are the kinetic products. It also seems likely that the first step in
the reaction is adduct formation to give a compound of the type
(RO)3W(µ-OR)3W(OR)2L where L = η1-PhNNPh or N2CH-
SiMe3. Then a redox reaction and terminal bridge exchange
leads to the observed bridged imido or hydrazonido complexes.
Further speculation at this time is not warranted.
Electronic structure calculations
Density functional theory (DFT) calculations with the
Gaussian 98 suite of programs13 utilized the B3LYP14–16
method and standard basis sets. C, H, O, and N were described
with the 6-31G* basis set (and 5 “pure” d functions), while W17
was represented by an SDD effective core potential. The geom-
etries were optimized under C1 symmetry starting from the
solid-state structure coordinates using the default optimization
criteria. Methoxide ligands were substituted for neopentoxides.
Stationary points were characterized as minima by vibrational
frequency calculations.
Experimental
All manipulations were carried out under an inert atmos-
phere of oxygen-free UHP-grade argon using standard Schlenk
techniques or under a dry and oxygen-free atmosphere of nitro-
gen in a Vacuum Atmospheres Co. Dry Lab System. Hexane
was degassed and distilled from potassium under nitrogen.
Toluene-d8 was degassed, stirred over sodium for 24 h and
vacuum transferred to an ampoule. Azobenzene and trimethyl-
silyldiazomethane were purchased form Aldrich and used as
received. NMR spectra were recorded on a 400 MHz Bruker
Crystallographic studies
t
t
W2(ꢀ-N2C(H)SiMe3)(ꢀ-OCH2 Bu)2(OCH2 Bu)6. The data
collection crystal was an amber colored, triangular plate.
Examination of the diffraction pattern on a Nonius Kappa
CCD diffractometer indicated
a triclinic crystal system.
1
DPX Avance400 spectrometer. All H NMR chemical shifts are
All work was done at 200 K using an Oxford Cryosystems
Cryostream Cooler. The data collection strategy was set up to
measure a hemisphere of reciprocal space with a redundancy
factor of 3, which means that 90% of the reflections were meas-
ured at least 3 times. A combination of ꢀ and ω scans with a
frame width of 1.0Њ was used. Data integration was done with
Denzo.18 Scaling and merging of the data were done with
Scalepack;18 application of an absorption correction is inherent
in this treatment and is reflected in the scale factor range of
9.14 to 12.56. Merging the data and averaging the symmetry
equivalent reflections results in an R(int) value of 0.051. The
1
reported in ppm relative to the H impurity in toluene-d8 at
t
δ 2.09. [W2(OCH2 Bu)8]n was prepared according to literature
procedures.3
t
Preparation of W2(ꢀ-NPh)(ꢀ-OCH2tBu)2(OCH2 Bu)6
t
To a 25 mL round-bottomed flask was added [W2(OCH2 Bu)8]n
(0.100 g, 0.094 mmol) and PhNNPh (9.5 mg, 0.094 mmol).
Hexanes (10 mL) were added to give a purple slurry and the
reaction was heated to 50 ЊC. After stirring for 24 h, the solu-
tion was filtered through a medium glass frit with a Celite pad.
Stripping the solvent gave a red solid identified as W2(µ-NPh)-
teXsan19 package indicated the space group to be P1 based on
the intensity statistics.
¯
t
t
(µ-OCH2 Bu)2(OCH2 Bu)6 in 95% crude yield. X-Ray quality
crystals were obtained upon crystallization from hexanes at
The positions of the two W atoms were determined by the
Patterson method in SHELXS-86.20 The rest of the non-hydro-
gen atoms were located by a combination of DIRDIF21 and
standard Fourier methods. Full-matrix least-squares refine-
ments based on F 2 were performed in SHELXL-93.22
For both structures, the methyl group hydrogen atoms were
added at calculated positions using a riding model with U(H) =
1.5 × Ueq(bonded C atom). For each methyl group, the torsion
angle which defines its orientation about the Si–C or C–C bond
was refined. The other hydrogen atoms were included in the
model at calculated positions using a riding model with U(H) =
1.2 × Ueq(bonded C atom). Neutral atom scattering factors
were used and include terms for anomalous dispersion.23
Crystallographic details of both structures are in Table 3.
1
Ϫ20 ЊC. H NMR (400 MHz, toluene-d8, 27 ЊC): δ 0.98 (s,
36 H), 1.02 (s, 18 H), 1.24 (s, 18 H), 4.22 (d, 4 H, JH–H = 10 Hz),
4.36 (d, 4 H, JH–H = 10 Hz), 4.63 (s, 4 H), 4.83 (s, 4 H), 7–8 (m,
Ph). 13C{1H} NMR (100 MHz, toluene-d8, 27 ЊC): δ 26.72 (s,
C(CH3)3), 27.70 (s, C(CH3)3), 28.00 (s, C(CH3)3), 33.85 (s,
CH2), 34.26 (s, CH2), 34.90 (s, CH2), 81.30 (s, C(CH3)3, 83.12
(s, C(CH3)3, 85.09 (s, C(CH3)3, 120–140 (Ph). Anal. Calc. for
W2O8NC46H93: C, 47.77; H, 8.10; N, 1.27 Found: C, 47.25; H,
8.01; N, 1.25%.
Preparation of W2(ꢀ-NNC(H)SiMe3)(ꢀ-OCH2tBu)2(OCH2tBu)6
t
To a 25 mL round-bottomed flask was added [W2(OCH2 Bu)8]n
(0.100 g, 0.094 mmol). Hexanes (10 mL) were added to give a
purple slurry and the solution was cooled to Ϫ30 ЊC. NNC(H)-
SiMe3 (37.5 µL, 0.094 mmol).was added by syringe and the reac-
tion was allowed to warm to room temperature. After stirring for
24 h the solution was filtered through a medium glass frit with a
Celite pad. Stripping the solvent gave a red solid identified as
W2(µ-NNC(H)SiMe3)(µ-OCH2 Bu)2(OCH2 Bu)6 in 95% crude
yield. X-Ray quality crystals were obtained upon crystallization
from hexanes at Ϫ20 ЊC. 1H NMR (400 MHz, toluene-d8, 27 ЊC):
δ 0.33 (s, 9 H), 0.99 (s, 36 H), 1.02 (s, 18 H), 1.21 (s, 18 H), 4.38 (d,
t
W2(ꢀ-NPh)(ꢀ-OCH2tBu)2(OCH2 Bu)6ؒ½hexane. The data
collection crystal was an orange-red rectangular rod. Examin-
ation of the diffraction pattern on a Nonius Kappa CCD dif-
fractometer indicated a triclinic crystal system. All work was
done at 200 K using an Oxford Cryosystems Cryostream Co-
oler. The data collection strategy was set up to measure a hemi-
sphere of reciprocal space with a redundancy factor of 2.7,
which means that 90% of the reflections were measured at least
2.7 times. A combination of ꢀ and ω scans with a frame width
of 1.0Њ was used. Data integration was done with Denzo.18 Scal-
ing and merging of the data were done with Scalepack;18 appli-
cation of an absorption correction is inherent in this treatment
and is reflected in the scale factor range of 9.41–12.39. Merging
the data and averaging the symmetry equivalent reflections
resulted in an R(int) value of 0.046. The teXsan19 package indi-
t
t
4 H, JH–H = 11 Hz), 4.52 (s, 4 H), 4.71 (s, 4 H), 4.82 (d, 4 H, JH–H
=
11 Hz), 8.54 (s, 1 H). 13C{1H} NMR (100 MHz, toluene-d8,
27 ЊC): δ 0.00 (s, SiMe3), 28.69 (s, C(CH3)3), 29.47 (s, C(CH3)3),
29.72 (s, C(CH3)3), 35.67 (s, CH2), 36.14 (s, CH2), 36.50 (s, CH2),
81.05 (s, C(CH3)3, 85.38 (s, C(CH3)3, 88.56 (s, C(CH3)3, 166.50 (s,
NNC(H)SiMe3). IR (νmax/cmϪ1): 1477 (s), 1463 (vs), 1419(vw),
1392 (s), 1377 (s), 1363 (s), 1372 (s), 1301 (vw), 1289 (vw), 1260
(w), 1250 (m), 1217 (w), 1058 (vs), 1031 (s), 1022 (vs), 1005 (vs),
994 (s), 932 (m), 905 (w), 864 (m), 845 (m), 805 (w), 754 (m), 721
(w), 690 (s), 680 (s), 658 (s), 639 (m), 600 (m), 577 (w), 492 (vw),
458 (m), 436 (vw), 403 (m), 375 (s). UV-vis (THF, λmax/nm): 475
(ε/dm3 molϪ1 cmϪ1 = 300), 268 (37,000), 255 (74,000), 249 (64,000)
244 (52,000), 239 (40,000), 234 (30,000). Anal. Calc. for W2N2O8-
C44H98: C, 44.82; H, 8.39; N, 2.38. Found: C, 44.72; H, 8.39; N,
2.32%.
¯
cated the space group to be P1 based on the intensity statistics.
The structure was solved by the Patterson method in
SHELXS-86.20 Besides the W complex, the asymmetric unit
contains a hexane molecule positioned on an inversion
center. Full-matrix least-squares refinements based on F 2 were
performed in SHELXL-93.22 The hexane molecule was refined
isotropically, and only the hydrogen atoms bonded to the CH2
groups were included in the model.
CCDC reference numbers 192252 and 192253.
D a l t o n T r a n s . , 2 0 0 3 , 3 2 0 5 – 3 2 1 0
3209