Ruthenium complexes of tripodal ligands with pyridine and triazole arms: Subtle tuning of thermal, electrochemical, and photochemical reactivity

Article abstract of DOI:10.1002/chem.201303640

Electrochemical and photochemical bond-activation steps are important for a variety of chemical transformations. We present here four new complexes, [Ru(Lⁿ)(dmso)(Cl)]PF₆ (1-4), where Lⁿ is a tripodal amine ligand with 4-n

Full text of DOI:10.1002/chem.201303640

DOI: 10.1002/chem.201303640
Full Paper
&
Ruthenium Complexes
Ruthenium Complexes of Tripodal Ligands with Pyridine and
Triazole Arms: Subtle Tuning of Thermal, Electrochemical, and
Photochemical Reactivity
Fritz Weisser,[a] Stephan Hohloch,[a] Sebastian Plebst,[b] David Schweinfurth,[a] and
Biprajit Sarkar*[a]
Abstract: Electrochemical and photochemical bond-activa-
tion steps are important for a variety of chemical transforma-
tions. We present here four new complexes, [Ru(Ln)-
(dmso)(Cl)]PF6 (1–4), where Ln is a tripodal amine ligand with
4Àn pyridylmethyl arms and nÀ1 triazolylmethyl arms.
Structural comparisons show that the triazoles bind closer to
the Ru center than the pyridines. For L2, two isomers (with
respect to the position of the triazole arm, equatorial or
axial), trans-2sym and trans-2un, could be separated and com-
pared. The increase in the number of the triazole arms in
the ligand has almost no effect on the RuII/RuIII oxidation po-
tentials, but it increases the stability of the RuÀSdmso bond.
Hence, the oxidation waves become more reversible from
trans-1 to trans-4, and whereas the dmso ligand readily dis-
sociates from trans-1 upon heating or irradiation with UV
light, the RuÀS bond of trans-4 remains perfectly stable
under the same conditions. The strength of the RuÀS bond
is not only influenced by the number of triazole arms but
also by their position, as evidenced by the difference in
redox behavior and reactivity of the two isomers, trans-2sym
and trans-2un. A mechanistic picture for the electrochemical,
thermal, and photochemical bond activation is discussed
with data from NMR spectroscopy, cyclic voltammetry, and
spectroelectrochemistry.
generate high-valent metal–oxo species,[1l] which play an im-
portant role as active intermediates in various oxidases and
oxygenases.[1l,m] RuIV=O complexes with TPA showed high reac-
tivity towards the oxidative functionalization of alkanes.[1h,i]
This kind of functionalization is key to a more economical use
of natural resources like oil, coal, and gas. Various Ru com-
plexes of TPA catalyze the oxidation, epoxidation, and hydroxy-
lation of alkanes and alkenes.[1h–j,2]
Introduction
The tetradentate tripodal ligand, TPA [tris(2-pyridylmethyl)-
amine], and its derivatives have been widely used in coordina-
tion chemistry during the last decades.[1] Its s-donating central
amine, its three p-accepting pyridylmethyl arms, and the
mobility of these arms allow TPA to accommodate a variety of
metal centers with very different coordination geometries and
electronic structures.[1] These properties make it an ideal ligand
for the synthesis of redox or catalytically active species, be-
cause TPA can adapt to electronic and steric changes at the
metal center.
Yamaguchi et al. used complexes of the formula, [(L)Ru-
(dmso)(Cl)]+ where L is TPA or one of its derivatives, to cata-
,
lyze the conversion of adamantane into the corresponding al-
cohols.[3] They investigated the effect of different substituents
in position 5 of the pyridyl moieties of TPA on the reactivity of
the complexes. They could show that the addition of methoxy-
carbonyl groups increased the catalytic activity of the com-
plexes. This kind of complex screening through ligand tuning
is required for the development of effective catalytic systems.
The most common efforts to tune the electronic and steric
properties of TPA have included adding and varying substitu-
ents in positions 5 and 6 of the pyridyl arms.[1]
Therefore, TPA and its derivatives have often been used to
synthesize model complexes of metalloenzymes[1a,l,m] and to
[a] F. Weisser, S. Hohloch, Dr. D. Schweinfurth, Prof. Dr. B. Sarkar
Institut fꢀr Chemie und Biochemie, Freie Universitꢁt Berlin
Fabeckstraße 34–36, 14195 Berlin (Germany)
Fax: (+49)30-838-53310
E-mail: biprajit.sarkar@fu-berlin.de
[b] S. Plebst
An exceptionally efficient method for the easy variation of li-
gands and the modular build-up of compound libraries is the
copper-catalyzed azide–alkyne cycloaddition, one of the click
reactions.[4] In recent years, this synthetic method has gained
a lot of attention, and the resulting triazoles have found more
and more applications as ligands in coordination chemistry.[5]
For example, triazole analogues of bipyridine (bpy) and terpyri-
dine (terpy) have been used to synthesize complexes that
Institut fꢀr Anorganische Chemie, Universitꢁt Stuttgart
Pfaffenwaldring 55, 70550 Stuttgart (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201303640 and includes H NMR spectra of
1
the complexes in different solvents, cyclic voltammograms, differential
pulse voltammograms, CV simulations, EPR spectra, UV/Vis spectra, spec-
troelectrochemistry, structure-refinement parameters, reactivity experiments
1
monitored with H NMR and UV/Vis spectroscopy, and kinetic studies of the
light-driven DMSO exchange.
Chem. Eur. J. 2014, 20, 781 – 793
781
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
plexes to study the effect of the triazole–pyridine substitution
on the reactivity and redox behavior of this system.
Results and Discussion
1
Synthesis and H NMR spectroscopy
The ligands, TPA (L1)[9] and TBTA (L4),[10] were synthesized by re-
ported procedures. Even though the mixed-donor ligands L2
and L3 (Scheme 1) were reported previously,[7a] we have used
an alternate synthetic route for their synthesis (see the Sup-
porting Information). The click reaction for the synthesis of L2
and L3 was carried out by using CuI–N-heterocyclic carbene
complexes, as pioneered in our research group as catalysts.[11]
This method required low catalyst loadings and provided the
ligands in good yields.
Scheme 1. Tripodal ligands L1–L4.
Previous work by Yamaguchi et al. showed that the complex
[Ru(L)(dmso)(Cl)]+ with L=TPA forms two isomers with respect
to the position of the chlorido and the dmso ligand.[3] One
isomer has the dmso ligand trans to the central amine nitro-
gen atom of TPA (from here on referred to as “trans isomer”,
Scheme 2) and the other isomer has the dmso ligand posi-
tioned cis to the central nitrogen atom (“cis isomer”,
Scheme 2). Under reaction conditions with short heating (heat-
ing at reflux in methanol for 2 h), both the cis and the trans
isomer are formed, and they can be isolated by fractioned re-
crystallization in 47 and 29% yield, respectively. The two iso-
mers can be clearly distinguished by the resonances of the
dmso ligand in the 1H NMR spectrum. The dmso methyl
groups of the cis isomer show a singlet at 2.9 ppm in CDCl3,
whereas the corresponding singlet of the trans isomer appears
at 3.5 ppm. Yamaguchi et al. could also show that the cis
isomer is thermodynamically more stable than the trans
isomer. When the trans isomer is heated in DMSO at 1008C, it
transforms into the cis isomer.
show interesting differences to their bpy and terpy counter-
parts.[5,6] Although this method has been used to create TPA
analogues and derivatives,[7] there has been no systematic in-
vestigation of the effect of the pyridine–triazole substitution
on the properties of metal complexes.
In our work with triazole ligands, we have investigated the
coordination chemistry of tris[(1-benzyl-1H-1,2,3-triazol-4-yl)-
methyl]amine (TBTA),[8] a tripodal, tetradentate ligand, which
looks very similar to TPA (Scheme 1). In TBTA, the pyridyl
groups of TPA are replaced with N-benzyltriazolyl groups. By
using click chemistry, the substituents at the triazole ring can
be readily changed, thus allowing for the easy variation of
steric demand, the introduction of additional donors, or the
anchoring of complexes on surfaces. When we investigated 3d
transition metal complexes of TBTA, we noticed significant, at
times drastic, differences in their electronic and magnetic prop-
erties, as well as their reactivity,
compared to their TPA counter-
parts.[8]
The prospect of easily tunable
ligand derivatives led us to
apply this ligand system to the
field of Ru–TPA complexes. By
successively replacing the pyridyl
arms of TPA with triazolyl
groups, it should be possible to
tune the electronic properties
and the reactivity of the corre-
sponding Ru complexes. In this
work, we present a series of RuII
complexes of the formula,
[Ru(Ln)(dmso)(Cl)]PF6
(1–4),
where Ln (n=1–4) is a tripodal,
tetradentate ligand with nÀ1 tri-
azole arms and 4Àn pyridyl arms
(Scheme 1 and 2). We investigat-
ed the electronic and spectro-
1
Scheme 2. Complexes [Ru(Ln)(dmso)(Cl)]PF6. H NMR resonances assigned to the indicated protons (a–x) are
scopic properties of these com- shown in Figure 1.
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
782
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The conditions used for the complex synthesis in this work
were slightly different, that is, the Ru precursor and the tripo-
dal ligand were heated at reflux overnight (ca. 12 h). Under
1
these conditions, the trans–cis ratios, determined from H NMR
spectra of the crude products, vary with the tripodal ligand.
With L1 (TPA, complex 1), the cis isomer was formed exclusively,
whereas with L4 (TBTA, complex 4), the trans isomer was
formed exclusively, and the complexes of L2 and L3 (com-
plexes 2 and 3) have cis–trans ratios of 1:4 and 1:15, respec-
tively. In Table 1, the positions of the signals assigned to the
dmso methyl groups of the isomers are listed (see Figures S19
and S20 in the Supporting Information). For the complexes
with L2–L4, only the trans isomers were isolated after column
chromatography.
Figure 1. 1H NMR spectra measured in CD3CN. For assignment of the signals
see Scheme 2.
Table 1. 1H NMR chemical shifts of the methyl groups of the dmso li-
gands in crude products of 1–4 in CDCl3. Percentage of the respective
isomers (see Scheme 1) is given in parentheses.
(at 4.37 and 5.40 ppm) is observed (Figure 1, signals e and f).
Finally, because of the symmetrical coordination environment,
the dmso methyl groups of trans-2sym only show one singlet,
which integrates for six protons (Figure 1, signal g).
Trans/un
Trans/sym
Cis/un
Cis/sym
[ppm], ([%])
[ppm], ([%])
[ppm], ([%])
[ppm], ([%])
1
2
3
4
3.50[a]
3.68 (27)
3.56 (8)
3.66 (100)
2.92 (100)
2.93 (7)
2.97 (3)
3.47, 3.59 (55)
3.51, 3.84 (86)
2.89, 3.01 (11)
2.92, 2.99 (3)
The unsymmetrical isomer, trans-2un, has two different pyri-
dine arms, one in an axial and one in an equatorial position.
1
Therefore, its H NMR spectrum shows two sets of pyridine sig-
[a] Obtained under different reaction conditions, see Ref. [13].
nals (four doublets, four triplets; Figure 1, signalsh and i) in the
aromatic region. All four CH2 groups are different, and three of
them have two protons with different chemical environments.
This difference leads to geminal coupling and results in six
pseudo doublets with large coupling constants (JAB =14–15 Hz)
and one singlet in the range 4.50–5.80 ppm (Figure 1, sig-
nals m–o). Finally, because of the unsymmetrical coordination
environment, the dmso methyl groups of trans-2un show two
singlets, which each integrate for three protons (Figure 1, sig-
nal p). For trans-3, the symmetrical and the unsymmetrical iso-
mers could not be separated, and the resonances of the major
isomer, trans-3un, are assigned in Scheme 2. By comparing the
integrals of the resonances of the dmso methyl group, in the
1H NMR spectra of crude 2 and 3, the ratios of the symmetrical
to the unsymmetrical isomers could be determined. The ratio
trans-2sym/trans-2un is approximately 1:2, and the ratio trans-
Complexes 2 and 3, with their mixed tripodal ligands, also
form isomers with respect to the position of the individual
arms. Together with the chlorido and the dmso ligand, the trip-
odal ligand forms an octahedral coordination environment
around the Ru center. This results in two distinct positions of
the tripod arms. One arm is positioned in a plane with the
chloride, the dmso, and the central nitrogen atom of the
tripod, and it is referred to in this work as the “equatorial arm”.
The remaining two arms are positioned trans to each other, or-
thogonal to the abovementioned plane, and they are referred
to as the “axial arms”. For 2, where L2 has one triazole arm, this
triazole can be in the equatorial or in an axial position. The
isomer with the triazole in the equatorial position has a mirror
plane through ClÀRuÀSÀNamine and is therefore referred to as
the symmetrical isomer, trans-2sym (compared to the unsym-
metrical isomer trans-2un, Scheme 2).
3
sym/trans-3un is approximately 1:10.
Crystal structures
In the case of trans-2, the two isomers could be separated
by column chromatography. The structure of each isomer
could be determined by their 1H NMR spectra in CD3CN
(Figure 1). Trans-2sym has two identical pyridine arms and,
therefore, exhibits only one set of pyridine signals (two dou-
blets, two triplets; Figure 1, signals a) in the aromatic region.
The CH2 groups of the two identical 2-methylpyridyl arms have
two hydrogen atoms with different chemical environments,
thus resulting in geminal coupling, which can be observed in
the form of two pseudo doublets with large coupling con-
stants (JAB =15 Hz) in the range 4.70–5.80 ppm (Figure 1, sig-
nals d). The two CH2 groups of the triazole arm each have two
chemically identical protons, and, for each CH2 group, a singlet
Single crystals of trans-2sym (Figure 2), trans-2un (Figure 3), and
trans-3un (Figure 4) suitable for X-ray diffraction structure analy-
ses could be obtained. The orientations of the ligands in the
crystal structures match the structures of the complexes pre-
1
dicted from the H NMR data. Table 2 shows a comparison of
important distances and angles of the three new structures
(trans-2sym, trans-2un, and trans-3un) and two structures from the
literature (cis-1,[3] trans–1[3]). In the order trans-1<trans-2<
trans-3, that is, with an increasing number of triazole arms, the
RuÀNamine distance increases from 2.095(2) to 2.134(4) ꢁ, where-
as the distance RuÀS (trans to Namine) slightly decreases in the
same order, from 2.2653(8) to 2.2359(14) ꢁ. The RuÀCl distan-
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
783
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
positions trans to each other and have RuÀN bond lengths of
2.100(2) and 2.101(2) ꢁ, which are among the longest RuÀN
distances in all five structures. In contrast, in the crystal struc-
ture of unsymmetrical trans-2un, the three arms show similar
RuÀN distances in the range 2.066–2.076 ꢁ. The RuÀN dis-
tances of the axial triazole and pyridine arm are 2.069(3) and
2.076(3) ꢁ, respectively. Therefore, the comparison of the two
trans-2 isomers points to a push–pull effect between the mu-
tually trans disposed s-donating triazole ring and the p-accept-
ing pyridine ring in trans-2un (Table 2).
All five crystal structures show the Ru center in a distorted
octahedral coordination environment, in which the axis NaxÀ
RuÀNax is the most distorted, angles being in the range
160.15(10)–163.45(10) ꢁ. For the compounds trans-1, trans-2sym
,
trans-2un, and trans-3un, the axis NamineÀRuÀS is the least distort-
ed, with angles close to 1808, specifically, 176.13(7)–179.52(8)8.
For these four compounds, the two axial rings (triazole or pyri-
dine) are not coplanar, but they are bent away from the chlo-
rido ligand. In trans-2sym and trans-3un, the triazole rings in the
equatorial position are tilted out of the RuÀClÀSÀNamine plane
by 18.6(1) and 19.0(1)8, respectively. In trans-1 and trans-2un,
there are pyridine rings in the equatorial position, and they are
almost coplanar with RuÀClÀSÀNamine, and the tilt angles are
1.8(5) and 2.8(1)8, respectively. It could be argued, whether the
tilt of the equatorial ring is due to packing forces or the nature
of the ring. However, for cis-1, there are three different struc-
tures reported,[3,12] two of which have strongly tilted pyridine
rings in the equatorial position.
Figure 2. ORTEP view of trans-2sym·acetone. Ellipsoids are drawn at the 50%
probability level. H atoms, solvent molecules, and anions are omitted for
clarity.
Finally, the C and O atoms of the dmso ligand in trans-1,
trans-2sym, trans-2un, and trans-3un are positioned in an ecliptic
fashion with respect to the CH2 groups of the tripodal ligand.
In the structures with mixed tripodal ligands, the Odmso atom
always points to a pyridyl ring, and there are short CpyÀ
Hpy···Odmso contacts with the H atom in the 6-position of that
pyridyl ring.
Figure 3. ORTEP view of trans-2un. Ellipsoids are drawn at the 50% probabili-
ty level. H atoms, solvent molecules, and anions are omitted for clarity.
ces of all five structures show no
trend and vary only slightly
around 2.43 ꢁ.
Table 2. Comparison of important bond lengths [ꢁ] and angles [8].
In the order trans-1>trans-
cis-1[a]
trans-1[a]
trans-2sym·acetone
trans-2un
trans-3un
2
sym >trans-2un >trans-3un,
the
RuÀNamine [ꢁ]
RuÀNax [ꢁ]
2.070(3)
2.060(3)
2.078(3)
2.106(3)
6.244(9)
2.095(2)
2.091(2)
2.079(2)
2.095(2)
6.265(6)
2.119(3)
2.101(2)
2.100(2)
2.031(3)[b]
6.232(7)
2.112(3)
2.069(3)[b]
2.076(3)
2.066(3)
6.211(9)
2.134(4)
2.057(4)
2.061(4)[b]
2.052(4)[b]
6.170(12)
sum of the RuÀN distances of
the tripodal ligand decreases
from 6.265(6) to 6.170(12) ꢁ.
With an increasing number of
triazole arms, the arms of the tri-
podal ligand come closer to the
metal center. A remarkable dif-
ference in bond lengths is ob-
served for the two isomers trans-
2sym and trans-2un. In the crystal
structure of trans-2sym, the tria-
zole in the equatorial position
RuÀNeq [ꢁ]
ꢀ RuÀNarms [ꢁ]
RuÀS [ꢁ]
2.2385(10)
2.4321(9)
2.2653(8)
2.4319(9)
1.479(3)
2.2432(9)
2.4285(12)
1.492(2)
2.28
78.14(9)
82.01(10)
160.15(10)
179.52(8)
172.40(7)
87.36(4)
2.2470(10)
2.4462(9)
1.484(3)
2.22
80.27(11)
80.52(11)
160.55(11)
176.89(8)
172.43(8)
87.10(3)
2.2359(14)
2.4347(13)
1.485(4)
RuÀCl [ꢁ]
SÀO [ꢁ]
S=O···HÀCpyr [ꢁ]
NamineÀRuÀNax [8]
2.57
82.90(11)
80.55(10)
163.45(10)
98.48(8)
92.62(7)
88.87(3)
82.02(16)
78.81(16)
160.77(17)
177.07(12)
170.08(12)
85.96(5)
NaxÀRuÀNax [8]
NamineÀRuÀS [8]
NeqÀRuÀCl [8]
SÀRuÀCl [8]
176.13(7)
85.75(3)
exhibits
a
RuÀN distance of
2.031(3) ꢁ, which is the shortest
RuÀN distance observed in all
five structures. The two pyridine
rings of trans-2sym are in the axial
aringax–ringax [8]
aringeq–(RuÀClÀNamineÀS) [8]
24.02(8)
18.63(8)[b]
25.7(1)
2.81(9)
26.2(2)
19.0(1)[b]
1.80(5)
[a] Ref. [3b]. [b] Bond or angle involves a triazole ring.
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
784
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. Cyclic voltammogram of trans-1 in CH2Cl2 at 100 mVsÀ1 with 0.1 m
Bu4NPF6.
Figure 4. ORTEP view of trans-3un Ellipsoids are drawn at the 50% probabili-
ty level. H atoms, solvent molecules, and anions are omitted for clarity.
trans-2sym. In the literature, the appearance of additional re-re-
duction peaks with similar anodic shifts (ca. 0.5 V) has been re-
ported for other Ru–dmso complexes. In these cases, the new
re-reduction peaks were attributed to isomers with O-bound
dmso ligands instead of S-bound ones. The formation of such
isomers through an EC mechanism upon oxidation of the Ru
center has been observed before.[13] Kojima et al. proposed
that a key factor for the stability of the RuÀS bond is p back
donation from a dp Ru orbital to the S=O moiety of dmso. Ac-
cording to this explanation, the decreased electron density at
the RuIII center upon oxidation would lead to a labilization of
the RuÀS bond and hence result in a dissociation or coordina-
tive rearrangement of the dmso ligand.[14]
Electrochemistry
Cyclic voltammetry of all six complexes was measured in
CH2Cl2 and acetonitrile. All complexes show a quasireversible
or reversible one-electron oxidation from RuII to RuIII. The oxi-
dation potentials versus the ferrocene–ferrocenium couple are
listed in Table 3. In CH2Cl2, the potentials vary in the range
0.58–0.71 V, and there is a trend towards slightly more catho-
dic potentials with an increasing number of triazole arms in
the tripodal ligand (0.71 V for trans-1; 0.58 and 0.60 V for trans-
3 and trans-4, respectively). In acetonitrile, the variation of the
oxidation potential is even smaller (0.58–0.62 V).
We simulated the CVs of trans-1 in acetonitrile with an EC
mechanism at different scan rates by using the program Dig-
iElch7. The scan-rate-dependent behavior of the compound
could be best reproduced by assuming a dissociation–associa-
tion mechanism, in which the S-bound dmso ligand of trans-
1+ (one-electron-oxidized trans-1) dissociates and then re-co-
ordinates through the oxygen atom. Working with a mecha-
nism, in which dmso dissociates and an excess component
(acetonitrile solvent molecules) coordinates instead, the simu-
lations could not reproduce the scan-rate-dependent behavior
of trans-1. This seems to contrast with the findings of NMR
studies of temperature-driven and UV-light-driven reactions of
complexes 1–4 (see below), in which the dissociated dmso
ligand was replaced by a solvent molecule.
Although the oxidation waves in CH2Cl2 look reversible at
scan rates of 100 mVsÀ1 or higher, the complexes trans-1 and
trans-2sym show additional re-reduction peaks with cathodic
peak potentials, Epc, of À0.06 and 0.05 V, respectively (Figure 5
and, in the Supporting Information, Figure S21). In the second
cycle, there are new oxidation waves corresponding to these
additional re-reduction processes (Figure 5). Because the whole
effect is observed both in CH2Cl2 and acetonitrile, the effect is
not due to coordinating solvent molecules. It is more pro-
nounced at low scan rates, at which the intensity of the first
re-reduction wave is very much diminished for trans-1 and
However, the re-coordination of dmso can be rationalized by
the fact that the current in the cyclic voltammograms is due to
species and processes at the electrode surface, where there is
a highly increased concentration of complex molecules, a con-
dition that in return would result in an elevated concentration
of dissociated dmso molecules. Further away from the elec-
trode, a replacement of dmso by acetonitrile might occur, but
this would not be reflected in the CV current. The simulations
afforded an equilibrium constant of Keq =8 for the combined
dissociation–recoordination process of dmso, data that is in
good agreement with results obtained for a RuII–dmso poly-
pyridyl complex by Benet-Buchholz et al.[13] Figure 6 shows
a comparison of the simulated and the experimental CV of
Table 3. Oxidation potentials (E1/2) and reduction peak potentials (Ep)
measured in CH2Cl2 and acetonitrile at 100 mVsÀ1 with 0.1m Bu4NPF6 at
room temperature.
E1/2(Ox)
in CH2Cl2
[V]
E1/2(Ox)
in MeCN[a]
[V]
Ipc/Ipa
in MeCN[b]
Ep(Red)
in MeCN[a]
[V]
[a]
cis-1
0.65
0.71
0.65
0.68
0.62
0.60
0.58
0.62
0.61
0.62
0.60
0.59
0.92
0.45
0.63
0.77
0.99
0.99
À2.41
À2.31
À2.36
À2.35
À2.40
À2.57
trans-1
trans-2sym
trans-2un
trans-3un
trans-4
[a] vs. Fc/Fc+. [b] For the oxidation at 10 mVsÀ1
.
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
785
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
they show that the oxidized species of trans-1, trans-1+, is
much less stable than the isomer cis-1+. The stability of the
one-electron-oxidized species, and hence the reversibility of
the oxidation process, increases with the number of triazole
arms at the tripodal ligand (Ipc/Ipa =0.99 for trans-3 and trans-
4). Remarkably, the oxidation of trans-2un is more reversible
than that of trans-2sym, which means that the position of the
triazole arm has an effect on the redox stability of the com-
plex. According to the abovementioned explanation of the ob-
served EC phenomenon, the increased redox reversibility of
the triazole-rich compounds points to a higher stability of the
RuÀS bond in the oxidized state. This points to a higher elec-
tron density at the Ru center for complexes with more triazole
arms. Hence, the substitution of triazolyl for pyridyl moieties
has little effect on the oxidation potential but great effect on
the redox stability of the complex.
Figure 6. Comparison of a CV of trans-1 measured in acetonitrile at
250 mVsÀ1 and a simulation.
In acetonitrile, all six complexes show an irreversible reduc-
tion close to the edge of the solvent window (see the Support-
ing Information, Figure S24). DPV measurements show that
this reduction involves the same number of electrons as the
oxidation (see the Supporting Information, Figure S25). There-
fore, we attribute these redox waves to a one-electron reduc-
tion at the tripodal ligand. In agreement with this assignment,
the reduction potentials shift to more negative values with an
increasing number of triazole arms, a behavior that is due to
the stronger donating and weaker accepting nature of the tria-
zoles compared to the pyridines. As a result of this shift, the
potential gap between oxidation and reduction, which to
a first approximation often correlates to the HOMO–LUMO gap
of a complex, increases from trans-1 to trans-4.
trans-1 in acetonitrile at 250 mVsÀ1. Details of the simulation
parameters and voltammograms simulated at other scan rates
are presented in the Supporting Information (Figure S26).
The cyclic voltammograms of all six complexes, measured in
acetonitrile at 10 mVsÀ1, were used to determine the corre-
sponding peak-current ratios, Ipc/Ipa, which can be taken as
a measure for the reversibility of the oxidation process
(Figure 7 and, in the Supporting Information, Figure S22). The
ratio is 1 for ideal systems and close to 0 for highly irreversible
systems. The values of this ratio are also listed in Table 3, and
EPR and UV/Vis spectroelectrochemistry (SEC)
The measurement of EPR spectra of the oxidized species of 1–
4 was hampered by the redox dynamic behavior discussed
above. Only the two most stable complexes, trans-3un and
trans-4, showed good EPR spectra after electrolysis in CH2Cl2 at
room temperature. For example, Figure 8 shows the EPR spec-
trum of oxidized trans-3un. Apart from an isotropic signal at a g
value close to that of the free electron, a signal that is proba-
bly due to residual dmso impurities in the sample, the spec-
Figure 7. Cyclic voltammograms of trans-1 (top) and trans-4 (bottom) mea-
sured in acetonitrile at 10 mVsÀ1 with 0.1m Bu4NPF6 at room temperature.
Figure 8. EPR spectrum of one-electron-oxidized trans-3un recorded in CH2Cl2
at 110 K. The asterisk indicates the signal of an organic impurity.
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
786
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
of the pyridyl moieties. The intensity of this band decreases
with the number of pyridine arms, and it completely disap-
pears for trans-4.
Table 4. g values of the one-electron-oxidized complexes 1–4 measured
in CH2Cl2 at 110 K with 0.1m Bu4NPF6.
On the basis of assignments for similar complexes,[3,14] the
absorption bands between 300 and 450 nm can be attributed
to metal-to-ligand charge transfer (MLCT) from Ru dp orbitals
to pp* orbitals of the pyridine and triazole arms. From trans-
1 to trans-4, the MLCT bands shift to shorter wavelengths,
thus correlating to the increasing potential gap between oxi-
dation and reduction in the CV (see above). In agreement with
their assignment to MLCT transitions, the absorption bands be-
tween 300 and 450 nm decrease in intensity or disappear
upon oxidation (for example, see Figure 10). The oxidation-
stable complexes, cis-1, trans-3un, and trans-4, show isosbestic
points in the SEC plots (see the Supporting Information, Fig-
ure S29). The complexes with more labile dmso ligands (trans-
1, trans-2sym, trans-2un) only initially show isosbestic points that
are blurred in the course of the electrolysis by additional spec-
troscopic shifts owing to the dmso rearrangement (see the
Supporting Information, Figure S29). The observation that the
redox stability of the complexes increases with the number of
triazole arms is further corroborated by the UV/Vis spectra re-
corded before and after the SEC cycle for every complex (see
the Supporting Information, Figure S30). With more triazole
arms, the overlap of the two spectra becomes almost ideal,
which points to a reversible oxidation process.
g1
g2
g3
Dg
g*
cis-1
2.517
–
2.462
2.489
2.472
2.454
2.269
–
2.203
2.182
2.182
2.178
1.800
–
1.814
1.813
1.821
1.821
0.717
–
0.648
0.676
0.651
0.633
2.068
2.062
2.064
2.060
–
trans-1
trans-2sym
trans-2un
trans-3un
trans-4
–
trum shows an anisotropic signal with three g values, g1 =
2.472, g2 =2.182, and g3 =1.821. This spectrum is typical for
a metal-centered spin. For the complexes cis-1, trans-1, trans-
2
sym, and trans-2un, EPR spectra could only be measured when
the electrolysis was carried out at À408C, so as to slow down
the chemical reaction following the oxidation. In this way, cis-
1, trans-2sym, and trans-2un showed anisotropic spectra with g
values similar to those of trans-3un and trans-4 (Table 4). The
less redox stable complexes show an additional signal with a g
value (g*) in the range 2.06–2.07, a signal that we tentatively
attribute to the product of the dmso rearrangement upon oxi-
dation. The least oxidation stable complex, trans-1, exclusively
shows an EPR signal with g*=2.062 (see the Supporting Infor-
mation, Figures S27 and S28).
The UV/Vis spectra of 1–4 were recorded in CH2Cl2 (Figure 9)
and acetonitrile (see the Supporting Information, Figure S31),
showing identical bands in both solvents (see the Supporting
Information, Table S1). The main features of the spectra are
a band around 250 nm and bands between 300 and 450 nm.
The band around 250 nm can be assigned to p–p* transitions
Figure 10. Change in the UV/Vis absorption of trans-3un in the course of
a one-electron oxidation in CH2Cl2.
Thermo- and photoreactivity of 1–4
Prompted by the results of the CV measurements and previous
reports about the thermo- and photo-induced reactivity of cis-
1, trans-1, and related complexes,[2b,3,15] we investigated the re-
activity of all six complexes under heating or irradiation with
UV light. The experiments were carried out in CD3CN and
1
[D6]DMSO and monitored by H NMR spectroscopy. When com-
plexes cis-1 and trans-1 are heated in CD3CN at 808C for seven
days, they both form identical mixtures of several new species,
and the resonance of free DMSO appears at 2.47 ppm (see the
Supporting Information, Figure S33). Yamaguchi et al. were
able to show that trans-1 converts to cis-1 upon heating in
Figure 9. UV/Vis absorption spectra of the complexes 1–4 measured in
CH2Cl2.
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
787
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[D6]DMSO.[3] When cis-1 and trans-1 are irradiated at 365 nm (a
wavelength in the range of their MLCT bands) in CD3CN, the
NMR spectra of both show a clean conversion to product 1a,
and the signal at 2.47 ppm points to a loss of the DMSO
ligand (see the Supporting Information, Figure S34). In another
work, Yamaguchi et al. were able to show by mass spectrome-
try that, in 1a, DMSO is substituted by acetonitrile. Hence, the
MLCT transition leads to a decrease in the electron density at
the metal center, a decrease that in turn weakens the RuÀS
back bonding and leads to the dissociation and substitution of
the dmso ligand.
For trans-1, the conversion to 1a takes only 5 min, whereas
for cis-1, which has a much more stable RuÀSdmso bond (see
CV), irradiation for 100 min is needed to achieve a conversion
of greater than 90% (see the Supporting Information, Fig-
ure S34). For cis-1, the corresponding irradiation experiment in
1
[D6]DMSO leads to the disappearance of the H NMR signal of
the dmso ligand (2.8 ppm) and the appearance of the free-
DMSO resonance at 2.5 ppm (see the Supporting Information,
Figures S35 and S48). All other resonances remain almost un-
changed; only after prolonged irradiation, signals of trans-
1 appear. For trans-1, irradiation in [D6]DMSO leads to an
1H NMR spectrum identical to that of cis-1, except for the
signal of free DMSO and some remaining weak signals of
trans-1 (see the Supporting Information, Figures S35 and S48).
A closer look at the spectra shows that the prolonged irradia-
tion of either cis-1 or trans-1 results in a cis-1/trans-1 ratio of
approximately 95:5. Therefore, in the case of cis-1, irradiation
primarily leads to a simple substitution of the dmso ligand by
[D6]dmso, whereas in the case of trans-1, irradiation over-
whelmingly leads to a substitution and to the isomerization to
cis-1. Therefore, we assume that, analogously, the acetonitrile
ligand in 1a is in a position cis to the amine nitrogen atom
(that is, cis-1a; see Scheme 3).
We analyzed the kinetics of the light-driven DMSO exchange
in [D6]DMSO more thoroughly. The complexes cis-1, trans-1,
trans-2sym
,
and trans-2un were dissolved in [D6]DMSO
1
(3 mmolLÀ1) and irradiated at 365 nm. H NMR spectra were re-
corded after 10, 70, 190, 340, 520, and 760 min, and the con-
centrations of the original species with non-deuterated dmso
ligands were determined (see the Supporting Information, Fig-
ures S44–S47 and S49, Table S3). The DMSO exchange of all
four complexes follows pseudo-first-order kinetics (Figure 11
and, in the Supporting Information, Figures S49–S54), and the
rate constants, k, for cis-1, trans-1, trans-2sym, and trans-2un are
1.36ꢂ10À4, 2.07ꢂ10À4, 0.35ꢂ10À4, and 0.62ꢂ10À4 sÀ1, respec-
tively (see the Supporting Information, Table S3). These data
show that trans-1 is more reactive toward light-driven DMSO
exchange than cis-1, behavior that is in agreement with CV
measurements and thermal reactivity experiments (see above).
Furthermore, the two isomers of trans-2, which have one tria-
zole arm, are less reactive, that is, have more tightly bound
dmso ligands, than complex trans-1, behavior that also agrees
with CV and thermal-reactivity results. However, the fact that
trans-2un is more reactive in the light-driven DMSO exchange
than the isomer trans-2sym is in contrast to the CV results (see
above) and to the thermal reactivity in CD3CN (see below).
1
Scheme 3. Reaction Scheme proposed to explain the changes in the H NMR
spectra of 1–4 in CD3CN observed upon heating and irradiation with UV
light.
Figure 11. Concentrations of complexes with non-deuterated dmso ligands
after different periods of irradiation at 365 nm in [D6]-DMSO.
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
788
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
Figure 13. Changes in the H NMR spectra of trans-2sym and trans-2un in
CD3CN upon irradiation with UV light.
1
Figure 12. Changes in the H NMR spectrum of trans-2sym in CD3CN upon
heating and subsequent irradiation with UV light.
(see the Supporting Information, Figure S37). Corresponding to
this intensity loss, signals of a new symmetrical species
emerged. This means that 66% of the molecules that lost their
non-deuterated ligand (100%) substituted it for a deuterated
dmso ligand trans to the Namine atom, hence keeping their con-
formation. 33% of the molecules substituted non-deuterated
for deuterated DMSO and also changed their conformation,
thus forming a new species with two identical pyridine arms
(see aromatic region in Figure S37 in the Supporting Informa-
tion). We assume that this species is the isomer with the dmso
For trans-2sym, heating in CD3CN at 808C for four days leads
to the formation of one major new species. The corresponding
1H NMR spectrum (Figure 12 and, in the Supporting Informa-
tion, Figure S36) shows the signal of free DMSO at 2.47 ppm
and only one set of pyridine signals (two doublets, two trip-
lets) in the aromatic region. Therefore, the new species must
also be symmetrical with two identical pyridine arms and, pre-
sumably, an acetonitrile ligand. The unsymmetrical isomer,
trans-2un, which showed a slightly stronger binding of the
dmso ligand than trans-2sym in the CV, does not react under
the same heating conditions (Scheme 3). Remarkably, the posi-
tion of the triazole arm in trans-2 has a pronounced effect on
the thermal stability of the complex.
ligand in the position cis to the Namine atom, that is, cis-2sym
.
When trans-2sym was irradiated in [D6]DMSO for 760 min, the
1H NMR signal of coordinated dmso (3.60 ppm) lost 80% of its
original intensity, whereas the other signals lost 44% of their
intensity (see the Supporting Information, Figure S41). This
result means that of the molecules that substituted the dmso
ligand for [D6]dmso (80%), more than half changed their con-
formation (44%), thus forming cis-2sym. Hence, in the case of
trans-2sym, heating favors the substitution of the dmso ligand
with a retention of the original complex conformation, where-
as irradiation favors substitution with concomitant isomeriza-
When trans-2sym and trans-2un are irradiated at 365 nm in
CD3CN for 160 min, both complexes lose their dmso ligand
1
(signal of free DMSO in the H NMR spectrum) and each form
one major new species (see the Supporting Information, Fig-
ures S39 and S40). Mass spectra show that for both isomers of
trans-2, the dmso ligand is substituted by a CD3CN molecule
(see the Supporting Information, Figures S42 and S43). In both
cases, the major species has the same orientation of the tripod
arms as the dmso precursor: the species formed from trans-
2sym upon irradiation shows one set of pyridine signals in the
1H NMR spectrum and is, therefore, also symmetrical; the spe-
cies formed from trans-2un shows two sets of pyridine signals
and is, therefore, also unsymmetrical (Figure 13). The product
formed from trans-2sym upon irradiation is different from the
product formed upon heating (Figure 12 and, in the Support-
ing Information, Figures S36 and S39). When the heating prod-
uct is further irradiated for 2 h, it converts to the irradiation
product (Figure 12), but heating the irradiation product has no
effect.
tion to cis-2sym
.
We therefore assume that the product formed from trans-
2sym in CD3CN upon heating is the acetonitrile complex, trans-
2asym (Scheme 3). Because irradiation with UV light seems to
favor the isomerization of trans-2sym and trans-1 (see above) in
[D6]DMSO, we assume that the complexes formed in CD3CN
upon irradiation are cis-2asym and cis-2aun. Scheme 3 summa-
1
rizes our explanation of the observed reactivity. The H NMR
spectra also show that upon irradiation, some unsymmetrical
cis-2aun is formed from symmetrical trans-2sym, and some sym-
metrical cis-2asym is formed from unsymmetrical trans-2un (see
dashed lines in Figure 13). The irradiation of trans-2sym and
trans-2un in CD3CN leads to the following: (a) the substitution
of the dmso ligand for acetonitrile; (b) the isomerization of the
complexes with respect to the position of the chlorido ligand;
(c) a slight scrambling of the complex conformation with re-
spect to the individual positions of the pyridine and triazole
arms. Finally, complexes trans-3un and trans-4 do not show
To further investigate the difference between the heating
and irradiation product of trans-2sym, the experiments were re-
peated in [D6]DMSO. When a [D6]DMSO solution of trans-2sym
1
was heated at 808C for seven days, the H NMR signal of the
coordinated DMSO completely disappeared, but the other sig-
nals of the complex only lost a third of their original intensity
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
789
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
not have a pronounced effect on the oxidation potential of
the RuII center. However, the substitution greatly affects the
stability of the RuÀSdmso bond. This stability largely depends on
p back donation from the Ru center to the S=O moiety, and it
is, therefore, intimately linked to the electron density at the
metal center.
When the electron density at the metal center is lowered by
either electrochemical oxidation to RuIII or through an MLCT
transition in the range 300–450 nm, the dmso ligand becomes
labile and can dissociate. This results in a rearrangement of the
dmso ligand from the S-bound to the O-bound form in CV ex-
periments, behavior that was observed and simulated for
trans-1. Furthermore, it results in ligand substitution upon irra-
diation at 365 nm in CD3CN, a change that could be monitored
1
by H NMR. The electron density at the Ru center and, hence,
the stability of the RuÀS bond increases with the number of
triazole arms from complex trans-1 to trans-4. This increase is
evidenced by the increasing reversibility of the oxidation pro-
cess in the CV and an increasingly inert behavior under irradia-
tion or heating in coordinating solvents. Interestingly, the sta-
bility of the RuÀS bond is not only sensitive to the number of
triazole arms but also to their position. Complex trans-2un,
which has one triazole arm in an axial position, shows a much
more reversible oxidation wave in the CV than its isomer,
trans-2sym (triazole in the equatorial position). Whereas trans-
2un does not react upon heating in CD3CN, trans-2sym forms the
Scheme 4. Proposed mechanisms for the reaction of trans-2sym in CD3CN
under heating and irradiation with UV light.
clean reactions to new species neither upon heating nor upon
irradiation with UV light for an extended amount of time. They
acetonitrile complex trans-2asym
.
1
only show a slight decrease in intensity of their H NMR signals
We have thus shown that by the stepwise substitution of tri-
azolylmethyl arms for pyridylmethyl arms, the redox properties
and reactivities of metal complexes with tripodal N4 ligands
can be finely tuned. Even though the complex trans-1 has
been known in the literature for a while,[3,14] this is the first
time that redox-induced ligand exchange in that complex has
been thoroughly investigated and quantified through simula-
tions of the cyclic voltammograms. In combination with the
advantages offered by triazole click chemistry (preparational
ease, chemical stability, substituent versatility), this ligand
family is a promising platform for catalyst design and tuning,
as well as for photocatalysis and redox catalysis. Further re-
search to thoroughly characterize the ligand-exchange prod-
ucts, to broaden the ligand exchange beyond solvent mole-
cules, and to apply the complexes as oxidation catalysts are
currently under way in our laboratories.
(for example, 10% intensity loss for trans-4 after 10 h of irradia-
tion at 365 nm).
The formation of different products from trans-2sym upon
heating and irradiation in CD3CN can be rationalized by assum-
ing different transition states for the two reactions (Scheme 4).
We hypothesize that thermal activation should lead to a weak-
ening of the RuÀS bond and the formation of a 7-coordinate
transition state, which should favor the retention of the com-
plex conformation. An MLCT transition, like the oxidation to
RuIII, should lead to a breaking of the RuÀS bond and to the
formation of a 5-coordinate transition state. Such a transition
state would favor an attack of the acetonitrile molecule from
the less crowded side opposite to the equatorial arm, thus re-
sulting in a change in conformation with respect to the chlo-
rido ligand. It could also give rise to the observed scrambling
of the tripod arms.
Experimental Section
Conclusion
General procedures
Summarizing, we have presented an efficient synthetic route
for the generation of mixed-donor tripodal ligands containing
both pyridyl and triazolyl arms. With these and other ligands,
we have synthesized four new complexes of the series [Ru(Ln)-
(dmso)Cl](PF6) (1–4), in which the three pyridylmethyl arms of
L1 (TPA) are successively substituted with triazolylmethyl arms
(L4 =TBTA, three triazole arms). For L2, two isomers, with re-
spect to the position of the single triazole arm (equatorial or
axial, trans-2sym or trans-2un), were separated and structurally
characterized. The substitution of triazole for pyridine arms did
All chemicals were commercially available and used as purchased
unless otherwise noted. Solvents used for synthesis were dried
with appropriate drying agents. All manipulations were carried out
under a nitrogen atmosphere unless otherwise noted. The precur-
sor [Ru(dmso)4Cl2],[16] complex trans-1,[3] and the ligands TPA[9] and
TBTA,[10] were synthesized according to literature procedures. The
ligands L2 [N-(1-benzyl-1H-1,2,3-triazol-4-yl)methyl-N,N-di(2-pyridyl-
methyl)amine] and L3 [N,N-di(1-benzyl-1H-1,2,3-triazol-4-yl)methyl-
N-(2-pyridyl)methylamine] were prepared according to modified lit-
erature procedures, which are described in the Supporting Infor-
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
790
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
mation. 1H NMR data were recorded with a Jeol Lambda 400
(400 MHz) by using the chemical shift of the solvent as an internal
standard. X-band EPR spectra were recorded at 110 K with an EMX
Bruker system connected to an ER 4131 VT (variable-temperature)
accessory. The EPR samples were electrolyzed with a platinum-wire
working electrode and a platinum-wire counter electrode. Elemen-
tal analyses (CHN) were measured with an Elementar Vario EL III
and a PerkinElmer Analyzer 240. Mass spectra were recorded with
an Agilent 6210 ESI-TOF spectrometer (Agilent Technologies, Santa
Clara, CA, USA).
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_requests/cif.
General procedure for the preparation of the complexes
[Ru(dmso)4Cl2] (1 equiv), the tripodal ligand, Ln (1 equiv), and KPF6
(1.5 equiv) were heated at reflux in methanol (20 mL) overnight
(ca. 12 h). The work-up was carried out under ambient conditions
with commercially available non-dried solvents. After cooling to
room temperature, the reaction mixture was concentrated (to ca.
5 mL) by evaporation of the solvent, and Et2O (ca. 30 mL) was
added to precipitate the crude product. The crude product was
dried in a vacuum and purified by column chromatography on alu-
minum oxide (5 wt.-% water) with CH2Cl2/methanol (99:1 v/v) as
an eluent.
cis-1: L1 (TPA, 90 mg, 0.31 mmol), [Ru(dmso)4Cl2] (145 mg,
0.30 mmol), and KPF6 (83 mg, 0.45 mmol) were reacted in metha-
nol (20 mL). After work-up and purification by column chromatog-
raphy, the complex was precipitated from the concentrated eluent
by addition of diethyl ether, and it was obtained as a yellow
powder (130 mg, 0.195 mmol, 65% yield). 1H NMR (CD3CN,
400 MHz): d=2.82 (s, 6H; dmso), 4.46 (s, 2H; CH2(eq)), 4.64 (ABq,
Electrochemistry
Cyclic voltammograms were recorded with a PAR VersaStat 4 po-
tentiostat (Ametek) by working in anhydrous dichloromethane
(H2Oꢀ0.005%, puriss., Sigma Aldrich) or acetonitrile (H2Oꢀ0.01%,
puriss., Sigma Aldrich) distilled from calcium hydride or calcium
chloride, respectively. A three-electrode setup was used with
a glassy-carbon working electrode, a coiled platinum wire as the
counter electrode, and a coiled silver wire as the pseudo reference.
Ferrocene or decamethylferrocene were used as an internal stan-
dard, and 0.1m NBu4PF6 (Fluka, ꢁ99.0%, electrochemical grade)
was used as an electrolyte. Simulations of the CV data were carried
out with the software, DigiElch Professional (Version 7.FD), and de-
tails of the simulation are given in the Supporting Information.
J
AB =16 Hz, 2H; CH2(ax)), 5.37 (ABq, JAB =16 Hz, 2H; CH2(ax)), 7.09
(d, J=8 Hz, 1H; py(eq)), 7.27 (t, J=8 Hz, 2H; py(ax)), 7.31 (t, J=
7 Hz, 1H; py(eq)), 7.40 (d, J=8 Hz, 2H; py(ax)), 7.64 (dt, J1 =8 Hz,
J2 =1.6 Hz, 1H; py(eq)), 7.74 (dt, J1 =8 Hz, J2 =1.6 Hz, 2H; py(ax)),
8.73 (d, J=8 Hz, 2H; py(ax)), 9.67 ppm (d, J=8 Hz, 1H, py(eq)); UV/
Vis (CH2Cl2): l (e/104 mÀ1 cmÀ1)=250 (1.62), 316 (0.70), 366 (1.06),
425 nm (sh; 0.08); HRMS (ESI): m/z calcd for C18H18N6RuCl,
[Ru(L1)Cl(N2)]+: 455.0325 and C20H24N4RuClSO, [Ru(L1)Cl(dmso)]+,
505.0403; found: 455.0341 and 505.0417; elemental analysis calcd
(%) for C20H24N4RuSOClPF6·0.25Et2O: C 37.73, H 4.00, N 8.38; found:
C 37.85, H 3.99, N 8.31.
trans-2: L2 (195 mg, 0.5 mmol), [Ru(dmso)4Cl2] (242 mg, 0.5 mmol),
and KPF6 (137 mg, 0.75 mmol) were reacted in methanol (20 mL).
After work-up and purification by column chromatography, the
complex was precipitated from the concentrated eluent by addi-
tion of diethyl ether, and it was obtained as a yellow powder. At
this stage, the compound still contained both isomers, trans-2sym
and trans-2un. It was submitted to column chromatography once
more, and, this time, the eluent gradient was carefully raised from
pure CH2Cl2 to CH2Cl2/methanol 98:2 v/v. The first of two fractions
afforded pure trans-2un as a yellow powder after concentration and
precipitation by addition of diethyl ether (130 mg, 0.159 mmol,
32% yield). The second fraction contained trans-2sym (72 mg,
0.088 mmol, 18% yield). Single crystals of trans-2sym·acetone suit-
able for X-ray diffraction analysis were obtained by condensing di-
ethyl ether onto a concentrated solution of the complex in ace-
tone. Single crystals of trans-2un suitable for X-ray diffraction analy-
sis were obtained by condensing diethyl ether onto a concentrated
UV/Vis spectroscopy and spectroelectrochemistry
UV/Vis spectra were recorded with an Avantes spectrometer con-
sisting of a light source (AvaLight-DH-S-Bal), a UV/Vis detector
(AvaSpec-ULS2048), and a NIR detector (AvaSpec-NIR256-TEC). UV/
Vis spectroelectrochemistry measurements were carried out in an
optically transparent thin-layer electrochemical (OTTLE) cell[17] with
a platinum-mesh working electrode, an platinum-mesh counter
electrode, and a silver-foil pseudo reference.
Reactivity experiments
The reactivity experiments of the complexes were carried out in
deuterated solvents in NMR tubes sealed with parafilm with com-
plex concentrations of 3 mmolLÀ1. The NMR tubes were either
heated at 808C in an oil bath or irradiated with a 4 W UV lamp at
365 nm.
X-ray crystallography
Single-crystal X-ray diffraction data were collected with a Stoe X-
Area or a Bruker Smart AXS diffractometer. Data were collected at
100(2) K by using graphite-monochromated Mo Ka radiation (l=
0.71073 ꢁ). The strategy for the data collection was evaluated by
using the CrysAlisPro CCD software. The data were collected by
the standard y–w scan techniques and were scaled and reduced
by using CrysAlisPro RED software. The structures were solved by
direct methods using SHELXS-97 and refined by full-matrix least-
squares with SHELXL-97, refining on F2.[18]
solution of the complex in ethyl acetate. Data for trans-2sym
1H NMR (CD3CN, 400 MHz): d=3.60 (s, 6H; dmso), 4.37 (s, 2H;
CH2(eq)), 4.77 (ABq, AB =15 Hz, 2H; CH2(ax)), 5.40 (s, 2H;
:
J
CH2(benzyl,eq)), 5.69 (ABq, JAB =15 Hz, 2H; CH2(ax)), 6.96–7.01 (m,
2H, aryl), 7.15 (t, J=8 Hz, 2H; py(ax)), 7.25–7.31 (m, 3H; aryl), 7.36
(d, J=8 Hz, 2H; py(ax)), 7.45 (s, 1H; triazole), 7.66 (dt, J1 =8 Hz,
J2 =1.6 Hz, 2H, py(ax)], 8.70 ppm (d, J=8 Hz, 2H; py(ax)); UV/Vis
(CH2Cl2): l (e/104 mÀ1 cmÀ1)=255 (1.02), 318 (0.65), 389 (sh; 0.64),
412 nm (0.73); HRMS (ESI): m/z calcd for C22H22N8RuCl
[Ru(L2)Cl(N2)]+ 535.0699; found 535.0699; elemental analysis calcd
(%) for C24H28N6RuSOClPF6·CH2Cl2 (mixture of both isomers): C
36.84, H 3.71, N 10.31; found: C 37.25, H 3.94, N 10.33. Data for
The positions of all the atoms were obtained by direct methods.
All non-hydrogen atoms were refined anisotropically. The remain-
ing hydrogen atoms were placed in geometrically constrained po-
sitions and refined with isotropic temperature factors, generally
1.2 times the Ueq values of their parent atoms. Table S2 contains
the parameters for the data collection and refinement.
CCDC-918970 (for trans-2sym·acetone), 918971 (for trans-3un), and
949830 (for trans-2un) contain the crystallographic information for
this paper. All these data can be obtained free of charge from the
1
trans-2un: H NMR (CD3CN, 400 MHz): d=3.35 (s, 3H; dmso), 3.49 (s,
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
791
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
e) K. S. Min, A. G. DiPasquale, A. L. Rheingold, H. S. White, J. S. Miller, J.
Am. Chem. Soc. 2009, 131, 6229; f) J. Tao, H. Maruyama, O. Sato, J. Am.
Chem. Soc. 2006, 128, 1790; g) K. S. Min, A. G. DiPasquale, J. A. Golen,
A. L. Rheingold, J. S. Miller, J. Am. Chem. Soc. 2007, 129, 2360; h) S. Miya-
zaki, K. Ohkubo, T. Kojima, S. Fukuzumi, Angew. Chem. 2007, 119, 923;
Angew. Chem. Int. Ed. 2007, 46, 905; i) Y. Hirai, T. Kojima, Y. Mizutani, Y.
Shiota, K. Yoshizawa, S. Fukuzumi, Angew. Chem. 2008, 120, 5856;
Angew. Chem. Int. Ed. 2008, 47, 5772; j) T. Kojima, T. Morimoto, T. Saka-
moto, S. Miyazaki, S. Fukuzumi, Chem. Eur. J. 2008, 14, 8904; k) F. Weiss-
er, R. Huebner, D. Schweinfurth, B. Sarkar, Chem. Eur. J. 2011, 17, 5727;
l) K. Chen, L. Que, Jr., Am. Chem. Soc. 2001, 123, 6327; m) K. Chen, M.
Costas, J. Kim, A. K. Tipton, L. Que, Jr., J. Am. Chem. Soc. 2002, 124,
3026; n) J.-H. Ye, L. Duan, C. Yan, W. Zhang, W. He, Tetrahedron Lett.
2012, 53, 593.
3H; dmso), 4.58 (ABq, JAB =14 Hz, 1H; CH2), 4.59 (s, 2H; CH2), 4.79
(ABq, JAB =15 Hz, 1H; CH2), 5.10 (ABq, JAB =14 Hz, 1H; CH2), 5.41
(ABq, JAB =15 Hz, 1H; CH2), 5.47 (ABq, JAB =15 Hz, 1H; CH2), 5.72
(ABq, JAB =15 Hz, 1H; CH2), 6.94 (d, J=8 Hz, 1H; py), 6.79–7.01 (m,
2H; aryl), 7.10 (t, J=8 Hz, 1H; py), 7.21–7.33 (m, 4H; aryl and py),
7.44 (d, J=8 Hz, 1H; py), 7.48 (dt, J1 =8 Hz, J2 =1.6 Hz, 1H; py),
7.71 (dt, J1 =8 Hz, J2 =1.6 Hz, 1H; py), 7.80 (s, 1H; triazole), 8.82 (d,
J=8 Hz, 1H; py), 9.70 ppm (d, J=8 Hz, 1H; py); UV/Vis (CH2Cl2):
l (e/104 mÀ1 cmÀ1)=252 (1.01), 315 (sh; 0.53), 350 (0.74), 380 nm
(0.80).
trans-3: L3 (225 mg, 0.5 mmol), [Ru(dmso)4Cl2] (242 mg, 0.5 mmol),
and KPF6 (137 mg, 0.75 mmol) were reacted in methanol (20 mL).
After work-up and purification by column chromatography, the
complex was precipitated from the concentrated eluent by addi-
tion of diethyl ether, and it was obtained as a yellow powder
(260 mg, 0.293 mmol, 58% yield). Single crystals of trans-3un suit-
able for X-ray diffraction analysis were obtained by condensing di-
ethyl ether onto a concentrated solution of the complex in ethyl
[2] a) T. Kojima, Y. Hirai, T. Ishizuka, Y. Shiota, K. Yoshizawa, K. Ikemura, T.
Ogura, S. Fukuzumi, Angew. Chem. 2010, 122, 8627; b) M. Yamaguchi, T.
Kumano, D. Masui, T. Yamagishi, Chem. Commun. 2004, 798; c) S. Ohzu,
T. Ishizuka, Y. Hirai, S. Fukuzumi, T. Kojima, Chem. Eur. J. 2013, 19, 1563.
[3] M. Yamaguchi, H. Kousaka, S. Izawa, Y. Ichii, T. Kumano, D. Masui, T. Ya-
magishi, Inorg. Chem. 2006, 45, 8342.
[4] a) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew. Chem.
2002, 114, 2708; Angew. Chem. Int. Ed. 2002, 41, 2596; b) C. W. Tornøe,
C. Christensen, M. Meldal, J. Org. Chem. 2002, 67, 3057.
[5] a) L. Liang, D. Astruc, Coord. Chem. Rev. 2011, 255, 2933; b) H. Struthers,
T. L. Mindt, R. Schibli, Dalton Trans. 2010, 39, 675; c) J. D. Crowley, D.
McMorran in Topics in Heterocyclic Chemistry, Vol. 22 (Ed.: J. Kosmrlj),
Springer, Berlin/Heidelberg, 2012, pp. 31; d) D. Schweinfurth, N. Deibel,
F. Weisser, B. Sarkar, Nachr. Chem. 2011, 59, 937.
[6] For selected examples, see: a) Y. Li, J. C. Huffman, A. H. Flood, Chem.
Commun. 2007, 2692; b) D. G. Brown, N. Sanguantrakun, B. Schulz, U. S.
Schubert, C. P. Berlinguette, J. Am. Chem. Soc. 2012, 134, 12354; c) D.
Schweinfurth, R. Pattacini, S. Strobel, B. Sarkar, Dalton Trans. 2009, 9291;
d) D. Schweinfurth, S. Strobel, B. Sarkar, Inorg. Chim. Acta 2011, 374,
253; e) D. Schweinfurth, N. Bꢃttner, S. Hohloch, N. Deibel, J. Klein, B.
Sarkar, Organometallics 2013, 32, 5834; f) C. B. Anderson, A. B. S. Elliot,
J. E. M. Lewis, C. J. McAdam, K. C. Gordon, J. D. Crowley, Dalton Trans.
2012, 41, 14625; g) M. Ostermeier, M.-A. Berlin, R. M. Meudtner, S. De-
meshko, F. Meyer, C. Limberg, S. Hecht, Chem. Eur. J. 2010, 16, 10202;
h) B. Pinter, D. Urankar, A. Pevec, F. de Proft, J. Kosmrlj, Inorg. Chem.
2013, 52, 4528; i) D. Schweinfurth, C.-Y. Su, S.-C. Wei, P. Braunstein, B.
Sarkar, Dalton Trans. 2012, 41, 12984; j) M. Mydlak, C. Bizzarri, D. Hart-
mann, W. Sarfert, G. Schmid, L. De Cola, Adv. Funct. Mater. 2010, 20,
1812.
acetate. Data for trans-3un:
1H NMR (CD3CN, 400 MHz): d=3.35 (s,
3H; dmso), 3.76 (s, 3H; dmso), 4.22 (ABq, JAB =17 Hz, 1H; CH2),
4.46 (ABq, JAB =18 Hz, 1H; CH2), 4.64 (ABq, JAB =14 Hz, 1H; CH2),
4.75 (ABq, JAB =15 Hz, 1H; CH2), 5.13 (ABq, JAB =14 Hz, 1H; CH2),
5.44 (ABq, JAB =18 Hz, 1H; CH2), 5.46 (s, 2H; CH2(benzyl)), 5.56
(ABq, JAB =14 Hz, 1H; CH2), 6.93–6.99 (m, 2H; aryl), 7.14 (t, J=8 Hz,
1H; py), 7.21–7.38 (m, 9H; aryl and py), 7.46 (s, 1H; triazole), 7.61
(dt, J1 =8 Hz, J2 =1.6 Hz, 1H; py), 7.83 (s, 1H; triazole), 9.29 ppm (d,
J=8 Hz, 1H; py); UV/Vis (CH2Cl2): l (e/104 mÀ1 cmÀ1)=254 (0.66),
313 (0.79), 367 (0.73), 391 nm (0.54); HRMS (ESI): m/z calcd. for
C26H26N10RuCl, [Ru(L3)Cl(N2)]+ 615.1055 and C28H32N8RuClSO,
[Ru(L3)Cl(dmso)]+, 665.1131; found: 615.1074 and 665.1152; ele-
mental analysis calcd. (%) for C28H32N8RuSOClPF6·C2H6SO: C 40.56, H
4.31, N 12.61; found: C 40.56, H 4.46, N 12.80.
trans-4: L4 (TBTA, 205 mg, 0.38 mmol), [Ru(dmso)4Cl2] (184 mg,
0.38 mmol), and KPF6 (105 mg, 0.57 mmol) were reacted in metha-
nol (20 mL). After work-up and purification by column chromatog-
raphy, the complex was precipitated from the concentrated eluent
by addition of diethyl ether, and it was obtained as a yellow
powder (228 mg, 0.150 mmol, 62% yield). 1H NMR (CD3CN,
400 MHz): d=3.44 (s, 6H; dmso), 4.27 (s, 2H; CH2(eq)), 4.61 (ABq,
[7] a) H. A. Michaels, C. S. Murphy, R. J. Clark, M. W. Davidson, L. Zhu, Inorg.
Chem. 2010, 49, 4278; b) S. I. Presolski, V. Hong, S.-H. Cho, M. Finn, J.
Am. Chem. Soc. 2010, 132, 14570; c) J. Rosenthal, S. J. Lippard, J. Am.
Chem. Soc. 2010, 132, 5536; d) Y. Zhou, K. Liu, J.-Y. Li, Y. Fang, T.-C. Zhao,
C. Yao, Org. Lett. 2011, 13, 1290; e) S. Huang, R. J. Clark, L. Zhu, Org.
Lett. 2007, 9, 4999; f) C. Orain, N. LePoul, A. Gomila, J.-M. Kerbaol, N.
Cosquer, O. Reinaud, F. Conan, Y. LeMest, Chem. Eur. J. 2012, 18, 594;
g) T. J. Simmons, J. R. Allen, D. R. Morris, R. J. Clark, C. W. Levenson,
M. W. Davidson, L. Zhu, Inorg. Chem. 2013, 52, 5838; h) N. Sꢄgaud, J.-N.
Rebilly, K. Sꢄnꢄchal-David, R. Guillot, L. Billon, J.-P. Baltaze, J. Farjon, O.
Reinaud, F. Banse, Inorg. Chem. 2013, 52, 691.
J
AB =14 Hz, 2H; CH2(ax)), 5.10 (ABq, JAB =14 Hz, 2H; CH2(ax)), 5.34
(s, 2H; CH2(benzyl,eq)), 5.43 (ABq, JAB =15 Hz, 2H; CH2(benzyl,ax)),
5.47 (ABq, JAB =15 Hz, 2H; CH2(benzyl,ax)), 7.13–7.20 (m, 6H; aryl),
7.30–7.37 (m, 9H; aryl), 7.44 (s, 1H; triazole(eq)), 7.77 ppm (s, 2H;
triazole(ax)); UV/Vis (CH2Cl2): l (e/104 mÀ1 cmÀ1)=255 (0.31), 308
(sh; 0.83), 335 (0.98), 382 nm (sh; 0.13); HRMS (ESI): m/z calcd (%)
for C32H36N12RuClSO, [Ru(L4)Cl(dmso)]+: 745.1526; found: 745.1540;
elemental analysis calcd. (%) for C32H36N8RuSOClPF6·C2H6SO: C
42.17, H 4.37, N 14.46; found: C 42.83, H 4.44, N 14.62.
[8] a) D. Schweinfurth, F. Weisser, D. Bubrin, L. Bogani, B. Sarkar, Inorg.
Chem. 2011, 50, 6114; b) D. Schweinfurth, J. Krzystek, I. Schapiro, S. De-
meshko, J. Klein, J. Telser, A. Ozarowski, C.-Y. Su, F. Meyer, M. Atanasov,
F. Neese, B. Sarkar, Inorg. Chem. 2013, 52, 6880; c) D. Schweinfurth, J.
Klein, S. Hohloch, S. Dechert, S. Demeshko, F. Meyer, B. Sarkar, Dalton
Trans. 2013, 42, 6944; d) D. Schweinfurth, S. Demeshko, M. M. Khus-
niyarov, S. Dechert, V. Gurram, M. R. Buchmeiser, F. Meyer, B. Sarkar,
Inorg. Chem. 2012, 51, 7592.
[9] A. L. Ward, L. Elbaz, J. B. Kerr, J. Arnold, Inorg. Chem. 2012, 51, 4694.
[10] T. R. Chan, R. Hilgraf, K. B. Sharpless, V. V. Fokin, Org. Lett. 2004, 6, 2853.
[11] a) S. Hohloch, C.-Y. Su, B. Sarkar, Eur. J. Inorg. Chem. 2011, 3067; b) S.
Hohloch, D. Scheiffele, B. Sarkar, Eur. J. Inorg. Chem. 2013, 3956; c) S.
Hohloch, B. Sarkar, L. Nauton, F. Cisnetti, A. Gautier, Tetrahedron Lett.
2013, 54, 1808.
Acknowledgement
Felix Fꢃhrer is acknowledged for experimental work in the syn-
thesis of the complexes.
Keywords: click chemistry
substitution · ruthenium · tripodal ligands
· cyclic voltammetry · ligand
[1] For selected examples, see: a) J. Shearer, C. X. Zhang, L. N. Zakharov,
A. L. Rheingold, K. D. Karlin, J. Am. Chem. Soc. 2005, 127, 5469; b) A.
Beni, A. Dei, S. Laschi, M. Rizzitano, L. Sorace, Chem. Eur. J. 2008, 14,
1804; c) Y.-M. Chiou, L. Que, Jr., J. Am. Chem. Soc. 1995, 117, 3999; d) R.-
J. Wei, J. Tao, R.-B. Huang, L.-S. Zheng, Inorg. Chem. 2011, 50, 8553;
[12] a) J. Bjernemose, A. Hazell, C. J. McKenzie, M. F. Mahon, L. P. Nielsen, P. R.
Raithby, O. Simonsen, H. Toftlund, J. A. Wolny, Polyhedron 2003, 22, 875;
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
792
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
b) D. Masui, M. Yamaguchi, T. Yamagishi, Acta Crystallogr. E Struct. Rep.
Online 2003, 59, m308.
[16] I. P. Evans, A. Spencer, G. Wilkinson, J. Chem. Soc. Dalton Trans. 1973,
204.
ˇ
[13] J. Benet-Buchholz, P. Comba, A. Llobet, S. Roeser, P. Vadivelu, S. Wiesner,
Dalton Trans. 2010, 39, 3315.
[17] M. Krejcik, M. Danek, F. Hartl, J. Electroanal. Chem. 1991, 317, 179.
[18] G. M. Sheldrick, SHELX-97, Program for Crystal Structure Refinement,
[14] T. Kojima, T. Amano, Y. Ishii, M. Ohba, Y. Okaue, Y. Matsuda, Inorg. Chem.
1998, 37, 4076.
Gçttingen, Germany, 1997.
[15] a) C. E. Welby, C. R. Rice, P. I. P. Elliott, Angew. Chem. 2013, 125, 11026;
Angew. Chem. Int. Ed. 2013, 52, 10826; b) K. Garg, J. T. Engle, C. J. Zieg-
ler, J. J. Rack, Chem. Eur. J. 2013, 19, 11686; c) B. Boff, M. Ali, L. Alexan-
drova, N. ꢅ. Espinosa-Jalapa, R. O. Saavedra-Dꢆaz, R. Le Lagadec, M.
Pfeffer, Organometallics 2013, 32, 5092.
Received: September 15, 2013
Revised: October 30, 2013
Chem. Eur. J. 2014, 20, 781 – 793
www.chemeurj.org
793
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim