Synthesis and characterization of three-coordinate Ni(III)-imide complexes

Article abstract of DOI:10.1021/ja2024993

A new family of low-coordinate nickel imides supported by 1,2-bis(di-tert-butylphosphino)ethane was synthesized. Oxidation of nickel(II) complexes led to the formation of both aryl- and alkyl-substituted nickel(III)-imides, and examples of both types have

Full text of DOI:10.1021/ja2024993

ARTICLE
pubs.acs.org/JACS
Synthesis and Characterization of Three-Coordinate
Ni(III)-Imide Complexes
Vlad M. Iluc,† Alexander J. M. Miller,† John S. Anderson,† Marisa J. Monreal,‡ Mark P. Mehn,§ and
Gregory L. Hillhouse*,†
†Gordon Center for Integrative Science, Department of Chemistry, University of Chicago, Chicago, Illinois 60637, United States
‡Department of Chemistry & Biochemistry, University of California, Los Angeles, California 90095, United States
§Arnold and Mabel Beckman Laboratories of Chemical Synthesis, Division of Chemistry and Chemical Engineering,
California Institute of Technology, Pasadena, California 91125, United States
S Supporting Information
b
ABSTRACT: A new family of low-coordinate nickel imides
supported by 1,2-bis(di-tert-butylphosphino)ethane was synthe-
sized. Oxidation of nickel(II) complexes led to the formation of
both aryl- and alkyl-substituted nickel(III)-imides, and examples
of both types have been isolated and fully characterized. The aryl
substituent that proved most useful in stabilizing the Ni(III)-
imide moiety was the bulky 2,6-dimesitylphenyl. The two Ni(III)-
imide compounds showed different variable-temperature mag-
netic properties but analogous EPR spectra at low temperatures.
To account for this discrepancy, a low-spin/high-spin equilibrium
was proposed to take place for the alkyl-substituted Ni(III)-imide complex. This proposal was supported by DFT calculations. DFT
calculations also indicated that the unpaired electron is mostly localized on the imide nitrogen for the Ni(III) complexes. The results
of reactions carried out in the presence of hydrogen donors supported the findings from DFT calculations that the adamantyl
substituent was a significantly more reactive hydrogen-atom abstractor. Interestingly, the steric properties of the 2,6-dimesitylphenyl
substituent are important not only in protecting the NidN core but also in favoring one rotamer of the resulting Ni(III)-imide, by
locking the phenyl ring in a perpendicular orientation with respect to the NiPP plane.
’ INTRODUCTION
of a multiple bond between nickel and nitrogen. Herein we
present a comprehensive study on the synthesis and character-
ization of Ni(III)-imide complexes supported by the dtbpe ligand.
The use of a large substituent at the nitrogen, 2,6-(2,4,6-
Me3C6H2)2C6H3 (dmp), increased the stability of the corre-
sponding Ni(III)-imide. One of the Ni(III)-imide complexes
displayed interesting magnetic properties that were investigated
in both solution and the solid state. Accordingly, the alkyl- and
aryl-substituted Ni(III)-imides display distinct reactivity patterns:
the alkyl imide is a potent H-atom abstractor, reacting rapidly with
ethereal solvents, while the aryl imide is stable in ethers but does
accept a hydrogen atom from a tin hydride.
Although a plethora of nickel complexes has been studied and
characterized, isolating low-coordinate species that feature multi-
ple bonds between nickel and a light element is still challenging.
A strategy used by our group has been to employ the 1,2-bis(di-
tert-butylphosphino)ethane ligand (dtbpe), which increases the
steric congestion around the metal center. This ancillary ligand,
which is also a strong σ donor, is efficient in stabilizing electron-
deficient nickel centers.1ꢀ8 DFT calculations suggest that the
lowest unoccupied molecular orbital of the [(dtbpe)Ni2+] frag-
ment, a 12eꢀ species, is a py/dx2ꢀy2 hybrid of π symmetry. This
low-lying, empty orbital can interact with π donors to generate
multiple bonds with the metal center. On the basis of this
scaffold, three representative complexes, with the generic for-
mula (dtbpe)NidX (X = NR, PR, or CR2), were synthesized.1ꢀ3
Those unusual examples of terminal imide, phosphinidene, and
carbene ligands for a late-transition metal allowed insight into the
intermediates proposed in group-transfer reactions and in cou-
pling processes.4ꢀ6 They also showed unique reactivity in small-
molecule activation, including olefin functionalization.7,8
’ RESULTS AND DISCUSSION
Synthesis and Characterization of Ni(III)-Imide Com-
plexes. Prior to this work, only one example, the neutral
Ni(III)-imide (Me3NN)NidNAd (Me3NNH = N,N-(1,3-
dimethylpropanediylidene)dimesitylaniline), had been reported.9
Initial attempts to isolate a cationic Ni(III)-imide by one-electron
Although Ni(II)dX moieties were synthesized and character-
ized, Ni(III)-imide complexes had been predicted to be unstable
on the basis of the high oxidation state of nickel and the existence
Received: March 19, 2011
Published: July 28, 2011
r
2011 American Chemical Society
13055
dx.doi.org/10.1021/ja2024993 J. Am. Chem. Soc. 2011, 133, 13055–13063
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Synthesis of the Adꢀamantyl-Substituted
Ni(III)-Imide 3 (Aꢀ = B(ArF)4
)
Figure 2. Cyclic voltammogram of (dtbpe)NidNAd (1), 10 mM in
0.3 M [nBu4N][PF6] in DFB, Cp2Fe/Cp2Fe+ corrected.
Figure 3. Thermal-ellipsoid (50% probability) representation of the
complex cation of 3 (H atoms are omitted for clarity). Selected metrical
parameters: NiꢀN = 1.657(5), P(2)ꢀNi = 2.1992(12), P(1)ꢀ
Ni = 2.2041(12), and NꢀC(3) = 1.424(7) Å; P(2)ꢀNiꢀP(1) =
91.17(5), P(2)ꢀNiꢀN = 132.28(13), P(1)ꢀNiꢀN = 135.94(13),
and NiꢀNꢀC(3) = 165.2(4)°.
Figure 1. Thermal-ellipsoid (50% probability) representation of the
complex cation of 2 (only one of the two independent molecules from
the unit cell is shown; irrelevant H atoms are omitted for clarity).
Selected metrical parameters: Ni(1)ꢀN(1) = 1.771(4), P(1)ꢀNi(1) =
2.1609(14), P(4)ꢀNi(1) = 2.1726(14), N(1)ꢀC(50) = 1.459(7), and
N(1)ꢀH(1) = 0.98(4) Å; P(1)ꢀNi(1)ꢀP(4) = 90.04(5), P(1)ꢀ
Ni(1)ꢀN(1) = 127.56(16), P(4)ꢀNi(1)ꢀN(1) = 141.57(16), Ni(1)ꢀ
N(1)ꢀC(50) = 134.6(4), Ni(1)ꢀN(1)ꢀH(1) = 115(2), and C(50)ꢀ
H(1)ꢀN(1) = 107(2)°.
the electronic density map at 0.903 Å from nitrogen and
refined isotropically.
A cyclic-voltammetry study of 1 in THF showed complex
redox behavior. In order to avoid side reactions (Scheme 1) and
eliminate the presence of possible hydrogen-atom sources, the
solvent was replaced with 1,2-difluorobenzene (DFB), which,
although polar, is less reactive than THF. In DFB, the cyclic
voltammogram of 1 (Figure 2) features two reversible waves,
which were assigned to the Ni(III)/Ni(II) and Ni(II)/Ni(I)
couples (ꢀ1.07 and ꢀ1.43 V, respectively; both values are
relative to Cp2Fe/Cp2Fe+). These observations indicated that
the isolation of a Ni(III)-imide species might be achieved in the
absence of reactive solvents.
The oxidation of 1 was repeated in extensively purified DFB.
Extreme caution was used to avoid the presence of any con-
taminant (including diethyl ether and THF). Consequently,
the absence of any possible hydrogen atom sources allowed
the isolation of a paramagnetic, purple, crystalline powder,
[(dtbpe)NidNAd+][B(ArF)4ꢀ] (3), in good yield (Scheme 1).
Complex 3 was unstable in the presence of diethyl ether or THF,
in which it converted rapidly and quantitatively to the Ni(II)-
amide 2 (Scheme 1). It is important to note that 3 represents the
first ionic terminal Ni(III)-imide, and it shows high reactivity
versus ether solvents, unlike (Me3NN)NidNAd.9
oxidation of the Ni(II)-imide (dtbpe)NidNAd (1)10 with the
ferrocenium salt [Cp2Fe+][B(ArF)4ꢀ] (ArF = 3,5-bis-
(trifluoromethyl)phenyl) in a diethyl ether solution (Scheme 1)
led to the formation of [(dtbpe)NiNHAd+][B(ArF)4ꢀ] (2), a
Ni(II)-amide complex. A reactive Ni(III)-imide, 3, is proposed as
an intermediate, which converts to 2 in the presence of an
external hydrogen atom donor (Scheme 1). The same product, 2,
was also isolated from the reaction of the Ni(I) triflate
(dtbpe)NiOTf (4)11 with adamantyl azide in diethyl ether,
followed by [TfOꢀ]/[B(ArF)4ꢀ] anion exchange (Scheme 1).
A Ni(I) cationic azide complex was envisioned to give 3, which
led again to 2.
The Ni(II)-amide 2 was characterized by single-crystal
X-ray diffraction (Figure 1). Two independent molecules
were found in the unit cell. The metal center and the nitrogen
atom are trigonal planar (the sum of angles around Ni is
359.4° and around N is 356.0°). The NiꢀN average distance
of 1.765(5) Å is longer than the analogous distance in the
Ni(II)-imide (dtbpe)NidNAd (1.673(2) Å).10 The NiNC
angle of 135.77° (average value) indicates an sp2 hybridiza-
tion of the nitrogen atom and a possible π interaction be-
tween nickel and nitrogen. The amide hydrogen was found in
Complex 3 was characterized by single-crystal X-ray diffraction
(Figure 3). The metal center is trigonal planar, and the nitrogen
13056
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063

Journal of the American Chemical Society
ARTICLE
Scheme 2. Organoazide Route for the Synthesis of
(dtbpe)NidN(dmp) (7)
Figure 4. Cyclic voltammogram of (dtbpe)NidN(dmp) (7) at
150 mV/s, 10 mM in 0.3 M [nBu4N][PF6] in THF, Cp2Fe/Cp2Fe+
corrected.
atom lies in the P(2)ꢀNi(1)ꢀP(1) plane (the sum of angles
around Ni is 359.4°). The NiꢀN distance of 1.657(5) Å is only
0.016 Å shorter than in the parent Ni(II)-imide, (dtbpe)Nid
NAd (1, 1.673(2) Å), and is similar to the corresponding
distance in (Me3NN)NidNAd (NiꢀN = 1.662(2) Å).9 Other
metrical parameters were also similar for 3 and 1: the NimideꢀC
bond (1.424(7) Å for 3 and 1.417(3) Å for 1) and the NiꢀNꢀC
angle (164.2(5)° for 3 and 163.0(2)° for 1). The deviation of
this angle from linearity was also observed for an iron alkyl-
substituted imide, 159.6(3)° in (iPrPDI)FeNAd (iPrPDI = (2,6-
iPrC6H3NdCMe)2C5H3N),12 although it was 178.51(6)° in
(PhBPhP3)FeNAd (PhBPhP3 = [PhB(CH2PPh2)3]ꢀ).13,14 In
general, aryl-substituted imides tend to have a straight NiꢀNꢀC
arrangement as a consequence of the extended conjugation
provided by the phenyl ring, as was reported for (dtbpe)Nid
NMes (NiꢀNꢀC = 180.0°).10
Scheme 3. Synthesisof[(dtbpe)NidN(dmp)+][B(ArF)4ꢀ] (8)
ꢀ
(Aꢀ = B(ArF)4
)
were observed for this process, and 7 was obtained analytically
pure. 1H, 31P, and 13C spectroscopic studies of 7 indicated a C2v-
symmetric complex in solution, with magnetically equivalent
phosphorus nuclei (δ 119.23) matching the previously reported
data.16 Complex 7 is one of the few examples of selective photo-
lytic synthesis of an imide from a discrete azide complex.17,18 Such
conversions of azides to imides are not common for organome-
tallic complexes, likely because side-bound azides are rare.10,19ꢀ25
Organoazides, RN3, also undergo photochemical conversion to free
nitrenes.26
Faced with the instability of 3 in ethereal solvents, it was
reasoned that a larger substituent on the nitrogen might protect
the NiꢀN core better than the adamantyl group and decrease
its reactivity. We previously employed a similar strategy to
stabilize a nickel(II) phosphinidene, (dtbpe)NidP(dmp),2
after several failures to isolate a phosphinidene with a smaller
substituent.
Reaction of the Ni(0)-benzene adduct {(dtbpe)Ni}2(μ-C6H6)
15
(5) with (dmp)N3 led to the isolation, in high yield, of the
Ni(0) side-bound azide complex (dtbpe)Ni{η2-N3(dmp)} (6)
In order to determine the feasibility of oxidizing 7 to the
desired Ni(III)-imide, its electrochemical behavior was studied.
A cyclic voltammogram in THF (Figure 4, the use of DFB was
not necessary) featured two reversible waves, one at E1/2 = ꢀ0.76
V (vs Fc/Fc+), corresponding to the NiIII/NiII couple, and one at
E1/2 = ꢀ2.81 V, for the NiII/NiI redox couple.
(Scheme 2). Complex 6 was characterized by 1H, 31P, and 13
C
NMR, infrared spectroscopy, and elemental analysis. The NMR
spectra show characteristics of a planar, unsymmetrical Ni(0)
complex, with magnetically nonequivalent phosphorus nuclei
(JPP = 52 Hz). Its solution structure was assigned by analogy to
those of the previously characterized η2-bound azides (dtbpe)-
Ni(η2-N3Ad) and (dtbpe)Ni(η2-N3Mes).10
The bulky dmp substituent increased the thermal stability of 6;
both adamantyl and mesityl analogues decomposed when heated
for 24 h at 80 °C. Their thermal conversion to the corresponding
imides, 1 and (dtbpe)NidNMes, required additive (5%)
amounts of 5 to avoid the dissociation of the side-bound azide
to 5 and RN3.10 Complex 6 is thermally stable and can be heated
at 80 °C (C6D6) for long periods of time (24 h) without
conversion to the corresponding imide.
Accordingly (Scheme 3), the one-electron oxidation of ꢀ7,
carried out in diethyl ether at ꢀ35 °C with [Cp2Fe+][B(ArF)4
]
as the oxidant, led to the isolation, in high yield, of a purple,
paramagnetic powder of the Ni(III) complex [(dtbpe)Nid
N(dmp)+][B(ArF)4ꢀ] (8). The same product could be obtained
independently from 4 by the reaction with (dmp)N3, followed by
[TfOꢀ]/[B(ArF)4ꢀ] anion exchange (Scheme 3). Complex 8 is
stable in common solvents like diethyl ether, THF, and dichlor-
omethane. As expected because of its paramagnetic nature, 8 was
1
31P and 13C NMR silent, and its H NMR spectrum was very
Interestingly, crystals of 6 are sensitive to ambient light, which
prevented us from determining its solid-state structure but hinted
at photochemical reactivity. Exposure of 6 to 254 nm light
furnished the corresponding imide, (dtbpe)NidN(dmp) (7),
in quantitative yield after 12 h at room temperature. Com-
plex 7 has been previously prepared by H• abstraction from
(dtbpe)NiꢀNH(dmp).16 No decomposition or side products
broad and difficult to assign.
The single-crystal X-ray structure of 8 (Figure 5) revealed that
the cation has pseudo-C2v symmetry, with equal NiꢀP distances
(2.2314(11) and 2.2318(11) Å) and similar PꢀNiꢀN angles
(133.81(10)° and 135.82(10)°). TheNiꢀN distance is 1.674(3) Å,
about 1% longer than the analogous distance found in
[(dtbpe)NidNAd+] (3) or (Me3NN)NidNAd.9 A major
13057
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063

Journal of the American Chemical Society
ARTICLE
Figure 6. Plot of magnetic moment versus temperature for solid
[(dtbpe)NiNAd+][B(ArF)4ꢀ] (3). The solid line represents the fitted
data for the two spin states' equilibrium mixture.
Figure 5. Thermal-ellipsoid (50% probability) representation of the
complex cation of 8 (H atoms are omitted for clarity). Selected metrical
parameters: NiꢀN = 1.674(3), NiꢀP(1) = 2.2314(11), Niꢀ
P(2) = 2.2318(11), and NꢀC(31) = 1.359(5) Å; NiꢀNꢀC(31) =
178.4(3), P(1)ꢀNiꢀP(2) = 90.37(2), P(1)ꢀNiꢀN = 133.81(10), and
P(2)ꢀNiꢀN = 135.82(10)°.
difference between the two adamantyl-substituted Ni(III)-
imides (3 and (Me3NN)NidNAd) and 8 is the NiꢀNꢀC angle;
this angle is almost linear (178.4(3)°) in 8, perhaps indicative of
π delocalization that is not possible in 3 or (Me3NN)NidNAd.
A similar pattern was observed for the Ni(II)-imides, with the
aryl-substituted imides having a larger NiꢀNꢀC angle than the
alkyl-substituted complexes.1,10,16 The greater degree of electron
delocalization is also manifested in the slightly shorter NꢀC
bond found in 8 (1.359(5) Å) than in 3 (1.424(7) Å) and
(Me3NN)NidNAd (1.422(3) Å).9
Magnetic Susceptibility and EPR Spectroscopy Studies of
Ni(III)-Imide Complexes. The H NMR spectrum (CD2Cl2,
1
Figure 7. EPR spectrum of 3 (3.5 mM frozen glass in CH2Cl2, 3.7 K);
the solid line represents the simulated spectrum.
25 °C) of the adamantyl complex 3 showed broad resonances,
indicative of a paramagnetic species. The solution magnetic
moment (μeff = 2.79 μB), measured by the Evans method27,28
at room temperature, was too high for a low-spin, d7-Ni(III)
complex.29 Moreover, the solution magnetic moment increased
to 3.50 μB when the temperature was raised to 345 K. It is
noteworthy that the value measured at 345 K is lower than
A temperature dependence of the magnetic moment correspond-
ing to the two spin states was calculated by roughly fitting the data
with expected values of the magnetic moments. A value of 1.9 μB
was employed for the doublet state and 4.8 μB for the quartet state.
The molar fraction ([LS] and [HS]) of these two species was
calculated as a function of temperature; therefore, a temperature-
dependent equilibrium constant, K, could be found (eq 1).
that expected for a quartet state (spin only, μS = 3.87 μB; μS+L
=
5.20 μB; commonly observed, μobs = 4.30ꢀ5.20 μB).29
The temperature dependence of the magnetic moment ob-
μ2exp ꢀ μ2LS
1
½HSꢁ
½LSꢁ
½HSꢁ
served by H NMR spectroscopy for 3 was corroborated by a
K ¼
¼
,
where ½HSꢁ ¼
1 ꢀ ½HSꢁ
μ2HS ꢀ μ2LS
SQUID solid-state study. The magnetic moment increased from
1.6 μB (4 K) to 2.8 μB (300 K, Figure 6). The magnetic molar
susceptibility did not obey the CurieꢀWeiss law expected for
a mononuclear Ni(III) paramagnet (see the Supporting In-
formation for details). In order to explain the value of the
magnetic moment at low temperatures, two scenarios are
proposed: (1) the magnetic moment characterizes a dimeric
species, in which the unpaired electrons are antiferromagne-
tically coupled, and (2) a low-spin/high-spin (LS/HS) equi-
librium exists. Since crystallographic data (Figure 3) indicated
that 3 is a monomer in the solid state, the second interpreta-
tion is preferred.
ð1Þ
The enthalpy change, ΔH = 2.5 kJ/mol, and the entropy
change, ΔS = 1.5 J molꢀ1 ꢀ1, can be calculated by interpolat-
K
3
3
ing the linear part of the ln K vs 1/T plot with a first-order
equation (see Supporting Information). The enthalpy corre-
sponds to a small energy gap between the two states (about
210 cmꢀ1), hence the low concentration of the high-spin species;
the very small entropy is typical for a system in equilibrium (no
sharp spin crossover temperature). The resulting equilibrium
corresponds to a HS/LS ratio of about 1:4 at room temperature.
13058
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063

Journal of the American Chemical Society
ARTICLE
Figure 8. Plot of magnetic moment versus temperature for solid
[(dtbpe)NidN(dmp)+][B(ArF)4ꢀ] (8).
Figure 10. Diagram of free-energy differences (kcal/mol) between the
low- and high-spin states for [(dmpe)NidNPh+]. Left, the rotamer with
the imide phenyl ring perpendicular to the PNiP plane; right, the
rotamer with the imide phenyl ring coplanar to the PNiP plane.
aryl-substituted Ni(III)-imides: clean products were not ob-
tained upon oxidation of (dtbpe)NidNAr (Ar = 2,6-di-iso-
propylphenyl) or (dtbpe)NidNMes, possibly due to aryl
radical coupling.31
DFT calculations were performed on a phenyl-substituted
Ni(II)-imide model compound, (dmpe)NidNPh (dmpe =
Me2PCH2CH2PMe2), such that the role of steric factors is
minimized. The geometry-optimization results indicated that
the electronic barrier to rotation around the NꢀCipso bond is
3.5 kcal/mol. Using the same calculation parameters, we investi-
gated both an alkyl-substituted Ni(III)-imide model, [(dmpe)Nid
Figure 9. EPR spectrum (frozen glass 1.0 mM in toluene) for 8:
experimental, at 4 K (solid line), and simulated (dashed line).
This interpretation is supported by the fact that the EPR
X-band spectrum of 3 (3.5 mM in CH2Cl2) at 3.7 K is consistent
with the presence of a Ni(III) center with a low-spin, d7 elec-
tronic configuration (Figure 7).30 The spectrum exhibits a single
feature lacking a hyperfine structure and is nearly axial, with
g1 (2.11) > g2 (2.02). Another axial feature of low intensity was
observed in the half-field region (g ≈ 4).
NC(CH3)3 ], and an aryl-substituted model, [(dmpe)NidNPh+].
+
For the latter, both the high-spin and the low-spin energies of the
two rotamers were calculated.
The free-energy difference between the high-spin and the low-
spin isomers of the alkyl-substituted Ni(III) cation is 0.8 kcal/mol
(the calculated enthalpy for this process is ΔH = 0.6 kcal/mol).
For the aryl-substituted Ni(III) cation, the energy difference
between the two spin states is 4.8 kcal/mol for the rotamers with
the aryl ring perpendicular to the NiPP plane and 3.4 kcal/mol
for the rotamers with the ring coplanar (Figure 10). The trend
indicated by these values is consistent with the fact that the high-
spin state can be accessed for the adamantyl Ni(III)-imide at
room temperature, while only the low-spin state is observed for
the corresponding dmp complex, as indicated by the higher
energy difference between the low- and high-spin states of
Unlike 3, complex 8 has a magnetic moment of 1.9 μB,
measured by the Evans method27,28 in CD2Cl2 at room tem-
perature, indicative of one unpaired electron and a low-spin
Ni(III) species. The solid-state SQUID measurements con-
firmed the solution data. The magnetic moment (μeff = 1.8 μB,
4 K) did not vary greatly with the temperature (Figure 8), and the
magnetization followed the CurieꢀWeiss law for a mononuclear
paramagnetic species, with a Curie temperature, TC, of 2.78 K
and a magnetization constant of 0.357 emu.
+
Similar to what was observed for the Ni(III) cationic adaman-
tyl imide 3, the EPR spectrum of 8 (1 mM frozen glass in toluene,
Figure 9) shows a single feature, characterized by the parameters
g1 = 2.17, g2 = 2.06, and g3 = 1.97. No hyperfine coupling with the
nitrogen atom was observed, and the signal lost intensity with
increasing temperature.
DFT Calculations. The difference in the magnetic behavior of
the alkyl- (3) and the aryl-substituted (8) Ni(III)-imides may
arise from the electron-donating properties of the dmp group,
which can lower the energy of the bonding π orbitals and,
therefore, increase the LS/HS energy gap. In order to verify
this hypothesis, DFT studies were undertaken. In addition,
DFT studies might provide insight into whether the differ-
ence in the stability of the two complexes is a consequence of
steric or electronic factors. We were unable to isolate other
[(dmpe)NidNPh+] than of [(dmpe)NidNC(CH3)3 ]. Inter-
estingly, the free-energy difference between the low-spin state of
the rotamer with the phenyl ring perpendicular to the PNiP plane
and the high-spin state of the other rotamer was calculated to be
only 0.2 kcal/mol. The fact that a transformation between the
two was not observed experimentally indicates that the dmp
substituent does not rotate freely in solution.
DFT calculations were also used to investigate why the Ni(II)-
imide (dtbpe)NidNAd (1) and the corresponding Ni(III)-
imide [(dtbpe)NidNAd+][B(ArF)4ꢀ] (3) show, within experi-
mental error, identical metrical parameters. Three bonding
interactions (one having σ symmetry, an in-plane π bond, and
an out-of-plane π bond) and an antibonding π interaction
(HOMO) were observed for the model compound (dmpe)Nid
NtBu (Figure 11). A one-electron oxidation would decrease the
13059
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063

Journal of the American Chemical Society
ARTICLE
Scheme 4. Reactivity of 8 (Aꢀ = B(ArF)4
)
ꢀ
the nitrogen atom (69%) is higher than for the aryl complex
(53%), consistent with no change in the metrical parameters of
the Ni(II)- and the corresponding Ni(III)-imide compounds.
The observed high affinity of the cationic [(dtbpe)NidNAd+] (3)
for hydrogen atoms supports this interpretation. Similar reactiv-
ity was observed for the neutral Ni(III)-imide complex
(Me3NN)NidNAd,9 confirming the high radical character of
the nitrogen atom. Rhodium(I)- and iron(III)- and (IV)-imide
complexes have been shown to react with hydrogen donors at
the nitrogen center,32 and in the case of a cationic iron system a
detailed thermodynamic analysis suggests the NꢀH bond dis-
sociation energy to be ∼88 kcal/mol.32b Definitive conclusions
about the reactivity behavior of these complexes cannot be drawn
easily, however, because a Ni(III) center is also expected to show
high reactivity and because other reaction pathways involving a
radical localized on the metal cannot be excluded. In addition, the
fact that the alkyl-substituted imide complex undergoes a HS/LS
equilibrium can affect the reactivity, since high-spin complexes
are generally more reactive than low-spin analogues.
Figure 11. Frontier molecular orbitals for a (dmpe)NidNtBu, model
showing the orbital interactions between nickel and the nitrogen imide.
With DFT calculations and the reactivity of 3 with ethers both
suggesting that there is less spin density localized on the imide
nitrogen of the aryl-substituted Ni(III) complex 8 than for the
alkyl-substituted model 3, we set out to probe the hydrogen atom
abstraction ability of 8 with a more reactive H-donor substrate,
nBu3SnH (Scheme 4).33 Accordingly, the known cationic Ni(II)-
amide complex [(dtbpe)Ni-NH(dmp)+] (9) was obtained after
the reaction mixture was stirred for 12 h at room temperature in
diethyl ether. Complex 9 was also independently synthesized
from the Ni(II)-imide 7by protonation with [H(OEt2)2][B(ArF)4]
(Scheme 4). Interestingly, 7 can be accessed from 8 by reduction
with KC8 in diethyl ether (Scheme 4).
+
Figure 12. Spin-density plots for [(dmpe)NidNC(CH3)3 ] (top) and
[(dmpe)NidNPh+] models (bottom).
number of electrons in the antibonding orbital (HOMO) and
increase the bond order by 0.5. This bond-order increase should
be associated with a decrease in the NiꢀN distance in the
resulting Ni(III)-imide. If oxidation of the nitrogen center
(to form [Ni(II)ꢀAdN•ꢀ]) occurs, however, instead of the
metal center ([Ni(III)ꢀAdN2-]), then the unpaired electron
would be localized on the nitrogen atom of the imidyl radical,
AdN•ꢀ. Such an electronic structure would not impact drama-
tically the metrical parameters associated with the nickel center
and could explain the similarity observed between the solid-state
structures of (dtbpe)NidNAd (1) and [(dtbpe)NidNAd+]
[B(ArF)4ꢀ] (3).
’ CONCLUSIONS
A new family of amides and imides supported by the bulky
dtbpe ligand and containing a nickel center in different oxidation
states was synthesized and characterized. Oxidation of Ni(II)-
imides allowed the synthesis of the corresponding Ni(III)-imide
cations. Two such complexes were described, an alkyl- (3) and an
aryl-substituted (8) imide. The 2,6-dimesitylphenyl substituent
(dmp) proved useful in stabilizing the Ni(III)-imide complex 8.
The two complexes 3 and 8 showed different variable-tempera-
ture magnetic behaviors but analogous EPR spectra at low
temperatures. The effective magnetic moment for 3 increased
with increasing temperature, while it remained constant for 8. In
order to account for this difference, a low-spin/high-spin
Consequently, an investigation of the electronic structure for
the alkyl- and aryl-substituted Ni(III)-imide complexes (Figure 12)
shows that the spin density is divided between the nickel and the
nitrogen centers. For the alkyl imide model, the spin density on
13060
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063

Journal of the American Chemical Society
ARTICLE
equilibrium was proposed to take place for 3 but not for 8. That
proposal was supported by DFT calculations. DFT calculations
were also used to explain the similarity between the metrical
parameters observed for the adamantyl-substituted Ni(II)- and
Ni(III)-imides. Calculations on the model complex (dmpe)Nid
NtBu showed that three bonding interactions (one having σ
symmetry, an in-plane π bond, and an out-of-plane π bond) and
an antibonding π interaction (HOMO) constitute the frontier
orbitals, resulting in a bond order of 2 for the NiꢀN moiety.
Removal of an electron from the HOMO would result in the
shortening of the nickelꢀnitrogen distance if the unpaired
electron is localized on the nickel center, but it would not induce
a change in metrical parameters if the unpaired electron is
localized on the imide nitrogen. Calculations on an alkyl-imide
Ni(III) model complex indicated that the spin density was shared
by nickel and the imide nitrogen, with a large component on the
nitrogen atom. A similar situation was observed for the aryl-imide
Ni(III) simplified model complex, but with a diminished com-
ponent of the spin density on the nitrogen atom. Reactivity
studies in the presence of hydrogen donors supported the results
of DFT calculations and implied that the alkyl-substituted imide
3 is a much more potent H-atom abstractor than the aryl
derivative 8. Interestingly, the steric properties of the dmp
substituent are important not only in protecting the NidN core
but also in favoring one rotamer of the Ni(III)-imide; this effect is
the consequence of locking in the phenyl ring to be perpendicular
with respect to the NiPP plane. When the aryl ring is in the NiPP
plane (in conjugation with the NiꢀN π bond), radical deloca-
lization into the aromatic ring results in arylꢀaryl coupling.31
That conformation was correlated with a higher energy differ-
ence between the low-spin and the high-spin states than in the
case of the alkyl-substituted Ni(III)-imide complex.
a Nicolet 20-SXB spectometer with a TGS detector. Solution mag-
netic susceptibilities were calculated following the Evans method27,28
using trimethylsilyl ether as an internal standard. Elemental analyses
were performed by Columbia Analytics (Tucson, AZ) or by Midwest
Microlab (Indianapolis, IN). Cyclic voltammetry experiments were
carried out on an Eco-Chemie Autolab potentiostat using a 0.3 M
THF or difluorobenzene (DFB) solution of [nBu4N][PF6] as the
electrolyte. DFB was refluxed over CaH2 and distilled under nitrogen.
A ceramic patterned electrode with a platinum working electrode and
Ag/AgCl reference electrode from Pine Research Instrumentation was
used for data collection. Electrochemical response was recorded using an
Eco-Chemie Autolab potentiostat (pgstat20) and GPES 4.3 software.
The IR correction drop was not utilized, as there was no significant
resistance in solution. The spectra were recorded under a N2 atmo-
sphere, and for 1 the drybox was purged extensively to avoid any
exposure to Et2O or THF residual vapors. EPR spectra were collected on
a Bruker EMX EPR spectrometer. Magnetic susceptibility measure-
ments were carried out on batches obtained independently until at least
two different experiments gave superimposable results; the magnetic
susceptibility measurements were recorded using a SQUID magnet-
ometer at 5000 G. X-ray diffraction data were collected on a Siemens
platform goniometer with a charged-coupled device detector. Structures
were solved by direct methods using SHELXTL.38 All atoms were
refined anisotropically, and hydrogen atoms were placed in calculated
positions unless specified otherwise. Tables with atomic coordinates and
equivalent isotropic displacement parameters are available in the Sup-
porting Information. The crystals were coated with oil (STP Oil
Treatment) on a glass slide, which was brought outside the glovebox.
Synthesis of [(dtbpe)Ni-NHAd+][B(ArF)4ꢀ] (2). Method A
from (dtbpe)NidNAd (1) by Oxidation in Et2O. In a typical experiment,
a scintillation vial was charged in the drybox with a stirring bar, 40 mg
(0.076 mmol) of (dtbpe)NidNAd (1), and 5 mL of diethyl ether and
ꢀ
then cooled to ꢀ35 °C. A cold, blue solution of [Cp2Fe+][B(ArF)4
]
(80 mg, 0.076 mmol) in diethyl ether was added to the first solution, and
the reaction mixture was stirred for 1 h, dried under reduced pressure,
and washed with petroleum ether. The products were extracted with
diethyl ether, and green crystals were obtained by layering this solution
with petroleum ether (96 mg, 80% yield).
’ EXPERIMENTAL SECTION
Unless noted otherwise, all operations were performed under a
purified nitrogen atmosphere in a standard MBraun Lab Master 130
drybox or under an argon atmosphere using high-vacuum and Schlenk
techniques. Hexanes, petroleum ether, and toluene were dried by
passage through activated alumina and Q-5 columns under nitrogen.34
CH2Cl2 was dried by distillation from CaH2 or passage twice through
activated alumina columns. THF was distilled from a dark-purple THF
solution of sodium benzophenone ketyl. Anhydrous diethyl ether was
stirred for 24 h over metallic sodium and filtered under nitrogen through
an activated alumina plug. Benzene was refluxed under nitrogen over
CaH2 and distilled. CD2Cl2 and C6D6 were purchased from Cambridge
Isotope Laboratories, degassed, dried over CaH2, transferred under
vacuum, and stored over 4 Å molecular sieves. The 4 Å molecular sieves,
alumina, silica, and Celite were dried under dynamic vacuum overnight
at 180 °C. Q-5 was activated by heating at 200 °C under a 5% H2-in-N2
atmosphere. Unless noted, chemicals were purchased from commercial
sources and used without further purification. KC8,35 (dmp)N3,15
[Cp2Fe][B(3,5-(CF3)2C6H3)4],36 and [(dtbpe)Ni]2(C6H6)37 were
prepared according to the literature methods. All NMR spectra were
recorded using a Bruker DRX-400 or DRX-500 spectrometer. 1H NMR
spectra were referenced to solvent residual peaks at δ 7.15 for C6D6 and
δ 5.32 for CD2Cl2. 31P NMR spectra were measured using a 250 ppm
window and offset referenced to an external standard, 85% phosphoric
acid (δ 0.00). 13C NMR spectra were acquired using the pulse-Fourier
technique. Proton-decoupled spectra were collected with Waltz decou-
pling. 13C NMR spectra were referenced to solvent residual peaks,
δ 128.0 for C6D6 and δ 53.8 for CD2Cl2. Infrared spectra were measured
as Fluorolube-S20 or Nujol mulls between CaF2 or KBr plates using
Method B from [(dtbpe)NidNAd+][B(ArF)4ꢀ] (3)). First, 50 mg (0.036
mmol) of 3 was dissolved in 5 mL of Et2O or THF. The reaction solution
was stirred at room temperature for 30 min. Removal of the solvent under
reduced pressure yielded 50 mg of analytically pure 2 (100% yield).
For 2: 1H NMR (22 °C, 500 MHz, CD2Cl2) δ 8.21 (br, 1H, NH),
7.73 (s, 8H, C6H3(CF3)2), 7.56 (s, 4H, C6H3(CF3)2), 2.00 (s, 3H, Ad),
1.79ꢀ1.58 (m, 12H, Ad), 1.41 (d, 36H, C(CH3)3), 1.3 (m, 4H, C2H4);
31P{1H} NMR (22 °C, 161.97 MHz, CD2Cl2) δ 117.0 (s); 13C{1H}
NMR (22 °C, 100.61 MHz, CD2Cl2) δ 160.31 (q, CAr), 135.11 (s, CAr),
130.21 (s, CAr), 129.41 (m, CF3), 128.90 (t, CAr), 48.1 (s, CAd), 38.3
(t, CAd), 36.2 (s, CAd), 36.38 (d, C(CH3)3), 35.21 (d, C(CH3)3), 32.4
(s, CAd), 30.52 (s, C(CH3)3), 30.06 (s, C(CH3)3), 22.3 (m, CH2); IR
3224 cmꢀ1 (m, νNH). Anal. Calcd for C60H68F24BNNiP2: C, 51.82; H,
4.93; N, 1.01. Found: C, 51.72; H, 4.81; N, 0.91.
Synthesis of [(dtbpe)NidNAd+][B(ArF)4ꢀ] (3). In a scintilla-
tion vial in the drybox, crystals of (dtbpe)NidNAd (1; 38 mg, 0.0795
mmol) were dissolved in 2 mL of DFB, and the dark-red solution was
cooled to ꢀ35 °C. To the cold solution of 1 was added a cold, dark-blue
DFB (2 mL) solution of [Cp2Fe+][B(ArF)4ꢀ] (70 mg, 0.063 mmol).
The reaction mixture darkened to gray and was stirred for 15 min. The
mixture was then dried under reduced pressure and washed with
petroleum ether (never exposed to diethyl ether) to extract ferrocene.
The remaining dark-gray solid 3 was extracted with 1.5 mL of DFB. The
resulting gray solution was layered with 3 mL of petroleum ether (diethyl
ether-free) and cooled overnight to yield 82 mg (0.0590 mmol, 75%) of
dark crystals of paramagnetic 3 (purple under microscope).
13061
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063

Journal of the American Chemical Society
ARTICLE
For 3: 1H NMR (22 °C, 500 MHz, CD2Cl2) δ 7.72 (s, 8H, B(ArF)4),
7.56 (s, 4H, B(ArF)4), 5.32 (br s, Δν1/2 = 349 Hz, 36H, ꢀC(CH3)3),
3.42 (br s, Δν1/2 = 236 Hz, 8H, Ad), 0.08 (br s, Δν1/2 = 104 Hz, 5H, Ad),
ꢀ1.25 (br s, Δν1/2 = 179 Hz, 4H, -C2H4-); mp =160ꢀ165 °C. Anal.
Calcd for C60H67F24BNNiP2: C, 51.86; H, 4.86; N, 1.01. Found: C,
51.78; H, 4.84; N, 0.93.
Method B from (dtbpe)NiOTf (4) and (dmp)N3). A 20-mL scintilla-
tion vial was charged with 52 mg (0.1 mmol) of (dtbpe)NiOTf (7) and
5 mL of Et2O and cooled to ꢀ35 °C. To the cold suspension was
added dropwise, with constant stirring, a cold solution of (dmp)N3
(35.5 mg, 0.1 mmol in 3 mL of Et2O at ꢀ35 °C) over an interval of
5 min. The mixture was stirred for an additional 10 min, and a solution of
Na[B(ArF)4ꢀ] (89 mg, 0.1 mmol) in 3 mL of Et2O was added at once.
The mixture was warmed to room temperature and stirred for 30 min.
After removal of the volatiles under reduced pressure, the residue was
extracted with diethyl ether and filtered. Pure crystalline 8 was isolated
by cooling to ꢀ35 °C a concentrated solution of 8 in Et2O layered with
equal volumes of n-pentane (115 mg, 75% yield).
Synthesis of (dtbpe)Ni{η2-N3(dmp)} (6). To a suspension
of {(dtbpe)Ni}2(C6H6) (5; 166.4 mg, 0.2 mmol) in 5 mL of n-pentane
was added a solution of (dmp)N3 (142.2 mg, 0.4 mmol) in 5 mL of
n-pentane at once under stirring at room temperature. The orange-red
suspension turned yellow and formed a yellow precipitate. After being
stirred for 30 min at room temperature, the mixture was filtered and the
precipitate washed twice with 2 mL of n-pentane. The solids were dried
under reduced pressure, and a second crop was isolated by cooling the
combined mother liquors to ꢀ35 °C (combined yield 265 mg, 90%).
For 6: 1H NMR (20 °C, 400.13 MHz, C6D6) δ 7.23 (m, 3H,
C6H3Mes2), 6.89 (s, 4H, C6H2(CH3)3), 1.07 (m, 4H, C2H4), 1.01
(d, 18H, JHP = 12 Hz, C(CH3)3), 0.97 (d, 18H, JHP = 12 Hz, C(CH3)3);
31P{1H} NMR (20 °C, 202.47 MHz, C6D6) δ 96.64 (d, JPP = 52 Hz),
95.22 (d, JPP = 52 Hz); 13C{1H} NMR (20 °C, 125.77 MHz, C6D6)
δ 136.96 (s), 135.91 (s), 133.43(s), 129.26 (s), 128.93 (s), 127.92 (s),
For 8: 1H NMR (20 °C, 400.13 MHz, CD2Cl2) δ 10.49 (br s, Δ1/2
=
90 Hz, 7H, -C6H3(C6H2(CH3)3)2), 5.87 (br s, Δ1/2 = 140 Hz, 12H,
-C6H3(CF3)2), 4.51 (br s, Δ1/2 = 320 Hz, 36H, ꢀC(CH3)3), 3.213 (br s,
Δ1/2 = 180 Hz, 12H, -C6H2(CH3)3), 0.08 (br s, Δ1/2 = 130 Hz, 4H,
-C2H4-); magnetic moment (Evans, CD2Cl2, 20 °C) μeff = 2.2 μB. Anal.
Calcd for C74H77BF24NNiP2: C, 56.69; H, 4.95; N, 0.89. Found C,
56.40; H, 5.03; N, 0.82.
Synthesis of [(dtbpe)Ni-NH(dmp)+][B(ArF)4ꢀ] (9). Method
A from [(dtbpe)NidN(dmp)+][B(ArF)4ꢀ] (8) by Hydrogen Atom Addi-
tion. To a stirring solution of 8 (80 mg, 0.05 mmol in 5 mL of diethyl
ether at room temperature) was added at once a solution of HSnBu3
(15 mg, 0.05 mmol in 1 mL of Et2O). The mixture turned from dark
green to purple. After the mixture was stirred for 30 min at room
temperature, the volatiles were removed under reduced pressure, and
the residue was triturated three times with 5 mL of n-pentane. Pure,
crystalline 9 was isolated by cooling a concentrated diethyl ether solution
layeredwithan equal volume of n-pentane toꢀ35 °C (75 mg, 95% yield).
Method B from (dtbpe)NidN(dmp) (7) by Protonation . A 20-mL
scintillation vial was charged with 70.5 mg (0.1 mmol) of (dtbpe)Nid
N(dmp) (7) and 5 mL of Et2O, and the solution was cooled to ꢀ35 °C.
125.63 (s), 123.76 (s), 33.96 (d, JCP = 10 Hz, CH2), 33.72 (d, JCP
10 Hz, CH2), 30.06 (s, C(CH3)3), 30.00 (s, -C(CH3)3), 22.10 (d, JCP
=
=
10 Hz, C(CH3)3), 21.81 (d, JCP = 10 Hz, C(CH3)3), 21.33 (s,
o-C6H2(CH3)3), 21.28 (s, p-C6H2(CH3)3); IR (CaF2, Fluorolube)
2215 (m), 2107 (s), 2095 (s), 1451 (m), 1313 (m), 1604 (m), 1203
(s), 1182 (m), 1154 (s), 1102 (m), 889 (w), 851 (w), 801 (w) cmꢀ1
.
Yellow, single crystals were isolated by cooling a concentrated solution
of 6 in n-pentane at ꢀ35 °C. While crystals were inspected under a
microscope, vigorous gas evolution was observed. Several attempts were
made to collect crystallographic data on these crystals. Although unit cell
parameters were determined, extensive decomposition was observed
during the data collection at 100 K. Anal. Calcd for C42H65N3NiP2: C,
68.85; H, 8.94; N, 5.74. Found: C, 70.02; H, 8.69; N, 5.45.
Synthesis of (dtbpe)NidN(dmp) (7). Method A from (dtbpe)
Ni{η2-N3(dmp)} (6) by N2 Extrusion . A solution of 6 (220 mg,
0.3 mmol) in 20 mL of hexanes was placed in a glass Schlenk tube.
The tube was cooled at ꢀ196 °C in liquid nitrogen and evacuated for
30 min. The Schlenk tube was exposed to a high-energy ultraviolet
lamp (254 nm, 400 W) with air cooling for 12 h with stirring. The yellow
solution turned olive green. The volatiles were removed under reduced
pressure to yield analytically pure 7 (210 mg, 100% yield).
+
A cold solution of [H(OEt2)2 ][B(ArF)4ꢀ] (101 mg, 0.1 mmol in 3 mL
of Et2O at ꢀ35 °C) was added with constant stirring, and the mixture
was allowed to warm to room temperature. The volatiles were removed
under reduced pressure, and the dark-purple residue was triturated three
times with n-pentane. Pure, crystalline 9 was isolated by cooling a
concentrated diethyl ether solution layered with an equal volume of
n-pentane to ꢀ35 °C (140 mg, 90% yield). The purity of the product
was verified by NMR spectroscopy, and the data were compared with
previously reported data.16
X-ray Crystal Structure of 2. X-ray-quality crystals were obtained
by slow crystallization at ꢀ35 °C from a concentrated Et2O solution
layered with pentanes. A 0.05 ꢂ 0.03 ꢂ 0.02 mm3, purple block was
chosen and mounted on the diffractometer. Two independent mole-
cules are present in the unit cell, and some of the CF3 groups present
some degree of thermal disorder; this disorder was not modeled. A total
of 40 000 reflections (ꢀ22 e h e 22, ꢀ24 e k e 24, ꢀ26 e l e 27)
were collected at T = 100(2) K with θmax = 28.27°, of which 28 323 were
unique (Rint = 0.0514). The residual peak and holeꢀelectron density
were 2.850 and ꢀ1.057 e Åꢀ3. The least-squares refinement converged
normally with residuals of R1 = 0.0981 (I > 2σ(I)) and GOF of 1.119.
Crystal and refinement data for 2: C60H68BF24NiP2, space group P1, a =
16.6160(19) Å, b = 19.441(2) Å, c = 20.680(2) Å, R = 95.168°, β =
100.073(2)°, γ = 103.594(2)°, V = 6332.8(12) Å3, Z = 4, μ = 0.466
mmꢀ1, F(000) = 2856, R1 = 0.1379, wR2 = 0.2238 (based on all data).
X-ray Crystal Structure of 3. X-ray-quality crystals were obtained
by slow crystallization at ꢀ35 °C from a concentrated 1,2-C6H4F2
solution layered with pentanes. A 0.3 ꢂ 0.5 ꢂ 0.6 mm3, dark-purple
block was chosen and mounted on the diffractometer. A total of 32 564
reflections (ꢀ16 e h e 21, ꢀ19 e k e 18, ꢀ27 e l e 27) were
collected at T = 100(2) K with θmax = 25.00°, of which 11 176 were
unique (Rint = 0.0381). The residual peak and holeꢀelectron density
were 0.685 and ꢀ0.351 e Åꢀ3. The least-squares refinement converged
Method B from [(dtbpe)NidN(dmp)+][B(ArF)4ꢀ] (8) by Reduction.
To a cold solution of 8 (158 mg, 0.1 mmol) in 5 mL of Et2O at ꢀ35 °C
was added dropwise a cold suspension of KC8 (13.5 mg, 0.1 mmol, 2 mL
of Et2O, ꢀ35 °C) over an interval of 5 min. The black suspension
was warmed to room temperature and stirred for an additional 30 min.
After removal of the volatiles under reduced pressure, the residue was
extracted with hexanes, filtered through a plug of Celite, and concen-
trated. Cooling of this concentrated solution to ꢀ35 °C yielded
analytically pure 7 as green crystals (65 mg, 90% yield). The purity of
the product was verified by NMR spectroscopy, and the data were
compared to the previously reported data.16
Synthesis of [(dtbpe)NidN(dmp)+][B(ArF)4] (8). Method A
from (dtbpe)NidN(dmp) (7) by Oxidation. To a cold solution of 7
(141 mg, 0.2 mmol) in 5 mL of Et2O at ꢀ35 °C was added dropwise a
cold suspension of [Cp2Fe+][B(ArF)4ꢀ] (210 mg, 0.2 mmol, 5 mL of
Et2O, ꢀ35 °C) over an interval of 5 min. After the addition was finished,
the mixture was warmed to room temperature and stirred for another
30 min. The volatiles were removed under reduced pressure, and the
residue was triturated three times with 10 mL of pentanes. The isolated
solids were dried, and analytically pure 8 was isolated by cooling a
concentrated diethyl ether solution layered with an equal volume of
n-pentane to ꢀ35 °C (280 mg, 90% yield).
13062
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063

Journal of the American Chemical Society
ARTICLE
normally with residuals of R1 = 0.0717 (I > 2σ(I)) and GOF of 1.236.
Crystal and refinement data for 3: C60H67BF24NiP2, space group P21/n,
a = 17.736(5) Å, b = 16.476(4) Å, c = 23.481(6) Å, β = 112.082(4)°,
V = 6358.5(3) Å3, Z = 4, μ = 0.465 mmꢀ1, F(000) = 2852, R1 = 0.0837,
wR2 = 0.1500 (based on all data).
(9) Kogut, E.; Wiencko, H. L.; Zhang, L.; Cordeau, D. E.; Warren,
T. H. J. Am. Chem. Soc. 2005, 127, 11248–11249.
(10) Waterman, R.; Hillhouse, G. L. J. Am. Chem. Soc. 2008, 130,
12628–12629.
(11) Iluc, V. M.; Miller, A. J. M.; Hillhouse, G. L. Chem. Commun.
2005, 5091–5093.
X-ray Crystal Structure of 8. X-ray-quality crystals were obtained
byslow crystallization atꢀ35 °C froma concentratedpentanessolution. A
0.06 ꢂ 0.05 ꢂ 0.05 mm3, olive-green block was chosen and mounted on
the diffractometer. A total of 41 170 reflections (ꢀ18 e h e 18,
ꢀ17 e k e 10, ꢀ49 e l e 52) were collected at T = 100(2) K with
θmax = 28.28°, of which 16471 were unique (Rint = 0.0774). The residual
(12) Bart, S. C.; Lobkovsky, E.; Bill, E.; Chirik, P. J. J. Am. Chem. Soc.
2006, 128, 5302–5303.
(13) Brown, S. D.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003,
125, 322–323.
(14) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 1913–1923.
(15) Gavenonis, J.; Tilley, T. D. J. Am. Chem. Soc. 2002, 124,
8536–8537.
(16) Iluc, V. M.; Hillhouse, G. L. J. Am. Chem. Soc. 2010, 132,
15148–15150.
(17) Proulx, G.; Bergman, R. G. J. Am. Chem. Soc. 1995, 117,
peak and holeꢀelectron density were 0.775 and ꢀ0.439 e Åꢀ3
.
The least-squares refinement converged normally with residuals of
R1 = 0.0714 (I > 2σ(I)) and GOF of 0.972. Crystal and refinement data
for 8: C74H77BF24NNiP2, space group P21/n, a = 14.2589(12) Å,
b = 13.0889(11) Å, c = 39.706(3) Å, β = 90.907(2)°, V = 7409.5(11) Å3,
Z = 4, μ = 0.408 mmꢀ1, F(000) = 3228, R1 = 0.0775, wR2 = 0.1433 (based
on all data).
6382–6383.
(18) Proulx, G.; Bergman, R. G. Organometallics 1996, 15, 684–692.
(19) Fickes, M. G.; Davis, W. M.; Cummins, C. C. J. Am. Chem. Soc.
1995, 117, 6384–6385.
(20) Guillemot, G.; Solari, E.; Floriani, C.; Rizzoli, C. Organometal-
lics 2001, 20, 607–615.
(21) Hanna, T. A.; Baranger, A. M.; Bergman, R. G. Angew. Chem.
Int. Ed 1996, 35, 653–655.
(22) Barz, M.; Herdtweck, E.; Thiel, W. R. Angew. Chem., Int. Ed.
1998, 37, 2262–2265.
(23) Dias, H. V. R.; Polach, S. A.; Goh, S.; Archibong, E. F.;
Marynick, D. S. Inorg. Chem. 2000, 39, 3894–3901.
(24) Albertin, G.; Antoniutti, S.; Baldan, D.; Castro, J.; Garcia-
Fontan, S. Inorg. Chem. 2008, 47, 742–748.
DFT Calculations. Gaussian 0339 was used for all reported calcula-
tions. The B3LYP (DFT) method was used to carry out the geometry
optimizations on the model compounds specified in text using the
LANL2DZ basis set. The validity of the true minima was checked by the
absence of negative frequencies in the energy Hessian. Tables with
atomic coordinates of the optimized geometries can be found in the
Supporting Information.
’ ASSOCIATED CONTENT
(25) Munro, G. A. M.; Pauson, P. L. J. Organomet. Chem. 1978, 160,
177–181.
S
Supporting Information. Experimental details for com-
b
pound characterizations, DFT calculation details, full crystal-
lographic descriptions (CIF), and complete ref 39. This material
is available free of charge via the Internet at http://pubs.acs.org.
(26) Gritsan, N. P.; Platz, M. S. Chem. Rev. 2006, 106, 3844–3896.
(27) Sur, S. K. J. Magn. Reson. 1989, 82, 169–173.
(28) Evans, D. F. J. Chem. Soc. 1959, 2003–2005.
(29) Carlin, R. L.; van Duyneveldt, A. In Magnetic Properties of
Transition Metal Compounds; Springer: Berlin, 1977.
(30) (a) Carlin, R. L. In Magneto-chemistry; Springer-Verlag: Berlin,
1986. (b) Henry, W. E. Phys. Rev. 1952, 88, 559.(c) Casimir, H. B. G.
Magnetism and Very Low Temperatures; Dover Publications: New York,
1961.
’ AUTHOR INFORMATION
Corresponding Author
g-hillhouse@uchicago.edu
(31) Bai, G.;Stephan, D. W.Angew. Chem., Int. Ed. 2007, 11, 1856–1859.
(32) (a) Buttner, T.; Geier, J.; Frison, G.; Harmer, J.; Calle, C.;
Schweiger, A.; Schonberg, H.; Gr€utzmacher, H. Science 2005,
307, 235–238. (b) Nieto, I.; Ding, F.; Bontchev, R. P.; Wang, H.; Smith,
J. M. J. Am. Chem. Soc. 2008, 130, 2716–2717. (c) King, E. R.; Hennessy,
E. T.; Betley, T. A. J. Am. Chem. Soc. 2011, 133, 4197–4123.
(33) For nBu3SnH, the Sn-H BDE is ∼74 kcal/mol: (a) Davies, A. G.
Organotin Chemistry, 2nd ed.; Wiley-VCH: Weinheim. 2004. For
O(CH2CH3)2, the C-H BDE is ∼93 kcal/mol: (b) McMillen, D. F.;
Golden, D. M. Annu. Rev. Phys. Chem. 1982, 33, 493–532.
(34) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518–1520.
(35) Schwindt, M. A.; Lejon, T.; Hegedus, L. S. Organometallics
1990, 9, 2814–2819.
’ ACKNOWLEDGMENT
This work was supported by the National Science Foundation
through grants CHE-0615274 and CHE-0957816 (to G.L.H.), a
Beckman Scholars Fellowship from the Arnold and Mabel
Beckman Foundation (to J.S.A.), and a NIH postdoctoral fellow-
ship GM-072291 (to M.P.M.).
’ REFERENCES
(1) Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc. 2001, 123,
4623–4624.
(2) Melenkivitz, R.; Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc.
2002, 124, 3846–3847.
(36) Chavez, I.; Alvarez-Carena, A.; Molins, E.; Roig, A.; Maniukiewicz,
W.; Arancibia, A.; Arancibia, V.; Brand, H.; Manriquez, J. M. J. Organomet.
Chem. 2000, 601, 126–132.
(3) Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc. 2002, 124,
9976–9977.
(37) Bach, I.; Porschke, K.; Goddard, R.; Kopiske, C.; Kruger, C.;
Rufinska, A.; Seevogel, K. Organometallics 1996, 15, 4959–4966.
(38) All software and sources of scattering factors are contained in
the SHELXTL (version 5.1) program library: Sheldrick, G. SHELXTL;
Bruker Analytical X-ray Systems: Madison, WI.
(4) Waterman, R.; Hillhouse, G. L. J. Am. Chem. Soc. 2003, 125,
13350–13351.
(5) Waterman, R.; Hillhouse, G. L. Organometallics 2003, 22,
5182–5184.
(6) Harrold, N. D.; Waterman, R.; Hillhouse, G. L.; Cundari, T. R.
J. Am. Chem. Soc. 2009, 131, 12872–12873.
(7) Kitiachvili, K. D.; Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem.
Soc. 2004, 126, 10554–10555.
(39) Frisch, M. J.; et al. Gaussian 03, revision D.02; Gaussian, Inc.:
Wallingford, CT, 2004.
(8) Curley, J. J.; Kitiachvili, K. D.; Waterman, R.; Hillhouse, G. L.
Organometallics 2009, 28, 2568–2571.
13063
dx.doi.org/10.1021/ja2024993 |J. Am. Chem. Soc. 2011, 133, 13055–13063