Light-Induced Activation of a Molybdenum Oxotransferase Model within a Ru(II)-Mo(VI) Dyad

Article abstract of DOI:10.1021/acs.inorgchem.6b01485

Nature uses molybdenum-containing enzymes to catalyze oxygen atom transfer (OAT) from water to organic substrates. In these enzymes, the two electrons that are released during the reaction are rapidly removed, one at a time, by spatially separated electron transfer units. Inspired by this design, a Ru(II)-Mo(VI) dyad was synthesized and characterized, with the aim of accelerating the rate-determining step in the cis-dioxo molybdenum-catalyzed OAT cycle, the transfer of an oxo ligand to triphenyl phosphine, via a photo-oxidation process. The dyad consists of a photoactive bis(bipyridyl)-phenanthroline ruthenium moiety that is covalently linked to a bioinspired cis-dioxo molybdenum thiosemicarbazone complex. The quantum yield and luminescence lifetimes of the dyad [Ru(bpy)₂(L²)MoO₂(solv)]²⁺ were determined. The major component of the luminescence decay in MeCN solution (?., = 1149 ± 2 ns, 67%) corresponds closely to the lifetime of excited [Ru(bpy)₂(phen-NH₂)]²⁺, while the minor component (?., = 320 ± 1 ns, 31%) matches that of [Ru(bpy)₂(H₂-L²)]²⁺. In addition, the (spectro)electrochemical properties of the system were investigated. Catalytic tests showed that the dyad-catalyzed OAT from dimethyl sulfoxide to triphenyl phosphine proceeds significantly faster upon irradiation with visible light than in the dark. Methylviologen acts as a mediator in the photoredox cycle, but it is regenerated and hence only required in stoichiometric amounts with respect to the catalyst rather than sacrificial amounts. It is proposed that oxidative quenching of the photoexcited Ru unit, followed by intramolecular electron transfer, leads to the production of a reactive one-electron oxidized catalyst, which is not accessible by electrochemical methods. A significant, but less pronounced, rate enhancement was observed when an analogous bimolecular system was tested, indicating that intramolecular electron transfer between the photosensitizer and the catalytic center is more efficient than intermolecular electron transfer between the separate components.

Full text of DOI:10.1021/acs.inorgchem.6b01485

This is an open access article published under an ACS AuthorChoice License, which permits
copying and redistribution of the article or any adaptations for non-commercial purposes.
Article
pubs.acs.org/IC
Light-Induced Activation of a Molybdenum Oxotransferase Model
within a Ru(II)−Mo(VI) Dyad
Aurelien B. Ducrot, Ben A. Coulson, Robin N. Perutz, and Anne-Kathrin Duhme-Klair*
́
Department of Chemistry, University of York, Heslington, York YO10 5DD, United Kingdom
S
* Supporting Information
ABSTRACT: Nature uses molybdenum-containing enzymes to catalyze
oxygen atom transfer (OAT) from water to organic substrates. In these
enzymes, the two electrons that are released during the reaction are
rapidly removed, one at a time, by spatially separated electron transfer
units. Inspired by this design, a Ru(II)−Mo(VI) dyad was synthesized
and characterized, with the aim of accelerating the rate-determining step
in the cis-dioxo molybdenum-catalyzed OAT cycle, the transfer of an oxo
ligand to triphenyl phosphine, via a photo-oxidation process. The dyad
consists of a photoactive bis(bipyridyl)-phenanthroline ruthenium moiety
that is covalently linked to a bioinspired cis-dioxo molybdenum
thiosemicarbazone complex. The quantum yield and luminescence
lifetimes of the dyad [Ru(bpy)2(L2)MoO2(solv)]2+ were determined.
The major component of the luminescence decay in MeCN solution (τ = 1149 2 ns, 67%) corresponds closely to the lifetime
of excited [Ru(bpy)2(phen-NH2)]2+, while the minor component (τ = 320 1 ns, 31%) matches that of [Ru(bpy)2(H2-L2)]2+.
In addition, the (spectro)electrochemical properties of the system were investigated. Catalytic tests showed that the dyad-
catalyzed OAT from dimethyl sulfoxide to triphenyl phosphine proceeds significantly faster upon irradiation with visible light
than in the dark. Methylviologen acts as a mediator in the photoredox cycle, but it is regenerated and hence only required in
stoichiometric amounts with respect to the catalyst rather than sacrificial amounts. It is proposed that oxidative quenching of the
photoexcited Ru unit, followed by intramolecular electron transfer, leads to the production of a reactive one-electron oxidized
catalyst, which is not accessible by electrochemical methods. A significant, but less pronounced, rate enhancement was observed
when an analogous bimolecular system was tested, indicating that intramolecular electron transfer between the photosensitizer
and the catalytic center is more efficient than intermolecular electron transfer between the separate components.
additional redox centers built into their design. In these
models, the oxidative and reductive half-reaction take place at
the Mo center (Figure 1, right). The catalytic activity of these
first-generation models is often limited by slow kinetics due to
charge buildup at the Mo center and the formation of inactive
Mo(V) dimers as a result of the comproportionation of
Mo(IV) and Mo(VI).8
In our search for an advanced design, we were inspired by
catalytic systems that exploit photoredox processes, since they
offer the opportunity of harnessing solar energy. Because of
their high quantum yields, long excitation wavelengths,
photostability, and long lifetimes, polypyridine complexes of
metal cations with d6 low-spin configuration, such as Ru(II),
Re(I), and Ir(III) are frequently used as photoactive units in
heterometallic systems to convert light energy to chemical
energy. Inspired by photosystem II, manganese model
complexes have been linked to [Ru(bpy)3]2+-type photo-
sensitizers to target the four-electron oxidation of H2O by
taking advantage of photoinduced electron transfer (PeT).9
Similarly, bioinspired dyads that mimic hydrogenases have been
INTRODUCTION
■
Metal-containing oxidoreductases play an important role in
biological processes by catalyzing selective oxygenations, in
particular, those that involve O2 activation or oxygen atom
3
transfer (OAT) from water to an organic substrate. To be able
to utilize these “green oxygen sources” effectively and under
mild conditions, metalloenzymes rely on efficient electron
transfer chains that involve spatially separated electron transfer
units, such as Fe/S clusters and/or cytochromes to regenerate
their active sites (Figure 1, left).1
In the search for functional bioinspired catalysts the
importance of an overall design that includes electron transfer
components and, in particular, proton-coupled electron transfer
is increasingly recognized,3,4 and a number of second-
generation model complexes that contain enzyme active-site
mimics that are covalently linked to redox-active units have
been reported. Electrochemical approaches, for example, have
utilized rapid one-electron transfer components, such as
ferrocene.5−7
These second-generation models represent a significant
advancement over conventional molybdoenzyme models that
catalyze the thermodynamically favorable OAT reaction from
sulfoxides to tertiary phosphines, but they do not have
Received: June 23, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Structures of Bioinspired oxo-Mo Complexes with
Appended Electron-Transfer Units
Figure 1. (left) Schematic electron-transfer chain in a metalloenzyme,
as exemplified by xanthine dehydrogenase (pdb code 3am9, Mo =
molybdenum-containing active site, X = xanthine). (right) Conven-
tional “bioinspired” OAT cycle in which both the oxidative and
reductive half-reaction take place at the Mo center of a synthetic
model complex (DMSO + Ph3P → DMS + Ph3PO: ΔH = −47 kcal
mol−1, gas phase2).
studied with the aim of catalyzing the light-driven production of
H2.10,11
developed in bioinspired approaches to model the function of
the molybdenum enzymes20−24 we chose thiosemicarbazones,
whose synthesis is straightforward and allows a variety of
substituents to be incorporated.25−27 We recently reported the
electronic fine-tuning of series of six cis-dioxo-Mo(VI)
complexes with thiosemicarbazone ligands for OAT and
found that electron-withdrawing substituents increase the rate
of the OAT to triphenylphosphine, while electron-donating
groups have the opposite effect.28 This observation is consistent
with the generally accepted OAT mechanism, which starts with
the nucleophilic attack of the phosphine lone pair on a vacant
π* orbital of one of the oxo-ligands, leading to the formation of
an oxo-Mo(IV) species and phosphine oxide via a phosphine
oxide-coordinated intermediate in the rate-determining
step.29,30 Accordingly, a decrease of the electron density on
the cis-dioxo-Mo(VI) center renders the oxo ligand more
susceptible to attack and lowers the energy of the transition
state by ameliorating negative charge buildup, giving rise to
faster reaction rates.
Here, we report the synthesis and the characterization of the
Ru(II)−Mo(VI) dyad [Ru(bpy)2(L2)MoO2(solv)](BF4)2, in
which one coordination site on the Mo(VI) center is occupied
by a solvent molecule (Scheme 1c). The photoactive
bis(bipyridyl)-phenanthroline ruthenium moiety is linked to
the Mo-based OAT catalyst by a methylamine group to define
the spatial separation while allowing flexibility to facilitate
intramolecular electron transfer from the Mo-coordinated
thiosemicarbazone ligand to the photo-oxidized RuIII-bpy/
phen•−-unit. The photophysical, spectroscopic, and electro-
chemical properties of the dyad and the results of photo-
catalytic tests are reported.
Recently, there has been increasing interest in photoredox
catalysis12−14 and its scope in the photocatalyic oxygenation of
3
organic substrates using clean O atom sources, such as O2 or
H2O, which is highly desirable on both environmental and
economic grounds. For example, Hamelin and Torelli et al.
used a Ru(II)−Cu(II) dyad to oxidize sulfides, phosphines, and
3
alkenes photocatalytically with O2 as the oxygen atom source
and triethanolamine as sacrificial electron donor.15 Aukauloo,
Banse, and Leibl et al., reported a Ru(II)−Fe(II) dyad that
allows the photogeneration of a highly reactive oxo-Fe(IV)
center from an Fe(II)-coordinated H2O molecule using
[Co(NH3)5Cl]2+ as sacrificial electron acceptor, followed by
transfer of the oxygen atom to PPh3.16 To drive the
photoactivated OAT, the two-electron oxidation of the
substrate had to be coupled with two rapid proton-coupled
electron transfer steps.
Since Mo is chosen by nature as OAT catalyst par excellence
based on its ability to redox cycle between oxidation states VI,
V, and IV,17 we are interested in utilizing bioinspired Mo
complexes as catalysts in dyads designed to facilitate substrate
oxidation via a photoredox process. Enemark and Kirk et al.
already demonstrated in an early study that a Mo center can be
photoactivated via an antenna-mediated electron transfer
process by covalently linking an oxo-Mo(V) unit to
porphyrin-Fe(III) or Zn(II) complexes (Scheme 1a).18,19
Although catalytic investigations were not reported for these
dyads, the study highlighted photoactivation as a potential way
of initiating intercomponent electron transfer. More recently,
Heinze et al. appended two redox-active ferrocene units to a cis-
dioxo-Mo(VI) complex to mimic the electron-transfer chain in
the molybdenum enzyme sulfite oxidase (Scheme 1b).7 It was
observed that chemical oxidation to the diferrocenium dication
greatly accelerated OAT to PMe3, consistent with an
accelerated attack of the phosphine at the oxo ligand, coupled
with intramolecular electron transfer.
EXPERIMENTAL SECTION
■
Materials and Instruments. Ammonium tetrafluoroborate,
paraformaldehyde, 1,10-phenanthroline-5-amine, 4-phenylthiosemicar-
bazide, and sodium molybdate dihydrate were purchased from Aldrich.
Acetylacetic acid was purchased from BDH chemicals, ammonium
hexafluorophosphate from Alfa Aesar, salicylaldehyde from Fluka,
triethylamine from Fisher Chemicals, and cis-bis(2,2′-bipyridine)-
dichlororuthenium hydrate from Acros organics. [MoO2(acac)2]31 and
5-chloromethyl-2-hydroxylbenzaldehyde32 were synthesized according
to reported procedures. Solvents for syntheses were dried and stored
Drawing on these precedents, we anticipated that the
addition of a photoactivatable electron acceptor to a cis-
dioxo-Mo(VI) center could promote the attack of a phosphine
at the oxo ligand in a similar way, thereby increasing catalytic
activity. Of the many synthetic ligand systems that were
B
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Synthesis of the Dyad [Ru(bpy)2(L2)MoO2(MeOH)](BF4)2 and the Structure of [MoO2(LMe)MeOH)]
1
over 3 Å molecular sieves. H and 13C{1H} decoupled NMR spectra
13C NMR (CD3CN, 125.8 MHz): δ 158.3, 158.0, 153.14, 152.97,
were recorded on a Jeol ECS400 instrument (1H NMR 400 MHz, 13
C
152.88, 152.81, 149.3, 148.5, 145.5, 142.9, 138.7, 138.6, 134.7, 133.9,
132.5, 128.48, 128.44, 128.35, 126.8, 125.7, 125.20, 125.15, 125.11,
125.0, 104.4.
NMR 100.6 MHz, 19F 376.5 MHz), a Bruker 500 (1H NMR 500 MHz,
13C NMR 125.8 MHz, 31P 202.5 MHz), and a Bruker 700 (1H NMR
700 MHz, 13C NMR 176.0 MHz). Assignment of resonances was
confirmed based on COSY, DEPT135, and HSQC spectra.
Decomposition points were measured with a Stuart Scientific SMP3
apparatus. Elemental analyses of compounds were performed with an
Exeter CE-440 elemental analyzer. Electron spray ionization mass
spectrometry (ESI-MS) and high-resolution mass spectra were
recorded with a Bruker microTOF electrospray mass spectrometer,
and mass values are quoted for 102Ru and 98Mo. UV−visible absorption
spectra were measured using an Agilent 8453 spectrometer. Emission
spectra were recorded on a Hitachi F-4500 fluorescence spectropho-
tometer. Excited-state lifetimes were measured on an Edinburgh
Instruments FLS980 fluorescence lifetime instrument.
19F NMR (CD3CN, 376.5 MHz): δ −151.6.
MS-ESI: m/z 304.5595, [M2+] calcd. for C32H25N7Ru 304.5606.
C32H25B2F8 N7Ru + 1.5 H2O: Anal. Calcd C 47.49, H 3.49, N
12.12; found C 47.56, H 3.22, N 12.00.
[Ru(bpy)2(L1)](BF4)2. [Ru(bpy)2(phen-NH2)](BF4)2 (1.11 g, 1.42
mmol) was suspended in a solution of 5-chloromethyl-2-hydrox-
ybenzaldehyde (0.29 g, 1.70 mmol) in dichloromethane (30 mL)
under an argon atmosphere. Triethylamine (600 μL, 4.33 mmol) was
added dropwise, and the resultant mixture was stirred overnight.
Ammonium tetrafluoroborate (0.45 g, 4.29 mmol) was added, and the
mixture was stirred vigorously for 5 min prior to washing with water.
The organic layer was dried over magnesium sulfate and concentrated
to dryness. The product was triturated with diethyl ether to yield a red
solid (1 g, 72%).
Synthesis. The synthetic method is shown in Scheme 2.
[Ru(bpy)2(phen-NH2)](BF4)2.33 cis-Bis(2,2′-bipyridine)-
dichlororuthenium hydrate (1.06 g, 2.04 mmol) and 1,10-phenanthro-
line-5-amine (0.465 g, 2.38 mmol) were dissolved in a mixture of
ethanol (160 mL) and water (80 mL). The resulting solution was
degassed and refluxed under Ar for 20 h. The ethanol was removed
under reduced pressure, and a solution of ammonium tetrafluor-
oborate (2.14 g, 20.4 mmol) in water was added to precipitate the
product as tetrafluoroborate salt. The precipitate was isolated by
filtration and washed with water. The crude product was purified by
column chromatography on alumina (toluene/acetonitrile 1:1). The
product was solubilized in a minimum volume of acetonitrile and
reprecipitated with diethyl ether. The bright orange precipitate was
isolated and dried under vacuum (1.164 g, 73%).
Decomposition point 283 °C.
1H NMR (CD3CN, 400 MHz): δ 9.93 (s, 1H, CHO), 8.72 (d, 3J =
3
8.56 Hz, 1H, CHphen), 8.54−8.47 (m, 4H, CHbpy), 8.16 (d, J = 8.36
Hz, 1H, CHphen), 8.11−8.04 (m, 3 CHbpy + 1CHphen), 7.99 (t, 3J = 7.8
Hz, 2H, CHbpy), 7.85−7.80 (m, 3H, 2 CHbpy + 1CHAr), 7.71 (dd, 3J =
4
3
8.6 Hz, J = 3.64 Hz, 2H, 1 CHphen + 1CHbpy), 7.64 (d, J = 5.12 Hz,
1H, CHphen), 7.55 (dd, 3J = 12.4 Hz, 3J = 5.64 Hz, 2H, CHbpy), 7.47−
7.40 (m, 3H, 2 CHbpy + 1CHphen), 7.24 (m, 2H, CHbpy), 6.99 (d, 3J =
8.6 Hz 1H, CHAr), 6.90 (s, 1H, CHphen), 6.56 (t, 1H, NH), 4.66 (d, 3J
= 4.44 Hz, 2H, CH2).
13C NMR (CD3CN, 100.6 MHz): δ 198.3; 161.6; 158.24; 158.23;
158.04; 158.00; 153.13; 152.97; 152.85; 152.78; 152.75; 149.3; 148.3;
144.4; 142.8; 138.72; 138.70; 138.59; 137.4; 134.9; 133.96; 133.74;
131.9; 130.8; 128.50; 128.46; 128.44; 128.34; 127.0; 125.9; 125.4;
125.22; 125.14; 125.13; 125.09; 121.7; 118.5; 100.7; 47.0.
19F NMR (CD3CN, 376.5 MHz): δ −151.6.
Decomposition point 215 °C.
1H NMR (CD3CN, 500 MHz): δ 8.61 (d, 3J = 8.55 Hz, 1H,
CHphen), 8.52 (t, 3J = 6.65 Hz, 2H, CHbpy), 8.48 (d, 3J = 8.25 Hz, 2H,
CHbpy), 8.21 (d, 3J = 8.3 Hz, 1H, CHphen), 8.10−8.04 (m, 3H, 2 CHbpy
+ 1CHphen), 7.99 (t, 3J = 8 Hz, 2H, CHbpy), 7.83 (dd, 3J = 13.6 Hz, 3J =
MS-ESI: m/z 371.5775, [M2+] calcd. for C40H31N7O2Ru 371.5791.
C40H31B2F8N7O2Ru + H2O: Anal. Calcd C 51.41, H 3.56, N 10.49;
found C 50.84, H 3.53, N 10.10%.
3
5.55 Hz, 2H, CHbpy), 7.71−7.67 (m, 2H, CHphen), 7.55 (dd, J = 16.9
3
3
3
Hz, J = 5.3 Hz, 2H, CHbpy), 7.49 (dd, J = 8.4 Hz, J = 5.2 Hz, 1H,
CHphen), 7.43 (m, 2H, CHbpy), 7.23 (m, 2H, CHbpy), 7.17 (s, 1H,
CHphen), 5.57 (s, 2H, NH).
[Ru(bpy)2(H2-L2)](BF4)2. [Ru(bpy)2(L1)](BF4)2 (0.962 g, 1.05
mmol) and 4-phenylthiosemicarbazide (0.175 g, 1.05 mmol) were
C
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
refluxed for 16 h in a solvent mixture of ethanol and acetonitrile (2:1)
under nitrogen. After evaporation to dryness the residue was
solubilized in a minimum of acetonitrile, and the product was
precipitated by addition of ethanol. The product was collected by
filtration and dried under vacuum (0.83 g, 74%).
Decomposition point 245 °C.
1H NMR (CD3CN, 400 MHz): δ 8.70 (d, 3J = 8.04 Hz, 1H,
CHphen), 8.26−8.18 (m, 4H, CHbpy), 8.26 (s, 1H, CHN), 8.14 (d, 3J =
7.56 Hz, 1H, CHAr), 8.10−8.03 (m, 3H, CHAr), 8.00−7.93 (m, 2H,
CHAr), 7.81 (dd, 3J = 16.08 Hz, 3J = 4.84 Hz, 2H, CHAr), 7.68 (dd, 3J =
3
3
8.2 Hz, J = 5.04 Hz, 1H, CHAr), 7.62 (d, J = 4.76 Hz, 1H, CHAr),
3
3
7.53 (dd, J = 8.68 Hz, J = 4.92 Hz, 2H, CHAr), 7.49−7.39 (m, 7H,
3
CHAr), 7.33 (t, J = 7.28 Hz, 2H, CHAr), 7.24−7.17 (m, 3H, CHAr),
6.93 (d, 3J = 8.08 Hz, 1H, CHAr), 6.91 (s, 1H, CHAr), 6.50 (t, 3J = 5.32
Hz, 1H, NH), 4.61 (m, 2H, CH2).
13C NMR (CD3CN, 125.8 MHz): δ 158.2, 158.0, 153.0, 152.9,
152.8, 152.7, 149.3, 148.2, 144.6, 142.7, 138.7, 138.6, 134.8, 134.0,
131.9, 130.3, 129.4, 128.50, 128.45, 128.3, 126.9, 126.8, 126.2, 125.8,
125.4, 125.22, 125.14 125.09, 100.6, 47.4.
19F NMR (CD3CN, 376.5 MHz): δ −151.5
MS-ESI: 446.0982, [M2+] calcd. for C47H38N10ORuS 446.0997.
C47H38F8N10OB2RuS + 2 H2O: Anal. Calcd C 51.24, H 3.84, N
12.71; found C 50.92, H 3.33, N 12.45%.
[Ru(bpy)2(L2)MoO2(MeOH)](BF4)2. [Ru(bpy)2(H2-L2)](BF4)2 (100
mg, 0.094 mmol) and [MoO2(acac)2] (31 mg, 0.095 mmol) were
refluxed for 18 h in a mixture of acetonitrile and methanol (1:1, 10
mL) and under an atmosphere of nitrogen. After filtration, the reaction
mixture was evaporated to dryness. The residue was solubilized in
acetonitrile and filtered. The solvent was evaporated, and the product
was triturated with diethyl ether and washed with a small volume of
dichloromethane, followed by diethyl ether, and dried under vacuum
to yield a red crystalline solid (84 mg, 75%).
1H NMR (CD3CN, 500 MHz): δ 8.73 (s, 1H, CHN), 8.70 (d, 3J =
8.6 Hz, 1H, CHphen), 8.51−8.45 (m, 4H, CHbpy), 8.16 (dd, 3J = 8.4 Hz,
4J = 1.1 Hz, 1H, CHphen), 8.10−8.04 (m, 3H, 2 CHbpy + 1CHphen),
7.99−7.95 (m, 2H, CHbpy), 7.88 (s, 1H, CHAr), 7.84−7.79 (m, 2H,
Figure 2. (a) Electronic absorption (red) and emission spectrum
(black, λexc = 470 nm) of [Ru(bpy)2(L2)MoO2(solv)]2+ in comparison
with the absorption spectra of [Ru(bpy)2(phen-NH2)]2+ (blue) and
[MoO2(LMe)(solv)] (cyan) and the sum of their absorption spectra
(purple), recorded at RT in deaerated MeCN, 10 μM concentration.
(b) Plot of integrated emission intensity vs absorbance at 470 nm for
solutions of [Ru(bpy)3]2+ (black), [Ru(bpy)2(phen-NH2)]2+ (blue),
[Ru(bpy)2(H2-L2)]2+ (green), and [Ru(bpy)2(L2)MoO2(solv)]2+
(red), where solv = coordinated solvent molecule, at a range of
concentrations (1−6 μM, deaerated MeCN).
3
CHbpy), 7.71 (dd, J = 8.55 Hz, 3J = 5.25 Hz, 1H, CHphen), 7.68−7.65
3
4
(m, 3H, CHAr), 7.63 (dd, J = 5.15 Hz, J = 1.2 Hz, 1H, CHphen),
7.57−7.52 (m, 2H, CHbp3y), 7.45−7.39 (m, 3H, 2 CHbpy + 1CHphen),
3
7.33 (dd, J = 8.45 Hz, J = 7.5 Hz, 2H, CHAr), 7.24−7.21 (m, 2H,
3
3
CHbpy), 7.09 (t, J = 7.4 Hz, 1H, CHAr), 6.98 (d, J = 8.4 Hz, 1H,
3
CHAr), 6.90 (s, 1H, CHphen), 6.54 (t, J = 5.9 Hz, 1H, NH), 4.67 (m,
2H, CH2).
13C NMR (CD3CN, 125.8 MHz): δ 162.5, 159.7, 158.25, 158.23,
158.03, 158.00, 157.1, 153.1, 152.96, 152.84, 152.80, 152.7, 149.4,
148.2, 144.4, 142.8, 141.2, 138.72, 138.69, 138.58, 134.97, 134.85,
134.0, 133.7, 131.9, 129.8, 128.50, 128.45, 128.3, 127.0, 125.9, 125.4,
125.2, 125.14, 125.08, 124.4, 121.4, 121.2, 119.5, 100.7, 47.0.
19F NMR (CD3CN, 376.5 MHz): δ −151.7.
and 200 kHz for the shorter time range of 200 ns. The emission was
detected with a photomultiplier tube with maximum sensitivity in the
visible part of the spectrum. The excited-state decay was monitored at
different emission wavelengths (580, 600, 620, 640, 660, and 680 nm).
The slit width for detection was set to 10 nm, while the laser beam
profile was 10 nm.
MS-ESI: m/z 510.0399, [M2 + -MeOH] calcd. for
C47H36MoN10O3RuS 510.0399.
C48H40B2F8MoN10O4RuS + 2H2O: Anal. Calcd C 45.77, H 3.52, N
11.12; found C 46.32, H 3.08, N 10.79%.
Electrochemistry. All electrochemical measurements were per-
formed under nitrogen at RT. Cyclic voltammetry was performed with
an EC epsilon potentiostat from BASi with a standard three-electrode
electrochemical cell. Potentials were referenced to Ag/AgCl in NaClaq
(3 M). The working electrode was a carbon disk. The auxiliary
electrode was a platinum wire. Tetrabutylammonium tetrafluoroborate
(0.1 M) in dimethylformamide (DMF) was used as electrolyte.
Spectro-Electrochemistry. UV−vis spectro-electrochemical ex-
periments were performed in the dark in a N2-purged quartz cuvette of
1 mm path length equipped with a strip of carbon veil (Optiveil 12
g.m2 area weight) as working electrode, a Pt wire, and a Ag/AgCl
reference electrode. Tetrabutylammonium tetrafluoroborate 0.1 M in
DMF was used as electrolyte. The potential was controlled by an
EmStat potentiostat.
Determination of Quantum Yields. The emission quantum
yields were determined at room temperature (RT) by using
[Ru(bpy)3]Cl2 as reference and equation ΦX = Φref × (slope (X)/
2
slope (ref)) × (ηX /ηref2), where ΦX is the quantum yield of
compound X, Φref is the quantum yield of the reference compound
([Ru(bpy)3]CL2, Φ = 0.094 0.004),34 and η is the refraction index
of the solvent. However, the term depending on the solvent was
omitted, because acetonitrile was used as the solvent for all samples.
The slope for each compound was obtained from the plot of integrated
fluorescence intensity versus absorbance at five different concen-
trations (Figure 2b).
Time-Correlated Single-Photon Counting. The excited-state
lifetimes were measured by time-correlated single-photon counting
(TCSPC) in acetonitrile. Samples were deaerated and put under an
argon atmosphere by freeze−pump−thaw technique in a cuvette with
a degassing bulb. The light source was a pulsed laser at 470 nm (PEL-
470) with a repetition rate set at 50 kHz for measurements up to 5 μs
Photocatalytic Tests. Five samples (A−E) were tested in parallel.
Solution A contained Ph3P (300 mM), [Ru(bpy)2(L2)MoO2(solv)]-
(BF4)2 (1 mM), and methyl viologen dihexafluorophosphate (1 mM)
in CD3CN/deuterated dimethyl sulfoxide (DMSO-d6;1:1 v/v, 0.7
mL). The solution was transferred into a gas-tight NMR tube and
D
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
degassed via a minimum of three freeze−pump−thaw cycles using an
acetone/liquid nitrogen bath for cooling and research-grade argon for
equilibration. Four control samples were prepared in the same way.
Sample B had the same composition as sample A but was kept in the
dark. Sample C contained [Ru(bpy)2(H2-L2)] (BF4)2 (1 mM), 300
equiv of PPh3, and 1 equiv of methyl viologen dihexafluorophosphate.
Sample D contained [Ru(bpy)2(L2)MoO2(solv)](BF4)2 (1 mM) and
300 equiv of PPh3 but no methyl viologen dihexafluorophosphate.
Sample E contained [Ru(bpy)3)]Cl2 (1 mM) plus [MoO2(LMe)(solv)]
(1 mM), 300 equiv of PPh3, and 1 equiv of methyl viologen
dihexafluorophosphate.
Electronic Absorption and Emission Spectra. The
electronic absorption spectrum of [Ru(bpy)2(L2)-
MoO2(MeOH)](BF4)2, dissolved in MeCN, shows an intense
band in the range of 235−250 nm (ε242 = 71 900 M−1 cm−1),
ligand-centered (LC) transitions at 285 nm (ε285 = 98 000 M−1
cm−1) and 345 nm (ε345 = 32 900 M−1cm−1), and a Ru-based
metal-to-ligand (MLCT) transition at 455 nm (ε455 = 21 700
M−1 cm−1). Figure 2a shows the absorption spectrum of
[Ru(bpy)2(L2)MoO2(solv)]2+ in comparison with the absorp-
tion spectra of the two individual components of the dyad,
namely, [Ru(bpy)2(phen-NH2)]2+ and [MoO2(LMe) (solv)],
Samples A, C, D, and E were irradiated with a Philips 125 W
medium-pressure mercury vapor lamp equipped with a water infrared
filter and a cutoff filter (λ > 435 nm) for 17 h at a time followed by 7 h
of storage in the dark, until almost complete conversion was reached
by sample A (9 d). At the beginning and end of each irradiation
with H2 -LM e
= para-methylbenzaldehyde-N(1)-
ethylsemicarbazone28 and solv = MeOH or MeCN, (Scheme
2), and the sum of their spectra. It is evident that the
absorbance above 375 nm is dominated by transitions
associated with the Ru-based component of the dyad.
Excitation into the low-energy MLCT band at 470 nm gives
rise to a broad emission band with a maximum at 620 nm
(Figure 2a). As expected, the attachment of the Mo-based
catalytic unit leads to a slight change in the position and
intensity of the absorption and emission bands compared with
those reported for the parent photosensitizer [Ru-
(bpy)2(phen)]2+ (λmax,abs = 449 nm, ε449 = 15 700 M−1 cm−1,
λmax,em = 618 nm, MeCN).36,37
1
period, the formation of OPPh3 was monitored by H and 31P NMR
spectroscopy (1H NMR 500 MHz, 31P 202.5 MHz) and by a
photograph taken with and without flash to document changes in color
and apparent emission intensity.
To account for the different lengths of time taken for the
preparation of samples A−E, which required exposure to air and
light, the turnover numbers were set to zero at the start of the first
irradiation period. This allowed the direct comparison of the samples.
In an additional control sample that contained Ph3P (300 mM) in
CD3CN/DMSO-d6 (1:1 v/v, 0.7 mL) under Ar and was left standing
at RT for 9 d, only a trace amount of OPPh3 could be detected by 31
NMR.
P
If solutions with equal absorbance at 470 nm are compared,
the emission intensity of the molybdenum-containing dyad
[Ru(bpy)2(L2)MoO2(solv)]2+ is lower than that of the
thiosemicarbazone-appended photosensitizer [Ru(bpy)2(H2-
L2)]2+ (Figure 2b). The linear relationship between the
integrated luminescence intensity and the absorbance of
solutions of both [Ru(bpy)2(H2-L2)]2+ and [Ru(bpy)2(L2)-
MoO2(solv)]2+ at a range of different concentrations confirms
the absence of intermolecular quenching (Figure 2b).
Electron Paramagnetic Resonance Spectroscopy. The elec-
tron paramagnetic resonance (EPR) spectra were recorded at X band
on a Jeol JES-REIX spectrometer. Samples were irradiated in situ with
a 100 W mercury vapor arc equipped with a water infrared filter and a
cutoff filter (λ > 420 nm). The general EPR parameters were: power 1
mW, field 336.2 mT, scan width 7.5 mT, modulation width 0.1 mT,
time constant 0.1 s, scan time 60 s, modulation frequency 100 kHz,
and average of three scans at RT. Sample compositions and
concentrations were as follows. Sample 1: [Ru(bpy)2(H2-L2)]2+
(0.25 mM) and MV2+ (0.25 mM) in acetonitrile; sample 2:
[Ru(bpy)2(L2)MoO2(solv)]2+ (0.25 mM), PPh3 (75 mM) and MV2+
(0.25 mM) in acetonitrile; and sample 3: [Ru(bpy)2(L2)-
MoO2(solv)]2+ (0.25 mM), PPh3 (75 mM) and MV2+ (0.25 mM)
in acetonitrile/DMSO (1:1 v/v). All measurements were performed in
glass capillaries at RT and under Ar.
Time-Correlated Single-Photon Counting. The lumi-
nescence decay kinetics of [Ru(bpy)2(L2)MoO2(solv)]2+,
[Ru(bpy)2(phen-NH2)]2+, and [Ru(bpy)2(H2-L2)]2+ were
investigated by TCSPC. In an additional control experiment,
[Ru(bpy)2(L2)MoO2(solv)]2+ was investigated in the presence
of Ph3P (100-fold excess) to assess whether or not Ph3P
quenches the emission of the dyad. The samples were dissolved
in MeCN and excited at 470 nm, and time-resolved emission
data were collected at the maximum of the emission band at
620 nm (Figure 3 and Table 1). The emission decay curve
obtained for [Ru(bpy)2(phen-NH2)]2+ could be fitted using a
single exponential. The lifetime τ of 1100 1 ns obtained is
slightly longer than that reported for the parent photosensitizer
[Ru(bpy)2(phen)]2+, which was obtained under similar
conditions and also fitted to a single exponential (τ = 830 ns,
λexc = 450 nm, λem = 600 nm, MeCN, RT).37,37
RESULTS AND DISCUSSION
■
Synthesis and Characterization of [Ru(bpy)2(L2)-
MoO2(MeOH)](BF4)2, with H2-L2 = 2-Hydroxy-5-([1,10]-
phenanthroline-5-ylaminomethyl)-benzaldehyde-N(1)
phenylthiosemicarbazone. The photosensitizer [Ru-
(bpy)2(phen-NH2)](BF4)2 was prepared by adapting a
procedure reported for the synthesis of the corresponding
PF6 salt33 and was used as a platform for the sequential
−
The emission decay of [Ru(bpy)2(H2-L2)]2+ required a
synthesis of the catalytic unit (Scheme 2). 5-Chloromethyl-2-
hydroxybenzaldehyde was prepared32 and attached to the
ruthenium complex via nucleophilic substitution. The resulting
intermediate was condensed with commercially available 4-
phenylthiosemicarbazide by adapting a published procedure.35
The phenyl group was chosen to increase the hydrophobic
character of the compound and to facilitate its isolation by
precipitation. The photosensitizer-appended ligand [Ru-
(bpy)2(H2-L2)](BF4)2 obtained was reacted with
[MoO2(acac)2] to produce dyad [Ru(bpy)2(L2)-
MoO2(MeOH)](BF4)2. The product and intermediates were
3
double exponential fit, indicating that there are two MLCT
excited states contributing to the emission. The lifetime of the
major, longer-lived component of 1100
2 ns (60%) is
identical with the lifetime of [Ru(bpy)2(phen-NH2)]2+, while
the shorter-lived, minor component has a lifetime of 303 1 ns
(40%). Hence, the two excited states may tentatively be
assigned to bpy- and H2-L2-localized 3MLCT states,
respectively, taking into account that the distribution deviates
slightly from the statistical distribution of 66% and 33%. Dual
emission is frequently observed in heteroleptic Ru(II)
complexes of this type.37,38
1
characterized by H and 13C NMR spectroscopy, ESI mass
spectrometry, and elemental analyses (see Experimental
Section).
The emission decay of the dyad [Ru(bpy)2(L2)-
MoO2(solv)]2+ required a three-component fit. The lifetime
E
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Emission decay profiles obtained at 620 nm (λexc = 470 nm)
under Ar for 1 × 10−5 M solutions of [Ru(bpy)2(phen-NH2)]2+
(blue), [Ru(bpy)2(H2-L2)]2+ (green), [Ru(bpy)2(L2)MoO2(solv)]2+
(red), and [Ru(bpy)2(L2)MoO2(solv)]2+ with 100-fold excess of Ph3P
(black) in MeCN.
Figure 4. CV of [Ru(bpy)2(L2)MoO2(solv)](BF4)2 (scan range: from
0 to +1.7 V and from −2.0 to 0 V (red); between 0 and +1.7 V (green)
and between 0 and −2 V (blue); 1 mM with 0.1 M [Bu4N](BF4) in
anhydrous DMF, potentials vs Ag/AgCl, scan rate 100 mV s−1.
Table 1. Quantum Yields (Φ) of Luminescence and
3
Lifetimes (τ) of the MLCT States in MeCN
electrolyte, shows several redox processes that are also evident
in the CVs of its components (Figure 4, Figures S1−S3, and
Table 2). In the positive region, an irreversible oxidation wave
is observed that involves the transfer of approximately two
electrons, based on the comparison with the peak currents
associated with the reversible processes in the negative
potential region. One of the overlapping oxidation steps can
be assigned to the oxidation of Ru(II) to Ru(III) by
comparison with electrochemical data obtained for [Ru(bpy)3]-
Cl2 under analogous conditions (Table 2 and Figure S1),
previously reported oxidation potentials for similar Ru(II)
complexes,33 and spectro-electrochemical investigations (see
below). The overlapping second irreversible oxidation process
is associated with the molybdenum-coordinated thiosemicarba-
zone. A comparison with the CVs obtained for [MoO2(LMe)-
(solv)] and H2-LMe under analogous conditions (Figures S2
and S3), and observations reported for similar Mo(VI)-
semicarbazone complexes,40,41 provide further evidence for
this assertion. It is known that a one-electron oxidation of
thiosemicarbazones can lead to dimerization via disulfide-bond
formation, while a two-electron oxidation can lead to
cyclization and formation of 1,2,4-triazole-3-thione or 1,3,4-
thiadiazole rings.42 The latter is in accordance with the
irreversibility and high peak current of the oxidation process
observed at a peak potential (Ep,ox) of 1.43 V in the CV of
[MoO2(LMe)(solv)] (Table 2).
a
b
Φ470
τ (ns)
[Ru(bpy)2(phen-
NH2)]2+
0.078 0.003
1100
1
[Ru(bpy)2(H2-L2)]2+
0.076 0.003
0.044 0.002
303 1 (40%), 1100 2 (60%)
[Ru(bpy)2(L2)
11.9 0.1 (2%), 320 1 (31%),
1149 2 (67%)
MoO2(solv)]2+
[Ru(bpy)2(L2)
not determined 8.1 0.1 (1%), 290 1 (31%),
1037 1 (68%)
MoO2(solv)]2+
Ph3P
+
a
Argon atmosphere, 1 × 10−5 M in MeCN, [Ru(bpy)3]Cl2 reference:
Φ = 0.094 0.004,34 λexc = 470 nm, λem = 620 nm, error margins
represent one standard deviation and are propagated from the
b
standard error in the fitted gradients (plot shown in Figure 2b). λexc
470 nm, λem = 620 nm. Error margins represent one standard
deviation.
=
of the longest-lived component of 1149 2 ns resembles those
obtained for [Ru(bpy)2(phen-NH2)]2+ and [Ru(bpy)2(H2-
L2)]2+, and the close to statistical contribution of 67% suggests
that it is due to decay from a bpy-localized excited state. The
second lifetime of 320 1 ns resembles that of the shorter-
lived component obtained for [Ru(bpy)2(H2-L2)]2+, and again,
the close to statistical contribution of 31% indicates decay from
a H2-L2-centered excited state. An additional minor (2%) short-
lived component with τ = 11.9 0.1 ns had to be added to fit
the data satisfactorily.
The spectroscopic investigations show that the covalently
linked catalytic unit of the dyad decreases the quantum yield of
the photoactive unit. On the one hand, in similar systems,
quenching via intramolecular photoinduced electron transfer
from an appended electron-rich phenolate ligand to a
photoexcited metal-based luminophore, such as RuIII-bpy•−,
has been observed using time-resolved absorption spectrosco-
py.39 On the other hand, the addition of a 100-fold excess of
PPh3 does not have a significant effect on the three lifetimes
obtained for [Ru(bpy)2(L2)MoO2(solv)]2+ (Figure 3 and Table
1), confirming the absence of an intermolecular quenching
effect of the substrate.
In addition, there are weak irreversible waves at Ep,ox = 0.58
and 1.07 V and Ep,red = −0.27, −0.72, and −1.12 V that can be
assigned to the Mo-containing unit and/or its degradation
products as similar waves are observed in the CV of
[MoO2(LMe)(solv)] (Figure S2). While the shoulder at −1.12
V lies within the potential range typically associated with the
reduction of Mo(VI) to Mo(V),40,41 the irreversible reduction
waves at −0.27 and −0.72 V are only present if the positive
potential range is scanned first (Figure 4). Hence, these two
reduction processes appear to be associated with chemical
reaction products formed upon oxidation.
By comparison with the CV of [Ru(bpy)3]Cl2 (Table 2 and
Figure S1) and literature data,33 the three reversible waves at
E1/2 = −1.22, −1.41, and −1.75 V can be assigned to three
successive one-electron reductions of the Ru-polypyridine unit.
Electrochemistry. The cyclic voltammogram (CV) of
[Ru(bpy)2(L2)MoO2(solv)](BF4)2, recorded between −2 and
1.7 V in dry DMF using [Bu4N](BF4) as the supporting
F
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Electrochemical Data
E1/2/V (ΔEp/mV)
[Ru(polypy)3]0/−
E1/2/V (ΔEp/mV)
[Ru(polypy)3]+/0
E1/2/V (ΔEp/mV)
[Ru(polypy)3]2+/+
Ep,ox/V
(RuII/RuIII)
complex
[Ru(bpy)2(L2)-MoO2(solv)](BF4)2
−1.75 (120)
−1.68 (90)
−1.41 (80)
−1.37 (70)
−1.22 (60)
−1.19 (60)
1.46
1.41
[Ru(bpy)3]Cl2
compound
E
p,red/V (Mo unit)
Ep,ox/V (Mo unit)
[Ru(bpy)2(L2)-MoO2(solv)](BF4)2
[MoO2(LMe)(solv)]
−1.12, −0.72, −0.27
−1.14, ∼0.7, ∼0.3
∼1.0, 1.46
1.07, 1.43
E1/2/V (ΔEp/mV) MV+/2+
compound
E
1/2/V (ΔEp/mV) MV0/+
methyl viologen (MV) dichloride
−0.73 (80)
−0.35 (80)
1
1 mM solutions in degassed anhydrous DMF with 0.1 M [Bu4N](BF4) as supporting electrolyte, potentials vs Ag/AgCl, scan rate 100 mV s−1.
Spectro-Electrochemical Oxidation. [Ru(bpy)2(L2)-
MoO2(solv)](BF4)2 was oxidized electrochemically at 1.5 V
using a carbon veil electrode, and the difference absorption
spectra were recorded (Figure 5). Two negative bands
Figure 6. Difference spectra obtained during the spectro-electro-
chemical oxidation of [MoO2(LMe)MeOH] (0.5 mM) recorded at Eapp
= 1.5 V after 1, 2, 3, 4, 6, 8, 10, 12, 15, 18, 21, 25, 30, 35, and 40 min
(carbon veil electrode).
the clean interconversion of two defined chromophores. Hence,
Figure 5. Difference spectra obtained during the spectro-electro-
the initial oxidation of the phenol−thiosemicarbazone ligand
triggers subsequent chemical change(s), consistent with the
irreversible redox wave seen in the CV of [MoO2(LMe)-
(MeOH)] (Figure S2). It is conceivable that the initial ligand
oxidation causes a change in the coordination environment,
which may be followed by an oxidative dimerization or, more
likely, an intramolecular cyclization of the ligand, as reported
for structurally related thiosemicarbazones.42
While the bleach observed at 323 nm is consistent with the
gradual disappearance of the isosbestic point at 323 nm and the
bleach in the dyad spectra, the absorbance increase at ∼359 nm
is not reflected in the spectra of dyad. Hence, not all of the
spectro-electrochemical features of the dyad can be attributed
to contributions from its two components.
chemical oxidation of [Ru(bpy)2(L2)MoO2(solv)](BF4)2 (0.5 mM),
recorded at Eapp = 1.5 V after 5 s and after 1, 2, 3, 4, 5, 6, 7, 8, 10, 12,
14, 16, 18, 20, 23, 26, 30, and 35 min (carbon veil electrode).
(bleaches) developed over time at 360 and 470 nm, while
the absorbance below 340 nm increased, and a weak and broad
band developed above 500 nm. Bleaching in the visible region
of the broad MLCT band of the dyad is consistent with the
oxidation of Ru(II) to Ru(III). However, the relative intensity
of the two apparent bleaches at 360 and 470 nm differs from
that observed by Lomoth et al. for [Ru(bpy)3]2+ in
acetonitrile43 and indicates the presence of another chromo-
phore. Likewise, the fact that the isosbestic point at 332 nm
gradually disappears after ∼15 min of electrolysis is consistent
with the formation of another oxidized species.
Spectro-Electrochemical Reduction. The spectroscopic
profiles obtained upon electrochemical reduction (−1.2 V) are
shown in Figure 7. Two positive bands rise at 390 and 510 nm
as well as two negative bands at 350 and 450 nm.
For comparison, the reference compound [MoO2(LMe)
(MeOH)], which can be considered as representative of the
catalytic moiety of the dyad, was oxidized under conditions
analogous to those used for the oxidation of the dyad (Figure
6). In [MoO2(LMe) (MeOH)], the molybdenum is already in
its group oxidation state +VI and cannot oxidize further, yet
significant spectroscopic changes are observed in the visible
region of the spectra. Overall, the absorbance at 295 and 350
nm increases, while the absorbance at 320 nm and between 370
and 500 nm decreases.
Similarly, [Ru(bpy)3]Cl2 and [MoO2(LMe)(MeOH)] were
reduced at the same potential (Figures S5 and S6). A
comparison of the spectra obtained for the complexes shows
that the dyad behaves similarly to [Ru(bpy)3]Cl2 above 400
nm, while the spectral changes observed below 400 nm are
consistent with the reduction of both the ruthenium and the
molybdenum moieties.
A reduction of the dyad was then performed at −1 V (Figure
8a) to investigate if it is possible to selectively reduce only the
Mo-based catalytic center. The profile obtained is very different
Interestingly, after 10 min of electrolysis, four isosbestic
points emerge at 282, 310, 332, and 370 nm and point toward
G
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
from that recorded for the dyad upon reduction at the more
negative potential of −1.2 V (Figure 7). A positive band at 490
nm and three overlapping negative bands in the region between
300 and 450 nm grow in over time. The positive band
resembles the corresponding band in the spectrum of the
molybdenum complex (Figure 8b). Despite the absence of the
small positive band at 360 nm, the absorbance between 300 and
450 nm decreases, whereas it increases when [Ru(bpy)3]Cl2 is
reduced.
The spectrum obtained upon electrochemical reduction of
[Ru(bpy)2(L2)MoO2(solv)](BF4)2 at −1 V was then compared
with the spectrum obtained after the stoichiometric chemical
reduction of the dyad, which was performed at a 25 μM
concentration in MeCN in the absence of an O atom donor. To
ensure complete reduction of the Mo-based unit (turnover
number = 0.5) a 100-fold excess of the O atom acceptor PPh3
was used. The initial spectrum of [Ru(bpy)2(L2)MoO2(solv)]-
(BF4)2 was subtracted from the spectrum recorded after
chemical reduction, to obtain a difference spectrum (Figure
8b). The profiles of the spectra obtained after electrochemical
and chemical reduction are very similar. Therefore, the spectro-
electrochemical measurements confirm that the molybdenum-
based moiety can be selectively reduced at −1 V, while the
reduction of the ruthenium-based moiety requires a more
negative potential.
Figure 7. Difference absorption spectra obtained during the spectro-
electrochemical reduction of [Ru(bpy)2(L2)MoO2(solv)](BF4)2 (0.5
mM), recorded at Eapp = −1.2 V after 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12,
14, and 18 min (carbon veil electrode).
Photoactivated Oxygen Atom Transfer Catalysis. The
catalytic activity of the dyad was investigated using triphenyl
phosphine as the oxygen acceptor and DMSO as the oxygen
donor. Under an atmosphere of argon, a solution containing
[Ru(bpy)2(L2)MoO2(solv)]2+ (1 mM), triphenyl phosphine
(300 equiv), and methyl viologen (MV2+, 1 equiv) in a mixture
of deuterated acetonitrile/DMSO (1:1 v/v; sample A) was
subjected to alternating cycles of irradiation with >435 nm light
(17 h) followed by dark storage (7 h). At the beginning and
end of each irradiation period, the formation of triphenyl
phosphine oxide was monitored by 1H and 31P NMR
spectroscopy. As evident from Figure 9, the reaction proceeded
significantly faster during the periods of light exposure than in
the dark. After a total irradiation time of 154 h, 95% completion
was reached, and the presence of dimethyl sulfide in the sample
was clearly detectable by its characteristic smell. The catalyst
proved to be remarkably stable under these conditions, with the
turnover number only limited by the solubility of triphenyl
1
phosphine in DMSO/acetonitrile (1:1 v/v). The H NMR
spectra of sample A were examined before and after each
irradiation period to monitor the integrity of the catalyst. Even
after a total irradiation time of 154 h, the proton resonances of
the dyad remained essentially unchanged (Figure S8).
A control experiment in which an analogous solution was
kept in the dark all the time (sample B, Figure 9) produced
much less of the oxidation product, indicating that the thermal
conversion rate is much slower than the photochemical
1
conversion. Again, the H NMR spectra recorded during the
thermal reaction showed no evidence for either catalyst
decomposition or the formation of LMo(O)-O-Mo(O)L
dimers (Figure S9). A control sample containing [Ru-
(bpy)2(L2)MoO2(solv)]2+ (1 mM) and 300 equiv of
triphenylphosphine but no methyl viologen (sample D) yielded
a similar amount of product to the dark control after irradiation
for a total of 154 h (Figure 9), demonstrating that methyl
viologen accelerates the reaction in the presence of light. A
control sample containing [Ru(bpy)2(H2-L2)]2+ but no
molybdenum (sample C) was found to produce less
Figure 8. (a) Difference spectra obtained during the spectro-
electrochemical reduction of [Ru(bpy)2(L2)MoO2(solv)](BF4)2 (0.5
mM), recorded at Eapp. = −1 V after 10, 30, 60, and 90 s and 2, 3, 4,
and 6 min (carbon veil electrode); (b) electronic absorption spectrum
of [Ru(bpy)2(L2)MoO2(solv)](BF4)2 (25 μM) in MeCN before
(black) and after reaction with Ph3P (2.5 mM) (red) and the
corresponding difference spectrum (green).
H
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 10. Photographs of samples A−E taken immediately after
irradiation (a) and after storage in the dark (b), both without (left)
and with (right) flash (sample A: [Ru(bpy)2(L2)MoO2(solv)]2+, 300
equiv of PPh3 and 1 equiv of MV2+; sample B: control kept in the dark;
sample C: [Ru(bpy)2(H2-L2)]2+, 300 equiv of PPh3 and 1 equiv of
MV2+; sample D: [Ru(bpy)2(L2)MoO2(solv)]2+, 300 equiv of PPh3
but no MV2+; sample E: [Ru(bpy)3]2+ plus [MoO2(LMe)(solv)], 300
equiv of PPh3 and 1 equiv of MV2+).
Figure 9. Performance of selected catalysts and control systems in the
oxidation of PPh3 during alternating cycles of irradiation with >435 nm
light (17 h, indicated by yellow bars) followed by storage in the dark
(7 h). Conditions used: deuterated acetonitrile/DMSO (1:1 v/v), Ar
atmosphere, RT, (sample A: [Ru(bpy)2(L2)MoO2(solv)]2+, 300 equiv
of PPh3 and 1 equiv of MV2+; sample B: control kept in the dark;
sample C: [Ru(bpy)2(H2-L2)]2+, 300 equiv of PPh3 and 1 equiv of
MV2+; sample D: [Ru(bpy)2(L2)MoO2(solv)]2+, 300 equiv of PPh3
but no MV2+; sample E: [Ru(bpy)3)]2+ plus [MoO2(LMe) (solv)], 300
equiv of PPh3, and 1 equiv of MV2+).
These observations are consistent with a mixture that contains
the orange and luminescent Ru(II)-based photosensitizer unit
and dark blue-green MV•+, one of the products of the oxidative
quenching of the RuIII-bpy/phen•− 3MLCT excited state by
MV2+, which initially produces RuIII-bpy/phen and MV•+.
While in the absence of other reductants, back electron transfer
from MV•+ to RuIII occurs on the nanosecond time scale,43 in
our case the MV•+ radical is much longer-lived (see below). In
accordance with the irreversible oxidation wave seen in the CV
of the dyad and its emission lifetimes, we propose that fast
reduction of the RuIII center occurs via electron transfer from
the thiosemicarbazone ligand of the Mo complex rather than
back electron transfer from MV•+. To confirm this proposal, the
absorption spectra of sample A were recorded before
irradiation, immediately after selected irradiation intervals,
and after storage in the dark. The characteristic absorption
bands due to MV•+ at λmax 397 and 609 nm emerge during
triphenylphosphine oxide after irradiation for the same period
of time than the dark control. The last experiment confirms
that the molybdenum(VI) dioxo unit increases the reaction rate
by catalyzing the OAT between dimethyl sulfoxide and
triphenylphosphine.
When [Ru(bpy)3]2+ and [MoO2(LMe)(solv)] were tested in
a solution that contained the two individual components
separately in a 1:1 ratio (sample E), a significant amount of
triphenyl phosphine oxide was produced but ∼35% less than
with the dyad. The comparison of the catalytic activity of the
dyad with the two-component system indicates that the
intramolecular electron transfer between the photosensitizer
and the catalytic center is more efficient than intermolecular
electron transfer between two individual components.
irradiation (Figure 11).43 Bleaching of the 3MLCT band (λmax
=
Mechanistic information can be gathered from a comparison
of the color and luminescence of the samples before and after
each irradiation cycle. Therefore, after each irradiation and dark
period, photographs of samples A, C, D, and E were taken both
with and without flash to document any changes in emission
intensity and color, respectively, in comparison with sample B,
which was kept in the dark (Figure 10).
455 nm), which would be expected upon oxidation of RuII to
RuIII, is not observed. Consequently, the rereduction of the
photooxidized RuIII to RuII has occurred on the time scale that
it took to record the absorption spectrum (∼2 min). However,
the MV•+ radical is still present and hence is not the source of
the reducing equivalent.
During the storage periods in the dark, the color of both
NMR samples (A and E) returned to orange, and the emission
was restored, indicating that the resting state of the Ru(II)-
based photosensitizer is regenerated and the MV•+ radical
cation is reoxidized to MV2+ (Figure 10b). Light is not required
for this reaction to proceed. The electronic absorption spectra
confirm a decrease in the MV·+ concentration in the dark, with
the spectrum resembling that obtained after 1 h of irradiation
(Figure 11).
Electron Paramagnetic Resonance Spectroscopy. In
addition, EPR spectroscopy was used to monitor radical
formation. During in situ irradiation of a solution containing
[Ru(bpy)2(H2-L2)]2+ (0.25 mM) and MV2+ (0.25 mM) in
acetonitrile, a strong EPR signal due to MV·+ was observed
(Figure 12, Sample 1).44 Even 1 h after stopping the irradiation,
Initially, all five solutions were bright orange (Figure S7).
Upon irradiation, the sample containing [Ru(bpy)2(H2-L2)]2+
(sample C) changed color from orange to dark green, and the
emission was lost due to the oxidative quenching of the Ru(II)*
excited state by MV2+ leading to the formation of MV•+, which
strongly absorbs at the emission wavelength. (Figure 10a).43 In
the absence of the cis-dioxo molybdenum center, the dark green
color remained, even after storage of the sample for 7 h in the
dark. In this sample, a subsequent OAT could not take place,
since the catalytic Mo center was missing.
Upon irradiation, samples A and E, [Ru(bpy)2(L2)-
MoO2(solv)]2+ and [Ru(bpy)3]2+ plus [MoO2(LMe) (solv)],
respectively, changed color from bright orange to brown, and
the intensity of the apparent emission decreased (Figure 10a).
I
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Subsequently, DMSO was added to the dyad-containing
sample, and the EPR spectrum was recorded, while the sample
was irradiated. In this case, a weaker signal due to MV•+ was
observed (Figure 12, Sample 3). The decrease is consistent
with the expected decrease of the MV•+ concentration under
turnover conditions. In the dark, the signal decayed, with ∼33%
of the original intensity remaining after 1 h. Again, ligand- or
Mo-based signals could not be detected, since the spectrum was
dominated by the MV•+ signal.
Proposed Catalytic Cycle. On the basis of our
observations and the information obtained from photophysical,
spectroscopic, and electrochemical investigations, the following
catalytic mechanism is proposed (Scheme 3). In the first step of
Scheme 3. Proposed Catalytic Mechanism That Combines
Photoredox Catalysis Involving Single Electron Transfer
(SET) with Oxygen Atom Transfer (OAT)
Figure 11. Electronic absorption spectra of a dilute solution of sample
A recorded before irradiation, immediately after selected irradiation
intervals, and after subsequent storage in the dark. Conditions:
[Ru(bpy)2(L2)MoO2(solv)]2+ (0.25 mM), PPh3 (75 mM), and MV2+
(0.25 mM) in acetonitrile/DMSO 1:1 v/v, RT, Ar, 0.2 cm path length.
the photoredox cycle, the photoexcited dyad [RuII*(bpy)2(L2)-
MoVIO2(solv)]2+ is oxidatively quenched by MV2+ to give
[RuIII(bpy)2(L2)MoVIO2(solv)]2+ and MV•+, as evidenced by
the change in color from orange to dark blue-green, the EPR
spectra, and the loss of emission. The oxidized dyad then
undergoes fast intramolecular electron transfer to form
[RuII(bpy)2(L2)+MoVIO2(solv)]2+, where L+ represents the
oxidized form of Mo-coordinating phenol−thiosemicarbazone
ligand. This proposed step of the photoredox cycle is supported
by both the visible spectra obtained shortly after irradiation and
the irreversible oxidation wave in the cyclic voltammogram of
the dyad. As a consequence, back electron transfer from MV•+
to RuIII is outcompeted. The presence of MV•+ is clearly seen in
Sample C, in which the deep blue-green color of MV•+ persists,
since the catalytic OAT cycle cannot turn over in the absence of
molybdenum.
In the presence of molybdenum (sample A), the subsequent
OAT to PPh3 gives OPPh3, which was detected by NMR
spectroscopy, and a species with an electronic absorption
spectrum that closely resembles that obtained for the species
generated by electrochemical reduction, tentatively assigned as
[RuII(bpy)2(L2)MoVO(solv)]2+. The reductive half-reaction
with DMSO produces DMS and regenerates MV2+ and RuII-
(L)MoVI(O)2. While in-depth investigations are required to
determine the exact mechanistic and kinetics details, in
particular, of the reductive half-reaction, our tests show that
the photoactivated OAT cycle runs significantly faster than the
thermal OAT cycle, which proceeds in the absence of light.
Figure 12. EPR spectra obtained during in situ irradiation of Sample 1:
[Ru(bpy)2(H2-L2)]2+ (0.25 mM) and MV2+ (0.25 mM) in acetonitrile,
Sample 2: [Ru(bpy)2(L2)MoO2(solv)]2+ (0.25 mM), PPh3 (75 mM)
and MV2+ (0.25 mM) in acetonitrile, and Sample 3: [Ru(bpy)2(L2)-
MoO2(solv)]2+ (0.25 mM), PPh3 (75 mM) and MV2+ (0.25 mM) in
acetonitrile/DMSO (RT, Ar); simulated spectra are shown in red;
simulation parameters are listed in Table S1.
66% of the original MV•+ signal remained, confirming that the
radical is long-lived and back electron transfer from MV•+ to
the complex, most likely an oxidized form of the ligand H2-L2,
is very slow. However, a H2-L2-based radical could not be
detected, since the spectrum was dominated by the intense
MV•+ signal.
During in situ irradiation of a sample containing [Ru-
(bpy)2(L2)MoO2(solv)]2+ (0.25 mM), PPh3 (75 mM), and
MV2+ (0.25 mM) in acetonitrile (no DMSO), a clear signal due
to MV•+ was observed (Figure 12, Sample 2). In line with the
lower quantum yield of the dyad, the signal was weaker than
that observed with [Ru(bpy)2(H2-L2)]2+. An L2-based radical
or paramagnetic Mo species could not be detected. Again, the
strong signal of the exceptionally stable MV•+ radical is likely to
mask any signals of shorter-lived radical species.
J
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
support and Drs. A. Parkin and V. Chechik for helpful
discussions. Experimental support provided by K. Heaton
(mass spectrometry), H. Fish (NMR), and P. Groves (EPR) is
gratefully acknowledged.
CONCLUSIONS
■
The Ru(II)−Mo(VI) dyad [Ru(bpy)2(L2)MoO2(solv)]2+ was
synthesized, fully characterized, and shown to accelerate the
rate-determining step in the cis-dioxo molybdenum-catalyzed
OAT cycle, the transfer of an oxo ligand to PPh3, via a
photoredox process. The energy provided by visible light is
sufficient to generate a highly reactive, oxidized form of the
molybdenum-centered catalyst, in which the noninnocent
phenol−thiosemicarbazone ligand plays key role. The single
electron transfer (SET) step in the photoredox process is
essential in making the reactive one-electron oxidized form of
the thiosemicarbazone complex accessible. In contrast, electro-
chemical methods trigger an irreversible two-electron oxidation
process. Methylviologen acts as a mediator in the photoredox
cycle, but it is regenerated and hence only required in catalytic
rather than sacrificial amounts. By combining the ruthenium-
polypyridyl-based photoredox cycle with the molybdenum-
catalyzed OAT cycle, the rate acceleration of the OAT can be
switched on and off via the light source.
REFERENCES
■
(1) Harel, A.; Bromberg, Y.; Falkowski, P. G.; Bhattacharya, D.
Evolutionary history of redox metal-binding domains across the tree of
life. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 7042−7047.
(2) Holm, R. H.; Donahue, J. P. A thermodynamic scale for oxygen
atom transfer reactions. Polyhedron 1993, 12, 571−589.
(3) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. Proton-coupled
electron flow in protein redox machines. Chem. Rev. 2010, 110, 7024−
7039.
(4) Warren, J. J.; Mayer, J. M. Moving protons and electrons in
biomimetic systems. Biochemistry 2015, 54, 1863−1878.
(5) Ghosh, S.; Hogarth, G.; Hollingsworth, N.; Holt, K. B.; Kabir, S.
E.; Sanchez, B. E. Hydrogenase biomimetics: Fe2(CO)4(μ-dppf)(μ-
pdt) (dppf = 1,1′-bis(diphenylphosphino)ferrocene) both a proton-
reduction and hydrogen oxidation catalyst. Chem. Commun. 2014, 50,
945−947.
The dyad can be regarded as an advanced bioinspired model
for Mo-containing oxotransferases, since it demonstrates the
importance of both spatially separated one-electron transfer
units and the participation of a noninnocent S-containing
ligand in OAT catalysis.
(6) Camara, J. M.; Rauchfuss, T. B. Combining acid-base, redox and
substrate binding functionalities to give a complete model for the
[FeFe]-hydrogenase. Nat. Chem. 2011, 4, 26−30.
(7) Huttinger, K.; Forster, C.; Heinze, K. Intramolecular electron
̈
̈
transfer between molybdenum and iron mimicking bacterial sulphite
dehydrogenase. Chem. Commun. 2014, 50, 4285−4288.
(8) Whiteoak, C. J.; Britovsek, G. J. P.; Gibson, V. C.; White, A. J. P.
Electronic effects in oxo transfer reactions catalyzed by salan
molybdenum(VI) cis-dioxo complexes. Dalton Trans. 2009, 2337−
2344.
Interestingly, significant rate enhancement was also observed
when two separate sensitizer and catalyst components were
tested. Even though the effect seen with the bimolecular system
was less pronounced than that observed with the covalently
linked dyad, the reduced synthetic effort allows for the testing
of different catalyst-sensitizer combinations and a higher-
throughput approach could facilitate the discovery and
optimization of new photoredox-catalytic systems in the future.
(9) Herrero, C.; Quaranta, A.; Leibl, W.; Rutherford, A. W.;
Aukauloo, A. Artificial photosynthetic systems. Using light and water
to provide electrons and protons for the synthesis of a fuel. Energy
Environ. Sci. 2011, 4, 2353−2365.
ASSOCIATED CONTENT
* Supporting Information
(10) Ott, S.; Borgstrom, M.; Kritikos, M.; Lomoth, R.; Bergquist, J.;
̈
■
Akermark, B.; Hammarstrom, L.; Sun, L. Model of the iron
̈
S
hydrogenase active site covalently linked to a ruthenium photo-
sensitizer: synthesis and photophysical properties. Inorg. Chem. 2004,
43, 4683−4692.
(11) Summers, P. A.; Calladine, J. A.; Ghiotto, F.; Dawson, J.; Sun, S.-
Z.; Hamilton, M. L.; Towrie, M.; Davies, E. S.; McMaster, J.; George,
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.6b01485.
CVs of [Ru(bpy)3]Cl2, [MoO2(LMe)(solv)], H2-LMe, and
methyl viologen dichloride; spectro-electrochemical
information provided by control compounds; photo-
graph of samples A−E before irradiation; EPR simulation
M. W.; Schroder, M. Synthesis and photophysical study of a [NiFe]
̈
hydrogenase biomimetic compound covalently linked to a Re-diimine
photosensitizer. Inorg. Chem. 2016, 55, 527−536.
1
parameters, H and 31P NMR spectra recorded after
(12) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Visible light
photoredox catalysis with transition metal complexes: applications in
organic synthesis. Chem. Rev. 2013, 113, 5322−5363.
irradiation for samples A, C, D, and E; NMR spectra and
mass spectra of synthesized compounds. (PDF)
(13) Angnes, R. A.; Li, Z.; Correia, C. R. D.; Hammond, G. B. Recent
synthetic additions to the visible light photoredox toolbox. Org. Biomol.
Chem. 2015, 13, 9152−9167.
AUTHOR INFORMATION
Corresponding Author
*E-mail: anne.duhme-klair@york.ac.uk.
■
(14) Lang, X.; Zhao, J.; Chen, X. Cooperative photoredox catalysis.
Chem. Soc. Rev. 2016, 45, 3026−3038.
(15) Iali, W.; Lanoe, P.-H.; Torelli, S.; Jouvenot, D.; Loiseau, F.;
Funding
́
Lebrun, C.; Hamelin, O.; Menage, S. A ruthenium(II)-copper(II) dyad
for the photocatalytic oxygenation of organic substrates mediated by
EPSRC (EP/J019666/1), COST Action CM1003.
Notes
dioxygen activation. Angew. Chem., Int. Ed. 2015, 54, 8415−8419.
́ ́
(16) Herrero, C.; Quaranta, A.; Sircoglou, M.; Senechal-David, K.;
The authors declare no competing financial interest.
Experimental data sets associated with this paper were
deposited with the Univ. of York library (www.york.ac.uk/
library/info-for/researchers/datasets/).
Baron, A.; Marín, I. M.; Buron, C.; Baltaze, J.-P.; Leibl, W.; Aukauloo,
A.; Banse, F. Successive light-induced two electron transfers in a Ru-Fe
supramolecular assembly: from Fu-Fe(II)-OH2 to Ru-Fe(IV)-oxo.
Chem. Sci. 2015, 6, 2323−2327.
(17) Hille, R. The mononuclear molybdenum enzymes. Chem. Rev.
1996, 96, 2757−2816.
(18) Basu, P.; Raitsimring, A. M.; LaBarre, M. J.; Dhawan, I. K.;
Weibrecht, J. L.; Enemark, J. H. Covalently linked oxomolybdenum-
(V) and iron(III) porphyrin centers: Synthetic models for the
ACKNOWLEDGMENTS
■
We thank the EPSRC (EP/J019666/1), COST Action
CM1003 (Biological oxidation reactionsmechanisms and
design of new catalysts) and the Univ. of York for financial
K
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
molybdenum-iron interaction in sulfite oxidase. J. Am. Chem. Soc.
1994, 116, 7166−7176.
(19) Wall, M. H.; Basu, P.; Buranda, T.; Wicks, B. S.; Findsen, E. W.;
Ondrias, M.; Enemark, J. H.; Kirk, M. L. Potoinduced electron transfer
in covalently linked oxomolybdenum(V) porphyrin systems. Inorg.
Chem. 1997, 36, 5676−5677.
(20) Heinze, K. Bioinspired functional analogs of the active site of
molybdenum enzymes: Intermediates and mechanisms. Coord. Chem.
Rev. 2015, 300, 121−141.
(21) Holm, R. H. Metal-centered oxygen atom transfer reactions.
Chem. Rev. 1987, 87, 1401−1449.
(39) Zhang, M.-T.; Irebo, T.; Johansson, O.; Hammarstrom, L.
̈
Proton-coupled electron transfer from tyrosine: A strong rate
dependence of intramolecular proton transfer distance. J. Am. Chem.
Soc. 2011, 133, 13224−13227.
̈
(40) Duman, S.; Kizilcikli, I.; Koca, A.; Akkurt, M.; Ulkuseven, B.
̈
ONN-complexes of dioxomolybdenum(VI) with 2-hydroxy-1-naph-
thaldehyde S-ethyl-4-H/phenyl-thiosemicarbazones: Crystal structure,
electrochemistry and in situ spectroelectrochemistry. Polyhedron 2010,
29, 2924−2932.
(41) Mondal, J. U.; Zamora, J. G.; Kinon, M. D.; Schultz, F. A. Six-
coordinate monooxo molybdenum(VI) complexes with catecholate
and salicylaldehyde thiosemicarbazone ligands. Inorg. Chim. Acta 2000,
309, 147−150.
(22) Enemark, J. H.; Cooney, J. J. A.; Wang, J.-J.; Holm, R. H.
Synthetic analogues and reaction systems relevant to the molybdenum
and tungsten oxotransferases. Chem. Rev. 2004, 104, 1175−1200.
(23) Sugimoto, H.; Tsukube, H. Chemical analogues relevant to
molybdenum and tungsten enzyme reaction centres toward structural
dynamics and reaction diversity. Chem. Soc. Rev. 2008, 37, 2609−2619.
(24) Schulzke, C. Molybdenum and tungsten oxidoreductase models.
Eur. J. Inorg. Chem. 2011, 2011, 1189−1199.
́
(42) Pedrido, R.; Romero, M. J.; Bermejo, M. R.; Gonzalez-Noya, A.
M.; García-Lema, I.; Zaragoza, G. Metal-catalyzed oxidation processes
in thiosemicarbazones: New complexes with the ligand N-{2-([4-N-
ethylthiosemicarbazone]-methyl)phenyl}-p-toluenesulfonamide.
Chem. - Eur. J. 2008, 14, 500−512.
(43) Lomoth, R.; Haupl, T.; Johansson, O.; Hammarstrom, L. Redox-
̈
̈
switchable direction of photoinduced electron transfer in an
Ru(bpy)32+-viologen dyad. Chem. - Eur. J. 2002, 8, 102−110.
(44) Dunham, W. R.; Fee, J. A.; Harding, L. J.; Grande, H. J.
Application of fast Fourier transforms to EPR spectra of free radicals in
solution. J. Magn. Reson. 1980, 40, 351−359.
(25) Lobana, T. S.; Sharma, R.; Bawa, G.; Khanna, S. Bonding and
structure trends of thiosemicarbazone derivatives of metals - An
overview. Coord. Chem. Rev. 2009, 253, 977−1055.
(26) Stelzig, L.; Kotte, S.; Krebs, B. Molybdenum complexes with
̈
tridentate NS2 ligands. Synthesis, crystal structures and spectroscopic
properties. J. Chem. Soc., Dalton Trans. 1998, 2921−2926.
(27) Eierhoff, D.; Tung, W. C.; Hammerschmidt, A.; Krebs, B.
Molybdenum complexes with tridentate NS2 ligands. Synthesis, crystal
structures and spectroscopic properties. Inorg. Chim. Acta 2009, 362,
915−928.
(28) Ducrot, A.; Scattergood, B.; Coulson, B.; Perutz, R. N.; Duhme-
Klair, A.-K. Electronic fine-tuning of oxygen atom transfer reactivity of
cis-dioxomolybdenum(VI) complexes with thiosemicarbazone ligands.
Eur. J. Inorg. Chem. 2015, 2015, 3562−3571.
(29) Pietsch, M. A.; Hall, M. B. Theoretical studies on models for the
oxo-transfer reaction of dioxomolybdenum enzymes. Inorg. Chem.
1996, 35, 1273−1278.
(30) Kail, B. W.; Perez, L. M.; Zaric, S.; Millar, A. J.; Young, C. G.;
Hall, M. B.; Basu, P. Mechanistic investigation of the oxygen-atom-
transfer reactivity of dioxo-molybdenum(VI) complexes. Chem. - Eur. J.
2006, 12, 7501−7509.
(31) Gehrke, H.; Veal, J. Acetylacetonate complexes of molybdenum-
(V) and molybdenum(VI). I. Inorg. Chim. Acta 1969, 3, 623−627.
(32) Bagherzadeh, M.; Zare, M. Synthesis, characterization, and
catalysis of recyclable new piperazine-bridged Mo(VI) polymers
[MoO2(Salen) (piperazine)]n in highly selective oxygenation of
alkenes and sulphides. J. Coord. Chem. 2013, 66, 2885−2900.
(33) Ellis, C. D.; Margerum, L. D.; Murray, R. W.; Meyer, T. J.
Oxidative electropolymerisation of polypyridyl complexes of ruthe-
nium. Inorg. Chem. 1983, 22, 1283−1291.
(34) Suzuki, K.; Kobayashi, A.; Kaneko, S.; Takehira, K.; Yoshihara,
T.; Ishida, H.; Shiina, Y.; Oishi, S.; Tobita, S. Reevaluation of absolute
luminescence quantum yields of standard solutions using a
spectrometer with an integrating sphere and a back-thinned CCD
detector. Phys. Chem. Chem. Phys. 2009, 11, 9850−9860.
(35) Purohit, S.; Koley, A. P.; Prasad, L. S.; Manoharan, P. T.; Ghosh,
S. Chemistry of molybdenum with hard-soft donor ligands. 2.
Molybdenum(VI), -(V), and − (IV) oxo complexes with tridentate
Schiff base ligands. Inorg. Chem. 1989, 28, 3735−3742.
(36) Staniewicz, R. J.; Sympson, R. F.; Hendricker, D. G. Preparation
and investigation of the spectral and electrochemical properties of
mixed-ligand ruthenium(II) complexes containing 1,8-naphthyridines.
Inorg. Chem. 1977, 16, 2166−2171.
(37) Stark, C. W.; Schreier, W. J.; Lucon, J.; Edwards, E.; Douglas, T.;
Kohler, B. Interligand electron transfer in heteroleptic ruthenium(II)
complexes occurs on multiple time scales. J. Phys. Chem. A 2015, 119,
4813−4824.
(38) Magde, D.; Magde, M. D., Jr.; Glazer, E. C. So-called “dual
emission” for 3MLCT luminescence in ruthenium complex ions: What
is really happening? Coord. Chem. Rev. 2016, 306, 447−467.
L
DOI: 10.1021/acs.inorgchem.6b01485
Inorg. Chem. XXXX, XXX, XXX−XXX