Heterobimetallic Ti/Co Complexes That Promote Catalytic N-N Bond Cleavage

Article abstract of DOI:10.1021/acs.inorgchem.5b01962

Treatment of the tris(phosphinoamide) titanium precursor ClTi(XylNPⁱPr₂)₃ (1) with CoI₂ leads to the heterobimetallic complex (η²-ⁱPr₂PNXyl)Ti(XylNPⁱPr₂)₂(μ-Cl)CoI (2). One-electron reduction of 2 affords (η²-ⁱPr₂PNXyl)Ti(XylNPⁱPr₂)₂CoI (3), which can be reduced by another electron under dinitrogen to generate the reduced diamagnetic complex (THF)Ti(XylNPⁱPr₂)₃CoN₂ (4). The removal of the dinitrogen ligand from 4 under vacuum affords (THF)Ti(XylNPⁱPr₂)₃Co (5), which features a Ti-Co triple bond. Treatment of 4 with hydrazine or methyl hydrazine results in N-N bond cleavage and affords the new diamagnetic complexes (L)Ti(XylNPⁱPr₂)₃CoN₂ (L = NH₃ (6), MeNH₂ (7)). Complexes 4, 5, and 6 have been shown to catalyze the disproportionation of hydrazine into ammonia and dinitrogen gas through a mechanism involving a diazene intermediate.

Full text of DOI:10.1021/acs.inorgchem.5b01962

Article
pubs.acs.org/IC
Heterobimetallic Ti/Co Complexes That Promote Catalytic N−N Bond
Cleavage
Bing Wu, Kathryn M. Gramigna, Mark W. Bezpalko, Bruce M. Foxman, and Christine M. Thomas*
Department of Chemistry, Brandeis University, 415 South Street, Waltham, Massachusetts 02454, United States
S
* Supporting Information
ABSTRACT: Treatment of the tris(phosphinoamide) titanium
precursor ClTi(XylNPiPr2)3 (1) with CoI2 leads to the
heterobimetallic complex (η2-iPr2PNXyl)Ti(XylNPiPr2)2(μ-
Cl)CoI (2). One-electron reduction of 2 affords (η2-iPr2PNXyl)-
Ti(XylNPiPr2)2CoI (3), which can be reduced by another
electron under dinitrogen to generate the reduced diamagnetic
complex (THF)Ti(XylNPiPr2)3CoN2 (4). The removal of the
dinitrogen ligand from 4 under vacuum affords (THF)Ti-
(XylNPiPr2)3Co (5), which features a Ti−Co triple bond.
Treatment of 4 with hydrazine or methyl hydrazine results in
N−N bond cleavage and affords the new diamagnetic complexes
(L)Ti(XylNPiPr2)3CoN2 (L = NH3 (6), MeNH2 (7)). Complexes 4, 5, and 6 have been shown to catalyze the
disproportionation of hydrazine into ammonia and dinitrogen gas through a mechanism involving a diazene intermediate.
metals.6 The zwitterionic ZrIVCo−I heterobimetallic complexes
INTRODUCTION
■
(THF)Zr(MesNPiPr2)3CoN2 (Mes = 2,4,6-trimethylphenyl)
The transition metal-catalyzed production of ammonia from
and (THF)Zr(MesNPiPr2)3Co feature metal−metal double
dinitrogen under mild conditions remains one of the greatest
and triple bonds, respectively,6b,7 and have been shown to
challenges in small molecule activation chemistry.1 Extensive
activate alkyl halides,8 CO2,9 diaryl ketones,10 and organic
studies have been carried out in efforts to understand the
mechanisms of the two well-established nitrogen fixation
processes, namely, the Haber−Bosch process and the biological
fixation of dinitrogen by nitrogenase enzymes. Several recent
azides11 via one-, two-, or four-electron transformations. In
addition, the reaction of hydrazine and its derivatives with
(THF)Zr(MesNPiPr2)3CoN2 results in cleavage of an N−H
bond via a one-electron dissociative electron transfer process
reports have elegantly demonstrated the catalytic production of
(Scheme 1).12
ammonia from dinitrogen using well-defined iron and
molybdenum complexes.2 The operative catalytic cycles are
Scheme 1
thought to involve a Chatt-type “distal” mechanism in which
one transition metal atom coordinates N2, and subsequent
reduction and protonation steps first occur at the distal
nitrogen atom (MnNN → Mn+3N + NH3 → Mn + NH3).
However, detailed spectroscopic studies of the nitrogenase
enzyme have revealed intermediates involving N2H2 and N2H4
bound to the FeMo cofactor in the middle or late stages of the
catalytic cycle in the N2-fixation process (MnNN →
MnHNNH → MnH2NNH2 → Mn + NH3), and both
hydrazine and diazene are substrates for nitrogenase.3 Within
this context, reductive cleavage of the N−N bonds in hydrazine
to form ammonia is of considerable importance and there are a
number of transition metal complexes known to promote the
transformation stoichiometrically4 or catalytically.5 In some
cases, the transition metal species mediate the disproportiona-
tion of hydrazine into ammonia and dinitrogen.
From both a steric and an electronic perspective, zirconium
and its lighter congener titanium are quite different. In light of
titanium’s ability to undergo one-electron redox processes, we
have recently extended our studies to heterobimetallic Ti/M
complexes. A series of bis(phosphinoamide) Ti−Co complexes
featuring Ti/Co multiple bonds was recently reported, and the
reduced complex ClTi(XylNPiPr2)2CoPMe3 (Xyl = 3,5-
Our group has been investigating heterobimetallic complexes
featuring polar metal−metal multiple bonds in an effort to
promote multielectron redox processes and/or uncover new
stoichiometric and catalytic transformations of small molecules
facilitated by cooperative reactivity involving both transition
Received: August 25, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
dimethylphenyl) was shown to promote the reductive coupling
of aryl ketones into alkenes.13 Herein, we build upon this work
by synthesizing a series of heterobimetallic Ti/Co complexes
featuring metal−metal bonds. To showcase the enhanced
reactivity of the Ti/Co combination compared to Zr/Co, we
explore the reactivity of reduced Ti/Co complexes toward
hydrazine, demonstrating the ability of these complexes to
cleave N−N bonds and catalytically disproportionate hydrazine
to ammonia and dinitrogen.
were carried out to investigate both the one- and two-electron
reduced products.
One-electron reduction of 2 with KC8 led to the formation of
a new paramagnetic complex (η2-iPr2PNXyl)Ti-
(XylNPiPr2)2CoI (3) shown in Scheme 2. Complex 3 adopts
an asymmetric geometry in which one of the phosphinoamide
ligands is bound η2 to the titanium center in the solid state (vide
infra). However, the 1H NMR spectrum of complex 3 in THF-
d8 shows six broad resonances, which is more consistent with a
1
C3-symmetric geometry in solution (Figure S6). The H NMR
RESULTS AND DISCUSSION
spectrum in C6D6 displays fewer and broader resonances
(Figure S5). We hypothesize that weak THF binding to the Ti
center of 3 favors a C3-symmetric structure in solution, but that
the phosphinoamide ligands are rapidly exchanging on the
NMR time scale in benzene via reversible phosphine
dissociation, resulting in coalescence behavior.15 Measurement
of the solution magnetic moment of 3 reveals an S = 1/2
ground state (μeff = 1.98 μB, C6D6). An analogous one-electron
reduced product has not been isolated for tris-
(phosphinoamide) Zr/Co complexes, which favor two electron
reductions.7a,14,16
■
Synthesis and Spectroscopic Characterization of 1−5.
We previous reported the inability to install three [MesN-
PiPr2]− ligands around a Ti center, presumably due to the steric
hindrance of the mesityl group.13 However, changing the N-aryl
group from 2,4,6-trimethylphenyl to 3,5-dimethylphenyl
permits the formation of the desired tris(phosphinoamide)
precursor ClTi(XylNPiPr2)3 (1) via treatment of TiCl4 with
three equivalents of in situ-generated Li[XylNPiPr2] (Scheme
2).
Complex 3 can be further reduced using additional KC8
under a dinitrogen atmosphere, and from this reaction the
diamagnetic complex (THF)Ti(XylNPiPr2)3CoN2 (4) was
isolated. Alternatively, complex 4 can be synthesized directly
by reducing 2 with 2.5 equiv of Na/Hg under dinitrogen
(Scheme 2). The 31P NMR spectrum of 4 shows one broad
resonance at 45.5 ppm, consistent with a C3-symmetric
environment. The ν(N2) stretch at 2084 cm−1 in the IR
spectrum of 4 is considerably higher in energy than the N2
vibrational frequency of the Zr/Co analogue (THF)Zr-
(XylNPiPr2)3CoN2 (ν(N2) = 2045 cm−1).14 This indication
of less π back-bonding from cobalt to dinitrogen suggests a less
electron rich cobalt center in 4 and, in turn, a stronger donation
from Co to the more Lewis acidic early metal. The N2 molecule
is so weakly bound to cobalt that it easily dissociates upon
exposure to vacuum or argon atmosphere to generate a new
coordinatively unsaturated product, (THF)Ti(XylNPiPr2)3Co
(5) (Scheme 2), as indicated by a dramatic color change from
red to green. The 31P NMR resonance of 5 is also diagnostic
and shifts from 45.5 to 39.6 ppm upon removal of the N2
ligand. Consistent with the weak binding of N2 to Co in 4, a
related heterobimetallic Ti/Co complex, CoTi[N(o-(NCH2P-
(iPr)2C6H4)3], was recently reported to not bind N2 at all.17
Structural Characterization of Complexes 2−5. Com-
plexes 1−5 were characterized in the solid state using single
crystal X-ray diffraction. The solid state structure of
monometallic complex 1 was determined and is shown in
Scheme 2
1
Figure S21. Consistent with the H NMR data, the solid state
structure of 2 features two bridging phosphinoamide ligands, a
terminal iodide ligand bound to cobalt, and one chloride ligand
bridging titanium and cobalt. The third phosphinoamide ligand
is bound η2 to titanium (Figure 1). The Ti−Co intermetallic
distance in 2 is 2.6530(4) Å, which corresponds to a “formal
shortness ratio” (the ratio of the metal−metal interatomic
distance to the sum of the single bond atomic radii of the two
metal ions, FSR18) of 1.07 and suggests a weak metal−metal
dative interaction (Table 1). One-electron reduction signifi-
cantly decreases the Ti−Co intermetallic distance to 2.2735(8)
Å and 2.2734(8) Å in the two independent molecules in the
asymmetric unit of complex 3, indicative of a metal−metal
double bond (FSR = 0.92). The Ti−Co bond length in
complex 3 is slightly longer than that in the previously reported
Treatment of 1 with 1 equiv of CoI2 afforded a
heterobimetallic paramagnetic Ti/Co complex (η2-iPr2PNXyl)-
Ti(XylNPiPr2)2(μ-Cl)CoI (2). The 1H NMR spectrum of
complex 2 displays ten distinct broad paramagnetically shifted
resonances between 23 and −10 ppm, suggesting an
asymmetric structure similar to the previously reported Zr/
Co analogue (η2-XylNPiPr2)Zr(XylNPiPr2)2(μ-I)CoI.14 The
solution magnetic moment of 2 is 3.20 μB, consistent with an
S = 1 ground state. Cyclic voltammetry of complex 2 reveals a
quasi-reversible reduction at −1.7 V and an irreversible
reduction at −2.4 V (Figure S29). Given that these reductive
features were well-separated, bulk chemical reductions of 2
B
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. Displacement ellipsoid (50%) representations of 4 and 5.
For clarity, all hydrogen atoms and solvate molecules are omitted.
Only one of two independent molecules in the asymmetric unit of 4
and 5 is shown.
Figure 1. Displacement ellipsoid (50%) representations of 2 and 3.
For clarity, all hydrogen atoms and solvate molecules are omitted.
Only one of two independent molecules in the asymmetric unit of 3 is
shown.
decrease in the distance between the two metal centers in 5 to
2.0252(5) Å and 2.0271(5) Å (FSR = 0.81) indicates an
increase in Ti−Co bond order upon removal of the N2 ligand.
The FSR of complex 5 compares well with other hetero-
bimetallic complexes from our group that have been assigned as
having a metal−metal triple bond, including (THF)Zr-
bis(phosphinoamide) complex [(μ-Cl)Ti(XylNPiPr2)2CoI]2
(2.2051(4) Å, FSR = 0.89), which was described as having a
Ti−Co double bond (Table 1).
Both 4 and 5 are C3-symmetric in the solid state with trigonal
bipyramidal geometries at the Ti centers if the metal−metal
bond is taken into consideration (Figure 2). The Co center in
complex 4 has a distorted trigonal bipyramidal geometry with
Co 0.60 Å above the plane of the three phosphorus atoms. In
the absence of the dinitrogen ligand in complex 5, the Co atom
moves 0.2 Å closer to the plane of phosphine donors but is
likely restricted from adopting a true trigonal monopyramidal
geometry by the orientation of the phosphorus lone pairs. The
Ti−Co distances in complex 4 are 2.2371(3) Å and 2.2268(3)
Å in the two molecules in the asymmetric unit, corresponding
to an FSR of 0.90 that is in line with the previously reported
Zr/Co analogue (THF)Zr(XylNPiPr2)3CoN2 (Table 1).14 The
(MesNPi P r2 )3 Co (FSR
=
0. 82)7 a and ClTi-
(XylNPiPr2)2CoPMe3 (FSR = 0.81).13 However, the Ti−Co
distance in 5 is significantly shorter than that of Lu’s recently
reported C3-symmetric CoTi[N(o-(NCH2P(iPr)2C6H4)3] com-
plex (2.1979(8) Å, FSR = 0.88), which was assigned a TiCo
double bond.17
Upon examining the metal−ligand distances in complexes
2−5, a general trend emerges. With each sequential reduction
event, the Co−P distances contract and the Ti−N distances
elongate (Table 1). This trend is consistent with increased
Co→P back-bonding and decreased N→Ti π donation as the
a
Table 1. Metal−Cobalt and Metal−Ligand Distances and FSRs of Complexes 2−7 and a Selection of Similar Heterobimetallic
M/Co Complexes (M = Ti, Zr)
b
distance (av)
compound
core
M−Co distance
FSR
Co−P
M−N
c
(η2-iPr2PNXyl)Ti(XylNPiPr2)2(μ-Cl)CoI (2)
(η2-XylNPiPr2)Zr(XylNPiPr2)2(μ-Br)CoBr14
(η2-iPr2PNXyl)Ti(XylNPiPr2)2CoI (3)
(TiCo)5+
(ZrCo)5+
(TiCo)4+
2.6530(4) Å
2.7602(3) Å
2.2735(8) Å
2.2734(8) Å
2.2051(4) Å
2.2371(3) Å
2.2268(3) Å
2.2461(3) Å
2.2703(5) Å
2.3683(9) Å
2.3615(9) Å
2.3778(5) Å
2.3854(5) Å
2.0252(5) Å
2.0271(5) Å
2.0234(9) Å
2.1979(8) Å
2.1468(6) Å
2.1395(7) Å
1.07
1.06
0.92
2.27 Å
2.28 Å
2.25 Å
1.96 Å
2.10 Å
1.98 Å
c
c
13
[(μ-Cl)Ti(XylNPiPr2)2CoI]2
(TiCo)4+
(TiCo)3+
0.89
0.90
2.24 Å
2.21 Å
1.96 Å
2.00 Å
(THF)Ti(XylNPiPr2)3CoN2 (4)
(H3N)Ti(XylNPiPr2)3CoN2 (6)
(TiCo)3+
(TiCo)3+
(ZrCo)3+
0.90
0.91
0.90
2.22 Å
2.21 Å
2.22 Å
1.99 Å
1.99 Å
2.15 Å
(MeH2N)Ti(XylNPiPr2)3CoN2 (7)
7a
(THF)Zr(MesNPiPr2)3CoN2
14
(THF)Zr(XylNPiPr2)3CoN2
(ZrCo)3+
(TiCo)3+
0.91
0.81
2.23 Å
2.24 Å
2.12 Å
2.03 Å
(THF)Ti(XylNPiPr2)3Co (5)
13
ClTi(XylNPiPr2)2CoPMe3
(TiCo)3+
(TiCo)3+
(ZrCo)3+
0.81
0.88
0.82
2.24 Å
2.26 Å
2.22 Å
1.99 Å
1.95 Å
2.19 Å
CoTi[N(o-(NCH2P(iPr)2C6H4)3]17
(THF)Zr(MesNPiPr2)3Co7a
a
b
FSR = DMCo/(R1M + R1Co).18 Average distances are within 0.01 Å of the individual Co−P and Ti−N distances in all molecules in the asymmetric
c
units. Only Ti−N distances associated with the bridging phosphinoamide ligands are included in the average value.
C
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
complexes become more reduced. The notable outlier is the
N2-free complex 5, whose average Co−P distance (2.24 Å) is
0.03 Å longer than N2-bound complex 4. This phenomenon
could possibly result from attenuation of the electron density at
Co by stronger metal−metal bonding or from the geometric
constraints that could weaken the overlap of the sp3-hybridized
phosphine lone pair with the Co center as it moves into the
plane of the three phosphorus atoms; however, it should be
noted that a similar phenomenon was not observed for the Zr/
Co analogues (Table 1).7a
EPR Spectroscopy. In contrast to previously reported Zr/
Co complexes, both metals in the series of Ti/Co complexes
are potentially redox active. Therefore, EPR spectroscopy was
used to evaluate the localization of the unpaired electron in the
S = 1/2 complex 3. The X-band EPR spectrum of complex 3
was collected in toluene glass at 3 K. An axial signal is observed
with g∥ = 2.12 and g⊥ = 2.01 (Figure 3) that can be simulated
features of related heterobimetallic compounds.7,13 A table
comparing selected calculated interatomic distances and angles
derived from DFT geometry optimizations with those
determined crystallographically is provided in Table S2,
showing good agreement.
Consistent with the Co−Ti multiple bond character
insinuated by the short metal−metal distance in 3, the
calculated frontier molecular orbital diagram of 3 reveals a
Co→Ti σ bond along with two polarized π bonds involving the
Co and Ti dxz and dyz orbitals (Figure S30). Further, as
suggested by the hyperfine coupling in the EPR spectrum of 3,
Mulliken population analysis predicts that significantly more of
the unpaired spin density resides on the Co atom (Mulliken
spin densities: 1.20 (Co), −0.44 (Ti), Figure 4B), consistent
with a TiIVCo0 formulation.
Figure 4. (A) Depictions and energies of the frontier molecular
orbitals of 5 derived from DFT calculations (BP86/LANL2TZ(f)/6-
311+G(d)/D95 V). (B) Calculated unpaired spin density surface of
complex 3. Mulliken spin densities: 1.20 (Co); −0.44 (Ti).
The calculated frontier molecular orbital diagrams of
complexes 4 and 5 are shown in Figures S31 and 4A,
respectively. In addition to a σ bond between the Co and Ti dz
2
Figure 3. Experimental (black) and simulated (red) X-band EPR
spectra for 3 (A) and Ti(XylNPiPr2)3 (B) in toluene glass obtained at
3 K (frequency = 9.43 GHz, modulation to 5 G, power = 0.6325 mW).
orbitals, two π bonds involving the Co and Ti dxz and dyz
orbitals are predicted to contribute to the metal−metal bonding
in both complexes. Similar to 5, the previously reported
b i s ( p h o s p h i n o a m i d e ) T i − C o c o m p l e x C l T i -
with both hyperfine coupling to the I = 7/2 59Co nucleus (A∥ =
442 MHz) and superhyperfine coupling to the I = 5/2 127I
nucleus (A⊥ = 150 MHz) (Figure 3A). The large hyperfine
coupling to 59Co suggests that the unpaired electron in 3 is
localized on the Co atom, consistent with a TiIVCo0
assignment. The g values and hyperfine coupling values are
consistent with similar phosphine-ligated S = 1/2 cobalt iodide
complexes in which the unpaired electron is cobalt-centered.19
For comparison, an EPR spectrum of the monometallic
tris(phosphinoamide) TiIII complex, Ti(XylNPiPr2)3,13 was
also collected. The EPR signal of Ti(XylNPiPr2)3 is isotropic
with much weaker coupling that is simulated as superhyperfine
coupling to three equivalent 14N nuclei (I = 1), resulting in a
seven-line splitting pattern (g = 1.98, A = 47 MHz) (Figure
3B).
13
(XylNPiPr2)2CoPMe3 adopts a (σ)2(π)4(Conb)4 electronic
configuration, which corresponds to a formal bond order of 3.
In the case of complex 4, competitive π back-bonding from Co
to the N2 ligand partially disrupts the metal−metal π bonding,
leading to a longer Ti−Co distance and estimated bond order
closer to 2.
Natural bond orbital (NBO) calculations were used to assess
the electronic structure of complexes 3−5. Wiberg bond indices
(WBIs) were also calculated and used to assess trends in
metal−metal bonding, although we have found that the WBIs
calculated for metal−ligand and metal−metal bonds are always
lower than the formal bond orders we would predict based on
structural parameters and idealized molecular orbital consid-
erations.20 As expected based on their similar Ti−Co bond
distances, the calculated WBIs of complexes 3 and 4 are similar
(1.14 and 1.18, respectively), which confirms that the additional
electron in complex 4 occupies a Co-based orbital that is
nonbonding with respect to Ti. Natural population analysis
(NPA) also reveals a decrease in the natural charge on the Co
atom upon reduction from 3 to 4, suggesting that the redox
Computational Studies of 3−5. To better understand the
metal−metal bonding in complexes 3−5, a theoretical
investigation was carried out using density functional theory
(DFT). The BP86 functional and a mixed LANL2TZ(f)/6-
311+G(d)/D95 V basis set were chosen based on the previous
success of this method in accurately predicting the structural
D
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
event is cobalt-centered and that complex 4 is best described
using a zwitterionic TiIVCo−I formalism. The Ti−Co WBI
increases upon removal of the N2 ligand in complex 5 (WBI =
1.60), and becomes nearly identical to that reported for
ClTi(XylNPiPr2)2CoPMe3 (WBI = 1.59) (Table 2), consistent
Scheme 3
Table 2. Computed Natural Charges and Wiberg Bond
Indices (WBIs) of Heterobimetallic Complexes
natural charges
Ti/Zr
Co
WBI
(η2-iPr2PNXyl)Ti(XylNPiPr2)2CoI (3)
(THF)Ti(XylNPiPr2)3CoN2 (4)
(THF)Ti(XylNPiPr2)3Co (5)
0.38
0.45
0.44
0.26
0.40
1.39
−0.38
−1.14
−0.66
−0.25
−0.64
−1.22
1.14
1.18
1.60
1.26
1.59
0.95
13
[(μ-Cl)Ti(XylNPiPr2)2CoI]2
13
ClTi(XylNPiPr2)2CoPMe3
7b
(THF)Zr(MesNPiPr2)3CoN2
with the complexes having the same TiCo triple bond order.
The natural charges on Ti and Co in 5 and ClTi-
(XylNPiPr2)2CoPMe3 are similar, and in both cases the
difference in the natural charge between the two metal atoms
is significantly smaller than in the ZrIV/Co−I complex
(THF)Zr(MesNPiPr2)3CoN2, suggesting a more covalent
metal−metal bond between the two first row transition metals,
Ti and Co (Table 2). While we have assigned formal oxidation
states here that are consistent with spectroscopic and
computational data, a more detailed spectroscopic study will
be needed to confirm these assignments.
Stoichiometric Reactions of 4 with Hydrazine and
Methylhydrazine. We previously reported that the activation
of the N−H bonds of hydrazine and its derivatives by
(THF)Zr(MesNPiPr2)3CoN2 affords a series of S = 1/2
zirconium hydrazido complexes (RNNH2 )Zr-
(MesNPiPr2)3CoN2 (R = H, Me, and Ph) via one-electron
dissociative electron transfer processes (Scheme 1).12 In
contrast, addition of hydrazine to 4 quantitatively generates a
1
new diamagnetic complex 6 (Scheme 3). Both the H and 31P
NMR spectra are indicative of a symmetric complex similar to
4, with an additional broad signal at −1.01 ppm and the notable
absence of the resonances corresponding to the bound THF
protons in the 1H NMR spectrum of 6. The presence of an N2
vibration in the IR spectrum of 6 (ν(N2) = 2062 cm−1)
confirms that dinitrogen remains bound to Co and the 22 cm−1
decrease compared to 4 (ν(N2) = 2084 cm−1) suggests a more
electron rich cobalt center in 6, as would result from weaker
bonding to a less Lewis acidic titanium center. Taken together,
these data indicate that the Ti-bound THF molecule has been
replaced by a more basic neutral ligand. As expected from the
spectroscopic results, the solid-state structure of 6 reveals that
the THF ligand in 4 was replaced by an ammonia molecule to
give (H3N)Ti(XylNPiPr2)3CoN2 (6) (Figure 5). Since the
oxidation state of the two metal centers remains unchanged in
this reaction, we speculate that hydrazine disproportionates to
form the bound ammonia molecule with concomitant
elimination of N2, based on the balanced equation depicted
in Scheme 3. While the N2 released from the reaction was not
quantified, its formation was verified via a control reaction
between N2-free complex 5 and N2H4 (vide infra). Similarly,
methylhydrazine is disproportionated by 4 to afford a 1:3
mixture of 6 and (MeH2N)Ti(XylNPiPr2)3CoN2 (7) (Scheme
3, Figure S20). As an alternative synthetic method, addition of
Figure 5. Displacement ellipsoid (50%) representation of 6. For
clarity, all solvate molecules and hydrogen atoms, with the exception of
those bound to ammonia, are omitted. The cobalt-bound dinitrogen
ligand is disordered over two positions, and only one is shown for
clarity.
ammonia or methylamine to 4 cleanly generates 6 or 7,
respectively, via displacement of THF.
The structures of complexes 6 and 7 have been confirmed by
X-ray crystallographic structure determination, and the
structures of 6 and 7 are shown in Figures 5 and S27,
1
respectively. As predicted by the H and 31P NMR spectra,
complex 6 is C3-symmetric in the solid state with an NH3 ligand
bound to titanium in place of THF. The hydrogen atoms of the
ammonia and methylamine ligands in 6 and 7 were located and
refined and are shown in Figures 5 and S27. The Ti−Co
interatomic distances in complexes 6 and 7 are slightly longer
than that of the THF-bound precursor 4 (2.2461(3) Å for 6
and 2.2703(5) Å for 7), suggesting similar metal−metal
multiple bonding (Table 1).
E
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Catalytic Disproportionation of Hydrazine. Inspired by
the stoichiometric reactions of 4 with hydrazines, we
investigated the activity of 4 as a catalyst for the
disproportionation of hydrazine at ambient temperature
without the addition of a reductant or proton source. In a
typical reaction, complex 4 was dissolved in Et2O followed by
the addition of N2H4. After 30 min, the solution was quenched
by HCl, and the product was quantified using the indophenol
assay.21 The product yields summarized in Table 3 clearly
complex first reacts with two molecules of hydrazine, affording
two ammonia molecules and a diazene-coordinated adduct. In
the following step, the diazene complex can further react with
another molecule of hydrazine to generate N2 and NH3
(Scheme 4). The isolation or spectroscopic observation of
Scheme 4
Table 3. Catalytic Disproportionation of Hydrazine by
Heterobimetallic Ti/Co Complexes 4−6
a
N2H4
(equiv)
NH3 yield
(mmol)
NH3 yield (equiv)
[theor]
entry catalyst
turnover
1
2
3
4
5
6
7
4
4
4
4
4
5
6
1
5
0.026
0.13
0.26
0.42
0.47
0.38
1.3 [1.3]
6.6 [6.7]
13 [13]
1.0
5.0
10
10
20
50
20
20
21 [27]
16
any intermediate complexes was unsuccessful. Furthermore, the
3:1 mixture of 6 and 7 observed when MeNHNH2 was used is
difficult to explain, although similar product distributions have
been observed in dispoportionation reactions of substituted
hydrazines that proceed through diazene intermediates in the
literature.5b
22 [67]
18
b
19 [27]
14
0.46
b
23 [27]
17
Determined by colorimetry.21 The reaction was carried out in an
argon filled glovebox.
a
Since diazene itself, HNNH, is not stable, we attempted to
access a related intermediate using a stoichiometric reaction of
4 with azobenzene, PhNNPh (Scheme 5). Treatment of 4
suggest that a maximum of 16−18 turnovers can be reached.
The N2-free complex 5 and the ammonia coordinated complex
6 show comparable activity to complex 4 (entries 4, 6, and 7),
implying that product inhibition with either NH3 or N2 is not
problematic. However, the efficiency of 4 as a catalyst for this
reaction and the limited number of turnovers are most likely
hampered by the degradation of 4 under the reaction
Scheme 5
1
conditions. Analysis of the reaction mixture by 31P and H
NMR spectroscopy upon the completion of the reaction reveals
that most of the catalyst has decomposed to free XylNHPiPr2
ligand. In fact, when 20 equiv of NH3 are added to complex 4, a
substantial quantity of XylNHPiPr2 is formed along with the
NH3 complex 6, suggesting that the buildup of NH3 in the
catalyst mixture is responsible for catalyst deactivation.
A series of control reactions demonstrated that monometallic
complexes including the monometallic TiIV complex 1, a
monometallic TiIII complex Ti(XylNPiPr2)3, and a mono-
metallic CoI complex ICo(Ph2PNHiPr)3 were inactive catalysts
and produced no ammonia upon treatment with hydrazine.
Furthermore, a 1:1 mixture of Ti(XylNPiPr2)3 and ICo-
(Ph2PNHiPr)3 showed no catalytic activity for hydrazine
disproportionation.
While the ammonia formed via hydrazine disproportionation
could be easily quantified, the formation of dinitrogen in these
reactions was confirmed qualitatively via the reaction of N2-free
complex 5 with hydrazine. Addition of 3 equiv of hydrazine to a
C6D6 solution of 5 in a J. Young NMR tube under an argon
atmosphere resulted in the quantitative generation of 6, as
confirmed by NMR spectroscopy and IR spectroscopy (ν(N2)
= 2062 cm−1).
with stoichiometric azobenzene cleanly generates a para-
magnetic complex 8, rather than the expected diamagnetic
azobenzene adduct. The same paramagnetic complex 8 was
identified by 1H NMR spectroscopy as a product of the
reaction of 4 with 1,2-diphenylhydrazine. We were able to
determine the formulation and connectivity of complex 8 as
[(κ2-XylNPiPr2NPh)Ti(XylNPiPr2)2(μ-NPh)Co]2(μ-N2) using
single crystal X-ray diffraction (Figure S28); its formation in
both reactions suggests that azobenzene is an intermediate in
the reaction of 4 with 1,2-diphenylhydrazine, supporting the
mechanistic hypothesis shown in Scheme 4. The exact role of
Ti and Co in each N−H and N−N bond activation step and
the binding mode of hydrazine or diazene intermediates is
unknown, and given that one of the phosphinoamide ligands
has been oxidized in complex 8, this complex is likely an off-
cycle byproduct and not an intermediate in the hydrazine
disproportionation reaction. However, it is clear that both
metals play an essential role in substrate binding during the
catalytic disproportionation process.
A number of transition metal complexes promote hydrazine
disproportionation reactions to generate diazene,4c,5a,b,d which
can be considered the first step in the disproportionation of
hydrazine to ammonia and dinitrogen. Based on this precedent,
we propose a reaction mechanism for the disproportionation of
hydrazine by 4−6, in which the heterobimetallic Ti/Co
CONCLUSION
■
In conclusion, we have prepared a series of tris-
(phosphinoamide) heterobimetallic Ti/Co complexes featuring
F
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
TiCl4 (220 μL, 2.0 mmol) was added. The reaction mixture was
warmed to room temperature and stirred for 12 h to ensure
completion of the reaction. LiCl was removed by filtration, and the
volatiles were removed from the filtrate in vacuo. The crude red
material was extracted with pentane, and slow evaporation of a
concentrated pentane solution of 1 at room temperature yielded
orange crystals of 1 (1.01 g, 64%). 31P{1H} NMR (C6D6, 293 K, 162
MHz): δ −1.6 (s). 1H NMR (C6D6, 293 K, 400 MHz): δ 6.90 (s, 6H,
o-Ar), 6.46 (s, 3H, p-Ar), 2.61 (m, 6H, CH(CH3)2), 2.23 (s, 18H, Ar−
CH3), 1.25 (m, 18H, CH(CH3)2), 1.16 (m, 18H, CH(CH3)2).
13C{1H} NMR (C6D6, 293 K, 100.63 MHz): δ 152.8, 137.6, 125.8,
120.8, 29.0, 21.3, 20.1, 19.5. UV−vis (Et2O, λmax, nm (ε, mol−1
cm−1)): 250 (5100), 292 (2300), 605 (310). Anal. Calcd for
C42H69N3P3TiCl: C, 63.67; H, 8.78; N, 5.30. Found: C, 63.59; H,
8.71; N, 5.19.
metal−metal multiple bonds. Several notable differences
between these complexes and their Zr/Co analogues have
emerged from this study:
(1) The metal−metal distances are substantially shorter in
the Ti/Co series; however, the difference is relatively
small when corrected for the smaller atomic radius of Ti.
DFT calculations suggest that the metal−metal bonds in
the Ti/Co series are generally more covalent.
(2) A comparison of the infrared ν(N2) stretches of the
(THF)M(XylNPiPr2)3CoN2 (M = Zr, Ti (4)) complexes
suggests substantially more Co→N2 π back-bonding in
the Zr/Co case, which, in turn, is consistent with weaker
Co→M donation. The more weakly bound N2 ligand in
the Co/Ti complex may permit more facile small
molecule activation processes involving the Co center.
(3) The reduced C3-symmetric heterobimetallic Ti/Co
complex 4 reacts with hydrazine and its derivatives
quite differently from the Zr/Co analogue (THF)Zr-
(MesNPiPr2)3CoN2. While the Zr/Co complex under-
goes a one-electron dissociative electron transfer process
Synthesis of (η2-iPr2PNXyl)Ti(XylNPiPr2)2(μ-Cl)CoI (2). Solid 1
(0.792 g, 1.00 mmol) and solid CoI2 (0.313 g, 1.00 mmol) were
combined in THF (15 mL) and stirred for 4 h at room temperature.
The resulting brown solution was filtered through Celite, and solvent
was removed from the filtrate in vacuo. The crude dark material was
extracted with Et2O, and slow evaporation of a concentrated Et2O
solution of 2 at room temperature yielded brown crystals of 2 (0.802 g,
1
82%). H NMR (400 MHz, C6D6): δ 22.0 (323 Hz), 11.9 (101 Hz),
−
to form a metal-bound N2H3 hydrazido species and
8.3 (11 Hz), 7.5 (8 Hz), 3.3 (16 Hz), 2.2 (12 Hz), 1.3 (9 Hz), −1.5
(195 Hz), −6.0 (214 Hz), −9.8 (258 Hz). Evans method (C6D6): 3.20
μB. UV−vis (Et2O, λmax, nm (ε, mol−1 cm−1)): 255, (27000), 296
(12000), 370 (4100), 578 (670). Anal. Calcd for C42H69N3P3TiCoICl:
C, 51.57; H, 7.11; N, 4.30. Found: C, 51.52; H, 7.03; N, 4.19.
Synthesis of (η2-iPr2PNXyl)Ti(XylNPiPr2)2CoI (3). A solution of 2
(0.978 g, 1.00 mmol) in THF (15 mL) was added to solid KC8 (0.128
g, 0.95 mmol). The reaction mixture was stirred at room temperature
for 2 h to ensure completion of the reaction. The insoluble byproducts
were removed by filtration. The volatiles were removed from the
filtrate in vacuo, and the remaining dark solid was washed with
pentane. Slow evaporation of a concentrated Et2O solution of 3 at
NH3, the Ti/Co complex proceeds further down the N−
H and N−N bond cleavage pathway to disproportionate
hydrazine into ammonia and dinitrogen catalytically.
Several mechanistic questions regarding the latter catalytic
transformation remain unanswered, including the binding mode
of the diazene molecule in the purported diazene intermediate,
which could be either terminal at Ti or bridging between the Ti
and Co centers. However, it is clear that both Ti and Co play
an important role in the reaction mechanism and, more
importantly, are a more reactive pair than Zr and Co in the
tris(phosphinoamide) heterobimetallic scaffold.
1
room temperature yielded brown crystals of 3 (0.790 g, 84%). H
NMR (400 MHz, THF-d8): δ 33.7 (1600 Hz), 7.22 (129 Hz), 6.69
1
(14 Hz), 3.82 (193 Hz), 2.28 (12 Hz), −0.17 (396 Hz); H NMR
EXPERIMENTAL SECTION
(400 MHz, C6D6): δ 8.05 (1600 Hz), 6.48 (1360 Hz), 2.31 (41 Hz),
■
−0.06 (595 Hz). Evans method (C6D6): 1.98 μB. UV−vis (Et2O, λmax
,
General Considerations. All manipulations were carried out
under an inert atmosphere using a nitrogen-filled glovebox or standard
Schlenk techniques unless otherwise noted. All glassware was oven or
flame-dried immediately prior to use. Diethyl ether and THF were
obtained as HPLC grade without inhibitors; pentane and benzene
were obtained as ACS reagent grade. All protio solvents were degassed
by sparging with ultrahigh purity argon and dried via passage through
columns of drying agents using a Seca solvent purification system from
Pure Process Technologies. THF-d8 and dichloromethane-d2 were
dried with CaH2 and degassed before use. All NMR spectra were
obtained using a Varian Inova or MR 400 MHz instrument, and all
chemical shifts are reported in ppm. 1H and 13C NMR chemical shifts
were referenced to residual solvent, and 31P NMR chemical shifts were
referenced to 85% H3PO4. For the paramagnetic molecules (2 and 3),
the 1H NMR data are reported with the chemical shift, followed by the
peak width at half-height in parentheses. iPr2PNHXyl,14
nm (ε, mol−1 cm−1)): 259 (39000), 291 (15860), 411 (5620), 627
(740). Anal. Calcd for C42H69N3P3TiCoI: C, 53.51; H, 7.38; N, 4.46.
Found: C, 53.38; H, 7.51; N, 4.40.
Synthesis of (THF)Ti(XylNPiPr2)3CoN2 (4). A 0.5% Na/Hg amalgam
was prepared from 0.057 g of Na (2.5 mmol) and 11.0 g of Hg. To this
vigorously stirred amalgam in 10 mL of THF was added a solution of 2
(0.978 g, 1.00 mmol) in THF (50 mL). The solution immediately
began to change color from brown to red. After 2 h, the resulting red
solution was filtered away from the amalgam, and the solvent was
removed from the filtrate in vacuo. The resulting solid was extracted
back into Et2O and filtered through Celite. Slow evaporation of the
concentrated Et2O solution of 4 at −35 °C yielded pure red crystals of
4 (0.750 g, 82%). 31P{1H} NMR (C6D6, 293 K, 162 MHz): δ 45.5 (br
1
s). H NMR (400 MHz, C6D6): δ 6.71 (s, 6H, o-Ar), 6.46 (s, 3H, p-
Ar), 3.32 (br, 6H, CH(CH3)2), 2.79 (br m, 4H, THF), 2.15 (s, 18H,
Ar−CH3), 1.64 (br m, 18H, CH(CH3)2), 1.51 (br m, 18H,
CH(CH3)2), 0.72 (br, 4H, THF). 13C{1H} NMR (C6D6, 100.63
MHz): δ 152.6, 138.2, 124.0, 123.8, 69.5, 35.8, 25.3, 22.2, 21.9, 21.4.
UV−vis (Et2O, λmax, nm (ε, mol−1 cm−1)): 250 (9100), 299 (2200),
443 (300), 680 (150). IR: 2084 cm−1 (KBr solution cell, C6D6). Anal.
Calcd for C46H77N5P3TiCoO: C, 60.33; H, 8.47; N, 7.65. Found: C,
60.17; H, 8.46, N, 7.42.
16
(iPr2PNXyl)3Ti,13 and ICo(Ph2PNHiPr)3 were synthesized using
literature procedures. All other reagents and solvents were obtained
from commercial sources and used without further purification. IR
spectra were recorded on a Varian 640-IR spectrometer controlled by
Resolutions Pro software. UV−vis spectra were recorded on a Cary 50
UV−vis spectrophotometer using Cary WinUV software. Elemental
microanalyses were performed by Complete Analysis Laboratories,
Inc., Parsippany, NJ. Caution! Anhydrous hydrazine is a highly toxic,
volatile, and flammable liquid. While we did not encounter any issues,
proper precautions were taken.
Synthesis of (THF)Ti(XylNPiPr2)3Co (5). The solvent was removed
from a Et2O solution of 4 (0.091 g, 0.10 mmol) in vacuo, and the
solution began to change color from red to green under vacuum. The
resulting green solid was extracted back into Et2O under argon
atmosphere. Slow evaporation of the concentrated Et2O solution of 5
at room temperature yielded pure green crystals of 5 (0.074 g, 83%).
31P{1H} NMR (C6D6, 162 MHz): δ 39.6 (br s). 1H NMR (400 MHz,
C6D6): δ 6.86 (s, 6H, o-Ar), 6.49 (s, 3H, p-Ar), 3.47 (m, 6H,
Synthesis of (iPr2PNXyl)3TiCl (1). A solution of iPr2PNHXyl (1.42 g,
6.0 mmol) in Et2O (150 mL) was cooled to −78 °C. To this was
n
added BuLi (3.7 mL, 1.6 M in hexanes, 6.0 mmol) dropwise over 10
min. The resulting yellow solution was warmed to room temperature
and stirred for 2 h. The solution was then cooled again to −78 °C, and
G
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
CH(CH3)2), 2.71 (br m, 4H, THF), 2.22 (s, 18H, Ar−CH3), 1.67 (br,
18H, CH(CH3)2), 1.52 (br, 18H, CH(CH3)2), 0.69 (br, 4H, THF).
13C{1H} NMR (C6D6, 293 K, 100.63 MHz): δ 154.3, 138.0, 124.5,
123.2, 69.1, 32.7, 25.4, 22.9, 22.3, 20.5. UV−vis (Et2O, λmax, nm (ε,
mol−1 cm−1)): 250 (3100), 291 (910), 499 (170). Owing to the
oxygen sensitivity of this complex, repeated elemental analysis samples
anaylzed as the oxidized compound (THF)Ti(XylNP(O)iPr2)3Co.
Anal. Calcd for C46H77N5P3TiCoO4: C, 59.04; H, 8.29; N, 4.49.
Found: C, 59.11; H, 8.19, N, 4.70.
were tested quantitatively for ammonia by using the indophenol test.21
Caution! The generation of dinitrogen gas may lead to the buildup of
positive pressure. All the catalytic hydrazine disproportionation reactions
were performed on very small scales. Large scale reactions would require the
appropriate venting equipment or pressure resistant containers.
X-ray Crystallography. All operations were performed on a
Bruker-Nonius Kappa Apex2 diffractometer, using graphite-mono-
chromated Mo Kα radiation. All diffractometer manipulations,
including data collection, integration, scaling, and absorption
corrections, were carried out using the Bruker Apex2 software.22
Fully labeled diagrams and data collection and refinement details are
included in Table S1 and on pages S14−S22 of the Supporting
Information.
Computational Details. All calculations were performed using
Gaussian09, Revision A.02 for the Linux operating system.23 Density
functional theory calculations were carried out using a combination of
Becke’s 1988 gradient-corrected exchange functional24 and Perdew’s
1986 electron correlation functional25 (BP86). A mixed-basis set was
employed, using the LANL2TZ(f) triple-ζ basis set with effective core
potentials for cobalt, iodine, and titanium,26 Gaussian09’s internal 6-
311+G(d) for heteroatoms (nitrogen, phosphorus, oxygen), and
Gaussian09’s internal LANL2DZ basis set (equivalent to D95 V27) for
carbon and hydrogen. Using crystallographically determined geo-
metries as a starting point, the geometries were optimized to a
minimum, followed by analytical frequency calculations to confirm that
no imaginary frequencies were present. XYZ coordinates of optimized
geometries are provided on Supporting Information pages S25−S32.
EPR Spectroscopy. X-band EPR spectra were obtained on a
Bruker ElexSys E500 EPR spectrometer (fitted with a cryostat for
measurements at 3 K). EPR samples were crystalline samples, and
spectra were measured as frozen toluene glasses at 3 K. The spectra
were referenced to diphenylpicrylhydrazyl (DPPH; g = 2.0037) and
modeled using EasySpin for MATLAB.
Synthesis of (H3N)Ti(XylNPiPr2)3CoN2 (6). In a 20 mL scintillation
vial containing a stir bar, complex 4 (91 mg, 0.10 mmol) was dissolved
in 5 mL of Et2O. To this solution was added H2NNH2 (2.4 μL, 0.075
mmol). The reaction mixture was stirred at room temperature for 10
min, and then the volatiles were removed in vacuo. The remaining red
solid was extracted with benzene and filtered through Celite. The
solvent was removed in vacuo to yield spectroscopically pure product
as a red solid (75 mg, 88% yield). Crystals suitable for X-ray diffraction
were grown from a concentrated Et2O solution at room temperature.
31P{1H} NMR (C6D6, 162 MHz): δ 39.6 (br s). 1H NMR (400 MHz,
C6D6): δ 6.83 (s, 6H, o-Ar), 6.47 (s, 3H, p-Ar), 3.20 (m, 6H,
CH(CH3)2), 2.15 (s, 18H, Ar−CH3), 1.71 (br, 18H, CH(CH3)2), 1.47
(br, 18H, CH(CH3)2), −1.01 (s, 3H, NH3). 13C{1H} NMR (C6D6,
100.63 MHz): δ 153.7, 137.5, 123.8, 123.3, 35.7, 21.1, 21.0, 19.9. UV−
vis (Et2O, λmax, nm (ε, mol−1 cm−1)): 252 (33000), 298 (12000), 443
(1500), 688 (170). IR: 2062 cm−1 (KBr solution cell, C6H6). Mutilple
elemental analysis samples failed owing to the oxygen sensitivity of
complex 6 and the lability of the dinitrogen ligand.
Synthesis of (MeH2N)Ti(XylNPiPr2)3CoN2 (7). In a 20 mL
scintillation vial containing a stir bar, complex 4 (91 mg, 0.10
mmol) was dissolved in 5 mL of Et2O. To this solution was added a
THF solution of MeNH2 (50 μL, 2 M THF solution, 0.1 mmol). The
reaction mixture was stirred at room temperature for 10 min, and then
the volatiles were removed in vacuo. The remaining red solid was
extracted with benzene and filtered through Celite. The solvent was
removed from the filtrate in vacuo to yield spectroscopically pure
product as a red solid (74 mg, 84% yield). Crystals suitable for X-ray
diffraction were grown from a concentrated Et2O solution at room
temperature. 31P{1H} NMR (C6D6, 162 MHz): δ 39.6 (br s). 1H
NMR (400 MHz, C6D6): δ 6.92 (s, 6H, o-Ar), 6.50 (s, 3H, p-Ar), 3.21
(br m, 6H, CH(CH3)2), 2.17 (s, 18H, Ar−CH3), 1.72 (br, 18H,
CH(CH3)2), 1.47 (br, 18H, CH(CH3)2), 0.5 (br, 3H, CH3), −0.31
(br, 2H, NH2). 13C{1H} NMR (C6D6, 100.63 MHz): δ 152.4, 137.8,
123.5, 123.3, 35.3, 21.5, 21.3, 20.6. UV−vis (Et2O, λmax, nm (ε, mol−1
cm−1)): 250 (14000), 297 (6100), 518 (1100), 483 (580), 681 (83).
IR: 2061 cm−1 (KBr solution cell, C6D6). Anal. Calcd for
C43H74N6P3TiCo: C, 59.04; H, 8.53; N, 9.61. Multiple elemental
analysis samples failed owing to the oxygen sensitivity of complex 7
and the lability of the dinitrogen ligand.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b01962.
■
S
Computational and spectral data and crystallographic
details (PDF)
Crystallographic data (CIF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: thomasc@brandeis.edu.
■
Synthesis of [(κ2-XylNPiPr2NPh)Ti(XylNPiPr2)2(μ-NPh)Co]2(μ-N2)
(8). In a 20 mL scintillation vial containing a stir bar, complex 4 (91
mg, 0.10 mmol) was dissolved in 5 mL of Et2O. To this solution was
added azobenzene (18.2 mg, 0.10 mmol). The reaction mixture was
stirred at room temperature for 10 min, and then the volatiles were
removed in vacuo. The remaining red solid was extracted with benzene
and filtered through Celite. The solvent was removed in vacuo to yield
analytically pure product as a red solid (80 mg, 79% yield). Crystals
suitable for X-ray diffraction were grown from a concentrated Et2O
solution at room temperature. 1H NMR (400 MHz, CD2Cl2): δ 36.28,
18.59, 16.06, 10.14, 8.73, 8.00, 6.94, 6.77, 4.63, 4.31, 3.77, 2.59, −0.12,
−1.59, −2.98, −3.24, −3.63, −7.17, −62.13, −69.72. Evans method
(CD2Cl2): 5.10 μB. UV−vis (Et2O, λmax, nm (ε, mol−1 cm−1)): 250
(8500), 300 (2000), 650(300). Anal. Calcd for: C, 64.09; H, 7.87; N,
8.30. Found: C, 64.24; H, 7.70; N, 8.17.
Catalytic Disproportionation of Hydrazine. To a solution of 4
(18.2 mg, 0.0200 mmol) in diethyl ether (5 mL) in a 20 mL
scintillation vial was added anhydrous hydrazine, and the mixture was
stirred for 30 min at room temperature. 200 μL of concentrated HCl
was added to the reaction mixture by syringe. The solvent was then
immediately removed in vacuo, and the residue was extracted with
distilled H2O (30 mL). Aliquots of this solution, diluted if necessary,
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This material is based upon work supported by the U.S.
Department of Energy under Award No. DE-SC0004019. The
authors are also grateful for access to the Brandeis University
high performance computing cluster. The authors also thank
́
́ ́
Dr. Raul Hernandez Sanchez (Harvard University) for
assistance with EPR data collection and Keith J. Fritzsching
(Brandeis University) for assistance with EPR simulation.
REFERENCES
■
(1) (a) Hazari, N. Chem. Soc. Rev. 2010, 39, 4044−4056. (b) Hidai,
M.; Mizobe, Y. Chem. Rev. 1995, 95, 1115−1133. (c) Rodriguez, M.
M.; Bill, E.; Brennessel, W. W.; Holland, P. L. Science 2011, 334, 780−
783.
(2) (a) Schrock, R. R. Acc. Chem. Res. 2005, 38, 955−962.
(b) Yandulov, D. V.; Schrock, R. R. Science 2003, 301, 76−78.
(c) Arashiba, K.; Miyake, Y.; Nishibayashi, Y. Nat. Chem. 2011, 3,
H
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
120−125. (d) Tian, Y.-H.; Pierpont, A. W.; Batista, E. R. Inorg. Chem.
2014, 53, 4177−4183. (e) Kuriyama, S.; Arashiba, K.; Nakajima, K.;
Tanaka, H.; Kamaru, N.; Yoshizawa, K.; Nishibayashi, Y. J. Am. Chem.
Soc. 2014, 136, 9719−9731. (f) Anderson, J. S.; Rittle, J.; Peters, J. C.
Nature 2013, 501, 84−87.
(22) Apex 2, Version 2 User Manual, M86-E01078; Bruker Analytical
X-ray Systems: Madison, WI, 2006.
(23) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; et al. Gaussian 09, Revision A. 2; Gaussian, Inc.:
Wallingford, CT (see Supporting Information for full reference), 2009.
(24) Becke, A. D. Phys. Rev. A: At. Mol. Opt. Phys. 1988, 38, 3098−
3100.
(25) Perdew, J. P. Phys. Rev. B: Condens. Matter 1986, 33, 8822−
8824.
(26) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310.
(b) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270−283.
(27) Dunning, T. H.; Hay, P. J. In Modern Theoretical Chemistry;
Schaeffer, H. F., III, Ed.; Plenum: New York, 1976; Vol. 3.
(3) (a) Hoffman, B. M.; Dean, D. R.; Seefeldt, L. C. Acc. Chem. Res.
2009, 42, 609−619. (b) Hoffman, B. M.; Lukoyanov, D.; Dean, D. R.;
Seefeldt, L. C. Acc. Chem. Res. 2013, 46, 587−595.
(4) (a) Schrock, R. R.; Glassman, T. E.; Vale, M. G.; Kol, M. J. Am.
Chem. Soc. 1993, 115, 1760−1772. (b) Powers, T. M.; Betley, T. A. J.
Am. Chem. Soc. 2013, 135, 12289−12296. (c) Sellmann, D.; Hille, A.;
Rosler, A.; Heinemann, F. W.; Moll, M.; Brehm, G.; Schneider, S.;
̈
Reiher, M.; Hess, B. A.; Bauer, W. Chem. - Eur. J. 2004, 10, 819−830.
(d) Saouma, C. T.; Lu, C. C.; Peters, J. C. Inorg. Chem. 2012, 51,
10043−10054. (e) Nakajima, Y.; Inagaki, A.; Suzuki, H. Organo-
metallics 2004, 23, 4040−4046.
(5) (a) Umehara, K.; Kuwata, S.; Ikariya, T. J. Am. Chem. Soc. 2013,
135, 6754−6757. (b) Kuwata, S.; Mizobe, Y.; Hidai, M. Inorg. Chem.
1994, 33, 3619−3620. (c) Rozenel, S. S.; Arnold, J. Inorg. Chem. 2012,
51, 9730−9739. (d) Saouma, C. T.; Moore, C. E.; Rheingold, A. L.;
Peters, J. C. Inorg. Chem. 2011, 50, 11285−11287. (e) Chen, Y.; Zhou,
Y.; Chen, P.; Tao, Y.; Li, Y.; Qu, J. J. Am. Chem. Soc. 2008, 130,
15250−15251. (f) Chang, Y.-H.; Chan, P.-M.; Tsai, Y.-F.; Lee, G.-H.;
Hsu, H.-F. Inorg. Chem. 2014, 53, 664−666. (g) Szklarzewicz, J.;
́
Matoga, D.; Kłys, A.; Łasocha, W. Inorg. Chem. 2008, 47, 5464−5472.
(h) Chu, W.-C.; Wu, C.-C.; Hsu, H.-F. Inorg. Chem. 2006, 45, 3164−
3166. (i) Demadis, K. D.; Malinak, S. M.; Coucouvanis, D. Inorg.
Chem. 1996, 35, 4038−4046.
(6) (a) Thomas, C. M. Comments Inorg. Chem. 2011, 32, 14−38.
(b) Krogman, J. P.; Thomas, C. M. Chem. Commun. 2014, 50, 5115−
5127.
(7) (a) Greenwood, B. P.; Rowe, G. T.; Chen, C.-H.; Foxman, B. M.;
Thomas, C. M. J. Am. Chem. Soc. 2010, 132, 44−45. (b) Krogman, J.
P.; Gallagher, J. R.; Zhang, G.; Hock, A. S.; Miller, J. T.; Thomas, C.
M. Dalton Trans. 2014, 43, 13852−13857.
(8) (a) Thomas, C. M.; Napoline, J. W.; Rowe, G. T.; Foxman, B. M.
Chem. Commun. 2010, 46, 5790−5792. (b) Zhou, W.; Napoline, J. W.;
Thomas, C. M. Eur. J. Inorg. Chem. 2011, 2011, 2029−2033.
(9) Krogman, J. P.; Foxman, B. M.; Thomas, C. M. J. Am. Chem. Soc.
2011, 133, 14582−14585.
(10) (a) Zhou, W.; Marquard, S. L.; Bezpalko, M. W.; Foxman, B.
M.; Thomas, C. M. Organometallics 2013, 32, 1766−1772.
(b) Marquard, S. L.; Bezpalko, M. W.; Foxman, B. M.; Thomas, C.
M. J. Am. Chem. Soc. 2013, 135, 6018−6021. (c) Marquard, S. L.;
Bezpalko, M. W.; Foxman, B. M.; Thomas, C. M. Organometallics
2014, 33, 2071−2079.
́ ́
(11) Wu, B.; Hernandez Sanchez, R.; Bezpalko, M. W.; Foxman, B.
M.; Thomas, C. M. Inorg. Chem. 2014, 53, 10021−10023.
(12) Napoline, J. W.; Bezpalko, M. W.; Foxman, B. M.; Thomas, C.
M. Chem. Commun. 2013, 49, 4388−4390.
(13) Wu, B.; Bezpalko, M. W.; Foxman, B. M.; Thomas, C. M. Chem.
Sci. 2015, 6, 2044−2049.
(14) Zhou, W.; Saper, N. I.; Krogman, J. P.; Foxman, B. M.; Thomas,
C. M. Dalton Trans. 2014, 43, 1984−1989.
(15) Attempts to collect meaningful 1H NMR data at low
temperature using toluene-d8 were unsucessful, as complex 3
decomposes over time in toluene-d8.
(16) Greenwood, B. P.; Forman, S. I.; Rowe, G. T.; Chen, C.-H.;
Foxman, B. M.; Thomas, C. M. Inorg. Chem. 2009, 48, 6251−6260.
(17) Clouston, L. J.; Bernales, V.; Carlson, R. K.; Gagliardi, L.; Lu, C.
C. Inorg. Chem. 2015, 54, 9263−9270.
(18) Cotton, F. A.; Murillo, C. A.; Walton, R. A. Multiple Bonds
Between Metal Atoms; Springer Science and Business Media, Inc.: New
York, 2005.
(19) Jenkins, D. M.; Di Bilio, A. J.; Allen, M. J.; Betley, T. A.; Peters,
J. C. J. Am. Chem. Soc. 2002, 124, 15336−15350.
(20) For example, the average Co−P and Ti−N WBIs calculated for
complex 4 are 0.61 and 0.78, respectively.
(21) Chaney, A. L.; Marbach, E. P. Clin. Chem. 1962, 8, 130−132.
I
DOI: 10.1021/acs.inorgchem.5b01962
Inorg. Chem. XXXX, XXX, XXX−XXX