Spontaneous formation in the dark, and visible light-induced cleavage, of a Ru-S bond in water: A thermodynamic and kinetic study

Article abstract of DOI:10.1021/ic401105v

In this work the thermal and photochemical reactivity of a series of ruthenium complexes [Ru(terpy)(N-N)(L)](X)₂ (terpy = 2,2′;6′,2″-terpyridine, L = 2-(methylthio)ethanol (Hmte) or water, and X is Cl^- or PF₆^-) with four different bidentate chelates N-N = bpy (2,2′-bipyridine), biq (2,2′-biquinoline), dcbpy (6,6′-dichloro-2,2′-bipyridine), or dmbpy (6,6′-dimethyl-2,2′-bipyridine), is described. For each chelate N-N the thermodynamic constant of the dark equilibrium between the aqua-and Hmte-complexes, the Hmte photosubstitution quantum yield, and the rate constants of the thermal interconversion between the aqua and Hmte complexes were measured at room temperature. By changing the steric hindrance and electronic properties of the spectator N-N ligand along the series bpy, biq, dcbpy, dmbpy the dark reactivity clearly shifts from a nonlabile equilibrium with N-N = bpy to a very labile thermal equilibrium with N-N = dmbpy. According to variable-temperature rate constant measurements in the dark near pH = 7 the activation enthalpies for the thermal substitution of H₂O by Hmte are comparable for all ruthenium complexes, whereas the activation entropies are negative for bpy and biq, and positive for dcbpy and dmbpy complexes. These data are indicative of a change in the substitution mechanism, being interchange associative with nonhindered or poorly hindered chelates (bpy, biq), and interchange dissociative for more bulky ligands (dcbpy, dmbpy). For the most labile dmbpy system, the thermal equilibrium is too fast to allow significant modification of the composition of the mixture using light, and for the nonhindered bpy complex the photosubstitution of Hmte by H₂O is possible but thermal binding of Hmte to the aqua complex does not occur at room temperature. By contrast, with N-N = biq or dcbpy the thermodynamic and kinetic parameters describing the formation and breakage of the Ru-S bond lie in a range where the bond forms spontaneously in the dark, but is efficiently cleaved under light irradiation. Thus, the ratio between the aqua and Hmte complex in solution can be efficiently controlled at room temperature using visible light irradiation.

Full text of DOI:10.1021/ic401105v

Article
pubs.acs.org/IC
Spontaneous Formation in the Dark, and Visible Light-Induced
Cleavage, of a Ru−S Bond in Water: A Thermodynamic and Kinetic
Study
Azadeh Bahreman,† Bart Limburg,† Maxime A. Siegler,‡ Elisabeth Bouwman,† and Sylvestre Bonnet*,†
†Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, P.O. Box 9502, Leiden, 2300 RA, The Netherlands
‡Small Molecule X-ray Facility, Department of Chemistry, Johns Hopkins University, Baltimore, Maryland 21218, United States
S
* Supporting Information
ABSTRACT: In this work the thermal and photochemical
reactivity of a series of ruthenium complexes [Ru(terpy)(N−
N)(L)](X)2 (terpy = 2,2′;6′,2″-terpyridine, L = 2-
(methylthio)ethanol (Hmte) or water, and X is Cl− or PF6 )
−
with four different bidentate chelates N−N = bpy (2,2′-
bipyridine), biq (2,2′-biquinoline), dcbpy (6,6′-dichloro-2,2′-
bipyridine), or dmbpy (6,6′-dimethyl-2,2′-bipyridine), is
described. For each chelate N−N the thermodynamic constant
of the dark equilibrium between the aqua- and Hmte- complexes, the Hmte photosubstitution quantum yield, and the rate
constants of the thermal interconversion between the aqua and Hmte complexes were measured at room temperature. By
changing the steric hindrance and electronic properties of the spectator N−N ligand along the series bpy, biq, dcbpy, dmbpy the
dark reactivity clearly shifts from a nonlabile equilibrium with N−N = bpy to a very labile thermal equilibrium with N−N =
dmbpy. According to variable-temperature rate constant measurements in the dark near pH = 7 the activation enthalpies for the
thermal substitution of H2O by Hmte are comparable for all ruthenium complexes, whereas the activation entropies are negative
for bpy and biq, and positive for dcbpy and dmbpy complexes. These data are indicative of a change in the substitution
mechanism, being interchange associative with nonhindered or poorly hindered chelates (bpy, biq), and interchange dissociative
for more bulky ligands (dcbpy, dmbpy). For the most labile dmbpy system, the thermal equilibrium is too fast to allow significant
modification of the composition of the mixture using light, and for the nonhindered bpy complex the photosubstitution of Hmte
by H2O is possible but thermal binding of Hmte to the aqua complex does not occur at room temperature. By contrast, with N−
N = biq or dcbpy the thermodynamic and kinetic parameters describing the formation and breakage of the Ru−S bond lie in a
range where the bond forms spontaneously in the dark, but is efficiently cleaved under light irradiation. Thus, the ratio between
the aqua and Hmte complex in solution can be efficiently controlled at room temperature using visible light irradiation.
activate “caged” bioactive ruthenium complexes or li-
gands.29,33,36,38,43,62
INTRODUCTION
■
Visible light is an efficient tool to control molecular and
supramolecular metal-based systems,1−10 for applications in
material science,11−14 nanotechnologies,15−25 or medicine.26−45
Among the vast family of photosensitive compounds
ruthenium(II) polypyridyl complexes certainly play a prom-
inent role.18,46 Whereas [Ru(bpy)3]2+-type complexes are
notorious for their luminescence,47−50 complexes bearing
terpyridyl-like ligands, or sterically hindered chelates, have
emerged for their ability to selectively photosubstitute one of
the ligands of the coordination sphere by solvent mole-
cule(s).25,49,51−54 Such reactivity is based on low-lying, metal-
centered (3MC) excited states with dissociative character that
are thermally populated from the photochemically generated
metal-to-ligand charge-transfer (3MLCT) excited states. In such
systems, the photosubstitution reaction can be used to power a
molecular machine20,23,24,55−60 or trigger molecular
switches.12−14,61 More recently, visible light-induced photo-
substitution reactions have been proposed as a new way to
It has been clearly demonstrated, notably by Sauvage et al.,
that in solution the steric properties of the spectator ligands
influence dramatically the quantum efficiency of photo-
substitution reactions.24,63 This phenomenon is interpreted as
a cause of the distortion of the coordination octahedron
induced by steric bulkyness, which in turn lowers the ligand
field splitting energy of the complex and brings the 3MC states
closer in energy to the photogenerated 3MLCT states.
However, the electronic and steric properties of the ligand set
also influence the thermal reactivity of the metal complex. In
principle, the thermal coordination of sterically hindered
ligands requires more thermal energy than that of unhindered
ligands.24 Two decades ago however, Takeuchi et al. reported
the reverse phenomenon in a family of complexes [Ru(terpy)-
(N−N)(L)]2+ (terpy = 2,2′;6′,2″-terpyridine, L = H2O or
Received: May 3, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Thermal Equilibrium between [Ru(terpy)(N−N)(OH2)]2+ and [Ru(terpy)(N−N)(Hmte)]2+, and the
a
Photosubstitution of Hmte by an Aqua Ligand
a
ki are second-order rate constants for the thermal substitution of H2O by Hmte (unit: M−1 s−1), k‑i are first-order rate constants for the thermal
substitution of Hmte by H2O (unit: s−1), Ki the thermodynamic equilibrium constants (unit: M−1), and φi are the photosubstitution quantum yields
(dimensionless). Indexes i refer to the complexes with N−N = bpy (i = 1), N−N = biq (i = 2), N−N = dcbpy (i = 3), and N−N = dmbpy (i = 4).
Scheme 2. Synthesis and Numbering Scheme of [Ru(terpy)(bpy)(Hmte)](PF6)2 ([5](PF6)2)
CH3CN), where the rate of the thermal substitution of the aqua
ligand by acetonitrile at room temperature was increased with
more sterically hindered spectator diimine ligands N−N.64 This
work introduced a quantitative measure of the steric bulkiness
of diimine chelates, but it remained elusive on the reasons for
the higher lability of the aqua ligand observed with hindered
spectator chelates. The reaction was studied at a single
temperature, and based on earlier work65 a dissociative-
interchange substitution mechanism was proposed without
variable-temperature kinetic measurements.
Inspired by these results we recently studied the substitution
reaction of [Ru(terpy)(dcbpy)(OH2)]2+ (dcbpy = 6,6′-
dichloro-2,2′-bipyridine) with 2-methylthioethanol (hereafter,
Hmte) in pure water. At room temperature, binding of the
thioether ligand to afford [Ru(terpy)(dcbpy)(Hmte)]2+ is a fast
reaction.22 We realized that considering the high photo-
substitution quantum yield of the Hmte complex (0.13 at
465 nm) to afford the starting aqua complex [Ru(terpy)-
(dcbpy)(OH2)]2+, this system represented a very interesting
tool in supramolecular chemistry, as the chemical equilibrium
between the aqua and the Hmte ruthenium complexes can be
shifted by visible light, while re-establishing itself in the dark.
We now broaden this work by studying in water the thermal
coordination of Hmte to [Ru(terpy)(N−N)(OH2)]2+ (here-
after, RuOH2) with a series of three bidentate ligands having
different steric demands, namely N−N = bpy (2,2′-bipyridine),
biq (2,2′- biquinoline), or dmbpy (6,6′-dimethyl-2,2′-bipyr-
idine, see Scheme 1). The aim of the present work was double:
first, we wanted to obtain a complete overview of the effect of
sterically hindered substituents on the thermal and photo-
chemical reactivity of Ru(II) complexes in water (Scheme 1).
Second, we aimed at unraveling the mechanism of the thermal
coordination of Hmte to the aqua complex, and to understand
the counterintuitive observation that ligand binding to more
hindered complexes is faster. We demonstrate here that the
higher thermal lability of the hindered complexes is due to a
change in the ligand substitution mechanism, from interchange
associative with the nonhindered bpy N−N chelate, to
interchange dissociative with the most hindered dcbpy and
dmbpy complexes.
B
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
RESULTS
Table 1. Selected Bond Lengths (Å) and Angles (deg) for
[5](PF6)2 and [7](PF6)2
■
22
Synthesis and Crystal Structure. The new complex
[Ru(terpy)(bpy)(Hmte)](PF6)2 ([5](PF6)2) was synthesized
by silver-induced removal of the chloride ligand of [Ru(terpy)-
(bpy)Cl]Cl ([9]Cl) in presence of Hmte at elevated temper-
a
b
[5](PF6)2
[7](PF6)2
Ru1−S1
Ru1−N1
Ru1−N2
Ru1−N3
Ru1−N4
Ru1−N5
Ru1−N4−C20−C21
Ru1−N5−C21−C20
Ru1−N1−C5−C6
Ru1−N3−C11−C10
Ru1−N2−C6−C5
2.3690(5)
2.061(1)
1.961(1)
2.066(1)
2.092(1)
2.064(1)
2.3(2)
2.3819(6)
2.084(2)
1.962(2)
2.074(2)
2.126(2)
2.115(2)
21.5(3)
22.0(3)
2.4(3)
1
atures (see Scheme 2). [5](PF6)2 was characterized by H
NMR and 13C NMR spectroscopy (Supporting Information,
Figure S2), electrospray mass spectrometry (ES-MS), elemental
1
analysis, and electron absorption spectroscopy (UV−vis). H
NMR spectroscopy in acetone-d6 showed that the protons of
the Hmte ligand (3.55 ppm, 2.00 ppm, 1.53 ppm) are shielded
in [5](PF6)2 compared to free Hmte (3.89 ppm, 2.58 ppm,
2.07 ppm) because of coordination to the ruthenium
polypyridyl complex. Single crystals of [5](PF6)2 were obtained
by slow vapor diffusion of toluene into a solution of [5](PF6)2
in Hmte. The crystal structure of the complex was determined
by single-crystal X-ray diffraction (see Figure 1). As expected,
10.5(2)
1.8(2)
5.0(2)
7.6(3)
2.7(2)
4.9(3)
Ru1−N2−C10−C11
2.8(2)
0.7(3)
a
b
This work. Taken from reference 22.
and Hmte in water, was followed by precipitation, but the
resulting salts [6](PF6)2 and [8](PF6)2 were always impure
after chromatography. Preparation of [6]2+ and [8]2+ in
aqueous solution is straightforward, however, as they
spontaneously and quantitatively form upon mixing [10]Cl or
[12]Cl and an excess of Hmte in pure water, thus without
1
addition of AgPF6. According to H NMR in such conditions
[6]2+ or [8]2+ are the only ruthenium species present in
solution (see Figure 2 and Supporting Information, Figure S3).
1
Both complexes were fully characterized in solution by H and
13C NMR, electrospray mass spectrometry, and UV−vis
spectroscopy (see Experimental Part).
Dissolution of the nonhindered bpy complex [9]Cl in water
notoriously leads to a slow equilibrium between the chlorido
complex [9]+ and the aqua complex [Ru(terpy)(bpy)(OH2)]2+
([1]2+).33,66 This equilibrium establishes within hours at room
temperature. By contrast, the chlorido complexes [10]Cl or
[12]Cl are, within minutes at room temperature, fully
hydrolyzed into the aqua species [2]2+ or [4]2+, respectively.
Figure 1. Displacement ellipsoid plot (at 50% probability level) of
complex [5](PF6)2. Hexafluorophosphate counterions and H atoms
were omitted for clarity.
1
Indeed, according to H NMR adding increasing amount of
D2O to CD3OD solutions of [10]Cl or [12]Cl leads, within the
1
time necessary for recording a H NMR spectrum, to the
formation of a second species (see Supporting Information,
Figure S4). In pure D2O, the H NMR spectrum of [10]Cl or
1
the Hmte ligand is coordinated to ruthenium(II) via its soft
sulfur atom. The bpy ligand in [5](PF6)2 is positioned almost
perpendicular to the terpy. The comparison of the crystal
structure of [5](PF6)2 to that of the previously reported
complex [Ru(terpy)(dcbpy)(Hmte)](PF6)2 ([7](PF6)2),22
shows that the torsion angles Ru1−N4−C20−C21 and Ru1−
N5−C21−C20 for the bpy derivative are much smaller than
those of the dcbpy derivative (see Table 1), which suggests that
the coordination sphere is less distorted in [5](PF6)2.
Moreover, the Ru1−S1 bond in [5](PF6)2 is slightly shorter
(2.3690(5) Å) than that in [7](PF6)2 (2.3819(6) Å, see Table
1), also indicating less steric hindrance in [5](PF6)2. These
results are similar to those reported for [Ru(terpy*)(phen)-
(dms)](PF6)2 and [Ru(terpy*)(dmp)(dms)](PF6)2 (terpy* =
4′-(3,5-dit-butylphenyl)-2,2′;6′;2″-terpyridine, phen = 1,10-
phenanthroline, dmp = 2,9-dimethyl-1,10-phenanthroline, and
dms = dimethyl sulfide).63
[12]Cl shows a unique A8 or A5 doublet at 6.75 ppm or 6.78
ppm (see Figure 2 and Supporting Information, Figure S1),
respectively. Aqua Ru(II) complexes are very weak acids in
water, with typical pKa values above 9.5. The pKa of [2]2+ and
[4]2+ were unknown; UV−vis titration led to values of 9.5 and
10.5, respectively (see Supporting Information, Figure S5),
which is comparable to that of [1]2+ (9.7) and [3]2+ (10.9).67,68
As a consequence, complexes [1]2+−[4]2+ are not deprotonated
in pure water near pH 7, and dissolving in Milli-Q water [10]Cl
or [12]Cl produces only the aqua complex [2]2+ or [4]2+,
respectively. We recently published a similar observation for the
dcbpy system.22 Thus, the hydrolysis of the Ru−Cl bond in
water is fast at room temperature with hindered N−N ligands
(biq, dmbpy, or dcbpy), and the hindered chlorido compounds
are good precursors for the corresponding aqua complexes in
nonbasic solutions.
Unlike [5](PF6)2 and [7](PF6)2,22 the RuHmte complexes
[Ru(terpy)(biq)(Hmte)](PF6)2 ([6](PF6)2) and [Ru(terpy)-
(dmbpy)(Hmte)](PF6)2 ([8](PF6)2) could not be isolated in
the solid state. Mixing [Ru(terpy)(biq)(Cl)]Cl ([10]Cl) or
[Ru(terpy)(dmbpy)(Cl)]Cl ([12]Cl), respectively, with AgPF6
As noted above, with hindered complexes (N−N = biq,
dmbpy, or dcbpy22), addition of an excess of Hmte to a
solution of the chlorido precursor complex [Ru(terpy)(N−
N)Cl]Cl (hereafter noted [RuCl]Cl) in pure water leads, in the
dark and at room temperature, to an equilibrium between the
C
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. 1H NMR of a solution of [2]Cl2 (top) and [6]Cl2 (down) in pure D2O near pH 7 (aromatic region, N−N = biq). Conditions: [Ru]tot = 12
mM, [Hmte] = 0 (top) or 0.93 M (bottom), MilliQ water (pH ∼ 7), 298 K. See Supporting Information, Figure S1 for proton attributions.
Figure 3. Evolution of the 1H NMR spectra, at the equilibrium between RuOH2 and RuHmte, with different initial concentrations of Hmte for (a)
the equilibrium between [2]2+ and [6]2+ (N−N = biq); (b) the equilibrium between [4]2+ and [8]2+ (N−N = dmbpy). Condition: (a) [Ru]tot = 5.13
mM, (b) [Ru]tot = 12.7 mM, in D2O, pH ∼ 7 (pure water), T = 297 K, in the dark. The initial amounts of Hmte are indicated on each spectrum.
corresponding aqua species [Ru(terpy)(N−N)(OH2)]2+
([2]2+−[4]2+, noted RuOH2) and the S-bonded Hmte
ruthenium complexes [Ru(terpy)(N−N)(Hmte)]2+ ([6]2+−
[8]2+, noted RuHmte). Thioether ligands are not basic and,
unlike for amine or pyridine ligands where a buffer is required,
here the addition of even large excesses of Hmte to solutions of
the aqua complex [2]2+−[4]2+ does not lead to significant
deviations of the pH from 7. This was also observed upon
adding Hmte to [1]2+, which can be introduced in solution in
the form of [1](PF6)2. Typically, in presence of 0.1 M Hmte a
10−4 M solution of [1](PF6)2, [10]Cl, [11]Cl, or [12]Cl in
Milli-Q water has a pH of 7.2−7.4, that is, the aqua complex
[1]2+, [2]2+, [3]2+, or [4]2+ is not deprotonated. The
substitution of the aqua ligand in [1]2+ by Hmte can be
D
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
studied above 50 °C, whereas for the hindered biq, dcbpy, and
dmbpy system it was studied at room temperature and above
(see below). The overall equilibria for the four systems are
summarized in Scheme 1.
dmbpy); the values were found to be 143(10) M−1 and 37(2)
M−1, respectively, at 297 K, in pure water and in the dark.
These values are both slightly lower than that of the dcbpy
system (K3= 151(8) M−1 in the same conditions,22 see Table
2).
Thermodynamic Study. 1H NMR experiments were
performed in D2O to measure the equilibrium constants K2
and K4 for the equilibria between [2]2+ and [6]2+ (N−N = biq),
and between [4]2+ and [8]2+ (N−N = dmbpy), respectively
(see Scheme 1). For each reaction, NMR samples containing
the RuCl precursor [10]Cl or [12]Cl and different initial
amounts of free Hmte were prepared. After equilibration at 24
[RuHmte]
[RuOH2]
= Ki·[Hmte]
(1)
Knowing the equilibrium constant for each reaction and
using ΔG°i =−RT·ln(Ki), the free Gibbs energies ΔG°2, ΔG°3,
and ΔG°4 were calculated at 297 K to be −12(2) kJ mol−1,
−13(2) kJ mol−1, and −9(1) kJ mol−1, respectively, showing a
lower thermodynamic driving force toward the formation of
RuHmte for the most hindered dmbpy system, in water and at
room temperature (see Table 3). The establishment of the
1
°C in the dark, the H NMR spectrum of each sample was
measured. Integration of the two A8 doublets at 6.35 ([6]2+)
and 6.75 ([2]2+) ppm (N−N = biq), or of the two A5 doublets
at 6.86 ([8]2+) and 6.78 ([4]2+) ppm (N−N = dmbpy), allowed
for calculating the relative amount of RuHmte and RuOH2
present in solution (see Figure 3 and Supporting Information,
Figure S1). A plot of the ratio [RuHmte]/[RuOH2] vs [Hmte]
is shown in Figure 4, where [RuHmte], [RuOH2], and [Hmte]
Table 3. Activation Parameters for the Thermal
Coordination of Hmte to RuOH2 (i) and Thermal
a
Hydrolysis of RuHmte (−i)
ΔG°i (297
ΔG⧧i (297
ΔG⧧
K) (kJ
mol−1
−i
ΔH⧧
ΔS⧧
K)
(297 K)
i
N−N
(kJ·moli−1
)
(J·mol−1·K−1
)
(kJ·mol−1
)
(kJ·mol−1
)
K−1
)
bpy
83(1)
79(3)
93(1)
85(1)
−48(9)
−20(8)
+38(4)
+20(2)
97(5)
85(4)
82(3)
79(3)
117(20)
97(6)
−20(2)
−12(2)
−13(2)
−9(1)
biq
dcbpy
dmbpy
94(4)
88(4)
a
N−N is bpy (i = 1), biq (i = 2), dcbpy (i = 3), or dmbpy (i = 4).
Condition: T = 297 K, in the dark, Milli-Q water, pH ∼ 7.
thermodynamic equilibrium for the unhindered N−N = bpy
system is too slow at room temperature to be measured, and
the corresponding equilibrium constant K1 could not be
obtained directly (see below).
Kinetic Study. Kinetic measurements were performed using
UV−vis spectroscopy to compare the rate of the thermal
substitution of the aqua ligand in [1]2+, [2]2+, and [4]2+ by
Hmte in pure water. After adding a large excess of Hmte to an
aqueous solution of [10]+ or [12]+, the UV−vis spectrum of
each solution with absorption maximum at 549 or 486 nm,
respectively, gradually evolved within minutes in the dark to
give rise to a new absorption maximum at 519 or 463 nm,
corresponding to the Hmte complex [6]2+ or [8]2+,
respectively. Clear isosbestic points (see Figures 5b and 5c)
indicated a selective reaction involving only RuOH2 and
RuHmte. Remarkably, a solution of [1]2+ containing large
excess of the Hmte ligand is kinetically stable at room
temperature, and coordination of the thioether ligand only
takes place at temperatures above 323 K. At such high
Figure 4. Plots of the ratio [RuHmte]/[RuOH2] at the equilibrium at
297 K, as a function of the equilibrium concentration in free Hmte.
[RuHmte] and [RuOH2] represents the concentrations (in mol L−1)
in [6]2+ and [2]2+, respectively (N−N = biq), or in [8]2+ and [4]2+,
respectively (N−N = dmbpy).
represent the concentrations of the thioether complex, of the
aqua complex, and of the free thioether ligand, respectively. For
both reactions straight lines were obtained. According to eq 1
the slope of each line corresponds to the thermodynamic
equilibrium constant K2 (N−N = biq) and K4 (N−N =
Table 2. Thermodynamic and Kinetic Data at 297 K for the Interconversion between [Ru(terpy)(N−N)(OH2)]2+ and
a
[Ru(terpy)(N−N)(Hmte)]2+ Complexes
d
i
N−N
Ki (M−1
)
ki (M−1 s−1
)
t1/2(i) (min)
k−i (s−1
)
t1/2(−i) (min)
b
b
1
bpy
6.8(8) × 10+3
143(10)
151(8)
1.5(9) × 10−8
6.4(1) × 10−3
2.5(1) × 10−2
1.2(5) × 10−1
590(60)
9.0(9)
1.5(4) × 10−8
4.5(9) × 10−5
1.6(9) × 10−4
3.3(9) × 10−3
7.7(7) × 10+5
257(80)
74(9)
2
3
4
biq
c
c
dcbpy
2.3(1)
dmbpy
37(2)
0.43(5)
6.5(5)
a
b
N−N is bpy, biq, dcbpy, and dmbpy. Conditions: in the dark, pure water, pH ∼ 7. Data extrapolated at 297 K from the temperature-dependent
kinetic measurements above 323 K (see text and Table 3). Uncertainties are high but the low rate constant obtained confirms the absence of
c
d
measurable binding of Hmte to the unhindered aqua complex [1]2+ at room temperature. Data taken from reference 22 for comparison. Calculated
for [Hmte] = 0.2 M (t1/2(i)).
E
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Time evolution of the UV−vis spectra of aqueous solutions initially containing (a) [1]2+, (b) [2]2+, and (c) [4]2+, and a large excess of
Hmte in Milli-Q water (pseudofirst order conditions). Conditions: (a) T = 323 K, [Ru]tot = 6.6 × 10−5 M, [Hmte] = 0.067 M, (b) T = 297 K, [Ru]tot
= 6.6 × 10−5 M, [Hmte] = 0.067 M, and (c) T = 297 K, [Ru]tot = 2.1 × 10−4 M, [Hmte] = 0.032 M.
Figure 6. Plots of ln([RuOH2]/[Ru]tot) vs time at different temperatures for the thermal coordination, in the dark and in pure water (pH ∼ 7), of
Hmte to (a) [1]2+, (b) [2]2+, (c) [3]2+,22 and (d) [4]2+. All the numerical values of k′i and ki are given in Supporting Information, Tables S2 and S3.
Conditions: T = 297 K, (a) [Ru]tot = 6.6 × 10−5 M, [Hmte] = 0.067 M, (b) [Ru]tot = 6.6 × 10−5 M, [Hmte] = 0.067 M, (c) [Ru]tot = 1.4 × 10−4 M,
[Hmte] = 0.16 M, and (d) [Ru]tot = 2.1 × 10−4 M, [Hmte] = 0.032 M.
temperatures, the Hmte complex [5]2+ forms selectively, as
shown by the clear isosbestic point at 455 nm and the final λmax
at 450 nm, which is identical to that of the isolated complex
(see Figure 5a). For the two systems N−N = bpy and biq the
plots of ln([RuOH2]/[Ru]tot) vs time were found linear at 323
and 297 K, respectively (Supporting Information, Figure S8),
where [RuOH2] is the concentration in [1]2+ or [2]2+, and
[Ru]tot is the total ruthenium concentration. The pseudo first-
order rate constants k′i (i = 1 or 2) were extracted from the
slopes of these lines (see Figure 6), and a plot of k′i vs [Hmte]
was found linear (Supporting Information, Figure S9), thus
showing that the coordination of Hmte to [1]2+ and [2]2+ is
first order in the ligand Hmte.
For N−N = dmbpy the plot of ln([RuOH2]/[Ru]tot) vs time
at 297 K was not linear (see Figure 6d and Supporting
Information, Figure S8) because with such a sterically hindered
chelate the thermal back-substitution of Hmte by water cannot
be neglected, that is, k−4 becomes comparable to k′4. Equation
2 and 3 give the general expression of the rate of the thermal
formation of the RuHmte complex in pseudo first-order
conditions. Equation 4 was obtained by integration, which
was used to fit the plot ln([RuOH2]/[Ru]tot) vs time and
extract the values of kobs = k−4 + k′4 (see Supporting
Information). Finally, a plot of kobs vs [Hmte] afforded a
straight line, showing that also for N−N = dmbpy the
coordination of Hmte to [4]2+ is first order in Hmte (see
Supporting Information, Figure S9 for the full treatment).
F
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Overall, like for N−N = dcbpy22 the three rate laws for N−N =
bpy, biq, and dmbpy were found to be first order in the Hmte
ligand (see Supporting Information, Figures S8). The second-
order rate constants ki and the half-reaction time t1/2(i)
(calculated with [Hmte] = 0.2 M) are given in Table 2. At
room temperature the N−N = biq and N−N = dmbpy systems
are slower and faster, respectively, compared to the N−N =
dcbpy system. With N−N = bpy Hmte does not coordinate to
[1]2+ at 297 K, but k1 and t1/2(1) can be measured at 323 K
(8.2(5) × 10−4 M−1 s−1 and 71 min at [Hmte] = 0.2 M,
respectively). Even at such high temperatures the rate of the
coordination reaction was found to be 8 times slower compared
to the rate of the N−N = biq system at 297 K (all other
conditions being identical), which highlights the low lability of
the nonhindered bpy system, compared to the sterically
hindered ones.
coordination, but also faster hydrolysis of the Hmte ligand,
while the thermodynamic driving force for Hmte binding to
ruthenium is lowered.
Activation Parameters for the Coordination of Hmte.
To obtain mechanistic information the rate of the thermal
substitution of the aqua ligand by Hmte in [1]2+, [2]2+, [3]2+,
or [4]2+ was studied at different temperatures using UV−vis
spectroscopy. In pseudo first-order conditions the plot of
ln([RuOH2]/[Ru]tot) vs time at different temperatures afforded
straight lines for N−N = bpy, biq, and dcbpy complexes
(Figure 6a−c), which allowed determining the second-order
rate constants ki at different temperatures for all three reactions
(Supporting Information, Table S3). For N−N = dmbpy the
ln([RuOH2]/[Ru]tot) vs time data set was found non linear as
explained above (Figure 6d). They were modeled using eq 4
and the values k4 and k‑4 could also be determined at five
different temperatures (see Supporting Information, Table S2).
The activation enthalpy ΔH⧧ , activation entropy ΔS⧧ , and
d[RuHmte]
= k′4[RuOH2] − k−4[RuHmte]
i
i
activation Gibbs energy at 297 K, ΔG⧧ , are defined, for each
(2)
dt
i
reaction, by the Eyring equation (eq 7). In this equation ki
represents the second-order rate constant, kB is the Boltzmann
constant (1.38 × 10−23 J K−1), h is Planck’s constant (6.63 ×
10−34 J s), and R is the gas constant (8.314 J·mol−1 K−1). An
Eyring plot of ln(ki/T) vs 1/T for the four systems afforded
d[RuHmte]
= k′4[Ru]tot − [RuHmte](k′4 + k−4)
(3)
dt
4
c·e−(k′ +k
−4)t
k′4[Ru]tot
[RuHmte] =
−
straight lines (see Figure 7), from which the values of ΔH⧧ and
(k′4 + k−4)
(k′4 + k−4)
(4)
i
The pseudo first-order rate constants k−i, and half-reaction
times t1/2(−i) for the thermal hydrolysis of the RuHmte
complexes with N−N = biq, dcbpy, and dmbpy, in the dark
and at 297 K, were determined from the knowledge of the
thermodynamic equilibrium constants Ki, and the second-order
rate constants ki (see Table 2). Equation 5, written at the
equilibrium, indeed rearranges into eq 6:
k−i·[RuHmte]eq = ki·[Hmte]eq ·[RuOH2]eq
(5)
ki
Ki
k−i
=
(6)
For the N−N = bpy system measuring K1 was not possible at
room temperature and k−1 could not be calculated. However,
k−1 could be obtained experimentally by heating an aqueous
solution of [5](PF6)2 at high temperatures (>343 K), and
monitoring by UV−vis spectroscopy the thermal substitution of
Hmte by water at different temperatures. Subsequently, the
activation parameters for the thermal hydrolysis of [5]2+ were
extracted via an Eyring plot (see Supporting Information,
Figure S10 and Table S4): values of 110(6) kJ mol−1 and
Figure 7. Eyring plots for the thermal substitution of H2O by Hmte
for [Ru(terpy)(N−N)(OH2)]2+ in pure water, where N−N is (a) bpy,
(b) biq, (c) dcbpy, and (d) dmbpy. The slope of the line is −ΔH⧧ /R,
i
and the y-intercept is ln(kb/h) + ΔS⧧ /R. See Table 3 for numerical
i
values.
ΔS⧧ could be extracted. The activation Gibbs energies, ΔG⧧ ,
i
i
were calculated at 297 K using the equation ΔG⧧ = ΔH⧧
−22(15) J mol−1 K−1 were found for ΔH⧧ and ΔS⧧
,
−1
i
i
−1
−TΔS⧧ (see Table 3).
i
respectively. By extrapolation of the values of k−1 at T > 323 K,
the value of ΔG⧧ and k−1 at 297 K were calculated to be
117(10) kJ mol−1−1and 1.5(9) × 10−8 s−1, respectively. The
equilibrium constant K1 at room temperature (6.8(8) × 10+3
M−1) was obtained using eq 6 and the extrapolated value of k1
at 297 K (see below and Table 2). These extrapolated values
for N−N = bpy are less precise than the direct measurements
done for N−N = biq, dcbpy, and dmbpy considering the
significant error on ΔS⧧−1. However, they give a qualitative
information about how stable and inert the nonhindered
complex [5]2+ is. Finally, comparing the kinetic data in Table
2 shows that the thermal lability of both species RuOH2 and
RuHmte increases along the series bpy, biq, dcbpy, dmbpy, that
is, upon increasing the steric hindrance of the spectator N−N
ligands. Such higher lability results in faster thermal
ki
T
−ΔHi⧧
kB
h
ΔSi⧧
R
1
T
ln
=
·
+ ln
+
(7)
R
Quite surprisingly the four activation enthalpies were found
too similar to account for the clear differences in reactivity
between the four N−N ligands. By contrast, unexpected
differences in activation entropies were observed: the values for
the less hindered N−N = bpy and N−N = biq bidentate ligands
were found negative, whereas for the more hindering chelates
N−N = dcbpy and N−N = dmbpy the values were found
positive. When both contributions of enthalpy and entropy are
taken into account, a clear trend was observed: the activation
Gibbs energies ΔG⧧ decrease along the series bpy, biq, dcbpy,
i
dmbpy. Such acceleration of the coordination of Hmte to the
G
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 4. Photochemical and Thermal First-Order Rate Constant Values for a Typical Visible Light Irradiation Experiment with
a
Interconversion between [Ru(terp)(N−N)(OH2)]2+ and [Ru(terpy)(N−N)(Hmte)]2+
N−N
[Hmte] (M)
k′i (s−1
7.3 × 10−5
2.2 × 10−4
1.8 × 10−2
)
k−i (s−1
)
kφi (s−1
)
Φ (Einstein s−1
)
φi
biq
0.011
0.010
0.20
4.4 × 10−5
1.1 × 10−4
4.5 × 10−2
4.2 × 10−4
1.1 × 10−3
2.0 × 10−3
9.8 × 10−10
3.9 × 10−9
3.9 × 10−9
0.12(5)
0.13(5)
0.30(6)
dcbpy
dmbpy
a
(N−N = biq, dcbpy, or dmbpy). Conditions: T = 297 K, solvent = Milli-Q water (pH ∼ 7). The photon flux Φ is indicated.
aqua complex appears to be a consequence of a drastic increase
[RuHmte]eq
[RuOH2]eq
ki[Hmte]
of the activation entropy ΔS⧧ , that is, a change in the
(a)
=
i
k−i
substitution mechanism, rather than a simple destabilization of
RuOH2, which would lead to a decrease of the activation
[RuHmte]ss
[RuOH2]ss
ki[Hmte]
k−i + kφi
(b)
=
enthalpy ΔH⧧ (see discussion).
i
(8)
(9)
These variable-temperature measurements also allowed us to
obtain the values of ΔG⧧−i for the thermal substitution of Hmte
Φ·ϕ·(1 − 10−A
)
e
by water in [6]2+, [7]2+, or [8]2+, from the values of ΔG⧧ and
i
i
kφi
=
ΔG°i, and using the equation ΔG⧧ = ΔG⧧ − ΔG°i. Upon
n
−i
i
Ru(tot)
notably increasing the steric hindrance of the bidentate chelate,
ΔG⧧−i was found to decrease as well (see Table 3), that is, the
coordinated Hmte ligand becomes more and more labile in
water. Overall, our data clearly indicate that increasing the
bulkiness of the substituent on the bidentate chelate N−N
increases the lability of both monodentate ligands (H2O and
Hmte), whereas it decreases the thermodynamic driving force
for the formation of the RuHmte species.
First the value of k‑i was obtained in the dark from eq 8(a)
knowing the value of ki . Then the value of kφi can be obtained
under irradiation using eq 8(b), and from the values of kφi the
photosubstitution quantum yields φi were calculated using eq 9.
Numerical values φ2 = 0.12(5) (at 520 nm), φ3 = 0.13(5) (at
465 nm), and φ4 = 0.30(6) (at 465 nm) were found for the biq,
dcbpy, and dmbpy systems, respectively, at 297 K. These values
are significantly higher than φ1, as expected for sterically
hindered complexes. The value of φ3 found by this method was
close to that obtained by us using a more direct method
(0.097(9)).22
Photochemistry. Quantum Yield Determination. Ruthe-
nium polypyridyl complexes are known for their ability to
photosubstitute a ligand of the coordination sphere by a solvent
molecule upon visible light irradiation.20,21,25,46,51 The Ru−S
bond of [5]2+, [6]2+, [7]2+, or [8]2+ can indeed be cleaved by
visible light irradiation in water, to afford the ruthenium aqua
complexes [1]2+, [2]2+, [3]2+, or [4]2+, respectively (see
Scheme 1). This photochemical process comes in addition to
the thermal hydrolysis of the Hmte complex, the kinetics of
which significantly varies depending on the steric hindrance of
the bidentate chelate N−N (see above and Table 2). Different
methods were used for measuring the photosubstitution
quantum yields φi of the four ruthenium compounds [5]2+−
[8]2+ (see Supporting Information). For [5]2+ full conversion
to [1]2+ is obtained after 30 min irradiation at 452 nm using a
1000 W Xe lamp fitted with a bandpath filter. The
photochemical reaction can be followed by UV−vis spectros-
copy (Supporting Information, Figure S11), and a value of
0.022(6) was found for φ1 at room temperature and at 452 nm,
which is consistent with previous work.33
On the other hand, measuring the photosubstitution
quantum yields φ2, φ3, and φ4 for [6]2+, [7]2+, and [8]2+,
respectively, was challenging because of the rapid equilibrium
between RuHmte and RuOH2.22 For these compounds
standard measurements cannot be realized, so that another
method was used consisting in the perturbation with light of
the thermal equilibrium between RuOH2 and RuHmte (see
Supporting Information). In short, the ratio [RuHmte]eq/
[RuOH2]eq is measured by UV−vis at the equilibrium in the
dark (eq), and compared to the ratio [RuHmte]ss/[RuOH2]ss at
the steady state under visible light irradiation (ss). Both ratios
can be expressed as a function of k′i, k‑i, and kφi (eq 8a and 8b),
where kφi is a first-order rate constant for the photochemical
substitution of Hmte by H2O (unit: s−1, see eq 9 and
Supporting Information).
Interestingly, comparing (Table 4) the pseudo first-order rate
constant for the thermal substitution of H2O by Hmte, k′i, and
the first-order rate constants k‑i and kφi for the thermal and
photochemical substitution of Hmte by H2O, respectively,
highlights that with N−N = biq or N−N = dcbpy the values of
kφi are 1 order of magnitude higher than that of k′i and k−i. In
contrast, for N−N = dmbpy kφ4 is 1 order of magnitude lower
than k−4 and k′4. Thus, by increasing too much the steric
hindrance of the spectator diimine bidentate ligand (N−N =
dmbpy), the thermal lability of Hmte increases to a point where
the light-induced shifting of the thermal equilibrium between
RuOH2 and RuHmte becomes difficult to realize. For such
compounds shifting appreciably the equilibrium in favor of the
aqua complex would require much higher light intensities. For
N−N = biq and N−N = dcbpy low light intensities efficiently
perturb the thermal equilibrium between RuOH2 and RuHmte.
As shown in Supporting Information, Figure S12b and S12c,
during light irradiation the ratio [RuHmte]/[RuOH2] varies
significantly: a steady state can be reached where Ru is mostly
bound to H2O, whereas in the dark it is mostly bound to Hmte.
Thus, moderately hindered compounds such as those with biq
and dcbpy represent a better compromise between thermal and
photochemical lability, and afford a light-sensitive Ru−S
coordination bond in water. By contrast, the thermal reactivity
of nonhindered (N−N = bpy) or too hindered (N−N =
dmbpy) complexes is either too low, or too high, respectively.
Reversibility of the Light-Induced Equilibrium Shift.
We recently showed that the blue light-induced shifting of the
equilibrium between RuOH2 and RuHmte in water for the N−
N = dcbpy system could be repeated at least up to four cycles at
room temperature.22 Considering the similar kinetic properties
of the N−N = biq system, we repeated these studies for [6]2+
using green light. The thermal equilibrium between [6]2+ and
H
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
[2]2+ in water was perturbed by light irradiation (λe = 520 nm)
for a period of 45 min, followed by a dark period of 90 min.
This cycle was repeated four times, and the state of the system
was monitored by UV−vis spectroscopy. The time evolution of
the ratio [RuOH2]/[Ru]tot is shown in Figure 8. Like for N−N
system is robust and only contains the two ruthenium
complexes [2]2+ and [6]2+ that interconvert upon switching
on and off a source of green light. Like for N−N = dcbpy, the
Ru−S coordination bond forms spontaneously in the dark and
is cleaved by visible light irradiation.
DISCUSSION
■
Following previous work of Takeuchi,64,65 Rack,12−14 or
Sauvage25,63 on the influence of steric hindrance on the
photoreactivity of polypyridyl ruthenium(II) compounds we
more recently realized22 that in the dark the Ru−S coordination
bond of hindered complexes such as [Ru(terpy)(dcbpy)-
(Hmte)]2+ spontaneously forms at room temperature and in
neutral aqueous solutions, while still keeping a very high
sensitivity to visible light irradiation. As dark formation and
photochemical breakage can both occur, such systems open
new possibilities for building supramolecular systems driven by
visible light irradiation. However, the higher lability observed
with the dcbpy complex seemed counterintuitive: for other
light-sensitive complexes such as [Ru(phen)2(dmbpy)]2+ (phen
= 1,10-phenanthroline, dmbpy = 6,6′-dimethyl-2,2′-bipyridine),
steric hindrance leads to efficient photosubstitution25 indeed,
but also to a difficult thermal binding of the hindered chelate to
the Ru center. The present study was undertaken to understand
the relationship between thermal lability and steric hindrance
for ruthenium complexes of the [Ru(terpy)(N−N)(Hmte)]2+
family, and to gather temperature-dependent kinetic data that
had been overlooked in the past.
Figure 8. Plot of the ratio [RuOH2]/[Ru]tot vs time upon switching
ON and OFF several times a source of green light (λe = 520 nm) in
presence of [6]2+ and [2]2+. Conditions: T = 297 K, Milli-Q water (pH
∼ 7); photon flux Φ = 9.8(5) × 10−9 Einstein s−1; [Ru]tot = 8.6 × 10−5
M, [Hmte] = 0.011 M, spectra measured every 1 min.
First, it might be noticed that the substitutents in ortho
position to the coordinated nitrogen atoms of N−N = biq,
dcbpy, and dmbpy do not only increase the steric bulk of the
coordination sphere around the metal, but they also exert
electronic effect on the metal center, which may in turn
= dcbpy, the N−N = biq system shows reversible light-induced
shift of the equilibrium between [6]2+ and [2]2+, and no sign of
degradation was observed after four cycles. The composition of
the solution varies between 45% of [2]2+ in the dark and up to
85% of [2]2+ after irradiation. These results show that the biq
Scheme 3. Proposed Transition States for the Substitution of the Aqua Ligand in [Ru(terpy)(N−N)(H2O)]2+ by Hmte, Where
a
b
(a) N−N = bpy, biq and (b) N−N = dcbpy, dmbpy
a
More compact transition state with hydrogen-binding contributing to a loose hepta-coordinated transition state.
Less compact transition state.
b
I
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
influence the rates of ligand substitutions. This effect can be
seen for example on the absorption maximum of the RuHmte
complexes [5]2+−[8]2+, which lies at significant higher
wavelength for N−N = biq (λmax = 519 nm) than for N−N
= bpy, dcbpy, or dmbpy (λmax = 450, 467, and 463 nm,
respectively). These electronic effects might play a role in fine-
tuning the activation enthalpies and entropies of the thermal
substitution reactions. However, one substituent of the
bidentate chelate and the monodentate ligand coordinated to
the metal center lie in very close spatial proximity, thus leading
to significant distortion of the geometry in the ground state
(compare for example the X-ray structures of [5]2+ and [7]2+).
Thus, in the following discussion we mostly interpret the
change in mechanism along the series bpy, biq, dcbpy, dmbpy
as a consequence of the increasing steric demands of the
spectator diimine chelate.
CONCLUSION
■
The thermodynamic, kinetic, and photochemical properties of a
series of polypyridyl ruthenium complexes [Ru(terpy)(N−
N)(L)]2+ where N−N is bpy, biq, dcbpy, or dmbpy, and L is
H2O or Hmte, have been determined in water near neutral pH.
Our data provide a global understanding of the influence of the
N−N chelate on the reactivity of these systems. Qualitatively, a
global acceleration of all thermal and photochemical ligand
exchange processes is observed when the steric hindrance of the
spectator diimine chelate is increased. Variable-temperature
kinetic data show that the increased lability of the monodentate
ligand with hindered N−N chelates is due to entropy, and that
the mechanism of the thermal ligand substitution reaction
changes from interchange associative to interchange dissociative
following the series N−N = bpy, biq, dcbpy, dmbpy. Analysis of
the relative values of the rate constants for the thermal and
photochemical ligand substitution reactions also shows that by
increasing the steric hindrance too much (N−N = dmbpy) the
lability in the dark becomes so high that no appreciable change
of the composition of the solution can be obtained by light
irradiation, unless exceptionally intense light would be used.
With intermediate steric hindrance (N−N = biq or dcbpy) the
Ru−S bond forms spontaneously in the dark at room
temperature, but it is efficiently cleaved under mild irradiation,
which will allow using these systems in supramolecular
chemistry. With the nonhindered ligand N−N = bpy, the
photosensitivity of the Hmte complex is lower, and the
monodentate ligands (Hmte and H2O) are nonlabile at room
temperature. Overall, changing the N−N bidentate ligand
appears as an efficient means to tune the thermal and
photochemical reactivities of [Ru(terpy)(N−N)L]2+ com-
plexes.
Usually, the higher thermal lability for sterically hindered
complexes is explained in terms of destabilization of the
ground-state hexacoordinated species, compared to the
transition state of the thermal substitution reaction. In such
interpretation, the reaction always follows a dissociative
interchange mechanism.64,69−72 Applied to our system, this
explanation should lead to enthalpy (ΔH⧧ ) being the main
i
reason for the decreased activation Gibbs energies (ΔG⧧ )
i
when going from N−N = bpy to N−N = dmbpy. However, our
data suggest that the increased lability of the hindered
complexes in water is due to variations of the entropic term
(ΔS⧧ ) in Eyring’s equation. Although ΔS⧧ values are known to
i
i
contain significant experimental error and may be less accurate
than, for example, activation volumes ΔV⧧ , the similarities in
i
ΔH⧧ for the four systems and the clear differences in ΔS⧧ , as
i
i
seen in Figure 7, allow drawing some mechanistic conclusions.
Considering that for all four systems the rate law is first order in
Hmte we conclude that there is a shift in the mechanism of the
thermal substitution of H2O by Hmte, from interchange
associative with N−N = bpy and biq, marked by ΔSi⧧ < 0, to
interchange dissociative with N−N = dcbpy and dmbpy,
marked by ΔSi⧧ > 0.73−82 As shown in Scheme 3, H2O is still
present in the coordination sphere when the Ru−S bond-
making occurs, and in an interchange mechanism bond making
occurs before the second coordination sphere has had time to
relax. For less bulkier chelates (N−N = bpy, biq) the Ru−S
bond-making is essentially synchronous with the Ru−O bond-
breaking (Ia mechanism). Hydrogen bonding between Hmte
and the aqua ligand may also contribute to stabilizing the hepta-
coordinated transition state. Thus, a more compact transition
state and more constraints for the unhindered chelates N−N =
bpy and biq lead to negative values for the activation entropy,
and thus to significantly (bpy) or slightly (biq) lower
substitution rate constants. In contrast, for bulkier systems
the Ru−S bond making only occurs when Ru−OH2 is already
partially broken, but before H2O exits from the second
coordination sphere (Id mechanism). Thus, there is no
formation of a coordinatively unsaturated and potentially
highly reactive pentacoordinated state, which would cancel
the dependence of the substitution rate law in [Hmte]. The less
compact transition state for N−N = dcbpy and dmbpy
increases the degrees of freedom of both incoming and leaving
monodentate ligands, thus resulting in positive activation
entropies for the substitution process, which significantly
enhances its rate constants.65,69,77,82−87
EXPERIMENTAL SECTION
■
Synthesis. 1H and 13C NMR spectra were recorded using a Bruker
DPX-300 spectrometer; chemical shifts are indicated in ppm relative to
TMS. Electrospray mass spectra were recorded on a Finnigan TSQ-
quantum instrument using an electrospray ionization technique (ESI-
MS). UV−vis spectra were obtained on a Perkin-Elmer Lambda 900
spectrophotometer or on a Varian Cary 50 UV−visible spectrometer.
The classical routes for synthesizing [Ru(terpy)(biq)(Cl)]Cl ([10]-
Cl),64 [Ru(terpy)(dmbpy)(Cl)]Cl ([12]Cl), and [Ru(terpy)(dcbpy)-
(Cl)]Cl ([11]Cl),67 were modified (see Supporting Information).
[Ru(terpy)Cl3],88 6,6-dichloro-2,2′-bipyridine,89 [Ru(terpy)(bpy)-
(Cl)]Cl ([9]Cl), [Ru(terpy)(bpy)(OH2)](PF6)2 ([1](PF6)2),33,68
and [Ru(terpy)(dcbpy)(Hmte)](PF6)2 ([7](PF6)2),22 were synthe-
sized following literature procedures. 2,2′;6′,2″-terpyridine was
purchased from ABCR GmbH & Co.KG. 2,2′-bipyridine, 6,6′-
dimethyl-2,2′-bipyridine, 2,2′-biquinoline, 2-(methylthio)-ethanol
(Hmte), and AgPF6 were purchased from Sigma-Aldrich and used as
such.
[Ru(terpy)(bpy)(Hmte)](PF6)2 ([5](PF6)2). [9]Cl (56 mg, 0.10
mmol) and AgPF6 (57 mg, 0.22 mmol) were dissolved in 3:5
acetone/H2O mixture (16 mL). To this solution was added Hmte (90
μL, 1.0 mmol). The mixture was refluxed under argon for 8 h in the
absence of light, after which it was filtered hot over Celite. Evaporation
of the filtrate gave an orange solid, which was taken up in acetone and
reprecipitated with Et2O. Filtration of the suspension yielded
[5](PF6)2 as an orange powder (69 mg, 79%).1H NMR (300 MHz,
Acetone, 297 K, see Supporting Information, Figure S2) δ 9.95 (d, J =
5.6 Hz, 1H, A6), 9.03−8.87 (m, 3H, A3+T3′), 8.78 (d, J = 8.1 Hz, 2H,
T3), 8.72 (d, J = 8.2 Hz, 1H, B3), 8.59−8.42 (m, 2H, A4+T4′), 8.26−
8.09 (m, 3H, A5+T4), 8.09−7.94 (m, 3H, B4+T6), 7.63−7.47 (m, 3H,
B6+T5), 7.31 (t, J = 6.6 Hz, B6, B5), 3.55 (t, J = 5.5 Hz, 2H, S-CH2−
CH2), 2.03−1.97 (m, 2H, S−CH2), 1.53 (s, 3H, S-Me). 13C NMR (75
MHz, Acetone, 297 K) δ 158.48 + 157.94 + 157.23 + 157.18
J
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(B2+A2+T2+T2′), 153.89 (T6), 152.61 (A6), 150.56 (B6), 139.34
(T4), 138.67 + 138.58 (B4+A4), 137.42 (T4′), 129.03 (T5), 128.37
(A5), 127.71 (B5), 125.45 (T3), 125.16 (A3), 124.76 (T3′), 124.30
(B4), 58.37 (S-CH2-CH2), 37.17 (S-CH2), 14.38 (S-Me). UV−vis:
λmax (ε in L·mol−1·cm−1) in pure H2O: 450 nm (6600). ES MS m/z
(calc): 728.0 (727.7 [M − PF6]+), 582.1 (581.7 [M − 2 PF6 − H]+),
261.5 (261.3 [M − 2PF6 − Hmte + MeOH]2+). Anal. Calcd for
C28H27F12N5OP2RuS: C, 38.54; H, 3.12; N, 8.03; S, 3.67. Found: C,
38.25; H, 3.41; N, 7.94; S, 3.78. Crystal growing: Large single crystals of
compound [5](PF6)2 were grown by vapor diffusion of toluene into a
solution of [1](PF6)2 in Hmte (∼10 mg in 0.5 mL mte). Crystal
structure data: [C28H27N5ORuS](PF6)2; Fw = 872.62, red block, 0.45
× 0.25 × 0.24 mm3, monoclinic, C2/c (no. 15), a = 24.06815(17), b =
10.86063(8), c = 24.69614(19) Å, β = 93.6407(7)°, V = 6442.43(8)
Å3, Z = 8, Dx = 1.799 g cm−3, μ = 0.755 mm−1, abs. corr. range: 0.769−
0.867. 31610 Reflections were measured up to a resolution of (sin θ/
and 160 μL of D2O were added successively to each NMR tube, and
1H NMR spectra were recorded after each addition (see Supporting
Information, Figure S4).
Equilibrium Constant Determination. (a) For N−N = biq ([2]Cl2
⇆ [6]Cl2): A stock solution A of [10]Cl (17 mg in 5.0 mL D2O, 5.1
mM) and a stock solution B of Hmte (92 mg Hmte in 2.0 mL D2O,
0.50 M) were prepared. Eight NMR tubes containing 0.50 mL of
solution A (2.5 μmol [10]Cl) were prepared, and to each tube was
added 2.5 μL, 5.0 μL, 8.0 μL, 10 μL, 15 μL, 26 μL, 34 μL, or 35 μL
solution B, resulting in 0.50, 1.0, 1.6, 2.0, 3.0, 5.2, 6.8, or 7.0
equivalents of Hmte, respectively. The NMR tubes were put in a water
bath for 30 min at 50 °C and left standing overnight at room
1
temperature. After equilibration, H NMR spectra of all samples were
measured at room temperature, to determine the relative integral of
[6]2+ and [2]2+. Then the ratios [RuHmte]/[RuOH2] were
determined by integration of the peaks at 6.35 and 6.75 ppm
corresponding to [6]2+ and [2]2+, respectively, where [RuHmte]
represents the concentration in [6]2+ and [RuOH2] the concentration
in [2]2+. A plot of [RuHmte]/[RuOH2] as a function of equilibrium
concentration in Hmte was made. The slope of the plot numerically
corresponds to K2 (see Figure 4 and eq 1).
λ)max = 0.65 Å−1. A total of 5673 reflections were unique (Rint
=
0.0367), of which 5375 were observed [I > 2σ(I)]. 511 Parameters
were refined with 195 restraints. R1/wR2 [I > 2σ(I)]: 0.0207/0.0517.
R1/wR2 [all refl.]: 0.0226/0.0525. S = 1.051. Residual electron density
found between −0.55 and 0.37 eÅ−3.
[Ru(terpy)(biq)(Hmte)]2+ ([6]2+). [10]Cl (4.0 mg, 6.0 μmol) was
dissolved in D2O (0.50 mL). To this solution a large excess of Hmte
(50 μL, 0.51 mmol) was added and stirred for 5 min. The mixture was
(b) For N−N = dmbpy ([4]Cl2 ⇆ [8]Cl2): A stock solution C of
[12]Cl (40 mg in 5.0 mL D2O, 13 mM) was prepared. NMR samples,
each containing 0.50 mL of stock solution C (6.4 μmol [12]Cl) were
prepared. To each NMR tube was added a known amount of pure
Hmte (0.60 μL, 1.2 μL, 1.8 μL, 2.4 μL, 3.0 μL, 4.5 or 6.0 μL) to give
1.0, 2.0, 3.0, 4.0, 5.0, 7.5, 10, or 20 equivalents, respectively. Each
NMR tube was stirred for 5 min and then left to stand for more than
1
kept for 3 h at 80 °C in a water bath. According to H NMR and ES
MS, [6]2+ is the only ruthenium species in solution. (For atom
1
numbering see Supporting Information, Figure S1) H NMR (300
MHz, D2O, 297 K) δ 8.97 (dd, J = 19.4, 8.8 Hz, 2H, B3+B4), 8.66 (d, J
= 8.2 Hz, 4H, B8+T3′+A3), 8.47 (d, J = 8.0 Hz, 2H, T3), 8.43−8.33
(m, 3H, B5+A4+T4′), 8.14−7.92 (m, 6H, T4+B6+B7+T6), 7.85 (d, J
= 7.4 Hz, 1H,A5), 7.55−7.40 (m, 3H, T5+A6), 7.25 (t, J = 7.9 Hz,
1H,A7), 6.50 (d, J = 8.8 Hz, 1H,A8), 3.30 (t, J = 7.1, 4.4 Hz, 2H,S-
CH2-CH2), 1.54 (t, J = 5.8 Hz, 2H, S-CH2-CH2), 1.03 (s, 3H, CH3-S).
13C NMR (75 MHz, D2O, 297 K) δ 159.83 + 159.80 (T2+T2′),
158.17 + 158.00 (A2+B2), 153.46 (T6), 150.19 + 149.91 (A8a+B8a),
140.32 + 139.36 (B8+A8), 139.46 (T4), 137.94 (T4′), 133.09
(B6+A6), 130.60 + 130.05 (A7+B7), 129.75 + 128.80 (B4a+ A4a),
129.64 + 129.04 (A4+B4), 128.76 (T5), 126.92 + 123.06 (B5+A5),
125.03 + 124.34 (T3+T3′), 121.24 + 120.87 (A3+B3), 57.32 (S-CH2-
CH2), 46.78 (S-CH2-CH2), 8.48 (CH3-S).UV−vis (see Supporting
Information): λmax (ε in L·mol−1·cm−1) in pure H2O: 519 nm (5600).
ES MS m/z (calc): 682.0 (682.1 [M − 2Cl−H]+), 295.5 (295.3 [M −
2Cl−Hmte]2+).
1
10 min at room temperature. After equilibration, H NMR spectra of
all samples were measured at room temperature. The ratio [RuHmte]/
[RuOH2] were determined by integration of the peaks at 6.86 and 6.78
ppm, where [RuHmte] represents the concentration in [8]2+ (δ = 6.86
ppm) and [RuOH2] the concentration in [4]2+ (δ = 6.78 ppm). A plot
of [RuHmte]/[RuOH2] as a function of equilibrium concentration in
Hmte was made. The slope of the plot numerically corresponds to K4
(see Figure 4 and eq 1).
The values for Gibbs free energy ΔG°i at 297 K were calculated for
both reactions using the equation ΔGi°= −RT ln Ki.
Kinetics. A Perkin-Elmer Lambda 900 UV−vis spectrometer
equipped with stirring and temperature control was used for kinetic
experiments. The measurement procedure and values of the extinction
coefficients of all aqua and Hmte complexes used in the kinetic study is
described in the Supporting Information, Figure S6 and Table S1. The
experimental procedure for the calculation of the rate constants at 297
K from the slope of a plot of k′i vs [Hmte] is explained in the
Supporting Information.
[Ru(terpy)(dmbpy)(Hmte)]2+ ([8]2+). [12]Cl (4.0 mg, 6.8 μmol)
was dissolved in D2O (0.50 mL). To this solution a large excess of
Hmte (24 μL, 0.28 mmol) was added. The mixture was stirred for 5
min. The compound was not isolated as it would react back to [12]Cl
upon evaporation of water. According to 1H NMR and MS [8]2+ is the
only ruthenium species present in solution. (For atom notations see
Thermal Substitution of H2O by Hmte on RuOH2 Complexes
(ki, ΔH⧧ , ΔS⧧ , ΔG⧧ ). Stock solutions D of complex [1](PF6)2 (2.0
i
i
i
mg in 25 mL of H2O, 1.0 × 10−4 M), E of [10]Cl (1.6 mg in 25 mL of
H2O, 1.0 × 10−4 M), F of [11]Cl (3.7 mg in 25 mL of H2O, 2.1 ×
10−4 M), G of [12]Cl (3.5 mg in 25 mL of H2O, 2.2 × 10−4 M), and
H and I of Hmte (460 mg in 25.0 mL of H2O, 2.00 × 10−1 M (H), and
438 mg in 10.0 mL of H2O, 4.70 × 10−1 M (I)) were prepared. 2.0 mL
of D, E, F, or G was added to a UV−vis cuvette, which was placed in
the UV−vis spectrometer. The temperature was set at 50, 60, 70, or 80
°C for D, 24, 28, 35, 42, or 50 °C for E, and 10, 15, 20, 24, or 28 °C
for F and G. After obtaining a constant temperature in each cuvette,
1.0 mL of H was added to D and E, or 1 mL of I to F, or 0.8 mL of
H2O plus 0.2 mL of I to G, for each experiment at each temperature
(final Hmte and Ru concentrations for each experiments are given in
Supporting Information, Tables S2 and S3). In such conditions, Hmte
is in large excess (pseudo first-order condition). After addition of
Hmte, a UV−vis spectrum was taken every 60 s for D and every 30 s
for E, F, or G. For each spectrum, the concentrations in [RuHmte]
and [RuOH2] were determined by deconvolution of the UV−vis
spectra knowing the extinction coefficients of both RuHmte and
RuOH2 species (see Supporting Information and Table S1). The
pseudo first order rate constants k′i at each temperature for each
sample D, E, F, or G were determined from the slope of the plot of
ln([RuOH2]/[Ru]tot) vs time, and ki were then calculated knowing the
1
Supporting Information, Figure S1) H NMR (300 MHz, D2O, 297
K) δ 8.57 (d, J = 8.2 Hz, 2H, T3′), 8.45 (d, J = 8.0 Hz, 3H, B3+T3),
8.26 (t, J = 8.1 Hz, 1H, T4′), 8.21−8.01 (m, 6H, B4+A3+T4+T6),
7.81 (d, J = 7.6 Hz, 1H, B5), 7.61 (t, J = 7.9 Hz, 1H, A4), 7.55−7.46
(m, 2H, T5), 6.88 (d, J = 7.6 Hz, 1H, A5), 3.26 (t, J = 5.7 Hz, 2H, S-
CH2−CH2), 3.09 (s, 3H, H7), 1.39 (t, J = 5.6 Hz, 2H, S−CH2), 1.27
(s, 3H, H7′), 0.87 (s, 3H, S-Me). 13C NMR (300 MHz, D2O) δ 165.61
+ 164.45 (B6+A6), 158.94 + 158.38 (T2+T2′), 158.12 + 158.06
(B2+A2), 153.82 (T6), 138.99 (T4), 138.33 (A4), 138.15 (B4),
136.95 (T4′), 128.58 (T5), 127.76 (A5), 127.32 (B5), 124.61 (T3),
123.77 (T3′), 121.80 (A3), 121.38 (B3), 56.46 (HO-CH2-), 34.71
(Me-S-CH2-), 26.86 (A7), 22.00 (B7), 11.70 (Me-S). UV−vis (see
Supporting Information): λmax (ε in L·mol−1·cm−1) in pure H2O: 463
nm (5700). ES MS m/z (calc): 610.1 (609.8 [M − 2Cl − H]+), 305.6
(305.3 [M − 2 Cl]2+).
General Procedure for the Hydrolysis in CD3OD of [Ru(terpy)(N−
N)(Cl)]Cl ([10]Cl, [11]Cl, or [12]Cl, N−N = biq, dcbpy, or dmbpy).
Three NMR samples of compound [10]Cl (2.2 mg, 3.3 × 10−3
mmol), [11]Cl (2.8 mg, 4.8 × 10−3 mmol), or [12]Cl (2.9 mg, 4.6 ×
10−3 mmol) were dissolved in MeOD (500 μL). An 1H NMR
spectrum was recorded for each sample. Then, 20 μL, 40 μL, 80 μL,
K
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
concentration of Hmte in the solution (see Supporting Information,
Tables S2 and S3). By plotting ln(ki/T) as a function of 1/T for each
sample, the activation enthalpy and entropy were calculated from the
slope and y-intercept of the Eyring plot, respectively. ΔGi⧧ at 297 K
was calculated for each reaction using the equation ΔGi⧧ = ΔHi⧧ − T
× ΔSi⧧ (see Table 3).
1 mL of a solution of Hmte (31 mg in 10 mL of H2O, 0.030 M) was
prepared and stirred overnight to reach equilibrium at 24 °C. Then,
UV−vis spectra were measured, once in the dark, and then during 45
min under irradiation using LED lamp at λe= 520 nm. After 45 min the
LED lamp was switched off, and UV−vis spectra were measured for 90
min in the dark (1 min interval between each spectrum, either under
irradiation or in the dark). The cycle was repeated 3 more times for a
total experimental time of 9 h (see Figure 8). For each spectrum
[RuHmte] and [RuOH2], that is, the concentration in [6]2+ and [2]2+,
respectively, were determined by deconvolution, knowing the
extinction coefficients of both species (see Supporting Information,
Table S1). By calculating the ratio [RuHmte]/[RuOH2] at the
equilibrium in the dark (eq 8a) and at the photochemical steady state
(eq 8b), reporting the second order rate constant k2 and the photon
flux ϕ, the quantum yield φ2 was calculated using eq 9, to be 0.12(5)
(see Supporting Information, Table S5 for all numerical values).
Determination of the Photosubstitution Quantum Yield for
[8]2+ (φ4). 2.0 mL of a stock solution of [8]Cl (7.0 mg in 50 mL H2O,
2.2 × 10−4 M) was put in a UV−vis cuvette and 1 mL of a solution of
Hmte (277 mg in 5.00 mL of H2O, 0.600 M) was added. After
equilibration at 24 °C in the dark, UV−vis spectra of the sample were
measured in the dark and then 10 times during 10 min irradiation with
an LED lamp at λe = 465 nm to calculate φ4 in the same procedure as
that for [6]2+. A value of 0.30(6) was found for φ4 (see Supporting
Information, Table S5).
Thermal Substitution of Hmte by H2O in [5]2+ (k−1). Three
milliters of a solution of [5]2+ (5.6 mg of [5](PF6)2 in 25 mL of H2O,
2.5 × 10−4 M) was placed in a UV−vis cuvette, which was placed at t =
0 in the UV−vis spectrometer pre-equilibrated at 70, 75, 80, 85, or 90
°C. UV−vis spectra were measured every 60 s. The concentrations in
[RuHmte] and [RuOH2] were determined by deconvolution of the
UV−vis spectra knowing the extinction coefficients of both RuHmte
and RuOH2 species (see Supporting Information and Table S1). The
first-order rate constant k−1 at each temperature was determined by
plotting ln([RuOH2]/[Ru]tot) vs time. The slope and y-intercept of an
Eyring plot afforded the activation enthalpy and entropy, respectively
(see Supporting Information, Figure S10). k−1 at 24 °C was extracted
from extrapolating the Eyring equation down to room temperature; a
value of 1.5(4) × 10−8 s−1 was found.
Thermal Substitution of Hmte by H2O on [6]2+, [7]2+, and
[8]2+ (k−2, k−3, k−4). At the thermodynamic equilibrium between
RuOH2, free Hmte, and RuHmte in water, the rates for the formation
and hydrolysis of RuHmte complex are equal.
k−i·[RuHmte]eq = ki·[RuOH2]eq ·[Hmte]eq
ASSOCIATED CONTENT
* Supporting Information
■
Thus the first-order rate constant k−i for the thermal substitution
free Hmte by water is numerically given by k−i = ki/K . The activation
S
⧧
The synthesis of [10]Cl, [12]Cl, proton attribution schemes,
1H NMR spectra in D2O and MeOD/D2O mixtures, procedure
for X-ray crystal structure determination, extinction coefficient
determination procedure and values, mathematical modeling of
the fast equilibrium between [4]2+ and [8]2+, numerical values
of first-order and second-order rate constant for all four
systems, determination of the rate law (order of Hmte) of the
thermal coordination reaction for N−N = bpy, biq, and dmbpy,
Eyring plot for the thermal hydrolysis of [5]2+, pKa measure-
ments for [2]2+ and [4]2+, and photosubstitution quantum yield
measurements. This material is available free of charge via the
Internet at http://pubs.acs.org.
Gibbs energy ΔG−i for the thermal substitution of Hmte by H2O
⧧
were calculated using the equation ΔG−i = ΔGi⧧ − ΔGi° (see Table
3).
Photochemistry. The photochemical quantum yield for [5]2+ was
measured using a Varian Cary 50 UV−visible spectrometer and a LOT
1000 W xenon arc lamp, fitted with a water filter and a 450FS10-50
Andover interference filter (λe = 452 nm, Δλ1/2 = 8.9 nm). Irradiation
was thus performed close to the isosbestic point of the reaction, which
was at 449 nm. The photochemistry measurements for [6]2+, [7]2+,
and [8]2+ were done using a Perkin-Elmer Lambda 900 spectrometer
equipped with a custom-made LED lamp fitted to the top of a 1 cm
quartz UV−vis cuvette, using an OSRAM Opto electronics LEDs LB
W5KM-EZGY-35 (λe = 465 nm or λe = 520 nm, Δλ1/2 = 25 nm). In
these cases, UV−vis measurements of a sample during irradiation was
superimposable with a spectrum of the sample when the LED lamp
was switched off, which means that the light used to irradiate the
sample perpendicularly to the optical axis of the spectrophotometer
was not detected by the spectrometer. Photon fluxes of the three
irradiation setups were measured using the ferrioxalate actinometer;90
a value Φ = 6.4(6) × 10−8 Einstein s−1 was measured at 452 nm for the
filtered LOT lamp; Φ = 3.9(4) × 10−9 Einstein s−1 was found for the
LED at 465 nm, and Φ = 9.8(8) × 10−10 Einstein s−1 was found for the
LED at 520 nm. In the latter two cases, the irradiation path length was
3 cm, and the volume of the irradiated solution was 3 mL.
AUTHOR INFORMATION
Corresponding Author
*E-mail: bonnet@chem.leidenuniv.nl.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Leiden University is kindly acknowledged for financial support.
The Dutch Organization for Scientific Research (NWO−CW/
Vernieuwingsimpuls) is kindly acknowledged for a Veni grant
to S.B.
Photosubstitution Quantum Yield Determination for Com-
plex [5]2+ (φ1). 0.75 mL of a stock solution of the complex [5](PF6)2
(5.0 mg in 10 mL of H2O, 5.7 × 10−4 M) was put in a UV−vis cuvette.
The volume of the solution was completed to 3 mL with H2O (Final
concentration: 1.5 × 10−4 M). The sample was irradiated using the
same setup as was used for actinometry (Φ = 6.4(6) × 10−8 Einstein
s−1). After each irradiation period (1 min) a UV−vis spectrum was
measured until a total irradiation time of 10 min. The concentrations
in [5]2+ and [1]2+ were determined by deconvolution knowing the
extinction coefficients of both species (see Supporting Information
and Table S1). The evolution of ln([RuHmte]/[Ru]tot) was plotted as
a function of irradiation time, and from the slope S of the plot and
using Equation 11, the quantum yield φi was determined to be
0.022(6) (see Supporting Information, Table S5).
REFERENCES
■
(1) Ryabov, A. D.; Kuzmina, L. G.; Dvortsova, N. V.; Stufkens, D. J.;
van Eldik, R. Inorg. Chem. 1993, 32, 3166.
(2) Nishihara, H. Coord. Chem. Rev. 2005, 249, 1468.
(3) Wenger, O. S.; Henling, L. M.; Day, M. W.; Winkler, J. R.; Gray,
H. B. Polyhedron 2004, 23, 2955.
(4) Han, M.; Hirade, T.; Hara, M. New J. Chem. 2010, 34, 2887.
(5) Pratihar, P.; Mondal, T. K.; Patra, A. K.; Sinha, C. Inorg. Chem.
2009, 48, 2760.
(6) Mitsuoka, T.; Sato, H.; Yoshida, J.; Yamagishi, A.; Einaga, Y.
Chem. Mater. 2006, 18, 3442.
Irradiation of an Equilibrated Sample of the Biq System
([2]Cl2⇆[6]Cl2) and Photosubstitution Quantum Yield Deter-
mination for [6]2+ (φ2). A UV−vis cuvette containing 2 mL of a
stock solution of [10]Cl (1.5 mg in 10 mL of H2O, 2.3 × 10−4 M) and
(7) Kopelman, R. A.; Snyder, S. M.; Frank, N. L. J. Am. Chem. Soc.
2003, 125, 13684.
L
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(8) Bannwarth, A.; Schmidt, S. O.; Peters, G.; Sonnichsen, F. D.;
Thimm, W.; Herges, R.; Tuczek, F. Eur. J. Inorg. Chem. 2012, 2012,
2776.
(9) Takahashi, K.; Hasegawa, Y.; Sakamoto, R.; Nishikawa, M.;
Kume, S.; Nishibori, E.; Nishihara, H. Inorg. Chem. 2012, 51, 5188.
(10) Segarra-Maset, M. D.; van Leeuwen, P. W. N. M.; Freixa, Z. Eur.
J. Inorg. Chem. 2010, 2075.
(11) Boillot, M. L.; Zarembowitch, J.; Sour, A. Top. Curr. Chem.
2004, 234, 261.
(44) Respondek, T.; Garner, R. N.; Herroon, M. K.; Podgorski, I.;
Turro, C.; Kodanko, J. J. J. Am. Chem. Soc. 2011, 133, 17164.
(45) Lutterman, D. A.; Fu, P. K. L.; Turro, C. J. Am. Chem. Soc. 2006,
128, 738.
(46) Balzani, V.; Bergamini, G.; Campagna, S.; Puntoriero, F. Top.
Curr. Chem. 2007, 280, 1.
(47) Paris, J. P.; Brandt, W. W. J. Am. Chem. Soc. 1959, 81, 5001.
(48) Anderson, C. P.; Salmon, D. J.; Meyer, T. J.; Young, R. C. J. Am.
Chem. Soc. 1977, 99, 1980.
(49) Van Houten, J.; Watts, R. J. J. Am. Chem. Soc. 1976, 98, 4853.
(50) Durham, B.; Caspar, J. V.; Nagle, J. K.; Meyer, T. J. J. Am. Chem.
Soc. 1982, 104, 4803.
(51) Hecker, C. R.; Fanwick, P. E.; McMillin, D. R. Inorg. Chem.
1991, 30, 659.
(52) Baranoff, E.; Collin, J. P.; Furusho, Y.; Laemmel, A. C.; Sauvage,
J. P. Chem. Commun. 2000, 1935.
(53) Durham, B.; Walsh, J. L.; Carter, C. L.; Meyer, T. J. Inorg. Chem.
̈
(12) Lutterman, D. A.; Rachford, A. A.; Rack, J. J.; Turro, C. J. Phys.
Chem. Lett. 2010, 1, 3371.
(13) Rachford, A. A.; Petersen, J. L.; Rack, J. J. Inorg. Chem. 2006, 45,
5953.
(14) Rack, J. J.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. 2001,
123, 2432.
(15) Sauvage, J. P. Chem. Commun. 2005, 1507.
(16) Balzani, V.; Clemente-Leon, M.; Credi, A.; Ferrer, B.; Venturi,
M.; Flood, A. H.; Stoddart, J. F. Proc. Natl. Acad. Sci. U. S. A. 2006,
103, 1178.
(17) Agrawal, U. C.; Nigam, H. L. J. Indian Chem. Soc. 2008, 85, 677.
(18) Campagna, S.; Puntoriero, F.; Nastasi, F.; Bergamini, G.;
Balzani, V. Photochem. Photophys. Coord. Compd. I 2007, 280, 117.
(19) Balzani, V.; Credi, A.; Venturi, M. Chem. Soc. Rev. 2009, 38,
1542.
(20) Bonnet, S.; Collin, J. P.; Koizumi, M.; Mobian, P.; Sauvage, J. P.
Adv. Mater. 2006, 18, 1239.
(21) Bonnet, S.; Collin, J.-P. Chem. Soc. Rev. 2008, 37, 1207.
(22) Bahreman, A.; Limburg, B.; Siegler, M. A.; Koning, R.; Koster,
A. J.; Bonnet, S. Chem.Eur. J. 2012, 18, 10271.
(23) Collin, J. P.; Heitz, V.; Bonnet, S.; Sauvage, J. P. Inorg. Chem.
Commun. 2005, 8, 1063.
1980, 19, 860.
(54) Bonnet, S.; Collin, J. P.; Gruber, N.; Sauvage, J. P.; Schofield, E.
R. Dalton Trans. 2003, 4654.
(55) Ashton, P. R.; Ballardini, R.; Balzani, V.; Credi, A.; Dress, K. R.;
Ishow, E.; Kleverlaan, C. J.; Kocian, O.; Preece, J. A.; Spencer, N.;
Stoddart, J. F.; Venturi, M.; Wenger, S. Chem.Eur. J. 2000, 6, 3558.
(56) Ashton, P. R.; Ballardini, R.; Balzani, V.; Constable, E. C.; Credi,
A.; Kocian, O.; Langford, S. J.; Preece, J. A.; Prodi, L.; Schofield, E. R.;
Spencer, N.; Stoddart, J. F.; Wenger, S. Chem.Eur. J. 1998, 4, 2413.
(57) Baranoff, E.; Collin, J. P.; Furusho, J.; Furusho, Y.; Laemmel, A.
C.; Sauvage, J. P. Inorg. Chem. 2002, 41, 1215.
(58) Mobian, P.; Kern, J. M.; Sauvage, J. P. J. Am. Chem. Soc. 2003,
125, 2016.
(59) Nguyen, T. D.; Leung, K. C. F.; Liong, M.; Liu, Y.; Stoddart, J.
F.; Zink, J. I. Adv. Funct. Mater. 2007, 17, 2101.
(60) Collin, J. P.; Jouvenot, D.; Koizumi, M.; Sauvage, J. P. Inorg.
Chem. 2005, 44, 4693.
(61) Bonnet, S.; Collin, J. P.; Sauvage, P. Inorg. Chem. 2006, 45, 4024.
(62) Gianferrara, T.; Bergamo, A.; Bratsos, I.; Milani, B.; Spagnul, C.;
Sava, G.; Alessio, E. J. Med. Chem. 2010, 53, 4678.
(63) Bonnet, S.; Collin, J. P.; Sauvage, J. P.; Schofield, E. Inorg. Chem.
2004, 43, 8346.
(64) Bessel, C. A.; Margarucci, J. A.; Acquaye, J. H.; Rubino, R. S.;
Crandall, J.; Jircitano, A. J.; Takeuchi, K. J. Inorg. Chem. 1993, 32,
5779.
(65) Leising, R. A.; Ohman, J. S.; Takeuchi, K. J. Inorg. Chem. 1988,
27, 3804.
(66) Wasylenko, D. J.; Ganesamoorthy, C.; Koivisto, B. D.;
Henderson, M. A.; Berlinguette, C. P. Inorg. Chem. 2010, 49, 2202.
(67) Che, C. M.; Ho, C.; Lau, T. C. J. Chem. Soc., Dalton Trans. 1991,
1901.
(68) Takeuchi, K. J.; Thompson, M. S.; Pipes, D. W.; Meyer, T. J.
Inorg. Chem. 1984, 23, 1845.
(24) Mobian, P.; Kern, J. M.; Sauvage, J. P. Angew. Chem., Int. Ed.
2004, 43, 2392.
(25) Laemmel, A. C.; Collin, J. P.; Sauvage, J. P. Eur. J. Inorg. Chem.
1999, 383.
(26) Bednarski, P. J.; Mackay, F. S.; Sadler, P. J. Anti-Cancer Agents
Med. Chem. 2007, 7, 75.
(27) Farrer, N. J.; Salassa, L.; Sadler, P. J. Dalton Trans. 2009, 10690.
(28) Howerton, B. S.; Heidary, D. K.; Glazer, E. C. J. Am. Chem. Soc.
2012, 134, 8324.
(29) Wachter, E.; Heidary, D. K.; Howerton, B. S.; Parkin, S.; Glazer,
E. C. Chem. Commun. 2012, 48, 9649.
(30) Indelli, M.; Chiorboli, C. Top. Curr. Chem. 2007, 280, 215.
(31) Mahnken, R. E.; Billadeau, M. A.; Nikonowicz, E. P.; Morrison,
H. J. Am. Chem. Soc. 1992, 114, 9253.
(32) Schatzschneider, U.; Niesel, J.; Ott, I.; Gust, R.; Alborzinia, H.;
Woelfl, S. ChemMedChem 2008, 3, 1104.
(33) Goldbach, R. E.; Rodriguez-Garcia, I.; van Lenthe, J. H.; Siegler,
M. A.; Bonnet, S. Chem.Eur. J. 2011, 17, 9924.
(34) Moucheron, C.; KirschDeMesmaeker, A.; Kelly, J. J. Photochem.
1997, 40, 91.
(69) Swaddle, T. W. Coord. Chem. Rev. 1974, 14, 217.
(70) Mahanti, B.; De, G. S. Transition Met. Chem. 1994, 19, 201.
(71) Aebischer, N.; Laurenczy, G.; Ludi, A.; Merbach, A. E. Inorg.
Chem. 1993, 32, 2810.
(72) Aebischer, N.; Churlaud, R.; Dolci, L.; Frey, U.; Merbach, A. E.
Inorg. Chem. 1998, 37, 5915.
(35) Higgins, S. L. H.; Brewer, K. J. Angew. Chem., Int. Ed. 2012, 51,
2.
(36) Garner, R. N.; Gallucci, J. C.; Dunbar, K. R.; Turro, C. Inorg.
Chem. 2011, 50, 9213.
(37) De Candia, A. G.; Marcolongo, J. P.; Etchenique, R.; Slep, L. D.
Inorg. Chem. 2010, 100625124914047.
(73) Rotzinger, F. P. J. Phys. Chem. A 2000, 104, 8787.
(74) Bunten, K. A.; Farrar, D. H.; Poe, A. J.; Lough, A. J.
Organometallics 2000, 19, 3674.
(75) Czap, A.; Heinemann, F. W.; van Eldik, R. Inorg. Chem. 2004,
43, 7832.
(38) Zayat, L.; Calero, C.; Albores, P.; Baraldo, L.; Etchenique, R. J.
Am. Chem. Soc. 2003, 125, 882.
(39) Fry, N. L.; Mascharak, P. K. Acc. Chem. Res. 2011, 44, 289.
(40) Rose, M. J.; Mascharak, P. K. Coord. Chem. Rev. 2008, 252,
2093−2114.
(76) Gamasa, M. P.; Gimeno, J.; GonzalezBernardo, C.; MartinVaca,
B. M.; Monti, D.; Bassetti, M. Organometallics 1996, 15, 302.
(77) Kojima, T.; Morimoto, T.; Sakamoto, T.; Miyazaki, S.;
Fukuzumi, S. Chem.Eur. J. 2008, 14, 8904.
(41) Barragan
Habtemariam, A.; Moreno, V.; Sadler, P. J.; Marchan
Soc. 2011, 133 (35), 14098−14108.
(42) Salassa, L.; Garino, C.; Salassa, G.; Gobetto, R.; Nervi, C. J. Am.
Chem. Soc. 2008, 130, 9590.
́
, F.; Lop
́
ez-Senín, P.; Salassa, L.; Betanzos-Lara, S.;
, V. J. Am. Chem.
́
(78) Rilak, A.; Petrovic, B.; Grguric-Sipka, S.; Tesic, Z.; Bugarcic, Z.
D. Polyhedron 2011, 30, 2339.
(43) Howerton, B. S.; Heidary, D. K.; Glazer, E. C. J. Am. Chem. Soc.
2012, 134, 8324.
(79) Goswami, A. M.; De, K. Transition Met. Chem. 2007, 32, 419.
M
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(80) Shoukry, M. M.; Shehata, M. R.; Hamza, M. S. A.; van Eldik, R.
Dalton Trans. 2005, 3921.
(81) Naik, R. M.; Singh, R.; Asthana, A. Int. J. Chem. Kinet. 2011, 43,
21.
(82) Thiel, V.; Hendann, M.; Wannowius, K.-J.; Plenio, H. J. Am.
Chem. Soc. 2012, 134, 1104.
(83) Wilkins, R. G. In Kinetics and Mechanism of Reactions of
Transition Metal Complexes, 2nd ed.; Wiley-VCH: Weinheim,
Germany, 1991; p 199.
(84) Langford, C. H.; Muir, W. R. J. Am. Chem. Soc. 1967, 89, 3141.
(85) Mukherjee, A.; De, K. Transition Met. Chem. 2005, 30, 677.
(86) Krishnan, R.; Schultz, R. H. Organometallics 2001, 20, 3314.
(87) Wang, F.; Chen, H. M.; Parsons, S.; Oswald, L. D. H.; Davidson,
J. E.; Sadler, P. J. Chem.Eur. J. 2003, 9, 5810.
(88) Leising, R. A.; Kubow, S. A.; Churchill, M. R.; Buttrey, L. A.;
Ziller, J. W.; Takeuchi, K. J. Inorg. Chem. 1990, 29, 1306.
(89) Constable, E. C.; Seddon, K. R. Tetrahedron 1983, 39, 291.
(90) Calvert, J. G.; Pitts, J. N., Photochemistry; Wiley and Sons: New
York, 1967; p 780.
N
dx.doi.org/10.1021/ic401105v | Inorg. Chem. XXXX, XXX, XXX−XXX