Charge transfer and intraligand excited state interactions in platinum-sensitized dithienylethenes

Article abstract of DOI:10.1021/ic200200j

The photophysical behavior for two photochromic Pt-terpyridine acetylide complexes containing pendant dithienylethenes (DTEs) bound to the metal through the alkynyl linkage is presented. Selective excitation of the Pt complex with visible light resulted i

Full text of DOI:10.1021/ic200200j

ARTICLE
pubs.acs.org/IC
Charge Transfer and Intraligand Excited State Interactions in
Platinum-Sensitized Dithienylethenes
Matthew N. Roberts,† Jeffrey K. Nagle,‡ Marek B. Majewski,† Jeremy G. Finden,§ Neil R. Branda,*,§ and
Michael O. Wolf *,†
†Department of Chemistry, University of British Columbia, Vancouver, British Columbia V6T 1Z1, Canada
‡Department of Chemistry, Bowdoin College, Brunswick, Maine 04011, United States
§Department of Chemistry, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada
S Supporting Information
b
ABSTRACT: The photophysical behavior for two photochro-
mic Pt-terpyridine acetylide complexes containing pendant
dithienylethenes (DTEs) bound to the metal through the
alkynyl linkage is presented. Selective excitation of the Pt
complex with visible light resulted in the metal-sensitized ring
closing of the DTE unit. The central purpose of this study was to
understand how excited state interactions govern the photo-
physics by correlating differences in the linkage of the two
components with differences in the intramolecular energy transfer processes that occur between the Pt complex and the DTE.
A series of model complexes without photochromic ligands were prepared and studied to elucidate the contributions of the triplet
metal-to-ligand charge transfer and triplet intraligand states. It is demonstrated that reducing the orbital overlap of the metal-based
and intraligand states by lengthening the linkage and eliminating a conjugated pathway is effective at dramatically decreasing the
efficiency of intramolecular energy transfer. This is evidenced by the appearance of Pt-terpyridine based phosphorescence and a
significant decrease in the observed rate of metal-sensitized ring closing of the DTE.
’ INTRODUCTION
these complexes. These absorptions occur in the visible region and
result in long-lived excited states that often decay radiatively.
Notably, it has been demonstrated that either functionalizing the
pyridyl ligand with electron-donating substituents or coordinating
aromatic alkynes to the Pt center results in significant increases in
excited state lifetimes.17ꢀ19 The enhanced excited state lifetime
makes them useful as inorganic photosensitizers20ꢀ23 because
sufficiently long-lived charge-separated states can drive a variety of
intermolecular electron transfer reactions.24
When choosing the organic photochromic component, it is
difficult not to focus on the dithienylethene (DTE) architecture.
This particular class of photoswitch has garnered much attention
because it undergoes reversible ring-closing reactions between
two isomers, each having its own opto-electronic properties,
when exposed to UV and visible light (Scheme 1).25,26 The
molecular backbone is also relatively easy to modify and functio-
nalize, and the reactions tend not to occur in the dark. In addition
to pursuing systems in which the photoswitch modulates proper-
ties such as absorption, emission, or redox potential, we are
interested in desiging systems in which photoswitching can
control chemical reactivity and function.27ꢀ35
Photoresponsive materials that integrate photochromic or-
ganic compounds with metal coordination complexes are sig-
nificant because the combination of the electronic and optical
properties of each component gives rise to characteristics not
found in either the organic or inorganic species alone.1ꢀ3 The
ground and excited state interactions between the components
govern the overall electro-optical properties of the hybrid
system; therefore, the choice of each building block is an essential
consideration when designing this type of material. Successful
modulation of a metal complex’s properties by toggling the
photoswitch between its two metastable states has been
demonstrated,4ꢀ9 while others have focused on the ability of a
metal complex to affect the switching behavior of the photo-
chromic unit.10ꢀ14 In both cases, the identity of the metal and the
linkage between the two components are primary factors in
determining the types of interactions that exist.
Platinum polypyridyl complexes are attractive choices to act as
the metal component. Their rich photophysical behavior15 has
been extensively studied for applications in biological sensing and
photocatalysis, and in color-based sensors. The photobehavior of
complexes belonging to this versatile family can be tuned by
tailoring the polypyridyl ligands or by varying the ancillary ligands
in the square-planar Pt(II) coordination sphere.16 Of particular
interest are the metal-to-ligand and ligand-to-ligand charge trans-
fer (MLCT and LLCT) transitions that are commonly observed in
Covalently attaching the DTE backbone to Pt terpyridyl
complexes is an appealing route to regulate their function as
Received: January 28, 2011
Published: May 06, 2011
r
2011 American Chemical Society
4956
dx.doi.org/10.1021/ic200200j Inorg. Chem. 2011, 50, 4956–4966
|

Inorganic Chemistry
ARTICLE
Scheme 1
’ EXPERIMENTAL SECTION
General Procedures. All solvents and reagents including those for
NMR analysis (Cambridge Isotope Laboratories) were obtained from
commercial sources and used as received except where noted. 1H NMR
spectra were recorded on a Bruker AV400-Direct (400 MHz) or Bruker
AV400-Indirect (400 MHz) spectrometer. All chemical shifts are
referenced to residual solvent signals, which were previously referenced
to tetramethylsilane and splitting patterns are designated as s (singlet),
d (doublet), t (triplet), m (multiplet), and br (broad). IR spectra were
obtained on a Thermo Nicolet 6700 FT-IR spectrometer. ESI-MS
(Bruker Esquire), EI-HRMS (Concept IIHQ), and elemental analysis
were completed at the UBC Microanalysis facility.
General Procedure to Prepare Pt Complexes. A Schlenk flask
was charged with [Pt(40-tolyl-trpy)Cl]Cl, the appropriate TMS-pro-
tected acetylene, and KOH/CH3OH. After sparging the reaction
mixture with N2 for 30 min, CuI (∼10 mol %) was added, and the
reaction was left to stir under N2 overnight.
photosensitizers because the lifetime of the charge-separated
excited state should depend on which isomeric form of the DTE
is present. Previously, the potential of this approach was dem-
onstrated in a bimolecular system with [Ru(bpy)3]2þ as the
photosensitizer.36 Covalent attachment of the photoswitch to the
photosensitizer is advantageous because it strengthens the interac-
tion between the two components. In a preliminary communica-
tion, we reported the metal-sensitized photocyclization of
a DTE ligand attached to a Pt terpyridyl complex through a
conjugated alkyne linker.37 This cyclization occurs because a
photoactive intraligand triplet (3IL) state localized on the DTE is
populated by energy transfer from triplet charge transfer (3CT)
states delocalized over the complex. By indirectly accessing the 3IL
state, photoswitching can occur using light of significantly lower
energy than required to directly populate the singlet excited state of
the DTE. In this case of sensitized photoreactivity, the involvement
of the Pt with a large spinꢀorbit coupling is necessary to populate
the excited-state triplet manifold. This type of photobehavior is
consistent with other DTE-containing Re,38,39 Ru,40,41 and Pt42
complexes that also efficiently produce triplet excited states.
Despite the appeal of being able to trigger the photoreac-
tions of DTEs in an indirect manner, this type of sensitized
photocyclization is undesirable for applications involving
utilization of the metal-based excited states, since these
require independent excitation of the metal complex and
photoswitch. One strategy to eliminate metal-sensitized
photocyclization involves modifying the alkynyl linkage to
electronically isolate the Pt(trpy) and DTE chromophores by
introducing a nonconjugated bridge. In this article, the
excited-state interactions of ligand-localized and charge-
transfer states are discussed for two hybrid photoswitching
systems (1 and 2), and are compared to those for model
components 3ꢀ7. Variable temperature absorption and lu-
minescence spectroscopy, bimolecular quenching experi-
ments, transient absorption, and DFT calculations are used
to probe how the molecular architecture impacts the excited-
state interactions.
Detailed synthetic procedures and characterization for complexes 1ꢀ4
and the ethynyl DTE precursors have been reported previously.37,43
Propargyl 4-(2-Thienyl)phenyl Ether. 4-(2-Thienyl)phenol
(0.138 g, 0.789 mmol), potassium bicarbonate (0.16 g, 0.011 mol),
and propargyl bromide (0.15 mL, 1.0 mmol) were dissolved in
anhydrous dimethylformamide (DMF). The reaction mixture was
stirred at room temperature for 20 h, at which time the product was
extracted into diethyl ether and subsequently washed with H2O and
brine. The organic layer was isolated, dried with MgSO4, filtered, and the
solvent removed in vacuo. Purification by column chromatography
(SiO2, hexanes/ethyl acetate 19:1) afforded 0.147 g (87%) of the title
1
compound as a white powder. H NMR (400 MHz, CDCl3) δ 2.54
(t, J = 2.0 Hz, 1H), 4.73 (d, J = 2.2 Hz, 2H), 7.00 (d, J = 8.5 Hz, 2H), 7.06
(m, 2H), 7.22 (m, 2H), 7.56 (d, J = 8.7 Hz, 2H).
[(40-Tolyl-trpy)Pt(CtCꢀCH2ꢀOꢀC6H4ꢀC4H3S)]PF6 (5).
Propargyl 4-(2-thienyl)phenyl ether (0.056 g, 0.263 mmol) and KOH
(0.044 g, 0.79 mmol) were added to 50 mL of N2-sparged CH3OH. The
reaction was stirred at room temperature for 30 min after which it was
treated with [Pt(40-tolyl-trpy)Cl]Cl (0.135 g, 0.24 mmol) and CuI
(0.007 g, 0.036 mmol). The reaction mixture was stirred under a N2
atmosphere for 48 h at room temperature. The CH3OH was removed in
vacuo leaving an orange precipitate, which was dissolved in a minimal
amount of DMF and added dropwise to an aqueous solution of NH4PF6
(0.300 g in 150 mL H2O). After stirring for 1 h, the resulting precipitate
was filtered off and washed with H2O, CH3OH, and diethyl ether. The
isolated orange solid was added to a solution of EtOH:CH3CN (20:1)
that was heated to reflux. Any undissolved solid was filtered off yielding
0.100 g (60%) of the product, which was recrystallized by slow diffusion
of diethyl ether into a CH3CN solution. 1H NMR (400 MHz, acetone-
d6) δ 2.46 (s, 3H), 5.08 (s, 2H), 7.11 (dd, J = 5.1 Hz, 1H), 7.20
(d, J = 8.7 Hz, 2H), 7.35 (d, J = 2.8 Hz, 1H), 7.41 (d, J = 5.2 Hz, 1H), 7.49
4957
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
(d, J = 8.1 Hz, 2H), 7.66 (d, J = 8.7 Hz, 2H), 7.73 (t, J = 6.5 Hz, 2H), 8.04
(d, J = 8.1 Hz, 2H), 8.49 (t, J = 7.9 Hz, 2H), 8.77 (d, J = 8.3 Hz, 2H), 8.91
(s, 2H), 8.94 (d, J = 5.9 Hz, 2H). ESI-MS m/z 731.3 [MþꢀPF6]. FT-IR
(solid, cmꢀ1): 2135 (s, νCtC). Anal. Calcd for C35H26F6N3OPSPt: C,
47.95; H, 2.97; N, 4.79. Found: C, 47.58; H, 3.10; N, 4.44.
Photophysical Measurements. For UV light irradiations, either
an unfiltered 302 nm Hg lamp (18 mW/cm2) or filtered TLC hand-lamp
(UVP Model:UVGL-58) was used, unless otherwise noted. For broad-
band visible light irradiation, a hand-held lamp with a tungsten bulb
(65 mW/cm2) fitted with the appropriate low-pass filter was used. To
determine the active wavelengths for photocyclization, a 75-W arc lamp
with a double monochromator was used to irradiate the sample.
Samples for all spectroscopic measurements were prepared using
HPLC-grade Fisher solvents, which were used as received. Absorption
spectra were obtained on a Varian Cary 5000 spectrometer. Fluores-
cence spectra were collected on a Photon Technology International
fluorimeter using a 75-W arc lamp as the source. Emission quantum
yields were measured using an integrating sphere. Time-resolved emis-
sion data were collected using a Horiba Yvon Fluorocube TCSPC
apparatus. A 470 nm NanoLED source pulsing at a repetition rate of
50 kHz was used for excitation. Broadband emission was monitored by a
CCD detector at λ > 500 nm, using a low pass filter. Data were fit using
the DAS6 Data Analysis software package. The reported lifetimes
are averages of 3 measurements. All samples for luminescence studies
were prepared in CH2Cl2 that had been previously dried by passing
through an alumina column and deaerated by no less than three
freezeꢀpumpꢀthaw cycles. Sample solutions were maintained under
N2 atmosphere for the duration of the experiment in 1 cm2 anaerobic
quartz cells (NSG PCI cells) fitted with a PTFE valve.
Figure 1. UVꢀvis absorption spectra of CH3CN solutions of
(a) complexes 1o and 3, and (b) complexes 2o, 4, and 5.
density functional theory (TDDFT)50,51 calculations of electronic
transitions with the SAOP model potential.
’ RESULTS AND DISCUSSION
Ground State Electronic Absorption Properties. The ab-
sorption spectra of the ring-open form of each of the photo-
chromic DTE-Pt complexes (1o and 2o) and the comparative
spectra of the respective model complexes that do not contain
the DTE components (3, 4 and 5) are shown in Figure 1. The
absorption spectra of Pt terpyridyl complexes typically originate
from two sets of transitions. The higher energy πfπ* transitions
involve ligand-localized orbitals of the terpyridine and acetylide
ligands. The lower energy transitions arise from charge transfer
transitions from orbitals that have significant mixing of metal and
acetylide ligand character to low-lying π* orbitals localized on the
terpyridine.16,19 Conjugation of aromatic groups directly to the
acetylide group increases the participation of the alkynyl
ligand in the CT transitions. Strongly electron-donating
ligands give rise to CT transitions that are more ligand-to-
ligand in nature and occur at lower energies than in cases
where the highest occupied molecular orbital (HOMO)
primarily involves the metal atom. This accounts for the lower
energy CT absorptions in 1o and 3 compared to 2o, 4, and 5.
Throughout this article, we generally refer to those transitions
that have both MLCT and LLCT character as CT transitions.
A comparison of the spectroscopic properties of the Pt com-
plexes is summarized in Table 1.
In the UV region of the spectrum, both photochromic com-
plexes have higher molar absorptivities because of the presence of
DTE-localized thiophene-based πfπ* transitions which are
absent in the model compounds. The nearly identical molar
absorptivities of the lowest energy bands, which are assigned to
charge transfer transitions in the DTE and model complexes, are
indicative that these transitions do not significantly involve the
whole photoswitch component. A comparison of the absorption
spectra of hybrid system 1o to its model 3 demonstrates that the
lowest energy CT transitions of the complexes are not perturbed
by the presence of the DTE component. The CT bands for
the ether-linked complexes 2o, 4, and 5 appear between
∼380ꢀ450 nm, which are blue-shifted significantly from the
corresponding bands in 1o and 3. The lowest unoccupied
molecular orbital (LUMO) energies are expected to be similar
for all the complexes, so these trends are attributed to differences
in the HOMO energies. Notably, the CT bands in complexes 4
and 5 are similar suggesting that the HOMOs are the same in
these complexes, and are localized on the metal acetylide moiety
with little influence of the aromatic groups located on the distal
Quantum yields of the ring-closing reactions (φrc) were obtained
using a ferrioxalate actinometer to determine photon flux. Irradiations
were carried out on samples dissolved in CH3CN using a 75-W arc lamp
as the source, with a double monochromator to isolate the desired
wavelength (425 or 302 nm). Full details of the calculations are given in
the Supporting Information.
Transient absorption experiments were performed using a Nd:YAG
laser (EKSPLA PL2241) that generated pulses of 35 ps (fwhm) in
duration. The third harmonic output (355 nm) was employed as the
pump beam. The fundamental, 1064-nm output pumped a Xe-filled
quartz cell to generate the white light continuum used as the probe
beam. The pump and probe beams were aligned to pass through the
sample at a 90° angle and focused so that the probe beam was completely
enclosed by the pump beam within the cuvette. The probe beam was
detected by first passing through a monochromator (Princeton Instru-
ments SpectraPro 2300i) equipped with a 150 g/mm diffraction grating.
The grating was centered at 470 nm, 540 nm, or 640 nm to collect data
over the 350ꢀ800 nm spectral region. This was coupled to a streak
camera (Hamamatsu C7700) and a CCD detector (Hamamatsu
C8484), which digitized the images from the streakscope. The trigger
delay between the streak camera and the pulse firing was controlled
using a passive delay unit (Hamamatsu C1097ꢀ01). Computer control
of data acquisition was via HPD-TA software (ver. 8.3) from Hama-
matsu. The data consists of a 200-shot average. Samples for transient
absorption were prepared in HPLC grade CH3CN (Fisher) that had
been previously deaerated by three freezeꢀpumpꢀthaw cycles and were
maintained under a N2 atmosphere for the duration of the experiment.
Steady-state UVꢀvis absorption spectra were collected before and after
laser photolysis to confirm sample degradation was not significant
during the course of the experiment.
TDDFT and DFT Calculations. ADF 2007.0144,45 all-electron
calculations were performed with TZ2P basis sets, scalar relativistic
effects included through the ZORA46ꢀ48 formalism and solvation effects
(CH2Cl2) through the COSMO49 formalism. Geometry optimizations were
carried out with the Becke-Perdew GGA potential and time-dependent
4958
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
Table 1. Spectroscopic Data for Compounds 1ꢀ5
compound
c
a
abs λmax (nm)a
Em λmax (nm)
τem (μs)c
φem
φrc
1o
1c
2o
2c
3
287, 383, 404, 465
568, 608b
0.14 ( 0.02,d 0.41 ( 0.07e
0.03 ( 0.004d
266, 286, 310, 328, 382, 404 (sh), 602
286, 336 (sh), 396 (sh), 417
268, 286, 314, 336, 380, 416, 592
288, 384, 409, 466
4
286, 336, 396 (sh), 418
548, 582 (sh)c
558c
2.92 ( 0.006
3.50 ( 0.012
0.076 ( 0.003
0.048 ( 0.003
5
286, 336, 396 (sh), 418
a At room temperature in CH3CN solution. b At 90K in EtOH:MeOH (4:1) glass. c At room temperature in CH2Cl2 solution. λex = 425 nm.
d
e λex = 302 nm.
Figure 2. Changes in the absorption spectra of CH3CN solutions of (a) 6o, (b) 1o, (c) 7o, and (d) 2o upon irradiation with 302 nm light.
transitions by comparison to the absorption spectra of 6o and 7o
(Figure 2).
Excited-State Interactions Involved in Metal-Sensitized
Ring-Closure. In our recent report describing the metal-sensi-
tized photocyclization of DTE photoswitches, we highlighted
that visible light irradiation of the Pt sensitizer induces the ring-
closing reaction, and that the two components (the metal
complex and the DTE) must be covalently linked through a
conjugated pathway for this to occur.37 Here we present a
detailed understanding of the excited-state interactions between
the two components that lead to the visible light-sensitized ring-
closing. The transient absorption (TA) difference spectra of
complexes 1o, 1c, and 3 measured at various delay times after
excitation at 355 nm are shown in Figure 3. The TA spectrum of
the model complex 3 shows a bleach in the region between
450ꢀ500 nm between strong positive absorption bands at higher
and lower energies (Figure 3a). This bleach is coincident with the
ground state absorption bands assigned to the CT transitions.
The broad, positive absorption at lower energy clearly consists of
at least two overlapping bands and is similar to the TA spectrum
Figure 3. Transient absorption difference spectra for CH3CN solutions
of (a) 3 and (b) 1o (black) and 1c (red) measured in at 800 ps following
a 355 nm laser pulse.
of the phenyl acetylide complex.52 On the basis of excited-state
spectra of other Pt terpyridyl complexes,53 we assign the absorp-
tion bands centered at 540 and 690 nm to transitions in the
transiently formed terpyridyl anion and the band at 400 nm to
end of the nonconjugated linkage. This is also consistent with the
transitions in the transient cation localized on the metal acetylide.
The transient signal of complex 3 decays simultaneously over the
wavelength range with a lifetime τ = 8 ns (Supporting Informa-
tion, Figure S1).
Excitation of the hybrid system 1o with 355 nm light results in
photocyclization of the DTE photoswitch either by direct
excitation or through energy transfer from the metal-based CT
states; however, these two processes occur on significantly
different time scales. Whereas direct excitation of the DTE is
expected to result in photocyclization within picoseconds,54
changes observed in the absorption spectra upon ring-closing of
the appended DTE components in hybrid systems 1o and 2o. In
1o, where the DTE is attached through a conjugated linker, the
CT bands red-shift upon conversion of the DTE to the ring-
closed form (1c) supporting the presence of ground-state
interaction of the metal complex and the photoswitch
(Figure 2b). This behavior contrasts with that of complex 2,
where no shift of the CT bands is observed when 2o is converted
to 2c (Figure 2d). The increase in absorption from 300ꢀ400 nm
upon photocyclization of 1o and 2o is assigned to localized DTE
4959
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
Figure 5. UVꢀvis absorption spectra (a) and changes in emission
spectra (b) when an EtOH:CH3OH (4:1) glass of hybrid complex 1o is
warmed from 85 to 125 K. Excitation = 470 nm.
Figure 4. (a) Changes in absorbance at λ = 600 nm of CH3CN
solutions of 1o (1.4 ꢁ 10ꢀ5 M) without quencher (O) or with varying
concentrations of DABCO, 1.0 ꢁ 10ꢀ3 M ((), 1.0 ꢁ 10ꢀ2 M (0), and
0.10 M (b), upon irradiation with light of wavelengths greater than
420 nm. (b) Changes in absorbance at λ = 600 nm when an EtOH:
CH3OH solution of 1o is irradiated with light of wavelengths greater
than 420 nm at 150 K (O), 170 K ((), 200 K (0), and 295 K (b).
Formation of the ring-closed isomer (1c) was evident by the
increased absorbance at 600 nm, where only 1c absorbs. In the
ring-open isomer (1o) the charge-separated state that is formed
initially after excitation transfers its energy to the DTE compo-
nent, so efficient quenching of the charge-separated state with
DABCO is expected to prevent sensitized photocyclization.
At sufficiently high concentrations of DABCO, diffusion-
controlled quenching of the excited state of 1o occurs when
irradiating with visible light, as evidenced by suppression of the
cyclization of 1o to produce 1c. In this case, the rate of inter-
molecular reductive quenching is faster than the rate of intra-
molecular energy transfer to the DTE component. At a 0.10 M
DABCO concentration, quenching almost completely sup-
presses photocyclization of the photoswitch. Significantly, irra-
diation of this solution with UV light results in direct excitation of
the DTE component and 1o photocyclizes at a comparable rate
to that when quencher is not present. This observation demon-
strates the difference in the rate of photoswitching with respect to
the excited state pathways following excitation with visible light
or UV light. The metal-sensitized ring closure, initiated by visible
light excitation, is a slow excited-state process, occurring on the
nanosecond time scale, and is susceptible to intermolecular
quenching. Irradiation with UV light excites the DTE directly,
resulting in fast (picoseconds) photocyclization of the DTE core.
The rate of energy transfer was also reduced by cooling the
system to lower temperature. An EtOH:CH3OH (4:1) solution
of 1o was frozen to a glass at 85 K. Photocyclization of 1o was
monitored by measuring the changes in absorbance at 600 nm
corresponding to the appearance of ring-closed 1c. At 85 K,
irradiation with light of wavelengths greater than 420 nm does
not result in an appreciable amount of 1c being formed after
20 min of irradiation (Figure 4b). Warming the glass through its
melting point (130ꢀ150 K), results in the onset of cyclization of
1o and the appearance of 1c. Raising the temperature further
resulted in an increase in the rate at which photocyclization
occurred. It is important to note that 1o is photoactive at
temperatures down to 90 K when irradiated with UV light. This
confirms that the DTE is still photoactive in the frozen matrix,
eliminating the restriction of rotation as an explanation for the
lack of visible light-induced photocyclization below 130 K.
The observed temperature dependence of photoactivity with
UV or visible light irradiation further supports the conclusion
that there are at least two separate excited state pathways that
lead to photocyclization of 1o. Irradiation with UV light results in
direct population of the photoactive excited singlet state of the
DTE component. In solution, the photoswitch in 1o adopts two
conformations. When in the antiparallel conformation,55,56 in
energy transfer processes occurring within the triplet manifold
are more likely to occur on a time scale on the order of tens of
nanoseconds.39 In this experiment, a fresh sample was continu-
ously passed through a flow cell; however, accumulation of the
ring-closed form of the system (1c) in the beam path still
occurred. The bleach centered at 600 nm is indicative of some
accumulation of 1c, based on comparison to the TA spectrum of
1c, produced directly by 355 nm excitation. There are, however,
some notable excited state spectral differences between the
hybrid system 1o and the model compound 3. The TA spectrum
of 1o does not exhibit a bleach overlapping with the ground state
CT absorption, and the positive absorption at ∼430 nm is
considerably red-shifted compared to its counterpart in 3. This
3
spectrum is assigned to the IL state of the DTE, which must
form faster than the time resolution of the instrument since this
was the only signal observed.
The TA spectrum observed after excitation of the ring-closed
isomer 1c is notably different than that observed for its ring-open
counterpart (1o) or model compound 3 (Figure 3b). An intense
bleach appears between 560ꢀ700 nm, the spectral region that
coincides with the πfπ* absorption of the ring-closed DTE.
Cyclization of the photoswitch results in the HOMO becoming
predominantly localized on the DTE component and conse-
quently the lowest energy absorptions are ascribed to LLCT
transitions. The positive absorption appearing in the TA spec-
trum of ring-closed 1c is slightly red-shifted compared to 1o. It is
unclear whether this feature is more appropriately assigned to
the 3LLCT state or the 3IL state of the photocyclized DTE. The
transient signal of complex 1c decays simultaneously over the
wavelength range with a lifetime τ = 10 ns (Supporting Informa-
tion, Figure S2).
Recently, Castellano and co-workers demonstrated that 1,4-
diazabicyclo-[2.2.2]octane (DABCO) efficiently reduces the
transient hole localized on the Pt atom and aryl acetylide ligand
in the charge-separated excited state of several Pt terpyridyl
complexes.53 DABCO is a suitable reductive quencher since it
does not absorb in the spectral region where the Pt complex is
excited. We used this bimolecular quenching reaction as a means
to provide further evidence for the existence of a charge-
separated state involved in the photosensitized ring closure of
1o. The rate of photocyclization (1of1c) by irradiation with
light of wavelengths greater than 420 nm was monitored over
time with various concentrations of DABCO (Figure 4a).
4960
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
Figure 6. Proposed Jablonski diagram rationalizing the low tempera-
ture emission of hybrid complex 1o. The shaded box represents limits on
the energy of the 3CT state.
Figure 7. Percent conversion to the photostationary states as a function
of time for CH3CN solutions of 1o and 2o when irradiated at
wavelengths between 415 and 425 nm. Linear fits are shown.
which the orientation of the thiophene rings results in a C2
symmetric conformation with the reactive carbons of each
thiophene in close proximity to each other, ring-closing occurs.
Upon freezing, a small amount of reorganizational energy is
required for the conversion of 1o to 1c compared to if the
photoswitch was frozen in the photoinactive parallel conforma-
tion (the orientation of the thiophene rings results in a mirror
plane) and required interconversion to the photoactive antipar-
allel conformation. This fact accounts for its photoactivity at low
temperatures when irradiated with UV light. On the other hand,
photocyclization of 1o with visible light, via intramolecular
energy transfer, occurs by a pathway with additional energy
barriers and competing deactivation pathways (such as phos-
energies than those at room temperature (Figure 5a). The
appearance of these lower energy transitions suggests that CT
states might lie at lower energies at low temperature. A change in
the relative energies of the 3CT and the 3IL states would affect the
rates of energy transfer between them, resulting in radiative decay
3
from the CT state as the primary pathway of excited-state
deactivation rather than energy transfer to the ligand (Figure 6).
The most likely scenario is that contributions from both the
adjustments in the relative energies of the excited states and the
contributions of thermal energy play a role in the photophysical
behavior at low temperature.
3
3
phorescence). Energy transfer from the CT to the IL state
requires sufficient thermal energy to overcome the barrier
between states. DFT calculations estimate that in the ring-open
complex (1o), the DTE 3πf3π* orbital energy difference
(localized on the remote acetylene-thiophene moiety) that leads
Excited-State Interactions for the Ether-Linked System.
Hybrid complex 2o was designed with the intent to limit the
3
3
interactions of the CT and IL states. In addition to conven-
tional ring-closing of the DTE component in 2o using UV light,
cyclization is also triggered with selective irradiation into the
MLCT band in this complex. Although this photobehavior is
similar to that of complex 1o, it is clear that there is a significant
difference in the efficiency of photocyclization using visible light.
To compare the relative efficiencies of ring-closing when excited
with visible light, solutions of complexes 1o and 2o were
irradiated under the same conditions with light between wave-
lengths 415 and 425 nm.
3
to ring closing is only 0.02 eV higher than the lowest πf3π*
orbital energy difference localized on the adjacent acetylide
moiety. At room temperature sufficient thermal energy is present
3
to populate the IL state capable of photocyclization. Below
140 K, the cyclization reaction no longer occurs, likely because
there is no longer sufficient thermal energy to populate the 3IL
state that leads to ring closing.
An important result correlated to the suppression of photo-
activity at low temperature is that strong emission is observed
from complex 1o (Figure 5b) below 125 K. The spectral profile
of the orange emission is characterized by multiple peaks with
well-defined vibronic spacing of approximately 1200 cmꢀ1 at-
tributed to the ring “breathing” of the terpyridyl scaffold.
Comparison of the energy and vibronic spacing to other Pt
terpyridyl acetylide complexes strongly supports the conclusion
that the emission originates from a state localized on the Pt
terpyridine moiety.18,52 As the temperature is increased closer to
the melting range of the glass, the emission profile becomes less
structured and intense until it completely disappears at tempera-
tures above 125 K.
The progress of the photoreactions was monitored by UVꢀvis
spectroscopy. To ensure each sample absorbed the same number
of photons, the solutions were prepared with absorbances greater
than 2.5 at the irradiation wavelength so that the majority of light
was absorbed by both samples. Conversions of the DTE com-
ponents to their ring-closed isomers were kept below 10% for
both samples so that absorbance by the ring-closed isomers does
not interfere significantly with the cyclization photoreaction. The
amount converted to the ring-closed form was determined by
calculating the ratio of absorbance at either 602 nm (1o) or
592 nm (2o) relative to the absorbance at 286 nm after each time
interval of irradiation. The ratios at various time intervals were
then divided by the same ratio, representative of the total changes
in the samples, to give the percentages photocyclized of the total
amounts achievable at the photostationary states. The relative
rates of both photoreactions (1of1c and 2of2c) are illustrated
in Figure 7. Clearly, system 1o demonstrates more efficient
photocyclization using visible light and although ring-closing is
not completely quenched for 2o, it is evident the linkage has a
substantial effect. The difference in rate is also reflected in the
larger ring-closing quantum yield for compound 1o (Table 1)
than that for compound 2o. The fact that the quantum yield is
smaller for 1o when 425 nm light is used than when 302 nm light
The efficiency of energy transfer is governed by the energy gap
between donor and acceptor energy levels, and by the height of
the energy barrier that exists between them.57 The diminished
thermal energy available at low temperatures might not be
3
enough to overcome the energy barrier between the CT and
3IL states (Figure 6). Although the 3IL excited state may still lie at
an accessible energy level near the 3CT state, the energy barrier
forces deactivation from the 3CT state, which is characterized by
orange phosphorescence. Additionally, examination of the ab-
sorption spectrum at low temperature reveals transitions at lower
4961
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
Figure 8. Corrected emission (dashed line) and excitation (solid line)
spectra for CH2Cl2 solutions of complexes 4 (black) and 5 (red).
Emission spectra were collected by exciting the solutions at 425 nm.
Excitation spectra were collected by monitoring emission at 550 nm for
complex 4 and 560 nm for complex 5.
is used is a consequence of the presence of additional pathways
for decay of the 3CT state involved in the indirect ring-closing.
The lower efficiency of sensitized ring-closing of 2o compared
to 1o suggests that other pathways are involved in its excited-
state decay. The model complexes 4 and 5 generally exhibit the
same structured emission, where both the wavelength and band
3
shape of emission are characteristic of Pt terpyridyl MLCT-
based phosphorescence (Figure 8). Since there is virtually no
effect on the emission, regardless of how the alkynyl ligand is
terminated, it appears that the CT state for these complexes
involves transfer of an electron from a mixed Pt-alkynyl orbital to
a low-lying terpyridyl π* orbital. This interpretation is also
supported by the excitation spectra, which are identical to the
respective absorption spectra for the respective complexes.
Comparatively weak emission was detected for CH2Cl2 solutions
of complex 2o or 2c relative to the model compounds. Despite
no indication that 2 was impure in the characterization data
(NMR or elemental analysis), this emission was determined to at
least partly originate from trace [Pt(trpy)Cl]PF6 by comparison
of the excitation spectrum with the absorption spectrum of 2o.
Therefore, emission from neither complex 2o nor 2c was
conclusively established.
The excited-state lifetimes and fluorescence quantum yields
for complexes 4 and 5 are listed in Table 1. The emission decay
profiles for both complexes were fit to monoexponential models
(Supporting Information, Figure S3), supporting the conclusion
that only one excited state was responsible for the observed
emission in each case. Their excited-state lifetimes are both on
the order of a few microseconds, a result consistent with the
notion that the same emitting state is responsible for emission.
The complexes also exhibit similar emission efficiency. Complex
4 has a slightly higher quantum yield than 5.
Figure 9. Transient absorption difference spectra of CH3CN solutions
of (a) complexes 3 (black), 4 (red), and 5 (blue), (b) complexes 2o
(black) and 5 (blue), (c) complexes 1o (blue) and 2o (black), and
(d) complexes 1c (blue) and 2c (black). All data were collected at 800 ps
after excitation at 355 nm.
3
lowering of a IL state could funnel energy from the emissive
3MLCT state. If a 3IL state were populated, the observed decay for
the excited-state species would be dependent on the relative rates
of relaxation from either the 3MLCT or the 3IL states and the rates
of energy transfer between them. The differences in rates would
account for some fluctuation in the excited-state lifetime and
quantum efficiency. This would explain the appearance of emis-
sion fromthesame3MLCT state, withslightlyperturbed dynamics
affected by the substitution of the alkynyl ligand.
Given the slight differences in the lifetimes and quantum yields,
there are several possible interpretations of these data. Minor
differences might result because varying the substitution on the
3
alkynyl ligand could perturb the MLCT relaxation processes
differently. Alternatively, additional states may be present, depend-
ing on how the alkynyl ligand is substituted, that could be
accessibleor interactwith theemitting3MLCT state. Thearomatic
groups incorporate low-lying 3IL states that could be populated by
energy transfer from the 3MLCT state. For instance, this would be
expected comparing 4 to 5, where extension of the conjugated
backbone and addition of thiophene (which often has a low-lying
nonemissive triplet state) is consistent with the notion that
The transient absorptions of the excited state species of the
complexes were measured by picosecond pumpꢀprobe laser
spectroscopy to further elucidate the excited state dynamics
of the complexes. Upon excitation of complex 4 at 355 nm,
the resulting transient species has a similar spectral profile to
the directly linked thienyl model complex 3 (Figure 9a). The
difference spectrum of 4 is characterized by a positive absorption
centered at 365 nm, a bleach from 390ꢀ450 nm, and a broad,
4962
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
Scheme 2
absorption, a bleach does not appear as it does in the spectrum
of 4 because the concurrent positive absorption is more intense.
Despite the alteration of the spectrum attributed to the transient
cation, the appearance of the lower energy band corresponding
to the terpyridyl-localized anion remains unchanged compared
to 4. If direct excitation resulted in ligand-to-ligand charge
transfer, one would expect this difference to manifest itself in
the ground-state absorption spectra, for which 4 and 5 are
identical in terms of their CT absorptions. Therefore, the
formation of this state would likely be a two-step process
similar to the behavior of the Pt chromophores reported by
Eisenberg and designed to undergo photoinitiated electron
transfer cascades.52,58 Within the limitations of the picosecond
TA experimental setup, however, the excited-state species re-
sponsible for this transient spectrum shown in Figure 9a is the
only one observed.
The other explanation is that the signal arises from a 3IL state
localized on the phenyl-thiophene subunit, shown as “path 2” in
Scheme 2. DFT calculations estimate that with the addition of
3
the thiophene in complex 5, the IL state falls considerably
(≈ 1 eV) in energy relative to 4. The lowering of the 3IL state
3
could make it accessible from the MLCT state. These two
excited states could be in equilibrium, which would explain the
appearance of 3MLCT-based emission in the spectrum of 5. The
nearly identical ground state absorption spectra and similar
phosphorescence of complexes 4 and 5 support that the same
excited state is responsible for radiative decay in all of those
3
complexes. The appearance of the IL state in the transient
spectrum and emission from the 3MLCT state is consistent with
observations for other metal complexes with approximately
isoenergetic 3MLCT and 3IL states.59 The excited-state lifetime
is not, however, observed to be significantly longer as one might
expect if the two states were in equilibrium.17,60,61 It is possible
that the rate of decay from the 3IL state is comparable to that of
intense positive absorption throughout the visible region greater
than 450 nm. The appearance of a bleach coinciding with ground
state CT absorption, and its similarity to spectra observed for
other Pt terpyridyl complexes, provides strong supporting evi-
dence that this spectrum originates from the 3MLCT state. The
transient signal decays monoexponentially at all wavelengths
with τ = 6.3 ns (Supporting Information, Figure S4).
3
the MLCT state and therefore there is no enhancement of
3
excited-state lifetime. The ability to populate the IL state in
3
complex 5 would be supporting evidence that a low-lying IL
Despite having nearly identical ground state absorption spec-
tra and similar emission profiles, the TA difference spectra for
complexes 4 and 5 indicate different excited-state behavior. The
difference spectra measured at 800 ps after 355-nm excitation for
CH3CN solutions are shown in Figure 9a. For complex 5, an
intense, positive absorption centered at 440 nm appears in the
region coinciding with the bleach observed for complex 4. The
positive absorption that appears at lower energy for 5 is con-
sistent in band shape and energy with that for 4. Comparing the
absorbance decay absorption from 650ꢀ680 nm, complex 5
exhibits a shorter lifetime than 4, τ = 2.0 ns compared to τ = 6.3 ns,
respectively (Supporting Information, Figures S4 and S5).
The assignment of the transient signal for 5 is complicated by
the nature of a difference spectrum, for which the data may be a
combination of signals of different signs. As such, concurrent
overlapping positive and negative changes can mask the appear-
ance of one or the other based on their relative intensities. With
this in mind, there are two possible states to which the observed
signal could be attributed (Scheme 2). The spectrum could be
assigned as a charge-separated state (3CS) with the cation
localized over the phenyl and thienyl rings instead of the Pt-
state exists within the molecular linker and could act as a conduit
to a 3IL state responsible for Pt-sensitized photocyclization of the
DTE component.
To acquire a TA spectrum for the excited state species of 2o,
fresh sample solution was continuously pumped through a flow
cell to prevent buildup of the photoproduct 2c in the probe
beam. Although 355-nm excitation addresses both the CT
1
transition and the photoswitch directly, the photoactive π*
DTE state, which is only accessible by direct UV excitation, is
assumed to have completely decayed within tens of picoseconds,
an assumption supported by other work.54 Therefore, the
transient signal observed on the nanosecond time scale is
interpreted as being directly related to the excited-state dynamics
resulting from excitation of the CT transition.
The TA spectra for complexes 5 and 2o are similar (Figure 9b).
A positive absorption appears centered at 450 nm, red-shifted
15 nm from the similar absorption observed for 5. Fitting the
decay of this absorption to a monoexponential model indicates
similar lifetimes (τ = 2.4 ns for 2o and τ = 1.4 ns for 5) for the two
excited-state species (Supporting Information, Figures S5 and
S6). The very shallow bleach centered at 600 nm is attributed to
the accumulation of some 2c (Figure 9d) in the sample beam
despite being replenished continually in the flow cell. The bleach
appears to diminish the intensity of a coincident positive
absorption that would be consistent with the spectrum appearing
3
acetylide unit (the CT state). This would account for the
bathochromic shift and change in band shape of the absorption
band attributed to the cation relative to complex 4. Since this
band coincidentally overlaps with the ground state CT
4963
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
Figure 10. Contour plots of the LUMO, HOMO, and HOMO-1 for complexes 4 (left) and 5 (right).
for 5. Therefore, our interpretation is that this excited-state
absorption is attributed to the same state for both 5 and 2o.
Evidence that these spectra might be attributed to a 3IL state is
found by comparison to the TA signal observed for 1o
(Figure 9c). Although the positive absorption exhibited by 2o
is slightly red-shifted compared to that observed for 1o, the
similar band shape and spectral features that appear throughout
the visible region suggest that these features originate from the
same state. For hybrid system 1o, the intense bleach centered at
about 600 nm is indicative of accumulation of the ring-closed
isomer (1c). Assignment of this signal as a 3IL state is consistent
with the observed photoswitching behavior of 2o, which under-
goes ring-closing with lower energy light when coordinated to
the Pt complex than when not coordinated.
populated first, have a short lifetime, and subsequently lead to
population of the 3IL state.
DFT and TDDFT calculations were carried out to aid in
understanding the distribution of electron density and the
orbitals involved in the lowest energy CT transitions. Contour
plots of the LUMO, HOMO, and HOMO-1 are shown in
Figure 10.
Since the lowest energy transitions that appear in the absorp-
tion spectra of complexes 2o, 4, and 5 are identical, the orbitals
involved in those transitions must be similar. The contour plots
show the orbitals relevant to the lowest energy transition. Since
the majority of the electron density of the HOMO is located on
the aromatic rings, which are distorted out of the Pt terpyridyl
plane, the oscillator strength of the HOMO to LUMO transition
is diminished. The next lowest allowed transition from HOMO-1
to LUMO is more favorable and occurs more efficiently. The
similarity of the HOMO-1 and LUMO of complexes 4 and 5
supports the nearly identical CT absorption bands observed for
these complexes.
A comparison of the TA spectra for CH3CN solutions of 1c
and 2c is shown in Figure 9d. The essentially identical spectra
support that the same state is populated by excitation of either
complex. The only excited state in common between 1c and 2c
that should generate the same spectral signature is the
3
DTE-localized IL state. Additionally, the comparable excited-
Since the lowest 3CT energy of 2o is comparable to that of 1o,
3
state lifetimes (τ = 10 ns for 1c and τ = 8.8 ns for 2c) support the
assignment of these spectra as originating from the same excited
state (Supporting Information, Figures S2 and S7). The TA
there should be sufficient energy to populate the IL state if
energy transfer is possible. DFT calculations of the CT energies,
together with observed absorption spectra of 1o and 2o, indicate
CT energies to be g2.0 eV, in agreement with experimental and
DFT-calculated values for related compounds.62 The triplet
energy of the DTE photoswitch state that leads to ring-closing
is estimated from DFT calculations to be about 1.9 eV for 1o and
1.8 eV for 2o, values that are in good agreement with an
experimental value of 1.85 ( 0.16 eV for the same photoswitch
in a related Ru(II) complex.40
3
spectrum of the DTE-localized IL state is characterized by an
intense bleach coinciding with the ground state πfπ* absorp-
tion of the ring-closed DTE. This is accompanied by a positive
absorption appearing at higher energy. The traces shown in
Figure 9d represent the only spectral signature observed within
the time resolution of our experimental setup. Considering that
the ring-closed isomer of the DTE component exhibits appreci-
able absorption at 355 nm (Figures 2a and 2c), one possibility is
that direct excitation of the 1IL state is followed by rapid
For the model complexes 4 and 5, the trends of DFT-cal-
3
culated energies for their alkynyl-ligand-based IL states are
3
intersystem crossing to the IL state, a process facilitated by
consistent with the differences observed in their luminescence
behavior and TA spectra. These empirical differences have been
attributed to the changes in the relative energies of the 3CT and
the spinꢀorbit coupling of the appended Pt atom. Although not
3
observed, it cannot be discounted that the CT state could be
4964
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
behavior. A key result that has emerged is that the excited states
involved in the metal sensitized photocyclization of 1o (where
the linker is a short, conjugated group) and 2o (where the linker
is longer and nonconjugated), are similar. The differences in
photocyclization rate between the two complexes can be attrib-
uted to the poorer orbital overlap in 2o. The photophysical
behavior of the model complexes 4 and 5 provides interesting
guidance for the design of complexes in which pendant DTE
groups can be used to modulate the photophysical behavior of
metal complexes. This is not possible in 1o and 2o because
excitation of the metal results in energy transfer to the DTE and
subsequent photocyclization. The excited state in complex 4 is a
MLCT state, while addition of the thiophene in 5 results in
population of either a 3CS or a 3IL state (or a combination of the
two). The similarity in the TA spectra of 5 and 1o or 2o indicates
that the presence of the phenylthiophene group is sufficient to
cause the excited state to shift from the metal (3MLCT) to ligand
(phenylthiophene or DTE photoswitch). Either removing the
phenyl group or breaking the conjugation between phenyl and
thienyl groups could prevent energy transfer across the linker
from metal to the uncyclized DTE. For this strategy to succeed,
however, energy transfer to the photocyclized DTE would still
have to occur to allow modulation of excited state chemistry at
the metal.
Figure 11. Proposed Jablonski diagrams comparing the excited-state
interaction of the 3CT state and varying 3IL states for complexes (a) 4
and 5 or (b) 2o after excitation of the CT transition. The shaded box
represents limits on the energy of the 3IL state.
3IL states, which interact with each other if they are close enough
in energy. Whereas the energy of the 3CT state remains relatively
3
constant for the models, the energy of the IL state drops
considerably based on the functionality of the alkynyl ligand.
The estimated values of the 3IL state energy for complexes 4 and
5 are 3.2 and 2.5 eV, respectively. If the 3IL state is accessible from
3
the CT state, it might act as an alternate pathway for decay.
3
Otherwise, emission from the CT state is the predominant
decay pathway. This model is summarized in the Jablonski
diagram shown in Figure 11a.
Unlike in the directly linked system 1o, the nonconjugated,
ether linkage seems to largely remove interactions between the
two chromophores in the ground state. This is supported by the
lack of any significant shifts of absorption bands in 2o compared
to the model complexes of its constituents. In addition, there are
no shifts observed in the 1H NMR resonances of the DTE upon
coordination, as were observed for 1o. TDDFT-calculated
oscillator strengths for transitions between the Pt and the DTE
moiety in 2o compared to 1o (e.g., f < 0.001 for 2o vs f > 0.1
for 1o) are indicative of the diminished orbital overlap of the two
components when linked by the nonconjugated ether linkage.
’ ASSOCIATED CONTENT
S
Supporting Information. Fitted time decay profiles of
b
phosphorescence for 4 and 5, and fitted time decay profiles of the
TA signals for 1c, 2o, 2c, 3, 4, and 5. This material is available free
of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
3
Observation of the MLCT-based emission for the models is
Corresponding Author
*E-mail: nbranda@sfu.ca (N.R.B.), mwolf@chem.ubc.ca (M.O.W.).
consistent with the reduced interaction between the chromo-
3
phores, as opposed to efficient energy transfer to a IL state as
observed in 1o. The transient absorption data, however, show
that interactions between the two chromophores exist in the
excited state. DFT calculations support a model in which low-
’ ACKNOWLEDGMENT
This research was supported in part by the Natural Sciences
and Engineering Research Council of Canada, the Canada
Research Chairs Program, and Simon Fraser University. The
assistance of Dr. Saeid Kamal (Laboratory for Advanced Spec-
troscopy and Imaging Research) is acknowledged.
3
3
ering of the IL state permits its population from the MLCT
3
state. Population of a IL state also explains the ring-closing
observed for 2o when selectively exciting its MLCT absorption.
An energy diagram for 2o illustrating the proposed model is
shown in Figure 11b.
’ REFERENCES
(1) Kume, S.; Nishihara, H. Dalton Trans. 2008, 3260–3271.
(2) Guerchais, V.; Le Bozec, H. Top. Organomet. Chem. 2010, 28,
171–225.
(3) Ko, C.-C.; Yam, V. W.-W. J. Mater. Chem. 2010, 20, 2063–2070.
(4) Fernandez-Acebes, A.; Lehn, J.-M. Chem.—Eur. J. 1999, 5,
3285–3292.
(5) Norsten, T. B.; Branda, N. R. Adv. Mater. 2001, 13, 347–349.
(6) Lee, I.; You, Y.; Lim, S.-J.; Park, S. Y. Chem. Lett. 2007,
36, 888–889.
(7) Green, K. A.; Cifuentes, M. P.; Corkery, T. C.; Samoc, M.;
Humphrey, M. G. Angew. Chem., Int. Ed. 2009, 48, 7867–7870.
(8) Nakagawa, T.; Hasegawa, Y.; Kawai, T. Chem. Commun. 2009,
5630–5632.
’ CONCLUSIONS
Efficient metal sensitized photocyclization of the DTE in 1o
has been shown to occur via population of the 3IL state, as well as
direct excitation of the 1IL state with UV light. These pathways
have different energy barriers, as shown by the different tem-
perature dependences of the photocyclization rates. Addition of a
reductive quencher, DABCO, results in a slower rate of DTE
photocyclization with irradiation with light at wavelengths great-
er than 420 nm, showing that this pathway involves an initially
formed charge transfer site at the metal acetylide. With direct UV
excitation the rate of photocyclization is unaffected by the
addition of DABCO, supporting direct ring closing from the
1IL excited state in this case.
(9) Aubert, V.; Guerchais, V.; Ishow, E.; Hoang-Thi, K.; Ledoux, I.;
Nakatani, K.; Le Bozec, H. Angew. Chem., Int. Ed. 2008, 47, 577–580.
(10) Tanaka, Y.; Inagaki, A.; Akita, M. Chem. Commun. 2007,
1169–1171.
An important goal of this study was to examine the effect of the
linker between the metal and the DTE on the photophysical
4965
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966

Inorganic Chemistry
ARTICLE
(11) Guirado, G.; Coudret, C.; Launay, J. P. J. Phys. Chem. C 2007,
111, 2770–2776.
(46) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
1993, 99, 4597–4610.
(12) Lee, P. H. M.; Ko, C. C.; Zhu, N. Y.; Yam, V. W. W. J. Am. Chem.
Soc. 2007, 129, 6058–6059.
(47) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
1994, 101, 9783–9792.
(13) Motoyama, K.; Koike, T.; Akita, M. Chem. Commun. 2008,
5812–5814.
(48) van Lenthe, E.; Ehlers, A.; Baerends, E. J. J. Chem. Phys. 1999,
110, 8943–8953.
(14) Liu, Y. F.; Lagrost, C.; Costuas, K.; Tchouar, N.; Le Bozec, H.;
Rigaut, S. Chem. Commun. 2008, 6117–6119.
(49) Pye, C. C.; Ziegler, T. Theor. Chem. Acc. 1999, 101, 396–408.
(50) Rosa, A.; Baerends, E. J.; van Gisbergen, S. J. A.; van Lenthe, E.;
Groeneveld, J. A.; Snijders, J. G. J. Am. Chem. Soc. 1999, 121,
10356–10365.
(51) van Gisbergen, S. J. A.; Snijders, J. G.; Baerends, E. J. Comput.
Phys. Commun. 1999, 118, 119–138.
(52) Chakraborty, S.; Wadas, T. J.; Hester, H.; Flaschenreim, C.;
Schmehl, R.; Eisenberg, R. Inorg. Chem. 2005, 44, 6284–6293.
(53) Shikhova, E.; Danilov, E. O.; Kinayyigit, S.; Pomestchenko,
I. E.; Tregubov, A. D.; Camerel, F.; Retailleau, P.; Ziessel, R.; Castellano,
F. N. Inorg. Chem. 2007, 46, 3038–3048.
(54) Tamai, N.; Miyasaka, H. Chem. Rev. 2000, 100, 1875–1890.
(55) Uchida, K.; Nakayama, Y.; Irie, M. Bull. Chem. Soc. Jpn. 1990,
63, 1311–1315.
(15) Castellano, F. N.; Pomestchenko, I. E.; Shikhova, E.; Hua, F.;
Muro, M. L.; Rajapakse, N. Coord. Chem. Rev. 2006, 250, 1819–1828.
(16) Han, X.; Wu, L. Z.; Si, G.; Pan, J.; Yang, Q. Z.; Zhang, L. P.;
Tung, C. H. Chem.—Eur. J. 2007, 13, 1231–1239.
(17) Michalec, J. F.; Bejune, S. A.; McMillin, D. R. Inorg. Chem. 2000,
39, 2708–2709.
(18) Yang, Q. Z.; Wu, L. Z.; Wu, Z. X.; Zhang, L. P.; Tung, C. H.
Inorg. Chem. 2002, 41, 5653–5655.
(19) Guo, F. Q.; Sun, W. F.; Liu, Y.; Schanze, K. Inorg. Chem. 2005,
44, 4055–4065.
(20) Du, P. W.; Schneider, J.; Jarosz, P.; Eisenberg, R. J. Am. Chem.
Soc. 2006, 128, 7726–7727.
(21) Du, P. W.; Schneider, J.; Jarosz, P.; Zhang, J.; Brennessel,
W. W.; Eisenberg, R. J. Phys. Chem. B 2007, 111, 6887–6894.
(22) Feng, K.; Zhang, R. Y.; Wu, L. Z.; Tu, B.; Peng, M. L.; Zhang,
L. P.; Zhao, D.; Tung, C. H. J. Am. Chem. Soc. 2006, 128, 14685–14690.
(23) Wang, X.; Goeb, S.; Ji, Z.; Pogulaichenko, N. A.; Castellano,
F. N. Inorg. Chem. 2011, 50, 705–707.
(56) Irie, M.; Sakemura, K.; Okinaka, M.; Uchida, K. J. Org. Chem.
1995, 60, 8305–8309.
(57) Wang, X. Y.; Del Guerzo, A.; Schmehl, R. H. J. Photochem.
Photobiol., C 2004, 5, 55–77.
(58) McGarrah, J. E.; Eisenberg, R. Inorg. Chem. 2003,
42, 4355–4365.
(24) Yoon, T. P.; Ischay, M. A.; Du, J. N. Nat. Chem. 2010, 2,
527–532.
(59) Tyson, D. S.; Luman, C. R.; Zhou, X. L.; Castellano, F. N. Inorg.
Chem. 2001, 40, 4063–4071.
(25) Irie, M. Chem. Rev. 2000, 100, 1685–1716.
(26) Tian, H.; Wang, S. Chem. Commun. 2007, 781–792.
(27) Lemieux, V.; Branda, N. R. Org. Lett. 2005, 7, 2969–2972.
(28) Samachetty, H. D.; Branda, N. R. Chem. Commun. 2005,
2840–2842.
(60) Hissler, M.; Harriman, A.; Khatyr, A.; Ziessel, R. Chem.—Eur. J.
1999, 5, 3366–3381.
(61) Ji, S. M.; Wu, W. H.; Wu, W. T.; Song, P.; Han, K. L.; Wang,
Z. G.; Liu, S. S.; Guo, H. M.; Zhao, J. Z. J. Mater. Chem. 2010,
20, 1953–1963.
(29) Sud, D.; Norsten, T. B.; Branda, N. R. Angew. Chem., Int. Ed.
2005, 44, 2019–2021.
(62) Liu, X. J.; Feng, J. K.; Meng, J.; Pan, Q. J.; Ren, A. M.; Zhou, X.;
Zhang, H. X. Eur. J. Inorg. Chem. 2005, 1856–1866.
(30) Odo, Y.; Matsuda, K.; Irie, M. Chem.—Eur. J. 2006, 12,
4283–4288.
(31) Samachetty, H. D.; Branda, N. R. Pure Appl. Chem. 2006, 78,
2351–2359.
(32) Sud, D.; Wigglesworth, T. J.; Branda, N. R. Angew. Chem., Int.
Ed. 2007, 46, 8017–8019.
(33) Lemieux, V.; Spantulescu, M. D.; Baldridge, K. K.; Branda, N. R.
Angew. Chem., Int. Ed. 2008, 47, 5034–5037.
(34) Samachetty, H. D.; Lemieux, V.; Branda, N. R. Tetrahedron
2008, 64, 8292–8300.
(35) Vomasta, D.; Hogner, C.; Branda, N. R.; Konig, B. Angew.
Chem., Int. Ed. 2008, 47, 7644–7647.
(36) Kozlov, D. V.; Castellano, F. N. J. Phys. Chem. A 2004, 108,
10619–10622.
(37) Roberts, M. N.; Nagle, J. K.; Finden, J. G.; Branda, N. R.; Wolf,
M. O. Inorg. Chem. 2009, 48, 19–21.
(38) Yam, V. W.-W.; Ko, C.-C.; Zhu, N. J. Am. Chem. Soc. 2004, 126,
12734–12735.
(39) Ko, C.-C.; Kwok, W.-M.; Yam, V. W.-W.; Phillips, D. L. Chem.
—Eur. J. 2006, 12, 5840–5848.
(40) Jukes, R. T. F.; Adamo, V.; Hartl, F.; Belser, P.; De Cola, L.
Inorg. Chem. 2004, 43, 2779–2792.
(41) Indelli, M. T.; Carli, S.; Ghirotti, M.; Chiorboli, C.; Ravaglia,
M.; Garavelli, M.; Scandola, F. J. Am. Chem. Soc. 2008, 130, 7286–7299.
(42) Lee, J. K.-W.; Ko, C.-C.; Wong, K. M.-C.; Zhu, N.; Yam,
V. W.-W. Organometallics 2007, 26, 12–15.
(43) Finden, J.; Kunz, T. K.; Branda, N. R.; Wolf, M. O. Adv. Mater.
2008, 20, 1998–2002.
(44) Guerra, C. F.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor.
Chem. Acc. 1998, 99, 391–403.
(45) Velde, G. T.; Bickelhaupt, F. M.; Baerends, E. J.; Guerra, C. F.;
Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001,
22, 931–967.
4966
dx.doi.org/10.1021/ic200200j |Inorg. Chem. 2011, 50, 4956–4966