Low coordinate germanium(II) and tin(II) hydride complexes: Efficient catalysts for the hydroboration of carbonyl compounds

Article abstract of DOI:10.1021/ja5006477

This study details the first use of well-defined low-valent p-block metal hydrides as catalysts in organic synthesis. That is, the bulky, two-coordinate germanium(II) and tin(II) hydride complexes, L^?(H)M: (M = Ge or Sn, L^? = -N(Ar^?)(SiPrⁱ ₃), Ar^? = C₆H₂{C(H)Ph ₂}₂Prⁱ-2,6,4), are shown to act as efficient catalysts for the hydroboration (with HBpin, pin = pinacolato) of a variety of unactivated, and sometimes very bulky, carbonyl compounds. Catalyst loadings as low as 0.05 mol % are required to achieve quantitative conversions, with turnover frequencies in excess of 13 300 h^-1 in some cases. This activity rivals that of currently available catalysts used for such reactions.

Full text of DOI:10.1021/ja5006477

Communication
pubs.acs.org/JACS
Low Coordinate Germanium(II) and Tin(II) Hydride Complexes:
Efficient Catalysts for the Hydroboration of Carbonyl Compounds
Terrance J. Hadlington,† Markus Hermann,‡ Gernot Frenking,*,‡ and Cameron Jones*,†
†School of Chemistry, Monash University, P.O. Box 23, Melbourne, VIC, 3800, Australia
‡Fachbereich Chemie, Philipps-Universitat Marburg, 35032, Marburg, Germany
̈
S
* Supporting Information
variety of aldehydes and ketones with the mild borane reagent,
ABSTRACT: This study details the first use of well-
defined low-valent p-block metal hydrides as catalysts in
organic synthesis. That is, the bulky, two-coordinate
germanium(II) and tin(II) hydride complexes, L†(H)M:
(M = Ge or Sn, L† = −N(Ar†)(SiPri3), Ar† = C6H2{C-
(H)Ph2}2Pri-2,6,4), are shown to act as efficient catalysts
for the hydroboration (with HBpin, pin = pinacolato) of a
variety of unactivated, and sometimes very bulky, carbonyl
compounds. Catalyst loadings as low as 0.05 mol % are
required to achieve quantitative conversions, with turnover
frequencies in excess of 13 300 h−1 in some cases. This
activity rivals that of currently available catalysts used for
such reactions.
HBpin (pin = pinacolato), was chosen. This choice was made,
first because the selective and efficient conversion of carbonyl
compounds to alcohols is of great importance to organic
synthesis.4 Moreover, it would allow for comparisons to be
drawn with the many transition metal (e.g., RhI,5 RuII,6 MoIV,7
TiII/IV8) complexes, and the small number of “normal oxidation
state” main group systems (e.g., {(DipNacnac)MgH}2, DipNacnac
= [(DipNCMe)2CH]−, Dip = C6H3Pri2-2,6),9 that are known to
effect hydroborations of carbonyl compounds under mild
conditions. Another reason for the selection of carbonyl
substrates in the present study was that a recent computational
investigation predicted that related hydroelementations of the
activated ketone, OC(Ph)(CF3), catalyzed by the three-
coordinate germanium(II) hydride, (DipNacnac)GeH, should
be both thermodynamically and kinetically viable.10 Here, we
show that the markedly more reactive two-coordinate group 14
metal(II) hydrides, 1 and 2, can efficiently and selectively
catalyze the hydroboration of even unactivated and bulky
ketones/aldehydes (Chart 1). Although the intermediacy of
tin(II) hydrides in carbonyl hydroelementations has been
previously proposed, this work represents the first use of well-
defined low-valent p-block metal hydride species as catalysts in
organic synthesis.11
At the outset of this study, the stoichiometric reactions of 1
and 2 with either HBpin or the bulky carbonyl substrates, O
C(Pri)2 and OC(H)(PhOMe-4), were carried out in order to
assess the viability of the proposed catalytic protocols.12 No
noticeable reactions were observed between the metal(II)
hydrides and HBpin, as evidenced by the fact that the
characteristic low field hydride resonances (1H NMR) of 1 and
2 shifted negligibly upon addition of HBpin to their solutions. In
contrast, the reactions with 1 equiv of the carbonyl substrates
were rapid at ambient temperature and afforded the hydro-
metalation products, L†MOC(H)(Pri)2 (M = Ge 3, Sn 4) and
L†MOC(H)2(PhOMe-4) (M = Ge 5, Sn 6), in a matter of
seconds, and in essentially quantitative yields (as determined by
1H NMR spectroscopic analyses of the reaction mixtures). All of
3−6 were crystallographically characterized (see Figure 1 for
molecular structures of 3 and 5) and were found to be
monomeric in the solid state.
he advances that have been made in the chemistry of low
Toxidation state p-block compounds over the past several
decades have been remarkable. More than being mere chemical
curiosities, it has begun to be realized that the low coordination
numbers and electronic structure of many of these compounds
can lead to them displaying reactivity that closely mimics that of
transition metal complexes. This is especially so for the facile
activation of catalytically relevant small molecules (e.g., H2, NH3,
CO2, alkenes, ketones, etc.), the regular demonstration of which
has led to a recent drive to develop low-valent p-block
compounds as replacements for transition metal based catalysts
in a variety of synthetic processes.1 Although efficient “transition
metal-like” catalysis involving well-defined low-valent p-block
metal systems is still essentially unknown,2 any progress in this
field would offer the enticing possibility of substituting often
expensive, toxic, and increasingly scarce d-block metals with
cheaper, less toxic, and earth abundant p-block metals in both the
academic and industrial settings.
Recently we reported that the amido-digermyne, L†GeGeL†
(L† = −N(Ar†)(SiPri3), Ar† = C6H2{C(H)Ph2}2Pri-2,6,4),
activates dihydrogen at temperatures as low as −10 °C to
quantitatively yield the hydrido-digermene, L†(H)GeGe(H)-
L†.3 Furthermore, because of the extreme steric bulk of its amide
ligand, this compound, and its isomeric tin analogue, L†Sn(μ-
H)2SnL†, were shown to exist in equilibrium with significant
amounts of the unprecedented monomeric, two-coordinate
metal(II) hydrides, L†(H)M: (M = Ge 1 or Sn 2), in solution.3
Given the unsaturated nature of hydridic 1 and 2 (they possess an
empty p-orbital at M), it seemed plausible that they could act as
catalysts for the hydroelementation of unsaturated substrates. In
order to test this hypothesis, the catalytic hydroboration of a
The reactivity of 1 and 2 can be compared to that of the only
other group 14 metal(II) hydrides, (DipNacnac)MH (M = Ge or
Sn), that are known to hydrometallate ketones.13,14 While 1 and
Received: January 21, 2014
Published: February 13, 2014
© 2014 American Chemical Society
3028
dx.doi.org/10.1021/ja5006477 | J. Am. Chem. Soc. 2014, 136, 3028−3031

Journal of the American Chemical Society
Communication
Chart 1. Catalysts Used in This Study
Table 2. Hydroborations of Ketones, (R1)(R2)CO, Catalyzed
by 1 or 2 (see Scheme 1)
loading
(mol %)
time
b
yield
c
TOF
d
a
cat.
R1
R2
(h)
(%)
(h−1
)
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
0.5
0.5
0.5
0.5
0.5
0.5
1
Ph
Ph
CF3
Ph
<0.15
2.5
>99
95
1330
80
PhOMe
PhEt-4
Ph
Me
0.25
0.25
0.33
1.75
0.5
98
800
800
600
115
200
400
80
Me
>99
96
e
C(O)Ph
f
Cy
96
g
CyMe-2
99
eh
,
0.5
0.5
2
2-cyclohexene
0.5
>99
95
i
Ad
2.5
Pri
Ph
Pri
24
95
1.7
160
1.7
30
2.5
1.25
2.5
2.5
5
CF3
Ph
0.25
48
>99
>99
>99
>99
94
Ph
Figure 1. Thermal ellipsoid plot (20% probability surface) of (a)
L†GeOC(H)(Pri)2 3 and (b) L†GeOC(H)2(PhOMe-4) 5 (hydrogen
atoms omitted). Selected bond lengths (Å) and angles (deg) for 3/5:
Ge(1)−N(1) 1.883(3)/1.877(2), Ge(1)−O(1) 1.797(3)/1.8120(19),
N(1)−Ge(1)−O(1) 99.44(14)/97.49(9).
PhOMe
PhEt-4
Ph
Me
1.33
1
Me
40
e
C(O)Ph
0.4
50
f
5
Cy
<0.15
0.5
>99
>99
>99
>99
80
130
40
g
5
CyMe-2
eh
,
2.5
1.25
1.25
2-cyclohexene
1
40
Table 1. Hydroborations of Aldehydes, RC(H)O, Catalyzed
by 1 or 2 (see Scheme 1)
i
Ad
4
20
Pri
Pri
168
0.47
b
a
b
c
d
cat.
loading (mol %)
R
time (h)
yield (%) TOF (h−1
)
a
Catalyst 2 generated in situ using the precatalyst, L†SnOBut. All
reactions carried out in d6-benzene at 20 °C using 1 equiv of HBpin
(unless stated otherwise). Obtained by integration of R2CHOBpin
signal against tetramethylsilane internal standard. Turnover frequency
- average value for complete reaction. 2 equiv of HBpin.
Cyclohexanone. 2-Methylcyclohexanone. 2-Cyclohexen-1-one.
Adamantanone.
2
2
2
2
2
2
1
1
1
1
1
1
0.05
0.05
0.05
0.05
0.05
0.05
1
Ph
2.5
4.5
5
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
800
c
PhBr-4
PhOMe
Et
450
d
400
e
e
<0.15
<0.15
<0.15
1.5
4
>13 300
>13 300
>13 300
67
f
g
h
e
e
Pri
i
Cy
Ph
Scheme 1. Proposed Cycle for the Hydroboration of Carbonyl
Compounds, (R1)(R2)CO (R1/R2 = Alkyl, Aryl or H),
Catalyzed by L†MH (M = Ge or Sn)
1
PhBr-4
PhOMe
Et
25
1
6
17
0.05
0.05
0.05
1
2000
5000
Pri
0.4
0.33
Cy
6000
a
b
Catalyst 2 generated in situ using the precatalyst, L†SnOBut. All
reactions carried out in d6-benzene at 20 °C using 1 equiv of HBpin.
c
Obtained by integration of RCH2OBpin signal against tetramethylsi-
d
lane internal standard. Turnover frequency - average value for
complete reaction. TOF lower limit.
e
2 rapidly hydrometallate the bulky, unactivated ketone, O
C(Pri)2, at ambient temperature, (DipNacnac)GeH reacts only
slowly (over 12 h) with activated ketones (e.g., OC(Ph)-
(CF3)),13 and forcing conditions (110 °C, 12 h) are required for
(see Supporting Information (SI)) was used for the tin based
studies. Prior to the catalytic runs, it was confirmed that this does
not itself react with the carbonyl substrates, but does react rapidly
and cleanly with excess HBpin to generate 2 in situ. One other
advantage of the use of L†SnOBut is that it is considerably more
moisture and oxygen tolerant than 2. With regard to the
hydroboration of aldehydes (see Table 1), (pre)catalyst loadings
of only 0.05 mol % were typically required to generate the borate
ester product, essentially quantitatively, and often in under 1 h.
As the experiments were carried out in NMR tubes and
(
DipNacnac)SnH to hydrostannylate the bulky aliphatic ketone,
OC(cyclopropyl)2.14 Undoubtedly, the empty p-orbital
available at the metal centers of 1 and 2 give rise to their
markedly enhanced reactivity, relative to intramolecularly base
stabilized (DipNacnac)MH. The two-coordinate nature of 3−6 is
also the most likely reason why they subsequently react cleanly
with 1 equiv of HBpin at ambient temperature (though at slower
rates than the hydrometalation reactions) to give the borate
esters, pinBOC(H)(Pri)2 and pinBOC(H)2(PhOMe-4)2, and to
regenerate the metal hydrides, 1 and 2. Furthermore, the
generated borate esters are unreactive toward regenerated 1 or 2.
These results gave a strong indication that 1 and 2 could act as
efficient catalysts for the hydroboration of carbonyl compounds.
This was systematically explored, though given the fact that 2
slowly decomposes in solution at ambient temperature (over 2
days),3a the thermally stable, monomeric precatalyst, L†SnOBut
1
monitored by H NMR spectroscopy, a number of the tin
catalyzed reactions (i.e., those marked time <0.15 h) were
complete before the reaction mixture could be analyzed. Indeed,
a qualitative visual assessment of all such reactions suggested that
they were complete in <1 min and, therefore, that their turnover
frequencies are likely to be considerably higher than the already
impressive quoted lower limit of 13 300 h−1.
3029
dx.doi.org/10.1021/ja5006477 | J. Am. Chem. Soc. 2014, 136, 3028−3031

Journal of the American Chemical Society
Communication
Figure 2. Calculated (BP86+D(BJ)/def2-TZVPP) free energy profile for the reaction of OC(Pri)2 with HBpin, catalyzed by L†SnH 2. Selected bond
lengths (Å), and free and electronic energies (kcal/mol), are shown.
It is clear from the results presented in Table 1 that the
hydroboration of aliphatic aldehydes is significantly more
efficient than that of the aromatic aldehydes. This is likely due
to a combination of steric and electronic factors. That is, the
bulkier the substrate, the more difficult it is for the O-center of its
hydrometalated product, L†MOC(H)2R, to approach the boron
center of HBpin in the rate-determining, σ-bond metathesis step
of the catalytic cycle (vide infra). Similarly, aryl substituents (R)
likely reduce the Lewis basicity of the O-center of L†MOC(H)2R,
relative to aliphatic substituents (R) on the intermediate.
Furthermore, it is apparent that reactions employing the tin
precatalyst, L†SnOBut (and therefore the catalyst, 2), are more
rapid than those carried out with the Ge catalyst, 1. This can be
explained by the larger, more Lewis acidic metal center, and more
polar δ+M−Oδ− bonds, in the hydrostannylated intermediates,
thus making these intermediates more reactive toward HBpin
than their Ge counterparts.15
Not surprisingly, to achieve near-quantitative hydroborations
of ketones (Table 2), higher catalyst loadings were required than
for the less bulky aldehyde substrates. In addition, the ketone
hydroboration reactions were slower, though, as was the case for
aldehydes, the reactions catalyzed by tin were more rapid than
those employing the germanium hydride catalyst. With that said,
the hydroborations of the aldehydes and ketones reported here
are typically more efficient than all similar, previously reported
transition metal catalyzed hydroborations utilizing HBpin.6,8a It
should be noted, however, that the group 14 catalysts show
similar efficiencies to the highly active magnesium hydride
hydroboration catalyst, {(DipNacnac)MgH}2, recently reported
by Hill et al.9a
chemoselective and quantitative reduction of the ketone
functionality, leaving the alkene fragment intact. With respect
to diastereoselectivity, the hydroboration of 2-methylcyclo-
hexanone was carried out with both catalysts. While little
selectivity was observed for the tin catalyst, significant cis-/trans-
selectivity (ca. 72:28) was reproducibly achieved for the Ge
catalyst under several catalyst loadings. Although the origin of
this selectivity is not yet clear, it is intriguing that this is the
opposite of the (less pronounced) trans-/cis-diastereoselectivity
typically observed for the hydroboration of this substrate.16
In order to shed some light on the mechanism of the observed
reactions, a kinetic analysis of the hydroboration of 4-
ethylacetophenone catalyzed by the germanium hydride
compound, 1, was undertaken using the initial rates method
(see SI for full details).17 The results of that analysis clearly
indicate first-order dependence of the reaction in both HBpin
and catalyst and zero-order dependence in ketone. This implies
that the rate-determining step of the catalytic cycle is the reaction
of the alkoxide intermediate, L†GeOC(H)(Me)(PhEt-4), with
HBpin, and therefore that the intermediate is the resting state in
the cycle. This conclusion is consistent with the preliminary
stoichiometric reaction studies mentioned above and also implies
that the equilibrium between L†(H)GeGe(H)L† and
L†(H)Ge, 1, should have little effect on the overall reaction
rate. While it cannot be certain, it seems very likely that the active
species in that equilibrium is two-coordinate 1, which is rapidly
consumed by reaction with the ketone,12 and is thus only present
in the reaction mixture in negligible amounts throughout the
cycle.
We propose that the mechanism of the Ge- and Sn-catalyzed
reactions initially involves attack of the O-center of the substrate
at the two-coordinate metal center, with hydrometalation
subsequently proceeding via a four-membered transition state
(Scheme 1). The monomeric, two-coordinate metal alkoxide
intermediate then undergoes a σ-bond metathesis reaction with
HBpin to generate the borate ester and return the catalyst. It is
noteworthy that this mechanism is reminiscent of that proposed
by Hill et al. for {(DipNacnac)MgH}2 catalyzed ketone
hydroborations.9a So as to assess the feasibility of our proposal,
Although all hydroborations are completely regioselective for
the formation of borate esters, preliminary investigations were
carried out to determine the chemo- and diastereoselectivity of
the ketone reductions. First, it was found that dihydroborations
of the α-diketone, benzil, were rapid with both catalysts (using
200 mol % of HBpin) and that there was no evidence for the
formation of the singly hydroborated product in either reaction.
In contrast, the catalyzed reactions of 2-cyclohexen-1-one with 2
equiv of HBpin were complete in under 1 h, but led only to the
3030
dx.doi.org/10.1021/ja5006477 | J. Am. Chem. Soc. 2014, 136, 3028−3031

Journal of the American Chemical Society
Communication
(2) N.B. Excellent progress has been made in the development of
catalysis involving “normal oxidation state” p-block frustrated Lewis
pairs. See, for example: Erker, G.; Stephan, D. W. Frustrated Lewis Pairs
II: Expanding the Scope; Springer: Heidelberg, 2013.
(3) (a) Hadlington, T. J.; Hermann, M.; Li, J.; Frenking, G.; Jones, C.
Angew. Chem., Int. Ed. 2013, 52, 10199. (b) Hadlington, T. J.; Jones, C.
Chem. Commun. 2014, 50, 2321. For the first activations of H2 by group
14 centers, see: (c) Spikes, G. H.; Fettinger, J. C.; Power, P. P. J. Am.
Chem. Soc. 2005, 127, 12232. (d) Frey, G. D.; Lavallo, V.; Donnadieu, B.;
the free energy profile of the tin hydride catalyzed reaction of
HBpin with the very bulky substrate, OC(Pri)2, was calculated
using DFT, with the inclusion of dispersion interactions
(BP86+D(BJ)/def2-TZVPP). The calculated profile (Figure
2)18 closely matches the proposed general mechanism and shows
that both the hydrostannylation and σ-bond metathesis reactions
are exergonic, by −16.7 and −3.4 kcal/mol, respectively.
Furthermore, the fact that these reactions exhibit kinetic barriers
of 10.5 and 16.1 kcal/mol, respectively, is fully consistent with
the experimental observation that the σ-bond metathesis reaction
is the rate-determining step in the catalytic cycle. Considering
that these calculations were carried out on the experimentally
most difficult substrate to hydroborate, it would be expected that
the kinetic barriers to the hydroboration of less bulky substrates
would be significantly lower. Accordingly, the computational
study clearly highlights the thermodynamic and kinetic viability
of the proposed general mechanism.
In conclusion, the use of well-defined low-valent p-block metal
hydrides as catalysts in organic synthesis has been demonstrated
for the first time. In this respect, the highly reactive, two-
coordinate germanium(II) and tin(II) hydride compounds, 1
and 2, have effected the catalytic addition of HBpin to a range of
carbonyl compounds, with efficiencies that rival the most active
catalysts presently available for such reactions. It seems
reasonable that low-valent group 14 metal compounds, such as
1 and 2, will find a range of other catalytic applications in organic
synthetic methodologies (e.g., alkene hydrosilylations, CO2
reductions, etc.). We are currently investigating this possibility
and will report on our findings in due course.
Scholler, W. W.; Bertrand, G. Science 2007, 316, 439.
̈
(4) Relevant reviews: (a) Cho, B. T. Chem. Soc. Rev. 2009, 38, 443.
(b) Togni, A.; Grutzmacher, H. Catalytic Heterofunctionalization; Wiley-
̈
VCH: Weinheim, 2001.
(5) Fu, G. C.; Evans, D. A. J. Org. Chem. 1990, 55, 5678.
(6) Koren-Selfridge, L.; Query, I. P.; Hanson, J. A.; Isley, N. A.; Guzei, I.
A.; Clark, T. B. Organometallics 2010, 29, 3896.
(7) Khalimon, A. Y.; Farha, P.; Kuzmina, L. G.; Nikonov, G. I. Chem.
Commun. 2012, 48, 455.
(8) (a) Oluyadi, A. A.; Ma, S.; Muhoro, C. N. Organometallics 2013, 32,
70. (b) Almqvist, F.; Torstensson, L.; Gudmundsson, A.; Frejd, T.
Angew. Chem., Int. Ed. Engl. 1997, 36, 376. (c) Giffels, G.; Dreisbach, C.;
Kragl, U.; Weigerding, M.; Waldmann, H.; Wandrey, C. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2005.
(9) (a) Arrowsmith, M.; Hadlington, T. J.; Hill, M. S.; Kociok-Kohn, G.
Chem. Commun. 2012, 48, 4567. (b) Blake, A. J.; Cunningham, A.; Ford,
A.; Teat, S. J.; Woodward, S. Chem.Eur. J. 2000, 6, 3586.
(10) Takagi, N.; Sakaki, S. J. Am. Chem. Soc. 2013, 135, 8955.
(11) N.B. (a) One report on the reduction of ketones using high
loadings (10 mol%) of tin(II) triflate as a catalyst in the presence of a
silane invoked the intermediacy of a tin(II) hydride in the reaction. No
evidence for this proposal was presented. Lawrence, N. J.; Bushell, S. M.
Tetrahedron Lett. 2000, 41, 4507. (b) Tilley et al. have described several
cationic transition metal hydridosilylene complexes ([LnTMSi(H)
R]+) that catalyze hydrosilylations of unsaturates. For a recent example,
see: Fasulo, M. E.; Lipke, M. C.; Tilley, T. D. Chem. Sci. 2013, 4, 3882.
(12) We assume that the monomeric species, 1 and 2, which are in
equilibrium with dimeric L†(H)GeGe(H)L† and L†Sn(μ-H)2SnL† in
solution, are the active species in the stoichiometric and catalytic
reactions reported here (see ref 3 and Supporting Information).
(13) Jana, A.; Roesky, H. W.; Schulzke, C. Dalton Trans. 2010, 39, 132.
ASSOCIATED CONTENT
■
S
* Supporting Information
Details of the synthesis and characterizing data for all new
compounds. Full details and references for the catalysis, kinetic,
crystallographic, and computational studies. Crystallographic
data in CIF format. This material is available free of charge via the
Internet at http://pubs.acs.org.
(14) (a) Jana, A.; Roesky, H. W.; Schulzke, C.; Doring, A. Angew.
̈
Chem., Int. Ed. 2009, 48, 1106. (b) Jana, A.; Roesky, H. W.; Schulzke, C.
Inorg. Chem. 2009, 48, 9543.
AUTHOR INFORMATION
■
(15) All aldehyde and ketone substrates were found to be either
unreactive or poorly reactive (max. 5% conversion) towards HBpin in
catalyst-free control experiments. No enolization byproducts were
observed for the catalyzed hydroborations of any substrate.
(16) Query, I. P.; Squier, P. A.; Larson, E. M.; Isley, N. A.; Clark, T. B. J.
Org. Chem. 2011, 76, 6452.
(17) Sewell, L. J.; Huertos, M. A.; Dickinson, M. E.; Weller, A. S. Inorg.
Chem. 2013, 52, 4509.
(18) The weakly bonded adducts 2Sn and 4Sn, which are minima on
the potential energy surface, are energetically lower-lying than 1Sn and
3Sn, respectively. The former species are less stable than the latter
molecules when temperature and entropic effects are considered.
Corresponding Authors
frenking@chemie.uni-marburg.de.
cameron.jones@monash.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support from the Australian Research Council (C.J.,
DP120101300), the USAF Asian Office of Aerospace Research
and Development (C.J.), and the Deutsche Forschungsgemein-
schaft (G.F.) are acknowledged. The EPSRC National Mass
Spectrometry Facility is also thanked. We are also grateful to
Prof. Andrew S. Weller, Oxford University, for many helpful
discussions.
REFERENCES
■
(1) Selected recent reviews: (a) Power, P. P. Chem. Rec. 2012, 12, 238.
(b) Mandal, K. S.; Roesky, H. W. Acc. Chem. Res. 2012, 45, 298.
(c) Power, P. P. Acc. Chem. Res. 2011, 44, 627. (d) Asay, M.; Jones, C.;
Driess, M. Chem. Rev. 2011, 111, 354. (e) Yao, S.; Xiong, Y.; Driess, M.
Organometallics 2011, 30, 1748. (f) Martin, D.; Soleilhavoup, M.;
Bertrand, G. Chem. Sci. 2011, 2, 389. (g) Power, P. P. Nature 2010, 463,
171.
3031
dx.doi.org/10.1021/ja5006477 | J. Am. Chem. Soc. 2014, 136, 3028−3031