Importance of co-donor field strength in the preparation of tetradentate α-diimine nickel hydrosilylation catalysts

Article abstract of DOI:10.1039/c3dt52419a

Although bis(α-diimine)Ni complexes were prepared when amine-substituted chelates were added to Ni(COD)₂, the incorporation of strong-field phosphine donors allowed the isolation of (κ⁴- N,N,P,P-DI)Ni hydrosilylation catalysts. The c

Full text of DOI:10.1039/c3dt52419a

Dalton
Transactions
View Article Online
View Journal | View Issue
COMMUNICATION
Importance of co-donor field strength in the
preparation of tetradentate α-diimine nickel
hydrosilylation catalysts†
Cite this: Dalton Trans., 2013, 42, 14689
Received 2nd September 2013,
Accepted 4th September 2013
Tyler M. Porter,a Gabriel B. Hall,b Thomas L. Groya and Ryan J. Trovitch*a
DOI: 10.1039/c3dt52419a
www.rsc.org/dalton
Although bis(α-diimine)Ni complexes were prepared when amine- chelate was prepared using a slightly modified procedure and
substituted chelates were added to Ni(COD)2, the incorporation of a stoichiometric quantity was added to Ni(COD)2. Analysis of
strong-field phosphine donors allowed the isolation of (κ4-N,N,P,P-DI)Ni the resulting red solution by 1H NMR spectroscopy revealed
hydrosilylation catalysts. The crystallographic investigation of two only partial conversion to
a new compound featuring
different (κ4-N,N,P,P-DI)Ni compounds revealed that the geometry an upfield shifted resonance at −0.73 ppm, indicative of
about nickel influences the observed degree of α-diimine reduction.
bis(ligand) complex formation.4,13,14 Adding 2 equivalents of
Me NPrDI to Ni(COD)2 allowed for the isolation and full charac-
2
Redox-active ligands, which assist first-row transition metal
catalysis by mediating reversible intramolecular electron trans-
fer, have continued to gain in popularity.1,2 The coordination
chemistry of 1,4-diaza-1,3-butadiene (α-diimine or DI) ligands
has been particularly well-investigated3 as they are structurally
simple, easy to prepare, and highly modular non-innocent
chelates.4,5 While these characteristics have enabled the
design of several (DI)Ni α-olefin polymerization catalysts,6,7
selectivity for polymerization over oligomerization has been
strongly linked to the incorporation of sterically demanding DI
imine substituents.6,8 Although the inclusion of bulky substi-
tuents has remained a guiding force in the development of
redox-active ligand-supported catalysts,1,2 it is believed that
integrating secondary donors into the chelate framework may
offer the same benefits that steric bulk provides while poten-
terization of (κ2-N,N-Me NPrDI)2Ni (1, eqn (1)).
2
ð1Þ
Because the dimethylamino groups of Me NPrDI did not
2
allow for κ4-DI ligation, a formal Ni(0) complex supported by
PyEtDI (eqn (1), left) was sought. This ligand has been reported
to bind Cu(II) in a tetradentate fashion15 and the glyoxal-
derived variant of this chelate has been found to adopt
κ4-coordination to Ni(II).16 Although PyEtDI has not been pre-
viously isolated, this ligand was prepared following the conden-
sation of 2,3-butanedione with 2 equivalents of 2-(2-aminoethyl)-
pyridine in a sealed reaction vessel at 100 °C. Unfortunately,
adding a single equivalent of PyEtDI to Ni(COD)2 also resulted
in limited formation of the respective bis(ligand) complex,
(κ2-N,N-PyEtDI)2Ni (2, eqn (1)), as judged by 1H and 13C NMR
spectroscopy. As with 1, complex 2 was prepared following the
direct addition of two PyEtDI equivalents to Ni(COD)2. Further-
more, the bidentate coordination of each PyEtDI moiety was
confirmed by single crystal X-ray diffraction (Fig. 1).
tially
facilitating
the
stabilization
of
high-energy
intermediates.9
Following this methodology, low-valent nickel complexes
featuring a κ4-DI ligand were targeted so their ability to
mediate catalytic transformations, including hydrosilyl-
ation,10,11 could be investigated. Knowing that Me NPrDI
2
(eqn (1), left) is capable of κ4-coordination to iron,12 this
aDepartment of Chemistry & Biochemistry, Arizona State University, Tempe, Arizona
85287, USA. E-mail: ryan.trovitch@asu.edu; Fax: +1 480 965 2747;
Tel: +1 480 727 8930
bDepartment of Chemistry & Biochemistry, The University of Arizona, Tucson,
Arizona 85721, USA
The geometry about the metal centre in 2 is best described
as a distorted tetrahedron, having an acute dihedral angle
between the two 5-membered chelate planes of 74.8°. Notably,
this complex features elongated N(1)–C(2) and N(2)–C(3) bond
distances of 1.344(3) and 1.341(3) Å, respectively, along with a
†Electronic supplementary information (ESI) available: Full experimental
details, supporting figures for the NMR, UV-vis, and CV data discussed within
the text, and the crystallographic data for complexes 2,
3 and 4. CCDC
947972–947974. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c3dt52419a
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 14689–14692 | 14689

View Article Online
Communication
Dalton Transactions
Investigation of the isolated product by multinuclear NMR
spectroscopy confirmed the formation of (κ4-N,N,P,P-Ph PPrDI)
2
Ni (4, eqn (2)), which also exhibits a single 31P resonance at
39.03 ppm. Although both the 1H and {31P}1H NMR spectra
1
recorded for 4 featured 6 methylene resonances, the H NMR
resonances found for 3 collapsed into two multiplets upon 31P
decoupling, suggesting that 3 possesses ligand arms that are
equivalent from top to bottom in benzene-d6 solution at 23 °C.
Fig. 1 The solid state structure of 2 at 30% probability ellipsoids. Hydrogen
atoms are omitted for clarity. Relevant bond distances and angles are provided
throughout the text and in Table S2 of the ESI.†
significantly shortened C(2)–C(3) distance of 1.416(4) Å. Based
on prior crystallographic studies,4,5 the α-diimine bond dis-
tances determined for 2 strongly suggest that each ligand is
singly reduced and that the electronic structure of this
complex is best described as having a high-spin Ni(II) centre
(SNi = 1) that is antiferromagnetically coupled to each ligand
based radical. Additional evidence that 2 possesses this elec-
tronic structure was obtained by cyclic voltammetry, as
two one-electron oxidation waves were observed at
E11/2 = −0.84 and E12/2 = −0.56 V (vs. Fc+/Fc, Fig. S21 of the ESI†).
These potentials directly correlate to those previously
described for a related alkyl-substituted (DI)2Ni complex fea-
ð2Þ
The different colours observed for 3 (green) and 4 (red) in
solution prompted the investigation of each complex by UV-
visible spectroscopy. Although both complexes exhibited a
spectrum dominated by charge transfer bands (Fig. S20 of the
ESI†), the transitions observed for 3 [420 nm (ε = 5600 M−1
cm−1) and 714 nm (ε = 5000 M−1 cm−1)] were higher in energy
than the same bands observed for 4 [498 nm (ε = 11 800 M−1
cm−1) and 780 nm (ε = 1000 M−1 cm−1)]. Notably, the MLCT
band associated with backbonding into the phosphine donors
of 3 was found to have a much higher extinction coefficient
than the same transition observed for 4, suggesting that geo-
metric differences might be influencing the relative degree of
overlap between the Ni(3d) orbitals and the P(σ*) and DI(π*)
orbitals of the chelate.
turing two monoanionic DI ligands (E11/2 = −0.82 and E12/2
−0.53 V relative to Fc+/Fc in THF).5
=
Although κ4-DI coordination to Ni(0) could not be achieved
when amines were built into the chelate framework, it was
hypothesized that the incorporation of strong-field co-donors
would more effectively compete for metal-based electron
density through backbonding, therefore preventing bis(ligand)
complex formation. For this reason, DI ligands featuring phos-
phinoalkyl imine substituents were pursued. Since the optimal
carbon-chain length for tetradentate DI coordination was
unclear, chelates featuring either ethylene (Ph PEtDI, eqn (2)) or
2
propylene (Ph PPrDI, eqn (2)) bridges to diphenylphosphine
2
substituents were synthesized in a similar fashion to PyEtDI
Moreover, the molecular structures of 3 and 4 were deter-
mined by single crystal X-ray diffraction (Fig. 2).19 The geome-
try about the Ni centre in 3 can be described as distorted
square planar, as the acute dihedral angle between the planes
defined by Ni(1)–N(1)–C(2)–C(3)–N(2) and Ni(1)–P(1)–P(2) is
only 32.5°. In contrast, the geometry about the Ni centre in 4 is
best defined as distorted tetrahedral; the same dihedral angle
for this complex is 56.3°. While this observation confirms that
the ethylene-bridged donor arms of 3 restrict this complex
from achieving approximate tetrahedral coordination about
Ni, this characteristic also appears to influence the relative
degree of DI reduction observed between the two complexes.
The N(1)–C(2) and N(2)–C(3) bond distances in 4 were deter-
mined to be 1.340(3) and 1.341(3) Å, respectively, while the
C(2)–C(3) distance was found to be shortened to 1.414(3) Å.
These bond lengths are statistically indistinguishable from the
same distances in 2 and strongly indicate4,5 that the electronic
(see ESI† for experimental details). Whereas an alternate
preparation of Ph PEtDI has been previously reported,17 litera-
2
ture discussion of Ph PPrDI has remained limited.18
2
The stoichiometric addition of Ph PEtDI to Ni(COD)2 in
2
benzene-d6 resulted in an immediate colour change from
yellow to a dark green solution. Analysis by 1H NMR spec-
troscopy revealed the complete disappearance of Ni(COD)2 and
the formation of a complex featuring a single DI backbone
methyl resonance that was split into a well-defined triplet
(1.95 ppm, 4.3 Hz). This resonance collapsed into a singlet
upon 31P decoupling (Fig. S13 of the ESI†) and 31P NMR spec-
troscopy further confirmed that the newly formed compound,
(κ4-N,N,P,P-Ph PEtDI)Ni (3, eqn (2)), possesses a single phos-
2
phorous environment (56.36 ppm). In contrast, adding a
single equivalent of Ph PPrDI to Ni(COD)2 resulted in an instan-
2
taneous colour change from yellow to a dark red solution.
14690 | Dalton Trans., 2013, 42, 14689–14692
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Communication
each complex were investigated using cyclic voltammetry, and
upon scanning from negative to positive potentials, 3 and 4
exhibited fully reversible one electron oxidation waves at E11/2
=
−0.54 V and −0.46 V (relative to Fc+/Fc in acetonitrile), respect-
ively. Additionally, 4 was found to exhibit a significantly broad-
ened, yet reversible, oxidation at E21/2 = 0.21 V while the cyclic
voltammogram of 3 featured a partially reversible (scan rate =
50 mV s−1
) =
and irregularly shaped oxidation at E12/2
−0.13 V. While this 0.30 V difference in E21/2 values is signifi-
cant, the CV data collected for 3 and 4 is otherwise analogous
(Fig. S22 of the ESI†), suggesting these complexes may possess
the same electronic structure (as proposed in eqn (2)). It
should be appreciated that as tetradentate complexes of this
type approach their square planar geometric limit, the energy
of the highest lying Ni d-orbital will increase, boosting the like-
lihood that the ligand LUMO (π*) will become doubly popu-
lated. A complete DFT investigation of 3 and 4 is underway,
which may more accurately address the question of when (at
what dihedral angle) the redox-active DI core of the chelate
becomes a true dianion. Neither 3 nor 4 was found to exhibit
an X-band EPR signal in frozen toluene solution at 130 K or
140 K, respectively.
Having studied the electronic structure of two different
(κ4-DI)Ni complexes, we sought to determine if either complex
was capable of mediating the hydrosilylation of unsaturated
organic substrates. Fortunately, in spite of their geometric
differences, both 3 and 4 were found to catalyse the hydrosilyl-
ation of cyclohexanone, diisopropyl ketone, and phenyl-
acetylene (Table 1). Using an equimolar quantity of PhSiH3
Fig. 2 The molecular structure of 3 (top) and 4 (bottom) at 30% probability
ellipsoids. Hydrogen atoms are omitted for clarity and the aryl groups were
removed to generate the core representations shown at the bottom left of each
structure. Relevant bond distances and angles are provided throughout the text
and in Tables S3 and S4 of the ESI.†
Table 1 The hydrosilylation of unsaturated substrates using 3 or 4
structure of 4 is best described as having a Ni(I) centre that is
antiferromagnetically coupled to a DI radical monoanion.
In contrast, complex 3 features N(1)–C(2) and N(2)–C(3)
bond lengths of 1.360(4) and 1.373(3) Å, respectively, which
are longer than the imine distances determined for 2 and 4.
Concurrently, the C(2)–C(3) bond distance of 1.388(4) Å estab-
lished for 3 is shortened by approximately 0.027 Å relative to
2 and 4, indicating a greater degree of DI ligand reduction.
Although the metrical parameters associated with the imine
bonds in 3 are consistent with those determined for other
complexes featuring a doubly reduced α-diimine ligand,20 the
C(2)–C(3) bond distance of 1.388(4) Å is longer than one might
expect to observe for this level of chelate reduction
(1.36–1.37 Å).4,5 Further crystallographic evidence that the Ni
centre in 3 may possess a higher physical oxidation state than
4 is also observed upon inspection of the Ni–N bond distances.
The Ni(1)–N(1) and Ni(1)–N(2) distances of 1.896(2) and
1.885(2) Å, respectively, for 3 are shorter than the bond lengths
of 1.9369(17) and 1.9250(18) Å found for 4.
Conv.a
(%)
Entry Cat. Substrate
Products (Ratio)
1
2
3
3
4
3
PhSiH(OCy)2
PhSiH2(OCy) (20 : 1)
93
PhSiH2(OCy)
PhSiH(OCy)2 (2 : 1)
>99
>99
PhSiH(OCH(iPr)2)2
PhSiH2(OCH(iPr)2) (11 : 1)
4
4
PhSiH(OCH(iPr)2)2
>99
PhSiH2(OCH(iPr)2) (1 : 1)
5
6
3
4
trans-(Ph)HCvCH(SiH2Ph)b >99
trans-(Ph)HCvCH(SiH2Ph)b >99
a Conversion determined by NMR spectroscopy at 24 h. b Major
product.
Complimentary methods to probe the electronic structure
of 3 and 4 have also been utilized. The redox properties of
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 14689–14692 | 14691

View Article Online
Communication
Dalton Transactions
and 5 mol% of either 3 or 4, the near complete hydrosilylation
of each substrate was observed after 24 h at ambient tempera-
ture. Efforts to optimize the performance of these catalysts,
while determining their substrate scope and functional group
tolerance, are underway.
8 S. A. Svejda and M. Brookhart, Organometallics, 1999, 18,
65.
9 H. Ben-Daat, G. B. Hall, T. L. Groy and R. J. Trovitch,
Eur. J. Inorg. Chem., 2013, 4430–4442.
10 S. Chakraborty, J. A. Krause and H. Guan, Organometallics,
2009, 28, 582, and references therein.
This work was supported as part of the Center for Bio-
Inspired Solar Fuel Production, an Energy Frontier Research 11 S. Chakraborty and H. Guan, Dalton Trans., 2010, 39,
Center funded by the U. S. Department of Energy, Office of 7427.
Science, Basic Energy Sciences under Award Number 12 N. J. Hardman, X. Fang, B. L. Scott, R. J. Wright,
DE-SC0001016. We would like to thank Prof. Anne K. Jones R. L. Martin and G. J. Kubas, Inorg. Chem., 2005, 44, 8306.
(ASU) and Prof. Amanda C. Bowman (Transylvania U.) for 13 M. Svoboda and H. tom Dieck, Z. Naturforsch., B: Chem.
helpful discussions and Dr. Sue Roberts (U. of Arizona) for
Sci., 1981, 36, 814.
assisting with the collection of X-ray diffraction data.
14 H. tom Dieck, M. Svoboda and T. Greiser, Z. Naturforsch.,
B: Chem. Sci., 1981, 36, 823.
15 M. L. P. Santos, I. A. Bagatin, E. M. Pereira and
A. M. D. C. Ferreira, J. Chem. Soc., Dalton Trans., 2001,
838.
Notes and references
16 A. Kriza, L. Pricop, A. Meghea and N. Stanica, J. Indian
Chem. Soc., 2001, 78, 448.
1 O. R. Luca and R. H. Crabtree, Chem. Soc. Rev., 2013, 42, 1440.
2 P. J. Chirik and K. Wieghardt, Science, 2010, 327, 794.
3 G. Van Koten and K. Vrieze, Adv. Organomet. Chem., 1982,
21, 151.
4 M. M. Khusniyarov, K. Harms, O. Burghaus and
J. Sundermeyer, Eur. J. Inorg. Chem., 2006, 2985.
5 N. Muresan, K. Chlopek, T. Weyhermüller, F. Nesse and
K. Wieghardt, Inorg. Chem., 2007, 46, 5327.
17 T. D. DuBois, Inorg. Chem., 1972, 11, 718.
18 J. W. Corr, The synthesis of a series of ligands containing
nitrogen and phosphorus or arsenic donor atoms within the
same moiety. An investigation of the donor properties of these
ligands toward certain transition metals, Dissertation,
University of North Carolina, Chapel Hill, 1967.
19 The needle of 3 analyzed by X-ray diffraction appeared red
in colour.
6 L. K. Johnson, C. M. Killian and M. Brookhart, J. Am. Chem.
Soc., 1995, 117, 6414.
20 T. Spaniel, H. Görls and J. Scholz, Angew. Chem., Int. Ed.,
1998, 37, 1862.
7 C. M. Killian, D. J. Tempel, L. K. Johnson and
M. Brookhart, J. Am. Chem. Soc., 1996, 118, 11664.
14692 | Dalton Trans., 2013, 42, 14689–14692
This journal is © The Royal Society of Chemistry 2013