One photon yields two isomerizations: Large atomic displacements during electronic excited-state dynamics in ruthenium sulfoxide complexes

Article abstract of DOI:10.1021/ja409262r

Photochromic compounds efficiently transduce photonic energy to potential energy for excited-state bond-breaking and bond-forming reactions. A critical feature of this reaction is the nature of the electronic excited-state potential energy surface and how this surface facilitates large nuclear displacements and rearrangements. We have prepared two photochromic ruthenium sulfoxide complexes that feature two isomerization reactions following absorption of a single photon. We show by femtosecond transient absorption spectroscopy that this reaction is complete within a few hundred picoseconds and suggest that isomerization occurs along a conical intersection seam formed by the ground-state and excited-state potential energy surfaces.

Full text of DOI:10.1021/ja409262r

Article
pubs.acs.org/JACS
One Photon Yields Two Isomerizations: Large Atomic Displacements
during Electronic Excited-State Dynamics in Ruthenium Sulfoxide
Complexes
Komal Garg,‡ Albert W. King,‡ and Jeffrey J. Rack*
Nanoscale and Quantum Phenomena Institute, Department of Chemistry and Biochemistry, Ohio University, Athens, Ohio 45701,
United States
S
* Supporting Information
ABSTRACT: Photochromic compounds efficiently transduce pho-
tonic energy to potential energy for excited-state bond-breaking and
bond-forming reactions. A critical feature of this reaction is the
nature of the electronic excited-state potential energy surface and
how this surface facilitates large nuclear displacements and
rearrangements. We have prepared two photochromic ruthenium
sulfoxide complexes that feature two isomerization reactions
following absorption of a single photon. We show by femtosecond
transient absorption spectroscopy that this reaction is complete
within a few hundred picoseconds and suggest that isomerization
occurs along a conical intersection seam formed by the ground-state and excited-state potential energy surfaces.
bonded and the MS is O-bonded.39−41 Our study of these
compounds reveals that the isomerization (time constants
ranging from 45 ps to 1.5 ns) and efficiency (quantum yields
INTRODUCTION
■
Photochromic compounds selectively and specifically transduce
photonic energy to potential energy for excited-state bond-
breaking and bond-forming reactions. Accordingly, they are of
fundamental interest in pursuit of understanding photophysical
ranging from 0.03 to 0.9) of the photochemical transformation
can be tuned through judicious choice of ligand structure.42,43
For example, we recently reported that for [Ru-
(bpy)2(pySO)]2+ (bpy = 2,2′-bipyridine; pySO = 2-(isopropyl-
sulfinylmethyl)pyridine), S→O and O→S isomerizations are
phototriggered with two different wavelengths of light.44
Consistent with this observation, we found that upon
irradiation of this complex with white light in solution a
photostationary state was produced. In another case,
spectroscopic data of [Ru(bpy)2(OSO)]+ (OSO = 2-methyl-
sulfinylbenzoate) are consistent with computational results
showing that isomerization occurs through a conical
intersection.2,45 In contrast to the first example above,
[Ru(bpy)2(OSO)]+ features only a phototriggered S→O
isomerization pathway and not a photochemical O→S
isomerization. Furthermore, we have investigated a series of
bis-sulfoxide complexes with the long-term aim of determining
how two isomerizations may be triggered by a single-photon
absorption.46−50 The fundamental purpose of this study is to
reveal the nature of the interaction between the electronic wave
function and large atomic displacements. Due to the distinct
absorption spectra of these isomers, bis-sulfoxides are ideally
suited for observing this interaction. In this article, we show
that two sulfoxide isomerizations can be triggered from a single-
photon absorption in chelating bis-sulfoxide complexes.
and photochemical reactions involving the coupling of atomic
motions to electronic transitions. Moreover, due to the
disparate electronic and molecular structures exhibited by the
ground state (GS) and metastable state (MS), they provide a
unique chemical environment for investigating how large
atomic motions are coupled with electronic transitions. Indeed,
photochromic compounds have been the focus of certain
computational studies,1−11 which are related to a larger class of
photochemical reactions, where electronic transitions occur
non-adiabatically through conical intersections or conical
intersection seams.12−19 Photoinduced or phototriggered
molecular motions are important design aspects in light-driven
molecular machines,20−23 in photomechanical effects,24−27 in
certain photoresponsive materials,28−33 and in solar thermal
energy storage.34−36
A critical feature of photochromic compounds (and photo-
chemical reactions in general) is the shape and electronic
character of the excited-state potential energy surface that
promotes bond-breaking and bond-forming reactions. In
principle, such knowledge permits optimization of these
surfaces through synthetic design with respect to specific
parameters, such as temporal response, quantum yield,
photoreversibility, or magnitude of color change. Recent
reports on concerted multiple cis−trans isomerizations in
polyenes demonstrate these challenges.37,38 We have created
a class of photochromic compounds based on a phototriggered
isomerism of a coordinated sulfoxide ligand, where the GS is S-
Received: September 6, 2013
Published: January 17, 2014
© 2014 American Chemical Society
1856
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863

Journal of the American Chemical Society
Article
RESULTS AND DISCUSSION
Shown in Chart 1 are bond line sketches of the two complexes
investigated in this study. The compounds, [Ru-
■
Chart 1. Bond Line Drawings of [Ru(bpy)2(OSSO)]2+ (Left)
and [Ru(bpy)2(F-bpSO)]2+ (Right)
(bpy)2 (OSSO)]2 +
, where OSSO = dimethylbis-
(methylsulfinylmethyl)silane, and [Ru(bpy)2(F-bpSO)]2+,
where F-bpSO = 1,2-bis(2-fluorophenylsulfinyl)ethane, were
produced according to literature procedures.46,48,51 Briefly,
[Ru(bpy)2Cl2] was mixed with the bis-thioether ligand to yield
the bis-thioether ruthenium complex, which was subsequently
oxidized to the bis-sulfoxide ruthenium complex. Elemental
analysis and 1H NMR spectra are consistent with the
formulation of each compound.
The electronic spectra of [Ru(bpy)2(OSSO)]2+ and [Ru-
(bpy)2(F-bpSO)]2+ are shown in Figures 1a and 2a,
respectively. For [Ru(bpy)2(OSSO)]2+, the lowest energy
absorption maximum is observed at 355 nm and is assigned
to a Ru dπ→bpy π* metal-to-ligand charge-transfer (MLCT)
transition on the basis of its position and intensity (ε = 4560
M−1 cm−1). The MLCT absorbance maximum for [Ru-
(bpy)2(F-bpSO)]2+ appears as a shoulder on the bipyridine-
centered π→π* transition at ∼340 nm. This shift to the blue
for the MLCT transition is due to stabilization of the sulfur
lone pair by the aromatic phenyl ring. These MLCT absorption
maxima are strongly indicative of Ru dπ orbital stabilization by
the sulfoxides.
Figure 1. (a) Steady-state electronic spectra of S,S-[Ru-
(bpy)2(OSSO)]2+ (blue) and O,O-[Ru(bpy)2(OSSO)]2+ (red). The
difference trace (black) is obtained from subtraction of the S,S
spectrum from the O,O spectrum. (b) Spectral changes in the solution
obtained from bulk photolysis of S,S-[Ru(bpy)2(OSSO)]2+ (blue,
bottom), producing O,O-[Ru(bpy)2(OSSO)]2+ (red, top). Total
irradiation time is 40 min from 400 nm excitation. (c) Transient
spectra collected at pump−probe delays of 0.3, 1.0, 5.0, 15, and 65 ps
(black to magenta), obtained from magic angle pump excitation at 400
nm. Complex concentration is 1.19 × 10−4 M. (d) Transient spectra
collected at pump−probe delays of 65, 150, 350, 600, 1500, and 2800
ps (magenta to bold black), obtained from magic angle pump
excitation at 400 nm.
Irradiation of both [Ru(bpy)2(OSSO)]2+ and [Ru(bpy)2(F-
bpSO)]2+ reveals striking changes in their electronic spectra.
For [Ru(bpy)2(OSSO)]2+, 400 nm excitation shows a rise in
the optical density of the solution resulting in the formation of
two new peaks at 350 and 500 nm (ΦSS→OO = 0.28; Figure 1b).
In accord with literature precedence, these new absorption
maxima are assigned to the bis-O-bonded isomer, in which S→
O isomerization of both sulfoxides is triggered by light.49,52 For
comparison, the closely related [Ru(bpy)2(OH2)2]2+ com-
pound (containing two O-atom donors) features absorption
maxima of 360 and 498 nm.53 Importantly, only a single
isosbestic point is observed, suggesting that the bis-O-bonded
isomer is produced from the S,S-ground state without the
formation of a ground-state reaction intermediate.
We employed femtosecond transient absorption spectrosco-
py to reveal kinetic details of the excited-state reactivity of these
compounds. Shown in Figure 1c,d are the early (<65 ps) and
late (>65 ps) spectral changes for [Ru(bpy)2(OSSO)]2+ in
propylene carbonate at different pump−probe delays following
magic angle excitation at 400 nm. Figure 1a also displays the
difference spectrum obtained from subtraction of the S,S-
isomer from the O,O-isomer (ΔOD = At=t − At=0 = O,O −
S,S). If the O,O-isomer is created following excitation of the
S,S-isomer, then this difference spectrum should be observed in
the transient spectra. The early time traces (Figure 1c, from 0.3
to 65 ps) show a sharp feature at ∼360 nm and a broad excited-
state absorption at λ > 600 nm. The data near the excitation
wavelength have been removed for clarity. The 360 nm feature
has been assigned to a π*→π* absorption of the reduced
Similarly, 355 nm excitation of [Ru(bpy)2(F-bpSO)]2+
results in the formation of a new isomer with absorption
maxima at 360 and 495 nm (ΦSS→OO = 0.10; Figure 2b). Again,
we assign this new spectrum to the bis-O-bonded isomer. In
accord with [Ru(bpy)2(OSSO)]2+, these bulk photolysis data
also show only a single isosbestic point. In both compounds,
the spectral changes are reversible at room temperature, with
complete re-formation of the ground-state bis-S-bonded
isomers occurring over a period of days to weeks. The spectral
changes of this transformation are consistent with two
sequential O→S isomerizations on the ground-state potential
energy surface.
1857
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863

Journal of the American Chemical Society
Article
may be due to the proximity of the laser line, or, more likely, is
due to the formation of the single isomerization product,
namely the S,O-isomer (see below). The appearance of the
isosbestic point in the 65−2800 ps transient spectra indicates
3
transition of a complex in the MLCT state with S-bonded
coordination to a singlet GS with O-bonded coordination
without the formation of a ground-state intermediate. These
data strongly suggest that both S→O isomerizations are
triggered by a single-photon absorption.
A combination of single wavelength and global fitting
analyses were employed to reveal dynamics and isomerization
kinetic data. Global analysis yields two lifetimes of 1.64 0.14
and 233.4 6.9 ps. On the basis of our previous studies, we
assign the 1.64 ps kinetic phase to formation of an isomerizing
3MLCT state, which features a geometry poised for isomer-
ization.56 We ascribe the long component to formation of the
both ground-state isomers (S,O and O,O) from a triplet
excited-state surface. The nature of this surface and this kinetic
component will be discussed later. The single wavelength
kinetics are consistent with the global fitting results. We chose
to fit the data at four wavelengths: 340, 630, 450, and 606 nm.
These first two wavelengths, 340 and 630 nm, correspond to
the reduced bpy π*→π* transition and to the bpy π→Ru3+ dπ
3
transition, respectively, both of the MLCT excited state. This
latter wavelength (630 nm) features essentially no absorbance
from any of the ground-state isomers and thus reports on the
excited-state dynamics, as well as the loss of excited state. Due
to the stimulated emissions features ranging from ∼460 to
∼580 nm, we were unable to obtain reliable kinetic data in this
region. Thus, the data at 450 nm report on formation of
ground-state S,O-, and O,O-bonded isomers. We also fit data at
the isosbestic point (606 nm), which reports on the early time
dynamics of the complex, prior to isomerization. These data are
displayed in Figures S1−S4. The fit at the isosbestic point
reveals a biexponential decay of 1.02 0.3 and 5.0 1.8 ps.
The first time constant is consistent with that obtained from
the global fit analysis. At present, we do not have an assignment
for the 5.0 ps kinetic component, nor do we understand why it
is not present in the global analysis. This kinetic trace does not
contain the long time kinetic phase corresponding to formation
of ground-state products. We fixed the shortest time
component (1.0 ps) for the modeling of the kinetic traces at
the other wavelengths in order to obtain reliable fits of the data.
For 340 nm, a triexponential fit with a fixed 1.0 ps phase
Figure 2. (a) Steady-state electronic spectra of S,S-[Ru(bpy)2(F-
bpSO)]2+ (blue) and O,O-[Ru(bpy)2(F-bpSO)]2+ (red). The differ-
ence trace (black) is obtained from subtraction of the S,S spectrum
from the O,O spectrum. (b) Spectral changes in the solution obtained
from bulk photolysis of S,S-[Ru(bpy)2(F-bpSO)]2+ (blue, bottom),
producing O,O-[Ru(bpy)2(F-bpSO)]2+ (red, top). Total irradiation
time is 30 min from 355 nm excitation. Complex concentration is 1.61
× 10−4 M. (c) Transient spectra collected at pump−probe delays of
0.75, 1.0, 5.0 10, 20, 75, 500, and 3000 ps (red to bold black).
bipyridine in the MLCT excited state, while the low-energy
feature has been previously ascribed to both an LMCT
transition from the unreduced bipyridine to Ru(III) and π*→
π* absorption of the reduced bipyridine.54,55 The sharp,
negative peaks near 500 nm are due to stimulated emission, as
the S,S-isomer shows no ground-state absorbance at this
wavelength. Over the next 15 ps, the 360 nm feature reduces in
intensity, with only minor changes elsewhere in the spectrum.
Dramatic changes occur from 15 to 2800 ps, where the
absorption at λ > 600 nm collapses to zero, thus indicating
conversion from a triplet excited-state potential energy surface
to a singlet ground-state potential energy surface. The 360 nm
feature reduces in intensity concomitant with a rise of the
absorption near 500 nm, with the appearance of an isosbestic
point near 600 nm. The 2800 ps transient spectrum (bold black
trace, Figure 1d) and the ground-state difference spectrum
(black trace, Figure 1a) are similar in that both show an intense
absorption at ∼500 nm ascribed to the bis-O-bonded isomer.
However, the absorption near 400 nm in the 2800 ps transient
is not duplicated in the ground-state difference spectrum. This
revealed time constants of 2.8
1.1 and 250
51 ps. We
ascribe the large error for the 250 ps time constant to the poor
signal-to-noise ratio at this wavelength. At 450 nm, we found
time constants of 5.5 2.0 and 213 8.2 ps. At 630 nm, we
found time constants of 4.9 1.8 and 220 29 ps. In accord
with the global fitting results, the long time component is
assigned to the formation of ground-state S,S-, S,O-, and O,O-
3
bonded isomers from the isomerizing MLCT excited state.
Surprisingly, neither global fitting nor single-wavelength kinetic
fits reveal a short sub-picosecond time constant. This kinetic
3
component is typically attributed to formation of the MLCT
1
excited state from the initially populated MLCT excited state
and, in closely related [Ru(bpy)3]2+, is known to occur in as
short a time as 30 fs.57−59 We assume that this transition occurs
within the pulse width of our instrument (150 fs). It is striking
that the kinetic analysis returns a relatively simple description of
a photochemical reaction involving two isomerizations.
The femtosecond transient absorption spectroscopic data of
[Ru(bpy)2(F-bpSO)]2+ are qualitatively similar to those
1858
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863

Journal of the American Chemical Society
Article
observed for [Ru(bpy)2(OSSO)]2+. Shown in Figure 2c are the
spectral traces collected at different pump−probe delays
obtained from magic angle excitation at 355 nm. The data
near 355 and 710 nm (wavelength employed to create 355 nm)
are omitted for clarity. The 0.75 ps trace exhibits a broad
absorption from nearly 500 nm well past 700 nm, which is
ascribed to an LMCT transition and a ligand-centered
and monitored the absorbance at 500 nm, the absorption
maxima of the bis-O-bonded isomers for both complexes. We
found a linear correlation (Figures S7 and S8) between the
pump intensity (ranging from 300 to 600 μW), which is
expected for a one-photon absorption mechanism. Moreover,
we do not observe kinetic traces at any wavelength that are
consistent with a mechanism in which a ground-state isomer is
formed after the initial excitation and is then re-excited. For
example, the transient spectra in the red region of the spectrum
and resultant kinetic traces illustrate this point. This excited-
state absorption is assigned to bpy π→Ru3+ dπ LMCT on the
basis of literature precedence.54,55 Importantly, there is no
ground-state absorption at these long wavelengths (∼650 nm
for both compounds, see Figures 1a and 2a) for either of the
bis-S-isomers discussed here. Thus, the kinetic changes at these
wavelengths are due only to excited-state transformations. That
kinetic traces only show a sub-picosecond rise (excited-state
formation), followed by a long time decay to zero absorbance
(ground state), indicates that all ground-state isomers are
formed from the same excited state after the initial excitation.
That is, if the bis-O-bonded isomer were formed from two
successive photon absorptions from ground-state isomers, then
one would expect a kinetic trace with two successive rise
(excited-state formation) and fall (ground-state formation)
cycles in the red region of the spectrum. The linear power
dependence and kinetic traces argue against the absorption of
two photons to trigger this photochemistry.
3
transition of the reduced bipyridine of the MLCT excited
state. There are also two excited-state absorption features near
360 and 450 nm. We assign the 360 nm absorption to the
intraligand π*→π* absorption of the reduced bipyrdine, but are
uncertain of the origin of the 450 nm transition. There appears
to be a relatively weak stimulated emission feature near 500 nm,
similar to [Ru(bpy)2(OSSO)]2+ described above. As time
evolves over the next 500 ps, the far visible absorption at 650
nm reduces to zero, concomitant with the appearance of two
new ground-state absorption transitions at 400 and 500 nm.
Again, the 500 ps transient spectrum (bold black trace, Figure
2c) is in good agreement with the steady-state difference
spectrum obtained from the spectral subtraction of the S,S-
isomer from the photoproduct generated by bulk photolysis
(black trace, Figure 2a). In accordance with the spectral data of
[Ru(bpy)2(OSSO)]2+ and other isomerizing bis-sulfoxides, the
500 nm absorption is assigned to the bis-O-bonded isomer.
The more intense 400 nm transition is assigned to a mixed S,O-
bonded bis-sulfoxide, where one sulfoxide is S-bonded and the
other is O-bonded. Such a species is observed in the
photochemical reactivity of [Ru(bpy)2(dmso)2]2+,49,52 and
[Os(bpy)2(dmso)2]2+ (dmso = dimethylsulfoxide).47 More-
over, we recently discovered the photochemical formation of
such an isomer in [Ru(bpy)2(bpSO)]2+, where bpSO = 1,2-
bis(phenylsulfinyl)ethane.46 In that study, we demonstrated
through pump−repump−probe spectroscopy that each sulf-
oxide isomerization was triggered sequentially through two
separate single-photon absorptions. In contrast, the transient
absorption spectra for [Ru(bpy)2(F-bpSO)]2+ clearly indicate
the formation of two isomeric products from a single excitation
event, namely S,O-[Ru(bpy)2(F-bpSO)]2+ and O,O-[Ru-
(bpy)2(F-bpSO)]2+.
For both compounds, the data indicate that multiple
photoproducts formed from phototriggered S→O isomer-
ization occur from a single excitation or photon absorption.
While the excited-state (MLCT) symmetry of each complex is
C1 (localization of the promoted electron on one bipyridine
ligand), the local ground-state symmetry is C2v (see NMR
spectra in Supporting Information). In this local symmetry, the
σ-bonding framework of the S−Ru−S moiety is formed by an
a1 and b2 SALC (symmetry-adapted linear combination) with
2
the Ru dz and dyz orbitals (the z-axis bisects the S−Ru−S
chelate; Figure 3). This simple picture demonstrates that in
Similar to [Ru(bpy)2(OSSO)]2+, single-wavelength and
global fitting analyses were performed to characterize the
dynamics and isomerization kinetics of [Ru(bpy)2(F-bpSO)]2+.
Triexponential fits were observed at 400 and 600 nm (Figures
S5 and S6). These wavelengths correspond to an absorption
maximum of the photoisomerization products and to loss of the
3MLCT excited state, respectively. We were unable to obtain
reliable fits at 500 nm, an absorption maximum for the bis-O,O-
isomer, due to the stimulated emission features ranging from
∼475 to ∼525 nm. Fitting at 400 nm yields lifetimes of 0.23
0.02, 10.6 0.3, and 121 25 ps, with similar values obtained
at 600 nm. The observed time constants found by global fitting
analysis are in good agreement with single-wavelength fits,
providing values of 0.45 0.06, 11.1 1.3, and 114 60 ps.
We ascribe the sub-picosecond component to formation of the
3MLCT state from the 1MLCT state, consistent with numerous
photophysical reports of [Ru(bpy)3]2+. The ∼11 ps kinetic
local C2v (and excited-state C1 symmetry, though with different
3
component is assigned to formation of an isomerizing MLCT
state, and the longest time component (115−120 ps) is
attributed to formation of ground-state S,S-, S,O-, and O,O-
isomers, with the S,O- and O,O-isomers displaying new
absorption maxima near 400 and 500 nm, respectively.
In order to further test the hypothesis that one absorbed
photon yields two isomerizations, we varied the pump intensity
Figure 3. Sketch of orbital interactions depicting the S−Ru−S σ- and
σ*-bonding manifolds.
1859
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863

Journal of the American Chemical Society
Article
Mulliken symbols) symmetry, both sulfoxides share common σ-
bonding (and antibonding) Ru orbitals, suggesting that two
isomerizations from a single-photon absorption is possible.
Depopulation of these σ-bonding orbitals or population of their
antibonding counterparts will affect both Ru−S bonds
simultaneously, but not necessarily to the same degree, due
to effects associated with the reduced bipyridine ligand. Thus, if
the excited-state potential energy surface contains the σ* orbital
pictured here, then both Ru−S bonds will be broken or
weakened upon excitation. Moreover, as the ground-state
orbital description involves Ru dπ stabilization by the sulfoxides
(see above), depopulation of these occupied orbitals from light
excitation will also weaken both Ru−S bonds. Accordingly,
visible light excitation is expected to alter Ru−S bonding
through both σ and π interactions. The first molecular events in
isomerization must involve Ru−S bond breaking.
transformations and often accompanies a rapid transition from
an excited-state potential energy surface to a ground-state
potential energy surface.60,61 We observe a red-shifting of the
stimulated emission in the early time (0.3−15 ps) transient
spectra of [Ru(bpy)2(OSSO)]2+. This shift is explained as
emission from an excited-state surface to a rapidly rising
ground-state surface (from loss of Ru dπ stabilization) in the
vicinity of the conical intersection (seam). We suspect that the
strong overlap of these signals prevents our clear observation of
these effects in the later time spectra (15−2800 ps) of
[Ru(bpy)2(OSSO)]2+ and on any time interval in [Ru(bpy)2(F-
bpSO)]2+.
Another description for the formation of multiple products
from sequential isomerizations is from a single, electronic
excited state or potential energy surface. For example, excitation
of the S,S ground state produces the Franck−Condon states on
1
3
the MLCT surface. Relaxation to the MLCT surface occurs
on the sub-picosecond time scale with S-bonded character,
which subsequently yields the isomerizing state or surface in 1−
10 ps (Figure 4). As stated above, the reaction pathway from
One may view this reaction as occurring from an excited-
state potential energy surface that permits the formation of
multiple photoproducts from sequential isomerizations. From
the bonding analysis above, this surface must involve the σ*-
bonding manifold, also known as the ligand field (LF) or metal-
centered (MC) states. The surface must also accommodate
orthogonal vibrations and nuclear motions involving each
sulfoxide isomerization separately. Indeed, the motions leading
to the first isomerization will not lead to the second
isomerization, though there must exist vibrations common to
both isomerizations. Also, the surface must contain contribu-
3
tions from the MLCT state, which is reached upon excitation
and triggers both isomerizations (metal oxidation is required
for isomerization). Thus, the excited-state potential energy
surface that leads to isomerization must involve multiple
orthogonal vibrations, comprise both LF (or MC) and MLCT
character, and intersect with the ground-state potential energy
surface. Such a mechanism suggests the reaction occurs in a
non-adiabatic fashion, in which the productive isomerization
vibrations are strongly coupled with electronic relaxation from
the triplet excited-state surface to the singlet ground-state
surface. The intersection of the excited-state surface with the
ground-state potential energy surface is more commonly known
as a conical intersection seam. The observed product ratio is
sensitive to where decay takes place along this seam.
Apparently, the conical intersection seams between the
ground-state and excited-state potential energy surfaces in
[Ru(bpy)2(OSSO)]2+ and [Ru(bpy)2(F-bpSO)]2+ are different
enough to alter the S,O- to O,O-isomer product ratio (Scheme
1).
Figure 4. Energy diagram describing photochemical reactivity for
[Ru(bpy)2(OSSO)]2+ and [Ru(bpy)2(F-bpSO)]2+. The vertical arrows
connecting the triplet excited-state surface with the single ground-state
surface represent both radiative and nonradiative deactivation
pathways.
Scheme 1. Photochemical Reaction Producing Two Isomers
such a state must involve orthogonal reaction coordinates, as
certain of the motions producing the first isomerization are
distinct from the second isomerization. That is, once on the
triplet excited-state surface, which comprises both LF and
MLCT character, the thermodynamic barriers for atomic
rearrangement are small (ΔH⧧ < kBT), indicating that largely
different structures can be accessed with little energy required
to reach a new conformation. Thus, all three isomeric structures
on the excited-state surface are in fast equilibrium in
comparison to decay to the ground-state potential energy
surface. We strongly suspect that these isomerizations are
kinematically coupled (lowering the energy barrier between
isomers), but these data do not address this hypothesis. In
contrast, the ground-state potential energy surface, which
allows formation of the S,S-isomer through sequential isomer-
izations from the O,O-isomer, shows large thermodynamic
The spectroscopic data presented here strongly support this
interpretation, in particular the observation of isosbestic points
3
connecting spectral features of MLCT excited states with
singlet ground states. Indeed, a non-adiabatic isomerization
mechanism was found in the computational study of a related
photochromic ruthenium sulfoxide compound.2 Also, stimu-
lated emission has been observed in certain non-adiabatic
1860
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863

Journal of the American Chemical Society
Article
3.5 W) where one beam is directed to a Light Conversion TOPAS-C
OPA to generate the pump beam used in the experiment, and the
other beam is used to generate the white light continuum (∼330−850
nm) probe beam by passing the 800 nm pulse through a CaF2 crystal.
The probe beam is ∼0.2 mm in diameter, and slightly smaller than the
pump beam diameter. Typical pump beam power at the sample ranges
from 200 to 600 μW. Each spectrum was integrated for 2 s, and the
sample was flowed through a fluid pump (Lab Pump Jr., model RHSY
by Fluid Metering, Inc., Syosset, NY) flowing at ∼7.5 mL/min.
Transient data were corrected using Surface Xplorer Pro 1.1.5 software
(Ultrafast Systems) by subtracting spectral background features that
persisted from the previous pulse and appeared pre-pulse, and applying
chirp and t0 corrections. For magic angle excitation experiments, the
pump beam was directed through a polarizer (Thor Laboratories
WPH10M) set in an optical mount and rotated to 18.2°.
barriers between these isomers. Such an observation is
consistent with the observed slow thermal reversion rates at
room temperature. This mechanism does not require that both
isomerizations occur in a concerted or simultaneous fashion;
rather, both isomerizations must occur from the triplet excited-
state surface.
There are few ultrafast studies of structural rearrangements in
transition metal complexes. The recent results of Cu(I) diimine
compounds are relevant in that they feature a flattening of the
angle formed from two diimine complexes from 90° to 68° with
a time constant of 10−15 ps, which is surprisingly independent
of diimine and solvent.62,63 We have recently observed a solvent
dependence on our isomerization, but not on the internal
conversion (IC) process involving two separate 3MLCT
states.42 Woike and co-workers reported from optical transient
absorption spectroscopy an isomerization time constant of 300
20 fs in sodium nitroprusside (Na2[Fe(CN)5(NO)]) crystals,
where NO rotates from N-bonded to side-on NO (η2).64 These
measurements are consistent with time-resolved IR studies.65
The studies on copper and iron both involve a strong role for
the Jahn−Teller distortion in the structural rearrangement or
isomerization. Finally, the elongation of Fe−N bonds in light-
induced spin-crossover compounds is on the order of 150 fs as
measured by a variety of optical and X-ray techniques.66 These
examples, in conjunction with our studies, demonstrate that a
variety of large-amplitude motions can occur on very short time
scales.
These results on ruthenium bis-sulfoxides represent a current
optimization in excited-state nuclear rearrangements triggered
by visible light irradiation as well as in storage of potential
energy from photonic energy. Both of these parameters are
important in the design of photoactive materials for photo-
tropic or photomechanical effects as well as in photo-origami
constructions. The more efficiently a complex (or material) can
absorb light and transduce this energy to specific structural and
shape deformations, the greater the utility of the complex.
Similarly, the efficiency of solar thermal energy storage is
predicated on maximizing the conversion of photonic energy to
potential energy, which is stored in chemical bonds. For these
bis-sulfoxide compounds, one photon leading to two isomer-
izations indicates more energy is stored per absorbed photon.
That we have observed these results in two separate complexes
suggests that the unusual reactivity described here may be more
general than originally expected. The goal of future studies is to
further investigate these and related complexes to help
understand the coupling of large atomic displacements with
electronic transitions in excited-state reactions.
[Ru(bpy)2(SS)](PF6)2. The complex Ru(bpy)2Cl2·2H2O (50 mg,
0.096 mmol), dithioether (SS, 20 μL, 0.108 mmol), and AgPF6 (52.1
mg, 0.02 mmol) were dissolved in 50 mL of 1,2-dichloroethane and
refluxed for 3 h under N2 atmosphere. The solution changed color
from purple to deep red, and the precipitation of AgCl occurred. The
solution was stored in the refrigerator overnight to induce complete
precipitation of AgCl and was filtered by using a fine frit. The filtrate
solution was evaporated and then washed with diethyl ether. The
resulting red solid product was isolated via vacuum filtration, washed
with diethyl ether (3 × 15 mL), and air-dried. Yield: 63 mg (75%).
1
UV−vis (CH3CN): λmax = 422 nm. H NMR (CD2Cl2, 300 MHz): δ
(ppm) = 9.63 (d, bpy, 2 H), 8.53 (d, bpy, 2 H), 8.38 (m, bpy, 4 H),
8.0 (m, bpy, 4 H), 7.43 (d, bpy, 2 H), 7.39 (t, bpy, 2 H), 1.92 (d, CH2,
2 H), 1.85 (d, CH2, 2 H), 1.2 (s, SCH3, 6 H), 0.42 (s, SiCH3, 6 H).
13C NMR (CD3CN, 75 MHz): δ (ppm) = 158.7 (bpy), 157.5 (bpy),
153.4 (bpy), 152.2 (bpy), 139.9 (bpy), 139.7 (bpy), 129.3 (bpy),
128.6 (bpy), 126.0 (bpy), 125.4 (bpy), 20.8 (SCH2), 20.7 (SCH3), 2.8
(SiCH3). Elemental analysis (%), calcd for [RuC26H32F12N4P2S2Si]: C,
35.33; H, 3.65; N, 6.34. Found: C, 35.69; H, 3.79; N, 6.50.
[Ru(bpy)2(OSSO)](PF6)2. Oxidation of [Ru(bpy)2(SS)](PF6)2 was
achieved by dissolving [Ru(bpy)2(SS)](PF6)2 (50 mg, 0.0577 mmol)
and mCPBA (21.5 mg, 0.115 mmol) in 30 mL of acetonitrile. The
reaction was allowed to stir at room temperature in the dark for 3 h,
and the progress of the reaction was observed by the blue shift of
MLCT transition in the UV−vis spectrum. The solution was
evaporated and reduced to <5 mL, and the product was isolated by
addition of diethyl ether. The yellow colored product was isolated by
vacuum filtration, washed with diethyl ether, and air-dried. Yield: 45
mg (90%). UV−vis (CH3CN): λmax = 355 nm. IR (KBr pellet): S-
bonded, νasym(SO) = 1072 cm−1, νsym(SO) = 1022 cm−1; O-bonded,
νasym(SO) = 986 cm−1, νsym(SO) = 935 cm−1. 1H NMR (CD3CN, 300
MHz): δ (ppm) = 10.05 (d, bpy, 2 H), 8.56 (d, bpy, 2 H), 8.45 (d,
bpy, 2 H), 8.38 (t, bpy, 2 H), 8.13 (t, bpy, 2 H), 7.98 (t, bpy, 2 H),
7.44 (t, bpy, 2 H), 7.25 (d, bpy, 2 H), 3.43 (d, CH2, 2 H), 2.75 (d,
CH2, 2 H), 2.11 (s, SCH3, 6 H), 0.53 (s, SiCH3, 6 H). 13C NMR
(CD3CN, 75 MHz): δ (ppm) = 157.8 (bpy), 156.4 (bpy), 155.9
(bpy), 151.2 (bpy), 141.9 (bpy), 141.7 (bpy), 130.4 (bpy), 129.8
(bpy), 126.8 (bpy), 126.1 (bpy), 44.2 (SCH2), 15.6 (SCH3), −0.48
(SiCH3). Elemental analysis (%), calcd for [RuC26H32F12N4P2S2O2Si]:
C, 34.10; H, 3.52; N, 6.12. Found: C, 34.58; H, 3.60; N, 6.40.
1,2-Bis(2-fluorophenylthio)ethane (F-bpte). The starting ma-
terial 2-fluorothiophenol (2.6160 g, 20.41 mmol) was magnetically
stirred with sodium hydroxide (1.1610 g, 29.03 mmol) at room
temperature for 30 min in 25 mL of methanol. Solvent was removed
by rotary evaporation, and the resulting white solid was redissolved in
25 mL of acetone, after which 1,2-dibromoethane (1.6677 g, 8.88
mmol) was added, and the reaction mixture was brought to reflux
while stirring for 4 h. The reaction mixture was cooled to room
temperature and filtered. Solvent was removed from the filtrate by
rotary evaporation; the resulting white solid was dissolved in 25 mL of
toluene and washed with two 20 mL portions of water. The organic
phase was dried over magnesium sulfate, and solvent was removed by
rotary evaporation. The resultant thin, pale yellow oil was dried on a
Schlenk line overnight to yield a pale yellow solid. Yield: 667.5 mg
EXPERIMENTAL SECTION
■
The starting materials (RuCl3·xH2O) and silver hexafluorophosphate
(AgPF6) were purchased from Strem. The reagents 2-fluorothiophe-
nol, 1,2-dibromoethane, and 3-chloroperoxybenzoic acid (mCPBA)
were purchased from Sigma-Aldrich. ACS-grade solvents were
purchased from Fisher Scientific and used without further purification.
The starting material Ru(bpy)2Cl2·2H2O was produced in accord with
the literature procedure.51 One of the dithioether ligands (dimethyl
bis(dimethylthiomethylene)silane, SS) was received as a gift from Dr.
Rabinovich at UNC-Charlotte.
The details of our transient absorption instrumentation have been
described elsewhere.56 Briefly, Ultrafast transient absorption data were
collected using an Ultrafast Systems HELIOS spectrometer, pumped
by a Spectra Physics Solstice containing a one-box regenerative
amplifier, Mai Tai fs oscillator and an Empower pump laser. Light
from the Solstice is split (50:50) to produce pulses of 800 nm (1 kHz,
1861
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863

Journal of the American Chemical Society
Article
1
(26.6%). H NMR (acetone-d6, 300 MHz): δ (ppm) = 7.43 (t, 2 H),
(5) Boggio-Pasqua, M.; Bearpark, M. J.; Robb, M. A. J. Org. Chem.
2007, 72, 4497.
7.34 (m, 2 H), 7.18 (t, 4 H), 3.18 (s, 4 H).
[Ru(bpy)2(F-bpte)](PF6)2. The difluorophenyl dithioether ruthe-
nium complex was synthesized in a manner similar to that used for
[Ru(bpy)2(bpte)](PF6)2. For this synthesis, Ru(bpy)2Cl2·2H2O
(399.9 mg, 0.77 mmol), AgPF6 (392.4 mg, 1.55 mmol), and F-bpte
(449.2 mg, 1.60 mmol) were used. After 3 h, additional AgPF6 (191.5
mg, 0.76 mmol) was added. The reaction was allowed to continue for
an additional 2 h, at which time the reaction mixture was moved to a
freezer to facilitate the complete precipitation of AgCl overnight. The
precipitate was removed by filtration through a fine frit, and solvent
was removed from the filtrate by rotary evaporation. After being
precipitated from a concentrated solution in methanol by the addition
of a saturated aqueous solution of NH4PF6, the ruthenium complex
was recrystallized twice from methanol using diethyl ether to
precipitate a bright yellow powder. Yield: 0.5238 g (69.2%). UV−vis
(PC): λmax = 388 nm. 1H NMR (acetone-d6): δ (ppm) = 9.80 (d, bpy,
2 H), 8.39 (t, bpy, 4 H), 8.28 (t, bpy, 2 H), 8.14 (t, bpy, 2 H), 8.04 (m,
bpy, 4 H), 7.52 (t, bpy, 2 H), 7.42 (m, F-bpte, 2 H), 6.94 (m, F-bpte, 4
H), 6.85 (t, F-bpte, 2 H), 4.19 (d, CH2, 2 H), 3.64 (d, CH2, 2 H).
Elemental analysis (%), calcd for [Ru(C10H8N2)2(C14H12F2S2)](PF6)2·
2H2O: C, 39.97; H, 3.16; N, 5.48. Found: C, 39.95; H, 3.02; N, 5.43.
[Ru(bpy)2(F-bpSO)](PF6)2. Oxidation of [Ru(bpy)2(F-bpte)]2+
(51.3 mg, 0.052 mmol) was achieved by the method previously
described.46 Upon completion of the oxidation (monitored by UV−
vis), the product was precipitated by the addition of diethyl ether, and
the precipitate was filtered on a fine frit. The product was mixed with
approximately 3 mL of DCM in a small vial and brought to ambient
temperature in a freezer. The remaining solid was filtered on a fine frit.
This process was repeated twice to yield the pure product. Yield: 38.5
mg (75.0%). UV−vis (PC): λmax = 338 nm. IR (KBr pellet): S-bonded,
νasym(SO) = 1101 cm−1, νsym(SO) = 1085 cm−1; O-bonded, νasym(SO)
= 960 cm−1, νsym(SO) = 937 cm−1. 1H NMR (acetone-d6): δ (ppm) =
10.01 (d, bpy, 2 H), 8.49 (t, bpy, 4 H), 8.45 (m, bpy, 2 H), 8.30 (t,
bpy, 2 H), 8.13 (t, bpy, 2 H), 7.79 (d, bpy, 2 H), 7.68 (m, bpy/F-
bpSO, 4 H), 7.23 (m, F-bpSO, 4 H), 7.04 (m, F-bpSO, 2 H), 5.12 (d,
CH2, 2 H), 4.78 (d, CH2, 2 H). Elemental analysis (%), calcd for
[Ru(C10H8N2)2(C14H12F2O2S2)](PF6)2: C, 40.13; H, 2.77; N, 5.50.
Found: C, 40.12; H, 3.02; N, 5.68.
(6) Boggio-Pasqua, M.; Ravaglia, M.; Bearpark, M. J.; Garavelli, M.;
Robb, M. A. J. Phys. Chem. A 2003, 107, 11139.
(7) Celani, P.; Bernardi, F.; Olivucci, M.; Robb, M. A. J. Am. Chem.
Soc. 1997, 119, 10815.
(8) Gomez, I.; Reguero, M.; Robb, M. A. J. Phys. Chem. A 2006, 110,
3986.
(9) Nakamura, S.; Yokojima, S.; Uchida, K.; Tsujioka, T.; Goldberg,
A.; Murakami, A.; Shinoda, K.; Mikami, M.; Kobayashi, T.; Kobatake,
S.; Matsuda, K.; Irie, M. J. Photochem. Photobiol. A 2008, 200, 10.
(10) Sanchez-Lozano, M.; Estevez, C. M.; Hermida-Ramon, J.;
Serrano-Andres, L. J. Phys. Chem. A 2011, 115, 9128.
(11) Tomasello, G.; Ogliaro, F.; Bearpark, M. J.; Robb, M. A.;
Garavelli, M. J. Phys. Chem. A 2008, 112, 10096.
(12) Martinez, T. J. Nature 2010, 467, 412.
(13) Robb, M. A.; Bernardi, F.; Olivucci, M. Pure Appl. Chem. 1995,
67, 783.
(14) Yarkony, D. R. Rev. Mod. Phys. 1996, 68, 985.
(15) Yarkony, D. R. J. Phys. Chem. A 2001, 105, 6277.
(16) Domcke, W.; Yarkony, D. R. Annu. Rev. Phys. Chem. 2012, 63,
325.
(17) Levine, B. G.; Martinez, T. J. Annu. Rev. Phys. Chem. 2007, 58,
613.
(18) Matsika, S.; Krause, P. Annu. Rev. Phys. Chem. 2011, 62, 621.
(19) Yarkony, D. R. Acc. Chem. Res. 1998, 31, 511.
(20) Balzani, V.; Credi, A.; Venturi, M. Chem. Soc. Rev. 2009, 38,
1542.
(21) Bonnet, S.; Collin, J.-P. Chem. Soc. Rev. 2008, 37, 1207.
(22) Feringa, B. L. J. Org. Chem. 2007, 72, 6635.
(23) Saha, S.; Stoddart, J. F. Chem. Soc. Rev. 2007, 36, 77.
(24) Lee, K. M.; White, T. J. Macromolecules 2012, 45, 7163.
(25) Shimamura, A.; Priimagi, A.; Mamiya, J.; Ikeda, T.; Yu, Y. L.;
Barrett, C. J.; Shishido, A. ACS Appl. Mater. Interfaces 2011, 3, 4190.
(26) Morimoto, M.; Irie, M. J. Am. Chem. Soc. 2010, 132, 14172.
(27) Zhu, L.; Al-Kaysi, R. O.; Bardeen, C. J. J. Am. Chem. Soc. 2011,
133, 12569.
(28) Allampally, N. K.; Stassert, C. A.; De Cola, L. Dalton Trans.
2012, 41, 13132.
ASSOCIATED CONTENT
* Supporting Information
NMR spectra and kinetic data. This material is available free of
charge via the Internet at http://pubs.acs.org.
(29) Liu, D.; Bastiaansen, C. W. M.; den Toonder, J. M. J.; Broer, D.
J. Macromolecules 2012, 45, 8005.
(30) Pieroni, O.; Fissi, A.; Angelini, N.; Lenci, F. Acc. Chem. Res.
■
S
2001, 34, 9.
(31) Koshima, H.; Ojima, N.; Uchimoto, H. J. Am. Chem. Soc. 2009,
131, 6890.
(32) O’Bryan, G.; Wong, B. M.; McElhanon, J. R. ACS Appl. Mater.
Interfaces 2010, 2, 1594.
(33) Toshchevikov, V.; Saphiannikova, M.; Heinrich, G. J. Phys.
Chem. A 2012, 116, 913.
AUTHOR INFORMATION
Corresponding Author
rackj@ohio.edu
■
Author Contributions
‡K.G. and A.W.K. contributed equally.
(34) Kucharski, T. J.; Tian, Y.; Akbulatov, S.; Boulatov, R. Energy
Environ. Sci. 2011, 4, 4449.
Notes
(35) Moth-Poulsen, K.; Coso, D.; Borjesson, K.; Vinokurov, N.;
Meier, S. K.; Majumdar, A.; Vollhardt, K. P. C.; Segalman, R. A. Energy
Environ. Sci. 2012, 5, 8534.
(36) Kolpak, A. M.; Grossman, J. C. Nano Lett. 2011, 11, 3156.
(37) Saltiel, J.; Bremer, M. A.; Laohhasurayotin, S.; Krishna, T. S. R.
Angew. Chem., Int. Ed. 2008, 47, 1237.
(38) Saltiel, J.; Papdimtriou, D.; Krishna, T. S. R.; Huang, Z.-N.;
Krishnamoorthy, G.; Laohhasurayotin, S.; Clark, R. J. Angew. Chem.,
Int. Ed. 2009, 48, 8082.
(39) McClure, B. A.; Rack, J. J. Eur. J. Inorg. Chem. 2010, 3895.
(40) Rack, J. J. Z. Kristallogr. 2008, 223, 356.
(41) Rack, J. J. Coord. Chem. Rev. 2009, 253, 78.
(42) Garg, K.; Engle, J. T.; Ziegler, C. J.; Rack, J. J. Chem.Eur. J.
2013, 19, 11686.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This material is based upon work supported by the National
Science Foundation under Grants CHE 0809699, 0947031, and
1112250.
REFERENCES
■
(1) Bearpark, M. J.; Boggio-Pasqua, M.; Robb, M. A.; Ogliaro, R.
Theor. Chem. Acc. 2006, 116, 670.
(2) Gottle, A. J.; Dixon, I. M.; Alary, F.; Heully, J. L.; Boggio-Pasqua,
M. J. Am. Chem. Soc. 2011, 133, 9172.
(3) Indelli, M. T.; Carli, S.; Ghirotti, M.; Chiorboli, C.; Ravaglia, M.;
Garavelli, M.; Scandola, F. J. Am. Chem. Soc. 2008, 130, 7286.
(4) Boggio-Pasqua, M.; Bearpark, M. J.; Hunt, P. A.; Robb, M. A. J.
Am. Chem. Soc. 2002, 124, 1456.
(43) McClure, B. A.; Abrams, E. R.; Rack, J. J. J. Am. Chem. Soc. 2010,
132, 5428.
(44) McClure, B. A.; Rack, J. J. Angew. Chem., Int. Ed. 2009, 48, 8556.
1862
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863

Journal of the American Chemical Society
Article
(45) McClure, B. A.; Mockus, N. V.; Butcher, D. P.; Lutterman, D.
A.; Turro, C.; Petersen, J. L.; Rack, J. J. Inorg. Chem. 2009, 48, 8084.
(46) King, A. W.; Jin, Y.; Engle, J. T.; Ziegler, C. J.; Rack, J. J. Inorg.
Chem. 2013, 52, 2086.
(47) Mockus, N. V.; Petersen, J. L.; Rack, J. J. Inorg. Chem. 2006, 45,
8.
(48) Mockus, N. V.; Rabinovich, D.; Petersen, J. L.; Rack, J. J. Angew.
Chem., Int. Ed. 2008, 47, 1458.
(49) Rack, J. J.; Mockus, N. V. Inorg. Chem. 2003, 42, 5792.
(50) Rachford, A. A.; Petersen, J. L.; Rack, J. J. Dalton Trans. 2007,
3245.
(51) Sullivan, B. P.; Salman, D. J.; Meyer, T. J. Inorg. Chem. 1978, 17,
3334.
(52) Smith, M. K.; Gibson, J. A.; Young, C. G.; Broomhead, J. A.;
Junk, P. C.; Keene, F. R. Eur. J. Inorg. Chem. 2000, 1365.
(53) Dobson, J. C.; Meyer, T. J. Inorg. Chem. 1988, 27, 3283.
(54) Damrauer, N. H.; McCusker, J. K. J. Phys. Chem. A 1999, 103,
8440.
(55) Braterman, P. S.; Song, J.-I.; Peacock, R. D. Inorg. Chem. 1992,
31, 555.
(56) McClure, B. A.; Rack, J. J. Inorg. Chem. 2011, 50, 7586.
(57) Bhasikuttan, A. C.; Suzuki, M.; Nakashima, S.; Okada, T. J. Am.
Chem. Soc. 2002, 124, 8398.
(58) Damrauer, N. H.; Cerullo, G.; Yeh, A.; Boussie, T. R.; Shank, C.
V.; McCusker, J. K. Science 1997, 275, 54.
(59) Yoon, S.; Kukura, P.; Stuart, C. M.; Mathies, R. A. Mol. Phys.
2006, 104, 1275.
(60) Polli, D.; Altoe, P.; Weingart, O.; Spillane, K. M.; Manzoni, C.;
Brida, D.; Tomasello, G.; Orlandi, G.; Kukura, P.; Mathies, R. A.;
Garavelli, M.; Cerullo, G. Nature 2010, 467, 440.
(61) Shim, S.; Joo, T.; Bae, S. C.; Kim, K. S.; Kim, E. J. Phys. Chem. A
2003, 107, 8106.
(62) Iwamura, M.; Watanabe, H.; Ishii, K.; Takeuchi, S.; Tahara, T. J.
Am. Chem. Soc. 2011, 133, 7728.
(63) Shaw, G. B.; Grant, C. D.; Shirota, H.; Castner, E. W., Jr.;
Meyer, G. J.; Chen, L. X. J. Am. Chem. Soc. 2007, 129, 2147.
(64) Schaniel, D.; Nicoul, M.; Woike, T. Phys. Chem. Chem. Phys.
2010, 12, 9029.
(65) Galle, G.; Nicoul, M.; Woike, T.; Schaniel, D.; Freysz, E. Chem.
Phys. Lett. 2012, 552, 64.
(66) Cannizzo, A.; Milne, C. J.; Consani, C.; Gawelda, W.; Bressler,
C.; van Mourik, F.; Chergui, M. Coord. Chem. Rev. 2010, 254, 2677.
1863
dx.doi.org/10.1021/ja409262r | J. Am. Chem. Soc. 2014, 136, 1856−1863