Photocatalytic H2-Evolution by Homogeneous Molybdenum Sulfide Clusters Supported by Dithiocarbamate Ligands

Article abstract of DOI:10.1021/acs.inorgchem.9b02252

Irradiation at 460 nm of [Mo₃(μ₃-S)(μ₂-S₂)₃(S₂CNR₂)₃]I ([2a]I, R = Me; [2b]I, R = Et; [2c]I, R = ⁱBu; [2d]I, R = CH₂C₆H₅) in a mixed aqueous-polar organic medium with [Ru(bipy)₃]²⁺ as photosensitizer and Et₃N as electron donor leads to H₂ evolution. Maximum activity (300 turnovers, 3 h) is found with R = ⁱBu in 1:9 H₂O:MeCN; diminished activity is attributed to deterioration of [Ru(bipy)₃]²⁺. Monitoring of the photolysis mixture by mass spectrometry suggests transformation of [Mo₃(μ₃-S)(μ₂-S₂)₃(S₂CNR₂)₃]⁺ to [Mo₃(μ₃-S)(μ₂-S)₃(S₂CNR₂)₃]⁺ via extrusion of sulfur on a time scale of minutes without accumulation of the intermediate [Mo₃S₆(S₂CNR₂)₃]⁺ or [Mo₃S₅(S₂CNR₂)₃]⁺ species. Deliberate preparation of [Mo₃S₄(S₂CNEt₂)₃]⁺ ([3]⁺) and treatment with Et₂NCS₂^1- yields [Mo₃S₄(S₂CNEt₂)₄] (4), where the fourth dithiocarbamate ligand bridges one edge of the Mo₃ triangle. Photolysis of 4 leads to H₂ evolution but at ～25% the level observed for [Mo₃S₇(S₂CNEt₂)₃]⁺. Early time monitoring of the photolyses shows that [Mo₃S₄(S₂CNEt₂)₄] evolves H₂ immediately and at constant rate, while [Mo₃S₇(S₂CNEt₂)₃]⁺ shows a distinctive incubation prior to a more rapid H₂ evolution rate. This observation implies the operation of catalysts of different identity in the two cases. Photolysis solutions of [Mo₃S₇(S₂CNⁱBu₂)₃]⁺ left undisturbed over 24 h deposit the asymmetric Mo₆ cluster [(ⁱBu₂NCS₂)₃(μ₂-S₂)₂(μ₃-S)Mo₃](μ₃-S)(μ₃-η²,η¹-S′,η¹-S″-S₂)[Mo₃(μ₂-S)₃(μ₃-S)(S₂CNⁱBu₂)₂(μ₂-S₂CNⁱBu₂)] in crystalline form, suggesting that species with this hexametallic composition and core topology are the probable H₂-evolving catalysts in photolyses beginning with [Mo₃S₇(S₂CNR₂)₃]⁺. When used as solvent, N,N-dimethylformamide (DMF) suppresses H₂-evolution but to a greater degree for [Mo₃S₄(S₂CNEt₂)₄] than for [Mo₃S₇(S₂CNEt₂)₃]⁺. Recrystallization of [Mo₃S₄(S₂CNEt₂)₄] from DMF affords [Mo₃S₄(S₂CNEt₂)₄(η¹,κO-DMF)] (5), implying that inhibition by DMF arises from competition for a Mo coordination site that is requisite for H₂ evolution. Computational assessment of [Mo₃S₄(S₂CNMe₂)₃]⁺ following addition of 2H⁺ and 2e^- suggests a Mo(H)-μ₂(SH) intermediate as the lowest energy species for H₂ elimination. An analogous pathway may be available to the Mo₆ cluster via dissociation of one end of the μ₂-S₂CNR₂ ligand, a known hemilabile ligand type, in the [Mo₃S₄]⁴⁺ fragment.

Full text of DOI:10.1021/acs.inorgchem.9b02252

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Photocatalytic H2‑Evolution by Homogeneous Molybdenum Sulfide
Clusters Supported by Dithiocarbamate Ligands
Patricia R. Fontenot,† Bing Shan,† Bo Wang,† Spenser Simpson,† Gayathri Ragunathan,†
Angelique F. Greene,† Antony Obanda,† Leigh Anna Hunt,‡ Nathan I. Hammer,‡
Charles Edwin Webster,§ Joel T. Mague,† Russell H. Schmehl,*,† and James P. Donahue*,†
†Department of Chemistry, Tulane University, 6400 Freret Street, New Orleans, Louisiana 70118-5698, United States
‡Department of Chemistry and Biochemistry, University of Mississippi, Oxford, Mississippi 38655, United States
§Department of Chemistry, Mississippi State University, P.O. Box 9573, Mississippi State, Mississippi 39762-9573, United States
S
* Supporting Information
ABSTRACT: Irradiation at 460 nm of [Mo3(μ3-S)(μ2-
S2)3(S2CNR2)3]I ([2a]I, R = Me; [2b]I, R = Et; [2c]I, R =
iBu; [2d]I, R = CH2C6H5) in a mixed aqueous−polar organic
medium with [Ru(bipy)3]2+ as photosensitizer and Et3N as
electron donor leads to H2 evolution. Maximum activity (300
i
turnovers, 3 h) is found with R = Bu in 1:9 H2O:MeCN;
diminished activity is attributed to deterioration of [Ru(bipy)3]2+.
Monitoring of the photolysis mixture by mass spectrometry
suggests transformation of [Mo3(μ3-S)(μ2-S2)3(S2CNR2)3]+ to
[Mo3(μ3-S)(μ2-S)3(S2CNR2)3]+ via extrusion of sulfur on a time
scale of minutes without accumulation of the intermediate
[Mo3S6(S2CNR2)3]+ or [Mo3S5(S2CNR2)3]+ species. Deliberate
preparation of [Mo3S4(S2CNEt2)3]+ ([3]+) and treatment with
1−
Et2NCS2 yields [Mo3S4(S2CNEt2)4] (4), where the fourth dithiocarbamate ligand bridges one edge of the Mo3 triangle.
Photolysis of 4 leads to H2 evolution but at ∼25% the level observed for [Mo3S7(S2CNEt2)3]+. Early time monitoring of the
photolyses shows that [Mo3S4(S2CNEt2)4] evolves H2 immediately and at constant rate, while [Mo3S7(S2CNEt2)3]+ shows a
distinctive incubation prior to a more rapid H2 evolution rate. This observation implies the operation of catalysts of different
identity in the two cases. Photolysis solutions of [Mo3S7(S2CNiBu2)3]+ left undisturbed over 24 h deposit the asymmetric Mo6
cluster [(iBu2NCS2)3(μ2-S2)2(μ3-S)Mo3](μ3-S)(μ3-η2,η1-S′,η1-S″-S2)[Mo3(μ2-S)3(μ3-S)(S2CNiBu2)2(μ2-S2CNiBu2)] in crystal-
line form, suggesting that species with this hexametallic composition and core topology are the probable H2-evolving catalysts in
photolyses beginning with [Mo3S7(S2CNR2)3]+. When used as solvent, N,N-dimethylformamide (DMF) suppresses H2-
evolution but to a greater degree for [Mo3S4(S2CNEt2)4] than for [Mo3S7(S2CNEt2)3]+. Recrystallization of
[Mo3S4(S2CNEt2)4] from DMF affords [Mo3S4(S2CNEt2)4(η1,κO-DMF)] (5), implying that inhibition by DMF arises from
competition for a Mo coordination site that is requisite for H2 evolution. Computational assessment of [Mo3S4(S2CNMe2)3]+
following addition of 2H+ and 2e− suggests a Mo(H)−μ2(SH) intermediate as the lowest energy species for H2 elimination. An
analogous pathway may be available to the Mo6 cluster via dissociation of one end of the μ2-S2CNR2 ligand, a known hemilabile
ligand type, in the [Mo3S4]4+ fragment.
experimentally5 and computationally6 as the loci for H2
formation and activation.
This insight has stimulated a considerable body of recent
INTRODUCTION
■
At an annual production of ∼23 million tons, hydrogen gas
ranks among the greatest volume commodity chemicals
consumed globally.1 Its centrality arises not only from its
research aimed at preparing MoS2 in various ways that
maximize surface area and therefore reactive sites. Distinctive
uses as a feedstock chemical for processes such as hydro-
desulfurization2 and ammonia production3 but also from its
forms of MoS2 that have been prepared and studied as
electrocatalysts are amorphous thin films,7,8 nanoplates,9
nanoparticles,10 nanosheets exfoliated and activated chemi-
cally11 or by ultrasonication,12 and nanocubed structures
produced through lithographic techniques.13 Hybrid materials
increasing significance as a clean energy source.4 Among
materials known to catalyze H2 formation from H+ and e−,
MoS2 has drawn attention as one of the few materials that
might meet the economics of scale and therefore have practical
impact. As a bulk material, however, MoS2 is only a moderately
active hydrogenase catalyst. Reactive edge sites that present
terminal disulfide ligation have been implicated both
Received: July 25, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
in which MoS2 is deposited onto reduced graphene oxide,14
carbon nanoparticles,15 or multiwalled carbon nanotubes16
show enhanced catalytic activity due to the conducting nature
of the support. Molybdenum sulfide has also been used in
photoelectrochemical approaches to water splitting, both as a
reduction catalyst and as a photoelectrode/catalytic surface.17
Early work of Tributsch detailed the behavior of MoS2 as a
photoelectrode;18 a brief review, including examination of
MoS2 as a photocathode was recently published by Laursen et
al.17 More recently, the focus has been to employ MoS2 in
various forms as a deposit/thin film on a variety of
photoelectrodes for proton reduction.19−23
Interest in molybdenum sulfides as H2-evolving catalysts
includes well-defined molecular species as well, which offer the
advantages of tunability by variation of the ancillary ligand set
and of amenability to study by solution spectroscopic methods.
Studies by Dubois, Rakowski-Dubois, and co-workers have
shown that [CpMo(μ-S)2(μ-S2CH2)MoCp] (Figure 1(a)),
by showing that Mu
̈
ller’s classic anion, [Mo3S13]2−, which has a
surfeit of disulfide ligands both terminal and bridging, can be
adsorbed onto graphite electrodes and function as H2-
electrocatalyst with good stability and turnover frequency.27
In the context of work aimed at use of the well-known
[Mo3S7(S2CNEt2)3]I cluster, which is prepared from
[Mo3S13]2−, as a precursor to MoS2 films on Cu2O photo-
cathodes,28 we were motivated by the foregoing reports to
assess the activity of [Mo3S13]2− and [Mo3S7(S2CNEt2)3]I as
homogeneous H2-evolving catalysts. Prior reports of the
catalytic H2-evolving ability of small molecule molybdenum
sulfides have been electrocatalytic studies. Furthermore, no
homogeneous catalytic system has, to our knowledge,
originated with a cluster bearing the [Mo3S7]4+ core. In this
study, we have observed both electrocatalytic and photo-
catalytic H2 evolution by these soluble clusters but have
focused the attention of this report upon the latter system. The
photolysis system implemented here used a H2O/MeCN
mixture as solvent, [Ru(bipy)3]2+ as chromophore, Et3N as
sacrificial electron donor, and either tris(p-tolyl)amine or p-
(dimethylamino)toluene as electron transfer intermediary
(ETI) (Scheme 1). This system operates sacrificially to
Scheme 1. Photocatalytic System Employed in These
a
Studies
a
Conditions: [Ru(bpy)3]2+ = 260 μM; [ETI] = 50 mM; [TEA] = 0.4
M; solvent = 1:2:1 THF:MeCN:H2O, 4 mL total; headspace = 4.7
mL.
produce reducing equivalents ([Ru(bpy)3]+) with high
quantum efficiency because of its effectiveness in obviating
unproductive back-electron transfer reactions.29 Leading
aspects of this survey of the photocatalytic H2-evolving activity
of triangular Mo3 sulfides are summarized herein.
Figure 1. Well-defined molecular sulfides of molybdenum reported as
electrocatalysts for H2 production.
EXPERIMENTAL SECTION
■
Spectroscopic and Analytical Methods. All NMR spectra were
recorded at 25 °C with a Varian Unity Inova spectrometer and
referenced to the protonated solvent residual. Electronic absorption
spectra were obtained at ambient temperature with a Hewlett-Packard
8452A diode array spectrophotometer. Raman spectra were obtained
with a Horiba LabRAM HR Evolution spectrometer. The excitation
source was a 633 nm line with an 1800 grooves/mm grating and an
ultralow frequency filter (ULF). Mass spectra were obtained using a
MALDI-TOF-MS (Matrix Assisted Laser Desorption Ionization -
Time of Flight) instrument in positive ion mode without the use of
any matrix. Spectra were acquired by spotting the stainless-steel plate
with a solution containing the compound of interest. Elemental
analyses were obtained through either Galbraith Laboratories of
Knoxville, TN, or Kolbe Microanalytical Laboratory of Oberhausen,
Germany.
and variants thereof with at least two bridging monosulfide
ligands, are competent for the electrocatalytic reduction of H+
to H2 with high Faradaic efficiency and mild overpotential
(120 mV).24 Long, Chang, and co-workers have detailed the
activity of a mononuclear MoS2 species, [MoS2(2,6-bis(1,1-
bis(2-pyridyl)ethyl)pyridine)]2+, for the electrocatalytic pro-
duction of H2 from acidic media (Figure 1(b)).25 This cation
supports the injection of two reducing equivalents, the first of
which is associated with a protonation, prior to the onset of a
catalytic current. At a mercury electrode, protracted electrolysis
over 23 h produced 3500 mol H2/mol catalyst with minimal
degradation. Wu and associates have shown that the complexes
[(4,4′-R2-bipy)MoO(S2)2] (R = H, tBu, or OMe) can
electrocatalytically reduce H+ to H2 with rate constants and
overpotentials that vary with the identity of R (Figure 1(c)).26
A recent report by Kibsgaard, Jaramillo, and Besenbecher
conceptually connects polymeric MoS2 with molecular sulfides
Electrochemistry. Cyclic voltammetry measurements were done
with a glassy carbon working electrode, AgCl/Ag reference electrode,
Pt wire counter electrode, and [nBu4N][PF6] as supporting electro-
lyte. Anhydrous DMF was most often used as solvent, in which
B
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
medium the [Cp2Fe]+/Cp2Fe couple was observed at +0.51 V vs
AgCl/Ag. Bulk electrolyses were conducted in an H-cell with quartz
frit separating the counter electrode from the working electrode
chamber, the underside of which held a Pt wire fed through the glass
to enable electrical contact between the Hg pool working electrode
and the potentiostat (Figure S1). The top of the working electrode
half of the cell was fitted with a rubber stopper through which a AgCl/
Ag reference electrode was held. A hermetic seal around the reference
electrode and stopper was ensured with silicone polymer. A side arm
near the top of the working electrode was closed with a septum
through which aliquots of the headspace could be drawn for GC
analysis. A Pt wire was employed as counter electrode and
[nBu4N][Ph4B] as supporting electrolyte. Proper functioning of this
system was demonstrated by bulk electrochemical reduction of methyl
viologen, which was accomplished with a Faradaic efficiency of 97%.
X-ray Crystallography. Descriptions of the procedures used for
X-ray diffraction data collection and crystal structure solution and
refinement are deferred to the Supporting Information. Unit cell and
refinement data are collected in Tables S1 and S2, and figures with full
atom labeling are presented in Figures S2−S18.
Photolyses. Samples were illuminated under N2 in a custom-built,
multiwell photoreactor comprised of an Al cylinder equipped with
blue LEDs (Solid Apollo, 24 W, 460 nm) mounted inside the cylinder
wall in a uniform, spiral pattern. The illumination was directed into a
two-layer 16-well sample holder, which was centered inside the
cylinder, and the distance between vials and LEDs was kept uniform
at 35 mm. Actinometry was carried out using the photooxidation of
[Ru(bpy)3]2+ by [S2O8]2−.30 A typical sample for photoreaction
contained 2.0 mL of MeCN, 1.0 mL of H2O, and 1.0 mL of THF with
concentrations of 260 μM for [Ru(bpy)3]Cl2, 0.05 M for tris(p-
tolyl)amine or p-(dimethylamino)toluene, 0.40 M for Et3N, and
25−100 μM for the Mo-based catalysts. Before illumination, samples
were thoroughly degassed by bubbling with N2 and sealed with
screwcaps having PTFE/silicone septa. The headspace volume was
4.7 mL for each sample. After irradiation for variable periods of time,
a 50 μL sample of gas was extracted and injected into a gas
chromatograph (Gow-Mac CG, molecular sieve column, T = 35 °C,
carrier gas = N2) for quantitative determination of the H2 produced.
The quantum yield for H2 production per absorbed photon was
determined as
Tetra-iso-butylthiuram Disulfide. Carbon disulfide (5.44 g,
0.071 mol) was added in rapid dropwise fashion to a solution of di-iso-
butylamine (18.5 g, 25.0 mL, 0.143 mol) in 100 mL of absolute
EtOH. This addition was attended by an exotherm of 10 °C (26 to 36
°C) and by the induction of a pale yellow color to the solution.
Stirring was continued for 1 h at ambient temperature, and the
solution was then warmed to 50 °C in a water bath. Solid I2 (9.02 g,
0.0355 mol) was added to the EtOH as a single portion, and its dark
color was rapidly extinguished to afford a pale yellow solution again.
Stirring was continued for a further 30 min, and the solvent was
removed under reduced pressure to afford a pasty, yellowish solid
residue. This residue was extracted with Et2O (3 × 100 mL), which
left behind a white microcrystalline mass of [iBu2NH2]I. The Et2O
extracts were filtered through paper, and the filtrate was taken to
dryness under reduced pressure. The resulting pale yellow solid was
dried under vacuum for 1 h. Yield: 13.64 g, 93%. Recrystallization
from a warm hexanes solution (1.2 g solid/10 mL hexanes) that was
slowly cooled to −20 °C yielded nearly colorless block-shaped crystals
(8.63 g). MP = 66−67 °C. 1H NMR (δ, ppm in DMSO-d6): 3.83 (d,
8H, −CH2CH(CH3)2), 2.39 (br m, 4H, −CH2CH(CH3)2), 0.98 (d,
12H, −CH2CH(CH3)2), 0.87 (d, 12H, −CH2CH(CH3)2) (Figures
S19−S22). 13C NMR (δ, ppm in CDCl3): 194.3, 65.4, 61.7, 29.0,
26.1, 20.4, 20.3 (Figure S23). Anal. Calcd for C18H36N2S4: C, 52.89;
H, 8.88; N, 6.85; S, 31.38. Found: C, 52.91; H, 8.43; N, 6.87; S,
30.15.
[Mo3S7(S2CNMe2)3]I, [2a]I. The procedure described by He-
getschweiler and co-workers was employed on scale involving 0.305 g
(0.411 mmol) of [NH4]2[Mo3S13] and 0.307 g (1.25 mmol) of
1
tetramethylthiuram disulfide. Yield: 0.219 g (0.219 mmol), 53%. H
NMR (δ, in CDCl3): 3.39 (s, 9H, N(CH3)), 3.34 (s, 9H, N(CH3))
(Figure S24). Absorption spectrum (CH2Cl2) λmax nm (εM): ∼360
(unresolved sh, ∼8000) (Figure S25). MS (MALDI-TOF): 865.302
(M+) (Figures S26−S27). Anal. Calcd for [Mo3S7(S2CNMe2)3]I·N,N-
DMF, C12H25IMo3N4OS13: C, 13.43; H, 2.35; N, 5.22. Found: C,
14.10; H, 2.38; N, 4.62.
[Mo3S7(S2CNEt2)3]I, [2b]I. The procedure described by Hegetsch-
weiler and co-workers was employed on scale involving 0.300 g (0.405
mmol) [NH4]2[Mo3S13]. Yield: 0.256 g (0.256 mmol), 58%. 1H NMR
(δ, ppm in DMSO-d6): 3.82 (q, 6H, J = 8 Hz, −S2CN(CH2CH3)2),
3.77 (q, 6H, J = 8 Hz, −S2CN(CH2CH3)2), 1.22, 1.21 (overlapping t,
18H, J = 8 Hz, −S2CN(CH2CH3)2) (Figures S28−S30). Absorption
spectrum (2:1 THF:H2O): 300−450 nm, unresolved trailing edge
(Figure S31). Raman spectrum, cm−1: 532.1 (m), 510.9 (s), 475.1 (s),
393.8 (w), 358.7 (m), 306.9 (m), 267.6 (s), 254.8 (s), 229.3 (m),
219.6 (vs), 152.2 (s), 105.9 (m), 61.6 (vs), 48.2 (vs), 31.4 (vs), 16.9
(vs) (Figure S32). MS (MALDI-TOF): 957.67 (M+), 861.61 ([M −
3S]+) (Figure S33).
ΦH = 2(moles H2 produced)/(moles photons)
2
= 2PVH2/(R·T·I·t)
where VH is the volume of H2 produced in the cell headspace, P is the
2
pressure in the headspace of the photolysis vial, R is the gas constant,
T is the temperature, I is the light intensity (quanta/s from
actinometry), and t is the irradiation time. Turnover numbers for
H2 production per catalyst were determined as
[Mo3S7(S2CNEt2)3][PF6], [2b][PF6]. The procedure and scale
reported for [Mo3S7(S2CNEt2)3]I were implemented but with NaPF6
used in place of NaI.33
[Mo3S7(S2CNiBu2)3]I, [2c]I. A 100 mL Schlenk flask with stir bar
was charged with [NH4]2[Mo3S13] (0.300 g, 0.405 mmol), tetra-iso-
butylthiuram disulfide (0.630 g, 1.54 mmol) and anhydrous N,N-
dimethylformamide. This mixture was stirred in the open air for 30
min and then warmed to 50 °C. A solution of NaI (2.00 g, 13.3
mmol) in 50 mL of absolute EtOH was added, which induced a color
change from dark red to light orange. Stirring and mild heating were
continued for a period of 12 h, at which point the mixture was allowed
to stand and cool overnight. A modest, yellow-white precipitate of
crystalline sulfur, identified by its unit cell, was observed. The
supernatant was filtered, and the volume of the filtrate was reduced to
∼15 mL. A 100 mL portion of CH2Cl2 was added to the concentrate,
which solubilized the target compound but precipitated the other
inorganic salts. These salts were removed by filtration, and the filtrate
was taken to dryness under reduced pressure. The orange solid
residue was redissolved in a minimal volume of 1,2-dichloroethane,
and the concentrated extract was filtered through a packed Celite pad.
Large, red-orange rectangular block crystals were obtained by the
diffusion of tBuOMe vapor into the 1,2-dichloroethane solution
TONH = (moles H2 produced)/(moles Mo‐catalyst)
2
= PVH /(R·T·nMo
)
2
where nMo is the number of moles of Mo-catalyst in each sample.
SYNTHESES
■
General Considerations. Solvents employed for synthesis were
dried with a system of drying columns from the Glass Contour
Company (CH2Cl2, THF, Et2O), freshly distilled according to
standard procedures (CH3CN, ClCH2CH2Cl),31 purchased in an
ultradry grade (N,N-dimethylformamide, EtOH), or simply used as
received from commercial sources (tBuOMe). Literature procedures
were used in the preparations of [NH4]2[Mo3S13],32
[Mo3S7(S2CNEt2)3]I,33 [Mo3S7(S2CNEt2)3][PF6],33 and Na[BArF]
([BArF]− = tetrakis(3,5-bis(trifluoromethyl)phenyl)borate(1−)),34
while [nBu4N]2[Mo3S13] was prepared from [NH4]2[Mo3S13] by
treatment with [nBu4N][HO].35 All other reagents from commercial
sources were used without further purification.
1
(Figure S34). Yield: 0.170 g, 0.118 mmol, 30%. H NMR (δ, ppm in
C
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Summary of Cluster Syntheses
DMSO-d6): 3.66 (d, 6H, −CH2CH(CH3)2), 3.61 (d, 6H, −CH2CH-
(CH3)2), 2.23 (m, 6H, −CH2CH(CH3)2), 0.905 (d, 18H, −CH2CH-
(CH3)2), 0.885 (d, 18H, −CH2CH(CH3)2) (Figure S35). 13C NMR
(δ, ppm in CD2Cl2): 204.7, 57.5, 58.0 27.7, 27.6, 20.3 (Figures S36−
S37). Absorption spectrum (CH3CN) λmax nm: ∼280 (unresolved
sh), ∼360 (unresolved sh), ∼420 (unresolved sh) (Figure S38). MS
(MALDI-TOF): 1125.748 (M+) (Figure S39). Anal. Calcd for
C27H54IMo3N3S13: C, 25.90; H, 4.35; N, 3.36; S, 33.29. Found: C,
26.14; H, 4.51; N, 3.33; S, 33.21.
[Mo3S4(S2CNEt2)3][BArF], [3][BArF]. A 50 mL Schlenk flask was
charged with [Mo3S7(S2CNEt2)3]I (0.100 g, 0.090 mmol), PPh3
(0.145 g, 0.55 mmol), and Na[BArF] (0.082 g, 0.090 mmol) under
an atmosphere of N2. Distilled acetonitrile (25 mL) was delivered to
the flask via syringe, and a color change from green to brown occurred
upon mixing. The reaction mixture was heated to 50 °C, stirred
overnight, and taken to dryness under reduced pressure. The solid
residue was washed with 5 × 6 mL portions of hexanes, which were
removed by filter cannulation. The remaining solid was extracted with
Et2O to afford a green-brown solution that was then separated from a
white solid, presumably NaI, via filter cannulation. The filtrate was
layered with hexanes in a Schlenk tube and left for 48 h, after which
time a deposit of bronze-colored needles was obtained. The identity
of [3][BArF] was confirmed by a mass spectrum (MALDI) and UV−
vis absorption spectrum that were the same as those for [3][PF6].
[Mo3S4(S2CNEt2)4], 4. Under an atmosphere of N2, a 50 mL
Schlenk flask was charged with [Mo3S7(S2CNEt2)3]I (0.100 g, 0.092
mmol), PPh3 (0.145 g, 0.552 mmol), Na(S2CNEt2)·3H2O (0.021 g,
0.093 mmol), and MeCN (25 mL). This dark mixture was heated to
60 °C and stirred for 12 h, after which time it was cooled to ambient
temperature without mechanical agitation. A dark red-black
precipitate was observed, which was collected by vacuum filtration
on a Hirsch funnel and washed with ∼10 mL MeCN followed by
copious amounts of Et2O to remove PPh3/SPPh3. Yield: 0.079 g, 86%.
1H NMR (δ, ppm in DMSO-d6): 4.18−4.09 (m, 6H, −CH2CH3),
3.76 (q, J = 8 Hz, 6H, −CH2CH3), 3.70 (q, J = 8 Hz, 4H, −CH2CH3),
1.42 (t, J = 8 Hz, 3H, −CH2CH3), 1.41 (t, J = 8 Hz, 6H, −CH2CH3),
1.17 (t, J = 8 Hz, 9H, −CH2CH3), 1.05 (t, J = 8 Hz, 6H, −CH2CH3)
(Figures S49−S51). Absorption spectrum (CH2Cl2) λmax nm (εM):
∼385 (sh, 6000), 480 (5000), ∼558 (sh, 2900) (Figure S52). MS
[Mo3S7(S2CNBn2)3]I, [2d]I. The same scale and procedure as
implemented for [Mo3S7(S2CNEt2)3]I (vide supra) were employed
1
here. Yield: 61%. H NMR (δ, ppm in DMSO-d6): 7.41−7.33 (m,
30H, −CH2C6H5), 5.27 (d, J = 18 Hz, 12H, −CH2C6H5) (Figure
S40). 13C NMR (δ, ppm in CD2Cl2): 206.0, 134.3, 134.2, 129.7,
129.6, 129.13, 129.06, 129.04, 128.7, 52.7, 52.3. (Figure S41).
Absorption spectrum (CH2Cl2) λmax nm: ∼340 (unresolved sh),
∼380 (unresolved sh) (Figure S42). MS (MALDI-TOF): 1329.527
(M+) (Figure S43). Anal. Calcd for C45H42IMo3N3S13: C, 37.11; H,
2.91; N, 2.89. Found: C, 36.89; H, 2.83; N, 2.38.
[Mo3S4(S2CNEt2)3][PF6], [3][PF6]. A 50 mL Schlenk flask was
charged with [Mo3S7(S2CNEt2)3][PF6] (0.100 g, 4.5 mmol) and
PPh3 (0.143 g, 27 mmol), and the solids were dissolved in MeCN (20
mL). A color change from light orange to dark reddish brown was
observed upon mixing, and the mixture was then heated to 50 °C
under N2. The reaction mixture was stirred overnight (∼12 h) and
produced a reddish-brown solid precipitate. After it was cooled to
ambient temperature, the solid was collected by gravity filtration
through a fluted filter paper and washed with MeCN, EtOAc, and
then Et2O. Yield: 0.040 g, 0.0397 mmol, 44%. 1H NMR (400 MHz; δ,
ppm in DMSO-d6): 4.14 (m, ∼2H, −CH2CH3), 4.07−3.98
(overlapping m, ∼5H, −CH2CH3), 3.76 and 3.70 (overlapping
quartets, 5H, −CH2CH3), 1.44−1.40 (t, ∼4H, −CH2CH3), 1.35 (t,
∼7H, −CH2CH3), 1.17 (t, ∼4H, −CH2CH3), 1.09−1.03 (over-
lapping triplets, ∼3H, −CH2CH3) (Figures S44−S46). Absorption
spectrum (DMF) λmax nm (εM): 430 (5000), ∼525 (sh, 2900) (Figure
S47). Raman spectrum, cm−1: 472.7 (vs), 431.8 (m), 217.9 (vs),
152.1 (s), 80.6 (m), 48.0 (s), 40.4 (s), 24.4 (m) (Figure S48). MS
(MALDI-TOF): 862.305 (M+).
−
(MALDI-TOF): 861.819, ([M − Et2NCS2 ]+) (Figure S53). Anal.
Calcd for C20H40Mo3N4S12: C, 23.80; H, 4.00; N, 5.55. Found: C,
24.33; H, 3.91; N, 5.48.
[Mo3S4(S2CNEt2)4(η1,κO-DMF)], 5. Recrystallization of
[Mo3S4(S2CNEt2)4] by diffusion of Et2O vapor into N,N-
dimethylformamide solution quantitatively afforded compound with
1
DMF as ligand. H NMR (δ, ppm in DMF-d7): 4.24−4.16 (m, 6H,
−CH2CH3), 3.87−3.78 (m, 10H, −CH2CH3), 2.96 (s, η1,κO-DMF),
2.79 (s, η1,κO-DMF), 1.48 (t, J = 8 Hz, 6H, −CH2CH3), 1.44 (t, J = 8
D
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 1. Selected Interatomic Distances (Å) and Angles (deg) for Cations [2]+ and Compound 4. Averaged Values Are
Presented for Distances and Angles That Are Chemically Identical
[2a]+
[2b]+
[2c]+
[2d]+
4
Mo−Mo
Mo−μ3-S
Mo−Sax
Mo−Seq
S−S
2.7162[2]
2.3801[4]
2.4081[3]
2.4854[3]
2.0515[5]
2.4822[4]
2.5194[4]
69.587[11]
68.658[11]
66.245[10]
85.12[1]
170.82[1]
49.540[8]
110.270[8]
85.415[8]
88.3
2.7173[2]
2.3725[5]
2.4156[4]
2.4814[4]
2.0553[8]
2.4678[6]
2.5198[6]
69.87[2]
68.45[2]
66.40[1]
85.47[2]
170.30[2]
49.61[1]
110.22[1]
85.16[1]
89.0
2.7190[3]
2.3807[6]
2.4101[8]
2.4855[4]
2.0573[7]
2.4709[6]
2.5145[6]
69.65[2]
68.68[2]
66.32[2]
85.16[2]
170.52[2]
49.67[1]
110.23[1]
85.26[1]
88.5
2.7160[2]
2.3749[3]
2.4103[2]
2.4854[3]
2.0603[5]
2.4743[4]
2.5228[3]
69.753[11]
68.584[17]
66.239[10]
85.19[1]
170.44[1]
49.745[8]
110.228[8]
85.222[8]
88.2
Mo(1)−Mo(2)
Mo(3)−Mo(2)
Mo(1)−Mo(3)
Mo(2)−μ3-S
Mo(1)−μ3-S, Mo(3)−μ3-S
Mo(2)−μ2-S
2.7502(8)
2.7516(8)
2.7050(8)
2.3493(15)
2.3374[11]
2.2496[11]
2.3054[11]
2.3358[11]
2.4196(16)
2.5154(16)
2.5149[11]
2.5299[11]
2.5565[11]
71.84(4)
b
c
Mo−Sdtc,syn
Mo−Sdtc,anti
d
e
Mo(1)−μ2-S, Mo(3)−μ2-S
Mo(1)−μ2-S, Mo(3)−μ2-S
Mo−μ3-S−Mo
Mo−Sax−Mo
Mo−Seq−Mo
Sax−Mo−Sax
Seq−Mo−Seq
Seq−Mo−Sax
μ3-S−Mo−Sax
μ3-S−Mo−Seq
b
Mo(2)−Sdtc,syn
c
Mo(2)−Sdtc,anti
bf
,
Mo−Sdtc,syn
Mo−Sdtc,anti
cf
,
g
Mo−Sdtc,bridging
Mo(1)−S(4)−Mo(3)
h
Mo−μ2-S−Mo
θ1, θ2, θ3
71.89[3]
77.1, 60.9, 44.3
i
j
k
l
θ
a
For averaged values, uncertainties are determined using the general formula for error propagation as described by Taylor50 and are enclosed with
b
c
square brackets. Mo−S bond length for dithiocarbamate sulfur atom on the same side of the Mo3 plane as the μ3-S ligand. Mo−S bond length for
d
dithiocarbamate sulfur atom on the opposite side of the Mo3 plane as the μ3-S ligand. Average Mo−μ2-S distance along edge with bridging
e
f
dithiocarbamate ligand. Average of Mo(1)−μ2-S and Mo(3)−μ2-S distances along edges without bridging dithiocarbamate ligand. Average of two
g
h
values for chelating dithiocarbamate ligands bound to Mo(1), Mo(3). Average Mo−S bond length for bridging dithiocarbamate ligand. Average
i
j
for the two edges without bridging dithiocarbamate ligand. Average angle between Mo3 and S2CN planes. Angle between Mo3 plane and S2CN
k
plane for dithiocarbamate chelated to Mo(2). Average angle between Mo3 plane and S2CN plane for dithiocarbamate ligands chelated to Mo(1),
Mo(3). Angle between Mo3 plane and S2NC plane for bridging dithiocarbamate.
l
Hz, 3H, −CH2CH3), 1.23 (t, J = 8 Hz, 9H, −CH2CH3), 1.10 (t, J = 8
Hz, 6H, −CH2CH3). Raman spectrum, cm−1: 993.9 (vs), 820.1 (vs),
668.4 (m), 662.4 (m), 376.3 (m), 333.9 (m), 281.1 (s), 236.4 (m),
193.6 (m), 145.0 (s), 120.6 (s), 110.3 (s), 89.6 (m), 76.6 (m) (Figure
S54).
[2c]I, the workup and isolation steps are substantially modified
because its greatly enhanced solubility does not permit an
immediate, convenient precipitation from the reaction mixture.
Characterization of [2a]I−[2d]I was undertaken to bring a
more clear focus to the properties inherent to the compound
type. Structural identification by X-ray crystallography reveals a
[Mo3(μ3-S)(μ2-S2)3]4+ core that is highly similar in its metrical
details across the set (Table 1, Figures S3−S8, S10−S11).
[(iBu2NCS2)3(μ2-S2)2(μ3-S)Mo3](μ3-S)(μ3-η2,η1-S′,η1-S″-S2)-
[Mo3(μ2-S)3(μ3-S)(S2CNiBu2)2(μ2-S2CNiBu2)], 6. A photolysis sol-
ution of [Mo3S7(S2CNiBu2)3]I was composed as described earlier but
with 1:9 H2O:MeCN. Slow cooling following the typical 3 h
photolysis yielded small but reproducible amounts of crystalline 6·
2.5MeCN (Figure S55). Absorption spectrum (Et2O) nm: ∼320 (sh),
∼480 (sh) (Figure S56). MS (MALDI-TOF): 1984.34 ([M −
Previous X-ray diffraction studies of halide salts of
33,40−46
[Mo3S7(S2CNEt2)3]+
were conducted at room temper-
ature and consequently are lower in their resolution than the
structure reported here. A distinctive feature of these structures
is the close positioning of the soft I− counteranion on the
underside of the cluster in proximity to the axial sulfur atoms
of the bridging disulfide ligands (Figure 2). As observed in
recent reviews specific to [Mo3S7]4+ clusters,47,48 such close
interactions arise from an emphatic electrophilic character to
the Sax atoms. Other noteworthy structural characteristics are
(1) near orthogonality of the NCS2 atom set of the
dithiocarbamate ligands with the Mo3 plane; (2) moderate
−
iBu2CNS2 ]+), 1125.75 ([M − Mo3S5(S2CNiBu2)3]+) (Figures S57−
S58).
RESULTS AND DISCUSSION
■
Syntheses and Structures. While simple tetraalkylthiur-
am disulfides, which are oxidized forms of dithiocarbamate
anions, are widely supplied and inexpensive, variants such as
iBu2NC(S)SSC(S)NiBu2 that are less available commercially
are nonetheless readily prepared. Procedures describing the
synthesis of iBu2NC(S)SSC(S)NiBu2 from the reaction
36−38
i
between Bu2NH and CS2 in aqueous base
have, in our
hands, been found to be attended by significant competing
reaction pathways. This observation motivated our implement-
ing a clean, effective modification that simply employs excess
iBu2NH in CS2/EtOH (Scheme 2). Recrystallization of
iBu2NC(S)SSC(S)NiBu2 from hexanes produced a monoclinic
form (Figure S2, Table S1) with structural features (e.g., C−
S−S−C torsion angles of −89.1°, 87.8° for two independent
molecules) otherwise similar to those of a triclinic polymorph
obtained earlier by evaporation of a solution in CH2Cl2.39
The synthesis of [Mo3S7(S2CNR2)3]I proceeds readily from
R2NC(S)SSC(S)NR2 and Mu
̈
ller’s classic [Mo3S13]2− anion32
Figure 2. Illustration of the structural distinction between equatorial
and axial sulfur in μ-S2(2-) ligands. The former are removed by
nucleophilic abstraction to produce [Mo3S4]4+ voided cubanes.
(Scheme 2), a procedure that was first reported for
[Mo3S7(S2CNEt2)3]I.33 In the case of [Mo3S7(S2CNiBu2)3]I,
E
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Thermal ellipsoid plots of [Mo3S7(S2CNMe2)3]+ (a), [Mo3S7(S2CNiBu2)3]+ (b), [Mo3S4(S2CNEt2)4] (c), and
[Mo3S4(S2CNEt2)4(η1,κO-DMF)] (d) at the 50% probability level. All H atoms are removed for clarity. An iso-butyl group in
[Mo3S7(S2CNiBu2)3]+ (top right) is disordered over two positions.
lengthening for the Mo−Sdithiocarbamate bond that is anti to the
μ3-S2− ligand; and (3) appreciable lengthening by ∼0.08 Å for
the Mo−Seq bonds, as compared to the Mo−Sax bonds (Table
1, Figure 3).
to clean crystallization, possibly because the halide counter-
anion can no longer pack in the same felicitous way near the
S2− ligands as with the S2 ligands. Introduction of a fourth
2−
dithiocarbamate ligand leads to a charge-neutral cluster, 4, with
one edge to the Mo3 triangle bridged and one MoIV ion left
five-coordinate, as confirmed by structural identification of
crystals grown from noncoordinating CH2Cl2 (Figure 3(c)). A
related [Mo3S4]4+ cluster of the same structure type has been
A defining reactivity for clusters with the [Mo3(μ3-S)(μ2-
S2)3]4+ core is facile excision of the equatorial sulfur atom from
2−
the μ2-S2 ligands (Figure 2) by nucleophiles such as
phosphines or NC−, thereby transforming the cluster core
to [Mo3(μ3-S)(μ2-S)3]4+ by reduction of ligand rather than
metal (Scheme 2).49 This change is marked by a darkening in
color from orange-red to dark maroon. Although it appears to
have good chemical stability in the open air, the [Mo3(μ3-
S)(μ2-S)3(S2CNEt2)3]+ cation has proven decidedly intractable
described with Et2PS2 as supporting ligand.51 Noteworthy
1−
aspects of the structure of 4 are a pronounced shortening of
the Mo−μ2-S bond lengths by ∼0.11 Å, as compared to the
Mo−Sax bond lengths in cations [2]+ (Table 1), and significant
twisting of the chelating dithiocarbamate ligands from
F
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4. (a) Photocatalytic generation of H2 (μmol H2 vs time) by [nBu4N]2[Mo3S13] and [Mo3S7(S2CNEt2)3]I in the system of Scheme 1. (b)
Photocatalytic generation of H2 (μmol H2 vs time) by [Mo3S7(S2CNEt2)3]I as a function of catalyst concentration. (c) Turnover number (mol H2/
mol catalyst vs time) for [Mo3S7(S2CNEt2)3]I as a function of catalyst concentration. (d) Photocatalytic generation of H2 (μmol H2 vs time) by
[Mo3S7(S2CNiBu2)3]I as a function of catalyst concentration.
orthogonality with respect to the Mo3 plane. The bridging
dithiocarbamate ligand is folded by 44.3° from the Mo3 plane,
while its Mo−S distances are longer by 0.03−0.14 Å than those
of the other dithiocarbamate ligands and reflect the trans
influence of the μ2-S2− ligands (Table 1).
Although it lacks the terminal disulfide ligands that, in some
molecular systems, appear to be involved in the mechanism of
H2 evolution, [Mo3S7(S2CNEt2)3]I was surveyed for its activity
under the same conditions implemented for
[nBu4N]2[Mo3S13]. Over the same time frame (Figure 4(a)),
this cation demonstrated a 10-fold greater activity than
[Mo3S13]2−, producing some 100 turnovers of H2 without
the same conspicuous dwindling in activity after 1 h. Several
characteristics of [Mo3S7(S2CNEt2)3]I may contribute to its
appreciably greater activity and lifetime. First, the absence of a
dinegative charge, as found in [Mo3S13]2−, likely enables the
infusion of the necessary reducing equivalents at milder
cathodic potential. This cation is also lighter in color and
produces less competing absorption with the [Ru(bipy)3]2+
chromophore at 460 nm (Figure S31). In a series of catalytic
runs in which [Mo3S7(S2CNEt2)3]I concentration was varied
from 25 to 100 μM, H2 quantities increased with increasing
concentration (Figure 4(b)). However, catalyst efficiency as
gauged by turnover numbers was maximized at catalyst
concentration of ∼50 μM (Figure 4(c)), reflecting the fact
that the particular photosystem used is not producing
[Ru(bpy)3]+ at a rate that keeps pace with turnover of
available catalyst.
The slow diffusion of various ethers into N,N-dimethylform-
amide solutions of [Mo3S4(S2CNEt2)4] produced several
crystalline forms of [Mo3S4(S2CNEt2)4(η1,κO-DMF)] in
which the solvent is coordinated to the unique MoIV ion.
One form is without interstitial solvent in the unit cell, while
the others differ in the identity of solvent molecules
incorporated into the lattice. Other polar, donor solvents
such as THF, MeCN, and DMSO showed no similar tendency
to coordinate to the cluster, although a structure of
[Mo3S4(S2CNEt2)4(py)]·2py·H2O has been described,52 and
synthesis of a H2O-ligated form, [Mo3S4(S2CNEt2)4(H2O)],
has been reported from [Mo3S4(S2P(OEt)2)4(H2O)] by direct
Et2NCS21−-for-(EtO)2PS2 ligand substitution.53
1−
Photocatalytic H2 Evolution. The marginal solubility of
[NH4]2[Mo3S13] even in strongly polar organic solvents
necessitated its conversion to the more tractable [nBu4N]+
salt. In a mixed 1:2:1 H2O/MeCN/THF medium at 35 μM
concentration, [nBu4N]2[Mo3S13] supported only modest
turnovers of ∼10 for H2 formation over the course of several
hours of photolysis (Figure 4(a)). Apparent deactivation, as
indicated by leveling of the curve, occurs after ∼60 min.
Higher concentrations of [nBu4N]2[Mo3S13] beyond 50 μM
failed to produce proportionately greater amounts of H2(g).
In the course of some photolysis runs with
[Mo3S7(S2CNEt2)3]I, a turbid aspect to the solution was
observed that indicated a departure from complete solution
homogeneity in the solvent mixture. Recourse was therefore
made to the much more soluble [Mo3S7(S2CNiBu2)3]I to
G
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. (a) Turnover number (mol H2/mol catalyst vs time) for [Mo3S7(S2CNiBut2)3]I as a function of catalyst concentration. (b)
Photocatalytic generation of H2 (μmol H2 vs time) by [Mo3S7(S2CNEt2)3]I as compared to [Mo3S4(S2CNEt2)4]. (c) Hydrogen generated in
photolyses of [Mo3S7(S2CNEt2)3]I vs [Mo3S4(S2CNEt2)3][BArF] in the early first moments of irradiation. (d) Photocatalytic generation of H2
(μmol H2 vs time) by [Mo3S7(S2CNEt2)3]I vs [Mo3S4(S2CNEt2)4] in 3:1 N,N-dimethylformamide:H2O.
ensure that all species remained fully dissolved during active
photolysis. An approximate quintupling of H2-evolving
catalytic activity was observed for [Mo3S7(S2CNiBu2)3]I, as
compared to [Mo3S7(S2CNEt2)3]I, with some 300 turnovers
over a 3 h period under identical photolysis conditions (Figure
5(a)). Quantum efficiencies, defined as moles of stored
electrons per mole of photoproduced electrons, measured for
the two catalysts at 100 μM catalyst and 100 min irradiation,
were 0.10 for the diethyl catalyst and 0.42 for the di-iso-butyl
complex. For all the catalysts examined, the quantum
efficiencies were time dependent and reached a maximum
typically around 30 min into a photolysis with the photosystem
used. These results clearly indicate that at least half of the
photogenerated reducing equivalents are reacting in ways that
do not involve hydrogen production. For the isobutyl complex,
no cloudiness was observed at any point during experimenta-
tion. Tightly reproducible data were obtained with multiple,
successive photolyses with [Mo3S7(S2CNiBu2)3], affirming
measurement reliability. We are inclined to the view that the
greater activity found for [Mo3S7(S2CNiBu2)3]I vs
[Mo3S7(S2CNEt2)3]I is attributable to enhanced solubility
rather than to intrinsically greater activity or a different
mechanism of action.
signal of appreciable intensity is observed at 861.54 mass units
in conjunction with the peak for [Mo3S7(S2CNEt2)3]+ at
957.51 amu (Figure 6). The mass at 861.54 is attributable to
[Mo3S4(S2CNEt2)3]+ formed by loss of the three equatorial
sulfur atoms of the disulfide bridging ligands. Additional assays
over the ensuing few moments reveal rapid growth of the
861.54 signal with corresponding diminution of the 957.51
signal such that the latter is completely gone after three
minutes (Figure 6, lower right panel). Peaks corresponding to
the intermediates with one and two sulfur atoms extruded are
observed (cf. panel at 1 min) but without significant signal
intensity, implying an accelerated loss of the second and third
sulfur atoms to form [Mo3S4(S2CNEt2)3]+ once the first is lost.
Importantly, mass spectrometry (MALDI) upon solutions of
analytically pure samples of [Mo3S7(S2CNEt2)3]I reveals a
prominent peak for [Mo3S7(S2CNEt2)3]+ with the correct
isotopic distribution profile and only a minor fragmentation
peak corresponding to [Mo3S4(S2CNEt2)3]+ with intensity that
is invariant over time (Figure S33). Thus, the
[Mo3S4(S2CNEt2)3]+ that is observed to grow in the photolysis
cell is authentic to the sample and not merely a fragment from
the parent [Mo3S7(S2CNEt2)3]+ in the mass spectrometer.
The implication from the time-dependent mass spectrom-
etry data summarized in Figure 6 is that [Mo3S7(S2CNEt2)3]I
is a precursor that readily gives way to [Mo3S4(S2CNEt2)3]I,
which suggests itself then as being the functional catalyst. This
supposition appears reasonable in light of several reports
As a means of gauging catalyst integrity during the course of
photolysis, a series of mass spectrometry assays was undertaken
of the catalytic mixture with [Mo3S7(S2CNEt2)3]I. At a time
interval as early as 30 s from the initiation of photolysis, a
H
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. Time-dependent mass spectrometric assay (MALDI-TOF) of a photolysis run using [Mo3S7(S2CNEt2)3]I. At 30 s, an appreciable
amount of [Mo3S4(S2CNEt3)3]+ has formed (upper left). At 3 min, the original [Mo3S7(S2CNEt2)3]+ cation at 957 amu is no longer detected
(lower right). At 1 min, minor peaks between A and B, corresponding to clusters with the [Mo3S6]4+ and [Mo3S5]4+ core structures, are visible.
Figure 7. Thermal ellipsoid plot of [(iBu2NCS2)3(μ2-S2)2(μ3-S)Mo3](μ3-S)(μ3-η2,η1-S′,η1-S″-S2)[Mo3(μ2-S)3(μ3-S)(S2CNiBu2)2(μ2-S2CNiBu2)]
(6) at the 50% probability level. All hydrogen atoms are removed for clarity.
describing the H2 -evolving catalytic ability of
[Mo3S4(H2O)9]4+ adsorbed onto electrocathodes.54−56 How-
ever, other key observations point toward more complex
solution processes in which rapidly generated
I
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
[Mo3S4(S2CNEt2)3]I participates in the formation of other
species. Importantly, when deliberately prepared and isolated
s a m p l e s o f [ M o 3 S 4 ( S 2 C N E t 2 ) 3 ] [ A n i o n ] o r
[Mo3S4(S2CNEt2)4] are employed in photolysis runs under
the same conditions as used for [Mo3S7(S2CNEt2)3]I, an
emphatically lowered activity level is found (Figure 5(b)).
Furthermore, rapid assays of H2 in the early stages of
photolysis of both [Mo3 S7 (S2 CNEt2 )3 ]I and
[Mo3S4(S2CNEt2)3][BArF] (Figure 5(c)) reveal that the latter
immediately catalyzes H2-formation at constant rate while the
former undergoes a brief, but distinctive, incubation period
with a delayed onset of H2 evolution. This H2 evolution rate
quickly overtakes and surpasses H2 evolution by
[Mo3S4(S2CNEt2)3][BArF]. The conclusion to which these
observations lead is that, while photolysis of both
[Mo3S7(S2CNEt2)3]I and [Mo3S4(S2CNEt2)3][BArF] solu-
tions leads to H2 evolution, either the catalyst identity is
different in the two cases or the ability to form the catalytically
active species differs.
A systematic effort to optimize the experimental conditions
for photolysis included an examination of the effect of N,N-
DMF/H2O as medium. Although DMF has good solubilizing
power for these clusters, it proved to have a striking inhibitory
effect upon H2-evolution. In 1:2:1 DMF:MeCN:H2O, H2
generated by [Mo3S7(S2CNEt2)3]I was attenuated to 33% of
the level observed when 1:2:1 THF:MeCN:H2O was
employed. In 3:1 DMF:H2 O, H2 evolution by
[Mo3S7(S2CNEt2)3]I was further decreased to ∼8% of the
activity found in 1:2:1 THF:MeCN:H2O, while for
[Mo3S4(S2CNEt2)4] catalysis was completely suppressed
(Figure 5(d)). It should be noted that the solubility of H2 in
MeCN and DMF is very similar at 298 K, so it is unlikely that
the absence of H2 in the headspace of photolyses in DMF
containing solutions is due to differences in the amount of
dissolved H2.57 Taken in conjunction with crystal structures
showing DMF as a ligand to the unique MoIV ion in
[Mo3S4(S2C2NEt2)4] (vide supra), the impeding effect exerted
by DMF on H2 evolution strongly suggests that a vacant
coordination site at a MoIV center is required for catalysis. By
virtue of being a good ligand and present at high
concentration, DMF effectively removes the availability of
the coordination site.
Figure 8. Structure of [Mo3S7(S2CNEt2)3]2(μ-S).
sometimes referred to as an “A-type” structure,59 occurs with
some frequency in transition metal sulfide clusters.59−66
Disposed at an angle of ∼11.9°, the two triangular [Mo3]
fragments are closer to planar than otherwise. The resulting
distinctive [[Mo3S5](μ3-S)(μ3-η2,η1-S′,η1-S″-S2)[Mo3S4]]6+
core topology that is produced in 6 has been found only
once previously in a mixed dithiophosphate/carboxylate
cluster.67
As noted earlier, a defining characteristic of the [Mo3S7]4+
clusters is an electrophilic quality to the Sax atoms of the μ-S2
2−
ligands that is evident in their pronounced tendency to closely
associate with soft anions. Upon transformation from the
[Mo3S7]4+ structure to a [Mo3S4]4+ core, the Sax2− ligands, now
2−
fully reduced relative to the original S2 ligands, possess a
pronounced nucleophilicity that is clearly demonstrated by a
propensity for [Mo3S4]4+ clusters to bind a range of cations to
afford heterometal cubanes.68−70 The separate [Mo3S7]4+ and
[Mo3S4]4+ fragments of 6 therefore have a reciprocal nature.
This complementarity is manifested in a crystal packing
arrangement that juxtaposes pairs of molecules in a parallel
planar, “head-to-tail” fashion. The electrophilic Sax atoms of the
In photolysis studies with the much more soluble
[Mo3S7(S2CNiBu2)3]I, cells left to stand and cool overnight
following photolysis deposited well-formed, needle-shaped
orange crystals. An X-ray diffraction data set revealed this
compound to be a hexa-molybdenum cluster (6 in Scheme 2,
F i g u r e 7 ) a s y m m e t r i c a l l y c o n s t i t u t e d f r o m
[Mo3S7(S2CNiBu2)3]+ and [Mo3S4(S2CNiBu2)3]+ fragments
with one additional μ3-S2− anion that renders the entire
assembly charge neutral. Compound 6 contrasts sharply with
hexanuclear [Mo3S7(S2CNEt2)3]2(μ-S), which may be de-
scribed as two [Mo3S7(S2CNEt2)3]+ cations held by a S2−
counteranion via weak Sax···μ-S interactions (Figure 8).58 In 6,
S2 ligands of the [Mo3S7]4+ fragment of one molecule are
2−
2−
thus presented toward the nucleophilic Sax atoms of the
[Mo3S4]4+ fragment of the other, thereby producing multiple
intermolecular S···S close contacts in the range of 3.00−3.39 Å
(Figure 9), which are appreciably shorter than twice the van
der Waals radius of sulfur.71 This centrosymmetric pairing of
molecules, in conjunction with overall charge-neutrality and a
i
−
hydrophobic periphery of Bu2NCS2 ligands, provides for
good ordering and growth of high-quality crystals from this
relatively polar mixed MeCN−H2O medium.
2−
one S2 ligand from the [Mo3S7(S2CNiBu2)3]+ portion is
rotated ∼90° from the original position orthogonal to the Mo3
The presence of 6 was not noted in time-dependent mass
spectrometry assays because its charge neutrality does not
enable it to fly well as an intact molecule. Instead, only
fragments of 6 (vide supra) are observed when isolated samples
are examined by the MALDI technique in positive ion mode.
Nevertheless, the chemical significance of 6 is that it strongly
suggests itself as the active catalyst in photolyses beginning
with [Mo3S7(S2CNR2)3]I and provides plausible explanation
plane such that it bridges to one molybdenum ion of the
i
1−
[Mo3S4]4+ moiety. Further, one of the three Bu2NCS2
ligands of the [Mo3S4(S2CNiBu2)3]+ fragment is moved from
a chelating mode to a bridging mode between two MoIV ions
such that coordination sites are opened through which
interactions with μ3-S2− and μ3-S2 ligands are then made.
2−
The μ3-η2,η1-S′,η1-S″ binding motif of this disulfide ligand,
J
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
the μ3-S2− ligand. The diminished activity found for
[Mo3S4(S2CNR2)3][Anion] is attributed to the cluster “as
is.” (3) For both [Mo3S4(S2CNR2)3]+ and hexamolybdenum
cluster 6, H2 formation demands the availability of a vacant
coordination site at molybdenum. The greater suppression of
activity by DMF upon [Mo3S4(S2CNR2)3]+, as compared to 6,
may arise in part from its positive charge and smaller size,
which would sharpen its Lewis acidic character and strengthen
the MoIV···OC(H)NMe2 interaction. In contrast, the
charge-neutral character of 6 would tend to attenuate its
Lewis acidic character and dampen the effect of solvent
interaction. Further, if an open coordination site in 6 is
rendered available for catalysis by dissociation of one end of
1−
the μ-S2CNiBu2 ligand (cf. the Mo−Sdtc,bridging bond lengths,
Table 1, which are appreciably longer than the other Mo−Sdtc,
bond lengths), then its hemilabile character could offset the
inhibiting effect of coordinating solvent by continuous
displacement of it.
ELECTROCHEMISTRY
■
Figure 9. Thermal ellipsoid plot (50%) of juxtaposed molecules of 6
illustrating the close intermolecular Sax2−···S22− contacts. Sulfur−sulfur
close contacts involving dithiocarbamate ligands are also apparent.
The cyclic voltammogram of [Mo3S7(S2CNEt2)3]I reveals an
irreversible reduction with current maximum at −0.78 V that is
followed by a reversible reduction at −1.30 V, the current
amplitude for which is somewhat smaller than that for the
preceding feature (Figure 10(a)). Further reduction, irrever-
sible in nature, occurs with a cathodic maximum at −1.93 V
(Figure S59). Under the same conditions, the cyclic
voltammogram of [Mo3S7(S2CNiBu2)3]I shows qualitative
similarity to that of [Mo3S7(S2CNEt2)3]I (Figure 10(b)) but
with a conspicuous diminishment of current for the second,
reversible feature. The resolution of this second wave shows an
apparent sensitivity to concentration. At 0.50 mM, Ea for the
return scan is scarcely discernible, but at 2 mM, a current
maximum can be identified (Figure S60). Owing to the
irreversible nature of the preceding reductions, these second
i
For clarity, some dithiocarbamate Bu groups have been truncated.
for our collected observations of the H2-evolving behavior of
[Mo3S7(S2CNR2)3]I versus [Mo3S4(S2CNR2)3][Anion]/
[Mo3S4(S2CNR2)4]: (1) Photolyses of [Mo3S7(S2CNiBu2)3]I
provide both of the necessary [Mo3S7(S2CNiBu2)3]+ and
[Mo3S4(S2CNiBu2)3]+ fragments, as the latter is readily
generated from the former by facile extrusion of sulfur
which, under the photoreducing conditions, is transformed
to S2− for bridging the Mo3S7 and Mo3S4 fragments. (2)
Photolyses of isolated [Mo3S4(S2CNR2)3][Anion] cannot yield
the hexamolybdenum cluster, as no exogenous S0/Sx can be
n−
made available to provide for the [Mo3S7]4+ fragment or for
Figure 10. Cyclic voltammetry in DMF at 100 mV/s of [Mo3S7(S2CNEt2)3]I (a), [Mo3S7(S2CNiBu2)3]I (b), and [Mo3S4(S2CNEt2)3][PF6] (c).
Panel (d) shows the Faradaic yield for H2 in controlled potential bulk electrolysis of [Mo3S7(S2CNEt2)3]I at −1.60 V.
K
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 11. Calculated B3LYP/BS1 MO energy level diagram for [Mo3S7(S2CNEt2)3]+. Orbital images are rendered at the 0.05 contour level.
reduction features necessarily arise from species that are
different either structurally or compositionallyor both
f r o m t h e o r i g i n a l [ M o 3 S 7 ( S 2 C N E t 2 ) 3 ] I a n d
[Mo3S7(S2CNiBu2)3]I.
sibility of such a process is plausible, as S1− would likely either
rapidly disproportionate to S0 and S2− or rapidly dimerize to
2−
S2
.
Although [Mo3(μ3-S)(μ2-S2)3−x(μ2-S)x(S2CNEt2)3]+ (x = 1
Complex, irreversible behavior upon cathodic scanning has
been noted for related compounds bearing the [Mo3(μ3-S)(μ2-
S2)3]4+ core and has been attributed to facile extrusion of sulfur
from the bridging disulfide ligands.72−74 However, the
reversible wave at ∼ −1.30 V does not result from
[Mo3S4(S2CNEt2)3][PF6] or from [Mo3S4(S2CNEt2)4]. The
cyclic voltammetry of these compounds, performed upon
deliberately prepared and isolated samples, shows distinctly
different behavior with the initial cathodic feature being a
reversible wave at −0.83 V (Figure 10(c)). The similarity in
potential for the first reductions in [Mo3S4(S2CNEt2)3][PF6]
and [Mo3S4(S2CNEt2)4] (Figure S61) may arise from close
association of the [PF6]− anion (also present from the
electrolyte) with [Mo3S4(S2CNEt2)3]+, such that its cationic
charge is mitigated and it acts more as a charge-neutral species,
or 2) would be relatively reactive owing to the nucleophilicity
of the μ2-S2− ligand(s), several considerations suggest one or
the other or a mixture of both as the species formed after the
irreversible cathodic wave at ∼ −0.75 V: (1) At the 30 and 60 s
time intervals, the mass spectrometry assays of photolysis
solutions of [Mo3S7(S2CNEt2)3]+ show minor peaks corre-
sponding to both [Mo3(μ3-S)(μ2-S2)2(μ2-S)(S2CNEt2)3]+ and
[Mo3(μ3-S)(μ2-S2)(μ2-S)2(S2CNEt2)3]+ (Figure 6). These
peaks are not observed in the mass spectrum (MALDI-TOF)
of an unphotolyzed sample of [Mo3S7(S2CNEt2)3]+ (Figure
S33), suggesting therefore that these species are authentic to
the solution and its reducing environment and not mere
fragments formed in the spectrometer. (2) An ESI mass
spectrometry study of [Mo3(μ3-S)(μ2-S2)3(bdt)3]2− (bdt =
benzene-1,2-dithiolate(2−)),75 an anion isoelectronic with
or from dissociation of the μ-Et2NCS2 ligand (cf. its longer
[Mo3(μ3-S)(μ2-S2)3(S2CNEt2)3]+ shows a prominent peak
−
,
bond lengths to Mo, Table 1), such that it is identical to
[ M o 3 S 4 ( S 2 C N E t 2 ) 3 ] [ P F 6 ] i n s o l u t i o n . F o r
[Mo3S4(S2CNEt2)3][PF6], an additional reduction, irreversible
in nature, is observed at ∼ −1.7 V (Figure 10(c)).
corresponding to [Mo3(μ3-S)(μ2-S2)(μ2-S)2(bdt)3]2− formed
via collision induced dissociation with Ar gas. Although
detected under nonreducing conditions, the observation of
the [Mo3S5]4+ species implies a modest stability that could
enable its transient appearance in other reaction settings. (3)
In a prior instance,76 an [Mo3S6]4+ fragment has been trapped
as an adduct with a [Mo3NiS4] cluster by a dative Sax → Ni
A
gas phase DFT geometry optimization of
[Mo3S7(S2CNEt2)3]+ proceeds readily with commonly im-
plemented basis sets and level of theory and produces good
agreement with the observed structure. The HOMO for
[Mo3S7(S2CNEt2)3]+ is a doubly degenerate set of MOs with
both Mo−Mo σ-bonding and Mo d−S p (dithiocarbamate) π*
character, while the LUMO and LUMO+1 are essentially
degenerate despite corresponding to different symmetry
species (Figure 11). One is predominantly metal-based
2−
interaction that is reinforced with S2−···S2 interactions
analogous to those found for 6 (Figure 12). Indeed, instead
of being described as composed of [Mo3S7]+ and [Mo3S4]+
fragments with a charge-balancing S2−, 6 itself could be
alternatively viewed as a [Mo3S6]+ cluster and joined to a
2−
[Mo3S4]+ cluster with an S2 ligand capping the assembly.
(70%), being the σ* combination of dxy, dx −y orbitals, while
the other is largely comprised of the σ* combination of sulfur p
orbitals on the μ-S22− ligands. The infusion of electron density
Invoking the transient existence of [Mo3S6]+ provides a
common rationalization to both systems despite their
ostensible dissimilarities.
2
2
2−
into the latter MO would promote fragmentation of a S2
Following the irreversible process or processes at ∼ −0.7 V,
ligand, presumably affording [Mo3(μ3-S)(μ2-S2)2(μ2-S)-
continued cathodic scanning of solutions of
(S2CNEt2)3]+ and S1− as immediate products. The irrever-
[Mo3S7(S2CNR2)3]I (R = Me, Et, Bu, Bn) shows a reversible
i
L
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Faradaic efficiency in the [Mo3S7(S2CNiBu2)3]I system, as
compared to that for [Mo3S7(S2CNEt2)3]I, is not clear.
MECHANISM
■
The elimination of H2 from the molybdenum sulfide catalyst,
irrespective of its identity, is preceded by reduction twice and
protonation twice, the ordering of which admits of a variety of
possible permutations. Although the reducing power (−1.4 V)
of the one electron reduced chromophore, [Ru(bipy)3]+, is
strong enough to deliver only one electron to
[Mo3S4(S2CNR2)3]+ (Figure 9(c)), it is plausible that
hexamolybdenum cluster 6 could sustain two reductions
because it is in effect two Mo3 clusters and its larger size
would render it a larger electron sink. All molybdenum ions in
6 occur in a six-coordinate environment of sulfur. The
[Mo3S4]4+ fragment in 6 (Figure 7, right side) bears the
i
1−
compound’s only bridging Bu2NCS2 ligand, which has
substantially longer Mo−S bond lengths than the chelating
dithiocarbamate ligands. Taken with the known hemilabile
character of dithiocarbamates77 and related ligands,78 which
readily enables a bidentate-to-monodentate shift in binding,
this observation suggests that, following reduction, a
coordination site in 6 could become available by dissociation
of one end of the dithiocarbamate ligand that bridges Mo(5)
and Mo(6) (Figure 7). Subsequent protonation steps leading
to H2 generation from [6]2− would likely be similar to those
that would occur with [3]1− and [4]2−, were they to exist.
Assuming [3]+ to be pertinent as a model for the catalytically
operative fragment of 6, computational assessment of the
lowest energy site for initial protonation of [3]1− identifies a 5-
coordinate MoIV site as energetically preferable to μ2-S2− by
6.49 kcal/mol (Scheme 3). A second protonation then occurs
at an adjacent μ-S2− ligand. Relative to this cis-MoIV(H)(μ-SH)
intermediate, a modest transition state of +6.7 kcal/mol that
brings an incipient H···H interaction is traversed and followed
by release of H2 with a calculated exotherm of −32.1 kcal/mol
(Scheme 3). Regardless of the specific sequence of e−
transfer−H+ transfer events, the result of two electron
reduction and double protonation is implicated as being cis-
MoIV(H)(μ-SH). The interfering effect exerted by DMF upon
catalysis thus appears to arise from its competition with H+ for
the coordination site at molybdenum. The cis-MoIV(H)(μ-SH)
intermediate invoked here differs from the species proposed to
give way to stoichiometric H2 evolution from the hydrido
clusters [M3S4H3(diphosphine)3]+ (M = Mo or W; diphos-
phine = 1,2-bis(dimethylphosphino)ethane or (+)-1,2-bis-
(2R,5R)-(dimethylphospholan-1-yl)ethane) when treated
with HX (X = Cl or Br). In this system, kinetic data have
been interpreted as being consistent with a hydrogen-bonded
intermediate of the form M−H···H−X, which is then activated
toward H2 release by peripheral H−X···H−X hydrogen bond
formation.79−81
Figure 12. [Mo3S6][Mo3NiS4] assembly that is supported in part by
2−
interactions between Sax of the S2 ligands (top fragment) with an
2−
Sax ligand of the bottom fragment.
or quasireversible wave at ∼ −1.3 V that differs appreciably in
its current amplitude among the members of the set. We
speculate that a reduction that produces [Mo3Sx(S2CNR2)3]+
(x = 5 or 6) may be followed by condensation with
[Mo3S7(S2CNR2)3]+ or [Mo3Sx(S2CNR2)3]+ to a hexanuclear
species that is a precursor to 6, rather than 6 itself, that such
condensation is facile or somewhat impeded according to the
steric envelopment afforded by the dithiocarbamate sub-
stituents, and that it is this putative species to which the −1.33
V reversible reduction may be attributed. The modest current
amplitude for this reduction that is found in the voltammo-
gram of [Mo3S7(S2CNiBu2)3]I, compared to the significant
relative current observed in [Mo3S7(S2CNEt2)3]I, is plausible
in this light. Consistent with this suggestion, the cyclic
voltammogram of [Mo3S7(S2CNMe2)3]I shows a reversible
reduction at −1.31 V with full current amplitude (Figure S63),
while that for [Mo3S7(S2CNBn2)3]I shows only a poorly
defined feature in this same potential (Figure S64).
The known hemilability of dithiocarbamate(1−), which
likely is accentuated upon reduction, is undoubtedly important
for vacating the coordination sites at the Mo ion that assumes
the all-sulfide environment at the nexus of this cluster 6. The
obviously complex series of transformations that must occur
(e.g., movement of R2NCS2(1−) into a bridging mode,
2−
transformation of μ-S2 to μ-S2− in one Mo3 fragment,
incorporation of additional S2− for bridging) argues against 6
itself being formed on the cyclic voltammetry time scale.
However, that 6 can be formed in the reducing conditions of
the photolysis cell suggests that a reducing environment
generated by different means could still enable its formation.
Hydrogen evolving catalysis by [Mo3S7(S2CNR2)3]I (R =
SUMMARY AND CONCLUSIONS
i
■
Et, Bu) was also examined in bulk electrolysis at a controlled
potential of −1.6 V in 1:2:1 THF:MeCN:H2O and 9:1
MeCN:H2O, respectively. Over a two hour period, the
corresponding Faradaic efficiencies for H2 production by
these two clusters were 61% (Figures 9(d)) and 37%.
Assuming some degree of transformation of [Mo3S7]4+ to
[Mo3S4]4+ + 3S0 in these electrolyses, competition for
electrons by S0 would be expected to undercut Faradaic
efficiencies. However, the basis for the appreciably lower
This study reports photocatalytic H2-evolution at moderate
levels by soluble molybdenum sulfide clusters supported with
dithiocarbamate ligands, which are initially of the form
[Mo3S7(S2CNR2)3]I (R = alkyl). The isobutyl variant provides
more consistent, reproducible H2-evolution data because its
enhanced solubility ensures solution homogeneity. With a
modification of the classic [Ru(bipy)3]2+/Et3N sacrificial
photoreduction system, the [Mo3S7(S2CNiBu2)3]I catalytic
M
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
r
Scheme 3. Proposed Catalytic Cycle for H2 Evolution for [Mo3S4(S2CNMe2)3]+, Assuming Initial 2e− Reduction
r
Results from SMD-PBEPBE//PBEPBE/BS2 computations. The [Mo3S4(S2CNMe2)3]+ cation is taken as a model for the [Mo3S4(S2CNiBu2)3]+
fragment of hexamolybdenum cluster 6, which is likely the functional site for H2 evolution.
system in 1:9 H2O:MeCN demonstrated several hundred
turnovers in H2 formation in a period of several hours with
tight reproducibility.
Our collective observations lead to the following con-
clusions:
coordinating solvent significantly hinders formation of a
Mo−H intermediate (vide infra).
(3) A plausible sequence for the catalysis cycle is reduction
of the cluster by two electrons followed by successive H+
transfer steps. A computational survey of the energetic
landscape of possible intermediates points toward a
Mo(H)−SH−Mo species (metal hydrido−bridging
hydrosulfido) as the likely intermediate just prior to
H2 elimination.
(1) The [Mo3S7(S2CNR2)3]I clusters appear to be catalyst
precursors rather than the active catalysts themselves.
Mass spectrometry assays of the photolysis solutions
show the transformation of [Mo3S7(S2CNR2)3]+ to
[Mo3S4(S2CNR2)3]+ over a time scale of minutes via
(4) The most active catalytic system of those investigated
here produces a hexametallic cluster that deposits in
2−
facile extrusion of sulfur from the bridging S2 ligands.
i
(2) An open coordination site at molybdenum is requisite
for catalyst function. When N,N-dimethylformamide
(DMF) is employed as photolysis solvent with any of the
clusters studied, H2-evolving activity is suppressed.
Crystallization of [Mo3S4(η2-S2CNEt2)3(μ2-S2CNEt2)]
from DMF produces [Mo3S4(η2-S2CNEt2)3(μ2-
S2CNEt2)(η1,κO-DMF)], strongly suggesting that the
crystalline form from the photolysis cells when R = Bu
on the dithiocarbamate ligand. This cluster appears to
form via condensation of an equivalent of the original
[Mo3S7(S2CNR2)3]+ cation with the sulfur-extruded
derivative, [Mo3S4(S2CNR2)3]+, to give an asymmetric
hexanuclear cluster comprised of [Mo3S7(η2-
S2CNiBu2)3]+ and [Mo3S4(η2-S2CNiBu2)(μ2-
N
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
S2CNiBu2)]+ “halves”, which are held in part by a μ6-S2−
Article
authors (N.I.H., C.E.W., R.H.S, J.P.D) gratefully acknowledge
support for this project from the National Science Foundation
(OIA-1539035) and very helpful commentary from the
manuscript’s reviewers.
ligand that also provides overall charge neutrality.
I s o l a t e d [ M o 3 S 4 ( S 2 C N R 2 ) 3 ] [ A n i o n ] o r
[Mo3S4(S2CNR2)4] samples, while able to produce H2
under photolysis, are only active at ∼25% the level of
[Mo3S7(S2CNR2)3]I, which readily extrudes sulfur under
the reducing conditions of photolysis (vide supra). Thus,
the H2-evolving activity levels observed cannot be
accounted for by [Mo3S4(S2CNEt2)3][Anion] alone or
by intact [Mo3S7(S2CNR2)3]I but are explained by the
formation of hexamolybdenum clusters such as 6.
Further solution speciation to other, unidentified
[MoxSy(S2CNR2)z]n clusters cannot be excluded, nor
can their participation in H2-evolution.
REFERENCES
■
(1) Quadrennial Technology Review; U.S. Department of Energy:
Washington, DC, 2015; p. 256.
(2) Angelici, R. J. Hydrodesulfurization and Hydrodenitrogenation.
In Encyclopedia of Inorganic Chemistry; King, R. B., Ed.; Wiley: New
York, 1994; Vol. 3, pp 1433−1443.
(3) Smil, V. Enriching the Earth; MIT Press: Cambridge, MA, 2001.
(4) Brandon, N. P.; Kurban, Z. Clean Energy and the Hydrogen
Economy. Philos. Trans. R. Soc., A 2017, 375, 20160400/1−
20160400/17.
(5) Jaramillo, T. F.; Jørgensen, K. P.; Bonde, J.; Nielsen, J. H.;
Horch, S.; Chorkendorff, I. Identification of Active Edge Sites for
Electrochemical H2 Evolution from MoS2 Nanocatalysts. Science
2007, 317, 100−102.
(6) Hinnemann, B.; Moses, P. G.; Bonde, J.; Jørgensen, K. P.;
Nielsen, J. H.; Horch, S.; Chorkendorff, I.; Nørskov, J. K. Biomimetic
Hydrogen Evolution: MoS2 Nanoparticles as Catalyst for Hydrogen
Evolution. J. Am. Chem. Soc. 2005, 127, 5308−5309.
(7) Lu, Z.; Zhang, H.; Zhu, W.; Yu, X.; Kuang, Y.; Chang, Z.; Lei, X.;
Sun, X. In Situ Fabrication of Porous MoS2 Thin-Films as High-
Performance Catalysts for Electrochemical Hydrogen Evolution.
Chem. Commun. 2013, 49, 7516−7518.
In continuing work, our laboratories aim at a deliberate
synthesis of 6 so that its inherent catalytic activity and other
properties can be more definitively established. Additionally,
we are developing an expanded set of [M3E7(E2CNR2)3]+
clusters (M = Mo, W; E = S, Se; R = alkyl) to learn what
effects these differing compositions have upon catalytic
activity.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.inorgchem.9b02252.
■
S
(8) Merki, D.; Fierro, S.; Vrubel, H.; Hu, X. Amorphous
Molybdenum Sulfide Films as Catalysts for Electrochemical Hydro-
gen Production in Water. Chem. Sci. 2011, 2, 1262−1267.
(9) Yan, Y.; Xia, B.; Ge, X.; Liu, Z.; Wang, J.-Y.; Wang, X. Ultrathin
MoS2 Nanoplates with Rich Active Sites as Highly Efficient Catalyst
for Hydrogen Evolution. ACS Appl. Mater. Interfaces 2013, 5, 12794−
12798.
(10) Wang, D.; Wang, Z.; Wang, C.; Zhou, P.; Wu, Z.; Liu, Z.
Distorted MoS2 Nanostructures: An Efficient Catalyst for the
Electrochemical Hydrogen Evolution Reaction. Electrochem. Commun.
2013, 34, 219−222.
(11) Voiry, D.; Salehi, M.; Silva, R.; Fujita, T.; Chen, M.; Asefa, T.;
Shenoy, V. B.; Eda, G.; Chhowalla, M. Conducting MoS2 Nanosheets
as Catalysts for Hydrogen Evolution Reaction. Nano Lett. 2013, 13,
6222−6227.
(12) Ji, S.; Yang, Z.; Zhang, C.; Liu, Z.; Tjiu, W. W.; Phang, I. Y.;
Zhang, Z.; Pan, J.; Liu, T. Exfoliated MoS2 Nanosheets as Efficient
Catalysts for Electrochemical Hydrogen Evolution. Electrochim. Acta
2013, 109, 269−275.
Procedures for crystal growth and computational
methods, unit cell refinement data, photograph of the
H-cell, thermal ellipsoid plots, NMR, UV−vis, mass, and
Raman spectra, and cyclic voltammograms (PDF)
Accession Codes
CCDC 1906107−1906112 and 1919684−1919687 contain
the supplementary crystallographic data for this paper. These
data can be obtained free of charge via www.ccdc.cam.ac.uk/
data_request/cif, or by emailing data_request@ccdc.cam.ac.
uk, or by contacting The Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44
1223 336033.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: russ@tulane.edu.
*E-mail: donahue@tulane.edu.
(13) Maijenburg, A. W.; Regis, M.; Hattori, A. N.; Tanaka, H.; Choi,
K.-S.; ten Elshof, J. E. MoS2 Nanocube Structures as Catalysts for
Electrochemical H2 Evolution from Acidic Aqueous Solutions. ACS
Appl. Mater. Interfaces 2014, 6, 2003−2010.
(14) Li, Y.; Wang, H.; Xie, L.; Liang, Y.; Hong, G.; Dai, H. MoS2
Nanoparticles Grown on Graphene: An Advanced Catalyst for the
Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2011, 133, 7296−
7299.
ORCID
Bing Shan: 0000-0002-6802-3095
Antony Obanda: 0000-0003-3532-0402
Nathan I. Hammer: 0000-0002-6221-2709
Russell H. Schmehl: 0000-0002-5467-9069
James P. Donahue: 0000-0001-9768-4813
Notes
(15) Bian, X.; Zhu, J.; Liao, L.; Scanlon, M. D.; Ge, P.; Ji, C.; Girault,
H. H.; Liu, B. Nanocomposite of MoS2 on Ordered Mesoporous
Carbon Nanospheres: A Highly Active Catalyst for Electrochemical
Hydrogen Evolution. Electrochem. Commun. 2012, 22, 128−132.
(16) Lin, T.-W.; Liu, C.-J.; Lin, J.-Y. Facile Synthesis of MoS3/
Carbon Nanotube Nanocomposite with High Catalytic Activity
Toward Hydrogen Evolution Reaction. Appl. Catal., B 2013, 134−
135, 75−82.
(17) Laursen, A. B.; Kegnæs, S.; Dahl, S.; Chorkendorff, I.
Molybdenum Sulfides−Efficient and Viable Materials for Electro-
and Photoelectrocatalytic Hydrogen Evolution. Energy Environ. Sci.
2012, 5, 5577−5591.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The Louisiana Board of Regents (LEQSF-(2002-03)-ENH-
TR-67) and the National Science Foundation (MRI: 1228232
and 0619770) are thanked for funding of Tulane University’s
X-ray crystallography and mass spectrometry instrumentation,
and Tulane University is acknowledged for its ongoing
assistance with operational costs for the X-ray diffraction
facility. The Horiba LabRam spectrometer at the University of
Mississippi was funded by the NSF (MRI CHE-1532079). The
O
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(18) Tributsch, H.; Bennett, J. C. Electrochemistry and Photo-
chemistry of MoS2 Layer Crystals. I. J. Electroanal. Chem. Interfacial
Electrochem. 1977, 81, 97−111.
(36) Neelakantan, L. Disulfides. J. Org. Chem. 1958, 23, 938−939.
(37) Kapanda, C. N.; Muccioli, G. G.; Labar, G.; Poupaert, J. H.;
Lambert, C. M. Bis(dialkylaminethiocarbonyl)disulfides as Potent and
Selective Monoglyceride Lipase Inhibitors. J. Med. Chem. 2009, 52,
7310−7314.
(38) Lal, N.; Jangir, S.; Bala, V.; Mandalapu, D.; Sarswat, A.; Kumar,
L.; Jain, A.; Kumar, L.; Kushwaha, B.; Pandey, A. K.; Krishna, S.;
Rawat, T.; Shukla, P. K.; Maikhuri, J. P.; Siddiqi, M. I.; Gupta, G.;
Sharma, V. L. Role of Disulfide Linkage in Action of Bis-
(dialkylaminethiocarbonyl)disulfides as Potent Double-Edged Micro-
bicidal Spermicide: Design, Synthesis and Biology. Eur. J. Med. Chem.
2016, 115, 275−290.
(39) Fontenot, P.; Wang, B.; Chen, Y.; Donahue, J. P. Crystal
Structure of Tetraisobutylthiuram Disulfide. Acta Crystallogr., Sect. E
2017, 73, 1764−1769.
(40) Fedin, V. P.; Sokolov, M. N.; Geras’ko, O. A.; Virovets, A. V.;
Podberezskaya, N. V.; Federov, V. Y. Diethyldithiocarbamate
Triangular Thio Complexes of Molybdenum: Synthesis and an X-
ray Structural Study of [Mo3S7(S2CNEt2)3]Cl. Inorg. Chim. Acta
1992, 192, 153−156.
(41) Fedin, V. P.; Mueller, A.; Boege, H.; Armatazh, A.; Sokolov, M.
N.; Yarovoi, S. S.; Fedorov, V. E. Synthesis of Dithiocarbamate
Complexes Containing the Cluster Fragment [Mo3(S)(S2)3]4+. X-ray
Structural Study of [Mo3S7(S2CNEt2)3]I·1.5C6H6. Zh. Neorg. Khim.
1993, 38, 1677−1682.
(42) Lu, C.-Z.; Tong, W.; Zhuang, H.-H.; Wu, D.-M. Crystal
Structure of Trinuclear Molybdenum Cluster Mo3(μ3-S)(μ2-
S2)3(Et2dtc)3I Synthesized by Solid State Reaction. Jiegou Huaxue
1993, 12, 124−128.
(19) Seger, B.; Laursen, A. B.; Vesborg, P. C. K.; Pedersen, T.;
Hansen, O.; Dahl, S.; Chorkendorff, I. Hydrogen Production using a
Molybdenum Sulfide Catalyst on a Titanium-Protected n+p-Silicon
Photocathode. Angew. Chem., Int. Ed. 2012, 51, 9128−9131.
(20) Benck, J. D.; Lee, S. C.; Fong, K. D.; Kibsgaard, J.; Sinclair, R.;
Jaramillo, T. F. Designing Active and Stable Silicon Photocathodes for
Solar Hydrogen Production using Molybdenum Sulfide Nanomateri-
als. Adv. Energy Mater. 2014, 4, 1400739/1−1400739/8.
́
(21) Ding, Q.; Meng, F.; English, C. R.; Caban-Acevedo, M.;
Shearer, M. J.; Liang, D.; Daniel, A. S.; Hamers, R. J.; Jin, S. Efficient
Photoelectrochemical Hydrogen Generation Using Heterostructures
of Si and Chemically Exfoliated Metallic MoS2. J. Am. Chem. Soc.
2014, 136, 8504−8507.
̈
(22) Morales-Guio, C. G.; Tilley, S. D.; Vrubel, H.; Gratzel, M.; Hu,
X. Hydrogen Evolution from a Copper(I) Oxide Photocathode
Coated with an Amorphous Molybdenum Sulphide Catalyst. Nat.
Commun. 2014, 5, 3059.
(23) Zhang, L.; Liu, C.; Wong, A. B.; Resasco, J.; Yang, P. MoS2-
Wrapped Silicon Nanowires for Photoelectrochemical Water
Reduction. Nano Res. 2015, 8, 281−287.
(24) Appel, A. M.; DuBois, D. L.; Rakowski DuBois, M.
Molybdenum−Sulfur Dimers as Electrocatalysts for the Production
of Hydrogen at Low Overpotentials. J. Am. Chem. Soc. 2005, 127,
12717−12726.
(25) Karunadasa, H. I.; Montalvo, E.; Sun, Y.; Majda, M.; Long, J.
R.; Chang, C. J. A Molecular MoS2 Edge Site Mimic for Catalytic
Hydrogen Generation. Science 2012, 335, 698−702.
(43) Wang, M.-F.; Guo, G.-C.; Huang, J.-S.; Zhuang, H.-H.; Zhang,
Q.-E.; Lu, J.-X. Synthesis and Crystal Structure of Trinuclear
Molybdenum-Sulfur Cluster by Solid State Reaction at Low Heating
Temperature. Jiegou Huaxue 1994, 13, 221−225.
(26) Garrett, B. R.; Polen, S. M.; Click, K. A.; He, M.; Huang, Z.;
Hadad, C. M.; Wu, Y. Tunable Molecular MoS2 Edge-Site Mimics for
Catalytic Hydrogen Production. Inorg. Chem. 2016, 55, 3960−3966.
(27) Kibsgaard, J.; Jaramillo, T. F.; Besenbacher, F. Building an
Active-Site Motif into a Hydrogen-Evolution Catalyst with
Thiomolybdate [Mo3S13]2‑ Clusters. Nat. Chem. 2014, 6, 248−253.
(28) Shinde, P. S.; Fontenot, P. R.; Donahue, J. P.; Waters, J. L.;
Kung, P.; McNamara, L. E.; Hammer, N. I.; Gupta, A.; Pan, S.
Synthesis of MoS2 from [Mo3S7(S2CNEt2)3]I for Enhancing Photo-
electrochemical Performance and Stability of Cu2O Photocathode
Toward Efficient Solar Water Splitting. J. Mater. Chem. A 2018, 6,
9569−9582.
́
(44) Mayor-Lopez, M. J.; Weber, J.; Hegetschweiler, K.;
Meienberger, M. D.; Joho, F.; Leoni, S.; Nesper, R.; Reiss, G. J.;
Frank, W.; Kolesov, B. A.; Fedin, V. P.; Fedorov, V. P. Structure and
Reactivity of [Mo3-μ3S-(μS2)3]4+ Complexes. Quantum Chemical
Calculations, X-ray Structural Characterization, and Raman Spectro-
scopic Measurements. Inorg. Chem. 1998, 37, 2633−2644.
(45) Il’inchik, E. A.; Polyanskaya, T. M.; Myakishev, K. G.; Basova,
T. V.; Kolesov, B. A. Synthesis and Structure of the Complex
[Mo3(μ3-S)(μ2-S2)3(Et2NCS2)3]+Br. Russ. J. Gen. Chem. 2002, 72,
1862−1866.
(29) Shan, B.; Schmehl, R. Photochemical Generation of Strong
One-Electron Reductants via Light-Induced Electron Transfer with
Reversible Donors Followed by Cross Reaction with Sacrificial
Donors. J. Phys. Chem. A 2014, 118, 10400−10406.
(46) Il’inchik, E. A.; Polyanskaya, T. M.; Myakishev, K. G. Synthesis
and Structure of the Complex [Mo3(μ3-S)(μ2-S2)3(Et2NCS2)3]I. Russ.
J. Gen. Chem. 2007, 77, 1148−1154.
(30) Shan, B.; Baine, T.; Ma, X. A. N.; Zhao, X.; Schmehl, R. H.
Mechanistic Details for Cobalt Catalyzed Photochemical Hydrogen
Production in Aqueous Solution: Efficiencies of the Photochemical
and Non-Photochemical Steps. Inorg. Chem. 2013, 52, 4853−4859.
(31) Armarego, W. L. F.; Chai, C. L. L. Purification of Laboratory
Chemicals, 6th ed.; Butterworth-Heinemann: Oxford, U.K., 2009.
(47) Virovets, A. V.; Podberezskaya, N. V. Specfic Nonbonded
Interactions in the Structures of M3X74+ and M3X44+ (M = Mo, W; X
= O, S, Se) Clusters. J. Struct. Chem. 1993, 34, 306−322.
(48) Llusar, R.; Vicent, C. Trinuclear Molybdenum Cluster Sulfides
Coordinated to Dithiolene Ligands and Their Use in the Develop-
ment of Molecular Conductors. Coord. Chem. Rev. 2010, 254, 1534−
1548.
(49) Halbert, T. R.; McGauley, K.; Pan, W.-H.; Czernuszewicz, R.
S.; Stiefel, E. I. Synthesis, Structure, and Reactivity of
[(C2H5)4N]2[Mo3(μ3-S)(μ2-S)3(SCH2CH2S)3]: A Cluster with Sul-
fur “Vacancies” and Resonance Raman Spectral Similarity to Fe3S4
Proteins. J. Am. Chem. Soc. 1984, 106, 1849−1851.
(50) Taylor, J. R. An Introduction to Error Analysis, 2nd ed.;
University Science Books: Sausalito, CA, 1997; pp 73−77.
(51) Wunderlich, H. Structure of μ-Diethyldithiophosphinato-
tris(diethyldithiophosphinato)-tri-μ-thio-μ3-thio-triangulo-
trimolybdenum(IV). Acta Crystallogr., Sect. C: Cryst. Struct. Commun.
1987, 43, 24−26.
(52) Huang, M.; Lu, S.; Huang, J.; Huang, J. Replacement Reaction
of Trinuclear Molybdenum Clusters − Synthesis and Crystal
Structure of Two Molybdenum Clusters [Mo3S4(DTC)4(DMF)]-
(EtOH) and [Mo3S4(DTC)4(Py)](Py)2(H2O). Acta Chim. Sin. 1989,
47, 121−127.
(32) Muller, A.; Bhattacharyya, R. G.; Pfefferkorn, B. Eine Einfache
̈
2−
̈
Darstellung der Binaren Metall-Schwefel-Cluster [Mo3S13] und
2−
[Mo2S12]2− aus MoO4 in Praktisch Quantitativer Ausbeute. Chem.
Ber. 1979, 112, 778−780.
(33) Zimmermann, H.; Hegetschweiler, K.; Keller, T.; Gramlich, V.;
Schmalle, H. W.; Petter, W.; Schneider, W. Preparation of Complexes
Containing the [Mo3S(S2)3]4+ Core and Structure of Tris-
(diethyldithiocarbamato)tris(μ-disulfido)(μ3-thio)-triangulo-
trimolybdenum(IV) Iodide. Inorg. Chem. 1991, 30, 4336−4341.
(34) Reger, D. L.; Little, C. A.; Lambda, J. J. S.; Brown, K. J.;
Krumper, J. R.; Bergman, R. G.; Irwin, M.; Fackler, J. P., Jr. Main
Group Compounds. Sodium Tetrakis(3,5-bis(trifluoromethyl)-
phenyl)borate, Na[B(3,5-(CF3)2C6H3)4]. Inorg. Synth. 2004, 34, 5−8.
(35) McDonald, J. W.; Friesen, G. D.; Rosenhein, L. D.; Newton, W.
E. Syntheses and Characterization of Ammonium and Tetraalkylam-
monium Thiomolybdates and Thiotungstates. Inorg. Chim. Acta 1983,
72, 205−210.
P
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(53) Huang, J. Q.; Huang, J. L.; Shang, M. Y.; Lu, S. F.; Lin, X. T.;
Lin, Y. H.; Huang, M. D.; Zhuang, H. H.; Lu, J. X. Structure and
Reactivity of Molybdenum Clusters with Loose Coordination Site,
Mo3S4[S2P(OEt)2]4L. Pure Appl. Chem. 1988, 60, 1185−1192.
(54) Jaramillo, T. F.; Bonde, J.; Zhang, J.; Ooi, B.-L.; Andersson, K.;
Ulstrup, J.; Chorkendorff, I. Hydrogen Evolution on Supported
Incomplete Cubane-type [Mo3S4]4+ Electrocatalysts. J. Phys. Chem. C
2008, 112, 17492−17498.
(55) Seo, S. W.; Park, S.; Jeong, H.-Y.; Kim, S. H.; Sim, U.; Lee, C.
W.; Nam, K. T.; Hong, K. S. Enhanced Performance of NaTaO3
Using Molecular Co-catalyst [Mo3S4]4+ for Water Splitting into H2
and O2. Chem. Commun. 2012, 48, 10452−10454.
(56) Seger, B.; Herbst, K.; Pedersen, T.; Abrams, B.; Vesborg, P. C.
K.; Hansen, O.; Chorkendorff, I. Mo3S4 Clusters as an Effective H2
Evolution Catalyst on Protected Si Photocathodes. J. Electrochem. Soc.
2014, 161, H722−H724.
(57) Brunner, E. Solubility of Hydrogen in 10 Organic Solvents at
298.15, 323.15 and 373.15 K. J. Chem. Eng. Data 1985, 30, 269−273.
̈
(58) Meienberger, M. D.; Hegetschweiler, K.; Ruegger, H.;
(72) Garriga, J. M.; Llusar, R.; Uriel, S.; Vicent, C.; Usher, A. J.;
Lucas, N. T.; Humphrey, M. G.; Samoc, M. Synthesis and Third-
Order Nonlinear Optical Properties of [Mo3(μ3-S)(μ2-S2)3]4+
Clusters with Maleonitriledithiolate, Oxalate, and Thiocyanate
Ligands. Dalton Trans. 2003, 4546−4551.
́
(73) Llusar, R.; Triguero, S.; Polo, V.; Vicent, C.; Gomez-García, C.
́
J.; Jeannin, O.; Foumigue, M. Trinuclear, Mo3S7 Clusters Coordi-
nated to Dithiolate or Diselenolate Ligands and Their Use in the
Preparation of Magnetic Single Component Molecular Conductors.
Inorg. Chem. 2008, 47, 9400−9409.
(74) Alberola, A.; Llusar, R.; Triguero, S.; Vicent, C.; Sokolov, M.
́
N.; Gomez-García, C. Structural Diversity in Charge Transfer Salts
Based on Mo3S7 and Mo3S4Se3 Cluster Complexes and Bis-
(ethylenedithio)tetrathiafulvalene (ET). J. Mater. Chem. 2007, 17,
3440−3450.
(75) Llusar, R.; Polo, V.; Velez, E.; Vicent, C. Sulfur-Based Redox
4+
4+
Reactions in Mo3S7 and Mo3S4 Clusters Bearing Halide and 1,2-
Dithiolene Ligands: a Mass Spectrometric and Density Functional
Theory Study. Inorg. Chem. 2010, 49, 8045−8055.
(76) Hernandez-Molina, R.; Gonzales-Platas, J.; Vicent, C. Isolation
of a New Cs-Symmetrized Mo3(μ3-S)(μ-S)(μ-S2)2 Structural Type
Through Complementary Association with a Cubane-Type Mo3NiS4
Cluster. Eur. J. Inorg. Chem. 2012, 2012, 1278−1284.
Gramlich, V. The Reactivity of Complexes Containing the
[Mo3(μ3S)(μS2)3]4+ Core. Ligand Substitution, Sulfur Elimination,
and Sulfide Binding. Inorg. Chim. Acta 1993, 213, 157−169.
(59) Zank, G. A.; Jones, C. A.; Rauchfuss, T. B.; Rheingold, A. L.
Oxygenated Titanium Sulfide Clusters. Synthesis and Structures of
(CH3C5H4)2Ti4S8Ox (x = 1,2). Inorg. Chem. 1986, 25, 1886−1891.
(60) Dupre, N.; Hendriks, H. M. J.; Jordanov, J.; Gaillard, J.; Auric,
P. Synthesis and Structural Characterization of [Fe4(μ3-S)3(μ3-
S2)Cp4][MoOCl4(thf)]. A New Fe−S Cluster with Different Modes
of Coordination at the Iron Sites. Organometallics 1984, 3, 800−802.
(61) Brunner, H.; Meier, W.; Wachter, J.; Nuber, B.; Ziegler, M. L.
(77) Fackler, J. P., Jr; Thompson, L. D.; Lin, I. J. B.; Stephenson, A.;
Gould, R. O.; Alison, J. M. C.; Fraser, A. J. F. Phosphine Adducts of
Palladium(II) and Platinum(II) 1,1-Dithiolates. Single-Crystal X-ray
Structures of Bis(diethyldithiocarbamato)(triphenylphosphine)-
platinum(II), Pt(S2CNEt2)2PPh3, Bis(diphenylphosphinodithioato)-
(triphenylphosphine)palladium(II), Pd(S2PPh2)2PPh3, Bis(O,O-di-
ethyl dithiophosphato)(triphenylphosphine)platinum(II), Pt(S2P-
( O E t )
)
P P h ,
a n d
̈
2
2
3
Neue Aspekte in der Chemie von Ubergangsmetallpolysulfidkomplex-
(Diphenylphosphinodithioato)bis(triethylphosphine)palladium(II)
Diphenylphosphinodithioate, [Pd(S2PPh2)(PEt3)2]S2PPh2. Solution
31P NMR Studies of the Dithiophosphate Complexes. Inorg. Chem.
1982, 21, 2397−2403.
(78) Roberts, S. A.; Young, C. G.; Cleland, W. E., Jr.; Ortega, R. B.;
Enemark, J. H. Synthesis, Characterization, and Atom Transfer
Reactions of {HB(Me2C3N2H)3}MoO{S2P(OR)2} and {HB-
(Me2C3N2H)3}MoO2{η1-S2P(OEt)2}. Inorg. Chem. 1988, 27, 3044−
3051.
́
́
en: Synthese und Kristallstrukturen von Cp3Nb3S12 und Cp3Nb3S10O
́
(Cp = t-BuC5H4). J. Organomet. Chem. 1990, 381, C7−C12.
(62) Venturelli, A.; Rauchfuss, T. B.; Verma, A. K. Sulfido−
Persulfido Equilibria in Sulfur-Rich Metal Clusters: The Case of
(C5Me5)3RhRu2S42+. Inorg. Chem. 1997, 36, 1360−1365.
(63) Yuki, M.; Kuge, K.; Okazaki, M.; Mitsui, T.; Inomata, S.;
Tobita, H.; Ogino, H. Cluster Synthesis by the Reactions of
́
́
[Cp2Fe2S4] with Transition-Metal Complexes (Cp = C5Me5, 1,3-
C5H3(SiMe3)2). Inorg. Chim. Acta 1999, 291, 395−402.
(64) Adams, R. D.; Boswell, E. M.; Captain, B.; Miao, S.; Beddie, C.;
Webster, C. E.; Hall, M. B.; Dalal, N. S.; Kaur, N.; Zipse, D. Disulfio
Iron-Manganese Carbonyl Cluster Complexes: Synthesis, Structure,
Bonding and Properties of the Radical CpFeMn2(CO)7(μ3-S2)2. J.
Organomet. Chem. 2008, 693, 2732−2738.
́
(79) Basallote, M. G.; Feliz, M.; Fernandez-Trujillo, M. J.; Llusar, R.;
Safont, V. S.; Uriel, S. Mechanism of the Reaction of the
[W3S4H3(dmpe)3]+ Cluster with Acids: Evidence for the Acid-
Promoted Substitution of Coordinated Hydrides and the Effect of the
Attacking Species on the Kinetics of Protonation of the Metal-
Hydride Bonds. Chem. - Eur. J. 2004, 10, 1463−1471.
(65) Bar-Nahum, I.; Gupta, A. K.; Huber, S. M.; Ertem, M. Z.;
Cramer, C. J.; Tolman, W. B. Reduction of Nitrous Oxide to
Dinitrogen by a Mixed Valent Tricopper-Disulfido Cluster. J. Am.
Chem. Soc. 2009, 131, 2812−2814.
(66) Han, S.-D.; Miao, X.-H.; Liu, S.-J.; Bu, X.-H. A Series of Cobalt
and Nickel Clusters Based on Thiol-Containing Ligands Accom-
panied by in situ Ligand Formation. Dalton Trans. 2015, 44, 560−567.
(67) Lu, S.-F.; Huang, J.-Q.; Wu, Q.-J.; Huang, X.-Y. Synthesis and
Crystal Structure of an Unusual Hexanuclear Molybdenum Cluster
{[Mo3(μ3-S)(μ-S2)2(dtp)3](μ3-S)(η3-S2)[Mo3(μ3-S)(μ-S)3(μ-OAc)-
(dtp)2]}. Jiegou Huaxue 1996, 15, 415−421.
(68) Hernandez-Molina, R.; Sykes, A. G. Reactions of the
Heterometallic Cuboidal Clusters Mo3MS4 (M = Co, Ni, Pd, Cu)
and Mo3NiSe4 with CO: Electron Counts and Kinetic/Thermody-
namic Studies with M = Ni, Pd. Coord. Chem. Rev. 1999, 187, 291−
302.
́
(80) Algarra, A. G.; Basallote, M. G.; Feliz, M.; Fernandez-Trujillo,
M. J.; Llusar, R.; Safont, V. S. New Insights into the Mechanism of
Proton Transfer to Hydride Complexes: Kinetic and Theoretical
Evidence Showing the Existence of Competitive Pathways for
Protonation of the Cluster [W3S4H3(dmpe)3]+ with Acids. Chem. -
Eur. J. 2006, 12, 1413−1426.
́
(81) Algarra, A. G.; Basallote, M. G.; Fernandez-Trujillo, M. J.; Feliz,
́
M.; Guillamon, E.; Llusar, R.; Sorribes, I.; Vicent, C. Chiral
[Mo3S4H3(diphosphine)3]+ Hydrido Clusters and Study of the Effect
of the Metal Atom on the Kinetics of the Acid-Assisted Substitution of
the Coordinated Hydride: Mo vs W. Inorg. Chem. 2010, 49, 5935−
5942.
(69) Hidai, M.; Kuwata, S.; Mizobe, Y. Synthesis and Reactivities of
Cubane-Type Sulfido Clusters Containing Noble Metals. Acc. Chem.
Res. 2000, 33, 46−52.
(70) Hernandez-Molina, R.; Sokolov, M. N.; Sykes, A. G. Behavioral
Patterns of Heterometallic Cuboidal Derivatives of [M3Q4(H2O)9]4+
(M = Mo. W; Q = S, Se). Acc. Chem. Res. 2001, 34, 223−230.
(71) Batsanov, S. S. Van der Waals Radii of Elements. Inorg. Mater.
2001, 37, 871−885.
Q
DOI: 10.1021/acs.inorgchem.9b02252
Inorg. Chem. XXXX, XXX, XXX−XXX