Vapochromic Luminescent π-Conjugated Systems with Reversible Coordination-Number Control of Hypervalent Tin(IV)-Fused Azobenzene Complexes

Article abstract of DOI:10.1002/chem.202100571

The dynamic and reversible changes of coordination numbers between five and six in solution and solid states, based on hypervalent tin(IV)-fused azobenzene (TAz) complexes, are reported. It was found that the TAz complexes showed deep-red emission owing t

Full text of DOI:10.1002/chem.202100571

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
Vapochromic Luminescent π-Conjugated Systems with
Reversible Coordination-Number Control of Hypervalent
Tin(IV)-Fused Azobenzene Complexes
Masayuki Gon,[a] Kazuo Tanaka,*[a] and Yoshiki Chujo[a]
Abstract: The dynamic and reversible changes of coordina-
tion numbers between five and six in solution and solid
states, based on hypervalent tin(IV)-fused azobenzene (TAz)
complexes, are reported. It was found that the TAz complexes
showed deep-red emission owing to the hypervalent bond
composed of an electron-donating three-center four-electron
(3c–4e) bond and an electron-accepting nitrogen–tin (N–Sn)
coordination. Furthermore, hypsochromic shifts in optical
spectra were observed in Lewis basic solvents because of
alteration of the coordination number from five to six. In
particular, vapochromic luminescence was induced by attach-
ment of dimethyl sulfoxide (DMSO) vapor to the coordination
point at the tin atom accompanied with a crystal–crystal
phase transition. Additionally, the color-change mechanism
and degree of binding constants were well explained by
theoretical calculation. To the best of our knowledge, this is
the first example of vapochromic luminescence by using
stable and variable coordination numbers of hypervalent
bonds.
Introduction
tuning systems to realize next generation of optoelectronic
sensors.
Vapochromic luminescent properties are the class of stimuli-
responsive phenomena where compounds change luminescent
color by exposing solvent vapor.[1,2] Development of organic
materials are strongly required for realizing intelligent optoelec-
tronic organic devices, such as chemical sensors because tuning
of optical properties including luminescent colors is expectable
by chemical modification and substituent effects.[3] Vapochro-
mic luminescence is usually regarding induction of inter- or
intramolecular structural changes with π-conjugated
systems.[4–14] In many cases, the luminescence-color change is
triggered by insertion of solvent molecules into the crystal
structure as guest molecules. By modifying with organic bulky
substituents and long alkyl chains which are capable of causing
drastic structural rearrangements of molecular distributions by
solvent absorbing, electronic structures at the luminophores are
critically perturbed. The limited examples of the vapochromic
luminescence are caused by the direct attachment of solvent
molecules to the metal center in the luminescent
complexes.[15–27] In this approach, transition metal complexes
have been often used for capturing vapor molecules. Transition
metals can readily change the valence number and subse-
quently show different optical properties in each state. If stable
coordination states can be constructed, desired behaviors are
obtained. Thus, it is desired for constructing tailor-made color-
Heteroatom-containing building blocks are effective in
constructing functional and unique materials.[28–30] As examples,
we discovered cubic silica and boron cluster compounds named
polyhedral oligomeric silsesquioxane (POSS)[31,32] and o-
carborane,[33,34] respectively, have been applicable to main
scaffolds for creating functional materials showing superior
luminescence behaviors with high durability or stimuli-respon-
siveness. We recently reported the series of solid-state
luminescent complexes and polymers containing boron-fused
azomethine (À C=NÀ )[35–39] and azo (À N=NÀ )[40–44] complexes with
the tridentate ligands, o-salicylideneaminophenol or 2,2’-dihy-
droxyazobenzne scaffold, respectively.[45] These materials
showed intense emission in the condensed state, and highly-
efficient polymer film emission[38] and near-infrared (NIR)
absorptive and emissive characteristics[40–42,44] were obtained. It
should be emphasized that unique environment-sensitive
luminescent properties, such as aggregation-induced emission
(AIE),[46,47] crystallization-induced emission enhancement
(CIEE),[48] in which enhanced emission can be observed only in
aggregation and crystal, respectively. Thus, we regarded these
complexes as a solid-state luminescent building block having
stimuli-responsivity.
Herein, we illustrate that the hypervalent tin(IV)-fused
azobenzene (TAz) complexes are the robust platform for
realizing color-tunable vapochromic luminescent π-conjugated
systems. To the best of our knowledge, we found that TAz
systems can show enough stability for isolation and structural
analyses of both five and six coordinated structures and
reactivity to offer the first example of color-tunable vapochro-
mic luminescence based on the dynamic and reversible
coordination-number changes. We synthesized the series of the
TAz complexes and found the formation of five coordinated
[a] Dr. M. Gon, Prof. Dr. K. Tanaka, Prof. Dr. Y. Chujo
Department of Polymer Chemistry
Graduate School of Engineering,
Kyoto University
Katsura, Nishikyo-ku, Kyoto 615-8510 (Japan)
E-mail: tanaka@poly.synchem.kyoto-u.ac.jp
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202100571
Chem. Eur. J. 2021, 27, 7561–7571
7561
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
structures with the distorted trigonal bipyramidal geometry.[49,50]
Additionally, the TAz complexes exhibited emission in the
deep-red regions both in solution and solid states owing to the
hypervalent bond of tin atom including an electron-donating
three-center four-electron (3c–4e)[51] bond and an electron-
accepting nitrogen-tin (NÀ Sn) coordination. Moreover, it was
shown that hypsochromic luminescence was observed in the
presence of some kinds of Lewis basic solvents. In particular,
the dimethyl sulfoxide (DMSO)-coordinated complexes showed
high stability enough for further structural and optical analyses,
and it was revealed that chromism and luminochromism
originating from the coordination number changes from five to
six at the tin atom. The binding affinities followed by
vapochromic luminescent behaviors were varied by the sub-
stituent effect in the azobenzene ligand. Moreover, the order of
binding constants measured in solution and the color-change
mechanism based on the π-conjugated system were good
explained by theoretical investigation. In other words, from the
theoretical calculations, it should be possible to estimate not
only binding constants but also vapochromic luminescence
behaviors in this system. Finally, by using the powder samples,
we can demonstrate good reversibility. This is the first example,
to the best of our knowledge, to offer color-tunable vapochro-
mic luminescence based on the dynamic and reversible
changes of the coordination number in the hypervalent tin
complexes.
enough for structural and optical measurements. It should be
noted that all products show good stability to light, air and
moisture in solution and solid, meaning that degradation
should be negligible in further analyses.
NMR spectra
The structures of TAz complexes in solution were investigated
1
1
with H, 13C and 119Sn NMR spectrometry. From H and 13C NMR
spectra in chloroform-d (CDCl3), the signal peaks attributable to
five- and six-membered rings were clearly detected. According
to the previous reports, the fast conformational change of the
coordination position at the N=N bond in the azobenzene
ligand, called as a “pedal motion”,[52] is proposed in the silicon-
azobenzene complexes.[53,54] In our study, because both
benzene rings were distinguishable in H and 13C NMR spectra
1
at high temperature (Figure S1), structural fluctuation might be
ignorable even in solution. The chemical shifts of 119Sn NMR
spectra (À 365.4 ppm for TAz-H, À 366.5 ppm for TAz-F,
À 367.3 ppm for TAz-OBu and À 360.2 ppm for TAz-CF3) were
similar to that of the five coordinated tin atom with two phenyl
groups (À 328.4 ppm).[55] Interestingly, the large upfield shifts in
the 119Sn NMR spectra were observed in DMSO-d6 (À 467.3 ppm
for TAz-H, À 471.1 ppm for TAz-F, À 465.0 ppm for TAz-OBu and
À 475.9 ppm for TAz-CF3, Figure S2), strongly suggesting that
the DMSO coordination should proceed at the tin atom
followed by the formation of the six coordinated state.[56–59] This
result also means that the asymmetric five- and six-coordinated
structures can be obtained by selecting the solvent between
CDCl3 and DMSO-d6, respectively. Noted that the complexes
involving six-coordinated states showed high stability. There-
fore, we tried isolation of these DMSO-coordinated complexes.
Results and Discussion
Synthesis
Scheme 1 shows the syntheses of the TAz complexes with
hydrogens (TAz-H), halogens (TAz-F), electron-withdrawing
groups (TAz-CF3) and electron-donating groups (TAz-OBu). All
of the complexes were prepared by the reaction between
azobenzene tridentate ligands and diorganotin(IV) oxide under
reflux conditions with acetone in almost quantitative yields.[49]
The ligand, 2,2’-dihydroxyazobenzene (L-H), is commercially
available, and the detailed synthetic methods for the other
ligands (L-F, L-CF3 and L-OBu) and model compounds, Az-H,
Az-F, Az-OBu, Az-CF3, are described in the Supporting
Information. The structures of new compounds in this study
Crystal structures
The structures of the TAz complexes were confirmed by single
crystal X-ray diffraction (SC-XRD) analyses (Figures 1A and S3–
S6). Representative bond lengths, angles and dihedral angles
are listed in the distorted trigonal bipyramidal geometry
(Table S1).[49] The formation of the five-coordinated tin atom is
proved from the structural data. Around the tin atom, two
oxygen atoms (O(1) and O(2)) were at the apical positions, and
nitrogen (N(1) or N(2)) and two carbon atoms in the phenyl
groups (C(13) and C(19)) were at the equatorial positions. The
1
were confirmed by H, 13C and 119Sn NMR spectroscopies, high-
resolution mass spectrometry (HRMS), and elemental analyses.
From these characterization data, we concluded that the
products should have desired structures with high purity
°
°
angles of O(1)À Sn(1)À O(2) were 157.5 for TAz-H, 159.3 for
°
TAz-F, and 157.8 for TAz-CF3, indicating that the structural
°
distortion should be involved (ideal angle, 180 ). The dihedral
°
°
angles of C(1)À N(1)À N(2)À C(7) were À 179.3 for TAz-H, À 178.7
°
for TAz-F and À 179.5 for TAz-CF3 and the azobenzene
moieties were highly planar regardless of substituent effects.
The bond lengths of N(1)À N(2) were 1.26 Å for TAz-H, 1.25 Å for
TAz-F and 1.27 Å for TAz-CF3. These values were within a
category of the N=N bond of azobenzene (1.25–1.27 Å).[49,60] The
structure of TAz-OBu was not able to be determined due to
large disorders at linear butyl chains (Figure S5). Regarding the
Scheme 1. Synthesis of TAz complexes from each ligand.
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7562
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
tin complexes with the DMSO coordination.[57] The crystals were
sufficiently grown in the saturated DMSO solutions of TAz-F
and TAz-CF3. Figures 1A, S8 and S9 show ORTEP drawings of
the obtained DMSO complexes, TAz-F-DMSO and TAz-CF3-
DMSO. Representative bond lengths, angles and dihedral
angles are also listed in Table S1. The oxygen atom of DMSO
was perpendicularly attached to the tin atom against the
azobenzene surface and formed an orthogonal geometry.[55]
The angles of O(1)À Sn(1)À O(2), dihedral angles of C(1)À N(1)À N-
°
N(2)À C(7) and the bond lengths of N(1)À N(2) were 155.1 ,
°
°
°
À 177.6 and 1.25 Å for TAz-F-DMSO, 155.5 , À 176.8 and 1.25 Å
for TAz-CF3-DMSO, respectively. These values mean that the
geometry of the azobenzene ligand is hardly changed during
the transition between five and six coordination states. That is,
π-planarity, which is the origin of luminescence, should be
preserved after the coordination. The DMSO attachment
induced a slight bond elongation of Sn(1)À N(1), Sn(1)À C(13)
and Sn(1)À C(19) (+0.10 Å, +0.04 Å and +0.02 Å from TAz-F,
+0.10 Å, +0.05 Å and +0.03 Å from TAz-CF3, respectively),
suggesting that bond strengths could be released by the DMSO
coordination. Although the single crystal of TAz-H-DMSO was
obtained, the unit cell was too large to obtain reliable crystal
data (Figure S7). The difficulty in obtaining a distinct crystal
packing structure of TAz-H-DMSO is probably because there
were no dimeric interactions between DMSO moieties as seen
in TAz-F-DMSO and TAz-CF3-DMSO (Figures S8B and S9B). In
addition, there are few effects of the donor and acceptor
substituents or DMSO attachments on the distribution of the π-
electrons because distinct alteration of the bond lengths of
benzene rings in azobenzene moiety was hardly observed
among TAz complexes (Figure S10). In summary, critical struc-
tural changes influenced on the optical properties, such as
distortion of π-planes, were hardly induced by the DMSO
coordination.
Another structural feature is the distances between each
molecule in the crystalline cell, which critically influence
emission efficiency in solid. Most of luminescent organic dyes
show poor luminescence in solid due to intermolecular
interaction, and this concentration quenching process is one of
critical issues to be solved for obtaining luminescent materials
and devices.[46,62] Therefore, it is known that condensed
molecular distribution is unfavorable for receiving emission in
the solid state. From the crystal of TAz-F, relatively tight
packing was obtained (3.81 Å), whereas the much longer
intermolecular distance was observed from TAz-CF3 (6.63 Å)
(Figure 1B). The sparse packing is favorable for obtaining solid-
state luminescence by suppressing concentration quenching.
Surprisingly, by the DMSO coordination, the opposite changes
were observed. The longer distance was detected by +0.33 Å
from TAz-F, meanwhile the close packing was obtained in the
TAz-CF3-DMSO crystal (À 2.37 Å). These changes in molecular
distribution would play a critical role in emission efficiency in
solid.
Figure 1. (A) ORTEP drawings of TAz-H, TAz-F, TAz-CF3, TAz-F-DMSO and
TAz-CF3-DMSO (50% probability for thermal ellipsoids). (B) Crystal packing
structures of TAz-F and TAz-F-DMSO, TAz-CF3 and TAz-CF3-DMSO with the
distances of the closest two benzene rings. Hydrogen and a part of
disordered atoms are omitted to clarify. All crystallographic data are shown
in the Supporting Information.
ambiguity of the nitrogen atoms, two different nitrogen
positions in crystal packings were detected from the complexes,
especially TAz-CF3. This result implies that existence of a “pedal
motion” might occur in crystal.[52–54,61] Meanwhile, the NMR data
suggest that the conformational changes should hardly proceed
even in solution. Therefore, we assume that the ambiguity of
nitrogen positions in the crystalline state should be categorized
in a static disorder,[52] not a dynamic one, which is caused by
the difficulty in the distinguishment of both structures replaced
the five- and six-membered rings from each other in the SC-
XRD analysis.
Fortunately, in the case of TAz-F and TAz-CF3, we
successfully isolated the single crystals of the six-coordinated
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7563
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
Solution-state UV-vis absorption measurements
dichloromethane (CH2Cl2) which are classified into non-coordi-
nating solvents (Figure S11A). Meanwhile, clear hypsochromic
shifts in the absorption spectra in the series of coordinating
(Lewis basic) solvents, such as tetrahydrofuran (THF), 1,4-
dioxane, ethyl acetate (EtOAc), acetone, ethanol (EtOH),
acetonitrile (MeCN), dimethyl formamide (DMF) and DMSO,
were observed, indicating that these solvents can drastically
perturb electronic structures of the azobenzene ligands by the
coordination to tin atom (Figure S11B). From the structural
analyses as shown in previous section (Figure 1), significant
changes in the π-plane were not obtained. Therefore, it is
proposed that electronic properties at the tin atom should be
responsible for these band shifts. It is noteworthy that the
solvent-coordinated TAz complexes hardly showed any degra-
dation under ambient conditions. Stable six-coordinated com-
plexes were able to be obtained. Furthermore, as demonstrated
later, the coordination and detachment can reversibly proceed
without degradation. These stability of both hypervalent states
are advantageous in the usages as practical sensors.[64]
Firstly, UV–vis absorption spectra of the TAz complexes were
measured in solution for investigating electronic structures in
the ground state (Figures 2, S11 and Tables 1, S11–S13).
Azobenzene derivatives are known to show a strong π-π*
transition band around 370 nm and a weak n-π* band around
450 nm,[61] while the tin complexes showed the absorption
bands attributed to π–π* transition in the longer wavelength
region (500 nm~). This means that drastic stabilization effects
on electronic structures are obtained by the tin
coordination.[44,49,63] Thus, the TAz complexes can have a narrow
energy gap despite that this molecule has small-sized π-
conjugated system. The detailed mechanism will be discussed
later with the theoretical calculation data.
Interestingly, hypsochromic shifts of the absorption bands
were observed by the DMSO coordination. As representative
examples, comparing to the spectra in CHCl3 and DMSO/
CHCl3 =99/1 v/v solutions (1.0×10À 5 M), we found that the
absorption bands were blue-shifted in the presence of DMSO
(Figures 2 and S12). The shapes and positions of the absorption
bands were hardly changed in hexane, toluene, CHCl3 and
Photoluminescence
To examine electronic structures in the excited state, emission
properties were evaluated. In photoluminescence (PL) spectra,
the bands of the TAz complexes were observed from orange to
NIR regions (Figures 3, S13 and Table 1). The emission proper-
ties were largely affected by the substitution effects.[63,65] The
peak top wavelengths (λPLs) in CHCl3 were obtained in the red
region. Larger emission efficiencies (ΦPLs) were observed from
TAz-F and TAz-OBu. To obtain further information on emission
mechanisms, we estimated radiative and non-radiative rate
constants (kr and knr) from lifetime measurements (Figure S14).
Relatively smaller kr and larger knr were observed TAz-H.
Significant differences in the substituent effect were hardly
obtained between the electron-donating (TAz-OBu) and accept-
ing groups (TAz-F). It is suggested that π-conjugation effects in
relation to delocalization of the lone-pair electrons of fluorine
or oxygen atoms seem to be positive for enhancing emission.
This is because expansion of π-conjugation should increase
transition probability and structural rigidity which lead to rising
the value of kr and reducing the value of knr, respectively.
Figure 2. UV–vis absorption spectra and photos under room lights of TAz-H,
TAz-F, TAz-OBu and TAz-CF3 (1.0×10À 5 M) in CHCl3 (red line) and DMSO/
CHCl3 =99/1 v/v (blue line). The concentration of the samples of photos
were 5.0×10À 5 M for clear recognition of color differences.
Table 1. Spectroscopic data of TAz complexes in solutions.[a]
[c]
Solvent
λabs/nm
λPL[c]/nm
ΦPL
τ
[d] (α)/ns
kr[e]/108 sÀ 1
knr[e]/108 sÀ 1
TAz-H
TAz-F
CHCl3
547
528
523
506
678
626
640
596
0.035
0.041
0.21
0.74
0.69
3.1
0.34 (89%)
1.9 (11%)
3.1
2.7
1.0
0.47
0.59
0.69
0.34
13
14
2.6
10
DMSO[b]
CHCl3
DMSO[b]
0.032
TAz-OBu
TAz-CF3
CHCl3
543
515
551
538
643
622
681
638
0.33
0.32
0.057
0.012
1.0
1.2
0.55
0.12
2.2
2.6
9.2
9.9
DMSO[b]
CHCl3
DMSO[b]
0.16 (57%)
1.2 (43%)
[a] 1.0×10À 5 M. [b] In mixed solvent, DMSO/CHCl3 =99/1 v/v. [c] Absolute PL quantum efficiency excited at λabs. [d] PL lifetime monitored at λPL. [e] kr =ΦPL/
τav, knr =(1À ΦPL)/τav, τav =Σαiτi2/Σαiτi. α: relative amplitude.
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7564
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
support the existence of the 1:1 binding. Moreover, the VT
measurements proved that the alteration between the five and
six coordination states can reversibly proceed without signifi-
cant degradation. On the basis of these analyses, we estimated
a binding constant (K) in 1:1 binding with the results of
titration experiments (Table 2). Accordingly, the KDMSO values of
the complexes were calculated as 3.9 MÀ 1 for TAz-H, 6.6 MÀ 1 for
TAz-F, 1.3 MÀ 1 for TAz-OBu and 25 MÀ 1 for TAz-CF3 (Figur-
es S15–S18). As suggested in the PL measurements in the
presence and absence of DMSO, it was also shown that the
electron-withdrawing groups tended to enhance binding abil-
ity. We also calculated the binding constant with DMF (KDMF) in
the toluene/DMF mixed solvent. The values of KDMF were
estimated to be 0.55 MÀ 1 for TAz-H, 0.92 MÀ 1 for TAz-F, 0.40 MÀ 1
for TAz-OBu and 3.2 MÀ 1 for TAz-CF3, indicating that KDMF is
much smaller than KDMSO (Figures S19–S22). The binding con-
stants of other solvents were too small to be estimated by the
titration experiments. The order of binding affinity was similar
to the reported Lewis basicity in nonprotogenic solvents by
using a zinc Schiff-base complex.[66] Therefore, the TAz com-
plexes also have a potential to investigate the Lewis basicity by
using a tin atom center.
Figure 3. PL spectra and photos irradiated by a UV lamp (365 nm) of TAz-H,
TAz-F, TAz-OBu and TAz-CF3 in CHCl3 (red line) and DMSO/CHCl3 =99/1 v/v
(blue line), excited at wavelengths of absorption maxima. The concentration
of the samples of photos were 5.0×10À 5 M for clear recognition of color
differences.
In DMSO/CHCl3 =99/1 v/v solution, hypsochromic shifts in
the PL spectra were also detected in all TAz complexes, similarly
to the absorption spectra (Figure 3). After the DMSO coordina-
tion, intense emissions with high ΦPLs were still observed. These
data mean that the DMSO coordination should be retained in
the excited state. If the DMSO detachment occurred in the
excited state, the emission components of the five coordinated
system should be observed with large bathochromic effect
accompanied with drastic structural relaxation. That is, DMSO
detachment is negligible in the photoluminescence processes.
According to the orders of λPL and ΦPL values, it was found that
the electron-withdrawing groups lowered emission efficiencies.
In addition, this tendency was observed in the other solvents
(Tables S9–S12). Furthermore, the smaller krs and larger knrs,
which are negative tendencies for giving luminescence, were
detected after the DMSO coordination especially in TAz-F and
TAz-CF3 (Table 1). In comparison to only π-conjugation effects
in the absence of DMSO, luminescent properties of the hyper-
valent tin complexes can be influenced by the substituent
effect. These results suggest that tuning of optical properties is
acceptable in this system.
Cyclic voltammetry
To confirm the narrow-energy-gap properties of the TAz
complexes, experimental values of HOMO (highest occupied
molecular orbital) and LUMO (lowest unoccupied molecular
orbital) energy levels were estimated by cyclic voltammetry
(Figure S24, Table S13). The HOMO and LUMO energy levels of
TAz-H were estimated to be À 5.60 and À 3.56 eV, respectively.
On the other hand, those of the methylated ligand, Az-H, were
calculated to be À 5.73 and À 3.00 eV, respectively. These results
indicate that tin coordination plays a significant role in the
elevation of energy levels of HOMO as well as lowering of those
of LUMO. The opposite effects by the tin coordination on the
energy levels of HOMO and LUMO could originate from
position-dependent different natures of the bond properties in
the trigonal bipyramidal structure containing 3c–4e bonds at
apical positions and sp2 hybrid orbital at equatorial positions.
That is, increases in electron density at the non-bonding orbital
of the 3c–4e bond contribute to the enhancement of electron-
donating ability of oxygen atoms, leading to elevating the
HOMO energy level.[67,68] In addition, acceptance of nitrogen
electron lone pair by the tin atom decreased the LUMO energy
level.[44] Consequently, the energy levels of HOMO and LUMO
should be independently shifted to the opposite directions. The
Binding constants
The crystal structures obtained from SC-XRD analyses indicated
1:1 binding between the five coordinated TAz complex and
DMSO. To investigate whether the same 1:1 binding occurs in
solution, we carried out titration experiments (Figures S15–S22)
and variable temperature (VT) measurements (Figure S23) of
absorption in toluene/DMSO mixed solutions. The results from
both of the titration experiments and the VT measurements
suggested the existence of the isosbestic points which should
Table 2. Binding constants (K) of TAz complexes and DMSO or DMF.[a]
K
DMSO/MÀ 1
K
DMF/MÀ 1
TAz-H
TAz-F
TAz-OBu
TAz-CF3
3.9
6.6
1.3
25
0.55
0.92
0.40
3.2
[a] Determined in toluene at ambient temperature.
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7565
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
HOMO and LUMO energy levels of the other TAz complexes
were À 5.60 and À 3.57 eV for TAz-F, À 5.23 and À 3.38 eV for
TAz-OBu, À 5.78 and À 3.88 eV for TAz-CF3, respectively. These
data can be explained by the conventional substituent effect,
that is, electron-donating and withdrawing groups elevate and
lower HOMO and LUMO energy levels, respectively.
Solid-state coordination
We investigated solid-state optical properties and particularly
vapochromism between five and six coordinations. To obtain
high sensitivity by enlarging surface areas, we prepared the
powder samples through the evaporation with CHCl3 solutions.
The resulting powders of the TAz complexes were exposed by
various vapors at ambient temperature with a vapor-saturated
closed condition (Figure S25). Almost all samples including
doping film in a polymer matrix were dissolved or not changed,
meanwhile the color-changed powders were stably obtained in
TAz-F and TAz-CF3 with the DMSO vapor (Figure S25). Preser-
vation of DMSO in the crystal packing structure with a dimer
should contribute to stable capturing of the solvent vapor.
PXRD measurements
Figure 4. PXRD patterns of (A) TAz-F and (B) TAz-CF3 in each state. Purple
and red lines show simulated pattern from SC-XRD analyses. Light blue and
We carried out a powder X-ray diffraction (PXRD) measurement
to confirm the molecular distributions in the powder samples of
TAz-F and TAz-CF3 before and after exposure of the DMSO
vapor (Figures 4 and S26). In summary, almost identical
molecular conformation was obtained between the single
crystals and the powder samples both in the presence and
absence of DMSO. The pristine samples of TAz-F and TAz-CF3
prepared by evaporation from the CHCl3 solutions showed the
similar PXRD patterns obtained by the simulations from SC-XRD.
After exposure to DMSO, the sample color was clearly varied
from dark brown to orange for TAz-F and from dark brown to
red for TAz-CF3. The PXRD patterns were changed to those
which were good agreements with the simulated ones with
TAz-F-DMSO and TAz-CF3-DMSO. Furthermore, once the
obtained TAz-F-DMSO and TAz-CF3-DMSO were heated at
°
green lines show the patterns from the heated powder samples at 100 C for
30 min and exposed samples to DMSO vapor for 4 h at an ambient
temperature under a vapor-saturated closed condition, respectively. The
inserted photos show the samples in each state under room light.
ric analysis (TGA) was carried out. The samples of TAz-F-DMSO
and TAz-CF3-DMSO were prepared by exposure with the
powder samples of TAz-F and TAz-CF3 to saturated DMSO
atmosphere for 4 h at ambient temperature under a vapor-
saturated closed condition, respectively. The TGA profiles with
the single decomposition step were obtained from the pristine
samples, whereas two decomposition steps were detected from
the DMSO complexes (Figure S27). The first decomposition
temperature (Td1) and the percentages of weight loss were
°
100 C, the sample colors were returned to those at the initial
°
°
state. The PXRD patterns were almost recovered to the original
ones although partially broadening was observed. Those results
indicate two important issues. Firstly, the DMSO coordination
can proceed even in solid. Secondary, the DMSO coordination
and detachment can reversibly proceed with the crystalÀ crystal
transitions. It should be noted that this exposing and heating
cycle can be stably repeated at least 3 times (Figure S26). By
using powder samples, homogeneous states could be main-
tained during the cycles.
103 C and 12.7 wt% for TAz-F-DMSO and 111 C and 11.6 wt%
for TAz-CF3-DMSO, respectively (Table S14). These are almost
identical to theoretical values calculated from chemical formula,
13.0 wt% for TAz-F-DMSO and 11.2 wt% for TAz-CF3-DMSO.
Therefore, Td1 was attributable to the loss of DMSO. The higher
Td1 and fine agreement of DMSO weight loss percentages in
TAz-CF3-DMSO are corresponded to the larger binding con-
stant than that of TAz-F-DMSO.
Solid-state photoluminescence
Thermogravimetric analysis (TGA)
We performed PL measurements with the solid samples (Fig-
ure 5 and Table 3). Initially, we compared luminescent proper-
ties between crystal and powder samples. Larger emission
To estimate the amount of adsorption by DMSO exposure to
the powder samples of TAz-F and TAz-CF3, a thermogravimet-
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7566
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
Figure 5. PL spectra of TAz-H, TAz-F, TAz-OBu and TAz-CF3 in the solid state, excited at wavelengths of absorption maxima under solution conditions.
Table 3. Optical data of the TAz complexes in solid.
mic shifts were observed from all samples, similarly to the
Crystal
Powder
results with the solution samples. In the powder samples of
TAz-H and TAz-F, we investigated the reversible alteration of
the PL properties corresponding to repeated cycles of five and
six coordinations triggered by the DMSO adsorption. These data
clearly indicate that DMSO vapor should be captured by the tin
atom, and the six-coordinate complexes were constructed,
followed by luminochromism. Owing to appropriate binding
constants, the coordinated DMSO can be released by heating
with crystalÀ crystal transitions. Thus, the reversible changes
should be achieved. The relatively-larger difference in ΦPLs of
TAz-CF3 can be explained by the crystal packing. According to
the SC-XRD data, the drastic change of intermolecular distance
was revealed during the DMSO capturing and detaching
(Figure 1B). Corresponding to these structural alterations,
degree of electronic interaction should be varied, followed by
emission efficiency changes. Optical properties of the initial
samples and the first DMSO exposed ones were a little bit
difference from those of the samples in the latter cycles
(Figures 6 and S28, Table S15). This might be by the heat history
caused in the evaporation (Figure S26). Indeed, after the first
exposed samples were heated to remove DMSO, good repeat-
able situation was provided. The coordination number of the
hypervalent tin atom can be dynamically and reversibly altered
by vapor fuming. Moreover, both states can be isolated and
examined by the series of analyses. Thus, optical properties at
[b]
[b]
λPL[a]/nm
ΦPL
λPL[a]/nm
ΦPL
TAz-H
TAz-H-DMSO
TAz-F
TAz-F-DMSO
TAz-OBu
TAz-OBu-DMSO
TAz-CF3
699
628
682
624
680
0.035
0.065
0.044
0.098
0.058
692
0.034
[c]
[c]
–
–
692
627
667
0.023
0.021
0.010
[c]
[c]
[c]
[c]
–
–
–
–
699
642
0.058
0.021
713
647
0.035
0.013
TAz-CF3-DMSO
[a] Excited at absorption maxima in solution (λabs). [b] Absolute PL quantum
efficiency excited at λabs. [c] Not isolated as a solid sample.
efficiencies were observed from the crystalline samples than
those from the powder ones. It is likely that vigorous molecular
tumbling occurs around the surfaces of particles and amor-
phous regions in the powder samples. Thus, decreases in
emission efficiencies were detected from the powder samples
although almost the same molecular packing should be formed.
However, we performed luminescent vapochromic experiments
with the powder samples because of high sensitivity in
emission color change and reversibility observed in the
chromism investigation. The emission bands with the powder
samples were obtained in the deep-red region. By the DMSO
exposing under a vapor-saturated closed condition, hypsochro-
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7567
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
°
Figure 6. Reversible controls of the PL properties of (A) TAz-F and (B) TAz-CF3 with heating (100 C for 30 min)-DMSO exposing (an ambient temperature for
2 h under a vapor-saturated closed condition) processes. The inserted photos show the samples irradiated by a UV lamp (365 nm).
each state can be explained. There should be almost no special
electronic intermolecular interactions originating from the
dimer, such as J and H-type aggregations,[69] in crystal because
the variation of the optical properties including before and after
DMSO attachment was similar to that in solution.
by critically stabilizing the non-bonding MO (Figure 7A). As a
result, the S0!S1 transitions of the TAz complexes are assigned
to the π-π* allowed transition (f=0.2354 for TAz-H; 0.2316 for
TAz-H-DMSO) (Figure 7B). These theoretical results are corre-
sponded to the experimental data. The absorption bands from
the azobenzene ligands in the longest wavelength regions were
very small, meanwhile the magnitudes of these absorption
bands were drastically enhanced in the TAz complexes
(Figures 2 and S12). Moreover, the energy diagrams also
proposed tin coordination caused both increase and decrease
in HOMO and LUMO energy levels, respectively, in comparison
to the azobenzene derivatives. From the CV data with the TAz
complexes, these narrower energy gap effects were obviously
confirmed (Figure S24 and Table S13).
To get deep insight about MO, we performed a natural
bond orbital (NBO) calculation (Figures S32 and S33).[70,71] The
results suggest that NBO energy levels of lone pairs of oxygen
atoms (LP(O1) and LP(O2)) (for example, average energy:
À 8.05 eV for TAz-H) critically increased compared to those of
the methylated ligand (for example, average energy: À 8.41 eV
Theoretical calculations
To support the optical properties of TAz complexes, density
functional theory (DFT) and time-dependent (TD)-DFT calcula-
tions were executed (Figures 7, S29–S31 and Table S16). The
most significant point is that the intrinsic forbidden nature of
the HOMO-LUMO transition in the azobenzene ligand is
changed to the allowed one by the tin coordination. The energy
diagram of molecular orbitals (MOs) suggests that the S0!S1
transitions of the azobenzene derivatives were assigned to the
n–π* forbidden transition (oscillator strength: f=0.000 for
azobenzene; 0.0261 for Az-H). By the tin coordination, the
HOMO is transformed from non-bonding MO to π-bonding one
Figure 7. (A) Energy diagram, selected MOs and (B) oscillator strengths (f) of selected transition bands of azobenzene, Az-H, TAz-H and TAz-H-DMSO obtained
with DFT and TD-DFT calculations at TDÀ B3LYP/6-311G(d,p)//B3LYP/6-311G(d,p) level (isovalue=0.02).
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7568
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
for Az-H) (Figure S32). In other words, the electron donating
ability of two oxygen atoms at the apical positions should be
stronger than those of the methoxy groups owing to the 3c–4e
bond character. Indeed, the contribution of 3c–4e bond was
observable in HOMO-11 (bonding orbital), HOMO-7 (non-
bonding orbital) and LUMO+14 (anti-bonding orbital) (Fig-
ure S34).[72] In addition, a strong donor–acceptor interaction
occurred between a lone pair of nitrogen atom and the tin
atom (for example, LP(N1)!LP*(Sn1), stabilization energy:
76.35 kcalmolÀ 1 for TAz-H) (Figure S33), leading to decrease in
the LUMO energy level. Consequently, different bond natures in
the trigonal bipyramidal structure enable to coexist the two
opposite properties, electron-donating and accepting abilities.
We investigated the geometry in the excited state of TAz-H
with TD-DFT calculations (Figure S35). The TAz-H preserved
planar structures even in the excited states in spite of
elongation of nitrogen–nitrogen bonds; dihedral angle of
the HOMO was responsible for the wider energy gap. It should
be mentioned that substitution effects on optical properties
were good agreement with the experiment results (Figures S29
and S30).
The NBO calculation also performed with the DMSO
complex. It was shown that the DMSO attachment decreased
electron-accepting ability of tin atom by enhancing electron
density at the tin atom (for example, LP(N1)!LP*(Sn1),
stabilization energy: 55.58 kcalmolÀ 1 for TAz-H-DMSO) (Fig-
ure S33). The upfield shift of the 119Sn NMR signal peak by a
shielding effect in DMSO-d6 solvent experimentally supported
this calculation result (Figure S2). Subsequently, the decrease in
electron-accepting ability should induce elevation of LUMO and
HOMO energy levels. Indeed, the energy levels of LP(O1) and
LP(O2) of six coordinated geometry (for example, average
energy: À 7.71 eV for TAz-H-DMSO) were larger than those of
five one (for example, average energy: À 8.05 eV for TAz-H)
(Figure S32). Considering the energy-gap relationships between
TAz-H and TAz-H-DMSO as mentioned above, the LUMO
energy level was effectively risen by increased electron density
at the tin atom compared to the HOMO energy level. Notably,
the shapes of HOMO and LUMO were preserved between five
and six coordinated geometries (Figures 7A, S31). Those results
are corresponded to the experimental data that the chromisms
to shorter wavelength regions in both absorption and lumines-
cence spectra can proceed through not distortion at the
azobenzene moiety but drastic changes in electronic properties
at the tin atom. In the TAz complexes, the electronic characters
only at the tin atom dominate not only optical properties but
also binding affinities. This is the uniqueness of this system.
Finally, we evaluated applicability of theoretical calculations
for estimating binding affinity of DMSO and DMF to the tin
atom (Figures 8 and S36). Accordingly, the significant linear
relationship between the Sn(1)À O(3) bond length (Sn(1): TAz
°
C(1)À N(1)À N(2)À C(7) and bond length of N(1)À N(2) are 179.97
°
and 1.275 Å in the ground state, 179.99 and 1.327 Å in the
excited state, respectively. In the previous works, we mentioned
that non-planar boron-fused azomethine and azo derivatives
did not show emission in solution due to intramolecular
motions in the excited state.[44] Conversely, the high planarity of
TAz-H inhibited molecular motion in the excited state, therefore
we successfully received emission from the azobenzene moiety
even in solution.
The TD-DFT calculation also provided speculation on the
peak shifts of optical bands by the DMSO coordination.
Correspondingly, hypsochromic shifts of the absorption bands
were observed from the comparison with the complexes in the
~
presence and absence of DMSO ( ES0!S1 =2.47 eV for TAz-H
and 2.66 eV for TAz-H-DMSO). Although both energy levels of
HOMO and LUMO were elevated by the DMSO coordination,
larger degree of increase in the LUMO energy level than that of
Figure 8. (A) The relationship between simulated Sn(1)À O(3) bond lengths of TAz-DMSO complexes and binding constants (KDMSO) measured in toluene
solution at ambient temperature. (B) The optimized structures and simulated Sn(1)À O(3) bond lengths of TAz-DMSO complexes with DFT calculations at a
B3LYP/6-311G(d,p) level.
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7569
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
complexes, O(3): DMSO or DMF) and the natural logarithm of Conflict of Interest
binding constant (lnK) was observed. This might be based on
the relationship between a Gibbs energy change and an
The authors declare no conflict of interest.
°
°
equilibrium constant, ΔG =À RTlnK (ΔG : standard Gibbs free
energy change, R: universal gas constant, T: temperature). In
this case, the bond length of SnÀ O should be approximately
Keywords: Azobenzene · hypervalent · luminescence · tin ·
vapochromism
°
proportional connection to ΔG . Although the proof of the
relationship is not clear, bond lengths might be simple
parameters for the estimation of binding constants by theoret-
ical calculation.
[1] O. S. Wenger, Chem. Rev. 2013, 113, 3686–3733.
[2] X. Zhang, B. Li, Z.-H. Chen, Z.-N. Chen, J. Mater. Chem. 2012, 22, 11427–
11441.
[3] E. Li, K. Jie, M. Liu, X. Sheng, W. Zhu, F. Huang, Chem. Soc. Rev. 2020, 49,
1517–1544.
[4] C. Dou, L. Han, S. Zhao, H. Zhang, Y. Wang, J. Phys. Chem. Lett. 2011, 2,
666–670.
Conclusion
[5] B. Prusti, M. Chakravarty, Dyes Pigm. 2020, 181, 108543.
[6] M. Kumar, S. J. George, Chem. Eur. J. 2011, 17, 11102–11106.
[7] Y. Akune, R. Hirosawa, H. Takahashi, M. Shiro, S. Matsumoto, RSC Adv.
2016, 6, 74506–74509.
[8] S. Hatanaka, T. Ono, Y. Hisaeda, Chem. Eur. J. 2016, 22, 10346–10350.
[9] H. Naito, Y. Morisaki, Y. Chujo, Angew. Chem. Int. Ed. 2015, 54, 5084–
5087; Angew. Chem. 2015, 127, 5173–5176.
[10] S. Ito, A. Hirose, M. Yamaguchi, K. Tanaka, Y. Chujo, J. Mater. Chem. C
2016, 4, 5564–5571.
[11] R. Yoshii, K. Suenaga, K. Tanaka, Y. Chujo, Chem. Eur. J. 2015, 21, 7231–
We synthesized
a series of TAz complexes having five
coordinated tin atoms with the distorted trigonal bipyramidal
molecular geometry. Owing to high stability of the complexes
with and without solvent molecules, a series of analyses were
applicable. Coexistence of electron-donating and accepting
abilities originating from 3c–4e bond and Sn–N coordination,
respectively, realize outstanding narrow energy gap in spite of
the small azobenzene-based π-conjugated system. Further-
more, we found the luminochromic behaviors by the DMSO
vapor fuming and subsequently by heating for the reverse
process. Since dynamic and reversible shifts can proceed
between five and six coordination numbers at the hypervalent
tin atom, these behaviors can be obtained. From the SC-XRD
and PXRD, it was revealed that the captured DMSO altered the
coordination number from five to six with crystalÀ crystal
transitions. Hypsochromic shifts of absorption and emission
bands were monitored by the solvent coordination, and these
optical changes were also able to be well explained by the
theoretical calculation. In summary, emission color changes of
π-conjugated systems should originate from the electronic
density change at the tin atom. Our findings propose that the
hypervalent system is expected to open up the field of clear
vapochromic system with color-tunable π-conjugated system.
7237.
[12] R. Yoshii, A. Hirose, K. Tanaka, Y. Chujo, Chem. Eur. J. 2014, 20, 8320–
8324.
[13] H. Sasaki, H. Imoto, T. Kitao, T. Uemura, T. Yumura, K. Naka, Chem.
Commun. 2019, 55, 6487–6490.
[14] K. Naka, T. Kato, S. Watase, K. Matsukawa, Inorg. Chem. 2012, 51, 4420–
4422.
[15] B. Orwat, M. J. Oh, M. Zaranek, M. Kubicki, R. Januszewski, I. Kownacki,
Inorg. Chem. 2020, 59, 9163–9176.
[16] H. Ohara, T. Ogawa, M. Yoshida, A. Kobayashi, M. Kato, Dalton Trans.
2017, 46, 3755–3760.
[17] T. Hasegawa, A. Kobayashi, H. Ohara, M. Yoshida, M. Kato, Inorg. Chem.
2017, 56, 4928–4936.
[18] A. Kobayashi, R. Arata, T. Ogawa, M. Yoshida, M. Kato, Inorg. Chem.
2017, 56, 4280–4288.
[19] A. Kobayashi, K. Komatsu, H. Ohara, W. Kamada, Y. Chishina, K. Tsuge,
H.-C. Chang, M. Kato, Inorg. Chem. 2013, 52, 13188–13198.
[20] Y. Wu, X. Zhang, L.-J. Xu, M. Yang, Z.-N. Chen, Inorg. Chem. 2018, 57,
9175–9181.
[21] J. Forniés, N. Giménez, S. Ibáñez, E. Lalinde, A. Martín, M. T. Moreno,
Inorg. Chem. 2015, 54, 4351–4363.
[22] S. Tsukamoto, S. Sakaki, Dalton Trans. 2013, 42, 4809–4821.
[23] C. E. Strasser, V. J. Catalano, J. Am. Chem. Soc. 2010, 132, 10009–10011.
[24] E. J. Fernández, J. M. López-de-Luzuriaga, M. Monge, M. E. Olmos, R. C.
Puelles, A. Laguna, A. A. Mohamed, J. P. Fackler, Inorg. Chem. 2008, 47,
8069–8076.
Experimental Section
[25] E. J. Fernández, J. M. López-de-Luzuriaga, M. Monge, M. Montiel, M. E.
Olmos, J. Pérez, A. Laguna, F. Mendizabal, A. A. Mohamed, J. P. Fackler,
Inorg. Chem. 2004, 43, 3573–3581.
[26] J. M. López-de-Luzuriaga, M. Monge, M. E. Olmos, J. Quintana, M.
Rodríguez-Castillo, Inorg. Chem. 2019, 58, 1501–1512.
[27] L.-Y. Zhang, L.-J. Xu, X. Zhang, J.-Y. Wang, J. Li, Z.-N. Chen, Inorg. Chem.
2013, 52, 5167–5175.
Crystallographic data: Deposition numbers 2052296, 2052299,
2052300, 2052301, 2052302, 2052305, and 2052306 contain the
supplementary crystallographic data for this paper. These data are
provided free of charge by the joint Cambridge Crystallographic
Data Centre and Fachinformationszentrum Karlsruhe Access Struc-
tures service.
[28] Y. Chujo, K. Tanaka, Bull. Chem. Soc. Jpn. 2015, 88, 633–643.
[29] M. Gon, K. Tanaka, Y. Chujo, Polym. J. 2018, 50, 109–126.
[30] K. Tanaka, Y. Chujo, Polym. J. 2020, 52, 555–566.
[31] M. Gon, K. Sato, K. Kato, K. Tanaka, Y. Chujo, Mater. Chem. Front. 2019, 3,
314–320.
Acknowledgements
[32] M. Gon, K. Kato, K. Tanaka, Y. Chujo, Mater. Chem. Front. 2019, 3, 1174–
This work was partially supported by the Inamori Foundation
(for M.G.) and a Grant-in-Aid for Early-Career Scientists (for M.G.)
(JSPS KAKENHI Grant numbers 20 K15334), for Scientific Re-
search (B) (for K.T) (JP17H03067), for Challenging Research
(Pioneering) (JP20 K20368), for Scientific Research on Innovative
Areas “New Polymeric Materials Based on Element-Blocks
(No.2401)” (JP24102013).
1180.
[33] J. Ochi, K. Tanaka, Y. Chujo, Angew. Chem. Int. Ed. 2020, 59, 9841–9855;
Angew. Chem. 2020, 132, 9925–9855.
[34] J. Ochi, K. Tanaka, Y. Chujo, Dalton Trans. 2021, 50, 1025–1033.
[35] S. Ohtani, Y. Takeda, M. Gon, K. Tanaka, Y. Chujo, Chem. Commun. 2020,
56, 15305–15308.
[36] S. Ohtani, M. Gon, K. Tanaka, Y. Chujo, Crystals 2020, 10, 615.
[37] S. Ohtani, M. Nakamura, M. Gon, K. Tanaka, Y. Chujo, Chem. Commun.
2020, 56, 6575–6578.
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7570
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100571
Chemistry—A European Journal
[38] S. Ohtani, M. Gon, K. Tanaka, Y. Chujo, Macromolecules 2019, 52, 3387–
3393.
[39] S. Ohtani, M. Gon, K. Tanaka, Y. Chujo, Chem. Eur. J. 2017, 23, 11827–
[57] A. Esparza-Ruiz, A. Peña-Hueso, I. Ramos-García, A. Flores-Parra, R.
Contreras, J. Organomet. Chem. 2008, 693, 2739–2747.
[58] A. Esparza-Ruiz, A. Peña-Hueso, I. Ramos-García, A. Vásquez-Badillo, A.
Flores-Parra, R. Contreras, J. Organomet. Chem. 2009, 694, 269–277.
[59] C. Camacho-Camacho, A. Esparza-Ruiz, A. Peña-Hueso, E. Mijangos, I.
Ramos-García, R. Contreras, A. Flores-Parra, Z. Anorg. Allg. Chem. 2013,
639, 1122–1128.
[60] J. Harada, K. Ogawa, S. Tomoda, Acta Cryst. Sect. B 1997, 53, 662–672.
[61] E. Merino, M. Ribagorda, Beilstein J. Org. Chem. 2012, 8, 1071–1090.
[62] S. A. Jenekhe, J. A. Osaheni, Science 1994, 265, 765–768.
[63] J. Yoshino, N. Kano, T. Kawashima, Chem. Commun. 2007, 559–561.
[64] N. Singh, K. Kumar, N. Srivastav, R. Singh, V. Kaur, J. P. Jasinski, R. J.
Butcher, New J. Chem. 2018, 42, 8756–8764.
11833.
[40] M. Gon, J. Wakabayashi, M. Nakamura, K. Tanaka, Y. Chujo, Chem. Asian
J. 2021, 16, 696–703.
[41] M. Gon, J. Wakabayashi, M. Nakamura, K. Tanaka, Y. Chujo, Macromol.
Rapid Commun. 2021, https://doi.org/10.1002/marc.202000566.
[42] J. Wakabayashi, M. Gon, K. Tanaka, Y. Chujo, Macromolecules 2020, 53,
4524–4532.
[43] M. Gon, J. Wakabayashi, K. Tanaka, Y. Chujo, Chem. Asian J. 2019, 14,
1837–1843.
[44] M. Gon, K. Tanaka, Y. Chujo, Angew. Chem. 2018, 130, 6656–6661;
Angew. Chem. Int. Ed. 2018, 57, 6546–6551.
[45] M. Gon, K. Tanaka, Y. Chujo, Chem. Rec. 2021, https://doi.org/10.1002/
[65] J. Yoshino, A. Furuta, T. Kambe, H. Itoi, N. Kano, T. Kawashima, Y. Ito, M.
Asashima, Chem. Eur. J. 2010, 16, 5026–5035.
tcr.202000156.
[66] I. P. Oliveri, G. Maccarrone, S. Di Bella, J. Org. Chem. 2011, 76, 8879–
8884.
[67] P. A. Cahill, C. E. Dykstra, J. C. Martin, J. Am. Chem. Soc. 1985, 107, 6359–
[46] J. Mei, N. L. C. Leung, R. T. K. Kwok, J. W. Y. Lam, B. Z. Tang, Chem. Rev.
2015, 115, 11718–11940.
[47] J. Luo, Z. Xie, J. W. Y. Lam, L. Cheng, H. Chen, C. Qiu, H. S. Kwok, X. Zhan,
Y. Liu, D. Zhu, B. Z. Tang, Chem. Commun. 2001, 1740–1741.
[48] Y. Dong, J. W. Y. Lam, A. Qin, Z. Li, J. Sun, H. H.-Y. Sung, I. D. Williams,
B. Z. Tang, Chem. Commun. 2007, 40–42.
[49] K. E. Bessler, J. A. dos Santos, V. M. Deflon, S. de Souza Lemos, E. Niquet,
Z. Anorg. Allg. Chem. 2004, 630, 742–745.
6362.
[68] Y. V. Torubaev, F. M. Dolgushin, I. V. Skabitsky, A. E. Popova, New J.
Chem. 2019, 43, 12225–12232.
[69] N. Hestand, F. C. Spano, Chem. Rev. 2018, 118, 7069–7163.
[70] E. D. Glendening, C. R. Landis, F. Weinhold, WIREs Comput. Mol. Sci.
2012, 2, 1–42.
[50] H. Preut, F. Huber, R. Barbieri, N. Bertazzi, Z. Anorg. Allg. Chem. 1976,
423, 75–82.
[71] A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899–926.
[72] T. Atsumi, T. Abe, K. Akiba, H. Nakai, Bull. Chem. Soc. Jpn. 2010, 83, 892–
899.
[51] M. L. H. Green, G. Parkin, Dalton Trans. 2016, 45, 18784–18795.
[52] J. Harada, K. Ogawa, Chem. Soc. Rev. 2009, 38, 2244–2252.
[53] U. Böhme, S. Jähnigen, Acta Crystallogr. 2008, C64, o364-o366.
[54] A. S. Soldatenko, I. V. Sterkhova, N. F. Lazareva, J. Organomet. Chem.
2019, 903, 120997.
[55] C. Pettinari, F. Marchetti, R. Pettinari, D. Martini, A. Drozdov, S. Troyanov,
Inorg. Chim. Acta 2001, 325, 103–114.
[56] J. S. Zugazagoitia, M. Maya, C. Damián-Zea, P. Navarro, H. I. Beltrán, J.
Peon, J. Phys. Chem. A 2010, 114, 704–714.
Manuscript received: February 14, 2021
Accepted manuscript online: March 29, 2021
Version of record online: April 22, 2021
Chem. Eur. J. 2021, 27, 7561–7571
www.chemeurj.org
7571
© 2021 Wiley-VCH GmbH