The RuIVO-catalyzed sulfoxidation: A gated mechanism where O to S linkage isomerization switches between different efficiencies

Article abstract of DOI:10.1039/b924614b

Two Ru^IVO catalysts with either a pentadentate bispidine ligand L¹ or a bidentate pyrazolate L²/terpy L³ combination of ligands have very different efficiencies as oxygen transfer catalysts for the selective oxidation of sulfides to sulfoxides: the [Ru ^II(L¹)(solvent)]²⁺/iodosyl benzene system has an initial TOF of approx. 40 h^-1 and quantitative yield, with [Ru^II(L²)(L³)(solvent)]⁺ the initial TOF is approx. 12 h^-1 with a maximum yield of approx. 60%. By experiment (cyclovoltametry) it is shown that there is S- to O-linkage isomerization of the Ru^II sulfoxide product complex, and this may partially switch off the catalytic cycle for the L²/L ³-based catalyst. It emerges that the reasons for the reduced efficiency in the case of the L²/L³, in comparison with the L¹-based catalyst, are a more efficient linkage isomerization, a more stable S-bonded, in comparison with the O-bonded, Ru^II-based isomer, and inefficient ligand exchange in the product (hydrolysis produces the free sulfoxide and the Ru^II precatalyst). These interpretations are qualitatively in good agreement with preliminary DFT-based data. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/b924614b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
The RuIV O-catalyzed sulfoxidation: a gated mechanism where O to S
linkage isomerization switches between different efficiencies†
=
Jordi Benet-Buchholz,a Peter Comba,*b Antoni Llobet,*a Stephan Roeser,a Prabha Vadivelub and
Sebastian Wiesnerb
Received 23rd November 2009, Accepted 22nd January 2010
First published as an Advance Article on the web 16th February 2010
DOI: 10.1039/b924614b
1
Two RuIV O catalysts with either a pentadentate bispidine ligand L or a bidentate
=
pyrazolate L2/terpy L3 combination of ligands have very different efficiencies as oxygen transfer
catalysts for the selective oxidation of sulfides to sulfoxides: the [RuII(L1)(solvent)]2+/iodosyl benzene
system has an initial TOF of approx. 40 h-1 and quantitative yield, with [RuII(L2)(L3)(solvent)]+ the
initial TOF is approx. 12 h-1 with a maximum yield of approx. 60%. By experiment (cyclovoltametry) it
is shown that there is S- to O-linkage isomerization of the RuII sulfoxide product complex, and this may
partially switch off the catalytic cycle for the L2/L3-based catalyst. It emerges that the reasons for the
reduced efficiency in the case of the L2/L3, in comparison with the L1-based catalyst, are a more
efficient linkage isomerization, a more stable S-bonded, in comparison with the O-bonded, RuII-based
isomer, and inefficient ligand exchange in the product (hydrolysis produces the free sulfoxide and the
RuII precatalyst). These interpretations are qualitatively in good agreement with preliminary
DFT-based data.
Introduction
Due to the versatility of sulfoxides, the efficient and selective
oxidation of sulfides is of importance in preparative organic
chemistry, specifically in the area of asymmetric synthesis and
enantioselective catalysis.1-5 Ruthenium- and iron-based catalysts
with pentadentate ligand systems have been found to be partic-
ularly efficient. We have used high-valent iron complexes with
a variety of bispidine ligands in oxidation catalysis,6,7 and more
recently have also studied the corresponding ruthenium chemistry8
to compare the relative reactivities and selectivities, specifically
because with ruthenium there are no ambiguities with respect to
the spin states.9 Bispidines (see Chart 1 for ligand structures) are
very rigid and widely variable ligand systems10,11 and, specifically
in the area of non-heme iron oxidation catalysis, have yielded a
wealth of unique results.12-14
Chart 1 Structures of the three ligands used.
Here, we report the preparation and characterization of
two RuIV O-based catalysts (see Fig. 1 for the structures of
=
[Ru(L1)(dmso-S)]2+ (computed structure, see below; the X-ray
structures of the corresponding Cl- and OH2 complexes have been
reported8) and [Ru(L2)(L3)(dmso-S)]+ (X-ray crystal structure);
for L1, L2, L3, see Chart 1, dmso = dimethylsulfoxide), which
both selectively oxidize thioanisol to the corresponding sulfoxide
but with very different efficiencies. Mechanistic studies reported
here indicate that the strikingly different efficiencies are due to a
Fig. 1 Plots of the molecular structures of [Ru(L2)(L3)(dmso-S)]+ (X-ray
single crystal structure; ellipsoids are drawn at the 25% probability level)
and [Ru(L1)(dmso-S)]2+ (calculated; the corresponding structure of the
aqua complex has been reported8).
aInstitute of Chemical Research of Catalonia (ICIQ), Av. Pa¨ısos Catalans
16, E-43007 Tarragona, Spain
bUniversita¨t Heidelberg, Anorganisch-Chemisches Institut, INF 270, D-
69120 Heidelberg, Germany. E-mail: peter.comba@aci.uni-heidelberg.de;
Fax: +49-6221-546617
gated mechanism, involving for the less efficient catalyst an O to S
linkage isomerization, which stabilizes the RuII-sulfoxide product
and therefore prevents fast ligand exchange to allow reoxidation
of RuII to the catalytically active high-valent species: if the
† Electronic supplementary information (ESI) available: The electrochem-
ical analysis of [Ru(L2)(L3)(OH2)]2+, the DFT-based computational study
and crystal data. CCDC reference number 755687. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/b924614b
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3315–3320 | 3315
©

View Article Online
yielding a RuII-sulfoxide-O complex, which may isomerize to the
corresponding RuII-sulfoxide-S intermediate;1 by ligand exchange,
both linkage isomers then produce the metal-free sulfoxide
product and the RuII-OH2 precursor complex, which, in our
=
◦
1
2+ a
˚
Table 1 Selected bond distances (A) and angles ( ) for [Ru(L )(dmso)]
and [Ru(L2)(L3)(dmso)]+. Structural data of [Ru(L1)(OH2)]2+ 8 appear for
comparison
[Ru(L2)(L3)-
experiments, is reoxidized by iodosyl benzene to the active RuIV
O
(dmso)]+
[Ru(L1)(dmso)]2+ [Ru(L1)(OH2)]2+
form. RuIV O has been proposed before to be the active oxidant
in catalytic oxidation reactions and, for other similar ligand
systems, high-valent ruthenium complexes have been trapped and
characterized, e.g. by ESI mass spectrometry.9 Moreover, iodosyl
benzene is an oxygen atom transfer agent, leading to a two
electron oxidation and, in the absence of a Ru-based catalyst
under otherwise identical conditions, we have shown that it is not
able to oxidize sulfide substrates. The pH dependence of the one-
electron potentials (Pourbaix plots) of RuII complexes has been
used to show the formation of RuIII and RuIV intermediates,16,17
and the corresponding electrochemistry of [Ru(L1)(OH2)]2+ has
=
M–N3
M–N7
M–Npy1
M–Npy2
M–Npy3
M–X
1.967 (4)
2.070 (4)
2.086 (5)
2.077 (5)
2.128 (5)
2.265 (2)
2.104
2.190
2.040
2.049
2.086
2.267
2.069 (2)
2.119 (2)
2.053 (3)
2.040 (3)
2.040 (3)
2.134 (2)
(X = dmso, OH2)
Npy1–M–Npy2
N3–M–Npy3
N7–M–X
N3 ◊ ◊ ◊ N7
159.5 (2)
170.7 (2)
173.7 (1)
2.953 (6)
163.8
164.1
170.2
2.92
163.5 (1)
172.2 (1)
175.1 (1)
2.91 (1)
a Calculated structure; geometry optimization was done using SVWN/
LACVP** as implemented in the Jaguar 6.5 package.
been described in detail and unambiguously demonstrates the
8
formation of RuIV O. Similar differential pulse voltammetric
=
measurements of the second catalyst, [Ru(L2)(L3)(OH2)]2+, at two
different pH values (DPV; see ESI, Fig. S1†) show two signals for
the RuIII/II and RuIV/III redox-couples. These potentials are proton
coupled one electron processes, as shown by a shift of the potential
according to the Nernst equation, i.e. approx. 59 mV per pH unit,
RuII-sulfoxide product isomerizes to an inert S-bonded form,
it switches off the efficient catalytic pathway. The linkage iso-
merization is analyzed in detail and strategies to prevent the
partial inhibition and thus increase the efficiency are discussed.
Preliminary data of a qualitative DFT calculation support this
interpretation.
and therefore involve RuII-OH2, RuIII-OH and RuIV O. A similar
behavior was observed for [Ru(L1)(OH2)]2+.8
=
Ligand exchange (solvation, formation of the aqua complex
in the reaction studied here) of the two isomeric RuII-sulfoxide
complexes (S- and O-bonded) is believed to follow a mechanism
with a dissociative activation mode (Id),18,19 and the linkage isomer
withthe softer S-bondeddonor is expected tobe considerably more
stable and, therefore, to significantly decrease the exchange rate
and partially inhibit catalysis. In fact, in [(bpy)(terpy)RuII(dmso-
S)]2+ (bpy = 2,2¢-bipyridine; terpy = L3), the substitution of dmso
with water has the exceedingly slow rate of kaq = (1.46 0.04)
Results and discussion
The RuII complexes were obtained in good yield from equimolar
amounts of [Ru(Cl)2(dmso)4] and the ligands, refluxed for 18 h
under Ar in pure MeOH.15 The structural properties of the
two complexes are rather similar to each other (see Fig. 1
and Table 1; note that the geometric parameters of the com-
puted structure of the [Ru(L1)(dmso-S)]2+ complex are in good
agreement with the earlier reported experimental data of the
corresponding aqua complex8). The only possibly significant
difference is that the site of the coordinated sulfoxide (dmso in
the reported structures and in the electrochemical experiments,
thioanisoloxide in the catalytic experiments and DFT calculations)
is open in the [Ru(L2)(L3)(substrate)]+-based system, while in the
[Ru(L1)(substrate)]2+ system the N3-appended methyl substituent
may lead to some steric congestion (see Fig. 1).
◦
10-5 s-1 at 50 C.20 The inertness of the two linkage isomers
of the two catalysts and the corresponding exchange rates were
probed experimentally by scan-rate-dependent CV of the RuIII/II
dmso complexes (catalysis product analogs; see Fig. 2). For the
[Ru(L1)(dmso)]2+-based system there is a much larger amount of
O- than S-bonded isomer on the returning scan, when scans were
at 100 mV s-1. In sharp contrast, for the [Ru(L2)(L3)(dmso)]+-based
system, the amount of S-bonded isomer is much larger than the
O-bonded form. The quantitative analysis of the CV’s, based on
a square scheme involving the linkage isomer equilibria of the
oxidized and reduced forms and the two electron transfer steps15
also shown in Fig. 2, indicates that, in the catalytically relevant RuII
oxidation state (see Scheme 1), the L1-based bispidine complex
remains primarily O-bonded, while the L2,L3-based complex
isomerizes to the more inert sulfur-bonded form (see Table 2),
and this may partially block catalysis.
The catalytic cycle studied here (see Scheme 1) involves the
usually observed oxygen transfer from the high-valent metal–
oxygen fragment, RuIV O in the present case, to the sulfur atom,
=
Indeed, the differing stability and reactivity of the O-bonded
linkage isomer of the two RuII complexes leads to a remark-
able difference in the catalytic efficiency (see Fig. 3). While
[Ru(L1)(solvent)]2+, with a largely isomerization-inert RuII-dmso-
O complex, is one of the most efficient Ru-based sulfoxidation
pre-catalysts, [Ru(L2)(L3)(solvent)]+ leads to a slower and less
efficient catalytic transformation, and this is assumed to be
due to the relatively fast isomerization to the substitution-inert
S-bonded isomer. Both catalysts selectively yield the sulfoxide
Scheme 1 Catalytic cycle for the RuII-catalyzed oxidation of sulfides to
sulfoxides, involving the linkage isomerization gate RuII-O → RuII-S.
3316 | Dalton Trans., 2010, 39, 3315–3320
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 2 Cyclic voltammograms (CV) in CH2Cl2 vs. SSCE, containing 0.1 M TBAPF6, 100 mV s-1 of (a) 1 mM [Ru(L2)(L3)(dmso)]+ + 4.5 mM t-BuOK;
(b) 1 mM [Ru(L1)(dmso)]+; (c,d) CV’s at different scan rates (in mV s-1, see color code), starting at 1.2 V (c) and 1.6 V (d; before starting a scan, a constant
potential of 1.2 V (a,c) and 1.6 V (b,d) for 180 s was applied to ensure complete equilibration); (e) square scheme of the electron transfer and linkage
isomerization processes.
product but, under the conditions of the experiment (see Fig. 3
and Experimental), the bispidine-based catalyst has a 3 times
higher initial TOF and leads to 100% conversion, while the yield
with [Ru(L2)(L3)(solvent)]+ as catalyst is only about 60% (see
Fig. 3). This indicates that part of the L2/L3-based catalyst is
inhibited and this is consistent with the hypothesis that the O- to
S-isomerization of the coordinated sulfoxide is responsible for the
reduced efficiency.
We have tried to support this interpretation with a density-
functional-theory-based (DFT) analysis. Despite a careful valida-
tion of functionals and basis sets, the error limit, specifically with
respect to the energies of the transition states and intermediates,
is too large for a quantitative interpretation, and this is not
unexpected and has been observed with other ruthenium-S-donor
systems.21-23 Therefore, the detailed computational results are not
discussed here in detail, and these are presented, together with their
interpretation, as ESI.† However, a short and qualitative analysis
is warranted, specifically because this is based on relative energies
and structural as well as electronic differences between analogous
transition states or intermediates for very similar catalyst systems.
The two main conclusions are: (i) direct ligand exchange of
the RuII-sulfoxide-O product complex to complete the catalytic
cycle is more efficient than linkage isomerization in both catalyst
systems. This is in agreement with the observation that both RuII
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3315–3320 | 3317
©

View Article Online
With [Ru(L1)(solvent)]2+, we have developed a highly efficient
sulfoxidation catalyst which selectively oxidizes thioanisole to
the corresponding sulfoxide in quantitative yield. The lower
efficiency observed with other Ru-based catalyst systems has
been analyzed by a comparison of [Ru(L1)(solvent)]2+ with
[Ru(L2)(L3)(solvent)]+. The analysis reveals that there is basically
no difference with respect to the oxidation power excerted by the
two catalysts but that the reduced efficiency is primarily due to
a gated product release step, where linkage isomerization of the
coordinated product may lead to inhibition. It is known that
ligand systems with strong s donors lead to a stabilization of
the S-bonded isomer,21,23 and it emerges that this may lead to
inhibition of the catalytic activity. Subtle steric effects may be used
to prevent the formation of this inactive form (destabilization of
the corresponding transition state), and this is an important aspect
for future developments because asymmetric ligands ideally have
sterically demanding groups adjacent to the site where the Ru-O
group interacts with the substrate and product.
Table 2 Thermodynamic and kinetic parameters for the linkage iso-
merization of [Ru(L1)(dmso-S/O)]2+ and [Ru(L2)(L3)(dmso-S/O)]+, deter-
mined by scan-rate-dependent CV (see Fig. 2)
[Ru(L1)(dmso)]2+
[Ru(L2)(L3)(dmso)]+
E
E
1/2, S(V)
1/2, O(V)
O→S
1.24
0.68
0.98
0.41
KII
7.4 ¥ 107
6. 0 ¥ 102
3.6 ¥ 10-2
5.0 ¥ 10-10
1.1 ¥ 10-2
6.5
5.5 ¥ 108
7.81
KIII
kII
S→O
(s-1)a
2.5 ¥ 10-1
4.6 ¥ 10-10
7.7 ¥ 10-2
6.0 ¥ 10-1
O→S
S→O
kII
(s-1)a
(s-1)a
(s-1)a
kIII
kIII
O→S
S→O
a Due to the exceedingly high stability of the O-bonded isomer, the values
for [Ru(L1)(dmso)]2+ could only be obtained by simulation with DIGISIM.
Experimental
General
Chemicals (Aldrich, Fluka) and solvents were of the highest pos-
sible grade and used as purchased. The bispidine ligands and com-
plexes were prepared as described before.8,24 [Ru(L2)(L3)(dmso)]2+
was prepared with a method slightly modified to that described be-
fore for similar compounds.15 Elemental analyses were performed
by the analytical laboratories of the chemical institutes of the
University of Heidelberg.
Fig. 3 Time dependence of the sulfoxidation of thioanisole, with
[Ru(L1)(OH2)](ClO4)2 and [Ru(L2)(L3)(OH2)]ClO4 as catalysts and io-
dosylbenzene diacetate as oxidant (anaerobic conditions, in acetone;
catalyst : oxidant : substrate = 1 : 100 : 1000, maximum TON of 100); initial
TOF for [Ru(L1)(OH2)](ClO4)2 = 39.5, for [Ru(L2)(L3)(OH2)]ClO4 = 12.1.
Electrochemistry
Cyclic voltammetry was measured with a CH Instruments 660c
potentiostat using a standard three electrode cell; a glassy carbon
electrode (5 mm diameter) was used as working electrode, a Pt
wire as auxiliary electrode and a SSCE electrode as reference.
The samples were dissolved in degassed solvents containing the
required supporting electrolyte (potassium perchlorate for water,
tetrabutylammoniumperchlorate for MeOH, tetrabutylammoni-
umhexafluorophosphate for CH2Cl2; 0.1 M). Unless otherwise
stated, the concentrations of the complexes were approx. 1 mM.
complexes discussed here are catalytically active. (ii) There is a
significant difference between the two systems with respect to
the relative stability of the S-bonded linkage isomer: in the L1-
based system, the O-bonded isomer is the more stable, and in
the L2/L3-based complex, it is the less stable linkage isomer. This
has some consequences with respect to the corresponding energy
barriers (isomerization back to the O-bonded isomer and ligand
exchange from the S-bonded isomer to complete the catalytic
cycle): for the more efficient L1-based catalyst with an instable S-
bonded linkage isomer, ligand exchange to reform the pre-catalyst
and produce the free sulfoxide is only slightly less efficient than
with the O-bonded isomer, i.e. the two pathways in Scheme 1
are feasible. For the L1/L2-based catalyst, both energy barriers for
isomerizationbacktothe O-bonded isomer and for product release
are exceedingly high. That is, for the L2/L3-based system, there is
competition between hydrolysis of the initially formed O-bonded
isomer of the product complex to complete the catalytic cycle
and linkage isomerization which inhibits product formation, and
this is as observed experimentally, where the yield of the catalytic
transformation is limited to 60%. In the L1-based system, linkage
isomerization is not a dead end, and experimentally, one therefore
observes 100% transformation.
Catalysis
Gas chromatography was performed by capillary GC with a Var-
ian 8000 instrument with flame ionisation detection, and equipped
with a ZB-1701 column. All products from catalytic runs were
identified by retention time on GC relative to authentic samples.
The quantitative determination was achieved by GC, calibrated
by authentic samples and with naphthalene as internal standard.
General conditions for the oxidation experiments of thioanisol are
given here for the reaction with PhI(OAc)2 as oxidant. A known
amount (approx. 68 mg (210 mmol)) of PhI(OAc)2 was added at
once to a solution of thioanisol (2100 mmol), the catalyst (2.1 mmol)
and the internal standard naphthalene (10 mM) in acetone (4 ml)
at 25 ◦C. Quantification of the sulfoxide product for both catalysts
was done every 60 min for 7 h, using the GC setup discussed above.
3318 | Dalton Trans., 2010, 39, 3315–3320
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
X-Ray crystallography
ESI for the atom numbering scheme used†): d = 10.15 (d, 3J1–2
=
5.5 Hz, 1H; H1), 8.93 (d, 3J21–22 = 8.4 Hz, 2H; H21, H23), 8.78 (d,
3J18–17 = 8.1 Hz, 2H; H18, H26), 8.63 (m, 2H; H4, H22), 8.49 (t,
Data collection was performed on a Bruker Nonius FR 591 system
equipped with a multilayer Montel 200 mirror monochromator
3
3
3J3–4 = J3–2 = 7.4 Hz, 1H; H3), 8.29 (t, 3J17–18 = J17–16 = 8.0 Hz,
˚
Mo Ka (l = 0.71073 A) radiation and an Apex II CCD detector.
2H; H17, H27), 8.18 (d, 3J15–16 = 5.5 Hz, 2H; H15, H29), 8.03 (t,
The molecular structure was solved by direct methods25,26 and
refined on F2 by full matrix least squares techniques using the
SHELX TL package with anisotropic thermal parameters.27,28
Non-hydrogen atoms were refined anisotropically, the hydrogen
atoms were placed in ideal positions.
3
3J2–3 = J2–1 = 6.5 Hz, 1H; H2), 7.76 (s, 1H; H7), 7.68 (t, 3J16–17
=
3J16–18 = 6.5 Hz, 2H; H16, H28), 7.38 (m, 5H; H10, H11, H12, H13,
H14), 2.63 (s, 6H; –CH3). 13C{ H} NMR (acetone-d6): d = 158.4
1
(C5, C19), 158.0 (C20), 155.2 (C1), 154.3 (C15), 152.5 (C6), 149.2
(C8), 139.7 (C17), 139.0 (C3, C22), 129.1 and 126.3 (C2, C10,
C11, C12, C16), 127.0 (C9), 125.3 (C18), 124.8 (C21), 123.7 (C4),
102.9 (C7), 41.7 (CH3). UV-vis [CH2Cl2; lmax/nm (e/LM-1cm-1)]:
237 (34 270), 274 (44 470), 299 (28 930), 316 (25 930), 393 (6990),
475 (4015). E1/2 Ru(II)/Ru(III): (CH2Cl2 + 0.1 M TBAH) 0.41 V
(DMSO, O-bonded); 0.98 (DMSO, S-bonded) vs. SSCE.
Synthesis
[Ru(HL2)(L3)(Cl)](PF6). A 140 mg (0.63 mmol) sample of 3-
pyridyl-5-phenyl-1H-pyrazole (HL2) and 305 mg (0.63 mmol)
of Ru(Cl)2(DMSO)4 were dissolved in 20 ml of freshly distilled
methanol, and the resulting solution was refluxed for 18 h under a
static argon atmosphere. The resulting yellow solid was collected
and dried in vacuum. The product was dissolved, together with
110 mg (0.473 mmol) of 2,2¢:6¢,2¢¢-terpyridine, in 75 mL pure
methanol and refluxed for 18 h under argon. The solvent was
then removed in vacuum and the brown solid dissolved in
10 mL of a methanol/NH4OH (aq., 28%) mixture (100 : 1). A
purple solid was removed by filtration, redissolved in 60 mL
of a methanol/NH4PF6(aq., 3 M) mixture (20 : 1) and precipitated
by addition of some H2O to the solution and then decreasing
the volume to yield 167 mg of the product (0.227 mmol, 48%).
Elemental analysis (%) calcd for C29H22ClF6N6PRu·2H2O: C 45.1,
H 3.4, N 10.9; found: C 45.6, H 3.1, N 11.0. 1H NMR (400 MHz,
DMSO-d6, see ESI for the atom numbering scheme used†): d =
Acknowledgements
Generous financial support by the German Science Foundation
(DFG), MICINN and MEC of Spain are gratefully acknowledged
for grants CSD2006-0003 and CTQ2007-67918, and for the
allocation of a doctoral grant to S. R. respectively.
References
1 L. Roecker, J. C. Dobson, W. J. Vining and T. J. Meyer, Inorg. Chem.,
1987, 26, 779.
2 M. H. V. Huynh, L. M. Witham, J. M. Lasker, M. Wetzler, B. Mort,
D. L. Jameson, P. S. White and K. J. Takeuchi, J. Am. Chem. Soc., 2003,
125, 308.
3 O. Hamelin, S. Menage, F. Charnay, M. Chavarot, J.-L. Pierre, J. Pecaut
and M. Fontecave, Inorg. Chem., 2008, 47, 6413.
4 J. Legros and C. Bolm, Chem.–Eur. J., 2005, 11, 1086.
5 A. Correa, O. Garcia Mancheno and C. Bolm, Chem. Soc. Rev., 2008,
37, 1108.
6 M. R. Bukowski, P. Comba, C. Limberg, M. Merz, L. Que Jr. and T.
Wistuba, Angew. Chem., Int. Ed., 2004, 43, 1283.
7 M. R. Bukowski, P. Comba, A. Lienke, C. Limberg, C. Lopez de
Laorden, R. Mas-Balleste, M. Merz and L. Que Jr., Angew. Chem.,
Int. Ed., 2006, 45, 3446.
3
3
9.96 (d, J14–13 = 5.5 Hz, 1H; H14), 8.77 (d, J21–22 = 8 Hz, 2H;
H21, H23), 8.67 (d, 3J18–17 = 8 Hz, 2H; H18, H26), 8.51 (d, 3J11–12
=
8 Hz, 1H; H11), 8.32 (t, 3J12–11 = J12–13 = 7.7 Hz, 1H; H12), 8.18
3
(t, 3J22–21 = 8 Hz, 1H; H22), 7.98 (t, 3J17–16 = J17–18 = 7.8 Hz, 2H;
3
H17, H27), 7.9 (t, 3J1–12 = J13–14 = 6.6 Hz, 1H; H13), 7.71 (s, H8),
3
3
7.64 (d, 3J15–16 = J29–28 = 5.3 Hz, 2H; H15, H29), 7.45 (d, 3J1–2
=
3J5–4 = 6.9 Hz, 2H; H1, H5), 7.38 (m, 5H; H2, H3, H4, H16, H28).
NOESY (400 MHz, DMSO-d6): d = 7.71–7.45 (H8–H1 or H8–
8 J. Benet-Buchholz, P. Comba, A. Llobet, S. Roeser, P. Vadivelu, H.
Wadepohl and S. Wiesner, Dalton Trans., 2009, 5910.
9 S. N. Dhuri, M. S. Seo, Y.-M. Lee, H. Hirao, Y. Wang, W. Nam and S.
Shaik, Angew. Chem., Int. Ed., 2008, 47, 3356.
H5), 7.71–8.51 (H8–H11). 13C{ H} NMR (DMSO-d6): d = 159.7
1
(C19), 159.4 (C14), 153.7 (C9), 152.5 (C15), 146.9 (C7), 137.2
(C12, C17), 134.1 (C22), 129.5 and 127.7 (C2, C3, C6, C16), 126.5
(C1), 125.1 (C13), 123.6 (C18), 122.5 (C11, C21), 122.6 (C20),
102.9 (C8). UV-vis [CH2Cl2; lmax/nm (e/LM-1cm-1)]: 240 (46 610),
281 (37 990), 323 (34 440), 412 (7100), 501 (8320), 655 (1448). E1/2
Ru(II)/Ru(III): (CH2Cl2 + 0.1 M TBAH) 0.785 V vs. SSCE.
10 P. Comba, M. Kerscher and W. Schiek, Prog. Inorg. Chem., 2007, 55,
613.
11 P. Comba, B. Nuber and A. Ramlow, J. Chem. Soc., Dalton Trans.,
1997, 347.
12 J. Bautz, M. Bukowski, M. Kerscher, A. Stubna, P. Comba, A. Lienke,
E. Mu¨nck and L. Que Jr, Angew. Chem., Int. Ed., 2006, 45, 5681.
13 J. Bautz, P. Comba, C. Lopez de Laorden, M. Menzel and G.
Rajaraman, Angew. Chem., Int. Ed., 2007, 46, 8067.
14 P. Comba, M. Maurer and P. Vadivelu, J. Phys. Chem. A, 2008, 112,
13028.
15 C. Sens, R. M., I. Romero, A. Llobet, T. Parella, P. B. Sullivan and J.
Benet-Buchholz, Inorg. Chem., 2003, 42, 2040.
16 C. Sens, M. Rodriguez, I. Romero, T. Parella, J. Benet-Buchholz and
A. Llobet, Inorg. Chem., 2003, 42, 8385.
17 C. Sens, I. Romero, A. Llobet, T. Pasella and J. Benet-Buchholz, J. Am.
Chem. Soc., 2004, 126, 7798.
18 N. Aebischer, E. Sidorenkova, M. Ravera, G. Laurenczy, D. Osella, J.
Weber and A. E. Merbach, Inorg. Chem., 1997, 36, 6009.
19 D. De Vito, H. Siderenkova, F. P. Rotzinger, J. Weber and A. E.
Merbach, Inorg. Chem., 2000, 39, 5547.
20 M. J. Root and E. Deutsch, Inorg. Chem., 1985, 24, 1464.
21 N. S. Panina and M. Calligaris, Inorg. Chim. Acta, 2002, 334, 165.
22 M. Kato, T. Takayanagi, T. Fujihara and A. Nagasawa, Inorg. Chim.
Acta, 2009, 362, 1199.
[Ru(HL2)(L3)(dmso-S)](PF6)2. A 200 mg (0.272 mmol) sam-
ple of [Ru(HL2)(L3)(Cl)](PF6) and 61.2 mg (0.272 mmol) of
AgClO4·H2O was dissolved in 120 mL of a pre-degassed 3 : 1
mixture of acetone–H2O. The mixture was refluxed in the absence
of light and a static argon atmosphere for 4 h. After filtering the
formed AgCl through a pad of Celite, 100 eq. (1.9 ml) of DMSO
was added to the filtrate and the mixture was refluxed for an
additional 4 h period. After precipitation of the desired complex by
adding 1 mL of NH4PF6(aq., 3 M) and reducing the reaction volume,
the orange solid was filtered off, washed with a minimum amount
of H2O and Et2O and dried in vacuum, giving 170 mg (0.184 mmol,
68% yield) of a brown solid. Elemental analysis (%) calcd for
C31H28F12N6OP2RuS·H2O: C 39.5, H 3.2, N 8.9, S 3.4; found: C
1
39.8, H 3.2, N 8.7, S 3.2. H NMR (400 MHz, acetone-d6, see
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3315–3320 | 3319
©

View Article Online
23 M. Stener and M. Calligaris, THEOCHEM, 2000, 497, 91.
24 H. Bo¨rzel, P. Comba, K. S. Hagen, M. Merz, Y. D. Lampeka, A. Lienke,
G. Linti, H. Pritzkow and L. V. Tsymbal, Inorg. Chim. Acta, 2002, 337,
407.
25 P. T. Beurskens, in Crystallographic Computing 3, ed. G. M. Sheldrick,
C. Kru¨ger and R. Goddard, Clarendon Press, Oxford, UK, 1985, p.
216.
26 P. T. Beurskens, G. Beurskens, R. de Gelder, J. M. M. Smits, S.
Garcia-Granda and R. O. Gould, DIRDIF-2008, Radboud University
Nijmegen, The Netherlands, 2008.
27 G. M. Sheldrick, SHELXL-97, Program for refinement of crystal
structures, University of Go¨ttingen, Germany, 1997.
28 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112.
3320 | Dalton Trans., 2010, 39, 3315–3320
This journal is
The Royal Society of Chemistry 2010
©