Reversible copper(II)/(I) electrochemical potential switching driven by visible light-induced coordinated ring rotation

Article abstract of DOI:10.1021/ja3028873

We here describe the first metal complex system in which electronic signals can be repeatedly extracted by converting bistable states related to an intramolecular ligand rotational motion, which is fueled by visible light. The molecular structure for relating an electron transfer and a motion consists of a copper center and a coordinated unsymmetrically substituted pyrimidine derivative, whose rotational isomerization causes an electrochemical potential shift. To harness light energy effectively through metal-to-ligand charge transfer (MLCT) excitation, we prepared a simple copper(I) complex coordinated by a 4-methyl-2-(6′-methyl-2′-pyridyl)pyrimidine and a bulky diimine. The thermodynamic and kinetic parameters of redox and rotational reactions were analyzed by cyclic voltammograms at variable temperatures, by considering four stable isomers related to copper(II)/(I) states and rotational isomeric states. The key feature of this compound is that the rotation is frozen in the copper(I) state (rate constant for the rotation, k_Ii→o = 10^-4 s^-1) but is active in the copper(II) state (k _IIi→o = 10^-1 s^-1) at 203 K. The compound makes a bypass route to the isomeric metastable copper(I) state, via a tentative copper(II) state formed by photoelectron transfer (PET) in the presence of a redox mediator, decamethylferrocenium ion (DMFc⁺), or upon a partial oxidation of the complex. Light- and heat-driven rotation in the copper(I) state with a potential shift (ΔE°′ = 0.14 V) was analyzed by electrochemical measurements of the complex in the solution state. The rotor could be reset to the initial state by heating, thereby completing the cycle and enabling repeated operation fueled by light energy. A significant redox potential shift associated with the copper(II)/(I) transition accompanied the rotation, thereby providing a new type of molecular signaling system.

Full text of DOI:10.1021/ja3028873

Article
pubs.acs.org/JACS
Reversible Copper(II)/(I) Electrochemical Potential Switching Driven
by Visible Light-Induced Coordinated Ring Rotation
Michihiro Nishikawa, Kuniharu Nomoto, Shoko Kume,* and Hiroshi Nishihara*
Department of Chemistry, School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
S
* Supporting Information
ABSTRACT: We here describe the first metal complex
system in which electronic signals can be repeatedly extracted
by converting bistable states related to an intramolecular ligand
rotational motion, which is fueled by visible light. The
molecular structure for relating an electron transfer and a
motion consists of a copper center and a coordinated
unsymmetrically substituted pyrimidine derivative, whose
rotational isomerization causes an electrochemical potential shift. To harness light energy effectively through metal-to-ligand
charge transfer (MLCT) excitation, we prepared a simple copper(I) complex coordinated by a 4-methyl-2-(6′-methyl-2′-
pyridyl)pyrimidine and a bulky diimine. The thermodynamic and kinetic parameters of redox and rotational reactions were
analyzed by cyclic voltammograms at variable temperatures, by considering four stable isomers related to copper(II)/(I) states
and rotational isomeric states. The key feature of this compound is that the rotation is frozen in the copper(I) state (rate constant
for the rotation, kIi→o = 10−4 s−1) but is active in the copper(II) state (kIIi→o = 10−1 s−1) at 203 K. The compound makes a bypass
route to the isomeric metastable copper(I) state, via a tentative copper(II) state formed by photoelectron transfer (PET) in the
presence of a redox mediator, decamethylferrocenium ion (DMFc+), or upon a partial oxidation of the complex. Light- and heat-
driven rotation in the copper(I) state with a potential shift (ΔE°′ = 0.14 V) was analyzed by electrochemical measurements of the
complex in the solution state. The rotor could be reset to the initial state by heating, thereby completing the cycle and enabling
repeated operation fueled by light energy. A significant redox potential shift associated with the copper(II)/(I) transition
accompanied the rotation, thereby providing a new type of molecular signaling system.
we have focused on simple copper coordination compounds
with bidentate diimines, because of the well-established unique
relationship between reversible redox activities, photophysics,
and coordination structures in these compounds.43−52 The
copper(I) state prefers a tetrahedral geometry, whereas the
copper(II) state favors a square planar geometry or a 5- or 6-
coordinated form due to Jahn−Teller effects.36−38,43−56 The
structural changes associated with electron transfer events turn
out to play a significant role in the function of copper blue
proteins.57−59 As a result, crowded coordination geometry
generally renders the oxidation of copper(I) to copper(II)
thermodynamically less favorable due to destabilization by
steric repulsion in the copper(II) state.43,44 For example, the
oxidation potential of [Cu(dmp)2]+ (dmp = 2,9-dimethyl-1,10-
phenanthlorine) is much more positive than that of [Cu-
(phen)2]+ (phen = 1,10-phenanthlorine).43,44 In addition,
copper(I) polypyridyl complexes basically exhibit an absorption
band in the visible light region due to the metal-to-ligand
charge transfer (MLCT) transition.47−51 Introducing bulky
substituents, such as a methyl group,50,51 around the copper
center strongly enhances its luminescence on account of the
long-lived MLCT electronic excited state;47−51 application of
the luminessive copper(I) complexes into optical devices has
INTRODUCTION
■
Multistable molecules that are capable of intramolecular
structural or chemical transitions form a subclass of molecules
useful in nanotechnology applications, such as in single-electron
transistors,1,2 molecular electronics,3−7 magnetic ordering,8−10
artificial photosynthesis,11,12 photochromic materials,10,13−25
and molecular machines.6,7,25−38 Redox-active molecules, in
which oxidation states are reversibly switched by electronic
stimuli, are one of the key components in these molecular
devices. Redox-active molecules can be functionalized with
photoswitchable ability by attaching photochromic mole-
cules,18−24 such as azobenzene21,22 and diarylethene.23,24
The activities of multistable molecules as functional units
within single molecules are frequently observed in natural
systems.39−41 For example, proton gradient energies across
membranes can be converted to chemically high-energy ATP
molecules via rotational motion of the F0 unit of ATPase, where
proton receptors are ultimately fabricated. Ferrocene-attached
unidirectional motion driven by electronic voltages has been
proposed as an energy-converting function acting like a natural
motor.42 The central goal in this field is to harness the natural
motor functions of molecules that can convert the energy of
external stimuli into useful responses based on the capacities of
molecular motion.
To develop an electrochemical potential response from an
artificial molecular rotor with a stimulus-convertible function,
Received: March 24, 2012
Published: May 25, 2012
© 2012 American Chemical Society
10543
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
been developed by considering the steric effect around the
copper center.60−63 The MLCT state facilitates the photo-
electron transfer (PET) process,43,44,64 as in ruthenium(II)
polypyridyl complexes.65 Furthermore, bidentate diimines on
copper undergo ligand substitution reactions at minute-scale
rates at ambient temperatures.53,54
gand, which is similar to our previously described ligand but
includes an additional methyl group at the 6-position of the
pyridyl moiety. The additional methyl group is expected to
extend the MLCT excited state lifetime.47−51 2,9-Dimesityl-
1,10-phenanthroline (Lmes) was used as an auxiliary ligand
because the bulky mesityl group yielded a similar MLCT
lifetime extension; Lmes also helps prevent undesirable
formation of a homoleptic complex.74−77
These unique properties enable us to design a molecule
whose structural responses can be converted into a different
signal form. We previously examined azobenzene-appended
bipyridine ligands that undergo an electric potential shift via a
reversible photoisomerization through an intermediate stabi-
lized by ligand exchange at the copper center (i.e., the transition
was not intramolecular).66,67 To embed the ligand exchange
within a single molecular process, we introduced a bidentate
ligand that includes a coordinated pyrimidine moiety.68−73 The
ligand can effectively alternate between two possible pyrimidine
nitrogen atom coordination geometries at the copper center via
rotational isomerization. As the alpha substituents on the
alternate pyrimidine nitrogen atoms are different, rotational
isomerization alters the steric interactions within the coordina-
tion sphere at the copper center. The steric effects cause a shift
of the redox potential by favoring either a tetrahedral geometry
(preferred by copper(I)) or a square planar geometry (5- or 6-
coordinated form preferred by copper(II)).43,44 Rotational
isomerization of the bidentate ligand therefore displays dual
redox potentials, the electric signal of which can be detected in
the cyclic voltammogram for the copper(II)/(I) couple.68−71
Copper complexes that include unsymmetrically substituted
pyrimidines, such as 4-methyl-2-(2′-pyridyl)pyrimidine, can
undergo structural transitions based on the chemical energy
supplied by adding redox reagent solutions.68,69 The structural
transitions modulate the electrode potential68 and can
manipulate intramolecular electron transfer within a redox
array comprising a copper center and a bound ferrocene
moiety.69
RESULTS AND DISCUSSION
■
Synthesis and Characterization. MepmMepy was newly
synthesized by a condensation of a reported amidine78,79 with
acetylacetoaldehyde dimethylacetal, according to the modified
procedure described in the literature.80 A new compound,
[Cu(MepmMepy)(Lmes)]+BF4− (1+BF4 ), was synthesized by a
−
reaction of tetrakis(acetonitrile)copper(I) tetrafluoroborate
74,75
−
([Cu(MeCN)4]+BF4 ) with MepmMepy and Lmes
in
dichloromethane at room temperature,81 according to the
slightly modified procedures described in our previous
literature (Scheme 1).68 The interconversion between coordi-
−
Scheme 1. Synthesis of 1+BF4
The major problem to be solved in this system is how best to
drive the pyrimidine rotational isomerization in the energeti-
cally uphill direction to extract a responsive electric signal
repeatedly. In the present work, we succeeded in driving this
rotational isomeric system using visible light illumination
through a PET process for the first time (Figure 1). The
simplicity of the motion enabled us to accurately tailor the
dynamics of the system. The structure of the copper complex
was carefully designed to optimize sensitivity to photo-
illumination, by considering the sensitivity of the copper(I)
complex excited state to steric factors.47−51 We employed a 4-
methyl-2-(6′-methyl-2′-pyridyl)pyrimidine (MepmMepy) li-
nation isomers is described in Figure 1, where the notation of
inner (i-) and outer (o-) isomers indicates the direction of the
pyrimidine ring.
X-ray structural analysis of 1+BF4 ·CH2Cl2·0.5hexane reveals
−
that both i- and o-coordination forms were present in the same
crystal (Figure 2a and Supporting Information Figure S1). The
crystal structure was treated as a disorder between the i- and o-
Figure 2. (a) ORTEP view of 1+BF4−·CH2Cl2·0.5hexane. The
displacement ellipsoids are drawn at a 50% probability level. Hydrogen
atoms, counteranions, crystal solvents, and one complex cation are
omitted for clarity. (b) Temperature dependence of the methyl signal
1
Figure 1. Conceptual diagram showing the photodriven and heat-
on the pyrimidine moiety in the H NMR spectrum in CD2Cl2 in the
dark.
−
driven pyrimidine ring rotational isomerization of 1+BF4 .
10544
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
forms, considering that the isomeric structures are almost
identical, except for the position of the methyl group.
Occupancy refinement of the disorder gave a i-CuI/o-CuI
isomer ratio = 1:2.3.
−
The rotational behavior of 1+BF4 in a solution state was
investigated by nuclear magnetic resonance (NMR) spectros-
1
copy. The H NMR spectrum in dichloromethane-d2 at 273 K
shows two sets of signals, which were assigned to i-CuI and o-
CuI according to their chemical shifts, after consideration of the
shielding effects of the copper and Lmes moieties (Figure 2b and
Supporting Information Figure S2). Broadening of the peaks
upon heating indicates that the two isomers coexisted and
interconverted via rotational motions (Figures 2b and S3). The
molar ratio of i-CuI and o-CuI, estimated from signal
integration, remained fixed at 1:2.3 over the range 200−300
K (Figure S3).
Electrochemistry. Two requirements are required to
manipulate an electron transfer via rotation processes: (i) the
copper(II)/(I) redox potential between i- and o- isomers is
different, and (ii) the rotational interconversion can be frozen
and activated at an accessible temperature range.68,69 We
conducted cyclic voltammetry, as it provides information on
both redox properties and rotation dynamics.
−
Figure 4. Cyclic voltammograms of 1+BF4 (0.45 mM) in 0.1 M
−
Bu4N+BF4 −CH2Cl2 at a scan rate of 250 (purple line) and 50 mV s−1
(black line) at 225 K (a) and 275 K (b) in the dark.
CuII slightly decreased; the decrease was more apparent at a
scan rate of 50 mV s−1 (Figure 4a, black line). Further warming
to 275 K caused a transformation of the entire voltammogram,
as if there was one cathodic wave (Figure 4b, purple line). The
convergence into a single wave suggests that oxidation
promotes a rotation from i- to o- on the time scale of the
potential sweep.56,83 The preference of the square planar
geometry in the copper(II) state converted i-CuII into o-CuII
because of steric repulsion of the methyl group via a thermally
activated process. From the voltammograms, it appears as if
there is a single reversible redox process when both
interconversions in copper(II) and copper(I) states are
sufficiently rapid, as compared with the potential scan at 50
mV s−1 (shown in Figure 4b).
A voltammogram at 203 K in Bu4NBF4−CH2Cl2 at a scan
rate of both 250 and 50 mV s−1 showed two reversible waves
derived from two isomers at E°′ = 0.48 and 0.62 V vs Ag+/Ag
(Figure 3a). The waves at 0.48 and 0.62 V were assigned to the
Thermodynamics and Kinetics for the Rotation. As
described above, we deduced that the shape of the voltammo-
gram is mostly determined by the competence of the potential
sweep and the nature of the rotational processes. A quantitative
simulation analysis using cyclic voltammograms provides
further constraints on the experimental conditions at which
photodriven i-CuI → o-CuI rotation can be achieved.56,83 We
assumed a square scheme (Figure 5a) between four stable
isomers, i-CuI, o-CuI, i-CuII, and o-CuII, and reproduced the
experimental voltammograms at several temperatures. These
simulations provide thermal and kinetic parameters for the
redox potentials and rotations in both copper(I) and copper(II)
states, as described in Table 1 and Supporting Information
Figure S5 and Table S2.56,83
The equilibrium constant for the rotation in the copper(I)
state (KI), which was set as [o-isomer]/[i-isomer], remained 2.3
over the temperature range 200−300 K (Figure 5b). This result
indicates that the two isomers were comparably stable. In
contrast, a much larger equilibrium constant in the copper(II)
state (KII), obtained from the electrochemical analysis, indicates
that nearly all molecules were present as the o-isomer under
equilibrated conditions (Figure 5b).
The rate constant for the i- → o- rotation, estimated from the
Arrhenius plots in the copper(I) state (kIi→o), was less than
10−4 s−1 at 203 K, which indicates that the rotational motion
was substantially frozen, which we call the “rotation-OFF state”
(Figure 5c and d). When the temperature was raised to 250 K,
kIi→o increased to 10−1 s−1, indicating that the rotation was
sufficiently activated from OFF to ON in the copper(I) state. At
the intermediate temperature of 225 K, kIi→o was 10−3 s−1,
implying that the rotation proceeded slowly over a period of
several minutes, which enabled us to trace the time progression
of the rotational conversion. In contrast, the rate constant in
the copper(II) state (kIIi→o) was 10−1 s−1, even at 203 K,
−
Figure 3. (a) Cyclic voltammograms of 1+BF4 (0.45 mM) in 0.1 M
−
Bu4N+BF4 −CH2Cl2 at a scan rate of 250 (purple line) and 50 mV s−1
(black line) at 203 K. (b) Differential pulse voltammogram under the
same conditions as for part a.
o-CuII/I and i-CuII/I redox reactions, respectively, as a bulky
substituent around the copper center shifted the potential
toward positive values by preventing the square planar
geometry favored in the copper(II) state.43,44 The prominence
of the wave at E°′ = 0.48 V compared with that at E°′ = 0.62 V
was consistent with the molar ratios of isomers estimated from
1H NMR measurements. A differential pulse voltammogram
showed both o-CuII/I and i-CuII/I waves, where the peak
currents reflect the molar ratio of i-CuI and o-CuI in the
solution (Figure 3b). Repeated potential sweeps in cyclic
voltammetry gave identical current ratios of the isomers. The
nice reversibility of these two waves suggests that four states, i-
CuI, o-CuI, i-CuII, and o-CuII, are sufficiently stable without
subsequent chemical degradation. A reduction wave of 1+BF4
was not observed with potential sweeps down to −1.8 V
−
(Supporting Information Figure S4).
As the temperature was increased to 225 K at a scan rate of
250 mV s−1 (Figure 4a, purple line), the cathodic wave for i-
10545
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
These results suggest that we can thermally capture a
metastable copper(I) state out of the i-/o- equilibrium at 203 K,
if light converts a coordination isomer into another form. Also,
the metastable state can be reset to its initial state by heating to
250 K, at which thermal rotation proceeds to the equilibrium
state within several seconds. The unique heat-induced ON−
OFF switching of the rotation process in the copper(I) state is a
key phenomenon for photoinduction of our rotational system.
Photophysical Properties. To fuel the rotational isomer-
−
ization of 1+BF4 by light illumination, we examined the
−
absorption and luminescence spectra of 1+BF4 . The
−
absorption spectrum of 1+BF4 in CH2Cl2 is shown in Figure
6a. A characteristic absorption band in the visible region (λmax
=
Figure 5. (a) Illustration for an aim of the simulation analysis based on
the square scheme comprising the redox reaction (vertical) and the
pyrimidine ring rotation (horizontal). (b, c) The van’t Hoff plots (b)
and the Arrhenius plots (c) for the i- → o- rotational equilibrium in the
copper(I) (green dots) and copper(II) (purple dots) states in 0.1 M
−
Figure 6. (a) UV−vis spectra of 1+BF4 in CH2Cl2 at room
temperature: absorption spectrum in the dark (purple); absorption
spectrum in the dark, ×10 (dashed purple); emission spectrum
−
Bu4N+BF4 −CH2Cl2 in the dark. (d) The rate constants for the i- →
−
(green). (b) Absorption spectra of 1+BF4 in the dark upon addition
o- ligand geometry isomerization in the copper(I) and copper(II)
states as a function of temperature, estimated from the Arrhenius plots.
Values determined from a simulated fit to the experimental cyclic
voltammograms are shown as dots. Regression curves of the Arrhenius
plots are indicated by lines.
of 0, 0.14, 0.29, 0.43, 0.57, 0.71, 0.86, and 1.0 equiv of (NH4)+ [Ce-
2
(NO3)6]2− in acetone at 193 K. (c, d) Differences in absorption
spectra of 1+BF4− in acetone at 193 K in the dark upon oxidation with
(NH4)2[Ce(NO3)6] between 0.14 equiv and 0.29 equiv (c), and
between 0.71 equiv and 0.86 equiv (d).
Table 1. Selected Parameters Obtained from the Simulated
Cyclic Voltammograms of 1+BF4− at Scan Rates of 250, 100,
456 nm, ε = 4.8 × 103 M−1 cm−1) was attributed to a transition
from the ground state to the singlet MLCT excited state, which
appears generally in the bis(diimine)copper(I) complex
−
50, and 25 mV s−1 in 0.1 M Bu4N+BF4 −CH2Cl2
285 K
265 K
245 K
225 K
200 K
family.47−49 The absorption spectra of 1+BF4 in acetone at
−
a
E°′o/V
0.59
0.73
2.3
7 × 102
10
0.57
0.71
2.3
1 × 103
1
0.55
0.69
2.3
2 × 103
0.1
0.52
0.66
2.3
3 × 103
<0.1
1
0.48
both 293 and 193 K were similar to those in dichloromethane.
The rotational behavior in acetone was also very similar to that
in dichloromethane (Supporting Information Figures S6 and S7
and Table S3).
b
E°′i/V
0.62
c
KI
2.3
d
KII
8 × 103
<0.1
<0.1
e
kIi→o/s−1
IIi→o/s−1
The difference between the absorption spectra of o-CuI and i-
CuI was examined, on the basis of spectral changes upon
f
k
>10
>10
5
a
b
Redox potential for o-CuII/I, vs Ag+/Ag. Redox potential for i-CuII/I
,
oxidation of 1+BF4 , as follows. The absorption spectral
−
c
d
vs Ag+/Ag. Equilibrium constant, set as [o-CuI]/[i-CuI]. Equilibrium
−
changes of 1+BF4 upon addition of 0, 0.14, 0.29, 0.43, 0.57,
e
constant, set as [o-CuII]/[i-CuII]. Rate constant for the i-CuI→o-CuI
0.71, 0.86, and 1.0 equiv (NH4)+ [Ce(NO3)6]2− (ammonium
f
rotation. Rate constant for the i-CuII→o-CuII rotation.
2
hexanitratocerate (IV))85 in acetone at 193 K are shown in
Figure 6b. The oxidation diminished the MLCT absorption
band of copper(I) complexes and increased a weak d−d
transition band of copper(II) complexes.43−56 The oxidation
reaction of i-CuI and o-CuI (with the ratio, 1:2.3) proceeded
indicating that the rotation was active on the time scale of one
minute (rotation-ON state) (Figure 5c and 5d).84
10546
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
Figure 7. Photorotation experiments in the DMFc+ system with notes about the procedures. Experimental (a) and simulated (b) cyclic
−
voltammograms at a scan rate of 50 mV s−1, and experimental differential pulse voltammograms (c). Investigated DMFc+ systems comprise 1+BF4
(0.45 mM) in 0.1 M Bu4N+BF4 CH2Cl2 containing 1.8 mM DMFc+BF4− at 203 K in the dark. Top: initial state (purple lines). Middle: after 60 min
−
of visible light irradiation (λ > 400 nm) at 203 K (green lines). Bottom: after 2 min of heating at 250 K in the dark (pink lines).
almost independently, because the rotation was frozen at 193
K. As o-CuI is oxidized more easily than i-CuI, the early (from 0
to 0.7 equiv) and later (from 0.7 to 1.0 equiv) spectral changes
qualitatively show absorption changes of o-CuI and i-CuI upon
oxidation, respectively. The changes in the region from 400 to
500 nm reflect the absorption of o-CuI and i-CuI, as MLCT
absorption is much more intense than that of the d−d
transition of copper(II) complexes.
The energy difference between the ground and excited states
was estimated at ca. 1.6 eV from the maxima of the emission
spectra.86 The copper(II)/(I) redox potential of the ground
state of the i-isomer was 0.62 V vs Ag+/Ag; thus, the redox
potential of the excited i-CuI state was ca. −1.0 V, sufficient for
oxidation by the decamethylferrocenium ion (DMFc+) with a
reduction potential of −0.4 V. The transiently formed
copper(II) complex would be expected to return to the
copper(I) state by the subsequent back electron transfer
process. The copper(II)/(I) redox potential values for both the
i- and o-isomers were sufficiently positive to reoxidize DMFc0.
This estimation was also supported by the facts that similar
copper(I) polypyridyl complexes undergo PET fairly effi-
ciently,43,44 and DMFc+/0 works as an excellent redox
mediator.87
Figure 6c and d display the differences in absorption spectra
upon oxidation in the early (0.14 → 0.29 equiv, where only o-
CuI is oxidized) and later (0.71 → 0.86 equiv, where only i-CuI
is oxidized) spectral changes, respectively. As the absorption
change is constant during the addition of (NH4)+ [Ce-
2
(NO3)6]2−, the difference in absorption between o-CuI and i-
CuI would be small. The small absorption difference is similar
to that in our previous rotational system, based on a copper
complex coordinated by 4-methyl-2-(2′-pyridyl)pyrimidine and
2,9-bis(9-anthryl)-1,10-phenanthroline as ligands.68 It should
be noted that introducing methyl substituents into ligands of a
bis(diimine)copper(I) family slightly affects their absorption
spectra, within several nanometers of the maximum absorption
change.50,51 The absorption spectra of the solutions were
reversed by chemical reduction using decamethylferrocene
Photorotation of 1+ with Redox Mediator. The above
results provide a basis for controlling a photoinduced rotation.
The PET process is likely to form a tentative copper(II) state,
which can make a bypass from i-CuI to o-CuI, as the activation
energy for rotation in copper(II) is less than that in copper(I).
In addition, DMFc can be a potential electron acceptor to
complete the PET scheme.
−
The voltammograms of 1+BF4 in Bu4NBF4−CH2Cl2 at
−
(DMFc0), suggesting that oxidation of 1+BF4 proceeds
several temperatures in the dark were not altered by addition of
4 equiv of DMFc+BF4− (E°′ = −0.41 V vs Ag+/Ag at 200 K, and
−0.35 V vs Ag+/Ag at 275 K), suggesting that DMFc+ does not
disturb thermal rotational dynamics (Figures 3 and 7 and
Supporting Information Figures S5 and S9). Since DMFc+ does
without the subsequent chemical degradation (Supporting
Information Figure S8).
−
The redox potential of the excited state of 1+BF4 , which is
important for constructing a strategy for PET-driven i-CuI → o-
CuI rotational processes, can be estimated using its
not oxidize 1+BF4 (redox potentials: o-CuII/I = 0.89 V vs
−
luminescence spectra.86 A broad emission band from 1+BF4
DMFc+/0, i-CuII/I = 1.03 V vs DMFc+/0), the ratio of rotational
isomers, i-CuI/o-CuI/i-CuII/o-CuII, is 1:2.3:0:0. For clarity, we
hereafter describe only the rotational isomer ratio i-CuI/o-CuI.
In the presence of 4 equiv of DMFc+ (1.8 mM), the cyclic
voltammogram of a solution containing 1+BF4− (0.45 mM) was
−
in CH2Cl2 was observed in the red region (λmax = 750 nm)
(Figure 6a). It should be noted that this emission band may
have arisen predominantly from the i-isomer, as the steric
hindrance among substituents around the copper(I) center
significantly increases the emission efficiency.47−51
−
significantly altered upon photoirradiation in Bu4N+BF4 −
10547
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
CH2Cl2. The experimental procedures and results of this
DMFc+ system are summarized in Figure 7. Prior to
photoirradiation (Figure 7a top, purple lines), two redox
waves, one each for the i- and o-isomers in a ratio of 1:2.3, were
observed. The shape of the voltammograms remained constant
in the dark, demonstrating that there were no ongoing
processes in the absence of photoirradiation (Supporting
Information Figure S10). At 60 min, with photoirradiation
with visible light (λ > 400 nm) at 203 K, the redox waves
gradually converged to a wave corresponding to the o-isomer
(Figure 7a middle, green lines). The shapes of the voltammo-
grams after photoirradiation did not change with incubation at
203 K in the dark for 10 min, reflecting the persistence of the
irradiated state (Figure S10). Subsequent heating for 2 min at
250 K in the dark recovered the initial voltammogram,
indicating thermal relaxation of the light-induced metastable
state (Figure 7a bottom, pink lines). The ratios of i-CuI to o-
CuI in the initial, photoirradiated, and recovered states
estimated from a simulated model fit to the cyclic voltammo-
grams (Figure 7b) were 1:2.3, 1:8, and 1:2.3, respectively.83
Changes in the differential pulse voltammograms of the
solution during the process corresponded to the changes in
the cyclic voltammograms (Figure 7c).
Table 2. Photorotation Experiments under Several
Conditions Using Cyclic Voltammograms
t/
DMFc+/
c
T/
d
changes in i-CuI/o-
a
b
e
entry min
λ/nm
equiv
K
CuI
1
2
3
4
5
6
5
20
60
60
60
20
>400
>400
4
4
4
4
0
4
203
203
203
203
203
negligible
1:2.3 → 1:6
1:2.3 → 1:8
1:2.3 → 1:4
negligible
>400
450 40
>400
>400
225
negligible
a
b
c
Irradiation time. Irradiation wavelength. Chemical equivalent of
d
−
e
−
DMFc+BF4 to total 1+BF4 . Temperature. Changes in i-CuI/o-CuI
upon light irradiation, judging from a simulated model fit to the
−
experimental cyclic voltammograms of 1+BF4 (0.45 mM) containing
1.8 mM DMFc+BF4− in 0.1 M Bu4NBF4−CH2Cl2 at a scan rate of 50
mV s−1 in the dark at T K before and after light irradiation. All
voltammograms are displayed in Figure S12 of the Supporting
Information.
Figure 8 displays the changes in the peak current for the o-
CuII/I anodic wave in the experimental cyclic voltammograms
Figure 8. Changes in the anodic peak current at 0.55 V in the
−
experimental cyclic voltammograms of 1+BF4 (0.45 mM) in 0.1 M
−
−
Bu4N+BF4 −CH2Cl2 containing 1.8 mM DMFc+BF4 at 203 K.
Figure 9. (a, b) Schematic representation of the PET-driven i-CuI to o-
−
CuI ligand geometry isomerization of (a) 1+BF4 in the presence of
−
DMFc+ and (b) the partial oxidation of 1+BF4 by (NH4)+ [Ce-
2
(NO3)6]2−. The reversible changes in the molar ratios of the isomers
upon light irradiation and heating are illustrated in the bottom panels.
(Supporting Information Figure S11) upon repeated operations
with alternating light irradiation and heating. The process
worked repeatedly, demonstrating that the photoinduced
isomerization and the redox switching were reversible (Figure
8). These results clearly demonstrate a reversible change in the
molar ratios of isomers based on photodriven and heat-driven
rotational isomerization
ligand geometries of the copper(I) and copper(II) states,
obtained from simulation fits to the voltammograms (Figure 5).
The i- and o-isomers are present in the initial copper(I) state in
a ratio of 1:2.3 (KI = [o-CuI]/[i-CuI] = 2.3). Photoirradiation of
the solution induces PET from the excited i-CuI state (*i-CuI)
to the electron acceptor (DMFc+), as the oxidation potential of
*i-CuI is −1.0 V. The transiently formed i-CuII isomerizes to o-
CuII for two reasons: (i) the ligand geometric transition within
the copper(II) state was not frozen (kIIi→o = 10−1 s−1, in the ON
state); and (ii) i-CuII was thermodynamically unfavorable as
compared with o-CuII, on account of large steric repulsion (KII
= [o-CuII]/[i-CuII] > 102). Finally, the transiently formed o-CuII
is reduced to o-CuI through a back-electron transfer reaction
with DMFc0. The net process proceeds through the conversion
of i-CuI to o-CuI. If the forward conversion rate constant is
sufficiently larger than the thermally activated reverse rate
constant (from o-CuI to i-CuI, kIo→i = 10−4 s−1 at 203 K, in the
OFF state), the photodriven i-CuI → o-CuI ligand isomerization
proceeds. Therefore, the rate of i-CuI → o-CuI photorotation is
greater than 10−4 s−1 in the DMFc+ system at 203 K. The
photorotation induces the change in a ratio of i-CuI/o-CuI (in
The convergence of redox waves at 203 K in the DMFc+
system with visible light irradiation was negligible after 5 min,
considerable after 20 min, and nearly saturated after 60 min
(Table 2 and Supporting Information Figure S12). As selective
irradiation with blue light (λ = 450
40 nm) also induced
isomerization from the i- to the o-isomer, isomerization could
have resulted from excitation of the MLCT absorption band
(Table 2). Because photoisomerization did not occur in the
absence of DMFc+, a reaction pathway through PET from
copper(I) complexes to the electron acceptor (DMFc+)
appeared to be a key step for the i- to o-conversion (Table
2). It should be noted that 20 min of visible light irradiation at
225 K did not alter the voltammogram (Table 2); these results
are discussed in the final section.
Figure 9a shows the schematic representation of photo-
rotation in the DMFc+ system. The observed photodriven and
heat-driven isomerization mechanisms can be interpreted in
terms of the kinetics of the transition between the preferred
10548
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
the conditions of Figure 7, from 1:2.3 to 1:8). The reversion
from the light-induced metastable state by the thermal process
at 203 K would take on the order of 104 s, judging from the rate
constant value, kIo→i = 10−4 s−1. Heating to 250 K (kIo→i = 10−1
s−1, in the ON state) provides sufficient thermal energy to the
copper complex that the initial ratio of isomers is restored
through back-isomerization. Consequently, the results show
that the i-CuI → o-CuI rotation via the redox potential shift in
the DMFc+ system can be driven by light.88
Photorotation from i-CuI to o-CuI under Partial
Oxidation. We next attempted to construct a combined
photorotation-redox switching system without the need for an
electron mediator, to show its high versatility. The mechanism
described above predicts that a more efficient reversible
transition is expected if the copper(II) complex itself acts as
an electron acceptor (E°′ = 0.48 V), in place of the relatively
weaker DMFc+ state (E°′ = −0.41 V) (Figure 9b).
To test this speculation, 1+BF4− in Bu4N+BF4 −CH2Cl2 was
−
partially oxidized by 0.4 equiv of (NH4)+ [Ce(NO3)6]2−. The
2
changes in the partially oxidized system with light irradiation
and heating were basically similar to those of the DMFc+
system, judging from the anodic waves in the voltammograms.
In this case, the sample contains four copper components, as
it includes copper(II) species even in the dark condition. The
ratio of each component was estimated as follows.
a = [o‐CuII] + [i‐CuII]
(1)
1 − a = [o‐CuI] + [i‐CuI]
(2)
x(1 − a) = [o‐CuI]
(3)
[o‐CuI] + [i‐CuI] + [o‐CuII] + [i‐CuII] = 1
in the dark, upon repeated operation with 60 min of visible
light irradiation (λ > 400 nm) at 203 K and 2 min of heating at
250 K in the dark. The voltammograms are shown in Figure
S11 of the Supporting Information.
Here, the total amount of copper(II) species is expressed as
a, which is identical to an equivalent amount of oxidant. The
ratio of o-CuI to total copper(I) species is expressed as x. Since
the rotation in the copper(II) state is always under equilibrium,
the molar ratio of i-CuII is negligible, considering its large
equilibrium constant (KII = [o-CuII]/[i-CuII] > 102). Therefore,
(4)
Figure 10. (a) Experimental cyclic voltammograms at a scan rate of
−
−
100 mV s−1 of 1+BF4 (0.75 mM) in Bu4N+BF4 −CH2Cl2/acetone
(v/v 20:1) upon addition of 0.4 equiv of (NH4)+ [Ce(NO3)6]2− (0.3
2
mM) at 225 K in the dark. Top (purple line): initial state. Middle
(green line): after 3 min of visible light irradiation (λ > 400 nm) at 225
K. Bottom (purple line): after a 20-min interval at 225 K in the dark.
(b) Changes in the anodic peak current at 0.50 V at 225 K in the
experimental cyclic voltammograms of the solution upon repeated
operation with 3 min of visible light irradiation (λ > 400 nm) at 225 K
and 20-min intervals at 225 K in the dark. The voltammograms are
shown in Figure S14 of the Supporting Information.
a = [o‐CuII] + [i‐CuII] ≅ [o‐CuII]
(5)
suggesting that the expected self-exchange electron transfer
between the copper species induced negligible changes in the
ratios of rotational isomers (Figure S13 of the Supporting
Information). The redox waves almost completely converged
into a single wave corresponding to the o-isomer upon 3 min of
photoirradiation with visible light (λ > 400 nm) at 225 K
(Figure 10a middle, green line). The convergence can be
qualitatively explained by a change in molar ratio from i-CuI/o-
CuI = 1:2.3 to ca. 1:8, since the light-induced changes in the
cyclic voltammograms resemble those of the DMFc+ system
from the viewpoint of anodic waves; simulation of the anodic
waves also supported the molar ratio changes (Figure S13 and
Table S4). The voltammograms gradually recovered to the
initial state in the dark (Figure S14) and were completely
restored after a 20-min interval (Figure 10a bottom, purple).
Figure 10b displays the changes in anodic peak currents of the
o-isomer in the cyclic voltammograms at 225 K in the partially
oxidized system for 10 repeated operations (experimental
voltammograms are displayed in Figure S15). The result shows
As KI = [o-CuI]/[i-CuI] = 2.3 does not depend on temperature,
x in the initial state is
x = KI/(1 + KI) = 0.7
(6)
In the case of a = 0.4, the molar ratio in the initial state of the
partial oxidation system is i-CuI/o-CuI/i-CuII/o-CuII
=
1:2.3:0:2.2. Considering that the copper(II) state is always
under equilibrium and that i-CuII can be set to 0, we describe
the rotational isomer ratio only as i-CuI/o-CuI for clarity. Thus,
oxidation of 0.75 mM 1+BF4 by 0.4 equiv of (NH4)+ [Ce-
−
2
(NO3)6]2− (0.30 mM) in Bu4N+BF4 −CH2Cl2 yields 0.45 mM
−
−
of 1+BF4 solution with copper(II) complexes (0.30 mM) as
redox mediator.
Cyclic voltammograms of the partially oxidized system noted
above showed two anodic waves at 225 K, one each for the i-
and o-isomers, in a ratio of 1:2.3, which reflects the ratio of i-
CuI and o-CuI in the solution (Figure 10a top, purple line). No
deformation of voltammograms in the dark occurred,
10549
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
the excellent repeatability of the photorotation and redox
switching.
copper(II)/(I) species may contribute to the efficiency of
rotation processes.
Consequently, we succeeded in establishing two kinds of
systems in which light stimuli are reversibly converted into
copper(II)/(I) redox potentials via molecular rotation, but
which exhibited different photorotation behaviors.
The PET process of the partially oxidized system can be
described as follows (Figure 9b). Photoirradiation of i-CuI
followed by PET to o-CuII yields i-CuII and o-CuI, leading to a
rapid rotational isomerization transition of i-CuII to o-CuII. The
net scheme is i-CuI to o-CuI without back electron transfer to o-
CuII, unlike in the DMFc+-mediated system. It should be noted
that the photorotation proceeds even at 225 K, unlike in the
DMFc+ system, and in a very short time frame. The rate of i-
CuI → o-CuI photorotation is greater than 10−3 s−1 in the
partially oxidized system at 225 K, coincident with the
competitive thermal back-isomerization process, with rate
constant kIo→i = 10−3 s−1. The scheme contributes considerably
to changes in the molar ratio, i-CuI/o-CuI, from initial to
metastable irradiated states (in Figure 10, from 1:2.3 to ca. 1:8).
The rate constant of thermal reversion at 225 K is kIo→i = 10−3
s−1, which qualitatively implies that the light-induced
metastable state is restored to equilibrium after 103 s. The
estimation is consistent with the 20-min recovery process
displayed in the voltammogram at 225 K. Consequently, partial
oxidation turns out to be an effective way to achieve the
conversion of light stimuli to redox potentials through i-CuI →
o-CuI photorotation.
CONCLUSION
■
We demonstrated the conversion of light stimuli into
electrochemical potential via reversibly working artificial
molecular rotation. The energy-processing function was
acquired by adding a photoresponse to a copper-coordinated
pyrimidine ring rotation system, in which the molecular motion
can be extracted as an electrochemical potential of the
copper(II)/(I) redox reaction. A key feature of the present
system is that the population of bistable isomers in the
copper(I) state with different redox potentials (ΔE°′ = 0.14 V)
is reversibly converted by light and heat stimuli through a PET-
induced process. The bistability is acquired by the high
activation energy of the copper(I) state, which is drastically
decreased in the transiently formed copper(II) state. PET
processes can take a bypass route in the rotation of the
copper(I) complex. The system works not only with a redox
mediator but also upon partial oxidation, in which the copper
complex itself considerably assists the photorotation, in contrast
to the role of DMFc+ in the DMFc+ system.
Factors Dominating Photorotation Rate. Upon study-
ing the photorotation condition, we noticed that photorotation
under partial oxidation proceeds much faster than in the
presence of DMFc+. The rate difference is beyond that
attributable to concentration or temperature and, therefore,
must originate from the rotation mechanism.
Generally, photodriven bistable material changes are
accompanied by significant color changes, which involve light
absorption efficiency and reconstruction of the electronic
state.13−25 Our present photo- and heat-driven rotation system
1
When we used DMFc+ with 20 min of visible light irradiation
at 225 K and at 200 K (Table 2), the light-induced
voltammogram changes at 225 K were negligible, because the
rate of o-CuI → i-CuI thermal reversion at 225 K (kIo→i = 10−3
s−1) overcomes that of photodriven i-CuI → o-CuI. Photo-
rotation in the DMFc+ system, on the other hand, requires
cooling to 200 K (kIo→i = 10−4 s−1). As the conditions were
identical, with the exception of temperature, we determined
that the rate increment of kIo→i due to heating is larger than that
of photorotation. In other words, the slow thermal reversion
rate is a determining factor for photorotation at 200 K in the
DMFc+ system.
works without a significant color change or copper(I) MLCT
absorption, but it can induce electron transfer. It provides a
further methodology to construct photodriven materials.
Since molecular electronics components such as transistor1,2
and memory7 work by charge injection, the present redox
potential response can be progressed into other types of signals
via intramolecular electron transfer. The function of the light-
driven redox-synchronized molecular rotor can provide
electronic, magnetic, and other molecular signaling character-
istics.
EXPERIMENTAL SECTION
■
Materials. 6-Methyl-2-pyridinecarboximidamide monohydrochlor-
ide,78,79 tetrakis(acetonitrile)copper(I) tetrafluoroborate,89 2,9-dimesi-
tyl-1,10-phenanthroline (Lmes),74,75 and decamethylferrocenium tetra-
fluoroborate90 were also prepared according to methods described in
the literature. Other chemicals were used as purchased.
On the other hand, photorotation was observed at 225 K in a
partially oxidized system (Figure 10). Possible variables in the
experimental conditions of the two systems related to
photorotation behavior are as follows: (i) temperature (225
K), (ii) concentrations of total copper(I) species (0.45 mM),
(iii) light irradiation time (3 or 20 min), (iv) concentration of
the redox mediator (0.30 mM or 1.8 mM), and (v) type of
redox mediator (copper(II) complex or DMFc+). Conditions
(i) and (ii) are identical in both systems. Conditions (iii) and
(iv) are considered to work against rapid rotation in a partially
oxidized system, by decreasing the density of photons and the
probability of collisions during intermolecular electron transfer.
Therefore, enhancement of the photorotation rate must be
related to redox mediator processes. In a comparison of the two
photorotation schemes (Figure 9), the partial oxidation system
does not require back electron transfer processes to complete
the scheme. Also, the highly reducing driving force of o-CuII can
contribute to the higher efficiency of photorotation in the
partially oxidized system, as compared with the case in the
DMFc+ system. The expected slow rate of electron transfer in
X-ray Structural Analysis.
A red single crystal of
−
1+BF4 ·CH2Cl2·0.5hexane was obtained by diffusing hexane into a
−
dichloromethane solution of 1+BF4 . Diffraction data were collected
with an AFC10 diffractometer coupled with a Rigaku Saturn charge-
coupled device (CCD) system equipped with a rotating-anode X-ray
generator producing graphite-monochromated Mo Kα radiation (λ =
0.7107 Å). Lorentz polarization and numerical absorption corrections
were performed with the program Crystal Clear 1.3.6. The structure
was solved by the direct method using SIR 92 software91 and refined
against F2 using SHELXL-97.92 WinGX software was used to prepare
the material for publication.93 The crystallographic data are listed in
Table S1 of the Supporting Information. The isomeric structures were
almost identical, except for the position of the methyl group.
Occupancy refinement of the cyrstallographic disorder gave a i-/o- =
3:7 isomer ratio, as analyzed by the PART option of the SHELX-97
−
program. The disorders of the BF4 molecules in the crystals were
analyzed by the PART, ISOR, DELU, and SIMU options of the
SHELX-97 program.
10550
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
−
Instrumentation. Nuclear magnetic resonance spectra at several
temperatures in the dark were recorded using a Bruker DRX 500
spectrometer, using a ca. 20 min data-recording interval. The 13C
NMR spectrum of MepmMepy was recorded using a JEOL ECX 400
spectrometer. The reported chemical shifts of the solvent residual
peaks were used for calibration of the NMR spectra in chloroform-d1
(CDCl3, δ = 7.26 ppm for 1H, δ = 77.16 ppm for 13C) and
dichloromethane-d2 (CD2Cl2, δ = 5.32 ppm for 1H).82 UV−vis
absorption spectra in the dark were recorded with a JASCO V-570
spectrometer and a Hewlett-Packard 8453 spectrometer equipped with
a USP-203-A UNISOKU cryostat, using a quartz cell with a 1-cm
optical path length. Steady-state corrected emission spectra were
recorded with a HITACHI F-4500 spectrometer. Electrochemical
measurements were recorded with an ALS 750A electrochemical
analyzer (BAS. Co., Ltd.) and a HZ-3000 electrochemical analyzer
(Hokuto Denko Co. Ltd.), each equipped with a UNISOKU cryostat.
The working electrode was a 0.3 mm i.d. glassy carbon electrode; a
platinum wire served as the auxiliary electrode, and the reference
electrode was an Ag+/Ag electrode (a silver wire immersed in 0.1 M
Bu4N+ClO4−/0.01 M Ag+ClO4−/CH3CN). The solutions were
deoxygenated with pure argon prior to the electrochemical measure-
ments. Light irradiation was performed using a MAX-302 xenon lamp
(ASAHI SPECTRA) equipped with an optical fiber with a long-pass
filter (cut-on 400 nm) or a bandpass filter (central wavelength 450 nm,
full width at half-maximum 80 nm).
product, a deep red solid of 1+BF4 : yield, 42 mg (0.055 mmol, 65%).
1H NMR (500 MHz, CD2Cl2 273 K): δ 8.71−8.66 (m, i-3H + o-2H),
8.30 (d, J = 7.7 Hz, i-1H), 8.24 (d, J = 7.7 Hz, o-1H), 8.20 (s, i-2H),
8.19 (s, o-2H), 8.14 (d, J = 5.4 Hz, o-1H), 7.87−7.80 (m, i-3H + o-
3H), 7.30 (d, J = 4.1 Hz, i-1H), 7.29 (d, J = 7.7 Hz, o-1H), 7.25 (d, J =
5.6 Hz, o-1H), 7.22 (d, J = 4.8 Hz, i-1H), 6.39 (s, i-2H), 6.33 (s, i-2H),
6.32 (s, o-2H), 6.28 (s, o-2H), 2.64 (s, o-3H), 2.05 (s, i-3H), 1.97 (s, o-
3H), 1.95 (s, o-6H), 1.93 (s, i-6H), 1.90 (s, i-3H), 1.77 (s, o-6H), 1.69
(s, i-6H), 1.68 (s, o-6H), 1.61 (s, i-6H). Elemental analysis. Calculated
for C41H39N5CuBF4·C0.1H0.2Cl0.2: C 64.90, H 5.19, N 9.21. Found: C
64.71, H 5.33, N 8.92.
Monitoring of Photorotation from i-CuI to o-CuI in the
Redox Mediator System. The electrochemical cell, shielded from
room light by covering with aluminum foil, was equipped with
electrodes, a cryostat, and the optical fiber connected to the xenon
lamp. The electrochemical cells were interfaced with a cryostat
container, whose size and shape fitted the cells. The opening between
the cell and the container is filled with methanol. The electrochemical
measurements were performed on the static solution at 203 K in the
dark after vigorous stirring. Under an argon atmosphere, a 2 mL 0.1 M
Bu4N+BF4 −CH2Cl2 solution of 1+BF4− (0.45 mM) was cooled at 203
−
K in the dark, followed by electrochemical measurements. Then, the
solution was stirred and photoirradiated with visible light (λ > 400
nm) at 203 K for 60 min, followed by further measurements, which
showed a negligible change in the voltammograms. Then, a 100 μL
−
deoxygenated CH2Cl2 solution of DMFc+BF4 (1.8 mM, 4 equiv to
Thermodynamic and Kinetic Analysis. The molar ratios of
isomers of 1+BF4− in solution at several temperatures were determined
from the 1H NMR signal integration in the methyl group on the basis
of pyrimidine moieties. The broad spectra at around room
temperature were excluded from the thermodynamic analysis. The
experimental cyclic voltammograms were simulated using BASi
Digisim 3.03a software, based on two independent double-step
electron transfer systems. The isomer equilibrium constants in the
copper(II) state were derived from the experimentally obtained
equilibrium constants of isomers in the copper(I) state and the two
redox potentials of i- and o-isomers. The van’t Hoff plots were based
on an equilibrium constant of [o-isomer]/[i-isomer]. The molar ratios
of isomers at variable temperatures were calculated from extrapolations
of the van’t Hoff plots. The rate constants for rotational processes
from the i- to o-isomer were determined from the simulation analysis
of the cyclic voltammograms. The rates of the rotations at various
temperatures were calculated from the slopes and intercepts of the
Arrhenius plots. The good fit of simulated curves with the
experimental voltammograms confirmed the validity of the analyses.
Synthesis of 4-Methyl-2-(6′-methyl-2′-pyridyl)pyrimidine
(MepmMepy). Under a nitrogen atmosphere, 6-methyl-2-pyridine-
carboximidamide (1.0 g, 5.8 mmol) was dissolved in an ethanol
solution of sodium ethoxide (prepared from 175 mg of sodium and 60
mL of ethanol) at room temperature. Acetylacetoaldehyde dimethy-
lacetal (1 mL, 8 mmol) was added to the mixture. The reaction
mixture was refluxed for 3 days. The reaction mixture was cooled to
room temperature and acidified with acetic acid. The solvent was
evaporated and the residue dissolved in water and extracted with
CH2Cl2. The organic layer was dried over sodium sulfate, filtered, and
evaporated. The residue was chromatographed on an alumina gel
column eluted with ethyl acetate−hexane (1:1 v/v). The solvent was
evaporated to obtain a colorless powder: yield, 0.285 g (1.54 mmol,
−
1+BF4 ) was added to the solution at room temperature. The solution
was cooled at 203 K in the dark, followed by measurements in the
initial state. Then, the solution was stirred at 203 K in the dark for
several minutes, followed by measurements to check that no changes
were detected in the voltammogram with time. Then, the solution was
stirred and photoirradiated with visible light (λ > 400 nm) at 203 K for
60 min, followed by measurements which showed considerable
voltammogram changes. The solution was stirred at 203 K in the
dark for several minutes, followed by measurements for the persistence
of the irradiated state. After a 2-min heating of the solution to 250 K in
the dark with stirring, the solution was cooled to 203 K in the dark,
followed by measurements in the warmed state. The operations were
repeated three times, with a 60-min photoirradiation at 203 K and a 2-
min heating in the dark at 250 K, to achieve repeatable voltammogram
changes. Then, the operations were performed using blue light (λ =
450 40 nm) instead of visible light. The solution was heated to 225
K in the dark, followed by measurements performed at 225 K instead
of at 203 K. The solution was stirred and photoirradiated with 20 min
of visible light (λ > 400 nm) at 225 K, followed by measurements at
225 K, which showed a negligible change.
Monitoring of Photorotation from i-CuI to o-CuI in the
Partially Oxidized System. The measurements were performed
using the same electrochemical cell mentioned above. Electrochemical
measurements were performed for the static solution at 225 K in the
dark after vigorous stirring. Under an argon atmosphere, a 100 μL of a
deoxygenated acetone solution of ammonium hexanitratocerate (IV)
−
(0.3 mM, 0.4 equiv to 1+BF4 ) was added to 2 mL of 0.1 M
−
Bu4N+BF4 −CH2Cl2 solution of 1·BF4 (0.75 mM) at room temper-
ature. Then, the solution was cooled to 225 K in the dark, followed by
measurements at the initial state. The solution was stirred and
photoirradiated with visible light (λ > 400 nm) at 225 K for 3 min,
followed by measurements in the irradiated state, which gave a
considerable voltammogram change. The solution was stirred for 20
min at 225 K in the dark, followed by the measurements. The
operation was repeated 10 times, with 3 min of visible light (λ > 400
nm) irradiation at 225 K followed by a 20-min interval at 225 K in the
dark, to achieve a repeatable voltammogram change.
1
27%). H NMR (500 MHz, CDCl3): δ 8.76 (d, J = 5.1 Hz, 1H), 8.29
(d, J = 7.8 Hz, 1H), 7.74 (t, J = 7.7 Hz, 1H), 7.26 (d, 1H), 7.16 (d, J =
5.1 Hz, 1H), 2.72 (s, 3H), 2.65 (s, 3H). 13C NMR (100 MHz,
CDCl3): δ 24.7, 25.1, 119.9, 121.0, 124.7, 137.2, 154.7, 157.5, 159.1,
163.8, 168.0. Elemental analysis. Calculated for C11H11N3: C 71.33, H
5.99, N 22.69. Found C 71.53, H 6.19, N 22.40.
−
Synthesis of [Cu(MepmMepy)(Lmes)]+BF4 (1+BF4−). Under a
−
nitrogen atmosphere, [Cu(MeCN)4]+BF4 (28.3 mg, 0.0900 mmol)
ASSOCIATED CONTENT
■
was added to Lmes (37.1 mg, 0.0891 mmol) in 5 mL of CH2Cl2. To the
resultant orange solution, MepmMepy (15.7 mg, 0.0848 mmol) was
added, and the color of the solution changed immediately to dark red.
The reaction mixture was stirred for 30 min. The solution was filtered,
and diethyl ether (15 mL) was added to the solution to precipitate the
S
* Supporting Information
Crystal structure data (CIF), spectral data, and electrochemical
data. This material is available free of charge via the Internet at
http://pubs.acs.org.
10551
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
(28) Hernan
1532−1537.
́
dez, J. V.; Kay, E. R.; Leigh, D. A. Science 2004, 306,
AUTHOR INFORMATION
■
Corresponding Author
kume@chem.s.u-tokyo.ac.jp; nisihara@chem.s.u-tokyo.ac.jp
(29) Serreli, V.; Lee, C.-F.; Kay, E. R.; Leigh, D. A. Nature 2007, 445,
523−527.
(30) Balzani, V.; Credi, A.; Venturi, M. Molecular Devices and
Machines, 2nd ed.; Wiley-VCH: Weinheim, 2008.
(31) Kay, E. R.; Leigh, D. A.; Zerbetto, F. Angew. Chem., Int. Ed.
2007, 46, 72−191.
(32) Browne, W. R.; Feringa, B. L. Nat. Nanotechnol. 2006, 1, 25−35.
(33) Brouwer, A. M.; Frochot, C.; Gatti, F. G.; Leigh, D. A.; Mottier,
L.; Paolucci, F.; Roffia, S.; Wurpel, G. W. H. Science 2001, 291, 2124−
2128.
(34) Mobian, P.; Kern, J.-M.; Sauvage, J.-P. Angew. Chem., Int. Ed.
2004, 43, 2392−2395.
(35) Murakami, H.; Kawabuchi, A.; Kotoo, K.; Kunitake, M.;
Nakashima, N. J. Am. Chem. Soc. 1997, 119, 7605−7606.
(36) Armaroli, N.; Balzani, V.; Collin, J.-P.; Gavina, P.; Sauvage, J.-P.;
Ventura, B. J. Am. Chem. Soc. 1999, 121, 4397−4408.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Grants-in-Aid from MEXT of
Japan (20750044, 20245013, and 21108002, area 2107), JST
(Research Seeds Quest Program), and a Research Fellowship of
the Japan Society for the Promotion of Science for Young
Scientists.
REFERENCES
(1) Kubatkin, S.; Danilov, A.; Hjort, M.; Cornil, J.; Bred
Stuhr-Hansen, N.; Hedegard, P.; Bjørnholm, T. Nature 2003, 425,
■
́
as, J.-L.;
̊
(37) Collin, J.-P.; Dietrich-Buchecker, C.; Gavina, P.; Jimen
Molero, M. C.; Sauvage, J.-P. Acc. Chem. Res. 2001, 34, 477−487.
́
ez-
̃
698−701.
(2) Park, J.; Pasupathy, A. N.; Goldsmith, J. I.; Chang, C.; Yaish, Y.;
(38) Zahn, S.; Canary, J. W. J. Am. Chem. Soc. 2002, 124, 9204−9211.
(39) Dau, H.; Zaharieva, I. Acc. Chem. Res. 2009, 42, 1861−1870.
(40) Vale, R. D.; Milligan, R. A. Science 2000, 288, 88−95.
Petta, J. R.; Rinkoski, M.; Sthena, J. P.; Abruna, H. D.; McEuen, P. L.;
̃
Ralph, D. C. Nature 2002, 417, 722−725.
(3) Moth-Poulsen, K.; Bjørnholm, T. Nat. Nanotechnol. 2009, 4,
551−556.
(41) Junge, W.; Sielaff, H.; Engelbrecht, S. Nature 2009, 459, 364−
370.
(4) Joachim, C.; Gimzewski, J. K.; Aviram, A. Nature 2000, 408,
541−548.
(42) Vives, G.; Carella, A.; Launay, J.-P.; Rapenne, G. Coord. Chem.
Rev. 2008, 252, 1451−1459.
(5) de Silva, A. P. Nature 2008, 454, 417−418.
(6) Flood, A. H.; Stoddart, J. F.; Steuerman, D. W.; Heath, J. R.
Science 2004, 306, 2055−2056.
(43) Ruthkosky, M.; Kelly, C. A.; Castellano, F. N.; Meyer, G. J.
Coord. Chem. Rev. 1998, 171, 309−322.
(44) Ruthkosky, M.; Castellano, F. N.; Meyer, G. J. Inorg. Chem.
1996, 35, 6406−6412.
(7) Green, J. E.; Wook Choi, J.; Boukai, A.; Bunimovich, Y.;
Johnston-Halperin, E.; DeIonno, E.; Luo, Y.; Sheriff, B. A.; Xu, K.; Shik
Shin, Y.; Tseng, H.-R.; Stoddart, J. F.; Heath, J. R. Nature 2007, 445,
414−417.
(45) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. Inorg. Chem. 1998,
37, 2285−2290.
(46) Itoh, S.; Funahashi, S.; Koshino, N.; Takagi, H. D. Inorg. Chim.
Acta 2001, 324, 252−265.
́
(8) Simao, C.; Mas-Torrent, M.; Crivillers, N.; Lloveras, V.; Artes, J.
̃
M.; Gorostiza, P.; Veciana, J.; Rovira, C. Nat. Chem. 2011, 3, 359−364.
(9) Sato, O.; Iyoda, T.; Fujishima, A.; Hashimoto, K. Science 1996,
272, 704−705.
(47) Armaroli, N.; Accorsi, G.; Cardinali, F.; Listorti, A. Top. Curr.
Chem. 2007, 280, 69−115.
(48) Lavie-Cambot, A.; Cantuel, M.; Leydet, Y.; Jonusauskas, G.;
Bassani, D. M.; McClenaghan, N. D. Coord. Chem. Rev. 2008, 252,
2572−2584.
(10) Venkataramani, S.; Jana, U.; Dommaschk, M.; Sonnichsen, F.
̈
D.; Tuczek, F.; Herges, R. Science 2011, 331, 445−448.
(11) Imahori, H.; Guldi, D. M.; Tamaki, K.; Yoshida, Yu.; Luo, C.;
Sakata, Y.; Fukuzumi, S. J. Am. Chem. Soc. 2001, 123, 6617−6628.
(12) Meyer, T. J. Acc. Chem. Res. 1989, 22, 163−170.
(13) Irie, M.; Fukaminato, T.; Sasaki, T.; Tamai, N.; Kawai, T. Nature
2002, 420, 759−760.
(14) Kobatake, S.; Takami, S.; Muto, H.; Ishikawa, T.; Irie, M. Nature
2007, 446, 778−781.
(15) Gorostiza, P.; Isacoff, E. Y. Science 2008, 322, 395−399.
(16) Beharry, A. A.; Sadovski, O.; Woolley, G. A. J. Am. Chem. Soc.
2011, 133, 19684−19687.
(17) Uchida, K.; Yamanoi, Y.; Yonezawa, T.; Nishihara, H. J. Am.
Chem. Soc. 2011, 133, 9239−9241.
(18) Kume, S.; Nishihara, H. Dalton Trans. 2008, 3260−3271.
(19) Muratsugu, S.; Kume, S.; Nishihara, H. J. Am. Chem. Soc. 2008,
130, 7204−7205.
(20) Sakamoto, R.; Murata, M.; Nishihara, H. Angew. Chem., Int. Ed.
2006, 45, 4793−4795.
(21) Nishihara, H. Coord. Chem. Rev. 2005, 249, 1468−1475.
(22) Kurihara, M.; Hirooka, A.; Kume, S.; Sugimoto, M.; Nishihara,
H. J. Am. Chem. Soc. 2002, 124, 8800−8801.
(23) Fraysse, S.; Coudret, C.; Launay, J.-P. Eur. J. Inorg. Chem. 2000,
1581−1590.
(24) Tanaka, Y.; Inagaki, A.; Akita, M. Chem. Commun. 2007, 1169−
1171.
(25) Muraoka, T.; Kinbara, K.; Aida, T. Nature 2006, 440, 512−515.
(26) Ruangsupapichat, N.; Pollard, M. M.; Harutyunyan, S. R.;
Feringa, B. L. Nat. Chem. 2011, 3, 53−60.
(27) Fletcher, S. P.; Dumur, F.; Pollard, M. M.; Feringa, B. L. Science
2005, 310, 80−82.
(49) McMillin, D. R.; McNett, K. M. Chem. Rev. 1998, 98, 1201−
1219.
(50) Everly, R. M.; Ziessel, R.; Suffert, J.; McMillin, D. R. Inorg.
Chem. 1991, 30, 559−561.
(51) Cunningham, C. T.; Cunningham, K. L. H.; Michalec, J. F.;
McMillin, D. R. Inorg. Chem. 1999, 38, 4388−4392.
(52) Iwamura, M.; Watanabe, H.; Ishii, K.; Takeuchi, S.; Tahara, T. J.
Am. Chem. Soc. 2011, 133, 7728−7736.
(53) Desvergnes-Breuil, V.; Hebbe, V.; Dietrich-Buchecker, C.;
Sauvage, J.-P.; Lacour, J. Inorg. Chem. 2003, 42, 255−257.
(54) Meyer, M.; Albrecht-Gary, A. M.; Dietrich-Buchecker, C. O.;
Sauvage, J.-P. Inorg. Chem. 1999, 38, 2279−2287.
(55) Rorabacher, D. B. Chem. Rev. 2004, 104, 651−697.
(56) Le Poul, N.; Campion, M.; Douziech, B.; Rondelez, Y.; Le
Clainche, L.; Reinaud, O.; Le Mest, Y. J. Am. Chem. Soc. 2007, 129,
8801−8810.
(57) Solomon, E. I.; Szilagyi, R. K.; George, S. D.; Basumallick, L.
Chem. Rev. 2004, 104, 419−458.
(58) Lewis, E. A.; Tolman, W. B. Chem. Rev. 2004, 104, 1047−1076.
(59) Farver, O.; Pecht, I. Coord. Chem. Rev. 2011, 255, 757−773.
(60) Bessho, T.; Constable, E. C.; Gratzel, M.; Redondo, A. H.;
̈
Housecroft, C. E.; Kylberg, W.; Nazeeruddin, M. K.; Neuburger, M.;
Schaffner, S. Chem. Commun. 2008, 32, 3717−3719.
(61) Zhang, Q.; Zhou, Q.; Cheng, Y.; Wang, L.; Ma, D.; Jing, X.;
Wang, F. Adv. Funct. Mater. 2006, 16, 1203−1208.
(62) Smith, C. S.; Branham, C. W.; Marquardt, B. J.; Mann, K. R. J.
Am. Chem. Soc. 2010, 132, 14079−14085.
10552
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553

Journal of the American Chemical Society
Article
(63) Hsu, C.-W.; Lin, C.-C.; Chung, M.-W.; Chi, Y.; Lee, G.-H.;
Chou, P.-T.; Chang, C.-H.; Chen, P.-Y. J. Am. Chem. Soc. 2011, 133,
12085−12099.
the back electron transfer process, i-CuII + DMFc0 → i-CuI + DMFc+,
can compete with the rotation in the copper(II) state, i-CuII → o-CuII.
For another example, both radiative and nonradiative transition of the
photoexcited state, *i-CuI → i-CuI + hν′ or heat, can compete with the
PET process, *i-CuI + DMFc+ → i-CuII + DMFc0. Energy transfer in
the photoexcited states also contributes to the deactivation. One or
more of them can be the reason why the i-CuI → o-CuI photorotation
is slow, considering that the sufficient light-induced voltammograms
changes require long (60 min) irradiation time in the DMFc+ system.
However, these deactivation processes are not critical to stop the
photodriven rotation.
(64) Tran, D.; Ryu, C. K.; Ford, P. C. Inorg. Chem. 1994, 33, 5957−
5959.
(65) Campagna, S.; Puntoriero, F.; Nastasi, F.; Bergamini, G.;
Balzani, V. Top. Curr. Chem. 2007, 280, 117−214.
(66) Kume, S.; Murata, M.; Ozeki, T.; Nishihara, H. J. Am. Chem. Soc.
2005, 127, 490−491.
(67) Umeki, S.; Kume, S.; Nishihara, H. Chem. Lett. 2010, 39, 204−
205.
(89) Merrill, C. L.; Wilson, L. J.; Thamann, T. J.; Loehr, T. M.; Ferris,
N. S.; Woodruff, W. H. J. Chem. Soc., Dalton Trans. 1984, 2207−2221.
(90) Tabbi, G.; Cassino, C.; Cavigiolio, G.; Colangelo, D.; Ghiglia,
A.; Viano, I.; Osella, D. J. Med. Chem. 2002, 45, 5786−5796.
(91) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Burla, M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27,
435.
(68) Nomoto, K.; Kume, S.; Nishihara, H. J. Am. Chem. Soc. 2009,
131, 3830−3831.
(69) Kume, S.; Nomoto, K.; Kusamoto, T.; Nishihara, H. J. Am.
Chem. Soc. 2009, 131, 14198−14199.
(70) Kume, S.; Nishihara, H. Chem. Commun. 2011, 47, 415−417.
(71) Kume, S.; Nishihara, H. Dalton Trans. 2011, 40, 2299−2305.
(72) Nishikawa, M.; Nomoto, K.; Kume, S.; Inoue, K.; Sakai, M.;
Fujii, M.; Nishihara, H. J. Am. Chem. Soc. 2010, 132, 9579−9581.
(73) Photoresponsive coordinated ligand rotational isomerization,
such as based on a rheniumcarbon double bond, has been reported.
McCormick, F. B.; Kiel, W. A.; Gladysz, J. A. Organometallics 1982, 1,
405−408.
(92) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(93) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838.
(74) Schmittel, M.; Michel, C.; Wiegrefe, A.; Kalsani, V. Synthesis
2001, 10, 1561−1567.
(75) Schmittel, M.; Ganz, A. Chem. Commun. 1997, 999−1000.
(76) Schmittel, M.; Michel, C.; Liu, S.-X.; Schildbach, D.; Fenske, D.
Eur. J. Inorg. Chem. 2001, 1155−1166.
(77) Pallenberg, A. J.; Koenig, K. S.; Barnhart, D. M. Inorg. Chem.
1995, 34, 2833−2840.
(78) Hsieh, H.-Y.; Lin, C.-H.; Tu, G.-M.; Chi, Y.; Lee, G.-H. Inorg.
Chim. Acta 2009, 362, 4734−4739.
(79) Medwid, J. B.; Paul, R.; Baker, J. S.; Brockman, J. A.; Du, M. T.;
Hallett, W. A.; Hanifin, J. W.; Hardy, R. A.; Tarrant, M. E.; Torley, L.
W.; Wrenn, S. J. Med. Chem. 1990, 33, 1230−1241.
(80) Lafferty, J. J.; Case, F. H. J. Org. Chem. 1967, 32, 1591−1596.
−
(81) The sample of 1+BF4 , which contained a slight amount of
CH2Cl2, was employed for the experiments in the manuscript,
considering the elemental analysis results; the weight ratio of CH2Cl2
−
to 1+BF4 is 1%. 1H NMR measurement in CD2Cl2 revealed a
presence of the residual dichloromethane (Supporting Information
Figure S2), considering the reported values.82 For clarity, we omitted
the remaining solvent.
(82) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I.
Organometallics 2010, 29, 2176−2179.
(83) Electrochemical methods for a square scheme, such as
interpretation for convergence of redox waves and simulation of
voltammograms, are well-established. (a) Bard, A. J.; Faulkner, L. R.
Electrochemical Methods, Fundamentals and Applications, 2nd ed.;
Wiley: New York, 2001. (b) Jacq, J. J. Electroanal. Chem. 1971, 29,
149−180. (c) Carano, M.; Echegoyen, L. Chem.Eur. J. 2003, 9,
1974−1981. (d) Lerke, S. A.; Evans, D. H.; Feldberg, S. W. J.
Electroanal. Chem. 1990, 296, 299−315.
(84) The activation energies for the i- → o-isomerization, Ea, in the
copper(I) and copper(II) states, estimated from the slope of the
Arrhenius plots, are 67 and 35 kJ mol−1, respectively.
(85) The redox potential of cerium(IV)/(III) in water is known to be
1.36 V vs DMFc+/0 in CH2Cl2, which is enough positive to oxidize the
copper(I) complexes. Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996,
96, 877−910.
(86) The redox potential in the excited state was estimated according
to a well-established method. Juris, A.; Balzani, V.; Barigelletti, F.;
Campagna, S.; Belser, P.; von Zelewsky, A. Coord. Chem. Rev. 1988, 84,
85−277.
(87) Fery-Forgues, S.; Delavaux-Nicot, B. J. Photochem. Photobiol. A
2000, 132, 137−159.
(88) It is considered that several undesired reactions can contribute
to a deactivation for the i-CuI → o-CuI photorotation. For example,
10553
dx.doi.org/10.1021/ja3028873 | J. Am. Chem. Soc. 2012, 134, 10543−10553