Large-Amplitude Thermal Vibration-Coupled Valence Tautomeric Transition Observed in a Conductive One-Dimensional Rhodium–Dioxolene Complex

Article abstract of DOI:10.1002/chem.202004217

The exploration of dynamic molecular crystals is a fascinating theme for materials scientists owing to their fundamental science and potential application to molecular devices. Herein, a one-dimensional (1D) rhodium-dioxolene complex is reported that exhibits drastic changes in properties with the phase transition. X-ray photoelectron spectroscopy (XPS) revealed that the room-temperature (RT) phase is in a mixed-valence state, and therefore, the drastic changes originate from the mixed-valence state appearing in the RT phase. Another notable feature is that the mean square displacements of the rhodium atoms along the 1D chain dramatically increased in the RT phase, indicating a large-amplitude vibration of the Rh?Rh bonds. From these results, a possible mechanism for the appearance of the mixed-valence state in the RT phase was proposed based on the thermal electron transfer from the 1D d-band to the semiquinonato π* orbital coupled with the large-amplitude vibration of the Rh?Rh bonds.

Full text of DOI:10.1002/chem.202004217

Accepted Article
Accepted Article
Title: Large-Amplitude Thermal Vibration-Coupled Valence Tautomeric
Transition Observed in a Conductive One-Dimensional Rhodium–
Dioxolene Complex
Authors: Minoru Mitsumi, Yuuki Komatsu, Masahiro Hashimoto,
Koshiro Toriumi, Yasutaka Kitagawa, Yuji Miyazaki, Hiroki
Akutsu, and Haruo Akashi
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.202004217
Link to VoR: https://doi.org/10.1002/chem.202004217
01/2020

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
Large-Amplitude Thermal Vibration-Coupled Valence Tautomeric
Transition Observed in a Conductive One-Dimensional Rhodium–
Dioxolene Complex
Minoru Mitsumi,*[a] Yuuki Komatsu,[b] Masahiro Hashimoto,[b] Koshiro Toriumi,[b] Yasutaka Kitagawa,*[c]
Yuji Miyazaki,*[d] Hiroki Akutsu,[e] and Haruo Akashi[f]
Dedicated to the memory of Professor Gleb A. Abakumov, who passed away on August 29, 2019.
[a]
Prof. M. Mitsumi
Department of Chemistry, Faculty of Science, Okayama University of Science
1-1 Ridai-cho, Kita-ku, Okayama 700-0005 (Japan)
E-mail: mitsumi@chem.ous.ac.jp
[b]
[c]
Y. Komatsu, M. Hashimoto, Prof. Koshiro Toriumi
Department of Material Science, Graduate School of Material Science, University of Hyogo
3-2-1 Kouto, Kamigori-cho, Ako-gun, Hyogo 678-1297 (Japan)
Prof. Y. Kitagawa
Department of Materials Engineering Science, Graduate School of Engineering Science, Osaka University
1-3 Machikaneyama, Toyonaka, Osaka 560-8531 (Japan)
E-mail: kitagawa@cheng.es.osaka-u.ac.jp
[d]
Prof. Yuji Miyazaki
Research Center for Thermal and Entropic Science, Graduate School of Science, Osaka University
1-1 Machikaneyama, Toyonaka, Osaka 560-0043 (Japan)
E-mail: miyazaki@chem.sci.osaka-u.ac.jp
[e]
[f]
Prof. H. Akutsu
Department of Chemistry, Graduate School of Science, Osaka University
1-1 Machikaneyama, Toyonaka, Osaka 560-0043 (Japan)
Prof. H. Akashi
Research Institute of Natural Sciences, Okayama University of Science
1-1 Ridai-cho, Kita-ku, Okayama 700-0005 (Japan)
Supporting information for this article is given via a link at the end of the document.
Abstract: The exploration of dynamic molecular crystals is
a
electric or magnetic fields, has received increased attention owing
to their promising applications in molecular switches, molecular
memories, sensors, and so on.[1] These switchable physical
properties originate from the interconversion between two nearly
degenerated electronic states caused by charge transfer,[2] spin
transition,[1b,3] charge-transfer coupled spin transitions,[4] proton
transfer,[5] and cooperative proton–electron transfer[6] that occur
in response to external stimuli. In addition, the switching that
originates from the motion of constituted atoms or molecules in
molecular materials has also attracted attention in recent years.[7]
Such stimuli-responsive, bistable crystalline molecular materials
are summarized in a recent review by Sato as dynamic molecular
crystals.[1d] Valence tautomeric (VT) complexes derived from
redox-active ligands and metal centers, represented by metal–
dioxolene complexes, can take two valence isomers or redox
isomers with different localized electronic structures exhibiting
different physical properties.[8] VT transitions between these
isomers accompanying intramolecular electron transfer can not
only cause changes in the valence and spin states and radical
formation and disappearance but also lead to interesting mixed-
valence states[9] and spin transitions,[2a-c,10] and these features
can be tuned by controlling the VT transitions by external stimuli,
such as heat,[10a,10f,10g,11] light,[1g,10e,10f,12] X-ray,[13] pressure,[3g]
fascinating theme for materials scientists owing to their fundamental
science and potential application to molecular devices. Herein, we
report a one-dimensional (1D) rhodium–dioxolene complex that
exhibits drastic changes in properties with the phase transition. The
X-ray photoelectron spectroscopy (XPS) reveals that the room-
temperature (RT) phase is in a mixed-valence state, and therefore,
the drastic changes originate from the mixed-valence state appearing
in the RT phase. Another notable feature is that the mean square
displacements of the rhodium atoms along the 1D chain dramatically
increase in the RT phase, indicating a large-amplitude vibration of the
Rh–Rh bonds. From these results, we propose a possible mechanism
for the appearance of the mixed-valence state in the RT phase based
on the thermal electron transfer from the 1D d-band to the
semiquinonato π* orbital coupled with the large-amplitude vibration of
the Rh–Rh bonds.
Introduction
The development of bistable molecular materials that can control
their physical properties, such as their electronic, magnetic, and
optical properties, by external stimuli, such as heat, light, pressure,
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
(a)
(b)
(c)
e–
OC
O
O
OC
OC
Rh+
Rh+
Rh+
O
O
X
X
Rh+
chemical
modification
e–
OC
O
O
CO
CO
O
CO
CO
X
X
O
CO
Rh+
Rh2+
O
O
CO
O
O
X
X
OC
Rh+
molecule
O
O
OC
OC
OC
solid state
X = electron-withdrawing
substituent
SOMO (π*)
dz2
dz2 band
π* dz2 band
Scheme 1. Illustration of the frontier orbital control and energy level diagrams
of 1D rhodium–dioxolene complexes
Figure 1. Excess heat capacity of [Rh(3,6-DBDiox-4-NO2)(CO)2]∞ (4) as a
function of temperature. The measurements in the transition region were
repeated four times to determine the accurate transition temperature.
electric[14] and magnetic fields.[15] Therefore, VT transitions are
widely utilized as stimuli-responsive, controllable functions on
dynamic molecular crystals.
dimethoxy-1,2-benzosemiquinonato),
although
the
basic
To realize multifunctional molecular materials that can control
both electrical and magnetic properties by utilizing the VT
skeletons of the Rh(3,6-DBSQ)(CO)2 moieties are almost the
same among compounds 1–3.[10d]
transition, frontier orbital control is conducted for
a one-
Herein, we report the synthesis, crystal structure, and solid-
state properties of a conductive 1D rhodium–dioxolene complex,
[Rh(3,6-DBDiox-4-NO2]∞ (4), that exhibits drastic changes in the
solid-state properties with a first-order phase transition at 176.3 K.
In particular, the conducting behavior shows typical
semiconducting behavior in the low-temperature (LT) phase but
drastically changes to a relatively high electrical conductivity with
little temperature dependence in the room-temperature (RT)
phase. The X-ray photoelectron spectroscopy (XPS) spectrum
indicated that compound 4 in the RT phase exists in a Rh(I,II)
mixed-valence state due to valence tautomerism. The
temperature dependence of the X-ray crystal structures indicated
that the mean square displacements of the rhodium atoms along
the 1D chain dramatically increase in the RT phase, indicating the
occurrence of a large-amplitude vibration of the Rh–Rh bonds in
the RT phase. We performed density functional theory (DFT)
calculations to estimate the energy levels of the semiquinonato π*
orbital and 1D d-band. Finally, employing a combination of these
results, we propose a possible mechanism for the appearance of
the mixed-valence state in the RT phase from the viewpoint of
thermal electron transfer from the filled 1D d-band to the vacant
semiquinonato π* orbital coupled with the increase in the band
dispersion accompanying the large-amplitude vibration of the Rh–
Rh bonds.
dimensional (1D) rhodium(I)–semiquinonato complex [Rh(3,6-
DBSQ)(CO)2]∞
benzosemiquinonato)
(1;
3,6-DBSQ•−
by chemical
=
3,6-di-tert-butyl-1,2-
modification.[16]
Semiquinonate is an organic radical anion and is one of the redox
series of dioxolene, which can produce three different redox
isomers, i.e., benzoquinone (BQ), semiquinonate (SQ•−), and
catecholate (Cat2−), via two sequential one-electron transfers. In
our target system, the semiquinonato radical can serve not only
as a spin carrier of S = ½ but also as an electron acceptor for the
partial oxidation of a rhodium(I) linear chain. Thus, the linear
rhodium chain can produce the 1D conducting d-band via electron
transfer from the rhodium(I) ion to the semiquinonato ligand
(valence tautomerism), i.e., the partial oxidation of the rhodium
linear chain. In addition, the filling of the 1D d-band can be tuned
post-synthetically in cases where the application of external
stimuli, such as heat, light, or pressure, can control valence
tautomerism. The low-lying π* orbitals of the 3,6-DBSQ•− ligand
are relatively close in energy to the upper level of the filled 1D d-
band derived from electron-rich rhodium(I) ions, which facilitates
electron transfer from the rhodium(I) ion to the 3,6-DBSQ•− ligand
by chemical modification. In a previously reported 1D mixed-
valence
rhodium(I,II)–semiquinonato/catecholato
complex
[Rh(3,6-DBDiox-4,5-Cl2)(CO)2]∞ (2), where 3,6-DBDiox-4,5-Cl2
indicates 3,6-di-tert-butyl-4,5-dichloro-1,2-benzosemiquinonato
or 3,6-di-tert-butyl-4,5-dichlorocatecholato, electron transfer from
the rhodium(I) ion to the semiquinonato ligand was realized in the
solid state by introducing two chloro substituents into the 4 and 5
positions of the 3,6-DBSQ•− ligand.[9a] The average formal
oxidation state of the rhodium ion is +1.33, and the compound is
Results and Discussion
Synthesis and heat capacity
Dark brown needle crystals of compound 4 were prepared by a
redox reaction of [Rh4(CO)12] with 3,6-di-tert-butyl-4-nitro-1,2-
benzoquinone (3,6-DBBQ-4-NO2) in a mixture of hexane–toluene
(3:1, v/v). To examine whether the compound 4 undergoes a
phase transition, the heat capacity was measured from 7 K to 302
K by adiabatic calorimetry (Figure 1). A thermal anomaly was
observed at Ttrs = 176.3 K during heating, indicating the existence
of RT and LT phases. The enthalpy and entropy of the phase
transition were determined to be 580(16) J mol–1 and 3.39(11) J
K–1 mol–1, respectively. Generally, the transition entropy for the
a
paramagnetic semiconductor with a significantly higher
electrical conductivity of σRT = 17–34 S cm−1 at RT. However, the
band-filling control of 2 through the valence tautomerism by the
temperature could not be realized post-synthetically. On the other
hand, we have also found that the introduction of two methoxy
groups at the 4 and 5 positions of the 3,6-DBSQ•− ligand has led
to intriguing bistable multifunctionality and switchable very strong
ferromagnetic-to-antiferromagnetic
rhodium(I)–semiquinonato complex,
(MeO)2)(CO)2]∞ (3; 3,6-DBSQ-4,5-(MeO)2•− = 3,6-di-tert-butyl-4,5-
coupling
of
the
1D
[Rh(3,6-DBSQ-4,5-
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
Figure 2. Crystal structure of [Rh(3,6-DBDiox-4-NO2)(CO)2]∞ (4). a) Structure of the infinite chain built from a stack of planar complex molecules, and b) structural
view down the 1D chain showing an overlap mode of the complex molecules in the RT phase (288 K). c) Structure of an infinite chain, and d) structural view down
the 1D chain in the LT phase (101 K). Thermal ellipsoids are drawn at the 25 % probability level. Intermolecular C···O and C···C contacts shorter than the sum of
the van der Waals radii of the contacting atoms are represented by dashed lines. e) Packing diagram viewed down the b-axis, and f) packing diagram viewed
down the a-axis in the LT phase (101 K). Thermal ellipsoids are drawn at the 50 % probability level.
data and intensity data collection for all
structures are provided in the Supporting
Information (Table S1). The 1D chain structures
and packing diagrams in the RT and LT phases
are shown in Figure 2 and Figures S1, S2 in
Supporting Information, respectively. Although
compound 4 undergoes a first-order phase
transition, all the structures determined at each
temperature belong to the same orthorhombic
space group Pbca and the asymmetric unit
composed of only one complex molecule.
First, we describe the crystal structure of 4
at 288 K in the RT phase (Figure 2a,b). The
crystal structure is composed of neutral 1D
chains of the planar complex molecule stacked
in equal intervals with a staggered arrangement
along the a-axis, where the twisting angles
between adjacent molecules estimated from the
torsion angles of O1–Rh1–Rh1*–O2* and O2–
Rh1–Rh1*–O1* are approximately 139° at 288
K. The Rh1–Rh1* distance at 288 K is 3.0665(5)
Figure 3. a) Temperature dependence of the lattice parameters of [Rh(3,6-DBDiox-4-NO2)(CO)2]∞
Å. Regarding the intermolecular contacts in the
1D chain, the intermolecular C···O contacts
(3.079(4)–3.104(4) Å) and C···C contacts
(3.156(5)–3.377(5) Å) shorter than the sum of
the van der Waals radii of contacting atoms
(4). The a-axis is parallel to the 1D chain direction. The dashed line represents the phase transition
temperature. b) Temperature dependence of the anisotropic atomic displacement parameters
(ADPs; U11, U22, and U33) of the rhodium atom and the coordinated oxygen atoms of the
semiquinonato ligand and carbon atoms of the carbonyl ligands in 4.
order–disorder-type phase transition is known to be a large value,
exist between the dioxolene ligand and carbonyl ligand in the
adjacent stacked molecules; they are represented by dashed
lines in Figure 2a,b. The slight zigzag conformation is observed in
the Rh–Rh chain, and the bond angle of Rh1'–Rh1–Rh1* is
173.108(19)°. This result is due to the influence of these strong
intermolecular interactions. The positional disorder in the
dioxolene moiety was observed as follows (Figure S3 in
Supporting Information). The carbon atoms in the 4- and 5-
positions of the dioxolene ring are disordered in both the upper
and lower positions of the molecular plane, and their occupancy
factors were estimated to be 0.65 (C4, C5) and 0.35 (C7, C8),
respectively. Moreover, the nitro group binding to these carbon
such as ΔtrsS = R ln 2 = 5.76 J K–1 mol–1. Therefore, since the
observed transition entropy is smaller than 5.76 J K–1 mol–1, this
phase transition is presumably not of the order–disorder type.
However, the value is somewhat larger than that for the typical
displacive type phase transition.[17] This cause will be discussed
later.
X-ray crystal structure analyses
To investigate the details of the temperature dependence of the
crystal structure, the crystal structures of compound 4 were
determined at ten different temperatures from 288 K to 35 K by
single-crystal X-ray crystal structure analyses. Crystallographic
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
atoms is disordered over three sites with estimated occupancy
factors of 0.55 (N1), 0.10 (N3) and 0.35 (N2). This disorder model
was applied to all other temperature data sets. On the other hand,
the Rh1–Rh1* distance at 101 K in the LT phase is 2.9892(4) Å,
which is shortened by 0.077 Å compared to 288 K (Figure 2c,d).
The bond angle of Rh1'–Rh1–Rh1* is 173.371(16)°, which is very
slightly increased compared with that in the RT phase. The
intermolecular C···O contacts (2.980(4)–3.205(4) Å) and C···C
contacts (3.085(4)–3.307(4) Å) between dioxolene ligands and
carbonyl ligands in the adjacent stacked molecules are also
slightly shortened compared with those in the RT phase (Figure
2c,d).
The temperature dependence of the lattice parameters and
Rh–Rh distance of 4 are shown in Figure 3a and Figure S4 in
Supporting Information, respectively. In accordance with the
thermal anomaly observed in the heat capacity, the slope of each
lattice parameter has changed in the vicinity of 175 K. While the
a-axis parallel to the 1D Rh chain has elongated with the phase
transition, the c-axis has contracted. The Rh1–Rh1* distance at
288 K is 3.0665(5) Å, which is 0.21 Å shorter than that of [RhI(3,6-
DBSQ)(CO)2]∞ (1) (3.252(4) and 3.304(5) Å at 294−297 K)[16a] but
0.19 Å longer than that of [RhI,II(3,6-DBDiox-4,5-Cl2)(CO)2]∞ (2)
(2.8629(2)−2.8984(3) Å at 302 K) on average.[9a] Formally, there
is no bond between Rh atoms in the 1D Rh(I) chain because the
dz2 orbital involved in the bond is filled by two electrons. However,
when the rhodium atom is partially oxidized, the Rh–Rh distance
decreases owing to bond formation by removing the electrons
from the antibonding orbitals of the 1D Rh(I) chain. The shorter
Figure 4. Rh 3d5/2 and 3d3/2 core-level spectra for a) [Rh(3,6-DBDiox-4-
NO2)(CO)2]∞ (4) and b) [RhI(3,6-DBSQ)(CO)2]∞ (1). The solid lines are the
deconvoluted components and the sum of all the components.
Table 1. XPS Data for [RhI(3,6-DBSQ)(CO)2]∞ (1) and [Rh(3,6-DBDiox-4-
NO2)(CO)2]∞ (4)
binding energies, eV[a,b]
compound
Rh+ 3d5/2
Rh2+ 3d5/2
Rh+ 3d3/2
Rh2+ 3d3/2
309.31
(1.91)
308.94
(2.00)
309.17
(1.91)
313.99
(2.17)
314.88
(2.48)
313.86
(2.17)
1
4
4
310.49
(2.12)
315.18
(2.41)
Rh–Rh distance compared to that of
a
typical Rh(I)–
semiquinonato complex suggests the possibility that a slight
amount of charge transfer from the Rh(I) ion to semiquinonato
[a] Full width at half maximum (fwhm) values for peaks are given in
parentheses. [b] These values were corrected against the C1s peak using a
value of 284.6 eV
ligand
has
occurred
to
produce
a
Rh(I,II)–
atoms are dragged by the large-amplitude vibration of the rhodium
atoms and vibrate greatly along the 1D chain direction.
semiquinonato/catecholato mixed-valence state. The space
group in the LT phase is also Pbca, which is the same as the RT
phase, and no indication of lattice distortion, such as dimerization
due to the phase transition, was observed. However, a drastic
increase in the Rh–Rh distance (approximately 0.031 Å) is
observed in the vicinity of 175 K due to the phase transition.
The temperature dependence of the anisotropic atomic
displacement parameters (ADPs; U11, U22, and U33) of the
rhodium atom and the coordinated oxygen atoms of the dioxolene
ligand and carbon atoms of carbonyl ligands in 4 is shown in
Figure 3b. U11, U22, and U33 indicate the squares of the average
moving distance of the atom along the a-, b-, and c-axes from its
equilibrium position, respectively. For U11 corresponding to the 1D
chain direction, a drastic increase in all atoms was observed in
the vicinity of the transition point. In particular, the U11 of the
rhodium atom is remarkably increased by approximately 0.04 Å2
with the phase transition. However, U22 and U33 in the direction
perpendicular to the 1D chain increase almost linearly with
increasing temperature. For the cause of the drastic increase in
the temperature factor, either static structural disorder or dynamic
disorder of atoms may be considered. The observed increase in
U11 is attributed to the increase in the amplitude of the thermal
vibration since it occurs not in the LT phase but starts in the RT
phase where lattice distortion is generally eliminated. Therefore,
it is suggested that rhodium atoms in the RT phase vibrate with a
large amplitude along the 1D chain direction. Since the rates of
increase in U11 of the coordinated oxygen and carbon atoms are
smaller than that of the rhodium atom, it is considered that these
To discuss the lattice periodicity of the 1D chain in each phase,
X-ray diffraction photographs of 4 were taken in the LT and RT
phases (Figure S5 in Supporting Information). The X-ray
diffraction pattern in the RT phase exhibits diffuse lines at 0.5a*,
in addition to the main Bragg reflections, indicating that a new
periodic ordering along the a-axis (1D chain direction) with a 2-
fold repetition length of the a-axis has occurred in the RT phase.
Considering the large-amplitude vibration that appears only at the
RT phase, the observed diffuse scattering can be attributed to
dynamic distortion of the rhodium atoms. From these results, as
the possible phonon mode expected in the RT phase, we propose
phonon mode 2, which satisfies the two-fold periodicity (see
Figure S6c in Supporting Information).
X-ray photoelectron spectroscopy (XPS)
To examine the valence state of the rhodium ions, we measured
the XPS spectrum of 4 at RT together with that of [RhI(3,6-
DBSQ)(CO)2]∞ (1) (Figure 4 and Table 1). Rh(I)–semiquinonato
complex 1 shows sharp Rh 3d5/2 and 3d3/2 peaks, whereas the full
width at half maximum (fwhm) values of the Rh 3d5/2 and 3d3/2
peaks of compound 4 are slightly larger than those of 1. When the
curve deconvolution using a Gaussian–Lorentzian line shape fit is
performed by assuming compound 4 to be in the mixed-valence
state composed of Rh(I) and Rh(II), these peaks could be
resolved into two sets of Rh+ 3d5/2,3/2 and Rh2+ 3d5/2,3/2 doublets.
The peak area ratio of the Rh+ 3d5/2/Rh2+ 3d5/2 and Rh+ 3d3/2/Rh2+
3d3/2 doublets is approximately 10:1, which suggests that the
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
Figure 6. Temperature dependence of the electrical resistivity ρ and
electrical conductivity σ of [Rh(3,6-DBDiox-4-NO2)(CO)2]∞ (4) measured
along the chain axis a. The green solid lines indicate the fits to the data
using the Arrhenius equation.
Figure 5. Vis–NIR–mid-IR spectrum of [Rh(3,6-DBDiox-4-NO2)(CO)2]∞ (4) in
the solid-state (KBr pressed disk), together with those of [RhI(3,6-
DBSQ)(CO)2]∞ (1) and [RhI,II(3,6-DBDiox-4,5-Cl2)(CO)2]∞ (2). The Vis–NIR
and mid-IR spectra were connected smoothly at approximately 4000–5000
cm–1
.
formal oxidation state of rhodium ions at RT is approximately
Rh1.09+ and the ratio of [RhI(3,6-DBSQ-4-NO2)(CO)2] to [RhII(3,6-
DBCat-4-NO2)(CO)2] is approximately 10:1, where 3,6-DBSQ-4-
NO2 and 3,6-DBCat-4-NO2 indicate 3,6-di-tert-butyl-4-nitro-1,2-
benzosemiquinonato and 3,6-di-tert-butyl-4-nitrocatecholato,
respectively.
Electronic absorption spectra
To better understand the electronic structure, the visible-near-
infrared-mid-infrared (Vis–NIR–mid-IR) spectrum of
4 was
measured, together with the spectra of 1 and 2 (Figure 5). Each
compound exhibits intense absorption expanding from the NIR to
mid-IR region, which is the absorption that appears by adopting a
1D chain structure. The absorption band for 1 with the maximum
at 6700 cm–1 is assigned to the charge-transfer absorption from a
filled 1D d-band to the vacant semiquinonato π* levels.[10d] The
absorption band for 2 with the maximum at 8000 cm–1 extends
widely from the NIR to mid-IR region compared with that of 1 and
is considered to consist of several overlapping bands, including
the intervalence charge-transfer transitions of Rh2+ ← Rh+ and
π*(SQ•–) ← π*(Cat2–) transition, in addition to the π*(SQ•–) ← 1D
Figure 7. Temperature dependence of the molar magnetic susceptibility in the
form of χMT vs. T plots for [Rh(3,6-DBDiox-4-NO2)(CO)2]∞ (4) measured on a
polycrystalline sample under ordinary pressure and several static pressures
up to 8.4 kbar. Each measurement was performed during a cooling process.
on the temperature down to 220 K and surprisingly decreases to
the vicinity of the phase transition temperature and then increases
rapidly in the LT phase. The observed decrease in the resistivity
and the almost temperature-independent resistivity may indicate
that 4 in the RT phase is in a metallic state. However, the electrical
conductivity at RT is about two orders of magnitude smaller than
the metallic 1D d-electronic metal complex conductors.[18a,19] This
may be due to the fact that the carrier concentration is quite low
due to the very small degree of partial oxidation and the electrons
flowing in the 1D Rh chain are strongly affected by the large-
amplitude vibrations of Rh atoms. Further studies will still be
required to reveal whether the RT phase is in a metallic state or
not. Another important feature of the electrical property of 4 is the
crossover from a relatively high conducting state with a small
activation energy of Ea = 13.1 meV (299–230 K) in the RT phase
to a semiconductor with a large activation energy of Ea = 295 meV
(167–150 K) in the LT phase. The observed relatively high
electrical conductivity (~10–2 S cm–1) and a small activation
energy (13.1 meV) can be attributed to the Rh(I,II) mixed-valence
state, in agreement with the XPS results. On the other hand,
considering the typical semiconducting behavior with a large
2
d-band (dz ) transition.[9a] This broad band extends further from
the mid-IR to the far-IR region, indicating the presence of low-
energy electronic excitation in the Rh chain associated with –Rh+–
Rh2+– ← –Rh2+–Rh+–.[18] In fact, 2 exhibits a high electrical
conductivity of 17–34 S cm–1.[9a] The absorption band for 4 with
the maximum at 6700 cm–1 is clearly broader than that of 1 and
extends to the far-IR region, as in the case of the band observed
for 2, although the degree of expansion of the band is small.
Therefore, the origin of this absorption is considered to be the
same as those of 2, but the reason for the less broad feature is
that the ratio of Rh(II)–catecholato to Rh(I)–semiquinonato in 4 is
small. The broadening of the band reaching the far-IR region
suggests that compound 4 will exhibit a relatively high conductivity.
Electrical properties
The temperature dependence of the electrical resistivity of 4 was
measured along the chain axis a using a DC four-probe technique
(Figure 6). The electrical conductivity at RT is a relatively high
value of 2.9 × 10–2 S cm–1. The resistivity does not depend much
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
Figure 8. a) Frontier molecular orbitals and energies (eV) for the monomer and dimer models derived from the UB3LYP level calculation based on the structure
determined at 129K. The MIDI + polarization function, 6-31G*, and 6-31+G* basis functions were used for Rh atom, 3,6-DBSQ-4-NO2 ligand, and CO ligands,
respectively. b) Spin-density isosurface (threshold = 0.004) of the monomer model at 129 K derived from the UB3LYP calculation. Blue and green lobes represent
positive and negative spin densities, respectively.
activation energy in the LT phase, although complex molecules
are stacked with equal spacing, it is natural to consider that the
LT phase is a Rh(I)–semiquinonato state rather than a mixed-
valence state. This result is the first example succeeded in band-
filling control by using valence tautomerism.
neighbor magnetic interaction is very large at J/kB = –200.1 K.[10d]
This magnetic interaction is expected to have a strong effect even
in 4. Therefore, the details of the magnetic interactions in 4 could
not be well discussed. It is, however, apparent that not only the
conductivity but also the magnetism varies dramatically due to the
change in the electronic state accompanying the phase transition.
As shown in Figure 7, when pressure is applied, the transition
point disappears, which is probably because the transition
temperature shifts to a higher temperature by applying pressure.
A similar shift of the transition point to RT with applying pressure
has also been found in compound 3, in which the transition point
shifts to RT at a pressure of 1.6 kbar.[10d] As the applied pressure
is increased, the swelling of the χMT value near 170 K decreases,
and then the χMT value monotonically decreases with decreasing
temperature under pressures above 1.9 kbar. The disappearance
of the transition point with applied pressure is also confirmed by
the temperature dependence of the electrical conductivity under
pressure (Figure S8 in Supporting Information), and it is
considered that the LT phase shifts to room temperature by
applying the pressure. This behavior is because the LT phase with
a short Rh–Rh distance is stabilized under pressure.
Magnetic properties
The temperature dependence of the molar magnetic susceptibility
of 4 was examined under ordinary pressure and several static
pressures up to 8.4 kbar (Figure 7). Compound 4 exhibits a
dramatic change in magnetic behavior with the phase transition.
The χMT value at 300 K under ordinary pressure is 0.200 emu K
mol–1, which is 53 % of the spin-only value of S = 1/2 (0.375 emu
K mol–1). The χMT value increases very slightly with decreasing
temperature in the RT phase and then increases rapidly with the
phase transition up to the maximum of 0.603 emu K mol–1 at 140
K, before rapidly decreasing down to 0.050 emu K mol–1 at 2 K.
The observed magnetic behavior is similar to that observed for
compound 3 but does not exhibit a striking increase in χMT value,
as observed in the LT phase of 3.[10d]
Although compound 4 is an equally spaced molecule, the
magnetic data in the RT phase show a significantly different
behavior from the S = 1/2 1D Heisenberg antiferromagnetic chain
model (Figure S7 in Supporting Information). For this reason, it is
first considered that a little less than 10 % of this compound has
been changed to the Rh(II)–catecholato complex, as suggested
by the XPS spectrum. Second, although the χMT value at the RT
is 53 % of the spin-only value, the χMT value slightly increases
with decreasing temperature, suggesting that ferromagnetic
interactions coexist in addition to the strong antiferromagnetic
interactions. In the DFT calculations performed for the trimer
model for compound 3, we have already reported that the next-
Density functional theory (DFT) calculations
To obtain information regarding the electronic structure and
magnetic interaction for 4, we performed DFT calculations based
on the monomer and dimer models constructed from the atomic
coordinates of the RT and LT phases. Selected frontier orbitals
and their energies are shown in Figure 8a and Figure S9 in
Supporting Information. Here, the SOMO, SUMO, HOMO, and
LUMO denote the singly occupied, singly unoccupied, highest-
occupied, and lowest-unoccupied molecular orbitals, respectively,
and the molecular orbitals were obtained as the α and β orbitals
6
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
Figure 9. Band structure model predicted from the energy levels of the monomer and dimer models, and changes in energy levels accompanying the large-
amplitude vibrations of rhodium atoms.
due to the spin-unrestricted formalism. The spin densities of the
monomer models based on the structures at 129 K and 228 K are
ꢄꢂꢃ– ꢄꢄꢃ
ꢂꢀꢁ
=
ꢅ ꢄꢃ
ꢅ ꢂꢃ
shown in Figure 8b and Figure S10 in Supporting Information,
respectively. In either case, the spin density is delocalized to the
whole molecule, except for the t-butyl groups, but is not very
present on the rhodium atom. Since no obvious difference in the
electronic state due to the difference in the temperatures at 129
K and 228 K was confirmed, the results for the 129 K data will be
described.
(〈
2 ꢃ
〉
〈
〉 )
– ꢃ
where EX and <S2>X represent the total energies and the <S2>
values of the spin state X (X = HS or LS), respectively. Therefore,
the difference in the magnetic behavior between the LT and RT
phases could not be deduced by the DFT calculation based on
the dimer models.
The α-SOMO and β-SUMO of the monomer model are
predominantly the semiquinonato π* orbitals with some
contributions from the Rh dπ orbital and carbonyl ligands. The
HOMOs primarily comprise a semiquinonato π* orbital with some
contribution from a Rh dπ orbital. The dz2 orbital that consists of
the 1D d-band greatly contributes to the HOMO–1s. On the other
hand, unlike the monomer model, the HOMO of the dimer model
is composed of the HOMO–1s of the monomer model derived
from the dz2 orbitals, and the SOMO–HOMO energy level
conversion has occurred. This conversion is due to a large orbital
splitting of dσ and dσ* orbitals caused by the large overlap of dz2
orbitals between monomers. Moreover, the LUMO of the dimer
model originating from the SUMO of the monomer exists at
approximately –4.4 eV, and its energy level is almost the same as
the SUMO level of the monomer because there is no direct
overlap between semiquinonato ligands.
To estimate the effective exchange integral Jab, we calculated
the energies of the high-spin (HS) triplet state and the low-spin
(LS) spin-singlet state of the dimer units at 228 K and 129 K (Table
S2 in Supporting Information).[20] The effective exchange integrals
Jab/kB at 228 K and 129 K estimated from these energy values
based on the following equation were –433 K and –462 K,
respectively, which suggests that both intrachain magnetic
interactions are strongly antiferromagnetic:
Dynamic mixed-valence state
Although the DFT calculations have only been performed up to
the dimer model, it is expected that the splitting between the dσ
and dσ* orbitals becomes large with an increasing number of
interacting complex molecules. The band structure model
predicted from the DFT calculations based on the monomer and
dimer models is shown in Figure 9. In the polymeric structure, the
further overlap of dz2 orbitals is expected to form much wider 1D
d-bands and reduce the HOMO–LUMO gap since the LUMO
comprised of a semiquinonato π* orbital exists at almost the same
level as the SUMO of the monomer. Judging from the typical
semiconducting behavior, the LT phase is considered to be a
Rh(I)–semiquinonato complex, and therefore, the upper part of
the filled d-band is expected to be relatively far from the empty π*
level of the semiquinonato ligand. As already described, there
was no significant difference in the HOMO–LUMO gap estimated
for the dimer models at 129 K and 228 K. However, the obvious
structural difference between the LT and RT phases is the large-
amplitude vibration of the rhodium atoms and its surrounding
coordinated atoms in the 1D chain direction, which is only
observed in the RT phase. Since the influence of the molecular
vibration is not included in the DFT calculations, it is considered
that no difference between the two phases was observed.
The thermal vibration of rhodium atoms contains the
contribution of lattice vibration (including the libration and
vibration of the whole molecule), in addition to the large-amplitude
ꢀ = – ꢁꢂꢀꢁ ꢃꢀ ⋅ ꢃꢁ
ꢀ,ꢁ
7
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
vibration of the rhodium atoms along the 1D chain. As a very
rough approximation, assuming that the mean square
displacement values in another two directions perpendicular to
the 1D chain (U22 = 0.02948 (14), U33 = 0.03073 (13) Å2) are the
contributions from lattice vibration, its magnitude is estimated
from the average of these values. Subtracting this averaged value
from the mean square displacement in the 1D chain direction (U11
= 0.0676 (4) Å2) and calculating the square root of it, the deviation
in the 1D chain direction from the equilibrium position of the
rhodium atom is estimated to be 0.19 Å. As already mentioned
above, this large-amplitude vibration of the Rh–Rh bonds is
tentatively inferred to be a phonon mode like the phonon mode 2
(Figure S6c in Supporting Information), due to the existence of the
two-fold periodicity of the –Rh–Rh– period. Considering the effect
of the large-amplitude vibration of rhodium atoms on the thermally
induced metal–ligand electron transfer (valence tautomerism) and
vice versa, we propose a band structure model shown in Figure
9. When the Rh–Rh distance is largely reduced due to the
increase in the amplitude of the Rh–Rh stretching vibration by
thermal excitation, the 1D d-band expands, and the upper part of
the band is very close to the vacant semiquinonato π* level. This
situation will allow the thermally induced electron transfer from the
rhodium(I) ion to the semiquinonato ligand, resulting in the
Rh(I,II)–semiquinonato/catecholato mixed–valence state, which
further causes a decrease in the Rh–Rh distances by removing
the electrons from the antibonding orbitals of the 1D d-band. Thus,
it is considered that the coupling of the thermally induced metal–
ligand electron transfer (valence tautomerism) with the large-
amplitude vibration of the Rh–Rh bonds causes a dynamic mixed-
valence state, producing approximately 10 % of [RhII(3,6-DBCat-
4-NO2)(CO)2]. Finally, as described above, the present phase
transition exhibited a comparatively larger transition entropy than
that for the typical displacive type phase transition.[17] Such a large
entropy change would arise from the softening of the vibrational
modes due to the phase transition, which is consistent with the
above situation.
dynamic mixed-valence state that appears by the coupling of
electron transfer with molecular vibrations.
Experimental Section
General
All syntheses were performed under an argon atmosphere using standard
vacuum-line and Schlenk techniques. All solvents were dried using
appropriate drying agents and were freshly distilled under argon before
use.[21] 1H and 13C NMR spectra were obtained on a JEOL ECA-600
spectrometer using tetramethylsilane (TMS) as an internal standard.
Elemental analyses were performed at the Center for Organic Elemental
Microanalysis of the Graduate School of Pharmaceutical Sciences, Kyoto
University.
Syntheses
Dodecacarbonyltetrarhodium(0) ([Rh4(CO)12])[22], [Rh(3,6-DBSQ)(CO)2]∞
(1)[10d]
,
and [Rh(3,6-DBDiox-4,5-Cl2)(CO)2]∞ (2)[9a] were prepared
according to published procedures. The synthesis of 3,6-di-tert-butyl-4-
nitro-1,2-benzoquinone (3,6-DBBQ-4-NO2) was performed after improving
the literature method and once isolating the intermediate of 3,6-di-tert-
butyl-5,6-dinitro-3-cyclohexene-1,2-dione.[23]
Synthesis of 3,6-di-tert-butyl-5,6-dinitro-3-cyclohexene-1,2-dione: To
a solution of 3,6-di-tert-butylcatechol (2.202 g, 9.904 mmol) dissolved in
25 mL of acetic acid, a mixture of 2.00 mL (43.8 mmol) of nitic acid
(S.G.=1.38) and 5 mL of acetic acid was slowly added dropwise, with
stirring in an oil bath, at room temperature. The resulting black solution
was heated at 50 °C for 7 h to produce a bright yellowish orange
suspension. After cooling to room temperature, the suspension was cooled
overnight at 0 °C. The precipitate was filtered through a glass filter. The
collected solid was rinsed with ice-cooled hexane and dried in vacuo to
afford a bright yellow powder (902 mg, 29%). 1H NMR (600 MHz, CDCl3):
δ=1.15 (s, 9H, t-butyl), 1.25 (s, 9H, t-butyl), 6.29 (d, J = 7.6 Hz, 1H, –
CH(NO2)–CH=), 7.17 (d, J = 7.6 Hz, 1H, –CH(NO2)–CH=).
Synthesis of 3,6-di-tert-butyl-4-nitro-1,2-benzoquinone (3,6-DBBQ-4-
NO2): To a solution of 3,6-di-tert-butyl-5,6-dinitro-3-cyclohexene-1,2-dione
(1.706 g, 5.462 mmol) in 45 mL of dichloromethane, a solution of
diethylamine (0.7 mL, 6.77 mmol) in 10 mL of dichloromethane was slowly
added dropwise, with stirring, at room temperature. The color of the
reaction mixture changed from bright yellow to black. After 1 h, the black
reaction mixture was washed with 5 % NaCl solution (100 mL × 3) until the
washings became neutral, and then the aqueous phase was extracted with
dichloromethane (100 mL × 2). The combined organic phases were dried
over anhydrous Na2SO4. After filtration, the solution was concentrated until
dryness using a rotatory evaporator. The resulting residue was purified by
silica gel column chromatography using hexane–diethyl ether (10:1, v/v)
as the eluent to render 3,6-DBBQ-4-NO2 as a dark green powder. The
crude product was recrystallized twice by dissolving it in a minimal amount
of diethyl ether, adding hexane with a forming layer, and then cooling the
solution overnight in a freezer at –60 °C to produce pure 3,6-DBBQ-4-NO2
as a dark green powder (815 mg, 72%). 1H NMR (600 MHz, CDCl3):
δ=1.27 (s, 9H, t-butyl), 1.38 (s, 9H, t-butyl), 6.39 (s, 1H, –C(NO2)=CH–);
13C NMR (600 MHz, CDCl3): δ=28.7 (CH3), 28.8 (CH3), 35.6 (CMe3), 36.5
(CMe3), 131.2 (C(5)H), 139.3 (ring C(3)), 151.8 (ring C(6)), 153.1
(C(4)NO2), 178.8 (C(1)=O), 182.5 (C(2)=O); elemental analysis calcd for
C14H19NO4: C 63.38, H 7.22, N 5.28, found: C 63.22, H 7.21, N 5.29.
Conclusions
As mentioned in the introduction, since VT complexes have two
nearly degenerate electronic states, the interconversion of
electrical, magnetic, or optical properties using the switching of
valence tautomerism by external stimuli has been vigorously
studied. However, there is currently no example of a dynamic
mixed-valence state originating from the combination of valence
tautomerism with large-amplitude molecular vibrations. The
present compound 4 is the first example of the dynamic mixed-
valence state originating from the metal–ligand electron transfer
(valence tautomerism) coupled with the large-amplitude vibration
of the metal–metal bonds. We have also reported a 1D iodido-
bridged mixed-valence diplatinum(II,III) complex [Pt2(EtCS2)4I]∞
that exhibits a metallic state above 204 K originating from the
electron transfer between diplatinum units coupled with the large-
amplitude vibration of the bridging iodido ions.[19] Electron transfer
coupled with molecular vibrations is very important since it may
create new research fields in the chemistry and physics of
dynamic molecular crystals based on dynamic electronic states
arising from the mixing of different electronic states by nuclear
displacements. We will continue to search for molecular materials
that exhibit unique physical properties originating from the
Synthesis of [Rh(3,6-DBDiox-4-NO2)(CO)2]∞ (4): Bubnov et al. reported
the synthesis of compound 4 by the reaction of Tl(3,6-DBSQ-4-NO2) with
[RhCl(CO)2]2 in toluene.[24] We prepared compound 4 using our developed
redox reaction of [Rh4(CO)12] with the corresponding benzoquinone
derivative. [Rh4(CO)12] (37.35 mg, 50.0 μmol) and 3,6-DBBQ-4-NO2 (53.35
mg, 201 μmol) were placed in a small cut vial (10 mm i.d., 30 mm height)
and a large sealable vial (15.0 mm i.d., 50 mm height), respectively. The
small vial was placed inside the large sealable vial such that 3,6-DBBQ-4-
8
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
NO2 was not buried under the small vial. After filling the large sealable vial
with a mixture of hexane–toluene (3:1, v/v), well-shaped dark brown
needle crystals of 4 grew within 2 weeks via the slow diffusion of
[Rh4(CO)12] and 3,6-DBBQ-4-NO2. The resulting needle crystals of 4 were
collected by suction filtration and were rinsed with ice-cooled hexane
(57.58 mg, 68%). Elemental analysis calcd for C16H19NO6Rh: C 45.30, H
4.51, N 3.30, found: C 45.15, H 4.50, N 3.31.
UV–Visible–Near-IR and IR Spectroscopies
UV-visible-near-IR spectra of complexes 1, 2 and 4 as KBr pressed disks
were recorded on a Hitachi U-3500 spectrophotometer equipped with a
60-mm-i.d. integrating-sphere apparatus. IR spectra were recorded as KBr
pressed disks on a JASCO FT/IR-4100 spectrophotometer.
Magnetic Measurements
X-ray Photoelectron Spectroscopy (XPS)
Magnetic measurements were performed using
MPMS-5SH or MPMS-XL7 SQUID magnetometer.
a
Quantum Design
polycrystalline
XPS data were obtained on
a VG Scientific ESCALAB 220i-XL
A
spectrometer equipped with a non-monochromatized Mg Kα radiation
source (1253.6 eV) operated at 15 kV and 20 mA. Measurements were
conducted in the 10−6 Pa pressure range. Samples were ground into
powders and spread onto the conductive adhesive tape attached to
sample holders. The absence of X-ray beam effects was verified by the X-
ray power dependence of the XPS spectra. The carbon 1s binding energy
(284.6 eV) was used to calibrate the binding energies. Curve-fitting
analysis was performed with the iterative least squares computer program
XPSPEAK41 using a combination of Gaussian and Lorentzian line shapes.
sample (27.64 mg) of 4 held in a polyethylene film (19.97 mg) inside a
plastic straw was used. Magnetic susceptibility measurements on the
polycrystalline sample were performed from 350 K to 2 K under an applied
magnetic field of 1 kOe. The diamagnetic contributions of the polyethylene
film and plastic straw were measured from 350 to 2 K under an applied
magnetic field of 10 kOe, and the corresponding diamagnetic contribution
obtained by multiplying the measured value by a factor of 1000/10000 was
subtracted from the raw data. The paramagnetic component of the molar
magnetic susceptibility (χM) was obtained after subtraction of the
calculated diamagnetic core contribution (–181.5 × 10–6 emu mol–1
)
estimated from the measured diamagnetic susceptibility of 3,6-di-tert-
butyl-4-nitro-1,2-benzoquinone (3,6-DBBQ-4-NO2) (–121.5 × 10–6 emu
mol–1) and Pascal’s constants for a rhodium(0) atom and two carbon
monoxide ligands. Magnetic measurements under various static pressures
up to 8.4 kbar were performed using a piston-cylinder-type pressure cell
constructed from a Cu–Be alloy. The crystals (10.75 mg) were dispersed
into a pressure-transmitting medium, Daphne 7373, with a piece of tin as
the manometer in a Teflon bucket. The actual pressure was calibrated
using the superconducting transition temperature of tin.
Calorimetry
Heat capacity measurements were performed in the temperature range of
7–302 K with a laboratory-made LT adiabatic calorimeter for small
samples.[25] A sample of 0.11413 g after the buoyancy correction was
loaded into a gold-plated copper cell and sealed under helium gas at
ambient pressure using an indium gasket. The helium gas functions as a
heat-exchange medium. Thermometry was performed with a rhodium-iron
alloy resistance thermometer (nominal 27 Ω, Oxford Instruments)
calibrated on the basis of the international temperature scale of 1990 (ITS-
90).
Electrical Resistivity Measurements
X-ray Crystal Structure Analyses
Direct current electrical resistivity measurements along the 1D chain
direction were performed on several single crystals by the four-probe
method. The electrical contacts between the sample and 15 μm o.d. gold
wires were created using gold paint. The sample temperature was
controlled using a closed-cycle helium refrigerator (Iwatani, CryoMini
D105). The electrical resistivity under static pressure up to 12 kbar (1.2
GPa) was measured using a clamp-type pressure cell purchased from
Kitano Seiki Co., Ltd.
Single-crystal X-ray diffraction data were collected at 10 temperatures
ranging from 288 K to 35 K on two different crystals. Crystals of 4 were
mounted on glass fibers using Apiezon grease. X-ray diffraction
experiments in-house were performed using Mo Kα radiation of λ =
0.71075 Å on a Rigaku Saturn 724 HG CCD detector with a MicroMax007
HF DW X-ray generator installed at Okayama University of Science. The
temperature of the crystal was controlled by a Rigaku variable-temperature
apparatus based on a cold nitrogen gas stream method. X-ray diffraction
experiments using a synchrotron radiation source (λ = 0.68869 Å) were
performed at the Photon Factory (PF) BL-8A in the High Energy
Accelerator Research Organization (KEK), Japan. The diffraction data
were collected on a Rigaku imaging-plate area detector system equipped
with a cold helium gas flow-type system for the LT experiments. Data
collection, cell refinements, indexing, peak integrations, and the scaling of
the diffraction data were performed using CrystalClear software for the in-
house data and RAPID AUTO software for the synchrotron radiation data.
Lorentz, polarization, and multi-scan absorption corrections were applied
to the intensity data.
Density functional theory (DFT) calculations
We performed DFT calculations based on the monomer and dimer models
of 4 in the RT and LT phases constructed from the atomic coordinates at
228 K and 129 K, respectively. All calculations were performed using an
unrestricted B3LYP (UB3LYP) method with basis sets of MIDI
+
polarization function, 6-31G*, and 6-31+G* for the Rh atoms, 3,6-DBSQ-
4-NO2 ligand, and CO ligands, respectively. All calculations were carried
out using the Gaussian 09 program.[28]
The structures were solved by a direct method using the SHELXS
program and refined by a full-matrix least squares method on F2 for all
reflections utilizing the SHELXL-2018/1 or -2018/3 program.[26] The X-ray
analyses were performed using the free GUI software Yadokari-XG.[27] All
non-hydrogen atoms were refined anisotropically, whereas hydrogen
atoms on the methyl groups were placed in geometrically calculated
positions using a riding model and were refined using the isotropic
displacement parameters derived from their parent atoms. Further details
of the crystal data and crystal structure refinements for 4 are available in
the Supporting Information. Deposition Numbers CCDC-2027122–
2027131 contain the supplementary crystallographic data for 206 K, 101
K, 156 K, 173 K, 288 K, 228 K, 129 K, 254 K, 153 K, and 35 K data,
respectively. These data are provided free of charge by the joint
Cambridge Crystallographic Data Centre and Fachinformationszentrum
Karlsruhe Access Structures service www.ccdc.cam.ac.uk/structures.
Acknowledgements
This study was supported by JSPS KAKENHI Grant Number JP
JP23110723 (Molecular Degree of Freedom), 17H05389
(Coordination Asymmetry). The synchrotron X-ray diffraction
experiment was performed under the approval of the Photon
Factory Program Advisory Committee (Proposal No. 2014G710).
We also thank the Instrument Center at the Institute for Molecular
Science, for the use of an X-ray photoelectron spectrometer (VG
Scientific ESCALAB 220i-XL spectrometer) and
magnetometer (Quantum Design MPMS-XL7).
a SQUID
9
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
H. Mori, J. Am. Chem. Soc. 2014, 136, 12184–12192; c) K. Nakasuji, K.
Sugiura, T. Kitagawa, J. Toyoda, H. Okamoto, K. Okaniwa, T. Mitani, H.
Yamamoto, I. Murata, A. Kawamoto, J. Tanaka, J. Am. Chem. Soc. 1993,
113, 1862–1864.
Conflict of interest
The authors declare no conflict of interest.
[7]
[8]
a) D. Z. S. Yao, S. Q. Wu, Y. Kitagawa, S. Q. Su, Y. G. Huang, G. L. Li,
Z. H. Ni, H. Nojiri, Y. Shiota, K. Yoshizawa, S. Kang, S. Kanegawa, O.
Sato, Angew. Chem. Int. Ed. 2017, 56, 717–721; b) J. Harada, M. Ohtani,
Y. Takahashi, T. Inabe, J. Am. Chem. Soc. 2015, 137, 4477–4486.
a) T. Tezgerevska, K. G. Alley, C. Boskovic, Coord. Chem. Rev. 2014,
268, 23–40; b) E. Evangelio, D. Ruiz-Molina, C. R. Chimie 2008, 11,
1137–1154; c) E. Evangelio, D. Ruiz-Molina, Eur. J. Inorg. Chem. 2005,
2957–2971; d) D. N. Hendrickson, C. G. Pierpont, in Topics in Current
Chemistry, Vol. 234 (Ed.: H. A. G. Philipp Gütlich), Springer-Verlag 2004,
pp. 63–95; e) C. G. Pierpont, Coord. Chem. Rev. 2001, 216–217, 99–
125; f) P. Gütlich, A. Dei, Angew. Chem. Int. Ed. Engl. 1997, 36, 2734–
2736; g) C. G. Pierpont, C. W. Lange, Prog. Inorg. Chem. 1994, 41, 331–
442; h) G. A. Abakumov, V. K. Cherkasov, M. P. Bubnov, O. G. Ellert, Z.
V. Dobrokhotova, L. N. Zakharov, Y. T. Struchkov, Dokl. Akad. Nauk
SSSR 1993, 328, 332–335; i) G. A. Abakumov, V. K. Cherkasov, A. V.
Lobanov, Dokl. Akad. Nauk SSSR 1982, 266, 361–363; j) R. M.
Buchanan, C. G. Pierpont, J. Am. Chem. Soc. 1980, 102, 4951–4957.
a) M. Mitsumi, H. Goto, S. Umebayashi, Y. Ozawa, M. Kobayashi, T.
Yokoyama, H. Tanaka, S.-i. Kuroda, K. Toriumi, Angew. Chem. Int. Ed.
2005, 44, 4164−4168; b) O.-S. Jung, C. G. Pierpont, J. Am. Chem. Soc.
1994, 116, 2229–2230.
Keywords: conducting materials • mixed-valent compounds •
valence tautomerism • vibronic interaction • large-amplitude
vibration
[1]
a) Z. S. Yao, Z. Tang, J. Tao, Chem. Commun. 2020, 56, 2071–2086; b)
Y. S. Meng, T. Liu, Acc. Chem. Res. 2019, 52, 1369–1379; c) J.
Ferrando-Soria, J. Vallejo, M. Castellano, J. Martinez-Lillo, E. Pardo, J.
Cano, I. Castro, F. Lloret, R. Ruiz-Garcia, M. Julve, Coord. Chem. Rev.
2017, 339, 17–103; d) O. Sato, Nat. Chem. 2016, 8, 644–656; e) E.
Coronado, G. Minguez Espallargas, Chem. Soc. Rev. 2013, 42, 1525–
1539; f) V. I. Minkin, Russ. Chem. Bull. 2008, 57, 687–717; g) O. Sato,
J. Tao, Y.-Z. Zhang, Angew. Chem. Int. Ed. 2007, 46, 2152–2187; h) A.
Dei, D. Gatteschi, C. Sangregorio, L. Sorace, Acc. Chem. Res. 2004, 37,
827–835; i) O. Kahn, J. P. Launay, Chemtronics 1988, 3, 140–151.
a) J. Chen, Y. Sekine, A. Okazawa, H. Sato, W. Kosaka, H. Miyasaka,
Chem. Sci. 2020, 11, 3610–3618; b) J. Chen, Y. Sekine, Y.
Komatsumaru, S. Hayami, H. Miyasaka, Angew. Chem. Int. Ed. 2018, 57,
12043–12047; c) J. A. DeGayner, K. Wang, T. D. Harris, J. Am. Chem.
Soc. 2018, 140, 6550–6553; d) J. Guasch, L. Grisanti, S. Jung, D.
Morales, G. D’Avino, M. Souto, X. Fontrodona, A. Painelli, F. Renz, I.
Ratera, J. Veciana, Chem. Mater. 2013, 25, 808−814; e) H. Miyasaka, N.
Motokawa, T. Chiyo, M. Takemura, M. Yamashita, H. Sagayama, T. H.
Arima, J. Am. Chem. Soc. 2011, 133, 5338–5345; f) T. Mahfoud, G.
Molnar, S. Bonhommeau, S. Cobo, L. Salmon, P. Demont, H. Tokoro, S.
I. Ohkoshi, K. Boukheddaden, A. Bousseksou, J. Am. Chem. Soc. 2009,
131, 15049–15054; g) S. Horiuchi, Y. Okimoto, R. Kumai, Y. Tokura,
Science 2003, 299, 229–232; h) N. Kojima, W. Aoki, M. Itoi, Y. Ono, M.
Seto, Y. Kobayashi, Y. Maeda, Solid State Commun. 2001, 120, 165–
170; i) J. B. Torrance, J. E. Vazquez, J. J. Mayerle, V. Y. Lee, Phys. Rev.
Lett. 1981, 46, 253–257.
[2]
[9]
[10] a) G. K. Gransbury, B. N. Livesay, J. T. Janetzki, M. A. Hay, R. W. Gable,
M. P. Shores, A. Starikova, C. Boskovic, J. Am. Chem. Soc. 2020, 142,
10692–10704; b) B. Li, Y. M. Zhao, A. Kirchon, J. D. Pang, X. Y. Yang,
G. L. Zhuang, H. C. Zhou, J. Am. Chem. Soc. 2019, 141, 6822–6826; c)
S. Kanegawa, Y. Shiota, S. Kang, K. Takahashi, H. Okajima, A.
Sakamoto, T. Iwata, H. Kandori, K. Yoshizawa, O. Sato, J. Am. Chem.
Soc. 2016, 138, 14170–14173; d) M. Mitsumi, T. Nishitani, S. Yamasaki,
N. Shimada, Y. Komatsu, K. Toriumi, Y. Kitagawa, M. Okumura, Y.
Miyazaki, N. Górska, A. Inaba, A. Kanda, N. Hanasaki, J. Am. Chem.
Soc. 2014, 136, 7026–7037; e) C. Carbonera, A. Dei, J. F. Létard, C.
Sangregorio, L. Sorace, Angew. Chem. Int. Ed. 2004, 43, 3136–3138; f)
D. M. Adams, B. Li, J. D. Simon, D. N. Hendrickson, Angew. Chem. Int.
Ed. 1995, 34, 1481–1483; g) David M. Adams, Andrea Dei, Arnold L.
Rheingold, D. N. Hendrickson, J. Am. Chem. Soc. 1993, 115, 8221–8229.
[11] A. Lannes, Y. Suffren, J. B. Tommasino, R. Chiriac, F. Toche, L. Khrouz,
F. Molton, C. Duboc, I. Kieffer, J. L. Hazemann, C. Reber, A. Hauser, D.
Luneau, J. Am. Chem. Soc. 2016, 138, 16493–16501.
[3]
a) A. Diaconu, S. L. Lupu, I. Rusu, I. M. Risca, L. Salmon, G. Molnar, A.
Bousseksou, P. Demont, A. Rotaru, J. Phys. Chem. Lett. 2017, 8, 3147–
3151; b) K. Bairagi, O. Iasco, A. Bellec, A. Kartsev, D. Li, J. Lagoute, C.
Chacon, Y. Girard, S. Rousset, F. Miserque, Y. J. Dappe, A. Smogunov,
C. Barreteau, M. L. Boillot, T. Mallah, V. Repain, Nat. Commun. 2016,
7:12212; c) T. Matsumoto, G. N. Newton, T. Shiga, S. Hayami, Y. Matsui,
H. Okamoto, R. Kumai, Y. Murakami, H. Oshio, Nat. Commun. 2014,
5:3865; d) Spin-Crossover Materials: Properties and Applications, M. A.
Halcrow ed., John Wiley & Sons, Ltd., 2013; e) M. Clemente-León, E.
Coronado, M. López-Jordà, J. C. Waerenborgh, C. Desplanches, H.
Wang, J.-F. Létard, A. Hauser, A. Tissot, J. Am. Chem. Soc. 2013, 135,
8655−8667; f) A. Bousseksou, G. Molnar, L. Salmon, W. Nicolazzi, Chem.
Soc. Rev. 2011, 40, 3313–3335; g) P. Gütlich, A. B. Gaspar, Y. Garcia,
V. Ksenofontov, C. R. Chimie 2007, 10, 21–36; h) A. Hauser, in Topics
in Current Chemistry, Vol. 234 (Eds.: P. Gütlich, H. A. Goodwin),
Springer-Verlag, 2004, pp. 155–198; i) O. Kahn, C. J. Martinez, Science
1998, 279, 44–48.
[12] S. Q. Wu, M. Liu, K. Gao, S. Kanegawa, Y. Horie, G. Aoyama, H. Okajima,
A. Sakamoto, M. L. Baker, M. S. Huzan, P. Bencok, T. Abe, Y. Shiota, K.
Yoshizawa, W. Xu, H. Z. Kou, O. Sato, Nat. Commun. 2020, 11:1992.
[13] T. M. Francisco, W. J. Gee, H. J. Shepherd, M. R. Warren, D. A. Shultz,
P. R. Raithby, C. B. Pinheiro, J. Phys. Chem. Lett. 2017, 8, 4774–4778.
[14] O. Reu, S. Ostrovsky, S. Decurtins, S. X. Liu, S. Klokishner, Eur. J. Inorg.
Chem. 2017, 5356–5365.
[15] I. N. Markevtsev, M. P. Monakhov, V. V. Platonov, A. S. Mischenko, A.
K. Zvezdin, M. P. Bubnov, G. A. Abakumov, V. K. Cherkasov, J. Magn.
Magn. Mater. 2006, 300, e407–e410.
[16] a) C. W. Lange, M. Földeàki, V. I. Nevodchikov, V. K. Cherkasov, G. A.
Abakumov, C. G. Pierpont, J. Am. Chem. Soc. 1992, 114, 4220–4222;
b) V. I. Nevodchikov, G. A. Abakumov, V. K. Cherkasov, G. A. Razuvaev,
J. Organomet. Chem. 1981, 214, 119–124.
[4]
a) M. Nihei, Y. Yanai, I. J. Hsu, Y. Sekine, H. Oshio, Angew. Chem. Int.
Ed. 2017, 129, 606–609; b) N. Hoshino, F. Iijima, G. N. Newton, N.
Yoshida, T. Shiga, H. Nojiri, A. Nakao, R. Kumai, Y. Murakami, H. Oshio,
Nat. Chem. 2012, 4, 921–926; c) C. P. Berlinguette, A. Dragulescu-
Andrasi, A. Sieber, H.-U. Güdel, C. Achim, K. R. Dunbar, J. Am. Chem.
Soc. 2005, 127, 6766–6779; d) M. E. Itkis, X. Chi, A. W. Cordes, R. C.
Haddon, Science 2002, 296, 1443–1445.
[17] K. Nomoto, T. Atake, B. K. Chaudhuri, H. Chihara, J. Phys. Soc. Jpn.
1983, 52, 3475–3485.
[18] a) M. Mitsumi, H. Ueda, K. Furukawa, Y. Ozawa, K. Toriumi, M. Kurmoo,
J. Am. Chem. Soc. 2008, 130, 14102–14104; b) J. B. Torrance, B. A.
Scott, B. Welber, F. B. Kaufman, P. E. Seiden, Phys. Rev. B 1979, 19,
730–741.
[5]
[6]
a) M. Mitsumi, K. Ezaki, Y. Komatsu, K. Toriumi, T. Miyatou, M. Mizuno,
N. Azuma, Y. Miyazaki, M. Nakano, Y. Kitagawa, T. Hanashima, R.
Kiyanagi, T. Ohhara, K. Nakasuji, Chem. Eur. J. 2015, 21, 9682–9696;
b) S. Horiuchi, Y. Tokunaga, G. Giovannetti, S. Picozzi, H. Itoh, R.
Shimano, R. Kumai, Y. Tokura, Nature 2010, 463, 789–792.
[19] M. Mitsumi, T. Murase, H. Kishida, T. Yoshinari, Y. Ozawa, K. Toriumi,
T. Sonoyama, H. Kitagawa, T. Mitani, J. Am. Chem. Soc. 2001, 123,
11179–11192.
a) M. Tadokoro, H. Hosoda, T. Inoue, A. Murayama, K. Noguchi, A. Iioka,
R. Nishimura, M. Itoh, T. Sugaya, H. Kamebuchi, M. A. Haga, Inorg.
Chem. 2017, 56, 8513–8526; b) A. Ueda, S. Yamada, T. Isono, H. Kamo,
A. Nakao, R. Kumai, H. Nakao, Y. Murakami, K. Yamamoto, Y. Nishio,
[20] T. Soda, Y. Kitagawa, T. Onishi, Y. Takano, Y. Shigeta, H. Nagao, Y.
Yoshioka, K. Yamaguchi, Chem. Phys. Lett. 2000, 319, 223–230.
[21] D. D. Perrin, W. L. F. Armarego, Purification of laboratory chemicals,
Pergamon Press, New York, 1988.
10
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
[22] S. Martinengo, G. Gxordano, P. Chini, G. W. Parshall, E. R. Wonchoba,
Inorg. Synth. 1990, 28, 242–245.
[23] S. A. Chesnokov, V. K. Cherkasov, Y. V. Chechet, N. I. Nevodchikov, G.
A. Abakumov, O. N. Mamysheva, Russ. Chem. Bull. 2000, 49, 1506–
1511.
[24] M. P. Bubnov, I. A. Teplova, K. A. Kozhanov, G. A. Abakumov, V. K.
Cherkasov, J. Magn. Reson. 2011, 209, 149–155.
[25] Y. Kume, Y. Miyazaki, T. Matsuo, H. Suga, J.Phys. Chem. Solids 1992,
53, 1297–1304.
[26] a) G. M. Sheldrick, Acta Cryst. 2008, A64, 112–122; b) G. M. Sheldrick,
Acta Crystallogr., Sect. C 2015, 71, 3–8.
[27] a) C. Kabuto, S. Akine, T. Nemoto, E. Kwon, J. Cryst. Soc. Jpn. 2009,
51, 218–224; b) K. Wakita, 2001.
[28] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino,
G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T.
Vreven, J. Montgomery, J. A., J. E. Peralta, F. Ogliaro, M. Bearpark, J.
J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J.
Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J.
V. Ortiz, J. Cioslowski, D. J. Fox, Revision C.01; Gaussian, Inc.:
Wallingford, CT, Wallingford CT, 2009.
11
This article is protected by copyright. All rights reserved.

10.1002/chem.202004217
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
Vibronic interaction has the potential to create new fields on dynamic electronic states
arising from the mixing of different electronic states by nuclear displacements. A one-
dimensional rhodium–dioxolene complex undergoes drastic changes in properties with the
phase transition, which are ascribable to the appearance of the mixed-valence state due
to the metal–ligand electron transfer coupled with a large-amplitude vibration of Rh–Rh
bonds.
12
This article is protected by copyright. All rights reserved.