG0 in kJ mol−1
RAA·+/RAA RAA2+/RAA·+
Eox Ered Eox Ered
.
20
6
10
0
-10
E1/2
E1/2
ΔE1/2 ΔG0 Kdisp
30
7
15
1
2
3
4
5
6
7
8
−0.22 −0.35 −0.29 0.22 0.10 0.16 0.45
−0.34 −0.46 −0.40 0.28 0.15 0.22 0.62
−0.42 −0.57 −0.50 0.22 0.06 0.14 0.64
−0.50 −0.61 −0.57 0.22 0.12 0.13 0.70
−0.49 −0.65 −0.56 0.20 0.07 0.17 0.73
43.4 2.5∙10−8
59.8 3.3∙10−11
61.8 1.5∙10−11
67.5 1.5∙10−12
70.4 4.5∙10−13
57.0 1.0∙10−10
63.0 9.1∙10−12
55.0 2.3∙10−10
0
-15
30
8
15
0
-15
-1.0
-0.5
0.0
0.5
1.0
E / V
0.10
0.03
0.07
0.72 0.60 0.66 0.59
b)
−0.18 −0.30 −0.24 0.48 0.35 0.42 0.65
−0.03 −0.29 −0.16 0.55 0.28 0.41 0.57
80
40
0
scan rate
50 mV/s
100 mV/s
200 mV/s
500 mV/s
The difference DE1/2 between the first and second redox waves is
sensitive to modifications in the backbone. As one can see, the
separation is generally larger for RAAs with heterocyclic rings (2
and 4 - 7) than for acyclic ones (1 and 8). The DE1/2 values are
directly connected with the Gibbs free energy change and the
equilibrium constant Kdisp for disproportionation of RAA·+ into
RAA2+ and RAA. With the help of the equations
-40
-80
퐹
DG0 = F·DE1/2 and 퐾disp = 푒푥푝 (− Δ퐸1/2
)
푅푇
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
0.5
1.0
1.5
E / V
the DG0 and Kdisp values for disproportionation can be estimated.
All DG0 values are significantly positive (43.4 - 70.4 kJ mol−1),
indicating endergonic processes (see Table 2). The small Kdisp
values indicate that the degree of disproportionation is negligible
at ambient conditions. The radical 5·+ exhibits the highest DG0
value of 70.4 kJ mol−1.
Figure 4. a) CV curves recorded for 4 and 6 - 8 in CH2Cl2 solutions [Ag/AgCl
reference electrode, scan speed 100 mV s-1, 0.1 M N(nBu)4(PF6) as supporting
electrolyte]. Potentials given vs. ferrocenium/ferrocene (Fc+/Fc). b) CV curves
measured for 5 in CH2Cl2 solution for several scan rates.
From Figure 4b one can see that the peak cathodic current for the
reduction of 52+ to 5·+ decreases with decreasing scan rate,
eventually vanishing for a scan rate of 50 mV s−1. Simultaneously,
the peak cathodic current for reduction of 5·+ to neutral 5
increases, reaching at 50 mV s-1 almost twice the value of the
corresponding peak anodic current. This behaviour can be
rationalised by comproportionation to 5·+, enabled by diffusion of
significant amounts of neutral 5 into the diffusion layer at low scan
rates (Scheme 4). Consequently, the reduction current decreases
with decreasing scan rate as there is less 52+ left for
heterogeneous electrode reduction in the diffusion layer while the
concentration of 5·+ is probably almost doubled due to the
homogenous comproportionation process. Consequently, the
current for reduction of 5·+ to 5 increases. The effect is strongest
for compound 5, in line with its extremely low Kdisp value.
The E1/2 values for the asymmetric RAAs 7 and 8 are in between
those of the corresponding symmetric RAAs. A closer inspection
suggests that the redox potential of an asymmetric RAA
composed of two groups A-B (linked through the central N-N
bond) is comparable to the mean value of the redox potentials of
the corresponding symmetric RAAs A-A and B-B, allowing a fine-
tuning of the redox potentials by rational synthesis of asymmetric
RAAs. In the case of compound 7, the predicted redox potential
([E1/2(5) + E1/2(6)]/2 = −0.245 V) is very close to the experimental
value (−0.24 V). For compound 8 the difference is slightly larger
(predicted [E1/2(1) + E1/2(6)]/2 = −0.06 V, experimentally obtained
−0.16 V).
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
Chemical redox reactions. Having studied the redox properties
by CV measurements, we next carried out some chemical redox
reactions to obtain more information about the oxidized states.
First the RAAs were integrated in donor-acceptor materials,
where the differences in the redox potentials should lead to
different degrees of charge-transfer. TCNQ was chosen as
acceptor component. The solid state structure of compound
2(TCNQ) was already reported,[10] and shown to be composed of
mixed chains of 2·+ radical cations and TCNQ·− radical anions.
Reaction of compounds 4, 7 and 8 with TCNQ led to the new
donor-acceptor materials 4(TCNQ), 7(TCNQ) and 8(TCNQ). All
three materials were structurally characterized (see Figure 5 and
SI). Despite of the same donor:acceptor ratio, the structures and
consequently the degree of charge transfer are different (vide
infra). In compounds 4(TCNQ) and 8(TCNQ), single donor and
acceptor units are alternately arranged within the mixed chains.
By contrast, pairs of donors respectively acceptors build the
mixed chains in 7(TCNQ).
Scheme 4. Coupling of the heterogeneous electrode reactions with
homogeneous comproportionation, explaining the variations of the CV curves
with the scan rate. RAABulk denotes an RAA molecule from the bulk of the
solution (outside the diffusion layer).
Quantum-chemical calculations (B3LYP/TZVP) show that the
HOMO of the neutral RAAs is predominantly located on the
bisguandine unit; thus modifications in this part should be most
effective. The introduction of sulphur (6 - 8) with its mismatching
orbital energies and shapes leads to less electron-rich azines.
This trend gets also evident from the calculated first adiabatic
ionisation energies, showing higher values for the thiourea azines
a)
(Table 3).
A linear relationship was found between the
experimentally determined E1/2 values of the redox couples
RAA·+/RAA and the calculated first (adiabatic) ionization energies
I1 (see SI). On the other hand, E(HOMO) and E1/2 do not correlate
in such a simple manner.
4(TCNQ)
Table 3. Comparison between the experimental redox potentials E1/2 (first one-
electron oxidation, vs. Fc+/Fc) and calculated HOMO energies E(HOMO), first
adiabatic ionisation energies I1 and absolute (adiabatic) proton affinities PA
(B3LYP/TZVP, 298.15 K) for 1 - 8.
1
RAA
E1/2 / V
−0.57
−0.56
−0.50
−0.40
−0.29
−0.24
−0.16
+0.07
E(HOMO) / eV I1 / eV
PA / kJ mol−
1040
7(TCNQ)
8(TCNQ)
4
5
3
2
1
7
8
6
−4.16
−4.23
−4.76
−4.42
−4.66
−4.49
−4.68
−4.75
5.29
5.35
5.29
5.46
5.63
5.64
5.66
5.88
b)
1033
1047
1061
1076
Figure 5. a) Packing of the mixed donor-acceptor chains in the donor-acceptor
materials 4(TCNQ), 7(TCNQ) and 8(TCNQ). b) Empirical formula to estimate
the partial charge on the TCNQ units.
1024
1045
997
According to Kistenmacher et al., the partial charge 휉 (in e) on the
TCNQ units can be estimated from the experimentally derived
bond lengths b, c and d (see Figure 5b).[31] Using this formula, a
partial charge of −0.98 was derived for the TCNQ units in
4(TCNQ). By contrast, in 7(TCNQ) the partial charge is only −0.44
and in 8(TCNQ) it is further reduced to −0.18. Chappell et al.
established a simple linear relationship between the wavenumber
of the TCNQ nitrile stretching modes, (CN), and the partial
charge 휉 (see Figure 5b).[32] Using this formula, a partial charge
of −1.11 results for 4(TCNQ), shrinking to −0.52 for 7(TCNQ) and
−0.34 for 8(TCNQ). Hence the values derived from both formula
are in pleasing agreement. Moreover, the trend in the degree of
The calculated absolute proton affinities (PA) are higher than
those of amines (e.g. 981.8 kJ mol−1 for triethylamine) and
comparable with the PA values of alkyl-substituted guanidines
(e.g. pentamethylguanidine 1047.7 kJ mol−1).[30] Overall the non-
aromatic RAAs 1 - 3 exhibit the highest PA values. The PA values
of the aromatic urea azines 4 and 5 are slightly lower, while the
twofold sulphur-substituted derivative 6 exhibits the lowest PA
value. This trend is in line with the pKs values in aqueous media
for 4 (pKs = 7.5) and 6 (pKs = 2.5).[4a]
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
charge-transfer is fully consistent with the trend of the redox
potentials E1/2 (4 (−0.57 V) < 7 (−0.24 V) < 8 (−0.16 V)). Hence
the results demonstrate that the differences in the redox
properties could be applied to fine-tune the degree of
charge-transfer in donor-acceptor materials.
“local donor strength” (and the proton affinity) is reduced due to
the condensed aromatic core, but the enlarged π-system
increases the overall donor strength. The high donor strength of
4 favors electron-transfer to the CuII atom. From Table 4 one can
see that the cations in 4(CuCl2) and 4(TCNQ) exhibit similar
structures, confirming the high degree of charge transfer in the
donor-acceptor material.
a)
b)
Scheme 5. Reactions leading to one- and two-electron oxidation of 4.
Two-electron oxidation of 4 with AgBF4 in THF solution yielded
the salt 4(BF4)2 as bright red powder (Scheme 5). Crystals of this
salt were grown from a DMF/H2O solvent mixture (Figure 6a).
With respect to the neutral molecule, the central azine bond is
shortened from 1.381 Å in 4 to 1.272 Å in 42+ (see Table 4),
indicating the formation of an N=N double bond upon two-electron
oxidation (cf. N-N single bond distance in hydrazine of 1.450 Å,
and double bond distance in diazene of 1.250 Å).[33] While the
central N-N bond length decreases, the former C-N imine bonds
are elongated upon oxidation, in line with the expectations from
the Lewis structures.
Figure 6. Illustration of the structures of 4(BF4)2 (a) and 4(CuCl2) (b). Hydrogen
atoms omitted for clarity. Displacement ellipsoids drawn at the 50% probability
level. Selected bond parameters are given in Table 4.
Reaction of 7 with CuCl2 in THF led to a dark-blue solution,
indicating redox reaction (as for 4 and CuCl2). The salt 7(PF6) was
obtained as the product of the reaction between 7 and AgPF6
(structure illustrated in the SI). The presence of an unpaired
electron was proven by EPR spectroscopy (see SI for more
information). Comparing its critical structural parameters (CN and
NN bonds of RAA system) with 7 and 7(TCNQ) further supports
partial charge-transfer in the donor-acceptor compound. In the
series from 7 to 7(TCNQ) and to 7(PF6) one observes elongation
of the former CN-imine bonds and shortening of the NN-azine
bond with values in-between for 7(TCNQ).
Reaction of 4 with CuCl2 did not yield a coordination compound.
Instead blue crystals were isolated, composed of non-coordinated
−
CuCl2 anions and 4·+ radical cations as a result of a redox
reaction (Scheme 5 and Figure 6b). In difference to the results
obtained with compounds 1 and 2, a chelating coordination of 4 is
hampered by the unfavorable orientation of the lone pairs at the
amino N atoms (planar ligand backbone) and the reduction of the
Lewis basicity by π-interactions with the aromatic core. Hence the
Table 4. Selected experimentally determined bond lengths (in Å) for different redox states of 4 and 7.
RAA
4
CNA
CNB
NN
CN1
CN2
CN3
CN4
1.295(2) 1.295(2) 1.407(2) 1.400(2) 1.392(2) 1.400(2) 1.392(2)
4(CuCl2) 1.335(2) 1.335(2) 1.344(2) 1.368(2) 1.368(2) 1.368(2) 1.368(2)
4(TCNQ) 1.338(2) 1.338(2) 1.347(2) 1.369(2) 1.368(2) 1.369(2) 1.368(2)
4(BF4)2
RAA
7
1.381(2) 1.381(2) 1.272(2) 1.352(2) 1.349(2) 1.352(2) 1.349(2)
C=NA C=NB NN C-NA1 C-NA2 C-NB C-S
1.296(2) 1.286(2) 1.413(2) 1.396(2) 1.391(2) 1.381(2) 1.767(2)
7(TCNQ) 1.309(2) 1.295(2) 1.383(2) 1.383(2) 1.379(2) 1.368(2) 1.759(2)
7(PF6)
1.333(3) 1.333(3) 1.339(3) 1.374(3) 1.362(3) 1.351(3) 1.740(3)
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
Reaction of 8 with CuCl2 in THF resulted in a purple solution
indicating redox reaction. Interestingly, solutions of 8 and CuCl2 in
CH3CN bleached with time, also reaction of 8 with AgPF6 yielded
a purple solution that bleached within several days. Colourless
crystals grown from a MeCN/Et2O solution of 8 and CuCl2 were
identified as salt (8+H)(CuCl2). The formation of the reduced,
protonated RAA compound indicates decomposition initiated by
intramolecular HAT (as observed earlier for compound 1, see
Scheme 1).[10]
the potential to values in between +0.5 V and +1.5 V led to the
decrease of the bands due to 4·+ and appearance of a broad band
around 520 nm assigned to 42+. Finally, after the external potential
had been switched off, an intensified spectrum of 4·+ was
measured, indicating fast comproportionation of 42+ with 4, in line
with the CV results.
Spectro-electrochemical measurements for 6 and 7 gave similar
results. At low potentials the neutral form dominated, giving rise
to a band in the UV region (338 nm for 6 and 344 nm for 7).
Oxidation to the radical monocations at higher potentials led to a
red shift of the UV band (to ca. 400 nm for 6·+ and 389 nm for 7·+)
and the appearance of broad absorptions in the Vis-NIR region
(with maxima at 650 nm, 740 nm and 830 nm for 6·+ and 605, 680
and 760 nm for 7·+, see Figure 8). The spectrum received for
7(PF6) dissolved in CH2Cl2 showed high extinction coefficients
above 10000 L cm mol−1 at all wavelengths within the range
600 - 800 nm (see SI). With increasing redox potential of the RAA,
the low energy bands of the radical cationic forms are
bathochromic shifted (Figure 8).
Electrochromism. The discussion in the previous section
showed that the introduction of a phenylene group in the
backbone prohibits the HAT processes, leading to persistent
radical cationic RAAs. Compounds 4, 6 and 7 were therefore
chosen to study the electrochromism in more detail. Figure 7
shows UV/Vis spectra of 4 dissolved in CH2Cl2. The neutral
compound (pale-yellow) absorbs predominantly in the UV region,
with a strong band at 366 nm. However, with time the solution
changed its colour to slight purple-blue, indicating oxidation.[34] In
the UV/Vis spectra, the band at 366 nm shifted to 389 nm, and
broad absorptions covering almost the entire visible region, with
maxima at 521, 567, 631, and 704 nm, growed in (Figure 7). An
isosbestic point at 380 nm indicated clean oxidation. Prolonged
exposure to air did not lead to further oxidation to the dication,
showing that the radical monocation 4·+ is stable/persistent under
ambient conditions.
35
0 min
41 min
93 min
30
199 min
247 min
4
7
6
25
300 400 500 600 700 800 900 1000
20
15
10
5
l / nm
Figure 8. UV/Vis spectra of 4·+, 6·+ and 7·+ in CH2Cl2 derived from UV/Vis
spectroscopy. For 6·+ and 7·+, difference spectra obtained during the spectro-
electrochemical study are shown (spectra at +0.58 V for 6·+ respectively +0.48 V
for 7·+ minus spectra obtained at the beginning for the neutral compound).
0
300
400
500
600
700
800
900
l / nm
Additional spectro-electrochemical measurements for 8 disclosed
some differences. The neutral form, present at low potentials,
again gave rise to a band in the UV region (at 330 nm), shifting to
380 nm upon oxidation to 8·+. Radical cation 8·+ exhibits only one
broad band in the Vis region (around 550 nm). At higher potentials
further oxidation to 82+ occurred and a band at 410 nm formed
while all other bands vanished. In difference to the spectro-
electrochemical measurements for the other compounds, a
decrease of all absorptions was observed after several spectro-
electrochemical cycles, indicating degradation via HAT processes
of the radical cationic form, in line with the results of the chemical
redox reactions.
Figure 7. UV/Vis spectra of 4 (c = 2.04·10−5 M, d = 1 cm) in CH2Cl2 solution,
indicating its oxidation with time by dioxygen. The photo shows the sample after
247 min. No further changes were observed upon prolonged exposure to air,
showing that the radical monocation is air stable.
Electrochemical oxidation in combination with UV/Vis
spectroscopy (denoted spectro-electrochemistry in the following)
provided further information (see SI for details). UV/Vis spectra
measured at a negative potential of −1.3 V indicated the exclusive
presence of the neutral form, 4. Upon increasing the applied
potential to values in between −0.3 V and −0.2 V the strong band
at 366 nm due to 4 shifted to 389 nm and several bands in the
visible appeared, indicating formation of 4·+. Further increase of
TD-DFT calculations (B3LYP/def2-TZVP) gave additional
information about the orbitals involved in the electronic excitations.
For neutral 4, the intense UV band at 366 nm can be assigned to
a pure HOMO-LUMO excitation. The transition at 389 nm in the
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
spectrum of 4·+ involves similar orbitals (SOMO-LUMO excitation).
In addition, the calculations on 4·+ found two electronic transitions
in the Vis region at 549 nm and 665 nm, that belong to excitations
from occupied lower-energy orbitals into the SOMO (see SI for
details). The structure of the experimentally obtained broad
absorption with several maxima is therefore partially caused by a
vibration progression. Hence the energy difference between the
704 and 631 nm bands of 1643 cm−1 fits to vibrational progression
due to the (symmetric) imine stretching mode (CN). For
comparison, the asymmetric imine stretching vibration of neutral
4 in its electronic ground state showed at 1595 cm−1 (while
calculations predicted wavenumbers of 1688 and 1652 cm−1 for
the symmetric and asymmetric (CN) modes). Finally, TD-DFT
calculations for 42+ found an intense transition at 532 nm (close to
the observed band at 520 nm) due to excitation from the HOMO
and another lower-energy orbital into the LUMO, resembling the
SOMO in 4·+ and the HOMO in neutral 4). The results of similar
calculations for the different redox states of 6 and 7 are included
in the SI. In all cases, the simulations based on the TD-DFT
calculations are in good agreement with the experimental spectra.
a)
−
RAA
0
RAA∙+
RAA2+
+
b)
−
−
−
+
+
+
+
0,6
0,4
450.5 nm
558.9 nm
620.4 nm
0,2
0,0
-0,2
-0,4
0
50
100
t / s
150
200
rel. int.
0,70
800
700
600
500
400
300
0,42
0,14
Electrochromic test device. Finally, an electrochromic solute-
based cell was assembled under ambient conditions to
demonstrate the electrochromic properties, using a glass-ITO-
parylen C-ITO-glass setup, (see SI for further details). The cell
was soaked with a solution of 4(PF6) (10 mg mL−1) and NBu4PF6
in CH3CN and filled through capillary forces. Starting with the blue
radical 4∙+ as reference state, the colour changed to red within a
few seconds when a positive potential was applied, indicating
oxidation to the dicationic form 42+ (Figure 9a). Switching the
potential back close to zero restored the blue colour of 4∙+ within
seconds. The application of a negative potential let to decoloration,
indicating reduction to neutral 4. The blue colour was restored
upon switching the potential back to zero. As seen from Figure 9b
the electrochromic behaviour was completely reversible,
confirming the stability of the compound in all three redox states.
The fast change of colour upon potential change (e.g. starting at
the resting state 90% of the full absorption change at 450.5 nm is
achieved within ca. 9 s; switching directly form +3 V to −3 V
requires ca. 13 s to achieve 90% of maximum absorption change
and switching from −3 to +3 V takes ca. 10 s, respectively)
-0,14
-0,42
-0,70
l
0
50
100
t / s
150
200
Figure 9. a) Device for electrochromic measurements using 4(PF6) dissolved in
CH3CN (10 mg mL−1, 0.5 M Bu4NPF6), a picture with brezel and beer was
chosen as background. b) Absorbance at three characteristic wavelengths
(450.5, 558.9 and 620.4 nm) in dependence of the applied potential (varied
periodically with time) for 4(PF6) (5 mg mL−1) dissolved in CH3CN (with 0.5 M
Bu4NPF6). In each cycle, the potential was switched from +3 V to −3 V and back.
DA values are given relative to the devise with the radical-form (and
solvent + electrolyte).
Conclusions
Despite of their interesting redox chemistry and unprecedented
electrochromism, redox-active azines constitute a relatively
undeveloped class of compounds. In this work, the synthesis,
redox-properties, stabilities and electrochromism of redox-active
azines (RAAs) are systematically studied. It is shown that
symmetric as well as asymmetric redox-active azines could be
rationally synthesized via two routes, providing convenient access
to compounds with varying redox potentials. They could be
reversibly oxidized in two separated one-electron steps first to the
radical monocation and then to the dication. The radical
monocations, displaying broad and intense electronic transitions
in the visible region, are persistent if hydrogen-atom transfer
(HAT) processes are excluded by careful design of the
backbone.[37] The redox-states were not only generated
electrochemically, but also characterized in several products of
redox-reactions and in donor-acceptor materials. The analysis is
complemented by quantum-chemical calculations. Finally, the
distinct electrochromism and the possibility to switch with the
motivates the application of RAAs in electrochromic devices.[13b,
13c, 35, 36]
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
[14]
[15]
a) P. M. S. Monk, R. J. Mortimer, D. R. Rosseinsky,
Electrochromism : fundamentals and applications, VCH, Weinheim
[u.a.], 1995; b) Electrochromic materials and devices (Ed.: Editior),
Wiley-VCH, Weinheim, 2015; c) R. J. Mortimer, Electrochim. Acta
1999, 44, 2971-2981; d) R. J. Mortimer, A. L. Dyer, J. R. Reynolds,
Displays 2006, 27, 2-18; e) R. J. Mortimer, T. S. Varley, Chem.
Mater. 2011, 23, 4077-4082.
a) V. Balzani, A. Credi, F. M. Raymo, J. F. Stoddart, Angew. Chem.
Int. Ed. 2000, 39, 3348-3391; b) V. Balzani, A. Credi, F. M. Raymo,
J. F. Stoddart, Angew. Chem. 2000, 112, 3484-3530; c) P. Audebert,
F. Miomandre, Chemical Science 2013, 4, 575-584; d) I.
Pochorovski, F. Diederich, Acc. Chem. Res. 2014, 47, 2096-2105;
e) M. Irie, T. Fukaminato, K. Matsuda, S. Kobatake, Chem. Rev.
2014, 114, 12174-12277; f) J. Nagasaki, S. Hiroto, H. Shinokubo,
Chemistry – An Asian Journal 2017, 12, 2311-2317.
potential quickly and reversibly between the redox states is
demonstrated in a test device.
Experimental Section
General experimental details, the synthesis, protocols and analytical data
for all compounds, details of the spectro-electrochemical studies, results
of the quantum-chemical calculations, and
a description of the
electrochromic test device are provided in the Supporting Information.
[16]
[17]
a) J. Hankache, O. S. Wenger, Chem. Rev. 2011, 111, 5138-5178;
b) S. F. Nelsen, R. F. Ismagilov, D. R. Powell, J. Am. Chem. Soc.
1996, 118, 6313-6314; c) F. Wudl, D. Wobschall, E. J. Hufnagel, J.
Am. Chem. Soc. 1972, 94, 670-672.
a) L. Y. M. Eymann, A. G. Tskhovrebov, A. Sienkiewicz, J. L. Bila, I.
Živković, H. M. Rønnow, M. D. Wodrich, L. Vannay, C. Corminboeuf,
P. Pattison, E. Solari, R. Scopelliti, K. Severin, J. Am. Chem. Soc.
2016, 138, 15126-15129; b) Y. Kim, E. Lee, Chem. Eur. J. 2018, 24,
19110-19121.
Acknowledgements
The authors thank Dr. Jan Freudenberg (InnovationLab,
Heidelberg) for the donation of electrochromic devices and the
Deutsche Forschungsgemeinschaft (DFG) for financial support.
[18]
[19]
[20]
a) F. M. Chadwick, B. F. E. Curchod, R. Scopelliti, F. Fadaei Tirani,
E. Solari, K. Severin, Angew. Chem. Int. Ed. 2019, 58, 1764-1767;
Angew. Chem. 2019, 131, 1778-1781.
Y. Wang, H. P. Hickox, Y. Xie, P. Wei, S. A. Blair, M. K. Johnson, H.
F. Schaefer, G. H. Robinson, J. Am. Chem. Soc. 2017, 139, 6859-
6862.
a) J. K. Mahoney, D. Martin, C. E. Moore, A. L. Rheingold, G.
Bertrand, J. Am. Chem. Soc. 2013, 135, 18766-18769; b) J. K.
Mahoney, D. Martin, F. Thomas, C. E. Moore, A. L. Rheingold, G.
Bertrand, J. Am. Chem. Soc. 2015, 137, 7519-7525; c) M. M.
Hansmann, M. Melaimi, D. Munz, G. Bertrand, J. Am. Chem. Soc.
2018, 140, 2546-2554; d) M. M. Hansmann, M. Melaimi, G. Bertrand,
J. Am. Chem. Soc. 2018, 140, 2206-2213.
Keywords: organic radicals, azines, organic electron donors,
electrochromism, oxidation
[1]
a) I. V. Shelepin, O. A. Ushakov, N. I. Karpova, V. A. Barachevskii,
Elektrokhimiya 1977, 13, 32; b) R. B. Walker, J. D. Everette, J. Agric.
Food. Chem. 2009, 57, 1156-1161; c) J. Schreiber, R. Lachenicht,
Clin. Chem. Lab. Med. 1973, 11, 31; d) G. T. R. Palmore, H.-H. Kim,
J. Electroanal. Chem. 1999, 464, 110-117.
[2]
R. Bourbonnais, D. Leech, M. G. Paice, Biochim. Biophys. Acta
1998, 1379, 381-390.
[3]
[4]
J. Fei, A. Basu, F. Xue, G. T. R. Palmore, Org. Lett. 2006, 8, 3-6.
a) S. Hünig, H. Balli, H. Conrad, A. Schott, Liebigs Ann. Chem. 1964,
676, 36-51; b) S. Hünig, G. Kiesslich, F. Linhart, H. Schlaf, Liebigs
Ann. Chem. 1971, 752, 182-195.
C. Villiers, P. Thuery, M. Ephritikhine, Chem. Commun. 2007, 2832-
2834.
For some review articles on strong organic electron-donors, see:
a) M. Á. Herranz, L. Sánchez, N. Martín, Phosphorus, Sulfur, and
Silicon and the Related Elements 2005, 180, 1133-1148; b) J. A.
Murphy, in Encyclopedia of Radicals in Chemistry, Biology and
[21]
[22]
a) D. Rottschäfer, B. Neumann, H.-G. Stammler, M. van Gastel, D.
M. Andrada, R. S. Ghadwal, Angew. Chem. 2018, 130, 4855-4858;
Angew. Chem. Int. Ed. 2018, 57, 4765-4768.
a) D. Rottschäfer, N. K. T. Ho, B. Neumann, H.-G. Stammler, M. van
Gastel, D. M. Andrada, R. S. Ghadwal, Angew. Chem. 2018, 130,
5940-5944; Angew. Chem. Int. Ed. 2018, 57, 5838-5842; b) D.
Rottschäfer, B. Neumann, H.-G. Stammler, D. M. Andrada, R. S.
Ghadwal, Chemical Science 2018, 9, 4970-4976; c) D. Rottschäfer,
J. Busch, B. Neumann, H.-G. Stammler, M. van Gastel, R. Kishi, M.
Nakano, R. S. Ghadwal, Chem. Eur. J. 2018, 24, 16537-16542.
J. Chen, X. Zhou, T. Yuan, Synth. Commun. 2014, 44, 3630–3636.
S. M. Landge, E. Tkatchouk, D. Benítez, D. A. Lanfranchi, M.
Elhabiri, W. A. Goddard, I. Aprahamian, J. Am. Chem. Soc. 2011,
133, 9812-9823.
A. Padwa, Chem. Rev. 1977, 77, 37-68.
K. Akio, K. Daiki, M. Tomoo, A. Tooru, A. Nobuyuki, S. Kazuhiko,
Chem. Lett. 2010, 39, 230-231.
A. Ida, B. Cohen, T. Asaka, A. Kawai, J. A. Organero, K. Shibuya,
A. Douhal, Phys. Chem. Chem. Phys. 2011, 13, 20318-20325.
T. Yoshida, T. Monji, D. Kawamori, N. Akai, K. Shibuya, A. Kawai,
Chem. Lett. 2013, 42, 1490-1492.
C. I. Stassinopoulou, C. Zioudrou, G. J. Karabatsos, Tetrahedron
1976, 32, 1147-1151.
E. P. L. Hunter, S. G. Lias, J. Phys. Chem. Ref. Data 1998, 27, 413-
656.
T. J. Kistenmacher, T. J. Emge, A. N. Bloch, D. O. Cowan, Acta
Cryst. B 1982, 38, 1193-1199.
J. S. Chappell, A. N. Bloch, W. A. Bryden, M. Maxfield, T. O. Poehler,
D. O. Cowan, J. Am. Chem. Soc. 1981, 103, 2442-2443.
A. F. Holleman, E. Wiberg, Lehrbuch der anorganischen Chemie,
102. ed., de Gruyter, Berlin, 2007.
a) S. Hünig, G. Kieβlich, F. Linhart, H. Schlaf, Liebigs Ann. Chem.
1971, 752, 196-205; b) S. Hünig, H. Balli, H. Conrad, A. Schott,
Liebigs Ann. Chem. 1964, 676, 52-65.
D. Weng, Y. Shi, J. Zheng, C. Xu, Org. Electron. 2016, 34, 139-145.
S. N. Intorp, S. Kushida, E. Dmitrieva, A. A. Popov, F. Rominger, J.
Freudenberg, F. Hinkel, U. H. F. Bunz, Chem. Eur. J. 2019, 25,
5412-5415.
For review articles on other persistent organic radicals, see:
a) R. G. Hicks, Org. Biomol. Chem. 2007, 5, 1321-1338; b) R. G.
Hicks, Stable radicals: Fundamentals and Applied Aspects of Odd-
Electron Compounds, Wiley, Chichester, 2010; c) Y. Morita, S.
Suzuki, K. Sato, T. Takui, Nature Chem. 2011, 3, 197; d) S. Sanvito,
[5]
[6]
Materials, Vol.
2 (Eds.: C. Chatgilialoglu, A. Studer), Wiley,
[23]
[24]
Chichester, 2012, pp. 817-847; c) S. Zhou, H. Farwaha, J. A.
Murphy, Chimia 2012, 66, 418-424; d) N. Martín, Chem. Commun.
2013, 49, 7025-7027; e) E. Doni, J. A. Murphy, Chem. Commun.
2014, 50, 6073-6087; f) J. A. Murphy, J. Org. Chem. 2014, 79, 3731-
3746; g) J. Broggi, T. Terme, P. Vanelle, Angew. Chem. Int. Ed.
2014, 53, 384-413; Angew. Chem. 2014, 126, 392-423 h) H.-J.
Himmel, Synlett 2018, 29, 1957-1977.
B. Eberle, O. Hübner, A. Ziesak, E. Kaifer, H.-J. Himmel, Chem. Eur.
J. 2015, 21, 8578-8590.
H. Herrmann, M. Reinmuth, S. Wiesner, O. Hübner, E. Kaifer, H.
Wadepohl, H.-J. Himmel, Eur. J. Inorg. Chem. 2015, 2345-2361.
M. Reinmuth, Dissertation thesis, Ruprecht-Karls-Universität
Heidelberg (DE) 2012.
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[7]
[8]
[9]
[10]
[11]
[12]
H. Herrmann, E. Kaifer, H.-J. Himmel, Chem. Eur. J. 2017, 23, 5520-
5528.
M. Reinmuth, C. Neuhäuser, P. Walter, M. Enders, E. Kaifer, H.-J.
Himmel, Eur. J. Inorg. Chem. 2011, 83-90.
a) S. Wiesner, A. Wagner, O. Hübner, E. Kaifer, H.-J. Himmel,
Chem. Eur. J. 2015, 21, 16494-16503; b) S. Wiesner, A. Wagner, E.
Kaifer, H.-J. Himmel, Dalton Trans. 2016, 45, 15828-15839; c) D. F.
Schrempp, E. Kaifer, H. Wadepohl, H.-J. Himmel, Chem. Eur. J.
2016, 22, 16187-16199; d) D. F. Schrempp, S. Leingang, M.
Schnurr, E. Kaifer, H. Wadepohl, H.-J. Himmel, Chem. Eur. J. 2017,
23, 13607-13611; e) D. F. Schrempp, E. Schneider, E. Kaifer, H.
Wadepohl, H.-J. Himmel, Chem. Eur. J. 2017, 23, 11636-11648; f)
H.-J. Himmel, Inorg. Chim. Acta 2018, 481, 56-68; g) D. F.
Schrempp, E. Kaifer, H.-J. Himmel, Eur. J. Inorg. Chem. 2018,
3660-3667; h) A. Ziesak, L. Steuer, E. Kaifer, N. Wagner, J. Beck,
H. Wadepohl, H.-J. Himmel, Dalton Trans. 2018, 47, 9430-9441.
a) R. J. Mortimer, Chem. Soc. Rev. 1997, 26, 147-156; b) D. R.
Rosseinsky, R. J. Mortimer, Adv. Mater. 2001, 13, 783-793; c) R. J.
Mortimer, Annu. Rev. Mater. Res. 2011, 41, 241-268.
[35]
[36]
[37]
[13]
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
Chem. Soc. Rev. 2011, 40, 3336-3355; e) I. Ratera, J. Veciana,
Chem. Soc. Rev. 2012, 41, 303-349; f) H. Iwamura, Polyhedron
2013, 66, 3-14; g) S. Nishida, Y. Morta, in Organic Redox Systems
(Ed.: T. Nishinaga), Wiley, Hoboken, New Jersey, 2015, pp. 177-
243; h) K. Kato, A. Osuka, Angew. Chem. Int. Ed. 2019,
doi:10.1002/anie.201900307;
doi:10.1002/ange.201900307.
Angew.
Chem.
2019,
This article is protected by copyright. All rights reserved.

10.1002/chem.201902216
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Persistent radicals with bright
colours: Redox-active azines exhibit
preeminent electrochromism with fast
switching characteristics in
Marco Werr, Elisabeth Kaifer, Hubert
Wadepohl, Hans-Jörg Himmel*
Page No. – Page No.
electrochromic test devices; their
redox-potentials could be rationally
varied by synthesis.
Tuneable Redox-Chemistry and
Electrochromism of Persistent
Symmetric and Asymmetric Azine
Radical Cations
400
600
/ nm
800
1000
l
This article is protected by copyright. All rights reserved.