Subunit variation to uncover properties of polyazine-bridged Ru(II), Pt(II) supramolecules with low lying charge separated states providing insight into the functioning as H2O reduction photocatalysts to produce H 2

Article abstract of DOI:10.1021/ic4004406

Two new structurally diverse polyazine-bridged Ru(II),Pt(II) tetrametallic complexes, [{(Ph₂phen)₂Ru(dpp)}₂Ru(dpp) PtCl₂](PF₆)₆ (1a) and [{(Ph₂phen) ₂Ru(dpp)}₂Ru(dpq)PtCl₂](PF₆) ₆ (2a) (Ph₂phen = 4,7-diphenyl-1,10-phenanthroline, dpp = 2,3-bis(2-pyridyl)pyrazine, dpq = 2,3-bis(2-pyridyl)quinoxaline), as well as their trimetallic precursors have been synthesized to provide a comparison for detailed analysis to elucidate component effects in the previously reported photocatalyst [{(phen)₂Ru(dpp)}₂Ru(dpq)PtCl ₂](PF₆)₆ (4a) (phen = 1,10-phenanthroline). Electrochemistry shows terminal Ru based highest occupied molecular orbitals (HOMOs) with remote BL' (BL' = bridging ligand coupling central Ru and cis-PtCl₂ moiety) based lowest unoccupied molecular orbitals (LUMOs). Population of a lowest-lying charge separated (³CS) excited state with oxidized terminal Ru and reduced remote BL' via intramolecular electron transfer is predicted by electrochemical analysis and is observed through steady-state and time-resolved emission studies as well as emission excitation profiles which display unusual nonunity population of the lowest lying emissive Ru→dpp ³MLCT (metal-to-ligand charge transfer) state. Each tetrametallic complex is an active photocatalyst for H₂ production from H₂O with 2a showing the highest activity (94 TON (turnover number) in 10 h, where TON = mol H₂/mol catalyst). The nature of the bridging ligand coupling the trimetallic light absorber to the cis-PtCl ₂ moiety has a significant impact on the catalyst activity, correlated to the degree of population of the ³CS excited state. The choice of terminal ligand affects visible light absorption and has a minor influence on photocatalytic H₂ production from H₂O. Evidence that an intact supramolecule functions as the photocatalyst includes a strong dependence of the photocatalysis on the identity of BL', an insensitivity to Hg(l), no detectable H₂ production from the systems with the trimetallic synthons and cis-[PtCl₂(DMSO)₂] as well as spectroscopic analysis of the photocatalytic system.

Full text of DOI:10.1021/ic4004406

Article
pubs.acs.org/IC
Subunit Variation to Uncover Properties of Polyazine-Bridged Ru(II),
Pt(II) Supramolecules with Low Lying Charge Separated States
Providing Insight into the Functioning as H2O Reduction
Photocatalysts to Produce H2
Jessica D. Knoll, Samantha L. H. Higgins, Travis A. White, and Karen J. Brewer*
Department of Chemistry, Virginia Tech, Blacksburg, Virginia 24061-0212, United States
S
* Supporting Information
ABSTRACT: Two new structurally diverse polyazine-bridged
Ru(II),Pt(II) tetrametallic complexes, [{(Ph2phen)2Ru-
(dpp)}2Ru(dpp)PtCl2](PF6)6 (1a) and [{(Ph2phen)2Ru-
(dpp)}2Ru(dpq)PtCl2](PF6)6 (2a) (Ph2phen = 4,7-diphenyl-
1,10-phenanthroline, dpp = 2,3-bis(2-pyridyl)pyrazine, dpq =
2,3-bis(2-pyridyl)quinoxaline), as well as their trimetallic
precursors have been synthesized to provide a comparison
for detailed analysis to elucidate component effects in the
previously reported photocatalyst [{(phen)2Ru(dpp)}2Ru-
(dpq)PtCl2](PF6)6 (4a) (phen = 1,10-phenanthroline).
Electrochemistry shows terminal Ru based highest occupied
molecular orbitals (HOMOs) with remote BL′ (BL′ =
bridging ligand coupling central Ru and cis-PtCl2 moiety) based lowest unoccupied molecular orbitals (LUMOs). Population
of a lowest-lying charge separated (3CS) excited state with oxidized terminal Ru and reduced remote BL′ via intramolecular
electron transfer is predicted by electrochemical analysis and is observed through steady-state and time-resolved emission studies
as well as emission excitation profiles which display unusual nonunity population of the lowest lying emissive Ru→dpp 3MLCT
(metal-to-ligand charge transfer) state. Each tetrametallic complex is an active photocatalyst for H2 production from H2O with 2a
showing the highest activity (94 TON (turnover number) in 10 h, where TON = mol H2/mol catalyst). The nature of the
bridging ligand coupling the trimetallic light absorber to the cis-PtCl2 moiety has a significant impact on the catalyst activity,
correlated to the degree of population of the 3CS excited state. The choice of terminal ligand affects visible light absorption and
has a minor influence on photocatalytic H2 production from H2O. Evidence that an intact supramolecule functions as the
photocatalyst includes a strong dependence of the photocatalysis on the identity of BL′, an insensitivity to Hg(l), no detectable
H2 production from the systems with the trimetallic synthons and cis-[PtCl2(DMSO)2] as well as spectroscopic analysis of the
photocatalytic system.
complicated process which requires bond breaking, bond
making, and multielectron reactions. While many heteroge-
neous photocatalysis systems have been reported for each of
the half reactions and for H2O splitting into H2 and O2,
homogeneous molecular photocatalysts are advantageous as
they provide for tunable excited states and systems available to
study by electrochemistry and spectroscopy.6,7 In a general
scheme, a successful system for photocatalytic reduction of
H2O to generate H2 functions through photoexcitation of a
chromophore which undergoes inter- or intramolecular
electron and/or energy transfer to produce a charge separated
state with an appropriate driving force for H2O reduction, often
employing multiple excitations to generate the two electrons
needed to produce one molecule of H2.8−10
INTRODUCTION
■
Solar energy conversion via H2O splitting is an attractive forum
in the search for alternative fuel sources as the sun delivers
more energy to the surface of the earth in 1 h than the planet
consumes in one year.1 H2O is a clean, renewable source of H2,
and H2 fuel has the highest power density of any non-nuclear
based fuel.2−4 Using the sun’s energy to split H2O requires
complex, multielectron reactions which occur at the 1.23 V
driving force. The half reactions and overall reaction for H2O
splitting are given below for pH = 7 (V vs NHE).5
4H+ + 4e− → 2H2(g) E = −0.41 V
(1)
2H2O → O2(g) + 4H+ + 4e− E = −0.82 V
(2)
2H2O → 2H2(g) + O2(g) E = −1.23 V
(3)
Analogous one electron reactions require about 5 V, energy
not available in the solar spectrum. Solar H2O splitting is a
Received: February 19, 2013
Published: August 13, 2013
© 2013 American Chemical Society
9749
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
Figure 1. Structural representations of the tetrametallic complexes 1a, 2a, 3a, and 4a. Ph2phen = 4,7-diphenyl-1,10-phenanthroline, phen = 1,10-
phenanthroline, dpp = 2,3-bis(2-pyridyl)pyrazine, and dpq = 2,3-bis(2-pyridyl)quinoxaline.
Multicomponent photocatalytic H2O reduction systems have
been studied over several decades. One of the first reported
multicomponent systems consists of a [Ru(bpy)3]2+ (bpy =
2,2′-bipyridine) light absorber (LA), whose triplet metal-to-
ligand charge transfer (3MLCT) excited state possesses the
driving force necessary to undergo intermolecular electron
transfer to a methyl viologen (MV2+) electron relay.9 In the
presence of a heterogeneous Pt catalyst and H2O, H2 gas is
evolved. EDTA (ethylenediaminetetraacetic acid) serves as a
sacrificial electron donor (ED) by providing reducing
equivalents to [Ru(bpy)3]3+ to regenerate [Ru(bpy)3]2+ and
impede undesired back electron transfer from MV•+ to the
oxidized LA. Related systems employ a [Rh(bpy)3]3+ electron
relay.10−12
Supramolecular complexes covalently couple individual
components with specific functions into one molecule to
perform a complex function and are known as photochemical
molecular devices (PMDs) when that function is induced by
light.13 PMDs are a promising forum to design efficient H2O
reduction photocatalysts.4,7,14 The components necessary for
constructing PMDs for photoinduced H2O reduction include
LA units typically composed of terminal ligands coordinated to
a LA metal, bridging ligands to act as electron relays, and a
reactive metal center for substrate interaction.13 Commonly
used LAs are Ru(II) polyazine complexes with bidentate
polyazine terminal ligands.15,16 Bis-bidentate polyazine bridging
ligands are commonly used to couple Ru-based LAs to reactive
metal centers because of their π* acceptor orbitals which enable
electrons to move toward the reactive metal for delivery to a
substrate.17,18 Several reactive metal centers have been coupled
to LAs for H2O reduction photocatalysis, such as Rh,19−24 Fe,25
Co,26,27 Pt,28−33 and Pd.34−36 Supramolecular Ru(II),Pt(II),
and Ru(II),Pd(II) complexes reported to photocatalytically
reduce H2O to H2 are few, and their functioning is not well
understood. A series of Ru(II),Pt(II) bimetallic complexes
which feature a Ru(II) based LA linked to a cis-PtCl2 unit
through an amidate bridge, [(bpy)2Ru(phenNHCO(Rbpy))-
PtCl2]2+ (where phen = 1,10-phenanthroline, R = −COOH or
−COOEt) have been reported as active H2O reduction
catalysts in the presence of EDTA, with a quantum yield (Φ)
of 0.01 and 5 TON (Turnover Number) in 10 h for R =
−COOH,28,29,31,37 and a dimer of this architecture linked
through the Rbpy group was reported to increase efficiency.38
A
bimetallic supramolecular complex [(tBu2bpy)2Ru(tpphz)-
PdCl2]2+ (tBu2bpy = 4,4′-di-tert-butyl-2,2′-bipyridine, tpphz =
tetrapyrido[3,2-a:2′,3′-c:3″,2″-h:2‴,3‴-j]phenazine) was re-
ported to produce H2 with 56 TON in 30 h.39 The chemical
integrity of Ru(II),Pt(II) and Ru(II),Pd(II) complexes upon
photolysis has raised controversy as several reports demonstrate
colloidal Pt or Pd as the active catalyst.36,40 Detailed
investigations are necessary to understand how each mixed-
metal system functions. Systems with a Co reactive metal
coupled to Ru LAs produce H2 from H2O with [(bpy)2Ru(L-
pyr)Co(dmgBF2)2(OH2)]2+ (L-pyr = [(4-pyridine)oxazolo-
[4,5-f ]phenanthroline], dmgBF2 = (difluoroboryl)-
dimethylglyoximate) affording 56 TON in 4 h when photolyzed
in the presence of triethylamine (Et3N) and [Et3NH]+.26
PMDs that are able to collect multiple electrons in the
presence of visible light and an ED are known to undergo
photoinitiated electron collection (PEC). PMDs for PEC are of
particular interest in H2O reduction as the less thermodynami-
cally demanding mechanism of H2O reduction involves
multiple electrons; however, few systems are reported. The
first reported PMD to undergo PEC was [{(bpy)2Ru-
9750
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
(dpb)}2IrCl2]5+ (dpb = 2,3-bis(2-pyridyl)benzoquinoxaline)
which collects one electron on each of the μ-dpb ligands.41
Later [(phen)2Ru(BL)Ru(phen)2]4+ (BL = tatpp = 9,11,20,22-
tetraazatetrapyrido[3,2-a:2′3′-c:3″,2″-1:2‴,3‴-n]pentacene or
tatpq =9,11,20,22-tetraazatetrapyrido[3,2-a:2′3′-c:3″,2″-
1:2‴,3‴-n]pentacene-10,21-quinone) appeared which collects
two and four electrons on the BL π* orbitals, respectively.42,43
A monometallic complex [(bpy)2Ru(pbn)]2+ (pbn = 2-(2-
pyridyl)benzo[b]-1,5-naphthyridine) couples a Ru-based LA to
an NAD+ (nicotinamide adenine dinucleotide) model ligand,
which undergoes two proton coupled electron transfers in the
presence of Et3N.44,45 These complexes, however, are not active
photocatalysts for H2O reduction. The first photoinitiated
electron collector reported to act as an active photocatalyst in
H2O reduction, [{(bpy)2Ru(dpp)}2RhCl2]5+ (dpp = 2,3-bis(2-
pyridyl)pyrazine),20,46 has recently undergone terminal ligand
and halide modification to result in the more active complex
[{(Ph2phen)2Ru(dpp)}2RhBr2]5+ (Ph2phen = 4,7-diphenyl-
1,10-phenanthroline) with Φ = 0.073 and 610 TON in 20 h
in the presence of N,N-dimethylaniline (DMA).22
Several supramolecular complexes which couple Ru(II) or
Os(II) LAs to a Pt(II) reactive metal through bidentate
bridging ligands are reported, yet few systems have been
explored as photocatalysts for H2O reduction. A series of
Ru(II),Pt(II) bimetallic complexes with [(TL)2M(BL)PtCl2]2+
have been reported where M = Ru(II) or Os(II), TL = bpy,
phen, and Ph2phen and BL = dpp, dpq (2,3-bis(2-pyridyl)-
quinoxaline), dpb and bpm (2,2′-bipyrimidine).47−52 Their
redox and photophysical properties have been studied, and in
some cases their photoinduced interactions with DNA have
been studied. The structurally diverse tetrametallic and larger
complexes [Ru{(dpq)PtCl2}3]2+,53 [Os{(dpp)Ru[(dpp)-
PtCl2]2}3]8+,54 and [(bpy)2Os(dpp)Ru{(dpp)PtCl2}2]4+,54
were synthesized and their redox and photophysical properties
were investigated. Recently, the redox, spectroscopic, and
photophysical properties of a series of complexes of the form
[{(TL)2M(dpp)}2Ru(BL′)PtCl2]6+ (M = Ru(II) or Os(II), TL
= bpy or phen, BL′ = dpp, dpq, or bpm) were reported,55,56
and only one of these complexes, [{(phen)2Ru(dpp)}2Ru-
(dpq)PtCl2]6+ (4a), has yet been reported to undergo
photoreduction and deliver reducing equivalents to a H2O
substrate.32
architecture. The use of Ph2phen in related Ru,Rh,Ru
photocatalysts for H2 production from H2O and Ru,Pt
complexes as light activated PDT agents provides for enhanced
spectral coverage, extended excited state lifetimes, and
significant enhancement in reactivity despite the formally
Ru→BL 3MLCT and/or Ru→Rh 3MMCT nature of the
photoactive states in these complexes.22,52 These studies
prompted the incorporation of Ph2phen into the title
tetrametallic motif. Herein we present the first report of the
effect of bridging ligand and terminal ligand variation on this
structurally diverse supramolecular architecture through a
systematic, detailed study to expand upon preliminary results
for the reported photocatalyst [{(phen)2Ru(dpp)}2Ru(dpq)-
PtCl2](PF6)6,32 and uncover the photophysical and photo-
chemical properties imparted by component variation through
thorough analysis.
EXPERIMENTAL SECTION
■
Materials. [(Ph2phen)2RuCl2],57 [(Ph2phen)2Ru(dpp)](PF6)2,58
[{(phen)2Ru(dpp)}2RuCl2](PF6)4 (6),32 [{(phen)2Ru(dpp)}2Ru-
(dpp)](PF6)6 (3b),56 [{(phen)2Ru(dpp)}2Ru(dpq)](PF6)6 (4b),32
[{(phen)2Ru(dpp)}2Ru(dpp)PtCl2](PF6)6 (3a),56 [{(phen)2Ru-
(dpp)}2Ru(dpq)PtCl2](PF6)6 (4a),32 cis-[PtCl2(DMSO)2],59 and
dpq60 were synthesized as previously reported. AgSO3CF3, Bu4NCl,
and Ph2phen were purchased from Alfa Aesar. RuCl3·3H2O and
Bu4NPF6 were received from Fluka. Trifluoromethanesulfonic acid,
DMA, dpp, and phen were purchased from Aldrich. NH4PF6 and
K2PtCl4 were purchased from Strem. Solvents were HPLC grade.
Spectral grade CH3CN from Burdick and Jackson was used for
electrochemical and spectroscopic studies. All materials were used as
received without further purification.
Synthesis of [{(Ph2phen)2Ru(dpp)}2RuCl2](PF6)4 (5). Two
equivalents of the monometallic precursor [(Ph2phen)2Ru(dpp)]-
(PF6)2 (2.0 g, 1.6 mmol) were heated at reflux with RuCl3·3H2O (0.20
g, 0.79 mmol) and excess LiCl (0.50 g, 11 mmol) in 75 mL of EtOH
for 24 h. The reaction mixture was cooled to room temperature (RT),
and aqueous NH4PF6 was added to induce precipitation. The
precipitate was collected by vacuum filtration and washed with H2O
and diethyl ether. Purification was achieved by column chromatog-
raphy with deactivated alumina and a 3:2 toluene/CH3CN mobile
phase. The green band was collected, and the solvent was removed
under vacuum. The product was dissolved in 15 mL of CH3CN and
precipitated into 300 mL of diethyl ether, and the precipitate was
collected by vacuum filtration. Yield: 1.4 g, 0.51 mmol (65%). ESI-
TOF MS: [M−3PF6]3+, m/z = 772.4388.
Herein we report the synthesis and electrochemical,
spectroscopic, photophysical, and photochemical properties of
two new Ru(II),Pt(II) polyazine-bridged tetrametallic com-
plexes [{(Ph2phen)2Ru(dpp)}2Ru(dpp)PtCl2](PF6)6 (1a) and
[{(Ph2phen)2Ru(dpp)}2Ru(dpq)PtCl2](PF6)6 (2a) and their
trimetallic analogues [{(Ph2phen)2Ru(dpp)}2Ru(dpp)](PF6)6
(1b) and [{(Ph2phen)2Ru(dpp)}2Ru(dpq)](PF6)6 (2b). The
comparison of properties as a function of terminal ligand and
BL′ identity provides considerable insight into relative orbital
energetics in these supramolecules. The two title Ph2phen
tetrametallic complexes as well as their phen analogues (Figure
1) have been studied as photocatalysts in the visible-light driven
reduction of H2O to produce H2.
The stabilized dpq-based lowest unoccupied molecular
orbital (LUMO) in [{(TL)2Ru(dpp)}2Ru(dpq)PtCl2](PF6)6
(2a and 4a) provides for enhanced driving force for charge
separation, a larger driving force for reductive quenching of the
excited state, and enhanced H2 production in comparison with
the complexes with BL′ = dpp. Variation of the terminal ligand
from phen to Ph2phen results in fewer spectral gaps, enhancing
the visible light absorbing properties of this molecular
Synthesis of [{(Ph2phen)2Ru(dpp)}2Ru(dpp)](PF6)6 (1b). The
precursor 5 (0.30 g, 0.11 mmol) and 2 equiv of AgSO3CF3 (0.057 g,
0.22 mmol) were heated at reflux in 20 mL of EtOH for 2 h to remove
the chloride ligands. The reaction mixture was then added dropwise to
a hot, stirring solution of dpp (0.30 g, 1.3 mmol) in 25 mL of ethylene
glycol and heated at reflux for 18 h. The reaction mixture was cooled
to RT, and Bu4NCl (0.10 g, 0.36 mmol) was added to remove Ag+ as
AgCl(s) and aqueous NH4PF6 was added to induce precipitation as a
−
PF6 salt. The precipitate was collected by vacuum filtration.
Purification was achieved on a Sephadex LH-20 size exclusion column
with 2:1 EtOH/CH3CN mobile phase. The major purple-red band was
collected, and the solvent removed under vacuum. The product was
dissolved in 10 mL of CH3CN and syringed filtered into 200 mL of
diethyl ether to precipitate, and collected by vacuum filtration. Yield:
0.28 g, 0.087 mmol (80%). ESI-TOF MS: [M−4PF6]4+, m/z =
656.2841.
Synthesis of [{(Ph2phen)2Ru(dpp)}2Ru(dpq)](PF6)6 (2b). 2b
was prepared as above by substituting dpq (0.30 g, 0.10 mmol). Yield:
0.35 g, 0.11 mmol (75%). ESI-TOF MS: [M−4PF6]4+, m/z =
669.1125.
Synthesis of [{(Ph2phen)2Ru(dpp)}2Ru(dpp)PtCl2](PF6)6 (1a).
1b (0.10 g, 0.031 mmol) and excess cis-[PtCl2(DMSO)2] (0.10 g, 0.24
mmol) were heated at reflux in 25 mL of methanol for 48 h. The
9751
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
Figure 2. Synthetic scheme for 2a by the building block method. Ph2phen = 4,7-diphenyl-1,10-phenanthroline, dpp = 2,3-bis(2-pyridyl)pyrazine, dpq
a
b
= 2,3-bis(2-pyridyl)quinoxaline. From reference 57. From reference 58.
reaction mixture was cooled to RT and aqueous NH4PF6 was added to
induce precipitation. The solid was collected by vacuum filtration.
Purification was achieved by dissolving in 10 mL of CH3CN and
Yield: 0.11 g, 0.030 mmol (95%). ESI-TOF MS: [M−2PF6]2+, m/z =
1616.1486.
ESI-TOF Mass Spectrometry. High resolution mass spectral data
were collected by electrospray ionization time-of-flight mass
spectrometry performed in CH3CN with an Agilent Technologies
6220 Accurate-Mass TOF LC-MS (0.4 mL/min flow rate) with a dual
ESI source. Low resolution mass spectral analysis of the Pt containing
complexes was performed by direct infusion of the sample dissolved in
syringe filtering into 200 mL of diethyl ether several times. Yield: 0.10
g, 0.030 mmol (95%). ESI-TOF MS: [M−3PF6]3+, m/z = 1012.4365.
Synthesis of [{(Ph2phen)2Ru(dpp)}2Ru(dpq)PtCl2](PF6)6 (2a).
2a was prepared as above by substituting 2b (0.10 g, 0.031 mmol).
9752
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
CH3CN (10 μL/min flow rate) into the Thermo Instrument TSQ
triple quadrupole mass spectrometer equipped with an ESI source.
Electrochemistry. Cyclic voltammetry and square wave voltam-
metry were performed in a one compartment, three electrode cell
utilizing a glassy carbon working electrode, platinum wire auxiliary
electrode, and Ag/AgCl (NaCl, 3 M) reference electrode calibrated
the device, exploring the impact on basic chemical properties.
These supramolecules exhibit Ru-based oxidations and ligand-
based reductions, with bridging ligand reductions occurring at
potentials positive of −1.0 V vs Ag/AgCl. The reductive
electrochemistry at potentials more negative than −1.0 V is
complex with BL−/2−, TL0/−, and TL−/2− couples overlapping,
so discussion is limited to first bridging ligand reductions.
Oxidation and reduction potentials for the new Ph2phen
complexes and their previously reported phen analogues are
given in Table 1 with square wave voltammograms of the
+
against FeCp2/FeCp2 (0.46 V vs Ag/AgCl) with an Epsilon
potentiostat from Bioanalytical Systems, Inc. Measurements were
recorded at 0.1 V/s in CH3CN with 0.1 M Bu4NPF6 at RT
deoxygenated by bubbling with argon. The working electrode was
polished with alumina paste prior to analysis.
Electronic Absorption Spectroscopy. Electronic absorption
spectroscopy was performed using an Agilent 8453 UV−Vis
spectrophotometer with a diode array with 1 nm resolution and a
spectral range of 190 to 1100 nm. Spectra were measured in a 1 cm
quartz cuvette in CH3CN at RT. Extinction coefficient experiments
were performed in triplicate.
Table 1. Electrochemical Data for the Trimetallic and
Tetrametallic Complexes with TL = phen and Ph2phen.
a
E1/2 (V vs Ag/AgCl)
b
c
complex oxidation (assignment)
reduction (assignment)
Steady-State, Time-Resolved Emission Spectroscopy, and
Emission Excitation Spectroscopy. Steady-state emission spec-
troscopy and emission excitation spectroscopy were performed on a
QuantaMaster Model QM-200-45E fluorometer from Photon
Technologies, Inc. (PTI), and spectra were corrected for PMT
response. Emission quantum yields were measured vs [Os(bpy)3]-
(PF6)2 (Φem = 0.0046) in deoxygenated RT CH3CN.61 Excited state
emission lifetime measurements were performed with a PTI PL2300
nitrogen laser pumping a PTI PL 201 continuously tunable dye laser as
an excitation source (360−900 nm) with the signal digitized and
displayed on a LeCroy 9361 Dual 300 MHz oscilloscope (2.5 Gs/s).
RT spectra were measured in deoxygenated CH3CN, and 77 K spectra
were measured in a 4:1 (v/v) EtOH/MeOH glass.
5
0.86
(RuII/III
1.59
(2RuII/III
−0.68
−0.81
)
)
)
(dpp0/−
)
)
)
)
)
)
(dpp0/−
)
)
)
)
)
)
)
)
)
)
d
6
0.79
(RuII/III
1.56
(2RuII/III
−0.70
(dpp0/−
−0.84
)
(dpp0/−
1b
2b
3b
4b
1a
2a
1.55
(2RuII/III
−0.45
−0.60
(dpp0/−
−1.02
)
)
)
)
)
)
)
)
(dpp0/−
(dpp0/−
)
)
)
)
)
)
)
)
1.55
(2RuII/III
−0.42
(dpp0/−
−0.58
−0.79
(dpp0/−
(dpq0/−
e
1.61
(2RuII/III
−0.45
−0.59
(dpp0/−
−1.00
(dpp0/−
(dpp0/−
d
1.56
(2RuII/III
−0.42
(dpp0/−
−0.59
−0.80
(dpp0/−
(dpq0/−
1.56
(2RuII/III
−0.33
−0.53
(dpp0/−
−0.64
(dpp0/−
Photochemistry. H2 production experiments were performed
using a locally designed LED array as the light source (λirr = 470 nm)
with a photon flux of 2.3 × 1019 photons/min62 and H2 was quantified
in real time using HY-OPTIMA 700 in-line process H2 sensors from
H2scan (Valencia, CA).63 The photocatalyst in a CH3CN/H2O
solution was injected into an airtight cell with a septum cap and was
deoxygenated by bubbling with argon. The ED DMA was separately
deoxygenated and injected into the cell just prior to photolysis. The
final solution volume was 4.5 mL (50 μM photocatalyst, 0.62 M H2O,
(dpp0/−
)
)
)
)
1.57
(2RuII/III
−0.02
−0.50
−0.65
(dpq0/−
(dpp0/−
(dpp0/−
e
3a
4a
1.63
(2RuII/III
−0.32
−0.51
(dpp0/−
−0.63
(dpp0/−
(dpp0/−
d
1.58
(2RuII/III
−0.05
(dpq0/−
−0.42
−0.59
(dpp0/−
(dpp0/−
a
Measurements recorded in deoxygenated CH3CN at RT with 0.1 M
Bu4NPF6 electrolyte. phen = 1,10-phenanthroline, Ph2phen = 4,7-
diphenyl-1,10-phenanthroline, dpp = 2,3-bis(2-pyridyl)pyrazine, dpq =
2,3-bis(2-pyridyl)quinoxaline. Bold indicates central Ru. Bold
−
1.5 M DMA, 0.11 mM [DMAH+][CF3SO3 ]) with a headspace
volume of about 15.3 mL.
b
c
d
e
indicates BL′. From reference 32. From reference 56.
RESULTS AND DISCUSSION
■
Synthesis and Characterization. The tetrametallic
complexes were synthesized by a building block method
(Figure 2) which allows for purification and analysis of each
precursor. 1H NMR is not a useful characterization method for
such complexes because of the mixture of optical and geometric
isomers possible. In supramolecular complexes with several
polyazine ligands which possess a large number of C, H, and N
atoms, elemental analysis is expected to be too similar between
the reported compounds. The precursors and predicted
impurities at each step of the synthetic scheme exhibit
characteristic redox and spectroscopic properties, providing
square wave voltammetry and emission spectroscopy as
important tools to assay electrochemical and spectroscopic
purity. ESI-MS is used to identify the presence of the high mass
product, and the mass spectral data is given in Supporting
Information, Table S1. A combination of these diverse methods
suggests purity of the complexes. Mixtures of isomers have
been reported to have little effect on the spectroscopic
properties of Ru(II) polyazine-bridged supramolecular com-
plexes.64
tetrametallic complexes in Figure 3. For the dichloride
trimetallic precursors 6 and 5, two reversible oxidation couples
are observed, with the central RuII/III oxidation occurring at
0.79 and 0.86 V vs Ag/AgCl for the phen and Ph2phen
complexes, respectively, and at 1.56 and 1.59 V for the two
Electrochemistry. Cyclic voltammetry and square wave
voltammetry of the tetrametallic complexes and their
precursors give insight into relative orbital energetics. This is
facilitated by our synthetic ability to substitute each subunit of
Figure 3. Square wave voltammograms for (A) 3a, (B) 4a, (C) 1a, and
(D) 2a recorded in CH3CN with 0.1 M Bu4NPF6 at RT with potential
referenced to Ag/AgCl.
9753
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
a
Table 2. Electronic Absorption Spectroscopy Data
ε × 10−4 (M−1 cm−1
)
assignment
abs
complex
λmax (nm)
5
276
21.3
9.1
Ph2phen π→π*
dpp π→π*
Ru(dπ)→Ph2phen(π*) CT
Ru(dπ)→dpp(π*) CT
phen π→π*
315 (sh)
442
4.5
625
3.4
b
6
262
17.2
5.4
330
dpp π→π*
430
2.9
Ru(dπ)→phen(π*) CT
Ru(dπ)→dpp(π*) CT
Ph2phen π→π*
625
3.4
1b
2b
276
19.5
5.7
340 (sh)
432
dpp π→π*
4.2
Ru(dπ)→Ph2phen(π*) CT
Ru(dπ)→dpp(π*) CT
Ph2phen π→π*
dpp π→π*, dpq π→π*
Ru(dπ)→Ph2phen(π*) CT
554
4.2
276
14.4
5.5
340 (sh)
430
3.6
550
3.6
Ru(dπ)→dpp(π*) CT, Ru(dπ)→dpq(π*) CT
phen π→π*
dpp π→π*
c
3b
262
16.6
4.2
340
421
2.8
Ru(dπ)→phen(π*) CT
Ru(dπ)→dpp(π*) CT
phen π→π*
dpp π→π*, dpq π→π*
Ru(dπ)→phen(π*) CT
Ru(dπ)→dpp(π*) CT, Ru(dπ)→dpq(π*) CT
Ph2phen π→π*
543
3.4
b
4b
262
15.9
4.8
350 (sh)
415
2.6
541
3.8
1a
2a
276
17.0
7.7
326 (sh)
432
dpp π→π*
4.2
Ru(dπ)→Ph2phen(π*) CT
Ru(dπ)→dpp(π*) CT
Ph2phen π→π*
dpp π→π*, dpq π→π*
Ru(dπ)→Ph2phen(π*) CT
Ru(dπ)→dpp(π*) CT, Ru(dπ)→dpq(π*) CT
phen π→π*
550
4.2
276
13.9
5.2
340 (sh)
430
3.6
550
3.1
c
3a
262
15.5
5.6
330 (sh)
421
dpp π→π*
2.9
Ru(dπ)→phen(π*) CT
Ru(dπ)→dpp(π*) CT
phen π→π*
dpp π→π*, dpq π→π*
Ru(dπ)→phen(π*) CT
Ru(dπ)→dpp(π*) CT, Ru(dπ)→dpq(π*) CT
543
4.2
b
4a
262
15.0
4.8
350 (sh)
415
3.2
541
3.8
a
Measurements recorded in deoxygenated CH3CN at RT. phen = 1,10-phenanthroline, Ph2phen = 4,7-diphenyl-1,10-phenanthroline, dpp = 2,3-
b
c
bis(2-pyridyl)pyrazine, dpq = 2,3-bis(2-pyridyl)quinoxaline. From reference 32. From reference 56.
overlapping, one electron oxidations of the terminal Ru centers.
Coordination of a BL′ in place of the chloride ligands on the
central Ru results in stabilization of that Ru center, shifting its
oxidation potential outside the solvent window. The trimetallic
complexes [{(Ph2phen)2Ru(dpp)}2Ru(BL′)](PF6)6 (1b and
2b) possess terminal Ru-based highest occupied molecular
orbitals (HOMOs), with an oxidation potential at 1.55 V for
both 1b and 2b. Replacing Ph2phen with phen as the terminal
ligand results in the terminal RuII/III oxidation occurring at
slightly higher potentials, consistent with the slightly more
electron donating nature of Ph2phen (1.61 and 1.56 V for 3b
and 4b, respectively). Upon coordination of the cis-PtCl2
moiety, the oxidative processes remain similar to the trimetallic
precursors and the terminal Ru based HOMO is maintained.
Reductive electrochemistry is predictably complicated.
Cathodically, the first reduction of each BL is observed at
potentials between 0 V and about −1.0 V vs Ag/AgCl. 5
displays two subsequent one electron reductions at −0.68 and
−0.81 V, corresponding to the two μ-dpp ligands, consistent
with the previously reported phen analogue 6. Coordination of
the BL′ in place of chlorides results in a shift of the μ-dpp
reductions to −0.45 and −0.60 V for 1b and −0.42 and −0.58
V for 2b, attributed to the increased positive charge of the
complex from +4 to +6 and replacement of σ-donating chloride
ligands with a π-accepting BL′. The terminal BL′, prior to
coordination to cis-PtCl2, is reduced at −1.02 V (1b) and −0.79
V (2b), consistent with dpq’s stabilized acceptor orbitals with
respect to dpp. Coordination of the cis-PtCl2 unit results in
orbital inversion with the LUMO localized on the BL′ between
central Ru and Pt, consistent with similar systems which display
enhanced stabilization of a polyazine bridging ligand when
bound to Pt(II) vs Ru(II).53 These first reduction processes
occur at −0.33 V for 1a and −0.02 V for 2a. The two μ-dpp
reductions occur following the BL′ reduction. Similar trends are
9754
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
Figure 4. Electronic absorption spectra of (A) 4a (solid line) and 3a (dashed line), and (B) 2a (solid line) and 4a (dashed line) in CH3CN at RT.
observed in the TL = phen complexes (3a and 4a). BL′
variation between Ru and Pt has a great impact on the LUMO
energy of the tetrametallic complexes, while terminal ligand
variation finely tunes the HOMO energy. Relating electro-
chemical data to HOMO and LUMO localization is particularly
valid in the case of the [{(TL)2Ru(dpp)}2Ru(BL′)](PF6)6 and
[{(TL)2Ru(dpp)}2Ru(BL′)PtCl2](PF6)6 complexes, as the
overall charge of the complex and the HOMO energy level
are not impacted upon cis-PtCl2 coordination and affect
primarily the BL′ acceptor orbitals results in LUMO orbital
inversion. The electrochemistry of the tetrametallic complexes
suggests a lowest-lying charge separated state upon photo-
excitation with an oxidized terminal Ru and reduced remote
BL′; the notion of charge separation is important in solar
energy conversion schemes.
Electronic Absorption Spectroscopy. The Ph2phen
trimetallic and tetrametallic complexes exhibit electronic
absorption spectroscopy trends similar to the previously
reported complexes of this molecular architecture; however,
the use of Ph2phen results in enhanced absorptivity throughout
the visible region with fewer spectral gaps, typical of Ph2phen
Ru complexes.65 These complexes are very efficient LAs
throughout the UV and visible with ε between 4−20 × 104 M−1
cm−1. The electronic absorption spectroscopy and assignments
are summarized in Table 2. The dichloride precursor 5 displays
a lowest energy Ru(dπ)→dpp(π*) CT transition with similar
energy and absorptivity to the TL = phen analogue (λmax = 625
nm, ε = 3.4 × 104 M−1 cm−1). Utilization of the Ph2phen
terminal ligand results in a molar absorptivity of the Ru(dπ)→
Ph2phen(π*) CT transition about 440 nm of 4.5 × 104 M−1
cm−1, increased vs the molar absorptivity of the Ru(dπ)→
phen(π*) CT transition of 2.9 × 104 M−1 cm−1 for 6. The UV
region is dominated by Ph2phen- and dpp-based π→π*
transitions at 276 nm (ε = 21.3 × 104 M−1 cm−1) and 315
nm (ε = 9.1 × 104 M−1 cm−1), respectively. Replacing the two
σ-donating chloride ligands with a π-accepting BL′ (dpp or
dpq) stabilizes the central Ru(dπ) orbitals and results in a shift
of the lowest energy CT transition (assigned as overlapping
Ru(dπ)→dpp(π*) CT and Ru(dπ)→BL′(π*) CT transitions)
to 554 nm for 1b and 550 nm for 2b. These complexes absorb
light more efficiently than their TL = phen analogues between
380 and 500 nm because of the enhanced absorptivity of the
Ru(dπ)→Ph2phen(π*) CT transitions. Coordination of the cis-
PtCl2 moiety to the trimetallic complexes maintains similar
light absorbing properties in the UV and visible regions. In the
tetrametallic architecture, variation of BL′ from dpp to dpq
results in minor changes to the absorption spectra, while
variation of terminal ligand from phen to Ph2phen enhances the
absorptivity between 350 and 500 nm (Figure 4).
Emission Spectroscopy and Photophysics. The excited
state processes of the tetrametallic complexes are complicated
by the number of energetically close excited states and complex
photophysics, resulting in interesting properties. The emissive
3
nature of the lowest lying MLCT excited state provides a
probe into these processes. Typical of Ru(II) polyazine
1
complexes, population of the MLCT excited state leads to
intersystem crossing to populate a 3MLCT state with near unit
efficiency, and these states are known to undergo intra- and
intermolecular electron transfer.16 The trimetallic models and
tetrametallic systems display long-lived Ru→dpp 3MLCT
states, much longer than bimetallic systems with directly
coupled reactive metals, a design consideration for these
complexes.23,52 Emission spectroscopy for a series of this
Ru(II),Pt(II) supramolecular architecture with varied terminal
ligand and BL′ was recently reported, and the phen complexes
3a and 4a will be discussed in comparison to the Ph2phen
analogues to provide a more detailed analysis of terminal ligand
and BL′ effects. The steady-state and time-resolved emission
spectroscopy data for the trimetallic and tetrametallic
complexes at RT in CH3CN and at 77 K in a 4:1 EtOH/
MeOH rigid matrix are presented in Table 3. The excited state
lifetimes for all the trimetallics are quite similar whether the
terminal ligand is phen or Ph2phen, and the same trend is
observed for the tetrametallics. This is in contrast to the related
Ru,Rh,Ru and Ru,Pt systems where Ph2phen extends the
22,52
3
lifetime of the formally Ru→dpp MLCT state.
The phen
analogues were reported to populate the lowest-lying terminal
Table 3. Steady-State and Time-Resolved Emission
a
Spectroscopy Data
RT
77 K
kr ×
knr
×
ket ×
λem Φem
complex (nm)
103
x
10−3
10−6
10−6
λem
(nm) (μs)
τ
b
b
τ (μs) (s−1
)
(s−1
)
(s−1
)
1b
2b
3b
4b
1a
2a
764
764
760
752
766
760
756
752
1.04
1.09
1.01
1.01
0.71
0.37
0.71
0.32
0.11
9.3
8.9
8.9
9.1
9.1
8.9
8.9
9.1
9.1
705
715
705
715
715
715
705
715
2.0
2.0
2.1
2.0
2.0
2.0
2.1
2.0
0.11
9.7
9.2
9.2
9.3
9.7
9.2
9.2
c
0.11
d
0.11
0.077
0.075
0.083
0.080
4.1
4.4
3.0
3.4
c
3a
4a
d
a
Measurements at RT were performed on CH3CN solutions
deoxygenated with Ar. Measurements at 77 K were performed in a
b
4:1 EtOH/MeOH glass. Values corrected for PMT response. Values
for the tetrametallic complexes are assumed to be the same as the
c
values for the corresponding trimetallic complexes. From reference
d
32. From reference 56.
9755
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
Figure 5. State diagrams for 2b and 2a. Ru indicates central Ru. Ph2phen = 4,7-diphenyl-1,10-phenanthroline, dpp = 2,3-bis(2-pyridyl)pyrazine, dpq
= 2,3-bis(2-pyridyl)quinoxaline.
Ru(dπ)→dpp(π*) 3CT excited state with unit efficiency for 3b
and 4b, but for the tetrametallic complexes 3a and 4a a lowest-
lying nonemissive 3CS (charge separated) excited state
(oxidized terminal Ru and reduced BL′) is competitively
populated via intramolecular electron transfer from the emissive
terminal Ru(dπ)→dpp(π*) 3CT state (ket) with some
contribution from the higher lying central Ru(dπ)→BL′(π*)
3CT excited state (ket′). Similar trends are observed for the
Ph2phen analogues and are discussed below. A simplified state
diagram for the photophysical processes involved in the
trimetallic and tetrametallic complexes is shown in Figure 5.
The state diagrams are greatly simplified to display only the
emissive excited state and the state involving BL′ to highlight
the effect of Pt coordination on stabilizing the higher lying
excited state. The central Ru→μ-dpp CT, terminal Ru→TL
CT, and terminal ligand, dpp, and BL′ π→π* excited states can
also be populated; however, they are omitted in the state
diagrams as their energies, and subsequent impact on emissive
state population, are not expected to significantly change upon
Pt coordination. The predicted differences in electronic
absorption spectroscopy by primarily lowering the terminal
Ru→BL′ CT energy are greatly masked by strong transitions
because of multiple Ru→dpp CT and Ru→terminal ligand CT
excited states. This results in the expected small shifts in
electronic absorption spectroscopy upon cis-PtCl2 coordination
to BL′.
architecture is the emission that arises from the part of the
molecule that is remote from BL′ and Pt. The unusual
discrepancy in the effect on Φem and τ suggests that another
excited state is competitively populated, and this degree of
population is strongly influenced by the energy of the BL′
acceptor orbitals. The excited state lifetime is not reduced at 77
K upon platination of the trimetallic to produce the
tetrametallic complexes, and the emission profiles are nearly
identical, indicating that intramolecular electron transfer
observed at RT is impeded in low temperature rigid media.
To study the intramolecular electron transfer to populate the
3
3CS state from the emissive terminal Ru→dpp MLCT excited
state (ket) in the tetrametallic complexes, the trimetallic
analogues 1b and 2b were used as models. These models are
3
particularly valid as they possess the same Ru→μ-dpp MLCT
emissive state with the dpp coordinated to two Ru centers as
the title tetrametallic complex, the excited states are very similar
in energy and band shape, the μ-dpp is in the same
coordination environment, and the terminal ligands bound to
Ru are the same. Equations 4 and 5 give information on ket
based on the change in excited state lifetimes (τ).
1
τmodel
=
kr + knr
(4)
Both trimetallic complexes 1b and 2b emit at 764 nm from
1
3
τtetrametallic
=
the formally terminal Ru(dπ)→dpp(π*) MLCT excited state
kr + knr + ket
with Φem = 1.04 × 10−3 and 1.09 × 10−3 for 1b and 2b,
respectively, and both complexes have relatively long-lived
excited state lifetimes of 0.11 μs, similar to the values reported
for the TL = phen analogues. Upon platination of these
trimetallic complexes, the resulting tetrametallic complexes
emit at similar energies, yet a quenching of the emission is
observed (Φem = 7.1 × 10−4 for 1a and 3.7 × 10−4 for 2a), as
well as a reduction of the excited state lifetime, although to a
lesser extent (0.077 μs for 1a and 0.075 μs for 2a). While
enhanced nonradiative decay upon addition of the cis-PtCl2 unit
to BL′ may result in the shortened excited state lifetimes, the
discrepancy in the degree of quenching of Φem between BL′ =
dpp and dpq is not expected to be influenced only by enhanced
nonradiative decay. An important aspect of this molecular
(5)
The rate constants for radiative and nonradiative decay (kr
and knr, respectively) are assumed to be the same in the
trimetallic models and the tetrametallic analogues; these rate
constants and ket are given in Table 3. 4a and 2a have larger ket
values than their BL′ = dpp (3a and 1a) analogues, indicating
3
more efficient population of the CS excited state from the
lowest-lying emissive 3MLCT state. This is expected because of
the about 0.3 V increased driving force for intramolecular
electron transfer in the dpq acceptor systems relative to dpp.
The quantum yields of emission can also be related to the
rate constants for the processes deactivating the emissive state
with the trimetallic as a model in eqs 6 and 7.
9756
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
Figure 6. Overlaid absorption (dashed line) and excitation (solid line) spectra of (A) 1b, (B) 2b, (C) 1a, and (D) 2a in deoxygenated CH3CN at
RT.
⎛
⎞
suggest a higher-lying excited state, likely the central Ru→BL′
kr
k + k
Φe3m(model) = Φ3pop(model)
3
⎜
⎝
⎟
⎠
CT state, directly populates the CS state when BL′ = dpq.
MLCT
MLCT
(6)
(7)
r
nr
Photochemistry. The tetrametallic complex 4a was
recently reported as an active photocatalyst for H2O reduction
to generate H2,32 and herein we report the effects of terminal
ligand and BL′ variation on this tetrametallic supramolecular
architecture. The four tetrametallic complexes display photo-
catalysis reducing H2O to H2. A system containing 50 μM
[{(TL)2Ru(dpp)}2Ru(BL′)PtCl2](PF6)6 complex (1a−4a),
0.62 M H2O, 1.5 M DMA ED, and 110 μM [DMAH+]-
⎛
⎞
⎟
⎠
kr
k + k + ket
Φe3m(tetrametallic) = Φ3pop(tetrametallic)
⎜
MLCT
MLCT
⎝
r
nr
The tetrametallic complexes display slightly shortened
3MLCT excited state lifetimes but dramatically reduced
quantum yields for emission. In eq 6, the quantum yield of
−
[CF3SO3 ] irradiated at 470 nm produces H2 from the H2O
3
population of the lowest-lying MLCT state is assumed to be
substrate (Figure 7, Table 4). A dramatic difference in
unity, typical of Ru-polyazine complexes.16 The quantum yield
of population of this state in eq 7 is not unity in the
tetrametallic architecture, a very unusual property of these
3
systems. This is due to population of the CS state from a
higher lying state. The Φpop of the emissive state can be
calculated based on the excited state lifetimes and emission
quantum yields. The complexes with BL′ = dpp populate the
emissive excited state with Φ = 0.93 (3a) and 0.99 (1a) in line
with typical Ru complexes, suggesting minimal indirect
3
population of the nonemissive CS state from higher lying
3MLCT states. The BL′ = dpq complexes populate the emissive
state with Φ = 0.44 (4a) and 0.51 (2a), an uncommon property
imparted by this tetrametallic motif. This suggests the
population of the emissive state would depend on the
wavelength of excitation even though these complexes have a
variety of MLCT states that are very close in energy. The
excitation spectra for the Ph2phen tetrametallic complexes and
model trimetallic systems are provided in Figure 6, illustrating
that the excitation spectra follow the absorption profile for the
trimetallic systems (1b and 2b) and the BL′ = dpp tetrametallic
complexes (1a) but vary in the tetrametallics with BL′ = dpq
(2a). This is in agreement with the unity population of the
emissive state in the trimetallics and near unity population of
the emissive state when BL′ = dpp and enhanced population of
Figure 7. Hydrogen production profiles for photolysis systems with 2a
(circles), 4a (squares), 1a (triangles), and 3a (diamonds). Photolysis
systems contained 50 μM tetrametallic complex, 0.62 M H2O, 1.5 M
−
DMA, and 110 μM [DMAH+][CF3SO3 ] in deoxygenated CH3CN
and irradiated at 470 nm. TON = mol H2/mol catalyst.
photocatalytic activity is observed upon variation of the nature
of BL′, with 2a producing 21 μmol H2 (94 TON, where TON
= mol H2/mol catalyst) in 10 h compared to the dpp analogue
1a producing only 5.3 μmol H2 (23 TON). The same trend is
observed for the phen analogues, with 4a and 3a producing 15
μmol H2 (66 TON) and 4.0 μmol H2 (18 TON), respectively.
The [{(TL)2Ru(dpp)}2Ru(dpq)PtCl2](PF6)6 (2a and 4a)
3
the nonemissive CS state resulting in nonunity population of
the lowest energy 3MLCT state when BL′ = dpq. These results
9757
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
absorptivity of the photocatalyst throughout the visible range in
comparison to the phen analogues. The BL′ dependence
strongly supports intact supramolecules as active photo-
catalysts, as Pt decomplexation would revert the LUMO to
the dpp and the lowest lying excited state terminal Ru→dpp
3MLCT, independent of the dpq BL′.
Table 4. Hydrogen Production Data for Ru(II),Pt(II)
Tetrametallic Complexes
a
b
complex
μmol H2 at 10 h
TON at 10 h
1a
2a
3a
4a
5
1
23
94
18
66
2
6
1
4
21
4
1
1
1
The importance of BL′ in this motif for photocatalysis
supports intact supramolecules as being the active photocatalyst
which is further supported by additional control experiments
described below. Photocatalytic H2O reduction with 4a was not
impeded by addition of Hg(l), indicating the molecular
architecture does not dissociate to form the trimetallic LA 4b
and colloidal Pt.66 Further evidence for the tetrametallic
supramolecular architecture remaining intact during photolysis
is provided by experiments in which equimolar amounts of
[{(TL)2Ru(dpp)}2Ru(BL′)](PF6)6 (1b−4b) and cis-
[PtCl2(DMSO)2] were used in the photolysis system in place
of the analogous [{(TL)2Ru(dpp)}2Ru(BL′)PtCl2](PF6)6 (1a−
4a). The systems with [{(TL)2Ru(dpp)}2Ru(BL′)](PF6)6 and
cis-[PtCl2(DMSO)2] produced no detectable H2 after 10 h
photolysis for each combination of terminal ligand and BL′, and
in each case, solid precipitate was observed in the bottom of
each photolysis cell which indicates in situ formation of Pt0.
Additionally, the electronic absorption spectroscopy after
photolysis is noticeably different for each [{(TL)2Ru-
(dpp)}2Ru(BL′)PtCl2](PF6)6 and its analogous [{(TL)2Ru-
(dpp)}2Ru(BL′)](PF6)6 and cis-[PtCl2(DMSO)2] sample
(Figure 8), suggesting that the species present in the photolysis
samples are different.
15
c
1b + cis-[PtCl2(DMSO)2]
2b + cis-[PtCl2(DMSO)2]
3b + cis-[PtCl2(DMSO)2]
4b + cis-[PtCl2(DMSO)2]
c
c
c
a
Results correspond to 10 h photolysis time with 470 nm LED light
source (light flux = 2.36 × 1019 photons/min). Photolysis solutions
contained 50 μM tetrametallic complex (or 50 μM trimetallic complex
and 50 μM cis-[PtCl2(DMSO)2]), 0.62 M H2O, 1.5 M DMA, 110 μM
b
−
[DMAH+][CF3SO3 ] in deoxygenated CH3CN. TON = mol H2/mol
c
catalyst. No H2 detected.
system has a higher driving force for intramolecular electron
transfer and a lower lying π* acceptor orbital on the dpq ligand
bound to the catalytically active Pt site which enhances H2
production. These dpq containing systems also have reduced
3
population of the emissive MLCT state, Φpop
= 0.51 for
3MLCT
3
2a and 0.44 for 4a, providing enhanced population of the CS
state. This suggests an important role in photocatalysis of this
3CS state. The substantially stabilized dpq(π*) orbital relative
to dpp may also impede back electron transfer from BL′ to dpp
as this process is substantially uphill in the dpq systems and
nearly energy neutral in the dpp systems. The rate constants for
emission quenching by the ED DMA (Supporting Information,
Figure S7 and Table S2) for the four tetrametallic complexes
are in the range of 3.3−5.4 × 109 M−1 s−1, which is near the
diffusion controlled limit. This suggests that upon photo-
excitation to populate an MLCT excited state, reductive
quenching occurs very rapidly to afford electron collection on
the bridging ligands, and from this reduced complex H2O is
reduced to produce H2.
CONCLUSIONS
■
Two new mixed-metal, mixed-ligand complexes of the form
[{(Ph2phen)2Ru(dpp)}2Ru(BL′)PtCl2](PF6)6 as well as their
trimetallic precursors [{(Ph2phen)2Ru(dpp)}2Ru(BL′)](PF6)6,
where BL′ = dpp or dpq, were synthesized and their
spectroscopy, electrochemistry, photophysics, and photochem-
istry were studied in comparison to their phen analogues.
Detailed analysis of the Ph2phen systems, 3a, and the previously
explored 4a are provided, focusing on excited state and
photocatalytic properties. Introduction of Ph2phen into this
supramolecular architecture has a minor impact on the orbital
energetics of these complexes, while it has a significant impact
on the light absorbing properties with highly efficient visible
light absorptivity. Interestingly, the Ru→dpp 3MLCT lifetime is
not substantially lengthened in this motif unlike the
[{(Ph2phen)2Ru(dpp)}2RhBr2](PF6)5 and [(Ph2phen)2Ru-
(dpp)PtCl2]Cl2 motifs. Variation of BL′ from dpp to dpq
While the identity of BL′ has a strong impact on the
efficiency of the supramolecular architecture in photocatalytic
production of H2 from H2O, the terminal ligand only serves to
fine-tune the activity. This is somewhat surprising as the
[{(Ph2phen)2Ru(dpp)}2RhBr2](PF6)5 system was recently
shown by our group to be a far superior photocatalyst than
the phen analogue.22 In addition, the [(Ph2phen)2Ru(dpp)-
PtCl2]Cl2 system has been shown to provide DNA photo-
cleavage in stark contrast to the lack of activity of bpy and phen
analogues.52 Utilization of Ph2phen does result in enhanced
Figure 8. Electronic absorption spectroscopy after 10 h of photolysis in CH3CN in the presence of 1.5 M DMA, 0.62 M H2O, 110 μM
−
[DMAH+][CF3SO3 ] and (A) 50 μM 2a (solid line) or 50 μM 2b and 50 μM cis-[PtCl2(DMSO)2] (dashed line), (B) 50 μM 1a (solid line) or 50
μM 1b and 50 μM cis-[PtCl2(DMSO)2] (dashed line), (C) 50 μM 4a (solid line) or 50 μM 4b and 50 μM cis-[PtCl2(DMSO)2] (dashed line), and
(D) 50 μM 3a (solid line) or 50 μM 3b and 50 μM cis-[PtCl2(DMSO)2] (dashed line).
9758
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
results in a more stabilized LUMO and lowest-lying 3CS excited
state. A probe into the excited state dynamics of the molecular
architecture is provided by population of an emissive MLCT
bpy = 2,2′-bipyridine
CS = charge separated state
3
CT = charge transfer
excited state. This emissive state is populated with nonunit
efficiency as a result of competitive population of the 3CS state
with oxidized terminal Ru and reduced BL′, a very unusual
property displayed by these systems. The identity of BL′ has
the largest impact on Φpop of the emissive state, with near unity
population for [{(TL)2Ru(dpp)}2Ru(dpp)PtCl2](PF6)6 (1a
and 3a) and greatly decreased population for [{(TL)2Ru-
(dpp)}2Ru(dpq)PtCl2](PF6)6 (2a and 4a). Most supramolec-
ular H2O reduction photocatalysts have very short-lived excited
states, making excited state analysis more difficult. The
relatively long-lived excited state in this tetrametallic
architecture arising from spatial separation of the LA and
reactive metal provides a means to deeply study the excited
state dynamics to understand the complexes’ photoactivity.
Enhanced photocatalysis of H2 production from H2O is
observed for [{(TL)2Ru(dpp)}2Ru(dpq)PtCl2](PF6)6 in com-
parison to the BL′ = dpp analogue, suggesting that the
population of the 3CS excited state is important in this function.
Both complexes with Ph2phen provide slightly enhanced
photocatalysis compared to their phen analogues in stark
contrast to the dramatically enhanced photocatalysis by
[{(Ph2phen)2Ru(dpp)}2RhBr2](PF6)5 relative to the phen
analogue. Evidence supports the functioning of the tetrametallic
architecture as the active photocatalyst. Active supramolecular
photocatalysis is supported by the Hg test, the large impact of
the identity of the BL′ on photocatalyst activity, and the lack of
functioning of a [{(TL)2Ru(dpp)}2Ru(BL′)](PF6)6 (1−4b)
and cis-[Pt(DMSO)2Cl2] system as well as spectroscopic
differences of the supramolecular photocatalyst versus the
trimetallic and Pt system. The in depth analysis presented
herein is significant as few analyses of Ru,Pt or Ru,Pd
supramolecular photocatalysts for H2O reduction are reported.
Further studies into the factors impacting photocatalytic H2
production from H2O using these complexes are ongoing.
DMA = N,N-dimethylaniline
dmgBF2 = (difluoroboryl)dimethylglyoximate
dpb = 2,3-bis(2-pyridyl)benzoquinoxaline
dpp = 2,3-bis(2-pyridyl)pyrazine
dpq = 2,3-bis(2-pyridyl)quinoxaline
ED = sacrificial electron donor
EDTA = ethylenediaminetetraacetic acid
HOMO = highest occupied molecular orbital
ket = rate constant for electron transfer
kic = rate constant for internal conversion
kisc = rate constant for intersystem crossing
knr = rate constant for nonradiative decay
kr = rate constant for radiative decay
LA = light absorber
L-pyr = (4-pyridine)oxazolo-[4,5-f ]phenanthroline
LUMO = lowest unoccupied molecular orbital
MLCT = metal-to-ligand charge transfer
MMCT = metal-to-metal charge transfer
MV2+ = methyl viologen
NAD+ = nicotinamide adenine dinucleotide
NHE = normal hydrogen electrode
pbn = 2-(2-pyridyl)benzo[b]-1,5-naphthyridine
PEC = photoinitiated electron collection
Ph2phen = 4,7-diphenyl-1,10-phenanthroline
phen = 1,10-phenanthroline
PMD = photochemical molecular device
tatpp = 9,11,20,22-tetraazatetrapyrido[3,2-a:2′3′-c:3″,2″-
1:2‴,3‴-n]pentacene
tatpq = 9,11,20,22-tetraazatetrapyrido[3,2-a:2′3′-c:3″,2″-
1:2‴,3‴-n]pentacene-10,21-quinone
tBu2bpy = 4,4′-di-tert-butyl-2,2′-bipyridine
TON = turnover number
tpphz = tetrapyrido[3,2-a:2′,3′-c:3″,2″-h:2‴,3‴-j]phenazine
ε = molar absorptivity
τ = lifetime
ASSOCIATED CONTENT
* Supporting Information
ESI-MS data, steady-state emission spectra, time-resolved
emission profiles, and Stern−Volmer emission quenching
experiments. This material is available free of charge via the
Internet at http://pubs.acs.org.
Φ = quantum yield
■
S
REFERENCES
■
(1) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U. S. A. 2006,
103, 15729−15735.
(2) Lewis, N. S. ChemSusChem 2009, 2, 383−386.
(3) Lubitz, W.; Tumas, W. Chem. Rev. 2007, 107, 3900−3903.
(4) Tinker, L. L.; McDaniel, N. D.; Bernhard, S. J. Mater. Chem.
2009, 19, 3328−3337.
AUTHOR INFORMATION
Corresponding Author
*E-mail: kbrewer@vt.edu.
Notes
■
(5) Bard, A. J.; Fox, M. A. Acc. Chem. Res. 1995, 28, 141−145.
(6) Esswein, A. J.; Nocera, D. G. Chem. Rev. 2007, 107, 4022−4047.
(7) Inagaki, A.; Akita, M. Coord. Chem. Rev. 2010, 254, 1220−1239.
(8) Lehn, J.-M.; Sauvage, J.-P. Nouv. J. Chim. 1977, 1, 449−451.
The authors declare no competing financial interest.
(9) Kalyanasundaram, K.; Kiwi, J.; Gratzel, M. Helv. Chim. Acta 1978,
̈
61, 2720−2730.
ACKNOWLEDGMENTS
■
(10) Kirch, M.; Lehn, J.-M.; Sauvage, J.-P. Helv. Chim. Acta 1979, 62,
1345−1384.
Acknowledgment is made to Hannah Mallalieu for assistance
with this manuscript and to the Chemical Sciences, Geo-
sciences and Biosciences Division, Office of Basic Energy
Sciences, Office of Sciences, U.S. Department of Energy (DE
FG02-05ER15751) for their generous support of the develop-
ment of new LAs used in this research.
(11) Brown, G. M.; Chan, S. F.; Creutz, C.; Schwarz, H. A.; Sutin, N.
J. Am. Chem. Soc. 1979, 101, 7638−7640.
(12) Chan, S.-F.; Chou, M.; Creutz, C.; Matsubara, T.; Sutin, N. J.
Am. Chem. Soc. 1981, 103, 369−379.
(13) Balzani, V.; Moggi, L.; Scandola, F. In Supramolecular
Photochemistry; Balzani, V., Ed.; NATO Advanced Study Institute
Series 214; Reidel: Dordrecht, The Netherlands, 1987; p 1, and
references cited therein.
ABBREVIATIONS
BL′ = bridging ligand between Ru(II) and Pt(II)
bpm = 2,2′-bipyrimidine
■
(14) Wang, M.; Na, Y.; Gorlov, M.; Sun, L. Dalton Trans. 2009,
6458−6467.
9759
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760

Inorganic Chemistry
Article
(15) Durham, B.; Caspar, J. V.; Nagle, J. K.; Meyer, T. J. J. Am. Chem.
Soc. 1982, 104, 4803−4810.
(51) Swavey, S.; Fang, Z.; Brewer, K. J. Inorg. Chem. 2002, 41, 2598−
2607.
(52) Higgins, S. L. H.; White, T. A.; Winkel, B. S. J.; Brewer, K. J.
(16) Kalyanasundaram, K. Coord. Chem. Rev. 1982, 46, 159−244.
(17) Wallace, A. W.; Murphy, W. R., Jr.; Petersen, J. D. Inorg. Chim.
Acta 1989, 166, 47−54.
Inorg. Chem. 2010, 50, 463−470.
(53) Sahai, R.; Rillema, D. P. J. Chem. Soc., Chem. Commun. 1986,
1133−1134.
(18) Berger, R. M. Inorg. Chem. 1990, 29, 1920−1924.
(19) Arachchige, S. M.; Brown, J. R.; Chang, E.; Jain, A.; Zigler, D. F.;
Rangan, K.; Brewer, K. J. Inorg. Chem. 2009, 48, 1989−2000.
(20) Elvington, M.; Brown, J.; Arachchige, S. M.; Brewer, K. J. J. Am.
Chem. Soc. 2007, 129, 10644−10645.
(54) Sommovigo, M.; Denti, G.; Serroni, S.; Campagna, S.;
Mingazzini, C.; Mariotti, C.; Juris, A. Inorg. Chem. 2001, 40, 3318−
3323.
(55) Miao, R.; Mongelli, M. T.; Zigler, D. F.; Winkel, B. S. J.; Brewer,
K. J. Inorg. Chem. 2006, 45, 10413−10415.
(56) Knoll, J. D.; Arachchige, S. M.; Wang, G.; Rangan, K.; Miao, R.;
Higgins, S. L. H.; Okyere, B.; Zhao, M.; Croasdale, P.; Magruder, K.;
Sinclair, B.; Wall, C.; Brewer, K. J. Inorg. Chem. 2011, 50, 8850−8860.
(57) Basile, L. A.; Barton, J. K. J. Am. Chem. Soc. 1987, 109, 7548−
7550.
(21) Rangan, K.; Arachchige, S. M.; Brown, J.; Brewer, K. J. Energy
Environ. Sci. 2009, 2, 410−419.
(22) White, T. A.; Higgins, S. L. H.; Arachchige, S. M.; Brewer, K. J.
Angew. Chem., Int. Ed. 2011, 50, 12209−12213.
(23) White, T. A.; Whitaker, B. N.; Brewer, K. J. J. Am. Chem. Soc.
2011, 133, 15332−15334.
(58) Mongelli, M. T.; Brewer, K. J. Inorg. Chem. Commun. 2006, 9,
877−881.
(24) Miyake, Y.; Nakajima, K.; Sasaki, K.; Saito, R.; Nakanishi, H.;
Nishibayashi, Y. Organometallics 2009, 28, 5240−5243.
(25) Li, X.; Wang, M.; Zhang, S.; Pan, J.; Na, Y.; Liu, J.; Akermark, B.;
Sun, L. J. Phys. Chem. B 2008, 112, 8198−8202.
(59) Price, J. H.; Williamson, A. N.; Schramm, R. F.; Wayland, B. B.
Inorg. Chem. 1972, 11, 1280−1284.
(60) Goodwin, H. A.; Lions, F. J. Am. Chem. Soc. 1959, 81, 6415−
6422.
(26) Fihri, A.; Artero, V.; Razavet, M.; Baffert, C.; Leibl, W.;
Fontecave, M. Angew. Chem., Int. Ed. 2008, 47, 564−567.
(27) Li, C.; Wang, M.; Pan, J.; Zhang, P.; Zhang, R.; Sun, L. J.
Organomet. Chem. 2009, 694, 2814−2819.
(61) Caspar, J. V.; Kober, E. M.; Sullivan, B. P.; Meyer, T. J. J. Am.
Chem. Soc. 1982, 104, 630−632.
(62) Brown, J. R.; Elvington, M.; Mongelli, M. T.; Zigler, D. F.;
Brewer, K. J. Proc. SPIE - Optics and Photonics Solar Hydrogen
Nanotechnology; Lionel, V., Ed.; SPIE: Bellingham, WA, 2006; Vol.
6340, p 634017.
(63) Arachchige, S. M.; Shaw, R.; White, T. A.; Shenoy, V.; Tsui, H.-
M.; Brewer, K. J. ChemSusChem 2011, 4, 514−518.
(64) MacDonnell, F. M.; Bodige, S. Inorg. Chem. 1996, 35, 5758−
5759.
(65) Crosby, G. A.; Watts, R. J. J. Am. Chem. Soc. 1971, 93, 3184−
3188.
(66) Whitesides, G. M.; Hackett, M.; Brainard, R. L.; Lavalleye, J. P.
P. M.; Sowinski, A. F.; Izumi, A. N.; Moore, S. S.; Brown, D. W.;
Staudt, E. M. Organometallics 1985, 4, 1819−1830.
(28) Sakai, K.; Ozawa, H. Coord. Chem. Rev. 2007, 251, 2753−2766.
(29) Ozawa, H.; Sakai, K. Chem. Commun. 2011, 47, 2227−2242.
(30) Ozawa, H.; Haga, M.-a.; Sakai, K. J. Am. Chem. Soc. 2006, 128,
4926−4927.
(31) Masaoka, S.; Mukawa, Y.; Sakai, K. Dalton Trans. 2010, 39,
5868−5876.
(32) Knoll, J. D.; Arachchige, S. M.; Brewer, K. J. ChemSusChem
2011, 4, 252−261.
(33) Hirahara, M.; Masaoka, S.; Sakai, K. Dalton Trans. 2011, 40,
3967−3978.
(34) Singh Bindra, G.; Schulz, M.; Paul, A.; Soman, S.; Groarke, R.;
Inglis, J.; Pryce, M. T.; Browne, W. R.; Rau, S.; Maclean, B. J.; Vos, J.
G. Dalton Trans. 2011, 40, 10812−10814.
(35) Karnahl, M.; Kuhnt, C.; Ma, F.; Yartsev, A.; Schmitt, M.;
Dietzek, B.; Rau, S.; Popp, J. ChemPhysChem 2011, 12, 2101−2109.
(36) Lei, P.; Hedlund, M.; Lomoth, R.; Rensmo, H.; Johansson, O.;
Hammarstrom, L. J. Am. Chem. Soc. 2008, 130, 26−27.
̈
(37) Ozawa, H.; Sakai, K. Chem. Lett. 2007, 36.
(38) Ozawa, H.; Kobayashi, M.; Balan, B.; Masaoka, S.; Sakai, K.
Chem.Asian J. 2010, 5, 1860−1869.
(39) Rau, S.; Schafer, B.; Gleich, D.; Anders, E.; Rudolph, M.;
̈
Friedrich, M.; Gorls, H.; Henry, W.; Vos, J. G. Angew. Chem., Int. Ed.
̈
2006, 45, 6215−6218.
(40) Du, P.; Schneider, J.; Li, F.; Zhao, W.; Patel, U.; Castellano, F.
N.; Eisenberg, R. J. Am. Chem. Soc. 2008, 130, 5056−5058.
(41) Molnar, S. M.; Nallas, G. N. A.; Bridgewater, J. S.; Brewer, K. J.
J. Am. Chem. Soc. 1994, 116, 5206−5210.
(42) Kim, M.-J.; Konduri, R.; Ye, H.; MacDonnell, F. M.; Puntoriero,
F.; Serroni, S.; Campagna, S.; Holder, T.; Kinsel, G.; Rajeshwar, K.
Inorg. Chem. 2002, 41, 2471−2476.
(43) Konduri, R.; Ye, H.; MacDonnell, F. M.; Serroni, S.; Campagna,
S.; Rajeshwar, K. Angew. Chem., Int. Ed. 2002, 41, 3185−3187.
(44) Fukushima, T.; Fujita, E.; Muckerman, J. T.; Polyansky, D. E.;
Wada, T.; Tanaka, K. Inorg. Chem. 2009, 48, 11510−11512.
(45) Polyansky, D. E.; Cabelli, D.; Muckerman, J. T.; Fukushima, T.;
Tanaka, K.; Fujita, E. Inorg. Chem. 2008, 47, 3958−3968.
(46) Elvington, M.; Brewer, K. J. Inorg. Chem. 2006, 45, 5242−5244.
(47) Sahai, R.; Rillema, D. P. Inorg. Chim. Acta 1986, 118, L35−L37.
(48) Yam, V. W.-W.; Lee, V. W.-M.; Cheung, K.-K. J. Chem. Soc.,
Chem. Commun. 1994, 2075−2076.
(49) Milkevitch, M.; Shirley, B. W.; Brewer, K. J. Inorg. Chim. Acta
1997, 264, 249−256.
(50) Milkevitch, M.; Storrie, H.; Brauns, E.; Brewer, K. J.; Shirley, B.
W. Inorg. Chem. 1997, 36, 4534−4538.
9760
dx.doi.org/10.1021/ic4004406 | Inorg. Chem. 2013, 52, 9749−9760