Photophysical properties of Ru(II) bipyridyl complexes containing hemilabile phosphine-ether ligands

Article abstract of DOI:10.1021/ic050436l

Emission and absorbance spectra, along with low-temperature excited-state lifetimes, were obtained for the hemilabile complexes, [Ru(bpy) ₂L](PF₆)₂ [L = (2-methoxyphenyl) diphenylphosphine (RuPOMe) (1) and (2-ethoxyphenyl)-diphenylphosphine (RuPOEt) (2)] in solid 4:1 ethanol/methanol solution. Spectral data were evaluated with ground-state reduction potentials using Lever parameters. Lifetime data for these complexes were collected from 77 to 160 K, and the rate constant for the combined radiative and nonradiative decay process, k, the thermally activated process prefactor, k′⁰, the rate constant for the MLCT → d-d transition, k′, and the activation energy, ΔE′, were calculated from a plot of In(1/τ) versus 1/T for both (1) and (2). The low-temperature luminescence lifetimes of (1) were observed to decrease with increases in water concentration. The photophysical and kinetic data of (1) and (2) are compared to literature data for [Ru(bpy)₃](PF ₆)₂. The emission maxima of (1) and (2) are blue-shifted relative to [Ru(bpy)₃](PF₆)₂ due to the presence of the strong-field phosphine ligand, which enhances π back-bonding to the bipyridyl ligands. The thermal activation energy, ΔE′, is significantly larger for [Ru(bpy)₃](PF₆)₂ than for (1) and (2) resulting in a faster MLCT → d-d transition for (1) and (2). These results are discussed in the context of radiationless decay through thermally activated ligand-field states on the metal complex.

Full text of DOI:10.1021/ic050436l

Inorg. Chem. 2005, 44, 7377−7384
Photophysical Properties of Ru(II) Bipyridyl Complexes Containing
Hemilabile Phosphine Ether Ligands
−
Sarah E. Angell,‡ Yan Zhang,‡ Cerrie W. Rogers,†,§ Michael O. Wolf,*,† and Wayne E. Jones, Jr.*,‡
Chemistry Department and Institute for Materials Research, State UniVersity of New York at
Binghamton, Binghamton, New York 13902, and Department of Chemistry, UniVersity of British
Columbia, VancouVer, British Columbia, Canada V6T 1Z1
Received March 23, 2005
Emission and absorbance spectra, along with low-temperature excited-state lifetimes, were obtained for the hemilabile
complexes, [Ru(bpy)2L](PF6)2 [L (2-methoxyphenyl)diphenylphosphine (RuPOMe) (1) and (2-ethoxyphenyl)-
)
diphenylphosphine (RuPOEt) (2)] in solid 4:1 ethanol/methanol solution. Spectral data were evaluated with ground-
state reduction potentials using Lever parameters. Lifetime data for these complexes were collected from 77 to 160
K, and the rate constant for the combined radiative and nonradiative decay process, k, the thermally activated
process prefactor, k
calculated from a plot of ln(1/
′
0, the rate constant for the MLCT
) versus 1/T for both (1) and (2). The low-temperature luminescence lifetimes of (1)
f d−d transition, k′, and the activation energy, ∆E′, were
τ
were observed to decrease with increases in water concentration. The photophysical and kinetic data of (1) and
(2) are compared to literature data for [Ru(bpy)3](PF6)2. The emission maxima of (1) and (2) are blue-shifted relative
to [Ru(bpy)3](PF6)2 due to the presence of the strong-field phosphine ligand, which enhances
the bipyridyl ligands. The thermal activation energy, , is significantly larger for [Ru(bpy)3](PF6)2 than for (1) and
(2) resulting in a faster MLCT d transition for (1) and (2). These results are discussed in the context of
radiationless decay through thermally activated ligand-field states on the metal complex.
π back-bonding to
∆E′
f d−
Introduction
nificant challenge in these systems is the design of new
molecules for continuous analyte monitoring that have a
reversible receptor unit as part of the chemosensor.
The use of chemosensors as analytical testing devices has
been the subject of much attention in the chemistry
literature.1-2 Molecular sensors that are both sensitive and
selective offer significant advantages in the design of portable
detection systems. These chemosensors have recently taken
the form of redox-active sensors,3-4 small-molecule fluoro-
phores,5-6 chromophoric sensors,7-12 conducting poly-
mers,13,14 and conjugated fluorescent polymers.15-16 A sig-
Hemilabile ligands have appeared in the recent literature
with applications in catalysis, small-molecule activation,
small molecule sensing, and stabilization of transition metal
complexes.17 Complexes containing hemilabile ligands,
which are polydentate chelates that contain both substitution-
ally inert and substitutionally labile groups, provide a
* Authors to whom correspondence should be addressed. E-mail:
wjones@binghamton.edu (W.E.J.); mwolf@chem.ubc.ca (M.O.W.).
‡ State University of New York at Binghamton.
(7) Absalan, G.; Soleimani, M.; Asadi, M.; Ahmadi, M. B. Anal. Sci. 2004,
20, 1433.
(8) Spadavecchia, J.; Ciccarella, G.; Vasapollo, G.; Siciliano, P.; Rella,
R. Sens. Actuators, B 2004, B100, 135.
(9) Parry, D. A.; Sallah, M. M.; Miller, L. S.; Peterson, I. R.; Hollyoak,
R. Supramol. Sci. 1997, 4, 427.
(10) Wang, B.; Wasielewski, M. R. J. Am. Chem. Soc. 1997, 119, 12.
(11) Chen, L. X.; Jager, W. J. H.; Gostola, D. J.; Niemczyk, M. P.;
Wasielewski, M. R. J. Phys. Chem. B, 2000, 104, 1950.
(12) Chen, L. X.; Jager, W. J. H.; Gostola, D. J.; Niemczyk, M. P.;
Wasielewski, M. R. Synthetic Metals. 2001, 116, 229.
(13) Chen, J.; Burrell, A. K.; Collis, G. E.; Officer, D. L.; Swiegers, G. F.;
Too, C. O.; Wallace, G. G. Electrochim. Acta. 2002, 47, 2715.
(14) Lewis, T. W.; Wallace, G. G.; Smyth, M. R. Analyst 1999, 124, 213.
(15) Swager, T. M. Acc. Chem. Res. 1998, 31, 201.
† University of British Columbia.
§ Current address: Department of Chemistry and Biochemistry, Con-
cordia University, Montreal, Quebec, Canada H4B 1R6.
(1) McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem ReV. 2000, 100,
2537.
(2) Martinez-Manez, R.; Sancenon, F. Chem ReV. 2003, 103, 4419.
(3) Cui, X.; Carapuca, H. M.; Delgado, R.; Drew, M. G. B.; Felix, V. J.
Chem. Soc., Dalton Trans. 2004, 11, 1743.
(4) Severin, K. Coord.Chem.ReV. 2003, 245, 3.
(5) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley, A.
J. M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. ReV. 1997,
97, 1515.
(6) de Silva, A. P.; Fox, D. B.; Huxley, A. J. M.; Moody, T. S. Coord.
Chem. ReV. 2000, 205, 41.
(16) Tan, C.; Atas, E.; Muller, J. G.; Pinto, M. R.; Kleiman, V. D.; Schanze,
K. S. J. Am. Chem. Soc. 2004, 126, 13685.
10.1021/ic050436l CCC: $30.25
Published on Web 09/14/2005
© 2005 American Chemical Society
Inorganic Chemistry, Vol. 44, No. 21, 2005 7377

Angell et al.
taken. For the photophysical studies, anhydrous ethanol and
methanol were used as received in Sure Seal bottles from Aldrich
or alternatively dried by treatment with calcium hydride followed
by distillation and storage under nitrogen. Glass equipment was
purged with nitrogen and flame-dried before use. All solutions of
(1) and (2) were made 3.0 × 10-5 M in a 4:1 ethanol/methanol
solvent system under nitrogen, unless otherwise noted.
potential site for the binding of analytes to a transition metal
center. Hard-soft acid-base interactions provide the basis
for this hemilabile behavior. Importantly, hemilabile ligands
have been shown to allow the reVersible binding of small
molecules to metal complexes because of their dynamic
chelating ability.17-18 The labile group binds weakly to the
transition metal center in the absence of small-molecule
substrates and is easily displaced in the presence of a small
molecule with a strong binding affinity for the metal center.
However, the labile group remains in close proximity to the
metal because of the inert ligand anchor. Recoordination to
the transition metal center may occur if the small molecule
dissociates.
Photophysical Measurements. Room-temperature UV/visible
absorption measurements were performed using a HP 8253A diode-
array spectrometer from 190 to 820 nm or a Perkin-Elmer double-
beam Lambda 2 spectrometer. Luminescence spectra were measured
on an SLM 48000S fluorimeter with an Oxford Instruments liquid-
nitrogen-cooled cryostat and were excited at the MLCT absorbance
maxima. Emission lifetime measurements were carried out in the
aforementioned cryostat. A nitrogen pulsed laser (Laser Photonics)
was used as the excitation source (337 nm). Time-resolved emission
was collected through a one-stage monochromator at 90° from the
incident excitation beam. The emission was monitored at 620 nm
for both (1) and (2). The data were collected from a Hamamatsu
R4220P photomultiplier tube on a Tektronix TDS544A transient
digitizer. For lifetime determinations, at least 250 waveforms were
acquired and averaged and then fit to an exponential decay function
using a nonlinear least-squares fitting routine available in Microsoft
Excel 2000. All samples used in photophysical data collection were
freeze-pump-thaw degassed for 4-5 cycles prior to measurement
in sealed NMR tubes or extended cryogenics cells (NSG Precision).
Data Analysis. Temperature-dependent lifetimes were fit to the
expression below with use of a nonlinear least-squares procedure.25
Previously, we have reported the syntheses of hemilabile-
ligand complexes [Ru(bpy)2L](PF6)2 [L ) (2-methoxyphen-
yl)diphenylphosphine (RuPOMe) (1) and (2-ethoxyphenyl)-
diphenylphosphine (RuPOEt) (2)] and their responses to the
binding of various small molecules through changes in both
the absorption and low-temperature (77 K) emission spec-
tra.19 We found that (1) shows concentration-dependent shifts
in room-temperature absorbance and low-temperature lumi-
nescence spectra because (1) reversibly binds to residual
1
τ(T)
) k + k′0 exp(-∆E′/RT)
(1)
where k ) kr + knr; kr and knr are the rate constants for the nominally
temperature-independent radiative and nonradiative decay processes,
respectively; k′0 is the thermally activated process prefactor with
activation energy ∆E′; k′ is equal to this temperature-dependent
term, k′0 exp(-∆E′/RT), and R is the ideal gas constant. The error
present in the calculated terms, k, k′0, and ∆E is estimated to be
10% and is due largely to the nonlinear least-squares fitting
procedure.
Electrochemistry. Electrochemical measurements were per-
formed with a Pine Instruments AFCBP1 bipotentiostat using a
3-electrode cell (Pt disk working electrode, Pt wire coil counter
electrode, Ag wire quasi-reference electrode) in CH2Cl2 solution
that contained ∼0.1 M n-Bu4NPF6 as supporting electrolyte. Either
decamethylferrocene (Me10Fc) or ferrocene (Fc) was added to
samples as an internal standard in order to quote reduction potentials
versus the saturated calomel electrode, SCE (Me10Fc, E1/2 ) -0.120
V vs SCE; Fc, E1/2 ) 0.454 V vs SCE). Methylene chloride was
distilled from calcium hydride immediately before use in electro-
chemical experiments. n-Bu4NPF6 was recrystallized three times
from methanol, dried in vacuo at 110 °C for 3 days, and stored in
a desiccator.
water in the solvent to form an aquo complex, (1‚H2O).20
A
concerted mechanism for water substitution of (1) has been
discussed. To our knowledge, this is the first example of a
metal-to-ligand charge transfer (MLCT)-based sensor de-
signed to take advantage of the reversible binding provided
by a hemilabile ligand. As previously reported, MLCT
excited states are ideal in chemosensors because of their long
luminescence lifetime and proclivity to both electron- and
energy transfer-quenching processes.21-23 Here we report the
photophysical characterization of hemilabile complexes (1)
and (2).
Experimental Section
Materials. All chemicals were used as received unless otherwise
specified. RuPOMe (1) and RuPOEt (2) were synthesized as
described in the literature.24 RuPOEt was recrystallized in an
ethanol/methanol mixture before photophysical measurements were
(17) Slone, C. S.; Weinberger, D. A.; Mirkin, C. A. The Transition Metal
Coordination of Hemilabile Ligands. In Progress in Inorganic
Chemistry; Karlin, K. D., Ed.; John Wiley and Sons: New York, 1999;
Vol. 48, pp 233-350.
(18) Bader, A.; Linder, E. Coord. Chem. ReV. 1991, 108, 27.
(19) Rogers, C. W.; Wolf, M. O. Chem. Commun. 1999, 2297.
(20) Rogers, C. W.; Zhang, Y.; Patrick, B. O.; Jones, W. E., Jr.; Wolf, M.
O. Inorg. Chem. 2002, 41, 1162.
Results and Discussion
The photophysics and photochemistry of transition metal
complexes with MLCT excited states have been studied
extensively.26-31 Emission from an excited Ru(II) polypyridyl
(21) Kalyanasundaram, K.; Gratzel, M. Coord. Chem. ReV. 1998, 177, 347.
(22) de Silva, A. P.; Fox, D. B.; Huxley, A. J. M.; McClenaghan, N. D.;
Roiron, J. Coord. Chem. ReV. 1999, 186, 297.
(23) Jiang, B.; Yang, S. W.; Bailey, S. L.; Hermans, L. G.; Niver, R. A.;
Bolcar, M. A.; Jones, W. E., Jr. Coord. Chem. Rev. 1998, 171, 365.
(24) Rogers, C. W.; Patrick, B. O.; Rettig, S. J.; Wolf, M. O. J. Chem.
Soc., Dalton Trans. 2001, 1278.
(25) Caspar, J. V.; Meyer, T. J. Inorg. Chem. 1983, 22, 2444.
(26) Crosby, G. A.; Perkins, W. G.; Klassen, D. M. J. Chem. Phys. 1965,
43, 1498.
7378 Inorganic Chemistry, Vol. 44, No. 21, 2005

Photophysical Properties of Ru(II) Bipyridyl Complexes
Figure 2. Corrected absorption (a) (298 K) and emission (b and c) (77
K) spectra of 3.0 × 10-5 M RuPOMe (1) in 4:1 ethanol/methanol. The
emission maximum shifts depending on the relative amount of water in the
sample; (b) is a drier sample, (c) is a wetter sample due to fluctuation in
the relative amount of atmospheric humidity.
Figure 1. Energy-state diagram for (1) and (2) showing the possibility of
a thermally populated d-d state that may be tuned by variations in the
ligand-field, where ∆E′ is the activation energy barrier.
Absorption and Emission Spectra. RuPOMe (1) has
previously been shown to have analyte-dependent absorption
and emission behavior.19,20 The absorption spectrum at 298
K and two emission spectra at 77 K of complex (1) are shown
in Figure 2. Two sets of absorbance bands were observed
for (1). The lower-energy band (λmax ) 454 nm, ꢀ ) 4.6 ×
103 cm-1 M-1), labeled (a), is assigned to a spin-allowed
MLCT electronic transition labeled dπ(Ru) f π*(ligand),
wherein an electron is promoted from the HOMO, a
ruthenium centered orbital, into a ligand-centered π* orbital
by comparison with other ruthenium bipyridyl compounds.39
The higher-energy absorbance bands, not shown, are due to
ligand bipyridyl π f π* transitions (λmax ) 224 nm, ꢀ )
3.2 × 104 cm-1 M-1 and 292 nm, ꢀ ) 3.8 × 104 cm-1 M-1).
A small band, not pictured, at 334 nm (ꢀ ) 4.6 × 103 cm-1
M-1) is also observed.
Even though careful measures are taken to dry the solvent
used in these measurements, trace water is inevitably present
in a 4:1 ethanol/methanol solvent system. This results in an
equilibrium between RuPOMe (1) and a water-substituted
form of RuPOMe at the labile position. This aquo complex
is designated (1‚H2O). Evaluating the absorbance and
emission spectra involves determining whether the transition
originates from the hemilabile complex itself or the aquo-
substituted complex. The low-energy MLCT absorbance
band is attributed to (1‚H2O) by comparison with previous
work, which cites that at low concentrations (1‚H2O)
predominates.20 In more concentrated solutions, (1) absorbs
at ∼420 nm.20 This assignment is confirmed with a water
titration (Figure 3) of 2.0 × 10-4 M RuPOMe in 2:1 ethanol/
acetone monitored via absorbance spectra. Titration from 0%
to 5% (v/v) added H2O results in an 8 nm red-shift of the
absorbance maximum.
complex, such as [Ru(bpy)3](PF6)2 [(bpy) ) 2,2′- bipyridine],
commonly occurs from a triplet metal-to-ligand charge-
transfer (3MLCT) excited state.32-33 Such a 3MLCT state can
also undergo internal conversion by thermal deactivation to
a triplet, metal-centered ligand-field (d-d) state with rate,
k′ ) k2 and activation energy barrier, ∆E′ (Figure 1). These
low-energy states do not usually appear in the absorbance
or emission spectra because of very low extinction coef-
ficients or low emission quantum yields. However, these
states impact the photophysical and photochemical properties
of the complexes because they provide an efficient pathway
for nonradiative decay.
At ambient temperatures, the thermally populated d-d
states of (1) and (2) provide a very efficient mechanism for
nonradiative decay. Temperature-dependent lifetime mea-
surements are used to obtain information regarding this
metal-centered excited state, since thermal population of this
state leads to shorter luminescence lifetimes. Temperature-
dependent emission has been observed previously with
complexes of the type [Ru(bpy)2L2](PF6)2.25,32 The energy
of the ligand-field state, and likewise the rate and thermal
activation energy barrier of the MLCT f d-d transition,
may be tuned through variations in the ligand, L.25,34-38
Phosphine ligands, such as those in (1) and (2), are known
to increase the thermal accessibility of d-d states to the
extent that luminescence from phosphine complexes is
typically not measurable at room temperature.25
(27) Harrigan, R. W.; Crosby, G. A. J. Chem. Phys. 1973, 59, 3468.
(28) Felix, F.; Ferguson, J.; Gudel, H. C.; Ludi, A. J. Am. Chem. Soc. 1980,
102, 4096.
(29) Boletta, F.; Juris, A.; Maestri, M.; Sandrini, D. Inorg. Chem. Acta
1980, 44, L175.
(30) Demas, J. N.; Taylor, D. G. Inorg. Chem. 1979, 18, 3177.
(31) Van Houten, J.; Watts, R. J. J. Am. Chem. Soc. 1976, 98, 4853.
(32) ] Barigelletti, F.; Belser, P.; von Zelewsky, A.; Juris, A.; Balzani, V.
J. Phys. Chem. 1985, 89, 3680.
(33) Crosby, G. A. Acc. Chem. Res. 1975, 8, 231.
(34) Kinnunen, T. J. J.; Haukka, M.; Pesonen, E.; Pakkanen, T. J.
Organomet. Chem. 2002, 655, 31.
(35) Balzani, V.; Juris, A.; Barigelletti, F.; Belser, P.; Von Zelewsky, A.
Sci. Pap. Inst. Phys. Chem. Res. (Jpn.) 1984, 78, 78.
(36) Lumpkin, R. S.; Kober, E. M.; Worl, L. A.; Murtaza, Z.; Meyer, T. J.
J. Phys. Chem. 1990, 94, 239.
The highest-energy low-temperature emission band, seen
in Figure 2, of (1) at 77 K is observed at λ ) 558 nm (17.9
× 103 cm-1) and is labeled (b). There is a broad shoulder, λ
) 616 nm (16.2 × 103 cm-1), on the lower-energy side of
this band. While a low-energy shoulder is commonly
(38) Allen, G. H.; White, R. P.; Rillema, D. P.; Meyer, T. J. J. Am. Chem.
Soc. 1984, 106, 2613.
(37) Anderson, P. A.; Keene, F. R.; Meyer, T. J.; Moss, J. A.; Strouse, G.
F.; Treadway, J. A. J. Chem. Soc., Dalton Trans. 2002, 3820.
(39) Roundhill, D. M. Photochemistry and Photophysics of Metal Com-
plexes; Plenum Press: New York, 1994.
Inorganic Chemistry, Vol. 44, No. 21, 2005 7379

Angell et al.
Figure 4. Corrected absorption (a) (298 K) and emission (b and c) (77
K) spectra of 3.0 × 10-5 M RuPOEt (2) in 4:1 ethanol/methanol. The
emission maximum shifts depending on the relative amount of water in the
sample; (b) is a drier sample, (c) is a wetter sample due to fluctuation in
the relative amount of atmospheric humidity.
Figure 3. Absorption spectra of (1) (2 × 10-4 M) in 2:1 ethanol/acetone
with different volumes of water added.
Table 1. Lever Parameters (EL)48 for Selected Ligands Listed as
Potentials in Volts vs NHE ) -0.240 V vs SCE
observed for compounds of the type [Ru(bpy)2L2](PF6)2
containing an MLCT transition and has been assigned to
the V′ ) 0 f V ) 1 band,25 in this case, the shoulder is
also due to the presence of the aquo complex, (1‚H2O).
This assignment is made by analogy with previous work,
which cites that at low concentrations (1‚H2O) predomi-
nates, giving rise to an emission band that is red-shifted
compared to more-concentrated solutions of the complex
itself.20 This is confirmed by another 77 K emission
spectrum, with a higher relative concentration of water
present in solution prior to cooling, labeled (c). This spec-
trum has a dominant peak at 616 nm (1) and a shoulder at
558 nm (1‚H2O). This assignment is also supported by
low-temperature excitation spectra monitored at different
wavelengths. When the excitation is monitored at 558
nm, an excitation maximum is observed at 412 nm con-
sistent with the presence of the coordinated hemilabile
complex, (1). However, when the excitation is monitored at
616 nm, an excitation maximum is observed at 470 nm,
consistent with the absorbance of the aquo complex. The
absorbance peak that we observe at 454 nm is a convolution
of both (1) and (1‚H2O), as demonstrated by the excitation
spectra.
[Ru(bpy)3](PF6)2 has a long-lived excited state that visibly
luminesces at ambient temperatures.25 However, the excited-
state lifetimes of RuPOMe (1) and RuPOEt (2) are limited
by the presence of a thermally accessible d-d state, which
provides a very efficient pathway of nonradiative excited-
state decay. Thus, at ambient temperatures, no luminescence
is observed. At 77 K, luminescence is observed as thermal
population of the nonradiative d-d state is effectively
reduced due to the activation of the energy barrier. Emission
is observed for the transition from the ligand π* orbital back
to the ruthenium ground-state HOMO.
ligand
EL/V
ligand
EL/V
Cl-
H2O
Ph3P
-0.24
0.04
0.39
Me2O
POMe
bpy
0.45
0.63a
0.259
a Calculated from E1/2(RuIII/II
)
) 0.97[∑EL] + 0.04 47 using RuCl2-
calc
(POMe-P,O)2, E1/2(RuIII/II) ) 0.80 V vs NHE.49
Two emissive states (Figure 4b and c) are also observed
for RuPOEt (2) at 77 K depending on the relative concentra-
tion of water present in solution prior to cooling. Both the
MLCT emission band of (2) (λmax ) 584 nm) and the MLCT
emission band of the aquo-substituted RuPOEt (2‚H2O) (λmax
) 622 nm) are slightly red-shifted relative to their RuPOMe
counterparts.
Calculation of MLCT Absorption Energies. Ground-
state reduction potentials can be used to generate information
about the luminescent excited state.40,41 The energy of the
MLCT absorption, which corresponds to RuII(bpy) f
RuIII(bpy•-), scales linearly with the difference between the
42
RuIII/II and bpy0/- redox couples, ∆E(redox)
;
the same holds
true for the emission energies. These relationships have been
used as confirmation that, within families of Ru(II) bpy
complexes, the absorption and emission arise from analogous
MLCT and LMCT processes. Excellent discussions of these
relationships are available in reviews by Lever.43,44
The contributions of ligands to the reduction potentials of
complexes can be approximated through the use of Lever
parameters (EL). Lever parameters, derived from experimen-
tal electrochemical data47 and also recently from computa-
tional studies,48 are available in the literature for hundreds
of ligands.
It should be noted that Lever parameters are defined using
data from complexes that exhibit reversible redox chemistry
The absorption spectrum at 298 K and two emission
spectra at 77 K of complex (2) are shown in Figure 4. The
energy of the dπ(Ru) f π* (ligand) MLCT transition (λmax
) 446 nm, ꢀ ) 9.8 × 103 cm-1 M-1) of the aquo-substituted
RuPOEt (2‚H2O), is comparable to (1) (Figure 4a). As
expected, the ligand bipyridyl π f π* transitions are similar
to (1) (λmax ) 230 nm, ꢀ ) 4.2 × 104 cm-1 M-1 and 292
nm, ꢀ ) 5.1 × 104 cm-1 M-1, not shown).
(40) Kalyanasundaram, K. Coord. Chem. ReV. 1982, 46, 159.
(41) Kalyanasundaram, K. Photochemistry of Polypyridine and Porphyrin
Complexes; Academic Press: London, 1992.
(42) Dodsworth, E. S.; Lever, A. B. P. Chem. Phys. Lett. 1986, 124, 152.
(43) Lever, A. B. P.; Dodsworth, E. S. Electrochemistry, Charge-Transfer
Spectroscopy, and Electronic Structure. In Inorganic Electronic
Structure And Spectroscopy; Solomon, E. I., Lever, A. B. P., Ed.; John
Wiley and Sons: New York, 1999; Vol. II: Applications and Case
Studies, pp 227-289.
(44) Lever, A. B. P. Can. J. Anal. Sci. Spec. 1997, 42, 22.
7380 Inorganic Chemistry, Vol. 44, No. 21, 2005

Photophysical Properties of Ru(II) Bipyridyl Complexes
Table 2. Reduction Potentials and Transition Energies for (1) and Sensor-Analyte Complex, (1‚H2O)
observed (V vs SCE)
predicted
1/2(RuIII/II
)
calc
E
1/2(bpy0/-
)
calc
∆E(redox)calc
E
abs(MLCT)calc
Eem(0-0)calc
a
b
c
d
e
complex
Eox
Ered
∆E(redox)
E
1
1.56f
-
-1.27f
-
2.83
-
2.58
1.42
1.23
-1.35
-1.40
2.77
2.63
2.61h
2.98
2.84
2.74h
2.25
2.14
2.12h
1‚H2O
Ru(bpy)3
1.23g
-1.35g
2+
a E1/2(RuIII/II
)
) 0.97[∑EL] + 0.04.47 E1/2(bpy0/-
)
) 0.25((0.01)[∑EL(bpy)] - 1.40((0.03).49 c ∆E(redox)calc ) E1/2(RuIII/II
)
calc
- E1/2(bpy0/-
)
calc
.
b
calc
calc
d Eabs(MLCT) ) 1.00∆E(redox) + 0.21.45 Eem(0-0) ) 0.76((0.06)∆E(redox) + 0.14((0.04).46 f CH2Cl2, 0.1 M [n-Bu4N]PF6, 25 °C. g CH3CN, 0.1 M
e
[n-Et4N]PF6, 25 °C.53 h These values were calculated from emission and absorbance data.
and do not undergo ligand exchange during electrochemical
experiments; thus, application to systems that are not
similarly well-behaved must be done with caution.
Changes in the coordination sphere of (1) brought about
by displacement of the labile ether by water are expected to
impact the redox properties of (1), particularly the potential
at which the metal undergoes oxidation to Ru(III). In turn,
these changes are expected to correlate with the differences
in MLCT absorption and emission properties induced by the
binding of analytes. Using Lever parameters (Table 1) to
estimate the reduction potentials of the complexes, values
of ∆E(redox) were calculated for (1) and (1‚H2O). The solution
electrochemistry of the sensor complex was examined via
Figure 5. Differences in the ligand-field splitting parameter of
cyclic voltammetry for (1) to compare to the predicted value.
Ru(bpy)32+ and (1) and (2) and its subsequent effect on the activation energy
The experimentally observed and predicted values of ∆E(redox)
are listed in Table 2. From these ∆E(redox) values, the MLCT
absorption and LMCT emission energies for (1) and (1‚H2O),
were predicted; these are presented in Table 2.
of a MLCT f d-d transition.
) 579 nm; Eem(0-0)calc ) 2.14 eV). These values may be
compared to the experimental value for [Ru(bpy)3]2+ (λem
) 602 nm; Eem ) 2.06 eV).39
As cited in the previous section, the MLCT absorption
(λmax ) 420 nm; Eabs ) 2.95 eV) energy for (1) is higher
than that for [Ru(bpy)3]2+ (λmax ) 452 nm; Eabs ) 2.74 eV).50
The predicted MLCT transition energy based solely on Lever
parameters (Eabs(MLCT)calc ) 2.98 eV (λem ) 416 nm), using
EL ) 0.63 for POMe) is in close approximation of the
experimental value. The calculated absorbance energy for
(1‚H2O) based solely on Lever parameters (λmax ) 437 nm;
Eabs ) 2.84 eV) is in reasonable agreement with the
experimentally observed value reported in this paper for the
wet RuPOMe solution (λmax ) 454 nm; Eabs ) 2.73 eV).
These calculations support the assignment of the absorbance
band in Figure 2a to the aquo complex, as the experimental
value is closer to the calculated absorbance value for (1‚
H2O) than (1).
The highest-energy emission band observed for (1) (λem
) 558 nm; Eem ) 2.22 eV) corresponds in energy to the
value of Eem(0-0) predicted from Lever parameters
(λem(0-0)calc ) 552 nm; Eem(0-0)calc ) 2.25 eV). The
subsequent band (λem ) 616 nm; Eem ) 2.01 eV), corre-
sponding to (1‚H2O), is lower in energy than the value of
Eem(0-0) predicted from the Lever parameters (λem(0-0)calc
The Lever parameters confirm that (1) and (1‚H2O) behave
in an electronically similar way to other emissive Ru(II) bpy
complexes. The calculations predict that (1‚H2O) will red-
shift relative to (1) in both the absorbance and emission
spectra, and these red-shifts are observed experimentally even
if the magnitude of the shift differs from the calculated value
in some cases. This provides additional confirmation for the
peak assignments made in the previous section. It is
interesting to note that the calculated emission maxima do
not correlate as closely with the measured values as the
absorption maxima. It is likely that this difference in energy
is the result of distortions in the excited state and any
associated reorganizational energies associated with the
hemilabile ligand complexes. These differences would not
be included in the Lever parameters generated on the basis
of [Ru(bpy)3](PF6)2 analogues.
As predicted through Lever calculations and observed in
the emission spectra, the MLCT band of RuPOMe (1) is at
a higher energy than the MLCT emission band of [Ru(bpy)3]-
(PF6)2 (602 nm (16.6 × 103 cm-1)).39,51 This can be attributed
to the presence of the strong-field phosphine ligand, which
enhances π back-bonding to the bipyridyl ligands. This
strong-field ligand stabilizes the metal-based HOMO, result-
ing in a larger-energy MLCT transition (Figure 5). For
RuPOEt (2), the high-energy MLCT band is also consistent
with the enhanced π back-bonding effect due to stabilization
of the metal-based HOMO by the phosphine.
(45) Dodsworth, E. S.; Lever, A. B. P. Chem. Phys. Lett. 1986, 124, 152.
(46) Vlcek, A. A.; Dodsworth, E. S.; Pietro, W. J.; Lever, A. B. P. Inorg.
Chem. 1995, 34, 1906.
(47) Lever, A. B. P. Inorg. Chem. 1990, 29, 1271.
(48) Perrin, L.; Clot, E.; Eisenstein, O.; Loch, J.; Crabtree, R. H. Inorg.
Chem. 2001, 40, 5806.
(49) Rogers, Cerrie. A Hemilabile Ligand Approach to Ruthenium-based
Luminescent Molecular Sensors, Ph.D. thesis, University of British
Columbia, Vancouver, British Columbia, 2001.
(51) Tokel-Takvoryan, N. E.; Hemingway, R. E.; Bard, A. J. J. Am. Chem.
Soc. 1973, 95, 6582.
(50) Crutchley, R. J.; Lever, A. B. P. Inorg. Chem. 1982, 21, 2276.
Inorganic Chemistry, Vol. 44, No. 21, 2005 7381

Angell et al.
Table 3. Lifetimes of (1) and (2) (average) Measured for 3.0 × 10-5
M Solutions in 4:1 Ethanol/Methanol at 77, 120 (glass-fluid transition
temperature), and 160 K
τ(µs), 77 K
τ(µs), 120 K
τ(µs), 160 K
2+
Ru(bpy)3
5.21 ( 0.06 52
4.19 ( 0.05
3.24 ( 0.05
RuPOMe (1)
RuPOEt(2)
3.59 ( 0.05
2.61 ( 0.05
0.85 ( 0.05
0.98 ( 0.05
tion and causing destabilization of the ruthenium HOMO.
This lowers the energy of the MLCT emission band (Figure
6). Thus, the aquo complex (1‚H2O) emission band appears
at longer wavelengths than the emission of the parent P,O-
complex (1).
Lifetimes. Excited-state lifetimes were determined by
time-resolved emission measurements following laser excita-
tion at 337 nm. The data were fit to a single-exponential
decay, assuming unimolecular kinetics, where [I] is the
intensity of luminescence at time, t, and [I]t ) [I]0 × e-kt
(Figure 7). The lifetime was calculated by taking the inverse
of the rate constant. The luminescence lifetimes of (1) and
(2) at 77, 120 (glass-fluid transition temperature), and 160
K (highest temperature at which luminescence is observed)
are displayed in Table 3. As the temperature increases, a
decrease in the lifetime of luminescence is observed for both
(1) and (2). This result is consistent with thermal population
of the nonradiative d-d states. The lifetime of the MLCT
emission becomes shorter as the rate of energy transfer to
the d-d state increases.
The lifetime dependence of (2) at 77 K as a function of
the percentage of water present is shown in Figure 8. As
expected, the lifetime decreases with an increasing percentage
of water present, eventually reaching saturation at ∼0.12 M
H2O. When water binds to the ruthenium center of the
hemilabile complexes, the phosphine-ether ligand becomes
a monodentate phosphine. The increased molecular flexibility
appears to result in an increased manifold of d-d states
capable of coupling to the solvent (Figure 6). The rate of
vibrational relaxation from the d-d state is therefore more
Figure 6. Differences in the ligand-field splitting parameter of (1) and
(1‚H2O) due to changed overlap of the Ru-phosphine dπ* overlap, resulting
in less π back-bonding.
When the labile ether functional group is displaced by
water, the energy of the MLCT emission band is red-shifted
∼50 nm in both (1) and (2). The Lever parameters similarly
predict that (1‚H2O) will be red-shifted by ∼30 nm compared
to (1). This qualitative agreement suggests that the Lever
parameter used for POMe is reasonable; however, it is also
possible that the change from a bidentate to a monodentate
binding mode for the phosphine-ether causes additional
electronic changes not taken into account by Lever parameter
calculations. It is likely that as the bidentate P,O-chelate is
opened with the binding of water, the metal-ligand orbital
overlap of the ruthenium-phosphine dπ* state is compro-
mised. This change in overlap affects the electron density at
the ruthenium center, decreasing the π back-bonding interac-
Figure 7. Lifetime decay plot of RuPOEt (2) at 77 K fit to a single-exponential decay function; y ) 0.025 e-4.0E-5×t. The residuals display the deviation
of the experimental decay from the mathematically calculated exponential decay function.
7382 Inorganic Chemistry, Vol. 44, No. 21, 2005

Photophysical Properties of Ru(II) Bipyridyl Complexes
Table 4. Excited State Decay Parameters for (1) and (2)a
complex
concn, M
k, s-1
k′ 0, s-1
∆E′, cm-1
3.6 × 103
1.4 × 103
1.1 × 103
k′, s-1 (298 K)
1.5 × 106
3.7 × 108
1.2 × 108
k′, s-1 (77 K)
5.8 × 10-16
[Ru(bpy)3](PF6)2
-
4.1 × 105
2.5 × 105
2.5 × 105
4.5 × 1013
3.0 × 1011
3.0 × 1010
53
in CH2Cl2
RuPOMe(1) in 4:1
ethanol/ methanol
RuPOEt (2) in 4:1
ethanol/ methanol
3.0 × 10-5
3.0 × 10-5
1.6
1.5 × 101
a The values were obtained by fitting the lifetime to the expression in eq 1: 1/τ(T) ) k + k′0 exp(-∆E′/RT).
Figure 8. Lifetime dependence of RuPOEt (2) (3.0 × 10-5 M) on water
Figure 10. Lifetime temperature dependence of RuPOEt (2). Two separate
data sets are combined and normalized at one point. The solid curve
represents the calculated fit of the data with use of the expression for τ(T)
given in eq 1, y ) ln(2.5 × 105 + 3.0 × 1010 × e-1650x); SSR ) 0.74.
concentration at 77 K. The connecting line is shown to aid the eye.
in no emission at ambient temperatures. This increase in the
rate of conversion from an MLCT state to a d-d state is
dramatically observed in the differences in the rate constant,
k′, the temperature-dependent term that quantifies the rate
of the MLCT f d-d transition. Caspar and Meyer first
suggested that k′ ) k2 for [Ru(bpy)2L2](PF6)2 systems because
k3 is much greater than k-2.25 Thus, the rate of the MLCT
f d-d transition is 2 orders of magnitude greater for (1)
and (2) than for [Ru(bpy)2](PF6)2 at room temperature. At
77 K, the rate of this transition is 15-16 orders of mag-
nitude greater for (1) and (2) than for [Ru(bpy)2](PF6)2.
The experimentally observed rate constant, k′, is k′ ) k′0
exp(-∆E′/RT), where R is the ideal gas constant.25
Figure 9. Lifetime temperature dependence of RuPOMe (1). The solid
curve represents the calculated fit of the data with use of the expression
for τ(T) given in eq 1, y ) ln(2.5 × 105 + 3.0 × 1011 × e-2000.209x); SSR
) 0.143.
The value of the temperature-dependent term, k′, in eq 1
may be attributed to the ligand-field splitting parameter of
the ruthenium complex. As shown in Figure 5, the phos-
phine-ether ligand serves to stabilize the metal HOMO,
relative to a third bpy, and therefore decreases the thermal
activation required to reach the intersection region. This
results in a direct enhancement of the MLCT f d-d
transition rate, consistent with the observed change in k′ and
the pronounced temperature dependence of the luminescence.
rapid in the aquo complexes. This results in a more accessible
nonradiative d-d state and subsequently shortens the ob-
served lifetime of luminescence from the MLCT state.
Temperature-Dependent Lifetime Measurements. Cal-
culating the fluorescence lifetimes at different temperatures
allows for the determination of several kinetic factors. A plot
of ln(1/τ) for (1) as a function of the inverse temperature is
depicted in Figure 9. Fitting the three variables in eq 1 to
the experimental data yields the values shown in Table 4.
The experimentally observed rate constant, k′, is equal to
k′0exp(-∆E′/RT), where R is the ideal gas constant.25 A plot
of ln(1/τ) vs 1/T for (2) is shown in Figure 10. Here two
separate data sets are combined by normalization at one
point. The resulting kinetic data are in Table 4.
Conclusion
The photophysical and kinetic data of (1) and (2) show
that the replacement of a 2,2′-bipyridine ligand with a
phosphine-ether chelate stabilizes the metal HOMO d-state
and increases the MLCT gap. As a result, an emission peak
for (1) and (2) that is blue-shifted relative to [Ru(bpy)3](PF6)2
is observed. The larger MLCT gap results in a smaller value
for the thermal activation energy barrier between the MLCT
As expected, the MLCT f d-d activation energy barrier,
∆E′, is greater for [Ru(bpy)3](PF6)2 than for (1) or (2). The
efficient pathway of nonradiative decay through the d-d state
and the smaller activation energy barrier for (1) and (2) result
Inorganic Chemistry, Vol. 44, No. 21, 2005 7383

Angell et al.
and d-d state (∆E′) for (1) and (2) than observed for
[Ru(bpy)3](PF6)2. This accounts for the lack of measurable
luminescence of (1) and (2) at room temperature and also
explains the severalfold increase in the nonradiative decay
rate, k′, of (1) and (2) relative to [Ru(bpy)3](PF6)2. When
water is substituted for the labile ligand, π back-bonding is
weakened at the phosphorus due to a decrease in angular
orbital overlap. Thus, the ruthenium HOMO is destabilized,
resulting in red-shifted absorbance and emission maxima.
The increased degrees of freedom in the aquo complexes
results in both a broadened manifold of d-d states and faster
vibrational relaxation from the d-d state, with subsequently
shorter luminescence lifetimes from the MLCT states of the
aquo complexes.
This system shows promise as a chemosensor on the basis
of the reported photochemical changes caused by the binding
of water. These results successfully demonstrate that the rate
of nonradiative decay from the ligand-field d-d state in
hemilabile ligand complexes can be modified through
substitution at the labile portion of the bidentate ligand. We
describe this as “tunable ligand-field deactivation.” It has
also been shown that changing the chemical identity of the
labile portion of the chelating ligand changes the rate of
internal conversion.
In the future, we plan to integrate these hemilabile
complexes into a polymeric system for increased enhance-
ment in sensitivity.54-57 We also plan to test the sensitivity
of (1) and (2) toward other small molecules.
Acknowledgment. We thank Szu-Wei Yang of Bing-
hamton University for instrumental support and Dr. Brendan
Flynn, Dr. Cliff Timpson, and Dr. Stanley K. Madan for
useful discussions. S.E.A., Y.Z., and W.E.J. acknowledge
the NSF GK-12 program and NIH (1R15ES10106) for
funding. M.O.W. and C.W.R. acknowledge NSERC of
Canada for funding.
Supporting Information Available: Excitation spectra for (1).
This material is available free of charge via the Internet at
http://pubs.acs.org.
IC050436L
(52) Watts, R. J.; Crosby, G. A. J. Am. Chem. Soc. 1972, 94, 2606.
(53) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583.
(54) Zhang, Y.; Murphy, C. B.; Jones, W. E., Jr. Macromolecules 2002,
35, 630.
(55) Murphy, C. B.; Zhang, Y.; Troxler, T.; Ferry, V.; Martin, J. J.; Jones,
W. E., Jr. J. Phys. Chem. B. 2004, 108, 1537.
(56) Weinberger, D. A.; Higgins, T. B.; Mirkin, C. A.; Stern, C. L.; Liable-
Sands, L. M.; Rheingold, A. L. J. Am. Chem. Soc. 2001, 123, 2503.
(57) Allgeier, A. M.; Mirkin, C. A. Organometallics 1997, 16, 3071.
7384 Inorganic Chemistry, Vol. 44, No. 21, 2005