Dinitrogen complexation and extent of N-N activation within the group 6 "End-On-Bridged" dinuclear complexes, {(η5-C 5Me5)M[N(i-Pr)C(Me)N(i-Pr)]}2(μ- η1:η1-N2) (M = Mo and W)

Article abstract of DOI:10.1021/ja100469f

Chemical reduction of Cp*M[N(i-Pr)C(Me)N(i-Pr)]Cl₃ (Cp* = η⁵-C₅Me₅) (1, M = Mo) and (2, M = W) using 0.5% NaHg in THF provided excellent yields of the diamagnetic dinuclear end-on-bridged dinitrogen complexes {C

Full text of DOI:10.1021/ja100469f

Published on Web 08/13/2010
Dinitrogen Complexation and Extent of Nt N Activation within
the Group 6 “End-On-Bridged” Dinuclear Complexes,
{(η5-C5Me5)M[N(i-Pr)C(Me)N(i-Pr)]}2(µ-η1:η1-N2) (M ) Mo and W)
Philip P. Fontaine,† Brendan L. Yonke, Peter Y. Zavalij, and Lawrence R. Sita*
Department of Chemistry and Biochemistry, UniVersity of Maryland, College Park, Maryland 20742
Received January 18, 2010; E-mail: lsita@umd.edu
Abstract: Chemical reduction of Cp*M[N(i-Pr)C(Me)N(i-Pr)]Cl3 (Cp* ) η5-C5Me5) (1, M ) Mo) and (2, M ) W)
using 0.5% NaHg in THF provided excellent yields of the diamagnetic dinuclear end-on-bridged dinitrogen
complexes {Cp*M[N(i-Pr)C(Me)N(i-Pr)]}2(µ-η1:η1-N2) (6, M ) Mo) and (8, M ) W), respectively. Chemical
reduction of Cp*Mo[N(i-Pr)C(NMe2)N(i-Pr)]Cl2 (4) with 3 equiv of KC8 in THF similarly yielded diamagnetic
{Cp*Mo[N(i-Pr)C(NMe2)N(i-Pr)]}2(µ-η1:η1-N2) (7). Single-crystal X-ray analyses of 7 and 8 confirmed the dinuclear
end-on-bridged µ-η1:η1-N2 coordination mode and the solid-state molecular structures of these compounds
provided d(NN) values of 1.267(2) and 1.277(8) Å for 7 and 8, respectively. Based on a comparison of 15N
NMR spectra for 15N2 (99%)-labeled 6 and 15N2 (99%)-labeled 8, as well as similarities in chemical reactivity, a
dinuclear µ-η1:η1-N2 structure for 6 is further proposed. For comparison with a first-row metal derivative, chemical
reduction of Cp*Ti[N(i-Pr)C(Me)N(i-Pr)]Cl2 (9) with KC8 in THF was conducted to provide {Cp*Ti[N(i-Pr)C(Me)N(i-
Pr)]}2(µ-η1:η1-N2) (10) for which a d(NN) value of 1.270(2) Å was obtained through X-ray crystallography.
Compounds 6-8 were all found to be thermally robust in toluene solution up to temperatures of at least 100 °C,
and 6 and 8 were determined to be inert toward the addition of H2 or H3SiPh under a variety of conditions.
Single-crystal X-ray analysis of meso-{Cp*Mo(H)[N(i-Pr)C(Me)N(i-Pr)]}2(µ-η1:η1-N2) (meso-11), which was
serendipitously isolated as a product of attempted alkylation of Cp*Mo[N(i-Pr)C(Me)N(i-Pr)]Cl2 (3) with 2 equiv
of n-butyllithium, revealed a smaller d(NN) value of 1.189(4) Å that is consistent with two Mo(IV,d2) centers
connected by a bridging diazenido, [µ-N2]2-, moiety. Moreover, meso-11 was found to undergo clean
dehydrogenation in solution at 50 °C to provide 6 via a first-order process. Chemical oxidation of 8 with an
excess of PbCl2 in toluene solution at 25 °C provided a 1:1 mixture of rac- and meso-{Cp*W(Cl)[N(i-Pr)C(Me)N(i-
Pr)]}2(µ-η1:η1-N2) (12); both isomers of which provided solid-state structures through X-ray analyses that are
consistent with an electronic configuration comprised of two W(IV,d2) centers linked through a bridging [N2]2-
group [cf. for rac-12, d(NN) ) 1.206(9) Å, and for meso-12, d(NN) ) 1.192(3) Å]. Finally, treatment of 6 and 8
with either 4 equiv of CNAr (Ar ) 3,5-Me2C6H3) or an excess of CO in toluene provided excellent yields of
Cp*M[N(i-Pr)C(Me)N(i-Pr)](CNAr)2 (13, M ) Mo and 14, M ) W) and Cp*M[N(i-Pr)C(Me)N(i-Pr)](CO)2 (15, M
) Mo and 16, M ) W), respectively. Single-crystal X-ray analyses of 13-16, along with observation of reduced
IR vibrational νCN or νCO bond-stretching frequencies, provide strong support for the electron-rich character of
the Cp*M[N(i-Pr)C(Me)N(i-Pr)] fragment that can engage in a high degree of back-donation with moderate to
strong π-acceptors, such as N2, CNR, and CO. The collective results of this work are analyzed in terms of the
possible steric and electronic factors that contribute to preferred mode of µ-N2 coordination and the extent of
Nt N activation, including complete N-N bond scission, within the now completed experimentally-derived ligand-
centered isostructural series of {Cp*M[N(i-Pr)C(Me)N(i-Pr)]}2(µ-N2) compounds where M ) Ti, Zr, Hf, Ta, Mo,
and W.
Introduction
the steric and electronic environment experienced by the metal
centers can be held constant while the nature of µ-N2 coordina-
tion and the extent of Nt N activation is allowed to vary as a
Experimental and theoretical investigations have now con-
clusively established that substantial Nt N bond elongation of
dinitrogen, also referred to as N2 “activation”, can be ac-
complished through various bridging µ-N2 coordination modes
within different classes of dinuclear complexes of general
formula [LnM]2(µ-N2), where M is an early transition metal from
groups 4-6 of the periodic table.1-3 A remaining challenge
that has proven difficult to overcome, however, is the establish-
ment of a single common supporting ligand platform in which
(1) For reviews, see: (a) Fryzuk, M. D.; Johnson, S. A. Coord. Chem.
ReV. 2000, 200-202, 379–409. (b) Shaver, M. P.; Fryzuk, M. D. AdV.
Synth. Catal. 2003, 345, 1061–1076. (c) Gambarotta, S.; Scott, J.
Angew. Chem., Int. Ed. 2004, 43, 5298–5308. (d) MacKay, B. A.;
Fryzuk, M. D. Chem. ReV. 2004, 104, 385–401. (e) MacLachlan, E. A.;
Fryzuk, M. D. Organometallics 2006, 25, 1530–1543. (f) Kuganathan,
N.; Green, J. C.; Himmel, H. J. New. J. Chem. 2006, 30, 1253–1262.
(g) Ohki, Y.; Fryzuk, M. D. Angew. Chem., Int. Ed. 2007, 46, 3180–
3183. (h) Chirik, P. J. Dalton Trans. 2007, 16–25. (i) Fryzuk, M. D.
Acc. Chem. Res. 2009, 42, 127–133. (j) Chirik, P. J. Organometallics
2010, 29, 1500–1517.
† Present address: Dow Chemical Company, Freeport, TX.
9
10.1021/ja100469f  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 12273–12285 12273

A R T I C L E S
Fontaine et al.
η2:η2-N2) (Ia), where M ) Zr or Hf.8 Indeed, for the specific
case of M ) Hf, R ) R2 ) Me and R1 ) R3 ) Et in Ia, the
experimentally determined solid-state d(NN) value of 1.635(5)
Å now appears to be the longest known N-N bond distance
within a molecularly discrete [LnM]2(µ-N2) compound.1 For
comparison, the corresponding d(NN) values for free N2 and
free hydrazine (N2H4) are 1.0974 Å and 1.460 Å, respectively.9
On the other hand, in spite of these extreme values for d(NN)
in Ia, retention of a N-N bonding interaction in these
diamagnetic compounds is consistent with a ground-state
electronic structure in which two M(IV,d0) centers are linked
through a highly reduced, side-on-bridged hydrazido [µ-η2:η2-
N2]4- moiety. Moreover, an inherent corollary to this formaliza-
tion is that complete Nt N bond cleavage cannot occur in these
compounds as formation of a metal nitrido end-product would
require invoking the involvement of an inaccessible M(V)
oxidation state for these group 4 metals.
Chart 1
function of the metal’s group and row position, formal oxidation
state, and dn electron count.4 A highly desirable outcome of
having ready access to “ligand-centered” isostructural series of
[LnM]2(µ-N2) complexes is the experimental emergence and
successful theoretical analysis of metal-dependent trends for
dinitrogen activation that can aid the development of molecular
catalysts for the direct hydrogenation of dinitrogen to ammonia
that further operate with greater energy efficiencies than that
presently required by the high-temperature, high-pressure
Haber-Bosch commercial process.5
In contrast to group 4 derivatives of I, the paramagnetic, third-
row, group 5 metal analogue, {Cp*Ta[N(i-Pr)C(Me)N(i-Pr)]}2(µ-
η1:η1-N2) (Cp* ) η5-C5Me5) (Ib), was found to adopt the “end-
on-bridged” µ-η1:η1-N2 coordination mode shown in Chart 1
in which two formal M(IV,d1) centers are connected in linear
fashion through a highly reduced [µ-N2]4- bridge.7 Surprisingly,
however, although Ib possesses a smaller solid-state d(NN)
value of 1.313(4) Å, this compound quantitatively converts in
solution at temperatures above 0 °C to the dinuclear bis(µ-
nitrido) species {Cp*Ta[N(i-Pr)C(Me)N(i-Pr)(µ-N)]}2 (II), the
structure of which is also depicted in Chart 1. The solid-state
d(NN) value of 2.59 Å obtained for the planar [M(µ-N)]2 four-
membered ring of II provides strong evidence that complete
N-N bond cleavage has occurred with concomitant generation
of two M(V,d0) metal centers during the Ib f II thermal
process.6,10
Recently, we have reported that the η5-cyclopentadienyl/η2-
amidinate (CpAm) ligand combination can be used to access
several different group 4 and group 5 dinuclear dinitrogen
complexesofgeneralformula{(η5-C5R5)M[N(R1)C(R2)N(R3)]}2(µ-
N2) (I).6,7 With the second- and third-row group 4 metals, a
“side-on-bridged” µ-η2:η2-N2 coordination mode is obtained that
is characterized by having an extremely large value for the N-N
distance parameter, d(NN), that further correlates with the degree
of nonplanarity of the M2N2 four-membered ring that is defined
by the generic structure depicted in Chart 1 for the series of
compoundsrepresentedby{(η5-C5Me4R)M[N(R1)C(R2)N(R3)]}2(µ-
With successful results for group 4 and group 5 metal
derivatives of I in hand, a reasonable next step to undertake in
our efforts was to attempt to extend the experimental structural
database of compounds to include group 6 metal analogues.
Herein, we now report the successful realization of this goal
that serves to establish an important ligand-based isostructural
series for {Cp*M[N(i-Pr)C(Me)N(i-Pr)]}2(µ-N2) in which M can
now be an early transition metal chosen from among the group
4 (M ) Ti, Zr, and Hf), group 5 (M ) Ta), and group 6 (M )
Mo and W). Through comparisons of the solid-state structures,
thermal stabilities, and chemical reactivities of these various
derivatives of I, it is now possible to contemplate the nature of
the steric and electronic factors of the CpAm ligand platform
that govern the preference for a particular metal/µ-N2 coordina-
tion mode combination, the extent of Nt N activation, and
specific details of reaction energy profiles that are associated
(2) For recent experimental studies related to dinuclear N2 activation
leading to complete N-N bond cleavage, see: (a) Akagi, F.; Matsuo,
T.; Kawaguchi, H. Angew. Chem., Int. Ed. 2007, 46, 8778–8781. (b)
Nikiforov, G. B.; Vidyaratne, I.; Gambarotta, S.; Korobkov, I. Angew.
Chem., Int. Ed. 2009, 48, 7415–7419. (c) Knobloch, D. J.; Lobkovsky,
E.; Chirik, P. J. Nat. Chem. 2010, 2, 30–35.
(3) For relevant theoretical studies, see: (a) Fryzuk, M. D.; Haddad, T. S.;
Mylvaganam, M.; McConville, D. H.; Rettig, S. J. J. Am. Chem. Soc.
1993, 115, 2782–2792. (b) Cui, Q.; Musaev, D. G.; Svensson, M.;
Sieber, S.; Morokuma, K. J. Am. Chem. Soc. 1995, 117, 12366–12367.
(c) Khoroshun, D. V.; Musaev, D. G.; Morokuma, K. Organometallics
1999, 18, 5653–5660. (d) Basch, H.; Musaev, D. G.; Morokuma, K.
Organometallics 2000, 19, 3393–3403. (e) Pool, J. A.; Bernskoetter,
W. H.; Chirik, P. J. J. Am. Chem. Soc. 2004, 126, 14326–14327. (f)
Bobadova-Parvanova, P.; Wang, Q.; Morokuma, K.; Musaev, D. G.
Angew. Chem., Int. Ed. 2005, 44, 7101–7103. (g) Studt, F.; MacKay,
B. A.; Fryzuk, M. D.; Tuczek, F. Dalton Trans. 2006, 1137–1140.
(h) Graham, D. C.; Beran, G. J. O.; Head-Gordon, M.; Christian, G.;
Stranger, R.; Yates, B. F. J. Phys. Chem. A 2005, 109, 6762–6772.
(i) Martinez, S.; Morokuma, K.; Musaev, D. G. Organometallics 2007,
26, 5978–5986. (j) Tanaka, H.; Shiota, Y.; Matsuo, T.; Kawaguchi,
H.; Yoshizawa, K. Inorg. Chem. 2009, 48, 3875–3881.
(4) For instance, for a homologous series of transition metal dinuclear
N2 complexes, see: (a) Chomitz, W. A.; Arnold, J. Chem.Commun
2007, 4797–4799. For lanthanide metals, see: (b) Evans, W. J.; Fang,
M.; Zucchi, G.; Furche, F.; Ziller, J. W.; Hoekstra, R. M.; Zink, J. I.
J. Am. Chem. Soc. 2009, 131, 11195–11202, and references cited
therein.
(8) For discussions of the electronic and steric factors that govern “side-
on-bridged” µ-η2:η2-N2 complexation within group 4 dinuclear [LnM]2(µ-
N2) complexes, see refs 1g-j, 3a, and d and references cited therein.
(9) (a) Wilkinson, P. G.; Houk, N. B. J. Chem. Phys. 1956, 24, 528. (b)
Sutton, L. E., Ed.; Tables of Interatomic Distances and Configurations
in Molecules and Ions; Chemical Society Special Publications; The
Chemical Society: London, 1958, Vol. 11.
(10) For other early transition metal [M(µ-N)]2 compounds obtained through
formal dinitrogen activation and N-N bond cleavage, see: (a) Haddad,
T. S.; Aistars, A.; Ziller, J. W.; Doherty, N. M. Organometallics 1993,
12, 2420–2422. (b) Clentsmith, G. K. B.; Bates, V. M. E.; Hitchcock,
P. B.; Cloke, F. G. N. J. Am. Chem. Soc. 1999, 121, 10444–10445.
(c) Caselli, A.; Solari, E.; Scopelliti, R.; Floriani, C.; Re, N.; Rizzoli,
C.; Chiesi-Villa, A. J. Am. Chem. Soc. 2000, 122, 3652–3670. (d)
Fryzuk, M. D.; Kozak, C. M.; Bowdridge, M. R.; Patrick, B. O.; Rettig,
S. J. J. Am. Chem. Soc. 2002, 124, 8389–8397.
(5) (a) Travis, T. Chem. Ind. 1993, 15, 581–585. (b) Appl, M. Chem.
Tech. 1994, 46, 125–135. (c) Smil, V. Enriching the Earth: Fritz
Haber, Carl Bosch, and the Transformation of World Food Production;
MIT Press: Cambridge, MA, 2001. (d) Smil, V. Nature 1999, 401,
429. (e) DOE/EIA-0484 Report, International Energy Outlook, 2007;
OIAF: Washington, DC, 2007.
(6) Hirotsu, M.; Fontaine, P. P.; Zavalij, P. Y.; Sita, L. R J. Am. Chem.
Soc. 2007, 129, 12690–12692.
(7) Hirotsu, M.; Fontaine, P. P.; Epshteyn, A.; Zavalij, P. Y.; Sita, L. R.
J. Am. Chem. Soc. 2007, 129, 9284–9285.
9
12274 J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010

N2 Complexation and Extent of Nt N Activation
A R T I C L E S
Figure 1. Molecular structure (30% thermal ellipsoids) of (a) 3 and (b) 5. Hydrogen atoms have been removed for the sake of clarity.14
Scheme 1
with complete N-N bond cleavage and N-atom functionaliza-
for M ) W, the best means to obtain compound 5 (X ) Me) in
tion processes.
high yields was determined to be through chemical reduction
of the trichloride 2 using 1 equiv of 0.5% sodium amalgam
(NaHg) in tetrahydrofuran (THF) according to Scheme 1.
Results
Although not extensively relied upon in the present work,
the group 6 dichloride compounds 3 and 5 are proving to be
useful for exploring a broader range of midvalent group 6
organometallic chemistry. These compounds also now complete
a ligand-centered isostructural series for Cp*M[N(i-Pr)C(X)N(i-
Pr)]Cl2 that spans the early transition metal series and, more
specifically, where M ) Ti, Zr6, Hf6, Ta7, Mo, and W. Since,
to the best of our knowledge, CpMo[N(Ph)C(H)N(Ph)]Cl2 (Cp
) η5-C5H5) previously represented the only known example of
a group 6 metal CpAm derivative,12 compounds 3 and 5 were
subjected to further study. In this regard, both compounds proved
to be thermally stable in solution and the solid-state for extended
periods of time. In solution, both of the molybdenum derivatives,
3 and 4, were found to be high-spin and paramagnetic, while
the third-row tungsten congener 5 is low-spin and diamagnetic,
and these observations are consistent with expectations for a
square-pyramidal ligand-field geometry for a 16-electron,
M(IV,d2) complex.13 Single-crystal X-ray analyses of 3 and 5
confirmed the geometry of the structures depicted in Scheme 1
and Figure 1 further presents a side-by-side comparison of the
respective solid-state molecular structures.14 In brief, the gross
structural features of 3 and 5 are geometrically similar to those
A. Synthesis of Cp*M[N(i-Pr)C(X)N(i-Pr)]Cln (M ) Mo
and W, X ) Me or NMe2, n ) 2 or 3). Scheme 1 summarizes
the synthetic methods that were employed in the present work
to obtain group 6 metal chloride complexes that could serve as
viable precursors to the corresponding derivatives of I through
chemical reduction.6,7 To begin, synthesis of the M(V,d1)
trichlorides, Cp*M[N(i-Pr)C(Me)N(i-Pr)]Cl3, where M ) Mo
(1) and W (2), proceeded with good yields through reaction of
the respective Cp*MCl4 starting materials11 with 1 equiv of the
lithium amidinate salt, Li[N(i-Pr)C(Me)N(i-Pr)]; the latter
reagent being readily prepared through the addition of methyl
lithium to N,N-diisopropylcarbodiimide and subsequently iso-
lated in pure form. Although 1 and 2 were eventually discovered
to be excellent precursors to the desired dinuclear µ-N2
complexes, it was initially thought that higher yields might be
ensured through chemical reduction of the corresponding
M(IV,d2) dichloride precursors. As Scheme 1 reveals, a very
practical route to Cp*M[N(i-Pr)C(X)N(i-Pr)]Cl2 for M ) Mo
involved a “one-pot” procedure starting from Cp*MoCl4 that
employs a second equivalent of Li[N(i-Pr)C(X)N(i-Pr)] as an
in situ sacrificial reductant. In this way, compounds 3 (X )
Me) and 4 (X ) NMe2) were readily prepared in small but
sufficient quantities for synthetic evaluation. On the other hand,
(12) Romao, C. C.; Royo, B. J. Organomet. Chem. 2002, 663, 78–82.
(13) (a) Poli, R. Chem. ReV. 1996, 96, 2135–2204. (b) Baker, R. T.;
Calabrese, J. C.; Harlow, R. L.; Williams, I. D. Organometallics 1993,
12, 830–841.
(11) (a) Murray, R. C.; Blum, L.; Liu, A. H.; Schrock, R. R. Organome-
tallics 1985, 4, 953–954. (b) Gomez, S. P.; Jime´nez, I.; Mart´ın, A.;
Perdraz, T.; Royo, P.; Selle´s, A.; Va´zquez de Miguel, A. Inorg. Chim.
Acta 1998, 273, 270–278.
(14) Details are provided in the Supporting Information.
9
J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010 12275

A R T I C L E S
Fontaine et al.
Scheme 2
that we have previously reported for a number of related group
4 and group 5 Cp*M[N(R1)C(X)N(R2)]Cl2 derivatives.15 Be-
tween the two structures, the metal-related bond length and bond
angle metrical parameters for 3 and 5 are also quite similar [cf.
for 3 Mo(1)-N(1) ) 2.112(3) Å, Mo(1)-Cl(1) ) 2.4104(9)
Å, N(1)-Mo(1)-N(1A) ) 61.31(14)°, Cl(1)-Mo(1)-Cl(1A)
) 86.59(5)°; for 5 W(1)-N(11) ) 2.0707(15) Å, W(1)-N(12)
) 2.0764(15) Å, W(1)-Cl(1) ) 2.4130(5) Å, W(1)-Cl(2) )
2.4215(5)Å,N(11)-W(1)-N(12))63.27(6)°,Cl(1)-W(1)-Cl(2)
) 83.221(17)°]. However, the structure of the tungsten deriva-
tive 5 displays a significant nonplanarity associated with the
four-membered ring amidinate fragment, as evidenced by a value
of 21.1° for the angle between the two planes defined by
N(11)-W(1)-N(12) and N(11)-C(10)-N(12), whereas the
corresponding angle in 3 is only 5.1°. A possible manifestation
of this amidinate ring nonplanarity for 5 appears in the 1H NMR
(400 MHz, benzene-d6, 25 °C) spectrum of this compound in
which the resonance at δ 4.02 ppm for the methyl group of the
amidinate fragment, NC(CH3)N, is shifted far downfield from
the expected chemical shift range of 1.4-1.6 ppm that is
typically observed for the group 4 analogues.15
Scheme 3
Cp*Ti[N(i-Pr)C(Me)N(i-Pr)]Cl2 (9) that was prepared as part
of another investigation, the opportunity presented itself to
attempt the synthesis of a first-row group 4 analogue of I where
M ) Ti. Thus, as Scheme 3 reveals, low-temperature chemical
reduction of 9 employing 3 equiv of KC8 in THF did, in fact,
provide a gratifying 49% yield of a dark green, diamagnetic
crystalline material for which 1H NMR and elemental analyses
were satisfactorily consistent with the expected formulation for
{Cp*Ti[N(i-Pr)C(Me)N(i-Pr)]}2(µ-N2) (10).
C. Solid-State Molecular Structures of {Cp*M[N(i-Pr)-
C(X)N(i-Pr)]}2(µ-η1:η1-N2) (M ) Ti, Mo, and W; X ) Me or
NMe2) (7, 8, and 10). Unfortunately, due to high solubility in
pentane and poor crystallinity, single crystals of compound 6
suitable for crystallographic analysis could not be obtained even
after expending considerable effort. Thus, the main purpose for
including the η2-guanidinate, [N(i-Pr)C(NMe2)N(i-Pr)], series
of compounds, 4 and 7, in the present study was to possibly
obtain a crystalline analogue of 6 that would be suitable for
X-ray crystallography. Fortunately, single-crystals of 7 did,
indeed, prove to be much more amenable for X-ray analysis,
as did those for compounds 8 and 10. Figures 2-4 provide the
solid-state molecular structures of 7, 8, and 10, respectively,
while Table 1 presents a compilation of the selected metrical
B. Synthesis and Solution Characterization of {Cp*M[N-
(i-Pr)C(X)N(i-Pr)]}2(µ-η1:η1-N2) (M ) Ti, Mo, and W, X )
Me or NMe2). Following our previously published synthetic
procedure,6,7 we first sought to obtain the desired group 6
derivatives of I through chemical reductions of the respective
M(IV,d2) dichlorides 3-5. Accordingly, as Scheme 2 presents,
chemical reduction of either of the molybdenum compounds, 3
or 4, at low temperature under an atmosphere of N2 using 3
equiv of potassium graphite (KC8) in THF, provided, in both
cases, a modest yield of a diamagnetic, yellow-brown crystalline
1
material for which H NMR and elemental (C, H, and N)
analyses were fully consistent with formation of the expected
{Cp*Mo[N(i-Pr)C(X)N(i-Pr)]}2(µ-N2) products, where X ) Me
(6) and X ) NMe2 (7), respectively. It was subsequently
determined, however, that direct chemical reduction of the
trichloride 1 employing 3 equiv of 0.5% NaHg in THF at 25 °C
reproducibly provided a high yield of 6, and accordingly, this
latter route became the preferred one for conveniently accessing
larger quantities of this material. In similar fashion, the tungsten
trichloride 2 could also be reduced with 3 equiv of 0.5% NaHg
in THF according to Scheme 2 to provide a 92% yield of a
1
forest green, crystalline, diamagnetic compound for which H
NMR and elemental analyses were once again fully consistent
with the empirical formula expected for the desired product
{Cp*W[N(i-Pr)C(Me)N(i-Pr)]}2(µ-N2) (8).
Although group 6 metal derivatives of I were the primary
target goal of the present study, with the availability of
(15) See, for instance: (a) Sita, L. R.; Babcock, J. R. Organometallics 1998,
17, 5228–5230. (b) Zhang, Y.; Reeder, E. K.; Keaton, R. J.; Sita, L. R.
Organometallics 2004, 23, 3512–352. (c) Kissounko, D. A.; Fettinger,
J. C.; Sita, L. R. Inorg. Chim. Acta 2003, 345, 121–129. (d) Epshteyn,
A.; Zavalij, P. Y.; Sita, L. R. J. Am. Chem. Soc. 2006, 128, 16052–
16053.
Figure 2. Molecular structure (30% thermal ellipsoids) of 7. Hydrogen
atoms have been removed for the sake of clarity.14
9
12276 J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010

N2 Complexation and Extent of Nt N Activation
A R T I C L E S
Table 1. Comparison of MNNM Structural Parameters for [LnM]2(µ-η1:η1-N2) Complexes
compound
d(MN) (Å)
d(NN) (Å)
θ(MNN) (deg)
ref
7
1.7944(12)
1.816(4)
1.720(8)
1.7749(18), 1.7741(18)
1.899(2)
1.9032(17)
1.885(7)
1.807(2)
1.870(2), 1.872(2)
1.835(3), 1.841(1)
1.798(2)
1.819(2), 1.821(3)
1.742(17), 1.763(18)
1.907(8)
1.930(4)
1.845(11)
1.776(11), 1.735(11)
1.267(2)
1.277(8)
1.402(17)
1.270(2)
1.189(4)
1.192(3)
1.206(9)
1.313(4)
1.217(2)
1.239(4)
1.265(5)
1.236(3)
1.334(26)
1.20(2)
171.26(15)
176.7(5)
177.4(11)
166.71(15), 166.96(16)
167.4(3)
169.0(2)
177.7(6)
172.7(3)
179.87(14)
this work
this work
this work
this work
this work
this work
this work
7
16c
16c
16c
17e
17b, c
17g
17i
17j
17a
8
8a
10
meso-11
meso-12
rac-12
Ib
{[(t-Bu)ArN]3Mo}2(µ-N2) (III)
[III][B(ArF)4]
[III][B(ArF)4]2
{Cp*Me3Mo}2(µ-N2)
{Cp*Me3W}2(µ-N2)
{[R3SiN3N]Mo}2(µ-N2)
{[Ar′N3N]Mo}2(µ-N2)
{[NpN3N]W}2(µ-N2)
[W(PhCt CPh)(dme)Cl2(µ-N2)
172.0(2)
167.0(16) 170.2(16)
178(1)
1.186(7)
1.39(2)
1.292(16)
178.9(5)
175.6(10)
a Obtained as a 1:1 cocrystal with meso-12.14
Ib.7 On the other hand, 7, 8, and 10 all display somewhat smaller
d(NN) values of 1.267(2), 1.277(8), and 1.270(2) Å, respec-
tively, compared with that observed for Ib [d(NN) )1.313(4)
Å]. Although a more thorough analysis will be presented shortly,
the observed d(NN) bond lengths for 7, 8, and 10 qualitatively
imply that a moderate degree of Nt N activation is present
within the ground-state structures of these compounds.1,16-18
D. Solution Structures and Thermal Stabilities of
{Cp*M[N(i-Pr)C(X)N(i-Pr)]}2(µ-η1:η1-N2) (M ) Mo and W;
X ) Me or NMe2) (6-8). In order to assist with structural
assignments in solution, 15N2-labeled 6 and 15N2-labeled 8 were
(16) (a) Laplaza, C. E.; Cummins, C. C. Science 1995, 268, 861–863. (b)
Laplaza, C. E.; Johnson, M. J. A.; Peters, J. C.; Odom, A. L.; Kim,
E.; Cummins, C. C.; George, G. N.; Pickering, I. J. J. Am. Chem.
Soc. 1996, 118, 8623–8638. (c) Curley, J. J.; Cook, T. R.; Reece, S. Y.;
Mu¨ller, P.; Cummins, C. C. J. Am. Chem. Soc. 2008, 130, 9394–9405.
(17) (a) Churchill, M. R.; Li, Y. J.; Theopold, K. H.; Schrock, R. R. Inorg.
Chem. 1984, 23, 4472–4476. (b) Murray, R. C.; Schrock, R. R. J. Am.
Chem. Soc. 1985, 107, 4557–4558. (c) Churchill, M. R.; Li, Y. J. J.
Organomet. Chem. 1986, 301, 49–59. (d) O’Regan, M. B.; Liu, A. H.;
Finch, W. C.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc. 1990,
112, 4331–4338. (e) Schrock, R. R.; Kolodziej, R. M.; Liu, A. H.;
Davis, W. M.; Vale, M. G. J. Am. Chem. Soc. 1990, 112, 4338–4345.
(f) O’Regan, M. B.; Liu, A. H.; Finch, W. C.; Schrock, R. R.; Davis,
W. M. J. Am. Chem. Soc. 1990, 112, 4331–4338. (g) Shih, K. Y.;
Schrock, R. R.; Kempe, R.; Davis, W. M. J. Am. Chem. Soc. 1994,
116, 8804–8805. (h) Kol, M.; Schrock, R. R.; Kempe, R.; Davis, W. M.
J. Am. Chem. Soc. 1994, 116, 4382–4390. (i) Greco, G. E.; Schrock,
R. R. Inorg. Chem. 2001, 40, 3861–3878. (j) Scheer, M.; Mu¨ller, J.;
Schiffer, M.; Baum, G.; Winter, R. Chem.sEur. J. 2000, 6, 1252–
1257.
Figure 3. Molecular structure (30% thermal ellipsoids) of 8. Hydrogen
atoms have been removed for the sake of clarity.14
(18) (a) Bercaw, J. E. J. Am. Chem. Soc. 1974, 96, 5087–5095. (b) Sanner,
R. D.; Duggan, D. M.; McKenzie, T. C.; Marsh, R. E.; Bercaw, J. E.
J. Am. Chem. Soc. 1976, 98, 8358–8365. (c) de Wolf, J. M.; Blaauw,
R.; Meetsma, A.; Teuben, J. H.; Gyepes, R.; Varga, V.; Mach, K.;
Veldman, N.; Spek, A. L. Organometallics 1996, 15, 4977–4983. (d)
Hagadorn, J. R.; Arnold, J. J. Am. Chem. Soc. 1996, 118, 893–894.
(e) Beydoun, N.; Duchateau, R.; Gambarotta, S. J. Chem. Soc., Chem.
Commun. 1992, 244–246. (f) Mullins, S. M.; Duncan, A. P.; Bergman,
R. G.; Arnold, J. Inorg. Chem. 2001, 40, 6952–6963. (g) Hanna, T. E.;
Keresztes, I.; Lobkovsky, E.; Bernskoetter, W. H.; Chirik, P. J.
Organometallics 2004, 23, 3448–3458. (h) Morello, L.; Yu, P. H.;
Carmichael, C. D.; Patrick, B. O.; Fryzuk, M. D. J. Am. Chem. Soc.
2005, 127, 12796–12797. (i) Scherer, A.; Kollak, K.; Luetzen, A.;
Friedmann, M.; Haase, D.; Saak, W.; Beckhaus, R. Eur. J. Inorg.
Chem. 2005, 1003–1010. (j) Bai, G.; Wei, P.; Stephan, D. W.
Organometallics 2006, 25, 2649–2655. (k) Hanna, T. E.; Bernskoetter,
W. H.; Bouwkamp, M. W.; Lobkovsky, E.; Chirik, P. J. Organome-
tallics 2007, 26, 2431–2438. (l) Hanna, T. E.; Lobkovsky, E.; Chirik,
P. J. Organometallics 2009, 28, 4079–4088. Also, see: (m) Walter,
M. D.; Sofield, C. D.; Andersen, R. A. Organometallics 2008, 27,
2959–2970.
Figure 4. Molecular structure (30% thermal ellipsoids) of 10. Hydrogen
atoms have been removed for the sake of clarity.14
parameters for the M-N-N-M framework of these structures,
along with those for several other second- and third-row group
6 dinuclear [LnM]2(µ-N2) complexes that have appeared in the
literature.16,17 As is readily apparent from a quick survey of
the figures, all the new compounds, 7, 8, and 10, were found to
exist in the solid state as a dinuclear end-on-bridged µ-η1:η1-
N2 coordination complex in which the M-N-N-M framework
adopts a slight transoid “zigzag” conformation in a manner
similar to that previously discovered for the group 5 analogue
9
J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010 12277

A R T I C L E S
Fontaine et al.
Figure 5. Molecular structure (30% thermal ellipsoids) of meso-11. Hydrogen atoms have been removed for the sake of clarity except those for Mo(1) and
Mo(1A), which are represented by small spheres of arbitrary size.14
and co-workers (see Table 1).16c The disparity in stability among
Scheme 4
7, 8, and III serves as another reminder that bond length and
bond strength do not correlate in a simple fashion and one must
consider the broader picture of the potential energy landscape
for a compound, including the depth of energy minima, in order
to best assess the origin of stability. Finally, as a side note, a
high degree of stability in solution was found to extend to the
first-row group 4 metal derivative 10 as well.
E. Synthesis
and
Structural
Characterization
of
prepared from 3 and 2, respectively, by performing the low-
temperature KC8 chemical reductions of Scheme 2 under an
atmosphere of 15N2 (99%)-labeled N2. A 15N{1H} NMR (40.5
MHz, benzene-d6, liq. NH3 as an external reference) spectrum
of 15N2-labeled 6 obtained in this manner displayed a single
15N resonance with a chemical shift of δ 375 ppm, while that
for 15N2-labeled 8 likewise revealed a single 15N resonance with
a chemical shift of δ 351.4 ppm that was now further flanked
by two sets of 183W satellites [1J(183W-15N) ) 145 Hz and
2J(183W-15N) ) 10.5 Hz] in a manner that is most consistent
with a linear dinuclear end-on-bridged 183W-15N-15N-183W
spin system. Although not conclusive evidence, the observed
similarities in the 15N chemical shifts for 15N2-labeled 6 and
15N2-labeled 8 strongly suggests that the former compound also
exists as a dinuclear end-on-bridged µ-η1:η1-N2 coordination
complex, at least in solution. Further support for this structural
assignment is provided by considering that the chemical
reactivity profiles for 6 and 8, which will be presented herein,
are so far identical in nature.
{Cp*M(X)[N(i-Pr)C(X)N(i-Pr)]}2(µ-η1:η1-N2) (M ) Mo, X )
H and M )W, X ) Cl) (11 and 12). During the course of
investigations conducted to explore the chemical reactivity of
the group 6 metal dichlorides 3 and 5, two dinuclear CpAm
complexes closely related to I were serendipitously discovered,
and accordingly, details of these compounds are included here
as well. To begin, as presented in Scheme 4, reaction of 3 with
2 equiv of n-butyllithium in diethyl ether (Et2O) provided a 66%
yield of the dinuclear end-on-bridged µ-η1:η1-N2 dihydride
species meso-{Cp*Mo(H)[N(i-Pr)C(Me)N(i-Pr)])2(µ-η1:η1-N2)
(11) as a diamagnetic, crystalline material. Figure 5 displays
the solid-state molecular structure derived from single crystals
of meso-11 that were obtained through fractional crystallization
and subjected to single-crystal X-ray analysis.14 Table 1 also
includes for comparison purposes the following metrical pa-
rameters for the Mo-N-N-Mo framework of this structure:
Mo(1)-N(1), 1.899(2) Å, N(1)-N(1A), 1.189(4) Å, and N(1A)-
N(1)-Mo(1), 167.4(3)°.14 Here, the larger d(MN) and smaller
d(NN) values observed for meso-11, compared with the corre-
sponding values for 7, are strongly supportive of a ground-state
electronic configuration for the former compound that consists
of two Mo(IV,d2) centers that are now being linked through a
bridging diazenido, [µ-N2]2-, moiety as depicted in Scheme 4.
As Scheme 5 reveals, partial chemical reduction of the
tungsten trichloride 2 with 2.5 equiv of KC8 in THF produced
a complex mixture of products that included dichloride 5, 8,
and a set of two new diastereomeric compounds that were
eventually revealed through X-ray crystallography to be
rac,meso-{Cp*W(Cl)[N(i-Pr)C(Me)N(i-Pr)])2(µ-η1:η1-N2) (12).
Single crystals that were isolated from recrystallization of this
complex mixture at -30 °C were shown by X-ray analysis to
Of particular interest is that in contrast to the low thermal
stability displayed in solution by the group 5 derivative Ib,
compounds 6-8 are all thermally robust, and they remain
unchanged upon heating in toluene solution at 100 °C for 18 h
1
as determined by H NMR spectroscopy. This high degree of
thermal stability of 6-8 also stands in sharp contrast to {[(t-
Bu)ArN]3Mo}2(µ-η1:η1-N2) (Ar ) 3,5-Me2C6H3) (III), which
has been shown to undergo complete N-N bond cleavage in
solution at subambient temperatures to generate 2 equiv of the
terminal nitride species [[(t-Bu)ArN]3Mot N.16 A further in-
triguing aspect of this difference in stability is that both 7 and
8 possess d(NN) values that are significantly larger than that of
III, which is only 1.217(2) Å as recently reported by Cummins
9
12278 J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010

N2 Complexation and Extent of Nt N Activation
A R T I C L E S
Figure 6. Molecular structure (30% thermal ellipsoids) of rac-12. Hydrogen atoms have been removed for the sake of clarity.14
Scheme 5
Scheme 6
Scheme 7
possess a unit cell comprised of 8 and rac-12 in a 1: 1 ratio.
More intriguingly, structural parameters associated with 8 within
this 1: 1 cocrystal exhibited smaller d(MN) values and a
significantly larger d(NN) value than those determined for pure
8 [cf., 1.720(8) Å and 1.402(17) Å, respectively] (see Table 1).
The occurrence of two substantially different sets of d(MN) and
d(NN) values for 8 presumably arises as the result of subtle
differences in crystal packing forces, and this observation can
further be taken as strong evidence that 8, and perhaps other
third-row dinuclear µ-N2 complexes as well, resides in a
relatively shallow minimum on the potential energy surface with
respect to W-N and N-N bond length distortions.
Figure 6 presents the molecular structure of rac-12 for which
the following bond length and bond angle metrical parameters
were determined: W(1)-N(1), 1.885(7) Å; N(1)-N(1A), 1.206(9)
Å; and N(1A)-N(1)-W(1), 177.7(6)°.14 Once again, based on
the relative magnitudes of these bond length parameters for the
W-N-N-W framework of rac-12, vis-a-vis the corresponding
values for 8, a formal ground-state electronic structure for the
former compound can be assigned that consists of two W(IV,d2)
metal centers that are connected through an intervening diaz-
enido, [µ-N2]2-, bridge as depicted in Scheme 5.
F. Chemical Reactivity of {Cp*M[N(i-Pr)C(Me)N(i-Pr)]}2(µ-
η1:η1-N2) (M ) Mo and W) (6 and 8). Given some ambiguity
that exists regarding the formal oxidation state of the metal
centers in the diamagnetic compounds, 6-8, and more specif-
ically, whether to consider them to be M(II,d4) or M(IV,d2) (Vide
infra), a preliminary survey of the chemical reactivity of the
isostructural analogues 6 and 8 was undertaken. To begin, in
light of the successful isolation and structure determination of
meso-11, as well as our prior report that the group 5 derivative
Ib undergoes clean “1,4-addition” of H2 at 0 °C to yield
rac,meso-{Cp*Ta(H)[N(i-Pr)C(Me)N(i-Pr)])2(µ-η1:η1-N2) (IV)
for which a d(NN) value of 1.307(6) Å was obtained,7
hydrogenations of 6 and 8 were separately attempted in toluene
solution at temperatures within the range of 25 to 90 °C within
a sealed NMR tube. Unfortunately, as Scheme 6 presents, both
compounds remained inert and unchanged under all hydrogena-
tion conditions investigated. On the other hand, during the course
of these studies, it was determined that meso-11 readily
undergoes thermal conversion to 6 in toluene solution at 50 °C
through a process that is first-order in both disappearance of
starting material and appearance of product. An Eyring analysis
of this 11 f 6 transformation, conducted using variable-
temperature 1H NMR within a flame-sealed NMR tube, yielded
activation parameters of ∆Hq ) 35.3 kcal mol-1 and ∆Sq )
27.4 eu. Although evolution of H2 was not directly observed,
the positive entropic value is consistent with a mechanistic
pathway in which overall elimination of H2 occurs. Finally,
although previous success was achieved with the hydrosilylation
of both group 4 and group 5 derivatives of I,6,7 compound 6
remained inert toward H3SiPh in toluene solution up to
temperatures of 90 °C, while a mixture of 8 and H3SiPh
produced a complex mixture of products after several days in
toluene at 50 °C.
As presented in Scheme 7, treatment of compound 8 with an
excess of PbCl2 in toluene at 25 °C provided an 83% yield of
a 1:1 diastereomeric mixture of rac- and meso-12, from which
pure single crystals of meso-12 were obtained for X-ray analysis
through fractional crystallization at -30 °C. Figure 7 provides
the solid-state molecular structure of meso-12 for which the
9
J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010 12279

A R T I C L E S
Fontaine et al.
Figure 7. Molecular structure (30% thermal ellipsoids) of meso-12. Hydrogen atoms have been removed for the sake of clarity.14
Scheme 8
angles of the two metal-bonded CNAr ligands, one of which
possesses a C-N-Caryl bond angle of 142.0(2)° in 13 and a
corresponding value of 139.2(3)° in 14. In contrast, a more
obtuse bond angle is associated with the other CNAr group [cf.
a C-N-Caryl value of 167.2(3)° in 13 and 163.2(3)° in 14].
Such bond angle distortions for isocyanide ligands have
previously been attributed to a strong π-acceptor character in
which back-donation of electron density from the metal into an
antibonding π*-orbital of the Ct NR bond occurs.20 In keeping
with this π-acceptor model, the solution infrared spectra of 13
and 14 in pentane exhibited two reduced vibrational carbon-
nitrogen bond-stretching modes associated with the CNAr
ligands at νCN ) 1971 and 1966 cm-1 for 13 and at νCN ) 1952
and 1940 cm-1 for 14. Finally, despite the C1-symmetric nature
following metrical parameters were obtained: W(1)-N(1),
1.9032(17)Å; N(1)-N(1A), 1.192(3) Å; and N(1A)-N(1)-W(1),
169.0(2)°.14 Once again, the smaller d(NN) value observed for
meso-12 relative to that of 8 is in keeping with a diazenido,
[µ-N2]2-, group in this compound that is bridging two W(IV,
d2) centers.
1
of these compounds in the solid-state, H and 13C{1H} NMR
As part of a screen for group transfer and insertion chemistry,
it was discovered that both 6 and 8 reacted cleanly in toluene
solution with 4 equiv of the isocyanide, Ct NAr (Ar ) 2,6-
Me2C6H3), to provide excellent isolated yields of the diamag-
netic, crystalline bis(isocyanide) products, 13 and 14, respec-
tively, according to Scheme 8. This process is most likely driven
forward by the greater π-acceptor character of an isocyanide
ligand, compared with that of dinitrogen,19 but here it is also
interesting to note that utilization of only 2 equiv of CNAr
cleanly provided a 1:1 mixture of starting material and product,
with no evidence for any intermediates or other side products
spectroscopy support a dynamic process in solution that provides
a structure of higher Cs-symmetry in which the two isocyanide
groups become structurally (magnetically) equivalent.
The interaction of 6 and 8 with other strong π-acceptors was
further explored, and gratifyingly, it was discovered that both
compounds react rapidly with an excess of carbon monoxide
in toluene solution at room temperature to provide good yields
of the diamagnetic crystalline bis(carbonyl) derivatives, 15 and
16, according to Scheme 9. Once again, single-crystal X-ray
analyses served to confirm the structural assignments that are
depicted in the scheme.14 Solid-state infrared spectra (KBr) for
15 and 16 exhibit two reduced vibrational carbon-oxygen bond-
stretching modes associated with the CO ligands at νCO ) 1909
and 1806 cm-1 for 15 and at νCO ) 1892 and 1788 cm-1 for
16. For comparison, the IR spectrum of [Cp2Mo(CO)2][BF4]2
is characterized by νCO ) 2139 and 2108 cm-1 and that for
CpMo(η3-C5H5)(CO)2 by νCO ) 1950 and 1854 cm-1.21 Similar
1
as established by H NMR spectroscopy. Although a detailed
kinetic analysis of this ligand substitution reaction has not yet
been undertaken, it is reasonable to assume that simultaneous
coordination and competing back-donation of the first 2 equiv
of CNAr to 6 or 8 (i.e., one isocyanide per metal center) is
required to reduce π-back-donation from the metal to the µ-N2
group to the extent that the latter can then be rapidly displaced
by the next 2 equiv of CNAr. Single-crystal X-ray analyses of
13 and 14 further support the strong π-acceptor character of
the isocyanide ligands that are engaged in a high degree of back-
donation from the metal.14 Moreover, each compound displays
a significant deviation from nonlinearity of the C-N-Caryl bond
ν
CO values are reported for the tungsten congeners of these two
compounds, namely, νCO ) 2133 and 2087 cm-1 for
(20) (a) Martins, A. M.; Calhorda, M. J.; Roma˜o, C. C.; Vo¨lkl, C.; Kiprof,
P.; Filippou, A. C. J. Organomet. Chem. 1992, 423, 367–390. (b)
Filippou, A. C.; Dias, A. R.; Martins, A. M.; Roma˜o, C. C. J.
Organomet. Chem. 1993, 455, 129–135.
(21) Ascenso, J. R.; de Azevedo, C. G.; Goncalves, I. S.; Herdtweck, E.;
Moreno, D. S.; Pessanha, M.; Roma˜o, C. C. Organometallics 1995,
14, 3901–3919.
(19) Pombeiro, A. J. L.; Chatt, J.; Richards, R. L. J. Organomet. Chem.
1980, 190, 297–304.
9
12280 J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010

N2 Complexation and Extent of Nt N Activation
A R T I C L E S
Scheme 9
Scheme 10
be synthetically accessible, and if so, these should exist in the
form of dinuclear end-on-bridged µ-η1:η1-N2 complexes. On
the other hand, a first-row group 6 metal derivative of I,
where M ) Cr, may present a significant synthetic challenge
due to a metal covalent radius that is distinctly smaller at
1.39 Å than all the other early transition metals. Indeed, both
end-on-bridged and side-on-bridged µ-N2 dichromium com-
plexes are known, but these remain extremely rare.24
Moreover, Smith and co-workers25 have reported that chemi-
cal reduction of the CpAm Cr(III) monochloride,
CpCr[N(SiMe3)C(Ph)N(SiMe3)]Cl, under an atmosphere of
N2 does not provide a dinitrogen species, but rather simply
the Cr(II) product, CpCr[N(SiMe3)C(Ph)N(SiMe3)]. It re-
mains to be seen, therefore, whether a further decrease in
the steric environment of the CpAm ligand set might facilitate
µ-N2 complexation in the case of chromium. A similar
strategy is presently being explored to determine whether
dinuclear side-on-bridged µ-N2 derivatives of I for Mo and
W can likewise be obtained. In this regard, it is interesting
to note that the covalent radius of third-row group 5 Ta is
only slightly less than that for group 4 Zr and Hf (cf., 1.70
Å and 1.75 Å, respectively). Given this, a possible low-energy
pathway for Ib leading to N-N bond scission with formation
of II might actually involve an intramolecular µ-η1:η1-N2 f
µ-η2:η2-N2 structural isomerization according to Scheme 10.
The thermal stability displayed by the group 6 derivatives,
6-8, might then be simply due to an inability to similarly
adopt a dinuclear side-on-bridged µ-N2 structure as the result
of steric retardation.
[Cp2W(CO)2][BF4]2 and νCO ) 1955 and 1872 cm-1 for
CpW(η3-C5H5)(CO)2.22 Hence, these collective data support the
conclusion that the Cp*M[N(i-Pr)C(Me)N(i-Pr)] fragment is
formally quite electron-rich and it can engage in a very strong
degree of π-back-donation with π-acceptor ligands, such as CNR
and CO.
Discussion
The collective set of compounds experimentally synthesized
and characterized for {Cp*M[N(i-Pr)C(Me)N(i-Pr)]}2(µ-N2)
now encompasses M ) Ti (10), Zr (Ic),6 Hf (Id),6 Ta (Ib),7
Mo (6), and W (8). Importantly, this group represents perhaps
the most comprehensive series of ligand-centered isostructural
[LnM]2(µ-N2) complexes yet reported that span across the entire
early transition metals and that has representation from the first-
row down to the third-row metals. Not surprisingly, analyses
of solid-state structures and physiochemical properties that are
associated with this set of compounds are beginning to uncover
qualitative metal-dependent trends that may prove essential for
understanding the role of steric and electronic factors in
governing the mode of µ-N2 complexation, formal metal
oxidation state, and the extent of dinitrogen activation within
dinuclear [LnM]2(µ-N2) complexes supported by the CpAm
ligand set, which in some cases also includes examples of
complete N-N bond cleavage and N-atom functionalization
processes.6,7
A. Side-On Bridging vs End-On Bridging Dinuclear µ-N2
Coordination. While the steric and electronic effects of the
supporting η5-C5Me5/η2-[N(i-Pr)C(Me)N(i-Pr)] ligand environ-
ment might remain constant, the covalent radii for the metals
in this series are in the following order: group 4, Ti (1.60 Å),
Zr (1.75 Å) ≈ Hf (1.75 Å); group 5, V (1.53 Å), Nb (1.64 Å),
Ta (1.70 Å); group 6, Cr (1.39 Å), Mo (1.54 Å), W (1.62 Å).23
As Fryzuk1g,i and Chirik1h,j and co-workers have now amply
demonstrated and analyzed for group 4 dinuclear [LnM]2(µ-N2)
complexes, the extent of nonbonded steric interactions between
the ligand environments on the two metal centers can play a
critical role in directing the preference for dinuclear side-on-
bridged vs end-on-bridged µ-N2 coordination. Thus, within the
steric environment established by the η5-C5Me5/η2-[N(i-Pr)C-
(Me)N(i-Pr)] combination, it appears reasonable to assume that
only the second- and third-row group 4 metals, Zr and Hf, are
able to accommodate a dinuclear side-on-bridged µ-η2:η2-N2
bonding motif, whereas the smaller covalent radii for Ti, Ta,
Mo, and W now dictate that a dinuclear end-on-bridged µ-η1:
η1-N2 configuration be adopted.18k,l Although not yet a target
of synthetic effort by us, it is reasonable to predict that the first-
and second-row group 5 analogues of {Cp*M[N(i-Pr)C(Me)N(i-
Pr)]}2(µ-N2), where M ) V and Nb, respectively, should also
B. Formal Metal Oxidation States for {Cp*M[N(i-Pr)-
C(Me)N(i-Pr)]}2(µ-N2). A challenging, but useful, exercise
related to the investigation of [LnM]2(µ-N2) complexes is the
assignment of formal metal oxidation states that are a conse-
quence of the extent of N2 activation that has been achieved. In
making this assignment, the qualitative model that is commonly
adopted involves the concept of π-back-donation of electron
density from the metal orbital into the antibonding molecular
orbitals of appropriate symmetry on the µ-N2 fragment that
serves to lower N-N bond order through formal reduction of
the N2 bridge. Based on this model, the extent of µ-N2 activation
reported in the literature for a large variety of [LnM]2(µ-N2)
complexes ranges from virtually none at all, that is, the µ-N2
unit is functioning as only a σ-donor to the two metals, to a
partially reduced µ-diazenido, [N2]2-, fragment, and finally, to
the fully reduced µ-hydrazido, [N2]4-, linkage.1 Recently, Evans
and co-workers4b have added yet another member to this
spectrum of reduced forms for µ-N2 with experimental evidence
for a bridging radical anion, µ-[N2]3-
.
(24) (a) Denholm, S.; Hunter, G.; Weakley, T. J. R. J. Chem. Soc., Dalton
Trans. 1987, 2789–2791. (b) Vidyaratne, I.; Scott, J.; Gambarotta, S.;
Budzelaar, P. H. M. Inorg. Chem. 2007, 46, 7040–7049. (c) Monillas,
W. H.; Yap, G. P. A.; MacAdams, L. A.; Theopold, K. H. J. Am.
Chem. Soc. 2007, 129, 8090–8091. (d) Berben, L. A.; Kozimor, S. A.
Inorg. Chem. 2008, 47, 4639–4647.
(22) Goncalves, I. S.; Roma˜o, C. C. J. Organomet. Chem. 1995, 486, 155–
161.
(23) Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.; Echeverria,
J.; Cremades, E.; Barragan, F.; Alvarez, S. Dalton Trans. 2008, 2832–
2838.
(25) Gallant, A. J.; Smith, K. M.; Patrick, B. O. Chem. Commun. 2002,
23, 2914–2915.
9
J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010 12281

A R T I C L E S
Fontaine et al.
M(II,d4) centers are connected through a less-activated µ-N2
bridge. Importantly, while the observed d(NN) values for 7 and
8 appear to support the former formulation, the chemical
reactivity profiles displayed with strong π-acceptors certainly
suggest that the bridging µ-N2 group is easily displaced in 6-8,
and accordingly, these compounds can be viewed, to some
extent, as viable synthons for M(II,d4) (M ) Mo and W). In
reality, the true electronic configuration for 6-8 most likely
lies somewhere between the two limiting extremes. In this
regard, the present work has shown that the Cp*M[N(i-
Pr)C(Me)N(i-Pr)] fragment for M ) Mo and W can be viewed
as a very electron-rich fragment that can engage in a high degree
of π-back-donation with moderate to strong π-acceptors, such
as N2, CNR, and CO. Given the magnitude of the d(NN) values
in 7 and 8, which support a moderate to strong degree of N2
activation, the facile displacement of this N2 bridge by stronger
π-acceptors is then equivalent to a formal metal-centered
reductive process leading to formation of compounds 13-16
that unquestionably possess M(II,d4) metal centers. The closed-
shell electronic structures of 6-8 also beg the question of
whether a pathway for N-N bond scission might be more
accessible starting from an open-shell electronic configuration
for group 6 [LnM]2(µ-N2) complexes, as in the case of {[(t-
Bu)ArN]3Mo}2(µ-η1:η1-N2) (III), which is best described as
being comprised of two formal Mo(V,d1) metal centers.16 Here,
an evaluation of the one- and two-electron oxidation potentials
and oxidation products of 6 and 8 should prove very
informative,17c and these investigations are currently in progress.
Finally, it is interesting to consider changes in the nature and
extent of N2 activation that arise with chemical transformations
involving 6 and 8. In this respect, the PbCl2-mediated 8 f 12
process can be considered, on the basis of the relative
magnitudes of d(NN) values (see Table 1), as a formal oxidation
of the µ-N2 group that generates a diazenido linkage of smaller
d(NN) value in 12 and, thereby, preserves two M(IV,d2) metal
centers. A similar formal oxidation of the µ-N2 group in
{Cp′Hf[N(i-Pr)C(Me)N(i-Pr)]}2(µ-η2:η2-N2) (Cp′ ) η5-C5Me4H)
(Ic) occurs upon addition of Br2 to provide {Cp′Hf(Br)[N(i-
Pr)C(Me)N(i-Pr)]}2(µ-η2:η2-N2) (V) that likewise possesses a
significantly smaller d(NN) value consistent with a bridging
diazenido group [cf. for Ic d(NN) ) 1.630(4) Å vs for V d(NN)
) 1.253(3) Å].6
Scheme 11
The extreme limit for N2 activation leading to a fully reduced
µ-hydrazido, [N2]4-, bridge is perhaps best represented by the
collection of side-on-bridged group 4 derivatives presented by
Ia in which d(NN) values of >1.63 Å would appear to challenge
the concept of a N-N bonding interaction.6,26 However, as
previously noted, bond length and bond strength are parameters
that do not usually correlate in a simple fashion, and for Ia, it
is obvious that the M(IV,d0) metal centers that are intrinsic to
the side-on-bridged µ-N2 structure do not have sufficient
electron-reducing equivalents for complete N-N bond scission
to occur. Unfortunately, for dinuclear end-on-bridged [LnM]2(µ-
N2) complexes, particularly those for which the metal centers
have a greater range of formal oxidation states that can be
accessed, the common practice of utilizing the value of the
experimentally determined d(NN) parameter as a measure of
the extent of N2 activation and, from there, as an assessment of
formal metal oxidation state, is not a straightforward task due
to the spectrum of possibilities that exist between the two
limiting resonance structures presented in Scheme 11. Accord-
ingly, reasonable assignments of formal oxidation states for the
group 4, group 5, and group 6 derivatives of I must be formed
with the inclusion and evaluation of additional physiochemical
data, such as open vs closed ground-state configurations, thermal
stability, accessibility of higher (or lower) oxidation states, and
patterns of chemical reactivity. In this fashion, it is reasonable
to conclude on the basis of its diamagnetic character and
relatively long d(NN) value of 1.270(2) Å that the first-row
dititanium µ-η1:η1-N2 derivative 10 can best be described as
having a spin-coupled delocalized Ti-N-N-Ti framework
comprising two Ti(III,d1) that are linked through a reduced
[N2]2- bridge within the resonance structure A of Scheme 11.18
Likewise, we have previously presented the case that the
paramagnetic group 5 derivative Ib is also best described as
containing two Ta(IV,d1) centers linked through a highly reduced
[N2]4- bridge. With this formalization, the metal centers in Ib
still retain sufficient electron-reducing equivalents for final N-N
bond cleavage through a low-energy path leading to II, which
now incorporates two Ta(V,d0) metal centers in the cleaved
product. Moreover, the formal 1,4-addition of H2 to Ib that
provides the ditantalum dihydride IV as the final product can
be viewed as a Ta(IV,d1) f Ta(V,d0) metal-centered oxidation
that leaves the nature of the µ-[N2]4- bridge essentially
unchanged, as evidenced by the corresponding d(NN) values
for these two compounds (see Table 1).
Conclusion
The CpAm supporting ligand environment is proving to be
uniquely versatile for the preparation and experimental inves-
tigation of different families of isostructural dinuclear [LnM]2(µ-
N2) complexes based on I that span the entire set of early
transition metals comprising groups 4-6. While not yet
ubiquitous for all metals, it is reasonable to assume that synthetic
derivatives for M ) V and Nb should be accessible, while those
for M ) Cr may still present a significant challenge. Fortunately,
however, the CpAm ligand combination provides for a diverse
array of different families of derivatives that can present
substituents of varying steric size. Some degree of freedom
might also exist for fine-tuning the electronic features of the
CpAm platform through substituent modification (e.g., more
electron-rich vs more electron-poor). In this respect, the now
documented ability of the CpAm ligand environment to support
the synthesis, isolation, and characterization of [LnM]2(µ-N2)
complexes for metals across the entire early transition metal
series establishes a versatile new tool for systematically explor-
ing and analyzing metal-dependent trends that are essential for
In moving to a consideration of the diamagnetic group 6 metal
derivatives of I represented by compounds 6-8, the question
becomes whether these compounds are best described by the
ground-state electronic resonance structure A that consists of
two M(IV,d2) centers coupled through a bridging [N2]4- group
or, alternatively, the resonance form B that is dominated by
more reduced metal oxidation states in which two formal
(26) It can be noted here that dinitrogen tetroxide, N2O4, has a solid-state
d(NN) value of 1.7562(4) Å that is associated with a N-N bond
strength of only 13.6 kcal mol-1, see: Kvick, Å; McMullan, R. K.;
Newton, M. D. J. Chem. Phys. 1982, 76, 3754–3761.
9
12282 J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010

N2 Complexation and Extent of Nt N Activation
A R T I C L E S
The resulting solution was allowed to warm to room temperature
and stirred overnight. The solution was then pumped down to
dryness, extracted in toluene, and filtered through Celite. The
volume of the solvent was reduced under vacuum, followed by a
second filtration, and then the material was pumped down to
dryness. The residue was washed with pentane (2 × 10 mL) and
then recrystallized from toluene at -30 °C, furnishing a brownish
red crystalline material (0.250 g, 66% yield). For 4: 1H NMR (400
MHz, C6D6) 3.9 (br), 8.1 (br), 11.1 (br), 28.6 (br).
furthering current knowledge regarding dinitrogen activation
leading to N-N bond cleavage and N-atom functionalization.
Experimental Section
General Methods. All manipulations with air- and moisture-
sensitive compounds were carried out under dinitrogen or argon
atmospheres with standard Schlenk or glovebox techniques. All
solvents were dried (Na/benzophenone for pentane, Et2O, toluene,
and THF) and distilled under argon prior to use. Benzene-d6 and
toluene-d8 were dried over Na/K alloy and isolated by vacuum
transfer prior to use. Celite was oven-dried (150 °C for several
days) before use in the glovebox. Cooling and recrystallizations
were performed within the internal freezer of a glovebox. (η5-
C5Me5)MoCl4 and (η5-C5Me5)WCl4 were prepared according to
previously reported procedures.11 1H NMR and 15N NMR were
recorded at 400 and 40.5 MHz respectively, in benzene-d6 (C6D6)
or toluene-d8. 15N NMR chemical shifts are reported relative to
liquid ammonia as an external standard. Elemental analyses were
carried out by Midwest Microlab.
(η5-C5Me5)W[(i-Pr)NC(Me)N(i-Pr)]Cl2 (5). A solution of 2
(0.200 g, 0.353 mmol in 50 mL Et2O) was cooled to -30 °C, at
which point 0.5% NaHg (1.62 g, 0.352 mmol) was added, and the
solution was allowed to warm to room temperature and stirred
overnight. The solution was then pumped down to dryness, extracted
in toluene, and filtered through Celite. Recrystallization from toluene
at -30 °C furnished a brick red crystalline material (0.155 g, 83%
yield). For 5: Anal. Calcd for C18H32Cl2N2W C, 40.69; H, 6.07; N,
5.27; found C, 40.69; H, 6.01; N, 5.21. 1H NMR (400 MHz, C6D6)
0.94 (6H, d, J ) 6.8 Hz, CH(CH3)2), 1.45 (6H, d, J ) 6.8 Hz,
CH(CH3)2), 2.33 (15H, s, C5(CH3)5), 3.98 (2H, sp, J ) 6.8 Hz,
CH(CH3)2), 4.02 (3H, s, NC(CH3)N).
Synthesis and Characterization of New Compounds.
(η5-C5Me5)Mo[N(i-Pr)C(Me)N(i-Pr)]Cl3 (1). A solution of (η5-
C5Me5)MoCl4 (0.626 g, 1.68 mmol) in 100 mL of Et2O was cooled
to -30 °C, whereupon, a solution of Li[N(i-Pr)C(Me)N(i-Pr)]
(0.259 g, 1.75 mmol) in 10 mL of Et2O, precooled to -30 °C, was
added dropwise. After the reaction mixture was allowed to warm
to room temperature, it was stirred at this temperature for 10.5 h
to yield a black-colored solution. The volatiles were removed in
vacuo, and then the residue was taken up in a minimum amount of
toluene and filtered through a small pad of Celite. The resulting
black filtrate was collected, and the solvent was removed once more
in vacuo. The crude product was taken up in 10 mL of toluene, to
which was added 10 mL of pentane, and the solution was cooled
to -30 °C overnight, whereupon 0.284 g of 1 (35% yield) was
obtained in the form of brown-black crystals. Although crystalline
1 appears to noticeably change form and color in the solid state
upon drying under vacuum for extended periods of time (e.g., >1
h), this material can still be used for subsequent synthetic reactions.
For 1: H NMR (400 MHz, benzene-d6, 25 °C) 2.18 (br), 9.05 (br).
(η5-C5Me5)W[(i-Pr)NC(Me)N(i-Pr)]Cl3 (2). A solution of Li[N(i-
Pr)C(Me)N(i-Pr)] (0.225 g, 1.52 mmol in 20 mL of Et2O) was added
dropwise to a solution of (η5-C5Me5)WCl4 (0.648 mg, 1.41 mmol
in 100 mL of Et2O) at -30 °C over a period of 5 min. The resulting
solution was allowed to warm to room temperature and stirred
overnight. The solution was then pumped down to a volume of
∼20 mL and filtered through Celite. The volume of the solvent
was reduced under vacuum, followed by a second filtration, and
then the material was pumped down to dryness. Recrystallization
from toluene at -30 °C furnished a blue crystalline material (0.575
g, 72% yield). For 2: Anal. Calcd for C18H32Cl3N2W C, 38.15; H,
5.69; N, 4.94; found C, 37.19; H, 5.60; N, 4.94.
{(η5-C5Me5)Mo[(i-Pr)NC(Me)N(i-Pr)]}2(µ-η1:η1-N2) (6). i. From
Low-Temperature Chemical Reduction of 3 with KC8. A Schlenk
tube equipped with a Teflon valve was charged with 1 (93 mg,
0.21 mmol) and KC8 (83 mg, 0.61 mmol). The tube was evacuated
and cooled to -196 °C, and then THF (10 mL) was added by
vacuum transfer. The tube was filled with dinitrogen (1 atm), sealed,
and warmed to -78 °C. After stirring at -78 °C for 3 h, the
suspension was warmed to 0 °C and stirred for 15 h, followed by
warming to room temperature for 5 h. The resulting dark brown/
yellow suspension was filtered through Celite, and the solvent was
removed under vacuum. The resulting solid was dissolved in a small
amount of pentane, filtered, and cooled at -30 °C to afford yellow/
brown crystalline material (32 mg, 40%). For 6: Anal. Calcd for
C36H64N6Mo2 C, 55.95; H, 8.35; N, 10.87; found C, 56.16; H, 8.33;
1
N, 10.85. H NMR (400 MHz, C6D6) 0.88 (12H, d, J ) 6.4 Hz),
1.13 (12H, d, J ) 6.4 Hz), 1.87 (30H, s), 1.89 (6H, s), 3.45 (4H,
sp, J ) 6.4 Hz). Note: due to poor crystallinity and high solubility
in pentane, it was not possible to obtain single crystals of 6 that
were suitable for X-ray crystallography after several attempts.
Isotopically labeled 15N2-6 was prepared in the same manner as
described above under an atmosphere of 15N2 (99%). The 1H NMR
spectrum was identical to that of unlabeled 6. For 15N2-6: 15N NMR
(40.5 MHz, benzene-d6, liq. NH3 as an external reference) δ 374.9
ppm.
ii. From Chemical Reduction of 1 Using 0.5% NaHg. A
solution of 1 (309 mg, 0.700 mmol) in 40 mL of THF was cooled
to -30 °C, at which point 0.5% (w/w) of NaHg (13.169 g, 2.864
mmol) was added, and the solution was allowed to warm to room
temperature. The reaction mixture was stirred at room temperature
for 16 h to yield a yellow-brown colored solution, after which time
the volatiles were removed in vacuo and the solid residue was taken
up in pentane and filtered through Celite. The filtrate was then
concentrated and cooled to -30 °C whereupon 208 mg (77% yield)
of 6 was isolated as a yellowish-brown crystalline material.
iii. From Thermolysis of meso-11. A NMR tube equipped with
a Teflon valve was charged with a solution of meso-11 (36 mg,
0.046 mmol) in benzene-d6 (1.5 mL). The tube was then heated to
70 °C over a period of 1 week, with the reaction progress being
monitored periodically. After this period, the solution was drained
from the tube, the tube was rinsed with pentane, and the combined
solutions were pumped down to dryness in vacuo. The resulting
residue was taken up in a minimal amount of pentane and cooled
to -30 °C, which furnished yellow/brown crystals of 6 (19 mg,
51% yield).
(η5-C5Me5)Mo[(i-Pr)NC(Me)N(i-Pr)]Cl2 (3). A solution of
Li[N(i-Pr)C(Me)N(i-Pr)] (0.836 g, 5.64 mmol in 25 mL of Et2O)
was added dropwise to a solution of (η5-C5Me5)MoCl4 (1.00 g, 2.68
mmol in 175 mL of Et2O) at -30 °C over a period of 5 min. The
resulting solution was allowed to warm to room temperature and
stirred overnight. The solution was then pumped down to dryness,
extracted in toluene, and filtered through Celite. The volume of
the solvent was reduced under vacuum, followed by a second
filtration, and then the material was pumped down to dryness. The
residue was washed with pentane (2 × 10 mL) and then recrystal-
lized from toluene at -30 °C, furnishing a brownish crystalline
material (0. 908 g, 76% yield). For 3: Anal. Calcd for
C18H32Cl2N2Mo C, 48.77; H, 7.28; N, 6.32; found C, 48.86; H,
1
7.38; N, 6.60. H NMR (400 MHz, C6D6) 2.1 (br), 8.9 (br), 15.8
(br).
(η5-C5Me5)Mo[(i-Pr)NC(NMe2)N(i-Pr)]Cl2 (4). A solution of
Li[N(i-Pr)C(NMe2)N(i-Pr)] (0.400 g, 1.60 mmol in 15 mL Et2O)
was added dropwise to a solution of (η5-C5Me5)MoCl4 (0.300 g,
0.804 mmol in 75 mL Et2O) at -30 °C over a period of 5 min.
{(η5-C5Me5)Mo[N(i-Pr)C(NMe2)N(i-Pr)])2(µ-η1:η1-N2) (7). A
Schlenk tube equipped with a Teflon valve was charged with 4
(99 mg, 0.21 mmol) and KC8 (89 mg, 0.66 mmol). The tube was
evacuated and cooled to -196 °C, and then THF (10 mL) was
9
J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010 12283

A R T I C L E S
Fontaine et al.
added by vacuum transfer. The tube was filled with dinitrogen (1
atm), sealed, and warmed to -78 °C. After stirring at -78 °C for
5 h, the suspension was warmed to 0 °C and stirred for 15 h,
followed by warming to room temperature for 5 h. The resulting
dark brown/yellow suspension was filtered through Celite, and the
solvent was removed under vacuum. The resulting solid was
dissolved in a small amount of pentane, filtered, and cooled to -30
°C to afford yellow/brown crystalline material (33 mg, 38% yield).
For 7: Anal. Calcd for C36H64N6Mo2 C, 54.93; H, 8.49; N, 13.49;
green crystalline material (43 mg, 49% yield). For 10: Anal. Calcd
for C36H64N6Ti2 C, 63.89; H, 9.53; N, 12.42; found C, 63.40; H,
1
9.36; N, 11.97. H NMR (400 MHz, C6D6) 1.01 (12H, d, J ) 6.4
Hz, CH(CH3)2), 1.05 (12H, d, J ) 6.4 Hz, CH(CH3)2), 2.06 (30H,
s, C5(CH3)5), 2.09 (6H, s, NC(CH3)N), 3.97 (4H, sp, J ) 6.4 Hz,
CH(CH3)2).
{(η5- C5Me5)Mo(H)[(i-Pr)NC(Me)N(i-Pr)]}2(µ-η1:η1-N2) (meso-
11). A solution of 1 (110 mg, 0.248 mmol) in Et2O (25 mL) in a
screw-cap vial was cooled to -30 °C, and then nBuLi (0.30 mL,
1.70 M in hexanes, 0.51 mmol) was added dropwise via syringe.
The resulting solution was capped and allowed to warm to room
temperature while stirring overnight. The volatiles were removed
under vacuum, and the residue was extracted in pentane and filtered
through Celite. The volume of the solution was reduced under
vacuum, a second filtration was performed, and the resulting
solution was cooled to -30 °C to provide brown crystalline material
(63 mg, 66% yield). For 11: Anal. Calcd for C36H66N6Mo2 C, 55.95;
1
found C, 55.01; H, 8.45; N, 13.29. H NMR (400 MHz, C6D6)
0.85(12H, d, J ) 6.4 Hz, CH(CH3)2), 1.18 (12H, d, J ) 6.4 Hz,
CH(CH3)2), 1.90 (30H, s, C5(CH3)5), 2.67 (12H, s, N(CH3)2), 3.67
(4H, sp, J ) 6.4 Hz CH(CH3)2).
{(η5-C5Me5)W[(i-Pr)NC(Me)N(i-Pr)]}2(µ-η1:η1-N2) (8). i. From
Low-Temperature Chemical Reduction of 2 with KC8. A Schlenk
tube equipped with a Teflon valve was charged with 2 (100 mg,
0.176 mmol) and KC8 (95 mg, 0.703 mmol). The tube was
evacuated and cooled to -196 °C, and then THF (10 mL) was
added by vacuum transfer. The tube was filled with dinitrogen (1
atm), sealed, and warmed to -78 °C. After stirring at -78 °C for
3 h, the suspension was warmed to 0 °C and stirred for 15 h,
followed by warming to room temperature for 5 h. The resulting
dark green suspension was filtered through Celite, and the solvent
was removed under vacuum. The resulting solid was dissolved in
a small amount of pentane, filtered, and cooled to -30 °C to afford
green crystalline material (47 mg, 56% yield). For 8: Anal. Calcd
for C36H64N6W2 C, 45.58; H, 6.80; N, 8.86; found C, 45.74; H,
1
H, 8.35; N, 10.87; found C, 55.65; H, 8.36; N, 10.57. H NMR
(400 MHz, C6D6) -1.51 (2H, s), 1.06 (6H, d, J ) 6.4 Hz), 1.09
(6H, d, J ) 6.4 Hz), 1.20 (6H, d, J ) 6.4 Hz), 1.28 (6H, d, J ) 6.4
Hz), 1.60 (6H, s), 1.99 (30H, s), 3.50 (4H, m).
{(η5-C5Me5)W(Cl)[(i-Pr)NC(Me)N(i-Pr)]}2(µ-η1:η1-N2) (rac,meso-
12). A Schelnk tube equipped with a Teflon valve was charged
with 8 (45 mg, 0.047 mmol) and PbCl2 (13 mg, 0.047 mmol). The
tube was evacuated and cooled to -196 °C, and then Et2O (5 mL)
was added by vacuum transfer. The tube was warmed to room
temperature and stirred for 18 h, at which point it was pumped
down to dryness, then extracted with pentane and filtered through
Celite. The resulting yellow/brown solution was reduced in volume,
refiltered, and cooled to -30 °C to afford dark yellow crystalline
material (40 mg, 83% yield), which was seen to be a 1:1 mixture
of two diastereomers that were not readily separable by fractional
crystallization. For 12: Anal. Calcd for C36H64N6Cl2W2 C, 42.41;
H, 6.33; N, 8.24; found C, 42.60; H, 6.26; N, 8.26. 1H NMR (400
MHz, C6D6) 1.11 (6H, d, J ) 6.8 Hz, CH(CH3)2), 1.21 (6H, d, J )
6.8 Hz, CH(CH3)2), 1.24 (6H, d, J ) 6.8 Hz, CH(CH3)2), 1.25 (6H,
d, J ) 6.8 Hz, CH(CH3)2), 1.27 (6H, d, J ) 6.8 Hz, CH(CH3)2),
1.32 (6H, d, J ) 6.8 Hz, CH(CH3)2), 1.34 (6H, d, J ) 6.8 Hz,
CH(CH3)2), 1.61 (6H, d, J ) 6.8 Hz, CH(CH3)2), 1.66 (6H, s,
NC(CH3)N), 1.80 (6H, s, NC(CH3)N), 2.05 (30H, s, C5(CH3)5), 2.10
(30H, s, C5(CH3)5), 3.71 (2H, sp, J ) 6.8 Hz, CH(CH3)2), 3.82
(2H, sp, J ) 6.8 Hz, CH(CH3)2), 4.07 (2H, sp, J ) 6.8 Hz,
CH(CH3)2), 4.18 (2H, sp, J ) 6.8 Hz, CH(CH3)2).
1
6.81; N, 8.76. H NMR (400 MHz, C6D6) 0.98 (12H, d, J ) 6.4
Hz, CH(CH3)2), 1.24 (12H, d, J ) 6.4 Hz, CH(CH3)2), 1.71 (6H,
s, NC(CH3)N), 2.14 (30H, s, C5(CH3)5), 3.26 (4H, sp, J ) 6.4 Hz,
CH(CH3)2).
The isotopically labeled compound 15N2-8 was prepared in the
same manner as described above under an atmosphere of 15N2 gas.
1
The H NMR spectrum was identical to that of unlabeled 8. For
15N2-8: 15N NMR (40.5 MHz, C6D6) 351.4 [1J(183W-15N) ) 145
2
Hz, J(183W-15N) ) 10.5 Hz).
ii. From Chemical Reduction of 2 Using 0.5% NaHg. A
solution of 2 (0.307 g, 0.54 mmol) in 30 mL of THF was cooled
to -30 °C, at which point 0.5% (w/w) Na/Hg (9.964 g, 2.167 mmol)
was added and the solution was allowed to warm to room
temperature. The reaction mixture was stirred at room temperature
for 1.5 h to yield a dark green colored solution, after which time
the volatiles were removed in vacuo and the solid residue was taken
up in pentane and filtered through Celite. The resulting dark green
filtrate was then concentrated and cooled to -30 °C whereupon
257 mg (92% yield) of 8 was isolated as a dark green crystalline
material.
(η5-C5Me5)Mo[N(i-Pr)C(Me)N(i-Pr)](CNAr)2 (Ar)3,5-Me2-
C6H3) (13). A flask was charged with 6 (32 mg, 0.041 mmol), 2,6-
dimethylphenylisocyanide (20 mg, 0.16 mmol), and toluene (2 mL),
and the mixture was stirred at room temperature for 18 h, over
which time the solution turned a deep green. The solution was
pumped down to dryness under vacuum, taken up in pentane, and
filtered through Celite. The pentane solution was cooled to -30
°C to afford green crystalline material (38 mg, 93% yield). For 13:
Anal. Calcd for C36H50N4Mo C, 68.12; H, 7.94; N, 8.83; found C,
67.61; H, 7.68; N, 8.39. 1H NMR (400 MHz, C6D6) 1.05 (6H, d, J
) 6.4 Hz), 1.19 (6H, d, J ) 6.4 Hz), 1.58 (3H, s), 1.88 (15H, s),
2.43 (12H, s), 3.61 (4H, sp, J ) 6.4 Hz), 6.746 (2H, t, J ) 7.6
Hz), 6.89 (4H, d, J ) 7.6 Hz). 13C NMR (125.8 MHz, C6D6) 11.9,
14.7, 20.3, 23.2, 25.9, 49.9, 105.2, 124.3, 132.2, 134.8, 167.1, 247.5.
IR (pentane) νCN ) 1966 cm-1(s), 1971 cm-1(s).
(η5-C5Me5)TiCl2[N(i-Pr)C(Me)N(i-Pr)] (9). To a solution of
(η5-C5Me5)TiCl3 (0.791 g, 2.73 mmol) in Et2O (75 mL) was added
Li[N(i-Pr)C(Me)N(i-Pr)] (0.405 g, 2.73 mmol) at -35 °C. The
reaction was warmed to ambient temperature and stirred for 16 h,
during which time the solution changed in color from orange to
burgundy. The solvent was removed in vacuo, and the resulting
product was filtered through Celite and recrystallized from toluene
(0.939 g, 2.37 mmol, 87% yield). For 9: 1H NMR (400 MHz, C6D6)
δ 3.92 (2H, m, CH(CH3)2), 2.05 (15H, s, C5Me5), 1.57 (s, 3H, CH3),
1.14 (12H, d, CH(CH3)2). Anal, Calcd. for TiC18H32N2Cl2 C, 54.70;
H, 8.16; N, 7.09; found C, 54.74; H, 8.00; N, 7.16.
(η5-C5Me5)W[N(i-Pr)C(Me)N(i-Pr)](CNAr)2 (Ar ) 3,5-Me2C6H3)
(14). A flask was charged with 8 (25 mg, 0.026 mmol), 2,6-
dimethylphenylisocyanide (13 mg, 0.11 mmol), and toluene (2 mL),
and the mixture was stirred at room temperature for 18 h, over
which time the solution turned a darker green. The solution was
pumped down to dryness under vacuum, taken up in pentane, and
filtered through Celite. The pentane solution was cooled to -30
°C to afford green crystalline material (27 mg, 73% yield). For 14:
Anal. Calcd for C36H50N4W C, 59.83; H, 6.97; N, 7.75; found C,
59.37; H, 7.07; N, 7.56. 1H NMR (400 MHz, C6D6) 1.04 (6H, d, J
) 6.4 Hz), 1.14 (6H, d, J ) 6.4 Hz), 1.51 (3H, s), 1.92 (15H, s),
{(η5-C5Me5)Ti[N(i-Pr)C(Me)N(i-Pr)]}2(µ-η1:η1-N2) (10). A
Schlenk tube equipped with a Teflon valve was charged with 9
(102 mg, 0.258 mmol) and KC8 (104 mg, 0.769 mmol). The tube
was evacuated and cooled to -196 °C, and then THF (10 mL)
was added by vacuum transfer. The tube was filled with dinitrogen
(1 atm), sealed, and warmed to -78 °C. After stirring at -78 °C
for 3 h, the suspension was warmed to 0 °C and stirred for 15 h,
followed by warming to room temperature for 5 h. The resulting
dark green suspension was filtered through Celite, and the solvent
was removed under vacuum. The resulting solid was dissolved in
a small amount of pentane, filtered, and cooled to -30 °C to afford
9
12284 J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010

N2 Complexation and Extent of Nt N Activation
A R T I C L E S
2.50 (12H, s), 3.52 (4H, sp, J ) 6.4 Hz), 6.74 (2H, t, J ) 7.6 Hz),
6.96 (4H, d, J ) 7.6 Hz). 13C NMR (125.8 MHz, C6D6) 12.0, 14.7,
20.7, 25.3, 25.8, 50.0, 103.5, 124.4, 128.8, 135.5, 137.3, 168.0,
244.3. IR (pentane) νCN ) 1940 cm-1(s), 1952 cm-1(s).
dark brown filtrate was removed in vacuo to provide a brown solid
that was recrystallized from pentane containing a few drops of
toluene at -30 °C to provide 16 as orange-red crystals (39 mg,
59% yield). For 16: Anal. Calcd for C20H32N2O2W C, 46.34; H,
(η5-C5Me5)Mo[N(i-Pr)C(Me)N(i-Pr)](CO)2 (15). Within a
Schlenk tube, a solution of 6 (0.142 g, 0.184 mmol) in 6 mL of
toluene was charged with 5 psi of carbon monoxide (99.95%), and
the contents of the storage tube were stirred for 40 h at room
temperature. The volatiles from the resulting dark red solution were
then removed in vacuo, and the residue was taken up in pentane
and filtered through a pad of Kimwipe placed within a glass pipet.
The solvent of the dark red filtrate was removed in vacuo to provide
a dark red solid that was recrystallized from pentane at -30 °C to
provide 15 (0.112 g, 71% yield). For 15: Anal. Calcd for
C20H32N2O2Mo C, 56.05; H, 7.53; N, 6.54; found C, 56.57; H, 7.38;
N, 6.54. 1H NMR (400 MHz, C6D6) 0.99 (6H, d, J ) 6.3 Hz), 1.03
(6H, d, J ) 6.3 Hz), 1.42 (3H, s), 1.74 (15H, s), 3.48 (2H, sp, J )
1
6.22; N, 5.40; found C, 46.19; H, 6.11; N, 5.37. H NMR (400
MHz, C6D6) 0.99 (6H, d, J ) 6.4 Hz), 1.01 (6H, d, J ) 6.4 Hz),
1.34 (3H, s), 1.81 (15H, s), 3.35 (2H, sp, J ) 6.4 Hz). IR (KBr)
νCO ) 1892 and 1788 cm-1
.
Acknowledgment. Funding for this work was provided by the
Department of Energy, Basic Energy Sciences (Grant DE-
SC0002217), and in part, by the NSF (Grant CHE-0848293), for
which we are grateful.
Supporting Information Available: Details of single-crystal
X-ray analyses, including tables of bond lengths and angles,
and anisotropic displacement parameters for the solid-state
structures, and complete X-ray crystallographic data (CIF) for
compounds 3, 4, 7, 8, 10, meso-11, rac-12, meso-12, 13, 14,
15, and 16. This material is available free of charge via the
Internet at http://pubs.acs.org.
6.3 Hz). IR (KBr) νCO ) 1909 and 1806 cm-1
.
(η5-C5Me5)W[N(i-Pr)C(Me)N(i-Pr)](CO)2 (16). Within a Schlenk
tube, a solution of 8 (60 mg, 63 µmol) in 4 mL of toluene was
charged with 5 psi of carbon monoxide (99.95%), and the contents
of the storage tube were stirred for 40 h at room temperature. The
volatiles from the resulting dark brown solution were then removed
in vacuo, and the residue taken up in pentane and filtered through
a pad of Kimwipe placed within a glass pipet. The solvent of the
JA100469F
9
J. AM. CHEM. SOC. VOL. 132, NO. 35, 2010 12285