Excited-state structure and delocalization in ruthenium(II)-bipyridine complexes that contain phenyleneethynylene substituents

Article abstract of DOI:10.1021/jp013008j

A comprehensive photophysical study has been carried out on the two complexes [(bpy)₂Ru(4,4′-PE-bpy)]²⁺ and [(bpy)₂Ru(5,5′-PE-bpy)]²⁺ (44Ru and 55Ru, respectively, where bpy = 2,2′-bipyridine and PE = phenyleneethynylene). The objective of this work is to determine the effect of the phenyleneethynylene substituents on the properties of the metal-to-ligand charge-transfer excited state. The complexes have been characterized by using UV-visible absorption, photoluminescence, and UV-visible and infrared transient absorption spectroscopy. The results indicate that the MLCT excited state is localized on the PE-substimted bpy ligands. Moreover, the photophysical data indicate that in the MLCT excited state the excited electron is delocalized into the PE substituents and the manifestations of the electronic delocalization are larger when the substituents are in the 4,4′-positions.

Full text of DOI:10.1021/jp013008j

11118
J. Phys. Chem. A 2001, 105, 11118-11127
Excited-State Structure and Delocalization in Ruthenium(II)-Bipyridine Complexes That
Contain Phenyleneethynylene Substituents
Yingsheng Wang,† Shengxia Liu,† Mauricio R. Pinto,† Dana M. Dattelbaum,‡
Jon R. Schoonover,‡ and Kirk S. Schanze†,*
Department of Chemistry, UniVersity of Florida, P.O. Box 117200, GainesVille, Florida 32611-7200, and
Los Alamos National Laboratory, Materials Science and Technology DiVision, MS E549,
Los Alamos, New Mexico 87544
ReceiVed: August 3, 2001; In Final Form: October 4, 2001
A comprehensive photophysical study has been carried out on the two complexes [(bpy)2Ru(4,4′-PE-bpy)]2+
and [(bpy)2Ru(5,5′-PE-bpy)]2+ (44Ru and 55Ru, respectively, where bpy ) 2,2′-bipyridine and PE )
phenyleneethynylene). The objective of this work is to determine the effect of the phenyleneethynylene
substituents on the properties of the metal-to-ligand charge-transfer excited state. The complexes have been
characterized by using UV-visible absorption, photoluminescence, and UV-visible and infrared transient
absorption spectroscopy. The results indicate that the MLCT excited state is localized on the PE-substituted
bpy ligands. Moreover, the photophysical data indicate that in the MLCT excited state the excited electron
is delocalized into the PE substituents and the manifestations of the electronic delocalization are larger when
the substituents are in the 4,4′-positions.
The photophysical properties of metal-to-ligand charge
transfer (MLCT) excited states are strongly influenced by the
pattern of substitution on the organic ligand that acts as the
electron acceptor in the MLCT transition.1-8 As expected on
the basis of a Mulliken-type charge transfer formalism, the
energy of an MLCT absorption increases as the reduction
potential of the acceptor ligand shifts cathodically (i.e., as the
LUMO energy increases).9 Ligand substituent effects have been
used to delineate the energy gap law correlation, wherein the
nonradiative decay rate of the MLCT excited state increases as
the energy of the MLCT state decreases.1 Recently, it has also
been shown that increased delocalization in the acceptor ligand
decreases the nonradiative decay rate, presumably due to
decreased electron-vibrational coupling in the excited state.6,8
We have been exploring the excited-state properties in metal-
organic oligomers and polymers where the acceptor ligand for
an MLCT transition is an integral part of an extended π-con-
jugated electronic system.10-13 This line of investigation is being
pursued in order to provide a basis for understanding the excited-
state properties of more complex metal-organic π-conjugated
polymers that have been used for electrooptical device
applications.14-18 Most of our work has examined oligomeric
and polymeric phenyleneethynylenes that contain d6 rhenium(I),
ruthenium(II), and osmium(II) polypyridine complexes that are
strongly coupled to the π-conjugated backbone.10,11,13,19 Al-
though much has been learned from this work, in many cases
the structures of the oligomers and polymers are sufficiently
complicated such that it is difficult to extract quantitative
information concerning the effect of the π-conjugated system
on the properties of the MLCT excited state.
CHART 1
complexes, 44Ru and 55Ru (Chart 1). These complexes contain
the Ru(bpy)32+ chromophore (bpy ) 2,2′-bipyridine), but in each
case one of the bpy ligands is substituted with phenyleneethy-
nylene (PE) moieties that feature a carboxyamide substituent
in the 4-position on the phenylene ring. These complexes were
designed with several objectives in mind. First, we sought to
determine the effect that substitution of the PE substituents has
on the photophysics of the MLCT excited state in the complexes.
It was anticipated that the properties of the MLCT state would
be modified due to electron delocalization into the PE substit-
uents. Second, the carboxyamide groups were included in the
structures specifically as mid-IR chromophores to allow the
application of time-resolved infrared spectroscopy (TRIR) to
To address this problem, we have carried out a detailed
photophysical investigation of the relatively simple model
* To whom correspondence should be addressed. E-mail:
kschanze@chem.ufl.edu.
† University of Florida.
‡ Los Alamos National Laboratory.
10.1021/jp013008j CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/16/2001

Ru(II)-Bpy Complexes Containing Phenyleneethynylene
J. Phys. Chem. A, Vol. 105, No. 49, 2001 11119
probe for delocalization of the excited electron into the
phenyleneethynylene unit.20,21 The present manuscript provides
a report of the photophysical properties of the complexes 44Ru
and 55Ru. These complexes have been characterized by UV-
visible absorption, photoluminescence, polarized photolumi-
nescence as well as time-resolved UV-visible and infrared
spectroscopy. The results indicate that in both cases the PE-
substituted bpy is the chromophoric ligand. Although there is
clear evidence from absorption, emission, and emission polar-
ization spectroscopy for delocalization into the PE units, the
effect of delocalization on the excited-state decay parameters
is comparatively small.
added. The product precipitated as a brick-red powder. The
crude product was collected by filtration, washed with diethyl
ether, and then purified by column chromatography on alumina
eluting with mixtures of CH2Cl2 and CH3CN (gradient, CH2-
Cl2:CH3CN from 10:1 to 10:4 v:v). The collection was
monitored by UV-visible spectroscopy. The product-containing
fractions were combined and concentrated to dryness. The
chromatographed product was reprecipitated by dropping a
concentrated CH2Cl2 solution into excess diethyl ether and it
was collected by filtration. The final product was obtained as a
1
brick-red microcrystalline solid, yield 90 mg (50%). H NMR
(CD3CN, 300 MHz): δ 1.05 (br, 6H), 1.18 (br, 6H), 3.18 (br,
4H), 3.45 (br, 4H), 7.36 (d, 4H), 7.41 (q, 4H), 7.51 (d, 4H),
7.69 (d, 2H), 7.80 (d, 2H), 7.85 (d, 2H), 8.07 (q, 4H), 8.15 (dd,
2H), 8.49 (s, 2H), 8.50 (dd, 4H). HRMS (electrospray, FT-MS):
Calcd for C56H50O2N8F6PRu: 1113.275. Found: 1113.303. The
full HRMS along with calculated isotope distribution is available
as Supporting Information.
Experimental Section
General Synthesis. Reagent grade solvents and chemicals
were used for synthesis without purification unless otherwise
noted. THF was distilled from Na-benzophenone. Chromatog-
raphy was carried out using either silica gel (Merck, 230-400
mesh) or neutral Alumina (Fisher Chemical Company, 6% water
added, Brockman grade III). NMR spectra were obtained on a
Varian Gemini NMR spectrometer operating at 300 MHz.
Proton NMR spectra of 44L, 55L, 44Ru, and 55Ru and
electrospray mass spectra of 44Ru and 55Ru are available as
Supporting Information.
5,5′-Bis[(4-(N,N-diethylcarboxamido)phenyl)ethynyl]-2,2′-
bipyridinyl (55L). 5,5′-Diethynyl-2,2′-bipyridine22,23 (75 mg,
0.36 mmol), 4-iodo-N,N-diethylbenzamide24 (245 mg, 0.81
mmol), diethylamine (3 mL, 29 mmol), and THF (5 mL) were
mixed in a three-necked round-bottom flask. The solution was
purged with N2 for 10 min whereupon Pd(PPh3)2Cl2 (3.9 mg,
5.6 µmol) and CuI (1.9 mg, 9.9 µmol) were added. The solution
was stirred at room temperature under N2 for 14 h. During the
course of the reaction the product precipitated from the reaction
mixture as a yellow microcrystalline material. The product was
isolated by filtering the reaction mixture through a medium
porosity fritted funnel. The crude product was purified by
washing with ether, yield 150 mg (75%). 1H NMR (CDCl3, 300
MHz): δ 1.13 (br, 6H), 1.25 (br, 6H), 3.27 (br, 4H), 3.54 (br,
4H), 7.40 (d, 4H), 7.60 (d, 4H), 7.96 (d, 2H), 8.45 (d, 2H),
8.82 (s, 2H).
4,4′-Bis[(4-(N,N-diethylcarboxamido)phenyl)ethynyl]-2,2′-
bipyridinyl (44L). This compound was prepared according to
the procedure described for 55L, except that 4,4′-diethynyl-
2,2′-bipyridine25 was used in place of 5,5′-diethynyl-2,2′-
bipyridine. The reaction mixture was concentrated under reduced
pressure. The crude product was purified by column chroma-
tography on silica gel eluting with diethyl ether:methanol (100:1
v:v). The purified product was recovered as a white solid, yield
120 mg (60%). 1H NMR (CDCl3, 300 MHz): δ 1.14 (br, 6H),
1.26 (br, 6H), 3.28 (br, 4H), 3.56 (br, 4H), 7.40 (d, 4H), 7.43
(d, 2H), 7.60 (d, 4H), 8.55 (s, 2H), 8.70 (d, 2H).
55Ru. Ru(bpy)2Cl226 (70 mg, 0.135 mmol), AgCF3SO3 (138
mg, 0.54 mmol), and acetone (15 mL) were combined in a
round-bottom flask. The solution was refluxed under N2 for 2
h. TLC (silica, CH2Cl2:MeOH, 5:1 v:v) of the reaction mixture
shows a spot with Rf ) 0.15-0.25, which is believed to be the
product, Ru(bpy)2(CF3SO3)2. The reaction mixture was filtered
through a medium porosity fritted funnel and the filtrate was
concentrated to a volume of 3-4 mL. The concentrated Ru-
(bpy)2(CF3SO3)2/acetone solution was transferred to a clean
round-bottom flask, whereupon 55L (96 mg, 0.173 mmol) and
95% EtOH (14 mL) were added. The resulting solution was
refluxed under N2 for 17 h. The reaction mixture was cooled
and then an aqueous solution of NH4PF6 (450 mg, 10 mL) was
44Ru. This compound was prepared according to the
procedure described for 55Ru, except that 44L was used in place
of 55L. The crude product was purified by column chroma-
tography on alumina eluting with CH2Cl2:CH3CN (4:1 v:v). The
collection was monitored by UV-visible spectroscopy. The
product-containing fractions were combined and concentrated
to dryness. The chromatographed product was reprecipitated by
dropping a concentrated CH2Cl2 solution into excess diethyl
ether, and it was collected by filtration. The final product was
1
obtained as a brick-red microcrystalline solid, yield 50%. H
NMR (CD3CN, 300 MHz): δ 1.1 (br, 6H), 1.2 (br, 6H), 3.2
(br, 4H), 3.5 (br, 4H), 7.4 (m, 10H), 7.7 (m, 10H), 8.07 (m,
4H), 8.5 (d, 4H), 8.65 (s, 2H). HRMS (electrospray, FT-MS):
Calcd for C56H50O2N8F6PRu: 1113.275. Found: 1113.304. The
full HRMS along with calculated isotope distribution is available
as Supporting Information.
Spectroscopy and Electrochemistry. UV-visible absorption
spectra were obtained on a Varian Cary 100 spectrophotometer.
Corrected steady-state emission spectra were recorded on a
SPEX F-112 fluorescence spectrometer. Samples were contained
in 1 cm × 1 cm quartz cuvettes, and the optical density was
adjusted to approximately 0.1 at the excitation wavelength.
Emission quantum yields are reported relative to [Ru(bpy)32+]-
[Cl-]2 in water (Φem ) 0.055)27 and appropriate correction was
applied for the difference in refractive indices of the sample
and actinometer solvents.28 Low-temperature fluorescence ex-
periments were carried out by cooling the samples in an Oxford
Instruments DN-1704 optical cryostat. Samples for low-tem-
perature spectroscopy were dissolved in an ethanol/methanol
(4:1 v:v) solvent mixture. Time-resolved emission decays were
obtained by time-correlated single photon counting on an
instrument that was constructed in-house. Excitation was
effected by using a blue diode laser (IBH instruments, Edin-
burgh, Scotland, pulse width 800 ps). The time-resolved
emission was collected using a red-sensitive, photon-counting
PMT (Hamamatsu R928) and the light was filtered using 10
nm band-pass interference filters. Lifetimes were determined
from the observed decays with the DAS6 deconvolution
software (IBH instruments, Edinburgh, Scotland). Nanosecond
transient absorption spectra were obtained on previously de-
scribed instrumentation,29 with the third harmonic of a Nd:YAG
laser (355 nm, 10 ns fwhm, 5 mJ pulse-1) as the excitation
source. Time-resolved infrared spectroscopy was carried out on
a step-scan FTIR instrument that has been previously de-
scribed.30,31 Samples were dissolved in CH3CN and were purged

11120 J. Phys. Chem. A, Vol. 105, No. 49, 2001
Wang et al.
TABLE 1: Electrochemical Potentialsa
with Ar. Excitation was effected using the third harmonic output
of a Nd:YAG laser (355 nm).
complex
E1/2ox/V
E1/2red/V
The emission spectra were fitted using a single-mode Franck-
+1.30 (RuII/III
)
)
)
-1.34 (bpy0/•-
-1.53 (bpy0/•-
-1.79 (bpy0/•-
)
)
)
2+
Ru(bpy)3
4,4′-Ru
5,5′-Ru
Condon expression (eq 1a),
+1.33 (RuII/III
+1.35 (RuII/III
-1.11 (44L0/•-
)
E00 - νmpωm 3(Sm)νm
5
-1.46 (bpy0/•-
-1.71 (bpy0/•-
-0.97 (55L0/•-
)
)
I(νj) )
∑
{
(
)
E00
νm!
νm)0
)
-1.42 (bpy0/•-
)
2
νj - E00 + νmpωm
-1.58 (bpy0/•- or 55L•-/2-
)
exp -4 ln 2
(1a)
(1b)
}
]
(
)
[
a Potentials in CH3CN/0.1 M tetrabutylammonium perchlorate solu-
tion, Pt disk working electrode, Pt wire auxiliary electrode, Ag/AgCl
quasi-reference electrode. Potentials were calibrated by using a ferrocene
internal standard and are converted to SCE by assuming E1/2(Fc+/Fc)
) +0.40 V vs SCE.33,34
∆νj0,1/2
S ) 1/2(Mω/p)(∆Qe)2
where I(νj) is the relative emission intensity at energy νj, E00 is
the energy of the zero-zero transition, νm is the quantum number
of the average medium-frequency vibrational mode, pωm is the
average of medium-frequency acceptor modes coupled to the
MLCT transition (1300 cm-1 was assumed throughout the
series), Sm is the Huang-Rhys factor (i.e., the electron-vibration
coupling constant), and ∆νj0,1/2 is the half-width of the individual
vibronic bands. Sm is related to the difference in equilibrium
displacement for the normal modes (∆Qe) between the ground
and excited state by eq 1b. In eq 1b, M is the reduced mass and
ω is the angular frequency.
Polarized photoluminescence excitation spectroscopy was
carried out on a SPEX F-112 fluorescence spectrometer that
was fitted with Glans-Thompson polarizers (single-channel, right
angle format).32 Samples were dissolved in ethanol/methanol
(4:1 v:v) and cooled to 80 K in the DN-1704 optical cryostat.
Wavelength-resolved polarization anisotropies, r(λ), were com-
puted according to the following equation,
Results and Discussion
Electrochemistry. Cyclic voltammetry was carried out on
the PE-substituted metal complexes 44Ru, 55Ru, and on Ru-
(bpy)32+. Potentials for the first oxidation and first three
reduction waves for the three complexes are listed in Table 1.
Each complex features an anodic wave in the 1.30-1.35 V
region due to the RuII/III couple. This wave is shifted to slightly
more positive potentials in 44Ru and 55Ru due to the effect of
the PE substituents that exert an electron-withdrawing effect
on the bipyridine. All of the complexes feature three well-
resolved, reversible cathodic waves. For 44Ru, the first reduc-
tion occurs at -1.11 V; this wave clearly arises from reduction
of the coordinated 44L. The +240 mV shift of the ligand-
centered reduction (relative to the first reduction of Ru(bpy)3
2+
)
indicates that the PE substituents lower the LUMO energy of
44L relative to bpy. The second and third cathodic waves in
44Ru occur at slightly more positive potentials compared to
those for the corresponding waves in Ru(bpy)32+. Nevertheless,
given the reasonable correspondence between the potentials for
these waves, it is likely that the second and third waves for
44Ru arise from reduction of the ancillary bpy ligands. For
55Ru, the first cathodic wave is shifted to an even more positive
potential (+370 mV relative to Ru(bpy)32+), indicating that the
substituent effect of the PE groups is stronger when they are in
the 5,5′-positions. The second and third cathodic waves in 55Ru
are also shifted to more positive potentials. While it is likely
that for this complex the second cathodic wave arises from
reduction of an ancillary bpy ligand, it is possible that the third
reduction corresponds to a second one-electron reduction of the
coordinated 55L.
IVV(λ) - GIVH(λ)
r(λ) )
(2)
IVV(λ) + 2GIVH(λ)
where G is IHV(λ)/IHH(λ) and Ixy(λ) is the emission intensity at
wavelength λ with the excitation and emission polarizers
adjusted according to x and y, respectively. (For example, IHV
is the emission intensity with horizontally polarized excitation
and vertically polarized emission.) Four separate excitation scans
were required to obtain the necessary intensity arrays, IVV(λ),
IVH(λ), IHV(λ), and IHH(λ). These four intensity arrays were
imported into an Excel spreadsheet that was used to compute
r(λ) according to eq 2.
Electrochemical measurements were conducted on CH3CN
solutions with tetrabutylammonium hexafluorophosphate (TBAH,
Aldrich) as the supporting electrolyte. Cyclic voltammetry was
carried out on nitrogen bubble-degassed solutions (TBAH )
0.1 M) with a BAS CV-27 voltammograph and MacLab Echem
software. A platinum disk working electrode, platinum wire
auxiliary electrode, and silver wire quasi-reference electrode
were used, and potentials were corrected to values vs SCE via
an internal standard (ferrocene, E1/2(Fc+/Fc) ) +0.40 V vs
SCE).33,34 A scan rate of 100 mV s-1 was employed in all
measurements.
Calculations were carried out using the Gamess client
software35 running through the Chem 3D (Cambridge Software)
interface. The ligand structures were geometry optimized by
using Hartree-Fock SCF theory with an STO 3G basis set. In
the optimized structures of both ligands, the 2,2′-bipyridine unit
was in the s-cis conformation and all of the aryl rings were in
the xy plane.
Several conclusions can be drawn from the electrochemistry
of the PE-substituted complexes. First, it is clear that the PE
groups lower the LUMO energy in the substituted bpy ligands.
The effect is more pronounced in 55Ru, consistent with previous
studies which show that electron-withdrawing substituents exert
a greater stabilizing effect on the LUMO energy when they are
in the 5,5′-positions.2 Although it is unclear how the PE
substituents act to stabilize the LUMO in 44L and 55L,
molecular orbital calculations of PE-substituted 2,2-bipyridines
suggest that the effect arises from orbital delocalization into
the PE substituents (vide infra).36,37 Second, in a recent study
Swager and co-workers reported the electrochemical properties
of a series of tris(2,2′-bipyridine) complexes in which the ligands
were substituted in the 4,4′- and 5,5′-positions with bithiophene
(BT) moieties.38 By analogy with the present work, the BT
substituents stabilize the LUMO of the substituted bpy ligands
and the effect is larger when the BT groups are in the 5,5′-

Ru(II)-Bpy Complexes Containing Phenyleneethynylene
J. Phys. Chem. A, Vol. 105, No. 49, 2001 11121
TABLE 2: Photophysical Properties of Ru Complexesa
complex
λmaxabs/nm
ꢀmax/103 M-1 cm-1
assignment
λmaxem/nm
Φem
τ
em, µs
τ
TA/µs
kr /105 s-1
knr /105 s-1
44Ru
482
445
402
367
315
287
490(sh)
442
366
330
287
19.3
16.1
14.0
20.2
49.2
79.0
7.4
10.0
68.9
46.5
76.0
Ru f 44L MLCT
Ru f bpy MLCT
668
0.16
1.4
1.3
1.1
6.0
π,π* 44L
π,π* bpy
Ru f 55L MLCT
Ru f bpy MLCT
π,π* 55L
55Ru
693
0.035
0.52
0.45
0.67
18
π,π* bpy
a Argon degassed CH3CN solution.
intraligand bands that occur in the UV region, the metal
complexes feature a manifold of MLCT transitions in the visible
region. First, 55Ru features two prominent absorptions in the
UV: a narrow band at 287 nm arising from the π,π* absorption
of the ancillary bpy ligands and a second broader band at 366
nm that is due to the long-axis polarized π,π* absorption of
the coordinated 55L. Note that the π,π* absorption of 55L is
red-shifted in the complex relative to its position in the free
ligand. Similar behavior has been observed for other π-conju-
gated bipyridine ligands, and the effect has been attributed to
the effect of the (electrophilic) metal center on the energies of
the π and π* levels of the ligand.13,38,41 In the visible region,
55Ru features a broad MLCT absorption that appears to consist
of at least two overlapping bands, one at 442 nm and the second
at 490 nm. We assign the low-energy band to the dπ (Ru) f
π* (55L) MLCT transition and the higher energy band to the
dπ (Ru) f π* (bpy) MLCT transition. These assignments are
substantiated by the photoluminescence excitation polarization
spectroscopy described below.
In complex 44Ru, the UV absorption is dominated by the
π,π* transitions of the bpy ligands. The π,π* transition of the
coordinated 44L appears as a shoulder on the red side of the
bpy absorption. The MLCT band, which is in the visible region,
is considerably more intense compared that of 55Ru. It consists
of a strong band at 482 nm with a possible second band that
appears as a shoulder at 445 nm. The low-energy band is
assigned to the dπ (Ru) f π* (44L) MLCT transition, while
the higher energy band is attributed to the π (Ru) f π* (bpy)
MLCT transition.
Figure 1. UV-visible absorpton spectra in CH3CN solutions: (a) (-)
44L, (- - -) 55L; (b) (-) 44Ru, (- - -) 55Ru.
positions. However, the BT substituents have a significantly
smaller effect on the LUMO energy compared to the PE
substituents.
UV-Visible Absorption Spectroscopy. UV-visible absorp-
tion spectra were recorded for the free ligands 44L and 55L in
CHCl3 and for metal complexes 44Ru and 55Ru in CH3CN.
Figure 1 shows the spectra, and Table 2 provides a tabular listing
of the absorption bands and band assignments in the metal
complexes. The free ligands display relatively intense absorption
in the UV region. In 55L, which features a “linear” geometry,
the predominant band is shifted to lower energy and is more
intense compared to the absorption of 44L. The predominant
absorption in 55L arises from the long-axis polarized π,π*
absorption of the “oligomer”. This absorption is very similar
in intensity, band shape, and energy to the absorption of linear
oligomeric phenyleneethynylenes (OPEs) that have been previ-
ously reported.11,13,39,40 In addition to being blue-shifted and less
intense, the absorption of 44L is more structured. The blue shift
likely arises due to disruption in the long-axis conjugation that
arises due to the 4,4′-pattern of substitution on the bpy moiety
(i.e., with respect to the π-conjugated system, this molecule is
1,3- or “meta” linked).
The enhanced oscillator strength in the dπ (Ru) f π* (44L)
MLCT transition likely arises because of orbital delocalization
into the PE-substituents. This premise is supported by previous
studies of MLCT transitions in aryl- and PE-substituted diimine
metal complexes. The effect of orbital delocalization on the
intensity of MLCT transitions was first noted by McMillin and
Pfifer in a study of a series of Cu(I)-phenanthroline com-
plexes.42 These authors applied Mulliken charge transfer theory
to analyze the MLCT absorption of the Cu(I) complexes, and
their results showed that when the charge-transfer distance is
increased due to orbital delocalization in the acceptor ligand,
the oscillator strength of the MLCT transition increases.42 We
recently reported a related study on a Re(I) complex that
contains PE substituents in the 4,4′-positions of a bpy ligand
and concluded that the effective charge-transfer distance in the
MLCT transition is increased by more than 50% due to
delocalization of the excited electron into the PE substituents.36
Photoluminescence Spectroscopy. Free ligands 44L and 55L
feature a strong fluorescence in the near-UV region that arises
from the π,π* state with λmax at 350 and 385 nm, respectively.
In the metal complexes the fluorescence is quenched, and it is
As might be expected, the absorption spectra of the metal
complexes 44Ru and 55Ru are considerably more complex than
compared to the free ligand counterparts. In addition to the

11122 J. Phys. Chem. A, Vol. 105, No. 49, 2001
Wang et al.
C-C modes of the PE-substituted bpy ligands. Second, the
Huang-Rhys parameters (Sm) that are recovered from the 80
K fitted spectra indicate that in both complexes the electron-
vibrational coupling in the MLCT excited state is comparatively
low. This feature is quite clear from the appearance of the low-
temperature emission spectra, where the intensity of the 0,1
transition is considerably less intense compared to that of the
0,0 band. (For comparison, Sm for Ru(bpy)32+ is ∼0.95).28 Third,
Sm is greater for 55Ru than for 44Ru, despite the fact that the
energy of the MLCT state is larger for the latter complex. This
runs counter to expectation, since theory and experimental
evidence indicate that Sm typically increases with E00.45 Based
on the fact that Sm(44Ru) < Sm(55Ru), we tentatively conclude
that in the MLCT state the excited electron is more delocalized
in 44Ru. This point is discussed at more length in a later section.
Excited-State Decay Kinetics. In many Ru(II)-polypyridine
complexes, the decay kinetics of the MLCT state are compli-
cated by the existence of a metal-centered (dd) excited state
that is in close energetic proximity to the luminescent MLCT
state.46,47 The existence of the dd state is typically inferred from
the observation of a moderately strong temperature dependence
in the MLCT decay kinetics.47 This temperature dependence
arises from thermally activated MLCT f dd internal conversion
(IC). However, when the MLCT state is at a comparatively low
energy, thermally activated MLCT f dd IC is slower than the
normal radiative and nonradiative decay channels and, conse-
quently, the MLCT decay kinetics are relatively temperature
independent.47 In this case, the ambient temperature emission
decay lifetime (τem) and emission quantum yield (φem) provide
reasonable estimates for the intrinsic radiative and nonradiative
decay rates (kr and knr, respectively) of the MLCT excited state
according to eqs 3 and 4,
Figure 2. Intensity normalized photoluminescence spectra of 44Ru
(-) and 55Ru (- - -): (a) 80 K, ethanol/methanol (4:1 v:v) glasses; (b)
298 K, CH3CN solutions.
TABLE 3: Emission Spectral Fitting Parametersa
complex
T/K
E00/cm-1
pωm/cm-1
Sm
∆νj0,1/2/cm-1
44Ru
44Ru
55Ru
55Ru
298
80
298
80
15100
15700
14500
15200
1200
1200
1300
1300
0.7
0.7
0.85
0.85
1500
1000
1450
1000
a Ethanol/methanol (4:1 v:v) solvent.
kr ) φem/τem
(3)
(4)
replaced by a red photoluminescence (PL). The PL spectra of
the complexes obtained in CH3CN at 298 K and in ethanol/
methanol (4:1 v:v) at 80 K are illustrated in Figure 2. At room
temperature, the PL from the complexes appears as a broad band
with λmax ) 668 (44Ru) and 693 nm (55Ru) (Figure 2b). At
80 K, the emission from both complexes is blue-shifted and
appears as a broad manifold with a clearly resolved vibronic
progression (Figure 2a). The red PL is clearly due to the MLCT
excited states for both complexes. The MLCT assignment is
supported by the PL decay lifetimes, which are 1.4 and 0.52 µs
(298 K in CH3CN), for 44Ru and 55Ru, respectively. The blue
shift that occurs upon cooling is due to the rigidochromic effect,
which is well-documented for MLCT PL in transition metal
complexes.43,44
To extract information concerning the structure of the
luminescent excited states from the PL data, the spectra were
fitted using a single-vibrational mode Franck-Condon band
shape expression (eq 1, see Experimental Section). The fits were
first carried out on the 80 K spectra and in this procedure all
four fitting parameters (E00, pωm, Sm, and ∆νj0,1/2) were varied
to optimize the fits. The room temperature spectra were
subsequently fitted. In this case the parameters Sm and pωm were
set to the values recovered from the 80 K fitting procedure,
knr ) (1 - φem)/τem
Given the relatively low energy of the MLCT state in 44Ru
and 55Ru, it was anticipated that at ambient temperature
thermally activated MLCT f dd IC would not be competitive
with the normal excited-state decay channels. To confirm this
premise, the temperature dependence of τem for the two
complexes was determined in CH3CN solution over the tem-
perature range 235-295 K. For both complexes over this
temperature range τem varied by less than 10% (see Table in
Supporting Information). In view of this result we conclude that
reasonably accurate kr and knr values can be computed from
the ambient temperature τem and φem values. Thus, the τem and
φem values for 44Ru and 55Ru determined at ambient temper-
ature in CH3CN solution were used to compute kr and knr values
according to eqs 3 and 4 (see Table 2 for values).
There are several noteworthy features with respect to the
photophysical parameters of the PE-substituted complexes. First,
both τem and φem are considerably larger for 44Ru. This trend
is consistent with the energy gap law: since E00 is larger for
44Ru, one would expect knr for this complex to be lower (τem
and φem increase as knr decreases).1,48 However, inspection of
the kr and knr values for 44Ru and 55Ru indicates that the
considerably enhanced φem for 44Ru arises not only because
knr is smaller but also because kr is larger. This finding indicates
that radiative decay is enhanced when the PE substituents are
in the 4,4′-positions. This is consistent with a previous study
by McCusker that examined a series of aryl-substituted Ru-
polypyridine complexes in which the aryl groups were in the
4,4′-positions of the bpy ligands.8 In this study it was seen that
and the fits were optimized by varying only E00 and ∆νj0,1/2
.
The parameters recovered from the fitting procedure are listed
in Table 3, and the fitted spectra are available as Supporting
Information.
Several points are of interest with respect to the parameters
recovered from the fits. First, in both cases in the 80 K spectra
there is a clearly resolved vibrational progression arising from

Ru(II)-Bpy Complexes Containing Phenyleneethynylene
J. Phys. Chem. A, Vol. 105, No. 49, 2001 11123
state absorption band that is assigned to the π,π* transition of
55L, rather than in the region of the MLCT absorption (400-
500 nm). The lack of bleaching in the ground-state MLCT bands
is believed to be due to the fact that the excited-state absorption
is very strong in this region (i.e., ∆ꢀ is large and positive, which
offsets any bleaching due to the loss of absorptivity in the MLCT
bands). It is believed that the principal TA absorption features
observed for excited state 55Ru arise from to the reduced 5,5′-
PE substituted ligand, i.e., [(bpy)2RuIII(55L•-)]2+*. The strong
visible absorption band in the excited state apparently arises
from the π*-π* transitions for 55L•-; it is likely that because
of the strong conjugation in this ligand, the absorptivity of the
radical anion is large.
The dominant TA spectral features of MLCT excited states
arise from the absorbance of the reduced acceptor ligand.50,52,53
In the foregoing analysis of the TA spectra of 44Ru and 55Ru,
a similar argument is invoked to support the TA difference
spectral assignments. To provide further evidence in support
of the premise that the principal TA spectral features observed
for the MLCT states of 44Ru and 55Ru arise from the reduced
PE-substituted ligands, laser flash photolysis experiments were
carried out with 44Ru and 55Ru in CH3CN solution with the
reductive quencher N,N′-dimethylaniline (DMA). Under these
conditions, photoexcitation of the complexes affords the one-
electron reduced complexes in moderate to high yield,54
Figure 3. Transient absorption-difference spectra for 44Ru and 55Ru
in CH3CN solution: (a) 44Ru and (b) 55Ru. Legend: (filled symbols)
complex alone, c ) 2 × 10-5 M; (unfilled symbols) complex, c ) 2 ×
10-5 M, and dimethylaniline, c ) 30 mM. Spectra are the principal
components extracted from factor analysis of time-resolved absorpton
spectral data. The principal component spectra are normalized to
correspond to ∆A at time 0.
[(bpy)2RuII(55L)]2+ + DMA
[(bpy)2RuII(55L•-)]+ + DMA•+ (5)
9
hν8
kr increased systematically as the degree of delocalization into
the aryl subsitutent increased. The authors concluded that the
radiative decay rate is enhanced because the transition dipole
moment, bµ, is increased by orbital delocalization into the 4,4′-
aryl substituents.8 We conclude that a similar effect is operative
in 44Ru; that is, bµ for the MLCT state is enhanced by
delocalization into the 4,4′-PE substituents.
UV-Visible Transient Absorption Spectroscopy. Laser
flash photolysis was carried out on 44Ru and 55Ru in order to
obtain information concerning the (difference) absorption spectra
of the MLCT excited state. The transient absorption (TA)
difference spectra of 44Ru and 55Ru in degassed CH3CN
solution following 355 nm excitation are illustrated in Figure 3
(filled polygons) and the TA decay lifetimes are listed in Table
2. As expected, both complexes feature comparatively strong
transient absorption, and on the basis of the correspondence
between the TA and emission decay lifetimes, the TA difference
spectra are assigned to the MLCT excited states.
[(bpy)2RuII(44L)]2+ + DMA
hν8
9
[(bpy)2RuII(44L•-)]+ + DMA•+ (6)
Since DMA•+ absorbs only moderately in the visible region,55
under the conditions of these experiments the principal TA
features are expected to arise from the difference in absorption
of the corresponding ground state and one-electron-reduced
complexes. Figure 3 illustrates the TA difference spectra of
44Ru and 55Ru obtained in CH3CN solution with [DMA] )
30 mM (unfilled polygons). For both complexes, there is a clear
similarity in the difference spectra of the reduced complexes
and the MLCT excited states. For example, both reduced and
excited state 44Ru display an absorption band in the near-UV;
as noted above, this spectral feature is characteristic of the 2,2′-
bipyridine anion radical. The similarity between the absorption
of the excited state and reduced complexes is especially strong
in the case of 55Ru. This underscores the fact that the reduced
ligand 55L features a strong visible absorption, and this
absorption dominates the spectra of the reduced and excited-
state complexes.
The TA difference spectrum of 44Ru is characterized by
moderately strong absorption in the near-UV, strong ground-
state bleaching in the 400-500 nm region, and broad, moderate
absorption for λ > 500 nm. It is noteworthy that the TA
difference spectrum of 44Ru is qualitatively similar to that of
2+ 49
Emission Excitation Polarization Spectroscopy. Emission
excitation spectroscopy was carried out on the PE-substituted
complexes in order to obtain additional insight concerning the
nature of the transitions observed in the near-UV and visible
spectral regions. These experiments were carried out with the
samples dissolved in an ethanol/methanol solvent glass at 80 K
in order to prevent depolarization due to rotational diffusion.
Ru(bpy)3
,
and it has been shown that for the latter the
principal TA bands arise from intraligand (IL) absorptions due
to the reduced bpy ligand in the excited state.50 By analogy we
suggest that the principal TA features for 44Ru arise due the
reduced 4,4′-PE substituted ligand that is present in the excited-
state complex, [(bpy)2RuIII(44L•-)]2+*.
The TA difference spectrum of 55Ru is distinct from that of
44Ru. First, qualitatively the TA of 55Ru is considerably
stronger than that of 44Ru,51 which suggests that the ground-
excited-state difference absorptivity (∆ꢀ) is larger in the former
complex. The TA difference spectrum of 55Ru features strong
bleaching at λ ≈ 360 nm and strong absorption at λ ≈ 490 nm.
Interestingly, in this complex the bleaching occurs in the ground-
In these experiments the emission anisotropy as a function
of excitation wavelength, r(λex) was determined. The anisotropy
provides a measure of the angular displacement (â) between
the transition moments for the absorbing and emitting states.32
Specifically, when â ) 0° (the transition moments for the two
states are parallel), r ) 0.4 (the theoretical maximum), and when

11124 J. Phys. Chem. A, Vol. 105, No. 49, 2001
Wang et al.
absorption band that is assigned to the 44L IL transition. This
feature implies that the lowest energy π,π* IL transition for
44L is polarized along the long axis of the PE-substituted 2,2-
bipyridine moiety.
In the low-energy region the excitation polarization spectrum
of 55Ru is similar to that of 44Ru. The r value for the lowest
energy MLCT band is also ≈0.3 (â ≈ 20°), which underscores
the fact that the lowest energy absorption and the emission both
occur from a state that is localized on the PE-substituted ligand.
Interestingly, for 55Ru at higher excitation energies r < 0. At
excitation wavelengths that correspond to the π,π* IL transition
of 55L (λ ) 350-400 nm), r ≈ -0.15, which implies that the
absorbing and emitting transition dipoles are almost perpen-
dicular (â ≈ 75°). This is consistent with the hypothesis that
the intense IL absorption band corresponds to the long-axis
polarized π,π* transition, which is oriented perpendicular to
that of the Ru f 55L MLCT transition.
Time-Resolved Infrared Spectroscopy. As noted in the
Introduction, one objective of this project was to use time-
resolved infrared spectroscopy (TRIR) to probe for excited-state
delocalization in 44Ru and 55Ru.58 This idea was based on
TRIR experiments carried out by Meyer and co-workers on the
complexes [Ru(bpy)2(4,4′-(CO2Et)2bpy)]2+ and [Ru(bpy)2(4,4′-
(CONEt2)2bpy)]2+ in which one of the 2,2′-bipyridine ligands
is substituted with carboxyester and carboxyamide groups.20 In
this study it was observed that in the MLCT excited state the
frequency of the CdO stretching modes (νCdO) of the ester and
amide functional groups decreased due to the increased electron
Figure 4. Absorption (fine lines, scale at left) and excitation anisotropy
(bold lines, scale at right) spectra obtained at 80 K on samples in
ethanol/methanol (4:1 v:v) glass: (a) Ru(bpy)32+; (b) 44Ru; (c) 55Ru.
â ) 90° (the transition moments are perpendicular), r ) -0.2.
Intermediate values of r are observed when â is between 0 and
90°.32
density in the acceptor bpy ligand. The frequency shifts (∆νCdO
)
were -15 and -25 cm-1 for the amide- and ester-substituted
Figure 4 illustrates absorption spectra obtained at 80 K along
complexes, respectively.
2+
with the excitation polarization spectra of Ru(bpy)3 and the
By analogy to this work, we anticipated that it might be
possible to observe shifts in νCdO for the diethylamide substit-
uents in 44Ru and 55Ru. Moreover, it was hoped that
differences in ∆νCdO for the excited-state complexes might be
used to compare the extent of delocalization in the MLCT
excited state for the two complexes. Ground-state infrared
spectra of 44Ru and 55Ru (CD3CN solution) feature a band at
1629 cm-1, which is assigned to νCdO for the amide groups.
Unfortunately, TRIR experiments carried out in the same solvent
revealed no detectable change in νCdO for the excited-state
complexes. This result indicates that νCdO shifts by less than 3
cm-1 when the complexes are in the MLCT excited state. In
view of the fact that ∆νCdO ) -15 cm-1 for [Ru(bpy)2(4,4′-
(CONEt2)2bpy)]2+, this result suggests in 44Ru and 55Ru there
is comparatively little added charge density in the phenylene
rings when the complexes are in the MLCT excited state. An
alternative explanation is that due to orbital delocalization in
the PE-bpy ligands, the charge buildup at the amido carbonyl
groups in the MLCT state is insufficient to give rise to a
PE-substituted complexes.56 The excitation polarization spectrum
2+
for Ru(bpy)3 is provided to allow comparison with the data
for the PE-substituted complexes. Our data for Ru(bpy)32+ are
consistent with previous reports;57 the complex features a
maximum r value of ≈0.15 when the excitation wavelength
corresponds to the lowest MLCT transition. The fact that r <
0.4 is believed to be due to the fact that although initial
excitation produces a state that is localized on a single bpy
ligand, randomization from the original ligand-localized state
occurs prior to light emission. This randomization results in
emission from a population in which some molecules have their
emission dipoles displaced significantly from the absorption
dipole.57
Next we turn to consider the low-temperature absorption and
excitation polarization spectra of 44Ru and 55Ru. First, the
low-temperature absorption spectra of both complexes feature
improved resolution in all of the transitions. For both complexes
the MLCT transitions in the 400-520 nm region are clearly
resolved into two bands. In each case the lowest energy band
is due to the Ru f 44L (or 55L) transition, while the higher
energy band arises from the Ru f bpy transition. The IL bands
in the near-UV are also better resolved at 80 K.
The excitation polarization spectrum of 44Ru features r ≈
0.3 in the lowest energy MLCT band. The comparatively large
anisotropy signals that the absorption and emission dipoles are
nearly collinear (â ≈ 20°) in this transition. Moreover, the large
anisotropy confirms that emission occurs almost exclusively
from an MLCT state that is localized on the PE-substituted
ligand, i.e., the emitting state is best described as (bpy)2RuIII-
(44L•-)2+*. The anisotropy decreases for λex < 450 nm because
the transition dipoles for the Ru f bpy transitions are not
parallel to that of the emitting state (i.e., Ru f 44L). An
interesting point is that r ≈ -0.1 on the red edge of the
measurable shift in νCdO
.
Excited-State Delocalization in the PE-Substituted Com-
plexes. It is clear that in 44Ru and 55Ru in the MLCT state
the excited electron is localized on the PE substituted ligand.
This conclusion is substantiated by the PL and transient
absorption results, which indicate that the properties of the
excited-state correlate with the structure and LUMO energy of
the PE substituted ligands. It less clear, however, to what extent
the excited electron is delocalized into the PE substituents, and
whether the extent of delocalization is greater when the
substituents are in the 4,4′- or 5,5′-positions on the bpy ligand.
Electrochemistry of 44Ru and 55Ru indicates that the PE
substituents decrease the LUMO energy of the bpy ligand.
Furthermore, the LUMO stabilization is markedly greater when

Ru(II)-Bpy Complexes Containing Phenyleneethynylene
J. Phys. Chem. A, Vol. 105, No. 49, 2001 11125
Figure 5. Graphical display of pz orbital coefficients for LUMO and LUMO+1 molecular orbitals for 44L and 55L. Calculated using Hartree-
Fock SCF theory using the STO-3G basis set.
the PE substituents are in the 5,5′-positions. This is not
surprising, because with respect to the π-system of the PE
substituted bpy ligand, the conjugation length is greater for 55L
than for 44L. Improved π-conjugation in 55L is a natural
consequence of the fact that the pyridyl rings in the π-system
are 1,4-substituted. By contrast, in 44L the π-system is disrupted
by the 1,3- (or “meta”) pattern of substitution on the pyridine
rings. The increased conjugation length (and concomitant
lowering of the LUMO energy) in 55L is clearly signaled by
the fact that this compound features a lower energy π,π*
absorption and fluorescence.
All of the available evidence suggests that the LUMO in 55L
is more strongly delocalized than in 44L. However, despite this
fact, the photophysical data on 44Ru and 55Ru imply that in
the MLCT state the excited electron is more delocalized in
44Ru. First, the oscillator strength of the Ru f 44L transition
in 44Ru is significantly larger than that of the Ru f 55L
transition in 55Ru. It is also larger than that of the Ru f bpy
MLCT transition in Ru(bpy)32+. In addition, the radiative decay
rate is enhanced in 44Ru. These features imply that the MLCT
transition dipole is larger in 44Ru compared to that in 55Ru
and Ru(bpy)32+, a result which strongly implies that delocal-
ization into the PE substituents occurs in 44Ru. (It is important
to note that because the axis of π-conjugation in 55L is
perpendicular to the MLCT transition dipole, a priori the
transition dipole moment is not expected to be enhanced in this
complex by conjugation into the PE substitutents.) Second, the
low-temperature PL spectra clearly indicate that electron
vibrational coupling is lower in 44Ru compared to that of 55Ru
and other Ru-polypyridyl complexes with similar excited-state
energies.8,48 In addition, the nonradiative decay rate is low in
44Ru, again compared to 55Ru and other ruthenium(II)
complexes.8,48 Several groups have argued that enhanced
delocalization in the MLCT state gives rise to decreased
electron-vibrational coupling, and this in turn leads to a decrease
in the nonradiative decay rate.5,6,8 This effect may be manifested
in 44Ru and to a lesser extent in 55Ru.
the structure of the low-energy π* orbitals on each ligand, since
it is believed that these orbitals are the “acceptors” for the dπ
(Ru) f π* (44L or 55L) MLCT transition. Figure 5 illustrates
the coefficients of the pz orbitals on the C and N atoms for the
LUMO and (LUMO+1) orbitals of each ligand. The symmetry
of the LUMO is the same in each ligand (B1 in C2V local
symmetry) and is correct for overlap with the Ru dxz orbital
(which also transforms B1 in C2V local symmetry). This pattern
follows for the (LUMO+1) orbitals on each ligand, which
feature A2 symmetry, which would be appropriate for overlap
with the Ru dyz orbital (which transforms A2 in C2V local
symmetry). Qualitatively, it is clear that in both cases the LUMO
and (LUMO+1) orbitals are delocalized significantly into the
PE substituents. On a more quantitative level, the degree of
delocalization is virtually identical in both ligands. Focusing
on the LUMO, as assessed by the sums of the squares of the pz
orbital coefficients in both ligands, 75% of the electron density
in the LUMO is centered on the bpy unit and 25% is delocalized
into the PE substituents. Thus, by this measure, one would
expect the degree of delocalization to be the same in each ligand.
However, close inspection of the LUMO and (LUMO+1)
orbitals reveals one important distinction between 44L and 55L.
In 44L the pz coefficient on the pyridine nitrogen atoms is
significantly larger in both the LUMO and (LUMO+1) orbitals.
Indeed, the difference in the nitrogen pz coefficients in the
LUMO+1 is striking: 0.26 in 44L vs 0.02 in 55L. This
difference is significant, because in the Mulliken charge transfer
theory, the degree of charge transfer is directly proportional to
the overlap between the orbitals localized on the donor and
acceptor moieties.9,36,42 Since the nitrogen atoms are the point
of overlap between the π-system of the acceptor ligands and
the metal-centered orbitals, the overlap depends strongly upon
the coefficient of the pz orbital on the nitrogen atoms. On this
basis we believe that the MO calculations point to the possible
origin of the difference in charge delocalization between 44Ru
and 55Ru, that is, overlap between the Ru-based dπ orbitals
and the π-electron system is greater in the former complex. This
difference in overlap stems from the difference in structure of
the low-energy π* orbitals, which in turn is related to the
geometry of the ligands. In essence, metal-ligand orbital
overlap in 44Ru is improved by the spatial distribution of the
low-energy π* orbitals on the PE-substituted ligand.
Taken together, the results imply that delocalization in the
MLCT state is greater in 44Ru than in 55Ru. In an attempt to
understand this difference, molecular orbital calculations were
carried out on the free ligands 44L and 55L. The calculations
were carried out on the neutral free ligands using Hartree-
Fock SCF theory with a minimal basis set (STO-3G). The
objective of these calculations was to provide insight concerning
An important outcome of this study is that it demonstrates
that although the effect of substituents on the energy of the π*

11126 J. Phys. Chem. A, Vol. 105, No. 49, 2001
Wang et al.
(9) Mines, G. A.; Roberts, J. A.; Hupp, J. T. Inorg. Chem. 1992, 31,
125-126.
(10) Ley, K. D.; Whittle, C. E.; Bartberger, M. D.; Schanze, K. S. J.
Am. Chem. Soc. 1997, 119, 3423-3424.
(11) Ley, K. D.; Li, Y. T.; Johnson, J. V.; Powell, D. H.; Schanze, K.
S. Chem. Commun. 1999, 1749-1750.
orbital in 2,2′-bipyridine is stronger when these substituents are
in the 5,5′-positions, the effect of π-delocalization in the MLCT
state is greater when the π-acceptor substituents are located in
the 4,4′-positions. Both of these effects stem from the spatial
distribution of the low-energy π* orbitals on the 2,2′-bipyridine
unit. Thus, in the LUMO the orbital coefficients are larger on
the 5,5′-positions, which enhances the electronic effect of
substituents located in these positions. By contrast, the degree
of metal to ligand charge transfer is strongly modulated by the
size of the orbital coefficient on nitrogen,9,36,42 and our MO
calculations indicate that the presence of a π-acceptor in the
4,4′-positions leads to enhancement of the coefficient of the pz
nitrogen orbital, which in turn enhances the degree of charge
transfer in the MCLT state. We believe that these effects are
generalizable to any ligand substituents that function as π-ac-
ceptors.
(12) Li, Y. T.; Whittle, C. E.; Walters, K. A.; Ley, K. D.; Schanze, K.
S. Pure Appl. Chem. 2001, 73, 497-501.
(13) Walters, K. A.; Ley, K. D.; Cavalaheiro, C. S. P.; Miller, S. E.;
Gosztola, D.; Wasielewski, M. R.; Bussandri, A. P.; van Willigen, H.;
Schanze, K. S. J. Am. Chem. Soc. 2001, 123, 8329-8342.
(14) Wong, C. T.; Chan, W. K. AdV. Mater. 1999, 11, 455-459.
(15) Wang, J.; Wang, R.; Yang, J.; Zheng, Z.; Carducci, M. D.; Cayou,
T.; Peyghambarian, N.; Jabbour, G. E. J. Am. Chem. Soc. 2001, 123, 6179-
6180.
(16) Bernier, P., Lefrant, S., Bidan, G., Eds. AdVances in Synthetic
Metals. Twenty Years of Progress in Science and Technology; Elsevier
Science SA: Amsterdam, 1999.
(17) Peng, Z.; Gharavi, A. R.; Yu, L. J. Am. Chem. Soc. 1997, 119,
4622-4632.
(18) Wang, Q.; Wang, L.; Yu, L. J. Am. Chem. Soc. 1998, 120, 12860-
12868.
Conclusion
(19) Li, Y.; Whittle, C. E.; Walters, K. A.; Ley, K. D.; Schanze, K. S.
MRS Symp. Proc., in press.
A detailed photophysical study has been carried out on the
complexes 44Ru and 55Ru. These complexes feature a lowest
excited state that is based on dπ (Ru) f π* (44L or 55L) metal-
to-ligand charge transfer. The presence of the PE substituents
has a marked effect on the absorption, photoluminescence, and
transient absorption spectra of the complexes. There is clear
evidence that in the MLCT state the excited electron is
delocalized into the PE-substituents. Moreover, the data also
imply that the extent of delocalization is stronger in 44Ru, that
is, when the PE substituents are in the 4,4′-positions on the bpy
unit. This is interesting, in view of the fact that the LUMO level
is lower in 55L, a feature that implies that the extent of
intraligand delocalization is larger in this ligand. Molecular
orbital calculations imply that the extent of delocalization in
the MLCT state is related to the overlap between the low-energy
π* orbitals localized on the ligand with the metal-centered d
orbitals. The overlap is strongly dependent on the size of the
π* orbital coefficients on the pyridine nitrogens, which in turn
is modulated by the pattern of substituents on the bpy unit. We
conclude that in general, π-acceptor substituents that are
positioned in the 4,4′-positions will induce stronger delocaliza-
tion in the MLCT excited state.
(20) Chen, P. Y.; Omberg, K. M.; Kavaliunas, D. A.; Treadway, J. A.;
Palmer, R. A.; Meyer, T. J. Inorg. Chem. 1997, 36, 954-955.
(21) Omberg, K. M.; Smith, G. D.; Kavaliunas, D. A.; Chen, P.;
Treadway, J. A.; Schoonover, J. R.; Palmer, R. A.; Meyer, T. J. Inorg.
Chem. 1999, 38, 951-956.
(22) Grosshenny, V.; Ziessel, R. Tetrahedron Lett. 1992, 33, 8075-
8078.
(23) Grosshenny, V.; Romero, F. M.; Ziessel, R. J. Org. Chem. 1997,
62, 1491-1500.
(24) Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 1996, 61, 1133-1135.
(25) Sessler, J. L.; Brown, C. T.; Wang, R.; Hirose, T. Inorg. Chim.
Acta 1996, 251, 135-140.
(26) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978,
17, 3334-3341.
(27) Harriman, A. J. Chem. Soc., Chem. Commun. 1977, 777-778.
(28) Caspar, J. V. Ph.D. Dissertation, University of North Carolina,
Chapel Hill, NC, 1982.
(29) Wang, Y.; Schanze, K. S. Chem. Phys. 1993, 176, 305-319.
(30) Omberg, K. M.; Schoonover, J. R.; Treadway, J. A.; Leasure, R.
M.; Dyer, R. B.; Meyer, T. J. J. Am. Chem. Soc. 1997, 119, 7013-7018.
(31) Omberg, K. M.; Schoonover, J. R.; Meyer, T. J. J. Phys. Chem. A
1997, 101, 9531-9536.
(32) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Kluwer Academic/Plenum Publishers: Dordrecht, The Netherlands, 1999.
(33) Gennett, T.; Milner, D. F.; Weaver, M. J. J. Phys. Chem. 1985,
89, 2787-2794.
(34) Sahami, S.; Weaver, M. J. J. Electroanal. Chem. 1981, 122, 155-
170.
Acknowledgment. We gratefully acknowledge support from
the National Science Foundation (Grant No. CHE-9901862) for
support of this work.
(35) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;
Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.;
Su, S. J.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem.
1993, 14, 1347-1363.
(36) Walters, K. A.; Premvardhan, L. L.; Liu, Y.; Peteanu, L. A.;
Schanze, K. S. Chem. Phys. Lett. 2001, 339, 255-262.
(37) Walters, K. A. Ph.D. Dissertation, University of Florida, Gainesville,
FL, 2000.
(38) Zhu, S. S.; Kingsborough, R. P.; Swager, T. M. J. Mater. Chem.
1999, 9, 2123-2131.
(39) Ziener, U.; Godt, A. J. Org. Chem. 1997, 62, 6137-6143.
(40) Grummt, U. W.; Birckner, E.; Klemm, E.; Egbe, D. A. M.; Heise,
B. J. Phys. Org. Chem. 2000, 13, 112-126.
(41) Manas, E. S.; Chen, L. X. Chem. Phys. Lett. 2000, 331, 299-307.
(42) Phifer, C. C.; McMillin, D. R. Inorg. Chem. 1986, 25, 1329-1333.
(43) Wrighton, M.; Morse, D. L. J. Am. Chem. Soc. 1974, 96, 998-
1003.
(44) Lees, A. J. Chem. ReV. 1987, 87, 711-743.
(45) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. J. Phys.
Chem. 1986, 90, 3722-3734.
(46) Van Houten, J.; Watts, R. J. Am. Chem. Soc. 1976, 98, 4853-
4858.
(47) Wacholtz, W. F.; Auerbach, R. A.; Schmehl, R. H. Inorg. Chem.
1986, 25, 227-234.
(48) Barqawi, K. R.; Murtaza, Z.; Meyer, T. J. J. Phys. Chem. 1991,
95, 47-50.
(49) Ohno, T.; Yoshimura, A.; Prasad, D. R.; Hoffman, M. Z. J. Phys.
Chem. 1991, 95, 4723-4728.
(50) Watts, R. J. J. Chem. Educ. 1983, 60, 834-842.
1
Supporting Information Available: Mass and H NMR
spectra of 44L, 55L, 44Ru, and 55Ru, fitted photoluminescence
spectra of 44Ru and 55Ru, and temperature-dependent emission
lifetimes of 44Ru and 55Ru. This material is available free of
charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Caspar, J. V.; Meyer, T. J. J. Phys. Chem. 1983, 87, 952-957.
(2) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.;
von Zelewsky, A. Coord. Chem. ReV. 1988, 84, 85-277.
(3) Worl, L. A.; Duesing, R.; Chen, P.; Della Ciana, L.; Meyer, T. J.
J. Chem. Soc., Dalton Trans. 1991, 849-858.
(4) Schanze, K. S.; Macqueen, D. B.; Perkins, T. A.; Cabana, L. A.
Coord. Chem. ReV. 1993, 122, 63-89.
(5) Strouse, G. F.; Schoonover, J. R.; Duesing, R.; Boyde, S.; Jones,
W. E., Jr.; Meyer, T. J. Inorg. Chem. 1995, 34, 473-487.
(6) Treadway, J. A.; Loeb, B.; Lopez, R.; Anderson, P. A.; Keene, F.
R.; Meyer, T. J. Inorg. Chem. 1996, 35, 2242-2246.
(7) Grosshenny, V.; Harriman, A.; Romero, F. M.; Ziessel, R. J. Phys.
Chem. 1996, 100, 17472-17484.
(8) Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K.
J. Am. Chem. Soc. 1997, 119, 8253-8268.

Ru(II)-Bpy Complexes Containing Phenyleneethynylene
J. Phys. Chem. A, Vol. 105, No. 49, 2001 11127
(51) The spectra for the two complexes were obtained at the same laser
power and with samples having the same ground-state absorptivity at the
excitation wavelength.
(55) Zhang, X. M.; Yeh, S. R.; Hong, S.; Freccero, M.; Albini, A.;
Falvey, D. E.; Mariano, P. S. J. Am. Chem. Soc. 1994, 116, 4211-4220.
(56) The low-temperature absorption spectra feature a tailing baseline
on the long wavelength side of the MLCT bands. This feature is due to
light scattering that arises because the samples are contained in an optical
cryostat and consequently the light beam passes through four windows.
(52) Sun, H.; Hoffman, M. Z.; Mulazzani, Q. G. Res. Chem. Intermed.
1994, 20, 735-754.
(53) Chen, P.; Danielson, E.; Meyer, T. J. J. Phys. Chem. 1988, 92,
3708-3711.
(57) Blakley, R. L.; Myrick, M. L.; DeArmond, M. K. Inorg. Chem.
1988, 27, 589-590.
(58) Schoonover, J. R.; Strouse, G. F. Chem. ReV. 1998, 98, 1335-
1355.
(54) Stern-Volmer luminescence quenching experiments were carried
out using DMA. Both 44Ru and 55Ru were quenched efficiently, and the
following bimolecular quenching constants were extracted from the quench-
ing data: kq(44Ru) ) 5.2 × 108 M-1 s-1; kq(55Ru) ) 1.8 × 109 M-1 s-1
.