A heterobimetallic approach to stabilize the elusive disulfur radical trianion ( subsulfide ) S23-

Article abstract of DOI:10.1002/chem.201203642

A unique heterobimetallic disulfur monoradical, complex 2, with a diamond-shaped {NiS₂Pt} core has been synthesized by two-electron reduction of a supersulfido-(nacnac)nickel(II) complex (nacnac=β- diketiminato) with [Pt(Ph₃P)₂(η²-C ₂H₄)] as a platinum(0) source and isolated in 82 % yield. Strikingly, the results of DFT calculations in accordance with spectroscopic (EPR, paramagnetic NMR) and structural features of the complex revealed that the bonding situation of the S₂ ligand is between the elusive half-bonded S₂ radical trianion (S₂ ^3-) and two separated S^2- ligands. Accordingly, the Ni^II center is partially oxidized, whereas the Pt^II site is redox innocent. The complex can be reversibly oxidized to the corresponding Ni,Pt-disulfido monocation, compound 3, with a S-S single bond, and reacts readily with O₂ to form the corresponding superoxonickel(II) and disulfidoplatinum(II) (4) complexes. These compounds have been isolated in crystalline form and fully characterized, including IR and multi-nuclear NMR spectroscopy as well as ESI mass spectrometry. The molecular structures of compounds 2-4 have been confirmed by single-crystal X-ray crystallography. Copyright

Full text of DOI:10.1002/chem.201203642

DOI: 10.1002/chem.201203642
A Heterobimetallic Approach To Stabilize the Elusive Disulfur Radical
C3ꢀ
Trianion (“Subsulfide”) S2
Shenglai Yao,[a] Peter Hrobꢀrik,[b, c] Florian Meier,[b] Robert Rudolph,[a] Eckhard Bill,[d]
Elisabeth Irran,[a] Martin Kaupp,*[b] and Matthias Driess*[a]
Dedicated to Professor Karl Wieghardt on the occasion of his 70th birthday
ꢀ
Abstract: A unique heterobimetallic di-
sulfur monoradical, complex 2, with a
diamond-shaped {NiS2Pt} core has
been synthesized by two-electron
reduction of a supersulfido-(nacnac)-
nickel(II) complex (nacnac=b-diketi-
uation of the S2 ligand is between the
elusive “half-bonded” S2 radical trian-
pound 3, with a S S single bond, and
reacts readily with O2 to form the cor-
responding superoxonickel(II) and di-
sulfidoplatinum(II) (4) complexes.
These compounds have been isolated
in crystalline form and fully character-
ized, including IR and multi-nuclear
NMR spectroscopy as well as ESI mass
spectrometry. The molecular structures
of compounds 2–4 have been con-
firmed by single-crystal X-ray crystal-
lography.
and two separated S2ꢀ
C3ꢀ
ion (S2
)
ligands. Accordingly, the NiII center is
partially oxidized, whereas the PtII site
is redox innocent. The complex can be
reversibly oxidized to the correspond-
ing Ni,Pt-disulfido monocation, com-
(h2-C2H4)] as a
minato) with [PtACHTUNGTRENNUNG(Ph3P)2ACHTUGNTRENNUGN
platinum(0) source and isolated in
82% yield. Strikingly, the results of
DFT calculations in accordance with
spectroscopic (EPR, paramagnetic
NMR) and structural features of the
complex revealed that the bonding sit-
Keywords: dichalcogens · nickel ·
oxidation states · platinum · redox
ligands
Introduction
moiety X2 (X=O, S) into two chalcogenide centers (X2ꢀ) re-
quires in total four electrons (Scheme 1).
Akin to the metal-mediated activation of dioxygen, the re-
duction of elemental sulfur (commonly Sn rings such as S8)
with metals can lead to different metal disulfide species
bearing the supersulfide (Sꢀ2 ), disulfide (S22ꢀ), or bridging sul-
fide ligand (S2ꢀ). Reductive cleavage of a dichalcogen
Scheme 1. Step-by-step reduction of a dichalcogen X2.
[a] Dr. S. Yao, Dipl.-Chem. R. Rudolph, Dr. E. Irran, Prof. Dr. M. Driess
Technische Universitꢁt Berlin
Understanding these step-by-step transformations in the
environment of transition metals is of considerable interest
because of their importance in catalytic redox processes and
biorelevance.[1] Although the chemistry of transition-metal
complexes containing dichalcogenide X22ꢀ and bridging X2ꢀ
ligands has been well established, only a few striking exam-
ples featuring a superchalcogenide Xꢀ2 ligand could be isolat-
ed and structurally characterized.[2,3] Realization of a dichal-
Department of Chemistry
Metalorganics and Inorganic Materials
Sekr. C2, Strasse des 17. Juni 135
10623 Berlin (Germany)
E-mail: matthias.driess@tu-berlin.de
[b] Dr. P. Hrobꢂrik, Dipl.-Chem. F. Meier, Prof. Dr. M. Kaupp
Technische Universitꢁt Berlin
Department of Chemistry: Theoretical Chemistry
Sekr. C7, Strasse des 17. Juni 135
10623 Berlin (Germany)
C3ꢀ
cogenide radical ligand X2 (“subchalcogenides”) with a
ꢀ
ꢀ
X X s bond order of 0.5 (similar to H2 ) represents an even
E-mail: martin.kaupp@tu-berlin.de
more ambitious task because dichalcogen Xn2ꢀ ligands are
expected to act as non-innocent redox ligands.
[c] Dr. P. Hrobꢂrik
Institute of Inorganic Chemistry
Slovak Academy of Sciences
Dfflbravskꢂ cesta 9
Recently, a striking copper complex with a {Cu3S2}3+ core
engendered a vivid debate about the interpretation of its
electronic structure (Figure 1).[4] Unlike the oxygen conge-
84536 Bratislava (Slovakia)
ner {Cu3O2}3+
,
which has a local C2v symmetry originating
[5]
[d] Dr. E. Bill
Max-Planck Institute of Chemical Energy Conversion
Stiftsstrasse 34-36
45470 Müllheim/Ruhr (Germany)
ꢀ
from shorter Cu O distances for one Cu atom than for the
other two, and which features unambiguously two O2ꢀ ions,
the {Cu3S2}3+ core adopts a local D3h symmetry and contains
a S2 moiety with a S···S distance of 2.69 ꢀ. It may be consid-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201203642.
1246
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 1246 – 1253

FULL PAPER
gations revealed that complex 2 shows significant S23ꢀ radical
C
character (ꢁ41%), very similar to the {Cu3S2}3+ system but
with surprisingly unsymmetrical atomic spin densities at the
two sulfur atoms (34 vs. 7%). In addition, our calculations
C3ꢀ
show that “pure” S2 radical character would be present
upon replacement of the redox non-innocent NiII in complex
a
{Cu3S2}3+ core.
ꢀ
Figure 1. Possible S S bonding interactions in
2 by redox innocent ZnII.
L’=N,N,N’,N’-tetramethylethylenediamine (tmeda).
ered as a species having a weak attractive S···S interaction in
Results and Discussion
C3ꢀ
formal accordance with S2 (“subsulfide”) character[4] (see,
very recent discussion of a possible Se23ꢀ-type ligand in a
Synthesis of radical complex 2: Treatment of 1 with two
equivalents of bis(triphenylphosphine)ethylene platinum(0)
in toluene at ꢀ208C resulted in the formation of complex 2,
indicated by the color change from dark brown to dark
green (Scheme 2). After workup, complex 2 could be isolat-
ed in 82% yield as dark green crystals. Its composition was
proven by elemental analysis and ESI mass spectrometry. Its
molecular structure has been established by single-crystal
X-ray diffraction analysis (Figure 2).
C
{Ni2Se2} complex[6]). However, X-ray adsorption spectrosco-
py and computational studies by using various electronic-
structure analysis approaches clearly demonstrated the in-
trinsic difficulty in assigning the formal oxidation states to
Cu and S in the system, owing to the high symmetry of the
{Cu3S2}3+ core, the large covalent character of the Cu S
ꢀ
bonds, and significant S,S!Cu(3d) donation.[4] The latter
controversy over the oxidation state encouraged us to devel-
op a less symmetric M,M’-heterobimetallic disulfur complex,
featuring a diamond-shaped {M(S2)M’} core (see Scheme 2)
and an electron count, which might facilitate the formation
C3ꢀ
of the elusive disulfur radical ligand S2
.
Figure 2. Molecular structure of complex 2. Thermal ellipsoids are drawn
at 50% probability level. Hydrogen atoms are omitted. Two independed
molecules are in the unit cell. Selected distances [ꢀ] and angles [8]: mole-
Scheme 2. Formation of the {Ni(S2)Pt} radical complex 2 from reaction of
supersulfidonickel(II) with [Pt(Ph3P)2A
(h2-C2H4)] and its reversible
redox conversion into 3. The electronic structure of 2 is described by two
resonance forms with spin densities residing at the sulfur and the nickel
atoms, respectively.
ꢀ
ꢀ
ꢀ
1
G
CHTUNGTRENNUNG
cule 1: S1···S2 2.71(2), Pt1 P1 2.283(2), Pt1 P2 2.301(2), Pt1 S1 2.305(2),
ꢀ
ꢀ
ꢀ
ꢀ
Pt1 S2 2.314(2), Ni1 N1 1.942(6), Ni1 N2 1.946(5), Ni1 S1 2.153(2),
ꢀ
Ni1 S2 2.164(2), P1-Pt1-P2 100.49(6), P1-Pt1-S1 94.42(6), P2-Pt1-S1
163.53(6), P1-Pt1-S2 166.17(6), P2-Pt1-S2 93.34(6), S1-Pt1-S2 71.83(6),
N1-Ni1-N2 93.9(2), S1-Ni1-S2 77.75(7), Ni1-S1-Pt1 103.90(7), Ni1-S2-Pt1
ꢀ
ꢀ
103.26(7); molecule 2: S1···S2 2.70(2), Pt1 P2 2.286(2), Pt1 P1 2.299(2),
ꢀ
ꢀ
ꢀ
ꢀ
Pt1 S1 2.302(2), Pt1 S2 2.307(2), Ni1 N2 1.931(5), Ni1 N1 1.948(6),
As a promising approach we pursued the synthesis of a
ꢀ
ꢀ
Ni1 S2 2.150(2), Ni1 S1 2.162(2), P2-Pt1-P1 100.12(6), P2-Pt1-S1
166.37(6), P1-Pt1-S1 93.49(6), P2-Pt1-S2 94.86(6), P1-Pt1-S2 163.76(6),
S1-Pt1-S2 71.71(6), N2-Ni1-N1 93.5(2), S2-Ni1-S1 77.51(7), Ni1-S1-Pt1
102.29(7), Ni1-S2-Pt1 102.48(7).
C3ꢀ
heterobimetallic S2 complex through a two-electron reduc-
tion of the supersulfido ligand in the dimeric supersulfido-
ACHTUNGTRENNUNG
responding nickel(I) precursor [{NiACTHNUTRGNEUNG
(nacnac)}2]·toluene[7] with
elemental sulfur.[3c] After a series of unsuccessful experi-
ments employing several potential reducing agents (e.g.,
Mg, Ni0, and Pd0 complexes), we learned that platinum(0) in
The Ni S distances in 2 (2.150(2)–2.164(2) ꢀ) are close to
those found in 1 (2.163(1), 2.166(1) ꢀ) and slightly shorter
ꢀ
than those observed for the butterfly-like complex with a
[Pt
G
N
{NiII (m-h2:h2-S2)} core, ranging from 2.193(1) to
2
ducing agent to achieve the desired reduction of the super-
sulfido complex 1 to give compound 2 (Scheme 2). Here, we
report the synthesis, characterization, and reactivity of the
unique {Ni(S2)Pt} radical complex 2. Our theoretical investi-
2.205(1) ꢀ.[3c] The Pt S distances in
2
(2.302(2) and
ꢀ
2.314(2) ꢀ), are similar to the corresponding values ob-
served in complexes with
a
{Pt2ACTHNGUTERNNU(G m-S)2} core (2.33–
2.35 ꢀ),[8a–c] and the d8-Pt center is square-planar coordinat-
Chem. Eur. J. 2013, 19, 1246 – 1253
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1247

M. Kaupp, M. Driess et al.
ed. The most interesting geometric features are the non-
planar tetra-coordination of the Ni site and the S···S distance
of approximately 2.71 ꢀ. The latter is much shorter than the
31P NMR signals and aid in the spectra assignment (Ta-
bles S8 and S9 as well as Figure S1 in the Supporting Infor-
mation). It turned out that the best agreement with experi-
mental NMR shifts was obtained with a modified B3LYP**
functional with reduced exact-exchange admixture (10%),[11]
thus avoiding excessive spin contamination (see Tables S5
and S8 as well as Figure S3 in the Supporting Information).
This functional was used subsequently to establish spin-den-
sity distributions and the electronic structure of complex 2
and of related complexes. The electronic g tensor of 2 is also
well reproduced at this level, with giso =2.162 and g-tensor
principal values of 2.085, 2.166, and 2.235. Closer analysis
(Table S7 in the Supporting Information) reveals that it is
dominated by Ni spin–orbit coupling, thus confirming non-
negligible spin density at this metal center.
distances observed in complexes with a {Pt2ACTHUNTRGNE(UGN m-S)2} core
(3.060–3.134 ꢀ).[8a–d] It is also shorter than that observed for
the naphthalene-based bis(phosphine)nickel(II) complex
with a {Ni
longer than the values observed for complexes with a
{Ni2A
(h2:h2-S2)} core (2.051(1)–2.054(2) ꢀ).[3c]
Complex 2 is paramagnetic in the solid state as well as in
2ACHTUNGTRENNUNG
(m-S)2} core (2.839(4) ꢀ),[8e] but significantly
A
CHTUNGTRENNUNG
solution. Magnetic measurement performed in C6D6 (by
using the Evans method[9]) revealed a meff value of 1.83 mB at
room temperature, which is close to the spin-only value of
one unpaired electron. Correspondingly, the X-band EPR
spectrum recorded at 10 K from a frozen solution of com-
plex 2 in THF (Figure 3) shows three rhombically split de-
Various approaches have been used to analyze the elec-
tronic structure of complex 2, in an attempt to assign “physi-
cal oxidation states” to Ni, Pt, and the bridging S2 unit. To
C3ꢀ
have a “pure” S2 subsulfide reference point, the simple
model complex Na3S2 has been optimized and analyzed at
2ꢀ
the same levels, whereas the cation [2]+ is our disulfide S2
reference point. Additional calculations on the analogues
2-Pt and 2-Zn (replacing Ni in complex 2 by Pt and by tetra-
hedrally coordinated Zn, respectively), the cation [2-Zn]+,
the anion [2-Zn]ꢀ, on [Na3S2]+/ꢀ, and on the [(tmeda)3-
AHCTUNGTRENNUNG
Cu3S2]3+ complex[4] (see Figure 1) provide further points for
comparison.
Spin-density distributions (Figure 4, Table 1) provide a
particularly informative view: the natural orbital population
(NPA) spin densities near 0.5 on both sulfur atoms, and
rather small spin densities on the metal fragments in Na3S2
and 2-Zn agree with expectations for an S23ꢀ radical disulfide
C
Figure 3. X-band EPR spectrum of complex 2 in a frozen THF solution
measured at 10 K (frequency 9.4390 GHz, power 0.2 mW, modulation
1 mT per 100 kHz). The gray line is the result of a powder simulation
with anisotropic g values (g=2.365, 2.143, 2.048) and line widths (G=
38”10ꢀ4, 31”10ꢀ4, 38”10ꢀ4 cmꢀ1).
rivative lines (g=2.365, 2.143, 2.048) with an average g
value of 2.189 (corresponding to meff =1.896 mB). Both, the
1H and the 31P NMR spectrum of 2 measured in C6D6 exhib-
it paramagnetically shifted resonances (see the Supporting
Information), the latter with
a
broad signal at
d=ꢀ415 ppm. Among the 1H NMR shifts, the most pro-
nounced shielding due to the paramagnetic nature of the
complex is experienced at the g position of the nacnac
ligand (d=ꢀ39.3 ppm), indicating some spin delocalization
from the {NiS2Pt} core to this fragment.
DFT calculations for complex 2: Structure optimizations on
complex 2 at the RI-BP86/def2-TZVP level (see Computa-
tional methods section) provide good agreement with the
ꢀ
ꢀ
X-ray diffraction data (dACTHNUGRTENNUG(S S)=2.717, dACHTUNGERTN(NUNG Ni S)=2.136,
ꢀ
2.201, d
A
ing Information). Subsequent state-of-the-art DFT calcula-
tions of paramagnetic NMR chemical shifts[10] at this struc-
ture reproduce well the experimentally observed 1H and
Figure 4. Spin-density isosurface plots (ꢂ0.001 a.u.) in the vicinity of the
S2 unit in pertinent metal complexes (B3LYP**/TZVP results). Positive
spin density is indicated in blue, negative spin density in red.
1248
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 1246 – 1253

C3ꢀ
A Heterobimetallic Approach to Stabilize S2
FULL PAPER
Table 1. Crucial NPA atomic spin densities in selected systems.[a]
Table 2). The twisting survives even in smaller model com-
plexes, in which the alkyl and aryl substituents on the
nacnac ligand and on the Pt-bound phosphine ligands were
replaced by hydrogen atoms (see the Supporting Informa-
tion). This indicates that the asymmetry is not caused by
steric congestion but electronically preferred. Indeed, the
twisting is related to a partial oxidation of the NiII center
beyond the d8 configuration and respective partial reduction
Complex
NPA spin densities
of {MS(1)S(2)M’} core atoms
S(1)
S(2)
M
M’=Pt
A
R
2
0.346
0.352[b]
0.506
0.446
0.240
0.521
0.070
0.123[b]
0.429
0.282
0.240
0.521
0.499
0.028
0.026[b]
0.011
0.058
–
0.454[b]
0.000
E
R
2-Zn
2-Pt
N
0.204
R
0.272[c]
ꢀ0.014[c]
of the disulfur ligand to a state between the S2 and two S2ꢀ
C3ꢀ
C
–
moieties (Scheme 2). A genuine tetra-coordinate, low-spin
NiII has no reason to become non-planar. Consequently, the
twisting is absent in small truncated models of the one-elec-
[a] B3LYP**/TZVP//BP86/def2-TZVP results (see the Computational
methods section). M atoms marked in bold. L stands for the CH-
ACHTUNGTRENNUNG{(CMe)(2,6-iPr2C6H3N)}2 ligand, denoted also as “nacnac” throughout
the text. [b] Atomic spin densities calculated for the crystal structure of 2
with optimized hydrogen positions. [c] Average values over all metal
atoms (M=Cu or Na).
tron-oxidized cationic complex [2]+ (see compound
3
below), which represents a true d8 NiII situation. Consistent
with this picture, the Ni charge in complex 2 is clearly more
positive than that in [2]+. We furthermore note, that only
the spin densities of Na3S2 and of the Cu complex are con-
ion. These two compounds provide clear-cut models for
ꢀ
ꢀ
such an oxidation state. The S S distance in complex 2
(Table 2) is actually shorter than those in the two reference
sistent with a hole in the s* orbital of the S S bond. In all
other cases, the lower symmetry leads to substantial contri-
butions of pp-type atom orbitals (AOs) to the SOMO and
thus to the spin density on the S2 unit (see Figure 4 and
Tables S14 and S15 in the Supporting Information).
models and very similar to the values obtained for
[(tmeda)3Cu3S2]3+. This suggests that the Coulombic S S re-
ꢀ
pulsion is better compensated due to an increased covalency
ꢀ
of the M S bond, as also confirmed by NPA charges
Compared to the copper system, all other systems exhibit
much less spin polarization at a given computational level
(Figure 4), clearly ruling out a broken-symmetry-type wave
function discussed by Berry[4d] for [(tmeda)3Cu3S2]3+ (see
also natural orbital populations in Table S13 in the Support-
ing Information).
We have used further measures in an attempt to establish
the presence or absence of attractive S···S interactions in
complex 2 and in the other compounds studied. Neither in
the electron localization function (ELF), the electron local-
izability indicator (ELI), or in the Laplacian of the electron
(Table 2). Strikingly, the sulfur atoms in 2 exhibit much
lower and asymmetrical sulfur spin densities, overall also
somewhat less on average than that of the copper system
(Table 1). The inequality clearly reflects the asymmetric
bonding of the disulfur ligand to the non-planar tetra-coor-
dinate Ni center. Even based on the crystal structure with
ꢀ
ꢀ
its more similar Ni S distances (see Ni S distances in
Table S2 in Supporting Information), the spin density re-
mains strongly asymmetric (see Table 1). This is due to a
twisting of the S2 ligand relative to the N-Ni-N plane of the
(nacnac)Ni complex fragment: The S1 atom is significantly
ꢀ
density, complex 2 exhibits an S S bonding attractor (see
ꢀ
ꢀ
outside the plane, leading to a larger Ni S1 than Ni S2 co-
valency, consistent with respective NPA charges (see
Table S16 in the Supporting Information). However, such an
attractor is also absent in 2-Zn or in Na3S2, which may be
C3ꢀ
considered as “pure” S2 exam-
ples, whereas such attractors
Table 2. Optimized S···S distances in [ꢀ], S···S Mayer bond orders (MBO), and NPA atomic charges in perti-
nent complexes.[a]
are clearly present in any disul-
fide complex we looked at (see
the Supporting Information). It
is thus difficult to establish the
Complex
d
(S···S)
MBO
NPA charges
S(1)
S(2)
M
Pt
C3ꢀ
radical disulfides, S2
ꢀ
presence of S S bonding in the
A
R
2
2.717
2.895
2.943
2.703
3.421
0.215
0.339
0.218
0.206
0.430
ꢀ0.425
ꢀ0.702
ꢀ0.350
ꢀ1.267
ꢀ0.675
ꢀ0.555
ꢀ0.762
ꢀ0.459
ꢀ1.267
ꢀ0.675
0.875
1.546
0.129
0.141
0.155
–
radical disulfide trianion species
by such real-space quantities.
On the other hand, Mayer bond
orders (Table 2) indicate non-
E
R
2-Zn
2-Pt
E
R
0.542
R
1.020[b]
0.845[b]
C
–
ꢀ
negligible but weak S S bond-
disulfides, S22ꢀ
[LNiS2Pt
(PPh3)2]+
[LZnS2Pt
(PPh3)2]+
[Na3S2]+
ing in complex 2 as well as in
G
E
[2]+
A
2.065
2.141
2.226
0.947
0.942
0.894
ꢀ0.168
ꢀ0.389
ꢀ0.931
ꢀ0.180
ꢀ0.407
ꢀ0.931
0.777
0.103
0.112
–
[(tmeda)3Cu3S2]3+
,
thus con-
firming a situation between a
genuine S2 radical trianion sit-
E
G
[2-Zn]+
1.564
A
0.954[b]
C3ꢀ
sulfides, 2S2ꢀ
[LZnS2Pt
(PPh3)2]ꢀ
[Na3S2]ꢀ
uation and two separated S2ꢀ
bridging ligands, but with un-
symmetrical sulfur spin-density
distribution and bonding situa-
tion for complex 2. However, it
A
U
A
3.278
4.102
0.052
0.042
ꢀ1.025
ꢀ1.533
ꢀ1.037
ꢀ1.533
1.511
0.159
–
E
0.689[b]
[a] B3LYP**/TZVP//BP86/def2-TZVP results (see the Computational methods section). M atoms marked in
bold. L stands for the CH
Na).
ACHTUNGTRNEGNUN{(CMe)(2,6-iPr2C6H3N)}2 ligand. [b] Average values over all metal atoms (M=Cu or
Chem. Eur. J. 2013, 19, 1246 – 1253
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1249

M. Kaupp, M. Driess et al.
is clearly not a broken-symmetry case, and the spin polariza-
tion is much weaker. We note in passing that weak S···S in-
teractions have also been discussed in detail in S24ꢀ-type
ligands in certain rhodium and iridium complexes and relat-
ed model systems.[4b,12]
through one-electron oxidation, the Ni,Pt-subsulfido com-
plex 2 was converted into the Ni,Pt-disulfido cation in 3 as
confirmed by electronic-structure analyses (see data for [2]+
in Table 2). Interestingly, the oxidation is reversible as
proven by the chemical reduction of 3 with one equivalent
of bis(h5-pentamethylcyclopentadienyl)cobalt, affording
complex 2 in 73% yield (see Scheme 2).
Reactivity of radical complex 2: The one-electron oxidation
of complex 2 with [FeCp2][BACHTUNGTRENUNG(C6F5)4] (Cp=cyclopentadienyl)
leads to the formation of the ion-separated pair 3, which has
been readily obtained as black crystals in 89% yield
(Scheme 2). Compound 3 is diamagnetic and displays rea-
1
sonable resonances in the H NMR spectrum. The 31P nu-
cleus of the PPh3 ligand resonates at d=11.6 ppm with 195Pt
satellites (1J
ACHTUNGTRENNUNG(P,Pt)=3963 Hz). The composition of 3 has
been further confirmed by 13C and 19F NMR as well as by el-
emental analysis. Compound 3 is marginally soluble in hy-
drocarbons but well soluble in THF and dichloromethane.
The molecular structure of 3 has also been established by
X-ray diffraction analysis (Figure 5). No significant interac-
Scheme 3. Reaction of complex 2 with dioxygen to form the superoxo-
nickel(II) complex and the disulfidoplatinum(II) complex 4. Compound 4
can be reconverted into 2 by reduction with the corresponding (nacnac)-
nickel(I) precursor, [{NiACHTNUGTRENUNG(nacnac)}2]·toluene.
Complex 2 is sensitive toward air and moisture. It reacts
readily with dry dioxygen to give the corresponding
superoxoACTHNUTRGNEUNG
(nacnac)nickel(II)[2c] and disulfidobis(triphenyl-
phosphine)platinum(II) (4) complexes as sole products
(Scheme 3). Thus complex 4 can be readily prepared from 2
in 77% yield and has been structurally characterized
(Figure 6). Remarkably, such disulfide-platinum(II) com-
plexes are rare. The latter species has only previously been
detected and spectroscopically characterized by Ishiiand co-
workers.[13] Remarkably, compound 4 can be reconverted
Figure 5. Molecular structure of 3. Thermal ellipsoids are drawn at 50%
probability level. Hydrogen atoms are omitted. Selected distances [ꢀ]
ꢀ
ꢀ
ꢀ
ꢀ
and angles [8]: S1 S2 2.058(3), Pt1 P2 2.254(2), Pt1 P1,2.286(3), Pt1 S2
ꢀ
ꢀ
ꢀ
ꢀ
2.344(2), Pt1 S1 2.360(2), Ni1 N1 1.880(7), Ni1 N2 1.881(7), Ni1 S1
ꢀ
2.196(2), Ni1 S2 2.231(2), P2-Pt1-P1 100.43(9), P2-Pt1-S2 104.81(8),
P1-Pt1-S2 154.58(8), P2-Pt1-S1 151.27(9), P1-Pt1-S1 103.12(8), S2-Pt1-S1
51.90(8), N1-Ni1-N2 97.7(3), S1-Ni1-S2 55.40(9), Ni1-S1-Pt1 100.10(8),
Ni1-S2-Pt1 99.55(9).
tions were observed between the counterions (shortest Ni–F
distance: 5.184 ꢀ and shortest Pt–F distance: 6.347 ꢀ). The
cation adopts a conformation quite similar to that of its pre-
cursor 2 except for the folding angle of the four-membered
NiS2Pt ring (62.08 in 3 vs. 17.0 and 23.28 in 2). However, the
Figure 6. Molecular structure of compound 4. Thermal ellipsoids are
drawn at 50% probability level. Hydrogen atoms are omitted. Selected
ꢀ
ꢀ
ꢀ
distances [ꢀ] and angles [8]: Pt1 P1 2.255(2), Pt1 P2 2.284(2), Pt1 S1
ꢀ
S S distance in 3 (2.0583(3) ꢀ), which is significantly short-
ꢀ
ꢀ
2.324(2), Pt1 S2 2.339(2), S1 S2 2.092(3), P1-Pt1-P2 100.34(8), P1-Pt1-S1
103.06(8), P2-Pt1-S1 156.60(9), P1-Pt1-S2 156.34(8), P2-Pt1-S2 103.30(8),
S1-Pt1-S2 53.32(9), S2-S1-Pt1 63.72(9).
er than the corresponding value found in 2 (2.70 or 2.71 ꢀ),
ꢀ
falls in the ordinary range of S S single bond lengths. Thus,
1250
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 1246 – 1253

C3ꢀ
A Heterobimetallic Approach to Stabilize S2
FULL PAPER
2
reflections with I>2s(l), qmax =258, R(Fo)=0.0882 (I>2s(l)), wR
0.1570 (all data), 1388 refined parameters.
(Fo )=
into complex 2 by treatment of 4 with the nickel(I) precur-
sor [{Ni
(nacnac)}2]·toluene[7] (see Scheme 3).
ACHTUNGTRENNUNG
¯
Complex 4: Triclinic, space group P1, a=11.1674(7), b=12.4848(8), c=
16.558(1) ꢀ,
a=92.875(5),
b=109.131(6),
g=110.834(6)8,
V=
14782
2001.8(2) ꢀ3, Z=2, 1calcd =1.300 Mgmꢀ3
,
mG
,
Conclusion
collected reflections, 7017 crystallographically independent reflections
[Rint =0.0538], 5611 reflections with I>2s(l), qmax =258, R(Fo)=0.0536
2
(I>2s(l)), wR
N
The two-electron reduction of the dimeric supersulfido-
(nacnac)nickel(II) complex 1 with bis(triphenylphosphine)-
ethyleneplatinum(0) furnishes the unique Ni,Pt-subsulfido
EPR measurements: X-band EPR derivative spectra were recorded on a
Bruker ELEXSYS E500 spectrometer equipped with the Bruker stand-
ard cavity (ER4102ST) and a helium flow cryostat (Oxford Instruments
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
complex 2, which undergoes reversible one-electron oxida-
ESR 910). Microwave frequencies were calibrated with
a Hewlett–
tion to form the corresponding diamagnetic Ni,Pt-disulfido
Packard frequency counter (HP5352B), and the field control was cali-
brated with a Bruker NMR field probe (ER035M). The spectra were si-
mulated with the program GFIT (by E.B.) for the calculation of the
powder spectra with effective g values and anisotropic line widths (mixed
Lorentzian and Gaussian line shapes were used).
ꢀ
cation in compound 3 with a S S single bond. DFT calcula-
tions in accordance with spectroscopic (EPR, paramagnetic
NMR) and structural features of complex 2 revealed a weak
attractive S···S interaction in the S2 bridge, highlighting sig-
Synthesis of 2
C3ꢀ
nificant S2 radical character (ca. 41%), very similar to the
(h2-C2H4)] (0.83 g, 1.11 mmol) in tol-
Method A: A solution of [PtACHTNUGTRENNUG(Ph3P)2ACHTUGNTRENNUGN
{Cu3S2}3+ system but with strongly unsymmetrical sulfur spin
densities owing to a twisted coordination of the S2 ligand to
the Ni center. In fact, this asymmetry is related to the partial
uene (15 mL) was added to a cooled (ꢀ208C) solution of 1 (0.60 g,
1.11 mmol) in toluene (15 mL). After stirring for 10 min, the reaction
mixture was allowed to warm to room temperature and stirred further
for 2 h. In the course of stirring, the color of the solution changed from
brown to green. Volatiles were removed in vacuum and the residue was
extracted with diethyl ether (5”30 mL). After concentration and cooling
to ꢀ208C for 24 h, compound 2 crystallized from the solution as dark
green crystals (1.15 g, 0.91 mmol, 82%).
C3ꢀ
reduction of the disulfur ligand beyond the S2 state, and
thus partial oxidation of nickel beyond NiII. Furthermore,
our computational results predict that the replacement of
the redox non-innocent NiII center in 2 by ZnII generates a
C3ꢀ
Method B: [PtACHTUNGTRNEN(GU Ph3P)2S2] (0.31 g, 0.40 mmol) was added to a cooled
heterobimetallic subsulfido radical complex with “pure” S2
(ꢀ208C) solution of [(NiL)2]·toluene (0.21 g, 0.20 mmol) in toluene
(15 mL). After stirring for 30 min, the reaction mixture was allowed to
warm to room temperature and stirred further for 2 h. In the course of
stirring the color of the solution changed from red-brown to green. Vola-
tiles were removed in vacuum and the residue was extracted with diethyl
ether (3”30 mL). After concentration and cooling to ꢀ208C for 24 h,
compound 2 crystallized from the solution as dark green crystals (0.45 g,
0.36 mmol, 89%).
character.
Experimental Section
General considerations: All experiments and manipulations were carried
out under dry oxygen-free nitrogen by using standard Schlenk techniques
or in an MBraun inert atmosphere glovebox containing an atmosphere of
purified nitrogen. Solvents were dried by standard methods and freshly
distilled prior to use. The starting materials 1,[3c] [(NiL)2]·toluene[7] (L=
Method C: A solution of [CoACHTUNRGTNE(NUG C5Me5)2] (0.034 g, 0.10 mmol) in THF
(5 mL) was added to a cooled (ꢀ708C) solution of compound 3 (0.20 g,
0.10 mmol) in THF (15 mL). After stirring for 10 min, the reaction mix-
ture was allowed to warm to room temperature and stirred further for
2 h. Volatiles were removed in vacuum and the residue was extracted
with diethyl ether (3”20 mL). Cooling the concentrated solution to
ꢀ208C for 24 h afforded compound 2 as dark green crystals (0.095 g,
0.076 mmol, 73%). M.p. 2228C (decomp.); IR (KBr): n˜ =423 (w), 456
(w), 465 (w), 494 (m), 513 (m), 522 (s), 539 (m), 694 (s), 730 (m), 742
(m), 792 (w), 850 (w), 935 (w), 999 (w), 1028 (w), 1057 (w), 1059 (m),
1159 (w), 1179 (w), 1252 (w), 1317 (m), 1383 (m), 1436 (s), 1461 (w),
1479 (w), 1495 (w), 1525 (w), 1544 (w), 2863 (w), 2923 (m), 2958 (s),
CH
ACHTUNGTRENNUNG{(CMe)(2,6-iPr2C6H3N)}2), [FeCAHTUNTGREN(NGU C5H5)2][BAHCUTNGETRNNUNG ACHTUNGTRENN(UNG Ph3P)2-
(C6F5)4],[14] and [Pt
ACHTUNGTRENNUNG
resolution ESI-MS were measured on a Thermo Scientific LTQ orbitrap
XL. 1H and 31P NMR spectra were recorded on Bruker spectrometer
APX 200 and AV 400, and the corresponding chemical shifts are refer-
enced to the H NMR signal of tetramethylsilane and the 31P NMR signal
1
of 85% H3PO4.
Single-crystal X-ray structure determinations: Crystals were each mount-
ed on a glass capillary in perfluorinated oil and measured in a cold N2
flow. The data of compounds 2–4 were collected on an Oxford Diffrac-
tion Xcalibur S Sapphire at 150 K (MoKa radiation, l=0.71073 ꢀ). The
structures were solved by direct methods and refined on F2 with the
SHELX-97 software package.[16] The positions of the hydrogen atoms
were calculated and considered isotropically according to a riding model.
CCDC-896855 (2), 896856 (3), and 896857 (4) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
3051 cmꢀ1 (w); HR ESI-MS (ion spray voltage 5 kV, flow rate 5 mLminꢀ1
,
in THF): m/z calcd for C65H72N2NiS2P2Pt [M+H]+: 1259.36079; found:
1259.35039; elemental analysis calcd (%) for C65H71N2NiS2P2Pt: C 61.95,
H 5.68, N 2.22; found: C 61.70, H 5.75, N 2.15. See Figure S1 and S2 as
well as Tables S8 and S9 in the Supporting Information for 1H and
31P NMR data and signal assignment.
Synthesis of 3: A solution of [FeACTHNURTGENNG(U C5H5)2][BCAHTUNGTRNE(NUGN C6F5)4] (0.34 g, 0.40 mmol) in
THF (10 mL) was added to a cooled (ꢀ308C) solution of 2 (0.50 g,
0.40 mmol) in THF (10 mL) with stirring. After addition the reaction
mixture was allowed to warm to room temperature and stirred further
for 2 h. Volatiles were removed in vacuum and the dark brown residue
was washed with n-hexane (25 mL) and extracted with benzene (20 mL).
After concentration and cooling to 68C for 48 h, compound 3 crystallized
as black crystals (0.69 g, 0.36 mmol, 89%). M.p. 2318C (decomp);
1H NMR (400.13 MHz, [D8]THF, 258C): d=1.07 (d, 3J (H,H)=7 Hz,
12H; CHMe2), 1.18 (d, 3J (H,H)=7 Hz, 12H; CHMe2), 1.45 (s, 6H;
NCMe), 3.73–4.00 (m., 4H; CHMe2), 5.09 (s, 1H; g-CH), 6.81–7.43 ppm
(m, 36H, Ar-H); 13C{1H} NMR (100.61 MHz, [D8]THF, 258C): d=21.9,
24.0, 26.4, 29.4 (NCMe, CHMe2), 101.5 (g- C), 124.7, 127.7, 141.5, 154.1,
161.5 (NCMe, 2,6-iPr2C6H3), 129.6, 129.7, 129.8, 132.9, 134.9, 135.0,
¯
Complex 2: Triclinic, space group P1, a=12.8699(3), b=18.2620(4), c=
28.7662(6) ꢀ,
a=80.245(2),
b=79.438(2),
g=72.610(2)8,
V=
45219
6295.0(2) ꢀ3, Z=4, 1calcd =1.330 Mgmꢀ3
,
mU
,
collected reflections, 22134 crystallographically independent reflections
[Rint =0.0532], 16179 reflections with I>2s(l), qmax =25.008, R(Fo)=
0.0562 (I>2s(l)), wR
2
(Fo )=0.1257 (all data), 1335 refined parameters.
Complex 3: Orthorhombic, space group Pbcn, a=17.105(1), b=
29.104(1), c=39.338(2) ꢀ, a=90, b=90, g=908, V=19608(2) ꢀ3, Z=8,
ACHTUNGTRENNUNG
1calcd =1.495 Mgmꢀ3, m(MoKa)=1.781 mmꢀ1, 76731 collected reflections,
17212 crystallographically independent reflections [Rint =0.0981], 10891
Chem. Eur. J. 2013, 19, 1246 – 1253
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1251

M. Kaupp, M. Driess et al.
135.1 ppm (P
11.6 ppm (195Pt satellites, 1J
[D8]THF, 258C): d=ꢀ134.4 (brs, o-F), ꢀ166.8 (t, J
ꢀ170.2 (brt, J(F,F)=17.6 Hz, m-F); IR (KBr): n˜ =457 (w), 499 (w), 510
G
bitals (GIAO).[33] These calculations have been done with the same ex-
change-correlation functionals and basis sets as the g-tensor calculations
(see above). All temperature-dependent terms were computed at 298 K.
The calculated 1H and 31P NMR shieldings were converted to chemical
shifts (d in [ppm]) relative to the shieldings of tetramethylsilane (TMS)
and PH3 as a secondary standard (dACHTNUGTRENNUNG
ly, obtained at the same computational level. The calculated NMR shifts
were averaged over magnetically equivalent nuclei.
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
(m), 522 (m), 544 (m), 573 (w), 602 (w), 610 (w), 661 (w), 683 (m), 692
(m), 707 (w), 744 (w), 756 (w), 768 (w), 775 (w), 799 (w), 981 (s), 1028
(w), 1057 (w), 1093 (s), 1059 (m), 1163 (w), 1180 (w), 1254 (w), 1275 (m),
1316 (m), 1392 (m), 1436 (m), 1464 (s), 1514 (m), 1532 (w), 1588 (w),
1643 (w), 2869 (w), 2927 (w), 2963 (m), 3059 cmꢀ1 (w); elemental analysis
calcd (%) for C107H89N2NiS2P2Pt [M+3C6H6]+: C 59.13, H 4.13, N 1.29;
found: C 59.34, H 4.34, N 1.57.
Atomic charges and spin densities were evaluated at the B3LYP**/TZVP
level of theory by means of natural population analyses (NPA), by using
the built-in NBO subroutines of the Gaussian 09 program.[35] Spin densi-
ties are provided either as isosurface plots or as NPA atomic values.
Mayer bond orders[36] were evaluated by using the program BORDER.[37]
The wave functions were also analyzed in the DGrid program[38] by
means of the electron localization function (ELF),[39] the electron localiz-
ability indicator based on the parallel-spin electron pair density
Reaction of 2 with dioxygen and the formation of 4: A solution of 2
(0.54 g, 0.43 mmol) in toluene (20 mL) was cooled to 08C. The N2 atmos-
phere in the flask was exchanged with dried oxygen. After stirring for
10 min, the reaction mixture was allowed to warm to room temperature
and stirred further for 1 h under the oxygen atmosphere. In the course of
stirring a pink precipitate of compound 4 was formed. After filtration,
the filtrate was concentrated and cooled to ꢀ208C to afford [LNiO2][2c]
as green crystals (0.15 g, 0.29 mmol, 67%). The pink precipitate of 4 was
collected and dried in vacuum (0.26 g, 0.33 mmol, 77%). Single crystals
of 4 were obtained from a toluene solution at 48C. Analytical data for
compound 4: M.p. 2028C (decomp.); 1H NMR (200.13 MHz, CDCl3,
258C): d=7.13–7.41 ppm (m, Ar-H); 13C{1H} NMR (100.61 MHz, CDCl3,
258C): d=127.9, 128.0, 128.1, 130.4, 133.9, 134.0, 134.2 ppm (Ar-C);
31P{1H} NMR (161.97 MHz, CDCl3, 258C): d=22.5 ppm (195Pt satellites,
AHCTUNGTRENNUNG
(ELI-D),[40] and the quantum theory of atoms in molecules (QTAIM).[41]
For this purpose, the Kohn–Sham orbitals of the single-point calculations
were transferred to the DGrid and the examined property was calculated
on a grid with ten points per Bohr. The results of ELF, ELI-D, and
QTAIM analyses were visualized by using the ParaView program.[42] The
visualization of molecules, spin density distributions, and orbitals was
done with Molekel[43] and gOpenMol.[44]
1J
ACHTUNGTRENNUNG(P,Pt)=3943 Hz); HR ESI-MS (ion spray voltage 5 kV, flow rate
5 mLminꢀ1, in THF): m/z calcd for C36H30S2P2Pt [M]+: 783.09064; found:
Acknowledgements
783.09346.
Computational methods: All structures were optimized at the RI-BP86/
def2-TZVP level of theory by using the TURBOMOLE program pack-
age.[17] That is, the generalized gradient approximation (GGA) of the
Becke exchange functional[18] and the Perdew correlation functional[19]
(BP86), in conjunction with standard TURBOMOLE all-electron def2-
TZVP basis sets[20] for atoms with Z<36 were used, together with an ap-
proximation of the Coulomb term by density fitting[21] with a def2-TZVP
auxiliary basis set. For the Pt atom a quasirelativistic small-core pseudo-
potential[22] (ECP) with the corresponding valence basis set[20] of the
The authors acknowledge support from the Cluster of Excellence “Unify-
ing Concepts in Catalysis” (http://www.unicat.tu-berlin.de) funded by the
German Research Foundation—Deutsche Forschungsgemeinschaft
(DFG) and the Senate of Science and Education of the State of Berlin
(Program “Nachhaltige Chemie”). P.H. also thanks the Alexander von
Humboldt Foundation and the Slovak grant agency VEGA (project No.
2/0079/09) for financial support.
quality (8s7p6d1f)/ACHTUNGTRENNUNG[6s4p3d1f] was employed. The unrestricted Kohn–
[1] a) S. Ferguson-Miller, G. T. Babock, Chem. Rev. 1996, 96, 2889;
b) E. Kim, M. E. Helton, I. M. Wasser, K. D. Karlin, S. Lu, H.-W.
Huang, P. Moenne-Loccoz, C. D. Incarvito, A. L. Rheingold, M.
Honecker, S. Kaderli, A. D. Zuberbühler, Proc. Natl. Acad. Sci.
USA 2003, 100, 3623; c) A. Decker, E. I. Solomon, Curr. Opin.
Chem. Biol. 2005, 9, 152; d) A. Müller, W. Jaegermann, J. H. Ene-
mark, Coord. Chem. Rev. 1982, 46, 245; e) T. Punniyamurthy, S. Ve-
lusamy, J. Iqbal, Chem. Rev. 2005, 105, 2329; f) B. Meunier, Biomi-
metic Oxidations Catalyzed by Transition Metals, Imperial College
Press, London 2000; g) M. R. Dubois, Chem. Rev. 1989, 89, 1; h) I.
Bar-Nahum, J. T. York, V. G. Young Jr. , W. B. Tolman, Angew.
Chem. 2008, 120, 543; Angew. Chem. Int. Ed. 2008, 47, 533; i) S.
Yao, M. Driess, Acc. Chem. Res. 2012, 45, 276.
[2] Selected superoxo complexes: a) C. J. Cramer, W. B. Tolman, K. H.
Theopold, A. L. Rheingold, Proc. Natl. Acad. Sci. USA 2003, 100,
3635, and cited references therein; b) M. T. Kieber-Emmons, C. G.
Riordan, Acc. Chem. Res. 2007, 40, 618; c) S. Yao, E. Bill, C. Mils-
mann, K. Wieghardt, M. Driess, Angew. Chem. 2008, 120, 7218;
Angew. Chem. Int. Ed. 2008, 47, 7110; d) J. Shearer, N. Zhao, Inorg.
Chem. 2006, 45, 9637; e) R. Huacuja, D. J. Graham, C. M. Fafard,
C.-H. Chen, B. M. Foxman, D. E. Herbert, G. Alliger, C. M.
Thomas, O. V. Ozerov, J. Am. Chem. Soc. 2011, 133, 3820.
Sham formalism was used for all open-shell species. The crystal structures
of complex 2 and the [(tmeda)3Cu3S2]3+ complex were used as initial
structures for full optimization. For comparison, partial optimization of
only the hydrogen-atom positions for the X-ray-based structure of 2 was
also done at the RI-BP86/def2-TZVP level (denoted as “crystal/opt-H”
in the Supporting Information).
g-Tensor calculations were performed in the MAG-ReSpect program[23]
at one-component second-order perturbation level[24] by using a common
gauge origin (see Tables S5 and S6 in the Supporting Information) and a
coupled-perturbed Kohn–Sham (CPKS) treatment for hybrid function-
als,[25] together with a small-core scalar relativistic pseudopotential (ECP)
for the Pt atom in the SCF step and a corresponding spin–orbit (SO)
ECP in the calculation of the SO matrix elements (related to Pt SO cou-
pling). For Ni and Zn we employed modified TZVP all-electron basis
sets of the (15s11p6d)/ACHTUNGTRENNUNG
[9s7p4d] quality,[26] whereas ligand atoms were
treated by Huzinaga–Kutzelnigg-type IGLO-II basis sets.[27] To assess the
effect of exact Hartree–Fock (HF) exchange admixture, several ex-
change-correlation functionals have been tested: BP86 with 0%, the
B3LYP** global hybrid with 10%, and B3LYP with 20% exact-exchange
admixture.[28] All SO matrix elements (except for the SO ECP on Pt)
have been obtained within the accurate and efficient atomic mean-field
approximation (AMFI)[29,30] to the full many-electron Breit–Pauli SO
operator.
[3] Selected supersulfido complexes: a) R. C. Elder, M. Trkula, Inorg.
Chem. 1977, 16, 1048; b) M. M. Taqui Khan, M. R. H. Siddiqui,
Inorg. Chem. 1991, 30, 1157; c) S. Yao, C. Milsmann, E. Bill, K. Wie-
ghardt, M. Driess, J. Am. Chem. Soc. 2008, 130, 13536; superseleni-
do complex: d) S. L. Yao, Y. Xiong, X. H. Zhang, M. Schlangen, H.
Schwarz, C. Milsmann, M. Driess, Angew. Chem. 2009, 121, 4621;
Angew. Chem. Int. Ed. 2009, 48, 4551.
The NMR shielding tensor calculations for paramagnetic systems were
carried out with an in-house PNMR program,[31] based on our recent ap-
proach, which also includes all relevant terms, going beyond the domi-
nant contact shifts.[10] For this purpose, the orbital term sorb and EPR hy-
perfine coupling tensors were computed in the Gaussian 09 program
package,[32] the former by an open-shell generalization of the usual CPKS
procedures for the diamagnetic case, by using gauge-including atomic or-
[4] a) E. C. Brown, J. T. York, W. E. Antholine, E. Ruiz, S. Alvarez,
W. B. Tolman, J. Am. Chem. Soc. 2005, 127, 13752; b) C. Mealli, A.
1252
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 1246 – 1253

C3ꢀ
A Heterobimetallic Approach to Stabilize S2
FULL PAPER
Ienco, A. Poduska, R. Hoffmann, Angew. Chem. 2008, 120, 2906;
Angew. Chem. Int. Ed. 2008, 47, 2864; c) S. Alvarez, R. Hoffmann,
C. Mealli, Chem. Eur. J. 2009, 15, 8358; d) J. F. Berry, Chem. Eur. J.
2010, 16, 2719; e) R. Sarangi, L. Yang, S. G. Winikoff, L. Gagliardi,
C. J. Cramer, W. B. Tolman, E. I. Solomon, J. Am. Chem. Soc. 2011,
133, 17180; f) S. Alvarez, E. Ruiz, Chem. Eur. J. 2010, 16, 2726.
[5] A. P. Cole, D. E. Root, P. Mukherjee, E. I. Solomon, T. D. P. Stack,
Science 1996, 273, 1848.
[6] a) H. Sitzmann, D. Saurenz, G. Wolmershauser, A. Klein, R. Boese,
Organometallics 2001, 20, 700; b) S. A. Yao, K. M. Lancaster, A. W.
Gçtz, S. DeBeer, J. F. Berry, Chem. Eur. J. 2012, 18, 9179.
[7] G. Bai, P. Wei, D. W. Stephan, Organometallics 2005, 24, 5901.
[8] a) J. Chatt, D. M. P. Mingos, J. Chem. Soc. A 1970, 1243; b) V. W.-W.
Yam, P. K.-Y. Yeung, K.-K. Cheung, J. Chem. Soc. Chem. Commun.
1995, 267; c) M. Capdevila, Y. Carrasco, W. Clegg, R. A. Coxall, P.
Gonzꢂlez-Duarte, A. Lledós, J. Sola, G. Ujaque, Chem. Commun.
1998, 597; d) A. Ienco, M. Caporali, F. Zanobini, C. Mealli, Inorg.
Chem. 2009, 48, 3840; e) V. M. Iluc, C. A. Laskowski, C. K. Brozek,
N. D. Harrold, G. L. Hillhouse, Inorg. Chem. 2010, 49, 6817.
[9] a) D. F. Evans, J. Chem. Soc. 1959, 2003; b) T. Ayers, R. Turk, C.
Lane, J. Goins, D. Jameson, S. J. Slattery, Inorg. Chim. Acta 2004,
357, 202.
[10] P. Hrobꢂrik, R. Reviakine, A. V. Arbuznikov, O. L. Malkina, V. G.
Malkin, F. H. Kçhler, M. Kaupp, J. Chem. Phys. 2007, 126, 024107.
[11] M. Reiher, O. Salomon, B. A. Hess, Theor. Chem. Acc. 2001, 107,
48.
[12] A. Poduska, R. Hoffmann, A. Ienco, C. Mealli, Chem. Asian J. 2009,
4, 302.
[13] A. Ishii, M. Murata, H. Oshida, K. Matsumoto, J. Nakayama, Eur. J.
Inorg. Chem. 2003, 3716.
[14] a) A. Nafady, T. T. Chin, W. E. Geiger, Organometallics 2006, 25,
1654; b) N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877.
[15] R. A. Head, Inorg. Synth. 1990, 28, 132.
Vol. 23 (Eds.: P. Diehl, E. Fluck, H. Günther, R. Kosfeld, J. Seelig),
Springer, Berlin/Heidelberg, 1991, pp. 165–262.
[28] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) C. T. Lee, W. T.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785; c) P. J. Stephens, F. J.
Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys. Chem. 1994, 98,
11623.
[29] B. A. Hess, C. M. Marian, U. Wahlgren, O. Gropen, Chem. Phys.
Lett. 1996, 251, 365.
[30] AMFI, Atomic Spin-Orbit Mean-Field Integral Program, B. Schim-
melpfennig, Stockholms Universitet, Stockholm (Sweden), 1996.
[31] PNMR, a computer program for the calculation of paramagnetic
NMR shielding tensors, version 1.0, P. Hrobarik, Technical Universi-
ty of Berlin (Germany), 2011.
[32] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,
Inc., Wallingford CT, 2009.
[33] K. Wolinski, J. F. Hinton, P. Pulay, J. Am. Chem. Soc. 1990, 112,
8251.
[34] C. J. Jameson, A. Dedios, A. K. Jameson, Chem. Phys. Lett. 1990,
167, 575.
[35] A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899.
[36] a) I. Mayer, Chem. Phys. Lett. 1983, 97, 270; b) I. Mayer, Chem.
Phys. Lett. 1985, 117, 396.
[37] BORDER, version 1.0, I. Mayer, Chemical Research Center, Hun-
garian Academy of Sciences, Budapest, 2005; available from http://
occam.chemres.hu/programs.
[16] SHELX-97, Program for Crystal Structure Determination, G. M.
Sheldrick, Universitꢁt Gçttingen (Germany), 1997.
[17] TURBOMOLE, version 6.3.1,
a Development of University of
Karlsruhe and Forschungszentrum Karlsruhe GmbH, 1989–2007,
TURBOMOLE GmBH since 2007; available from http://www.tur-
bomole.com.
[38] DGrid, version 4.6, M. Kohout, Radebeul, 2011.
[18] A. D. Becke, Phys. Rev. A 1988, 38, 3098.
[19] a) J. P. Perdew, Y. Wang, Phys. Rev. B 1986, 33, 8822; b) J. P. Perdew,
Y. Wang, Phys. Rev. B 1986, 34, 7406.
[39] a) A. D. Becke, K. E. Edgecombe, J. Chem. Phys. 1990, 92, 5397;
b) A. Savin, O. Jepsen, J. Flad, O. K. Andersen, H. Preuss, H. G.
von Schnering, Angew. Chem. 1992, 104, 186; Angew. Chem. Int. Ed.
Engl. 1992, 31, 187; c) M. Kohout, A. Savin, Int. J. Quantum Chem.
1996, 60, 875.
[40] a) M. Kohout, Int. J. Quantum Chem. 2004, 97, 651; b) M. Kohout,
K. Pernal, F. R. Wagner, Y. Grin, Theor. Chem. Acc. 2004, 112, 453.
[41] R. F. W. Bader, Atoms in Molecules: A Quantum Theory, Oxford
University Press, Oxford, 1990.
[20] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297.
[21] F. Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057.
[22] D. Andrae, U. Haussermann, M. Dolg, H. Stoll, H. Preuss, Theor.
Chim. Acta 1990, 77, 123.
[23] MAG-ReSpect, version 2.1, V. G. Malkin, O. L. Malkina, R. Revia-
kine, A. V. Arbuznikov, M. Kaupp, B. Schimmelpfennig, I. Malkin,
M. Repisky, S. Komorovsky, P. Hrobarik, E. Malkin, T. Helgaker, K.
Ruud, 2005.
[42] ParaView, version 3.8, Kitware Inc., Clifton Park, New York (USA),
2010; available from http://www.paraview.org.
[24] O. L. Malkina, J. Vaara, B. Schimmelpfennig, M. Munzarova, V. G.
Malkin, M. Kaupp, J. Am. Chem. Soc. 2000, 122, 9206.
[25] M. Kaupp, R. Reviakine, O. L. Malkina, A. Arbuznikov, B. Schim-
melpfennig, V. G. Malkin, J. Comput. Chem. 2002, 23, 794.
[26] M. Munzarovꢂ, M. Kaupp, J. Phys. Chem. A 1999, 103, 9966.
[27] W. Kutzelnigg, U. Fleischer, M. Schindler, The IGLO-Method: Ab
Initio Calculation and Interpretation of NMR Chemical Shifts and
Magnetic Susceptibilities in NMR Basic Principles and Progress,
[43] Molekel, version 5.4, U. Varetto, Swiss National Supercomputing
Centre, Manno (Switzerland), 2009; available from: http://mole-
kel.cscs.ch.
[44] gOpenMol, version 3.0, L. Laaksonen, CSC, Espoo (Finland), 2005;
available from http://www.csc.fi/english/pages/g0penMol.
Received: October 11, 2012
Published online: December 18, 2012
Chem. Eur. J. 2013, 19, 1246 – 1253
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1253