Shining light on dinitrogen cleavage: Structural features, redox chemistry, and photochemistry of the key intermediate bridging dinitrogen complex

Article abstract of DOI:10.1021/ja8002638

The key intermediate in dinitrogen cleavage by Mo(N[t-Bu]Ar)₃, 1 (Ar = 3,5-C₆H₃Me₂), has been characterized by a pair of single crystal X-ray structures. For the first time, the X-ray crystal structure of (μ-N₂)[Mo(N[t-Bu]Ar)₃]₂, 2, and the product of homolytic fragmentation of the N-N bond, N≡Mo(N[t-Bu] Ar)₃, are reported. The structural features of 2 are compared with previously reported EXAFS data. Moreover, contrasts are drawn between theoretical predictions concerning the structural and magnetic properties of 2 and those reported herein. In particular, it is shown that 2 exists as a triplet (S = 1) at 20°C. Further insight into the bonding across the MoNNMo core of the molecule is obtained by the synthesis and structural characterization of the one- and two-electron oxidized congeners, (μ-N₂)[Mo(N[t-Bu]Ar) ₃]₂[B(Ar^F)₄], 2[B(Ar ^F)₄] (Ar^F = 3,5-C₆H ₃(CF₃)₂) and (μ-N₂)[Mo(N[t-Bu]Ar) ₃]₂[B(Ar^F)₄]₂, 2[B(Ar^F)₄]₂, respectively. Bonding in these three molecules is discussed in view of X-ray crystallography, Raman spectroscopy, electronic absorption spectroscopy, and density functional theory. Combining X-ray crystallography data with Raman spectroscopy studies allows the N-N bond polarization energy and N-N internuclear distance to be correlated in three states of charge across the MoNNMo core. For 2[B(Ar^F) ₄], bonding is symmetric about the μ-N₂ ligand and the N-N polarization is Raman active; therefore, 2[B(Ar^F)₄] meets the criteria of a Robin-Day class III mixed-valent compound. The redox couples that interrelate 2, 2⁺, and 2²⁺ are studied by cyclic voltammetry and spectroelectrochemistry. Insights into the electronic structure of 2 led to the discovery of a photochemical reaction that forms N≡Mo(N[t-Bu]Ar)₃ and Mo(N[t-Bu]Ar)₃ through competing N-N bond cleavage and N₂ extrusion reaction pathways. The primary quantum yield was determined to be Φ_p = 0.05, and transient absorption experiments show that the photochemical reaction is complete in less than 10 ns.

Full text of DOI:10.1021/ja8002638

Published on Web 06/25/2008
Shining Light on Dinitrogen Cleavage: Structural Features,
Redox Chemistry, and Photochemistry of the Key
Intermediate Bridging Dinitrogen Complex
John J. Curley, Timothy R. Cook, Steven Y. Reece, Peter Mu¨ller, and
Christopher C. Cummins*
Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts AVenue,
Room 6-325, Cambridge, Massachusetts 02139
Received January 11, 2008; E-mail: ccummins@mit.edu
Abstract: The key intermediate in dinitrogen cleavage by Mo(N[t-Bu]Ar)3, 1 (Ar ) 3,5-C6H3Me2), has been
characterized by a pair of single crystal X-ray structures. For the first time, the X-ray crystal structure of
(µ-N2)[Mo(N[t-Bu]Ar)3]2, 2, and the product of homolytic fragmentation of the NsN bond, NtMo(N[t-Bu]Ar)3,
are reported. The structural features of 2 are compared with previously reported EXAFS data. Moreover,
contrasts are drawn between theoretical predictions concerning the structural and magnetic properties of
2 and those reported herein. In particular, it is shown that 2 exists as a triplet (S ) 1) at 20 °C. Further
insight into the bonding across the MoNNMo core of the molecule is obtained by the synthesis and structural
characterization of the one- and two-electron oxidized congeners, (µ-N2)[Mo(N[t-Bu]Ar)3]2[B(ArF)4], 2[B(ArF)4]
(ArF ) 3,5-C6H3(CF3)2) and (µ-N2)[Mo(N[t-Bu]Ar)3]2[B(ArF)4]2, 2[B(ArF)4]2, respectively. Bonding in these
three molecules is discussed in view of X-ray crystallography, Raman spectroscopy, electronic absorption
spectroscopy, and density functional theory. Combining X-ray crystallography data with Raman spectroscopy
studies allows the NsN bond polarization energy and NsN internuclear distance to be correlated in three
states of charge across the MoNNMo core. For 2[B(ArF)4], bonding is symmetric about the µ-N2 ligand and
the NsN polarization is Raman active; therefore, 2[B(ArF)4] meets the criteria of a Robin-Day class III
mixed-valent compound. The redox couples that interrelate 2, 2+, and 22+ are studied by cyclic voltammetry
and spectroelectrochemistry. Insights into the electronic structure of 2 led to the discovery of a photochemical
reaction that forms NtMo(N[t-Bu]Ar)3 and Mo(N[t-Bu]Ar)3 through competing NsN bond cleavage and N2
extrusion reaction pathways. The primary quantum yield was determined to be Φp ) 0.05, and transient
absorption experiments show that the photochemical reaction is complete in less than 10 ns.
and structurally characterized,7,12–25 these complexes typically
require harsh conditions to promote reactions that productively
Introduction
use the N2 fragment.15 This situation is unfortunate because
linear µ-N2 complexes store N2 in a reduced form and therefore
should be susceptible to productive chemistry of the N2
fragment.1,2 In contrast to linear µ-N2 complexes, bimetallic
complexes that contain a side-bound µ-N2 ligand show enhanced
An outstanding goal in synthetic chemistry is the use of N2
as a synthon for nitrogen-containing functional groups in organic
molecules.1–5 In this regard, several procedures have been
developed to transfer a nitrogen atom from an N2-derived metal
nitride to an organic functional group.6–9 Limiting the develop-
ment of new nitrogen transfer reactions is the small number of
well-defined, transition-metal reagents that bind N2, form a µ-N2
complex, and subsequently cleave the NsN bond to form 2
equiv of a terminal nitride.10,11 While it is true that a large
number of linear, bimetallic µ-N2 complexes have been isolated
(10) 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.
(11) Laplaza, C. E.; Cummins, C. C. Science 1995, 268, 861–863.
(12) Manriquez, J. M.; Bercaw, J. E. J. Am. Chem. Soc. 1974, 96, 6229–
6230.
(13) Roddick, D. M.; Fryzuk, M. D.; Seidler, P. F.; Hillhouse, G. L.;
Bercaw, J. E. Organometallics 1985, 4, 97–104.
(14) Sanner, R. D.; Duggan, D. M.; McKenzie, T. C.; Marsh, R. E.;
Bercaw, J. E. J. Am. Chem. Soc. 1976, 98, 8358–8365.
(15) Schrock, R. R. Acc. Chem. Res. 2005, 38, 955–962.
(16) Peters, J. C.; Cherry, J.-P. F.; Thomas, J. C.; Baraldo, L.; Mindiola,
D. J.; Davis, W. M.; Cummins, C. C. J. Am. Chem. Soc. 1999, 121,
10053–10067.
(1) Chatt, J. Philos. Trans. R. Soc. London, Ser. B. 1977, 281, 243–248.
(2) MacKay, B. A.; Fryzuk, M. D. Chem. ReV. 2004, 104, 385–401.
(3) Hidai, M.; Mizobe, Y. Can. J. Chem. 2005, 83, 358–374.
(4) Hori, K.; Mori, M. J. Am. Chem. Soc. 1998, 120, 7651–7652.
(5) Hori, M.; Mori, M. J. Org. Chem. 1995, 60, 1480–1481.
(6) Curley, J. J.; Sceats, E. L.; Cummins, C. C. J. Am. Chem. Soc. 2006,
128, 14036–14037.
(7) Figueroa, J. S.; Piro, N. A.; Clough, C. R.; Cummins, C. C. J. Am.
Chem. Soc. 2006, 128, 940–950.
(17) Mindiola, D. J.; Meyer, K.; Cherry, J.-P. F.; Baker, T. A.; Cummins,
C. C. Organometallics 2000, 19, 1622–1624.
(8) Watanabe, D.; Gondo, S.; Seino, H.; Mizobe, Y. Organometallics
2007, 26, 4909–4920.
(18) Chatt, J.; Dilworth, J. R.; Richards, R. L. Chem. ReV. 1978, 78, 589–
625.
(9) Henderickx, H.; Kwakkenbos, G.; Peters, A.; van der Spoel, J.; de
Vries, K. Chem. Commun. 2003, 2050–2051.
(19) Vidyaratne, I.; Scott, J.; Gambarotta, S.; Budzelaar, P. H. M. Inorg.
Chem. 2007, 46, 7040–7049.
9
9394 J. AM. CHEM. SOC. 2008, 130, 9394–9405
10.1021/ja8002638 CCC: $40.75
2008 American Chemical Society

Features and Chemistry of Dinitrogen Cleavage
A R T I C L E S
Scheme 1
reactivity,26 and engage in productive reactions to liberate
NH3,27 or to form Si-N28,29 or CsN bonds.30–32 Additionally,
a tantalum compound featuring a side-on, end-on coordination
mode of the µ-N2 ligand participates in hydrosilylation and
hydroborylation reactions, making either a SisN33,34 or a
BsN35,36 bond while cleaving the NsN bond.
One potential though rarely investigated method to initiate
reactions of µ-N2 complexes is photochemical activation of the
NsN bond.37 To our knowledge only one experimentalist has
adopted this strategy; Floriani reported an example of photo-
chemical NsN bond cleavage involving the thermally stable
compound (µ-N2)[Mo(Mes)3]2 (Mes ) 2,4,6-Me3C6H2).38 Pho-
tolysis of this compound yields 1 equiv of (µ-N)[Mo(Mes)3]2
while liberating 0.5 equiv of N2. Although the photochemistry
of transition metal µ-N2 complexes is sparsely explored,
productive photochemistry of N2 has been developed. For
example, under photochemical conditions, either TiO2 doped
39–41
with Fe2O3
or C60 encapsulated in two γ-cyclodextrin
molecules42 are reported to catalyze the conversion of N2 to
NH3.
Three-coordinate molybdenum trisanilide, Mo(N[t-Bu]Ar)3,
1, has been important in defining thermal 6e reduction chemistry
of N2 by cleavage of the NsN triple bond.7,10,11,17 Dinitrogen
is bound by 2 equiv of Mo(N[t-Bu]Ar)3 to form the thermally
unstable intermediate, (µ-N2)[Mo(N[t-Bu]Ar)3]2, 2, which un-
dergoes homolytic fragmentation of the NsN bond to form 2
equiv of NtMo(N[t-Bu]Ar)3, 3 (Scheme 1). Although several
well-defined metal reagents will cleave the NsN bond of N2
to produce either terminal or bridging nitride products,10,11,38,43–47
the case for 1 remains exceptional in that the metal fragment
that binds N2, 1, and the dinitrogen complex intermediate, 2,
en route to nitride formation can be isolated and studied. These
criteria make 2 an ideal candidate for investigating homolytic
fragmentation of the NsN bond uncomplicated by other
processes. As a result, this reaction sequence has been the topic
of numerous mechanistic10,16,48 and theoretical investigations.49–56
(20) Smith, J. M.; Sadique, A. R.; Cundari, T. R.; Rodgers, K. R.; Lukat-
Rodgers, G.; Lachicotte, R. J.; Flaschenriem, C. J.; Vela, J.; Holland,
P. L. J. Am. Chem. Soc. 2006, 128, 756–769.
(21) Chomitz, W. A.; Arnold, J. Chem. Commun. 2007, 4797–4799.
(22) Kilgore, U. J.; Yang, X.; Tomaszewski, J.; Huffman, J. C.; Mindiola,
D. J. Inorg. Chem. 2006, 45, 10712–10721.
(23) Jolly, P. W.; Jonas, K.; Kru¨ger, C.; Tsay, Y.-H. J. Organomet. Chem.
1971, 33, 109–122.
(24) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 6252–
6254.
(39) Schrauzer, G. N.; Guth, T. D. J. Am. Chem. Soc. 1977, 99, 7189–
7193.
(25) Vidyaratne, I.; Gambarotta, S.; Korobkov, I.; Budzelaar, P. H. M.
Inorg. Chem. 2005, 44, 1187–1189.
(40) Rusina, O.; Eremenko, A.; Frank, G.; Strunk, H. P.; Kisch, H. Angew.
Chem., Int. Ed. 2001, 40, 3993–3995.
(26) MacLachlan, E. A.; Fryzuk, M. D. Organometallics 2006, 25, 1530–
1543.
(41) Rusina, O.; Linnik, O.; Eremenko, A.; Kisch, H. Chem.sEur. J. 2003,
9, 561–565.
(27) Pool, J. A.; Lobkovsky, E.; Chirik, P. J. Nature 2004, 427, 527–
530.
(42) Nishibayashi, Y.; Saito, M.; Uemura, S.; Takekuma, S.; Takekuma,
H.; Yoshida, Z. Nature 2004, 428, 279–280.
(28) Basch, H.; Musaev, D. G.; Morokuma, K.; Fryzuk, M. D.; Love,
J. B.; Seidel, W. W.; Albinati, A.; Koetzle, T. F.; Klooster, W. T.;
Mason, S. A.; Eckert, J. J. Am. Chem. Soc. 1999, 121, 523–528.
(29) Fryzuk, M. D.; Love, J. B.; Rettig, S. J.; Young, V. G. Science 1997,
275, 1445–1447.
(43) Clentsmith, G. K. B.; Bates, V. M. E.; Hitchcock, P. B.; Cloke,
F. G. N. J. Am. Chem. Soc. 1999, 121, 10444–10445.
(44) Kawaguchi, H.; Matsuo, T. Angew. Chem., Int. Ed. 2002, 41, 2792–
2794.
(45) Fryzuk, M. D.; Kozak, C. M.; Bowdridge, M. R.; Patrick, B. O.;
Rettig, S. J. J. Am. Chem. Soc. 2002, 124, 8389–8397.
(46) Berno, P.; Gambarotta, S. Organometallics 1995, 14, 2159–2161.
(47) Vidyaratne, I.; Crewdson, P.; Lefebvre, E.; Gambarotta, S. Inorg.
Chem. 2007, 46, 8836–8842.
(30) Bernskoetter, W. H.; Lobkovsky, E.; Chirik, P. J. Angew. Chem.,
Int. Ed. 2007, 46, 2858–2861.
(31) Morello, L.; Love, J. B.; Patrick, B. O.; Fryzuk, M. D. J. Am. Chem.
Soc. 2004, 126, 9480–9481.
(32) Bernskoetter, W. H.; Olmos, A. V.; Pool, J. A.; Lobkovsky, E.; Chirik,
P. J. J. Am. Chem. Soc. 2006, 128, 10696–10697.
(33) Fryzuk, M. D.; MacKay, B. A.; Patrick, B. O. J. Am. Chem. Soc.
2003, 125, 3234–3235.
(48) Tsai, Y.-C.; Cummins, C. C. Inorg. Chim. Acta 2003, 345, 63–69.
(49) Cui, Q.; Musaev, D. G.; Svensson, M.; Sieber, S.; Morokuma, K.
J. Am. Chem. Soc. 1995, 117, 12366–12367.
(50) Christian, G.; Stranger, R.; Yates, B. F.; Graham, D. C. Dalton Trans.
2005, 962–968.
(34) MacKay, B. A.; Munha, R. F.; Fryzuk, M. D. J. Am. Chem. Soc.
2006, 128, 9472–9483.
(51) Christian, G.; Driver, J.; Stranger, R. Faraday Discuss. 2003, 124,
331–341.
(35) Fryzuk, M. D.; MacKay, B. A.; Johnson, S. A.; Patrick, B. O. Angew.
Chem., Int. Ed. 2002, 41, 3709–3712.
(52) 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.
(53) Christian, G.; Stranger, R. Dalton Trans. 2004, 2492–2495.
(54) Neyman, K. M.; Nasluzov, V. A.; Hahn, J.; Landis, C. R.; Rosch,
N. Organometallics 1997, 16, 995–1000.
(36) MacKay, B. A.; Johnson, S. A.; Patrick, B. O.; Fryzuk, M. D. Can.
J. Chem. 2005, 83, 315–323.
(37) Reiher, M.; Kirchner, B.; Dieter, J. H.; Bernd, S.; Hess, A.
Chem.sEur. J. 2004, 10, 4443–4453.
(38) Solari, E.; Da Silva, C.; Iacono, B.; Hesschenbrouck, J.; Rizzoli, C.;
Scopelliti, R.; Floriani, C. Angew. Chem., Int. Ed. 2001, 40, 3907–
3909.
(55) Hahn, J.; Landis, C. R.; Nasluzov, V. A.; Neyman, K. M.; Rosch,
N. Inorg. Chem. 1997, 36, 3947–3951.
(56) Marcus, C. D. Chem.sEur. J. 2007, 13, 3406–3413.
9
J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008 9395

A R T I C L E S
Curley et al.
In particular, the molecular structure and dynamics of 2 have
been the subject of much speculation.49–56
In this work, two X-ray crystal structures of 2 have been
solved to address detailed structural aspects of this key
intermediate in thermal N2 cleavage. A new protocol was
developed for isolating 2 as a pure material, and with 2 in hand
its reaction chemistry has been studied. Oxidation of 2 by either
1e or 2e affords isolable products, both of which have been
structurally characterized. Comparison between these oxidation
products and 2 offers insight into bonding within the MoNNMo
core. Furthermore, a photochemical reaction of 2 is reported in
which competition between NsN bond cleavage and N2
extrusion forms both 1 and 3.
Figure 1. The molecular structure of 2 is shown with thermal displacement
ellipsoids at the 50% probability level. Hydrogen atoms have been omitted
for clarity. Crystal parameters are P2/n: a ) 19.3 Å, b ) 11.5 Å, c ) 21.1
Å, ꢀ ) 99.2°, and a crystallographically imposed two-fold axis relates the
two halves of the molecule. A chemically equivalent but crystallographically
distinct structure is described in detail in the Supporting Information.
Results and Discussion
Structure and Magnetism of 2. We have shown in a prior
EXAFS (extended X-ray absorption fine structure) study that 2
adopts a conformation wherein the six tert-butyl groups are
directed inward, toward a linear MoNNMo core of the molecule,
and the six aryl rings are directed away from the center of the
molecule. A number of computational studies have interrogated
our structural model for 2 as well as those conformers that arise
from rotation of one or more anilide ligands about the MosN
bonds. One such investigation showed that rotation of a single
anilide ligand by 180° away from the structure described is
energetically uphill.55 By further exploring the effects of anilide-
ligand rotation, a recent theoretical study predicted a stable
minimum for a diamagnetic conformer in which two opposing
anilide ligands are rotated by 90°.50–53 Moreover, such a quasi-
C2h structure was predicted to be lower in energy than the
trigonal arrangement in which all tert-butyl groups are directed
inward toward the center of the molecule. In addition to a 90°
rotation of two anilide ligands, this conformation was found to
possess a trans-bent MoNNMo core. The calculated bond angles
around the µ-N2 ligand (MosNsN) were 175° or 164° for 2
and a model complex, respectively. Interestingly, such a
geometry deviates from the linear MoNNMo core that is found
for related, structurally characterized µ-N2 compounds of
molybdenum,38,57,58 but recalls the zig-zag transition state
structure for the N2 cleavage reaction shown in Scheme 1.10,49
To obtain more detailed structural data for 2 than in the prior
EXAFS study, single crystals of 2 were grown for the purpose
of X-ray diffraction. When solutions of 1 in a mixture of
MesCN/MeCy/O(SiMe3)2 were stored under an atmosphere of
N2 at -35 °C, we obtained purple crystals of 2 suitable for
X-ray diffraction. In the crystal structure, 2 equiv of O(SiMe3)2
pack with each equiv of 2 (Figure 1). The six anilide ligands
are arranged with the tert-butyl groups directed toward the center
of the molecule that features a linear MoNNMo core: MosNsN,
179.87(14)°. Three anilide ligands are arranged in a trigonal
conformation around each Mo center, with the three aryl
substitutents forming a C3 propeller. The crystallographic C2
axis, that is orthogonal to the NsN vector, requires the two C3
propellers to turn with the same chirality. The metrical
parameters obtained via single crystal X-ray diffraction (Table
1) are in close agreement with previously obtained EXAFS data
for frozen solutions of 2. The EXAFS data indicated a linear
MoNNMo core defined by a MosMo separation of 4.94 Å and
NsN internuclear distance of 1.19(2) Å.10
Table 1. Comparison of Internuclear Distances (Å) for
[(µ-N2)[Mo(N[t-Bu]Ar)3]2]n+ (n ) 0, 1, 2)
compound
2
2
2[B(ArF)4]
2[B(ArF)4]2
space group
NsN
MosN
P2/n
1.212(2)
1.868(1)
P21/n
P21/n
P21/n
1.265(5)
1.798(2)
1.217(2)
1.870(2)
1.872(2)
4.958(2)
1.986(2)
1.239(4)
1.835(3)
1.841(3)
4.9151(6)
1.956(3)
MosMo
MosN[t-Bu]Ar
(avg)
4.9476(5)
1.985(2)
4.8599(5)
1.931(3)
We found that 2 could also be crystallized from solutions of
1 in n-pentane/O(SiMe3)2 when stored under N2 at -35 °C. The
X-ray structure obtained using crystals so obtained indicated
that n-pentane was incorporated into the crystal packing. This
second molecular structure of 2 has a pseudo-C2 axis, although
no crystallographically imposed symmetry element bisects the
molecule; each atom is crystallographically independent. Com-
parison between the two X-ray crystal structures reveals no
significant differences in the molecular structure of 2 (Table
1).
Computational studies predicted that 2 exists as a triplet (S
) 1) at 20 °C when the anilide ligands adopt the observed,
trigonal arrangement.49–51,54,55 These predictions agree with the
previous experimental observation that (µ-N2)[Mo(N[t-Bu]Ph)3]2
exists as a triplet (S ) 1) at 20 °C.10 However, the proposal
that 2 and (µ-N2)[Mo(N[t-Bu]Ph)3]2 should exhibit similar
magnetic properties was recently challenged.50 It was predicted
that the bulky Ar groups of 2 force a pair of inversion-related
anilide ligands to rotate such that their NC2 planes become
orthogonal to the MoNNMo axis, and that this ligand rotation
coincides with a change in the magnetic ground state. The
resulting change in the calculated singlet-triplet gap was
dramatic: 1922 cm-1 for 2 compared to 585 cm-1 for (µ-
N2)[Mo(N[t-Bu]Ph)3]2.50 To test this intriguing prediction,
magnetic studies now have been carried out on 2.
To address experimentally the magnetism of 2, the variable-
temperature bulk magnetism of 2 was measured by SQUID
magnetometry.59 A plot of µeff vs T plateaus at a maximum
value of µeff ) 2.40 µB (Figure 2). This value is lower than the
expected spin-only value of 2.83 µB corresponding to a triplet
(57) Greco, G. E.; Schrock, R. R. Inorg. Chem. 2001, 40, 3861–3878.
(58) Shih, K.-Y.; Schrock, R. R.; Kempe, R. J. Am. Chem. Soc. 1994,
116, 8804–8805.
(59) O’Connor, C. J. In Molecule-Based Magnetic Materials; Turnbull,
M. M., Sugimoto, T., Thompson, L. K., Eds.; American Chemical
Society: Washington, DC, 1996; p 44-66.
9
9396 J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008

Features and Chemistry of Dinitrogen Cleavage
A R T I C L E S
1) ground-state that is best described by a 42 cm-1 separation
between the low lying ms ) 0 and degenerate ms ) -1, 1 levels
in the absence of an applied field.
Mechanism of Thermal NsN Bond Cleavage. A zig-zag
transition state structure, 2‡, has been proposed as part of a
mechanism for the required transit between the triplet and singlet
electronic surfaces that relate 2 and 2 equiv of 3 (Scheme
1).10,49,54,68,69 This transition-state model is a common feature
among theoretical analyses of thermal NsN bond cleavage by
2. Additionally, analogous zig-zag structures have been proposed
for the dimerization of two terminal nitrides to form a µ-N2
complex.24,70–73 For these reasons, it is noteworthy that the
singlet molecule, featuring a pair of inversion-related anilide
ligands rotated such that their NC2 planes become orthogonal
to the MoNNMo axis as located by Stranger and co-workers,
contains both this MoNNMo zig-zag structure and a longer
NsN bond than the observed, linear geometry.50–53 Although
this computationally predicted conformation for 2 agrees with
neither the structural nor magnetic data obtained from authentic
samples of 2 we still found the structural relationship between
the putative zig-zag transition state en route to cleavage of the
NsN bond and this proposed geometry intriguing.50,51
Rotation of one or more anilide ligands has previously not
been considered as a structural feature of the zig-zag transition
state, 2‡. Via ligand rotation, one of the metal π-symmetry (with
respect to the µ-N2 ligand) orbitals is destabilized and the orbital
degeneracies that exist in pseudo-C3 coordination geometries
are lifted. If one could show that ligand rotation were coupled
to the formation of a zig-zag structure of 2‡ then this would be
of mechanistic interest with regard to N2 cleavage. In this regard,
one expects the barrier to ligand rotation in 2 to be greater in
energy than for less sterically crowded (µ-N2)[Mo(N[i-Pr]Ar)3]2.
Interestingly, (µ-N2)[Mo(N[i-Pr]Ar)3]2 has never been directly
observed as an intermediate in the N2-cleavage reaction carried
out by HMo(η2-Me2CNAr)(N[i-Pr]Ar)2.74,75 The accordingly
inferred instability of (µ-N2)[Mo(N[i-Pr]Ar)3]2 is in sharp
contrast with the properties of 2, which can be isolated and
studied as a pure material. We have suggested that facile N2
cleavage by (µ-N2)[Mo(N[i-Pr]Ar)3]2 indicates that this com-
pound more readily accommodates geometric changes along the
reaction coordinate than does 2.76,77
Figure 2. Dependence of the magnetic moment of 2 upon temperature as
measured by SQUID magnetometry. The inset shows molar magnetization
of 2 as a function of both temperature and field strength. The magnetic
data were fit to a spin Hamiltonian for an S ) 1 state with zero-field splitting.
The parameters for the fits shown above are g ) 1.69 and D ) 42 cm-1
.
Contributions made by temperature-independent paramagnetism were fit
and subtracted from the data. N is Avogadro’s number, ꢀ is the Bohr
magneton, and k is Boltzmann’s constant.
(S ) 1) spin system.60–62 Moreover, the magnetic moment
rapidly declines when the temperature is reduced below 50 K.
To pinpoint the physical origin of the magnetic behavior at low
temperatures, magnetic data were acquired on a single sample
at three fields (0.5, 2.5, and 5.0 T) over a temperature range of
0-250 K. The plot of molar magnetization (M) vs ꢀB/kT
reaches a different maximum value at each field strength (Figure
2, inset). This finding suggested that the decline of the magnetic
moment at low temperatures is due to sizable zero-field splitting
of a ground-state triplet.63 Accordingly, the variable-temperature,
variable-field data were fit to an S ) 1 spin Hamiltonian to
provide the zero-field splitting parameter, D ) 42 cm-1, and
the gyromagnetic ratio, g ) 1.69 using the julX software
package.64 The positive sign of the zero-field splitting parameter,
D, indicates that the ms ) 0 level of the triplet state lies lower
in energy than the ms ) -1, 1 level. At low temperatures and
low magnetic fields the ms ) -1, 1 levels are depopulated and
the magnetization is correspondingly decreased. The g value
of 1.69 is lower than the Lande´ ge, reflecting the axial anisotropy
that is suggested by the value of D.65 Moreover, the value
obtained for g predicts a maximum spin-only magnetic moment
of 2.39 µB (via µeff ) g × (S(S + 1))1/2) which agrees well
with the observed value near 300 K. Using an alternative model
provided by the Bleaney-Bowers equation66,67 resulted in a
less satisfactory fit of the magnetic data and fails to account
for the field-dependence of the molar magnetization (Figure
S19). Therefore, we conclude that 2 has an isolated triplet (S )
The structure of (µ-N2)[Mo[(R)N(CH2)2]3N]2 (R ) Me2[t-
Bu]Si) has been compared with the structure of 2 previously.10
The two compounds have magnetic moments consistent with
an S ) 1 state and similar structural parameters. However, (µ-
N2)[Mo[(R)N(CH2)2]3N]2 does not undergo thermal fragmenta-
tion with NsN bond cleavage.58 We suggest that the chelating
(68) Woodward, R. B.; Hoffmann, R. Angew. Chem., Int. Ed. 1969, 8,
781–853.
(60) O’Connor, C. J. In Prog. Inorg. Chem.; Lippard, S. J., Ed.; John
Wiley & Sons, Inc.: New York, 1982; Vol. 29, p 203-283.
(61) Van Vleck, J. H. The Theory of Electric and Magnetic Susceptibilities;
Oxford University Press: New York, 1965; p 282-297.
(62) Ballhausen, C. J. Introduction to Ligand Field Theory; McGraw-
Hill: New York, 1962; p 139-151.
(69) Veige, A. S.; Slaughter, L. M.; Lobkovsky, E. B.; Wolczanski, P. T.;
Matsunaga, N.; Decker, S. A.; Cundari, T. R. Inorg. Chem. 2003,
42, 6204–6224.
(70) Ware, D. C.; Taube, H. Inorg. Chem. 1991, 30, 4605–4610.
(71) Seymore, S. B.; Brown, S. N. Inorg. Chem. 2002, 41, 462–469.
(72) Seymore, S. B.; Brown, S. N. Inorg. Chem. 2006, 45, 9540–9550.
(73) Man, W. L.; Tang, T. M.; Wong, T. W.; Lau, T. C.; Peng, S. M.;
Wong, W. T. J. Am. Chem. Soc. 2004, 126, 478–479.
(74) Tsai, Y.-C.; Johnson, M. J. A.; Mindiola, D. J.; Cummins, C. C.;
Klooster, W. T.; Koetzle, T. F. J. Am. Chem. Soc. 1999, 121, 10426–
10427.
(63) Girerd, J.-J.; Journaux, Y. In Physical Methods in Bioinorganic
Chemistry; Lawrence Que, J., Ed.; University Science Books:
Sausalito, CA, 2000; p 336-342.
(64) Bill, E. Julx 1.4, A Program for the Simulation and Analysis of
Magnetic Susceptibiltiy Data; Max Plank Institute for Bioinorganic
Chemistry: Mu¨lheim an der Ruhr, 2008 (http://ewww.mpi-muelhe-
im.mpg.de/bac/logins/bill/julX_en.php).
(75) Cherry, J.-P. F.; Stephens, F. H.; Johnson, M. J. A.; Diaconescu,
P. L.; Cummins, C. C. Inorg. Chem. 2001, 40, 6860–6862.
(76) Hammond, G. S. J. Am. Chem. Soc. 1955, 77, 334–338.
(77) Carey, F. A.; Sundberg, R. J. AdVanced Organic Chemsitry Part A:
Structure and Mechanism; Kluwer Academic/Plenum Publishers:
New York, 2000; p 217-222.
(65) Kahn, O. Molecular Magnetism; VCH Publishers: New York, 1993.
(66) Bleaney, B.; Bowers, K. D. Proc. R. Soc. London, Ser. A 1952, 214,
451–461.
(67) Kahn, O. Molecular Magnetism; VCH Publishers: New York, 1993;
p 103-134.
9
J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008 9397

A R T I C L E S
Curley et al.
arms of the [(R)N(CH2)2]3N ligand inhibit rotation about the
MosNarm bond. If ligand rotation is important for accommodat-
ing transit between the triplet and singlet spin-states, then
fundamental differences in the reaction chemistry of these two
closely related molecules might be explained in terms of
ancillary ligand structural constraints. Alternatively, this dif-
ference might also be the result of the apical-nitrogen donor
that is present in (µ-N2)[Mo[(R)N(CH2)2]3N]2 but absent in 2.49
Solid-State Stability of 2 and Structure of Crystalline 3. In
solution, 2 readily fragments to yield 2 equiv of 3 (t1/2 ≈ 30
min, 25 °C).10 Having obtained for the first time samples of 2
as single crystals, we became curious as to whether the extended
crystal lattice would confer additional stability to 2.78–83 Crystals
of 2 (ca. 0.25 × 0.20 × 0.20 mm3), grown from n-pentane/
O(SiMe3)2, were stored at 20 °C under light petroleum oil for
24 h. After that time, the crystals continued to diffract and the
unit cell was measured. The measured unit cell parameters in
P21/n were unchanged in comparison to those found for fresh
crystals that had been analyzed 24 h earlier. These crystals were
additionally suspended in nujol oil and monitored by UV-vis
spectroscopy at 20 °C over 22 h. During that time no change
had occurred in the absorbance spectrum. These two observa-
tions suggest that in comparison to solutions or amorphous
solids, the extended structure of the crystal lattice stabilizes 2
with respect to fragmentation.84
Figure 3. The molecular structure of 3 is shown with thermal displacement
ellipsoids at the 50% probability level. Hydrogen atoms have been omitted
for clarity.
Scheme 2
The terminal nitride, 3, formed by thermal fragmentation of
2, has not been structurally characterized in the past due to a
severe crystallographic disorder. Here we report the first
satisfactory solution to X-ray diffraction data. Crystals of 3 were
grown from n-pentane/Et2O at -35 °C. It was found that slow
cooling of a pale yellow crystal of 3 from 20 to -173 °C was
essential to prevent cracking of the crystal, allowing X-ray
diffraction data to be collected. A satisfactory solution to the
X-ray diffraction data was found in the space group P21/n. This
solution contains four crystallographically independent mol-
ecules per asymmetric unit, two of which are highly disordered.
Each molecule of 3 approximates C3 symmetry in the solid state,
as the aryl rings are staggered in the same direction (Figure 3).
The analogous compound, NtMo(N[t-Bu]Ph)3, is not subject
to the same crystallographic disorder, and its X-ray crystal
structure has been previously reported.10 Therefore, the metrical
parameters of this compound may be compared to those of 3.
For 3, the MotN bond length is 1.651(4) Å while the average
MosN[t-Bu]Ar bond length is 1.964(4) Å. These values are
closely approximated by those found for NtMo(N[t-Bu]Ph)3
for which the MotN internuclear distance is 1.658(5) Å and
one MosN[t-Bu]Ph distance is 1.979(2) Å.10
C5H5, ArF ) 3,5-C6H3(CF3)2). When the oxidation is carried
out in Et2O, (µ-N2)[Mo(N[t-Bu]Ar)3]2[B(ArF)4]2, 2[B(ArF)4]2,
is collected by filtration as a red powder in 98% yield (Scheme
2). Powders or CH2Cl2 solutions of 2[B(ArF)4]2 may be stored
at 20 °C for greater than 5 d without significant decomposition.
Reduction of 2[B(ArF)4]2 to return 2 was effected by the
soluble reducing agent K2COT (COT ) C8H8).85 This reducing
agent was desirable because it is soluble in THF86,87 and has a
sufficient reduction potential to carry out both reduction steps
to form 2 while avoiding reduction of 2 to form 2 equiv of
[K][NNMo(N[t-Bu]Ar)3].88–91 Reduction of 2[B(ArF)4]2 by
K2COT proceeds rapidly in THF solutions to form 2 as the only
molybdenum-containing product (Scheme 2). Using this pro-
cedure, 2 may be obtained in 85% isolated yield, after separation
from the K[B(ArF)4] and COT byproducts.
Redox Chemistry of (µ-N2)[Mo(N[t-Bu]Ar)3]2. We envisioned
that by removing 2e from 2, the corresponding formal d2-d2
compound should be thermally stable because too few electrons
are present to engage in 6e reductive cleavage of the NsN bond.
Moreover, a thermally stable, dicationic compound might be
useful for the storage and in situ generation of 2. With isolated
2 in hand, the 2e oxidation was carried out by addition of 2 to
a solution containing 2 equiv of [Cp2Fe][B(ArF)4] (Cp ) η5-
(85) Katz, T. J. J. Am. Chem. Soc. 1960, 82, 3784–3785.
(86) Stevenson, G. R.; Ocasio, I. J. Phys. Chem. 1975, 79, 1387–1390.
(87) Cox, R. H.; Harrison, L. W.; Austin, W. K. J. Phys. Chem. 1973,
77, 200–204.
(78) Ramamurthy, V.; Venkatesan, K. Chem. ReV. 1987, 87, 433–481.
(79) Buergi, H. B.; Dunitz, J. D. Acc. Chem. Res. 1983, 16, 153–161.
(80) Ohashi, Y. Acc. Chem. Res. 1988, 21, 268–274.
(81) Ohashi, Y.; Sasada, Y. Nature 1977, 267, 142–144.
(82) Ohashi, Y.; Yanagi, K.; Kurihara, T.; Sasada, Y.; Ohgo, Y. J. Am.
Chem. Soc. 1981, 103, 5805–5812.
(88) Allendoerfer, R. D. J. Am. Chem. Soc. 1975, 97, 218–219.
(89) Allendoerfer, R. D.; Rieger, P. H. J. Am. Chem. Soc. 1965, 87, 2336–
2344.
(90) Katz, T. J.; Reinmuth, W. H.; Smith, D. E. J. Am. Chem. Soc. 1962,
84, 802–808.
(83) Paul, I. C.; Curtin, D. Y. Acc. Chem. Res. 1973, 6, 217–225.
(84) Ozerov, O. V.; Guo, C.; Papkov, V. A.; Foxman, B. M. J. Am. Chem.
Soc. 2004, 126, 4792–4793.
(91) We measured an E1/2 of-2.9 V (vs Cp2Fe+/0) for the first 1e oxidation
of K2COT in 0.2 M [(n-Bu)4N][PF6]/THF.
9
9398 J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008

Features and Chemistry of Dinitrogen Cleavage
A R T I C L E S
Figure 4. The cyclic voltammogram of 2[B(ArF)4]2 in 0.1 M [(N(n-Bu)4]-
[B(C6F5)4]/THF92,93 shows two reversible redox events. Assignment of each
event is given in the text.
Figure 5. A spectroelectrochemical stack plot shows the conversion 2f
2+ f 22+ in 0.5 M [N(n-Bu)4][PF6]/THF in 0.5-s intervals.
Scheme 3
We discovered that 2+ is formed when 2 is dissolved in CDCl3
as this solution quickly changes in color from purple to
cranberry-red. Proton NMR, UV-vis, and Raman spectroscopy
data support this conclusion, however the identity of the
counteranion has not been investigated.
For 2[B(ArF)4], the magnetic moment of 1.96 µB is consistent
with one unpaired electron. This value, which is higher than
the expected spin-only value of 1.73 µB, may indicate an orbital
contribution from a 2E ground state (approximate C3 symmetry
is assumed) to the observed magnetic moment;103–105 however,
care should be taken when interpreting these magnetic data
because the measurement requires a large diamagnetic correc-
tion, ꢁdia ) -1.29 × 10-3 cm3 Mol-1 (ꢁM ) ꢁobs - ꢁdia).60
To further investigate the redox couples interrelating 2, 2+,
and 22+, the 2e oxidation of 2 was monitored spectroelectro-
chemically, allowing the visible transition of each chromophore
to be compared in the same solvent and under the same
experimental conditions.96,106 The oxidation of 2 in 0.5 M [(n-
Bu)4N][PF6]/THF was monitored by UV-vis spectroscopy,
while holding the spectroscopic cell at a potential of -200 mV.
A stack plot of the UV-vis spectra, taken in 0.5-s intervals,
shows the decay of 2, which absorbs at 544 nm, accompanied
by the growth of 2+ absorbing at 506 nm (Figure 5). Changes
in the absorbance are less dramatic as 2+ is oxidized to form
Two reversible 1e processes are observed in the cyclic
voltammogram of 2[B(ArF)4]2 (Figure 4).92,93 These are assigned
to the 22+/2+ couple at -320 mV and 2+/2 couple at -1460
mV, where Cp2Fe+/0 ) 0 mV. The 2+/2 couple has previously
been observed in the cyclic voltammogram of 2.16 From a
qualitative point of view, 2 is about as reducing as Cp2Co.94
The separation between the two reversible 1e processes
observed in the cyclic voltammogram is related to the ∆G° for
the comproportionation between 2 and 22+ to form 2 equiv of
2+.95–97 From the peak separation of 1.14 V, the equilibrium
constant for this reaction was calculated to be Keq ) 1.9 ×
1019.96–99 Comproportionation served as a viable synthetic route
to (µ-N2)[Mo(N[t-Bu]Ar)3]2[B(ArF)4], 2[B(ArF)4]: Addition of
2[B(ArF)4]2 to a solution of 2 rapidly formed the mixed-valent
compound, 2[B(ArF)4] which was isolated as a cranberry-red
solid in 95% yield (Scheme 3). Proton NMR spectra of
2[B(ArF)4] show a broad tert-butyl resonance centered at 6.0
ppm. Addition of either 2 or 2[B(ArF)4]2 to solutions of
2[B(ArF)4] broadens the resonances observed by 1H NMR
spectroscopy, as is expected for degenerate electron transfer.100–102
2
2+ because both species absorb strongly at 506 nm; however,
the spectral features near 300 nm differ between the two. The
spectral assignment for 2+ was confirmed by measuring the
UV-vis spectrum of 2+ that was electrochemically generated
from 22+ in an optically transparent thin-layer electrochemical
cell (see Supporting Information).107,108
(92) LeSuer, R. J.; Buttolph, C.; Geiger, W. E. Anal. Chem. 2004, 76,
6395–6401.
Solid State Structure of 2[B(ArF)4]. Crystals of 2[B(ArF)4]
for X-ray analysis were grown by vapor diffusion of n-pentane
(93) LeSuer, R. J.; Geiger, W. E. Angew. Chem., Int. Ed. 2000, 39, 248–
250.
(94) Connelly, N. G.; Geiger, W. E. Chem. ReV. 1996, 96, 877–910.
(95) Demadis, K. D.; Hartshorn, C. M.; Meyer, T. J. Chem. ReV. 2001,
101, 2655–2686.
(102) Yang, E. S.; Chan, M.-S.; Wahl, A. C. J. Phys. Chem. 1980, 84,
3094–3099.
(103) Gray, H. B.; Hendrickson, D. N.; Sohn, Y. S. Inorg. Chem. 1971,
10, 1559–1563.
(96) Faulkner, L. R.; Bard, A. J. Electrochemical Methods: Fundementals
and Applications; 2nd ed.; Wiley: Hoboken, NJ, 2001; p 680-694.
(97) Save´ant, J.-M. Elements of Molecular and Biomolecular Electro-
chemistry: An Electrochemical Approach to Electron Transfer
Chemistry; John Wiley & Sons: Hoboken, NJ, 2007, p 62-75.
(98) Richardson, D. E.; Taube, H. Inorg. Chem. 1981, 20, 1278–1285.
(99) Barriere, F.; Geiger, W. E. J. Am. Chem. Soc. 2006, 128, 3980–
3989.
(104) Sohn, Y. S.; Hendrickson, D. N.; Gray, H. B. J. Am. Chem. Soc.
1970, 92, 3233–3234.
(105) Kramer, J. A.; Hendrickson, D. N. Inorg. Chem. 1980, 19, 3330–
3337.
(106) Rosenthal, J.; Luckett, T. D.; Hodgkiss, J. M.; Nocera, D. G. J. Am.
Chem. Soc. 2006, 128, 6546–6547.
(107) Pilkington, M. B. G.; Coles, B. A.; Compton, R. G. Anal. Chem.
1989, 61, 1787–1789.
(100) Dietrich, M. W.; Wahl, A. C. J. Chem. Phys. 1963, 38, 1591–1596.
(101) Yang, E. S.; Chan, M.-S.; Wahl, A. C. J. Phys. Chem. 1975, 79,
2049–2052.
(108) Heineman, W. R.; Norris, B. J.; Goelz, J. F. Anal. Chem. 1975, 47,
79–84.
9
J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008 9399

A R T I C L E S
Curley et al.
Figure 6. The molecular structure of 2[B(ArF)4] is shown with thermal
displacement ellipsoids at the 50% probability level. Hydrogen atoms have
been omitted for clarity.
into THF at 20 °C. All atoms are crystallographically indepen-
dent in the space group P21/n, and the two MosN distances
around the µ-N2 ligand, Mo(1)sN(7) and Mo(2)sN(8), are
equivalent, within statistical error (Table 1, Figure 6). The
MoNNMo core is nearly linear with MosNsN angles of
179.8(3)° and 178.7(3)°. A distortion breaking the trigonal
symmetry around the MoNNMo core might be either small or
a dynamic effect that is not observed.109–111
The symmetrical bonding about the MoNNMo core in
2[B(ArF)4] contrasts other mixed-valent µ-N2 complexes that
exhibit asymmetrical bonding about the µ-N2 ligand.112,113 Such
complexes typically have a strong absorbance in the IR
spectrum, whereas for 2[B(ArF)4] the NsN stretching mode is
only Raman active (vide infra).112 The symmetric bonding about
the µ-N2 ligand in 2[B(ArF)4] is comparable with [(µ-
N2)(M(NH3)5)2]5+ (M ) Ru, Os), which have Raman active
but IR inactive bands corresponding to the NsN normal mode.
Therefore, structural characterization of 2[B(ArF)4] is consistent
with a Robin-Day class III system in which the unpaired
electron resides in a molecular orbital that is delocalized across
the NsN bridge and is symmetrically shared by both metals.95,114
Absorbance bands in the near-IR region (700-3000 nm) were
not located for 2[B(ArF)4]. The lack of intense near-IR bands
is consistent with the classification of 2[B(ArF)4] as a class III
mixed valent complex, for which these bands can become
exceedingly weak, hence unobserved.95
Figure 7. The molecular structure of 2[B(ArF)4]2 is shown with thermal
displacement ellipsoids at the 50% probability level. Hydrogen atoms have
been omitted for clarity.
Solid State Structure of 2[B(ArF)4]2. Crystals of 2[B(ArF)4]2,
grown from THF/Et2O/n-hexane at -35 °C, were suitable for
X-ray diffraction. The molecular structure of 2[B(ArF)4]2 is
symmetric about a crystallographic inversion center that is
located on the NsN internuclear vector (Figure 7). The
inversion center relates the two C3 propellers formed by the
staggered aryl rings, conferring a pseudo-S6 geometry to the
22+ ion. The short MosMo distance of 4.86 Å suggests a
congested ligand environment about the MoNNMo core (Table
1). Steric crowding is also evident in the 1H NMR spectrum at
20 °C as two distinct resonances are observed for both the
aromatic ortho- and aryl-methyl protons. These protons become
magnetically inequivalent when the dynamic process that
interconverts the two possible enantiomers defined by the C3
aryl propellers is sufficiently hindered.
The NsN internuclear distance is a commonly used marker
of N2 activation when it is coordinated to a metal;2,115 therefore,
it is worthwhile to compare the structures of 2[B(ArF)4]2,
2[B(ArF)4], and 2. The NsN internuclear distance is longest in
2[B(ArF)4]2, 1.265(5) Å, while the NsN distance is ca. 0.026
Å shorter for 2[B(ArF)4] or ca. 0.051 Å shorter for 2. The
opposite trend is found for MosN internuclear distances; 2 has
the longest MosN separation, 1.87 Å, when compared to the
MosN distance in 2[B(ArF)4], 1.835(3) Å, or 2[B(ArF)4]2,
1.798(2) Å. Both trends indicate that the µ-N2 ligand becomes
more activated by removing 2e from the π bonding MoNNMo
core of 2.115
(109) Jahn, H. A.; Teller, E. Proc. R. Soc. London, A 1937, 161, 220–235.
(110) Sugano, S.; Tanabe, Y.; Kamimura, H. Mutiplets of Transition-Metal
Ions in Crystals; Academic Press: New York, 1970, p 241-248.
(111) Bersuker, I. B. Chem. ReV. 2001, 101, 1067–1114.
(112) Demadis, K. D.; Meyer, T. J.; White, P. S. Inorg. Chem. 1997, 36,
5678–5679.
(113) Demadis, K. D.; El-Samanody, E. S.; Coia, G. M.; Meyer, T. J. J. Am.
Chem. Soc. 1999, 121, 535–544.
(114) Robin, M. B.; Day, P. In AdV. Inorg. Chem. Radiochem.; Emele´us,
H. J., Sharpe, A. G., Eds.; Academic Press: New York, 1967; Vol.
10, p 247-423.
(115) Fryzuk, M. D.; Johnson, S. A. Coord. Chem. ReV. 2000, 200, 379–
409.
9
9400 J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008

Features and Chemistry of Dinitrogen Cleavage
A R T I C L E S
Activation of N2 is generally achieved by reducing the N2,
and it is noteworthy to recognize that a number of dicationic
µ-N2 compounds are described in the literature. Schrock has
reported
the
electrochemistry
of
(µ-N2)[M((4-t-
BuC6H4)NCH2CH2)3N]2, M ) Mo, W.57 For the case of W,
the neutral µ-N2 complex was cleanly oxidized by either 1e or
2e to afford isolable products; however, crystals suitable for
X-ray diffraction studies could not be grown for these com-
pounds. The compound [(µ-N2)(W(dmpe)Cp*)2][B(C6F5)4]2
(Cp* ) η5-(CCH3)5, dmpe ) 1,2-(Me2P)2C2H4), formed upon
exposure of [(η7-C5Me3(CH2)2)(dmpe)W(H)2][B(C6F5)4] to N2,
has an NsN separation of 1.22(1) Å and a WsN distance of
1.888(5) Å while the WsNsN angle is nearly linear, 176(7)°.116
The dizwitterionic compound (µ-N2)[Fe((i-Pr)2PCH2CH2)3BPh]2
shows significant activation of the N2 ligand in its NsN distance
of 1.138(6) Å and FesN separation of 1.815 Å (average of
2).24 A recent EPR/Mo¨ssbauer study concluded that this
compound is best described as an Fe(I)/Fe(I) ground-state
septet.117 These examples demonstrate that marked reduction
of the µ-N2 ligand may be present even when bound between
two positively charged metal centers.
Figure 8. Raman shifts plotted as a function of NsN internuclear distance.
The best fit line through four points is given by ν(r) ) -5859r + 8762, R2
) 1.000, σ ) 1.04 cm -1. Raman shifts are given in the text and bond
lengths are listed in Table 1.
the three data points may be extrapolated to the Raman shift
for gaseous N2.119–121 The correlation of Raman shifts with
NsN bond length is consistent with the observation that the
overall complex structure remains unchanged by removing either
1e or 2e from 2: only MosN and NsN distances of the
MoNNMo core are changed by altering the charge.
Raman Spectroscopy. Raman spectroscopy is a common
marker of activation of the N2 ligand,2,115 and Raman shifts
are known to correlate well with NsN internuclear separation
for a number of compounds containing the N2 unit.10,118 The
extent of activation of the N2 ligand is typically measured against
free N2 which has a NsN separation of 1.0975 Å and Raman
DFT and Electronic Absorption Spectroscopy. Using the
experimental X-ray diffraction structures of 2 and 22+ as a
starting geometry, density functional theory (DFT)124 optimized
geometries were obtained for both µ-N2 complexes (see Sup-
porting Information, Tables S2 and S3). The frontier orbitals
calculated for 22+ indicate that the HOMO corresponds to the
bonding interaction between the Mo d orbitals and the π* orbital
of N2 whereas the LUMO is NN π bonding in character (Figure
9). Such an analysis leads one to expect that that population of
the LUMO orbitals by either 1e or 2e would enhance the
bonding between the two nitrogen atoms. This interpretation is
nicely supported by the X-ray and Raman data described above:
upon reduction of 2[B(ArF)4]2 to either 2[B(ArF)4] or 2,
shortening of the NsN bond is observed by X-ray crystal-
lography and a higher energy νNN is observed by Raman
spectroscopy.
It is interesting that the LUMO+2 is a low-lying σ symmetry
orbital. This orbital contains NN antibonding character but may
be viewed as substantially metal-based. The LUMO+3 is a
π-symmetry orbital that is NN antibonding in character.
However, the metal-localized lobes in LUMO+3 are not
coplanar with the nitrogen-based orbitals. Strongly antibonding
π-symmetry orbitals of the MoNNMo core are actually much
higher in energy (see Supporting Information for orbital
energies).
shift of 2331 cm-1 119–121
To complement the trend in NsN
.
internuclear distance for 2n+ (n ) 0, 1, 2) as observed by X-ray
crystallography (Table 1), Raman spectroscopy was carried out
on these compounds.
Resonance Raman spectroscopy performed on 2[B(ArF)4]2
(λexcite ) 514.5 nm) showed only one intense feature at 1349
cm-1 which was assigned to the Stokes shift for NsN
polarization. This assignment was confirmed by the Raman shift
for [15N-2][B(ArF)4]2 which is found at 1305 cm-1, 44 cm-1
less energetic than the shift observed for the lighter isotopomer.
This decrease in the Raman shift upon isotopic labeling agrees
with the change calculated for a harmonic oscillator.122 Raman
spectroscopy of 2[B(ArF)4] (λexcite ) 785 nm) showed an intense
Stokes shift at 1503 cm-1 which was shifted to 1438 cm-1 for
the 15N isotopomer. The observed isotope shift of 65 cm-1 is
greater than the calculated shift of 51 cm-1, indicating a degree
of anharmonic character in the vibration. Anharmonic potentials
have been associated with a dynamic Jahn-Teller distortion.110,123
The Raman spectrum of 2 contains the NsN bond polariza-
tion at 1630 cm-1 10
The lower energy polarizations of 1503
.
and 1349 cm-1 for 2[B(ArF)4] and 2[B(ArF)4]2, respectively,
are consistent with a weakening of the NsN bond upon
oxidation of 2.2,115 These Raman shifts may be plotted as a
function of NsN internuclear distance (Figure 8),118 moreover
This molecular orbital analysis is consistent with previous
treatments.10,20,125–127 For example, the short NsN bond in [(µ-
N2)(Ru(NH3)5)2][BF4]4 (1.124 Å, νNN ) 2100 cm-1) has been
attributed to filled orbitals of RusN antibonding character.127–129
Accordingly, an intense absorption (λ ) 263 nm, ε ) 48000
(116) Mork, B. V.; Tilley, T. D. J. Am. Chem. Soc. 2004, 126, 4375–4385.
(117) Hendrich, M. P.; Gunderson, W.; Behan, R. K.; Green, M. T.; Mehn,
M. P.; Betley, T. A.; Lu, C. C.; Peters, J. C. Proc. Natl. Acad. Sci.
U.S.A. 2006, 103, 17107–17112.
(124) Baerends, E. J.; et al. ADF2004.01 ed. Theoretical Chemistry; Vrije
Universiteit: Amsterdam, The Netherlands, 2004; http://www.scm-
.com.
(118) Badger, R. M. J. Chem. Phys. 1935, 3, 710–714.
(119) Buthcer, R. J.; Jones, W. J. J. Chem. Soc., Faraday Trans. 1974, 70,
560–563.
(125) O’Regan, M. B.; Liu, A. H.; Finch, W. C.; Schrock, R. R.; Davis,
W. M. J. Am. Chem. Soc. 1990, 112, 4331–4338.
(126) Sanner, R. D.; Manriquez, J. M.; Marsh, R. E.; Bercaw, J. E. J. Am.
Chem. Soc. 1976, 98, 8351–8357.
(120) Jørgen, B. J. Raman Spectrosc. 1974, 2, 133–145.
(121) Jones, W. J. Can. J. Phys. 2000, 78, 327–390.
(122) Ebsworth, E. A. V.; Rankin, D. W. H.; Cradock, S. Structural
Methods in Inorganic Chemistry; Blackwell Scientific Publications:
Cambridge, MA, 1991, p 228-233.
(127) Richardson, D. E.; Sen, J. P.; Buhr, J. D.; Taube, H. Inorg. Chem.
1982, 21, 3136–3140.
(128) Harrison, D. F.; Weissberger, E.; Taube, H. Science 1968, 159, 320–
322.
(123) Bersuker, I. B. The Jahn-Teller Effect; Cambridge University Press:
New York, 2006.
(129) Treitel, I. M.; Flood, M. T.; Marsh, R. E.; Gray, H. B. J. Am. Chem.
Soc. 1969, 91, 6152–6153.
9
J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008 9401

A R T I C L E S
Curley et al.
Figure 10. The UV-vis stack plot shows the bleach at 544 nm that
accompanies 1-s periods of irradiation (λirr ) 546 nm) of 2.
contain more NsN antibonding character than the ground state.
These authors described a geometric distortion of the MNNM
(M ) Fe, Ru) core from a linear to a trans-bent, “diazenido-
like” structure in the first singlet excited state.37 We felt that
this “diazenido-like” core resembled the putative zig-zag
transition state, 2‡, for cleavage of the NsN bond. Having
acquired a more thorough picture of the electronic structure of
2 we were compelled to explore possible photoreactions
occurring from the MoNNMo core. One previously reported
example of such a reaction resulted in photochemical NsN bond
cleavage; irradiation of (µ-N2)[Mo(Mes)3]2 forms (µ-
N)[Mo(Mes)3]2 and 0.5 equiv N2.38
Motivated by these ideas, we explored the photochemistry
of 2 when irradiated in the region of the intense, visible
absorption band centered at 544 nm. To isolate this absorbance,
it was convenient to use the 546 nm Hg-line of an Hg-Xe arc
lamp. Irradiation of 2 in Et2O at this wavelength results in a
rapid bleach at 542 nm, accompanied by a slight increase in
absorbance near 430 nm (Figure 10). Bulk photolysis of 2 was
carried out using broadband light (λirr g 480 nm) at -78 °C, at
which temperature 2 is sufficiently thermally stable to isolate
the photochemical process. Treatment of 2 in this way forms a
mixture of 1 and 3 in an approximate 1:1 ratio, while releasing
0.5 equiv of N2 (Scheme 4).133 This procedure was repeated
with 15N2-2 to unambiguously demonstrate that 15N-3 (15N
NMR: δ ) 839.5 ppm) is formed from the µ-N2 ligand. The
efficiency of the photoreaction was assayed at -78 °C by both
ferrioxolate and Reinieke salt actinometry.134–136 Both methods
determined a primary quantum yield of Φp ) 0.05 with respect
to 2.
Figure 9. Selected frontier orbitals of 2[B(ArF)4]2 show NsN bonding
character of the LUMO. Orbitals are drawn at an isosurface value of 0.037.
For simplicity, only one orbital is drawn for pairs of orbitals that are close
in energy and transform under the same irreducible representation in the S6
point group. The all-bonding MoNNMo orbital is much lower in energy
and not shown.
M-1 cm-1) was assigned as the NN dπ f π* transition. At the
time, the energy of that transition was a suggested parameter
for NsN bond activation, hinting at the possibility of photo-
chemical reactions of the MNNM core (M ) transition
metal).129
In view of the molecular orbital analysis presented for
2[B(ArF)4]2, it is of interest to examine the origin of the intense
absorption band that appears in the visible region for 2,
2[B(ArF)4], and 2[B(ArF)4]2 (Figure 5). The visible absorbance
for each of the three has both similar energy and intensity: for
2, 544 nm (ε ) 41000 M-1 cm-1); for 2[B(ArF)4], 506 nm (ε
) 38000 M-1 cm-1); and for 2[B(ArF)4]2, 510 nm (ε ) 39800
M-1 cm-1) (Figure S10). Considering first 2[B(ArF)4]2, it is
expected that the intense red color of this compound arises from
a suitable electronic transition that promotes 1e from the HOMO
to one of the low-lying empty orbitals. Time-dependent
DFT130–132 found that promoting 1e from the HOMO into the
LUMO is a low-energy process; however, promotion of 1e to
the LUMO+2 was predicted to have much greater intensity in
the absorption spectrum. Assigning the UV-vis spectra of
2[B(ArF)4] and 2 is further complicated by the presence of
unpaired electrons that give rise to a large number of excited-
state electronic configurations. A detailed description of those
electronic transitions that give rise to the observed spectra
remains a topic for future detailed investigation.
In an attempt to observe an intermediate in the photore-
action, the electronic transient absorption spectrum was
acquired immediately following photolysis of 2 with a 544
nm, 10 ns laser pulse (Figure S12). The obtained transient
absorption spectrum matched a digitally subtracted spectrum
of 2 and products formed during photolysis, indicating that
(133) Following photolysis, the crude reaction mixture was treated with
NO to convert paramagnetic 1 into the previously characterized,
diamagnetic species ONMo(N[t-Bu]Ar)3. Integration of the 1H NMR
spectrum led to an observed ratio of 56:44 of nitrosyl to 3. The loss
of N2 is assumed on the basis of the observed product distribution
but not quantified.
Photochemistry of 2. Recent theoretical attention has been
focused on the possibility that inert, linear µ-N2 complexes might
engage in photochemical reactions that productively use the N2
fragment. One study suggested that low-lying excited states may
(134) Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, T. M. Handbook of
Photochemistry; 3rd ed.; Taylor & Francis: New York, 2006; p 600-
616.
(130) Kootstra, F.; Boeij, P. L. d.; Snijders, J. G. J. Chem. Phys. 2000,
112, 6517–6531.
(135) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. London, Ser. A 1956,
518–536.
(131) Romaniello, P.; Boeij, P. L. d. J. Chem. Phys. 2007, 127, 174111.
(132) Romaniello, P.; Boeij, P. L. d. Phys. ReV. B 2005, 71, 155108.
(136) Wegner, E. E.; Adamson, A. W. J. Am. Chem. Soc. 1966, 88, 394–
404.
9
9402 J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008

Features and Chemistry of Dinitrogen Cleavage
A R T I C L E S
Scheme 4
Scheme 5
the photochemical bleach is completed within the 10 ns laser
pulse. Single-wavelength kinetics, obtained by monitoring
the bleach at 530 nm, agree with this result (Figure S13).
These data indicate that transient absorption measurements
need to be performed on a faster time scale to capture the
absorption spectrum of possible intermediates in the
photoreaction.
In contrast to thermal NsN bond cleavage by 2, the
photochemical fragmentation of 2 produces both 1 and 3
(Scheme 4). The observed bifurcation of the reaction path
in the photoreaction of 2 is intriguing. Loss of N2 from 2 is
a process reminiscent of N2 extrusion from organic azo
complexes,137–140 and most likely proceeds from an electronic
excited state. The other process, homolytic fragmentation of
the NsN bond, resembles the thermally induced fragmenta-
tion of 2. Currently, we are unable to distinguish whether
NsN bond cleavage occurs from the electronic excited-state
that is responsible for the observed visible transition or from
a lower-energy state.141,142
Scheme 6
A competition between NsN bond fragmentation and N2
extrusion, as described above for 2, may be responsible for the
isolation of (µ-N)[Mo(Mes)3]2 upon bulk photolysis of (µ-
N2)[Mo(Mes)3]2. Floriani proposed that photolysis of (µ-
N2)[Mo(Mes)3]2 fragments the NsN bond, and that the 2 equiv
of NMo(Mes)3 so generated are quickly trapped by (µ-
N2)[Mo(Mes)3]2 forming (µ-N2)[Mo(Mes)3NMo(Mes)3]2 (Scheme
5). Loss of N2 from this putative tetramolybdenum intermediate
was the favored explanation for the observed products.38 An
alternative mechanism is that both NMo(Mes)3 and Mo(Mes)3
are rapidly formed during photolysis and recombine to form
(µ-N)[Mo(Mes)3]2 (Scheme 6). This latter mechanism agrees
well with our observed product ratio obtained by bulk photolysis.
The absence of a µ-nitrido product in our case is a consequence
of the large kinetic barrier toward forming a µ-nitrido when
bulky tert-butyl anilide ligands are present.75,143,144 In fact, we
have shown that 1 does not react with 3 to give a µ-nitride
product due to the steric constraints.10,143
Comparison Between Thermal and Photochemical NsN
Bond Cleavage. A key distinction between the thermal and
photochemical reactions of 2 is the observed product distribu-
tion. Upon photolysis 2 produces a mixture of 1 and 3, whereas
thermal NsN bond scission produces 3 as the sole product.
This change may reflect the preference for lower energy
pathways with respect to the ground-state for the thermal
reaction, in contrast to photochemical processes that involve
excited molecular states. For example, we have suggested that
rotation of an anilide ligand may be a feature of 2‡, providing
a means to lower the activation barrier for NsN thermal bond
cleavage. However, geometric effects that stabilize the thermal
(137) Closs, G. L.; Kaplan, L. R. J. Am. Chem. Soc. 1969, 91, 2168–2169.
(138) Fox, J. R.; Hammond, G. S. J. Am. Chem. Soc. 1964, 86, 4031–
4035.
(139) Bartlett, P. D.; Porter, N. A. J. Am. Chem. Soc. 1968, 90, 5317–
5318.
(140) Turro, N. J. Modern Molecular Photochemistry; University Science
Books: Sausalito, CA, 1991; p 544-550.
(141) Kasha, M. Discuss. Faraday Soc. 1950, 9, 14–19.
(142) Turro, N. J.; McVey, J.; Ramamurthy, V.; Lechtken, P. Angew.
Chem., Int. Ed. Engl. 1979, 18, 572–586.
(143) Johnson, M. J. A.; Lee, P. M.; Odom, A. L.; Davis, W. M.; Cummins,
C. C. Angew. Chem., Int. Ed. Engl. 1997, 36, 87–91.
(144) Stephens, F. H.; Johnson, M. J. A.; Cummins, C. C.; Kryatova, O. P.;
Kryatov, S. V.; Rybak-Akimova, E. V.; McDonough, J. E.; Hoff,
C. D. J. Am. Chem. Soc. 2005, 127, 15191–15200.
9
J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008 9403

A R T I C L E S
Curley et al.
-35 °C for an additional 7 d to afford each of the 3 additional
crops (Overall, 4.760 g, 3.730 mmol, 91.6%). Purple solids were
washed with MeCN, in which 2 is entirely insoluble, and dried to
constant mass prior to magnetic studies: µeff (SQUID) ) 2.42 µB.
UV-vis (Et2O, 20 °C): λ (ε) ) 299 (67600), 544 (41000) nm (M-1
cm-1).
transition state may not be required under higher-energy
photochemical conditions. Taking this view to an extreme, in
the photochemical reaction the structure of the N2 cleavage
transition state might be entirely altered, as it has been shown
that thermal and photochemical pathways can proceed through
transition states having distinct structures due to orbital-
symmetry effects.68,146
Preparation of (µ-N2)[Mo(N[t-Bu]Ar)3]2[B(ArF)4]2 (2[B(ArF)4]2)
by Oxidation of 2. At -116 °C, to a rapidly stirred, blue solution
of [Cp2Fe][B(ArF)4] (1.796 g, 1.712 mmol, 2 equiv) in 50 mL of
Et2O was added a solution of 2 (1.09 g, 0.856 mmol) in 40 mL of
Et2O over 2 min. The mixture became dark brown in color after
approximately 25 mL of the solution of 2 has been added. The
intensity of the brown color grew over the next 5 min. To the
mixture was added 40 mL of thawing n-pentane (-130 °C) to
quickly precipitate the product as a red powder. The product was
then isolated by filtration. The solids were washed with n-pentane
(30 mL × 4) and then Et2O (20 mL × 4). Drying solids under a
dynamic vacuum yields the product, 2[B(ArF)4]2 (2.530 g, 0.842
mmol, 98.4%), mp ) 135-140 °C (dec). 1H NMR (500 MHz, 20
°C, CD2Cl2): δ ) 7.70 (16 H, s, o-ArF), 7.54 (8 H, s, p-ArF), 7.17
(6 H, s, p-Ar), 6.71 (6 H, s, o-Ar), 4.50 (6 H, s, o-Ar), 2.35 (18 H,
s, Ar-Me), 2.17 (18 H, s, Ar-Me), 1.41 (54 H, s, t-Bu) ppm. 13C
NMR (125 MHz, 20 °C, CD2Cl2): δ ) 162.0 (q, ipso-ArF), 147.3
(s, ipso-Ar), 139.8 (s, Ar), 139.5 (s, Ar), 135.2 (s, o-ArF), 132.7 (s,
Ar), 129.5 (q, m-ArF), 127.9 (s, Ar), 126.3 (s, Ar), 125.2 (q, CF3),
117.9 (s, p-ArF), 74.0 (s, CMe3), 31.5 (s, NCCH3), 22.2 (s, Ar-
CH3), 21.4 (s, Ar-CH3) ppm. 19F NMR δ ) (376 MHz, 20 °C,
CD2Cl2): δ ) -63 ppm. UV-vis (CH2Cl2, 20 °C): λ (ε) ) 296
(79200), 382 (30600), 514 (39800) nm (M-1 cm-1). Anal. Calcd
for C136H132B2F48Mo2N8: C, 54.38; H, 4.43; N, 3.73; F, 30.36.
Found: C, 54.02; H, 4.34; N, 3.66; F, 30.22.
Conclusions
The X-ray crystal structure of 2 had previously eluded us.
Having obtained a suitable structure, we addressed a number
of outstanding questions concerning the structure and mag-
netism of this linear, bimetallic µ-N2 complex. Moreover,
we have suggested that ligand rotation about the MosNanilide
bond may play an important role in reducing the kinetic
barrier toward NsN bond cleavage by 2 under thermal
conditions.
To our knowledge, no other study has directly compared
X-ray crystal structures and Raman shifts of linear, bimetallic
µ-N2 complexes in three states of charge. Here, we have
shown that for a series of dinitrogen-derived µ-N2 complexes
the common markers for NsN bond activation indicate that
the NsN bond is activated further by removing 1e or 2e
from the MoNNMo core. The reduction in NsN bonding
that accompanies oxidation of 2 was rationalized by DFT
studies that show NsN bonding character in the LUMO of
2[B(ArF)4]2, corresponding to the occupied HOMO of 2.
Floriani previously reported photochemical cleavage of the
NsN bond in a linear, bimetallic µ-N2 complex to form a
µ-nitrido product.38 We have now demonstrated that this reaction
is not unique to Floriani’s system, and we have proposed a more
concise, revised mechanism on the basis of our isolation of a
terminal nitride product (Scheme 6). Further exploitation of this
strategy for NsN bond cleavage could make for interesting
future work. For instance, it is tempting to apply our finding of
photochemical NsN bond scission to a number of systems
(MoNNMo) that contain the requisite number of electrons for
6e cleavage but are thermally stable.57,147–150
The 15N labeled derivative was prepared in an analogous fashion
from 15N-labeled 2.10 For 15N-labeled 2[B(ArF)4]2, ESI-MS: m/z
(calcd, assignment) ) 640.3426 (640.3423, 22+), 1222.6277
(1222.6069, [2-(t-Bu)]+, 1279.6773 (1279.6907, 2+).
Preparation of (µ-N2)[Mo(N[t-Bu]Ar)3]2[B(ArF)4] (2[B(ArF)4])
by Comproportionation of 2 and 2[B(ArF)4]2. To a red solution of
2[B(ArF)4]2 (1.500 g, 0.500 mmol) in 25 mL of THF was added 2
(700 mg, 0.548 mmol, 1.1 equiv) as a purple solid at 20 °C. Upon
mixing the purple 2 with the opaque, red solution of 2[B(ArF)4]2,
a cranberry-colored solution formed. The vial containing 2 was
rinsed with 10 mL of THF which was added to the stirring solution.
Once the addition was completed, the solution was allowed to stir
for 1 min before the solvent was removed under a dynamic vacuum.
The remaining solids were then scraped onto a sintered glass frit
and thoroughly washed with toluene (20 mL × 5) to remove a
pale yellow solution (presumably containing 3) and then the
remaining solids were washed with n-pentane (20 mL × 4) before
the solids were dried under a dynamic vacuum. The cranberry-
colored solids of 2[B(ArF)4] were then collected (2.030 g, 0.948
Experimental Procedures
Details concerning general remarks for handling of air-
sensitive compounds, determination of molecular structure by
X-ray diffraction, computational details, and procedures for
photochemical reaction and physical measurements are contained
in the Supporting Information document. Given here are
procedures for the synthesis and isolation of 2, 2[B(ArF)4], and
2[B(ArF)4]2 alongside detailed characterization data for new
compounds.
Synthetic Procedures. Preparation of (µ-N2)[Mo(N[t-Bu]Ar)3]2.
(2).10,11,16 As an orange powder, Mo(N[t-Bu]Ar)3 (5.081 g, 8.143
mmol) was loaded into a 250 mL Erlenmeyer flask. This solid was
then dissolved in 125 mL of THF with rapid stirring. The stir bar
was removed from the flask, and the flask was capped with a rubber
septum. The flask was then stored at -35 °C for 10 d under N2.
The flask was removed from the freezer every 3 d to readmit N2
into the flask. Over the first 7 d, the color of the solution became
dark purple but few solids had precipitated. After 10 d, a large
amount of powder had precipitated and 3.145 g of 2 was isolated
by filtration of the mixture. The solids were washed with n-pentane
(10 mL × 2). The filtrate was concentrated to 50 mL and stored at
1
mmol, 95.0%), mp 140-142 °C (dec). H NMR (500 MHz, 20
°C, CDCl3): δ ) 7.70 (8 H, s, ∆ν1/2 ) 9.7 Hz, o-ArF), 7.51 (4 H,
s, ∆ν1/2 ) 5.2 Hz, p-ArF), 6.08 (27 H, s, ∆ν1/2 ) 75 Hz, t-Bu),
5.47 (6 H, s, ∆ν1/2 ) 72 Hz, o-Ar), 1.25 (3 H, s, ∆ν1/2 ) 6.8 Hz,
p-Ar), 1.74 (18 H, s, ∆ν1/2 ) 14.5 Hz, Ar-Me) ppm. µeff (SQUID)
) 1.96 µB. UV-vis (CH2Cl2, 20 °C): λ (ε) ) 295 (63800), 508
(38000) nm (M-1cm-1). Anal. Calcd for C104H120BF24Mo2N8: C,
58.35; H, 5.64; N, 5.23; F, 21.30. Found: C, 58.64; H, 5.66; N,
5.05; F, 20.91.
Acknowledgment. This paper is dedicated to the memory of
Professor Carlo Floriani. Emily Nytko-Lutz and Dr. Dino
Villagra´n are thanked for assistance with the acquisition of
(147) Ferguson, R.; Solari, E.; Floriani, C.; Osella, D.; Ravera, M.; Re,
N.; Chiesi-Villa, A.; Rizzoli, C. J. Am. Chem. Soc. 1997, 119, 10104–
10115.
(148) O’Donoghue, M. B.; Zanetti, N. C.; Davis, W. M.; Schrock, R. R.
J. Am. Chem. Soc. 1997, 119, 2753–2754.
(145) These two energies may be compared in cm-1 at 293 K: kBT is 207
cm-1 in comparison to 18382 cm-1 for 544 nm.
(146) Woodward, R. B.; Hoffmann, R. J. Am. Chem. Soc. 1965, 87, 395–
397.
9
9404 J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008

Features and Chemistry of Dinitrogen Cleavage
A R T I C L E S
magnetic data. We are indebted to Dr. Eckhard Bill and Prof.
Karsten Meyer for stimulating discussions and for suggesting a
suitable magnetic model for 2. Dr. Eckhard Bill is also thanked
for making available his wonderful julX program. We are
grateful to Professors Daniel G. Nocera, Andrei Tokmakoff, and
Troy Van Voorhis for helpful discussions. We acknowledge Tim
McClure for making available the Raman spectroscopy instru-
mentation in the Center for Materials Science and Engineering.
We gratefully thank the NSF for funding this work (Grant CHE-
719157).
Note Added after ASAP Publication. The version of this paper
published on June 25, 2008, contained an error on page 9401. The
N–N distance has been corrected to 1.138(6) Å. The version
published on July 16, 2008 has the correct information.
Supporting Information Available: Complete ref 124; full
experimental details and cif files. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA8002638
(150) Greco, G. E.; O’Donoghue, M. B.; Seidel, S. W.; Davis, W. M.;
(149) Greco, G. E.; Schrock, R. R. Inorg. Chem. 2001, 40, 3850–3860.
Schrock, R. R. Organometallics 2000, 19, 1132–1149.
9
J. AM. CHEM. SOC. VOL. 130, NO. 29, 2008 9405