Kinetic study of the photo-induced electron transfer reaction between ruthenium(II) complexes of 2,2′-bipyridine derivatives and methyl viologen. Effects of bulky substituents introduced onto 2,2′-bipyridine

Article abstract of DOI:10.1039/b207048k

Ruthenium(II) complexes of 2,2′-bipyridine derivatives. [Ru(Rbpy)₃]²⁺ (Rbpy = 4,4′-di(alkylaminocarbonyl)-2,2′-bipyridine; alkyl = propyl, hexyl or adamantyl), have been newly synthesized and their photo-induced electron transfer (ET) reaction with methyl viologen (MV²⁺) has been investigated. The rate constant of the ET reaction decreases in the order alkyl = propyl > hexyl > adamantyl, which is opposite to the increasing order of the size of the alkylaminocarbonyl substituent introduced onto 2,2′-bipyridine. After correction of the diffusion rate, the ET reaction in the exciplex. [*Ru(Rbpy)₃²⁺ ... MV²⁺] → [Ru(Rbpy)₃³⁺ ... MV⁺], was analyzed on the basis of Marcus' theory. The electronic coupling matrix element (H_rp) decreases in the order propyl (2.28 × 10^-3 eV) > hexyl (1.86 × 10^-3 eV) > adamantyl (1.37 × 10^-3 eV), while the reorganization energy (λ) depends little on the alkyl group of Rbpy; λ = 0.772 eV, 0.767 eV and 0.798 eV for R = propyl, hexyl and adamantyl, respectively. Thus, not the λ value but the H_rp value is responsible for the above-mentioned decreasing order of the rate constant. This means that the bulky substituent decreases the orbital overlap between donor and acceptor to suppress the ET reaction. The H_rp value exponentially decreases with increasing electron transfer distance (r_AB), as follows: H_rp = H°_rp exp[-β(r_AB - r°_AB)/2] with β = 11 nm^-1, where H°_rp is the electronic coupling matrix element when r_AB is the closest distance (r°_AB) for effective contact between donor and acceptor. This β value is almost the same as the value (12 nm^-1) reported for the thermal ET reaction between aromatic compounds.

Full text of DOI:10.1039/b207048k

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetic study of the photo-induced electron transfer reaction
between ruthenium(II) complexes of 2,2Ј-bipyridine derivatives and
methyl viologen. Effects of bulky substituents introduced onto
2,2Ј-bipyridine†
Taisuke Hamada,*a Sei-ichi Tanaka,a Hiroaki Koga,a Yumi Sakaia and Shigeyoshi Sakaki*b
a Department of Applied Chemistry and Biochemistry, Faculty of Engineering,
Kumamoto University, Kurokami, Kumamoto 860-8555, Japan
b Department of Molecular Engineering, Graduate School of Engineering, Kyoto University,
Kyoto 606-8501, Japan
Received 18th July 2002, Accepted 13th December 2002
First published as an Advance Article on the web 22nd January 2003
Ruthenium() complexes of 2,2Ј-bipyridine derivatives, [Ru(Rbpy)3]2ϩ (Rbpy = 4,4Ј-di(alkylaminocarbonyl)-2,2Ј-
bipyridine; alkyl = propyl, hexyl or adamantyl), have been newly synthesized and their photo-induced electron
transfer (ET) reaction with methyl viologen (MV2ϩ) has been investigated. The rate constant of the ET reaction
decreases in the order alkyl = propyl > hexyl > adamantyl, which is opposite to the increasing order of the size of
the alkylaminocarbonyl substituent introduced onto 2,2Ј-bipyridine. After correction of the diffusion rate, the ET
2ϩ
3ϩ
reaction in the exciplex, [*Ru(Rbpy)3 ؒ ؒ ؒ MV2ϩ
]
[Ru(Rbpy)3 ؒ ؒ ؒ MV ϩ], was analyzed on the basis of
ؒ
Marcus’ theory. The electronic coupling matrix element (Hrp) decreases in the order propyl (2.28 × 10Ϫ3 eV) > hexyl
(1.86 × 10Ϫ3 eV) > adamantyl (1.37 × 10Ϫ3 eV), while the reorganization energy (λ) depends little on the alkyl group
of Rbpy; λ = 0.772 eV, 0.767 eV and 0.798 eV for R = propyl, hexyl and adamantyl, respectively. Thus, not the λ
value but the Hrp value is responsible for the above-mentioned decreasing order of the rate constant. This means
that the bulky substituent decreases the orbital overlap between donor and acceptor to suppress the ET reaction.
The Hrp value exponentially decreases with increasing electron transfer distance (rAB), as follows: Hrp = HrpЊ
exp[Ϫβ(rAB Ϫ rABЊ)/2] with β = 11 nmϪ1, where HrpЊ is the electronic coupling matrix element when rAB is the
closest distance (rABЊ) for effective contact between donor and acceptor. This β value is almost the same as the
value (12 nmϪ1) reported for the thermal ET reaction between aromatic compounds.
The Hrp value is considered to decrease exponentially with
Introduction
increasing donor–acceptor separation (rAB), as shown by the
Photo-induced electron transfer (ET) reactions form one
of the important subjects of research over the last two decades.1
Since ruthenium() tris(2,2Ј-bipyridine), [Ru(bpy)3]2ϩ, was
successfully utilized for H2 evolution from water,2–5 many
studies have been carried out to elucidate the photochemical
following eqn. 2.8a
Hrp = HrpЊ exp[Ϫβ(rAB Ϫ rABЊ)/2]
(2)
In this equation, HrpЊ is the electronic coupling matrix element
when rAB is the closest distance (rABЊ) for effective contact
between donor and acceptor. The orbital parameter β is con-
sidered a measure of orbital expansion of the electron transfer
reaction system; when β is small, the orbital overlap between
donor and acceptor does not decrease very much as the donor–
acceptor distance increases. In other words, when β is small, the
orbitals participating in the ET reaction expand well and the
Hrp value is not very sensitive to the distance. Thus, the β value
is important for the ET reaction. Of these parameters, the ∆GЊ
value can be experimentally evaluated with redox potentials at
the ground state, quenching reaction rate, and absorption and
emission spectra. However, it is not easy to evaluate experi-
mentally the Hrp, λ and β values, although they are fund-
amental parameters in the ET reaction. Actually, only a few of
these values have been experimentally presented; for instance,
Hoffman and coworkers reported the λ value in the photo-
induced ET reaction between [Ru(bpy)3]2ϩ and methyl viologen
(MV2ϩ),9 and also we evaluated the Hrp and λ values in the
similar photo-induced ET reaction between [Ru(bpy)3]2ϩ and
such viologens as MV2ϩ and 4,4Ј-bipyridyl-1,1Ј-di(propionic
acid) and the reverse ET reaction between [Ru(bpy)3]3ϩ and
one-electron reduced viologens.10 Moreover, the β value has not
been experimentally evaluated yet in the photo-induced ET
reactions of ruthenium() complexes, to our best knowledge.
In the present work, we synthesized three Ru() complexes of
2,2Ј-bipyridine derivatives, namely [Ru(prbpy)3]Cl2 1 (prbpy =
properties of [Ru(bpy)3]2ϩ 1a,b,6,7
This complex is still one
.
of the best photosensitizers because of its long-lived photo-
excited state and significantly negative reduction potential
at the excited state. Detailed knowledge of the photo-
induced ET reaction of this complex is necessary for making
further developments of photochemical applications of this
complex.
According to Marcus,8 the rate constant for the ET reaction
is represented as a function of the free energy change (∆GЊ), the
electronic coupling matrix element (Hrp), and the reorganiz-
ation energy (λ), as shown in eqn. 1.
(1)
† Electronic supplementary information (ESI) available: Fig. 1S, 2S and
3S describe the single exponential decays of the emission spectra of
[Ru(prbpy)3]2ϩ, [Ru(chbpy)3]2ϩ and [Ru(adbpy)3]2ϩ, respectively; Fig. 4S
and 5S show the Stern–Volmer relationships in the quenching reaction
between the excited state of [Ru(Rbpy)3]2ϩ (R = pr and ad) and MV2ϩ
;
Fig. 6S and 7S give the relationship between Hrp2 and rRu, where εr and
η values correspond to pure EtOH and EtOH–H2O (8 : 2 v/v); Tables S1
and S2 present Hrp, λ and β values, where εr and η values correspond
to pure EtOH and EtOH–H2O (8 : 2 v/v). See http://www.rsc.org/
suppdata/dt/b2/b207048k/
D a l t o n T r a n s . , 2 0 0 3 , 6 9 2 – 6 9 8
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
692

View Article Online
4,4Ј-di(n-propylaminocarbonyl)-2,2Ј-bipyridine), [Ru(chbpy)3]-
Cl2 2 (chbpy = 4,4Ј-di(cyclohexylaminocarbonyl)-2,2Ј-bipyr-
idine) and [Ru(adbpy)3]Cl2 3 (adbpy = 4,4Ј-di(adamantyl-
aminocarbonyl)-2,2Ј-bipyridine), as shown in Scheme 1. We
Synthesis of [Ru(prbpy)3]Cl2 1. Complex 1 was synthesized
like [Ru(bpy)3]Cl2,13 as follows: a solution of RuCl3ؒ3H2O
(52.5 mg) and prbpy (397 mg) in EtOH (10 ml) was kept at 80
ЊC for 10 days in a sealed glass tube. The crude products were
separated by filtration and the purification was carried out
by column chromatography (silica gel: eluent = methanol–
chloroform (1 : 19 v/v)). Yield 113 mg (36.5%). The differential
thermal analysis clearly indicated that four water molecules
were included in the crystal. Found: C, 52.96; H, 5.90; N,
13.50%. RuC54H66N12O6Cl2ؒ4H2O requies C, 53.02; H, 6.10; N,
13.74%. δH (CDCl3, 400 MHz) 10.5 (s, 1H, NH), 8.95 (s, 1H,
6-py), 8.10 (s, 1H, 3-py), 7.70 (s, 1H, 5-py), 3.30 (d, 2H,
>N–CH2–), 1.75 (m, 2H, –CH2–) and 0.95 (s, 3H, –CH3).
Syntheses of [Ru(chbpy)3]Cl2 2 and [Ru(adbpy)3]Cl2 3. Com-
plexes 2 and 3 were prepared similarly to 1, except that DMF
was used as the solvent for the synthesis of 2. 2: Yield 167 mg
(46.6%). Found: C, 57.07; H, 6.62; N, 11.17%. RuC72H90-
N12O6Cl2ؒ7H2O requires C, 56.98; H, 6.91; N, 11.07%.
δH (CDCl3, 400 MHz) 9.60 (s, 6H, NH), 9.00 (s, 6H, 3-py), 7.90
(d, 6H, 5-py), 7.80 (d, 6H, 6-py), 3.80 (m, 6H, cyclohexyl) and
∼1.50 (m, 60H, cyclohexyl). 3: Yield 118 mg (26.3%). Found: C,
62.93; H, 6.60; N, 8.78%. RuC96H114N12O6Cl2ؒ7H2O requires C,
63.00; H, 7.05; N, 9.18%. δH (DMSO-d6, 400 MHz) 9.50 (d, 6H,
5-py), 8.60 (s, 6H, NH), 7.90 (s, 6H, 3-py), 7.80 (d, 6H, 6-py)
and ∼2.00 (m, 90H, adamantyl).
Scheme 1
investigated the photo-induced ET reaction between these
Ru() complexes and MV2ϩ, to clarify the dependence of ET
reaction rate on the intermolecular distance, since these com-
plexes have similar photochemical properties but different sizes.
It is our intention here to evaluate experimentally the Hrp, λ and
β values, to present a clear relationship between the Hrp value
and the size of Ru() complexes and to report the β value.
Experimental
Electrochemical measurements
The cyclic voltammograms of the ruthenium() complexes were
recorded at controlled temperature under nitrogen atmosphere
with a combined system of potentiostat and function generator
to which a glassy carbon electrode was attached as a working
electrode (Toho technical research; model PS-06). A sample
solution was prepared by dissolving the ruthenium() complex
(1.0 mmol dmϪ3) in CH2Cl2 containing tetraethylammonium
perchlorate (0.1 mol dmϪ3).
Materials
2,2Ј-Bipyridyl-4,4Ј-dicarboxylic acid (dcbpy) was synthesized
as reported.11 Propylamine, cyclohexylamine, and adamantyl-
amine were purchased from Nakarai Chemical Co. Ltd.
(guaranteed grade) and used after distillation. Methyl viologen
(1,1Ј-dimethyl-4,4Ј-bipyridinium dichloride) was purchased
from Nakarai Chemical Co. Ltd. (guaranteed grade) and used
without further purification. All the solvents were used after
distillation.
Quenching reaction
Solutions of the ruthenium() complexes (5.0 µmol dmϪ3) and
methyl viologen (0.10–9.0 mmol dmϪ3) in EtOH–H2O (4 ml;
9 : 1 v/v) were placed in a pyrex cell after five freeze–pump–thaw
cycles. The ionic strength of the reaction solution was adjusted
to 0.1 mol dmϪ3 by addition of tetraethylammonium chloride
((Et4N)Cl). Then, the emission intensity and the lifetime of the
Ru() complexes were measured at the controlled temperature
( 0.1 ЊC), under visible light irradiation corresponding to the
absorption maximum of these Ru() complexes. A Hitachi
fluorometer F3010 was used for the measurements of emission
intensity. The excited state lifetime was measured by laser
flash photolysis, which was carried out with 355 nm laser pulses
from a Continuum Surelite I-10 laser system. The emission
spectra of the ruthenium() complexes were monitored with a
photomultiplier (Hamamatsu R928) and a digital oscilloscope
Tektronix TDS-380P.
Synthesis of 4,4Ј-di(propylaminocarbonyl)-2,2Ј-bipyridine
(prbpy). Prbpy was prepared according to the synthesis of a
similar compound,12 as follows: dcbpy (480 mg) was reacted
with thionyl chloride (10 ml) at 85 ЊC for 3 h to afford 2,2Ј-bi-
pyridyl-4,4Ј-dicarbonyl chloride. After cooling to room tem-
perature, the remaining thionyl chloride was removed from the
solution by evaporation under reduced pressure. The residue
was dissolved in benzene (40 ml) and n-propylamine (2 ml) was
added. The solution was refluxed for 12 h. The crude solids
separated by filtration were washed with acetone and then puri-
fied by recrystallization from MeOH–chloroform (3 : 1 v/v).
Yield 280 mg (44%). Found: C, 65.77; H, 6.87; N, 17.03%.
C18H22N4O2 requires C, 66.24; H, 6.79; N, 17.17%. δH (CDCl3,
400 MHz) 8.80 (d, 2H, 5-py), 8.60 (s, 2H, 3-py), 7.80 (d, 2H,
6-py), 6.40 (s, 2H, NH), 3.50 (q, 4H, CH2), 1.70 (m, 4H, CH2)
and 1.00 (t, 6H, CH3).
Synthesis of 4,4Ј-di(cyclohexylaminocarbonyl)-2,2Ј-bipyridine
(chbpy). Chbpy was synthesized like prbpy, except that the
solids were recrystallized from DMF. Yield 470 mg (28.3%).
Found: C, 70.49; H, 7.35; N, 13.77%. C24H30N4O2Cl2 requires
C, 70.91; H, 7.44; N, 13.78%. δH (DMSO-d6, 400 MHz) 8.90
(d, 62, 5-py), 8.80 (s, 2H, NH), 8.70 (s, 2H, 3-py), 7.80 (d, 2H,
6-py), 3.80 (m, 6H, ch) and ∼1.50 (m, 20H, cyclohexyl).
Results and discussion
Photochemical properties of [Ru(prbpy)3]2؉ 1 (prbpy ؍ 4,4Ј-
di(propylaminocarbonyl)-2,2Ј-bipyridine), [Ru(chbpy)3]2؉
2
(chbpy ؍ 4,4Ј-di(propylaminocarbonyl)-2,2Ј-bipyridine) and
[Ru(adbpy)3]2؉ 3 (adbpy ؍ 4,4Ј-di(adamantylaminocarbonyl)-
2,2Ј-bipyridine)
Synthesis of 4,4Ј-di(adamantylaminocarbonyl)-2,2Ј-bipyridine
(adbpy). Adbpy was synthesized like prbpy. Yield 510 mg
(31.7%). Found: C, 74.40; H, 7.32; N, 10.94%. C32H38N4O2
requires C, 75.26; H, 7.50; N, 10.97%. δH (CDCl3, 400 MHz)
8.80 (d, 2H, 5-py), 8.60 (s, 2H, 3-py), 7.80 (d, 2H, 6-py), 6.00 (s,
2H, NH) and ∼2.00 (m, 30H, adamantyl).
Photochemical properties of 1, 2 and 3 are given in Table 1 with
those of [Ru(bpy)3]2ϩ 14
For 1–3 the metal to ligand charge
.
transfer (MLCT) absorption band is observed at λmax = 464–465
nm (ε = 21600–24500 molϪ1 dm3 cmϪ1) and the emission
spectrum is at λmax = 614–616 nm in EtOH–H2O (9 : 1 v/v),
where uncorrected values of emission maxima are given. These
D a l t o n T r a n s . , 2 0 0 3 , 6 9 2 – 6 9 8
693

View Article Online
Table 1 Photochemical and electrochemical properties of [Ru(prbpy)3]2ϩ 1 (prbpy = 4,4Ј-di(propylaminocarbonyl)-2,2Ј-bipyridine), [Ru(chbpy)3]2ϩ
2 (chbpy = 4,4Ј-di(propylaminocarbonyl)-2,2Ј-bipyridine) and [Ru(adbpy)3]2ϩ 3 (adbpy = 4,4Ј-di(adamantylaminocarbonyl)-2,2Ј-bipyridine) in
EtOH–H2O 9 : 1 (v/v) at 25 ЊC, compared with those of [Ru(bpy)3]2ϩ
Absorption
Emission
λmax/nm
ε/molϪ1 dm3 cmϪ1
λmax/nm
τ/ns
E(Ru3ϩ/2ϩ)b/V vs. SCE
E(Ru3ϩ/2ϩ*)/V vs. SCE
[Ru(prbpy)3]2ϩ
[Ru(chbpy)3]2ϩ
[Ru(adbpy)3]2ϩ
[Ru(bpy)3]2ϩ
464
464
465
452
24500
21700
21600
14900a
614
615
616
600
1400
1420
1290
592a
ϩ1.53
ϩ1.57
ϩ1.52
ϩ1.22c
Ϫ0.61
Ϫ0.60
Ϫ0.62
Ϫ0.81d
a In a buffer solution of maleic acid/tris(hydroxymethyl)aminomethane/sodium hydroxide (pH 7.0), µ = 0.1 at 25 ЊC. b In dichloromethane.
c This potential was measured in acetonitrile/tetrabutylammonium hexafluorophosphate (0.1 mol dmϪ3) with glassy carbon electrode. See ref. 14.
d Ref. 20.
absorption and emission maxima are shifted to longer wave-
lengths relative to those of [Ru(bpy)3]2ϩ by about 12 nm and 15
nm, respectively. The red-shift of absorption and emission
spectra is easily interpreted in terms of the π* orbital of 2,2Ј-
bipyridine becoming lower in energy upon the introduction of
the electron-withdrawing alkylaminocarbonyl (CONHR) group.
2ϩ
A similar observation was reported for [Ru(dmp)n(decb)3Ϫn
]
(dmp = 4,4Ј-dimethyl-2,2Ј-bipyridine; decb = 4,4Ј-di(ethoxy-
carbonyl)-2,2Ј-bipyridine; n = 1–3).15 The decay of the emission
spectra occurs in a single-exponential manner, which is ascer-
tained by deconvolution. From the decay curve, the lifetime (τ)
of the excited state was evaluated. The lifetimes of 1, 2 and 3 are
not sensitive to the alkyl group of CONHR, as expected (see
Table 1). Interestingly, their lifetimes are about twice as long as
that of [Ru(bpy)3]2ϩ, as was also found for similar ruthenium()
complexes [Ru(decb)3]2ϩ 15 and [Ru(menbpy)3]2ϩ (menbpy =
4,4Ј-di{(1R,2S,5R)-(Ϫ)-menthoxycarbonyl}-2,2Ј-bipyridine).16
Their longer lifetimes than that of [Ru(bpy)3]2ϩ are understood
by considering the following two factors: one is the energy sep-
aration between the 3MLCT and the 3MC (metal centered d–d)
excited states. Since this energy separation increases upon the
introduction of electron-withdrawing substituents (vide supra),
Fig. 1 Relationships of quenching rate constant (kr) and the redox
potentials of quenchers (2,6-dichloro-p-benzoquinone, p-dinitro-
benzene, o-dinitrobenzene, m-dinitrobenzene, methyl p-nitrobenzoate,
methyl m-nitrobenzoate) in [Ru(bpy)3]2ϩ (ref. 19) and [Ru(Rbpy)3]2ϩ
(Rbpy = 4,4Ј-di(alkylaminocarbonyl)-2,2Ј-bipyridine; alkyl = propyl,
hexyl or adamantyl).
3
3
the deactivation of the MLCT excited state through the MC
2ϩ 15
excited state is suppressed, like that of [Ru(dmp)n(decb)3Ϫn
] .
On the other hand, the energy separation between 3MLCT and
the ground states decreases upon the introduction of electron-
withdrawing substituents. This would lead to acceleration of
0 V at which the slope of the plot is 0.5. Such a potential giving
∆GЊ = 0 is easily determined, as shown in Fig. 1. Comparing
this potential of [Ru(bpy)3]2ϩ with that of [Ru(Rbpy)3]2ϩ, we
estimate that the E Ru()/*Ru() value of [Ru(Rbpy)3]2ϩ is differ-
ent from that of [Ru(bpy)3]2ϩ by about 0.2 V. On the basis of the
reported E Ru()/*Ru() value of [Ru(bpy)3]2ϩ (–0.81 V (vs. SCE)
in acetonitrile),¶ the E Ru()/*Ru() values of 1, 2 and 3 are
evaluated to be Ϫ0.61, Ϫ0.60 and Ϫ0.62 V, respectively.||
3
direct deactivation of the MLCT excited state to the ground
state by the energy gap law.17 There is another plausible factor,
the protection of the excited *[Ru(Rbpy)3]2ϩ by the bulky
groups against solvent approach, as follows: the bulky groups
introduced onto the Rbpy ligand do not allow solvent mole-
cules to approach the core moiety of excited *[Ru(Rbpy)3]2ϩ
,
The reduction potentials (E Ru()/Ru()) of 1, 2 and 3 at the
ground state were measured with cyclic voltammetry to be
ϩ1.53, ϩ1.57 and ϩ1.52 V (vs. SCE), respectively, in dichloro-
methane. Their reduction potentials at the ground state are
more positive (or less negative) than that of [Ru(bpy)3]2ϩ by
about 200 mV, like the reduction potential at the excited state.
The shifts of reduction potential at both ground and excited
states are interpreted in terms of the π and π* orbitals of
2,2Ј-bipyridine being stabilized in energy by introduction of the
electron-withdrawing CONHR group to 2,2Ј-bipyridine. A
which suppresses the deactivation by the solvent. At this
moment, it is difficult to decide which factor is more important
and we need further experiments.
Though the reduction potential (E Ru()/*Ru()) of the Ru()
complexes at the excited state was spectroscopically estimated
by Navon and Sutin,18 we could not adopt their method,
because the redox potential of [Ru(bpy)3]2ϩ in the ground state
could not be measured in EtOH–H2O. ‡ Therefore, we estimated
the E Ru()/*Ru() value with the method proposed by Meyer
et al.19 and Balzani et al.,20 where the oxidative quenching reac-
tion was carried out with such quenchers as nitrobenzene and
p-quinone derivatives.§ In Fig. 1, the values of (RT/F)lnkr are
plotted against the redox potential of the quencher. According
to the method proposed,19,20 the ∆GЊ value can be taken to be
similar explanation was previously presented for the reduction
2ϩ 15
potentials of [Ru(dmp)n(decb)3Ϫn
] .
The radii (r) of the Ru() complexes and MV2ϩ were esti-
mated with the semiempirical equation r = 1/2(dxdydz)1/3 21
where
,
¶ The E Ru()/*Ru() value of [Ru(bpy)3]2ϩ was reported to be Ϫ0.81 V
(vs. SCE) in acetonitrile.19
‡ In this method, the reduction potential, E Ru()/Ru(), at the ground state
is necessary. However, we failed to measure the E Ru()/Ru() values of 1, 2
and 3 in EtOH–H2O, because H2 evolution occurred during the CV
measurement.
§ 2,6-Dichloro-p-benzoquinone, p-dinitrobenzene, o-dinitrobenzene,
m-dinitrobenzene, methyl p-nitrobenzoate, methyl m-nitrobenzoate
were used here.
|| To estimate the reduction potential at the excited state according to
Meyer,20 the value of kq(∆GЊ = 0) is necessary. Though kq(∆GЊ = 0) is
not known for 1, 2 and 3, we could estimate correctly the reduction
potentials of 1, 2 and 3 at the excited state by comparing them with that
of [Ru(bpy)3]2ϩ, since the reduction potential of [Ru(bpy)3]2ϩ in the
excited state is known.
D a l t o n T r a n s . , 2 0 0 3 , 6 9 2 – 6 9 8
694

View Article Online
dx, dy and dz are the molecular sizes along the three axes of the
CPKspace-fillingmodel.**Theestimatedrvaluesare12.5, 12.9,
13.4 and 3.62 Å for 1, 2, 3 and MV2ϩ, respectively.
The above results indicate that the size of the Ru() com-
plexes depends on the alkyl group of CONHR, as expected,
while the redox potential and the lifetime of *[Ru(Rbpy)3]2ϩ are
little influenced by the alkyl group. Thus, we can expect to
investigate the dependences of electron transfer rate, Hrp and λ
on the size of the donor without interference of the other
factor.
where Ru() represents [Ru(Rbpy)3]2ϩ. In this mechanism, the
Stern–Volmer relationship is given by eqn. 8.
τ0/τ = I0/I = 1 ϩ krτ0[MV2ϩ
]
(8)
The quenching rate constant (kr) for each complex was evalu-
ated from the slope of the Stern–Volmer plot and the obtained
values are listed in Table 2, where τ0 was measured in the
absence of MV2ϩ. It is noted that the kr value decreases as the
size of the Ru() complex increases.
Quenching reaction with MV2؉
Correction of diffusion rate
The excited states of 1, 2 and 3 were efficiently quenched by
MV2ϩ in EtOH–H2O (9 : 1 v/v). Fig. 2 shows the relation
Since these kr values are considerably large, we must make a
diffusion rate correction. The ET reaction (eqn. 5) is considered
to consist of the diffusional formation of an exciplex (kD in eqn.
9), the diffusional dissociation of the exciplex (k–D in eqn. 9)
and an ET reaction in the exciplex (krcor in eqn. 10). Application
of the usual steady-state approximation to the concentration of
[*Ru()2ϩ ؒ ؒ ؒ MV2ϩ] leads to eqn. 11.23
(9)
(10)
krcor = krobsk–D/(kD Ϫ krobs
)
(11)
The kD and k–D values were calculated according to Debye–
Smoluchowski24 and Eigen,25 respectively, where the dielectric
constant (εr) and viscosity (η) of the medium were taken from
the reference†† and the radii (r) of the reactants estimated above
were employed. The evaluated kD, kϪD, and krcor values are
summarized in Table 2. The kD value increases in the order
1 < 2 < 3 and the kϪD value decreases in the order 1 > 2 > 3.
Fig.
2
The Stern–Volmer plot in the quenching reaction of
[Ru(chbpy)3]2ϩ by MV2ϩ in EtOH–H2O (9 : 1 v/v) at 25 ЊC.
Since these tendencies are not consistent with the order of krobs
,
between the τ0/τ value and MV2ϩ concentration observed in the
quenching reaction of [Ru(chbpy)3]2ϩ by MV2ϩ as an example,
where τ0 represents the lifetime in the absence of MV2ϩ (see
ESI,† Fig. 4S and 5S for the relations observed in [Ru(prbpy)3]2ϩ
and [Ru(adbpy)3]2ϩ). The I0/I value is almost the same as the τ0/
τ value, where I and I0 are the emission intensities in the pres-
ence and the absence of quencher, respectively. This result
clearly indicates that this quenching reaction proceeds not
through a static quenching mechanism but through a dynamic
quenching mechanism.
In the dynamic quenching mechanism, the Ru() complex
transforms to the excited state upon irradiation, independently
of MV2ϩ (eqn. 3). The excited Ru() complex collides with
MV2ϩ to form the encounter complex (eqn. 5), from which
either charge separation (eqn. 6) or back-electron transfer
(eqn. 7) occurs:
the diffusional formation and the diffusional dissociation of the
exciplex are not responsible for the decreasing order of krobs
,
1 > 2 > 3.‡‡ On the other hand, the krcor value decreases in the
order 1 > 2 > 3, which is consistent with the decreasing order of
the krobs value. Thus, it is concluded that the krcor value is
responsible for the decreasing order of krobs; in other words,
the ET reaction in the exciplex plays an important role in the
overall ET reaction rate.
Analysis of ET reaction based on Marcus’ theory:
We analyzed here the ET reaction in the exciplex with Marcus’
theory. Eqn. 1 given by Marcus’ theory is transformed to
eqn. 12,
(12)
(3)
(4)
(5)
(6)
(7)
which apparently shows that a linear relationship exists between
ln(krcor T 1/2) and 1/T if ∆GЊ is not sensitive to T . Actually, the
∆GЊ value varies little when the temperature changes from
10 ЊC to 40 ЊC, §§ and a linear relationship is obtained, as
†† The dielectric constants (εr) used are 33.0, 31.2, 29.4, 27.7, 26.0 and
24.4 for 5, 15, 25, 35, 45 and 55 ЊC, respectively. The viscosity (η) values
are 2.57, 1.96, 1.51, 1.19, 0.97 and 0.80 mPa s for 5, 15, 25, 35, 45 and
55 ЊC, respectively. These εr and η values were estimated from the
reported values.26a,b
‡‡ The final results, such as Hrp, λ and β values, depend little on the
estimated εr and η values (see ESI). Thus, the present conclusions are
reliable in spite of the uncertainties of εr and η which arise from the use
of mixed solvent.
** Chem 3D was used for MM calculations with standard parameters.
Chem 3D version 3.0 for Apple Macintosh, Cambridge Scientific
Computing, Inc., Massachusetts. In the MM calculations, the Ru–N
distance was assumed to be the same as the experimental bond
distance (2.056 Å).22
§§ The ∆GЊ value changes little when the temperature increases from
10
10 ЊC to 40 ЊC in the similar reaction between [Ru(bpy)3]2ϩ and MV2ϩ
.
D a l t o n T r a n s . , 2 0 0 3 , 6 9 2 – 6 9 8
695

View Article Online
Table 2 The observed rate constants (krobs) of the electron transfer reaction between methyl viologen (MV2ϩ) and the excited state of [Ru(prbpy)3]2ϩ
1
(prbpy = 2 (chbpy = 4,4Ј-di(propylaminocarbonyl)-2,2Ј-bipyridine) or
4,4Ј-di(propylaminocarbonyl)-2,2Ј-bipyridine), [Ru(chbpy)3]2ϩ
[Ru(adbpy)3]2ϩ 3 (adbpy = 4,4Ј-di(adamantylaminocarbonyl)-2,2Ј-bipyridine), the diffusion rate constant (kD) for the exciplex formation, the
diffusional dissociation rate constant (kϪD) of the exciplex, and the electron transfer rate constant (krcor) in the exciplexa
Ru()
10Ϫ8 krobs a/molϪ1 dm3 sϪ1
10Ϫ9 kD/molϪ1 dm3 sϪ1
10Ϫ8 kϪD/sϪ1
10Ϫ7 krcor/sϪ1
[Ru(prbpy)3]2ϩ
[Ru(chbpy)3]2ϩ
[Ru(adbpy)3]2ϩ
2.62 0.01
1.72 0.01
1.06 0.03
6.09
6.17
6.27
4.52
4.33
4.12
2.03 0.01
1.24 0.01
0.707 0.002
a In EtOH–H2O (9 : 1 v/v; µ = 0.1 ((Et4N)Cl)) at 25 ЊC, [Ru()] = 5.0 µmol dmϪ3, [MV2ϩ] = 0–9 mmol dmϪ3
.
Table 3 Rate constant (krcor) of the electron transfer reaction in the
and εr values are the refractive index and the dielectric constant
of the medium, respectively, e is the charge of an electron and
4πε0 is the dielectric constant in vacuum. The λout value calcu-
lated with this equation depends little on the nature of Rbpy;
λout = 0.846 eV, 0.848 eV and 0.851 eV, for 1, 2 and 3, respect-
ively.|||| In spite of the minimal dependence of λout on Rbpy, the λ
value of 3 is much larger than those of 1 and 2 (see Table 3),
which indicates that the λin value of 3 is larger than those of 1
and 2. This is probably because bulky substituents such as the
adamantyl group cause a marked change in steric repulsion
with changes in the Ru–N distance induced by the ET reaction.
2ϩ
exciplex, [*Ru(Rbpy)3 ؒ ؒ ؒ MV2ϩ], electronic coupling matrix element
(Hrp) and reorganization energy (λ)a
Ru()
10Ϫ7 krcor/sϪ1
103 Hrp/eV
λ/eV
[Ru(prbpy)3]2ϩ
[Ru(chbpy)3]2ϩ
[Ru(adbpy)3]2ϩ
2.03
1.24
0.707
2.28
1.86
1.37
0.772
0.767
0.798
a In EtOH–H2O (9 : 1 v/v, µ = 0.1 mol dmϪ3) at 25 ЊC.
½
However, the (2π/h){1/(4πkbλ exp{Ϫ(∆GЊ ϩ λ)2/(4kbTλ} term
which involves λ is calculated to be 3.80 × 1012, 3.25 × 1012 and
3.57 × 1012 (eVϪ2 sϪ1) for 1, 2 and 3, respectively, when the ∆GЊ
and λ values evaluated here are used. These values are not con-
sistent with the decreasing order of krcor, 1 > 2 > 3. On the other
hand, the Hrp2 value decreases in the order 1 > 2 > 3. Thus, it is
clearly concluded that not the λ value but the Hrp value is
responsible for the decreasing order of krcor, 1 > 2 > 3.
2
The value of lnHrp is plotted against the radii of the Ru()
complexes, as shown in Fig. 4. As expected from eqn. 2, a good
Fig. 3 Plot of ln(T 1/2krcor) against T Ϫ1 in the quenching reactions of
[Ru(prbpy)3]2ϩ 1 (prbpy = 4,4Ј-di(propylaminocarbonyl)-2,2Ј-bipyr-
idine), [Ru(chbpy)3]2ϩ 2 (chbpy = 4,4Ј-di(propylaminocarbonyl)-2,2Ј-
bipyridine) and [Ru(adbpy)3]2ϩ 3 (adbpy = 4,4Ј-di(adamantylamino-
carbonyl)-2,2Ј-bipyridine) with methyl viologen (MV2ϩ).
shown in Fig. 3. From its slope, the λ value was evaluated with
the independently measured ∆GЊ values; ∆GЊ is 0.05, 0.06 and
0.04 V for 1, 2 and 3, respectively.¶¶ From its intercept, the Hrp
value was evaluated with the above-estimated λ value. Their
values are listed in Table 3. The Hrp value decreases in the order
1 > 2 > 3, while the λ value depends little on the size of the
Ru() complex. The λ value is in general considered to be the
sum of λin and λout; λ = λin ϩ λout, where λin is the contribution
from geometry changes of the reactant and the substrate and
λout is the contribution from changes of solvent reorganization.
The λout value is in general estimated with eqn. 13:21,27
Fig. 4 Plot of lnHrp2 against the radii of the Ru() complexes, rRu
.
linear relationship was obtained, and the β value was estimated
to be 11 nmϪ1 from the slope. This value is similar to the
reported values which were estimated in nonbiological ET reac-
tions;8a for instance, the β value was 12 nmϪ1 in the ET reaction
of aromatic compounds.29 Considering that the excited electron
is in a diffuse orbital, one can expect that the β value could be
λout = {1/(2rA) ϩ 1/(2rD) Ϫ
1/(rAD)}(1/n2 Ϫ 1/εr){e2/(4πε0)} (13)
where rA and rD are the radii of methyl viologen and
[Ru(Rbpy)3]2ϩ, respectively, and rAD is the distance between
methyl viologen and [Ru(Rbpy)3]2ϩ in effective contact, the n
|||| We adopted here rA and rD evaluated with the semiempirical
equation 1/2(dxdydz)1/3 (see text).21 The rAD value was assumed to be the
sum of rA and rD, in the usual way. In the evaluation of λout, we adopted
n and εr values for EtOH–H2O (9 : 1 v/v) at 15 ЊC, because the n value
has not been reported for EtOH–H2O (9 : 1 v/v) at 25 ЊC; n = 1.36728
and εr = 31.19.26b Though the experiments were carried out at 25 ЊC,
the n value depends little on temperature as follows: for example,
n of EtOH are 1.359 at 25 ЊC and 1.361 at 20 ЊC.30 Thus, the present
discussion is correct, at least semiquantitatively.
¶¶ The ∆GЊ value is evaluated as follows: ∆GЊ = EЊ(Ru(Rbpy)33ϩ/2ϩ*) Ϫ
EЊ(MV2ϩ/ϩ) ϩ (ωp Ϫ ωr), where EЊ(MV2ϩ/ϩ) = Ϫ0.66 V (vs. SCE in
EtOH–H2O (9 : 1 v/v)), and ωp and ωr are the Coulombic work terms
(∼10Ϫ2 eV). Since the slope of the linear relationship in Fig. 3 is Ϫ(∆GЊ
ϩ λ)2/(4πkb), the λ value is estimated with the ∆GЊ value which is meas-
ured electrochemically. Then, the Hrp value is estimated from the inter-
cept with the estimated λ value.
D a l t o n T r a n s . , 2 0 0 3 , 6 9 2 – 6 9 8
696

View Article Online
smaller in the photo-induced ET reaction than in the thermal
ET reaction.8a The present β value is not consistent with this
expectation, seemingly. However, the present result is
reasonably interpreted in terms of orbitals participating in the
ET reaction, as follows. In the 3MLCT excited state of
[Ru(Rbpy)3]2ϩ, the excited electron is believed to exist in the π*
acceptor to suppress the photo-induced ET reaction of
[Ru(Rbpy)3]2ϩ. The λ value has little influence on the present
ET reaction. Though these conclusions meet our expectations,
they are experimentally presented for the first time in this work.
The β value was estimated to be about 11 nmϪ1, which is
similar to that of the thermal ET reaction between an aromatic
anion radical and an aromatic compound. This is because the
electron transfer occurs from the π* orbital of aromatic system
in both the photo-induced ET reaction of [Ru(Rbpy)3]2ϩ and
the thermal ET reaction of the aromatic anion radical.
2ϩ
orbital of Rbpy. In the exciplex, [*Ru(Rbpy)3 ؒ ؒ ؒ MV2ϩ], the
excited electron transfers from the π* orbital of Rbpy to the
LUMO of MV2ϩ which is also the π* orbital. In the ET reac-
tions of aromatic compounds investigated previously,29 an odd
electron in the π* orbital of an aromatic anion radical transfers
to the π* orbital of a different aromatic compound. Thus, elec-
tron transfer occurs from the π* orbital of one molecule to the
π* orbital of the other molecule in both the photo-induced ET
reaction between [Ru(Rbpy)3]2ϩand MV2ϩ and the thermal ET
reaction between an aromatic anion radical and a neutral aro-
matic compound. From the above discussion, it is reasonably
concluded that the π* orbital of Rbpy expands similarly to that
of aromatic compounds and the β value in the photo-induced
ET reaction of the Ru() complexes of 2,2Ј-bipyridine deriv-
atives is similar to that in the thermal ET reaction of aromatic
anion radicals.
Strictly speaking, we need to consider not only the size but
also the shape of the substituent, because a highly branched
substituent such as the adamantyl group occupies the space
around the bpy ligand to suppress the approach of methyl vio-
logen to the [Ru(bpy)3]2ϩ core. In the present work, the highly
branched adamantyl group gives a very small Hrp value, while
the linear propyl group also gives a smaller Hrp value than
expected; the Hrp value of 1 exists at a lower position than the
linear relationship of Fig. 4. We need to perform more detailed
studies with systematically synthesized [Ru(Rbpy)3]2ϩ to clarify
how the size and the shape of substituent influence the ET
reaction rate.
Acknowledgements
Financial support from the Ministry of Education, Culture,
Sports, and Science through Grants-in Aid (No.20094013 and
09044096) is greatly acknowledged.
References
1 For instance, (a) K. Kalyanasundaram, Coord. Chem. Rev., 1982, 46,
46; (b) A. Juris, V. Balzani, F. Barigelleti, S. Campagna, P. Belser and
A. von Zelewsky, Coord. Chem. Rev., 1988, 84, 85; (c) Electron
Transfer in Inorganic, Organic, and Biological Systems, ed. J. R.
Bolton, N. Mataga and G. McLendon, Adv. Chem. Ser. 228, ACS,
Washington DC, Canadian Society for Chemistry, Ottawa, Canada,
1991.
2 (a) C. Creutz and N. Sutin, Proc. Natl. Acad. Sci. USA, 1975, 72,
2858; (b) G. M. Brown, B. S. Brunschwig, C. Creutz, J. F. Endicott
and N. Sutin, J. Am. Chem. Soc., 1979, 101, 1298; (c) N. Sutin and
C. Creutz, Pure Appl. Chem., 1980, 52, 2717; (d ) C. V. Krishnan,
B. S. Brunschwig, C. Creutz and N. Sutin, J. Am. Chem. Soc., 1985,
107, 2005.
3 K. Kalyanasundaram, J. Kiwi and M. Grätzel, Helv. Chim. Acta,
1978, 61, 2720.
4 M. Kirch, J.-M. Lehn and J.-P. Sauvage, Helv. Chim. Acta, 1979, 62,
1345.
5 (a) P. J. DeLaive, B. P. Sullivan, T. J. Meyer and D. G.
Whitten, J. Am. Chem. Soc., 1979, 101, 4007; (b) W. J. Dressick,
T. J. Meyer, B. Durham and D. P. Rillena, Inorg. Chem., 1982, 21,
3451.
6 H. Yersin, W. Humbs and J. Straser, Coord. Chem. Rev., 1997, 159,
325 and references therein.
Conclusions
Ruthenium() complexes of 2,2Ј-bipyridine derivatives,
[Ru(Rbpy)3]2ϩ (Rbpy = 4,4Ј-di(alkylaminocarbonyl)-2,2Ј-bi-
pyridine; alkyl = propyl, hexyl or adamantyl), were newly
synthesized here to investigate the distance dependence of their
ET reactions with methyl viologen (MV2ϩ). These three Ru()
complexes exhibit absorption and emission maxima at longer
wavelength than those of [Ru(bpy)3]2ϩ. They have more positive
reduction potentials than that of [Ru(bpy)3]2ϩ by about 200 mV
in both ground and excited states. All these results are inter-
preted in terms of the π and π* orbitals of Rbpy becoming
lower in energy because of the electron-withdrawing CONHR
substituent introduced onto 2,2Ј-bipyridine, as previously dis-
cussed.15 The ET reaction becomes slower in the order alkyl =
propyl > hexyl > adamantyl, which is the opposite of the
increasing order of the size of the substituent introduced onto
2,2Ј-bipyridine. Correction of diffusion rate was made and the
rate constant (krcor) of the ET reaction in the exciplex was
evaluated. The krcor value also decreases in the order alkyl =
propyl > hexyl > adamantyl, while the diffusion rate for exci-
plex formation (kD) increases in the order propyl < hexyl <
adamantyl and the diffusional dissociation rate of the exciplex
(kϪD) decreases in the order propyl > hexyl > adamantyl. Thus,
the kD and kϪD values are not responsible for the above-men-
tioned decreasing order of the rate constant but the krcor value is
responsible for the decreasing order of the overall ET reaction.
The ET reaction in the exciplex was analyzed on the basis of
Marcus’ theory. The electronic coupling matrix element (Hrp)
and the reorganization energy (λ) were evaluated from the tem-
perature dependence of the rate constant. The decreasing order
of krcor, 1 > 2 > 3, is clearly interpreted in terms of the decreas-
ing order of the Hrp value, 1 > 2 > 3. This means that the bulky
substituent decreases the orbital overlap between donor and
7 C. D. Clark and M. Z. Hoffman, Coord. Chem. Rev., 1997, 159, 359
and references therein.
8 (a) R. A. Marcus and N. Sutin, Biochim. Biophys. Acta, 1985, 811,
265; (b) M. D. Newton and N. Sutin, Annu. Rev. Phys. Chem., 1984,
35, 437.
9 (a) C. D. Clark and M. Z. Hoffman, J. Phys. Chem., 1996, 106, 7526;
(b) C. E. Crowley, C. D. Clark and M. Z. Hoffman, Inorg. Chem.,
1998, 37, 5704.
10 T. Hamada, M. Tsukamoto, H. Ohtsuka and S. Sakaki, Bull. Chem.
Soc. Jpn., 1998, 71, 2281.
11 G. Sprintschnik, H. W. Sprintnik, P. P. Kirsch and D. G. Whitten,
J. Am. Chem. Soc., 1977, 99, 4947.
12 K. Ohkubo, T. Hamada, T. Inaoka and H. Ishida, Inorg. Chem.,
1989, 28, 2021.
13 I. Fujita and H. Kobayashi, Ber. Bunsen-Ges. Phys. Chem., 1972, 76,
115; R. A. Palmer and T. S. Piper, Inorg. Chem., 1966, 5, 864.
14 C. M. Elliott, R. A. Freitag and D. D. Blaney, J. Am. Chem. Soc.,
1985, 107, 4647.
15 W. F. Wacholtz, R. A. Auerbach and R. H. Schmell, Inorg. Chem.,
1986, 25, 227.
16 K. Ohkubo, T. Hamada and H. Ishida, J. Chem. Soc.,
Chem. Commun., 1993, 1423.
17 N. J. Turro, Modern Molecular Photochemistry, The Benjamin/
Cummings Publishing Co., Menlo Park, CA, 1978, p. 183.
18 G. Navon and N. Sutin, Inorg. Chem., 1974, 13, 2159.
19 C. R. Bock, J. A. Connor, A. R. Gutierrez, T. J. Meyer, D. G.
Whitten, B. P. Sullivan and J. K. Nagle, J. Am. Chem. Soc., 1979,
101, 4815.
20 V. Balzani, F. Scandra, G. Orlandi, N. Sabbatini and M. T. Indelli,
J. Am. Chem. Soc., 1981, 103, 3370.
21 B. S. Brunschwig, S. Ehrenson and N. Sutin, J. Phys. Chem., 1986,
90, 3657.
22 D. P. Rillema and D. S. Jones, J. Chem. Soc., Chem. Commun., 1979,
849.
23 M. Eigen, W. Kruse, G. Maass and L. De Maeyer, Prog. React.
Kinet., 1964, 2, 287.
D a l t o n T r a n s . , 2 0 0 3 , 6 9 2 – 6 9 8
697

View Article Online
24 P. Debye, Trans. Electrochem. Soc., 1942, 82, 265; M. Z.
Smoluchowski, Phys. Chem., 1917, 92, 129.
J. Chem. Phys., 1965, 43, 679; (c) R. A. Marcus and N. Sutin,
Inorg. Chem., 1975, 14, 213.
28 Landolt-Börnstein Tabellen, 5 Aufl., Springer-Verlag, Berlin, 1923.
29 J. R. Miller, J. V. Beitz and R. K. Huddleston, J. Am. Chem. Soc.,
1984, 106, 5057.
30 Organic Solvents, J. A. Riddick and W. B. Bunger, Eds., 3rd edn,
Wiley-Interscience, New York, 1970, p. 146.
25 M. Eigen, Z. Phys. Chem. (Munich), 1954, 1, 176.
26 (a) KagakuBinran, 4th edn., Maruzen, Tokyo, 1993, ch. 13, p. II-501;
(b) Lange’s Handbook of Chemistry, J. A. Dean, Ed., 11th edn.,
McGraw-Hill, New York, 1973, pp. 10–148 and pp. 10–287.
27 (a) R. A. Marcus, J. Chem. Phys., 1956, 24, 966; (b) R. A. Marcus,
D a l t o n T r a n s . , 2 0 0 3 , 6 9 2 – 6 9 8
698