Evidence for strong tantalum-to-boron dative interactions in (silox) 3Ta(BH3) and (silox)3Ta(η2-B,Cl- BCl2Ph) (silox = tBu3SiO)1

Article abstract of DOI:10.1021/ic0616885

Treatment of (silox)₃Ta (1, silox = ^tBu ₃SiO) with BH₃·THF and BCl₂Ph afforded (silox)₃Ta(BH₃) (2) and (silox)₃Ta- (η²-B,Cl-BCl₂Ph) (3), which are both remarkably stable Ta(III) compounds. NMe₃ and ethylene failed to remove BH₃ from 2, and no indication of BH₃ exchange with BH ₃·THF-d₈ was noted via variable-temperature ¹H NMR studies. Addition of BH₃·THF to (silox) ₃TaH₂ provided the borohydride-hydride (silox) ₃HTa(η³-BH₄) (5), and its thermolysis released H₂ to generate 2. Exposure of 2 to D₂ enabled the preparation of isotopologues (silox)₃Ta(BH_3-nD _n) (n = 0, 2; 1, 2-D; 2, 2-D₂; 3, 2-D₃) for isotopic perturbation of chemical shift studies, but these failed to distinguish between inverse adduct (i.e., (silox)₃Ta→BH ₃) or (silox)₃Ta(η²-B,H-BH₃) forms of 2. Computational models (RO)₃Ta(BH₃) (R = H, 2′; SiH₃,2^SiH SiMe_3,2^SiMe, and Si^tBu₃, 2^SiBu) were investigated to assess the relative importance of steric and electronic effects on structure and bonding. With small R, η²-B,H structures were favored, but for 2 ^SiMe and 2^SiBu, the dative structure proved to be similar in energy. The electonic iand vibrational features of both structure types were probed. The IR spectrum of 2 was best matched by the η²-B,H conformer of 2^SiBu. In related computations pertaining to 3, small R models favored the oxidative addition of a BCl bond, while with R = Si ^tBu₃ (3^SiBu), an excellent match with its X-ray crystal structure revealed the critical steric influence of the silox ligands.

Full text of DOI:10.1021/ic0616885

Inorg. Chem. 2007, 46, 1222−1232
Evidence for Strong Tantalum-to-Boron Dative Interactions in
(silox)3Ta(BH3) and (silox)3Ta(
2-B,Cl-BCl2Ph) (silox tBu3SiO)1
η
)
Jeffrey B. Bonanno,† Thomas P. Henry,† Peter T. Wolczanski,*,† Aaron W. Pierpont,‡ and
Thomas R. Cundari*,‡
Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell UniVersity, Ithaca,
New York 14853, and Department of Chemistry, Center for AdVanced Scientific Computing and
Modeling (CASCaM), UniVersity of North Texas, Denton, Texas 76203-5070
Received September 6, 2006
Treatment of (silox)3Ta (1, silox
2-B,Cl-BCl2Ph) (3), which are both remarkably stable Ta(III) compounds. NMe3 and ethylene failed to remove
BH3 from 2, and no indication of BH3 exchange with BH3·THF-d8 was noted via variable-temperature 1H NMR
studies. Addition of BH3·THF to (silox)3TaH2 provided the borohydride-hydride (silox)3HTa(
3-BH4) (5), and its
thermolysis released H2 to generate 2. Exposure of to D2 enabled the preparation of isotopologues
(silox)3Ta(BH3 nDn) (n 0, 2; 1, 2-D; 2, 2-D2; 3, 2-D3) for isotopic perturbation of chemical shift studies, but these
failed to distinguish between “inverse adduct” (i.e., (silox)3Ta BH3) or (silox)3Ta(
2-B,H-BH3) forms of 2. Computational
models (RO)3Ta(BH3) (R H, 2
; SiH3, 2SiH SiMe3, 2SiMe, and SitBu3, 2SiBu) were investigated to assess the relative
importance of steric and electronic effects on structure and bonding. With small R,
2-B,H structures were favored,
but for 2SiMe and 2SiBu, the dative structure proved to be similar in energy. The electonic and vibrational features
of both structure types were probed. The IR spectrum of 2 was best matched by the
2-B,H conformer of 2SiBu. In
) ‚THF and BCl2Ph afforded (silox)3Ta(BH3) (2) and (silox)3Ta-
tBu3SiO) with BH3
(
η
η
2
)
-
f
η
)
′
η
η
related computations pertaining to 3, small R models favored the oxidative addition of a BCl bond, while with R
)
SitBu3 (3SiBu), an excellent match with its X-ray crystal structure revealed the critical steric influence of the silox
ligands.
Introduction
BX3 (X ) halide, etc.) or BR3 (R ) H, alkyl, etc.) are likely
candidates. In fact, there are a number of isolable late metal
compounds claimed to possess purely dative interactions of
the LnMfBX3/R3 type,3 including [Et4N][Mn(BH3)(CO)4-
PPh3],4 Ir{B(C6F5)3}(CO)Cl(PEt3)2, and [trans-Rh(BF3)2-
(diphos)2][BPh4].5 Unfortunately, the lack of structural data
pertaining to these adducts has called their bonding descrip-
tions into question.
Adducts of the type LnMf(HBX2) have been prepared
and structurally characterized, but in all cases there is an
interaction with the hydrogen. In cases where the LnM
fragment is 16e-, the bonds have been construed as “normal”
dative interactionssones in which the H-B electron pair
Transition metal complexes are usually comprised of
ligands, both charged and neutral, that donate pairs of
electrons to the more electropositive metal, i.e., X-/LfM.2
In principle, electronic deficient groups can also be used to
accept electron density from the metal in forming the metal-
ligand bond. While this is common for π-interactionsshence
the term “π-backbonding” for describing π-bonds in CO,
CN-, olefins, etc.2sit is quite unusual for a group to act as
a σ-acceptor in the same vein.
Ligands whose core atom violates the octet rule have the
greatest potential to interact in a σ-accepting fashion, and
* To whom correspondence should be addressed. E-mail: ptw2@
cornell.edu (P.T.W.).
(3) Gilbert, K. B.; Boocock, S. K.; Shore, S. G. In ComprehensiVe
Organometallic Chemistry; Wilkinson, G.; Stone, F. G. A.; Abel, E.
W., Eds.; Pergamon: Oxford, 1982; Vol. 6, Chapter 41.
(4) Parshall, G. W. J. Am. Chem. Soc. 1964, 86, 361-364.
(5) (a) Scott, R. N.; Shriver, D. F.; Vaska, L. J. Am. Chem. Soc. 1968,
90, 1079-1080. (b) Scott, R. N.; Shriver, D. F.; Lehmann, D. D. Inorg.
Chim. Acta. 1970, 4, 73-78. (c) Lehmann, D. D.; Shriver, D. F. Inorg.
Chem. 1974, 13, 2203-2207.
† Cornell University.
‡ University of North Texas.
(1) Taken in part from: Bonanno, J. B., Ph.D. Thesis, Cornell University,
1996.
(2) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; University
Science Books: Sausilito, CA, 1987.
1222 Inorganic Chemistry, Vol. 46, No. 4, 2007
10.1021/ic0616885 CCC: $37.00
© 2007 American Chemical Society
Published on Web 01/24/2007

EWidence for Strong Tantalum-to-Boron DatiWe Interactions
donates to the metal center6sbut the boron atoms are
typically well within bonding distance, and the stability of
these complexes suggests that metal donation to boron is an
integral part of the bonding, i.e., 3-center-4-electron bonding.
Depending on metal and environment, the interactions can
range from being dominated by H-B donation to the
formation of boryl-hydrides with7,8 or without9 a long
distance H‚‚‚B remnant. Other examples reveal additional
diversity. For the 14e- bent sandwich fragment Cp2Ti, two
catecholborane ligands appear to bind via their BH bonds,
i.e., Cp2Ti(η2-H,B-HB(O2C6H4))2, but there are 6e- and 5
atoms involved in the bonding, and a rudimentary MO
assessment suggests titanium donation to both borons in a
single molecular orbital.10 Isotopic perturbation of chemical
shift11 studies have suggested that Cp2Nb(H2B(catechol))
exists in solution as an equilibrium between hydrido-borane
and borohydride forms, while crystallographic studies showed
a single species that appeared to reside on the structural
continuum between these limiting cases.12,13 In very few
instances is the boron so distant from the transition metal
that it is more correctly described as a bridging hydride in a
3-center-2-electron bond.14
Cl,19 {κ3-S,S,B-B(mimR)3}Ir(CO)(PPh3)H (R ) Me,21 Ph,
tBu),20 [{κ4-B(mimMe)3}Pt(PPh3)H]Cl, and {κ4-B(mimMe)3}-
Pt(PPh3),22 the boron is centered over the metal by the three
chelate mercaptoimidazole arms of the tripod in the κ4-
arrangement and two in the κ3-form. Parkin has reviewed
these species and has painstakingly delineated the bonding
in the pseudo-d8 Rh and Ir compounds through calculations
and by clearly defining valence and dn, where n ) no.
valence electrons(neutral atom) - valence. The latter cir-
cumvents discrepancies in rationalizing common structure
types vs dn when the oxidation state formalism is mistakenly
utilized,23 which is a problem for LnMfBX3 species, given
the neutral view of the metal-boron interaction.
The synthesis of (silox)3Ta(BH3) (2, silox ) tBu3SiO), an
isotopic perturbation of chemical shift study, derivatives, and
calculations are reported herein. The unique structural
features of the siloxide ligand apparently sway the bonding
of the borane toward the “inverse adduct” or TafBH3
structure, although it is still possible that the alternative
TafB(µ-H)H2 structure represents the true ground state. A
related η2-B,Cl configuration is found in (silox)3Ta(η2-B,Cl-
BCl2Ph) (3).
LnMfBX3 interactions devoid of any ambiguity regarding
η2-B,H or bridging atoms have recently been observed via
BH bond activation of the tris(2-mercapto-1-R-imidazolyl)-
hydroborato ligand (HB(mimR)3) to afford metalaboratrane
and related complexes. In such derivatives as {k4-B(mimtBu)3}-
Fe(CO)2,15 {κ4-B(mimMe)3}M(CO)PPh3 (M ) Ru,16 Os),17
[{κ4-B(mimtBu)3}Co(PPh3)]BPh4,18 {κ4-B(mimR)3}M(Cl)PPh3
Results
Boron Adducts of (silox)3Ta. 1. Synthesis. As Scheme
1 indicates, treatment of diethyl ether solutions of (silox)3Ta
(1)24 with excess BH3·THF (1 M in THF) at -78 °C resulted
in an immediate color change from blue to orange. The BH3
adduct (silox)3Ta(BH3) (2) was isolated from hexane solution
t
(R ) Me, M ) Rh;19 R ) Bu, M ) Rh, Ir),20 [{κ4-
1
as orange crystals in 67% yield. The H NMR spectrum of
B(mimMe)3}Rh(CNR)PPh3]Cl,19 [{κ4-B(mimMe)3}Rh(PMe3)2]-
2 in benzene-d6 exhibited a sharp resonance at δ 1.28 and a
broad 1:1:1:1 quartet at δ 3.93 (1JBH ) 98 Hz) in a ratio of
27:3, while the 11B NMR spectrum revealed a 1:3:3:1 quartet
at δ 27.89 (1JBH ) 98 Hz).25 Single-frequency decoupling
of 11B collapsed the proton signal to a sharp singlet. The
infrared spectrum of 2 in nujol revealed two terminal BH
stretching absorptions at νBH ) 2445 and 2395 cm-1, and a
weak absorption at 1290 cm-1 attributed to BH bending.25
In order to corroborate these assignments, a THF solution
of 1 was treated with 0.5 equiv of B2D6 (presumably
generating BD3·THF in situ) to provide (silox)3Ta(BD3) (2-
D3). Its IR spectrum (nujol) manifested sharp absorptions at
1845 and 1770 cm-1 (νBH/νBD ) 1.32 and 1.35, respectively)
(6) Schlecht, S.; Hartwig, J. F. J. Am. Chem. Soc. 2000, 122, 9435-9443.
(7) (a) Lantero, D. R.; Miller, S. L.; Cho, J. Y.; Ward, D. L.; Smith, M.
R., III. Organometallics 1999, 18, 235-247. (b) Lantero, D. R.; Ward,
D. L.; Smith, M. R., III. J. Am. Chem. Soc. 1997, 119, 9699-9708.
(c) Lantero, D. R.; Motry, D. H.; Ward, D. L.; Smith, M. R., III. J.
Am. Chem. Soc. 1994, 116, 10811-10812.
(8) Hartwig, J. F.; Cook, K. S.; Hapke, M.; Incarvito, C. D.; Fan, Y. B.;
Webster, C. E.; Hall, M. B. J. Am. Chem. Soc. 2005, 127, 2538-
2552.
(9) Hartwig, J. F.; He, X. M. Organometallics 1996, 15, 5350-5358.
(10) (a) Muhoro, C. N.; He, X. M.; Hartwig, J. F. J. Am. Chem. Soc. 1999,
121, 5033-5046. (b) Hartwig, J. G.; Muhoro, C. N. Organometallics
2000, 19, 30-38.
(11) (a) Saunders, M.; Telkowski, L.; Kates, M. R. J. Am. Chem. Soc. 1977,
99, 8070-8071. (b) Saunders, M.; Kates, M. R. J. Am. Chem. Soc.
1977, 99, 8071-8072. (c) Saunders, M.; Kates, M. R.; Wiberg, K.
B.; Pratt, W. J. Am. Chem. Soc. 1977, 99, 8072-8073.
(12) Hartwig, J. F.; De Gala, S. R. J. Am. Chem. Soc. 1994, 116, 3661-
3662.
accompanied by a weak absorption at 980 cm-1 (νBH/νBD
)
1.32). The ratios are in good agreement with the value
predicted by a reduced mass calculation (νBH/νBD ) 1.36)26
and are therefore considered simple BH vibrations without
strong coupling to other modes. The spectral characterizations
(13) Jensen, J. A.; Wilson, S. R.; Girolami, G. S. J. Am. Chem. Soc. 1988,
110, 4977-4982.
(14) Shimoi, M.; Nagai, S.; Ichikawa, M.; Kawano, Y.; Katoh, K.; Uruichi,
M.; Ogino, H. J. Am. Chem. Soc. 1999, 121, 11704-11712.
(15) Figueroa, J. S.; Melnick, J. G.; Parkin, G. Inorg. Chem. 2006, 45,
7056-7058.
(21) Crossley, I. R.; Hill, A. F.; Willis, A. C. Organometallics 2005, 24,
1062-1064.
(16) Foreman, M. R. St.-J.; Hill, A. F.; Owen, G. R.; White, A. J. P.;
Williams, D. J. Organometallics 2003, 22, 4446-4450.
(17) Foreman, M. R. St.-J.; Hill, A. F.; White, A. J. P.; Williams, D. J.
Organometallics 2004, 23, 913-916.
(22) Crossley, I. R.; Hill, A. F. Organometallics 2004, 23, 5656-5658.
(23) (a) Parkin, G. J. Chem. Educ. 2006, 83, 791-799. (b) Parkin, G.
Organometallics 2006, 25, 4744-4747.
(18) Mihalcik, D. J.; White, J. L.; Tanski, J. M.; Zakharov, L. N.; Yap, G.
P. A.; Incarvito, C. D.; Rheingold, A. L.; Rabinovich, D. Dalton Trans.
2004, 1626-1634.
(24) Neithamer, D. R.; LaPointe, R. E.; Wheeler, R. A.; Richeson, D. S.;
Van Duyne, G. D.; Wolczanski, P. T. J. Am. Chem. Soc. 1989, 111,
9056-9072.
(19) Crossley, I. R.; Hill, A. F.; Willis, A. C. Organometallics 2006, 25,
289-299.
(25) Marks, T. J.; Kolb, J. R. Chem. ReV. 1977, 77, 263-293.
(26) Silverstine, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric
Identification of Organic Compounds; John Wiley and Sons: New
York, 1981; Chapter 3.
(20) Landry, V. K.; Melnick, J. G.; Buccella, D.; Pang, K.; Ulichny, J. C.;
Parkin, G. Inorg. Chem. 2006, 45 2588-2597.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1223

Bonanno et al.
Scheme 1
suggested that 2 was an “inverse adduct” of the type
(silox)3TafBH3, but the possibility of a Ta‚‚‚H‚‚‚B interac-
tion(s) remained.
orange color was consistent with a dative interaction (i.e.,
(silox)3TafBCl2Ph), but assignment of ν(BCl) bands29 in
the IR proved to be difficult due to overlapping siloxide
absorptions. Without IR data pertaining to BCl connectivity,
it was not possible to unambiguously assign a structure to
3, especially with the potential for µ-Cl interactions. Ther-
molysis of 3 at 60 °C for 2 days afforded (silox)3TaCl224 as
∼60-70% of the product mixture, and the fate of the PhB
fragment was undetermined.
2. Structure of (silox)3Ta(η2-B,Cl-BCl2Ph). A single-
crystal X-ray structure determination of 3 revealed an η2-
B,Cl binding mode in the pseudo-tetrahedral structure
illustrated in Figure 1. The d(B1-Cl1) distance of 1.999(8)
Å is markedly longer than its terminal partner (i.e., B1-
Cl2), which is 1.828(8) Å. The Ta-B1-Cl1 plane is directly
aligned with the Ta-O3 bond and bisects the O1-Ta-O2
angle (i.e., Cl1-Ta-O1 ) 87.65(14)°, Cl1-Ta-O2 )
87.74(14)°). Steric and electronic influences of Cl1 provide
a reason for the greater spread of O1-Ta-O2 (123.5(2)°)
relative to O1-Ta-O3 (105.4(2)°) and O2-Ta-O3
(106.6(2)°). The TafB dative interaction of 2.308(8) Å is
The BH3 ligand of 2 is strongly bound to tantalum and is
quite inert. When 2 was heated in benzene-d6 at 90 °C for
24 h, no decomposition was noted. Under similar conditions,
2 was treated with 3 equiv of C2H4, but there was no evidence
of any BEt groups, which would have been expected if BH3
had been liberated.27 In the presence of ∼10 equiv of NMe3,
no degradation of 2 was noted after a week at 23 °C.28
Exchange experiments between 2 and BH3·THF in THF-d8
(1:1, 0.029 M) were monitored by 11B NMR spectroscopy,
and no hint of line broadening (toward coalescence) was
observed up to 80 °C, placing a very conservative lower limit
on dissociative (or pseudo-dissociative) exchange at >19
kcal/mol. Even in a donor solvent (THF) that could facilitate
BH3 loss in an associative (pseudo-dissociative in the case
of a donor solvent) fashion, the BH3 unit appears to to be
very difficult to remove. This is best rationalized in terms
of thermodynamic stability.
Efforts to extend the range of these adducts to include
BR3 (R ) Me, Ph) species failed, with no noticeable binding
detected, and simple HBR2 and H2BR species were avoided
due to probable complications from H/R exchange. Instead,
124 was treated with 1 equiv of PhBCl2 at 23 °C in toluene
(Scheme 1), and the solution quickly turned from blue to
deep red. The addition of hexane at -78 °C precipitated
orange crystals of (silox)3Ta(η2-B,Cl-BCl2Ph) (3) in 69%
yield. In C6D6, the 1H NMR spectrum of 3 displayed a sharp
singlet at δ 1.26 and three aryl resonances in a 1:2:2 ratio.
Three aryl 13C{1H} NMR spectral resonances were observed
in THF-d8; presumably the ipso resonance (BC) was too
broadened by 11B coupling (11B, I ) 3/2, 80%; 10B, I ) 3,
20%) to be clearly seen. While the oxidative addition of a
B-Cl bond was considered, these bonds are quite strong,
and the orange color of 3 belied this possibility since most
Ta(V) species characterized in this system are colorless. The
significantly longer than the sum of covalent radii (rTa
)
1.34 Å, rB ) 0.82 Å, ∑(rTa + rB) ) 2.16 Å), perhaps due to
the strain induced by the Ta-Cl1 interaction of 2.561 Å,
which is also substantially longer than a normal covalent
bond (rCl ) 0.99 Å, ∑(rTa + rCl) ) 2.33 Å). The Ta-B1
interaction clearly perturbs the planarity about boron, since
the Cl1-B-Cl2 (107.5(4)°), C1-B-Cl2 (114.4(5)°),
and C1-B-Cl1 (110.9(5)°) angles sum to only 332.8°.
Note that the boron is close to occupying a regular position
in a tetrahedron, since the B1-Ta-O3 angle of 101.2(2)°
is only somewhat less than the B1-Ta-O1 and B1-
Ta-O2 angles of 108.4(2)° and 109.4(2)°, respectively.
Borohydride (silox)3HTa(η3-BH4). 1. Synthesis. When
(silox)3TaH2 (4), generated from dihydrogen and 1 in diethyl
ether,30 was exposed to 1 equiv of BH3·THF at 23 °C, the
colorless borohydride complex (silox)3HTa(η3-BH4) (5) was
1
isolated in 77% yield (Scheme 1). The H NMR spectrum
(27) Zaidlewicz, M. In ComprehensiVe Organometallic Chemistry; Wilkin-
son, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon: Oxford, 1982;
Vol. 7, Chapter 45.2.
of 5 consisted of a singlet at δ 1.29 corresponding to the
(28) For, example, NMe3 abstracts 1 or 2 equiv BH3 from Cp2Zr(BH4)2 to
yield the hydride/borohydride or dihydride complexes: James. B. D.;
Nanda, R. K.; Wallbridge, M. G. H. Inorg. Chem. 1967, 6, 1979-
1983.
(29) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coor-
dination Compounds; Wiley: New York, 1986.
(30) Miller, R. L.; Toreki, R.; LaPointe, R. E.; Wolczanski, P. T.; Van,
Duyne, G. D.; Roe, D. C. J. Am. Chem. Soc. 1993, 115, 5570-5588.
1224 Inorganic Chemistry, Vol. 46, No. 4, 2007

EWidence for Strong Tantalum-to-Boron DatiWe Interactions
Figure 2. Possible “inverse adduct” and η2-B,H structures for (silox)3Ta-
(BH3) (2).
deformation, were characteristic of the η3-borohydride
ligand.25 A strong band at 1765 cm-1 was assigned to the
ν(TaH). Treatment of (silox)3TaD2 (4-D2) with B2D6 in THF
afforded (silox)3DTa(η3-BD4) (5-D5), which exhibited boro-
hydride absorptions at 1870, 1630, and 1580 cm-1 corre-
sponding to BD stretching in accord with νBH/νBD ratios of
1.33, 1.35, and 1.36, respectively. The ν(TaD) appeared at
1270 cm-1 (νBH/νBD ) 1.39).
2. Equilibration with (silox)3Ta(BH3). In contrast to 2,
exposure of 5 to 1 equiv of NMe3 at 23 °C for ∼3 h
completely regenerated dihydride 4 with concomitant forma-
tion of Me3N·BH3.28 The reversibility of η3-BH4 formation
was expected, but the thermal instability of 5 proved to be
due to H2 loss and not loss of BH3. White crystals of 5 stored
at 25 °C for a number of weeks discolored and 2 was
observed in the 11B NMR spectrum of the resulting material.
When 5 was heated under ∼3 atm of D2 (90 °C, 18 h), 2,
(silox)3Ta(BH2D) (2-D), (silox)3Ta(BHD2), and (silox)3Ta-
(BD3) were observed (IR, NMR, vide infra). The addition
of H2 (∼3 atm) to solutions of 2 at 90 °C established an
equilibrium with 5. Since 4 rapidly undergoes σ-bond
metathesis with D2 to afford (silox)3TaHD (4-D) and 4-D2,30
it is plausible that H/D exchange in 5 occurs via BH3 loss.
However, the conversion of 2 to 5 under H2 (and reverse
process in the solid state) suggests that there may be a direct
process involving an H2/D2 addition across the Ta-B bond
of 2, and the reverse H2/HD/D2 elimination from isotopomers
of 5.
Figure 1. Molecular and core illustrations of (silox)3Ta(η2-B,Cl-BCl2Ph)
(3). Bond distances (Å) are in black and bond angles (deg) in red.
Interatomic distances (Å): d(Ta-C1)
3.493(4).
)
3.486(11), d(Ta-Cl2)
)
silox ligand, a broad downfield quintet at δ 22.47 (JHH
)
Isotopic Perturbation of Chemical Shift Studies on
(silox)3Ta(BH3). The spectral characterization of 2 is con-
sistent with a symmetric C3V adduct containing a TafB
dative interactionsan “inverse adduct”. However, an η2-
B,H32 structure where rapid interconversion of the three
configurations illustrated in Figure 2 averages the bridge and
terminal boron hydride positions is certainly feasible. The
IR spectra for such species should have two bands in the
terminal BH stretching region and one in the bridging region.
The IR spectra do effectively rule out the possibility of
rapidly exchanging (silox)3HTaBH2 structures, since an
intense ν(TaH) absorption band is absent. The two bands
observed in the IR spectrum of 2 may account for the
terminal stretching absorptions, while a bridging stretch may
be broad and weak and thus difficult to observe. NMR
12 Hz) characteristic of the hydride, and a broad unresolved
1:1:1:1 quartet at δ 3.48 that sharpened at 50 °C to enable
identification of a 1JBH of 88 Hz. A quintet in the 11B NMR
spectrum at δ -16.72 also displayed the 88 Hz coupling.
Rapid interconversion of bridging and terminal hydrogens
on the NMR time scale is common in a borohydride
ligand,25,31 hence the observation of equivalent hydrogens
on boron.
One absorption at 2495 cm-1 in the terminal BH stretching
region of the IR spectrum, two in the BH bridging region
(2200, 2150 cm-1) and another broad weak absorption at
1160 cm-1, which was tentatively assigned to a BH bridge
(31) For, a classic example where terminal/bridged hydrogen exchange is
slow (Cp2V(η2-BH4), ∆Gqex(-87 °C) ≈ 8 kcal/mol) on the NMR
timescale, see: Marks, T. J.; Kennelly, W. J. J. Am. Chem. Soc. 1975,
97, 1439-1443.
(32) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983, 250, 395.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1225

Bonanno et al.
(silox)3Ta(η2-B,Hb-B(Ht)(Dt)) (2-Dt) {K1
\}
spectral techniques for detecting such species include vari-
able-temperature acquisition and the isotopic perturbation of
chemical shifts.12 1H{11B} NMR spectra of 2 obtained in
THF-d8 from 22 to -105 °C showed a single B-H signal
that shifts slightly (∆δ ) -0.1) and only broadens noticeably
at -90 °C. Although these results also point to a static C3V
structure, interconversion of η2-B,H-bridging and terminal
BH positions is expected to have a very small barrier and
may still be rapid at the lowest accessible temperature for
this experiment.
(silox)3Ta(η2-B,Db-B(Ht)2) (2-Db) (2)
δ1 ) (2K1δt + δt + δb)/(2K1 + 2) (3)
(silox)3Ta(η2-B,Hb-B(Dt)2) (2-(Dt)2) {K2
\}
(silox)3Ta(η2-B,Db-B(Ht)(Dt)) (2-DbDt) (4)
δ1 ) (K2δt + δb)/(K2 + 1) (5)
Originally applied in carbocation chemistry,12 isotopic
perturbation of chemical shift (initially isotopic perturbation
of resonance (IPR)) experiments were first successfully
employed in an organometallic system by Shapley et al.,33
who used it to deduce the η2-C,H-bridge of a methyl group
in HOs3(CO)10(µ1,2-H,C-CH3). The experiment takes advan-
tage of the greater propensity of deuterium to reside in
terminal sites due to their greater zero point energy vs that
of a bridging site. The analogy to the plausible (silox)3Ta-
(η2-B,H-BH3) structure of 2 is clear; in a set of isotopologues
(silox)3Ta(BHxD3-x), the deuterium should prefer the terminal
sites, and if the intrinsic 1H chemical shift difference between
the bridging and terminal sites is sufficient, the experiment
could provide proof of the η2-B,H structure. If the symmetric,
observables (δ0, δ1, and δ2), but if the “primary” equilibrium
isotope effects for K1 and K2, i.e., the bridge vs terminal
positioning of H vs D, are considered the same and the
secondary equilibrium isotope effects (H vs D in the adjacent
terminal position) are considered negligible or the same,35
then K2 ) 4K1. Unfortunately, the fact that δ0 - δ1 ) 0.025
ppm is equal to δ1 - δ2 imparts conlinearity on the solutions
to these equations. For comparative purposes, the equations
were solved using δ2 values of δ2 ( 0.005, a reasonable
error in the measured chemical shift of 2-D2. With δ2 ) 3.875
(i.e., δ1 - δ2 > δ0 - δ1), the equations yielded the solution
δb ) 3.38, δt ) 4.205, K1 ) 0.375, and K2 ) 1.50. Solving
with δ2 ) 3.885 (i.e., δ1 - δ2 < δ0 - δ1), the calculated
values were δb ) 4.38, δt ) 3.705, K1 ) 0.6875, and K2 )
2.75. The solutions are highly divergent. With δ1 - δ2 > δ0
- δ1, the chemical shift of the bridging BH appears upfield
of the chemical shift of the terminal BH, but their positions
are reversed for δ1 - δ2 < δ0 - δ1. The equilibrium
constants favor H-bridging in the former, and D-bridging in
the latter. Since one would expect chemical shifts of the
terminal positions to be downfield of those corresponding
to bridging positions and the K to favor µ-H over µ-D,36 the
first set of data are the better of the two guesses. However,
the results appear unlikely for several reasons: (1) the two
divergent solutions are within error of the shift measurement;
(2) K1 and K2 represent large deviations from statistical values
1
dative structure is the ground state, the H NMR signals of
each (silox)3Ta(BHxD3-x) should be shifted upfield by
small increments due to subtle changes in magnetic
environment.34
A sample for isotopic perturbation of chemical shift study
was prepared by thermolysis of 2 in C6D6 under ∼3 atm of
1
D2 for 18 h. Inspection of the H{11B} NMR spectrum of
this sample revealed resonances at δ 3.930, 3.905, and 3.880
that were attributed to 2, 2-D, and 2-D2. HD coupling was
obscured in this experiment as a result of boron decoupling,
and neither JHD nor JBH values could be extracted from the
1
corresponding H NMR spectrum due to the broad line
of 0.5 and 2.0, respectively (at 25 °C, ∆G°1 - ∆G°stat
)
widths of the overlapping quartets. Immediately, the validity
of the isotopic perturbation of chemical shift experiment
was called into question due to the proximity of the
three chemical shifts, but the analysis was carried out to
completion.
The equilibrating η2-B,H structures could have small or
large shifts upfield or downfield depending on the associated
equilibrium constants and the shifts of the terminal and
bridging positions. The equilibria and pertinent equations for
the chemical shift analysis are given in eqs 1-5, where the
δt and δb subscripts indicate intrinsic terminal and bridge
shifts, and δn (n ) 0-2) is the average chemical shift for a
particular number of deuteriums. The equations present four
unknowns (δt, δb, K1, and K2) and three
∆G°2 - ∆G°stat ) 189 cal/mol; (3) the chemical shift
differences between terminal and bridging sites are quite
modest. While the ∆G° values appear small in magnitude,
they are larger than bona fide cases of isotopic perturbation
of chemical shift (e.g., HOs3(CO)10(µ1,2-H,C-CH3), ∆G° ≈
130 cal/mol;33 [Tp(PMe3)Ir(H2)H]BF4, ∆G° ≈ 121 cal/mol)37
whose observed δn (n ) 0-2 values were substantially
greater than those for 2-Dn and whose δ1 - δ2 > δ0 - δ1.
The data, while inconclusive, support the unbridged, dative
structure of 2.
Discussion
General. The stability of 2 is remarkable in the context
of the dehydrogenation of 5 and its lack of BH3 transfer
(silox)3Ta(η2-B,Hb-B(Ht)2) (2) has δ0 ) (2δt + δb)/3
(1)
(35) Carpenter, B. K. Determination of Reaction Mechanisms; Wiley-
Interscience: New York, 1984.
(36) (a) Bigeleisen, J.; Mayer, M. G. J. Chem. Phys. 1947, 15, 261-267.
(b) Melander, L.; Saunders, W. H. Reaction Rates of Isotopic
Molecules; Wiley: New York, 1980.
(37) Heinekey, D. M.; Oldham, Jr., W. J. J. Am. Chem. Soc. 1994, 116,
3137-3138.
(33) Shapley, J.; Calvert, R. B. J. Am. Chem. Soc. 1978, 100, 7726-7727.
(34) (a) Lambert, J. B.; Greifenstein, L. G. J. Am. Chem. Soc. 1974, 96,
5120-5124. (b) Luo, X. L.; Crabtree, R. H. J. Am. Chem. Soc. 1990,
112, 4813-4821.
1226 Inorganic Chemistry, Vol. 46, No. 4, 2007

EWidence for Strong Tantalum-to-Boron DatiWe Interactions
Table 1. Relative Conformational Energies (∆E, kcal/mol) Pertaining to Calculational Models of (RO)3Ta(BH3)
a Rearranged to (RO)2Ta(η3-H,H,B-BH3) upon geometry optimization. b Only stable minimum obtained for 2′. c Conformation converts upon geometry
optimization to a transition state between dative, (RO)3TafBH3 and (RO)2Ta(η2-H,B-BH3) structures.
capabilities. Unfortunately, the conformation of the BH3 unit
is still ambiguous, hence computational methods were
employed in an effort to establish its ground state structure.
As for the Cl2BPh adduct, 3, its η2-B,Cl binding mode looks
like an arrested oxidative addition of the B-Cl bond, hence
calculations were utilized in examining some of the energet-
ics surrounding this plausible event.
Computational Models of 2. 1. Ground State Structure.
Density functional calculations (B3LYP/CEP_31G(d)) were
conducted on (RO)3Ta(BH3) with sequentially larger sub-
stituents (R ) H, 2′; SiH3, 2SiH SiMe3, 2SiMe, and SitBu3,
2SiBu) in order to assess the relative importance of steric and
electronic effects on structure and bonding. DFT geometry
optimizations on the small models 2′ and 2SiH indicated a
preference for structures with one and two η2-B,H interac-
Figure 3. DFT-optimized geometry of “inverse adduct” conformer of
tions, as Table 1 indicates. Since η2-B,H conformers contain
(silox)3TafBH3 (2SiBu); methyl groups have been omitted for clarity.
Interatomic distances (Å) and angles (deg): TaB ) 2.22; BH ) 1.23;
Ta‚‚‚H ) 2.78-2.80; O-Si ) 1.71; Ta-O ) 1.89; Ta-O-Si ) 179; TaBH
) 104-105; O-Ta-B ) 100 -101; O-Ta-O ) 117.
hydrogen atoms in close proximity to the metal and shorter
calculated BH bonds than the inverse adduct, an increase in
steric interactions should shift the energetic balance away
from the η2-B,H species and toward the purely dative
geometry. This was corroborated by calculations that showed
the single η2-B,H (i.e., (RO)3Ta(η2-B,H-BH3)) and dative
(i.e., (RO)3TafBH3) structures to be essentially degenerate
when 2SiMe and 2SiBu were the models, and the “double η2-
B,H” structure, (RO)3Ta(η3-H,H,B-BH3)) did not even
constitute a minimum in 2SiBu. Given the calculated energies,
it can be reasonably assumed that the interconversion of η2-
B,H and inverse adduct structures is likely to have a very
modest barrier, no matter which conformation represents the
true ground state.
TaB bonding orbitals are very low in energy and their
corresponding σ* orbitals are in the energy realm of the
remaining d orbitals. For the “inverse-adduct” model, the
TaB σb orbital is at -5.5 eV, well below the TaO π* orbitals
2
2
,
at -1.0 eV (dxz, dyz) and the TaO σ* orbitals at 0.2 eV (dx -y
2
dxy) and 0.3 eV (“dz ”, i.e., TaB σ*). To further probe the
nature of the TaB bond in (silox)3TafBH3, a natural bond
orbital (NBO) analysis was conducted on the more compu-
tationally tractable 2SiMe. It indicates a covalent TaB bond
that is 55% Ta and 45% B (primarily 2p) in character, with
2
the tantalum contribution split 50:50 between 6s and 5dz .
Optimized geometries of (silox)3TafBH3 and 2SiMe are
given in Figures 3 and 4, respectively, with the methyl groups
removed for clarity. The d(TaB) in the “inverse adduct” is
2.22 Å compared to 2.18 Å for the η2-B,H conformer. The
B-H bond length of the η2-B,H interaction is 1.27 Å, which
is substantially longer than the terminal BH bond distances
of 1.21-1.23 Å. Aside from the asymmetry of the η2-B,H
conformation, the core geometries appear similar.
The orbitals pertaining to the model 2SiBu are quite similar
but with greater delocalization, as expected. The lowest
orbital at -5.4 eV manifests the TaB bonding, and the next
two are TaO π* orbitals at -1.0 and -0.8 eV (dxz, dyz). One
2
2
of the TaO σ* orbitals (dx -y ) now possesses additional
TaBH σ* character and is raised to 0.5 eV relative to its
partner dxy at 0.1 eV. An orbital that appears to be mostly
2
TaB antibonding in character resembling dz is found at
2. Electronic Structure of 2SiBu. Figure 5 illustrates the
d-orbital splitting diagrams of (silox)3TafBH3 and 2SiBu and
their respective TaB bonding orbitals. It is clear from either
depiction that 2 can be considered a d0 species, since the
0.3 eV.
3. Vibrational Structure. In an effort to discern the η2-
B,H ground state from the “inverse adduct” conformer
pertaining to 2, IR spectra were first calculated on 2SiMe. The
Inorganic Chemistry, Vol. 46, No. 4, 2007 1227

Bonanno et al.
at the B3LYP/CEP_31G(d) level of theory yielded values
of 2466 (νa) and 2408 cm-1 (νs) for 11B-H stretches,
implicating scaling factors of 0.96 (νa(11BH)) and 0.94
(νs(11BH)) to compensate for anharmonic effects, the incom-
plete basis set, etc.39 For the purely dative conformer of 2SiBu
,
the calculated stretching frequencies are 2454 cm-1 (actually
2451 and 2457 cm-1 due to deviation from strict C3V
symmetry in the computation) for νa(11BH) and 2392 cm-1
for νs(11BH), with a 4:1 intensity ratio; the scaled bands are
at 2355 and 2248 cm-1. The η2-B,H conformer has 2558
(νa(11BH)) and 2496 cm-1 (νs(11BH)) absorptions in a 2:1
ratio that scale to 2456 and 2346 cm-1. The latter more
closely agrees with the 2445 and 2395 cm-1 absorptions
present in a nujol mull of 2.40 The calculated η2-B,H structure
of 2SiBu also possesses a stretch for the bridging Ta‚‚‚HB at
2139 cm-1, which should have an intensity in between the
terminal bands and would scale to roughly 2150-2010 cm-1,
provided the previous scale factors were applicable. In
reinvestigating the experimental IR spectrum of 2, an
asymmetric absorption at ∼2030 cm-1 can be seen, although
it is very broad, spanning a 2250-1950 cm-1 range. The
presence of this broad absorption and the nearly matching
frequencies of 2SiBu and the experimental spectrum of 2
suggest that the η2-B,H conformer is the ground state
structure.
Figure 4. DFT-optimized geometry of the η2-B,H conformer of (silox)3-
Ta(η2-H,B-BH3) (2SiBu); methyl groups have been omitted for clarity.
Interatomic distances (Å) and angles (deg): TaB ) 2.18; BHt ) 1.21;
BHb ) 1.27; Ta‚‚‚Ht ) 2.91 and 2.93; TaHb ) 2.22; O-Si ) 1.70-1.71;
Ta-O ) 1.88-1.90; Ta-O-Si ) 174-176; TaBHt ) 116 and 117; TaBHb
) 75; O-Ta-B ) 100-106; O-Ta-O ) 112-122.
With the vibrational structure of 2SiBu in hand, the energy
difference between (silox)3Ta(η2-B,H-BH2D) and (silox)3-
Ta(η2-B,D-BH2D) (2SiBu-D) could be calculated by simple
placement of deuterium in the model. At 296 K, the η2-B,H,
D-terminal structure was calculated to be more favorable than
the η2-B,D, H-terminal isotopomer by -78 cal/mol. This
modest value is consistent with the inability of the isotopic
perturbation of chemical shift experiment to discern the η2-
B,H structure, assuming the chemical shifts of η2-B,H and
terminal positions are not hugely different.
4. BH3 Binding Energies. The binding energy of BH3 to
(RO)3Ta was calculated for R ) SiH3 (2SiH) and SiMe3
(2SiMe). For both models the binding energy is exothermic
by g35 kcal/mol to either the η2-B,H or purely dative
structures. Given the similarity in binding energies for the
two models, calculations employing the post-Hartree-Fock
coupled clusters methodology (i.e., CCSD(T)/CEP-31G(d)//
B3LYP/CEP-31G(d)) were conducted on the smaller of the
two (2SiH) to assess the binding energy at this more extensive
electron correlation level. These calculations afforded com-
parable binding energies (∼ -35 to -37 kcal/mol depending
on borane conformation) and are fully consistent with the
stability accorded 2.
Figure 5. d-Orbital splitting diagrams plus the TaB bonding orbital for
calculational models (silox)3Ta-BH3 and (silox)3Ta(η2-H,B-BH3) (2SiBu);
some hydrogens have been omitted.
Both η2-B,H and “inverse adduct” structures confer
substantial stability to a dative interaction that can be
construed as having considerable covalent character (i.e.,
main group adduct, Me3NfBH3, was chosen to calibrate
computational vs experimental vibrational data because the
true structures of plausible transition metal candidates were
unknown.2-5 Carpenter and Ault38 reported B-H stretching
frequencies of 2367 (νa, 11B) and 2270 cm-1 (νs, 11B) for
Me3NBH3 in an argon matrix, while gas-phase geometry
optimization and energy Hessian calculations for the same
-
Ta+-BH3 ). It is tempting to assign a Ta(V) oxidation state
(39) Several, groups, most notably that of Pulay, have discussed the
appropriateness of having different scale factors for different vibrational
modes. See: Rauhut, G.; Pulay, P. J. Phys. Chem. 1995, 99, 3093-
3100.
(40) Using a single, averaged scaling factor of 0.95 does not change these
conclusions and actually makes the predicted IR bands for the η2-
B,H conformer much closer to the observed.
(38) Carpenter, J. D.; Ault, B. S. J. Phys. Chem. 1991, 95, 3507-3511.
1228 Inorganic Chemistry, Vol. 46, No. 4, 2007

EWidence for Strong Tantalum-to-Boron DatiWe Interactions
a
Table 2. Comparison of Experimental and Calculational Metric Parameters Pertaining to (silox)3Ta(η2-B,Cl-BCl2Ph) (3 and 3SiBu
)
bonds
(Å, calcd)
(Å, exptl)
angles
(deg, calcd)
(deg, exptl)
Ta(1)-O(7)
Ta(1)-O(8)
Ta(1)-O(9)
O(7)-Si(4)
O(8)-Si(5)
O(9)-Si(6)
Ta(1)-B(10)
B(10)-Cl(2)
B(10)-Cl(3)
B(10)-C(11)
1.907
1.898
1.890
1.725
1.732
1.739
2.324
2.029
1.875
1.596
1.885(4)
1.890(4)
1.875(5)
1.693(5)
1.699(4)
1.702(5)
2.308(8)
1.999(8)
1.828(8)
1.582(11)
Ta(1)-O(7)-Si(4)
Ta(1)-O(8)-Si(5)
Ta(1)-O(9)-Si(6)
O(7)-Ta(1)-B(10)
O(8)-Ta(1)-B(10)
O(9)-Ta(1)-B(10)
O(7)-Ta(1)-O(8)
O(7)-Ta(1)-O(9)
O(8)-Ta(1)-O(9)
Ta(1)-B(10)-Cl(2)
Ta(1)-B(10)-Cl(3)
Ta(1)-B(10)-C(11)
Cl(2)-B(10)-Cl(3)
Cl(2)-B(10)-C(11)
Cl(3)-B(10)-C(11)
172.7
172.6
172.4
109.6
107.0
102.0
121.2
106.3
109.0
76.1
111.1
129.0
108.1
111.7
113.2
173.6(3)
168.6(3)
168.6(3)
108.4(2)
109.4(2)
101.2(2)
123.5(2)
105.4(2)
106.6(2)
72.6(3)
114.7(4)
126.4(5)
107.5(4)
110.9(5)
114.4(5)
a
tBu groups have been removed from the silox ligand for clarity.
-
to 2 on the basis of its stability, but there is little calculational
reasonable calibration with known compounds (e.g., BH4 ,
BCl3, B2H6, BF3·OEt2), the 11B NMR shifts calculated for
the η2-B,H (δ 8) and dative (δ -3) structures are not close
to the observed value of δ 28, which is midrange between
related BX3 (sp2, δ 60-80) and BX4- (sp3, δ -20) species.42
2-
support for a formal Ta2+rBH3 distribution of charge.
The tantalum is not considered formally oxidized to Ta(V),
but its Valencesthe number of tantalum electrons utilized
in bondingsis five. As Parkin has argued,20,23 despite a
formal oxidation state of (III) for the tantalum, it is a d0 center
on the basis of valence, the same as it is in the formally
Ta(V) hydride-borohydride, 5. The molecular orbital dia-
grams support this depiction, since 5e- from tantalum are
utilized in bonding orbitals that lie well below the d-block
in energy.
5. Calculated NMR Parameters. 1H NMR chemical shifts
were calculated for the η2-B,H and dative conformers of 2SiMe
using the gauge-independent atomic orbital (GIAO)41 method
and the B3LYP functional. For the main group elements,
the extended Pople basis set 6-311+G(2d,p) was used, while
the triple-ú CEP-31G valence basic set/effective core po-
tential was maintained for tantalum. In the “inverse adduct”
conformer, a chemical shift of δ 2.7 was calculated for the
three equivalent hydrogens. The η2-B,H conformer of 2SiMe
is calculated to have a shift of δ 2.1 for the bridging
proton and δ 4.2 for the terminal protons, yielding a weighted
average of δ 3.5, which is marginally closer to the experi-
mental shift of δ 3.9. Calculations of JBH proved to be equally
insufficient at differentiating dative (JBH ) 101 Hz) and η2-
B,H (JBH(BHb) ) 59 Hz, JBH(BHt) ) 118 Hz; JBH(ave) )
{59 + 2(118)}/3 ) 98 Hz) structures, although both were
quite close to the experimental value of 98 Hz. Despite a
Computational Model of 3. Table 2 lists the experimental
core distances and angles referenced to those calculated for
the model (silox)3Ta(η2-B,Cl-BCl2Ph) (3SiBu), and there is
excellent agreement between them. The worst distance
comparison is for the terminal B-Cl bond, where the
calculated bond length is ∼0.05 Å greater than the 1.83 Å
observed, and most are within 0.03 Å. The same is true for
the bond angles, where the worst cases are still within ∼4°
of the experimental values.
Initial B3LYP/CEP-31G(d) optimization of (RO)3Ta(BCl2-
Ph) using R ) H (3′), SiH3 (3SiH), and SiMe3 (3SiMe) yielded
a trigonal bipyramidal boryl complex, d0 (RO)3ClTaBClPh.
In following the course of geometry optimizations, a shallow
minimum was observed in the vicinity of a (RO)3Ta(η2-B,Cl-
BCl2Ph) structure, which eventually stabilized through B-Cl
bond lengthening to afford the oxidative addition product,
the boryl-chloride. It was only until the full silox model (R
) SitBu, 3SiBu) was employed that the η2-B,Cl adduct
became stable. The results imply that the unusual geometry
of 3 is a consequence of the severe steric interactions of the
silox groups, which effectively arrest the B-Cl oxidative
addition.
(41) Gauss, J.; Stanton, J. F. AdV. Chem. Phys. 2002, 123, 355-422.
(42) Herma´nek, S. Chem. ReV. 1992, 92, 325-362.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1229

Bonanno et al.
Perkin-Elmer 299B grating IR, or PE 377 grating IR. Cryoscopic
molecular weight determination in benzene was performed with a
homemade device. Combustion analyses were performed by Oneida
Research Services (Whitesboro, NY), Robertson Microlit Labora-
tories (Madison, NJ), or Texas Analytical (Houston, TX).
Conclusions
The treatment of 1 with BH3 and BCl2Ph afforded two
adducts in which the principal interaction is a donation from
2
2
the filled dz orbital of the d tantalum center to the empty
2pz orbital of boron that is covalent in nature. The conforma-
tion of 2 could not be confirmed despite exhaustive
spectroscopic studies and the isotopic perturbation of chemi-
cal shift investigation, but calculations on (silox)3Ta(η2-B,H-
BH3) and (silox)3TafBH3 (2SiBu) provided vibrational fre-
quencies that more closely match that of the former η2-B,H
structure. The η2-B,H structure is favored as the GS geometry
in models containing small RO (R ) H, SiH3, SiMe3) but
the “inverse-adduct” conformer becomes energetically com-
petitive as the steric parameters are increased to R ) tBu3Si
because its Ta-B bond is longer than in the η2-B,H species.
The similar energies accorded the two conformations suggest
that their dynamic interconversion is quite likely. The
stabilization of the pair of electrons in the TafB bond
renders the tantalum effectively d0 and energetically com-
petitive with any BH oxidative addition product.
Syntheses. 1a. (silox)3Ta(BH3) (2). A 25 mL two-neck flask
was charged with 0.251 g of 1 (0.303 mmol), fitted with a septum,
and attached to a frit assembly. The apparatus was evacuated, and
Et2O (10 mL) was added via vacuum-transfer at -78 °C. The frit
assembly was filled with Ar, and 0.31 mL of 1.0 M (in THF, 0.31
mmol) BH3·THF was syringed into the cold solution. The blue color
was discharged immediately, and the resulting orange solution was
allowed to warm to 23 °C and was stirred for 30 min. After the
septum was replaced with a glass stopper, the volatiles were
removed in vacuo and the solid residue was dissolved in hexane.
Filtration, concentration to 3 mL, and cooling to -78 °C generated
pale orange crystals which were collected by filtration (0.141 g).
1
A second crop yielded 0.031 g (67% total). H NMR (C6D6) δ
t
1.28 (s, Bu, 81H), 3.93 (br quar, BH3,1JBH ) 98 Hz, 3H); (THF-
d8) δ 1.25, 3.30. 13C{1H} NMR (C6D6) δ 23.78 (SiC), 30.48
(C(CH3)3); (THF-d8) δ 24.36, 30.88. 11B NMR (C6D6) δ 27.89
(quar,1JBH ) 98 Hz); (THF-d8) δ 32.52. IR (nujol, cm-1) 2445 (br
m), 2395 (br m), 2030 (v br), 1475 (s), 1375 (s), 1290 (w), 1010
(m), 930 (s), 860 (br, s), 810 (s), 665 (w), 620 (s), 585 (m). Mr
found: 833(25); calcd: 841. Anal. Calcd for C36H84BO3Si3Ta: C,
51.41; H, 10.07. Found: C, 51.21; H, 10.23.
Steric influences dramatically affect the binding of Cl2-
BPh as well. According to the calculations, a trigonal
bipyramidal oxidative addition product, (RO)3ClTaBClPh,
becomes less favored with respect to the (RO)3Ta(η2-B,Cl-
b. (silox)3TaBD3 (2-D3). A sample of 1 (0.101 g, 0.122 mmol)
was placed in a 10 mL flask, which was then attached to a gas
bulb. THF (5 mL) was added at -78 °C, and the flask was cooled
to 77 K. B2D6 (12.3 Torr in 91 mL, 0.061 mmol, 0.5 equiv) was
condensed into the flask. Upon warming to -78 °C, the stirred
solution quickly turned from blue to orange. Stirring was maintained
for 1 h at -78 °C and for an additional 30 min at 23 °C. The THF
was removed in vacuo, and hexane (2 × 5 mL) was placed on the
orange solid and removed. The resulting material was judged >95%
BCl2Ph) adduct as the size of RO (R ) H, 3′; SiH3, 3SiH
;
SiMe3, 3SiMe; SitBu3, 3SiBu) is increased. It is somewhat
surprising that the steric features of silox push the equilibrium
away from a formal Ta(V) oxidative addition product to the
µ-Cl adduct, 3. In this structure, the tantalum can be
considered d0 as it contributes 2e- to the TafB dative bond,
and a strong B-Cl bond is retained. These factors and the
steric influence of silox help counteract the tendency toward
oxidative addition of the BCl bond. This is the first time
calculations have indicated that steric features of silox
changed the reactivity at the metal center aside from
preventing dimerization.
1
pure and >98% deuterated by a H NMR assay and was used for
IR studies without purification. IR (nujol, cm-1, BD vibrations)
1845 (s), 1770 (s), 980 (s).
2. (silox)3Ta(η2-B,Cl-BCl2Ph) (3). Into a two-neck flask was
placed 0.499 g of 1 (0.603 mmol). A sidearm addition tube was
charged with PhBCl2 (78.3 µL, 0.603 mmol). The flask was cooled
and evacuated, and 15 mL of toluene was added at -78 °C. The
solution was warmed to 23 °C, and the PhBCl2 was washed into
the flask by condensing toluene into the sidearm. After 1 h the
solution was filtered, and the volume of the filtrate reduced to 5
mL. After cooling to -78 °C, small portions of hexane (∼1 mL
total) were added until copious solid was observed. Orange crystals
were collected by filtration and washed with cold hexane (0.409 g,
Experimental Section
General Considerations. All manipulations were performed
using either glovebox or high vacuum line techniques. All glassware
was oven-dried. THF and ether were distilled under nitrogen from
purple sodium benzophenone ketyl and vacuum-transferred from
the same prior to use. Hydrocarbon solvents were treated in the
same manner with the addition of 1-2 mL/L tetraglyme. Benzene-
d6 was dried over sodium, activated 4 Å molecular sieves, vacuum-
transferred, and stored under nitrogen. THF-d8 was dried over
sodium and vacuum-transferred from sodium benzophenone ketyl
prior to use. H2 and D2 were passed over a column containing
activated 4 Å sieves and copper oxide. BH3·THF, PhBCl2, and
LiBD4 were purchased from Aldrich and used without further
purification. B2D6 was prepared by adding I2/tetraglyme solution
to NaBD4 in tetraglyme. D2 was separated by freeze(77K)/pump/
thaw degassing. Pure B2D6 was then transferred to a separate bomb
and stored at 77 K. 124 and 430 were prepared following published
procedures.
1
t
69%). H NMR (C6D6) δ 1.26 (s, Bu, 81H), 7.02 (t, p-H, 1H),
7.33 (t, m-H, 2H), 7.95 (d, o-H, 2H). 13C{1H} NMR (THF-d8) δ
24.42 (SiC), 31.33 (C(CH3)3), 126.59, 128.22, 134.99 (Ph), BC(ipso)
not observed. 11B NMR (C6D6) δ 61.79 (br s). IR (nujol, cm-1
)
1475 (s), 1375 (m), 1190 (m), 1175 (m), 1150 (w), 1035 (w), 1015
(m), 935 (br, s), 870-750 (br, s), 725 (s), 700 (m), 625 (s). Anal.
Calcd for C42H86BCl2O3Si3Ta: C, 51.16; H, 8.79. Found: C, 51.29;
H, 9.03.
3a. (silox)3HTa(η3-BH4) (5). A 0.236 g sample of 1 (0.285
mmol) was placed in a 25 mL two-neck flask. The flask was fitted
with a stopper, attached to a frit, and cooled to -78 °C. Diethyl
ether (15 mL) was added by vacuum transfer, and the frit was filled
with dry H2 (∼600 Torr). The solution was stirred for 11 h at
23 °C until all vestiges of blue color disappeared. The excess H2
was removed, and a septum was attached to the flask under Ar
1H, 13C{1H}, and 11B NMR spectra were obtained on Varian
XL-200 and XL-400 spectrometers. Chemical shifts are reported
relative to benzene-d6 (1H, δ 7.15; 13C{1H}, δ 128.00) and BF3·
OEt2 (11B, δ 0.00). IR spectra were recorded on a Mattson FT-IR,
1230 Inorganic Chemistry, Vol. 46, No. 4, 2007

EWidence for Strong Tantalum-to-Boron DatiWe Interactions
counterflow. BH3·THF (0.28 mL, 1 M in THF, 0.28 mmol) was
added via syringe to the ether solution at -78 °C. The flask was
allowed to warm to 23 °C, and stirring was maintained for 1.5 h.
The volatiles were removed, and the volume of the solution was
reduced to 10 mL. Filtration was followed by further concentration
to 3 mL, and cooling to -78 °C generated a white crystalline solid
8. 5 + D2. A sample of 5 was placed into an NMR tube attached
to a 14/20 joint. The tube was fixed to a 180° needle valve and
evacuated. C6D6 (0.6 mL) was added via vacuum-transfer. The tube
was filled with ∼600 Torr of D2 and sealed with a torch. After
thermolysis at 90 °C for 18 h, the 1H{11B} NMR spectrum showed
the presence of 2, 2-D, and 2-D2.
1
that was isolated by filtration (0.186 g, 77%). H NMR (C6D6) δ
Single-Crystal X-ray Structure Determination. 9. 3. An orange
crystal approximately 0.4 × 0.5 × 0.4 mm3 obtained from a toluene/
hexane solution was mounted in a capillary and placed in the
goniometer of a Siemens P4 diffractometer equipped with a fine-
focus molybdenum X-ray tube and graphite monochromator at 293-
(2) K. Preliminary diffraction data revealed a triclinic cell, a
hemisphere routine was used to collect the data, and precise lattice
constants (a ) 13.1510(1) Å, b ) 13.1990(10) Å, and c ) 17.441-
(2) Å; a ) 102.020(10)°, b ) 96.160(10)°, g ) 119.120(10)°; V )
2508.5(4) Å3) were determined from a least-squares fit of 15
measured 2θ values. The space group (P1h) was determined, and
after correction for Lorentz, polarization, and background effects,
unique data were judged observed according to |Fo| > 2σ|Fo| (6444
out of 7476; 464 parameters). All heavy atoms were located using
direct methods, and all non-hydrogen atoms were revealed by
successive Fourier syntheses. Full matrix, least-squares refinements
(minimization of ∑w(Fo - Fc)2 where w is based on counting
statistics modified by an ignorance factor, w-1) with anisotropic
heavy atoms and all hydrogens included at calculated positions led
to the final model. Refinement utilized SHELXL and w-1 ) σ2-
t
1.29 (s, Bu, 81H), 3.48 (br quar, BH4,1JBH ) 88 Hz, 4H), 22.47
(br quin, TaH,2JHH ) 12 Hz, 1H). 13C{1H} NMR (C6D6) δ 24.25
(SiC), 30.47 (C(CH3)3). 11B NMR (C6D6) δ -16.72 (quin,1JBH
)
88 Hz). IR (nujol, cm-1) 2495 (m), 2200 (br, m), 2150 (br, m),
1765 (br, s), 1470 (s), 1375 (m), 1160 (br, w), 1010 (w), 960 (w),
925 (w), 870 (br, s), 810 (s), 620 (s). Anal. Calcd for C36H86BO3-
Si3Ta: C, 51.29; H, 10.28. Found: C, 51.57; H, 10.50.
b. (silox)3DTa(η3-BD4) (5-D5). Into a 10 mL flask was placed 1
(0.102 g, 0.123 mmol). The flask was attached to a gas bulb,
evacuated, and cooled to -78 °C. THF (5 mL) was added followed
by 600 Torr of D2. The solution was stirred for 1 h at 23 °C and
freeze/pump/thaw degassed. The reaction was recharged with D2
and stirred for another 1.5 h. After the solution was degassed, B2D6
(13.5 Torr in 91 mL, 0.067 mmol, 0.54 equiv) was condensed into
the flask. The solution was stirred for 3 h, and the volatiles were
removed. The white solid obtained after addition and removal
of 2 × 5 mL hexane was found to be >95% pure and ∼60%
deuterated (Ta-H and BH4 sites), which was sufficient for the IR
study. IR (nujol, cm-1, B-D and Ta-D vibrations) 1870, 1630,
1580, 1270 (TaD).
(Fo ) + (0.0595p)2 + 7.1714p, where p ) (Fo2 + 2Fc2)/3). R1 [I >
2
2σ(I)] ) ∑||Fo| - |Fc||/∑|Fo| ) 0.0433, R1 (all data) ) 0.0560;
NMR Tube Reactions. 4. 2 + C2H4. An NMR tube attached to
a ground glass joint was charged with 2 (0.018 g, 0.021 mmol)
and attached to a gas bulb. C6D6 (0.6 mL) was added at 77 K
followed by C2H4 (57.4 Torr in 20.5 mL, 0.064 mmol, 3 equiv).
The tube was sealed with a torch under dynamic vacuum. No change
2 1/2
wR2 [I > 2σ(I)] ) [∑w(|Fo| - |Fc|)2/∑wFo ] ) 0.1047, wR2
(all data) ) 0.1174; GOF (all data, n ) number of independent
reflections, p ) number of parameters) ) [∑w(|Fo| - |Fc|)2/(n -
p)]1/2 ) 1.040.
1
Computational Methods. All calculations were carried out using
the Gaussian0343 suite of programs. The B3LYP functional (Becke’s
three-parameter hybrid functional44 using the LYP correlation
functional containing both local and nonlocal terms of Lee, Yang,
and Parr)45 and VWN (Slater local exchange functional43 plus the
local correlation functional of Vosko, Wilk, and Nusair)46 was
employed in conjunction with the CEP-31G(d) valence basis sets
and effective core potentials of Stevens et al.47 Closed-shell
(diamagnetic) and open-shell (paramagnetic) species were modeled
within the restricted and unrestricted Kohn-Sham formalisms,48
respectively. All systems were fully optimized without symmetry
constraint.
was noted in the H NMR spectrum of the sample after weeks at
room temperature and 20 h at 90 °C.
5. 2 + Me3N. An NMR tube on a 14/20 joint was charged with
a sample of 2 (0.010 g, 0.012 mmol). The tube was attached to a
gas bulb and evacuated. Benzene-d6 (0.6 mL) and Me3N (188 Torr
in 12.3 mL, 0.126 mmol, 10.5 equiv) were added at 77 K. The
tube was flame-sealed under dynamic vacuum. After a week at room
temperature, the NMR spectrum showed only 2 and Me3N.
6. 5 + Me3N. A 25 mL flask containing an ether solution of 4
(0.052 g, 0.063 mmol) was treated with BH3·THF (0.063 mL, 1 M
in THF, 0.063 mmol) and stirred for 18 h. The volatiles were
removed, and the NMR spectrum of the resulting solid showed only
5. The solid was transferred into an NMR tube on a ground glass
joint with 0.7 mL of C6D6. The tube was attached to a gas bulb
and freeze/pump/thaw degassed three times. Me3N (55.2 Torr in
20.5 mL, 0.062 mmol) was condensed into the tube, which was
(43) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G.
A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J.
W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;
Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.
(44) Becke, A. D. J. Chem. Phys. 1993, 98, 1372-1377.
1
sealed with a torch. The H NMR spectrum taken immediately
upon warming showed a 1:1 mixture of 4 and 5. Within 3 h, the
NMR showed no 5 and complete formation of the products, 4 and
Me3N·BH3.
7. 2 + BH3·THF. To an NMR tube attached to a ground glass
joint was added a sample of 2 (0.012 g, 0.014 mmol). The tube
was attached to a 180° needle valve and freeze/pump/thaw degassed.
THF-d8 (0.6 mL) was added via vacuum-transfer. Against a
counterflow of N2, BH3·THF (14 µL, 1 M in THF, 0.014 mmol)
was added to the tube. After degassing, the tube was frozen at 77
K and flame-sealed under dynamic vacuum. The room-temperature
11B NMR spectrum of this sample showed a sharp quartet at δ
30.5 for 2 and a broad signal at δ -0.1 corresponding to BH3·
THF, neither of which showed any sign of coalescing or broadening
up to 80 °C.
(45) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. 1998, B37, 785.
(46) Vosko, S. H. W., L.; Nusair, M. Can. J. Chem. 1980, 58, 1200.
(47) (a) Stevens, W. J.; Basch, H.; Krauss, M. J. Chem. Phys. 1984, 81,
6026-6033. (b) Stevens, W. J.; Basch, H.; Krauss, M.; Jasien, P. G.
Can. J. Chem. 1992, 70, 612-630.
Inorganic Chemistry, Vol. 46, No. 4, 2007 1231

Bonanno et al.
Acknowledgment. P.T.W. thanks the NSF (CHE-
0415506) and Cornell University, and T.R.C. thanks the DOE
(No. DEFG02-03ER15387) for research support. The U.S.
Department of Education is acknowledged for support of the
CASCaM facility, and the NSF (CHE-0342824) for support
of the UNT computational chemistry resource. A.W.P.
acknowledges the UNT Toulouse School of Graduate Studies
for a fellowship. We thank Elliott B. Hulley for experimental
assistance.
Supporting Information Available: Crystallographic data in
cif format. This material is available free of charge via the Internet
at http://pubs.acs.org.
IC0616885
(48) Kohn, W.; Sham, L. Phys. ReV. 1980, A140, 1133-1138.
1232 Inorganic Chemistry, Vol. 46, No. 4, 2007