Reaction of Ta(NMe2)5 with O2: Formation of aminoxy and unusual (aminomethyl)amide oxo complexes and theoretical studies of the mechanistic pathways

Article abstract of DOI:10.1021/ja075076a

Reaction of d⁰ Ta(NMe₂)₅ (1) with O ₂ yields two aminoxy complexes (Me₂N)_nTa(η ²-ONMe₂)_5-n (n = 4, 2; 3, 3) as well as (Me₂N)₄Ta₂[η²-N(Me)CH ₂NMe₂]₂(μ-O)₂ (4) and (Me ₂N)₆Ta₃[η²-N(Me)CH ₂NMe₂]_2- (η²-ONMe ₂)(μ-O)₃ (5) containing novel chelating (aminomethyl)amide - N(Me)CH₂NMe₂ ligands. The crystal structures of 2-5 have been determined by X-ray crystallography. (Me ₂N)₄Ta(η²-ONMe₂) (2) converts to (Me₂N)₃Ta(η²-ONMe₂)₂ (3) in its reaction with O₂. In addition, the reaction of Ta(NMe₂)₅ with 3 gives 2 only at elevated temperatures. Density functional theory (DFT) calculations have been used to investigate the mechanistic pathways in the reactions of Ta(NMe₂)₅ (1) with triplet O₂. Monomeric reaction pathways in the formation of 2-5 are proposed. A key step is the oxygen insertion into a Ta-N bond in 1 through an intersystem conversion from triplet to singlet energy surface to give an active peroxide complex (Me₂N)_4- Ta(η²-O-O- NMe₂) (A4). The DFT studies indicate that the peroxide ligand plays an important role, including oxidizing an amide to an imine ligand through the abstraction of a hydride. Insertion of Me-N=CH₂ into a Ta-amide bond yields the unusual -N(Me)CH₂NMe₂ ligands.

Full text of DOI:10.1021/ja075076a

Published on Web 10/30/2007
Reaction of Ta(NMe2)5 with O2: Formation of Aminoxy and
Unusual (Aminomethyl)amide Oxo Complexes and Theoretical
Studies of the Mechanistic Pathways
Shu-Jian Chen,† Xin-Hao Zhang,‡ Xianghua Yu,† He Qiu,† Glenn P. A. Yap,§
Ilia A. Guzei,| Zhenyang Lin,*,‡ Yun-Dong Wu,*,‡ and Zi-Ling Xue*,†
Contribution from the Department of Chemistry, UniVersity of Tennessee, KnoxVille, Tennessee
37996, Department of Chemistry, Hong Kong UniVersity of Science and Technology, Hong
Kong, China, Department of Chemistry and Biochemistry, UniVersity of Delaware, Newark,
Delaware 19716, and Department of Chemistry, UniVersity of Wisconsin,
Madison, Wisconsin 53706
Received July 9, 2007; E-mail: chzlin@ust.hk; chydwu@ust.hk; xue@utk.edu
Abstract: Reaction of d0 Ta(NMe2)5 (1) with O2 yields two aminoxy complexes (Me2N)nTa(η2-ONMe2)5-n
(n ) 4, 2; 3, 3) as well as (Me2N)4Ta2[η2-N(Me)CH2NMe2]2(µ-O)2 (4) and (Me2N)6Ta3[η2-N(Me)CH2NMe2]2-
(η2-ONMe2)(µ-O)3 (5) containing novel chelating (aminomethyl)amide-N(Me)CH2NMe2 ligands. The crystal
structures of 2-5 have been determined by X-ray crystallography. (Me2N)4Ta(η2-ONMe2) (2) converts to
(Me2N)3Ta(η2-ONMe2)2 (3) in its reaction with O2. In addition, the reaction of Ta(NMe2)5 with 3 gives 2 only
at elevated temperatures. Density functional theory (DFT) calculations have been used to investigate the
mechanistic pathways in the reactions of Ta(NMe2)5 (1) with triplet O2. Monomeric reaction pathways in
the formation of 2-5 are proposed. A key step is the oxygen insertion into a Ta-N bond in 1 through an
intersystem conversion from triplet to singlet energy surface to give an active peroxide complex (Me2N)4-
Ta(η2-O-O-NMe2) (A4). The DFT studies indicate that the peroxide ligand plays an important role, including
oxidizing an amide to an imine ligand through the abstraction of a hydride. Insertion of Me-NdCH2 into a
Ta-amide bond yields the unusual -N(Me)CH2NMe2 ligands.
Reactions of O2 with metal complexes are important to our
understanding of fundamental biological processes and design
of oxidation catalysts.1,2 Most studies of these reactions involve
dn complexes, and oxidation of the metals by O2 is often a key
step. There are, however, fewer studies of reactions of d0
complexes with O2.3,4 In the reactions of O2 with d0 alkyl and
silyl complexes, O insertion into M-R3 and M-Si4a bonds has
been reported. Preparation of d0 Mo(NMe2)6, W(NMe2)6, and
WMe6 requires adventitious O2.5 Reactions of d0 complexes with
O2 have recently been used to yield metal oxides as microelec-
tronic gate-insulating materials.6-8 With the rapid reduction of
the thickness of the gate-insulating layer to less than 2 nm in
integrated transistors, SiO2 with a dielectric constant (κ) of 3.9
is not adequate, leading to a large leakage current.6 Metal oxides
such as Ta2O5 (κ ) 26) with large dielectric constants have
been actively studied to replace SiO2 in new generations of
microelectronic devices.6-8 Reactions of O2 with, e.g., d0 Ta-
(NR2)5 have been used to yield Ta2O5 films as gate-insulating
materials.8
† University of Tennessee.
‡ Hong Kong University of Science and Technology.
§ University of Delaware.
| University of Wisconsin.
(1) See, e.g.: (a) Kopp, D. A.; Lippard, S. J. Curr. Opin. Chem. Biol. 2002,
6, 568. (b) Theopold, K. H.; Reinaud, O. M.; Blanchard, S.; Leelasube-
haroen, S.; Hess, A.; Thyagarajan, S. ACS Symp. Ser. 2002, 823, 75. (c)
Que, L.; Jr.; Tolman, W. B. Angew. Chem., Int. Ed. 2002, 41, 1114. (d)
Stahl, S. S. Science 2005, 309, 1824. (e) Keith, J. M.; Muller, R. P.; Kemp,
R. A.; Goldberg, K. I.; Goddard, W. A., III; Oxgaard, J. Inorg. Chem.
2006, 45, 9631. (f) Popp, B. V.; Stahl, S. S. J. Am. Chem. Soc. 2007, 129,
4410 and references therein.
We recently observed that reactions of d0 (Me2N)4Ta-SiR3
and M(NMe2)4 (M ) Zr, Hf) with O2 yielded (Me2N)3Ta(η2-
ONMe2)(OSiR3)9 and M3(NMe2)6(µ-NMe2)3(µ3-O)(µ3-ONMe2),10
respectively. The nature of the reactions between d0 amides and
(2) Sima´ndi, L. I., Ed. AdVances in Catalytic ActiVation of Dioxygen by Metal
(5) (a) Bradley, D. C.; Chisholm, M. H.; Extine, M. W. Inorg. Chem. 1977,
16, 1791. (b) Chisholm, M. H.; Hammond, C. E.; Huffman, J. C. J. Chem.
Soc., Chem. Commun. 1987, 1423. (c) Galyor, L.; Mertis, K.; Wilkinson,
G. J. Organomet. Chem. 1975, 85, C37.
(6) (a) Wallace, R. M.; Wilk, G. D. Crit. ReV. Solid State Mater. Sci. 2003,
28, 231. (b) Smith, R. C.; Ma, T.; Hoilien, N.; Tsung, L. Y.; Bevan, M. J.;
Colombo, L.; Roberts, J.; Campbell, S. A.; Gladfelter, W. L. AdV. Mater.
Opt. Electron. 2000, 10, 105.
(7) (a) Bastianini, A.; Battiston, G. A.; Gerbasi, R.; Porchia, M.; Daolio, S. J.
Phys. IV 1995, 5(C5), 525. (b) Ohshita, Y.; Ogura, A.; Hoshino, A.; Hiiro,
S.; Machida, H. J. Cryst. Growth 2001, 233, 292.
(8) Son, K.-A.; Mao, A. Y.; Sun, Y.-M.; Kim, B. Y.; Liu, F.; Kamath, A.;
White, J. M.; Kwong, D. L.; Roberts, D. A.; Vrtis, R. N. Appl. Phys. Lett.
1998, 72, 1187.
Complexes; Kluwer: Boston, 2003.
(3) (a) Labinger, J. A.; Hart, D. W.; Seibert, W. E.; Schwartz, J. J. Am. Chem.
Soc. 1975, 97, 3851. (b) Lubben, T. V.; Wolczanski, P. T. J. Am. Chem.
Soc. 1987, 109, 424. (c) Brindley, P. B.; Scotton, M. J. J. Chem. Soc.,
Perkin Trans. 1981, 419. (d) Wang, R.; Folting, K.; Huffman, J. C.;
Chamberlain, L. R.; Rothwell, I. P. Inorg. Chem. Acta 1986, 120, 81. (e)
Gibson, V. C.; Redshaw, C.; Walker, G. L. P.; Howard, J. A. K.; Hoy, V.
J.; Cole, J. M.; Kuzmina, L. G.; De Silva, D. S. J. Chem. Soc., Dalton
Trans. 1999, 161. (f) Van Asselt, A.; Trimmer, M. S.; Healing, L. M.;
Bercaw, J. E. J. Am. Chem. Soc. 1988, 110, 8254. (g) Brindley, P. B.;
Hodgson, J. C. J. Organomet. Chem. 1974, 65, 57. (h) Gibson, T.
Organometallics 1987, 6, 918. (i) Kim, S.-J.; Jung, I. N.; Yoo, B. R.; Cho,
S.; Ko, J.; Kim, S. H.; Kang, S. O. Organometallics 2001, 20, 1501.
(4) (a) Tilley, T. D. Organometallics 1985, 4, 1452. (b) A bisperoxo complex
was observed in the reaction of a Zr(IV) amide complex with O2. Stanciu,
C.; Jones, M. E.; Fanwick, P. E.; Abu-Omar, M. M. J. Am. Chem. Soc.
2007, 129, 12400.
(9) Wu, Z.-Z.; Cai, H.; Yu, X.-H.; Blanton, J. R.; Diminnie, J. B.; Pan, H.-J.;
Xue, Z.-L. Organometallics 2002, 21, 3973.
(10) Wang, R.; Zhang, X.-H.; Chen, S.-J.; Yu, X.; Wang, C.-S.; Beach, D. B.;
Wu, Y.-D.; Xue, Z.-L. J. Am. Chem. Soc. 2005, 127, 5204.
9
14408
J. AM. CHEM. SOC. 2007, 129, 14408-14421
10.1021/ja075076a CCC: $37.00 © 2007 American Chemical Society

Reaction of Ta(NMe2)5 with O2
A R T I C L E S
Scheme 1. Formation of 2-5 from the Reaction of Ta(NMe2)5 (1)
with O2
lization of the solid gave colorless crystals of 2. Dissolving the
pale-yellow residue of the sublimation in n-pentane and
subsequent crystallization yielded colorless crystals of 3.
The molecular structures of 2-5 are shown in Figure 1.
Crystallographic data for 2-5 are given in Table 1.14 Selected
bond lengths and angles are listed in Table 2.
(Me2N)4Ta(η2-ONMe2) (2) was also prepared from the
reaction of (Me2N)4TaCl with 1 equiv of LiONMe2 (eq 1). Its
reactivity toward O2 is discussed below. In the molecular
structure of 2, one O inserted into a Ta-N bond to form a
monomer, yielded an O-Ta covalent bond and a N-Ta dative
bond in a distorted octahedral geometry with a nearly linear
1
N(3)-Ta-N(4) angle of 172.9(4)°. Only one peak in the H
(3.25 ppm) or the 13C (48.80 ppm) NMR spectrum of 2 is
observed for the four -NMe2 ligands, suggesting that 2
undergoes a fast ligand-site exchange.
O2 is still largely unknown.6-8 We have found that the reaction
of Ta(NMe2)5 (1) with O2 gives unusual oxo complexes
(Me2N)4Ta2[η2-N(Me)CH2NMe2]2(µ-O)2 (4) and (Me2N)6Ta3-
[η2-N(Me)CH2NMe2]2(η2-ONMe2)(µ-O)3 (5) containing novel
chelating -N(Me)CH2NMe2 ligands11 as well as aminoxy
complexes (Me2N)4Ta(η2-ONMe2) (2) and (Me2N)3Ta(η2-
ONMe2)2 (3) (Scheme 1). This reaction is a rare case of the
formation of oxo ligands (in 4 and 5) in the reactions of d0
complexes with O2.10,12 Density functional theory (DFT) studies
suggest monomeric reaction pathways in the formation of 2-5
through peroxide intermediates. The abstraction of a hydride
from a -NMe2 ligand by a peroxide ligand leads to the
formation of imine MeNdCH2, which then inserts into a Ta-
NMe2 bond to give the novel η2-N(Me)CH2NMe2 ligands. Our
experimental and theoretical studies are reported.
(Me2N)3Ta(η2-ONMe2)2 (3) is thermally stable. It was also
prepared from the reaction of (Me2N)3TaCl2 with 2 equiv of
LiONMe2 (eq 2). 3 shows a distorted pentagonal bipyramidal
geometry with two axial -NMe2 ligands and a nearly linear
N(1)-Ta-N(3) angle of 178.2(2)°. The equatorial -NMe2 and
two η2-ONMe2 ligands are coplanar15 with two dative TarN
bonds. The two O atoms face each other. The 1H and 13C NMR
spectra of 3 at 23 °C show three peaks (1H NMR: 3.41, 3.05,
and 2.94 ppm in a 1:1:1 ratio; 13C NMR: 48.98, 47.05, 46.11
ppm) for the equatorial and two axial -NMe2 ligands. This
observation suggests that (1) the two axial -NMe2 ligands do
not undergo a fast rotation at 23 °C, probably as a result of
M-N d-p π bonding in this otherwise electron-deficient
complex; (2) there is no axial-equatorial exchange among the
three -NMe2 ligands on the NMR time scale at 23 °C. Thus,
the two methyl groups in each axial amide are not equivalent
(-NMeAMeB). They are close to the equatorial -NMe2 and O
atoms, respectively. EXSY NMR studies revealed exchanges
among the amide ligands.14 Variable-temperature (VT) NMR
studies showed coalescences among the three -NMe2 ligands
in 3 at high temperatures. The EXSY and VT NMR spectra are
given in the Supporting Information. At 60 °C, the peaks at
2.94 and 3.05 ppm merge into one broad peak at 3.00 ppm with
a rate constant of 100 s-1 (∆G333Kq ) 17 kcal/mol), suggesting
that these two peaks are those of Mea and Meb groups in the
axial amide ligands. At 70 °C this new peak further coalesced
with that of the equatorial -NMe2 with a rate constant of 163
Results and Discussion
Formation and Characterization of 2-5. Reaction of O2
(0.5 equiv, 650 mmHg) with Ta(NMe2)5 (1) in toluene gave 2,
3, and 4 in 46-52%, 8-9%,13 and 8-9% yields (based on O2),
respectively, by 1H NMR. Crystallization of the mixture,
discussed below, gave 2, 3, and 4 in 27%, 12%,13 and 7%
isolated yields. When 1.0 equiv of O2 (570 mmHg) was used
in an NMR tube, the yield of 2 dropped to 26-27%, while that
of 3 increased to 34-37%. The yield of 4 at 9-10% was similar
to 8-9% from 0.5 equiv of O2.
Initial crystallization of the reaction mixture at -32 °C
yielded colorless crystals of dimeric 4. A few crystals of 5 were
sometimes found along with those of 4, and X-ray crystal-
lography confirmed the identity of the crystals of 5 isolated from
1
the mixture. H NMR spectrum of the mixture of the crystals
in benzene-d6, however, failed to reveal the presence of 5,
suggesting that the yield of 5 in the reaction is perhaps below
the detection limit of 1H NMR. 5 was prepared separately from
the reaction of excess O2 (4 equiv) with Ta(NMe2)5 (1) and
characterized by NMR spectroscopy, which is discussed below.
After the crystallization to isolate 4, removing volatiles from
the supernatant solution gave a yellow, oily mixture, which upon
sublimation at 55 °C yielded a pale-yellow solid of 2. Crystal-
(11) Our searches of the SciFinder and Beilstein data bases for -N(Me)CH2-
NR2 (R ) Me, Et, Pri) ligands and their parent amines yielded only the
following Japanese patent about HN(Me)CH2NEt2: Aron-Samuel, J. M.
D. Jpn. Tokkyo Koho, JP 53017595 19780609, 1978. MeNHCH2NMe2 was
the subject of calculations in Carballeira, L.; Perez-Juste, I. J. Comput.
Chem. 2001, 22, 135.
q
s-1 (∆G343K ) 17 kcal/mol) for the axial f equatorial
exchange.14 An alternative explanation is that there is only one
(12) Chen, T.-N.; Wu, Z.-Z.; Li, L.-T.; Sorasaenee, K. R.; Diminnie, J. B.; Pan,
H.-J.; Guzei, I. A.; Rheingold, A. L.; Xue, Z.-L. J. Am. Chem. Soc. 1998,
120, 13519.
(14) See Supporting Information for details.
(13) The difference between the 8-9% yield of 3 estimated by 1H NMR and
its 12% isolated yield is probably due to the error in NMR integration.
(15) Distances of N2, N4, and O1 to the Ta-O2-N5 plane are 0.0460, 0.0352,
and 0.0151 Å, respectively.
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14409

A R T I C L E S
Chen et al.
Figure 1. ORTEP views of 2-5. Disorder was observed in 4.
exchange of the ligands: axial-equatorial exchange of the
-NMe2 ligands. As the equatorial -NMe2 ligand moves to the
axial position, its two methyl groups move to the Mea and Meb
positions. The fact that the activation free energies ∆G333Kq and
∆G343K are equal suggests that the three amide ligands are
exchanging via the same process.
(Me2N)4Ta2[η2-N(Me)CH2NMe2]2(µ-O)2 (4) exhibits a dimer-
ic, centrosymmetric structure with two oxo bridges between Ta
atoms (Figure 1). The Ta-O(1)-TaA angle of 103.6(4)° in the
Ta-O(1)-TaA-O(1A) four-member ring is much larger than
the O(1)-Ta-O(1A) angle of 76.4(4)°. The Ta-O bond lengths
of 1.932(9)-1.958(10) Å are close to 1.917(6)-1.928(6) Å in
[TaCl2(NMe2)2(HNMe2)]2O16 containing a single O bridge
between two Ta atoms.
ppm, downfield shifted from those of the methyl groups in 4.
In the 13C NMR spectrum at 23 °C, the resonance of the
methylene carbon at 83.06 ppm is also downfield shifted from
those of the methyl groups in 4. 4 is inert to O2. Its solution in
benzene-d6 was found to slowly decompose to HNMe2 and
unknown Ta species.
q
As noted earlier, crystals of (Me2N)6Ta3[η2-N(Me)CH2-
NMe2]2(η2-ONMe2)(µ-O)3 (5) were sometimes found along with
the crystals of 4 in the separation of 4 from the reaction of O2
(0.5 equiv) with Ta(NMe2)5 (1). The identity of 5 in the crystal
mixture was confirmed by single-crystal X-ray diffraction.
However, NMR spectra of the crystal mixture showed only the
resonances of 4. When 4 equiv of O2 was used to react with 1,
1
a mixture of crystals of 4 and 5 was again obtained. H NMR
Crystals of 4 are thermally stable at -32 °C, but 4 in benzene-
d6 under N2 was found to decompose in one week at 23 °C. In
spectra of the mixture gave the resonances of 5, indicating that
the reaction of 1 with 4 equiv of O2 gave a higher yield (0.52%)
of 5.
1
the H NMR spectrum of 4, the peaks of the -NMe2 and the
-NMe groups in the chelating -N(Me)CH2NMe2 ligands are
at 2.39 and 3.22 ppm, respectively. The former (-NMe2) is an
amine group and donates its lone pair electrons to the Ta atoms.
The latter (-NMe) is an amide ligand carrying a negative
charge, and its chemical shift is close to 3.51 ppm for the
terminal -NMe2 ligands in 4. The methylene protons in the
chelating -N(Me)CH2NMe2 ligands appear as a singlet at 4.19
5 is a trimetallic complex containing three oxo bridges (Figure
1). Two Ta atoms [Ta(1) and Ta(2)] each and the third Ta atom
[Ta(3)] are coordinated by one η2-N(Me)CH2NMe2 ligand and
a η2-ONMe2 ligand, respectively. In addition, two -NMe2
ligands on each Ta atom give a pseudo-octahedral environment
around the metal. There is no symmetry in the solid structure
of 5, making each ligand unique. The MeN- and -NMe2
moieties of both η2-N(Me)CH2NMe2 ligands are trans to an
(16) Chisholm, M. H.; Huffman, J. C.; Tan, L.-S. Inorg. Chem. 1981, 20, 1859.
9
14410 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

Reaction of Ta(NMe2)5 with O2
A R T I C L E S
Table 1. Crystal Data and Structure Refinement for 2-5
2
3
4
5
emp. formula
(fw)
temp (K)
crystal system
space group
unit cell
C10H30N5OTa
(417.34)
173(2)
orthorhombic
Pca2(1)
C10H30N5O2Ta
(433.34)
173(2)
monoclinic
P2(1)/n
C16H46N8O2Ta2
(744.51)
120(2)
monoclinic
C2/m
C25.5H68N11O4Ta3
(1135.76)
100(2)
triclinic
P1h
dimensions
(Å and deg)
a ) 17.079(6),
R ) 90
a ) 8.273(9),
R ) 90
a ) 11.435(6),
R ) 90
a ) 9.8785(5),
R ) 77.972(1)
b ) 11.5811(5),
â ) 82.666(1)
c ) 19.2869(9),
γ ) 66.618(1)
1978.35(16)
2
b ) 12.092(4),
â ) 90
b ) 13.077(14),
â ) 99.880(18)
c ) 16.250(17),
γ ) 90
b ) 13.712(8),
â ) 107.534(7)
c ) 8.616(5),
γ ) 90
c ) 16.206(5),
γ ) 90
3346.9(19)
8
volume (Å3)
Z
1732(3)
4
1288.1(12)
2
density (calcd, Mg/m3)
1.920
6.565
1.657
6.350
1.662
8.514
1.907
8.318
ab. coeff. (mm-1
)
F(000)
1648
856
720
1098
crystal size (mm3)
θ range (deg)
index ranges
0.38 × 0.32 × 0.27
2.06 to 28.32
-21 e h e 22
-15 e k e 15
-20 e l e 21
32990
7935 [R(int) ) 0.0596]
28.32°, 98.1%
0.2702/
0.25 × 0.14 × 0.12
2.01 to 28.29
-10 e h e 10
-17 e k e 17
-21 e l e 21
18076
4170 [R(int) ) 0.0372]
28.29°, 97.0%
0.5162/
0.20 × 0.10 × 0.04
2.39 to 28.21
-15 e h e 14
-17 e k e 18
-11 e l e 11
7190
1565 [R(int) ) 0.0529]
25.00°, 100.0%
0.7270/
0.38 × 0.34 × 0.31
1.95 to 29.50
-13 e h e 13
-16 e k e 16
-26 e l e 26
36051
10805 [R(int) ) 0.0262]
29.50°, 98.1%
0.1825/
reflections collected
indep. reflections
completeness to θ )
max./min.
transmission
0.1893
0.2997
0.2808
0.1442
data/restraints/param.
goodness-of-fit on F2
final R indices [I > 2σ(I)]a
7935/1/327
1.163
4170/0/173
1.167
1565/0/79
1.060
10805/0/381
1.045
R1 ) 0.0388,
wR2 ) 0.0945
R1 ) 0.0536,
wR2 ) 0.1229
3.786 and
R1 ) 0.0248,
wR2 ) 0.0620
R1 ) 0.0360,
wR2 ) 0.0882
1.128 and
R1 ) 0.0566,
wR2 ) 0.1576
R1 ) 0.0576,
wR2 ) 0.1591
2.691 and
R1 ) 0.0201,
wR2 ) 0.0493
R1 ) 0.0238,
wR2 ) 0.0505
2.316 and
R indices (all data)
largest diff. peak and hole
-2.130 e‚Å-3
-2.190 e‚Å-3
-3.269 e‚Å-3
-0.911 e‚Å-3
a R ) ∑||Fo| - |Fc||/∑|Fo|; Rw ) (∑[w(F2o - Fc2)2]/∑[w(Fo2)2])1/2
.
oxo and amide ligand, respectively. The O(4) atom in the η2-
ONMe2 ligand is nearly trans to bridging O(2) with the O(4)-
Ta(3)-O(2) angle of 143.82(8)°. The Ta(3)-O(2) bond length
of 1.8845(17) Å is shorter than other Ta-oxo bond lengths
[1.9404(17)-1.9988(17) Å] and the Ta(3)-O(4)NMe2 length
of 2.028(2) Å. In the six-membered ring formed by the three
oxo ligands and three Ta atoms, the Ta-O-Ta angles range
from 148.35(10) to 154.89(10)°, significantly larger than that
[103.6(4)°] in dimeric 4. The O-Ta-O angles are nearly 90°
[87.07(7)-92.92(8)°], as expected for the octahedral-coordinated
Ta atoms. 5 provides an accurate structure of the novel
η2-N(Me)CH2NMe2 ligand, as those in 4 are both disordered.
In 5, the Ta-N(Me) bond lengths of 2.037(2)-2.042(2) Å are
similar to those [2.003(2)-2.042(2) Å] in Ta-NMe2 (mono-
dentate); however, they are significantly smaller than those of
dative TarNMe2 bonds [2.393(2)- 2.394(2) Å] in the η2-
reveal the parent ion or [3 + H+] peak.14 Similarly the spectrum
of (Me2N)4Ta2[η2-N(Me)CH2NMe2]2(µ-O)2 (4) using EI did not
reveal its parent ion peak. Thus, no attempts were made to
conduct crossover studies using a Ta(NMe2)5-Ta(NMe2-d6)5
mixture or investigate whether the two O atoms in either 3 or
4 were from the same O2 molecules using a 16O2-18O2 mixture.
Such studies would rely on the characterization of isotopomers
by MS.
Reaction of Ta(NMe2)5 (1) with O2 at -50 °C or a Lower
Pressure of O2. The reaction was conducted at -50 °C with
1
0.5 equiv of O2 and monitored by H NMR. (Me2N)4Ta(η2-
1
ONMe2) (2) was the first obserVed product by H NMR in 23
min. However, this does not necessarily indicate that 2 was the
first product in the reaction, as we cannot rule out that other
products had formed at concentrations below the NMR detection
limit. 3 and 4 were observed later in 40 min.
1
N(Me)CH2NMe2 ligands. Both H and 13C NMR spectra of 5
Effect of O2 pressure on its reaction with 1 was studied as
well. The partial pressure of O2 (0. 5 equiv) in sealed reaction
flasks varied between 240 (Sample A), 500 (Sample B), and
760 mmHg (Sample C), respectively. In 24 min, 1 in Sample
C had all disappeared. After 32 min, 5.5% of unreacted 1 was
observed in Sample B. In comparison, after 39 min, 21.8% of
unreacted 1 was observed in Sample A. Although these studies
are limited in scope, the observations indicate that the reaction
at a higher O2 partial pressure is faster.
are consistent with its structure, and the chemical shifts of the
η2-N(Me)CH2NMe2 and η2-ONMe2 ligands in 5 are similar to
those in 4 and 2-3, respectively.
Repeated attempts were made to characterize (Me2N)3Ta(η2-
ONMe2)2 (3) by mass spectrometry (MS) using electron
ionization (EI), DART (direct analysis in real time),17 and
electrospray ionization (ESI). These spectra, however, did not
(17) Cody, R. B.; Laramee, J. A.; Durst, H. D. Anal. Chem. 2005, 77, 2297.
After 19 h, 1H NMR spectra show that the yields of
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14411

A R T I C L E S
Chen et al.
Table 2. Selected Bond Lengths and Angles in 2-5
2
3
4
5
O(1)-Ta 1.983(7)
Ta-N(1) 2.052(5)
Ta-O(1) 1.932(9)
Ta(1)-O(3) 1.9404(18)
Ta(1)-O(1) 1.9723(17)
Ta(2)-O(1) 1.9365(17)
Ta(2)-O(2) 1.9988(17)
Ta(3)-O(2) 1.8845(17)
Ta(3)-O(3) 1.9554(18)
Ta(3)-O(4) 2.028(2)
O(4)-N(9) 1.436(3)
Ta(1)-N(1) 2.394(2)
Ta(1)-N(2) 2.042(2)
Ta(1)-N(3) 2.042(2)
Ta(1)-N(4) 2.015(2)
Ta(2)-N(5) 2.393(2)
Ta(2)-N(6) 2.037(2)
Ta(2)-N(7) 2.031(2)
Ta(2)-N(8) 2.003(2)
N(1)-Ta 2.288(10)
Ta-N(2) 2.029(5)
Ta-O(1A) 1.958(10)
Ta-N(1) 2.234(8)
N(2)-Ta 1.969(10)
Ta-N(3) 2.065(5)
N(3)-Ta 2.064(10)
Ta-N(4) 2.203(5)
Ta-N(2) 1.957(14)
Ta-N(3) 2.040(11)
Ta-O(1)-TaA 103.6(4)
N(2)-Ta-O(1A) 104.0(5)
O(1)-Ta-N(1) 103.7(2)
N(2)-Ta-N(1) 76.3(2)
N(4)-Ta 2.057(10)
Ta-N(5) 2.234(5)
N(5)-Ta 2.002(11)
Ta-O(1) 2.014(5)
N(2)-Ta-O(1) 133.2(4)
O(1)-Ta-N(5) 116.6(5)
O(1)-Ta-N(4) 88.5(4)
O(1)-Ta-N(3) 85.3(4)
O(1)-Ta-N(1) 38.8(3)
O(1)-N(1)-Ta 59.1(4)
Ta-O(2) 2.016(5)
N(5)-O(2) 1.462(7)
N(4)-O(1) 1.448(6)
N(1)-Ta-N(2) 90.1(2)
N(1)-Ta-N(4) 90.7(2)
N(1)-Ta-N(5) 89.9(2)
N(1)-Ta-N(3) 178.2(2)
O(1)-Ta-N(1) 89.4(2)
O(1)-Ta-N(4) 39.85(18)
O(1)-Ta-O(2) 87.17(18)
O(3)-Ta(1)-O(1) 87.07(7)
O(1)-Ta(2)-O(2) 84.52(7)
O(2)-Ta(3)-O(3) 90.35(8)
O(3)-Ta(3)-O(4) 92.92(8)
O(2)-Ta(3)-O(4) 143.82(8)
Ta(2)-O(1)-Ta(1) 154.89(10)
Ta(3)-O(2)-Ta(2) 151.38(10)
Ta(1)-O(3)-Ta(3) 148.35(10)
Ta(2)-O(1)-Ta(1) 154.89(10)
Ta(3)-O(2)-Ta(2) 151.38(10)
Ta(1)-O(3)-Ta(3) 148.35(10)
N(9)-O(4)-Ta(3) 78.12(13)
(Me2N)4Ta(η2-ONMe2) (2) (eq 3) was formed, suggesting that
there is a larger barrier for this exchange. Our theoretical studies,
discussed below, are consistent with the observation. The DFT
calculations suggest that this exchange occurs in the presence
of a catalyst to give 2. On the basis of the results thus far, a
partial picture of the pathways in the formation of 2 and 3 is
provided in Scheme 2. The cycle in the conversions between 2
and 3 uses both 1 and O2 until the limiting reagent is consumed.
Scheme 2
Reaction of Ta(NMe2)5 (1) with O2 in the Presence of a
Radical Trap. The products of the reaction between 1 and O2
may also be derived from an autoxidation path involving radical
initiation and propagation. In other words, free radicals are
expected to be present during the reaction. We conducted the
reaction of Ta(NMe2)5 (1) with O2 in the presence of TEMPO
(1,1,5,5-tetramethylpentamethylene nitroxide) as a radical trap.14,18
1H NMR monitoring of the reaction revealed that (1) 1 slowly
reacts with TEMPO, giving unknown species;18 (2) most
TEMPO (90%) remained after the reaction between 1 and O2
was completed; (3) no products of radical trapping by TEMPO
could be identified; (4) products identical to those in the absence
of TEMPO (2-4, HNMe2) were observed in the reaction, except
for the unidentified species from the slow reaction of 1 with
TEMPO. The study here, limited in scope, could not rule out
autoxidation as a pathway in the reaction. It suggests, however,
that, if autoxidation is a pathway, it is perhaps not the main
pathway. Our theoretical studies of the pathways are given
below.
Ta(NMe2)3(η2-ONMe2)2 (3) and (Me2N)4Ta2[η2-N(Me)CH2-
NMe2]2(µ-O)2 (4) in the three samples were similar: 5-7% of
3 and 7-11% of 4. The yields of Ta(NMe2)4(η2-ONMe2) (2)
in Samples B and C were similar (49-55%) and higher than
35% in Sample A (that was exposed to 240 mmHg O2). As
discussed earlier, the yield of 5 was too small to be detected by
NMR in the reaction mixture. It should be noted that the yields
of 2-4 in Samples B (500 mmHg) and C (760 mmHg) are
similar to those (46-52% 2, 8-9% 3, 8-9% 4) yielded with
650 mmHg (0.5 equiv) O2 discussed earlier.
Reaction of (Me2N)4Ta(η2-ONMe2) (2) with O2 and an
Exchange between Ta(NMe2)5 (1) and (Me2N)3Ta(η2-
ONMe2)2 (3). Additional studies have been conducted to shed
light on the formation of 2 and 3. 2 was found to react with 1
equiv of O2 to give 3 in 43% yield (based on 2, Scheme 2),
suggesting that 2 is perhaps an intermediate to 3. However, this
observation cannot rule out the formation of 3 from direct
insertion of two O atoms of an O2 molecule into two Ta-N
bonds in 1. Our earlier theoretical studies of the reactions of
O2 with model complexes Zr(NR2)4 (R ) Me, H) in fact suggest
that such pathways are feasible.10 In other words, two parallel
pathways in Scheme 2 may both lead to the formation of 3.
Heating a mixture of Ta(NMe2)5 (1) and (Me2N)3Ta(η2-
ONMe2)2 (3) in a sealed Young NMR tube at 50 °C for 4 days
yielded no change in the 1H NMR spectrum. After heating the
mixture at 90 °C for another 3 days, a significant amount of
DFT Studies of the Reaction between d0 Ta(NMe2)5 (1)
and O2. To extend our understanding of the reactions of d0
transition metal amide complexes with O2,10 the mechanistic
pathways of the reaction of 1 with O2 have been investigated
by DFT calculations.
(18) Albe´niz, A. C.; Espinet, P.; Lo´pez-Ferna´ndez, R.; Sen, A. J. Am. Chem.
Soc. 2002, 124, 11278.
9
14412 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

Reaction of Ta(NMe2)5 with O2
A R T I C L E S
Figure 2. Reaction energies and reaction free energies at 298.15 K calculated for the reactions of 1 with O2 leading to the formations of 2-5.
Figure 3. Optimized structures with selected bond lengths (Å) and angles (deg) for 1-5. The corresponding X-ray structural parameters are given in
parentheses. The hydrogen atoms in 5 are omitted for clarity.19
We first studied the reaction of 1 with O2 to form the
experimentally observed products 2-5. The calculated reaction
energies and geometries of 2-5 are shown in Figures 2 and 3.
The calculated Ta-N(amide) bonds are in good agreement with
those of the X-ray structures. The large reaction free energies
calculated indicate that the reactions are very exergonic. The
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14413

A R T I C L E S
Chen et al.
Scheme 3. Proposed Mechanism for the Formation of 2-5
Chart 1
mechanisms are proposed here for the formation of 2-5 in the
reaction of Ta(NMe2)5 (1) with O2.
Proposed Reaction Mechanism. On the basis of the calcula-
tion results, we proposed the reaction mechanisms to account
for the formation of 2-5 (Scheme 3). The reaction is initiated
by the insertion of the triplet O2 into one of the Ta-N bonds in
Ta(NMe2)5 (1) through minimum energy crossing point (MECP)20
A2 to produce the tantalum peroxide A3. A3 easily isomerizes
to A4. Then the peroxide ligand of A4 oxidizes one amide ligand
by two competitive reaction pathways. The predominant one is
Path II, in which the two-coordinate, Ta-bonded O of the
peroxide ligand attacks the N of the nearby amide ligand to
form 3 directly. The other pathway is Path I, in which the two-
coordinate O abstracts a hydride from the nearby amide ligand
followed by a proton transfer to eliminate an amine to give the
Ta-oxo species B4. B4 sequentially undergoes coupling reac-
tion (B4 f B6 via B5), ligand exchange (B6 + 1 f B7 + 2),
and dimerization (2 B7 f 4) to form 4. 2 can be obtained from
the following three ligand-exchange reactions: B6 + 1 f B7
+ 2, X1 + 1 f X2 + 2, and X2 + 3 f X1 + 2 (Scheme 3).
Via imine insertion, the transient species X2 also affords B7,
which can dimerize to form 4. The complexation of 4 and X1
leads to the formation of 5. The detailed calculation results that
lead to the proposed mechanism are described below.
calculations show that, different from the group 4 analogues
M(NMe2)4 (M ) Zr, Hf), Ta(NMe2)5 (1), containing one more
amide ligand, cannot dimerize because of its crowded coordina-
tion sphere. An end-on O2 complex (Chart 114) could not be
located. On the basis of these results, it is concluded that Ta-
(NMe2)5 (1) can only react with O2 via a monomeric pathway.
The van der Waals complex A1 (Scheme 3 and Figure 4)
formed between Ta(NMe2)5 (1) and the triplet O2 has long
Ta-O distances (ca. 4.4 Å). To reach the singlet energy surface,
an intersystem conversion from triplet to singlet energy surface
occurs at the MECP. The MECP, A2, was located with a
program code developed by Harvey and co-workers.20d,e A2 is
not a stationary point on the potential energy surface, and its
energy cannot be corrected with ZPE correction. The MECP
(A2) is ca. 14.9 kcal/mol higher in electronic energy than 1 +
triplet O2, which is set to be the reference state. The entropy
loss for the formation of A2 is expected to be between those of
A1 and A3. Therefore, the relative free energy of A2 with
respect to the reference state is estimated to be ca. 25 kcal/mol.
The result suggests that formation of A3 is viable under the
reaction conditions. Through this triplet-singlet crossover at
the structure A2, the singlet Ta peroxide A3 is formed with a
reaction energy of -13.0 kcal/mol. In this process, the amide
ligand is oxidized. After the O-O insertion into the Ta-N bond,
the “leaving” amide forms a dative bond with the Ta center.
Insertion of O2 into a Ta-N Bond in Ta(NMe2)5 (1). In a
previous study, we found that O2 reacts with Zr(NMe2)4 via a
dimeric pathway rather than a monomeric pathway.10 Our
(19) The X-ray structural parameters of 1 are from its structure in Batsanov, A.
S.; Churakov, A. V.; Howard, J. A. K.; Hughes, A. K.; Johnson, A. L.;
Kingsley, A. J.; Neretin, I. S.; Wade, K. J. Chem. Soc., Dalton Trans. 1999,
3867.
(20) (a) Schro¨der, D.; Shaik, S.; Schwarz, H. Acc. Chem. Res. 2000, 33, 139.
(b) Poli, R.; Harvey, J. N. Chem. Soc. ReV. 2003, 32, 1. (c) Harvey, J. N.
Phys. Chem. Chem. Phys. 2007, 9, 331. (d) Harvey, J. N.; Aschi, M.;
Schwarz, H.; Koch, W. Theor. Chem. Acc. 1998, 99, 95. (e) Harvey, J. N.;
Aschi, M. Phys. Chem. Chem. Phys. 1999, 1, 5555.
9
14414 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

Reaction of Ta(NMe2)5 with O2
A R T I C L E S
Figure 4. Calculated relative free energies and optimized structures of A1, A2, A3, and A4. The relative free energies and electronic energies (in parentheses)
are given in kcal/mol. The bond lengths are in Å.
The peroxide ligand -O-O-NMe2 formed from the O2
insertion step can also form a dative bond with the Ta center
from the other donor, O atom. By rotating the O-O bond, A3
rearranges to A4, which has a similar stability with A3.
Coordination of both the two oxygen atoms of the peroxide
ligand to the Ta center elongates the O-O bond, giving a
situation wherein the O-O bond is easier to be cleaved in this
Figure 5. Four types of possible hydrogen migration transition states. The
activation free energies and electronic energies (in parentheses), which are
isomer.
Possible Pathways for Hydrogen Migration and Amine
Elimination. Reaction of O2 with 1 leads to the formation of
2-5 with the η2-N(Me)CH2NMe2 ligand in 4 and 5. To generate
this interesting ligand, it is convenient to think of a pathway
that involves a â-H migration from a NMe2 ligand to one of
the other ligands or to the metal center, leading to the formation
of a MeNdCH2 ligand, followed by a nucleophilic attack of
another NMe2 ligand on the carbon center of the newly formed
MeNdCH2 ligand. â-H abstraction by another amide ligand is
well-known and is attributed, e.g., to the formation of Ta(NEt2)3-
(EtNdCHMe)21a and in the D incorporation from DN(CH3)2,
in presence of M(NMe2)n, into the methyl group, yielding (CH3)-
(CH2D)NH.21b In the current case, several possible precursor
complexes, 1, A3, A4, 2, and 3, are able to undergo the â-H
migration followed by the nucleophilic attack. All of these possi-
ble hydrogen migration pathways were examined thoroughly.
Four types of hydrogen migration transition structures based
on the five precursor complexes mentioned above are shown
in Figure 5. Transition structure I, in a â-H abstraction process,
is based on the precursor complex 1. Since 1 adopts a square
pyramidal geometry, the amides are not equivalent. Different
relative to their corresponding precursor complexes, are given in kcal/mol.
isomeric transition structures were calculated. The lowest barrier
for this type of hydrogen migration is ca. 35 kcal/mol. The
barriers are similar to those we obtained previously for cases
with a hydrogen migration from a methyl group.22 In the five-
membered-ring transition structures (I), the metal center interacts
with all of the other four members. The hydrogen migration
resembles a σ-bond metathesis between a C-H bond and a
Ta-N bond.
The situation is different in transition structure II. The
hydrogen migration involves a hydride rather than a proton,
leading to the breaking of both the C-H and X-N bonds. The
hydride migration liberates HNMe2 and gives a Ta-oxo if X
) O or a Ta-peroxide if X ) O-O. The transition structure
II, in which the NMe2 group in the six-membered ring does
not have a bonding interaction with the metal center, are also
high in energy. The results are expected because a NMe2 group
without metal coordination is not a good hydride acceptor.
Transition structure III corresponds to a stepwise process,
â-H elimination followed by hydride migration. Formation of
the Ta-H intermediates is a very endothermic process, leading
to even higher barriers (>39 kcal/mol).
(21) (a) Takahashi, Y.; Onoyama, N.; Ishikawa, Y.; Motojima, S.; Sugiyama,
K. Chem. Lett. 1978, 525. (b) Nugent, W. A.; Ovenall, D. W.; Holmes, S.
J. Organometallics 1983, 2, 161. For other examples of â-H abstraction
reactions to give imine ligands, see: (c) Berno, P.; Gambarotta, S.
Organometallics 1995, 14, 2159. (d) Scoles, L.; Ruppa, K. B. P.;
Gambarotta, S. J. Am. Chem. Soc. 1996, 118, 2529. (e) 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. (f) Rothwell, I. P. Polyhedron 1985,
4, 177. (g) Ahmed, K. J.; Chisholm, M. H.; Folting, K.; Huffman, J. C. J.
Am. Chem. Soc. 1986, 108, 989. (h) Airoldi, C.; Bradley, D. C.; Vuru, G.
Transition Met. Chem. 1979, 4, 64. (i) Mayer, J. M.; Curtis, C. J.; Bercaw,
J. E. J. Am. Chem. Soc. 1983, 105, 2651. (j) Bu¨rger, H.; Neese, H.-J. J.
Organomet. Chem. 1970, 21, 381. (k) de Castro, I.; Galakhov, M. V.;
Go´mez, M.; Go´mez-Sal, P.; Royo, P. Organometallics 1996, 15, 1362.
The most favorable hydrogen migration is the â-H migration
process involving the transition structure IV, which corresponds
to the A4 to B2 conversion shown in Scheme 3 and Figure 6.
During this process, the peroxide ligand oxidizes the amide to
imine through the abstraction of a hydride, while the peroxide
(22) (a) Wu, Y.-D.; Chan, K. W. K.; Xue, Z.-L. J. Am. Chem. Soc. 1995, 117,
9259. (b) Wu, Y.-D.; Peng, Z.-H.; Xue, Z.-L. J. Am. Chem. Soc. 1996,
118, 9772. (c) Yu, X.; Bi, S.; Guzei, I. A.; Lin, Z.; Xue, Z.-L. Inorg. Chem.
2004, 43, 7111.
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14415

A R T I C L E S
Chen et al.
Figure 6. Calculated energy profile for the most favorable hydrogen migration reaction. The relative free energies and electronic energies (in parentheses)
are given in kcal/mol.
itself is reduced to a hydroxide ligand and an aminoxide ligand.
The proton of the hydroxide ligand in B2 is considered to be
quite acidic and can be easily transferred to a neighboring amide
to complete the HNMe2 elimination, resulting in the formation
of the stable intermediate, Ta-oxo complex B4 (Scheme 3 and
Figure 6). The -OH proton transfer leading to the elimination
of HNMe2 from Ta(NMe2)5 (1) was reported in other complexes
as well.23 Both the hydride migration (A4 f B2) and the proton
transfer (B2 f B4) have low activation energies and are very
exothermic, suggesting that the peroxide ligand plays an
important role. The importance of the peroxide ligand may
explain the other oxidation reactions between O2 and metal
amides that involve amine eliminations.
Formation of the η2-N(Me)CH2NMe2 Ligand in 4. The
dissociation of the imine ligand from the Ta complex B4 affords
a five-coordinate Ta complex X1. X1 is a coordinatively
unsaturated Ta-oxo complex and tends to form O-bridging
complexes with other Ta complexes, making it act as a catalyst
in ligand-exchange reactions. By exchanging the -ONMe2
ligand with 1, X1 transforms to another Ta-oxo complex X2.
The details of the ligand exchange will be discussed later. The
relatively spacious environment of X1 and X2 provides a
possibility for the coordination of imine ligand (B4 and B8)
and the following imine ligand insertion. In transition state B5
or B9, the imine inserts into one of the Ta-N bonds to give
the intermediate Ta-oxo complex, B6 or B7, containing the
interesting η2-N(Me)CH2NMe2 ligand. These insertion processes
are exothermic with activation free energies of ca. 31 kcal/mol
(Figure 6). The gas-phase calculation tends to overestimate the
(23) (a) Schweiger, S. W.; Tillison, D. L.; Thorn, M. G.; Fanwick, P. E.;
Rothwell, I. P. J. Chem. Soc., Dalton Trans. 2001, 2401. (b) Fei, Z.; Busse,
S.; Edelmann, F. T. J. Chem. Soc., Dalton Trans. 2002, 2587.
9
14416 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

Reaction of Ta(NMe2)5 with O2
A R T I C L E S
Scheme 4
molecules. Since complexation of O2 with 2 equiv of 1 is
unlikely to take place, the separation of the two O atoms into
two molecules can be achieved with a ligand exchange between
an O-containing Ta complex and an O-free Ta complex. A
careful examination of all mononuclear complexes or intermedi-
ates we have discussed thus far indicates that a few oxo
intermediates, which have a less crowded ligand environment,
are suitable to undergo ligand-exchange reactions with the five-
coordinated complex Ta(NMe2)5 (1). These oxo intermediates
can rearrange themselves to four-coordinate species, providing
a vacant coordination site for binding with 1. Another reason
to consider these oxo intermediates is that an early transition
metal-oxo bond can easily transform itself to a bridging oxo
that links two metal centers.24 As discussed above, 2 can be
generated from the ligand exchange between 1 and B6, an oxo
intermediate. Equation 6 in Scheme 5 shows the energetics
related to the ligand exchange. The complex Com-B6-1 is
formed with two bridging ligands; one is the oxo of B6, and
the other is an amide of 1. Com-B6-1 can transform to Com-
B7-2 by switching the bridging ligand from the amide to the
aminoxide. Then Com-B7-2 decomposes into two new species
B7 and 2, completing the ligand-exchange process. The fact
that the ligand-exchange reaction is exothermic together with
the stability of Com-B6-1 and Com-B7-2 suggests that the
ligand exchange is feasible.
entropy loss of bimolecular reactions, and consequently to
overestimate the activation free energies. Therefore, the barriers
of the insertion step are expected to be lower than 31 kcal/mol
and feasible. Similar to X1, the intermediate oxo complex B6
is flexible to create a coordination-unsaturated environment
around the metal center, allowing a ligand-exchange reaction
to occur. The ligand exchange, which will be discussed later,
occurs between B6 and 1 to form B7 and 2 (Scheme 4). B7 is
in fact a monomer of the dimer 4. The dimerization energy of
B7 is ca. 8.2 kcal/mol, enough to compensate the entropy loss
of the dimerization process. The calculated dimerization free
energy is close to zero, suggesting that B7 and 4 are in
equilibrium.
Formation of 3. In Path I, the reactive peroxide bond of A4
breaks, facilitating the O-H bond formation (A4 f B2). In
Path II, the peroxide bond cleavage also plays an important role
in the formation of 3. In the peroxide bond cleavage, the
pathway from A4 is much more favorable than that directly
from A3 (Figure 7). In the much more favorable pathway, the
three-coordinate O facilitates the peroxide bond cleavage.10
Ligand Exchange and Formation of 2. Formation of 2
requires that the two O atoms of O2 are separated into two
If the ligand-exchange reaction in eq 4 is the only reaction
that produces 2, then 2 is supposed to be two equiv of 4.
However, the experimental results of the yields of 2 and 4 are
26-27% and 9-10% (by NMR, 1 equiv of O2), respectively,
indicating that there may be other pathways for the formation
of 2. X1 generated from the imine dissociation of B4 (shown
in Figure 6) is another Ta-oxo intermediate that is able to
undergo ligand exchange with Ta(NMe2)5 (1). Since there is
Scheme 5. Calculated Relative Energies for Species Involved in the Ligand Exchange Reactions
a Relative free energies and electronic energies (in parentheses) are given in kcal/mol.
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14417

A R T I C L E S
Chen et al.
Conclusions
The formation of an unusual oxo-amino complex (Me2N)4-
Ta2[η2-N(Me)CH2NMe2]2(µ-O)2 (4) and (Me2N)6Ta3[η2-
N(Me)CH2NMe2]2(η2-ONMe2)(µ-O)3 (5) containing novel chelat-
ing-N(Me)CH2NMe2ligands,aminoxycomplexes(Me2N)4Ta(η2-
ONMe2) (2) and (Me2N)3Ta(η2-ONMe2)2 (3) implies three key
steps in the reactions of Ta(NMe2)5 (1) with triplet O2: (i)
formation of a peroxide ligand from the oxygen insertion into
Ta-N bonds; (ii) cleavage of O-O bond in the peroxide O-O-
NMe2 ligands to form two aminoxy ligands or one aminoxy as
well as one TadO ligand; (iii) exchange between amino and
aminoxy ligands. The ligands formed from these three steps
are elementary steps in the formation of metal oxides from the
reaction of d0 metal amides with triplet O2.
A detailed theoretical study indicates that the formation of
2-5 is initiated by O2 insertion into monomeric Ta(NMe2)5 (1),
through an intersystem conversion from triplet to singlet energy
surface to form the active peroxide complex A4. This peroxide
complex undergoes two competitive reactions: A majority of
A4 undergo peroxide bond cleavage via Path II to produce 3
which is the product with the highest yield (37% by NMR). A
minority of A4 proceed via Path I and ultimately gives 4 and 3
at the ratio of 1:2. A small amount of 2 is also obtained from
the catalytic ligand exchange reaction between 3 and 1. Although
a study using a radical trap did not reveal the formation of
trapped radicals, autoxidation involving radicals as an alternative
pathway needs to be further explored in detail in the future.
Figure 7. Calculated energy profile for the peroxide bond cleavage reaction.
The relative free energies and electronic energies (in parentheses) are given
in kcal/mol.
only a small amount of X1 in the solution, instead of dimer-
ization it can easily react with the relatively abundant 1.
Equation 7 in Scheme 5 shows that complexation of X1 with 1
gives Com-X1-1 and then Com-X2-2. The X1-1 binding
energies in these two complexes are greater than the B6-1
binding energies in Com-B6-1 and Com-B7-2, suggesting a
lower barrier for this ligand-exchange process. This ligand-
exchange reaction gives 2 and X2 with a reaction free energy
of ca. -6.0 kcal/mol. X2 is another Ta-oxo complex that has
a less crowded ligand environment. The ligand exchange
between X2 and the relatively abundant species 3 can give 2
and X1. The two ligand-exchange reactions (1 + X1 f 2 +
X2 and X2 + 3 f 2 + X1) form a thermodynamically favorable
catalytic cycle, leading to the formation of 2, as shown in eq 9
(see also Scheme 3).
Experimental Section
General Procedures. All manipulations were performed under a
dry and oxygen-free nitrogen atmosphere with the use of glovebox or
Schlenk techniques. All solvents were purified by distillation from
potassium/benzophenone ketyl. Benzene-d6 and toluene-d8 were dried
and stored over activated molecular sieves under nitrogen. TaCl5
(Strem), Me2NOH‚HCl (Aldrich), liquid ammonia (Air Products &
Chemicals, Inc.), O2 (National Welders Supply Co.), dried by passing
a P2O5 column, and LiNMe2 (Aldrich) were used as received.
Ta(NMe2)5 (1), Ta(NMe2)4Cl, Ta(NMe2)3Cl2, and LiONMe2 were made
according to the references.15,25-27 NMR spectra were recorded on a
Bruker AMX-400 Fourier transform spectrometer. Elemental analyses
were conducted by Complete Analysis Laboratories, Inc., Parsippany,
NJ. Caution: SeVeral studies below were conducted in heated, sealed
Young NMR tubes. New NMR tubes should be used, and the heating
should be performed behind a protectiVe shield.
Experimentally, the mixture of 3 and 1 gives 2 under a fairly
harsh condition, 90 °C, suggesting that the direct ligand
exchange15 is difficult to take place. Therefore, a catalyst, such
as the Ta-oxo intermediate X1 or X2, is necessary for this
process.
Formation of 5. Because of the high tendency to complex
with other species, the Ta-oxo intermediates X1 and X2 were
not detected. However, the existence of the transient speice X1
was supported by the observation of 5. 5 is composed of 4 and
X1 formally. The complexation of 4 and X1 is found to be quite
exothermic (eq 10).
Preparation of (Me2N)4Ta(η2-ONMe2) (2), (Me2N)3Ta(η2-ONMe2)2
(3), (Me2N)4Ta2[η2-N(Me)CH2NMe2]2(µ-O)2 (4), and (Me2N)6Ta3-
[η2-N(Me)CH2NMe2]2(η2-ONMe2)(µ-O)3 (5). A solution of Ta(NMe2)5
(1, 1.969 g, 4.91 mmol) in toluene or benzene (30.0 mL) was frozen
in liquid nitrogen, and the Schlenk flask was pumped for 10 min. The
frozen solid was warmed gradually till it melted, and 0.5 equiv of O2
(2.45 mmol) was then added. The yellow solution was stirred overnight
(24) (a) Abbenhuis, H. C. L.; Feiken, N.; Grove, D. M.; Jastrzebski, J. T. B.
H.; Kooijman, H.; der Sluis, P. V.; Smeets, W. J. J.; Spek, A. L.; Van
Koten, G. J. Am. Chem. Soc. 1992, 114, 9773. (b) Jernakoff, P.; de Bellefon,
C. D.; Geoffroy, G. L.; Rheingold, A. L.; Geib, S. J. Organometallics 1987,
6, 1362. (c) Sa´nchez-Nieves, J.; Frutos, L. M.; Royo, P.; Castan˜o, O.;
Herdtweck, E. Organometallics 2005, 24, 2004.
(25) Bradley, D. C.; Thomas, I. M. Can. J. Chem. 1962, 40, 1355.
(26) Chisholm, M. H.; Tan, L.-S.; Huffman, J. C. J. Am. Chem. Soc. 1982, 104,
4879.
(27) Mitzel, N. W.; Losehand, V.; Wu, A.; Cremer, D.; Rankin, D. W. H. J.
Am. Chem. Soc. 2000, 122, 4471.
9
14418 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

Reaction of Ta(NMe2)5 with O2
A R T I C L E S
at room temperature. A small amount of white precipitation was
observed. All volatiles were removed in vacuo to leave a yellow oil
which was extracted with n-pentane. The filtrate was concentrated and
recrystallized at -32 °C to first give colorless crystals of (Me2N)4Ta2-
[η2-N(Me)CH2NMe2]2(µ-O)2 (4, 0.128 g, 0.172 mmol, 7% yield based
on O2) in a few days. 1H NMR of 4 (benzene-d6, 400.18 MHz, 23 °C)
δ 4.19 (s, 4H, 2CH2), 3.51 (s, 24H, 4NMe2), 3.22 (s, 6H, 2NMe-CH2),
2.39 (s, 12H, 2CH2-NMe2). 13C NMR (benzene-d6, 100.63 MHz,
23 °C) δ 83.06 (2CH2), 48.06 (2CH2-NMe2), 46.88 (4NMe2), 38.66
(2NMe-CH2). Anal. Calcd for C16H46N8O2Ta2: C, 25.81; H, 6.23.
Found: C, 25.64; H, 6.19.
dry ice/ethanol. O2 (1 atm, 0.5 equiv, 0.082 mmol) was added to the
NMR tube, and the NMR tube was then placed in a precooled 400
MHz NMR spectrometer at -50 °C. (Me2N)4Ta(η2-ONMe2) (2)
1
appeared first in the H NMR spectrum in ca. 23 min after O2 was
added. (Me2N)3Ta(η2-ONMe2)2 (3) and (Me2N)4Ta2[η2-N(Me)CH2-
NMe2]2(µ-O)2 (4) appeared later in 40 min.
Reaction of Ta(NMe2)5 (1) with 0.5 equiv of O2 at Different
Pressures. Young NMR tubes were connected to glass tubings to give
the required volumes. Three samples containing 43.5-77.2 mg of 1 in
toluene-d8 (0.5 mL) were prepared. O2 (0.5 equiv) at 240 (Sample A),
500 (Sample B), and 760 mmHg (Sample C), respectively, was added
1
at room temperature. The reaction was monitored by H NMR.
After the crystals of 4 were collected, the supernatant solution was
concentrated to afford a yellow oil which was sublimed at 55 °C under
reduced pressure to give a solid of (Me2N)4Ta(η2-ONMe2) (2) (0.560
g, 1.342 mmol, 27% yield based on O2) on a cold finger. This pale-
yellow solid was dissolved in Et2O, and the solution was cooled to
Reaction of Ta(NMe2)5 (1) with O2 in the Presence of TEMPO
as a Radical Trap. TEMPO (57.2 mg, 0.36 mmol) and 4, 4′-
dimethoxybiphenyl (4.2 mg, 0.019 mmol) as the internal standard (IS)
were dissolved in benzene-d6 (0.5 mL) in a Young NMR tube. After
the 1H NMR spectrum was taken, Ta(NMe2)5 (1, 36.2 mg, 0.090 mmol)
1
-32 °C to give colorless crystals of (Me2N)4Ta(η2-ONMe2) (2). H
1
NMR of 2 (benzene-d6, 400.18 MHz, 23 °C) δ 3.25 (s, 24H, 4NMe2),
2.47 (s, 6H, ONMe2). 13C NMR (benzene-d6, 100.63 MHz, 23 °C) δ
48.80 (s, 4NMe2), 47.04 (s, ONMe2). Anal. Calcd for C10H30N5O2Ta:
C, 28.78; H, 7.25. Found: C, 28.49; H, 7.17.
was added. After 20 min, the H NMR spectrum of the mixture was
taken. Subsequently O2 (0.090 mmol, 1.0 atm, 2.2 mL) was added to
the mixture, and the NMR tube was shaken. After 60 min, the 1H NMR
spectrum was taken.
After sublimation of 2, the yellow residue was dissolved in a small
amount of n-pentane again. A few days later, some colorless crystals
of (Me2N)3Ta(η2-ONMe2)2 (3) (0.130 g, 0.300 mmol, 12% yield based
Preparation of 2 from Ta(NMe2)4Cl and LiONMe2. LiONMe2
(0.134 g, 2.00 mmol) in toluene (20.0 mL) was added dropwise with
stirring to 1 equiv of Ta(NMe2)4Cl (0.784 g, 2.00 mmol) in toluene
(20.0 mL) at room temperature. The solution was warmed up to 50 °C
and continued to stir at this temperature for 20 h. The solvent was
stripped off to give yellow solids, which were extracted with n-pentane.
The yellow filtrate was concentrated and cooled to -32 °C to give the
pale-yellow solid Ta(NMe2)4(η2-ONMe2) (2, 0.797 g, 95.5% yield).
Preparation of 3 from Ta(NMe2)3Cl2 and LiONMe2. LiONMe2
(0.268 g, 4.00 mmol) in toluene (20.0 mL) was added dropwise with
stirring to Ta(NMe2)3Cl2 (0.768 g, 2.00 mmol) in toluene (20.0 mL) at
room temperature. The solution was warmed to 50 °C and stirred at
this temperature for 20 h. The volatiles removed to give yellow solids,
which were extracted with hexanes. The pale-yellow filtrate was cooled
to -32 °C to give a white solid of Ta(NMe2)3(η2-ONMe2)2 (3, 0.747
g, 86.2% yield).
Preparation of 3 from the Reaction of (Me2N)4Ta(η2-ONMe2)
(2) with O2. (Me2N)4Ta(η2-ONMe2) (2, 24.6 mg, 0.0590 mmol) and
4,4′-bimethyldiphenyl (2.0 mg, 0.011 mmol) as internal standard were
placed into a Young NMR tube with benzene-d6 (ca. 0.5 mL) in a
drybox. The solution was frozen by liquid nitrogen and pumped. The
frozen solution was then warmed to liquid, and 1 equiv of O2 was
added. The NMR tube was shaken for 30 min at 23 °C and then
placed at room temperature for 99 h. 1H NMR spectrum of the solution
showed 2 had converted 3 (10.9 mg, 0.0252 mmol, yield 42.8% based
on 2).
1
on O2) came out at -32 °C. H NMR of 3 (benzene-d6, 400.18 MHz,
23 °C) δ 3.41 (s, 6H, NMe2), 3.05 (s, 6H, 2NMeAMeB), 2.94 (s, 6H,
2NMeAMeB), 2.56 (s, 12H, 2ONMe2). 13C NMR (benzene-d6, 100.63
MHz, 23 °C) δ 50.14 (s, 2ONMe2), 48.98 (s, NMe2), 47.05 (s, 2NMeA-
MeB), 46.11 (s, 2NMeAMeB). Anal. Calcd for C10H30N5O2Ta: C, 27.72;
H, 6.98. Found: C, 27.49; H, 6.75.
A few crystals of 5 were sometimes found along with those of 4
1
from the reaction Ta(NMe2)5 (1) with 0.5 equiv of O2. However, H
NMR spectrum of the crystals showed only resonances of 4. A higher
1
yield of 5 was obtained from the reaction of 4 equiv of O2 with 1. H
NMR spectra of the mixture of crystals gave the resonances of 5 (and
4). In this reaction, 4 equiv of O2 was added to a solution of 1 (0.75 g,
1.87 mmol) by a procedure similar to that in the reaction of 1 with 0.5
equiv of O2. Crystallization gave a mixture of crystals (67 mg) of 4
and 5 in a molar ratio of 4:5 ) 26:1 (5.5 wt % 5; 3.7 mg, 0.0033
mmol, 0.52% yield of 5 based on 1) by 1H NMR. Attempts to separate
the two crystals were not successful, as they appeared similar. 1H NMR
of 5 (toluene-d8, 400.18 MHz, 23 °C) δ 4.14-3.93 (m, 4H, CHAHB
and CHCHD), 3.71 (s, 6H, NMe2), 3.64 (s, 6H, NMe2), 3.60 (s, 6H,
NMe2), 3.39 (s, 6H, NMe2), 3.32 (s, 6H, NMe2), 3.31 (s, 6H, NMe2),
3.17 (s, 3H, NMe-CH2), 3.08 (s, 3H, NMe-CH2), 2.31 (s, 6H, ONMe2),
2.21 (s, 6H, CH2-NMe2), 2.12 (s, 6H, CH2-NMe2). 13C NMR (toluene-
d8, 100.63 MHz, 23 °C) δ 82.84 (CH2), 82.00 (CH2), 48.25-47.19
(6NMe2), 47.16 (ONMe2), 46.37 (NMe2-CH2), 46.09 (NMe2-CH2), 39.04
(NMe-CH2), 38.89 (NMe-CH2). Anal. Calcd for a mixture of 4 and 5
[Molar ratio 4: 5 ) 26:1; C22H64N11O4Ta3‚1/2(toluene) or C25.5H68N11O4-
Ta3 (Calcd C, 26.97; H, 6.03) for 5]: C, 25.85; H, 6.22. Found: C,
25.85; H, 6.16.
Ligand Exchange between Ta(NMe2)5 (1) and (Me2N)3Ta(η2-
ONMe2)2 (3). A Young NMR tube was added 1 (5.9 mg, 0.015 mmol),
2 (7.3 mg, 0.017 mmol), and bibenzyl (3.0 mg, 0.016 mmol, internal
standard) in benzene-d6 (0.5 mL). The mixture was heated at 50 °C
1
for 4 days. Afterward its H NMR spectrum revealed no significant
changes in the concentrations of the complexes and formation of 2.
The mixture was then heated at 90 °C for additional 3 days in the
sealed NMR tube. 1H NMR spectrum of the solution revealed the
presence of a significant amount of (Me2N)4Ta(η2-ONMe2) (2, 0.008
mmol).
Heating a Mixture of Ta(NMe2)5 (1) and LiONMe2. To a Young
NMR tube was added 1 (5.2 mg, 0.013 mmol), LiONMe2 (1.2 mg,
0.018 mmol), and bibenzyl (3.4 mg, 0.019 mmol, internal standard) in
benzene-d6 (0.5 mL). The mixture was heated at 50 °C for 6 days, and
no ligand exchange was observed between Ta(NMe2)5 (1) and Li-
ONMe2.
Reaction of excess O2 with Ta(NMe2)5 (1) was also studied. To a
solution of 1 (10.3 mg, 0.026 mmol) in toluene-d8 (0.5 mL) in a Young
NMR tube was added 4 equiv of O2 (0.103 mmol) at -65 °C. Si-
(SiMe3)4 (1.0 mg) was used as the NMR internal standard. The NMR
tube was inserted into a spectrometer precooled at -65 °C, and the
reaction was monitored by 1H NMR. Over a period of 1.5 h, the
temperature was gradually raised from -65 to -25 °C. 1H NMR spectra
of the reaction mixture gave the following yields (based on the limiting
reagent 1): 42% for 2, 2.6% for 3, 9.0% for dimeric 4, and 4.5% for
trinuclear 5.
Reaction of Ta(NMe2)5 (1) with 0.5 equiv of O2 at -50 °C. A
Young NMR tube was loaded with Ta(NMe2)5 (1, 66.2 mg, 0.164
mmol) in toluene-d8 (0.5 mL). After the solution was chilled by liquid
nitrogen and pumped for 10 min, it was placed in a -50 °C bath in a
Ligand Exchange between Ta(NMe2)5 (1), (Me2N)3Ta(η2-ONMe2)2
(3), and LiNMe2. To a Young NMR tube was added 1 (6.8 mg, 0.017
mmol), 3 (7.8 mg, 0.018 mmol), LiNMe2 (1 mg, 0.020 mmol), and
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14419

A R T I C L E S
Chen et al.
bibenzyl (3.4 mg, 0.019 mmol, internal NMR standard) in benzene-d6
(0.5 mL). After the mixture was heated at 50 °C for 21 h, (Me2N)4-
Ta(η2-ONMe2) (2) was observed in 1H NMR spectra. The concentrations
of Ta(NMe2)5 (1) and (Me2N)3Ta(η2-ONMe2)2 (3) had increased and
decreased, respectively. After heating the solution at 50 °C for another
The data were collected by using the full sphere data collection
routine to survey the reciprocal space to the extent of a full sphere to
a resolution of 0.72 Å. A total of 36051 data were harvested by
collecting four sets of frames with 0.36° scans in ω and one set with
0.45° scans in æ with an exposure time 20 s per frame. These highly
redundant datasets were corrected for Lorentz and polarization effects.
The absorption correction was based on fitting a function to the
empirical transmission surface as sampled by multiple equivalent
measurements.29
The systematic absences in the diffraction data were consistent for
the space groups P1h and P1. The E-statistics strongly suggested the
centrosymmetric space group P1h that yielded chemically reasonable
and computationally stable results of refinement.29
1
6 days, H NMR revealed that the concentrations of (Me2N)4Ta(η2-
ONMe2) (2) (0.0066 mmol) and 1 (0.021 mmol) had increased, and
the concentration of (Me2N)3Ta(η2-ONMe2)2 (3) (0.0080 mmol) had
decreased. The NMR tube was then placed at room temperature, and
no further change was observed.
Variable-Temperature NMR Studies of (Me2N)3Ta(η2-ONMe2)2
1
(3). H NMR spectra of a solution of 3 (15 mg) in benzene-d6 (0.5
mL) in a Young NMR tube were taken at 296, 313, 333, and 343 K,
respectively. These spectra are shown in Figure S4a-d.
A successful solution by the direct methods provided most non-
hydrogen atoms from the E-map. The remaining non-hydrogen atoms
were located in an alternating series of least-squares cycles and
difference Fourier maps. All non-hydrogen atoms were refined with
anisotropic displacement coefficients. All hydrogen atoms were included
in the structure factor calculation at idealized positions and were allowed
to ride on the neighboring atoms with relative isotropic displacement
coefficients.
There was one solvate molecule of toluene present in the asymmetric
unit. A significant amount of time was invested in refining this molecule
which was disordered over a crystallographic inversion center. Bond
length restraints were applied to model the molecules, but the resulting
isotropic displacement coefficients suggested the molecules were
mobile. In addition, the refinement was computationally unstable.
Option SQUEEZE of program PLATON30 was used to correct the
diffraction data for diffuse scattering effects and to identify the solvate
molecule. PLATON calculated the upper limit of volume that can be
occupied by the solvent to be 237 Å3, or 12% of the unit cell volume.
The program calculated 55 electrons in the unit cell for the diffuse
species. This approximately corresponds to one molecule of toluene
molecule in the asymmetric unit (50 electrons), or one-half molecule
of solvent per trinuclear organometallic complex. Note that all derived
results in the tables are based on the known contents. No data are given
for the diffusely scattering species.
Determination of X-ray Crystal Structures of 2-4. The data for
the X-ray crystal structures of 4 were collected on a Bruker-AXS APEX
diffractometer with Kryoflex low-temperature device, and the data of
2 and 3 were collected on a Smart 1000 X-ray diffractometer equipped
with a CCD area detector fitted with an upgraded Nicolet LT-2 low-
temperature device. The data were obtained using a graphite-mono-
chromated Mo source (KR radiation, 0.71073 Å). Suitable crystals were
coated with paratone oil (Exxon) and mounted on loops under a stream
of nitrogen at the data collection temperature. The structures were solved
by direct methods.
The systematic absences in the diffraction data were consistent with
space groups Pca21 and Pbcm for 2, uniquely, P21/n for 3, and C2,
Cm, and C2/m for 4. Only the noncentrosymmetric space group option
in 4 yielded chemically reasonable and computationally stable results
of refinement. Two symmetry unique but chemically similar molecules
of 4 were located in the asymmetric unit. An inspection of the packing
diagram does not show evidence of overlooked symmetry. The Flack
parameter refined to nil, indicating the true hand of the data had been
determined.
After thorough exploration of structural solutions in the space group
options for 4, only the solution in C2/m yielded chemically reasonable
and computationally stable results. The molecule in 4 was located at a
two-fold axis and a mirror plane. Each Ta atom has two bridging O
atoms, two terminal NMe2 groups, and a chelating η2-N(Me)CH2NMe2
ligand. One NMe2 group is disordered by the mirror plane with part of
the chelating η2-N(Me)CH2NMe2 ligand.
The final least-squares refinement of 381 parameters against 10805
data resulted in residuals R (based on F2 for I g 2σ) and wR (based on
F2 for all data) of 0.0201 and 0.0505, respectively.
Non-hydrogen atoms were anisotropically refined. All hydrogen
atoms were treated as idealized contributions. Empirical absorption
correction was performed with SADABS.28a In addition, the global
refinements for the unit cells and data reductions of the two structures
were performed using the Saint program (version 6.02). All calculations
were performed using SHELXTL (version 5.1) proprietary software
package.28b
Determination of X-ray Crystal Structure of 5. A colorless crystal
with approximate dimensions 0.38 × 0.34 × 0.31 mm3 was selected
under oil under ambient conditions and attached to the tip of a nylon
loop. The crystal was mounted in a stream of cold nitrogen at 100(2)
K and centered in the X-ray beam by using a video camera. The crystal
evaluation and data collection were performed on a Bruker CCD-1000
diffractometer with Mo KR (λ ) 0.71073 Å) radiation and the
diffractometer to crystal distance of 4.9 cm.
The initial cell constants were obtained from three series of ω scans
at different starting angles. Each series consisted of 20 frames collected
at intervals of 0.3° in a 6° range about ω with the exposure time of 10
s per frame. A total of 64 reflections were obtained. The reflections
were successfully indexed by an automated indexing routine built in
the SMART program. The final cell constants were calculated from a
set of 20191 strong reflections from the actual data collection.
Computational Details. All calculations were performed using the
Gaussian 03 package31 with density functional theory (DFT) at the
B3LYP level. Geometries were optimized with the following basis
set: LanL2DZ with f polarization functions for Ta32 and 6-31G* for
the rest of elements. Vibration frequency calculations were performed
on all the optimized structures with the same method to provide free
energies at 298.15 K. The optimized minima and the transition structures
have been confirmed by harmonic vibration frequency calculations. The
calculated relative electronic energies shown in the text have been
corrected with zero-point energies (ZPE). The minimum energy crossing
point (MECP) was located with the code from Harvey.20
Acknowledgment is made to the National Science Founda-
tion (CHE-0516928 to Z.X.), Research Grant Council of Hong
Kong (HKUST6083/02M to Y.D.W. and HKUST 602304 to
Z.L.), Camille Dreyfus Teacher-Scholar program (Z.X.), and
the Royal Society (United Kingdom) Kan Tong Po Visiting
Professorship for financial support of this work. We thank
(29) SADABS V.2.05, SAINT V.6.22, SHELXTL V.6.10, and SMART 5.622
Software Reference Manuals: Bruker-AXS: Madison, Wisconsin, U.S.A.,
2000-2003.
(30) Spek, A. L. Acta Crystallogr. 1990, A46, C34.
(28) (a) Sheldrick, G. M. SADABS, A Program for Empirical Absorption
Correction of Area Detector Data; University of Go¨ttingen: Go¨ttingen,
Germany, 2000. (b) Sheldrick, G. M. SHELXL-97, A Program for the
Refinement of Crystal Structures; University of Go¨ttingen: Go¨ttingen,
Germany, 1997.
(31) Frisch, M. J.; et al. Gaussian 03, Revision B.03; Gaussian, Inc.: Pittsburgh,
PA, 2003. A complete author list is given in the Supporting Information.
(32) Ehlers, A. W.; Bo¨hme, M.; Dapprich, S.; Gobbi, A.; Ho¨llwarth, A.; Jonas,
V.; Ko¨hler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G. Chem. Phys.
Lett. 1993, 208, 111.
9
14420 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

Reaction of Ta(NMe2)5 with O2
A R T I C L E S
Professor Jeremy Harvey for sharing his program code to locate
MECP, Prof. Liguo Song for the MS measurements, and Dr.
Ruitao Wang for help.
of the reactions between 1 and O2 in the presence of TEMPO,
crystallographic data of 2-5, the optimized structures with total
energies of the species in the proposed pathways, and complete
ref 31. This material is available free of charge via the Internet
at http://pubs.acs.org.
Supporting Information Available: Experimental Section for
MS analyses, 2D HMQC NMR spectra of 3 and 4, variable-
temperature and EXSY NMR spectra of 3, 1H NMR monitoring
JA075076A
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14421