Hydroalumination of a dinuclear tantalum dinitrogen complex: N-N bond cleavage and ancillary ligand rearrangement

Article abstract of DOI:10.1021/om050208z

The addition of diisobutylaluminum hydride (DIBAL) to the side-on end-on dinitrogen complex ([NPN]Ta)₂(μ-η¹: η²-N₂)(μ-H)², 1 (where [NPN] = (PhNSiMe₂CH₂)₂PPh), is described. The two end products are diastereomeric rotational isomers in which N-N bond cleavage has occurred with an Al(ⁱBu)H group attached to one of the nitride atoms. The reaction proceeds through addition of DIBAL to 1 to generate a thermally sensitive intermediate that has been characterized in solution as the result of Al-H addition across the TaN₂ moiety, namely, ([NPN]TaH)(μ- η¹:η²-NNAlⁱBu₂)(μ-H) ₂(Ta[NPN), 2. This material subsequently rear-ranges via a second thermally labile intermediate to ultimately generate two diastereomeric end products that show N-N bond cleavage, loss of H₂, loss of an aluminum isobutyl group, and NPN ligand migration from tantalum to aluminum. Both of these complexes have been isolated in crystalline form and analyzed by single-crystal X-ray diffraction. The second thermally sensitive intermediate has been characterized on the basis of multinuclear NMR spectroscopy as ([NPN]TaH)(μ-η¹:η²-NNAlⁱBu(μ-H)) (μ-H)₂(Ta[NPN), 3.

Full text of DOI:10.1021/om050208z

3836
Organometallics 2005, 24, 3836-3841
Hydroalumination of a Dinuclear Tantalum Dinitrogen
Complex: N-N Bond Cleavage and Ancillary Ligand
Rearrangement
Bruce A. MacKay, Brian O. Patrick, and Michael D. Fryzuk*
Department of Chemistry, University of British Columbia, 2036 Main Mall,
Vancouver, British Columbia, Canada V6T 1Z1
Received March 18, 2005
The addition of diisobutylaluminum hydride (DIBAL) to the side-on end-on dinitrogen
complex ([NPN]Ta)2(µ-η1:η2-N2)(µ-H)2, 1 (where [NPN] ) (PhNSiMe2CH2)2PPh), is described.
The two end products are diastereomeric rotational isomers in which N-N bond cleavage
has occurred with an Al(iBu)H group attached to one of the nitride atoms. The reaction
proceeds through addition of DIBAL to 1 to generate a thermally sensitive intermediate
that has been characterized in solution as the result of Al-H addition across the TaN2 moiety,
namely, ([NPN]TaH)(µ-η1:η2-NNAliBu2)(µ-H)2(Ta[NPN), 2. This material subsequently rear-
ranges via a second thermally labile intermediate to ultimately generate two diastereomeric
end products that show N-N bond cleavage, loss of H2, loss of an aluminum isobutyl group,
and NPN ligand migration from tantalum to aluminum. Both of these complexes have been
isolated in crystalline form and analyzed by single-crystal X-ray diffraction. The second
thermally sensitive intermediate has been characterized on the basis of multinuclear NMR
spectroscopy as ([NPN]TaH)(µ-η1:η2-NNAliBu(µ-H))(µ-H)2(Ta[NPN), 3.
Introduction
containing materials directly from dinitrogen, new
reaction types for coordinated N2 are especially impor-
tant since they offer other avenues for use of this
abundant, yet fundamentally unreactive molecule.18
Previous work from our laboratory19,20 has explored
addition reactions of simple hydride reagents (E-H) to
the dinuclear tantalum dinitrogen complex, ([NPN]Ta)2-
(µ-η1:η2-N2)(µ-H)2, 1 (where [NPN] ) (PhNSiMe2CH2)2-
PPh). Both hydroboration and hydrosilylation initially
result in addition of HBR2 and H3SiR across the
coordinated N2 unit to generate a terminal Ta-H and
a functionalized dinitrogen moiety with newly formed
N-B and N-Si bonds, respectively. These initial addi-
tion products are thermally unstable and rearrange by
processes that involve N-N bond cleavage and func-
tionalization and, in the case of boron, ancillary ligand
degradation. Both of these processes are summarized
in Scheme 1. It is proposed that a common intermediate
A forms via reductive elimination of H2 to generate a
species wherein the N-N bond has been cleaved. In the
E ) B manifold, loss of the N-Ph substituent as
benzene and silyl-group migration generate the ob-
served product. In the E ) Si manifold, addition of
another equivalent of silane, followed by H2 elimination,
results in the formation of the bis(silylimide) product.
Given the remarkably different outcomes for boron-
hydride versus silicon-hydride reagents, we examined
other readily available sources of E-H. Herein we
report the hydroalumination of 1 with diisobutylalumi-
In the 40 years since the first report that dinitrogen
can bind to metal complexes,1 considerable sophistica-
tion has been attained in the activation of N2 by
transition and lanthanide elements.2-5 Fundamental
advances in different synthetic methodologies, struc-
tural types, and levels of activation are now well
described, as are reactions between activated N2 and
electrophiles.6-8 The most recent fruits of this knowl-
edge include homogeneous systems that produce am-
monia stoichiometrically9 and catalytically,10 and syn-
thetic methods that use titanium reagents to incorporate
dinitrogen-derived N atoms into natural products.11-17
With respect to the formation of higher value nitrogen-
* To whom correspondence should be addressed. E-mail:
fryzuk@chem.ubc.ca.
(1) Allen, A. D.; Senoff, C. V. J. Chem. Soc., Chem. Commun. 1965,
621.
(2) Gambarotta, S. J. Organomet. Chem. 1995, 500, 117.
(3) Gambarotta, S.; Scott, J. Angew. Chem., Int. Ed. 2004, 43, 5289-
5308.
(4) Hidai, M.; Mizobe, Y. Chem. Rev. 1995, 95, 1115.
(5) Hidai, M. Coord. Chem. Rev. 1999, 185-186, 99.
(6) Fryzuk, M. D.; Johnson, S. A. Coord. Chem. Rev. 2000, 200-
202, 379-409.
(7) Shaver, M. P.; Fryzuk, M. D. Adv. Synth. Catal. 2003, 345, 1061.
(8) MacKay, B. A.; Fryzuk, M. D. Chem. Rev. 2004, 104, 385.
(9) Pool, J. A.; Lobkovsky, E.; Chirik, P. J. Nature 2004, 427, 527.
(10) Yandulov, D. V.; Schrock, R. R. Science 2003, 301, 76.
(11) Akashi, M.; Sato, Y.; Mori, M. J. Org. Chem. 2001, 66, 7873-
7874.
(12) Ueda, K.; Sato, Y.; Mori, M. J. Am. Chem. Soc. 2000, 122,
10722-10723.
(13) Akashi, M.; Nishida, M.; Mori, M. Chem. Lett. 1999, 465-466.
(14) Hori, M.; Mori, M. J. Org. Chem. 1995, 60, 1480.
(15) Hori, K.; Mori, M. J. Am. Chem. Soc. 1998, 120, 7651-7652.
(16) Mori, M.; Hori, K.; Akashi, M.; Hori, M.; Sato, Y.; Nishida, M.
Angew. Chem., Int. Ed. 1998, 37, 636-637.
(17) Mori, M.; Hori, M.; Sato, Y. J. Org. Chem. 1998, 63, 4832-
4833.
(18) Fryzuk, M. D. Chem. Rec. 2003, 3, 2-11.
(19) Fryzuk, M. D.; MacKay, B. A.; Johnson, S. A.; Patrick, B. O.
Angew. Chem., Int. Ed. 2002, 41, 3709. MacKay, B. A.; Johnson, S. A.;
Patrick, B. O.; Fryzuk, M. D. Can. J. Chem. 2005, 83, 315.
(20) Fryzuk, M. D.; MacKay, B. A.; Patrick, B. O. J. Am. Chem. Soc.
2003, 125, 3234.
10.1021/om050208z CCC: $30.25 © 2005 American Chemical Society
Publication on Web 06/28/2005

Hydroalumination of a Dinuclear Ta Complex
Organometallics, Vol. 24, No. 16, 2005 3837
Scheme 1
of 1 (lower), showing that the reaction to give 2 is
complete in the time taken to reinsert the sample into
the probe. This is a marked contrast to the hydrobora-
tion and hydrosilylation reactions shown in Scheme 1,
each of which requires several hours under normal
synthetic conditions (0.02-0.05 M, ambient tempera-
ture, aromatic solvent) for the dinitrogen starting
complex 1 to be consumed.
1
The H NMR spectra of 2 obtained at room temper-
ature and at 213 K each show eight discrete silyl methyl
resonances, two one-proton resonances at δ 10.1 and
10.8 indicative of chemically inequivalent bridging
hydrido ligands, and a new resonance at δ 19.2 that is
suggestive of a terminal hydride based on comparison
to the boryl and silyl congeners. These findings suggest
that the Al-H bond of DIBAL has added across the
Ta-N π bond of 1 to give C1 molecular symmetry, a new
N-Al bond, and a new terminal tantalum hydride (eq
1). In terms of chemical shift, the 31P NMR spectrum of
2 is intermediate between that of 1 and the hydrobo-
rated or hydrosilylated derivatives of 1, several of which
have been characterized in the solid state.19,20
Figure 1. 31P{1H} NMR spectra (C6D6, 300 K) of 1 before
(lower) and of 2 (upper) immediately after addition of 1
equiv of DIBAL. Almost complete conversion to 2 is
observed and only a minor amount of 1 remains. The
pseudoquartet at δ 7.8 for 1 (lower) and at δ 14.5 for
intermediate 2 (upper) is due to additional coupling from
the 14N nucleus of the bridging, trans dinitrogen moiety.21
Intermediate 2 was also characterized by 15N NMR
spectroscopy. The 15N{1H} NMR spectrum of 15N2-2
(prepared by addition of 1 equiv of DIBAL to a sample
of 15N2-1 in C7D8 at -60 °C) is similar to the hydrobo-
rated and hydrosilylated derivatives of 1, in that it
features two resonances coupled to each other by 18.7
Hz, diagnostic of an intact N-N bond. The resonance
at δ -26.23 shows couplings of JNN ) 18.7 Hz and JPN
) 26.0 Hz, and the resonance at δ -32.88 shows
couplings of JPN ) 6.5 Hz and JNN ) 18.7 Hz. These
resonances can be assigned as the bridging N and the
terminal (Al-bound) N of the dinitrogen moiety, respec-
tively, by comparison with the spectra of 15N2-1 and a
trimethylaluminum adduct of 1 (in which no addition
reaction has taken place).22
The instability of 2 toward further rearrangement
was examined by 31P NMR spectroscopy by following
its concentration as a function of time relative to an
internal standard. Three new species are observed in
solution over the 24 h duration of the experiment, and
each one is of sufficiently low symmetry that the
phosphines are chemically inequivalent and thus appear
num hydride (DIBAL), in which N-N bond cleavage is
again observed along with another type of ancillary
ligand rearrangement.
Results and Discussion
The reaction between DIBAL (either neat or as a
commercially prepared solution) and toluene or benzene
solutions of 1 is immediate, as gauged by the rapid color
change from red-brown to yellow-brown upon mixing.
Monitoring this mixture by 31P NMR spectroscopy shows
that an intermediate, 2, forms quantitatively but per-
sists for only a few hours at room temperature before it
converts into other products. Because of the thermal
lability of 2, only solution NMR spectroscopic charac-
terization has been possible. The 31P{1H} NMR spec-
trum of 2 comprises two resonances, δ 12.1 and 14.5,
coupled at JPP ) 18.3 Hz. The spectrum of 2 as shown
in Figure 1 (upper) was obtained immediately after
injection of 1 equiv of neat DIBAL into an NMR sample
(21) Fryzuk, M. D.; Johnson, S. A.; Patrick, B. O.; Albinati, A.;
Mason, S. A.; Koetzle, T. F. J. Am. Chem. Soc. 2001, 123, 3960.
(22) Studt, F.; MacKay, B. A.; Johnson, S. A.; Patrick, B. O.; Fryzuk,
M. D.; Tuczek, F. Chem., Eur. J. 2005, 11, 604.

3838 Organometallics, Vol. 24, No. 16, 2005
MacKay et al.
Figure 2. Relative concentrations of species 2, 3, 4, and
5 in solution over time. Concentrations are measured by
integration of 31P NMR resonances versus an internal
standard.
Figure 4. ORTEP drawing of the solid-state molecular
structure of 5 as determined by X-ray crystallography
(ellipsoids at 50% probability). Silyl methyls and phenyl
ring carbons other than ipso are omitted for clarity. H71
and H72 were located in the difference map and refined
isotropically. Selected bond distances and angles are
presented in Table 1.
[NPN] ligand on Ta1 shows that one amido donor, N3,
has migrated to Al1. This aluminum center is attached
to N1 and contains only one isobutyl group and bridges
to hydride H71. The bridging hydride ligand H71 was
modeled using X-HYDEX,23 while the terminal hydride,
H72, was located in the difference map and refined
isotropically. In solution, the terminal tantalum hydride
is indicated by the presence of a doublet at δ 17.36 (JHH
) 11.3 Hz) in the 1H{31P} NMR spectrum of 4. The
source of the coupling was not initially obvious. The 1H/
1H COSY spectrum of 4 shows a cross-peak between the
terminal hydride and a one-proton resonance at δ 1.23
(doublet, JHH ) 11.3 Hz), which is partially occluded
by overlap with [NPN] and isobutyl methylene reso-
nances. This latter resonance is assigned to the bridging
hydride, Ta(µ-H)Al. The Ta1‚‚‚Al1 distance is 2.9275-
(15) Å, which is substantially longer than in a previously
reported complex featuring a dative Ta-Al bond, a
bridging hydride, and a terminal aluminum hydride.24
Crystals of end product 5 were manually separated
from those of 4. Despite the differences in the 31P{1H}
NMR spectra of 4 and 5, the two complexes have
identical connectivities in the solid state (the ORTEP
drawing of 5 is shown in Figure 4). It is clear that the
difference between 4 and 5 is the rotational orientation
of the intact [NPN] ligand bound to Ta2. In 4, the
P-Ta-Ta-P dihedral angle is -115.32(5)°, whereas the
same angle in 5 is 79.94(4)°. The other significant
structural attributes already listed for 4 are all present
in 5. Important metric parameters for the two complexes
are compared in Table 1, and details of the diffraction
experiments are summarized in Table 2.
Figure 3. ORTEP drawing of the solid-state molecular
structure of 4 as determined by X-ray crystallography
(ellipsoids at 50% probability). Silyl methyls and phenyl
ring carbons other than ipso are omitted for clarity. H71
was modeled using X-HYDEX,23 and H72 was located in
the difference map; both were refined isotropically. Selected
bond distances and angles are presented in Table 1.
as discrete coupled resonances that can be correlated
to a particular species. A plot of the integral values of
the major species present versus time is shown in Figure
2.
As intermediate 2 decays, a new intermediate 3 is
observed that builds up to a maximum concentration
after 6 h and then eventually decreases in concentration.
Two additional species, 4 and 5, form in roughly
equimolar amounts as the end products of the reaction.
Before we discuss our attempts to determine the solu-
tion structure of the second intermediate 3, the solid-
state and solution structures of 4 and 5 will be dis-
cussed.
Isomers 4 and 5 are consistently obtained in roughly
equimolar amounts in approximately 88% combined
yield as end products of the reaction given in Scheme
2, which summarizes the chemistry starting from
intermediate 2. These complexes are diastereomeric
The solid-state molecular structure of 4 has been
obtained by X-ray crystallography; an ORTEP drawing
of 4 is shown in Figure 3. The N1‚‚‚N2 distance is
2.657(3) Å, clearly indicating that N-N bond cleavage
has occurred. Metric parameters for the Ta2N2 square
in 4 are similar to others reported previously in the
hydroboration and hydrosilylation of 1.19,20 The [NPN]
ligand attached to Ta2 is unchanged; however, the
(23) Orpen, A. G. J. Chem. Soc., Dalton Trans. 1980, 2509.
(24) Fryzuk, M. D.; Clentsmith, G. K. B.; Rettig, S. J. Inorg. Chim.
Acta 1997, 259, 51-59.

Hydroalumination of a Dinuclear Ta Complex
Organometallics, Vol. 24, No. 16, 2005 3839
Table 1. Selected Bond Lengths (Å), Angles (deg),
and Dihedral Angles (deg) for 4 and 5
Scheme 2
atoms involved:
4
5
Ta1
Ta1
Ta2
Ta2
N1
Ta1
Ta1
Ta1
Ta1
Ta2
Ta2
Ta2
N1
N1
N2
2.158(4)
2.098(3)
1.861(4)
1.875(3)
N1
1.892(4)
1.918(3)
N2
1.917(4)
1.907(3)
N2
2.657(3)
2.606(4)
Ta2
P1
2.8737(3)
2.6275(13)
2.037(4)
2.84043(19)
2.6316(9)
2.054(3)
N4
Al1
P2
2.9275(15)
2.7423(13)
2.053(4)
2.9105(10)
2.7273(9)
2.039(3)
N5
N6
2.049(4)
2.046(3)
Al1
N3
1.856(4)
1.856(3)
Al1
Ta1
Ta1
N1
1.855(4)
1.845(3)
N1
Ta2
Ta2
N2
N1
Al1
Al1
P1
N3
N1
N4
N6
N1
N5
N6
90.15(16)
99.03(17)
82.38(16)
88.43(17)
93.34(17)
165.0(2)
89.92(11)
97.34(13)
83.56(11)
87.80(12)
94.60(12)
156.79(17)
74.22(8)
N2
Ta1
Ta2
N1
N1
Ta1
All
N2
Ta1
Ta2
N4
77.07(11)
115.25(19)
88.66(10)
165.73(14)
116.10(17)
173.48(12)
73.76(12)
81.21(12)
179.7(2)
N1
117.97(14)
90.24(8)
P1
Ta1
Ta1
Ta2
Ta2
Ta2
Ta2
Ta2
Ta2
Ta1
N1
161.03(11)
116.00(12)
90.33(8)
N5
P2
P2
79.69(9)
P2
74.77(9)
Ta1
P1
N1
Ta1
N1
N2
P2
N2
166.99(17)
79.94(4)
atures led to degradation of the complexes. The process
that interconverts 4 and 5 effectively results from
rotation of one [NPN] ligand with respect to the Ta2N2
core; however, how this occurs has not been investi-
gated.
The formation of 4 and 5 from intermediates 2 and 3
was further investigated by adding DIBAL to the
deuterated isotopomer of the starting dinitrogen com-
plex, ([NPN]Ta)2(µ-η1:η2-N2)(µ-D)2, d2-1. In this case, end
products 4 and 5 form with concomitant production of
D2 (no HD) as detected by GC-MS; therefore, the
hydrides in 4 and 5 originate from the added DIBAL
and are not the bridging hydrides in 1 and 2. It is
important to note that loss of H2 (or D2) from intermedi-
ate 2 is necessary for N-N bond cleavage, as it provides
the additional two electrons required for this process,
as discussed previously (see Scheme 1).19,20
-115.32(5)
165.64(18)
Al1
165.90(14)
Table 2. Crystallographic Data and Structure
Refinement for
anti-[NPN-Al(H)C4H9]Ta(µ-N)(µ-N)Ta[NPN] (4) and
syn-[NPN-Al(H)C4H9]Ta(µ-N)(µ-N)Ta[NPN] (5)
4
5
formula
fw
C
52H72N6Ta2P2Si4Al
C52H72N6Ta2P2Si4Al
1344.34
orange, prism
0.30 × 0.15 × 0.0
monoclinic
P21/c
1344.34
yellow, chip
0.20 × 0.15 × 0.10
orthorhombic
Pbca
color, habit
cryst size, mm
cryst syst
space group
a, Å
20.4244(5)
19.9666(5)
29.2898(8)
90
19.7152(5)
13.3919(3)
22.6221(6)
90
b, Å
c, Å
R, deg
â, deg
90
90
101.282(2)
90
γ, deg
V, Å3
11944.6(5)
8
5857.4(2)
4
Z
F
calc, g/cm3
1.495
5368.00
3.848
1.52
Not including [NPN] ligand amide migration from Ta
to Al, the significant differences between the proposed
structure for 2 and that of 4 (or 5) are cleavage of the
N-N bond, the presence of only two hydride ligands,
and the presence of only one isobutyl substituent on Al.
The characteristic cross-peak that correlates the two
F(000)
2680.00
µ(Mo KR), mm-1
transmn factors
2θmax, deg
3.923
0.7124-1.0000
58.24
14 991
9395
0.69784-1.0000
55.8
total no. of reflns
no. of unique
reflns
Rmerge
no. of reflns with
I g 2σ(I)
no. of variables
R (F2, all data)a
Rw (F2, all data)a
R (F, I > 2σ(I))a
Rw (F, I > 2σ(I))a
gof
46 802
13 747
0.099
8192
0.045
2480
1
hydride ligands in the H/1H COSY experiment is still
present in spectra of complexes 4 and 5 derived from
d2-1. The loss of the isobutyl group during this process
can be correlated with the detection of isobutene by 1H
NMR spectroscopy. In addition, the use of D-AliBu2, d1-
DIBAL, in the reaction with 1 results in the isolation
of both end products 4 and 5 with the downfield
resonance at δ 17.44 missing, clearly indicating that this
terminal Ta-H originates from the initial addition of
DIBAL and is the same terminal hydride as found in
the first intermediate 2. By process of elimination the
bridging hydride in 4 and 5 results from â-elimination
of the isobutyl group to aluminum. The cleaved N-N
bond of both 4 and 5 was confirmed by 15N NMR
spectroscopy using ([NPN]Ta)2(µ-η1:η2-15N2)(µ-H)2, 15N2-
1, in the reaction with DIBAL. The resulting 15N-labeled
materials both showed resonances that did not display
910
729
0.077
0.086
0.034
0.079
0.828
0.040
0.067
0.026
0.061
0.87
a Rigaku/ADSC CCD diffractometer, R ) ∑||Fo | - |Fc ||/∑|Fo |;
2
2
2
2
2
Rw ) (∑w(|Fo | - |Fc |)2.
rotational isomers, and traces of one isomer become
visible in C6D6 solutions of the other if left for several
days. To demonstrate that 4 and 5 interconvert, an
NMR sample of 4 and 5 (2:1 ratio) with an internal 31
P
reference standard was subjected to heating; after 3
days at 50 °C, a 1:1 mixture was observed without a
decrease in the overall integration of 31P NMR active
products versus the internal standard. Higher temper-

3840 Organometallics, Vol. 24, No. 16, 2005
MacKay et al.
N-N coupling as observed in the starting material and
the first intermediate 2.
in which the functionalized dinitrogen complex sponta-
neously and rapidly undergoes N-N bond cleavage
triggered by the reductive elimination of H2. Set in the
context of the hydroboration reaction19 and hydrosilyl-
ation,20 it is clear that the addition of E-H and the
resulting cascade of chemical events are strongly influ-
enced by the nature of E. In our previous reports, the
initial addition reaction took place on a time scale of
hours, whereas hydroalumination with DIBAL takes
place in seconds under equivalent reaction conditions.
Whereas the hydroborated complex [NPN]Ta(H)(µ-H)2-
(µ-η1:η2-NNB(C8H14))Ta[NPN] undergoes N-N bond
cleavage over several days, both the hydrosilylated and
hydroaluminated congeners take only a few hours. The
irreversible and obviously undesirable [NPN] ligand
degradation found in the hydroboration sequence is not
observed here with hydroalumination. Instead, amide
ligand migration from tantalum to aluminum occurs.
Ligand rearrangments have been reported previously
especially when aluminum reagents are added to tran-
sition metal complexes.29,30 The fact that the N-N bond
cleavage is preserved across this “family” of reactions
indicates that the latent reducing power of the system
is a property of 1. Although the ability to undergo E-H
addition is also a property that is (so far) unique to 1,
it is worth noting that Lewis acid-base adducts of 1
are stable complexes that exhibit no tendency toward
N-N bond cleavage.22 We are continuing to examine
this system and related derivatives. An obvious tactic
is to make the ancillary ligand system more rigid and
remove N-Si type linkers in the backbone to try to
prevent the kinds of undesirable rearrangements seen
here in hydroalumination and, previously, in hydrobo-
ration. This is currently underway in our laboratory.31
The second intermediate 3 can be inferred to contain
an intact N-N bond from its 15N NMR spectrum
(acquired at low temperature), which shows JNN ) 19.1
Hz. Although 1H NMR spectroscopy of 15N2-3 in the
absence of 2, 4, and 5 has been impossible, mutually
coupled hydride resonances at δ 19.0 (terminal), 10.5,
and 12.4 (bridging) unique to 3 were identified. Thus,
3 forms when the first intermediate 2 loses isobutene
by â-elimination25,26 and has only one isobutyl group at
the Al center (eq 2). Migration of a ligand amide from
Ta to Al, reductive elimination of H2, and N-N bond
cleavage complete the overall scheme. We cannot con-
clusively rule out a [NPN] ligand amide migration to
Al from Ta at or prior to this point.
Aluminum has previously been shown to facilitate
intramolecular rearrangements in transition metal hy-
dride27 and amide complexes.28 It is worth noting that
in survey reactions with alkenes and alkynes neither 4
nor 5 react via insertion at the Al-H moiety but rather
at the terminal tantalum hydride, as shown by the
disappearance of the downfield resonance in clean
formation of what we believe are new tantalum alkyls.
Although these reactions are relatively clean as ob-
served by 31P NMR spectroscopy, they proceed over
several days and, unfortunately, no pure products have
been isolated. Simple 13C NMR experiments comparing
reactions of 4 and 15N2-4 with ethylene, propylene, and
acetylene showed no evidence for formation of N-C
bonds, a process of interest to us. Cyclohexene and
phenylacetylene did not react with 4 at all. This may
indicate that the isobutene elimination is driven par-
tially by steric congestion at Al.
Experimental Section
General considerations for these experiments are published
elsewhere,19 as are the syntheses of 1 and its isotopomers.21
DIBAL was purchased as a hexanes solution from Aldrich and
used either as received or after evaporation of hydrocarbon
solvent to a neat liquid. Diisobutylaluminum deuteride was
prepared by literature methods.26 Ethylene, propylene (Praxair),
and acetylene (Matheson) were used as obtained from com-
mercial gas suppliers. Phenylacetylene was distilled prior to
preparation of a toluene solution of known concentration,
which was used in synthesis.
Synthesis and NMR Characterization of [NPN]Ta(H)-
(µ-η1:η2-NNAl(C4H9)2)(µ-H)2Ta[NPN], 2. A solution of 40 mg
of 1 in roughly 1 mL of d8-toluene was added to 4.5 mg (0.32
mmol, 1 equiv) of neat DIBAL, and the two reagents were
mixed using a Pasteur pipet prior to transfer into a 9 in.
Wilmad NMR tube fitted with a Kontes valve. The sample was
frozen in liquid nitrogen and flame-sealed before being thawed
in an ethanol-liquid nitrogen slurry and inserted into the
probe of a Bruker AVA-400 NMR spectrometer that had been
1
cooled to -60 °C. H NMR (400 MHz, C7D8, 213 K): δ -0.61,
Conclusions
-0.55, -0.38, 0.17, -0.05, 0.48, 0.80, 0.83 (s, 3H each, 24H
total), SiCH3; 0.95 (b, 2H) (CH3)2CHCH2Al; 0.31, 0.41 (s, 3H
each) (CH3)2CHCH2Al); 0.66, 1.22, 1.34, 1.46, 1.53, 1.71, 1.79,
1.84 (AMX, 8H total), PCH2; 1.52 (s, 1H), Al-H; 6.32, 6.50,
6.64, 6.70, 6.83, 6.90, 6.94, 7.20, 7.32, 7.34 (phenyl protons,
This study extends our understanding of the 1,2-
addition of E-H bonds across the side-on Ta-N2 π bond
of dinitrogen complex 1 and provides another example
(25) Bundens, J. W.; Yudenfreund, J.; Francl, M. M. Organometallics
1999, 18, 3913-3930.
(29) Kickham, J. E.; Guerin, F.; Stephan, D. W. J. Am. Chem. Soc.
2002, 124, 11486-11494.
(30) Kickham, J. E.; Guerin, F.; Stewart, J. C.; Urbanska, E.; Ong,
C. M.; Stephan, D. W. Organometallics 2001, 20, 1175-1182.
(31) MacLachlan, E. A.; Fryzuk, M. D. Organometallics 2005, 24,
1112-1118.
(26) Egger, K. W. J. Am. Chem. Soc. 1969, 91, 2867-2871.
(27) Bruno, J. W.; Huffman, J. C.; Caulton, K. G. J. Am. Chem. Soc.
1984, 106, 444-445.
(28) Ison, E. A.; Abboud, K. A.; Ghiviriga, I.; Boncella, J. M.
Organometallics 2004, 23, 929-931.

Hydroalumination of a Dinuclear Ta Complex
Organometallics, Vol. 24, No. 16, 2005 3841
0.47 (3H) (s, 24H total, SiCH3), 1.33, 1.87 (d, 3H each, 3JHH
various multiplicities, total 19 H; some resonances obscured
by solvent), 7.66 (AMX, 2H, JHH ) 7.1 Hz, JPH ) 1.7 Hz, PPh
o-H) 7.77 (AMX, 2H, JHH ) 7.2 Hz, JPH ) 1.3 Hz, PPh o-H),
10.11 (dd, 1H, 2JHH ) 11.2 Hz, 2JHP ) 12.8 Hz), 10.75 (vq, 1H,
2JHH ) 11.2 Hz, 2JHH ) 11.1 Hz 2JHP ) 15.8 Hz), TaHTa; 19.20
)
7 Hz, (CH3)2CHCH2Al), 1.91 (m, 1H (CH3)2CHCH2Al), 0.95,
1
1.16 (AMX, H each, (CH3)2CHCH2Al), 0.87, 1.21, 1.32, 1.41,
1.69, 1.83 (AMX, 5H total, PCH2), 1.54 (d, 1H, JHH ) 12.4 Hz,
Al-H), 6.49, 6.77, 6.82, 6.89, 7.22, 7.34, 7.40, 7.95 (phenyl
protons, various multiplicities, total 19 H), 8.15 (AMX, 2H, JHH
2
2
(dd, 1H, JHH ) 11.1 Hz, JHP ) 48.2 Hz), TaH. 31P{1H}NMR
(161.9 MHz, C7D8, 213 K): δ 12.1 (d, JPP ) 18.31 Hz), 14.5 (d,
JPP ) 18.31 Hz); [NPN] ligand. The product was not isolable
and therefore no elemental analysis was obtained.
) 6.90 Hz, JPH ) 1.58 Hz, PPh o-H) 8.26 (AMX, 2H, JHH
)
7.15 Hz, JPH ) 1.21 Hz, PPh o-H), 15.96 (dd, 1H, 2JHH ) 12.4
Hz, 2JHP ) 17.8 Hz, TaH). Note some proton resonances were
eclipsed by solvent and silylmethyl resonances. 31P{1H}NMR
(C6H6, 300 K): δ -5.8 (s, [NPN] ligand), 12.4 (s [NPN] ligand).
Anal. Calcd for C52H73N6P2Si4Ta2Al: C, 47.42; H, 5.47; N, 6.25.
Found: C, 47.49; H, 5.49; N, 6.23.
Synthesis of 15N2-2. An 80 mg sample of 15N2-1 was treated
with 9 mg of neat DIBAL in a manner analogous to that
reported above for 2. 31P{1H}NMR (161.9 MHz, C7D8, 213 K):
δ 12.1 (ddd, JPP ) 18.31 Hz, JPN ) 26.0 and 6.5 Hz), 14.5 (d,
(d, JPP ) 18.31 Hz); [NPN] ligand. 15N{1H} NMR (40.6 MHz,
C7D8, 213 K): δ -26.23 (dd, JPN ) 26.0 Hz, JNN ) 18.7 Hz),
Nb; -32.88 (dd, JPN ) 6.5 Hz, JNN ) 18.7 Hz), Nt.
Spectroscopic Data for 15N2-3, ([NPN]TaH)(µ-η1:η2-
NNAliBu(µ-H))(µ-H)2(Ta[NPN). The sample of 2 described
above was allowed to stand at room temperature for 8 h before
being reinserted into the spectrometer. New resonances
present: 1H NMR: 10.50 (dd, 1H, 2JHH ) 11.1 Hz, 2JHP ) 15.0
Hz), 12.4 (vq, 1H, 2JHH ) 11.2 Hz, 2JHH ) 10.6 Hz 2JHP ) 13.6
Hz), TaHTa; 19.0 (d, 1H, 2JHH ) 10.6 Hz), TaH. 31P{1H} NMR
(161.9 MHz, C7D8, 213 K): δ -10.9 (d, JPN ) 4.06 Hz), 11.2
(s); [NPN] ligand. 15N{1H} NMR (40.6 MHz, C7D8, 213 K): δ
Synthesis of 15N2-4. A solution of 15N2-1 was treated in a
manner similar to the preparation of 4 and 5. No changes to
1
the H NMR spectrum were observed compared to that of 4.
31P{1H} NMR (161.9 MHz, C6D6, 300 K): δ -22.9 (dd, 2JNP
)
16.7 and 8.8 Hz), 19.8 (s), [NPN] ligand. 15N{1H} NMR (C6D6
2
2
300 K, 40 MHz): δ 57.9 (d, JNP ) 16.7 Hz), 291.8 (d, JNP
)
8.8 Hz).
Synthesis of 15N2-5. A solution of 15N2-1 was treated in a
manner similar to the preparation of 4 and 5. No changes to
1
the H NMR spectrum were observed compared to that of 3
and 4. 31P{1H} NMR (161.9 MHz, C6D6, 300 K): δ -5.8 (d,
2JNP ) 10.1 Hz), 12.4 (s), [NPN] ligand. 15N{1H} NMR (C6D6
-25.56 (dd, JPN ) 4.1 Hz, JNN ) 17.9 Hz), Nb; 31.40 (d, JNN
17.9 Hz), Nt.
)
2
300 K, 40 MHz): δ 54.8 (s), 313.9 (d, JNP ) 10.1 Hz).
31P NMR Spectroscopic Investigation of the Reaction
of 1 with DIBAL. A 9 in. Wilmad NMR tube was charged
with 40.2 mg of 1 in roughly 1 mL of C6D6 and a sealed glass
capillary tube containing neat P(OMe)3 as an internal refer-
ence. The tube was sealed with a 5 mm rubber septum and
wrapped with ParaFilm laboratory film and inserted into the
probe of a Bruker AVA-400 NMR spectrometer. The spectrom-
Synthesis of [NPN-Al(H)C4H9]Ta(H)(µ-N)(µ-N)Ta[N-
PN], 4 and 5. To a stirred 20 mL toluene solution of 1 (0.5862
g, 0.465 mmol) was added dropwise 0.47 mL of a commercially
prepared 1.0 M hexanes solution of diisobutylaluminum hy-
dride in a glovebox. A color change from reddish-brown to
yellowish-brown was immediate. The solution was stirred for
24 h, and the solvents were removed under vacuum. The
resulting dark yellow-brown solid was taken up in hexanes.
The solution was allowed to rest for 16 h undisturbed, and an
initial crop of yellow chip crystals was recovered (281.9 mg,
45.1%, of 4). The remaining brown hexanes solution was then
refrigerated at -60 °C for 2 days, allowing recovery of a crop
of orange prismatic crystals (270.0 mg, 43.2%, of 5, overall
yield 88.3%). The crystals were used for elemental, X-ray
crystallographic, and NMR spectroscopic analyses.
1
eter was programmed to observe consecutive sets of H{31P}
and 31P{1H} spectra every 15 min for 25 h. After initial
spectrometer calibration was performed, the sample was
ejected and 4.5 mg of neat DIBAL (0.32 mmol, 1 equiv) in
roughly 0.25 mL of C6D6 was added as a bolus through the
septum using a 20-gauge hypodermic needle. The reagents
were mixed by brief inversion of the tube before the sample
was returned to the probe and automated acquisition was
begun. Individual resonances were integrated with respect to
the internal standard. Similar experiments using d2-1 and
15N2-1 were also conducted.
Synthesis of d2-2 and Its Decomposition. A toluene
solution of d2-1 in a sealed flask was treated in a manner
similar to the preparation of 2, giving d2-2. The solution was
frozen in liquid nitrogen and sealed under vacuum, after which
it was allowed to stir overnight. The headspace gas was
analyzed by GC/MS, showing D2 gas and no HD gas.
Interconversion of 4 and 5. In a glovebox, 2:1 and 1:2
mixtures of 4 and 5 were dissolved in C7D8 and transferred to
an 8 in. Wilmad NMR tube, which was capped with a plastic
stopper and sealed with ParaFilm laboratory film. Samples
were observed initially and again after 24 h, after which no
change in integration values was observed. In both cases
heating overnight at 60 °C led to 1:1 mixtures with minimal
loss of 31P-active species.
Characterization of anti-[NPN-Al(H)C4H9]Ta(µ-N)(µ-
1
N)Ta[NPN], 4. H NMR (400 MHz, C6D6, 300 K): δ -0.62,
-0.27, -0.15, -0.12, 0.03, 0.15, 0.19, 0.59 (s, 3 H each, 24H
3
total, SiCH3), 1.06, 1.24 (d, 3H each, JHH ) 7 Hz, (CH3)2-
CHCH2Al), 1.05 (m, 1H (CH3)2CHCH2Al), 0.57, 0.68 (AMX, 1H
each, (CH3)2CHCH2Al), 0.62, 0.65, 1.19, 1.36, 1.42, 1.58, 1.76,
1.82 (AMX, 8H total, PCH2), 1.23 (d, 1H, JHH ) 11.3 Hz, Al-
H), 6.38, 6.52, 6.60, 6.84, 6.90, 6.95, 7.08, 7.12, 7.26, 7.32, 7.41
(phenyl protons, various multiplicities, total 22 H), 8.09 (AMX,
2H, JHH ) 6.95 Hz, JPH ) 1.70 Hz, PPh o-H) 8.21 (AMX, 2H,
2
JHH ) 6.95 Hz, JPH ) 1.28 Hz, PPh o-H), 17.36 (dd, 1H, JHH
2
) 11.3 Hz, JHP ) 16.6 Hz, TaH). 13C{1H} NMR (100.6 MHz,
C6H6, 300 K): δ -1.52, 0.56, 2.62, 3.36, 3.63, 4.33,5.12, 5.94
(s, SiCH3), 0.53, 1.02, 1.50, 2.60, 14.08, 15.44, 18.40 (d, PCH2),
24.61 (b, Al-CH2), 21.17, 26.73, 28.53 (Al-iBu), 136.71, 134.91
(ipso P-C6H5), 130.56, 133.00 (ortho P-C6H5), 119.25, 121.52,
122.70, 123.74, 124.04, 126.30, 128.33, 128.54, 128.91, 129.56
(phenyl ring carbons). Note some proton and carbon resonances
were eclipsed by solvent. 31P{1H} NMR (161.9 MHz, C6D6, 300
K): δ -22.9 (s, [NPN] ligand), 19.8 (s [NPN] ligand). 29Si-DEPT
NMR (79.5 MHz, C6D6, 300 K): δ 7.52 (d, 2JPSi ) 5.9 Hz), 8.77
Acknowledgment. We thank NSERC of Canada for
funding (Discovery Grant to M.D.F., PGS B Scholarship
to B.A.M.).
(d, 2JPSi ) 15.1 Hz), 9.24 (d, 2JPSi ) 14.2 Hz), 11.32 (d, 2JPSi
)
12.9 Hz); [NPN] ligand. Anal. Calcd for C52H73N6P2Si4Ta2Al:
C, 47.42; H, 5.47; N, 6.25. Found: C, 47.47; H, 5.56; N, 6.29.
Supporting Information Available: X-ray crystallo-
graphic data for 4 and 5 in CIF format. This material is
available free of charge via the Internet at http://pubs.acs.org.
Characterization of syn-[NPN-Al(H)C4H9]Ta(µ-N)(µ-
1
N)Ta[NPN], 5. H NMR (400 MHz, C6D6, 300 K): δ -0.58-
(3H), -0.22 (3H), -0.10 (6H), 0.16 (3H), 0.24 (3H), 0.36 (3H),
OM050208Z