Co6H8(PiPr3)6: A Cobalt Octahedron with Face-Capping Hydrides

Article abstract of DOI:10.1002/anie.201608262

A square-planar Co₄amide cluster, Co₄{N(SiMe₃)₂}₄(2), and an octahedral Co₆hydride cluster, Co₆H₈(PⁱPr₃)₆(4), were obtained from metathesis-type amide to hydride exchange reactions of a Co^IIamide complex with pinacolborane (HBpin) in the absence/presence of PⁱPr₃. The crystal structure of 4 revealed face-capping hydrides on each triangular [Co₃] face, while the formal Co^II₂Co^I₄oxidation state of 4 indicated a reduction of the cobalt centers during the assembly process. Cluster 4 catalyzes the hydrosilylation of 2-cyclohexen-1-one favoring the conjugate reduction. Generation of the catalytically reactive Co cluster species was indicated by a trapping experiment with a chiral chelating agent.

Full text of DOI:10.1002/anie.201608262

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201608262
German Edition: DOI: 10.1002/ange.201608262
Metal–Hydride Clusters
Co6H8(PiPr3)6: A Cobalt Octahedron with Face-Capping Hydrides
Yasuhiro Ohki,* Yuki Shimizu, Ryoichi Araake, Mizuki Tada, W. M. C. Sameera, Jun-Ichi Ito,
and Hisao Nishiyama
Abstract: A square-planar Co4 amide cluster, Co4{N(SiMe3)2}4
(2), and an octahedral Co6 hydride cluster, Co6H8(PiPr3)6 (4),
were obtained from metathesis-type amide to hydride exchange
reactions of a CoII amide complex with pinacolborane
(HBpin) in the absence/presence of PiPr3. The crystal structure
of 4 revealed face-capping hydrides on each triangular [Co3]
FeMo-cofactor revealed the presence of Fe-bridging
hydrides,[4] and it was proposed that the subsequent uptake
of N2 should occur via concomitant liberation of H2.[5] Studies
on the synthesis and reactivity of new classes of such hydride-
supported transition metal clusters should thus expand their
scope of stoichiometric and catalytic applications of hydride
clusters.
face, while the formal CoII CoI4 oxidation state of 4 indicated
2
a reduction of the cobalt centers during the assembly process.
Cluster 4 catalyzes the hydrosilylation of 2-cyclohexen-1-one
favoring the conjugate reduction. Generation of the catalyti-
cally reactive Co cluster species was indicated by a trapping
experiment with a chiral chelating agent.
Typically, transition metal hydride complexes are
obtained from salt metathesis-type reactions of metal halides
with hydride reagents, from the addition of H2 or protons to
low-valent metal complexes, or from the b-elimination or
hydrogenolysis of metal alkyls.[6] Here, we examined less
common metathesis-type reactions of a CoII amide complex,
[Co{N(SiMe3)2}2]2 (1),[7] with pinacolborane (HBpin) in the
presence/absence of PiPr3. Replacement of a bulky N(SiMe3)2
moiety with a small hydride ligand upon treatment with
HBpin furnished low-coordinate cobalt–hydride species,
which should assemble through hydride (or amide) bridges
to provide cobalt clusters. In some cases, such assembly
processes are anticipated to accompany the reductive elim-
ination of H2 from intermediates carrying multiple hydrides
followed by a further assembly of cobalt; disproportionation
of CoII hydrides into CoI/CoIII species may also occur. We
have previously reported a similar reaction between Cp*FeN-
(SiMe3)2 and HBpin, which resulted in the concomitant
formation of an Fe-hydride species and (Me3Si)2N-Bpin.[8]
Other related approaches include the borylation of N2 on
a dinuclear Ta complex,[9] the in situ generation of Ca- and
Mn-hydrides from amide complexes and boranes,[10a] the
synthesis of s-block hydride clusters from the reactions of
amide/alkyl-supported Mg/alkali-metal complexes with
PhSiH3,[10b] and the synthesis of nanoparticles from metal–
amide complexes.[11] Herein, we describe the synthesis and
structural characterization of a new square-planar Co4 amide
cluster, Co4{N(SiMe3)2}4 (2), and of an octahedral Co6 hydride
cluster, Co6H8(PiPr3)6 (4). As far as catalytic applications are
concerned, we found that 4 is able to mediate the conjugate
hydrosilylation of 2-cyclohexen-1-one.
R
eactive molecular transition metal clusters are attractive
for synthetic chemists, as they may facilitate redox processes
and potentially allow the use of multiple metals in a reaction,
which can lead to an efficient activation of small molecules.[1]
For example, trinuclear hydride clusters of Ru and Ti with
auxiliary cyclopentadienyl ligands use all three metal centers
to activate C H, C C, Si H, and N N bonds,[2] while arene-
supported Ru hydride clusters have been applied as hydro-
genation catalysts for aromatic compounds.[3] In these cases,
hydride ligands protect the metal centers until the approach
of the substrates, and subsequently they are transferred to the
substrates or released in the form of H2. An analogous role for
hydrides has been proposed in the context of the FeMo-
cofactor, which is a naturally occurring metal–sulfur cluster
that catalyzes the biological reduction of atmospheric N2.
Spectroscopic studies of the active state (E4 state) of the
À
À
À
À
[*] Prof. Y. Ohki, Y. Shimizu, R. Araake, Prof. M. Tada
Department of Chemistry, Graduate School of Science
Nagoya University
Furo-cho, Chikusa-ku, Nagoya 464-8602 (Japan)
E-mail: ohki@chem.nagoya-u.ac.jp
Prof. Y. Ohki
PRESTO, Japan, Science and Technology Agency (JST)
4-1-8 Honcho, Kawaguchi, Saitama 332-0012 (Japan)
Treatment of 1 with HBpin (1 equiv with respect to Co) in
hexane resulted in the formation of a black solution, from
which the tetrameric CoI amide 2 could be isolated in 9%
yield (Scheme 1). The low isolated yield of 2 is partly due to
the formation of the Co7 amide/hydride cluster byproduct
Co7H6{N(SiMe3)2}6 (3), which was crystallographically iden-
tified but not isolated. Notably, a recent DFT evaluation
showed that the formation of 2 from Co{N(SiMe3)2}2 (the
monomeric form of 1) through putative homolysis of one of
Prof. M. Tada
Research Center for Materials Science (RCMS) & Integrated
Research Consortium on Chemical Sciences (IRCCS)
Nagoya University (Japan)
Prof. W. M. C. Sameera
Department of Chemistry, Graduate School of Science
Hokkaido University
Sapporo 060-0810 (Japan)
Prof. J.-I. Ito, Prof. H. Nishiyama
Department of Applied Chemistry, Graduate School of Engineering
Nagoya University (Japan)
À
the Co N bonds followed by the formation of HN(SiMe3)2 via
hydrogen abstraction from ether should be highly endergonic
(154 kJmolÀ1),[12] and the spontaneous decomposition of
1 into 2 should thus be severely hampered. Nevertheless,
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
1002/anie.201608262.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 1. ESI-MS spectra for 2 (left) and 4 (right) in THF.
Some transition metal hydride clusters undergo H/D
exchange between hydride ligands and deuterated solvents
[2a]
À
À
via reversible C H(D) bond cleavage. An analogous C
H(D) bond cleavage of the solvent (and PiPr3) was suggested
during the synthesis of 4. The m/z values of deuterated 4
observed in the ESI-MS increased in the order of samples
obtained a) in the presence of DBpin in toluene, b) in the
presence of HBpin in C6D6, and c) in the presence of DBpin
in C6D6 (Figure S6). These results indicate that the incorpo-
rated deuterium stems from both DBpin and C6D6. The
highest m/z value observed for sample (c) was even larger
than that calculated for Co6D8(PiPr3)3, suggesting additional
incorporation of deuterium into PiPr3. Conversely, the
isolated cluster 4 exhibited no evidence for H/D exchange
in C6D6 at room temperature, indicating that the H/D
exchange should occur in an intermediary cobalt hydride
species.
Scheme 1. Synthesis of clusters 2–4.
the substitution of one of the amide ligands in Co{N(SiMe3)2}2
by a hydride may offer a facile route to reduce CoII to CoI,
probably via the release of 0.5 equiv of H2 per cobalt atom
(Scheme 1). Similar to the synthesis of 2, the octahedral Co6
cluster 4 was obtained in the form of greenish black crystals in
42% yield from 1, PiPr3 (1 equiv with respect to Co), and
HBpin (2 equiv with respect to Co). The clusters 2 and 4 are
The molecular structures of clusters 2–4 were determined
by single-crystal X-ray diffraction analysis (Figure 2), even
though the quality of crystals of 2 was low (R1/wR2 = 0.11/
0.38). The solid-state structures revealed that each edge of the
[Co4] square in 2 is bridged by N(SiMe3)2 ligands, similar to
the corresponding nickel and copper congeners M4{N-
(SiMe3)2}4 (M = Ni, Cu).[12,13] These congeners share the
same space group (P2/n) and exhibit comparable crystal
parameters. The Co–Co distances in 2 (2.566(2)–2.591(3) ꢀ)
are longer than the corresponding distances in the Ni
congener (2.4328(4), 2.4347(5) ꢀ), but shorter than those in
the Cu congener (2.6770(7), 2.6937(7) ꢀ). The Co–N(bridge)
distances in 2 (1.938(7)–1.973(6) ꢀ) are longer than those in
the Ni (1.9127(2)–1.9189(2) ꢀ) and Cu (1.917(3)–1.925(4) ꢀ)
complexes. These Co–N(bridge) distances are shorter than
the corresponding distance in CoII complex 1 (2.062(4) ꢀ),
even though the ionic radius of CoI should be larger than that
of CoII.
The planar Co7 cluster 3 reveals a hexagonal arrangement
of six Co atoms around a central Co atom. Each Co3 triangle
around the central Co atom is capped by a m3-hydride ligand,
which faces in alternate directions across the Co3 triangles.
Although refinement of hydrides by X-ray crystallography is
not always reliable, these hydrides should coordinate effi-
ciently to the central cobalt atom, and the Co–(m3-H)
distances (1.69(3)–1.75(3) ꢀ) fall into the previously reported
range for cobalt clusters with triply-bridging hydrides (1.56-
(4)–1.82(2) ꢀ).[14] Moreover, a planar Rh7 hydride cluster
[Rh7(PiPr3)6H18]2+, containing an analogous [Rh7(m3-H)6]
1
paramagnetic, and exhibited broad H NMR signals in C6D6
at d = À21.8 and 20.8 ppm (SiMe3 in 2), as well as at d = À0.9
and 149.9 ppm (iPr in 4) (Figures S1 and S2 in the Supporting
Information). However, the appearance of two SiMe3 signals
for 2 is inconsistent with the solid-state structure (see below)
wherein all cobalt and nitrogen atoms adopt a coplanar
arrangement, indicating a slightly different solution structure.
Variable-temperature 1H NMR measurements of
2 in
[D8]toluene did not show any coalescense of the two SiMe3
signals but led to decomposition at around 1008C (Figure S3).
Although an NMR assignment of the hydrides in 4 was
unsuccessful, the chemical formulas of clusters 2 and 4 are
supported by their electro-spray ionization mass (ESI-MS)
spectra, wherein signals were observed at m/z = 876.0 ([2]+)
and 1322.5 ([4]+) (Figure 1). The cationic species [2]+ and [4]+
were probably generated in the spectrometer upon exposure
to the applied potential. The tentative facile electronic
oxidation of 4 is supported by its cyclic voltammogram in
THF, which revealed, besides the rest potential, a redox
couple for quasi-reversible 1e oxidation at E1/2 = À1.53 V (vs.
Fc/Fc+; Figure S8), while the cyclic voltammogram of 2
indicated a quasi-reversible reduction process at E1/2
=
À2.29 V (Figure S7). In addition to these quasi-reversible
processes, other irreversible oxidation and reduction process-
es were found for both 2 and 4.
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 2. Molecular structures of 2–4 with atomic displacement parameters set at the 50% probability. Only selected atoms are labeled, and
hydrogen atoms, except for the hydride ligands, are omitted for clarity. Selected bond distances (ꢀ) and angles (8) for 2: Co–Co 2.566(2)–2.591(3),
Co–N 1.938(7)–1.973(6); Co-N-Co 81.7(3)–83.5(4); 3: Co–Co 2.5231(2)–2.5288(3), Co–N 2.0476(17)–2.0703(18), Co–H 1.69(3)–1.75(3); Co-N-Co
75.72(6)–75.60(6); 4: Co–Co 2.4667(6)–2.4684(5), Co–P 2.3173(5), Co–H 1.810(17)–1.86(3).
core,[15] and a planar Co7 cluster supported by pyridinyl-
methoxide[16] have previously been reported. Based on the
molecular structure, the oxidation state of the cobalt atoms
2, we chose the S = 4 state, as it possesses an even number of
d-electrons, while the S = 7 state was applied for the
calculation of 4. The fully optimized structures of 2 and 4
are in good agreement with their X-ray structures (Fig-
ures S15 and S16). The unpaired electrons in 2 are mainly
localized on Co, affording a spin density of 7.8 mB for all Co
atoms. In contrast, 4 reveals a significant spin delocalization
over the P atoms and the hydrides, resulting in a calculated
spin density of 11.8 mB for all Co atoms. To get some insights
for the metal–metal bonding, natural bond orbital (NBO)
analysis was performed for 2 and 4. The Co–Co NBOs
between the neighboring Co atoms were found for both Co4
and Co6 systems (see the Supporting Information), while
assignment of the exact bond order numbers was difficult with
complex data from open-shell systems. Also importantly, no
should be assigned either as CoII5CoI2 or as CoII Co0.
6
Considering the observed narrow range of Co–Co distances
(2.5231(2)–2.5288(3) ꢀ), we tentatively assigned an averaged
oxidation state, Co(+ 1.71).
In cluster 4, the [Co6] core adopts an octahedral geometry
with face-capping m3-hydrides. The presence of eight hydrides
is in agreement with the observed ESI-MS signals, while the
Co–(m3-H) distances in 4 (1.810(17)–1.86(3) ꢀ) are slightly
longer than the corresponding distances in previously
reported Co clusters.[14] Even though the formal oxidation
state of 4 is CoII CoI4, the six cobalt atoms are crystallo-
2
graphically equivalent and the Co–Co distances (2.4667(6)–
2.4684(5) ꢀ) are also almost identical, which suggests an
averaged oxidation state of Co(+ 1.33) for each cobalt atom.
The Co–P distances in 4 (2.3173(5) ꢀ) are longer than those
reported for phosphine-supported hydride complexes of CoII
and CoI (ꢀ 2.291(3) ꢀ),[17] indicating steric repulsion between
the PiPr3 ligands. Interestingly, 4 exhibits a structural analogy
to Strykerꢁs reagent, Cu6H6(PPh3)6, which mediates chemo-
selective conjugate reductions of a,b-unsaturated carbonyl
compounds.[18,19] The crystallographically analyzed analogues
Cu6H6(PR3)6 (R = for example, NMe2, p-tolyl, p-anisyl) con-
tain face-capping hydrides on six of the eight [Cu3] faces.[20] In
contrast to the virtually identical Co–Co distances determined
for 4, the Cu6H6(PR3)6 copper clusters exhibited short (av.
2.47–2.52 ꢀ; Cu3(m3-H) faces) and long (av. 2.68–2.80 ꢀ; Cu3
without hydride faces) Cu–Cu distances. The Co–P distances
in 4 are longer than the Cu–P distances in Cu6H6(PR3)6
(2.196(1)–2.315(2) ꢀ).
À
diagonal Co Co bond was found in the analysis.
The aforementioned analogy between cluster 4 and
Strykerꢁs reagent prompted us to examine the potential of 4
for the catalytic hydrosilylation of 2-cyclohexen-1-one
(Table 1). Addition of 150 equiv of 2-cyclohexen-1-one and
PhSiH3 to a benzene solution of 4 resulted mainly in the
conjugate reduction. The subsequent workup afforded cyclo-
Table 1: Reduction of 2-cyclohexen-1-one catalyzed by cobalt clusters.[a]
Entry
Catalyst
sub:Si:cat[b]
Product yields [%][c]
a
b
c
In solution at 298 K, the magnetic moments of 2 (10.0-
(5) mB) and 4 (14.5(7) mB) are close to the spin-only S = 4.5
(9.9 mB) and S = 7 (15.0 mB) states. In order to assess their
magnetism and the Co–Co interactions, theoretical analyses
of these clusters were carried out by density functional theory
(DFT), using the solution magnetism for selecting their spin
states (see the Supporting Information). For the calculation of
1
2
3
4
1
2
150:150:1
50:50:1
100:100:1
83
9
12
4
38
78
11
8
8
[a] TBAF=tetrabutylammonium fluoride. [a] sub=cyclohexenone,
Si=PhSiH3, cat=catalyst. The substrate/Co-atom ratio was 25:1 in all
entries. [b] Yields were estimated by GC analyses.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
hexanone (a) in 83% yield, 2-cyclohexen-1-ol (b) in 4% yield,
and cylcohexanol (c) in 11% yield (entry 1); the chemo-
selectivity was found to vary with the silanes (Table S1). A
similar silane-dependent selectivity was observed for the
conjugate reduction of 1-phenyl-l-buten-3-one mediated by
CuI fluoride.[21] The chemoselectivity decreased in THF
(Table S3), which may be due to the slow and partial
degradation of 4 through the coordination of THF. The
corresponding hydrosilylation catalyzed by CoI amide cluster
2 using the same total amount of cobalt favored the 1,2-
reduction, affording a, b, and c in 12%, 78%, and 8% yields,
respectively (entry 3).
component(s). However, at this stage, the results remain
preliminary and require further investigations.
In summary, we synthesized the reactive cobalt clusters
Co4{N(SiMe3)2}4 (2) and Co6H8(PiPr3)6 (4) through metathesis
reactions of [Co{N(SiMe3)2}2]2 (1) with HBpin in the presence
or absence, respectively, of PiPr3. We demonstrated that the
obtained octahedral Co6–hydride cluster is able to catalyze
the hydrosilylation of 2-cyclohexen-1-one, and that the Co
cluster(s) most likely represent the predominant catalytic
component(s).
In the presence of additives such as chiral diphosphines,
fragmentation of Strykerꢁs reagent into catalytically active
monomeric copper–hydride species has been proposed,[22] and
the reaction of Cu6H6(PPh3)6 with PPh3 and Ph3SiH was
reported to furnish the monomeric complex (Ph3P)3Cu(SiPh3)
in 82% yield.[23] In light of the suggested fragmentation of
Cu6H6(PPh3)6, we tried to gain further insight into the
possibility of generating monomeric cobalt species from 4.
As the detection of cobalt species in the catalytic reaction
mixture by ESI-MS was unsuccessful, we attempted a trapping
experiment with a chelating agent. Nishiyama et al. reported
that a catalytic amount (5 mol%) of cobalt complex bearing
a chiral N,N,N-bis(oxazolinylphenyl)amine ligand (Bopa)
mediates the enantioselective hydrosilylation of 4-
nBuC6H4COMe with (EtO)2MeSiH at 408C (48 h) in THF
to afford the reduced product in 99% yield and 98% ee.[24] We
used this reaction as a reference, and observed that 4
catalyzed the same hydrosilylation in the presence of
6 equiv Bopa (1 equiv with respect to Co), when using the
same total amount of cobalt as in the reference reaction
(Scheme 2). In the presence of Bopa, the reduction product
was obtained in 98% yield and 28% ee, while the racemic
product was obtained in the absence of Bopa (84% yield).
Although the Bopa ligand improved the product yield by
14% and induced some enantioselectivity, the low ee (28%)
suggests a limited contribution of monomeric cobalt species
to the catalysis. Possible partial degradation of 4 in THF may
also contribute to the generation of a part of Bopa-supported
Co species. Thus, the results shown in Scheme 2 indicate that
Co cluster(s) are the predominant catalytically active
Acknowledgements
This research was financially supported by the PRESTO
program of the Japan Science and Technology Agency and
grants-in-aid for Scientific Research (No. 15H00936,
15K13655, and 16H04116) from the Japanese Ministry of
Education, Culture, Sports, Science and Technology (MEXT).
We are grateful to Prof. Kazuyuki Tatsumi for providing
access to instruments, to Prof. Satoshi Muratsugu for gen-
erous advice on electrochemical measurements and GC
analyses, as well as to Prof. Yusuke Sunada for fruitful
discussions.
Keywords: clusters · cobalt · hydride · hydrosilylation
[1] For examples, see: a) Metal Clusters in Chemistry (Eds.: P.
Braunstein, L. A. Oro, P. R. Raithby), Wiley, Chichester, 1999;
b) R. D. Adams, J. Organomet. Chem. 2000, 600, 1 – 6; c) P. J.
Dyson, Coord. Chem. Rev. 2004, 248, 2443 – 2458; d) S. K.
Brayshaw, M. J. Ingleson, J. C. Green, J. S. Mclndoe, P. R.
Raithby, G. Kociok-Kçhn, A. S. Weller, J. Am. Chem. Soc.
2006, 128, 6247 – 6263; e) G. Hogarth, S. E. Kabir, E. Nord-
lander, Dalton Trans. 2010, 39, 6153 – 6174; f) T. M. Powers, T. A.
Betley, J. Am. Chem. Soc. 2013, 135, 12289 – 12296.
[2] a) H. Suzuki, Eur. J. Inorg. Chem. 2002, 1009 – 1023, and
references therein; b) T. Shima, S. Hu, G. Luo, X. Kang, Y.
Luo, Z. Hou, Science 2013, 340, 1549 – 1552; c) S. Hu, T. Shima,
Z. Hou, Nature 2014, 512, 413 – 415.
[3] G. Sꢂss-Fink in Synthetic Methods of Organometallic and
Inorganic Chemistry, Vol. 10 (Ed.: W. A. Herrmann), Georg
Thieme, 2001.
[4] D. Lukoyanov, Z.-Y. Yang, D. R. Dean, L. C. Seefeldt, B. M.
Hoffman, J. Am. Chem. Soc. 2010, 132, 2526 – 2527.
[5] a) Z.-Y. Yang, N. Khadka, D. Lukoyanov, B. M. Hoffman, D. R.
Dean, L. C. Seefeldt, Proc. Natl. Acad. Sci. USA 2013, 110,
16327 – 16332; b) D. Lukoyanov, Z.-Y. Yang, N. Khadka, D. R.
Dean, L. C. Seefeldt, B. M. Hoffman, J. Am. Chem. Soc. 2015,
137, 3610 – 3615.
[6] R. G. Crabtree in Organometallic Chemistry of The Transition
Metals, 6th ed., Wiley, Chichester, 2014, Chap. 3.
[7] a) B. D. Murray, P. P. Power, Inorg. Chem. 1984, 23, 4584 – 4588;
b) A. M. Bryan, G. J. Long, F. Grandjean, P. P. Power, Inorg.
Chem. 2013, 52, 12152 – 12160.
[8] Y. Ohki, Y. Takikawa, T. Hatanaka, K. Tatsumi, Organometallics
2006, 25, 3111 – 3113.
[9] M. D. Fryzuk, B. A. MacKay, S. A. Johnson, B. O. Patrick,
Angew. Chem. Int. Ed. 2002, 41, 3709 – 3712; Angew. Chem.
2002, 114, 3861 – 3864.
Scheme 2. Attempted trapping of catalytically active monomeric Co
species.
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[10] a) D. J. Liptrot, M. S. Hill, M. F. Mahon, A. S. S. Wilson, Angew.
Speldrich, J. van Leusen, P. Kçgerler, P. Braunstein, J. M. Poblet,
Chem. Int. Ed. 2015, 54, 13362 – 13365; Angew. Chem. 2015, 127,
13560 – 13563; b) D. J. Liptrot, M. S. Hill, M. F. Mahon, Chem.
Eur. J. 2014, 20, 9871 – 9874.
Chem. Eur. J. 2014, 20, 3769 – 3781.
[17] The Cambridge Structural Database lists the longest Co – P
distance for this class of compounds for the CoII complex
(Ph3P)3CoH(NCBH3); see; R. J. Barton, D. G. Holah, H.
Shengzhi, A. N. Hughes, S. I. Khan, B. E. Robertson, Inorg.
Chem. 1984, 23, 2391 – 2395.
[11] B. Cormary, F. Dumestre, N. Liakakos, K. Soulantica, B.
Chaudret, Dalton Trans. 2013, 42, 12546 – 12553, and references
therein.
[12] M. Faust, A. M. Bryan, A. Mansikkamꢃki, P. Basko, M. M.
Olmstead, H. M. Tuononen, F. Grandjean, G. J. Long, P. P.
Power, Angew. Chem. Int. Ed. 2015, 54, 12914 – 12917; Angew.
Chem. 2015, 127, 13106 – 13109.
[18] For seminal studies, see: a) W. S. Mahoney, D. M. Brestensky,
J. M. Stryker, J. Am. Chem. Soc. 1988, 110, 291 – 293; b) W. S.
Mahoney, J. M. Stryker, J. Am. Chem. Soc. 1989, 111, 8818 –
8823.
[13] A. M. James, R. K. Laxman, F. R. Fronczek, A. W. Maverick,
Inorg. Chem. 1998, 37, 3785 – 3791.
[19] For reviews, see: a) S. Rendler, M. Oestreich, Angew. Chem. Int.
Ed. 2007, 46, 498 – 504; Angew. Chem. 2007, 119, 504 – 510; b) C.
Deutsch, N. Krause, B. H. Lipshutz, Chem. Rev. 2008, 108, 2916 –
2927.
[20] For representative structural analyses, see: a) S. A. Bezman,
M. R. Churchill, J. A. Osborn, J. Wormald, J. Am. Chem. Soc.
1971, 93, 2063 – 2065; b) T. H. Lemmen, K. Folting, J. C. Huff-
man, K. G. Caulton, J. Am. Chem. Soc. 1985, 107, 7774 – 7775;
c) R. C. Stevens, M. R. McLean, R. Bau, T. F. Koetzle, J. Am.
Chem. Soc. 1989, 111, 3472 – 3473; d) M. S. Eberhart, J. R.
Norton, A. Zuzek, W. Sattler, S. Ruccolo, J. Am. Chem. Soc.
2013, 135, 17262 – 17265.
[21] A. Mori, A. Fujita, H. Kajiro, Y. Nishihara, T. Hiyama,
Tetrahedron 1999, 55, 4573 – 4582.
[22] B. H. Lipshutz, K. Noson, W. Chrisman, A. Lower, J. Am. Chem.
Soc. 2003, 125, 8779 – 8789.
[23] U. Schubert, B. Mayer, C. Ruß, Chem. Ber. 1994, 127, 2189 –
2190.
[24] T. Inagaki, L. T. Phong, A. Furutta, J.-I. Ito, H. Nishiyama,
Chem. Eur. J. 2010, 16, 3090 – 3096.
[14] a) E. Iiskola, T. Pakkanen, Acta Chem. Scand. Ser. A 1984, 38,
731 – 734; b) S. Rossi, J. Pursiainen, M. Ahlgren, T. A. Pakkanen,
Organometallics 1990, 9, 475 – 479; c) P. Braunstein, J. Rosꢄ, P.
Granger, J. Raya, S.-E. Bouaoud, D. Grandjean, Organometal-
lics 1991, 10, 3686 – 3693; d) P. Braunstein, L. Mourey, J. Rosꢄ, P.
Granger, T. Richert, F. Balegroune, D. Grandjean, Organo-
metallics 1992, 11, 2628 – 2634; e) H. Wadepohl, T. Borchert, K.
Bꢂchner, H. Pritzkow, Chem. Ber. 1993, 126, 1615 – 1618; f) P.
Braunstein, J. Rosꢄ, D. Toussaint, S. Jꢃꢃskelꢃinen, M. Ahlgren,
T. A. Pakkanen, J. Pursiainen, L. Toupet, D. Grandjean,
Organometallics 1994, 13, 2472 – 2479; g) H. Wadepohl, M. J.
Calhorda, M. Herrmann, C. Jost, P. E. M. Lopes, H. Pritzkow,
Organometallics 1996, 15, 5622 – 5634; h) H. Wadepohl, T.
Borchert, H. Pritzkow, Chem. Ber. 1997, 130, 593 – 603; i) H.-F.
Klein, M. Mager, A. Schmidt, M. Hꢂber, W. Haase, U. Flçke, H.-
J. Haupt, R. Boca, Inorg. Chem. 1997, 36, 4303 – 4306; j) R. Bau,
N. N. Ho, J. J. Schneider, S. A. Mason, G. J. Mclntyre, Inorg.
Chem. 2004, 43, 555 – 558.
[15] S. K. Brayshaw, J. C. Green, R. Edge, E. J. L. McInnes, P. R.
Raithby, J. E. Warren, A. S. Weller, Angew. Chem. Int. Ed. 2007,
46, 7844 – 7848; Angew. Chem. 2007, 119, 7990 – 7994.
[16] a) R. Pattacini, P. Teo, J. Zhang, Y. Lan, A. K. Powell, J.
Nehrkorn, O. Waldmann, T. S. A. Hor, P. Braunstein, Dalton
Trans. 2011, 40, 10526 – 10534; b) K. Y. Monakhov, X. Lꢅpez, M.
Received: August 24, 2016
Revised: October 11, 2016
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Metal–Hydride Clusters
Reactive cobalt clusters: The octahedral
Co6–hydride cluster Co6H8(PiPr3)6 and the
square-planar Co4 cluster Co4{N-
(SiMe3)2}4 were synthesized from the
reactions of a CoII amide complex with
pinacolborane in the presence and
absence of PiPr3, respectively. The Co6–
hydride cluster catalyzes the conjugate
hydrosilylation of 2-cyclohexen-1-one.
Y. Ohki,* Y. Shimizu, R. Araake, M. Tada,
W. M. C. Sameera, J.-I. Ito,
H. Nishiyama
&&&& — &&&&
Co6H8(PiPr3)6: A Cobalt Octahedron with
Face-Capping Hydrides
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!