Photocatalytic splitting of CS2 to S8 and a carbon-sulfur polymer catalyzed by a bimetallic ruthenium(II) compound with a tertiary amine binding site: Toward photocatalytic splitting of CO2?

Article abstract of DOI:10.1021/ic2008045

The catalytic photocleavage of CS₂ to S₈ and a (C_xS_y)_n polymer with visible light using a dinuclear ruthenium(II) compound with a bipyridine units for photoactivity and a vicinal tertiary amine binding site for CS₂ activation was studied. The catalyst was characterized by X-ray diffraction, ¹H NMR, and ¹³C NMR, ESI-MS and elemental analysis. CS₂ photocleavage was significant (240 turnovers, 20 h) to yield isolable S₈ and a (C_xS_y)_n polymer. A mononuclear catalyst or one without an amine binding site showed significantly less activity. XPS of the (C_xS_y)_n polymer showed a carbon/sulfur ratio ～1.5-1.6 indicating that in part both C-S bonds of CS₂ had been cleaved. Catalyst was also included within the polymer. The absence of peaks in the ¹H NMR verified the (C_xS_y)_n nature of the polymer, while ¹³C NMR and IR indicated that the polymer had multiple types of C-S and C-C bonds.

Full text of DOI:10.1021/ic2008045

COMMUNICATION
pubs.acs.org/IC
Photocatalytic Splitting of CS2 to S8 and a CarbonꢀSulfur Polymer
Catalyzed by a Bimetallic Ruthenium(II) Compound with a Tertiary
Amine Binding Site: Toward Photocatalytic Splitting of CO2?
Konstantin Livanov,† Vedichi Madhu,† Ekambaram Balaraman,† Linda J. W. Shimon,‡ Yael Diskin-Posner,‡
and Ronny Neumann*,†
†Department of Organic Chemistry and ‡Chemical Research Support Unit, Weizmann Institute of Science, Rehovot, Israel 76100
S Supporting Information
b
Scheme 1. Catalytic Design for Photocleavage of CS2
ABSTRACT: The catalytic photocleavage of CS2 to S8 and
a (CxSy)n polymer with visible light using a dinuclear
ruthenium(II) compound with a bipyridine units for
photoactivity and a vicinal tertiary amine binding site
for CS2 activation was studied. The catalyst was char-
acterized by X-ray diffraction, 1H NMR, and 13C NMR,
ESI-MS and elemental analysis. CS2 photocleavage was
significant (240 turnovers, 20 h) to yield isolable S8 and a
(CxSy)n polymer. A mononuclear catalyst or one without
an amine binding site showed significantly less activity.
XPS of the (CxSy)n polymer showed a carbon/sulfur ratio
∼1.5ꢀ1.6 indicating that in part both CꢀS bonds of CS2
had been cleaved. Catalyst was also included within the
on the photocatalytic splitting of isostructural CS2 to the
corresponding (CxSy)n polymer and molecular sulfur, S8, with
emphasis on catalytic design principles needed to effect such a
photocatalytic splitting reaction. Such a photocleavage of CS2 has
not been reported.
1
polymer. The absence of peaks in the H NMR verified
the (CxSy)n nature of the polymer, while 13C NMR and
IR indicated that the polymer had multiple types of CꢀS
and CꢀC bonds.
Carbon disulfide is a linear molecule with relatively strong
CdS double bonds. On the one hand, the electrophilic carbon
atom easily reacts with good nucleophiles such as amines; for
example, the reaction with ammonia and primary and secondary
amines yields dithiocarbamates. On the other hand, the forma-
tion of CS and elemental sulfur from CS2 is significantly
endoergic (ΔGr = +27.12 kcal/mol); thus, such a splitting
reaction seemingly should be carried out under photocatalytic
conditions. The catalytic module that was designed for the
photocatalytic splitting pathway of CS2 (Scheme 1) consists of
(a) a binuclear metal complex that can lead to CS2 insertion and
activation,8 (b) a nucleophilic binding site for CS2 complexation,
such as a tertiary amine, thatꢀwill lead to the formation of a
trigonal zwitterionic R3N+CS2 unstable transient species that
will, however, have lengthened (weaker) CꢀS bonds.9 A position
vicinal to the binuclear center will combine the advantages of the
amine binding site and the binuclearity for CꢀS bond activation.
(c) The catalyst ideally will be photoactive; that is, it should
be able to absorb visible light; ruthenium bipyridine based
compounds appear to be excellent candidates, and (d) metal
sulfide species formed in the carbonꢀsulfide bond splitting step
should be in a position to recombine to yield molecular sulfur
after a series of CS2 binding and CꢀS bond splitting reactions.
Therefore, the catalysts that were prepared were based on ditopic
1,2-bis(2,20-bipyridyl-6-yl) ligands to form dinuclear ruthenium(II)
complexes with tertiary amine containing a bridge. For verification
of the catalyst design, both an acetylene-bridged compound to
eduction of CO2 is a hot research topic in efforts to
R
reduce its environmental impact and recycle fossil fuels.
Besides higher temperature heterogeneous catalytic methods1
and application of electrochemical cells,2 five main low-tem-
perature and/or photocatalytic approaches can be identified:
(1) thermal reduction of CO2 with H2 often under basic
conditions via the formation of bicarbonate to yield formic acid
or formate salts;3 (2) photocatalytic reduction of CO2 to CO
using tertiary amines as sacrificial reducing agents and proton
donors in a “half-cell” approach;4 (3) the use of H2 in a
photocatalytic reverse waterꢀgas shift reaction;5 (4) photo-
reduction with H2O as the reducing agent;6 (5) catalytic and
noncatalytic reactions of CO2 with stoichiometric acceptors.7
These methods all have disadvantages such as the use of
nonrenewable reducing agents such as amines and H2, the
apparent need for UV light for the one-electron reduction of
CO2 with H2O, or the need for stoichiometric oxygen-acceptor
reagents.
The photocatalytic cleavage of CO2 to yield CO and O2 under
ambient conditions with solar energy would be beneficial because
CO formed could be stored and then combusted when needed.
Alternatively, CO could be reacted with H2O using waterꢀgas
shift technology to yield H2; the latter could then be reacted with
additional CO to yield methanol as a liquid fuel. As a prelude to
the photocatalytic splitting of CO2, here we describe our research
Received: April 19, 2011
Published: October 26, 2011
r
2011 American Chemical Society
11273
dx.doi.org/10.1021/ic2008045 Inorg. Chem. 2011, 50, 11273–11275
|

Inorganic Chemistry
COMMUNICATION
a
Scheme 2. Designed and Control Ligands
Table 1. Photocleavage of CS2 to S8 and (CxSy)n
catalyst
solvent
yield S8, mol %
TON
[Ru2L1(p-cymene)2Cl2]2+
[Ru2L1(p-cymene)2Cl2]2+a
[Ru2L1(p-cymene)2Cl2]2+b
[Ru2L2(p-cymene)2Cl2]2+
[Ru2L3(p-cymene)2Cl2]2+
[(p-cymene)RuCl2]2
none
THF/H2O
THF/H2O
THF/H2O
THF/H2O
THF/H2O
THF/H2O
THF/H2O
12.3
0
242
0
4.7
3.9
2.4
<0.5
0
92
77
48
∼10
0
a Reaction conditions: 5 mg of catalyst, 0.5 mL of CS2, 1 mL of solvent,
20 h, 70 °C, light from a 60-W tungsten lamp: (a) no light; (b) 22 °C.
Carbon disulfide was photocleaved in H2O/tetrahydrofuran
(THF), which dissolves both the catalysts and CS2 (Table 1).
There was no reaction in other solvents that dissolved both
components such as acetonitrile and dimethyl sulfoxide (DMSO).
In a typical procedure, a 15-mL pressure tube was loaded with 5 mg
(4.2 μmol, 0.05 mol %) of 1, 1 mL of 1:1 H2O/THF, and 0.5 mL
(8.28 mmol) of CS2. The pressure tube was purged with argon, in
order to remove all traces of oxygen, and then hermetically sealed
and stirred vigorously. An ordinary 60-W light bulb was used as a
photon source in all reactions. Typically, reactions were carried out
at 70 °C. After 3ꢀ4 h, a dark insoluble material started to form;
reactions were continued for 20 h. In the workup procedure, H2O
and toluene were added to the reaction mixture to form three
different phases. The lower, aqueous phase contained the catalyst,
which was partially decomposed as seen by 1H NMR. Also, a small
amount of H2 was detected, indicating some catalyst decompostion
(see also below). The upper, organic phase was analyzed by gas
chromatographyꢀmass spectrometry (GCꢀMS), and besides CS2,
elemental sulfur (S8) in isolable quantities after evaporation was
found (30 mg, 12.3% yield, 242 turnovers). S8 was identified by
GCꢀMS and by comparison to a reference standard. At the
interface between the aqueous and organic phases, there were dark,
opaque, and very thin films (Figure S4 in the SI), which strongly
suggested a polymeric nature of the product. Control experi-
ments with 2 and 3 as catalysts under identical conditions yielded S8
(10.5 mg, 3.9% yield, 81 turnovers and 6.4 mg, 2.4%, 48 turnovers,
respectively) in significantly smaller amounts, supporting the design
concept that an amine binding site and dinuclear catalyst would
significantly increase the activity for CS2 cleavage. There was no
reaction in the absence of catalyst or light, and only traces of S8 were
formed using [(p-cymene)RuCl2]2 as a catalyst. Reactions at 22 °C
were less effective.
There is not much published research on (CxSy)n polymers.
Initially, CS2 was polymerized at very high pressures of up to
40 000 bar and ∼180 °C.12 Later on, photopolymerization at
313 nm of CS2 in the gas phase13 and plasma-polymerized CS2
were described.14 The polymers all have a 1:2 carbon/sulfur
atom ratio except for the plasma-polymerized CS2, where various
ratios, C < S, were obtained depending on the polymerization
conditions. Here, elemental analysis of the (CxSy)n polymer was
not quantitative because only ∼61% of the total mass could be
accounted for, presumably because of incomplete pyrolysis of the
material. Elemental analysis did, however, reveal significant
amounts of nitrogen and hydrogen, indicating that the catalyst
had been at least partially encapsulated within the polymer. XPS
measurements were, therefore, carried out that also showed
catalyst encapsulation and a carbon/sulfur ratio of 1.5ꢀ1.6 after
correction for catalyst inclusion. This is the first (CxSy)n polymer
Figure 1. ORTEP representation (50% probability) of [Ru2L1(p-
cymene)2Cl2](BF4)2 1.5CH3CN. Anions, solvates, and hydrogen
3
atoms are not shown for clarity. Color code: C, black; N, blue; Ru,
magenta; Cl, green.
demonstrate the importance of a tertiary amine binding site for
CS2 and a mononuclear complex to show the advantage of
activation of both CꢀS bonds were prepared (Scheme 2).
The ligand N,N-bis([2,20-bipyridin-6-yl]methyl)butan-1-amine
(L1) was prepared by the reaction of 2 equiv of 6-(chloro-
methyl)-2,20-bipyridine11 with n-butylamine. The first control
ligand, 1,2-bis(2,20-bipyridyl-6-yl)ethyne (L2), was prepared as
recently described through consecutive Suzuki and Sonogashira
coupling reactions of 6-bromo-2,2-bipyridine with sodium
6
6
0
acetylide,10 and the second control ligand, N ,N -diethyl[2,20-
bipyridinyl]-6,60-diamine (L3), was prepared by the palladium-
catalyzed coupling of 6,60-dibromo-2,20-bipyridine with ethyl-
benzylamine, followed by debenzylation with concentrated
H2SO4. These ligands were then reacted with a ruthenium(II)
precursor, [(p-cymene)RuCl2]2, to yield the desired orange dinu-
clear ruthenium(II) complexes [Ru2L1(p-cymene)2Cl2]2+ (1),
[Ru2L2(p-cymene)2Cl2]2+ (2), and [RuL3(p-cymene)Cl (3).
Structures derived from X-ray diffraction measurements are
shown in Figure 1 and in the Supporting Information (SI). The
structures have the expected coordination spheres around the
ruthenium(II) centers with typical RuꢀN, RuꢀC, and Ruꢀ
Cl bond lengths (see the SI for full data). The lability of the
p-cymene and chloride ligands was also evaluated. Thus, the
chloride ligand could be replaced by a solvent such as acetonitrile
by the addition of AgBF4, and heating or irradiation with light
from a tungsten lamp led to the replacement of the p-cymene
ligandbya solvent. Representativestructures (Figures S2and S3in
the SI) show that after these treatments the binuclear ruthenium-
(II) nature of these compounds is retained. Notably, the relative
positions of the proposed ruthenium(II) active sites can vary
because of free rotation around the bridging unit between the
bipyridine groups.
11274
dx.doi.org/10.1021/ic2008045 |Inorg. Chem. 2011, 50, 11273–11275

Inorganic Chemistry
COMMUNICATION
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: Ronny.Neumann@weizmann.ac.il.
’ ACKNOWLEDGMENT
This research was supported by the Divadol Foundation, the
Peter and Bernice Catalysis Fund, and the Helen and Martin
Kimmel Center for Molecular Design. R.N. is the Rebecca and
Israel Sieff Professor of Chemistry.
’ REFERENCES
(1) (a) Novak, E.; Fodor, K.; Szailer, T.; Oszko, A.; Erdoehelyi, A.
Top. Catal. 2002, 20, 107–117. (b) Traynor, A. J.; Jensen, R. J. Ind. Eng.
Chem. Res. 2002, 41, 1935–1939. (c) Lim, H.-W.; Park, M.-J.; Kang,
S.-H.; Chae, H.-J.; Bae, J. W.; Jun, K.-W. Ind. Eng. Chem. Res. 2009,
48, 10448–10455.
(2) (a) Kaplan, V.; Wachtel, E.; Gartsman, K.; Feldman, Y.; Lubomirsky,
I. J. Electochem. Soc. 2010, 157, B552–B556. (b) Gattrell, M.; Gupta, N.; Co,
A. J. Electroanal. Chem. 2006, 594, 1–19.
(3) Jessop, P. G.; Joꢀo, F.; Tai, C. C. Coord. Chem. Rev. 2004,
248, 2425–2442. Leitner, W. Angew. Chem., Int. Ed. 1995, 34, 2207–2221.
(4) (a) Morris, A. J.; Meyer, G. J.; Fujita, E. Acc. Chem. Res. 2009,
42, 1983–1994. (b) Takeda, H.; Ishitani, O. Coord. Chem. Rev. 2010,
254, 346–354. (c) Fujita, E. Coord. Chem. Rev. 1999, 185ꢀ186, 373–384.
(5) (a) Ettedgui, J.; Diskin-Posner, Y.; Weiner, L.; Neumann, R.
J. Am. Chem. Soc. 2011, 132, 188–190. (b) Teramura, K.; Tsuneoka, H.;
Shishido, T.; Tanaka, T. Chem. Phys. Lett. 2008, 467, 191–194.
(c) Kohno, Y.; Ishikawa, H.; Tanaka, T.; Funabiki, T.; Yoshida, S. Phys.
Chem. Chem. Phys. 2001, 3, 1108–1114. (d) Lo, C.-C.; Hung, C.-H.;
Yuan, C.-S.; Wu, J.-F. Sol. Energy Mater. Sol. Cells 2007, 91, 1765–1774.
(6) (a) Lin, W.; Han, H.; Frei, H. J. Phys. Chem. B 2004, 108,
18269–18273. (b) Anpo, M.; Yamashita, H.; Ichihashi, Y.; Fujii, Y.;
Honda, M. J. Phys. Chem. B 1997, 101, 2632–2636. (c) Liu, Q.; Zhou, Y.;
Kou, J.; Chen, X.; Tian, Z.; Gao, J.; Yan, S.; Zou, Z. J. Am. Chem. Soc.
2010, 132, 14385–14387.
(7) (a) Bart, S. C.; Anthon, C.; Heinemann, F. W.; Bill, E.;
Edelstein, N. M.; Meyer, K. J. Am. Chem. Soc. 2008, 130, 12536–12546.
(b) Castro-Rodriguez, I.; Meyer, K. J. Am. Chem. Soc. 2005, 127, 11242–
11243.
(c) Whited, M. T.; Grubbs, R. H. J. Am. Chem. Soc. 2008, 130, 5874–
5875. (d) Lu, C. C.; Saouma, C. T.; Day, M. W.; Peters, J. C. J. Am. Chem. Soc.
2007, 129, 4–5. (e) Laitar, D. S.; Mueller, P.; Sadighi, J. P. J. Am. Chem. Soc.
2005, 127, 17196–17197. (f) Ashley, A. E.; Thompson, A. L.; O’Hare, D.
Angew. Chem., Int. Ed. 2009,48, 9839–9843. (g) Riduan, S. N.; Zhang, Y.; Ying,
J. Y. Angew. Chem., Int. Ed. 2009, 48, 3322–3325. (h) Silvia, J. S.; Cummins,
C. C. J. Am. Chem. Soc. 2010, 132, 2169–2171.
Figure 2. 13C NMR (top) and IR (bottom) spectra of the (CxSy)n
polymer.
where x > y and indicates that, in part, both CꢀS bonds of CS2
had been cleaved in the photocatalytic process.
The (CxSy)n polymers are only soluble in apolar solvents such
1
as DMSO. H NMR in DMSO-d6 again showed residue of
catalyst but no other hydrogen-atom-containing compound. The
13C NMR spectrum (Figure 2, top) showed seven peaks
associated with the (CxSy)n polymer at 25.57, 25.97, 29.02,
29.22, 29.78, and 33.00 ppm for CꢀS single bonds and 61.18
ppm for CdS double bonds. Clearly, the 13C NMR spectrum is
indicative of various different types of CꢀS bond connectivity. The
IR spectrum (Figure 2, bottom) likewise showed several distinctive
peaks at 1452, 1084, and 1018 cmꢀ1 attributable to the various
possible CꢀS bonds.13d,14 Multiple matrix-assisted laser desorp-
tion ionization time-of-flight MS measurements were of low quality
but showed mostly molecular weights between 350 and 580 amu.
The photocatalytic splitting of CO2 was also attempted using
[Ru2L1(p-cymene)2Cl2]2+ as the catalyst in THF/H2O in var-
ious ratios as the solvent. Although on one or two occasions we
observed the formation of CO, the experiments were not
sufficiently repeatable. More typically, the formation of molec-
ular hydrogen was observed, which we associate with catalyst
decomposition because it is well-known that photoirradiation of
tertiary amines leads to decomposition via the formation of
cation radicals.4 Therefore, although we have successfully de-
signed an efficient catalyst for the photocleavage of CS2 with
visible light to its components, molecular sulfur, and a (CxSy)n
polymer, similar photocleavage of CO2 to CO and O2 will require
a more robust catalyst design.
(8) Chen, Y.; Peng, Y.; Chen, P.; Zhao, J.; Liu, L.; Li, Y.; Chen, S.;
Qu, J. Dalton Trans. 2010, 39, 3020–3025.
(9) (a) Zahradnik, R.; Jungwirth, P.; Urban, J.; Polasek, M. Helv.
Chim. Acta 1994, 77, 1810–1816. (b) Khenkin, A. M.; Efremenko, I.;
Weiner, L.; Martin, J. M. L.; Neumann, R. Chem.—Eur. J. 2010, 16,
1356–1364.
(10) Madhu, V.; Ekambaram, B.; Shimon, L. J. W.; Diskin-Posner,
Y.; Leitus, G.; Neumann, R. Dalton Trans. 2010, 39, 7266–7275.
(11) (a) Schubert, U. S.; Eschbaumer, C.; Heller, M. Org. Lett. 2000,
2, 3373–3376. (b) Smith, A. P.; Lamba, J. J. S.; Fraser, C. L. Org. Synth.
2002, 78, 82–90.
(12) (a) Bridgman, P. W. Proc. Am. Acad. Arts Sci. 1941, 74, 399–424.
(b) Whalley, E. Can. J. Chem. 1960, 38, 2105–2108. (c) Butcher, E. G.;
Alsop, M.; Weston, J. A.; Gebbie, H. A. Nature 1963, 199, 756–758.
(d) Chan, W. S.; Jonscher, A. K. Phys. Status Solidi 1969, 32, 749–761.
(13) Zmolek, P. B.; Sohn, H.; Gantzel, P. K.; Trogler, W. C. J. Am.
Chem. Soc. 2001, 123, 1199–1207.
’ ASSOCIATED CONTENT
S
Supporting Information. X-ray crystallographic data in
b
CIF format, full experimental details, and additional structures,
spectra, and characterizations. This material is available free of
charge via the Internet at http://pubs.acs.org.
(14) Sadhir, R. K.; Schoch, K. F. Chem. Mater. 1996, 8, 1281–1286.
11275
dx.doi.org/10.1021/ic2008045 |Inorg. Chem. 2011, 50, 11273–11275