A planar three-coordinate vanadium(II) complex and the study of terminal vanadium nitrides from N2: A kinetic or thermodynamic impediment to N-N bond cleavage?

Article abstract of DOI:10.1021/ja303360v

We report the first mononuclear three-coordinate vanadium(II) complex [(nacnac)V(ODiiP)] and its activation of N₂ to form an end-on bridging dinitrogen complex with a topologically linear V(III)N ₂V(III) core and where each vanadium center antiferromagnetically couples to give a ground state singlet with an accessible triplet state as inferred by HFEPR spectroscopy. In addition to investigating the conversion of N₂ to the terminal nitride (as well as the microscopic reverse process), we discuss its similarities and contrasts to the isovalent d ³ system, [Mo(N[^tBu]Ar)₃], and the S = 1 system [(Ar[^tBu]N)₃Mo]₂(μ₂- η¹:η¹-N₂).

Full text of DOI:10.1021/ja303360v

Article
pubs.acs.org/JACS
A Planar Three-Coordinate Vanadium(II) Complex and the Study of
Terminal Vanadium Nitrides from N2: A Kinetic or Thermodynamic
Impediment to N−N Bond Cleavage?
Ba L. Tran,† Balazs Pinter,† Adam J. Nichols,† Felicia T. Konopka,† Rick Thompson,† Chun-Hsing Chen,†
J. Krzystek,‡ Andrew Ozarowski,‡ Joshua Telser,§ Mu-Hyun Baik,† Karsten Meyer,∥
and Daniel J. Mindiola*,†
†Department of Chemistry and Molecular Structure Center, Indiana University, Bloomington, Indiana 47405, United States
‡National High Magnetic Field Laboratory, Florida State University, Tallahassee, Florida 32310, United States
§Department of Biological, Chemical and Physical Sciences, Roosevelt University, Chicago, Illinois 60605, United States
∥Department of Chemistry and Pharmacy, Inorganic Chemistry, University of Erlangen-Nurnberg, Egerlandstrasse, 91058 Erlangen,
̈
Germany
S
* Supporting Information
ABSTRACT: We report the first mononuclear three-coordinate
vanadium(II) complex [(nacnac)V(ODiiP)] and its activation of N2
to form an end-on bridging dinitrogen complex with a topologically
linear V(III)N2V(III) core and where each vanadium center
antiferromagnetically couples to give a ground state singlet with
an accessible triplet state as inferred by HFEPR spectroscopy. In
addition to investigating the conversion of N2 to the terminal
nitride (as well as the microscopic reverse process), we discuss its
similarities and contrasts to the isovalent d3 system, [Mo(N[tBu]-
Ar)3], and the S = 1 system [(Ar[tBu]N)3Mo]2(μ2-η1:η1-N2).
Cr(OtBu)3.11,12 However, unlike unsaturated V(II) complexes
generated by chemical and electrochemical reduction,
INTRODUCTION
■
Low-coordinate and electron-rich transition metal complexes
[(nacnac)V(Ntolyl2)] fails to bind N2, presumably due to the
have been long sought due to their potential utility in small
blockade imposed by the tethered arene motif (eq 1). This
molecule activation. One that illustrates the desired set of
suggests that the resting state of [(nacnac)V(Ntolyl2)], even in
properties is the three-coordinate complex [Mo(N[tBu]Ar)3]
the solution phase, is not in equilibrium with a three-coordinate
(Ar = 3,5-Me2C6H3) reported by Cummins in 1995, which was
shown to activate and cleave the triple bond of nitrogen, at 1
fragment (eq 1), or that the V(II)−arene interaction is more
bar and 25 °C.1−5 In a bimolecular fashion, this species can
form two equivalents of nitride, [NMo(N[tBu]Ar)3]. Based
on some diagonal relationship between Mo and V (their roles
in nitrogenase enzymes, and some similar structural features in
both low and high oxidation states), one may anticipate that the
isovalent and three-coordinate V(II) complex will be
comparably reactive. For example, three-coordinate V(III)
complexes have been shown to reversibly sequester atmos-
pheric nitrogen while V(II) systems can bind and break its
triple bond, as well as convert nitrogen to ammonia.6−9
Surprisingly, discrete examples of three-coordinate V(II)
complexes have not been reported, with the closest reference
favorable than V(II)−N2 binding. The lack of bona fide
examples of discrete three-coordinate V(II) complexes is
surprising since such species would likely engage in N2
activation and cleavage reactions.
In this work, we present the first example of a three-
coordinate vanadium(II) complex. Its existence has been
unequivocally established by a combination of characterization
techniques including solid state magnetic studies (SQUID),
high-frequency and -field EPR (HFEPR), and a single crystal
structural analysis (XRD). In addition, we demonstrate this
complex to be a reactive template, capable of activating N2 to
form a dinuclear complex displaying a singlet ground state
(with an accessible triplet state) with an end-on dinitrogen
ligand that bridges the two metal centers. Surprisingly, the
divanadium dinitrogen complex does not fragment into two
vanadium(V) nitride molecules despite the presence of a
being the masked system [(nacnac)V(Ntolyl2)] (nacnac−
=
[ArNC(CH3)]2CH, tolyl = 4-MeC6H4, Ar = 2,6-iPr2C6H3).10
Complex [(nacnac)V(Ntolyl2)] has a resting state displaying an
arene interaction with the tolyl moiety (eq 1) and the system
does behave as a three-coordinate V(II) fragment when treated
with reagents such as PhCCPh, N3Ad, S8, P4, pyridine N-oxide,
PhXXPh (X = S or Se), and the N-atom source N
Received: April 13, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1
(2.11 g, 3.09 mmol). H NMR (25 °C, 400 MHz, C6D6): δ 10.26
(Δν1/2 = 1069 Hz), 9.31 (Δν1/2 = 176 Hz), 8.58 (Δν1/2 = 1004 Hz),
6.95 (Δν1/2 = 18 Hz), 4.04 (Δν1/2 = 1360 Hz), 3.82 (Δν1/2 = 1448
Hz), 3.25 (Δν1/2 = 25 Hz), 2.97 (Δν1/2 = 901 Hz), 1.66 (Δν1/2 = 74
Hz), 1.22 (Δν1/2 = 31 Hz), 1.10 (Δν1/2 = 20 Hz), 0.86 (Δν1/2 = 24
Hz), −5.80 (Δν1/2 = 221 Hz). μeff = 2.78 μB (Evans, 25 °C). Anal.
Calcd for C41H58ClN2OV: C, 72.28; H, 8.58; N, 4.11. Found: C,
72.55; H, 8.50; N, 3.99.
sufficient number of electrons to populate all the π* and σ*
orbitals of N2. The chemistry of this species contrasts that of
isovalent [Mo(N[tBu]Ar)3]. Lastly, we present theoretical
studies that point to the conversion of N2 to two N3-(nitride)
to being both kinetically and thermodynamically prohibited.
EXPERIMENTAL DETAILS
■
Synthesis of (nacnac)V(ODiiP) (2). Under an argon atmosphere, a
250 mL flask was charged with 1 (640 mg, 0.94 mmol), toluene (50
mL), and 0.5% Na/Hg (27 mg, 1.17 mmol Na and 5.40 g Hg) at room
temperature. After stirring for 8 h, the reaction mixture turned from
dark green to brown. The supernatant was transferred via glass pipet
(leaving behind the amalgam) to a 250 mL round-bottom flask and all
volatiles removed. The crude product was extracted into n-pentane (75
mL), filtered through a medium porosity frit containing Celite to
remove salt residues and washed with an additional 25 mL of n-
pentane. Subsequently, the filtrate was reduced to 5 mL under vacuum,
and storage of the concentrated solution at −35 °C afforded a dark-
brown crystalline product, which was collected and dried under
General Considerations. Unless otherwise stated, all operations
were performed in an M. Braun Lab Master double-drybox under an
atmosphere of argon or nitrogen, or using high vacuum standard
Schlenk techniques under a nitrogen or argon atmosphere. Anhydrous
hexanes, n-pentane, toluene, and benzene were purchased from
Aldrich in sure-sealed reservoirs (18 L) and dried by passage through
two columns of activated alumina and a Q-5 column. Diethyl ether and
CH2Cl2 were dried by passage through a column of activated alumina.
THF was distilled, under nitrogen, from purple sodium benzophenone
ketyl and stored under sodium metal. Distilled THF was transferred
under vacuum into bombs before being pumped into a drybox. C6D6
was purchased from Cambridge Isotope Laboratory (CIL), degassed
and vacuum transferred to 4 Å molecular sieves. Celite, alumina, and 4
Å molecular sieves were activated under vacuum overnight at 200 °C.
All other chemicals were used as received from Sigma-Aldrich unless
otherwise stated and 98% atom enriched 15N2 was purchased from
1
reduced pressure. Yield = 38% (245.4 mg, 0.38 mmol). H NMR (25
°C, 400 MHz, C6D6): δ 20.81 (Δν1/2 = 76 Hz), 11.61 (Δν1/2 = 246
Hz), 8.95 (Δν1/2 = 316 Hz), 3.47 (Δν1/2 = 338 Hz), −4.78 (Δν1/2
=
8759 Hz), −14.37 (Δν1/2 = 279 Hz). μeff = 3.77 μB (SQUID, 25−300
K); 3.83 μB (Evans, 25 °C). Anal. Calcd for C41H58N2OV: C, 76.25; H,
9.05; N, 4.34. Found: C, 76.55; H, 9.10; N, 4.11.
1
Cambridge Isotope Laboratories. H, 13C, 51V, and 15N NMR spectra
1
were recorded on Varian 400 and 500 MHz NMR spectrometers. H
and 13C NMR are reported with reference to solvent resonances of
C6D6 at 7.16 and 128.0 ppm, respectively. 51V NMR spectral values are
reported with reference to neat VOCl3. 15N NMR spectral values are
reported with reference to neat MeNO2 (380.2 ppm). X-ray diffraction
data were collected on a SMART6000 (Bruker) system under a stream
of N2(g) at 150 K. Elemental analysis was performed at Indiana
University, Bloomington (CE Elantech Flash EA1112 Elemental
Analysis System). All UV−vis spectroscopy experiments were recorded
using a Perkin-Elmer Lambda 650 UV−vis Spectrometer under PC
control equipped with Perkin-Elmer software and Peltier heating
system. Infrared spectroscopy was performed on the Thermo Nicolet
6700 FT-IR equipped with software under PC control. Complex
[(nacnac)VCl2] [nacnac− = [ArNC(Me)]2CH, Ar = 2,6-iPr2C6H3] was
prepared according to published literature.13 NaODiiP was obtained
from HODiiP and NaN(SiMe3)2 in pentane or diethylether followed
by filtration of the solid, washing with ether, and then drying under
reduced pressure. Magnetic moments in solution were obtained by the
method of Evans.14 Warning: Na/Hg should be handled cautiously
due to its potentially pyrophoric nature. NaN3 should be handled
cautiously due to its explosive nature. It is recommended that NaN3 be
purified by stirring the solids in dry THF overnight, filtering and
washing the solid with copious amounts of THF, and then drying the
white solid overnight under reduced pressure.
Synthesis of (nacnac)VCl(ODiiP) (1). In a 100 mL round-bottom
flask, to a chilled toluene (50 mL) solution of (nacnac)VCl2 (2.20 g,
4.08 mmol) was added solid NaODiiP (857 mg, 4.28 mmol) via
spatula over 15 min. The yellow-green solution turned dark forest
green. After stirring for 2 h, the reaction mixture was filtered through a
medium porosity frit containing Celite to remove salt residues.
Subsequently, all volatiles were removed under reduced pressure and
addition of 50 mL of hexanes led to the precipitation of pure green
solids which were collected on a medium porosity frit and dried under
reduced vacuum. Reducing the volume of the filtrate to 20 mL and
storage at −37 °C afforded additional dark green crystalline product,
which was collected, and dried under reduced pressure. Yield = 76%
Synthesis of (nacnac)V(DMAP)(ODiiP) (3). In a 100 mL round-
bottom flask was charged 1 (260.0 mg, 0.38 mmol), DMAP (49.0 mg,
0.40 mmol), and toluene (15 mL) and cooled to −37 °C. To the
chilled round-bottom flask was added 0.5% Na/Hg (10.5 mg, 0.45
mmol Na and 2.09 g Hg). After stirring for 12 h, the reaction mixture
turned from dark green to a darker color. The supernatant was
transferred via glass pipet (leaving behind the amalgam) to a 100 mL
round-bottom flask and all volatiles removed under reduced pressure.
The crude product was extracted into diethyl ether (20 mL), filtered
through a medium porosity frit containing Celite to remove salt
residues. Subsequently, the ethereal solution (the filtrate) was reduced
to 4 mL under vacuum and storage at −37 °C afforded a dark
crystalline product, which was collected, and dried under reduced
1
pressure. Yield = 65% (189.7 mg, 0.25 mmol). H NMR (25 °C, 300
MHz, C6D6): δ 32.93 (Δν1/2 = 176 Hz), 26.00 (Δν1/2 = 200 Hz),
14.34 (Δν1/2 = 70 Hz), 8.47 (overlap), 7.80 (overlap), 5.64 (Δν1/2
=
1676 Hz), 4.06 (Δν1/2 = 1240 Hz), 2.74 (Δν1/2 = 719 Hz), 2.10
(Δν1/2 = 10 Hz), 0.87 (Δν1/2 = 12 Hz), −1.39 (Δν1/2 = 146 Hz),
−14.12 (Δν1/2 = 47 Hz). μeff = 3.79 μB (Evans, 25 °C). Anal. Calcd for
C48H68N4OV: C, 75.06; H, 8.92; N, 7.29. Found: C, 75.25; H, 8.85; N,
7.52.
Synthesis of [(nacnac)(ODiiP)V]2(μ2-η1:η1-N2) (4). In a 250 mL
round-bottom flask was charged 1 (2.00 g, 2.94 mmol), toluene (50
mL), and 0.5% Na/Hg (84.3 mg, 3.67 mmol Na/16.8 g Hg) at room
temperature. After stirring for 12 h, the reaction mixture turned from
dark green to brown. The supernatant was transferred via glass pipet
(leaving behind the amalgam) to a 250 mL round-bottom flask and all
volatiles removed. The crude product was extracted into n-pentane (75
mL) filtered through a medium porosity frit containing Celite to
remove salt residues and washed with additional 25 mL of n-pentane.
(Note: the insoluble brown solid atop the porous frit is an undesired
side product). Subsequently, the filtrate was reduced to 15 mL under
vacuum, and storage of the concentrated solution at −35 °C afforded a
dark-brown crystalline product, which was collected and dried under
reduced pressure. Yield = 60% (1.16 g, 0.88 mmol). 1H NMR (25 °C,
B
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
400 MHz, C6D6): δ 7.71 (Δν1/2 = 27 Hz), 7.48 (Δν1/2 = 29 Hz), 7.26
(Δν1/2 = 20 Hz), 7.05 (Δν1/2 = 26 Hz), 6.97 (Δν1/2 = 55 Hz), 6.86
(Δν1/2 = 24 Hz), 6.72 (Δν1/2 = 20 Hz), 6.39 (Δν1/2 = 20 Hz), 5.69
(Δν1/2 = 28 Hz), 3.27 (Δν1/2 = 122 Hz), 3.10 (Δν1/2 = 132 Hz), 1.77
(Δν1/2 = 537 Hz), 1.62 (Δν1/2 = 429 Hz), 1.42 (Δν1/2 = 29 Hz), 0.65
(Δν1/2 = 555 Hz). 51V NMR (25 °C, 131.5 MHz, C6D6): δ 167 (Δν1/2
= 500 Hz). Raman (785 nm excitation): 1374 (14N2), 1330 (15N2)
cm−1. Anal. Calcd for C82H116N6O2V2: C, 74.63; H, 8.86; N, 6.37.
Found: C, 74.55; H, 8.75; N, 6.50.
NMR spectroscopy. Complex 2 was confirmed as the product by
comparison to a 1H NMR (25 °C, C6D6) spectrum of authentic 2. For
the photolysis experiment, a J-Young NMR tube containing 4 in C6D6
was placed into a circulating water bath (25 °C) equipped with a
mercury lamp and monitored by 1H NMR spectroscopy, which
confirmed formation of 2, as in the thermolysis experiment.
Cyclic Voltammetry Measurement of [(nacnac)VCl(Ntol2)].
Cyclic voltammetry was performed in 0.3 M of predried and
recrystallized n-tetrabutylammonium hexafluorophosphate (TBAPF6)
in anhydrous THF solution. Commercial TBAPF6 was recrystallized
from boiling ethanol and cooling to 25 °C gave large colorless needles.
The crystals were collected on a frit and washed with cold ethanol, and
pure crystals of TBAPF6 were dried under reduced pressure at 25 °C
for 48 h. A platinum disk (2.0 mm diameter), a platinum wire, and
silver wire were employed as the working electrode, the auxiliary, and
the reference electrode, respectively. A one compartment cell was used
in the CV measurements. The electrochemical response was collected
with the assistance of an E2 Epsilon (BAS) autolab potentiostat/
galvanostat under control by BAS software. All of the potentials were
reported against the Fc+/Fc couple (0.0 V). The IR drop correction
was applied when significant resistance was noted. The spectrum was
recorded under an N2 atmosphere in the glovebox. In a typical
experiment, 20 mg of crystalline sample of 1 or 4 was dissolved in 5
mL of a TBAPF6 solution in THF at 25 °C.
Synthesis of 4-15N2. Under Ar atmosphere a 100 mL Schlenk flask
was charged with 1 (500.0 mg, 0.73 mmol), toluene (25 mL), and
0.5% Na/Hg (21.0 mg, 0.92 mmol Na and 4.22 g Hg) at 25 °C. The
Schlenk flask was removed from the glovebox and attached to a
Schlenk line. The flask underwent two cycles of freeze, pump, and
thaw. To the frozen solution was added 15N2 (1 atm) and the reaction
mixture was allowed to reach 25 °C. After stirring for 12 h, the
reaction mixture turned from dark green to yellow-brown. The flask
was taken back into the argon glovebox for workup. The supernatant
was transferred via glass pipet (leaving behind the amalgam) to a 100
mL round-bottom flask and all volatiles were removed under reduced
pressure. The crude product was extracted into n-pentane (20 mL),
filtered through a medium porosity frit containing Celite to remove
salt residues. Subsequently, the n-pentane solution was reduced in
volume under vacuum and storage of the concentrated solution at −37
°C afforded a crystalline product, which was collected, and dried under
reduced pressure. Yield = 63% (607 mg, 0.46 mmol). The dinitrogen
product was confirmed by 1H and 51V NMR spectral matching that of
4 as well as by IR spectroscopic data.
Magnetic Susceptibility Measurements. Magnetic data were
collected on a Quantumn Design MPMS-XL SQUID magnetometer.
All dc susceptibility measurements were obtained on three
independently synthesized microcrystalline powders placed in gelatin
capsules. Measurements were collected with a temperature range of 2−
300 K under a dc field of 1 T. All samples were corrected for
diamagnetic contributions using Pascal’s constants.15
Synthesis of (nacnac)VN(ODiiP) (5). In a 100 mL round-bottom
flask was charged 1 (1.00 g, 1.47 mmol), 75 mL of THF, and solid
NaN3 (573.3 mg, 8.82 mmol). After stirring for 6 h, all volatiles from
the reaction mixture were removed under reduced pressure and the
crude material was extracted with 50 mL of toluene and filtered
through a medium porosity frit containing Celite to remove excess
NaN3 and salt residues. Subsequently, the resulting filtrate was
transferred into a thick walled glass reaction vessel equipped with a
Telfon cap, a large stirring bar, and heated to 60 °C for 12 h in an oil
bath. Removal of all volatiles and addition of hexanes (10 mL) led to
the precipitation of yellow crystalline material that was collected on a
medium porosity frit, rinsed with hexanes (5 mL) and dried under
reduced vacuum to yield a clean product. Yield = 70% (679 mg, 1.02
HFEPR Data Acquisition and Sample Preparation. High-
frequency and -field electron paramagnetic resonance (HFEPR)
spectra were recorded using the Electron Magnetic Resonance
(EMR) Facility at the National High Magnetic Field Laboratory
(NHMFL, Tallahassee, FL). The spectrometer employs a Virginia
Diodes (Charlottesville, VA) source operating at a base frequency of
12−14 GHz and multiplied by a cascade of multipliers in conjunction
with a 15/17 T superconducting magnet.16 Detection was provided
with an InSb hot-electron bolometer (QMC Ltd., Cardiff, U.K.). The
magnetic field was modulated at 50 kHz. A Stanford Research Systems
SR830 lock-in amplifier converted the modulated signal to DC voltage.
Low temperature was provided by a continuous flow cryostat with a
temperature controller (both Oxford Instruments (Oxford, U.K.)).
Well-ground solid powder samples of complex 2 or 4 (typically, 30−50
mg) were loaded into sample holders under argon atmosphere. During
this process, and in subsequent, albeit rapid, transfer to the sample
holder followed by freezing in liquid nitrogen, it is likely that some air
oxidation of V(II) to paramagnetic V(III) and V(IV) species (and
perhaps even to diamagnetic V(V)) occurred.
Raman Spectroscopy Details. Raman data were recorded using a
Renishaw inVia Raman microscope fitted with a diode laser
(Renishaw) for 785 nm excitation. The instrument was interfaced
with a light microscope fitted with a 50× objective (Leica; numerical
aperture 0.5) focused into a cuvette containing the solution (Starna,
Spectrosil grade quartz). The spectrometer was wavelength-calibrated
using silicon (520.5 cm−1). Laser powers used in the experiments were
adjusted with neutral density filters to ensure sample integrity for the
duration of each experiment, usually 30 mW. No corrections for
spectrometer sensitivity were conducted. The sample was recorded
based on the following parameters: 100−3300 cm−1; 10 s/1000 cm−1;
15 mW; λex = 785 nm.
1
mmol). H NMR (25 °C, 400 MHz, d8-toluene): δ 7.18−7.6.95 (m,
9H, Ar-H), 5.66 (br, (Δν1/2 = 149 Hz, 1H, O{2,6-CH(CH3)2-C6H3),
4.98 (s, 1H, α-H), 4.04 (septet, 2H, CH(CH3)2), 2.90 (br, (Δν1/2
=
200 Hz, 1H, O{2,6-CH(CH3)2-C6H3), 2.67 (septet, 2H, CH(CH3)2),
1.58 (s, 6H, ArN(CH3)CCHC(CH3)NAr), 1.47 (d, JH−H = 6 Hz, 6H,
CH(CH3)2), 1.25 (br, Δν1/2 = 98 Hz, 12H, O{2,6-CH(CH3)2-C6H3)),
1.18 (d, JH−H = 7 Hz, 6H, CH(CH3)2), 1.15 (d, JH−H = 7 Hz, 6H,
1
CH(CH3)2), 1.02 (d, JH−H = 7 Hz, 6H, CH(CH3)2). H NMR (−40
°C, 400 MHz, d-toluene): δ 7.28 (d, JH−H = 7 Hz, 1H, Ar-H), 7.07−
7.00 (m, 7H, Ar-H), 6.95 (d, JH−H = 7 Hz, 1H, Ar-H), 5.75 (m, 1H,
CH(CH3)2), 4.79 (s, 1H, α-H), 4.10 (m 2H, CH(CH3)2), 2.92 (m,
1H, CH(CH3)2), 2.63 (m, 2H, CH(CH3)2), 1.51 (d, JH−H = 6 Hz,
CH(CH3)2), 1.43 (s, 6H, ArN(CH3)CCHC(CH3)NAr), 1.39 (d, JH−H
= 6 Hz, CH(CH3)2), 1.18 (m, 18H, CH(CH3)2), 1.06 (d, JH−H = 6 Hz,
CH(CH3)2). 13C{1H} NMR (25 °C, 100 MHz, C6D6): δ 170.1
(ArN(CH3)CCHC(CH3)NAr), 146.1 (Ar), 142.4 (Ar), 142.3 (Ar),
125.1 (Ar), 124.5 (Ar), 121.7 (Ar), 98.8 (ArN(CH3)CCHC(CH3)-
NAr), 29.3 (CH(CH3)2), 28.8 (CH(CH3)2), 26.3 (CH(CH3)2), 24.9
(CH(CH3)2), 24.7 (ArN(CH3)CCHC(CH3)NAr)), 24.2 (CH-
(CH3)2), 24.0 (CH(CH3)2). 51V NMR (25 °C, 131.5 MHz, C6D6):
δ −219.7 (Δν1/2 = 1308 Hz). FT−IR (KBr, Nujol, cm−1): 1024 (V
N), 998 (VN15). 15N (25 °C, 50.6 MHz, C6D6): δ 1058.7 (Δν1/2
=
Computational Details. All calculations were carried out using
DFT as implemented in the Jaguar 7.7 suite17 of ab initio quantum
chemistry programs. Geometry optimizations were performed with the
B3LYP18−21 functional and the 6-31G** basis set. Vanadium was
represented using the Los Alamos LACVP basis22,23 that includes
effective core potentials. The energies of the optimized structures were
reevaluated by additional single point calculations on each optimized
geometry using Dunning’s correlation consistent triple-ζ basis set22 cc-
36.0 Hz). Anal. Calcd for C41H58N3OV: C, 74.63; H, 8.86; N, 6.37.
Found: C, 74.65; H, 8.75; N, 6.15. The synthesis of the 15N
15
isotopomer, [(nacnac)V N(ODiiP)] was prepared following the
same procedure above by using 50% enriched Na15N3.
Thermolysis and Photolysis Experiments. For the thermolysis
experiment, a J-Young NMR tube containing 4 in C6D6 was placed
into a preheated oil bath at 90 °C. The reaction was monitored by 1H
C
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
pVTZ(-f) that includes a double set of polarization functions. For V,
we used a modified version of LACVP, designated as LACV3P, in
which the exponents were decontracted to match the effective core
potential with triple-ζ quality. Solvation energies were evaluated by a
self-consistent reaction field25−27 (SCRF) approach based on accurate
numerical solutions of the Poisson−Boltzmann equation. In the results
reported, solvation calculations were carried out with the 6-31G**/
LACVP basis at the optimized gas-phase geometry employing the
dielectric constant of ε = 2.284 for benzene. As is the case for all
continuum models, the solvation energies are subject to empirical
parametrization of the atomic radii that are used to generate the solute
surface. We employed28 the standard set of optimized radii in Jaguar
for H (1.150 Å), C (1.900 Å), N (1.600 Å) and V (1.572 Å). Analytical
vibrational frequencies within the harmonic approximation were
computed with the 6-31G**/LACVP basis to confirm proper
convergence to well-defined minima or saddle points on the potential
energy surface.
Scheme 1. Syntheses of All Described Vanadium Complexes
from the (nacnac)VCl(ODiiP) Precursor, 1
Convergence to plausible electronic states that correspond to
conceptually meaningful electronic configurations was monitored by
carefully observing the Mulliken spin densities and visualizing the
frontier molecular orbitals. When multiple minima were encountered,
we compared the total energies and chose the structure with the lowest
energy. Antiferromagnetically (AF) coupled states were modeled using
Noodleman’s broken symmetry (BS) formalism without spin
projection,29−31 as the spin projection corrections are uniformly
found to be negligibly small. The energy components have been
computed with the protocol below. The free energy in solution phase
G(Sol) has been calculated as follows:
G(Sol) = G(gas) + Gsolv
(1)
H(gas) = E(SCF) + Et + Er + Ev + kBT
(2)
G(gas) is the free energy in gas phase; Gsolv is the free energy of
solvation as computed using the continuum solvation model; H(gas) is
the enthalpy in gas phase; T is the temperature (298.00K); E(SCF) is
the self-consistent field energy, that is, “raw” electronic energy as
computed from the SCF procedure and the contributions of
translational, rotational and vibrational motions to the enthalpy
respectively are Et = 3/2RT, Er = 3/2RT (RT, for linear molecules)
excluded. Solid state SQUID magnetization measurements (at
1 T) of 1 confirmed a high-spin S = 1 vanadium(III) center
with an invariable average μeff = 2.81 μB over a temperature
range of 4−300 K (Figure S1). Simulation based on a best fit of
the data points in the range 2−200 K yielded the following
parameters: S = 1, giso = 1.95, D = 4.00 cm−1, with a
temperature independent paramagnetism (TIP) = 141 × 10−6
emu.42
and Ev = RΣKΘv,K(1/2 + 1/(eΘ /T−1)), where Θv,K = hvK/kB. Note
v,K
that by entropy here we refer specifically to the vibrational/rotational/
translational entropy of the solute(s); the entropy of the solvent is
incorporated implicitly in the continuum solvation model. To locate
transition states, the potential energy surface was first explored
approximately using the linear synchronous transit (LST)32 method,
followed by a quadratic synchronous transit (QST)33 search using the
LST geometry as an initial guess.
Cyclic voltammetry studies of 1 (0.3 M solution of TBAPF6
in THF) showed a reversible vanadium(III/II) cathodic wave at
+
−2.14 V versus the FeCp2/FeCp2 reference at 0.0 V (Figure
S2).42 Chemical reduction of 1 with 0.5% Na/Hg under argon
afforded a novel three-coordinate vanadium complex
[(nacnac)V(ODiiP)] (2) (Scheme 1) as a brown solid in
38% isolated yield via crystallization from a saturated n-pentane
solution cooled to −37 °C. The XRD of 2, shown in Figure 1,
reveals a rare example of a trigonal planar geometry with overall
C2v symmetry in which the mirror plane bisects the nacnac γ-
carbon, vanadium center, and aryloxide ligand. Metrical
parameters about the (nacnac)V(II) scaffold in 2 are V1−N1,
2.000(2) Å; V1−O1, 1.836(2) Å; N1′−V−N1′, 86.72(6)°,
which are similar to that observed in zwitterion complex
[(nacnac)V(η6-PhBPh3)].43 Moreover, the aryloxide ligand
coordinates to the vanadium in a linear fashion suggestive of
π donation into the unsaturated vanadium center. This type of
coordination behavior between T- versus Y-shaped geometries
has been predicted by DFT with (nacnac)M(II) systems (M =
Cr,44 Fe,45,46 Co,45 Ni,45,47 Cu48,49) with monoanionic σ and π
donors.
RESULTS AND DISCUSSION
■
Complexes of vanadium(II)34,35 and vanadium(III)36−39 have
been reported to bind and activate dinitrogen; however, only
transient vanadium(II) species generated under reducing
conditions have been implicated in the cleavage of
dinitrogen.35,40,41 Against this backdrop, we geared our efforts
toward the isolation of a well-defined vanadium(II) species to
investigate its electronic structure and interaction with
dinitrogen. Recently, we reported the isolation of a masked
three-coordinate V(II) fragment, [(nacnac)V(Ntol2)], but such
a complex failed to bind dinitrogen, likely due to the interaction
of tolyl arene with the coordinately unsaturated metal center as
noted in the solid state structural diagram (eq 1).10
Consequently, we sought to disfavor the metal-arene
interaction by replacing the ditolylanilide with 2,6-diisopropyl-
phenoxide (ODiiP). The reaction of [(nacnac)VCl2]13 and
NaODiiP produced [(nacnac)VCl(ODiiP)] (1) in 70% isolated
yield as a green solid (Scheme 1). Complex 1 can be prepared
in multigram quantities so as long as water and oxygen are
Solid-state SQUID magnetization measurements of 2 over
the temperature range of 2−300 K are consistent with an S =
3/2 species (Figure 2). The average magnetic moment of μeff
=
3.77 μB is similar to that of [(nacnac)V(Ntolyl2)] and is
D
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Molecular structures of [(nacnac)V(ODiiP)] (2), [(nacnac)V(DMAP)(ODiiP)] (3), and [(nacnac)VN(ODiiP)] (5) are shown with
thermal ellipsoids depicted as 50% probability. All hydrogen atoms have been omitted for clarity. Selected bonds (Å) and angles (°) for 2: V(1)−
N(1) = 2.000(2), V(1)−O(1) = 1.836(2); N(1)−V(1)−N(1′) = 86.72(6), N(1)−V(1)−O(1) = 136.56(6), V(1)−O(1)−C(ipso) = 177.46(5). For
3: V(1)−N(1) = 2.054(3), V(1)−N(2) = 2.058(3), V(1)−N(3) = 2.096(2), V(1)−O(1) = 1.925(2); N(1)−V(1)−N(2) = 87.79(10), N(1)−V(1)−
O(1) = 125.62(9), N(2)−(1)−O(1) = 131.97(10), N(1)−V(1)−N(3) = 107.41(10), N(3)−V(1)−O(1) = 95.33(9), V(1)−O(1)−C(ipso) =
166.66(19). For 5: V(1)−N(1) = 1.956(3), V(1)−N(2) = 1.947(4), V(1)−N(3) = 1.565(4), V(1)−O(1) = 1.814(3), N(1)−V(1)−N(2) =
97.85(15), N(1)−V(1)−N(3) = 110.58(17), N(1)−V(1)−O(1) = 109.46(14), N(3)−V(1)−O(1) = 114.76(18), V(1)−O(1)−C(ipso) = 150.4(3).
Figure 3. HFEPR spectrum of a polycrystalline sample of 2 recorded
at 10 K and 328.8 GHz (black trace) with the V(IV) impurity
resonance at 12 T left out. Arrows point at resonances attributed to a
superposition of V(II) turning points and those originating from a
V(III) (S = 1) impurity. Simulations for 2 are provided employing the
spin Hamiltonian parameters as in text (blue trace, D < 0; red trace, D
> 0). The cyan trace is a simulation of the V(III) impurity using: S = 1,
giso = 1.93, D = −2.62 cm−1, E = −0.186 cm−1.
Figure 2. Temperature-dependent DC susceptibility of a powder of 2
measured at 1 T (red circles) with a simulation (black line) in the
range 2−300 K using the following best-fitted parameters: S = 3/2, giso
= 1.94, D = 3.75 cm−1.
invariant over 25−300 K. This value is also comparable to the
magnetic moment of 3.83 μB at 25 °C obtained in solution by
the Evans method. In the solid state, the magnetic moment
sharply decreases below 20 K due to zero-field splitting (zfs).
The magnetization data were fitted using a standard spin
Hamiltonian for S = 3/2 and an axial zfs parameter of D = 3.75
cm−1 and giso = 1.94 (Figure 2). Multifrequency HFEPR
measurements of powdered samples at low temperatures were
also consistent with a mononuclear complex having a quartet
ground state (Figure 3). Although the spectra show trace
amounts of V(III) and V(IV) impurities, the dominant signals
3
fits: S = /2; |D| = 4.066(10) cm−1, |E| = 0.225(5) cm−1; g =
[1.960(4), 1.932(3), 1.979(9)]. The sign of D was proved
positive by simulating single-frequency spectra, as shown in
Figure 3. The spin Hamiltonian parameters stand in good
agreement with the SQUID data. Complex 2 has a larger D
value than any other reported V(II) complex, including
prototypical vanadocene systems51 and the masked three-
coordinate complex [(nacnac)V(Ntolyl2)],10 which we tenta-
tively attribute to its unusual trigonal planar environment.
Given the unsaturated nature of complex 2, it can readily
coordinate one equiv of DMAP (4-dimethylaminopyridine) to
form [(nacnac)V(DMAP)(ODiiP)] (3) quantitatively. Com-
plex 3 can be prepared more conveniently in 60% yield via Na/
Hg reduction of 1 in the presence of DMAP (Scheme 1).
Unlike [Mo(N[tBu]Ar)3], which readily cleaves N2 in minutes
3
are clearly indicative of an S = /2 ground state. Even if the
agreement between single-frequency spectra and their simu-
lations could be already deemed very good, the actual spin
Hamiltonian parameters were obtained by least-squares fitting
of the resonance field versus microwave frequency depend-
encies measured over a wide frequency range, as shown in the
two-dimensional map of EPR resonances in Figure 4.50 The
following spin Hamiltonian parameters were obtained from the
E
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
reduction and formation of a topologically linear V=N=N=V
core.
Raman vibrational spectra of 4 also corroborate substantial
NN bond reduction, since a sharp peak is observed at 1374
cm−1 which shifts to 1330 cm−1 when the 15N2 isotopologue
[(nacnac)(ODiiP)V]2(μ2-η1:η1-15N2) is prepared from 98%
enriched 15N2.39,52 Taken together, the Raman data and the
solid-state structure suggest a significant reduction of the N−N
bond order when compared to free dinitrogen, which is
characterized by νe/c = 2358 cm−1. Exposure of the 15N2
isotopologue of 4 to a dinitrogen atmosphere over 24 h at 25
°C shows no exchange with unlabeled N2. In fact, compound 4
is indefinitely stable at room temperature as a solid or in
solution, which contrasts the lability of N2 in unstable V(III)
dinitrogen systems that expel this ligand especially upon
dissolution.34,37 Alternatively, the reduction of 1 with 0.5% Na/
Hg in toluene for 6 h under a dinitrogen atmosphere also
rendered formation of 4 in 60% isolated yield and without need
to isolate the reactive complex 2 (Scheme 1). To our surprise,
SQUID magnetization measurements of 4 suggest that this
complex is diamagnetic, corroborated by NMR spectroscopic
experiments. The 1H NMR (d8-toluene) spectra collected
between 25 and −80 °C, show no changes in the relatively
sharp resonances that fall well within the diamagnetic region
(0−12 ppm). 51V NMR (C6D6, 25 °C) spectroscopy reveals an
intense resonance at 167.4 ppm (Δν1/2 = 500 Hz, Figure S8).42
Multiple attempts to observe an 15N NMR resonance for the
15N2 enriched isotopomer of 4 were unsuccessful presumably
due to broadening because of coupling to 51V (I = 7/2, 99.75%)
and 14N nuclei (I = 1, 99.636%). Therefore, our data alludes
that complex 4 has two vanadium(III) centers that are
antiferromagnetically (AF) coupled to yield an overall
diamagnetic (S = 0) system with S = 1, 2 excited spin states.
Solid state SQUID measurements of crystalline 4 were
inconclusive presumably due to its diamagnetic nature. As a
result, we relied on variable temperature HFEPR spectroscopy
(between 150 and 300 K) collected at a frequency of 211 GHz.
At temperatures above 200 K, a thermally activated triplet
system is observed (albeit in low concentrations, ∼1%), which
is most probably an excited spin state of the dinuclear complex
4 (Figure S9).42 Simulation of the spectra yielded small zfs
parameters: D = 0.75 cm−1, E = 0.056 cm−1, and the intensity of
the signal decayed as the temperature was lowered suggesting
complex 4 to have a singlet ground state. The quintet was never
observed under the above conditions. The cyclic voltammo-
gram of complex 4 displays a reversible cathodic wave at −1.37
V (top of Figure 6) with a peak-to-peak separation of 82 mV
implying minimal rearrangement within the structural frame-
work upon introduction of a single electron to form the radical
anion 4−. The linear dependence of the current versus the
square root of the scan rate confirms the reversibility of the
electron transfer (bottom of Figure 6). A second, but
irreversible cathodic wave is observed at −2.62 V (Figure
S5).42 Complex 4 also displays an irreversible anodic wave at
0.55 V and scanning faster did not deflect the irreversibility.
Surprisingly, complex 4 does not cleave the triple bond of
dinitrogen to afford the nitride complex [(nacnac)V
N(ODiiP)] (5), despite the system having six electrons
available from the two V(II) ions to populate all the π* and
σ* MO’s of N2. Monitoring the thermolysis of 4 at 90 °C in d6-
Figure 4. Two-dimensional map of EPR resonances observed in 2 at
10 K. Squares: experimental turning points; curves: simulations using
spin Hamiltonian parameters as in the text. Red lines: turning points
with B0||x; blue: B0||y; black: B0||z; green: off-axis turning points. The
dashed line represents the frequency at which the spectrum shown in
Figure 3 was acquired (328.8 GHz).
at room temperature when exposed to DMAP or other N-
heterocyclic donors, the adduct 3 is far less reactive than
precursor 2.4 The Evans susceptibility measurement of 3 is 3.79
μB, in good agreement with a high-spin vanadium(II) complex,
analogous to 2. An XRD structural analysis of 3 shows a very
similar N2VO scaffold (N2 = nacnac, O = ODiiP) when
compared to 2 (Figure 1). On the contrary, when exposed to
an atmosphere of N2, a benzene solution of complex 2
gradually and quantitatively engendered a new compound over
6 h based on 1H NMR spectroscopy (Scheme 1).
Recrystallization of the crude product from a saturated solution
of n-pentane at −37 °C yielded a dark brown microcrystalline
compound. XRD studies identified this new species to be the
end-on dinitrogen divanadium complex [(nacnac)(ODiiP)-
V]2(μ2-η1:η1-N2) (4) (Figure 5). Unfortunately, a satisfactory
refinement model could not be found for the diffraction data so
we refrain from commenting on any metrical parameters.
Nevertheless all salient features indicate some degree of N2
1
benzene by H NMR spectroscopy led to regeneration of 2
within 1 h. Moreover, photolysis experiments also liberated N2
albeit more sluggishly (3 h for complete displacement of N2)
Figure 5. Ball and stick representation of the disordered solid state
structure of 4 with Pr groups and H-atoms omitted for clarity.
i
F
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Me3C6H2) that we reported recently.10,53Although we can
prepare the nitride, attempts to convert 5 to 4 by chemical
reduction, photolysis, or thermolysis have been unsuccessful.
We turned to DFT calculations to address questions
concerning the inability of our vanadium system to cleave
dinitrogen, and to gain insight into the electronic structure of 4
to explain its unusual magnetism. Figure 7 illustrates the
Figure 7. Computed reaction profile from 2 to 4 and then to the
nitride 5. Computed structures of 2-N2, 4 and 4-TS are shown with
aryl groups and the nacnac β-methyl omitted for the purpose of clarity.
Distances are reported in Å and angles in degrees. Some computed
metrical parameters for 2-N2: V1−N1, 1.972; V1−N2, 1.973; V1−N5,
1.950; V1−O1, 1.842; N5−N6, 1.130; V1−N5−N6, 176.85. For 4:
V1−N1, 2.011; V1−N2, 2.024; V1−N5, 1.836; V1−O1, 1.836; N5−
N6, 1.204; V1−N5−N6, 175.06; V2−N6−N5, 175.24. For 4-TS: V1−
N1, 1.989; V1−N2, 1.988; V1−N5, 1.613; V1−O1, 1.818; N5−N6,
1.675; V1−N5−N6, 143.26; V2−N6−N5, 143.30.
Figure 6. (Top) Cyclic voltammogram of complex 4 in a 0.30 M of
solution of [NBu4][PF6] in anhydrous THF solution at a scan rate of
100 mV/s. (Bottom) Linear dependence of current versus scan rate1/2
.
1
and yielded no detectable vanadium nitride as evinced by H
NMR spectroscopy or high resolution mass spectrometry.
Using excess amounts of reducing agents also failed to generate
a nitride product from 4. To understand why complex 4 did not
fragment into the corresponding nitride complex 5, we
prepared the latter compound independently from 1 and
NaN3 in 70% yield (Scheme 1). Complex 5 has been fully
characterized by multinuclear (1H, 13C, 15N, and 51V) NMR,
FT-IR, and combustion analysis.42 Notably, complex 5 is a rare
example of a mononuclear, and neutral four-coordinate
vanadium nitride.10,53 The VN functionality for 5 has been
unequivocally identified by infrared spectroscopy with an
intense stretch at 1024 cm−1, which in the isotopically enriched
5-15N (from 50% enriched NaN3) shifts to 998 cm−1. A 15N
NMR (C6D6, 25 °C) spectrum of 5-15N confirmed a highly
deshielded terminal nitride moiety at 1059 ppm (Δν1/2 = 36
Hz, Figure S7).54 At 25 °C, the 1H NMR (d8-toluene)
spectrum of 5 revealed broad resonances resulting from rapid
rotation of the aryloxide ligand, but cooling of the solution to
−40 °C resolved all assignable resonances.42 The molecular
structure of 5, shown in Figure 1, clearly displays the presence
of a terminal nitride ligand at 1.565(4) Å, in which the bond
distance is comparable to the few known mononuclear
vanadium nitride complexes.55−59 The geometry of 5 is that
of a distorted tetrahedron as judged by the N3−V1−O1 angle
of 114.76(18)°. The V1−O1−C(ipso) angle is bent at
150.4(3)° suggestive of pseudo-σ ligation. Overall, metrical
parameters in the molecular structure of 5 resemble the nitride
derivative [(nacnac)VN(N[R]tol)] (R = tol or 2,4,6-
computed solution phase free energy reaction coordinate
involving complex 2 and N2, and its conversion to 5 via
intermediates [(nacnac)V(ODiiP)(N2)] (2-N2), and 4, as well
as the zigzag-like transition state structure 4-TS. All computed
i
structures use a truncated version where the Pr on the nacnac
ligand have been replaced with a methyl group. Metrical
parameters for the most salient features of these compounds are
listed in the caption of Figure 7 and the computed structures
for complexes 2 and 5 are in good agreement with the metrical
parameters obtained from the crystallographic data. Our
calculations suggest that complex 2 and N2 first form a
terminal end-on N2 adduct, 2-N2, with a minimal energy
difference of 2.0 kcal mol−1 to the reactants. Coordination of
the second equivalent of 2 to 2-N2 produces the divanadium-N2
bridged product 4 with a relative energy of −16.0 kcal mol−1.60
As noted from spectroscopic and magnetic studies, complex 2
has a quartet ground state that we calculate to be 24.8 kcal
mol−1 lower in energy than the doublet state. However, spin-
flip occurs during the association of 2 and N2 to form the
doublet species 2-N2.61 Up to this point, the reactivity of 2
mirrors that of complex [Mo(N[tBu]Ar)3].2,3,5 However,
Figure 7 shows disappointingly that conversion of complex 4
to 5 is associated with a prohibitively high barrier: 83 kcal
mol−1. Similarly, the reductive coupling of two equivalents of 5
into one 4, that is, the reverse reaction, is also associated with
an extremely high activation energy of 50.79 kcal mol−1. These
results are in agreement with the experimental observations that
4 does not convert into 5 and vice versa.62 As shown in Figure 7,
G
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 8. Molecular orbitals representing the main differences between the electronic structure of triplet Mo and AF coupled 4. Only the most
important π type orbitals are shown. Due to the small pyramidalization around the Mo centers, d orbitals are D3h/ML3 like in Mo whereas they
resemble the d orbitals are in a tetrahedral field in 4. Note the different axis systems, accordingly, to follow conventions.
the reactive nitride 5 lies higher in solution phase energy than 4
and the starting material 2: a striking contrast to the dominant
thermodynamic stability of [NMo(N[tBu]Ar)3] over [Mo-
(N[tBu]Ar)3].1−5 In accord with the structure of 4, the
optimized geometry also adopts a gauche conformation with a
dihedral angle of ∠O1−V1−V2−O2 = 69.5° and with only
slight bending of the V−N−N−V (∠V1−N5−N6 = 175.2°)
moiety (Figure 7).42 Our calculations also predict complex 4 to
possess two AF coupled metal centers that result in an overall
singlet ground state lying about 6 and 50 kcal mol−1 lower in
solution-phase free energy than the lowest triplet and closed
shell singlet states, respectively (Figure S11). As noted
previously, our HFEPR data at high temperature are in
agreement with complex 4 populating the triplet state. Despite
the metals being isovalent, the resting state electronic structure
of 4 differs profoundly from that of complex [(Ar[tBu]-
N)3Mo]2(μ2-η1:η1-N2) (Mo) (Figure 8).2,5 The latter has a
triplet ground state, in which four of the six π electrons are
paired and occupy the strongly delocalized π*(N2)-eg(Mo)
bonding combinations (HOMO-6 and HOMO-7), whereas the
remaining two electrons stay unpaired and are delocalized
between the two Mo centers2 (SOMO1 and SOMO2).23 In 4,
calculations of 4 show significant excess α-spin density of 2.1
and 0.4 on V1 and N6 matched by the same excess of β-spin
density on V2 and N5, respectively. These values indicate that 4
contains two AF coupled vanadium(III) centers and a doubly
2−
reduced N2 ligand bridging the vanadium centers. The
−
Wiberg bond index of 1.82 for N2 is consistent with NN−
type ligand, further supporting our assignment. Within the
Broken Symmetry Orbital approach, the different α and β
orbitals located on V2 and V1 illustrate the AF coupling of the
two V(III) centers (HOMO-2 and HOMO-6) through the
N22− bridge via a superexchange mechanism. In addition, due to
the gauche arrangement of the OAr groups and the difference in
the ligand strength of the nacnac and ODiiP ligands, the N2
based π* MOs are not degenerate in 4 (Figure 8), which
facilitates the pairing of the remaining two π electrons in the
lower lying π*(N2) (HOMO). The departure in the electronic
structures of 4 and Mo stems from the different interaction
between the π* orbitals of the N2 moiety and the d orbitals of
2
2
2
the metal centers: in 4, the dz and dx −y are involved, whereas
dxy and dxz are dominant in the case of the Mo complex (Figure
8). Recently, the coexistence of the closed shell singlet and AF
coupled singlet states were argued for the four π−electron
37
t
{(RO)3V}2(μ2−η1:η1-N2) (R = Bu2MeC).
2
2 2
however, the dz and dx ‑y orbitals lie significantly lower in
energy than the π* of N2 resulting in weak orbital interaction
and metal based bonding configurations. Mulliken spin density
The N2 fragment in 4 and Mo seem similar when judged by
their respective bond lengths of 1.20 and 1.19 Å and Wiberg
H
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 9. Simplified frontier molecular orbitals of complex 2 and [Mo(N[tBu]Ar)3]. The low symmetry in 2 renders all orbitals nondegenerate, but
for the purpose of simplification, we depict these at similar energy levels. The energy levels for the d-orbitals with purely σ-donors are shown in the
middle. Note that orbital energies for Mo and V are different due to effective charges of their cores.
bond indices of 1.82 and 1.85. However, breaking the N−N
bond is much more difficult in 4 than in Mo with the respective
computed barriers, ΔG⧧ = 83.0 kcal mol−1and ΔG⧧ = 20.8 kcal
mol−1. The prerequisite for the N−N bond breaking is that the
N2 based antibonding orbitals are appreciably filled: two
π*(N2) orbitals as well as the σ*(N2) orbital. To utilize the
available six π electrons for such purposes, these systems have
to undergo geometrical alterations that allow the transition
between π and σ subspaces. The transition state 4-TS must
adopt a zigzag geometry, as shown in Figure 7, which is a salient
feature already characterized for the analogous molybdenum
systems.2,63−68 The contrast in reactivity is rather surprising
since analysis of the frontier orbitals of 2 and [Mo(N[tBu]Ar)3]
reveal subtle differences in the bonding scheme. Figure 9
depicts a comparison of their frontier orbitals, paying close
attention to the π-component resulting from the contrast of the
nacnac ligand from the anilide. As shown in Figure 9, the π* set
in [Mo(N[tBu]Ar)3] is more destabilized than in 2, due to the
presence of two linear combinations resulting from the anilide
in-plane π-donor set. In contrast, complex 2 does not provide
such a combination, and the nacnac π-framework results in a
slight destabilization of the dxz and dyz orbitals, with the dxy and
bridging end-on dinitrogen complex 4, and gives access to the
terminal nitride complex 5, but unfortunately not from N2. Our
work also demonstrates that access to the nitride from
dinitrogen is both thermally and kinetically prohibited. This
difficulty is imparted both by the weaker d−π* interaction of
vanadium with nitrogen in 4 (as N2) and 5 (as N3−) as well as
the larger structural change needed to achieve the zigzag-like
transition state en route to formation of the nitride. The more
constrained geometry in 2, imparted by the less flexible,
chelating nacnac ligand, might be precluding such a distortion
to achieve a zigzag-like structure. We speculate that distortion
of the molecule (such as ligand rotation) is a necessity for
electron pairing in the intersystem crossing. As a result, more
rigid conformations prevent cooperative N2 splitting such as in
the case of [N3N]Mo (N3N3− = [RNCH2CH2]3N, R = Me3Si,
t-BuMe2Si, C6F5).69−71 We are presently investigating how to
manipulate and understand the latter process since it appears to
be a key step in dinitrogen cleavage. Being able to generate
vanadium nitrides from N2 will provide an opportunity to
exploit the nitride group since these species are not as
thermodynamically stable as the molybdenum counterparts.
2
2
ASSOCIATED CONTENT
* Supporting Information
X-ray crystallographic information, computational information,
magnetic data and spectral data are provided. This material is
available free of charge via the Internet at http://pubs.acs.org.
dx −y orbitals remaining essentially intact. Not surprisingly, the
■
2
dz orbital remains the same for both systems consistent with
S
the planar, three-coordinate environment being retained in
both V(III) and Mo(III). Although complexes 2 and [Mo(N-
[tBu]Ar)3] are not isolobal, they are not too dissimilar
considering the presence of the nacnac ligand framework.
The only main difference is that complex 2 is more electron
deficient as a result of lacking in-plane π-donor ligands (in the
xy axis).
AUTHOR INFORMATION
Corresponding Author
*mindiola@indiana.edu
■
Notes
CONCLUSIONS
The authors declare no competing financial interest.
■
Our work reports the first example of a discrete three-
coordinate V(II) complex 2, as well as some of its preliminary
reactivity. We show that this complex can serve as a template
for small molecule activation, including the formation of a
ACKNOWLEDGMENTS
We thank the Chemical Sciences, Geosciences and Biosciences
Division, Office of Basic Energy Science, Office of Science, U.S.
■
I
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(35) Vidyaratne, I.; Crewdson, P.; Lefebvre, E.; Gambarotta, S. Inorg.
Chem. 2007, 46, 8836.
(36) Buijink, J.-K. F.; Meetsma, A.; Teuben, J. H. Organometallics
Department of Energy (no. DE-FG02-07ER15893), as well as
the NHMFL, which is funded by the NSF through Cooperative
Agreement no. DMR-0654118, the State of Florida, and the
DOE for financial support. D.J.M. acknowledges support from
the Alexander von Humbold Stiftung. A.J.N. acknowledges
Indiana University STARS program for a summer fellowship.
K.M. thanks the Deutsche Forschungsgemeinschaft (DFG SFB
583) and the University of Erlangen for financial support. We
thank Dr. Edward H. Witlicki and the Flood group for
assistance with the Raman measurements.
1993, 12, 2004.
(37) Groysman, S.; Villagran, D.; Freedman, D. E.; Nocera, D. G.
Chem. Commun. 2011, 47, 10242.
(38) Song, J.-I.; Berno, P.; Gambarotta, S. J. Am. Chem. Soc. 1994,
116, 6927.
(39) Kilgore, U. J.; Sengelaub, C. A.; Pink, M.; Fout, A. R.; Mindiola,
D. J. Angew. Chem., Int. Ed. 2008, 47, 3769.
(40) Clentsmith, G. K. B.; Bates, V. M. E.; Hitchcock, P. B.; Cloke, F.
G. N. J. Am. Chem. Soc. 1999, 121, 10444.
(41) Bates, V. M. E.; Clentsmith, G. K. B.; Cloke, F. G. N.; Green, J.
C.; Jenkin, H. D. L. Chem. Commun. 2000, 927.
(42) See Supporting Information.
(43) Basuli, F.; Kilgore, U. J.; Hu, X.; Meyer, K.; Pink, M.; Huffman,
J. C.; Mindiola, D. J. Angew. Chem., Int. Ed. 2004, 43, 3156.
(44) Fan, H.; Adhikari, D.; Saleh, A. A.; Clark, R. L.; Zuno-Cruz, F. J.;
Cabrera, G. S.; Huffman, J. C.; Pink, M.; Mindiola, D. J.; Baik, M.-H. J.
Am. Chem. Soc. 2008, 130, 17351.
REFERENCES
■
(1) Laplaza, C. E.; Cummins, C. C. Science 1995, 268, 861.
(2) 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.
(3) Peters, J. C.; Cherry, J.-P. F.; Thomas, J. C.; Baraldo, L.;
Mindiola, D. J.; Davis, W. M.; Cummins, C. J. J. Am. Chem. Soc. 1999,
121, 10053.
(4) Tsai, Y.-C.; Cummins, C. C. Inorg. Chim. Acta 2003, 345, 63.
(5) Curley, J. J.; Cook, T. R.; Reece, S. Y.; Muller, P.; Cummins, C.
C. J. Am. Chem. Soc. 2008, 130, 9394.
(6) Ferguson, R.; Solari, E.; Floriani, C.; Osella, D.; Ravera, M.; Re,
N.; Chiesi-Villa, A.; Rizzoli, C. J. Am. Chem. Soc. 1997, 119, 10104.
(7) Shilov, A.; Denisov, N.; Efimov, O.; Shuvalov, N.; Shuvalova, N.;
Shilova, A. Nature 1971, 231, 460.
(45) Holland, P. L.; Cundari, T. R.; Perez, L. L.; Eckert, N. A.;
Lachicotte, R. J. J. Am. Chem. Soc. 2002, 124, 14416.
(46) Andres, H.; Bominaar, E. L.; Smith, J. M.; Eckert, N. A.;
Holland, P. L.; Munck, E. J. Am. Chem. Soc. 2002, 124, 3012.
̈
(47) Eckert, N. A.; Bones, E. M.; Lachicotte, R. J.; Holland, P. L.
Inorg. Chem. 2003, 42, 1720.
(48) Holland, P. L.; Tolman, W. B. J. Am. Chem. Soc. 1999, 121,
7270.
(49) Jazdzewski, B. A.; Holland, P. L.; Pink, M.; Victor G. Young, J.;
Spencer, D. J. E.; Tolman, W. B. Inorg. Chem. 2001, 2001, 6097.
(50) Krzystek, J.; Zvyagin, S. A.; Ozarowski, A.; Trofimenko, S.;
Telser, J. J. Magn. Reson. 2006, 178, 174.
(51) (a) Prins, R.; Biloen, P.; van Voorst, J. D. W. J. Chem. Phys.
1967, 46, 1216. (b) Jacobsen, C. J. H.; Pedersen, E.; Villadsen, J.;
Weihe, H. Inorg. Chem. 1993, 32, 1216. (c) Laurance, N.; Lambe. J.
Phys. Rev. 1963, 132, 1029. (d) Chan, K. K.; Shields, L. J. Chem. Soc. A
1970, 2700. (e) Girolami, G. S.; Wilkinson, G.; Galas, A. M. R.;
Thornton-Pett, M.; Hursthouse, M. B. J. Chem. Soc., Dalton Trans.
1985, 1339. (f) Scott, M. J.; Wilisch, W. C. A.; Armstrong, W. H. J.
Am. Chem. Soc. 1990, 112, 2430.
(8) Zones, S. I.; Vickery, T. M.; Palmer, J. G.; Schrauzer, G. N. J. Am.
Chem. Soc. 1976, 98, 1289.
(9) Zones, S. I.; Palmer, M. R.; Palmer, J. G.; Doemeny, J. M.;
Schrauzer, G. N. J. Am. Chem. Soc. 1978, 100, 2113.
(10) Tran, B. L.; Singhal, M.; Park, H.; Lam, O. P.; Pink, M.;
Krzystek, J.; Ozarowski, A.; Telser, J.; Meyer, K.; Mindiola, D. J.
Angew. Chem., Int. Ed. 2010, 49, 9871.
(11) Chiu, H.-T.; Chen, Y.-P.; Chuang, S.-H.; Jen, J.-S.; Lee, G.-H.;
Peng, S.-M. Chem. Commun. 1996, 139.
(12) Tran, B. L.; Chen, C.-H.; Mindiola, D. J. Inorg. Chim. Acta 2011,
369, 215.
(13) Budzelaar, P. H. M.; Oort, A. B. v.; Orpen, A. G. Eur. J. Inorg.
(52) For a recent discussion of dinuclear N2 complexes having a
similar degree of reduction, see: Chomitz, W. A.; Arnold, J. Chem.
Commun. 2007, 4797.
(53) Tran, B. L.; Pink, M.; Gao, X.; Park, H.; Mindiola, D. J. J. Am.
Chem. Soc. 2010, 132, 1458.
Chem. 1998, 1485−1494.
(14) Evans, D. F. J. Chem. Soc. 1959, 2003.
(15) Bain, G. A.; Berry, J. F. J. Chem. Educ. 2008, 85, 532.
(16) Hassan, A. K.; Pardi, L. A.; Krzystek, J.; Sienkiewicz, A.; Goy, P.;
Rohrer, M.; Brunel, L.-C. J. Magn. Reson. 2000, 142, 300.
(54) The highly downfield chemical shift has been correlated to the
large chemical shielding anisotropy affiliated with the metal−ligand
mutliple bond and its origin has been attributed to the hybridization of
a high lying filled σ-symmetry molecular orbital with a low-lying π*
molecular orbital in a 3-fold symmetric environment under an applied
magnetic field. (a) Greco, J. B.; Peters, J. C.; Baker, T. A.; Davis, W.
M.; Cummins, C. C.; Wu, G. J. Am. Chem. Soc. 2001, 123, 5003.
(b) Wu, G.; Rovnyak, D.; Johnson, M. J. A.; Zanetti, N. C.; Musaev, D.
G.; Morokuma, K.; Schrock, R. R.; Griffin, R. G.; Cummins, C. C. J.
Am. Chem. Soc. 1996, 118, 10654.
(17) In Jaguar 7.0; Schrodinger LLC: New York, 2007.
(18) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(19) Becke, A. D. Phys. Rev. A 1988, 38, 309858.
̈
(20) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.
(21) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(22) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284.
(23) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
(24) Dunning, T. H. J. Chem. Phys. 1989, 90, 1007.
(25) Marten, B.; Kim, K.; Cortis, C.; Friesner, R. A.; Murphy, R. B.;
Ringnalda, M. N.; Sitkoff, D.; Honig, B. J. Chem. Phys. 1996, 100.
(26) Friedrichs, M.; Zhou, R. H.; Edinger, S. R.; Friesner, R. A. J.
Chem. Phys. B 1999, 103, 3057.
(55) Willing, W.; Christophersen, R.; Muller, U.; Dehnicke, K. Z.
Anorg. Allg. Chem. 1987, 555, 16.
(27) Edinger, S. R.; Cortis, C.; Shenkin, P. S.; Friesner, R. A. J. Chem.
Phys. B 1997, 101, 1190.
(56) Critchlow, S. C.; Lerchen, M. E.; Smith, R. C.; Doherty, N. M. J.
Am. Chem. Soc. 1988, 110, 8071.
(57) Song, J.-I.; Gambarotta, S. Chem.Eur. J. 1996, 2, 1258.
(58) Henderson, R. A.; Janas, Z.; Jerzykiewicz, L. B.; Richards, R. L.;
Sobota, P. Inorg. Chim. Acta 1999, 285, 178.
(59) Johnson, C. E.; Kysor, E. A.; Findlater, M.; Jasinski, J. P.; Metell,
A. S.; Queen, J. W.; Abernethy, C. D. Dalton Trans. 2010, 39, 3482.
(60) Scans for the approach of N2 to 2 and for the association of 2-
N2 with 2 reflect that both processes are barrierless on the potential
energy surfaces (PES), in accord with earlier findings for isovalent
Mo(N[tBu]Ar)3 systems. Locating the transition states for the
formations of 2-N2 and 4 is, unfortunately, impossible with standard
(28) Rashin, A. A.; Honig, B. J. Chem. Phys. B 1985, 89, 5588.
(29) Noodleman, L. J. Chem. Phys. 1981, 74, 5737.
(30) Noodleman, L.; Davidson, E. R. J. Chem. Phys. 1986, 109, 131.
(31) Noodleman, L.; Lovell, T.; Han, W.-G.; Li, J.; Himo, F. Chem.
Rev. 2004, 104, 459.
(32) Halgren, T. A.; Lipscomb, W. N. Chem. Phys. Lett. 1977, 49,
225.
(33) Peng, C. Y.; Schlegel, H. B. Isr. J. Chem. 1993, 33, 449.
(34) Edema, J. J. H.; Stauthamer, W.; Bolhuis, F. V.; Gamarotta, S.;
Smeets, W. J. J.; Spek, A. L. Inorg. Chem. 1990, 29, 1302.
J
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
quantum mechanical methods, because entropy plays a pivotal and
nontrivial role in determining these transition-state energies (listed
with an * in Figure 4).
(61) No stable isomer was found on the quartet PES.
(62) For coupling of terminal metal nitrides to generate N2, see:
(a) Ware, D. C.; Taube, H. Inorg. Chem. 1991, 30, 4605. (b) Lam, H.
W.; Che, C. M.; Wong, K. Y. J. Chem. Soc., Dalton Trans. 1992, 1411.
(c) Demadis, K. D.; El-Samanody, E.-S.; Coia, G. M.; Meyer, T. J. J.
Am. Chem. Soc. 1999, 121, 535. (d) Demadis, K. D.; Meyer, T. J.;
White, P. S. Inorg. Chem. 1997, 36, 5678. (e) Newton, C.; Edwards, K.
D.; Ziller, J. W.; Doherty, N. M. Inorg. Chem. 1999, 38, 4032.
(f) Seymore, S. B.; Brown, S. N. Inorg. Chem. 2002, 41, 462.
(g) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 6252.
(h) Man, W.-L.; Chen, G.; Yiu, S.-M.; Shek, L.; Wong, W.-Y.; Wong,
W.-T.; Lau, T.-C. Dalton Trans 2010, 39, 11163. (i) Yiu, S.-M.; Lam,
W. W. Y.; Ho, C.-M.; Lau, T.-C. J. Am. Chem. Soc. 2007, 129, 803.
̌
(j) Schoffel, J.; Susnjar, N.a; Nuckel, S.; Sieh, D.; Burger, P. Eur. J.
̌
̈
̈
Inorg. Chem. 2010, 2010, 4911. (k) Chan, P.-H.; Chen, S.-P.; Yeung,
K.-W.; Yeung, K.-W.; Cheung, K.-C.; Hong, S.; Sun, S.-G.; Wong, K.-
Y. Electrochem. Commun. 2005, 7, 1244. (l) 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.
(63) Cui, Q.; Musaev, D. G.; Svensson, M.; Sieber, S.; Morokuma, K.
J. Am. Chem. Soc. 1995, 117, 12366.
(64) Christian, G. J.; Stranger, R.; Yates, B. F. Inorg. Chem. 2006, 45,
6851.
(65) Cavigliasso, G.; Wilson, L.; McAlpine, S.; Attar, M.; Stranger, R.;
Yates, B. F. Dalton Trans. 2010, 39, 4529.
(66) Christian, G. J.; Terrett, R. N. L.; Stranger, R.; Cavigliasso, G.;
Yates, B. F. Chem.Eur. J. 2009, 15, 11373.
(67) Brookes, N. J.; Graham, D. C.; Christian, G.; Stranger, R.; Yates,
B. F. J. Comput. Chem. 2009, 30, 2146.
(68) Christian, G.; Stranger, R.; Yates, B. F. Chem.Eur. J. 2009, 15,
646.
(69) Shih, K.-Y.; Schrock, R. R.; Kempe, R. J. Am. Chem. Soc. 1994,
116, 8804.
(70) Kol, M.; Schrock, R. R.; Kempe, R.; Davis, W. M. J. Am. Chem.
Soc. 1994, 116, 4382.
(71) O’Donoghue, M. B.; Davis, W. M.; Schrock, R. R. Inorg. Chem.
1998, 37, 5149.
K
dx.doi.org/10.1021/ja303360v | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX