A structurally rigid bis(amido) ligand framework in low-coordinate Ni(I), Ni(II), and Ni(III) analogues provides access to a Ni(III) methyl complex via oxidative addition

Article abstract of DOI:10.1021/ja408151h

A structurally persistent bis-amido ligand framework capable of supporting nickel compounds in three different oxidation states has been identified. A highly unusual, isolable Ni(III) alkyl species has been prepared and characterized via a rare example of a two-electron oxidative addition of MeI to Ni(I).

Full text of DOI:10.1021/ja408151h

Communication
pubs.acs.org/JACS
A Structurally Rigid Bis(amido) Ligand Framework in Low-Coordinate
Ni(I), Ni(II), and Ni(III) Analogues Provides Access to a Ni(III) Methyl
Complex via Oxidative Addition
Michael I. Lipschutz,† Xinzheng Yang,‡,§ Ruchira Chatterjee,∥ and T. Don Tilley*,†
†Department of Chemistry and ‡Molecular Graphics and Computation Facility, University of California, Berkeley, Berkeley, California
94720, United States
§Institute of Chemistry, Chinese Academy of Sciences, Bejing, 100190, P.R. China
∥Physical Biosciences Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Rd, Berkeley, California 94720, United States
S
* Supporting Information
imidazol-2-ylidene) reacted with 1 upon heating at 80 °C to
ABSTRACT: A structurally persistent bis-amido ligand
framework capable of supporting nickel compounds in
three different oxidation states has been identified. A
highly unusual, isolable Ni(III) alkyl species has been
prepared and characterized via a rare example of a two-
electron oxidative addition of MeI to Ni(I).
form the yellow nickel(I) species (IPr)Ni[N(SiMe3)DIPP] (4)
in 61% isolated yield (Scheme 1). This interesting redox
process, which formally involves displacement of •N(SiMe3)-
DIPP, represents a convenient route to low-coordinate Ni(I)
amido complexes, which have been used as precursors to Ni-
imido complexes by Hillhouse et al.3d,f The unusual redox
process associated with this reaction prompted an examination
of 1 by cyclic voltammetry (CV), which reveals fully reversible
Ni(I/II) and Ni(II/III) couples at −1.28 V (E1/2 vs Fc/Fc+, ipa/
ue to their comparatively low toxicity, minimal cost, and
Dhigh abundance, catalysts based on first-row transition ipc = 0.98) and +0.18 V ( E1/2 vs Fc/Fc+, ipa/ipc = 0.97; see
metals are attracting considerable interest.1 Many attempts to
Supporting Information, SI). The relatively low and reversible
reduction potential for 1 suggested that the corresponding
anionic Ni(I) complex might be isolable.
develop such catalysts are based on well-known mechanisms
associated with second- and third-row metals; however,
alternative design strategies might leverage the chemical
transformations and properties that are unique to the 3d
metals. Among the many features that distinguish first-row
metals from their heavier congeners is their ability to adopt very
low-coordination numbers (e.g., two and three) that are
associated with electrophilic metal centers and high reactivi-
ty.2a−g
Whereas a number of low-coordination number complexes
are known, e.g., with Cr, Mn, Fe, Co, and Ni, the reaction
chemistry associated with this structural type is relatively
unexplored.2a Exceptions are found in the research of Hillhouse
et al., who have explored a number of novel transformations for
low-coordinate, neutral Ni(I) and Ni(0) complexes.3a−g
Applications of such complexes to catalysis are less well
studied, but this laboratory has recently described use of the
precatalyst Fe[N(SiMe3)2]2 for hydrosilations of carbonyl
compounds4 and Ni[N(SiMe3)DIPP]2 (1, DIPP = 2,6-
diisopropylphenyl) as an alkene hydrosilation catalyst.5 Herein,
we describe the redox chemistry of 1 and demonstrate that this
nickel bis(amido) framework readily supports low-coordinate
complexes in three different oxidations states, Ni(I/II/III), and
accommodates a two-electron oxidative addition to anionic
{Ni[N(SiMe3)DIPP]2}− (2) to produce the unusual nickel(III)
alkyl complex (Me)Ni[N(SiMe3)DIPP]2 (3).
The reduction of 1 by 1.1 equiv of KC8 in toluene occurred
over 20 min at −30 °C and workup of the solution provided
K{Ni[N(SiMe3)DIPP]2} (2a) as light-yellow crystals in 89%
yield (Scheme 1). The X-ray structure of 2a reveals that the
complex maintains a linear geometry about nickel with a N−
Ni−N bond angle of 178.05(9)° (Scheme 1) and a Ni−N bond
distance of 1.8436(2) Å. The potassium atom in this complex is
sandwiched via η6-coordination to the aryl rings of both amido
ligands (Scheme 1). The potassium was exchanged for
tetrabutylammonium (NnBu4 ) by reaction of 2a with
+
[NnBu4]Br in THF to give NnBu4{Ni[N(SiMe3)DIPP]2} (2b)
as yellow crystals in 70% yield. As shown by the crystal
structure (Scheme 1), 2 exists as a rigorously two-coordinate
bis(amido) complex with metrical parameters similar to those
of 2a and both of these complexes exhibit an eclipsed
conformation of the planar amido substituents. The solution
magnetic moments of 2a,b, determined by Evans’ method,6 are
1.66 and 2.03 μB, respectively, and are consistent with one
unpaired electron as expected for a d9 configuration.
To better understand the electronic configuration of this
eclipsed, bis-amido structural motif, DFT calculations were
performed on 1 and 2b, using both the hybrid functional
ωB97XD7 and the pure functional revTPSS,8 since these
functionals treat open-shell spin states in fundamentally
different ways. The optimized structures of 1 and 2b are in
close agreement with their respective crystal structures in
Previous reactivity studies of 1 demonstrated that simple
two-electron donors, such as MeCN and 4-(N,N-
dimethylamino)pyridine, add rapidly to form three-coordinate,
T-shaped adducts.5 However, the more sterically demanding N-
heterocyclic carbene IPr (N,N′-1,3-bis(2,6-diisopropylphenyl)-
Received: August 6, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja408151h | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Scheme 1. Reactions of Ni Bis(amido) Complexes
retaining a linear, eclipsed coordination geometry with bond
lengths and angles consistent with experimental values (see SI).
In both calculations of 1, the triplet spin state is favored over
the singlet state by a considerable margin (17 and 7 kcal/mol
for the ωB97XD and revTPSS calculations, respectively). This
is in agreement with the magnetic moment of 2.67 μB observed
for 1, consistent with two unpaired electrons.
In addition to the conformational and metrical similarities
between neutral 1 and anionic 2b, the DFT calculations
indicate that their electronic structures are closely related. A
single qualitative MO diagram can be used to represent the
electronic structures of both 1 and 2b as shown in Figure 1.
The HOMO for both complexes is a singly occupied N−Ni−N
π-antibonding orbital. This three-centered π system may be
viewed as resulting from combinations of Ni dyz and N py
orbitals to give bonding, nonbonding, and antibonding orbitals
populated by a total of five electrons. Single occupancy of the π-
antibonding HOMO results in a net bond order of one-half for
this π system, for both 1 and 2b. It is difficult to assess the
contribution of this π bonding to the stability of eclipsed
structures for 1 and 2a. However, it is worth noting that the
conversion of 1 to 2a is accompanied by a rotation about the
Ni−N bond to place the two aryl groups cis to one another.
That this occurs at ambient temperatures suggests that the
rotational barrier is minimal and thus that the π bonding
interaction is weak.
Given the striking similarities between the electronic
structures of the two-coordinate Ni(I) and Ni(II) complexes
1 and 2b, it was of interest to synthetically access the analogous
Ni(III) (d7) cation. The optimized geometry of the Ni(III)
cation {Ni[N(SiMe3)DIPP]2}+ (5), calculated by the DFT
methods described above,9 is predicted to feature an eclipsed
conformation and a linear coordination geometry. The more
electrophilic metal center in this case results in a canting of the
aryl groups within the coordination plane to provide close aryl-
Ni interactions and a smaller Ni−N−Cipso bond angle (88.15°
compared to 106.66(9)° for 1 and 121.62(19)° for 2b, see SI).
The hybrid calculation also predicts that the doublet spin state
is favored over the quartet spin state by 14 kcal/mol.
Observation of the reversible oxidation wave for 1 at +0.18 V
suggested that the Ni(III) cation might be chemically
accessible. However, this wave lies close to an irreversible
event at ∼0.67 V (Ep vs Fc/Fc+), with an onset potential of
∼0.40 V. The close proximity of these two oxidation events
places upper and lower bounds on the oxidation potential for
the oxidant required to isolate the Ni(III) cation. Attempts to
observe the oxidation of 1 with Ag(I) reagents led to an
intractable mixture of Ni-containing products (Ag(Et2O)[B-
(C6F5)4] and Ag[CH6B11Br6]) or no reaction at 60 °C (AgOTf,
AgBF4, and AgNTf2).
Given these results, the isoelectronic d7 Co bis(amido)
complex Co[N(SiMe3)DIPP]2 (6) was obtained by a
procedure analogous to that used for 1 (see SI). This complex
was isolated as red crystals with a solution μeff value of 5.67 μB,
reflecting a high-spin configuration with strong spin−orbit
coupling, as previously observed for related two-coordinate
Co(II) bis-amido complexes.10 Complexes 1 and 6 are
Figure 1. Qualitative MO diagram for Ni bis(amido) ligand framework
with selected calculated molecular orbitals. The red electron is present
only in Ni(I) d9 species 2a,b. Calculated MO figures are from the
hybrid calculations on compound 1.
B
dx.doi.org/10.1021/ja408151h | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
isostructural, with an eclipsed conformation of the amido
substituents. There are several structural and electronic
differences between the calculated properties of 5 and the
observed properties of 6, but this comparison is convoluted by
the difference in charge and the strong spin−orbit coupling
interactions observed in 6.
displays hyperfine coupling to the two nitrogen atoms,
indicating substantial delocalization of the unpaired spin
density onto those nitrogen atoms, which is in agreement
with the calculated HOMO of 3. Together, the calculations and
EPR data support the characterization of 3 as a Ni(III) complex
rather than a Ni(II) complex with an oxidized, ligand-based
radical.
Further analysis of the calculated electronic structures of 1
and 2b suggested an alternative route to complexes with the
Ni(III) oxidation state. The HOMO-1 orbital of both
While 3 is an unusual compound in and of itself, the
oxidative process which gives rise to it is also quite rare.
Analysis of 2a,b and 3 by DFT calculations and EPR
spectroscopy supports the oxidation state assignments of
Ni(I) (see SI) and Ni(III), respectively (vide supra); thus,
the formation of 3 from 2a is formally a two-electron oxidative
addition. Notably, the lack of well-defined, two-electron redox
processes for first-row metal complexes is thought to represent
a major impediment to the development of catalytic cycles for
these metals.1 While oxidative additions to nickel are known,
they are much more commonly observed in Ni(0)/Ni(II) or
Ni(II/IV) couples. Well-defined Ni(I)/Ni(III) oxidative
additions are rare,12 but similar, ligand-assisted two-electron
oxidative additions are implicated in various nickel-catalyzed
cross-coupling reactions and Ni(III) alkyl complexes like 3 are
proposed as intermediates in those reactions.13,14 To the best of
our knowledge, no other Ni(I)/Ni(III) oxidative additions
leading to an isolable Ni(III) product are known. Remarkably,
complex 3 was found to be unreactive toward both powerful
electrophiles (Me3SiOTf, triflic acid) and nucleophiles (KOtBu,
KOMe, LiSPh). No reaction was observed with the above
reagents in THF at ambient temperature prior to the complete
2
compounds is the σ-antibonding dz orbital, in which most of
2
the electron density has been pushed into the torus of the dz
2
2
2
orbital. Substantial mixing between dz and dz −y , expected for
the C2h symmetry of 1 and 2b, results in electron density being
distributed along the “in-plane” molecular x-axis instead of in a
radially symmetrical fashion (Figure 1). This creates a much
more directional orbital, which is more amenable to bond
formation. Prior work with 1 has shown that this orbital,
2
depicted in Figure 1 (and derived from dz ), acts as an acceptor
for neutral, two-electron donors.5 However, in 2b, this orbital is
doubly occupied, suggesting it may instead serve as a reasonable
nucleophile toward a small, organic substrate.
The reaction of 2a with methyl iodide occurred over 35 min
starting at −78 °C and warming to room temperature. Workup
of the reaction mixture and recrystallization from pentane
afforded the unusual Ni(III) alkyl complex (Me)Ni[N(SiMe3)-
DIPP]2 (3, Scheme 1) as green prisms in 80% yield.
Compound 3 is thermally unstable, decomposing in benzene
solution at ambient temperature over ∼24 h to 1, accompanied
1
by the formation of ethane (detected by H and 13C NMR
1
thermal decomposition of 3 (∼24 h, via H NMR spectrosco-
spectroscopy). However, 3 was found to be sufficiently stable
for single crystal X-ray diffraction studies and combustion
analysis and exhibited no signs of decomposition after three
months as a solid at −30 °C.
py). Similarly, 3 (0.0035 mM in THF) was found to be
unreactive toward saturated THF solutions of [NnBu4]X (X =
Cl, 0.022 mM; Br, 0.11 mM; I, 0.0050 mM), indicating that the
oxidative addition of methyl iodide is not readily reversible
under these conditions.
Complex 3 is a rare example of a Ni(III) organometallic
complex and appears to be only the third structurally
characterized Ni(III) alkyl complex.11a−d The X-ray structure
of 3 reveals two inequivalent molecules, each in a T-shaped
coordination geometry, with an average N−Ni−N angle of
167.4(1)° and an average Ni−CMe bond length of 1.923(4) Å.
Both molecules display a close contact between the Ni center
and the ipso carbon of an adjacent aryl ring, similar to the
interactions predicted in the calculated structure of 5. Although
the relevant distances and angles vary by a significant amount
between the two molecules (Ni−Cipso: 2.377(2) and 2.541(2)
Å, Ni−N−Cipso: 94.01(1) and 102.90(1)°, respectively), this
suggests that the predicted structure of 5 is qualitatively
reasonable. The Ni−C bond length of 1.923(4) Å is similar to
those reported for the two previously known Ni(III) alkyl
species (1.994(3) and 2.015(3) Å for −Me and −Et analogs,
respectively), although both of these examples are five-
coordinate, negatively charged species.11a The solution
magnetic moment of 3 was determined to be 1.78 μB,
consistent with one unpaired electron and a low spin, d7
electronic configuration.
In summary, a structurally persistent bis-amido ligand
framework for nickel compounds in three oxidation states has
been identified, and a highly unusual Ni(III) alkyl species has
been prepared and characterized. Further exploration of this
system should yield important mechanistic information relevant
to nickel-catalyzed coupling reactions and multi-electron redox
processes for first-row metals.
ASSOCIATED CONTENT
■
S
* Supporting Information
Text and figures giving further experimental and spectroscopic
details. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
tdtilley@berkeley.edu
Notes
Efforts to better understand the electronic structure of 3
involved EPR spectroscopy and DFT calculations using the
same two functionals employed in the analyses of 1 and 2a,b.
Both computations of 3 place the odd electron in a π* orbital
bearing substantial metal character (see SI) supporting the
assignment of 3 as Ni(III). The frozen glass EPR spectrum of 3
in toluene reveals an anisotropic signal with gxx, gyy, gzz = (2.32,
2.15, 2.13), indicating that the unpaired spin resides in an
orbital with significant metal character (see SI). The signal also
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was funded by the National Science Foundation
under grant no. CHE-1265674. The Molecular Graphics and
Computational facility (College of Chemistry, University of
California, Berkeley) is supported by the National Science
Foundation under grant no. CHE-0840505.
C
dx.doi.org/10.1021/ja408151h | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
REFERENCES
■
(1) Catalysts Without Precious Metals; Bullock, R. M., Eds.; Wiley-
VCH: Weinheim, 2010.
(2) (a) Power, P. P. Chem. Rev. 2012, 112, 3482. (b) Lei, H.;
Fettinger, J. C.; Power, P. P. Inorg. Chem. 2012, 51, 1821. (c) Dugan,
T. R.; Sun, X.; Rybak-Akimova, E. V.; Olatunji-Ojo, O.; Cundari, T. R.;
Holland, P. L. J. Am. Chem. Soc. 2011, 133, 12418. (d) Harrold, N. D.;
Hillhouse, G. L. Chem. Sci. 2013, 4, 4011. (e) Nia, C.; Power, P. P.
Chem. Commun. 2009, 5543. (f) Danopoulos, A. A.; Braunstein, P.
Dalton Trans. 2013, 42, 7276. (g) Laskowski, C. A.; Hillhouse, G. L.
Organometallics 2009, 28, 6114.
(3) (a) Laskowski, C. A.; Hillhouse, G. L. J. Am. Chem. Soc. 2008,
130, 13846. (b) Laskowski, C. A.; Miller, A. J. M.; Hillhouse, G. L.;
Cundari, T. R. J. Am. Chem. Soc. 2011, 133, 771. (c) Iluc, V. M.;
Hillhouse, G. L. J. Am. Chem. Soc. 2010, 132, 11890. (d) Laskowski, C.
A.; Morello, G. R.; Saouma, C. T.; Cundari, T. R.; Hillhouse, G. L.
Chem. Sci. 2013, 4, 170. (e) Iluc, V. M.; Hillhouse, G. L. J. Am. Chem.
Soc. 2010, 132, 15148. (f) Anderson, J. S.; Iluc, V. M.; Hillhouse, G. L.
Inorg. Chem. 2010, 49, 10203. (g) Mindiola, D. J.; Hillhouse, G. L. J.
Am. Chem. Soc. 2001, 123, 4623.
(4) Yang, J.; Tilley, T. D. Angew. Chem., Int. Ed. 2010, 49, 10186.
(5) Lipschutz, M. I.; Tilley, T. D. Chem. Commun. 2012, 48, 7146.
(6) Evans, D. F. J. Chem. Soc. 1957, 2003.
(7) Chai, J. D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10,
6615.
(8) Perdew, J. P.; Ruzsinszky, A.; Csonka, G. I.; Constantin, L. A.;
Sun, J. Phys. Rev. Lett. 2011, 106, 179902.
(9) Calculations of the Ni(III) cation based on the pure functional
would only converge using an unrestricted, quadratically convergent
SCF procedure. The optimized geometry derived from this calculation
was similar in all relevant aspects to that calculated using the hybrid
functional and the doublet spin state was also found to be the lowest
energy. Backsay, G. B. Chem. Phys. 1988, 61, 385.
(10) Bryan, A. M.; Merrill, W. A.; Reiff, W. M.; Fettinger, J. C.;
Power, P. P. Inorg. Chem. 2012, 51, 3366.
(11) (a) Lee, C.-M.; Chen, C.-H.; Liao, F.-X.; Hu, C.-H.; Lee, G.-H. J.
Am. Chem. Soc. 2010, 132, 9256. (b) Grove, D. M.; van Koten, G.;
Zoet, R.; Murrall, N. W.; Welch, A. J. Am. Chem. Soc. 1983, 105, 1379.
(c) Grove, D. M.; van Koten, G.; Mul, W. P.; van der Zeijden, A.;
Terheijden, J. Organometallics 1985, 5, 33. (d) Kelley, D. G.; Marchaj,
A.; Bakac, A.; Espenson, J. H. J. Am. Chem. Soc. 1991, 113, 7583.
(12) Tsou, T. T.; Kochi, J. K. J. Am. Chem. Soc. 1979, 101, 7547.
(13) Hu, X. Chem. Sci. 2011, 2, 1867.
(14) Jones, G. D.; Martin, J. L.; McFarland, C.; Allen, O. R.; Hall, R.
E.; Haley, A. D.; Brandon, R. J.; Konovalova, T.; Desrochers, P. J.;
Pulay, P.; Vicic, D. A. J. Am. Chem. Soc. 2006, 128, 13175.
D
dx.doi.org/10.1021/ja408151h | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX