Effect of substituents on the water oxidation activity of [Ru II(terpy)(phen)Cl]+ procatalysts

Article abstract of DOI:10.1021/ic402118u

A series of [Ru^II(terpy-R)(phen-X)Cl]PF₆ complexes was designed where terpy-R is the tridentate 4′-(4-methylmercaptophenyl)- 2,2′:6′2″-terpyridine ligand MeMPTP and phen-X is a substituted phenanthroline with hydro (1), 5-nitro (2), 5,6-dimethyl (3), and 3,4,7,8-tetramethyl (4). This series allows us to compare the reactivity of phenanthroline-containing procatalysts with that of its well-established bipyridine counterparts as well as to study the effects of electron-withdrawing and -donating substituents on water oxidation. These species were thoroughly characterized by spectroscopic and spectrometric methods, and the structures of 1, 3, and 4 were determined by single-crystal X-ray diffraction. The procatalysts 1-4 show opposite trends compared to known terpyridine/bipyridine species; the unsubstituted procatalyst 1 yields a turnover number (TON) of 410 followed by 250 and 150 for complexes 3 and 4 with electron-donating substituents. Species 2, with electron-withdrawing properties, yields the lowest TON of 60. Although the TONs decrease upon substitution, the presence of electron-donating methyl substituents enhances the rate of O₂ evolution during an early stage of catalysis. Interestingly, no evidence of conversion from chlorido-containing procatalysts into expected aqua-containing catalysts was observed for 1-4 by NMR and UV-visible spectroscopy during the induction period. This observation, along with reactivity toward (NH ₄)₂[Ce^IV(NO₃)₆], suggests that water nucleophilic attack happens to a high-valent ruthenium species rather than while at the Ru^II oxidation state. Reactivity follows a trend similar to the rate of O₂ evolution in all complexes. Furthermore, the electrospray ionization mass spectrometry and ¹H NMR analyses of 1, as recovered after catalysis, indicate the presence of a chlorido ligand.

Full text of DOI:10.1021/ic402118u

Article
pubs.acs.org/IC
Effect of Substituents on the Water Oxidation Activity of
[RuII(terpy)(phen)Cl]+ Procatalysts
Dakshika C. Wanniarachchi, Mary Jane Heeg, and Claudio N. Verani*
́
Department of Chemistry, Wayne State University, Detroit, Michigan 48202, United States
S
* Supporting Information
ABSTRACT: A series of [RuII(terpy-R)(phen-X)Cl]PF6 com-
plexes was designed where terpy-R is the tridentate 4′-(4-
methylmercaptophenyl)-2,2′:6′2″-terpyridine ligand MeMPTP
and phen-X is a substituted phenanthroline with hydro (1), 5-
nitro (2), 5,6-dimethyl (3), and 3,4,7,8-tetramethyl (4). This
series allows us to compare the reactivity of phenanthroline-
containing procatalysts with that of its well-established
bipyridine counterparts as well as to study the effects of
electron-withdrawing and -donating substituents on water
oxidation. These species were thoroughly characterized by
spectroscopic and spectrometric methods, and the structures of
1, 3, and 4 were determined by single-crystal X-ray diffraction.
The procatalysts 1−4 show opposite trends compared to known
terpyridine/bipyridine species; the unsubstituted procatalyst 1
yields a turnover number (TON) of 410 followed by 250 and 150 for complexes 3 and 4 with electron-donating substituents.
Species 2, with electron-withdrawing properties, yields the lowest TON of 60. Although the TONs decrease upon substitution,
the presence of electron-donating methyl substituents enhances the rate of O2 evolution during an early stage of catalysis.
Interestingly, no evidence of conversion from chlorido-containing procatalysts into expected aqua-containing catalysts was
observed for 1−4 by NMR and UV−visible spectroscopy during the induction period. This observation, along with reactivity
toward (NH4)2[CeIV(NO3)6], suggests that water nucleophilic attack happens to a high-valent ruthenium species rather than
while at the RuII oxidation state. Reactivity follows a trend similar to the rate of O2 evolution in all complexes. Furthermore, the
1
electrospray ionization mass spectrometry and H NMR analyses of 1, as recovered after catalysis, indicate the presence of a
chlorido ligand.
(PCET) leads to the formation of a [RuIIIOH]2+ intermediate
INTRODUCTION
■
that is further oxidized to [RuIVO]2+ and [RuVO]3+. In the
presence of water, the electrophilic pentavalent ruthenium oxo
species can then be converted into hydroperoxo species
[RuIIIOOH]2+, which antecedes the formation of [RuIVOO]2+
via another PCET step. At this point, O2 can evolve via
replacement by a water molecule, and the catalyst is then
regenerated.5−9 Optimization of the catalyst activity requires a
balance between the availability of electron density at the metal
Ruthenium-based catalysts have been avidly pursued in order to
circumvent the steep thermodynamic requirements involved in
water oxidation.1 One of the most promising mononuclear
topologies for such catalysts takes advantage of coordination of
the bivalent ruthenium ion in an octahedral environment to
polypyridine ligands such as terpyridine (terpy) and bipyridine
(bipy) along with a monodentate aqua ligand occupying the
sixth position. Therefore, RuII(terpy)(bipy)X]+ procatalysts
have been synthesized in which halogeno ligands X such as Cl,
Br, and I occupy the sixth position.2−4 In the presence of a
sacrificial oxidant such as CeIV ions, these [RuII(terpy)(bipy)-
X]+ procatalysts require an induction period in which H2O is
believed to be incorporated, either by direct substitution of the
halogen or by formation of a seven-coordinated intermedi-
ate.2−4 Recent evidence suggests that direct substitution is
favored by chorido-containing procatalysts, while seven
coordination might be favored when heavier halogens such as
iodides are present.2−4 An alternative and less explored
possibility is that partial displacement of flexible polypyridine
ligands takes place. After water is coordinated, accepted
mechanisms assume that proton-coupled electron transfer
center and limited π-back-bonding to the most labile ligand.10
A
certain balance can be achieved by the careful insertion of
electron-donating and -withdrawing groups to each of the
polypyridyl ligands. Similar groups can foster or prevent catalyst
stability, depending on their position on either the bipy or terpy
moiety, as seen in recent reports.10,11 According to these
findings, the insertion of electron-donating groups to terpy and
electron-withdrawing groups to bipy should lead to a situation
where the catalyst activity and stability are optimized. In this
regard, we are interested in the electronic effects associated
Received: August 22, 2013
Published: March 19, 2014
© 2014 American Chemical Society
3311
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
with the more rigid phenanthroline (phen) in [RuII(terpy)-
(phen)Cl]+ procatalysts, both to evaluate their reactivity and
because proper phen functionalization allows for the synthesis
of multimetallic topologies.12−16 While working on the
functionalization of mercaptoterpyridines for self-assembly on
gold electrodes, we noticed that the complex [RuII(MeMPTP)-
(phen)Cl]PF6 [MeMPTP = 4′-(4-methylmercaptophenyl)-
2,2′:6′2″-terpyridine)] showed catalytic activity toward water
oxidation. We realized that this is a unique platform to (i)
compare the effects of substituted phen with electron-donating
and -withdrawing groups on the catalytic activities of the series
of procatalysts summarized in Scheme 1, (ii) discuss the
reactivity of these complexes toward CeIV, and (iii) evaluate the
composition of the catalyst recovered after catalysis. The results
follow.
Scheme 1. [RuII(MeMPTP)(R-phen)Cl]PF6 Complexes with
Substituted phen
Figure 1. Molecular structure of the cations (a) [RuII(MeMPTP)-
(Me2-phen)Cl]+ (3+) and (b) [RuII(MeMPTP)(Me4-phen)Cl]+ (4+).
Selected bond lengths (Å) and angles (deg): for 3+, Ru−N1 =
2.087(7), Ru−N2 = 1.955(7), Ru−N3 = 2.058(8), Ru−N4 =
2.051(7), Ru−N5 = 2.072(7), Ru−Cl = 2.450(2) and N1−Ru−N2
= 79.1(3), N2−Ru−N3 = 79.9(3), N3−Ru−N5 = 98.5(3), N5−Ru−
N1 = 102.4(3), N4−Ru−Cl = 173.8(2); for 4+, Ru−N1 = 2.061(2),
Ru−N2 = 1.943(2), Ru−N3 = 2.056(2), Ru−N4 = 2.039(2), Ru−N5
= 2.075(2), Ru−Cl = 2.407(7) and N1−Ru−N2 = 79.58(9), N2−Ru−
N3 = 79.74(9), N3−Ru−N5 = 101.96(9), N5−Ru−N1 = 98.62(9),
N4−Ru−Cl = 171.08(6).
dral geometries around the ruthenium center, where the Nterpy
−
Ru bonds vary from 2.087(7) to 1.943(7) Å and relatively
shorter bond lengths are observed for the central N atom of the
terpy. Bond lengths for Nphen−Ru vary from 2.039(2) to
2.075(2) Å, where shorter bond lengths are observed for the N
atom trans to the chlorido coligand. The difference in the bond
lengths for the phen Npyridine located trans to Cl− of 3 is 0.012 Å
longer than that for 4. These bond lengths are in good
agreement with values reported in the literature.18 The Ru−Cl
bond length in complex 4 is shorter than that of 3 by ca. 0.05 Å.
The molecular structures of 1, 3, and 4 clearly illustrate that the
chlorido group is trans to the N4 atom of the phen.
RESULTS AND DISCUSSION
■
Synthesis and Characterization of Ligands and
Complexes. The ligand MeMPTP was synthesized as
previously reported17 and treated with 1 equiv of
[RuII(DMSO)4Cl2] (DMSO = dimethyl sulfoxide) to yield
the precursor [RuII(MeMPTP)(DMSO)Cl2]. Four complexes
were obtained by reaction of the precursor with the appropriate
phen ligand and further column-chromatographed to yield the
high-purity species [RuII(MeMPTP)(phen)Cl]PF6 (1),
[RuII(MeMPTP)(NO2-phen)Cl]PF6 (2), [RuII(MeMPTP)-
(Me2-phen)Cl]PF6 (3), and [RuII(MeMPTP)(Me4-phen)Cl]-
Electronic and Redox Behavior. The efficiency of water
oxidation in complexes 1−4 is sensitive to the electron density
around the ruthenium center,2,10,11 and the effect of electron-
donating and -withdrawing substituents is reflected in the
potential of the RuII/RuIII couple, as measured in ACN and
shown in Figure S2 in the SI. Complexes 1−4 display the RuII/
1
PF6 (4). Species 1−4 were characterized by H NMR and IR
spectroscopies, high-resolution electrospray ionization mass
spectrometry (ESI-MS), and elemental analysis prior to use,
with excellent agreement between the different methods.
Species 1, 3, and 4 were also characterized by X-ray
crystallography.
RuIII couple between 0.31 and 0.46 VFc /Fc. Another irreversible
+
+
Structural Characterization. X-ray-quality crystals of 1, 3,
and 4 were obtained by the slow evaporation of a complex
dissolved in acetonitrile (ACN) under dark conditions, as
shown in Figure 1. The partial solution obtained for 1 does not
allow for discussion of the bond lengths and angles, but it
suffices to indicate connectivity, confirming the identity of the
[RuII(MeMPTP)(phen)Cl]+ cation (Figure S1 in the Support-
ing Information, SI). Complexes 3 and 4 show pseudooctahe-
process consistently observed between 1.0 and 1.1 VFc /Fc is
attributed to oxidation of the −SCH3 group of MeMPTP. The
terpy/terpyridinium couple is observed around −1.94 to −2.14
+
VFc /Fc for complexes 1, 3, and 4. This process is absent for
+
species 2, which instead shows a process at −1.08 VFc /Fc
centered on the NO2-phen ligand. In Figure 2, details are
shown for the RuII/RuIII couple, evidencing that the electron-
withdrawing nitro group in complex 2 does not favor the
3312
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
withdrawing nitro group displays the lowest TON, whereas 3
and 4 with electron-donating methyl groups indicate higher
activity than that of complex 2 but slightly lower than that of
the unsubstituted 1. In general terms, it has been observed2,10
that electron-withdrawing groups in related [RuII(terpy)(R-
bipy)X]+ complexes (X = Cl or H2O) lead to enhancement of
the TONs. Similarly, bipy-containing electron-donating groups
led to a clear decrease in the TONs. The species [RuII(terpy)-
(NO2-bipy)Cl]+ is an exception to this observation because its
TON is slightly lower than that of the unsubstituted complex.2
Nonetheless, the nitro complex still displays higher TONs than
the equivalent species with electron-donating substituents.
Considering terpy substituents on the 4′ position, an opposite
trend was reported and electron-withdrawing groups led to
lower TONs.10,11 This observation points to subtle but
important differences in the catalytic potential of bipy- and
phen-based catalysts. On the other hand, the presence of an
electron-donating Ph-SMe mercapto substituent in complex 1
has fostered a higher TON than its unsubstituted counterpart
[RuII(terpy)(phen)Cl]PF6, in good agreement with the
literature.11
Figure 2. Cyclic voltammograms of the RuII/RuIII couple for 1−4 in
ACN/TBAPF6 at 100 mV s−1.
The amounts and rates of O2 generated in acidic aqueous
media were studied with a Clark electrode under argon for
complexes 1−4 with an excess of (NH4)2[CeIV(NO3)6] in a
Ru/Ce ratio of 1:5000 equiv. For complex 1, a first-order
kinetic behavior was determined from measurements at
different concentrations (Figure 3a,b), confirming that O2
generation is dependent on the molecular procatalyst. All of
the complexes required an induction period of 200−300 s prior
to O2 detection. The shortest induction period was observed
for the tetramethylated 4, while the longest was reported for
the nitro species 2. This induction period has been historically
associated with conversion of the halido (Cl and Br)
procatalyst4 into the more active aqua-substituted catalyst.10,20
Further evidence by the Thummel group suggests that the iodo
procatalyst [RuII(terpy)(bipy)I]+ and its aqua counterpart
[RuII(terpy)(bipy)H2O]+ do not require this induction period.4
This observation is particularly relevant for the iodo-substituted
species because it presents the highest TON of 570. The rates
of O2 evolution for 1−4 were determined by measuring the
linear portion of the curves after the induction period and
before 800 s (see Table 1). The highest rate of O2 evolution
was observed for 4, whereas complex 2 indicated the lowest
rate, thus in good agreement with other published [RuII(terpy)-
(bipy)Cl]+ species.2
Another important consideration in the assessment of the
catalyst is the possibility of the formation of heterogeneous
catalysts such as RuO2 nanoparticles2 or Ce···RuIVO
clusters.21 In order to rule out the presence of heterogeneous
species in solution, dynamic light scattering (DLS) was used to
analyze the solutions of CeIV + triflic acid and CeIV + triflic acid
+ complex 1 in ACN at the same concentrations as those used
for determination of the rates of O2 generation (Figure S5 in
the SI). If the induction period should involve the formation of
heterogeneous species, a noticeable increase in the size
distribution of the samples should become apparent. Because
no heterogeneous forms were observed for any of those
solutions over a period of 20 min, thus longer than the ∼5 min
induction times, the notion that catalysis is based on molecular
species is again reinforced. It has been accepted that the
formation of a more active aqua catalyst in aqueous acidic
media via substitution of the chlorido ligand for water can be
inspected by UV−visible changes.10 Furthermore, if the
generation of a trivalent RuIII state, as observed by the highest
+
oxidation potential of 0.46 VFc /Fc. The unsubstituted 1 displays
+
this couple at 0.40 VFc /Fc. An opposite trend is observed in the
presence of weakly electron-donating methyl substituents in
complexes 3 and 4, yielding more affordable potentials of 0.38
+
and 0.33 VFc /Fc, respectively. The results in dichloromethane
are shown in Figure S3 in the SI and show consistency with the
above-mentioned behavior.
The UV−visible absorption spectra of complexes 1−4 were
recorded in ACN and are shown in Figure S4 in the SI. The
complexes show a pronounced band at 515−520 nm (Δ = 5
nm) due to ruthenium-to-terpy metal-to-ligand charge transfer
(MLCT).2,18,19 A second band is observed between 422 and
435 nm (Δ = 13 nm) and attributed to a MLCT process
involving the ruthenium center and the phen ligand. This band
has not been detected for 2 and is blue-shifted in 4 because of
the presence of four electron-donating methyl groups. The fact
that the second MLCT presents larger variance associated with
the nature of the substituents on the phen ligand confirms that
MLCT is associated with the RuII ← phen process.
Catalytic Activity toward Water Oxidation. The
influence of electron-donating and -withdrawing groups in the
catalytic properties of [RuII(MeMPTP)(R-phen)Cl]PF6 com-
plexes was evaluated by measurement of the turnover number
(TON) = [generated O2]/[catalyst] in moles) and of the rate
of O2 generation in solution. The TONs of these complexes
were measured in 10 mL septum-capped round-bottomed flasks
using (NH4)2[CeIV(NO3)6] (550 mg, 1.0 mmol) and triflic acid
(CF3SO3H, 3.0 mL, pH = 1) stirred together. An ACN solution
of the complex under study (100 μL, 8 × 10−5 mmol) was
added, and experiments were carried out under atmospheric
conditions. The resulting mixture reacted 24 h, and then a
sample of the headspace gas was collected and injected (100
μL) into a gas chromatograph. Atmospheric N2 was used as an
internal standard in the gas chromatography (GC) determi-
nation of O2 percentiles. The higher the TON, the more stable
the catalyst. The TON of complex [RuII(terpy)(phen)Cl]PF6
was used as an internal standard for consistency.2 Because good
agreement was found between our experiments and those
previously reported, the TONs of complexes 1−4 were
measured with confidence. Species 2 with an electron-
3313
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
monitored for 1 excluding CeIV while keeping the other
reaction conditions similar to those used for evaluation of the
O2 rates. Because precipitation was observed (Figure S6a in the
SI), the conditions were changed to a 1:1 mixture of ACN/
triflic acid (aqueous). Under these conditions and in the
absence of light, conversion to the aqua complex was observed,
however with a half-life of 290 min (Figure S6b in the SI). This
is a 60-fold longer time than the observed 300 s, suggesting that
CeIV is necessary to initiate the catalytic cycle. Further evidence
of nontrivial conversion of the procatalyst 1 comes from time-
dependent 1H NMR data in ACN-d3/D2O following the
disappearance of the peak at 13.00 ppm. This peak is associated
with the phen H9 proton closest to the chlorido ligand (see
Scheme 1 for the proton numbering) and is expected to shift to
a more downfield value of 12.64 ppm when a water molecule
replaces the chlorido ligand. We noticed that the original
chlorido species remained unchanged in the first 24 h and
disappeared completely after 120 h. The aqua species can be
observed as a minor signal after 6 h, subsequently increasing in
intensity and then existing alone after approximately 120 h
(Figures 4 and S7 in the SI). A slower conversion was observed
when the ACN-d3/D2O ratio was changed to 4:1 (Figure S8 in
the SI). We also followed up the chloride-to-aqua conversion
for complexes 1−4 in 3:1 DMSO-d6/D2O with triflic acid (0.1
M). In this solvent mixture, the phen H9 proton appears at
10.19 ppm, and a new peak was observed at 8.96 ppm
associated with the formation of a triflate-substituted entity that
is further converted into the aqua species, as observed by a peak
at 9.00 ppm (Figures S9−S14 in the SI). Even in this case, the
“chlorido” peak at 10.19 ppm is observed and does not
disappear. Some back-conversion was observed within 3 h
when LiCl was added to the aged solutions. Interestingly, the
spectra of the nitro-substituted 2 do not vary over time.
Reactivity Studies. Oxidative Titration of 1 with CeIV.
UV−visible spectroscopy has been used to monitor the
reactivity of aqua-substituted catalysts toward CeIV,8 but no
spectroscopic studies have been carried out with chlorido-
containing procatalysts. Therefore, we have generated spectro-
scopic profiles for 1−4 associated with RuIII and RuIV species by
titrating against CeIV before study of the relative reactivities of
these complexes. The addition of 1 equiv of CeIV to complex 1
resulted in ca. 60% decrease in the 518 nm RuII → terpy MLCT
band (Figure S15 in the SI) at the same time that a new band
assigned to a RuIII species appeared at 413 nm. Because the 518
nm band did not disappear completely, this is indicative of
either incomplete or competitive oxidation, most likely of the
−SCH3 substituent on the terpy ligand, as observed for
aromatic thiols.22 As shown in Figure 5a, the addition of 2 equiv
of CeIV resulted in the complete disappearance of the 518 nm
band and the appearance of yet another band at 360 nm, in
addition to that at 413 nm. Available literature on [RuII(terpy)-
(bipy)H2O]2+ complexes assign this band to either the CeIV ion
or the formation of [RuIVO]2+ species.3,5a,b,21,8 The UV−
visible spectrum of (NH4)2[Ce(NO3)6] was taken in aqueous
acidic media (triflic or nitric acid, pH = 1) and revealed ill-
defined processes at 300 nm (abs ≤ 0.3), whereas an expected
∼350 nm CeIV-related band was not observed (Figure S16 in
the SI). This reinforces the idea that the 360 nm process in 1 is
associated with a high-valent ruthenium species. Titration was
continued for up to 5 equiv of CeIV to monitor changes during
one catalytic cycle.
Figure 3. (a) O2 generation with different concentrations of 1 over
time. (b) First-order plot of the initial rate data for 1. (c) Generation
of O2 as a function of time with a YSI Clark electrode for complexes
1−4.
induction period observed for 1−4 is exclusively associated
with ligand substitution in the RuII core, then the presence of
CeIV is irrelevant. Therefore, UV−visible spectral changes were
Reductive Titration of 1 with Ascorbic Acid (AA). In order
to verify that the observed spectroscopic profiles are associated
3314
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
Table 1. Electronic and Water Oxidation Properties of Complexes 1−4
d
rate × 10−4 (μmol of
a
b
c
complex
λmax, nm (ε, M−1cm−1
)
E1/2 (ΔE; |Ipa/Ipc|), V)
TON
410
O2 s−1
8.19
)
1
515 (16224), 435 (sh, 7716), 318
(39614)
−1.97 (0.054; |1.4|), −1.73 (irrv), 0.40 (0.058; |1.00|), 1.05 (irrv) Fc+/Fc =
0.44 (0.074; |1.00|)
2
3
4
518 (18017), 316 (40689)
−1.93 (irrv), −0.11 (irrv), 0.46 (0.072; |0.9|), 1.11 (irrv) Fc+/Fc = 0.44 (0.075;
|1.0|)
−2.01 (0.060, |0.4|), −1.74 (irrv), 0.38 (0.063; |0.9|), 1108 (irrv) Fc+/Fc =
0.44 (0.081; |1.0|)
60
250
150
2.40
6.92
27.3
515 (15673), 429 (sh, 7492), 316
(40579)
520 (16997), 422 (sh, 11018), 319 −2.14 (irrv), −1.77 (irrv), 0.33 (0.068; |0.9|), 1.07 (irrv) Fc+/Fc = 0.44 (0.076;
(45154)
|1.0|)
a
b
UV−visible spectra recorded in 5 × 10−5 M ACN. Cyclic voltammograms of 1−4 recorded as 4.0 × 10−4 mol L−1 in ACN with 0.1 M TBAPF6 as
the supporting electrolyte using a 100 mV s−1 scan rate at room temperature in an inert atmosphere; WE = glassy carbon, RE = Ag/AgCl, and CE =
c
platinum wire. All potentials listed versus Fc+/Fc. Measured in ACN. (NH4)2[Ce(NO3)6] (550 mg, 1 mmol) and CF3SO3H (3 mL, pH = 1) were
d
stirred together in the complex (100 μL, 8 × 10−5 mmol) for 24 h. CF3SO3H (pH = 1, 5 mL), (NH4)2[Ce(NO3)6] (550 mg), and the complex (4
× 10−5 M).
Figure 5. (a) UV−visible spectral changes for 1 (5 × 10−5 M in ACN,
2.5 mL) upon oxidation with CeIV (5 × 10−3 M in aqueous triflic acid,
25−125 μL, pH = 1). (b) UV−visible spectral changes for oxidized 1
upon reductive titration with AA in water (5 × 10−3 M, 25−125 μL).
1
Figure 4. H NMR analysis of 1 in a 1:1 ACN-d3/D2O mixture.
bivalent RuII is regenerated, the next question is concerned with
with the formation of high-valent ruthenium species, likely
RuIV, the same solution treated with 5 equiv of CeIV was titrated
with the reducing agent AA. As illustrated in Figure 5b, the
characteristic RuII → terpy MLCT band reappears at ca. 515
nm, suggesting the regeneration of RuII species. Assuming the
whether the chlorido ligand is replaced by water during the
catalytic cycle. A comparison between the UV−visible data for
previously published [RuII(terpy)(phen)Cl]+ and [RuII(terpy)-
(phen)H2O]2+ suggests a hypsochromic shift of 22 nm for the
aqua complex.4 Because the regenerated MLCT band at 515
3315
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
nm does not indicate significant shifts, it is possible that the
chlorido group is still attached to the catalytic core. Similarly,
the bands at 360 and 410 nm associated with a high-valent
ruthenium species disappear, and the resulting spectrum
resembles that of the starting complex 1. In summary, 1
seems to reach higher oxidation states, while bound to the
chlorido ligand. This might be possible either via the formation
nm band in 1−4 was calculated along with the first-order decay
plots in Figure 6b. Complex 4 shows the highest rate of decay
(11.4 × 10−3 s−1), followed by species 1 and 3 (9.9 × 10−3 and
8.6 × 10−3 s−1, respectively). The rate observed for 2 reaches
3.4 × 10−3 s−1 and is much slower than that of 4. This trend is
in good agreement with the observed rates of O2 evolution
discussed previously, and the half-lives for the decay of 1−4 are
comparable to the induction period observed for the rate of O2
evolution (Table T1 in the SI).
2
of a seven-coordinate [RuIVO]+ or via the bond-breaking
and flipping of one of the pyridine units in the terpy to form a
six-coordinate [RuIVO]+ species. Although further mecha-
nistic evaluation will be required, complexes 1−4 would most
likely follow the latter case. The former mechanism has been
verified in [RuII(pda)L2] (pda = 1,10 phenanthrolinedicarbox-
ylic acid; L = pyridine), where a large O−Ru−O angle grants
access of water to the ruthenium core.23
Time-dependent Decomposition of High-valent Ruthe-
nium Species. Changes in the electronic absorption spectra of
complex 1 upon reaction with 5 equiv of CeIV are shown in
Figure 6a. The MLCT band at 518 nm disappeared
Evaluation of the Recovered Procatalyst. The results
outlined in the previous sections suggest that the catalytic core
in these species is regenerated after catalysis. We have
recovered and isolated the catalyst in order to address this
question in further detail. Complex 1 was treated with a 60-fold
excess of CeIV in aqueous triflic acid, yielding an air-unstable
green precipitate after 24 h. This precipitate was isolated by
filtration, dissolved in DMSO, and analyzed by 1H NMR
spectroscopy and ESI-MS in the positive mode. Broad peaks
1
observed in the H NMR spectrum of the precipitate are
suggestive of a paramagnetic high oxidation state different from
that of the low-spin 4d6 RuII present in 1. In order to verify
whether the catalytic core is preserved during catalysis, AA was
added to this species, prompting an immediate color change
from green to red. The 1H NMR spectrum of the resulting red
species was recorded in DMSO and compared with 1 before
catalysis. Proton labeling for 1 is shown in Scheme 1 and Figure
7, and the phen H9 proton at 10.32 ppm remained unchanged
in the recovered catalyst. This fact is interpreted as having the
chlorido group still coordinated to the ruthenium center.
However, peak shifts at 9.20, 8.33, and 7.58 ppm suggest
structural alteration associated with the terpy of the recovered
catalyst. Because the proton counts of the aromatic region in
both spectra are equivalent, the structural change must be
associated with oxidation of the −SMe group into a −SO2Me
group. This oxidation has been confirmed by ESI-MS analysis
(Figure S17 in the SI), where a molecular ion peak at m/z
703.91 is observed. This peak is 32 mass units higher than that
of the parent peak [RuII(MeMPTP)(phen)Cl]+ for 1. As added
evidence for this conversion, the equivalent species
[RuII(terpy)(phen)Cl]+ failed to show such mass modification
and retained its original mass (Figure S18 in the SI).
Furthermore, the ligand MeMPTP was treated with CeIV,
yielding the oxidized product, as observed by IR spectroscopy
with an SO band at 1298 cm−1 and ESI-MS where peaks at
m/z 372.1337 and 388.1302 are observed for the −SOCH3 and
−SO2CH3 species (Figures S19 and S20 in the SI). The mother
liquor recovered after precipitation of the green species was also
isolated and analyzed by ESI-MS in the positive and negative
modes. No measurable quantities of any ruthenium species was
detected either as discrete catalysts or in Ru/Ce clusters.21
CONCLUSIONS
■
The water oxidation ability of ruthenium/MeMPTP complexes
with several substituted phen ligands was analyzed by
evaluation of their TONs and rates of O2 generation. Although
we treated these species as “procatalysts” to follow the
established nomenclature that considers the aqua-substituted
counterparts as “catalysts”, the observed TONs for both classes
of compounds are comparable. We can summarize our
observations as follows: (i) The catalytic O2 generation
requires the presence of CeIV. (ii) The catalytic activity was
enhanced by the presence of electron-donating groups on the
phen, while the presence of electron-withdrawing substituents
Figure 6. (a) Spectral changes for 1 over time with a RuII/CeIV ratio of
1:5, [RuII] = 1 × 10−4 M, 1.5 mL in ACN, [CeIV] = 5 × 10−4 M, 1.5
mL in triflic acid (pH = 1). (b) First-order rate profiles for 1−4.
immediately after the addition of 5 equiv of an excess of the
oxidant. A similar observation has been reported,21 where a
comparable band on [RuII(terpy)(bipy)H2O]2+ disappeared
within 1.2 s upon the addition of a 10-fold excess of CeIV. Over
time, the new bands at 360 and 410 nm for 1 decreased, while a
325 nm band increased. The relative rate of decay of the 360
3316
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
1
Figure 7. H NMR spectra in DMSO for 1 (a) before and (b) after catalysis.
decreases it drastically. This effect is explained by the fact that
the −NO2 group in 2 decreases the electronic density on the
phen ligand, making it difficult to achieve higher oxidation
states relevant for catalysis, whereas the weakly donating −CH3
groups in 3 and 4 activate the phen ligand, facilitating the
generation of those high-valent ruthenium states. As such,
TONs follow the sequence 1 > 3 > 4 > 2, while the rates of O2
evolution follow 4 > 1 > 3 > 2. (iii) All species require an
induction period for catalysis, and the rates of O2 generation
follow a first-order mechanism in the presence of excess CeIV.
(iv) Spectroscopic evidence suggests that, during the induction
period, high oxidation species (possibly RuIV at 360 nm) are
generated, but no evidence of chloride-by-water exchange was
observed. Heterogeneous forms of the catalyst were ruled out
by DLS. (v) A decrease of the 360 nm band, related to RuVO
generation, associates the highest rates of decay in the order 4 >
1 > 3 > 2; this may be an indication that the position, as well as
the nature of the substituent, plays a role in catalysis. (vi)
Rather than an obvious chloride-by-water exchange, the
catalytic core seems to remain unchanged even in the presence
of excess CeIV, thus favoring either a seven-coordinate
intermediate or a six-coordinate route where a pyridine group
on MeMPTP gets uncoordinated. (iv) The presence of the
electron-donating −PhSMe group on MeMPTP enhances the
catalytic activity, although it undergoes oxidation to −SO2Me
in the presence of CeIV. These results highlight significant
differences between phen- and bipy-based procatalysts, point to
unique electronic nuances of the phen unit, and will be pivotal
for the design and optimization of future water oxidation
catalysts. Future studies will focus on the mechanistic details of
the catalytic cycle for 1−4, the evaluation of the catalytic
activity in multimetallic topologies based on phen functional-
ization, and the use of MeMPTP for self-assembly on gold
electrodes.
EXPERIMENTAL SECTION
■
Materials and Methods. Reagents and solvents were used as
received from commercial sources. Methanol and ethanol were
distilled over CaH2. 1,10-Phenanthroline, 1-(pyridin-2-yl)ethanone,
and 4-(methylthio)benzaldehyde were purchased from Alfa Aesar,
RuCl3 was purchased from Strem Chemicals, and 5,6-dimethylphenan-
throline was purchased from GFS Chemicals Inc. 5-Nitrophenanthro-
line was synthesized according to known procedures.24 IR spectra were
measured from 4000 to 400 cm−1 as KBr pellets on a Tensor 27 FTIR
spectrophotometer. 1H NMR spectra were measured using Varian 400
mHz spectrometers. ESI-MS spectra (positive) were measured in
either a triple-quadrupole Micromass QuattroLC or a single-
quadrupole Waters ZQ2000 mass spectrometer with an electro-
spray/APCI or ESCi source. Experimental assignments were simulated
on the basis of the peak position and isotopic distributions. Elemental
analyses were performed by Midwest Microlab, Indianapolis, IN. UV−
visible spectroscopy from 5.0 × 10−5 M ACN solutions was performed
using a Cary 50 spectrophotometer in the range of 250−1100 nm.
Cyclic voltammetry experiments were performed using a BAS 50W
voltammetric analyzer. A standard three-electrode cell was employed
with a glassy-carbon working electrode, a platinum wire auxiliary
3317
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
electrode, and an Ag/AgCl reference electrode under an inert
atmosphere at room temperature. The salt TBAPF6 (TBA =
tetrabutylammonium) was used as the supporting electrolyte.
Potentials are presented using ferrocene25 as the internal standard.
DLS experiments were performed in a Zetasizer Nano Series
spectrophotometer using [CeIV] = 0.2 M in triflic acid, 2.5 mL, pH
= 1, and complex 1 at 5 × 10−5 M (1.98 mg/0.5 mL, injecting 25 μL).
Water Oxidation. Studies were carried out in a 10 mL round-
bottomed flask capped with a rubber septum under ambient
conditions. (NH4)2[Ce(NO3)6] (550 mg, 1 mmol) and CF3SO3H
(3 mL, pH = 1) were stirred together, the complex (100 μL, 8 × 10−5
mmol) dissolved in ACN was injected through the septum, and the
mixture was allowed to react for 24 h. The amount of O2 generated
was measured with a gas chromatograph by injecting 100 μL of a
headspace gas sample. The gas chromatograph is a Gow-Mac 400 with
Table 2. X-ray Data
[3]·1/3CH3CN
[4]·1/2CH3CN
formula
fw
C42H38ClF6N8PRuS
968.35
C42H39ClF6N7PRuS
955.35
space group
a (Å)
triclinic, P1
monoclinic, P21/c
15.7842(8)
13.1866(7)
19.5857(10)
90
̅
8.8581(7)
15.4618(11)
16.0397(12)
103.240(4)
105.443(4)
91.864(4)
2051.0(3)
2
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
93.825(2)
90
4067.5(4)
4
1
a thermal conductivity detector, and a 8 ft × /8 in., 5 Å molecular
T (K)
100(2)
100(2)
sieve column operating at 60 °C was used with helium as the carrier
gas. The calibration was carried out with air as the standard (21% O2).
The TON was calculated as the ratio of moles of O2 produced over
moles of catalyst used. Clarity software was used for data acquisition
from GC. The rate of O2 evolution was carried out in a three-necked
flask charged with triflic acid (pH = 1, 5 mL) and (NH4)2[Ce(NO3)6]
(550 mg) under an argon atmosphere, following well-established
protocols available in the literature.4 Prior to injection of the complex,
the YSI Clark-type electrode was calibrated with O2, argon, and air-
saturated solutions. After calibration, water saturated with air indicated
the O2 percentage in solution as 20 1%. Then the complex (50 μL, 4
× 10−5 M, AcN) was injected, and O2 percentage reading was recorded
every 10 s up to 40 min.
λ (Å)
0.71073
1.568
0.71073
1.560
ρ (mg m−3
)
μ (mm−1
)
0.608
0.611
R(F) (%)
12.27
0.0419
Rw(F) (%)
18.55
0.0504
a
R(F) = ∑|Fo| − |Fc||/∑|Fo||; Rw(F) = [∑w(Fo2 − Fc2)2/∑w(Fo )2]1/2
for I > 2σ(I).
2
CH3OH (3 × 10 mL). The purified 1 was obtained after column
chromatography over neutral alumina with CH2Cl2/CH3CN (1:1).
Yield: 183 mg (56%). Elem anal. Calcd for C34H25ClN5RuSPF6: C,
X-ray Structural Determinations. Diffraction data were meas-
ured on a Bruker X8 APEX-II Kappa geometry diffractometer with Mo
radiation and a graphite monochromator. Frames were collected at
100 K with the detector at 40 mm and 0.3° between each frame and
were recorded for 10 s. APEX-IIa and SHELXb software were used in
the collection and refinement of the models. Crystals of 3 appeared as
dark plates. A total of 23546 reflections were measured, yielding 9818
unique data (Rint = 0.104). Hydrogen atoms were placed in calculated
1
49.97; H, 3.18; N, 8.57. Found: C, 49.91; H, 3.23; N, 8.47. H NMR
(DMSO-d6): δ 10.32 (br s, 1H), 9.2 (m, 2H), 8.95 (m, 4H), 8.42 (d, J
= 5.7 Hz, 2H), 8.38 (d, J = 7.3 Hz, 1H), 8.31 (d, J = 8.1 Hz, 2H), 8.21
(m, 1H), 7.95 (m, 2H), 7.82 (d, J = 5.7 Hz, 1H), 7.56 (d, J = 8.1 Hz,
2H), 7.48 (d, J = 4.1 Hz, 2H), 7.42 (br s, 1H), 7.23 (d, J = 5.7 Hz,
2H), 2.63 (s, 3H). MS: m/z 672.06 ([C34H25ClN5RuS]+). FTIR (KBr,
cm−1): 1596.58, 1476.03, 1426.85, 1406.31 (pyridine rings), 841.43
−
positions. These cationic complexes 1 crystallized with 1 equiv of PF6
−
and 3 equiv of ACN. The anion showed typical PF6− disorder in the F
positions, as evidenced by the high thermal parameters for these
atoms. Crystals of 4 were dark fragments. A total of 72908 reflections
(PF6 ).
Complexes 2−4 were synthesized in a manner similar to that
described for 1. Analyses follow.
[RuII(MeMPTP)(NO2-phen)Cl]PF6 (2). Yield: 109 mg (42%). Elem
anal. Calcd for C34H24ClN6O2RuSPF6: C, 47.37; H, 2.81; N, 9.75.
Found: C, 47.38; H, 2.82; N, 9.51. 1H NMR (DMSO-d6): δ 10.46 (m,
1H), 9.28 (m, 2H), 9.2 (m, 3H), 8.94 (d, J = 7.3 Hz, 2H), 8.57 (m,
2H), 8.32 (d, J = 8.1 Hz, 2H), 8.07 (d, J = 4.1 Hz, 1H), 8.00 (m, 2H),
7.54 (m, 4H), 7.23 (m, 2H), 2.62 (s, 3H). MS: m/z 717.0417
([C34H24ClN6O2RuS]+). FTIR (KBr, cm−1): 1595.12, 1532.80,
1425.13, 1406.12, 1406.31 (pyridine rings), 1329.31 (−NO2),
were counted, which averaged to 10161 independent data (Rint
=
0.054). H atoms were placed at calculated positions. The cationic
−
complex crystallized with one PF6 anion and 2 equiv of ACN.
Disorder in the PF6− anion was handled by assigning partial occupancy
positions for four of the F atoms. These partial contributions were held
isotropic.26 Table 2 summarized the data for both structures.
Syntheses. Ligand 4′-(4-Methylmercaptophenyl)-2,2′:6′2″-ter-
pyridine (MeMPTP). The ligand MeMPTP was prepared according
to literature procedures17 by treating 2 equiv of 2-acetylpyridine with 1
equiv of 4-(methylthio)benzaldehyde. Figure S21 in the SI summarizes
the NMR data.
−
843.28 (PF6 ).
[RuII(MeMPTP)(Me2-phen)Cl]PF6 (3). Yield: 55 mg (32%). Elem
anal. Calcd for C36H29ClN5RuSPF6: C, 51.16; H, 3.46; N, 8.29. Found:
C, 50.96; H, 3.64; N, 8.27. 1H NMR (DMSO-d6): δ 10.31 (d, J = 4.86
Hz, 1H), 9.20 (s, 2H), 9.08 (d, J = 8.11 Hz, 1H), 8.93 (d, J = 8.11 Hz,
2H), 8.45 (m, 2H), 8.31 (d, J = 8.92 Hz, 2H), 7.96 (t, J = 7.70 Hz,
2H), 7.74 (d, J = 5.67 Hz, 1H), 7.56 (d, J = 8.11 Hz, 2H), 7.43 (d, J =
5.67 Hz, 2H), 7.40 (m, 1H), 7.22 (m, 2H), 2.9 (s, 3H), 2.71 (s, 3H),
2.62 (s, 3H). MS: m/z 700.0878 ([C36H29ClN5RuS]+). FTIR (KBr,
cm−1): 2921.46 (C−H stretch in the −CH3 group), 1595.13, 1501.43,
[RuII(MeMPTP)(DMSO)Cl2]. A mixture of [RuII(DMSO)4Cl2]26
(0.696 g, 1.44 mmol) and MeMPTP (0.512 g, 1.44 mmol) was
heated and refluxed in argon-degassed CH3OH (50 mL) for 7 h. The
solution turned brownish red, and a dark-brown precipitate was
formed. The precipitate was isolated by frit filtration and washed with
1
cold CH3OH (3 × 10 mL). Yield: 510 g (58%). H NMR (DMSO-
d6): δ 9.01 (d, J = 4.9 Hz, 2H), 8.84 (s, 2H), 8.77 (d, J = 8.1 Hz, 2H),
8.14 (m, 4H), 7.78 (m, 2H), 7.46 (d, J = 8.1 Hz, 2H), 3.59 (s, 3H).
MS: m/z 601.95 ([C24H23Cl2N3ORuS2]+). FTIR (KBr, cm−1):
1054.26 (SO stretch of DMSO), 2956.22 (C−H stretch of the
tert-butyl substituent). UV−visible [DMSO; λmax, nm (ε, M−1 cm−1)]:
291 (29681), 338 (36703), 400 (9323), 530 (11255), 680 (5471).
[RuII(MeMPTP)(phen)Cl]PF6 (1). A mixture of [RuII(MeMPTP)-
(DMSO)Cl2] (0.241 g, 0.4 mmol), 1,10-phenanthroline (0.072g, 0.4
mmol), and triethylamine (0.5 mL) in CH3OH (50 mL) was refluxed
in the dark overnight under argon. The volume of the resulting
solution was reduced to one-third, and NH4PF6 (0.50 g) was added.
The dark-red precipitate was isolated over a frit and washed with
−
1476.47 (pyridine rings), 844.16 (PF6 ).
[RuII(MeMPTP)(Me4-phen)Cl]PF6 (4). Yield: 95 mg (34%). Elem
anal. Calcd for C38H33ClN5RuSPF6: C, 52.27; H, 3.81; N, 8.02. Found:
C, 52.33; H, 4.03; N, 7.96. H NMR (DMSO-d6): δ 10.05 (m, 1H),
1
9.17 (m, 2H), 8.91 (d, J = 8.1 Hz, 2H), 8.54 (m, 1H), 8.32 (m, 3H),
7.94 (m, 2H), 7.56 (m, 2H), 7.47 (m, 3H), 7.22 (m, 2H), 3.00 (s, 3H),
2.79 (s, 3H), 2.62 (s, 3H), 2.56 (s, 3H), 2.08 (s, 3H). MS: m/z
728.1174 ([C38H33ClN5RuS]+). FTIR (KBr, cm−1): 2924.46 (C−H
stretch in the −CH3 group), 1592.64, 1466.68, 1425.27 (pyridine
−
rings), 842.68 (PF6 ).
3318
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319

Inorganic Chemistry
Article
(12) Lesh, F. D.; Allard, M. M.; Shanmugam, R.; Hryhorczuk, L. M.;
Endicott, J. F.; Schlegel, H. B.; Verani, C. N. Inorg. Chem. 2011, 50,
969−977.
ASSOCIATED CONTENT
■
S
* Supporting Information
X-ray crystallographic data in CIF format, table of rate data,
structure of complex 1, cyclic voltammograms, UV−visible and
FTIR spectra, DLS, precipitation of complex 1, H NMR and
MS analysis, and ESI (positive) isotope distribution. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(13) Lesh, F. D.; Shanmugam, R.; Allard, M. M.; Lanznaster, M.;
Heeg, M. J.; Rodgers, M. T.; Shearer, J. M.; Verani, C. N. Inorg. Chem.
2010, 49, 7226−7228.
1
(14) Lanznaster, M.; Heeg, M. J.; Yee, G. T.; McGarvey, B. R.;
Verani, C. N. Inorg. Chem. 2006, 46, 72−78.
(15) Mulfort, K. L.; Mukherjee, A.; Kokhan, O.; Du, P.; Tiede, D. M.
Chem. Soc. Rev. 2013, 42, 2215−2227.
(16) Knoll, J. D.; Arachchige, S. M.; Brewer, K. J. ChemSusChem
2011, 4, 252−261.
AUTHOR INFORMATION
■
(17) (a) Smith, C. B.; Raston, C. L.; Sobolev, A. N. Green Chem.
2005, 7, 650−654. (b) Tuccitto, N.; Torrisi, V.; Cavazzini, M.;
Morotti, T.; Puntoriero, F.; Quici, S.; Campagna, S.; Licciardello, A.
ChemPhysChem 2007, 8, 227−230. (c) Wang, J.; Hanan, G. S. Synlett
2005, 1251−1254.
Corresponding Author
*E-mail: cnverani@chem.wayne.edu. Phone: 313 577 1076.
Notes
The authors declare no competing financial interest.
(18) Bonnet, S.; Collin, J.-P.; Gruber, N.; Sauvage, J.-P.; Schofield, E.
R. Dalton Trans. 2003, 4654−4662.
(19) Tsai, C.-N.; Allard, M. M.; Lord, R. L.; Luo, D.-W.; Chen, Y.-J.;
Schlegel, H. B.; Endicott, J. F. Inorg. Chem. 2011, 50, 11965−11977.
(20) Masaoka, S.; Sakai, K. Chem. Lett. 2009, 38, 182−183.
(21) Kimoto, A.; Yamauchi, K.; Yoshida, M.; Masaoka, S.; Sakai, K.
Chem. Commun. 2012, 48, 239−241.
ACKNOWLEDGMENTS
■
This research was made possible by the Division of Chemical
Sciences, Geosciences, and Biosciences, Office of Basic Energy
Sciences, of the U.S. Department of Energy through the Single-
Investigator and Small-Group Research (Solar Energy Program
Grants DE-SC0001907 and DE-FG02-09ER16120 to C.N.V.),
including financial support for D.C.W. We thank Prof. John
Endicott at WSU for insightful discussions and suggestions,
Gayani Dedduwa Mudalige and Prof. Christine Chow for
measurement of the DLS profiles for complex 1, and Chenchen
He and Prof. M. T. Rodgers for the exact ESI-MS of the
(22) (a) Farras
Eur. J. 2013, 19, 7162−7172. (b) Guillo, P.; Hamelin, O.; Batat, P.;
Jonusauskas, G.; McClenaghan, N. D.; Menage, S. Inorg. Chem. 2012,
51, 2222−2230.
̀
, P.; Maji, S.; Benet-Buchholz, J.; Llobet, A. Chem.
́
(23) Tong, L.; Duan, L.; Xu, Y.; Privalov, T.; Sun, L. Angew. Chem.,
Int. Ed. 2011, 50, 445−449.
(24) (a) Bolger, J.; Gourdon, A.; Ishow, E.; Launay, J.-P. Inorg. Chem.
1996, 35, 2937−2944. (b) Liu, T.; Hu, J.; Yin, J.; Zhang, Y.; Li, C.; Liu,
S. Chem. Mater. 2009, 21, 3439−3446.
oxidized ligand MeMPTPSO CH
.
2
3
(25) Gagne, R. K. C.; Licenski, G. Inorg. Chem. 1980, 19, 2854.
(26) Dulier
2003, 22, 804−811.
̀
e, E.; Devillers, M.; Marchand-Brynaert, J. Organometallics
REFERENCES
■
(1) (a) Brimblecombe, R.; Dismukes, G. C.; Swiegers, G. F.; Spiccia,
L. Dalton Trans. 2009, 43, 9374−9384. (b) Wasylenko, D. J.; Palmer,
R. D.; Berlinguette, C. P. Chem. Commun. 2013, 49, 218−227.
(c) Concepcion, J. J.; Jurss, J. W.; Brennaman, M. K.; Hoertz, P. G.;
Patrocinio, A. O. T.; Murakami Iha, N. Y.; Templeton, J. L.; Meyer, T.
J. Acc. Chem. Res. 2009, 42, 1954−1965. (d) Lewis, N. S.; Nocera, D.
G. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15729−15735.
(2) Tseng, H.-W.; Zong, R.; Muckerman, J. T.; Thummel, R. Inorg.
Chem. 2008, 47, 11763−11773.
(3) Concepcion, J. J.; Jurss, J. W.; Templeton, J. L.; Meyer, T. J. J.
Am. Chem. Soc. 2008, 130, 16462−16463.
(4) Kaveevivitchai, N.; Zong, R.; Tseng, H.-W.; Chitta, R.; Thummel,
R. P. Inorg. Chem. 2012, 51, 2930−2939.
(5) (a) Lin, X.; Hu, X.; Concepcion, J. J.; Chen, Z.; Liu, S.; Meyer, T.
J.; Yang, W. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 15669−15672.
(b) Concepcion, J. J.; Tsai, M.-K.; Muckerman, J. T.; Meyer, T. J. J.
Am. Chem. Soc. 2010, 132, 1545−1557. (c) Chen, Z.; Concepcion, J. J.;
Luo, H.; Hull, J. F.; Paul, A.; Meyer, T. J. J. Am. Chem. Soc. 2010, 132,
17670−17673.
(6) Gilbert, J. A.; Eggleston, D. S.; Murphy, W. R.; Geselowitz, D. A.;
Gersten, S. W.; Hodgson, D. J.; Meyer, T. J. J. Am. Chem. Soc. 1985,
107, 3855−3864.
(7) Jurss, J. W.; Concepcion, J. C.; Norris, M. R.; Templeton, J. L.;
Meyer, T. J. Inorg. Chem. 2010, 49, 3980−3982.
(8) Wasylenko, D. J.; Ganesamoorthy, C.; Henderson, M. A.;
Koivisto, B. D.; Osthoff, H. D.; Berlinguette, C. P. J. Am. Chem. Soc.
2010, 132, 16094−16106.
̀
(9) Roeser, S.; Farras, P.; Bozoglian, F.; Martínez-Belmonte, M.;
Benet-Buchholz, J.; Llobet, A. ChemSusChem 2011, 4, 197−207.
(10) Wasylenko, D. J.; Ganesamoorthy, C.; Koivisto, B. D.;
Henderson, M. A.; Berlinguette, C. P. Inorg. Chem. 2010, 49, 2202−
2209.
(11) Yagi, M.; Tajima, S.; Komi, M.; Yamazaki, H. Dalton Trans.
2011, 40, 3802−3804.
3319
dx.doi.org/10.1021/ic402118u | Inorg. Chem. 2014, 53, 3311−3319