Catalytic Light-Driven Generation of Hydrogen from Water by Iron Dithiolene Complexes

Article abstract of DOI:10.1021/jacs.6b05040

The development of active, robust systems for light-driven hydrogen production from aqueous protons based on catalysts and light absorbers composed solely of earth abundant elements remains a challenge in the development of an artificial photosynthetic system for water splitting. Herein, we report the synthesis and characterization of four closely related Fe bis(benzenedithiolate) complexes that exhibit catalytic activity for hydrogen evolution when employed in systems with water-soluble CdSe QDs as photosensitizer and ascorbic acid as a sacrificial electron source under visible light irradiation (520 nm). The complex with the most electron-donating dithiolene ligand exhibits the highest activity, the overall order of activity correlating with the reduction potential of the formally Fe(III) dimeric dianions. Detailed studies of the effect of different capping agents and the extent of surface coverage of these capping agents on the CdSe QD surfaces reveal that they affect system activity and provide insight into the continued development of such systems containing QD light absorbers and molecular catalysts for H₂ formation.

Full text of DOI:10.1021/jacs.6b05040

Article
pubs.acs.org/JACS
Catalytic Light-Driven Generation of Hydrogen from Water by Iron
Dithiolene Complexes
Hongjin Lv, T. Purnima A. Ruberu, Valerie E. Fleischauer, William W. Brennessel, Michael L. Neidig,
and Richard Eisenberg*
Department of Chemistry, University of Rochester, Rochester, New York 14627, United States
S
* Supporting Information
ABSTRACT: The development of active, robust systems for light-driven
hydrogen production from aqueous protons based on catalysts and light
absorbers composed solely of earth abundant elements remains a challenge in
the development of an artificial photosynthetic system for water splitting.
Herein, we report the synthesis and characterization of four closely related Fe
bis(benzenedithiolate) complexes that exhibit catalytic activity for hydrogen
evolution when employed in systems with water-soluble CdSe QDs as
photosensitizer and ascorbic acid as a sacrificial electron source under visible
light irradiation (520 nm). The complex with the most electron-donating
dithiolene ligand exhibits the highest activity, the overall order of activity
correlating with the reduction potential of the formally Fe(III) dimeric dianions. Detailed studies of the effect of different capping
agents and the extent of surface coverage of these capping agents on the CdSe QD surfaces reveal that they affect system activity
and provide insight into the continued development of such systems containing QD light absorbers and molecular catalysts for
H2 formation.
of new, efficient earth-abundant transition metal catalysts and
photostable light-absorbers remains part of that challenge.
Significant progress has been made locally and elsewhere
within the past few years regarding both the design of novel
catalysts and photostable light-absorbers.22,32−34 In these studies,
water-soluble cadmium selenide (CdSe) semiconductor quan-
tum dots (QDs) have been used as an alternative chromophore
for visible-light-driven hydrogen production.35−41 Compared
INTRODUCTION
■
Visible light-driven water splitting into H2 and O2 via artificial
photosynthesis (AP) is a central theme of solar fuel production
that will prove critically important in light of projected increases
in global energy demand and the need to minimize carbon
emissions with regard to global climate change.1−5 In light of the
intrinsic complexity of water splitting (a four-electron, four-
proton redox reaction), AP systems for this energy-storing
reaction are generally studied in terms of two half-reactions:
water oxidation to O2 and aqueous proton reduction to H2. The
light-driven reduction of water to H2 by molecularly based
systems has been known since the late 1970s when researchers
developed systems comprising [Ru(bpy)3]2+ and/or its deriva-
tives as light absorbers in the presence of a suitable catalyst such
as Pt, an electron mediator, and a sacrificial electron.6−9 In recent
years, extensive efforts have been made to replace precious metal
components of aqueous proton reduction systems with more
active and durable constituents made solely of earth-abundant
elements. For possible catalysts in homogeneous water reduction
systems, the focus has been on potentially redox active transition
metal complexes of elements such as iron,10−13 cobalt,14−19
nickel,20−26 and molybdenum.27,28 Despite extensive exper-
imental investigations and mechanistic studies of systems
containing these water reduction catalysts (WRCs), many of
them have been found to suffer from low activity and/or
photoinstability due to decomposition of the chromophore, or
catalyst degradation due to structural instability.29−31 While
significant improvements have been reported,16,18−20,22−25 the
challenge still remains for an active and robust sunlight-driven
system for hydrogen production from water. The development
2+
with molecular light absorbers such as Ru(bpy)3 and the
organic dyes Fluorescein and Eosin Y, semiconductor QDs have
advantages of superior photostability, tunable redox potentials,
large absorption cross sections across the visible spectrum, and
the propensity for photoinduced electron transfer for cataly-
sis.35,36,39,42
With respect to the development of novel and stable catalysts,
one interesting family of WRCs are metal dithiolene
complexes,16,32,43 in which the dithiolene ligands can undergo
reversible electron transfers with minimal structural perturba-
tions and are resistant toward hydrogenation.44,45 These
attractive properties make them proper candidates for multi-
electron-transfer catalysis as in proton reduction to H2. To date,
the Co-based bis(benzenedithiolate) (bdt) complexes and
related Ni systems have demonstrated promising activity toward
both light-driven and electrocatalytic hydrogen produc-
tion,16,32,33,43 but Fe-based dithiolene complexes as catalysts
for hydrogen evolution have been largely unexplored.46,47 Given
the fact that Fe2S2 or FeNiS2 clusters are the active sites of [FeFe]
Received: May 16, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
CdSe QDs were performed in water. Elemental analyses were performed
using a PerkinElmer 2400 Series II Analyzer. Cyclic Voltammetry (CV)
measurements were performed with a CHI 680D potentiostat at room
temperature under Ar atmosphere, using a one-compartment cell with a
glassy carbon working electrode (3 mm diameter), Pt auxiliary electrode,
and saturated calomel reference electrode (SCE). Tetrabutylammonium
hexafulorophosphate (TBAPF6) was used as supporting electrolyte for
all CV measurements.
Synthesis of (NBu4)2[Fe(bdt)2]2 (1). In a Schlenk flask, 1,2-
benzenedithiol (237.1 mg, 1.67 mmol) and NaOMe (185 mg, 3.42
mmol) were dissolved in 25 mL of dry MeOH under an N2 atmosphere
and allowed to stir at room temperature for 10 min. To this solution, a
degassed mixture of Fe(NO3)3·9H2O (343 mg, 0.85 mmol) in MeOH
(5 mL) was added dropwise. The solution was stirred for 1 h at room
temperature while the color became black-red. To this solution, a
solution of NBu4Br (275.2 mg, 0.85 mmol) in 5 mL of MeOH was
added, then the solution was allowed to stir at room temperature
overnight under an air atmosphere. After reaction, the solvent was
completely removed under vacuum, and a black-red precipitate formed.
The black-red precipitate was extracted using tetrahydrofuran (3 × 15
mL); the combined organic layer was filtered and dried under rotary
evaporator to obtain a pure black-red solid (330 mg, 67% yield). Crystals
for X-ray diffraction were grown by diffusion of diethyl ether to a
concentrated solution of (NBu4)2[Fe(bdt)2]2 in dichloromethane.
Elemental analyses: Calcd: C, 58.11; H, 7.66; N, 2.42; Found: C,
58.34; H, 7.93; N, 2.57.
or [FeNi] hydrogenases (the most efficient catalysts for
hydrogen uptake and proton reduction),48−51 a number of
recent studies have been reported that use semiconductor
nanoparticles as light-absorbers and [FeFe] hydrogenase or their
mimics as the catalysts. For example, Dukovic, King and co-
workers reported systems using CdS-MPA nanorods (MPA =
the capping agent 3-mercaptopropionic acid) and [FeFe]-
hydrogenase I that promote light-driven H2 production with a
turnover frequency (TOF) of 380−900 s−1 upon irradiation at
405 nm.37 In another study, Wu and co-workers have described
the assembly of an [FeFe]-hydrogenase mimic with CdSe QDs
using an interface-directed approach, with resultant activity in the
photogeneration of H2 of 8781 turnovers (TONs) and an initial
turnover frequency (TOF) of 596 h−1 under visible light
irradiation.40b On the basis of these observations and work done
locally on the light-driven generation H2 using Co and Ni
complexes of redox active unsaturated ligands as catalysts,16,32,33
related iron complexes containing 1,2-dithiolene ligands
appeared to be potentially interesting candidates as cost-effective
catalysts for hydrogen production. An important clue to potential
activity was provided nearly 25 years ago when Sellmann and co-
workers reported that the complex [Fe(II)(bdt)2]2− catalyzed
proton reduction to H2 but the process was stoichiometric and
not shown to be catalytic.46
Synthesis of (NBu4)2[Fe(tdt)2]2 (2). This complex was synthesized in a
procedure analogous to that described for (NBu4)2[Fe(bdt)2]2 except
that toluene-3,4-dithiol (261 mg, 1.67 mmol) was used. The yield is 366
mg (71%). Crystals for X-ray diffraction were grown by diffusion of
diethyl ether to a concentrated solution of (NBu4)2[Fe(tdt)2]2 in
dichloromethane. Elemental analyses: Calcd: C, 59.37; H, 7.97; N, 2.31;
Found: C, 58.84; H, 7.93; N, 2.12.
In this context, we report here the synthesis and character-
ization of four closely related Fe dithiolene complexes and their
catalytic hydrogen evolution activity when using water-soluble
CdSe QDs as photosensitizer and ascorbic acid as the sacrificial
electron donor. Detailed studies on the influence of different
capping agents and the extent of surface coverage of CdSe QDs
by these capping agents also yield insight into the continuing
development of highly robust and superior efficient catalytic
systems for solar energy utilization.
Synthesis of (NBu4)2[Fe(dcbdt)2]2 (3). This complex was synthesized
in a procedure analogous to that described for (NBu4)2[Fe(bdt)2]2
except that 3,6-dichloro-1,2-benzenedithiol (352.5 mg, 1.67 mmol) was
used. The yield is 396 mg (65%). Crystals for X-ray diffraction were
grown by diffusion of diethyl ether to a concentrated solution of
(NBu4)2[Fe(bdt-Cl2)2]2 in dichloromethane. Elemental analyses Calcd:
C, 46.93; H, 5.62; N, 1.95; Found: C, 46.28; H, 5.38; N, 2.02.
Synthesis of 3,4,5,6-Tetrachlorobenzene-1,2-dithiol (tcbdt).
3,4,5,6-Tetrachloro-benzene-1,2-dithiol (tcbdt) was prepared following
a slightly modified literature method.55 In a typical experiment, a
mixture of hexachlorobenzene (25 g), iron powder (2.57 g), NaHS (19
g), and sulfur (0.22 g) in 90 mL dimethylformamide (DMF) was
refluxed for 14 h under N2 atmosphere. After reaction, the mixture was
cooled to room temperature and diluted with NaOH solution (0.375 M,
100 mL). The black crude tetrachlorobenzenedithiolate iron complex
was obtained under vacuum suction and dried at room temperature
overnight. Caution! Hexachlorobenzene is considered to be a probable
human carcinogen and dangerous for the environment; make sure this
chemical is handled with personal protective equipment! Then, the crude
tetrachlorobenzenedithiol-iron complex (∼32 g) was dispersed in 180
mL warm methanol, and to this mixture was added a solution of zinc
oxide (10.5 g) and NaOH (27 g) in 200 mL water at 50 °C. The
resulting mixture was refluxed at around 75 °C for 1.5 h; the solution was
filtered and the filtrate was added to a solution of concentrated sulfuric
acid (102 mL) in 350 mL water. After cooling to room temperature, the
yellow precipitate was filtered off, washed with deionized water for three
times and dried under vacuum overnight. The yield of this pale yellow
3,4,5,6-tetrachlorobenzene-1,2-dithiol was 4.55 g (18.5%).
EXPERIMENTAL SECTION
■
Materials. All starting chemicals and solvents for syntheses,
characterization, and catalytic studies were purchased from commercial
sources and were used as received without further purification unless
otherwise noted. The solvents for syntheses were dried using 4 Å
molecular sieves before use. For photocatalytic hydrogen evolution
experiments, deionized water and ethanol (1/1 volume ratio) was used.
Two water-solubilizing capping agents, dihydrolopic acid (DHLA) and
3-mercapto-2,2-bis(mercaptomethyl)propanoic acid (S3), were synthe-
sized following literature methods.32,52 Other water-solubilizing capping
agents, 3-mercaptopropionic acid (MPA) and 11-mercaptoundecanoic
acid (MUA), were purchased from Aldrich and used as received. The tri-
n-octylphosphine oxide-capped CdSe (CdSe-TOPO) QDs were
prepared using slightly modified procedures to that reported
previously,53 and the water-soluble CdSe nanocrystals (CdSe-DHLA,
CdSe-S3, CdSe-MPA, and CdSe-MUA) were synthesized through
ligand exchange reactions in methanol under N2 atmosphere (see
experimental details in the SI).22,32,39,41
Instrumentation. The 1H NMR spectra were recorded on a Bruker
Avance 400-MHz spectrometer and referenced to residual proton
resonances of the deuterated solvents. The magnetic susceptibility
measurements for all four Fe complexes were performed by the Evans
Method54 using an NMR tube with a coaxial insert. The NMR tube
contains a THF-d8 solution of the Fe-complex with 2% reference
substance (t-butanol) and the coaxial insert contains only the THF-d8
solvent and reference substance. The difference in chemical shift of the
reference compound caused by the paramagnetic compounds was
recorded and used for effective magnetic moment calculation. UV−vis
spectra were recorded on a Cary 60 UV/vis spectrophotometer, using a
1 cm or 2 mm path length quartz cuvette. The UV−vis spectra of all Fe-
dithiolene complexes were recorded in CH3CN solution; the spectra of
CdSe-TOPO QDs were measured in hexane and those of water-soluble
Synthesis of (NBu4)2[Fe(tcbdt)2]2 (4). This complex was synthesized
in a procedure analogous to that described for (NBu4)2[Fe(bdt)2]2
except that 3,4,5,6-tetrachlorobenzene-1,2-dithiol (280 mg, 1 mmol),
NaOMe (108 mg, 2 mmol), Fe(NO3)3·9H2O (202 mg, 0.5 mmol), and
NBu4Br (161 mg, 0.5 mmol) were used. The yield is 316 mg (74%).
Crystals for X-ray diffraction were grown by diffusion of hexane to a
concentrated solution of (NBu4)2[Fe(tcdt)2]2 in dichloromethane.
Elemental analyses Calcd: C, 39.37; H, 4.25; N, 1.63; Found: C, 40.39;
H, 4.29; N, 1.57.
B
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
X-ray Crystallography. Crystals were placed onto the tips of glass
capillary tubes or fibers and mounted on a Bruker SMART CCD
platform diffractometer for data collection. For each crystal a
preliminary set of cell constants and an orientation matrix were
calculated from reflections harvested from three orthogonal wedges of
reciprocal space. Full data collections were carried out using Mo Kα
radiation (0.71073 Å, graphite monochromator) with frame times
ranging from 25 to 60 s and at a detector distance of approximately 4 cm.
Randomly oriented regions of reciprocal space were surveyed: three to
six major sections of frames were collected with 0.50° steps in ω at three
to six different φ settings and a detector position of −38° in 2θ. The
intensity data were corrected for absorption.56 Final cell constants were
calculated from the xyz centroids of about 4000 strong reflections from
the actual data collections after integration.57
The formation of A and H2 from ascorbic acid has been determined to be
thermodynamically unfavorable by 0.41 V at pH 4.8
RESULTS AND DISCUSSION
■
Synthesis and Structural Studies. Complexes 1−4 were
synthesized by reacting the iron precursor (Fe(NO3)3·9H2O)
with the corresponding dithiolate ligands in methanol under an
N2 atmosphere. For the synthesis of complex 4, the 3,4,5,6-
tetrachlorobenzene-1,2-dithiol ligand was prepared separately as
described in the Experimental Section. All four complexes were
isolated as anions with tetrabutylammonium as the countercation
in relatively high yield (65−74%) and they are stable in the solid
state under ambient conditions. Their molecular structures are
shown in Scheme 1. The purities of complexes 1−4 have been
confirmed by elemental analyses which match well with
theoretically calculated values (see Experimental Section).
Structures were solved using SIR9758 and refined using SHELXL-
97.59 Space groups were determined on the basis of systematic absences,
intensity statistics, or both. Direct-methods solutions were calculated,
which provided most non-hydrogen atoms from the E map. Full-matrix
least-squares/difference Fourier cycles were performed which located
the remaining non-hydrogen atoms. All non-hydrogen atoms were
refined with anisotropic displacement parameters. All hydrogen atoms
were placed in ideal positions and refined as riding atoms with relative
isotropic displacement parameters. Full-matrix least-squares refine-
ments on F2 were run to convergence.
Scheme 1. Molecular Structures of Various Fe-Dithiolene
Catalysts
Mossbauer Spectroscopy. All solid samples for 57Fe Mossbauer
spectroscopy were run on nonenriched samples of the as-isolated
complexes. Each sample was prepared under ambient conditions and
̈
̈
loaded into a Delrin Mossbauer sample cup for measurements. Low
̈
temperature 57Fe Mossbauer measurements were performed using a See
̈
Co. MS4Mossbauer spectrometer integrated with a Janis SVT-400T
He/N2 cryostat for measurements at 80 K. Isomer shifts were
̈
determined relative to α-Fe at 298 K. All Mossbauer spectra were fit
̈
using the program WMoss (SeeCo).60
Hydrogen Evolution Studies. Photocatalytic hydrogen evolution
experiments were performed in a custom-built 16-well apparatus.
Samples in EtOH/H2O (1/1) containing varying amounts of Fe
dithiolene catalysts, water-soluble CdSe QDs with different capping
agents and ascorbic acid (AA) sacrificial electron donors were prepared
in 40 mL scintillation vials and protected from light before use. Total
volume of the solution in each sample was 5 mL, the pH of the reaction
solution was adjusted to 4.5 by the addition of NaOH or HCl. Control
experiments were carried out under similar conditions in the absence of
each component (e.g., CdSe QDs, AA, or catalysts). The samples were
placed into a temperature-controlled block at 15 °C and sealed with an
airtight cap fitted with a pressure transducer and a rubber septum. The
samples were then purged with a mixture of gas containing 4:1 N2/CH4
with methane serving as an internal reference for GC analysis at the end
of the experiment. The samples were irradiated from below the vials with
high-power Philips LumiLED Luxeon Star Hex green (520 nm) 700-mA
LEDs. The light power of each LED was set to 150 mW and measured
with an L30 A Thermal sensor and a Nova II power meter (Ophir-
Spiricon). The samples were swirled using an orbital shaker at 110 rpm.
The pressure changes in the vials were recorded using a Labview
program from a Freescale semiconductor sensor (MPX4259A series).
At the end of the experiment, the headspace of the vials was
characterized by gas chromatography to ensure that the measured
pressure change was a consequence of hydrogen generation and to
confirm that the amount of hydrogen calculated by the change in
pressure corresponded to the amount determined by GC analysis. The
amounts of hydrogen evolved were determined using a Shimadzu GC-
17A gas chromatograph with a 5-Å molecular sieve column (30 m, 0.53
mm) and a thermal conductivity detector, and were quantified by a
calibration plot relative to the internal CH4 standard.
The solid-state molecular structures of complexes 1−4 were
determined by single-crystal X-ray diffraction (see Figure 1). The
detailed crystallographical information is listed in Tables S1−S4.
As shown in Figure 1, all four formally Fe(III) bis dithiolene
complexes exist as dianionic dimers and exhibit similar structural
characteristics. The Fe2S2 rhombus in each dimer is located on a
crystallographic center of symmetry with each iron center having
a tetragonal-pyramidal coordination geometry. Two dithiolate
ligands occupy the basal sites, while the axial site contains a
bridging dithiolate S ligand that serves as a basal ligand for the
centrosymmetrically related Fe atom in the dimer. For all four
complexes, the axial Fe−S bonds holding the two Fe-
−
(dithiolene)2 moiteies together are longer than the basal Fe−
S bonds within the chelate rings (Table 1).44,61−63 The structural
results are thus entirely consistent with the notion previously that
in solution the dimers exist in equilibrium with the
corresponding monomers.62 Additionally, the S−Fe−S′ bond
angle in each Fe2S2 rhombus increases in going from complexes 1
and 2 to complexes 3 and 4, suggesting greater steric strain in the
latter caused by the chloride substituents. Another feature of the
structure of 1 that is recognized in the Mossbauer spectrum of
̈
the sample discussed below is that 1 crystallizes in the triclinic
space group P1-bar with two crystallographically independent
half-dimers in the asymmetric unit. The two dimers are
structurally similar except for the Fe···Fe separations which are
very significantly different (3.2066(6) Å in 1a and 3.0398(6) Å in
1b). In the other structures as well, the Fe···Fe separations exhibit
a significant variance being 3.2126(14), 2.9623(7) and
3.0424(11) Å in 2, 3 and 4, respectively.
Systems with Ascorbic Acid Sacrificial Donors. The net reaction
for photochemical hydrogen generation using ascorbic acid as the
sacrificial donor can be expressed by the following equation in which
H2A is ascorbic acid and A is dehydroascorbic acid:
Spectroscopic Characterization and Redox Chemistry
of Fe-Dithiolene Complexes. The UV−vis absorption spectra
hv
H2A → A + H2
C
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Structures of complexes 1−4. Complex 1 has two crystallographically independent dimeric dianions which are shown separately as 1a and 1b.
+
Dashed lines between Fe atoms indicate distances shorter than 3.05 Å. The counter cations (NBu4 ) are omitted for simplicity.
of complexes 1−4 were determined in acetonitrile at room
temperature. Complexes 1−4 all show intense high-energy bands
in the UV region, corresponding to spin-allowed intraligand
(π−π*) transitions.24 As shown in Figure 2, all four complexes
exhibit broad low-energy bands in the 450−650 nm region,
which can be described as charge transfer transitions based on
their relative intensities.45,64 The absorption envelopes of the
charge transfer bands indicate at least three transitions for each
complex in the 450−650 nm region. The two highest energy
bands for 1−3 are similar in value, but differ in relative intensity
(1: 490, 520 nm; 2: 490, 525 nm; 3: 492, 522 nm), whereas for
complex 4 the observed highest energy transition is red-shifted to
522 nm with a shoulder at 562 nm. In view of the existence of
multiple charge transfer transitions for each of the complexes,
and the fact that they exist in monomer−dimer equilibria in fluid
solution, attention turned to other methods of definitive
characterization of 1−4.
Mossbauer measurements on solid samples of complexes 1−4
̈
were performed at 80 K (see details in Experimental Section).
The spectra of all four complexes are clean with insignificant
impurity peaks, as shown in Figure 3. Complexes 2−4 exhibit
only a single quadrupole doublet, whereas complex 1 shows two
quadrupole doublets in an approximate ratio of 1:1. The latter
observation is attributed to the two crystallographically
independent dinuclear dianions in the asymmetric unit as
determined by X-ray diffraction (Table 1, Figure S1). For 1,
the two independent dianions have significantly different Fe···Fe
separations and therefore exhibit slightly different quadrupole
splitting parameters (δ = 0.34 mm/s, ΔEQ = 2.62 mm/s vs δ =
0.34 mm/s, ΔEQ = 3.05 mm/s). Complexes 2−4 possess similar
D
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Key Bond Distances and Angles in Complexes 1−4
bond distances (Å)
complex 1
complex 2
complex 3
complex 4
Fe(1)−S(1)
Fe(1)−S(2)
Fe(1)−S(3)
Fe(1)−S(4)
Fe(1)−S(4)#1
S(4)−Fe(1)#1
Fe(1)−Fe(1)#1
2.2350(6)
2.2250(6)
2.2289(6)
2.2365(6)
2.4945(6)
2.4945(6)
3.2066(6)
2.2301(6)
2.2242(6)
2.2348(6)
2.2317(6)
2.4247(6)
2.4247(6)
3.0398(6)
2.2315(14)
2.2155(16)
2.2241(15)
2.2312(15)
2.4813(13)
2.4812(13)
3.2126(14)
bond angles (deg)
2.2207(7)
2.2209(7)
2.2323(7)
2.2292(7)
2.4461(7)
2.4462(7)
2.9623(7)
2.2167(12)
2.2125(12)
2.2280(12)
2.2172(11)
2.4808(11)
2.4807(11)
3.0424(11)
complex 1
complex 2
complex 3
complex 4
S(2)−Fe(1)−S(1)
S(3)−Fe(1)−S(4)
S(4)−Fe(1)−S(4)#1
Fe(1)−S(4)−Fe(1)#1
88.84(2)
89.29(2)
88.68(6)
88.86(6)
94.21(5)
85.78(5)
88.68(3)
89.43(3)
101.52(2)
78.48(2)
89.00(5)
89.92(4)
99.49(4)
80.51(4)
89.65(2)
94.84(2)
85.15(2)
88.93(2)
98.61(2)
81.39(2)
dimeric compounds. Three of these complexes (1, 2, 4) have
previously been reported to have intermediate spin ferric ion (SFe
= 3/2). Since there are two identical Fe ions (SFe = 3/2) in each
dinuclear dianion complex, it is important to understand the total
spin state per dianion. Previous temperature-dependent
magnetic studies on solid-state samples were consistent with
the presence of intramolecular antiferromagnetic coupling
between the SFe = 3/2 ferric ions in each dianion, resulting in
an S = 0 diamagnetic ground state.44,63,65
In light of dimer dissociation into monomers in fluid solution,
the 1H NMR spectra of complexes 1−4 in deuterated
tetrahydrofuran (THF-d8) were examined. The complexes
were found to exhibit broad proton resonance signals, supporting
the notion of dissociation into paramagnetic monoanions in
solution. Magnetic susceptibility measurements for complexes
1−4 in solution were carried out using the Evans method.54 The
experimentally calculated effective magnetic moments for
complexes 1 and 2 are both 5.01 μB, consistent with four
unpaired electrons per dianion, whereas for complexes 3 and 4,
two sets of effective magnetic moment values are obtained (3: μeff
= 4.90 μB and 6.01 μB; 4: μeff = 5.03 μB and 5.82 μB). Since these
Figure 2. UV−vis absorption spectra of 0.1 mM complexes 1−4 in
acetonitrile.
Mossbauer parameters to those of 1 (2: δ = 0.32 mm/s, ΔE =
̈
Q
2.91 mm/s; 3: δ = 0.32 mm/s, ΔEQ = 2.82 mm/s; 4: δ = 0.32
mm/s, ΔEQ = 3.02 mm/s). These results are in agreement with
data in the literature44,62,63 on these and similar Fe dithiolene
Figure 3. 80 K Mossbauer spectra of solid complexes 1−4.
̈
E
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. (a) Cyclic voltammograms of 0.5 mM complex 2 at different scan rates in the range of 0 to −1.2 V; (b) the plot of maximum peak currents
versus the square root of scan rates; and (c) cyclic voltammograms of 0.5 mM complex 2 with varying concentrations of trifluoroacetic acid. Conditions:
0.5 mM complex 2, 0.1 M TBAPF6 electrolyte in acetonitrile, 1 M H2O, glassy carbon working electrode, Pt auxiliary electrode, SCE reference electrode.
results were obtained at a particular temperature and a single
concentration, they simply serve as a characterization of the
specific samples employed. The notion of a dimer−monomer
equilibrium is important in the solution chemistry of these
complexes61,62 and may well play a role in the catalytic behavior
discussed below.
shown below with thiolates coordinating to surface-bound Cd2+
and carboxylic acid groups significantly deprotonated at pH 4.5).
UV−vis Spectra of the CdSe QDs were recorded before and after
ligand exchange and are shown in Figure S6.
Cyclic voltammetric experiments were conducted in acetoni-
trile media for complexes 1−4 at room temperature under an Ar
atmosphere. Figure 4 shows the representative CV curves of 0.25
mM complex 2 at different scan rates. A reversible reduction peak
at −0.928 V vs SCE was observed, corresponding to the
[FeL2]22−/[FeL2]2 couple. Table S5 summarizes the first
3−
Although appropriate reaction time for ligand exchange does
not affect the size of CdSe QDs (Figure S6), longer reaction time
or the addition of large excesses of water-solubilizing capping
agents can result in a decrease of QD size. The size of CdSe QDs
prepared for H2-generating experiments is in the range of 2.5−
2.7 nm, corresponding to their first excitonic peaks in the range of
515−530 nm. CdSe QDs of this size have a conduction band
(CB) edge as negative as −1 V vs NHE,66 making photo-
generated electrons sufficiently negative to reduce complexes 1−
4 (Table S5).
In initial catalytic studies involving complexes 1−4, CdSe-
DHLA QDs were employed with 520 nm LED light leading to
the hydrogen evolution profiles depicted in Figure 5. Under
minimally optimized conditions, the turnover numbers (TONs)
per Fe atom after 80 h of irradiation reached 20 600, 29 400,
reversible reduction potentials of all four Fe dithiolene
complexes, which range from −0.723 V to −0.928 V vs SCE,
depending on the electron donating ability of the dithiolene
ligand. A plot of peak current vs (scan rate)1/2 for 2 exhibits a
linear dependence (R2 = 0.999, Figure 4b), consistent with a
diffusion-controlled interfacial redox process. To examine the
potential activity of these complexes for proton reduction,
aliquots of trifluoroacetic acid were added to solutions of the
complexes in deaerated CH3CN containing 1 M H2O. As shown
in Figure 4c for complex 2, addition of trifluoroacetic acid
triggers significant current enhancement near the onset of the
reversible redox couple, indicating that the reduced form of
complex 2 is active for proton reduction. Similar behaviors were
also observed for complexes 1, 3, and 4 (Figure S5), where the
current enhancement was triggered by addition of trifluoroacetic
acid. Interestingly, under otherwise identical experimental
conditions, complex 2 exhibits highest catalytic current enhance-
ment followed in order by 1, 4 and 3. These results reveal the
inherent differences of the four complexes toward catalytic
proton reduction.
Photocatalytic H2 Evolution Activity. Photocatalytic
experiments for hydrogen evolution were performed using
water-soluble CdSe QDs as the photosensitizer, ascorbic acid
(AA) as the sacrificial electron donor, and complexes 1−4 as
water reduction catalysts in deaerated EtOH/H2O (1/1 v/v)
solvent. The amount of H2 produced was monitored in real time
using a pressure transducer on each sample and quantified by GC
analysis of the headspace gas at the end of the irradiation period
(see Experimental Section). While AA is employed solely as an
electron source in these studies, we note that the net conversion
of AA to dehydroascorbic acid + H2 is actually energy storing by
approximately 20 kcal/mol.8,16 The water-soluble CdSe QDs
used for the photochemical experiments were prepared through
ligand exchange of CdSe-TOPO QDs with the corresponding
water-solubilizing capping agent (DHLA, S3, MPA, and MUA
Figure 5. Photocatalytic hydrogen production from systems containing
Fe-dithiolene complexes (0.5 μM), CdSe-DHLA QDs (525 nm, 2 μM),
AA (0.2 M) in 5 mL EtOH/H2O (1/1) at pH = 4.5 upon irradiation with
520 nm LED at 15 °C and 1 atm initial pressure of N2/CH4 (4/1) with
CH4 as an internal standard for H2 quantification by GC analysis.
F
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
15 200, and 8000 for catalysts 1−4, respectively, with complex 2
showing highest activity. In view of the fact that earlier studies
had shown that the DHLA capping agents on the CdSe QDs are
labile and dissociate into the solution even in the absence of light
irradiation,22 examination of the interaction of DHLA with
catalyst complexes 1−4 was examined by UV−vis spectroscopy,
the results of which are shown in Figure S7. For complexes 1 and
2, the benzenedithiolate ligands L are readily displaced by
DHLA, resulting in a pale yellow solution, indicative of the
formation of new species such as [Fe(L)x(DHLA)y]n− with x = 0
or 1. In contrast, substitution of the benzenedithiolate ligands L
by DHLA is much slower for complexes 3 and 4, although
complete substitution is observed after 24 h at room temper-
ature. The differences in the observed rates of substitution
(Figure S7) may relate to the tendency of the
[Fe2(dithiolene)4]2− anions to dissociate into monomeric
[Fe(dithiolene)2]− species even without reduction. The Co
analogues of 1 and 2 exist as monomers in solution, whereas
those of 3 and 4 exist in dimeric form. The rate of ligand
replacement by DHLA also depends on its concentration (Figure
S8), with a faster disappearance of the characteristic absorption
band of 2 with increased concentrations of DHLA, supporting
the formation of DHLA-containing species. In light of the
DHLA/dithiolene ligand exchange reactions occurring, the H2-
generating activity observed in Figure 5 does not accurately
reflect the catalytic activity of complexes 1−4 due to formation of
[Fe(L)x(DHLA)y]n− species during H2 formation.
dark generated little or no H2 (Chart S1) The observed order of
activity for complexes 1−4 was found to correlate with the
reduction potential for the [FeL2]22−/3− couple (−0.928, −0.898,
−0.764, −0.723 V vs SCE), and was thus similar to the order of
activity reported for CdSe-S3 systems having the analogous Co
bis dithiolene catalysts.32 The observed order of activity for
complexes 1−4 under photochemical conditions is consistent
with their electrocatalytic proton reduction activity as shown in
Figure 6 and Figure S5. Finally, for the system Fe(NO3)3/CdSe-
S3/AA in which no dithiolene ligands are present, irradiation
after 80 h gave insignificant amount of H2.
Changes in relative amounts of components in the systems 1−
4/CdSe-S3/AA yielded additional insight. A 8-fold greater
concentration of 2 in the reaction solution led to a corresponding
increase of evolved H2, with the amount increasing from ∼1.28
mL (57 μmol) to 2.78 mL (124 μmol) (Figure S9) although the
TON decreased from ∼22 800 to ∼6200, suggesting that at the
higher concentration, the catalyst is no longer the limiting
component. The concentrations of CdSe-S3 QDs and the
sacrificial electron donor (AA) also affect the rates and yields of
H2 production. An increase in the concentration of CdSe-S3
QDs from 1 μM to 4 μM results in growth of H2 yields from
∼1.29 mL (58 μmol) to 2.80 mL (125 μmol) corresponding to a
change in TON from ∼11 500 to 25 000 (Figure S10). Whereas
an increase in AA concentration also impacts positively H2
evolution, at concentrations greater than 0.4 M, saturation may
be approached (Figure S11).
The exchange inertness of the S3 capping agent from the
surface of CdSe QDs has previously been described in studies
involving Ni DHLA and Co dithiolene catalysts.32 Experiments
were therefore undertaken with CdSe-S3 light absorbers and
complexes 1−4, as well as with Fe(NO3)3 as a catalyst precursor,
to determine catalyst activity in the absence of DHLA
substitution. Figure 6 illustrates the photocatalytic hydrogen
The addition of DHLA to the S3-containing systems
(Fe(NO3)3/CdSe-S3/AA), results in significant increases in
H2-generating capability. For the Fe(NO3)3-containing system,
the addition of 80 equiv of DHLA gave great enhancement of H2
evolution in terms of both rate and yield (the TON reaches
∼15 300 in the presence of DHLA, compared to ∼3900 in the
absence of DHLA), indicating catalytic activity results from an in
situ-formed Fe DHLA complex (Figure S12). For the system
containing the Fe dithiolene complexes 1−4 and CdSe-S3/AA,
160 equiv of DHLA were added, and the systems were allowed to
age for 2 h prior to irradiation to permit simple ligand exchange.
The H2 production profiles for the four systems are similar to
each other (Figure S13), as well as to that of the Fe(NO3)3/
CdSe-S3/AA system (Figure S14). These results thus support
the notion of a common Fe DHLA catalyst produced in each
system.
The addition of smaller amounts of DHLA to the 2/CdSe-S3/
AA system, however, indicate the formation of a mixed ligand
catalyst. The results are summarized in Table S6. Specifically, the
addition of 1 or 2 equiv of DHLA relative to catalyst yields an
increase in H2 yields, but for further additions of 4−80 equiv
DHLA, the rate declines slightly and levels off to a value slightly
lower than that observed for the system Fe(NO3)3/CdSe-S3/AA
+ 80 equiv DHLA. The results thus indicate that for the most
active system in which the iron dithiolene complex 2 and DHLA
exist, the in situ-generated catalyst contains both DHLA and
dithiolene ligands for H2 production.
Figure 6. Photocatalytic hydrogen production from systems containing
Fe-dithiolene complexes (0.5 μM), CdSe-S3 QDs (520 nm, 2 μM), AA
(0.2 M) in 5 mL EtOH/H2O (1/1) at pH = 4.5 upon irradiation with
520 nm LED at 15 °C and 1 atm initial pressure of N2/CH4 (4/1) with
CH4 as an internal standard for H2 quantification by GC analysis.
Long-Term Stability and Reusability Evaluation. Several
experiments were conducted to assess long-term system
robustness and reusability of system components. Whereas the
rate and yield of H2 generation exhibited a stable profile in initial
catalytic runs over >80 h of irradiation (Figure 7), reuse of the
system from the initial run led to decreases in both rate and yield
of H2 production. The decline in system activity is attributed (a)
depletion of AA, (b) instability of catalyst 2, and (c) change in pH
of the reaction solution. With regard to (c), system pH following
evolution profiles for complexes 1−4 using CdSe-S3 QDs as the
photosensitizer. The overall activity of the four complexes is
lower than that of the systems using CdSe-DHLA light-
absorbers. In terms of TONs for these CdSe-S3 systems after
80 h irradiation, the activity order is 2 > 1 > 3 > 4 with TONs of
17 600, 11 100, 9100, and 8400, respectively. In control
experiments, removal of any component in the WRC system
(catalyst 1−4, AA or CdSe-S3 QDs) or running the catalysis in
G
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
S3 to CdSe QDs was used to denote different CdSe QDs light-
absorbers, for example, 100 S3 units per CdSe dot indicates the
ratio of S3 capping agents per CdSe dot is 100.
Figure 8 illustrates results of the photogeneration of hydrogen
using CdSe-S3 QDs photosensitizers that had been reacted with
Figure 7. Photocatalytic hydrogen production from systems containing
complex 2 (0.5 μM), CdSe-S3 QDs (523 nm, 2 μM), AA (0.2 M) in 5
mL EtOH/H2O (1/1) at pH = 4.5 upon irradiation with 520 nm LED at
15 °C and 1 atm initial pressure of N2/CH4 (4/1): (a) black curve: initial
run of the catalytic solution; (b) red curve: second run of the reaction
solution; and (c) blue line: the reaction solution following (b) to which
0.5 μM complex 2 and 1 mL 1 M AA stock solution were added.
Figure 8. Photocatalytic hydrogen production from systems containing
complex 2 (0.5 μM), 2 μM CdSe-S3 QDs (blue curve: 523 nm with 70
S3 units; black curve: 525 nm with 100 S3 units; red curve: 515 nm with
180 S3 units), AA (0.2 M) in 5 mL EtOH/H2O (1/1) at pH = 4.5.
Irradiation with 520 nm LED at 15 °C and 1 atm initial pressure of N2/
CH4 (4/1) with CH4 as a GC analysis internal standard for H2
quantification.
a second period of irradiation was found to increase to 5.6,
consistent with proton removal from the system to form H2 and
the declining capacity of the ascorbic acid/ascorbate buffer. The
addition of 1 mL of 1 M AA and 25 μL of 0.1 mM solution of
catalyst 2 to a previously irradiated system yielded a rate and yield
of H2 production during a third period of irradiation that were
almost the same as those of the initial run, as shown in Figure 7.
To assay further the robustness of the photosensitizers,
previously employed CdSe QDs were isolated from reaction
solutions by centrifugation and then redispersed in slightly basic
water for subsequent catalytic experiments. The redispersed
CdSe-S3 QDs solution is quite clear, while that of CdSe-DHLA
QDs appears slightly cloudy (possibly as a consequence of
DHLA ligand dissociation resulting in lower capping agent
coverage). The UV−vis spectra of solutions of isolated CdSe-S3
QDs and CdSe-DHLA QDs have similar absorption profiles,
both to each other and to ones obtained from the original QDs
solutions, respectively (Figure S15). Light-driven H2 evolution
experiments using either isolated CdSe-DHLA QDs or CdSe-S3
QDs plus fresh catalyst and AA solution give similar H2
generation rates and yields to those obtained from systems
using fresh CdSe QDs (Figure S16 and Figure S17), indicating
the relative stability of the QD light absorbers during the
examined periods of irradiation.
different amounts of S3 capping agent prior to irradiation and the
TONs obtained for light-driven hydrogen generating experi-
ments are summarized in Table S7. The results show that
decreasing the amount of S3 capping agents on CdSe from 180 to
100 to 70 units per CdSe dot increases the rate and final yield of
H2 generation (the TONs increase from 4600 to 17 600 to
20 700, respectively, for complex 2-catalyzed reactions).
Similarly, for catalytic systems containing DHLA-capped CdSe
QDs with complex 2 as the catalyst and AA, an increase in activity
is noted with a decrease in DHLA units per CdSe QD from 210
to 140 (Figure S18, Table S8); however, the difference between
the obtained TONs is much smaller than for the systems using
S3-capped CdSe QDs. The reduced magnitude of the effect with
DHLA capping agents relative to S3 is consistent with the lability
of DHLA for dissociation from the CdSe QD surface in contrast
with the inertness of the S3 capping agent.
In addition to the results on the variation of capping agent
amounts for CdSe-S3 and CdSe-DHLA catalytic systems, it was
also found that photogeneration of H2 using CdSe QDs with the
capping agents mercaptopropionoic acid (MPA) and mercap-
toundecanoic acid (MUA) exhibit different behaviors in activity.
Figure 9 shows hydrogen evolution data using CdSe-MPA,
complex 2 as catalyst and AA as electron donor. In the absence of
external DHLA, ∼2.4 mL (107 μmol) hydrogen gas was
produced after 120 h irradiation, corresponding to a TON of
∼21 300. Interestingly, addition of 80 equiv of DHLA to the
catalytic solution dramatically enhances the rate and yield of H2
generation with 12.4 mL (553 μmol) H2 generated (5 times that
of the system without DHLA), consistent with a TON of
110 600. Very recently, Wu and co-workers reported a related
photochemical hydrogen evolution system using CdSe-MPA
QDs as photosensitizer, an Fe2S2 hydrogenase mimic bound to
poly(acrylic acid) (PPA-g-Fe2S2) as the catalyst and AA as the
electron donor at pH 4.0.40a A blue LED at 450 nm was
Effect of Surface Ligand Coverage of CdSe QDs on H2
Generation Activity. From studies above, it is evident that both
the nature and amount of capping agent affect system activity, in
accord with studies on the redox chemistry involving semi-
conductor nanoparticles.67 To probe this notion further, several
experiments were conducted. In one study, CdSe-S3 and CdSe-
DHLA QDs with varying amounts of water-solubilizing capping
agents were prepared and characterized and the amounts of
1
capping agent on the CdSe QD surface were quantified by H
NMR spectroscopy. Specifically, the S3 and DHLA capping
agents were removed from the QD surface by addition of 6 M
HCl to pH 0, thus causing precipitation of the CdSe particles.68
The dissociated capping agents were separated from the
quantum dots, redissolved in DMSO-d6 and then quantified by
1H NMR spectroscopy using benzene as an internal standard
(see Experimental details in the SI). The molar ratio of DHLA or
H
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
exhibits the highest activity when using different types of CdSe
light-absorbers (CdSe-DHLA, CdSe-S3, CdSe-MPA, and CdSe-
MUA). The overall order of activity correlates with the reduction
potential of the formally Fe(III) dimeric dianions.
The catalytic activity of these complexes is found to depend on
the nature of the water-solubilizing capping agent for the CdSe
light absorbers. When S3, which has been established to be inert
to substitution, is employed, modest activities are found that
correlate with the reduction potential of the dimeric bis-
(dithiolene) formally Fe(III) complexes. Upon addition of ≥80
equiv of DHLA, catalytic activity increases greatly and shows
similar levels of activity for all four bis dithiolene complexes, as
well as for Fe(NO3)3 as an Fe(III) source, indicative of a
common in situ-formed Fe(DHLA)x catalyst. For the additions
of 1−4 equiv of DHLA to the CdSe-S3 system, the catalytic
activity is consistent with the formation of a mixed Fe-
(DHLA)x(dithiolene)yz− that is most active for the toluenedi-
thiolate system.
Figure 9. Photocatalytic hydrogen production from systems containing
CdSe-MPA QDs (518 nm, 1 μM), complex 2 (0.5 μM; black curve) or
complex 2 (0.5 μM) + DHLA (80 μM) (red curve), AA (0.2 M) in 5 mL
EtOH/H2O (1/1) at pH = 4.5 upon irradiation with 520 nm LED at 15
°C and 1 atm initial pressure of N2/CH4.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.6b05040.
■
employed for irradiation in those systems. Under optimal
conditions, a TON of ∼27 100 was obtained after 8 h irradiation,
but the system tended to stop working upon prolonged
irradiation. While the relative activity and final TONs obtained
in their systems are comparable to those reported in current
work, the results cannot be compared directly due to many
differences between them and the present results with respect to
experimental conditions and analyses.
S
Detailed experimental procedures for water-soluble ligand
exchange, X-ray crystallographic data, UV−vis spectra,
electrochemistry data, and additional photocatalytic
hydrogen production data (PDF)
Crystal data (CIF)
The system with MUA as the surface capping agent is less
active than the corresponding MPA system, but like the MPA
system, it exhibits an increase in activity upon the addition of
DHLA (Figure S19). In either case (with CdSe-MUA only or
with CdSe-MUA + DHLA), hydrogen evolution results are lower
(TONs of 11 400 without DHLA and 29 200 with 160 equiv
DHLA) than those obtained with the corresponding MPA
systems. It can be conjectured that the longer aliphatic chains of
MUA relative to MPA inhibit electron transfer to the catalyst
needed for hydrogen generation.
All of the results for photochemical hydrogen formation from
protons for the systems described here illustrate the importance
of surface capping agents and their dynamics in determining their
relative catalytic activity. It appears that for photochemical H2
production, surface capping agents may influence both exciton
lifetime in CdSe QDs upon irradiation and electron transfer rates
within the system for such H2-generating catalysis.67,69 More
detailed studies to better understand the mechanism on how
capping agents affect catalytic activity are in progress.
AUTHOR INFORMATION
Corresponding Author
*eisenberg@chem.rochester.edu
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by the Division of Chemical
Sciences, Geosciences, and Biosciences, Office of Basic Energy
Sciences, U.S. Department of Energy, Grant DE-FG02-
09ER16121 for all syntheses, characterizations and H2
generation studies. Mossbauer measurements were conducted
̈
with the support of National Science Foundation Grant CHE-
1454370 to M. L. N. The receipt of an Alfred P. Sloan Fellowship
to M. L. N. is gratefully acknowledged.
REFERENCES
■
(1) Graetzel, M. Acc. Chem. Res. 1981, 14, 376.
(2) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U. S. A. 2006, 103,
15729.
(3) Gray, H. B. Nat. Chem. 2009, 1, 7.
(4) Eisenberg, R. Science 2009, 324, 44.
(5) Lv, H.; Geletii, Y. V.; Zhao, C.; Vickers, J. W.; Zhu, G.; Luo, Z.;
Song, J.; Lian, T.; Musaev, D. G.; Hill, C. L. Chem. Soc. Rev. 2012, 41,
7572.
(6) Lehn, J. M.; Sauvage, J. P. Nouv. J. Chim. 1977, 1, 449.
(7) Kalyanasundaram, K.; Kiwi, J.; Gratzel, M. Helv. Chim. Acta 1978,
CONCLUSIONS
■
In this study, four closely related Fe bis(benzenedithiolate)
complexes have been synthesized, characterized and studied as
catalysts for the light-driven generation of H2. All four formally
Fe(III) bis dithiolene complexes are isolated as dianionic dimers
with similar structural characteristics in the solid state, although
significant differences in nonbonding Fe···Fe separations are
noted. Cyclic voltammetric experiments in the presence of
trifluoroacetic acid reveal that the reduced monomeric forms of
the complexesthat is, [Fe(dithiolate ligand)2]2−are active
for proton reduction. When used with water-solubilized CdSe
QDs as the light absorber and ascorbic acid as the sacrificial
electron donor, all four complexes exhibit catalytic hydrogen-
evolving activity under visible light irradiation (520 nm). The
complex with the most electron-donating dithiolene ligand
̈
61, 2720.
(8) Brown, G. M.; Brunschwig, B. S.; Creutz, C.; Endicott, J. F.; Sutin,
N. J. Am. Chem. Soc. 1979, 101, 1298.
(9) DeLaive, P. J.; Sullivan, B. P.; Meyer, T. J.; Whitten, D. G. J. Am.
Chem. Soc. 1979, 101, 4007.
(10) Streich, D.; Astuti, Y.; Orlandi, M.; Schwartz, L.; Lomoth, R.;
Hammarstroem, L.; Ott, S. Chem. - Eur. J. 2010, 16, 60.
I
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(11) Li, X.; Wang, M.; Chen, L.; Wang, X.; Dong, J.; Sun, L.
ChemSusChem 2012, 5, 913.
(43) McNamara, W. R.; Han, Z.; Yin, C.-J.; Brennessel, W. W.;
Holland, P. L.; Eisenberg, R. Proc. Natl. Acad. Sci. U. S. A. 2012, 109,
15594.
(12) Berggren, G.; Adamska, A.; Lambertz, C.; Simmons, T. R.;
Esselborn, J.; Atta, M.; Gambarelli, S.; Mouesca, J. M.; Reijerse, E.;
Lubitz, W.; Happe, T.; Artero, V.; Fontecave, M. Nature 2013, 499, 66.
(13) Artero, V.; Saveant, J.-M. Energy Environ. Sci. 2014, 7, 3808.
(14) Fihri, A.; Artero, V.; Razavet, M.; Baffert, C.; Leibl, W.; Fontecave,
M. Angew. Chem., Int. Ed. 2008, 47, 564.
(44) Sproules, S.; Wieghardt, K. Coord. Chem. Rev. 2010, 254, 1358.
(45) Sproules, S.; Kapre, R. R.; Roy, N.; Weyhermuller, T.; Wieghardt,
̈
K. Inorg. Chim. Acta 2010, 363, 2702.
(46) Sellmann, D.; Geck, M.; Moll, M. J. Am. Chem. Soc. 1991, 113,
5259.
(47) Begum, A.; Sarkar, S. Eur. J. Inorg. Chem. 2012, 2012, 40.
(48) Fontecilla-Camps, J. C.; Volbeda, A.; Cavazza, C.; Nicolet, Y.
Chem. Rev. (Washington, DC, U. S.) 2007, 107, 4273.
(49) Shima, S.; Pilak, O.; Vogt, S.; Schick, M.; Stagni, M. S.; Meyer-
Klaucke, W.; Warkentin, E.; Thauer, R. K.; Ermler, U. Science 2008, 321,
572.
(50) Armstrong, F. A.; Belsey, N. A.; Cracknell, J. A.; Goldet, G.;
Parkin, A.; Reisner, E.; Vincent, K. A.; Wait, A. F. Chem. Soc. Rev. 2009,
38, 36.
(51) Mulder, D. W.; Shepard, E. M.; Meuser, J. E.; Joshi, N.; King, P.
W.; Posewitz, M. C.; Broderick, J. B.; Peters, J. W. Structure 2011, 19,
1038.
(52) Gunsalus, I. C.; Barton, L. S.; Gruber, W. J. Am. Chem. Soc. 1956,
78, 1763.
(15) Lazarides, T.; McCormick, T.; Du, P.; Luo, G.; Lindley, B.;
Eisenberg, R. J. Am. Chem. Soc. 2009, 131, 9192.
(16) McNamara, W. R.; Han, Z.; Alperin, P. J.; Brennessel, W. W.;
Holland, P. L.; Eisenberg, R. J. Am. Chem. Soc. 2011, 133, 15368.
(17) Zhang, P.; Jacques, P.-A.; Chavarot-Kerlidou, M.; Wang, M.; Sun,
L.; Fontecave, M.; Artero, V. Inorg. Chem. 2012, 51, 2115.
(18) Singh, W. M.; Baine, T.; Kudo, S.; Tian, S.; Ma, X. A. N.; Zhou, H.;
DeYonker, N. J.; Pham, T. C.; Bollinger, J. C.; Baker, D. L.; Yan, B.;
Webster, C. E.; Zhao, X. Angew. Chem., Int. Ed. 2012, 51, 5941.
(19) Khnayzer, R. S.; Thoi, V. S.; Nippe, M.; King, A. E.; Jurss, J. W.; El
Roz, K. A.; Long, J. R.; Chang, C. J.; Castellano, F. N. Energy Environ. Sci.
2014, 7, 1477.
(20) Helm, M. L.; Stewart, M. P.; Bullock, R. M.; DuBois, M. R.;
DuBois, D. L. Science 2011, 333, 863.
(21) Small, Y. A.; DuBois, D. L.; Fujita, E.; Muckerman, J. T. Energy
Environ. Sci. 2011, 4, 3008.
(53) Talapin, D. V.; Mekis, I.; Gotzinger, S.; Kornowski, A.; Benson,
̈
O.; Weller, H. J. Phys. Chem. B 2004, 108, 18826.
(54) Evans, D. F. J. Chem. Soc. 1959, 2003.
(22) Han, Z.; Qiu, F.; Eisenberg, R.; Holland, P. L.; Krauss, T. D.
(55) Josef, P. US Patent No. US2842578, 1958.
(56) Bruker; SADABS V2008-1, BRUKER AXS Inc.: Madison, WI,
2008.
Science 2012, 338, 1321.
(23) Han, Z.; McNamara, W. R.; Eum, M.-S.; Holland, P. L.; Eisenberg,
R. Angew. Chem., Int. Ed. 2012, 51, 1667.
(24) Han, Z.; Shen, L.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R.
J. Am. Chem. Soc. 2013, 135, 14659.
(25) Lv, H.; Guo, W.; Wu, K.; Chen, Z.; Bacsa, J.; Musaev, D. G.;
Geletii, Y. V.; Lauinger, S. M.; Lian, T.; Hill, C. L. J. Am. Chem. Soc. 2014,
136, 14015.
(26) Lv, H.; Chi, Y.; van Leusen, J.; Kogerler, P.; Chen, Z.; Bacsa, J.;
Geletii, Y. V.; Guo, W.; Lian, T.; Hill, C. L. Chem. - Eur. J. 2015, 21,
17363.
(27) Karunadasa, H. I.; Chang, C. J.; Long, J. R. Nature 2010, 464,
1329.
(28) Karunadasa, H. I.; Montalvo, E.; Sun, Y.; Majda, M.; Long, J. R.;
Chang, C. J. Science 2012, 335, 698.
(29) Hawecker, J.; Lehn, J. M.; Ziessel, R. Nouv. J. Chim. 1983, 7, 271.
(30) Collin, J. P.; Sauvage, J. P. Coord. Chem. Rev. 1989, 93, 245.
(31) McCormick, T. M.; Han, Z.; Weinberg, D. J.; Brennessel, W. W.;
Holland, P. L.; Eisenberg, R. Inorg. Chem. 2011, 50, 10660.
(32) Das, A.; Han, Z.; Haghighi, M. G.; Eisenberg, R. Proc. Natl. Acad.
Sci. U. S. A. 2013, 110, 16716.
(57) Bruker; SAINT, V8.27b, BRUKER AXS Inc.: Madison, WI, 2012.
(58) Altomare, A.; Burla, M.; Camalli, M.; Cascarano, G.; Giacovazzo,
C.; Guagliardi, A.; Moliterni, A.; Polidori, G.; Spagna, R. SIR97: A new
program for solving and refining crystal structures; Istituto di
Cristallografia, CNR: Bari, Italy, 1999.
(59) Sheldrick, G. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64,
112.
̈
(60) Daifuku, S. L.; Al-Afyouni, M. H.; Snyder, B. E. R.; Kneebone, J.
L.; Neidig, M. L. J. Am. Chem. Soc. 2014, 136, 9132.
(61) Kanatzidis, M. G.; Coucouvanis, D. Inorg. Chem. 1984, 23, 403.
(62) Kang, B. S.; Weng, L. H.; Wu, D. X.; Wang, F.; Guo, Z.; Huang, L.
R.; Huang, Z. Y.; Liu, H. Q. Inorg. Chem. 1988, 27, 1128.
(63) Ray, K.; Bill, E.; Weyhermuller, T.; Wieghardt, K. J. Am. Chem. Soc.
̈
2005, 127, 5641.
(64) Ray, K.; Begum, A.; Weyhermuller, T.; Piligkos, S.; van Slageren,
̈
J.; Neese, F.; Wieghardt, K. J. Am. Chem. Soc. 2005, 127, 4403.
(65) Birchall, T.; Greenwood, N. N. J. Chem. Soc. A 1969, 0, 286.
(66) Jasieniak, J.; Califano, M.; Watkins, S. E. ACS Nano 2011, 5, 5888.
(67) Weinberg, D. J.; He, C.; Weiss, E. A. J. Am. Chem. Soc. 2016, 138,
2319.
(33) Das, A.; Han, Z.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R.
ACS Catal. 2015, 5, 1397.
(34) Ruberu, T. P. A.; Dong, Y.; Das, A.; Eisenberg, R. ACS Catal.
2015, 5, 2255.
(68) Aldana, J.; Lavelle, N.; Wang, Y.; Peng, X. J. Am. Chem. Soc. 2005,
127, 2496.
(69) Jia, Y.; Chen, J.; Wu, K.; Kaledin, A.; Musaev, D. G.; Xie, Z.; Lian,
T. Chem. Sci. 2016, 7, 4125.
(35) Amirav, L.; Alivisatos, A. P. J. Phys. Chem. Lett. 2010, 1, 1051.
(36) Brown, K. A.; Dayal, S.; Ai, X.; Rumbles, G.; King, P. W. J. Am.
Chem. Soc. 2010, 132, 9672.
(37) Brown, K. A.; Wilker, M. B.; Boehm, M.; Dukovic, G.; King, P. W.
J. Am. Chem. Soc. 2012, 134, 5627.
(38) Wang, F.; Wang, W.-G.; Wang, X.-J.; Wang, H.-Y.; Tung, C.-H.;
Wu, L.-Z. Angew. Chem., Int. Ed. 2011, 50, 3193.
(39) Zhu, H.; Song, N.; Lv, H.; Hill, C. L.; Lian, T. J. Am. Chem. Soc.
2012, 134, 11701.
(40) (a) Wang, F.; Liang, W.-J.; Jian, J.-X.; Li, C.-B.; Chen, B.; Tung,
C.-H.; Wu, L.-Z. Angew. Chem., Int. Ed. 2013, 52, 8134. (b) Li, C.-B.; Li,
Z.-J.; Yu, S.; Wang, G.-X.; Wang, F.; Meng, Q.-Y.; Chen, B.; Feng, K.;
Tung, C.-H.; Wu, L.-Z. Energy Environ. Sci. 2013, 6, 2597.
(41) Wu, K.; Chen, Z.; Lv, H.; Zhu, H.; Hill, C. L.; Lian, T. J. Am. Chem.
Soc. 2014, 136, 7708.
(42) Shemesh, Y.; Macdonald, J. E.; Menagen, G.; Banin, U. Angew.
Chem., Int. Ed. 2011, 50, 1185.
J
DOI: 10.1021/jacs.6b05040
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX