Pnictogen-hydride activation by (silox)3Ta (silox = tBu3SiO); Attempts to circumvent the constraints of orbital symmetry in N2 activation

8524 Inorg. Chem. 2010, 49, 8524–8544

DOI: 10.1021/ic101147x

Pnictogen-Hydride Activation by (silox)₃Ta (silox = ^tBu₃SiO); Attempts

to Circumvent the Constraints of Orbital Symmetry in N₂Activation

Elliott B. Hulley,^†Jeffrey B. Bonanno,^†Peter T. Wolczanski,*^,†Thomas R. Cundari,^‡and Emil B. Lobkovsky^†

^†Department of Chemistry & Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York

14853, and ^‡Department of Chemistry, University of North Texas, Box 305070, Denton, Texas 76203-5070

Received June 9, 2010

t

Activation of N₂by (silox)₃Ta (1, silox = Bu₃SiO) to afford (silox)₃TadN-NdTa(silox)₃(1₂-N₂) does not occur

despite ΔGꢀ_cald= -55.6 kcal/mol because of constraints of orbital symmetry, prompting efforts at an independent

synthesis that included a study of REH₂activation (E = N, P, As). Oxidative addition of REH₂to 1 afforded

(silox)₃HTaEHR (2-NHR, R = H, Me, ⁿBu, C₆H₄-p-X (X = H, Me, NMe₂); 2-PHR, R = H, Ph; 2-AsHR, R = H, Ph), which

underwent 1,2-H₂-elimination to form (silox)₃TadNR (1dNR; R = H, Me, ⁿBu, C₆H₄-p-X (X = H (X-ray), Me, NMe₂,

CF₃)), (silox)₃TadPR (1dPR; R = H, Ph), and (silox)₃TadAsR (1dAsR; R = H, Ph). Kinetics revealed NH bond-

breaking as critical, and As > N > P rates for (silox)₃HTaEHPh (2-EHPh) were attributed to (1) ΔGꢀ_calc(N) <

ΔGꢀ_calc(P) ∼ ΔGꢀ_calc(As); (2) similar fractional reaction coordinates (RCs), but with RC shorter for N < P∼As; and

(3) stronger TaE bonds for N > P∼As. Calculations of the pnictidenes aided interpretation of UV-vis spectra. Addition

of H₂NNH₂or H₂N-N(^cNC₂H₃Me) to 1 afforded 1dNH, obviating these routes to 1₂-N₂, and formation of

(silox)₃MeTaNHNH2 (4-NHNH₂) and (silox)₃MeTaNH(-^cNCHMeCH₂) (4-NH(azir)) occurred upon exposure to

(silox)₃TadCH₂(1dCH₂). Thermolyses of 4-NHNH₂and 4-NH(azir) yielded [(silox)₂TaMe](μ-N_RHN_β)(μ-N_γHN_δH)-

[Ta(silox)₂] (5) and [(silox)₃MeTa](μ-η²-N,N:η¹-C-NHNHCH₂CH₂CH₂)[Ta(κ-O,C-OSi^tBu₂CMe₂CH₂)(silox)₂]

(7, X-ray), respectively. (silox)₃TadCPPh₃(1dCPPh₃, X-ray) was a byproduct from Ph₃PCH₂treatment of 1 to give

1dCH₂. Addition of Na(silox) to [(THF)₂Cl₃Ta]₂(μ-N₂) led to [(silox)₂ClTa](μ-N₂) (8-Cl), and via subsequent

methylation, [(silox)₂MeTa]₂(μ-N₂) (8-Me); both dimers were thermally stable. Orbital symmetry requirements for N₂

capture by 1 and pertinent calculations are given.

Introduction

the siloxide ligand, it has been shown that ^tBu₃SiO (silox) is

unique in its ability to sterically protect the Ta(III) center of

1,⁴thereby permitting the investigation of these activations.

Interestingly, one substrate that is expected to react with 1,

dinitrogen, is noticeably absent.

t

Since its discovery, (silox)₃Ta (1, silox = Bu₃SiO) has

produced a remarkable array of products derived from small

molecule activation because of its proclivity toward oxidative

addition.¹For example, 1/2 CO is cleaved to afford 1/2 (silox)₃-

TaO and 1/4 (silox)₃TaCCTa(silox)₃,²and p-CF₃-C₆H₄-

NH₂undergoes a rare C-N bond scission to provide (silox)₃-

Ta(NH₂)(C₆H₄-p-CF₃).³Other activations include C-O^3,4

and E-H (E = N, P, As) bonds,⁵and the reversible addition

of a CH bond at the para-position of 2,6-lutidine.⁶By varying

Schrock’s seminal studies of N₂activation have their origins

7-9

in the synthesis of L_nX₃TadN-NdTaX₃L_n

and related

“dinitrogen” complexes, which were typically generated via

the reduction of X_nM in the presence of L and N₂.^10-14The

(7) Turner, H. W.; Fellmann, J. D.; Rocklage, S. M.; Schrock, R. R.;

Churchill, M. R.; Wasserman, H. J. J. Am. Chem. Soc. 1980, 102, 7809–7811.

(8) Rocklage, S. M.; Turner, H. W.; Fellmann, J. D.; Schrock, R. R.

Organometallics 1982, 2, 703–707.

*To whom correspondence should be addressed. E-mail: ptw2@cornell.

edu.

(1) (a) Wolczanski, P. T. Chem. Commun. 2009, 740–757. (b) Wolczanski,

P. T. Polyhedron 1995, 14, 3335–3362.

(2) 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.

(9) Rocklage, S. M.; Schrock, R. R. J. Am. Chem. Soc. 1982, 104, 3077–

3081.

(10) (a) Fryzuk, M. D. Acc. Chem. Res. 2009, 42, 127–133. (b) Fryzuk,

M. D.; Johnson, S. A.; Patrick, B. O.; Albinati, A.; Mason, S. A.; Koetzle, T. F.

J. Am. Chem. Soc. 2001, 123, 3960–3973.

(3) Bonanno, J. B.; Henry, T. P.; Neithamer, D. R.; Wolczanski, P. T.;

Lobkovsky, E. B. J. Am. Chem. Soc. 1996, 118, 5132–5133.

(4) Chadeayne, A. R.; Wolczanski, P. T.; Lobkovsky, E. B. Inorg. Chem.

2004, 43, 3421–3432.

(5) Bonanno, J. B.; Wolczanski, P. T.; Lobkovsky, E. B. J. Am. Chem.

Soc. 1994, 116, 11159–11160.

(11) (a) Gambarotta, S.; Scott, J. Angew. Chem., Int. Ed. 2004, 43, 5298–

5308. (b) Gambarotta, S. J. Organomet. Chem. 1995, 500, 117–126.

(12) Chirik, P. J. Organometallics 2010, 29, 1500–1517.

(13) Lee, T.-Y.; Wooten, A. J.; Luci, J. J.; Swenson, D. C.; Messerle, L.

Chem. Commun. 2005, 5444–5446.

(14) Dilworth, J. R.; Henderson, R. A.; Hills, A.; Hughes, D. L.;

MacDonald, C.; Stephens, A. N.; Walton, D. R. M. J. Chem. Soc., Dalton

Trans. 1990, 1077–1085.

(6) Covert, K. J.; Neithamer, D. R.; Zonnevylle, M. C.; LaPointe, R. E.;

Schaller, C. P.; Wolczanski, P. T. Inorg. Chem. 1991, 30, 2494–2508.

r

pubs.acs.org/IC

Published on Web 08/19/2010

2010 American Chemical Society

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8525

Scheme 1

species are most aptly described as diimido complexes, as

structural parameters clearly reveal the TaN multiple bonds

(BO ∼ 3) and N-N single bond. Given the amount of

reduction of N₂implied by the structures, it is not surprising

that the compounds do not relinquish the bound “dinitrogen”,

even under harsh conditions. However, it is surprising that

(silox)₃Ta (1, silox = ^tBu₃SiO), does not appear to bind dini-

trogen to form (silox)₃Ta(N₂)_xor reduce it to (silox)₃Tad

N-NdTa(silox)₃(1₂-N₂). As Scheme 1 shows, in the presence

of N₂, 1 cyclometallates to afford (silox)₂HTa(κ-O,C-OSi-

^tBu₂CMe₂CH₂) (1-cymet),^6,15and if sufficiently concentrated

in benzene solution, a small amount is converted to the bridg-

ing benzene complex, [(silox)₃Ta]₂(μ-C₆H₆) (1₂-C₆H₆).^6,16

Since the thermodynamics, at least according to Schrock’s

preceding studies, certainly suggests that N₂activation by 1

should be viable, it is likely that orbital symmetry con-

straints^17,18play a crucial role in the lack of observed N₂

chemistry.^19-24Even the cyclometalation process, which has

not been shown to exhibit reversible behavior, is likely to

be constrained by orbital symmetry. Herein are described

attempts to circumvent these constraints and synthesize 1₂-N₂

via alternate methods. In concert with these efforts, a sig-

nificant study of REH₂(E = N, P, As) oxidative addition and

1,2-H₂-elimination from (silox)₃HTaEHR is also presented.⁵

Computational studies support the contention that 1₂-N₂is

considerably more stable than 1 and N₂.

Results

Pnictogen-Hydride Reactivity. 1. Tantalum Pnicto-

genide-Hydrides.^25,26Routes to the desired dinitrogen

complex can be envisaged through the oxidative addition

of N-H bonds to (silox)₃Ta (1), and Scheme 2 shows that

there is ample evidence for this approach, not only for

nitrogen, but for P-H and As-H bonds as well.^5,27-31

Treatment of 1 with primary amines^3,5,25,26or ammo-

nia^25,32in hydrocarbon solvent (or C₆D₆in the case of

NMR tube experiments) led to the formation of (silox)₃-

n

HTaNHR (2-NHR; R = Me, Bu, C₆H₄-p-X (X = H,

Me, NMe₂)) and (silox)₃HTaNH₂(2-NH₂) in near quanti-

tative yield (¹H NMR), although only the latter was

isolated because of the subsequent 1,2-H₂-elimination

to form the imido derivatives. An alternative synthesis

involving the addition of RNH₂to (silox)₃TaH₂(2-H)

provided 2-NHR (R = C₆H₄-p-X (X = H, NMe₂, CF₃))

accompanied by the evolution of dihydrogen. IR spectra

of 2-NH₂and 2-NHPh revealed two and one ν(NH)

stretches, respectively, and ν(TaH) stretches at 1786

cm^-1for the former and 1810 cm^-1for the latter, con-

sistent with terminal tantalum hydrides.³³The amide-

hydrides appeared colorless, and their NMR spectra are

provided in Table 1. Diagnostic, broad, amide-hydrogen

resonances ranged from δ 4.56 for 2-NH₂to δ 7.41 for

2-NH(C₆H₄-NMe₂), and these were accompanied by

(15) Hirsekorn, K. F.; Veige, A. S.; Marshak, M. P.; Koldobskaya, Y.;

Wolczanski, P. T.; Cundari, T. R.; Lobkovsky, E. B. J. Am. Chem. Soc. 2005,

127, 4809–4830.

ꢀ

(16) Neithamer, D. R.; Parkanyi, L.; Mitchell, J. F.; Wolczanski, P. T.

J. Am. Chem. Soc. 1988, 110, 4421–4423.

(17) Veige, A. S.; Slaughter, L. M.; Lobkovsky, E. B.; Wolczanski, P. T.;

Matsunaga, N.; Decker, S. A.; Cundari, T. R. Inorg. Chem. 2003, 42, 6204–

6224.

(18) Hirsekorn, K. F.; Hulley, E. B.; Wolczanski, P. T.; Cundari, T. R.

J. Am. Chem. Soc. 2008, 130, 1183–1196.

(19) Cavigliasso, G.; Wilson, L.; McAlpine, S.; Attar, M.; Stranger, R.;

Yates, B. F. Dalton Trans. 2010, 39, 4529–4540.

(25) Ammonia oxidative addition originally reported in: Neithamer,

D. R., Ph.D. Thesis, Cornell University, Ithaca, NY, 1989

(26) Bonanno, J. B., Ph.D. Thesis, Cornell University, Ithaca, NY, 1996.

(27) Greenberg, S.; Stephan, D. W. Chem. Soc. Rev. 2008, 37, 1482-1489;

2798.

(20) (a) Laplaza, C. E.; Johnson, M. J. A.; Peters, J. C.; Odom, A. L.;

Kim, E.; Cummins, C. C.; George, G. N.; Pickering, I. J. J. Am. Chem. Soc.

1996, 118, 8623–8638. (b) Figueroa, J. S.; Piro, N. A.; Clough, C. R.; Cummins,

C. C. J. Am. Chem. Soc. 2006, 128, 940–950.

(28) (a) Masuda, J. D.; Hoskin, A. J.; Graham, T. W.; Beddie, C.; Fermin,

M. C.; Etkin, N.; Stephan, D. W. Chem. Eur. J. 2006, 12m, 8696–8707.

;

€

(21) Curley, J. J.; Cook, T. R.; Reece, S. Y.; Muller, P.; Cummins, C. C.

(b) Stephan, D. W. Angew. Chem., Int. Ed. 2000, 39, 314–329.

(29) Shaver, M. P.; Fryzuk, M. D. Organometallics 2005, 24, 1419–1427.

(30) Tanaka, M. Top. Curr. Chem. 2004, 232, 25–54.

(31) Ebsworth, E. A. V.; Gould, R. O.; Mayo, R. A.; Walkinshaw, M.

J. Chem. Soc., Dalton Trans. 1987, 2831–2838.

J. Am. Chem. Soc. 2008, 130, 9394–9405.

(22) Peters, J. C.; Cherry, J.-P. F.; Thomas, J. C.; Barlado, L.; Mindiola,

D. J.; Davis, W. M.; Cummins, C. C. J. Am. Chem. Soc. 1999, 121, 10053–

10067.

(23) Cui, Q.; Musaev, D. G.; Svensson, M.; Sieber, S.; Morokuma, K.

J. Am. Chem. Soc. 1995, 117, 12366–12367.

(24) Caselli, A.; Solari, E.; Scopelliti, R.; Floriani, C.; Re, N.; Rizzoli, C.;

Chiesi-Villa, A. J. Am. Chem. Soc. 2000, 122, 3652–3670.

(32) Zhao, J.; Goldman, A. S.; Hartwig, J. F. Science 2005, 307, 1080–

1082.

(33) 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.

8526 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

Scheme 2

2

dramatically lower J_PHfor 2-PH₂(3 vs 56 Hz) and a

typical downfield hydride shifts ranging from δ 19.96

(2-NHMe) to δ 21.76 (2-NH(C₆H₄-CF₃)). Their struc-

tures are likely to be trigonal bipyramidal (tbp) with

equatorial amide and hydride ligands or square pyrami-

dal with basal Ta-H and Ta-NHR groups as observed

for the C-N bond oxidative addition product, (silox)₃-

(H₂N)TaC₆H₄-p-CF₃, which was previously crystallogra-

phically characterized.³

3

related, smaller J_HH= 2 (vs 9 Hz for 2-PHPh) value

suggests that the geometries of the phosphide hydrides

may also be different at the metal.

Preparation of the arsenide complexes proved to be

trickier, as each species rapidly lost dihydrogen to form

the corresponding arsinidenes. Treatment of (silox)₃Ta

(1) with PhAsH₂at -78 ꢀC in toluene-d₈provided

(silox)₃HTaAsHPh (2-AsHPh), but the complex was

stable at low temperature only if 1,4-cyclohexadiene, a

radical trap, was present. Given the danger accorded the

use of toxic AsH₃, it was generated via hydrolysis of

Zn₃As₂with HCl, and used immediately after passing

through a cold trap. In this fashion, exposure of 1 to AsH₃

in toluene-d₈at -78 ꢀC, in the presence of 1,4-cyclohex-

adiene, afforded (silox)₃HTaAsH₂(2-AsH₂). Character-

The related exposure of (silox)₃Ta (1) to PhPH₂led

to the formation of bright yellow (silox)₃HTaPHPh

(2-PHPh), which was crystallized from pentane in only

39% yield because of its high solubility. The parent

(silox)₃HTaPH₂(2-PH₂) complex, a pale yellow species,

was isolated in 54% yield, but it was accompanied by an

insoluble brick red material, which may be the double

activation product (silox)₃HTa-PH-TaH(silox)₃, based

on the near insolubility of the size-related, crystallogra-

phically characterized (silox)₃NbdPH-Nb(silox)₃spe-

cies.³⁴Phosphide 2-PH₂manifests ν(PH) bands at 2275

and 2285 cm^-1in its IR spectrum, while 2-PHPh exhibits

only one at 2320 cm^-1, and these were accompanied by

tantalum hydride stretches at 1775 and 1790 cm^-1, re-

spectively. The phenyl phosphide exhibited a ddt at δ 55.5

in its ³¹P NMR spectrum, with ¹J_PH= 243 Hz, ²J_PH= 56

Hz, and a coupling to the o-Ph hydrogens of ³J_PH= 9 Hz.

The ³¹P NMR signal for 2-PH₂was very different, reso-

1

ization of the two arsenide hydrides was limited to H

NMR spectra at -78 ꢀC, and the hydride resonances at δ

24.47 (d, 2-AsHPh) and δ 24.98 (t, 2-AsH₂) that were

observed were the furthest downfield for this genre of

compounds. The hydrides were coupled to AsH protons

3

at δ 5.53 (d, J_HH= 8 Hz) in 2-AsHPh and at δ 3.13

(d, ³J_HH= 5 Hz) in 2-AsH₂.

2. Tantalum Pnictidene Derivatives.^35-46The pnictide-

hydrides addressed in the previous section all underwent

nating at -142.39 ppm, with ¹J_PH= 177 Hz and ²J_PH

=

(35) Cowley, A. H. Acc. Chem. Res. 1997, 30, 445–451.

(36) Johnson, B. P.; Balazs, G.; Sheer, M. Top. Curr. Chem. 2004,

232, 1–23.

(37) Cummins, C. C. Angew. Chem., Int. Ed. 2006, 45, 862–870.

(38) Piro, N. A.; Cummins, C. C. J. Am. Chem. Soc. 2009, 131, 8764–8765.

(39) Cossairt, B. M.; Cummins, C. C. Inorg. Chem. 2008, 47, 9363–9371.

(40) Zhao, G. Y.; Basuli, F.; Kilgore, U. J.; Fan, H. J.; Aneetha, H.;

Huffman, J. C.; Wu, G.; Mindiola, D. J. J. Am. Chem. Soc. 2006, 128, 13575–

13585.

3 Hz. The difference prompted a molecular weight study

of 2-PH₂, and this was fully consistent with a monomer:

M_r(found) 858(68); calcd 861. It is plausible that the

two phosphide complexes possess substantially different

Ta-P-R angles because of disparate steric factors, and

that their chemical shifts reflect these features in addition

to the electronic effects of H versus Ph. In addition, the

(41) (a) Waterman, R.; Hillhouse, G. L. Organometallics 2003, 22, 5182–

5184. (b) Melenkivitz, R.; Mindiola, D. J.; HIllhouse, G. L. J. Am. Chem. Soc.

2002, 124, 3846–3847.

ꢀ

(34) Hirsekorn, K. F.; Veige, A. S.; Wolczanski, P. T. J. Am. Chem. Soc.

2006, 128, 2192–2193.

(42) Sanchez-Nieves, J.; Sterenberg, B. T.; Udachin, K. A.; Carty, A. J.

J. Am. Chem. Soc. 2003, 125, 2404–2405.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8527

Table 1. NMR Spectral Assignments^afor (silox)₃M Pnictogen Derivatives and Selected IR Data

¹H NMR (δ (J(Hz))

¹³C{¹H} NMR (δ (J(Hz))

³¹P/¹⁵N NMR (δ (J(Hz))

and IR (cm^-1

compound

^tBu

H/ER

CMe₃,C(CH₃)₃

23.83, 30.83

ER/R

)

(silox)₃HTaNH₂(2-NH₂)

1.29

4.56

20.04

-229.37 (td, 72^b,5^c)

1786, ν(TaH)

3363, ν(NH)

3453, ν(NH)

(silox)₃HTaNHMe (2-NHMe)

(silox)₃HTaNHⁿBu (2-NHⁿBu)^d

1.31

1.32

3.34 (d,7)

5.42 (q,7)

19.96

0.85 (t, CH₃)

3.85 (dd, NCH₂)

5.66 (t, NH)

19.99

6.76 (m, p-H)

7.15-7.26 (m, o-,m-H)

7.29 (NH)

21.47

2.17 (p-Me)

6.99 (m, o-,m-H)

7.33 (NH)

21.38

2.55 (NMe₂)

6.66 (d, m-H)

7.03 (d, o-H)

7.41 (NH)

21.13

6.91 (d, m-H)

7.04 (NH)

7.43 (d, o-H)

21.76

(silox)₃HTaNHPh (2-NHPh)^e

1.26

1.28

1.30

1810, ν(TaH)

3350, ν(NH)

(silox)₃HTaNH(C₆H₄-p-Me)^e

(2-NHC₆H₄-p-Me)

(silox)₃HTaNH(C₆H₄-p-NMe₂)^e

(2-NHC₆H₄-p-NMe₂)

(silox)₃HTaNH(C₆H₄-p-CF₃)^e

(2-NHC₆H₄-p-CF₃)

1.23

(silox)₃Ta(NH₂)₂(3-(NH₂)₂)

(silox)₃HTaPH₂(2-PH₂)

1.30

1.28

4.14

-241.62 (t, 71^b)

2.40 (dd,177,^f2)

21.72 (dt,3^g, 2)

23.75, 30.65

23.58, 30.66

-142.39 (td,177^f, 3^g)

1775, ν(TaH)

2275, ν(PH)

2285, ν(PH)

(silox)₃HTaPHPh (2-PHPh)^e,i

1.26

6.88 (t, p-H)

7.21 (dd,243,^f9)

7.72 (t,o-H, 9)

21.62 (dd,56,^g9)

3.13 (d,5)

126.31

127.53

127.61

55.5 (ddt,243^f, 56^g,9^h)

1790, ν(TaH)

2320, ν(PH)

132.51 (8)^j

(silox)₃HTaAsH₂(2-AsH₂)

1.29

1.26

24.98 (t,5)

5.53 (d, 8)

(silox)₃HTaAsHPh (2-AsHPh)^c

6.96 (m, p-H)

7.84 (d, o-H)

24.47 (d,8)

(silox)₃TadNH (1dNH)

1.29

1.28

6.01 (br)^k

23.78, 30.65

-60.60 (¹J_NH= 76)

3470, ν(NH)

(silox)₃(H₃N)TadNH

0.80 (br, NH₃)

5.67 (br, NH)

(1dNH(NH₃))

(silox)₃TadNLi (1dNLi)^l

(silox)₃TadNCH₃(1dNMe)

(silox)₃TadNⁿBu (1dNⁿBu)^m

1.20

1.29

24.40, 31.43

23.42, 30.54

3.89

47.88

0.92 (t, CH₃)

1.60 (m, CH₂)

4.23 (t, NCH₂)

0.28

6.75 (m, p-H)

7.16-7.25 (o-,p-H)

(silox)₃TadNSiMe₃(1dNTMS)

(silox)₃TadNPh (1dNPh)^e

1.28

1.27

23.59, 30.66

23.36, 30.55

3.79

121.82

125.35

128.31

129.41

(silox)₃TadN(C₆H₄-p-Me)^e

(1dN(C₆H₄-p-Me))

1.28

1.31

2.14 (Me)

6.99 (d, m-H)

7.08 (d, o-H)

(silox)₃TadN(C₆H₄-p-NMe₂)^e

(1dN(C₆H₄-p-NMe₂))

2.50 (NMe₂)

6.62 (d, m-H)

7.13 (d, o-H)

23.41, 30.62

23.35, 30.50

40.92

113.04

125.85

146.31

150.07

123.05 (²J_CF= 32)

(silox)₃TadN(C₆H₄-p-CF₃)^e

1.24

7.01 (d, m-H)

8528 Inorganic Chemistry, Vol. 49, No. 18, 2010

Table 1. Continued

Hulley et al.

¹H NMR (δ (J(Hz))

¹³C{¹H} NMR (δ (J(Hz))

³¹P/¹⁵N NMR (δ (J(Hz))

compound

^tBu

H/ER

7.38 (d, o-H)

CMe₃,C(CH₃)₃

ER/R

and IR (cm^-1

)

(1dN(C₆H₄-p-CF₃))

125.21

125.45 (¹J_CF= 271)

125.78 (³J_CF= 4)

160.33

(silox)₃TadCH₂(1dCH₂)

1.28

1.31

6.66

23.75, 30.59

25.21, 31.31

185.02 (135)

(silox)₃TadCdPPh₃(1dCPPh₃)

7.71 (m, m,p-H)

7.78 (m, o-H)

128.75 (p)

131.15 (m)

133.94 (o)

-4.26

136.37 (d, 88)

175.01 (d, 53)

(silox)₃TadPH (1dPH)

(silox)₃TadPPh (1dPPh)^e

1.30

1.28

3.57 (d,82^f)

276.33 (d,82^f)

2150, ν(PH)

6.76 (m-H^h)

7.20 (o-H)

7.67 (p-H)

23.75, 30.61

125.79

127.67

128.29

137.27 (10^j)

334.6 (t,6^h)

(silox)₃TadPSiMe₃(1dPTMS)

1.34

0.47 (d,6^k)

196.07

(silox)₃TadAsH (1dAsH)

(silox)₃TadAsPh (1dAsPh)^e

1.30

1.29

2.04

1950, ν(AsH)

6.81 (m-H)

7.21 (o-H)

7.86 (p-H)

125.83

127.93

139.61

142.59

(silox)₃Ta(AsH₂)₂(3-(AsH₂)₂)

1.23

1.26

2.33

2180, ν(AsH)

2190, ν(AsH)

(silox)₃MeTaNHNH₂

(4-NHNH₂)

1.03 (Me)

3.63 (NH₂)

4.11 (NH)

1.06 (d,6,CH₃)

1.24 (d,8,CHMe)

1.44 (dd,8,4,CHH)

1.54 (d,4,CHH)

1.21 (Me)

23.78, 30.43

33.79

64.80 (NH₂, 78)ⁿ

148.47 (NH, 78)ⁿ

(silox)₃MeTaNH(^cNCH₂CMeH)

(4-NH(azir))

81.99 (^cN)ⁿ

244.46 (NH, 72)^n,o

[(silox)₂TaMe][(silox)₂Ta]

(μ-N_RHN_β)(μ-N_γHN_δH) (5)

1.24

1.26

1.28

1.30

1.26

24.01, 30.83

24.10, 30.88

24.15, 31.04

24.49, 31.11

23.76, 30.39

38.31

33.87

79.50 (N_γH, 72)^n,o

131.27 (N_RH, 72)^n,o

179.70 (N_δH, 75)^n,o

298.23 (N_β, 8)^n,p

3.00 (N_γH)

5.42 (N_RH)

6.38 (N_δH)

[(silox)₂ClTa]₂(μ-N₂) (8-Cl)

[(silox)₂MeTa]₂(μ-N₂) (8-Me)

1.24

23.79, 30.44

c 3

^aBenzene-d₆unless otherwise noted. ^{b 1}J_15NH

.

J

.

^dβ-CH₂and γ-CH₂obscured. ^eAll couplings between o-, m-, and p-Ph positions are 8(1) Hz.

15NH

f 1

g 2

h 3

J

.

J

HMBCAD correlation spectroscopy.

.

J

. ⁱm-H obscured. ^{j 1}J_PC

.

^kAt 80 ꢀC, a 1:1:1 triplet due to ¹J_14NH∼ 50 Hz was observed. ^lTHF-d₈. ^mCH₂obscured. ⁿObtained via

PH

o 1

p 2

J

.

J

15NH

.

15NH

1,2-H₂-elimination to form the corresponding imido

(nitrene) derivatives (silox)₃TadNR (1dNR; R = H,

a compound tentatively identified as (silox)₃Ta(AsH₂)₂

(3-(AsH₂)₂; ν(AsH) = 2180, 2190 cm^-1), which com-

prised ∼10% of the material. A red insoluble material

also formed, and this is likely to be a dinuclear substance,

as speculated for the related phosphorus chemistry

above.³⁴It is likely that 3-(AsH₂)₂forms from reac-

tion of excess AsH₃in solution with (silox)₃HTaAsH₂

(2-AsH₂) via the loss of H₂, or via the 1,2-AsH-addition to

the TadAs bond of 1dAsH. Phenylarsinidene⁴⁶(silox)₃-

TadAsPh (1dAsPh)⁵was obtained as green crystals in

85% yield from hexane.

n

Me, Bu, C₆H₄-p-X (X = H, Me, NMe₂, CF₃)), phos-

phinidenes (silox)₃TadPR (1dPR; R = H, Ph) and arsi-

nidenes (silox)₃TadAsR (1dAsR; R = H, Ph). An alter-

nate preparation of 1dNH from 2-methylaziridene

proved to be convenient. While the imido derivatives

were colorless, (silox)₃TadPH (1dPH) was isolated as

pale orange crystals (61%), and red-violet crystals of

(silox)₃TadPPh (1dPPh)⁵were obtained in 62% yield.

Parent arsinidene (silox)₃TadAsH (1dAsH) was isolated

as dark orange crystals, but proved to be contaminated by

¹H and ¹³C{¹H} NMR spectra of the pnictidenes

listed in Table 1 reveal standard silox resonances and

others corresponding to the R groups. The parent imide

(1dNH), phosphinidene (1dPH), and arsinidine (1dAsH)

possess varied pnictogen-hydrogen ¹H NMR spectral

shifts. The imide proton is at δ 6.01, and upon warming

to 80 ꢀC, manifests a 1:1:1 triplet due to ¹⁴N-coupling

(¹J_N14H∼ 50 Hz). The PH resonance in 1dPH is at δ 3.57

(43) Termaten, A. T.; Nijbacker, T.; Schakel, M.; Lutz, M.; Spek, A. L.;

Lammertsma, K. Organometallics 2002, 21, 3196–3202.

(44) Freundlich, J. S.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc.

1996, 118, 3643–3655.

(45) Breen, T. L.; Stephan, D. W. J. Am. Chem. Soc. 1995, 117, 11915–

11921.

(46) Mosch-Zanetti, N. C.; Schrock, R. R.; Davis, W. M.; Wanninger, K.;

Seidel, S. W.; O’Donoghue, M. B. J. Am. Chem. Soc. 1997, 119, 11037–11048.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8529

with ¹J_PH= 82 Hz, and the arsinidene proton is observed

at δ 2.04. The observation of large ¹⁴N-coupling is indi-

cative of the cylindrical symmetry expected for the triply

Table 2. UV-vis Absorptions for (silox)₃TadEPh (1dEPh; E = N, P, As)

1dNPh

1dPPh

1dAsPh

1

bonded imido group. In contrast, the J_PHfor 1dPH is

λ (nm) ε (M^-1cm^-1

)

λ (nm) ε (M^-1cm^-1

)

λ (nm) ε (M^-1cm^-1

)

substantially less than that found for the phosphide-

hydride complexes, and is taken as an indication of fur-

ther diminished s-character in the phosphinidene-hydro-

gen bond. Structural studies of (silox)₃TaX species have

historically been difficult to obtain or interpret because of

disorder and twinning issues common to the 3-fold core,

but it is likely that the TadEH angle (E = P, As) in 1dEH

is very close to 90ꢀ.

215 (sh)

246

280 (sh)

285

294 (sh)

17,000

13,300

9,700

9,750

9,300

230 (sh)

244 (sh)

275

300 (sh)

361

18,800

15,600

15,900

5,160

3,880

650

226

249

270

312

385

450 (sh)

602

21,300

18,900

14,800

2,100

3,950

910

434 (sh)

553

310

250

copicstudies. Phosphinidene³¹P chemicalshiftsfor1dPH,

1dPPh, and 1dPTMS were recorded at δ 276.33, 334.6,

and 196.07, respectively. While most of the pnictidene

complexes were isolated and subjected to elemental anal-

ysis (1dER, E = N, R = H, Me, SiMe₃, C₆H₄-p-X (X =

H, CF₃); E = P, R = H, Ph; ER = AsPh), the remaining

complexes were generated in situ for subsequent studies.

3. UV-vis Spectra of Pnictidenes and Calculations.

UV-vis spectra of (silox)₃TadEPh (1dEPh, E = N, P,

As) were taken for comparison and their UV-vis absorp-

tions are listed in Table 2. The higher energy absorption in

each case is taken as an IL (intraligand) band because of

The addition of excess NH₃to benzene solutions of

1

1dNH caused the formation of a white solid, and H

NMR spectral analysis of the reaction mixture implicated

the formation of (silox)₃Ta(NH₂)₂(3-(NH₂)₂). ¹⁵NH₃was

used to obtain ¹⁵N NMR data for 3-(¹⁵NH₂)₂, whose

nitrogens were observed as a triplet (¹J_15NH= 71 Hz) at δ

-241.62. This assignment was based on the ¹⁵N NMR

spectrum of 2-¹⁵NH₂, obtained from ¹⁵NH₃and (silox)₃-

Ta (1), which resonated as a triplet of doublets as δ

-229.37 (¹J_15NH= 72 Hz; ³J_15NH= 5 Hz). Attempts to

generate equilibrium information pertaining to 1dNH þ

NH₃a 3-(NH₂)₂were fraught with complications that

were attributed to complex equilibria,^47,48since different

van’t Hoff plots were obtained at different concentrations

of NH₃in benzene.²⁶It was suspected that aggregation of

NH₃in solution (1, 3, and 10 equiv) was at the origin of

the unusual behavior, and no further attempts were made.

Another complication concerned the formation of (silox)₃-

(H₃N)TadNH (1dNH(NH₃)), whose imide resonance

t

6

ꢀ

comparisons with NaOSi Bu₃, and the melange of bands

in the 240-300 nm region are likely to be ligand-to-metal

charge transfer (LMCT) bands of siloxide and pnictide

character. In the cases of 1dPPh and 1dAsPh, low energy

bands of relatively lower intensity are seen, whereas no

features are observed below ∼330 nm for 1dNPh. Cal-

culations reveal the highest occupied molecular orbitals

(HOMOs) of 1dPPh and 1dAsPh to be essentially the

TadE π-bonding orbitals (Figure 1). The weak 553 nm

(ε = 310 M^-1cm^-1) absorption of 1dPPh is roughly

9-10,000 cm^-1lower⁵⁰than the 361 nm absorption of

considerably greater intensity (ε=3,880 M^-1cm^-1), sug-

gesting that these are the triplet and singlet bands corre-

sponding to TaP(π^b)fTa(d_xz(π^nb)) LMCT. The modest

intensities speak to the minimal overlap of the roughly

orthogonal dπ_yz^band d_xzorbitals. The shoulder at 300 nm

and the low intensity band at 434 nm could be the TaP-

(π^b)fTa(d_yz(π*)) LMCT and its accompanying triplet

absorption, but computationally a P(3s/3p_z)^nbfTa(d_xz-

(π^nb)) LMCT would also be a reasonable fit from both

energy and intensity standpoints. Strong parallels are

observed in the low energy features of 1dAsPh, such as

a weak 602 nm band (ε = 250 M^-1cm^-1) that is ∼9-

10,000 cm^-1below a 385 nm band of modest intensity

(ε = 3,950 M^-1cm^-1), which are consistent with triplet

and singlet absorptions pertaining to the related to TaAs-

(π^b)fTa(d_xz(π^nb)) LMCT. The assignment of the re-

maining low energy transitions as TaP(π^b)fTa(d_yz-

(π^nb)) or As(4s/4p_z)^nbfTa(d_xz(π^nb)) singlet or triplet

LMCT bands also parallels the phosphinidene case.

Related interpretations of similar chalcogenide features

have been made.^51,52

1

was shifted to a broad signal at δ 5.67 in its H NMR

spectrum accompanied by a broad NH₃resonance at δ

0.80. It is suspected that the original white solid is a

combination of 3-(NH₂)₂and 1dNH(NH₃), but attempts

to isolate either invariably led to ammonia loss. The

1

¹⁵N NMR resonance at δ -60.60 (d, J_15NH= 76 Hz)

observed in solution is attributed to 1d¹⁵NH, but it is

conceivable it is slightly shifted because of equilibria with

the ammonia adduct. A switch to THF-d₈revealed four

distinct species, but the spectral signature for the fourth

species, presumably 1dNH(NH₃), continued to be ques-

tioned for lack of a signal that could be attributed with

confidence to the NH₃group.

Alternative syntheses of Me₃Si-capped imido and

phosphinidene derivatives are also given in Scheme 2.

Deprotonation of (silox)₃TadNH (1dNH) with neopen-

tyllithium afforded the lithio-imide, (silox)₃TadNLi

(1dNLi), in 78% yield as colorless crystals from tetrahy-

drofuran (THF). Subsequent silylation with TMSCl af-

forded the imide, (silox)₃TadNSiMe₃(1dNTMS, 50%),

again as colorless crystals. The corresponding route was

also employed in the preparation of (silox)₃TadPTMS

(1dPTMS), except that the phosphide anion^34,49was

generated in situ and silylated to produce a small amount

of the phosphinidene for the sake of ³¹P NMR spectros-

The surprising feature to the calculations of (silox)₃-

TadEPh (1dEPh, E = N, P, As) shown in Figure 1 are

(47) Roberts, J. H.; Bettignies, B. J. D. J. Phys. Chem. 1974, 78, 2106–

2109.

(50) Kuiper, D. S.; Douthwaite, R. E.; Mayol, A.-R.; Wolczanski, P. T.;

Lobkovsky, E. B.; Cundari, T. R.; Lam, O. P.; Meyer, K. Inorg. Chem. 2008,

47, 7139–7153.

(51) Paradis, J. A.; Wertz, D. W.; Thorp, H. H. J. Am. Chem. Soc. 1993,

115, 5308–5309.

(52) Murphy, V. J.; Parkin, G. J. Am. Chem. Soc. 1995, 117, 3522–3528.

(48) Wanna, J.; Menapace, J. A.; Bernstein, E. R. J. Chem. Phys. 1986, 85,

1795–1806.

(49) (a) Figueroa, J. S.; Cummins, C. C. Angew. Chem., Int. Ed. 2004, 43,

984–988. (b) Figueroa, J. S.; Cummins, C. C. J. Am. Chem. Soc. 2004, 126,

13916–13917.

8530 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

Figure 1. Truncated molecular orbital diagrams featuring pertinent TadE occupied levels and the ligand fields for (silox)₃TadER (1dER; E = N, P, As;

R = H, Ph). In all cases, the orbital parentage has been tentatively interpreted from orbital pictures, but the low symmetry (C_s) cases were difficult, and the

d_xz/d_yzset mixed with the d_x₂_-y₂/d_xyset in 1dNPh to such an extent that no assignment could be made. Note that the two π-bonds in 1dNH are spread

among four orbitals because of σ/π-mixing, and are further split in 1dNPh because of mixing with the phenyl orbitals.

the HOMOs of the phosphinidene and arsenidine, which

are the TadE π-bonding orbitals. The “lone pairs” or

“inert pairs” on P and As are expected to be relatively low

in energy, and the π-bonds in the system are so weak that

the TadE bonding orbitals are not driven below the inert

pairs. Clearly the large 3s/3p and 4s/4p energy gaps for

elemental phosphorus and arsenic, the origin of the so-

called inert pair effect,^53-56are revealed relative to the

π-interaction energies.

4. Pnictidene Structural Studies and Calculations.

Crystallographic studies of (silox)₃TadPPh (1dPPh,

˚

d(TaP) = 2.317(4) A, —TadP-C = 110.2(4)ꢀ) and (silox)₃-

˚

TadAsPh (1dAsPh, d(TaAs) = 2.428(2) A, —TadAs-

C = 107.2(4)ꢀ) have been reported.⁵A molecular view of

(silox)₃TadNPh (1dNPh,) and selected geometric para-

meters are given in Figure 2, and crystallographic details

are given in Table 3. In contrast to the Ta-E-C angles of

the phophinidene and arsinidene, the TaNPh linkage is

essentially linear, and the tantalum-nitrogen distance

Figure 2. Molecular view of (silox)₃TadNPh (1dNPh) with the per-

˚

ipheral Me groups removed for clarity. Selected distances (A) and angles

(deg): TaN, 1.767(10); TaO1, 1.894(7); TaO2, 1.891(7); TaO3, 1.893(6);

˚

is standard for a triple bond (d(TaN) = 1.767(10) A,

—TadN-C = 172.4(8)ꢀ).⁵⁷Utilization of the 2s orbital

of N in the imide linkage is therefore apparent, constitut-

ing persuasive evidence for sp hybridization. High level

quantum mechanical calculations of (silox)₃TadEPh

(1⁰dEPh, primes indicated calculated values) are consis-

tent with the reported structures: 1⁰dNPh, d(TaN)=1.81

SiO(ave), 1.663(11); NC1, 1.425(15); TaNC1, 172.4(8); NTaO1, 108.0(4);

NTaO2, 105.9(5); NTaO3, 109.6(3); O1TaO2, 111.4(3); O1TaO3, 111.0(3);

O2TaO3, 110.8(3); TaO1Si1, 169.2(5); TaO2Si2, 164.7(5); TaO3Si3,

172.9(5).

(1⁰dEH) show the TadE-H angles to be close to 90ꢀ for

E = P and As, therefore steric factors probably cause the

TadE-C angles for 1/1⁰dPPh and 1/1⁰dAsPh to deviate

0

˚

A, —TadN-C = 176.0ꢀ; 1 dPPh, d(TaP) = 2.36 A,

0

˚

—TadP-C=112.7ꢀ; 1 dAsPh, d(TaAs)=2.47 A, —Tad

As-C = 109.4ꢀ. Related calculations of (silox)₃TadEH

0

˚

from 90ꢀ: 1 dNH, d(TaN) = 1.79 A, —TadN-H =

0

˚

179.4ꢀ; 1 dPH, d(TaP) = 2.36 A, —TadP-H = 87.6ꢀ;

0

˚

1 dAsH, d(TaAs) = 2.46 A, —TadAs-H = 87.0ꢀ.

5. 1,2-H₂-Elimination from (silox)₃HTaEHR. An as-

sessment of the mechanism of 1,2-H₂-elimination from

(silox)₃HTaEHR (2-EHR) was made via an evaluation of

the rates as monitored by ¹H NMR spectroscopy in C₆D₆

(53) Drago, R. S. J. Phys. Chem. 1958, 62, 353–357.

(54) Power, P. P. Chem. Rev. 1999, 99, 3463–3503.

(55) Kutzelnigg, W. Angew. Chem., Int. Ed. Engl. 1984, 23, 272.

(56) Magnusson, E. J. Am. Chem. Soc. 1984, 106, 1177–1185.

(57) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; John

Wiley & Sons: New York, 1988.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8531

Table 3. Selected Crystallographic and Refinement Data for (silox)₃TadNPh (1dNPh), (silox)₃TadCPPh₃(1dCPPh₃), [(silox)₃MeTa](μ-η²-N,N:η¹-C-NHNHCH₂CH₂-

CH₂)[Ta(κ-O,C-OSi^tBu₂CMe₂CH₂)(silox)₂] (7), and [(silox)₂ClTa](μ-N₂) (8-Cl)

1dNPh

1dCPPh₃

7

8-Cl

formula

formula wt

space group

Z

C

918.34

Pna2₁

4

18.114(4)

13.436(3)

20.650(4)

90

₄₂H₈₆NO₃Si₃Ta

C₅₅H₉₆O₃PSi₃Ta

1101.51

P2₁/n

C₇₆H₁₇₂N₂O₆Si₆Ta₂

1740.60

P1

C₄₈H₁₀₈N₂O₄Cl₂Si₄Ta

1322.52

P2₁/c

4

2

4

˚

a, A

11.8368(2)

22.2055(4)

22.1197(4)

90

90.1200(10)

90

5813.97(18)

1.258

2.018

173(2)

0.71073

R₁= 0.0404

wR₂= 0.0739

R₁= 0.0674

wR₂= 0.0824

1.028

12.7517(6)

17.5127(9)

23.2899(11)

103.755(2)

90.992(2)

110.627(2)

4698.7(4)

1.230

2.446

173(2)

0.71073

R₁= 0.0611

wR₂= 0.1429

R₁= 0.0912

wR₂= 0.1549

1.059

14.6904(6)

27.0349(10)

17.4888(7)

90

113.637(2)

90

6363.0(4)

1.381

3.633

173(2)

0.71073

R₁= 0.0566

wR₂= 0.1309

R₁= 0.0980

wR₂= 0.1561

1.014

˚

b, A

˚

c, A

R, deg

β, deg

γ, deg

3

˚

V, A

F

5025.8(19)

1.214

2.291

_calc, g cm^-3

3

μ, mm^-1

temp, K

293(2)

˚

λ (A)

0.71073

R₁= 0.0415

wR₂= 0.0976

R₁= 0.0733

wR₂= 0.1186

1.080

R indices [I > 2σ(I)]^a,b

R indices (all data)^a,b

GOF^c

P

2 1/2

]

^aR₁= ||F_o| - |F_c||/ |F_o|. ^bwR₂= [ w(|F_o| - |F_c|)²/ wF_o

reflections, p = number of parameters.

.

^cGOF (all data) = [ w(|F_o| - |F_c|)²/(n - p)]^1/2, n = number of independent

Table 4. Rate Constants^a,bfor 1,2-H₂-Elimination from (silox)₃HTaEHR (2-EHR), with and without Added H₂ER and Radical Trap C₆H₈Obtained by ¹H NMR

Spectroscopic Monitoring

cmpd

(silox)₃HTaNH₂(2-NH₂)^c,d

[HTaEHR]

[H₂ER]

[C₆H₈]

T(ꢀC)

k (ꢀ10⁵s^-1

166(5)

42.2(7)

18.8(4)

7.3(1)

2.73(2)

2.49(6)

29.0(10)

2.60(4)

3.8(2)

5.2(1)

22(1)

23.2(6)

20.0(2)

1.14(2)

1.12(8)

1.06(1)

1.04(7)

1.07(2)

0.26(2)

0.272(6)

0.28(3)

0.275(4)

140(10)

33(1)

)

0.050

0.056

0.053

0.029

0.038^e

0.060

0.062

0.058^g

0.033^h

0.053

0.065^h,i

0.062

0.034

0.056

0.060

0.029^m

0.064^j

0.037^j

0.054^m

0.053

0.037

0.040

136.2(4)

119.0(4)

109.6(4)

99.1(4)

90.0(0)

24.8(4)

-76(1)

-9.5(10)

(silox)₃HTaNHMe (2-NHMe)

(silox)₃HTaNPh (2-NHPh)^f

0.363

0.373

(silox)₃HTaNH(C₆H₄-p-Me) (2-NHC₆H₄-p-Me)

(silox)₃HTaNH(C₆H₄-p-NMe₂) (2-NHC₆H₄-p-NMe₂)

(silox)₃DTaNH(C₆H₄-p-NMe₂) (2D-NHC₆H₄-p-NMe₂)

(silox)₃HTaNH(C₆H₄-p-CF₃) (2-NHC₆H₄-p-CF₃)

(silox)₃DTaNH(C₆H₄-p-CF₃) (2D-NHC₆H₄-p-CF₃)^j

(silox)₃HTaPH₂(2-PH₂)^k,l

0.060

0.054

(silox)₃HTaPHPh (2-PHPh)^m

(silox)₃HTaAsHPh (2-AsHPh)ⁿ

0.109

0.037

0.398

33(1)

^aDetermined from unweighted, non-linear, least-squares fitting of the exponential form of the rate expression (averaged where simultaneous runs

were obtained); concentrations are in M. ^b2-NHR generated from 1 and RNH₂unless otherwise indicated; monitored via disappearance of TaH. ^c2-NH₂

isolated prior to runs. ^dΔH^q= 25.3(8) kcal/mol; ΔS^q= -10(2) eu. ^eMonitored via loss of NH(CH₃). ^fGenerated from 2-H and RNH₂. ^gMonitored by

disappearance of ArCH₃. ^hMonitored via the appearance of imido N(CH₃)₂. ⁱGenerated from (silox)₃TaD₂(2D-D) and H₂NC₆H₄-p-NMe₂; run under

∼1 atm D₂. ^jGenerated from (silox)₃TaD₂(2D-D) and H₂NC₆H₄-p-CF₃; monitored by disappearance of NH. ^kIsolated prior to run. ^lMonitored via

appearance of PH. ^mMonitored via disappearance of TaH. ⁿ2-EHR generated from 1 and H₂ER.

or C₇D₈. Table 4 lists rate constants obtained (typically

an average of two or three runs) under various conditions.

Rates were measured by generating 2-EHR in situ when

isolation of the pure compound was not possible. The eli-

minations were found to be first order in metal-hydride,

and while the addition of H₂NMe was found to accelerate

the elimination rate in the case of 2-NHMe, the addition

of PhNH₂and PhPH₂to 2-NHPh and 2-PHPh, respec-

tively, failed to induce a change in the observed rates. It

may be that only small bases have the ability to catalyti-

cally assist in the H-transfer, perhaps via the formation of

(silox)₃Ta(EHR)₂and H₂, and subsequent elimination of

H₂ER. EH₃could not be added to 2-EH₂, since 3-(EH₂)₂

was found to form for E = N, As. Trace amounts of

EH₂R were common in NMR tube experiments where

hydrolysis of Ta-EHR bonds can typically occur during

8532 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

Figure 3. Hammett plot for 1,2-H₂-elimination from (silox)₃HTa(NHC₆H₄-p-X) (2-NHC₆H₄-p-X; X (σ_F) = NMe₂(-0.83), CH₃(-0.17), H (0.0), CF₃

(0.54)).

sealing of the tube. Fortunately, 2-NH₂and 2-PH₂could

be isolated and purified prior to kinetics runs, no evidence

of 3-(NH₂)₂and 3-(PH₂)₂was obtained during the runs,

and the data for these substrates was reproducible. The

results are thus interpreted as a straightforward 1,2-H₂-

elimination, but with the caveat that trace base interfer-

ence cannot be completely ruled out.⁵⁸

(i.e., k(2-NHC₆H₄NMe₂)/k(2D-NHC₆H₄NMe₂)) and

1.02(10) (i.e., k(2-NHC₆H₄CF₃)/k(2D-NHC₆H₄CF₃))

are not significant,⁶⁴since the hydrogen being transferred

is the one attached to N. No scrambling between hydride

and amide sites were observed. Unfortunately, measure-

ments of the complementary KIE for k(2-NHC₆H₄-

NMe₂)/k(2-NDC₆H₄NMe₂) were not conducted because

it proved difficult to prepare 2-NDC₆H₄NMe₂in suffi-

cient isotopic purity.

A separate problem was revealed upon reproduction

of data pertaining to the 1,2-H₂-elimination from (silox)₃-

HTaAsHPh (2-AsHPh). While an elimination rate of

1.4(1) ꢀ 10^-3s^-1was obtained at -76 ꢀC, it was subse-

quently found that this was due to processes that were

shut down via the addition of 1,4-cyclohexadiene, a

common radical trap. If ΔS^qfor H₂elimination from

2-AsHPh is also -10 eu, then its rate can be estimated at

24.8 ꢀC to be ∼2.0 ꢀ 10^-4s^-1, and the relative 1,2-H₂-

elimination rates from (silox)₃HTaEHPh (2-EHPh) are

As (5800) > N (14) > P (1).

Attempts to Synthesize [(silox)₃Ta](μ-N₂). 1. 1,2-H₂-

Elimination and Aziridine Strategies. As Scheme 2 implies,

the oxidative addition of a substrate such as hydrazine,

H₂N-NH₂, was envisaged to undergo 1,2-NH-additions

to sequential (silox)₃Ta (1) centers in concert with 1,2-H₂-

eliminations to afford the desired N₂complex, [(silox)₃-

Ta]₂(μ-N₂) (1₂-N₂). Unfortunately, elimination of NH₃

instead of H₂occurred independent of stoichiometry,

yielding (silox)₃TadNH (1dNH, eq 1).⁶⁵Negligible

amounts of

Some of the rate data are readily interpreted, as the

Hammett plot for 1,2-H₂-elimination from (silox)₃HTa-

(NHC₆H₄-p-X) (2-NHC₆H₄-p-X) in Figure 3 exemplifies.

Its slope of -0.95 suggests a stabilization of positive

charge at nitrogen in the transition state, perhaps an

indication that greater NfTa donation increases the

rate of reaction through both π- (p-NMe₂) and σ-dona-

tion (p-Me). A temperature dependent study (90.0(4)-

136.2(4)ꢀC) of 1,2-H₂-elimination from (silox)₃HTaNH₂

(2-NH₂) afforded Eyring parameters of ΔH^q= 25.3(8)

kcal/mol and ΔS^q= -10(2) eu.^59-61The parameters

permit the 1,2-H₂-elimination rate for 2-NH₂to be calcu-

lated at 24.8 ꢀC as 1.11 ꢀ 10^-8s^-1, thus the relative rates

for 2-NHR are R = Ph (3420), Me (2340), and H (1). It

appears that the π-system of Ph and the σ-donor ability of

Me speed up the elimination, and these relative rates are

significantly faster than H. The trend is consistent with

the expectedly weaker D(NH) for the pseudobenzylic

-NHPh group (D(NH) = 92 kcal/mol in H₂NPh)⁶²versus

the -NHMe (D(NH) = 100 kcal/mol in H₂NMe) and

-NH₂(D(NH) = 107 kcal/mol in NH₃)⁶³fragments, sug-

gesting that the transition state is early, which is typical

for an irreversible process. The trend also parallels pK_a’s

of the corresponding aniline (30.6),⁶²methylamine

(∼38-40), and ammonia (41), although the disparity

between the Me and H substituents in 2-NHR relative

to Ph is greater than predicted. The k_H/k_D’s of 1.10(6)

C₆D₆

ðsiloxÞ₃Ta þ H₂N-NH₂f ðsiloxÞ₃TadNH þ NH₃ð1Þ

23ꢀC

1

2dNH

C₆D₆

ðsiloxÞ₃Ta þ H₂N-Nð NC₂H₃MeÞ f ðsiloxÞ₃TadNH

c

23ꢀC

1

2dNH

(58) Baratta, W.; Siega, K.; Rigo, P. Chem. Eur. J. 2007, 13, 7479–7486.

;

þ H^cNC₂H₃Me

1dNH(NH₃) and (silox)₃Ta(NH₂)₂(3-(NH₂)₂) were ob-

served, as expected for the low concentration of product

ð2Þ

(59) Bennett, J. L.; Wolczanski, P. T. J. Am. Chem. Soc. 1997, 119, 10696–

10719.

(60) Schaller, C. P.; Cummins, C. C.; Wolczanski, P. T. J. Am. Chem. Soc.

1996, 118, 591–611.

(61) Schaller, C. P.; Wolczanski, P. T. Inorg. Chem. 1993, 32, 131–144.

(62) Bordwell, F. G.; Zhang, X.-M.; Cheng, J.-P. J. Org. Chem. 1993, 58,

6410–6416.

(63) McMillen, D. F.; Golden, D. M. Annu. Rev. Phys. Chem. 1982, 33,

493–532.

(64) Carpenter, B. K. Determination of Reaction Mechanisms; Wiley-

Interscience: New York, 1984.

(65) Shaver, M. P.; Fryzuk, M. D. J. Am. Chem. Soc. 2005, 127, 500–501.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8533

Scheme 3

NH₃. It appears that 1,2-H₂-elimination from putative

(silox)₃HTaNH(NH₂) (2-NHNH₂) is simply uncompeti-

tive with respect to 1,2-HN-elimination, although N-N

oxidative addition followed by NH₃loss cannot be ex-

cluded. A related approach, utilizing the addition of

N-amino-2-methylaziridine,⁶⁶was considered in the hope

that propene loss and formation of (silox)₃TadN-NH₂

(1dNNH₂) would permit a stepwise synthesis of 1₂-N₂.

Unfortunately, (silox)₃TadNH (1dNH) was the pro-

duct, consistent with 2-methylaziridine loss analogous

to that delineated for hydrazine (eq 2); byproducts (e.g.,

(silox)₃Ta(η-C₂H₃Me) as in Scheme 2) were consistent

with the reaction of 2-methylaziridine (shown as the

initial product in eq 2) with 1.

2. Synthesis of (silox)₃TadCH₂. In an attempt to obvi-

ate the presumed deleterious, swift 1,2-HN-elimination

reaction, routes to (silox)₃RTaNHNH₂and (silox)₃-

RTaNH(2-Me-aziridine) were sought in the hope that

1,2-RH-elimination to give the desired imides would be

controllable. As Scheme 3 illustrates, successful imple-

mentation of this route required the preparation of

(silox)₃TadCH₂(1dCH₂),²⁶which was accomplished

via methylene transfer to (silox)₃Ta (1) from Ph₃Pd

CH₂;⁶⁷related attempts with Me₃PCH₂,⁶⁸which would

yield the easily removable PMe₃as a byproduct, afforded

complex mixtures. As is typical for reactions with the

Wittig reagent, removal of byproduct Ph₃P was difficult,

but a 29% yield of 1dCH₂was obtained after multiple

crystallizations from hexane. During the course of opti-

mizing the synthesis, a curious byproduct was obtained

that was devoid of hydrogens other than those of phenyl

or silox. The compound, which could be isolated in 20%

yield, possessed a single ³¹P{¹H} NMR spectral shift at δ

-4.26 and a unique ¹³C{¹H} NMR spectral resonance at

δ 175.01 that appeared as a doublet with J_PC= 53 Hz,

consistent with the formulation (silox)₃TadCdPPh₃

(1dCPPh₃). Its structure was confirmed by X-ray crystal-

lography.

Figure 4. Molecular view of (silox)₃TadCdPPh₃(1dCPPh₃). Selected

˚

distances (A) and angles (deg): Ta-C37, 1.862(2); Ta-O1, 1.9122(16);

Ta-O2, 1.9072(15); Ta-O3, 1.9139; P-C37, 1.667(2); P-C^Ph(ave),

1.816(3); Si-O(ave), 1.655(5); Ta-C37-P, 177.83(17); O1-Ta-C37,

109.24(9); O2-Ta-C37, 108.66(9); O3-Ta-C37, 108.27(9); O1-Ta-

O2, 110.45(7); O1-Ta-O3, 108.81(7); O2-Ta-O3, 111.36(7); C37-

P-C^Ph(ave), 113.7(7); C^Ph-P-C^Ph(ave), 105.0(7); Ta-O-Si(ave),

163.6(19).

triple bond character.⁵⁹The accompanying PC distance

˚

of 1.667(2) A is slightly shorter than expected for a

˚

P-C(sp) single bond (1.78 A), consistent with consider-

able double bond character; hence, the unit can be con-

strued as having contributions from both TadCdP and

Ta^(-)tC-P^(þ)resonance structures. Similar constructs

have been useful in rationalizing phosphacarbene/yne

groups in related molecules.^69-72The remainder of the

core has standard (silox)₃Ta^Vparameters that are given in

the caption to Figure 3.

3. Structure of (silox)₃TadCdPPh₃. Selected crystal-

lographic details of (silox)₃TadCdPPh₃(1dCPPh₃) are

given in Table 3, and Figure 4 illustrates its near C_3vcore

configuration, with the silox groups staggered relative to

the phosphine phenyl substituents on the periphery. The

Ta-C-P linkage is virtually linear (177.83(17)ꢀ), and the

(69) (a) Jamison, G. M.; White, P. S.; Templeton, J. L. Organometallics

1991, 10, 1954–1959. (b) Bruce, A. E.; Gamble, A. S.; Tonker, T. L.; Templeton,

J. L. Organometallics 1987, 6, 1350–1352.

(70) List, A. K.; Hillhouse, G. L.; Rheingold, A. L. Organometallics 1989,

8, 2010–2016.

˚

d(TaC) is quite short at 1.862(2) A, signifying substantial

(71) (a) Li, X. Y.; Wang, A.; Wang, L.; Sun, H.; Harms, K.; Sundermeyer,

J. Organometallics 2007, 26, 1411–1413. (b) Li, X. Y.; Sun, H. J.; Harms, M.;

Sundermeyer, J. Organometallics 2005, 19, 4699–4701. (c) Li, X. Y.; Schopf, M.;

Stephan, J.; Harms, M.; Sundermeyer, J. Organometallics 2002, 21, 2356–2358.

(72) Cordiner, R. L.; Gugger, P. A.; Hill, A. F.; Willis, A. C. Organome-

tallics 2009, 28, 6632–6635.

(66) Hillers, S.; Eremeev, A. V.; Lidaks, M.; Pestunovich, V. A.; Liepin,

E.; Kalvins, I. Khim. Geterotsikl. Soedin. 1971, 7, 45–48.

(67) Sharp, P. R.; Schrock, R. R. J. Organomet. Chem. 1979, 171, 43–51.

(68) Klein, H. F. Inorg. Synth. 1978, 18, 139.

8534 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

Scheme 4

4. 1,2-RH-Elimination Attempts. As Scheme 3 reveals,

the addition of hydrazine to (silox)₃TadCH₂(1dCH₂)

afforded (silox)₃MeTaNHNH2 (4-NHNH₂) in 70%

yield. Although 4-NHNH₂was modestly thermally sensi-

tive, NMR spectral characterization was readily accom-

plished using ¹H-¹⁵N HSQCAD correlation spectros-

copy.⁷³In addition to the silox and Me proton resonances,

different nitrogens at δ 64.80 and δ 148.47 (J_NH= 78 Hz)

were observed to correspond with proton signals at δ 3.63

and δ 4.11, and these were assigned to the NH₂and NH

groups, respectively. Similarly, treatment of 1dCH₂with

amino-2-methylaziridine provided (silox)₃MeTaNH-

(-^cNCHMeCH₂) (4-NH(azir)) in 60% yield as a waxy

solid. The aziridine nitrogen was indirectly found at δ

81.99 (¹⁵N), and the accompanying NH was located at δ

244.46 with J_NH= 72 Hz. Subsequent thermolyses of the

methyl-hydrazides generated derivatives that proved to

be more complicated than those expected from simple

1,2-MeH-eliminations.

Thermal degradation of (silox)₃MeTaNHNH2 (4-NH-

NH₂) led to the loss of MeH, but evidence of (silox)H was

also obtained, and NMR spectroscopic data were more

consistent with a dinuclear species with four inequivalent

silox groups. Attempts to rid the solutions of byproduct

(silox)H invariably led to further degradation, thus it is

conceivable that the silanol is weakly hydrogen-bonded

to the product, especially when solutions are concen-

trated. Successful implementation of ¹H-¹⁵N HMBCAD

NMR correlation spectroscopy enabled the connectivity

of the N-containing bridges to be ascertained, and an

elusive Me group, partially masked by the silox reso-

1

nances, was brought to light by H-¹³C gHSQC NMR

correlation spectroscopy.⁷⁴The product is thought to

be [(silox)₂TaMe](μ-N_RHN_β)(μ-N_γHN_δH)[Ta(silox)₂] (5),

which is portrayed in Scheme 4. Attempts to crystallize 5

in the presence of (silox)H led to waxy solids of reason-

able purity, but these were not amenable to X-ray diffrac-

tion analysis. As such, the hapticity and orientation of

bridges and silox groups are undetermined. Nonetheless,

the NMR spectroscopy clearly shows a diamido μ-NH-

NH bridge accompanied by an imido-amide μ-NH-N

unit. The shift of one amide (N_γH, δ 79.50) is substan-

tially lower than the remaining two (N_RH, δ 131.27; N_δH,

δ 179.70), and may be indicative of a strong σ-donor (i.e.,

Me) opposite it in the complex. The imido nitrogen is

observed more downfield at δ 298.23. EXSY correlation

spectroscopy revealed exchange between N_γH and N_δH,

thus it is possible that this bridge exists as an η²,η^2/1-NH,

NH unit^12,75,76or this configuration is at least energeti-

cally feasible to permit rapid exchange, as Scheme 4

shows.

(74) Friebolin, H. Basic One- and Two-Dimensional NMR Spectroscopy;

Wiley-VCH: Weinheim, 2005.

(75) Fryzuk, M. D.; Love, J. B.; Rettig, S. J.; Young, V. G. Science 1997,

275, 1445–1447.

(76) Bernskoetter, W. H.; Pool, J. A.; Lobkovsky, E.; Chirik, P. J. J. Am.

Chem. Soc. 2005, 127, 7901–7911.

(73) Koshino, H.; Uzawa, J. Kagaku Seibutsu 1995, 33, 252–258.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8535

Figure 5. Molecular view of [(silox)₃MeTa](μ-η²-N,N:η¹-C-NHNHCH₂CH₂CH₂)[Ta(κ-O,C-OSi^tBu₂CMe₂CH₂) (silox)₂] (7) with the methyls of the ^tBu

˚

groups removed for clarity. Selected distances (A) and angles (deg): Ta1-O1, 1.862(8); Ta1-O2, 1.932(5); Ta1-O3, 1.905(5); Ta1-C24, 2.217(9);

Ta1-C73, 2.133(8); C73-C74, 1.523(11); C74-C75, 1.481(12); N1-C75, 1.441(11); N1-N2, 1.384(11); Ta2-N1, 2.283(7); Ta2-N2, 2.008(7); Ta2-C76,

2.211(9); Ta2-O4, 1.915(5); Ta2-O5, 1.951(5); Ta2-O6, 1.896(6); Si-O(ave), 1.662(11); O1-Ta1-O2, 100.4(2); O1-Ta1-O3, 104.8(2); O2-Ta1-O3,

154.6(2); O1-Ta1-C24, 123.8(3); O1-Ta1-C73, 111.0(3); O2-Ta1-C24, 77.0(3); O2-Ta1-C73, 86.0(3); O3-Ta1-C24, 86.1(3); O3-Ta1-C73,

88.3(3); O4-Ta2-N1, 108.7(3); O4-Ta2-N2, 85.0(3); O4-Ta2-O5, 152.7(2); O4-Ta2-O6, 100.4(2); O4-Ta2-C76, 83.1(3); O5-Ta2-N1, 79.4(2);

O5-Ta2-N2, 87.4(3); O5-Ta2-O6, 105.3(2); O5-Ta2-C76, 82.8(3); O6-Ta2-N1, 93.1(3); O6-Ta2-N2, 126.0(3); O6-Ta2-C76, 101.3(3);

N1-Ta2-N2, 36.9(3); N1-Ta2-C76, 159.6(3); N2-Ta2-C76, 132.6(3); Ta1-C73-C74, 118.2(6); C73-C74-C75, 114.5(8); C74-C75-N1, 115.8(8);

N2-N1-C75, 114.7(7); N1-N2-Ta2, 82.4(4); Ta2-N1-C75, 136.8(6); Ta2-N1-N2, 60.7(4); Ta1-O2-Si2, 131.5(4); O2-Si2-C21, 96.4(4);

Si2-C21-C24, 102.7(6); Ta1-C24-C21, 119.7(6).

The failure to generate (silox)₃TadNNH₂as a

stable entity^77,78prompted a turn toward the aziridine

chemistry, and (silox)₃MeTaNH(-^cNCHMeCH₂) (4-NH-

(azir)) was treated with 2 equiv of (silox)₃Ta (1) with the

hope of producing (silox)₃MeTaNH-NdTa(silox)₃and

(silox)₃Ta(η-C₃H₆). Unfortunately, no deolefination of

the aziridine was noted at 23 ꢀC, and thermolysis

4-NH(azir) under varied conditions could not be accom-

plished without significant cyclometalation of 1 to give

(silox)₂HTa(κ-O,C-OSi^tBu₂CMe₂CH₂) (1-cymet).^6,15Dur-

ing the thermolysis, several products were noted, includ-

ing those containing olefinic residues, consistent with

ring-opening of the aziridine⁷⁹to (silox)₃MeTaNHNH-

CH₂CHdCH₂(6). Yellow crystals that proved to be

nearly insoluble in common hydrocarbons were observed

to form during this process, and X-ray crystallographic

studies revealed this material to be [(silox)₃MeTa](μ-η²-

N,N:η¹-C-NHNHCH₂CH₂CH₂)[Ta(κ-O,C-OSi^tBu₂CMe₂-

CH₂)(silox)₂] (7). As Scheme 4 illustrates, 7 is best thought

of arising from insertion of the olefin of 6 into the Ta-H

bond of 1-cymet.¹⁵

6. Structure of [(silox)₃MeTa](μ-η²-N,N:η¹-C-NH-

NHCH₂CH₂CH₂)[Ta(K-O,C-OSi^tBu₂CMe₂CH₂) (silox)₂]

(7). With solubility precluding a definitive NMR analysis

of [(silox)₃MeTa](μ-η²-N,N:η¹-CNHNHCH₂CH₂CH₂)-

[Ta(κ-O,C-OSi^tBu₂CMe₂CH₂)(silox)₂] (7), an X-ray crys-

tallographic study (Table 3) was conducted on the

aforementioned yellow crystals. Despite one badly dis-

ordered set of silox ^tBu groups (on Si6), the study yielded

a reasonable model that illustrates the ring-opened azir-

idine product as illustrated by Figure 5. The cyclometa-

lated center is fairly well described as a pseudo-tbp

geometry with the two less-electronegative alkyl substitu-

ents in their expected equatorial positions.^3,80The chain

˚

linkage, Ta1-C73, is 2.133(8) A, and the cyclometalated

˚

d(Ta1-C24) is 2.217(9) A, and both are separated by

124.5(4)ꢀ. The remaining pseudo-equatorial position is

occupied by O1, which is 1.862(5) A from the tantalum,

˚

and 111.0(3)ꢀ and 123.8(3)ꢀ from C73 and C24, respec-

tively. Siloxide oxygens O2 and O3 are 1.932(5) and

˚

1.905(5) A from the tantalum, respectively, which are

slightly longer than O1 and consistent with their pseu-

doaxial positions. The bite angle of the cylometalated

silox is 77.0(3)ꢀ, and its oxygen, O2, is 100.4(2)ꢀ and

86.0(3)ꢀ from O1 and C73, respectively.

The treatment of (silox)₃TadCH₂(1dCH₂) with di-

imines was also considered in view of Schrock’s prepara-

tion of [(THF)₂Cl₃Ta]₂(μ-N₂).⁹Unfortunately, stable

diimines such as PhCHdN-NdCHPh failed to react

with the methylene derivative, presumably because of

deleterious steric interactions in the putative 4-membered

ring intermediate.

One can similarly assess the hydrazide-methyl core,

˚

taking TaO4 (1.915(5) A) and TaO5 (1.951(5) A)

˚

as pseudoaxial siloxides (—O4-Ta2-O5 = 152.7(2)ꢀ),

and Ta2O6 (1.896(6) A) as a pseudo-equatorial siloxide,

˚

100.4(2)ꢀ and 105.3(2)ꢀ from O4 and O5, respectively.

(77) Sebe, E.; Heeg, M. J.; Winter, C. H. Polyhedron 2006, 25, 2109.

(78) Tonks, I. A.; Bercaw, J. E. Inorg. Chem. 2010, 49, 4648–4656.

(79) Kashelikar, D. V.; Fanta, P. E. J. Org. Chem. 1961, 26, 1841–1842.

(80) Strazisar, S. A.; Wolczanski, P. T. J. Am. Chem. Soc. 2001, 123, 4728–

4740.

8536 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

Scheme 5

This assessment treats the NH-NHR ligand as occupying

˚

one site. The methyl is pseudo-equatorial (2.211(9) A)

to attack by nitride (silox)₃TaNLi (1dNLi), as indi-

cated by eq 4.⁸¹Unfortunately, attempts to halogenate

the nitride anion with Br₂, I₂, or NBS failed to elicit any

tractable material under a variety of conditions. Since

(silox)₃TadNH was a common product in virtually all of

the attempts, it appears that 1e^-reactivity is at the core of

the problem.

along with the amide of the hydrazide (d(Ta2-N2) =

˚

2.008(7) A; d(N1-N2) = 1.384(11) A), and the trans

influence of the donor nitrogen, N2 (2.283(7), —C76-

Ta2-N1 = 159.6(3)ꢀ), helps slightly weaken the methyl

interaction. The remaining core distances and angles are

listed in the Figure 5 caption. Distances and angles

pertaining to the bridge are consistent with a (CH₂)₃-

NHNH linkage derived from ring-opening of the aziri-

dene and trapping of the subsequent olefin by the hydride

of the 1-cymet, just as Scheme 4 portrays.

Compromise, [(silox)₂ClTa](μ-N₂). 1. Synthesis of

[(silox)₂ClTa](μ-N₂). Although a number of variants on

the preceding methods are plausible, Schrock’s diimide

complexes, in particular [(THF)₂Cl₃Ta]₂(μ-N₂),⁹were

considered as starting materials for metathesis reactions

to attain some measure of success in a reasonable period

of time. As Scheme 5 illustrates, treatment of the chloro-

diimide derivative with 4 equiv of Na(silox) afforded

[(silox)₂ClTa](μ-N₂) (8-Cl) is 56% yield. Further metath-

esis with Na(silox) or Tl(silox) on 8-Cl failed to provide

the desired [(silox)₃Ta]₂(μ-N₂) (1₂-N₂) complex, and ther-

molysis ultimately led to degradation. Prior work⁴sug-

gested that smaller siloxides/alkoxides would not sup-

port a 3-coordinate tantalum center. The chloride was

converted to the Me derivative, [(silox)₂MeTa](μ-N₂)

(8-Me, 70%) with MeMgBr, but subsequent attempts at

silanolysis reactions with (silox)H did not generate MeH,

and heating again led to decomposition. NMR spectra

of 8-Cl and 8-Me revealed one type of silox, consistent

with either C_2hsymmetry and/or free rotation about the

bridge.

7. Attempted Oxidative Coupling of (silox)₃TaNLi.

Conceptually, oxidation of the nitride anion, (silox)₃-

TaNLi (1dNLi), can lead to the desired N₂-complex,

but one electron oxidants invariably generated (silox)₃-

TadNH (1dNH), presumably via H-atom abstraction by

an incipient (silox)₃TaN species. A more probable strat-

3

egy would rely on N-N oxidative coupling induced by

reductive elimination from L_nM(NTa(silox)₃)₂, as eq 3

implicates. Utilization of (Ph₃P)₂PtCl₂, (Et₃P)₂PdCl₂,

(DME)NiCl₂, and other related complexes failed to in-

duce N-N bond formation, and no evidence for M-N

intermediates was found. Only for Cp₂TiCl₂was NMR

spectroscopic evidence for Cp₂ClTi-NdTa(silox)₃ob-

tained, and its thermolysis yielded 1dNH and Cp₂TiCl.

Attempts to oxidatively add the NH bond of 1dNH to

(Ph₃P)₄Pd and trap the product hydride with olefin also

failed to provide any evidence of (Ph₃P)_nPdH(NdTa-

(silox)₃).

2. Structure of [(silox)₂ClTa]₂(μ-N₂). Table 3 lists per-

tinent crystallographic information pertaining to the

structure of [(silox)₂ClTa]₂(μ-N₂) (8-Cl), and Figure 4

provides two views of the molecule. Unfortunately, while

the data confirm the C₂structure, the electron density

between the two heavy tantalum atoms would not yield a

reliable solution; using an isotropic model for the bridg-

˚

ing nitrogens, TaN distances of 1.780(7) and 1.739(8) A,

˚

and a d(NN) of 1.355(11) A were found, but these should

not be construed as accurately portraying the bridge.

Nonetheless, the conformation of 8-Cl was confirmed,

including a pseudo-eclipsed placement of the core sub-

stituents relative to the adjacent tantalum, as previously

seen in [(silox)₂XM]₂systems.⁸²This arrangement essen-

tially minimizes steric interactions across the bridge by

A potentially simpler route would require conversion

of the anionic nitride into an electrophile of the type

(silox)₃TadNX (1dNX), which would then be susceptible

(81) Figueroa, J. S.; Piro, N. A.; Clough, C. R.; Cummins, C. C. J. Am.

Chem. Soc. 2006, 128, 940–950.

(82) Miller, R. L.; Lawler, K. A.; Bennett, J. L.; Wolczanski, P. T. Inorg.

Chem. 1996, 35, 3242–3253.

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8537

Figure 6. (a) Molecular view of [(silox)₂ClTa]₂(μ-N₂) (8-Cl) with peripheral Me groups removed for clarity; note the N atoms refined isotropically; see text

˚

for comments pertaining to the TaNNTa bridge. (b) End on view of 8-Cl revealing typical C₂symmetry of dimeric species. Selected distances (A) and angles

(deg): Ta1-O1, 1.869(6); Ta1-O2, 1.871(7); Ta2-O3, 1.868(7); Ta2-O4, 1.854(6); Ta1-Cl1, 2.312(2); Ta2-Cl2, 2.307(3); Si-O(ave), 1.652(5);

N1-Ta1-O1, 109.2(3); N1-Ta1-O2, 111.4(3); N1-Ta1-Cl1, 107.5(3); O1-Ta1-O2, 110.7(3); O1-Ta1-Cl1, 109.5(2); O2-Ta1-Cl1, 108.5(2);

N2-Ta2-O3, 111.6(4); N2-Ta2-O4, 108.5(3); N2-Ta2-Cl2, 107.9(3); O3-Ta2-O4, 109.6(3); O3-Ta2-Cl2, 109.8(2); O4-Ta2-Cl2, 109.5(2).

trading two minimal silox/Cl interactions for one signifi-

cant silox/silox interaction. Only one type of silox

group is observed in NMR spectra of 8-Cl or [(silox)₂-

MeTa]₂(μ-N₂) (8-Me), presumably because of rapid rota-

tion about the TaNNTa linkage. Distances and angles of

the pseudo-tetrahedral cores (109.5(13)ꢀ) are listed in

Figure 6.

while the related experiments were not conducted in this

study, the absence of a significant isotope effect in 1,2-

HD-elimination from (silox)₃DTaNH(C₆H₄-p-X) (2D-

NHC₆H₄-p-X; X = CF₃, NMe₂) suggests that the activa-

tion is primarily one involving N-H breakage.

The Hammett study of 1,2-H₂-elimination from (silox)₃-

HTa(NHC₆H₄-p-X) (2-NHC₆H₄-p-X) in Figure 3, the

relative rates of dihydrogen loss from (silox)₃HTa-

(NHR) (2-NHR; R = Ph (3420), Me (2340), and H (1)),

and the modest correlation with pK_a’s of the correspond-

ing amines support the contention that positive charge on

N develops in the transition state. In 2-NHR, NfTa

π-bonding also justifies the acidic character of the NH

hydrogen, while the electropositive nature of tanta-

lum relative to hydrogen ensures that the TaH bond

is Ta(δ^þ)-H(δ^-). Two additional factors suggest that

N-H bond breaking, viewed as having a (NH)δ^þcom-

ponent and hence having some proton transfer character,

is the major bond activation. First, virtually no isotope

effect is observed in the hydride position, as assessed

via k(2-NHC₆H₄-p-X)/k(2D-NHC₆H₄-p-X)) (X = NMe₂,

KIE = 1.10(6); CF₃, KIE = 1.02(10)). Second, a modest

inverse correlation of rate versus D(H-NHR) is observed,

where the weaker the bond, the faster the 1,2-H₂-elimina-

tion.

3. 1,2-H₂-Elimination from (silox)₃HTaEHR. While

dihydrogen elimination from the nitrogen-containing

species was reasonably explained, and consistent with

the widely held view of charge distribution in early metal

complexes, inclusion of phosphorus and arsenic muddies

the waters. To reiterate, if ΔS^qfor H₂elimination from

2-AsHPh is assumed to be -10 eu, the relative 1,2-H₂-

elimination rates from (silox)₃HTaEHPh (2-EHPh) at

24.8 ꢀC are As (5800) > N (14) > P (1). The aforemen-

tioned thermodynamic D(E-H) or ΔH_fꢀ (EH₃) correla-

tions predict a standard periodic trend; As should be

faster than P, followed by N. Given the relative paucity

of information on the actual thermodynamics of such

events, high level quantum calculations were conducted.

Discussion

Pnictogen Chemistry. 1. H₂ER Oxidative Addition to

(silox)₃Ta. A previous study that employed substituents

on H₂N(C₆H₄-X) (X, k_rel: 3-CF₃, 0.30(7); 3-F, 0.63(12);

4-F, 0.89(17); H, 1; 4-Ph, 0.93(17); 4-Me, 1.2(2); 4-OMe,

1.9(2); 4-NMe₂, 2.5(2)) examined the relative rates

(aniline at k_rel= 1) of NH oxidative addition via compe-

tition experiments.³As the basicity of the aniline in-

creased (F = -0.69, R = 0.93), the relative rate of oxi-

dative addition increased, consistent with nucleophilic

attack by the amine at a vacant d_xz/d_yzorbital preceding

oxidative addition. The correlation was modest, and

separate Hammett parameters were indicative of slightly

greater resonance than inductive contributions. While

complementary studies of REH₂oxidative addition to

(silox)₃Ta (1) were not examined, they qualitatively cor-

related with basicity of the substrate modified by steric

factors.

2. 1,2-Dihydrogen-Elimination from (silox)HTaNHR.

The 1,2-dihydrogen elimination data listed in Table 4

represent a limited, but reasonably interpretable set with

regard to (silox)₃HTaNHR (2-NHR). The activation

parameters (ΔH^q= 25.3(8) and ΔS^q= -10(2) eu) for

H₂elimination from (silox)₃HTaNH₂(2-NH₂) are fully

consistent with previous 1,2-RH-eliminations of RH

(silox)_n(^tBu₃SiNH)_3-nMR species, whose ΔS^qvalues

were typically -12 to -13 eu.^59-61Since H does not need

to reorganize prior to elimination, in contrast to R, the

slightly less negative activation entropy observed is ex-

pected. Loss of RH/D from (silox)₂(^tBu₃SiNH/D)TiR

results in KIEs ranging from ∼7-15 at 24.8 ꢀC,⁵⁹and

8538 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

Figure 7. Calculated energies (y-axis, kcal/mol) of 1,2-H₂-elimination from (silox)₃HTaEHPh (2-EHPh; E = N, P, As) to give (silox)₃TadEPh (1dEPh,

E = N, P, As) þ H₂via transition states [2-E-1]^q(E = N, P, As). The reaction coordinate (x-axis) is described in Table 5 (RC(x)).

4. Calculations on 1,2-H₂-Elimination from (silox)₃-

HTaEHR. Figure 7 reveals the geometries and relative

energies of (silox)₃HTaEHPh (2-EHPh; E = N, P, As),

the products (silox)₃TadEPh (1dEPh, E = N, P, As) þ

H₂, and intervening transition states [2-E-1]^q(E = N, P,

length normalized to 1.00 with the pnictide-hydrides

2-EHPh at 0.00. Somewhat surprisingly, the RC for all of

the pnictogens is similar, where the TS is about 40% of the

way to the product pnictidene (N, 0.39; P, 0.40; As, 0.38),

despite a significant disparity in the sum of bond distance

As). Remarkably, the calculations (24.8 ꢀC: ΔG^q(P)_cald

=

changes: N, 6.12 A; P, 6.92 A; As, 7.05 A. One way to

rationalize these features is to accept the premise that the

TadN bond is likely to have a substantially greater force

constant than the remaining TadP and TadAs bonds,

thus affording a “steeper sided” free energy surface

(Figure 7) for the product 1dNPh relative to 1dPPh

and 1dAsPh (likely to be also true for 2-EHPh). A steeper

free energy surface would counteract the expectation ofan

earlier transition state for N because of the more favorable

ΔGꢀ, while allowing the overall reaction coordinate to be

more compressed. This can be observed in the changes in

the pnictide and pnictidene distances. From 2-EHPh to

˚

26.7 kcal/mol, ΔG^q(N)_cald= 25.6 kcal/mol, ΔG^q(As)_cald

24.5 kcal/mol) parallel the general experimental trend

although the activation free energies are somewhat dif-

ferent (24.8 ꢀC: ΔG^q(P) = 25.0 kcal/mol, ΔG^q(N) = 23.5

kcal/mol, ΔG^q(As) = 19.9 kcal/mol). One straightfor-

ward explanation for the lack of a periodic trend concerns

the thermodynamics for 1,2-H₂-elimination, which is

calculated to be 20.4 kcal/mol exoergic for N, but only

4.0 and 5.4 kcal/mol favorable for P and As, respectively.

The greater favorable free energy for elimination of H₂

from the amide-hydride suggests a lower transition state

than the P and As cases that have only a modest driving

force.

The argument that the more favorable free energy

change for N renders a lower transition state also suggests

that [2-N-1]^qshould occur earlier in the reaction coordi-

nate. Table 5 provides a view of the reaction coordinate

parametrized solely via bond distance changes: (1) as a

change in total bond lengths starting from (silox)₃HTa-

EHPh (2-EHPh; E = N, P, As; x = 0), given as RC(x);

(2) as a fractional change in a reaction coordinate in bond

q

˚

[2-E-1] , the Ta-E distances all change by -0.16 A, but

this is a far greater percentage change for the amide (7.8%

relative to ∼6% for P and As), while the percent change in

d(TaE) from [2-E-1]^qto 1dEPh are all roughly 5% for all

three cases. There is a corresponding greater percent

change in the d(EH) for N > P > As that complements

the d(TaE) explanation. Essentially, the greater bond

strengths in the N species, especially that of the imido

product, render the transition state later than it would

be based solely on the standard free energy change for

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8539

Table 5. Calculated Reaction Coordinate^aChanges from 2-EHPh to [2-E-1]^qto

1dEPh

provides a potent example of the entitled dinitrogen

activation, the forward reaction of N₂þ “(silox)₂ClTa”

cannot be examined because the bis-siloxchloride tanta-

lum monomer is not a stable entity; only (silox)₃Ta (1) has

been shown to possess the necessary sterics coupled to a

unique electronic, 3-coordinate environment to exist in

monomer form.⁴As a consequence, high level quantum

calculations were performed on the hypothetical dinitro-

gen activation illustrated in Scheme 1.

P

d(TaH) d(TaE) d(EH) d(HH)

d

RC(x)^bRC^c

2-NHPh

Δ

1.78

0.20

1.98

2.06

-0.16

1.90

1.02

0.24

1.26

3.81

8.67

0.00

2.36

0.00

0.39

-2.47 -2.36

1.17 6.31

[2-N-1]^q

Δ

-0.09

1.81

2.64

-0.16

2.48

-0.43 -3.76

1dNPh

2-PHPh

Δ

0.74

4.18

2.55

10.02

6.12

0.00

1.00

0.00

1.76

0.14

1.90

1.44

0.25

1.69

-2.97 -2.74

1.21 7.28

Figure 8 illustrates some of the complexity pertaining

to dinitrogen activation by (silox)₃Ta (1, Ta), first in

[2-P-1]^q

Δ

2.74

0.40

n

-0.12

2.36

2.76

-0.16

2.60

-0.47 -4.18

n

1dPPh

2-AsHPh

Δ

0.74

4.18

3.10

10.23

6.92

0.00

1.00

0.00

binding N₂to form (silox)₃TaN₂(1-N₂, Ta-N₂), and

1.75

0.13

1.88

1.54

0.21

1.75

second to form the diimide (silox)₃TadN-NdTa(silox)₃

-2.89 -2.71

1.29 7.52

-0.55 -4.34

0.74 3.18

n

(1₂-N₂, Ta-NN-ⁿTa) within the constraints of a linear

[2-As-1]^q

Δ

1dAsPh

2.71

7.05

0.38

1.00

reaction coordinate. 3-Coordinate 1 is a singlet (¹Ta,

-0.13

2.47

2

˚

(d_z2) ) with d(TaO) of 1.89 A and O-Ta-O angles of

120(1)ꢀ that is ∼19 kcal/mol below the triplet of nearest

^aThe reaction coordinate (RC) is roughly considered as the sum of the

interatomic distances involved in the elimination; the product RC is simply

defined as d(TaE) þ d(HH). ^bThe positional coordinate x indicates the

total distance changes along the reaction coordinate. ^cThe RC refers to the

fraction of reaction progress relative to the pnictide-hydrides at 0.00.

energy, Ta, whose electronic configuration is (d_z2)¹(d_xz

3

or d_yz)¹, that is, E⁰⁰in D_3h. Borrowing from previous

3

calculations,⁸⁴the barrier to convert from ¹Ta to ³Ta is a

few kcal/mol higher. Binding N₂requires intersystem

crossing to a triplet surface, as 1-N₂exists as a triplet

that is 7.2 kcal/mol above the separated reagents, and

1,2-H₂-elimination. Since the torsional motions of the

atoms in the 1,2-H₂-elimination events are likely to parallel

the bond strengths from the standpoint of energies (i.e.,

easierto bend X-E-H bonds asAs(easier) < P < N), this

parametrization of the RC is reasonable.

3

∼5.5 kcal/mol below its respective singlet. Triplet Ta-

˚

N₂possesses a d(TaN) of 1.96 A, an elongated d(NN) of

˚

1.18 A (relative to the calculated 1.108 A for free N₂), and

a slightly pyramidalized core suggesting only modest

reorganization energy. The electronic configuration of

Dinitrogen Binding. 1. Collateral Discoveries. While

the indirect synthesis of the hypothetical dinitrogen com-

plex, [(silox)₃Ta]₂(μ-N₂) (1₂-N₂), which is likely to be a

diimide,^7-9has not been realized, several discoveries

during the course of these investigations are noteworthy.

The synthesis of (silox)₃TadCdPPh₃(1dCPPh₃), for

example, may permit examination of the reactivity of an

incipient “(silox)₃TaC” (1-C) via phosphine dissociation.

Preliminary indications are that carbonylation yields

the known ketenylidene (silox)₃TadCdCdO (1dCCO),

suggesting that loss of PPh₃can occur under the right

circumstances. The possibility of Wittig-like reactivity of

the phosphaalkylidyne fragment is another plausible path

of discovery; studies are ongoing.

d_xzd_yzfor Ta-N₂(³A₂in C_3v) correlates with neither

1

3

the ground state (GS) nor the first excited state (ES) of 1,

3

and suggests that a higher lying triplet ( A₂, red dashed

00

line) must somehow be accessed for binding. In summary,

all transition states for ⁿTa þ N₂f ⁿTa-N₂require

intersystem crossing events that are forbidden by orbital

symmetry. Some also require spin crossovers, but pre-

vious calculations suggest that these are inconsequential,

especially within second and third row transition metal

systems.^83-88

The second step, binding of (silox)₃TaN₂(1-N₂, ³Ta-

N₂) to another (silox)₃Ta (1, ⁿTa) can be estimated to be

roughly the same energy as the initial binding event, but

the surface is now a quintet, since both tantalum centers

are triplets. One possible way for (silox)₃Ta (1) to bind

1-N₂is to intersystem cross to the same triplet that may be

necessary to bind N₂in the first step (i.e., the dashed blue

surface), incurring a large barrier in the process. Once the

binding has occurred to give (silox)₃Ta-N₂-Ta(silox)₃

(1₂-N₂, ⁿTa-NN-ⁿTa), electronic reorganization must

occur to reach the singlet surface of the GS product. Since

In the history of silox-based chemistry from these

laboratories, the loss of silanol, other than from degra-

dative hydrolysis, had not been cleanly observed until the

thermolysis of (silox)₃MeTaNHNH2 (4-NHNH₂) led

to [(silox)₂TaMe](μ-N_RHN_β)(μ-N_γHN_δH)[Ta(silox)₂] (5).

Unfortunately, the association of (silox)H with this com-

plex, presumably via hydrogen bonding as solutions were

concentrated, hampered purification of the unusual cyclic

ditantalum species. Nonetheless, the loss of (silox)H via

amination will be kept in mind as a future synthetic tool.

2. Compromise Diimide, [(silox)₂ClTa]₂(μ-N₂). The

structure and stability of [(silox)₂ClTa]₂(μ-N₂) (8-Cl) leave

little question that the hypothetical [(silox)₃Ta]₂(μ-N₂) (1₂-

N₂) complex should be stable. Even though the crystal-

lographic model precluded a true determination of d(TaN)

˚

the diimide product (d(TaN) = 1.84 A; d(NN) = 1.29 A;

—O-Ta-O = 109(1)ꢀ) is ∼55 kcal/mol lower than 2 1 þ

N₂, orbital symmetry constraints may be more readily

(83) Veige, A. S.; Slaughter, L. M.; Lobkovsky, E. B.; Wolczanski, P. T.;

Matsunaga, N.; Decker, S. A.; Cundari, T. R. Inorg. Chem. 2003, 42, 6204–

6224.

(84) Carreon-Macedo, J. L.; Harvey, J. N. J. Am. Chem. Soc. 2004, 126,

5789–5797.

(85) (a) Poli, R. J. Organomet. Chem. 2004, 689, 4291–4304. (b) Poli, R.

Acc. Chem. Res. 1997, 30, 1861–1866.

(86) (a) Harvey, J. N.; Poli, R.; Smith, K. M. Coord. Chem. Rev. 2003, 238,

347–361. (b) Poli, R.; Harvey, J. N. Chem. Soc. Rev. 2003, 32, 1–8.

(87) Harvey, J. N. Struct. Bonding 2004, 112, 151–183.

(88) Matsunaga, N.; Koseki, S. Rev. Comput. Chem. 2004, 20, 101–152.

˚

and d(NN), the ditantalum interatomic distance of 4.874 A

is more consistent with two tantalum-imido interactions

and an elongated dinitrogen distance than a bridging

dinitrogen complex. Again, its stability with regard to

dinitrogen loss supports the notion of the title quest.

3. Calculations on Dinitrogen Binding and Activation

by (silox)₃Ta. Although [(silox)₂ClTa]₂(μ-N₂) (8-Cl)

8540 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

Figure 8. Plausible linear reaction coordinate (RC) versus standard free energy diagram for 2 (silox)₃Ta (1) þ N₂((silox)₃Ta = ⁿTa) to hypothetical

(silox)₃TaN₂Ta(silox)₃(1₂-N₂, ⁿTa-NN-ⁿTa) via (silox)₃TaN₂(1-N₂, ⁿTa-N₂). Calculated standard free energies are in kcal/mol and the RC is based on the

calculated d(TaN) and d(NN) changes, with all surface representations parabolically equivalent. The energies of species pertaining to dashed surfaces have

been estimated.

Figure 9. Simple illustration of orbital symmetry constraints in the conversion of 2 (silox)₃Ta (1) þ N₂((silox)₃Ta = ⁿTa) to hypothetical

(silox)₃TaN₂Ta(silox)₃(1₂-N₂, ⁿTa-NN-ⁿTa) via (silox)₃TaN₂(1-N₂, ⁿTa-N₂).

overcome, but the reduction of dinitrogen to diimide is also

orbital symmetry forbidden from 1-N₂.

additional evidence, but one reaction coordinate pertinent

to the reorganization of (silox)₃TaN₂Ta(silox)₃(1₂-N₂, ⁿTa-

NN-ⁿTa) has already been vetted.

It is clear from Figure 8 that the constraints of orbital

symmetry must be overcome in both the initial and the

second binding events of N₂. One way to overcome the

electronic constraints depicted, that is, lower the inter-

system crossing barriers, is to consider non-linear paths

that effectively “mix” electron configurations that are

intrinsically disparate, namely, those of σ-character

with those of π-character. These paths are too varied to

individually address without extensive calculations or

The cleavage of N₂by (ArRN)₃Mo to GS singlet nitride

products, 2 (ArRN)₃MoN, is orbital symmetry forbi-

dden.^19-23In this case, the binding of N₂to form

[(ArRN)₃Mo]₂(μ-N₂) is aided by the S = 3/2 states of

the Mo(III) precursor, which require little promotional

energy (to Mo(III) S = 1/2 states) to achieve the S = 1

state of the dinitrogen intermediate. Compared to the

tantalum case, binding is readily accomplished, and it is

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8541

the cleavage of N₂that is orbital symmetry forbidden.

Scission eventually occurs because the parentage of

only one orbital pair needs to be converted from π- to

σ-character, and there is a considerably favorable free

energy to aid in the process. In the tantalum case, each

step (Figure 9) carries a significant energy penalty to

overcome orbital symmetry constraints, either in chan-

ging electronic surfaces or in the significant reorganiza-

tion energy required in lowering the symmetry to mix

σ- and π-character.

(m), 2275 (m), 1775 (br, s), 1450 (m), 1375 (m), 1065 (w), 1010

(w), 1000 (w), 970 (w), 930 (w), 880 (br, s), 800 (br, s), 625 (s). M_r

found: 858(68); calcd: 861. Combustion analysis was precluded

by thermal instability.

3. (^tBu₃SiO)₃HTaPHPh (2-PHPh). A 25 mL flask was

charged with (silox)₃Ta (1, 0.682 g, 0.824 mmol), and 10 mL

of pentane was added at -78 ꢀC. Phenyl phosphine (90.7 μL,

0.825 mmol) was transferred into a separate flask, and 10 mL

pentane was added at -78 ꢀC. The solutions were warmed and

stirred to ensure homogeneity. The solution of 1 was cooled to

-78 ꢀC, and the PhPH₂solution was added via syringe in two

portions. This mixture was stirred for 5 min and then warmed to

23 ꢀC as a color change from blue to yellow-orange was observed

(∼10 min); stirring was continued for another 10 min. The

volatiles were removed, and pentane (10 mL) was added. The

solution was filtered, concentrated to 5 mL, and cooled to

-78 ꢀC to afford bright yellow crystals (0.298 g, 39%). IR

(nujol, cm^-1) 2320 (m), 1790 (br, s), 1580 (m), 1470 (s), 1375 (m),

1005 (w), 960-800 (br, s), 725 (m), 690 (m), 620 (s). Combustion

analysis was precluded because of thermal instability.

4. (^tBu₃SiO)₃TadNH (2dNH). A 100 mL flask was charged

with 1 (0.979 g, 1.18 mmol), attached to a frit assembly, and

evacuated. Hexane (50 mL) was added via vacuum transfer.

2-Methylaziridine (360 Torr in 91 mL, 1.79 mmol, 1.5 equiv) was

added to the stirred solution at -78 ꢀC. As the solution was

warmed to room temperature, the blue color of the solution

quickly faded to pale yellow. After stirring for 30 min the

volatiles were removed and fresh hexane was added. The solu-

tion was filtered and concentrated to 10 mL. White crystals

formed upon cooling to -78 ꢀC and were isolated by filtration

(0.771 g). A second crop was obtained in a like manner (0.080 g,

85% total). IR (nujol, cm^-1) 3470 (m), 1375 (m), 1010 (w),

935 (w), 880 (br, s), 820 (s), 630 (s). Anal. Calcd for C₃₆H₈₂O₃N-

Si₃Ta: C, 51.34; H, 9.81; N, 1.66. Found: C, 51.38; H, 10.09; N,

1.54.

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-d₆and toluene-d₈were dried over sodium,

˚

activated 4 A molecular sieves, vacuum transferred, and stored

under nitrogen. THF-d₈was dried over sodium and vacuum

transferred from sodium benzophenone ketyl prior to use. H₂

˚

and D₂were passed over a column containing activated 4 A

sieves and copper oxide. Primary amines, TMSCl, 1,4-cyclohexa-

diene, and phenylarsine oxide were purchased from Aldrich and

˚

dried over activated 4 A sieves. NH₃(Matheson) was dried over

sodium and degassed prior to use. PH₃(Matheson) was used as

received. AsH₃(Caution! Danger, very toxic!) was generated in

small quantities from dilute mineral acid digestion of Zn₃As₂

(Cerac). Phenyl phosphine was purchased from Aldrich and

used without further purification. (^tBu₃SiO)₃Ta (1),²

(^tBu₃SiO)₃TaH₂(2-H),³³and phenyl arsine⁸⁹were prepared

following published procedures.

¹H, ¹³C{¹H}, and ³¹P NMR spectra were obtained on Varian

XL-200 and XL-400, and Inova 400, 500 and 600 MHz spectro-

meters and chemical shifts are reported relative to benzene-d₆

(¹H, δ 7.15; ¹³C{¹H}, δ 128.00) or external H₃PO₄(³¹P, δ 0.00).

UV-vis spectra were acquired on a Hitachi U-2000 spectro-

meter, and IR spectra were recorded on a Mattson FT-IR,

Perkin-Elmer 299B grating IR, or PE 377 grating IR. Combus-

tion analyses were performed by Oneida Research Services

(Whitesboro, NY), Robertson Microlit Laboratories (Madison,

NJ), or Texas Analytical (Houston, TX). Molecular weight deter-

minations via cryoscopy in benzene or vapor phase osmometry

were performed on home-built instruments.

Procedures. 1. (^tBu₃SiO)₃HTaNH₂(2-NH₂). To a 10 mL

flask containing 360 mg (silox)₃Ta (1, 0.435 mmol) was distilled

8 mL of hexanes at -78 ꢀC. The flask was opened to a calibrated

gas bulb containing NH₃(0.456 mmol, 1.05 equiv) and allowed

to warm to 23 ꢀC with stirring. The initial blue color discharged

completely after 30 min. Reduction of the volume to 3 mL,

cooling to -78 ꢀC, and filtration led to the isolation of 217 mg of

white crystals (59%). Anal. Calcd. for C₃₆H₈₄O₃NSi₃Ta: C,

51.22; H, 10.03; N, 1.66. Found: C, 51.31; H, 10.04; N, 1.41.

2. (^tBu₃SiO)₃HTaPH₂(2-PH₂). To a 100 mL flask contain-

ing (silox)₃Ta (1, 0.341 g, 0.412 mmol) was distilled 50 mL of

hexane at 77 K. PH₃was condensed into the flask from a

calibrated bulb (415 Torr in 91 mL, 2.06 mmol, 5 equiv), and

the solution was allowed to warm slowly to 23 ꢀC. The initial

blue color faded to green, then yellow over a period of 30 min,

and a red solid formed. After another 30 min, the excess PH₃was

removed, and the volume of the solution reduced to 10 mL. The

solution was filtered, concentrated, and cooled to -78 ꢀC, to

yield pale yellow crystals (0.192 g, 54%). IR (nujol, cm^-1) 2285

5. (^tBu₃SiO)₃TadNCH₃(1dNCH₃). Into a glass bomb re-

actor was transferred a solution of (silox)₃Ta (1, 0.438 g, 0.529

mmol) in hexane (10 mL). The solution was freeze/pump/thaw

degassed three times, and MeNH₂was condensed in from a

calibrated gas bulb (77 Torr in 125 mL, 0.53 mmol). As the

bomb warmed to 23 ꢀC, the blue color faded to pale yellow.

After stirring for 12 h, the solution was frozen and a copious

quantity of gas was evacuated from the bomb. The solution was

again warmed and stirred. This cycle was repeated intermit-

tently over a period of 3 d until no more gas was evolved. The

resulting solution was concentrated to 4 mL, cooled to -78 ꢀC,

and filtered to yield a white powder (0.228 g, 50%). IR (nujol,

cm^-1) 1475 (s), 1375 (m), 1320 (m), 1010 (w), 975 (m), 935 (w),

875 (br, s), 820 (s), 625 (s). Anal. Calcd for C₃₇H₈₄O₃NSi₃Ta: C,

51.90; H, 9.89; N, 1.64. Found: C, 51.79; H, 9.79; N, 1.45.

6. (^tBu₃SiO)₃TadNPh (1dNPh). To a glass bomb reactor

was added a solution of (silox)₃Ta (1, 0.486 g, 0.587 mmol)

dissolved in 6 mL of benzene. Aniline (53.5 μL, 0.587 mmol) was

dissolved in benzene (2 mL) and added to the bomb. This

mixture was freeze/pump/thaw degassed three times and stirred

at 23 ꢀC for 6 h, during which time a generous quantity of gas

evolved. The solution was degassed and warmed to 23 ꢀC with

continued stirring, and the process was repeated after an addi-

tional 12 h. The solution was then heated to 100 ꢀC for 30 min

and cooled to 23 ꢀC. After the volatiles were removed, 15 mL of

hexane was added, and the solution was degassed, filtered, and

the volatiles removed. Ether (5 mL) was added, and the solution

was cooled to -78 ꢀC to afford white crystals (0.346 g, 64%). IR

(nujol, cm^-1) 1590 (m), 1475 (s), 1385 (m), 1360 (s), 1010 (w), 985

(w), 950 (s), 875 (br, s), 820 (s), 750 (s), 685 (m), 620 (s). Anal.

Calcd for C₄₂H₈₆O₃NSi₃Ta: C, 54.93; H, 9.44; N, 1.53. Found:

C, 54.92; H, 9.45; N, 1.48.

(89) (a) Doak, G. O.; Freedman, L. D. Organometallic Compounds of

Arsenic, Antimony, and Bismuth; Wiley-Interscience: New York, 1970.

(b) Wiberg, K. B. Z. Naturforsch. 1957, 12B, 127–128.

7. (^tBu₃SiO)₃TadN(C₆H₄-p-CF₃) (1dNC₆H₄-p-CF₃). A sam-

ple of (silox)₃Ta (1, 0.208 g, 0.251 mmol) was dissolved in 5 mL

8542 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

of hexane and placed in a glass bomb reactor. Dihydrogen

(∼600 Torr at 77 K) was admitted to the reactor, and the

solution was stirred at 23 ꢀC for 20 h to ensure formation of

(silox)₃TaH₂(2-H), and degassed. In a glovebox, 4-aminoben-

zotrifluoride (31.6 μL, 0.252 mmol) was syringed into the hexane

solution, which was degassed. After thermolysis in a wax bath at

90 ꢀC for 10 h, the contents were transferred and the volatiles

removed. Ether was added, and the solution was filtered.

Concentration to 2 mL and cooling to -78 ꢀC afforded white

crystals (0.097 g, 39%). IR (Nujol, cm^-1) 1605 (m), 1510 (m),

1475 (s), 1375 (s), 1320 (s), 1200 (w), 1170 (m), 1160 (s), 1120 (s),

1110 (m), 1075 (m), 1010 (w), 960 (s), 880 (s, br), 840 (m), 820 (s),

645 (w), 635 (s). Anal. Calcd for C₄₃H₈₅O₃F₃NSi₃Ta: C, 51.14;

H, 8.69; N, 1.42. Found: C, 52.62; H, 8.86; N, 1.31.

8. (^tBu₃SiO)₃TadNLi (1dNLi). Glassware for this reaction

was silylated in the following manner. After drying in an oven

and cooling under vacuum, hexamethyldisilazane was intro-

duced into a frit assembly against N₂counterflow. The liquid

was freeze/pump/thaw degassed three times and then warmed to

reflux with a heat gun until it completely covered the inner

surface of the glassware. The liquid was cooled and removed in

vacuo, and the glassware transferred into a glovebox. One flask

was charged with (silox)₃TadNH (1dNH, 0.473 g, 0.562 mmol)

and neopentyllithium (0.066 g, 0.845 mmol, 1.5 equiv). THF

(25 mL) was transferred via vacuum transfer at -78 ꢀC. The

resulting solution went from yellow to colorless as it warmed to

23 ꢀC, and after stirring for 1.5 h, the solution was filtered,

concentrated to 5 mL, and cooled to -78 ꢀC to afford colorless

crystals (0.371 g, 78%). IR (nujol, cm^-1) 1460 (s), 1375 (s), 1010

(w), 900 (br, s), 820 (m), 625 (m). Anal. Calcd for C₃₆H₈₁O₃N-

LiSi₃Ta: C, 50.98; H, 9.63; N, 1.65. Found: C, 50.92; H, 9.84;

N, 1.41.

9. (^tBu₃SiO)₃TadNSiMe₃(1dNTMS). A flask attached to a

calibrated gas bulb was silylated as described above. The flask

was charged with (silox)₃TadNLi (1dNLi, 0.129 g, 0.152

mmol). THF (20 mL) was added at -78 ꢀC followed by TMSCl

(30 Torr in 91 mL, 0.15 mmol). Upon warming to 23 ꢀC while

stirring for 3 h, the volatiles were removed, and the solid was

extracted into hexane. The solution was filtered, concentrated to

3 mL, and cooled to -78 ꢀC, yielding white crystals (0.070 g,

50%). IR (nujol, cm^-1) 1475 (s), 1375 (m), 1245 (m), 1165 (s),

1010 (w), 950 (s), 870 (br, s), 820 (s), 745 (m), 630 (w), 625 (s).

Anal. Calcd for C₄₀H₉₀O₃NSi₄Ta: C, 51.23; H, 9.92; N, 1.53.

Found: C, 51.56; H, 9.94; N, 1.56.

10. (^tBu₃SiO)₃TadPH (1dPH). A solution of 1 (0.476 g,

0.0.575 mmol) in hexane (50 mL) was exposed to 5 equiv of PH₃,

which was admitted via a calibrated gas bulb. After stirring for

1 h, the solution was filtered and transferred to a glass bomb

reactor, where it was freeze/pump/thaw degassed three times,

warmed, and placed in an oil bath at 70 ꢀC. After 2 h the solution

was again frozen at 77 K and evacuated. A large quantity of gas

was observed. The process was repeated until no more gas was

evolved (two more cycles). The volatiles were removed, and

10 mL of hexane added. The solution was filtered, concentrated

to 5 mL, and cooled to -78 ꢀC, affording pale orange micro-

crystals (0.301 g, 61%). IR (nujol, cm^-1) 2150 (w), 1450 (m),

1375 (m), 1000 (w), 960 (m), 930 (w), 865 (s), 815 (m), 625 (m).

M_rfound: 857(50); calcd: 859. Anal. Calcd for C₃₆H₈₂O₃PSi₃Ta:

C, 50.32; H, 9.62. Found: C, 50.32; H, 10.05.

remaining Ta). IR (nujol, cm^-1) 1580 (w), 1475 (m), 1375 (m),

1010 (w), 950 (m), 850 (s), 820 (m), 730 (w), 695 (w), 625 (m).

Anal. Calcd for C₄₂H₈₆O₃PSi₃Ta: C, 53.93; H, 9.27. Found: C,

53.50; H, 9.44.

12. (^tBu₃SiO)₃TadAsPh (1dAsPh). To a flask charged with

1 (0.615 g, 0.743 mmol) was distilled 10 mL of toluene at -78 ꢀC.

A solution of phenyl arsine (84.5 μL, 0.744 mmol) in toluene

(5 mL) was added via syringe. The blue color of the solution

faded noticeably after 15 min and changed to yellow-orange

after 1 h. Stirring was continued for an additional hour, and

upon warming, the solution effervesced and turned red, then

green. After the reaction subsided, the volatiles were removed,

and the residue was triturated three times with 5 mL of hexane.

Hexane (10 mL) was added, and the solution was filtered. The

residual was washed, and the filtrate was concentrated to 6 mL,

and cooled to afford green microcrystals (0.511 g). A second

crop was also isolated (0.109 g, 85% total). IR (nujol, cm^-1

)

1576 (w), 1470 (m), 1375 (m), 1010 (w), 945 (s), 850 (br, s), 820

(s), 725 (w), 690 (w), 620 (m). M_rfound: 980(45); calcd: 979.

Anal. Calcd for C₄₂H₈₆O₃Si₃AsTa: C, 51.51; H, 8.85. Found: C,

51.17; H, 8.56.

13. (^tBu₃SiO)₃TadAsH (1dAsH). Conducted on a Schlenk

Line in a Hood. In a separate 2-neck flask attached to a needle

valve and fitted with a bent tube containing solid Zn₃As₂(34.4

mg, 0.0968 mmol, 1.2 equiv As) was frozen dilute sulfuric acid

which was carefully degassed. The Zn₃As₂was tapped into the

acid at 0 ꢀC to generate small quantities of AsH₃(Caution!

Danger, Toxic). The needle valve was opened and the AsH₃was

allowed to expand through a cold trap containing KOH and

admitted to another flask attached to a frit assembly containing

(silox)₃Ta (1, 0.200 g, 0.242 mmol) in hexane at -78 ꢀC. The

solution became orange immediately and the evolution of a gas,

presumably H₂was noted, along with the formation of a red

precipitate. The red solid was removed by filtration and dark

orange crystals were obtained from the solution. ¹H NMR

analysis revealed the material to be composed of (silox)₃-

TadAsH (1dAsH) and (silox)₃Ta(AsH₂)₂(3-(AsH₂)₂) in a 9:1

ratio. Any remaining AsH₃was exposed to a cold NaOCl

solution to oxidize the gas, and the Schlenk line was thoroughly

evacuated through an LN₂trap prior to further use.

14. (^tBu₃SiO)₃TadCH₂(1dCH₂). To a small bomb reactor

containing 1 (2.01 g, 2.40 mmol), 690 mg of PPh₃CH₂was added

40 mL of THF (40 mL). The reaction mixture was stirred for 1 h

at 23 ꢀC, degassed, and transferred for workup. The solution

was filtered, concentrated to 15 mL, and cooled to -78 ꢀC to

afford 1.43 g (∼70%) of yellow crystals that contained ∼6%

PPh₃by mass. This was often sufficient for further reactions.

Recrystallization from THF, then Et₂O, yielded pure yellow

crystals (29%) that were submitted for EA. Anal. Calcd for

H₈₃C₃₇O₃Si₃Ta: C, 52.83; H, 9.94. Found: C, 52.91; H, 9.84.

15. (^tBu₃SiO)₃TadCPPh₃(1dCPPh₃). To a 50 mL flask

charged with 1 (1.96 g, 2.40 mmol) and 687 mg of PPh₃CH₂at

-78 ꢀC was added 40 mL of THF by vacuum transfer. The

solution was allowed to slowly warm to 23 ꢀC over 12 h. The

solution was degassed, filtered, concentrated to 15 mL, cooled to

-78 ꢀC, and filtered to yield 1.15 g 1dCH₂(58%). The filtrate

residue was dissolved in 5 mL of pentane and cooled to -78 ꢀC

to yield 371 mg of an off-white powder identified (NMR) as an

equimolar mixture of PPh₃and 1dCPPh₃. This material was

dissolved in 5 mL of benzene and CH₃I (2 equiv, relative to

PPh₃) was added and stirred overnight. A white precipitate was

separated by filtration, and the solvent was removed in vacuo to

yield 256 mg (10%, based on 1) of 1dCPPh₃as a white powder.

X-ray quality crystals could be grown from cooling a saturated

pentane solution to -40 ꢀC.

11. (^tBu₃SiO)₃TadPPh (1dPPh). The residue from the pre-

paration of 2-HPPh was dissolved in hexane and transferred

into a glass bomb reactor. The contents were freeze/pump/thaw

degassed three times. The bomb was warmed to 23 ꢀC, and the

solution stirred for 18 h. The bomb was degassed and immersed

in a 55 ꢀC bath. It was stirred for 3 d with intermittent degass-

ing until gas evolution ceased and the solution was deep red. The

volatiles were removed and 10 mL of pentane was added. The

solution was transferred, concentrated to 3 mL, and cooled to

-78 ꢀC, yielding red-violet crystals (0.293 g, 62% based on

16. (silox)₃MeTaNHNH2 (4-NHNH₂). To a 10 mL round-

bottom flask charged with 1dCH₂(560 mg, 0.665 mmol) and

8 mL of benzene was added 100 μL of hydrazine (3.1 mmol,

4.7 equiv) via syringe. The solution was stirred for 45 min, and

Article

Inorganic Chemistry, Vol. 49, No. 18, 2010 8543

the solvent was removed in vacuo. The resultant off-white solid

was triturated twice with 5 mL of THF, and 5 mL of pentane was

added via vacuum transfer. The suspension was cold-filtered

(-78 ꢀC), and the filter cake washed with 5 mL of pentane. Upon

removal of the volatiles, the residual was dissolved in 20 mL of

benzene and filtered twice through Celite, each time followed by

a 20 mL benzene wash. The filtrates were combined, and the

solvent was removed in vacuo to yield 360 mg of 4-NHNH₂

(49%) as a white powder. Anal. Calcd for C₃₇H₈₇N₂O₃Si₃Ta: C,

50.89; H, 10.04; N, 3.21. Found C, 50.94; H, 10.28; N, 3.09.

17. (silox)₃MeTaNH(-^cNCHMeCH₂) (4-NH(azir)). A 10 mL

round-bottom flask was charged with 200 mg of 1dCH₂(0.24

mmol) and attached to a 180ꢀ valve adapter. The assembly was

evacuated, and 2 mL of diethyl ether was added via vacuum

transfer. To this solution was added 250 μL of N-amino-2-

methylaziridine (3 mmol) at room temperature under argon

counter-flow. The solution discolored over a period of 5 min-

utes, and the solvent and excess aziridine removed in vacuo, and

the resulting waxy solid was left under dynamic vacuum for

45 min. The solid was recrystallized from pentane (0.5 mL) at

-40 ꢀC allowing for slow evaporation and washed with 0.5 mL

of cold (-40 ꢀC) pentane to yield 70 mg (32%) of a thermally

sensitive waxy solid.

deuterated solvent (∼0.6-0.8 mL). Liquid reagents were added

via microliter syringe. The tube was attached to a needle valve,

and the tube was degassed and sealed on a vacuum line. Any

gaseous reagents were added via calibrated gas bulbs, and sol-

vents were added to solid reagents via vacuum distillation prior to

sealing.

22. (silox)₃HTaAsH₂(2-AsH). Conducted on a Schlenk

Line in a Hood. In a separate 2-neck flask attached to a needle

valve and fitted with a bent tube containing solid Zn₃As₂was

frozen dilute sulfuric acid which was carefully degassed. The

Zn₃As₂was tapped into the acid at 0 ꢀC to generate small

quantities of AsH₃(Caution! Danger, Toxic). The needle valve

was opened, and the AsH₃was allowed to expand through a cold

trap containing KOH and admitted to an NMR tube containing

(silox)₃Ta (1), an equiv of 1,4-cyclohexadiene (added via va-

cuum transfer), and toluene-d₈. After the tube was sealed, any

remaining AsH₃was exposed to a cold NaOCl solution to

oxidize the gas, and the Schlenk line was thoroughly evacuated

through an LN₂trap prior to further use. The NMR tube was

kept at -78 ꢀC for NMR monitoring.

General Kinetics. Solutions of (silox)₃HTaEHR (2-EHR)

were generated in 2.0 mL volumetric flasks by one of three

procedures: from isolated 2-EHR and C₆D₆, dissolving (silox)₃-

Ta (1) in C₆D₆and adding REH₂via syringe, or dissolving

(silox)₃TaH₂(2-H) in C₆D₆and adding REH₂via syringe. An

exception was (silox)₃HTaNHMe (2-HNMe), which required

addition of MeNH₂from gas bulbs to individual tubes of 1 in

C₆D₆. TMS₂O (∼0.5 μL) was added as an internal integration

standard. The samples (∼0.6 mL) were transferred to NMR

tubes sealed to 14/20 ground glass joints and attached to 180ꢀ

needle valves. The tubes were freeze/pump/thaw degassed three

times and sealed with a torch. Samples were kept in a constant

temperature oven at 24.8(4)ꢀC or were thermolyzed in a Tamson

TX9 constant temperature bath (2-NH₂). The rate of disap-

pearance of 2-EHR was typically monitored via disappearance

of the hydride ¹H NMR resonance (see Table 1 for exceptions)

for 5-6 half-lives. Single-transient spectra were used for repro-

ducibility of integration. In all cases where appearance of 1dER

was monitored, the rate was found to be the same as disap-

pearance of 2-EHR. Rates and uncertainties were obtained from

unweighted, non-linear, least-squares fitting of the exponential

form of the rate expression (averaged where simultaneous runs

were obtained).

Solutions of (silox)₃HTaAsPh (2-HAsPh) were generated in

the following manner. PhAsH₂was syringed into volumetric

flasks which were then attached to needle valves. The flasks were

cooled to 77 K, evacuated, and toluene-d₈was added via

vacuum transfer. The resulting solutions were added to NMR

tubes containing 1 via vacuum transfer at 77 K, and 1,4-

cyclohexadiene was added from gas bulbs. The tubes were sealed

with a torch and maintained at 77 K until immediately prior to

each run. The tubes were warmed to -78ꢀ in dry ice/acetone

baths and agitated until the blue color of 1 had bleached. The

samples were transferred to the NMR probe which had been

previously cooled to -78 ꢀC. Initial spectra were obtained to

confirm the complete formation of 14-AsHPh. The temperature

of the probe was then raised to -9.5 ꢀC and single-transient

spectra were recorded at measured time intervals.

18. [(silox)₂TaMe](μ-N_rHN_β)( μ-N_γHN_δH)[Ta(silox)₂] (5).

A small bomb was charged with 95 mg of 4-NHNH₂(0.11

mmol), and 2 mL of benzene was added. The bomb was heated

to 75 ꢀC for 24 h and subsequently evacuated of all volatiles. The

resulting oil was dissolved in pentane, removed from the bomb,

and placed in a 4-dram vial where the solvent was removed in

vacuo to yield a slightly yellow liquid (90 mg, 95%). ¹H NMR

analysis indicated complete conversion. The complex could not

be separated from 2 equiv of ^tBu₃SiOH, and was characterized

by multinuclear NMR spectroscopy.

19. [(silox)₃MeTa](μ-η²-N,N:η¹-C-NHNHCH₂CH₂CH₂)-

[Ta(K-O,C-OSi^tBu₂CMe₂CH₂)(silox)₂] (7). To a resealable

NMR tube was added 61 mg of 4-NH(azir) (0.67 mmol) and

0.7 mL of C₆D₆. The tube was heated for 4 h at 80 ꢀC until the

starting material had disappeared. The resulting solution was

removed from the tube and placed in a 1-dram vial, and a

solution of (silox)₃Ta (58 mg in 1 mL. 0.070 mmol) was added.

From the resulting green solution, yellow needles precipitated

which were isolated by decanting the mother-liquor. The needles

were washed with pentane and dried in vacuo to yield 15 mg of

effectively insoluble material that could be recrystallized from

hot THF (80 ꢀC).

20. [(^tBu₃SiO)₂TaCl]₂(μ-N₂) (8-Cl). To a 50 mL round-bot-

tom flask containing 1.000 g of [TaCl₃(THF)₂]₂N₂(1.12 mmol)

and 1.09 g of sodium silox (4.56 mmol, 4.06 equiv) at -78 ꢀC was

added 25 mL of THF via vacuum transfer. The flask was

allowed to warm to 23 ꢀC over 18 h during which time the color

changed from red-orange to yellow. The solvent was removed in

vacuo, and 10 mL of pentane was added. Filtering the suspen-

sion, concentrating the solution to 5 mL, and cooling to -78 ꢀC

afforded yellow microcrystals which were dried in vacuo. After

collection of a second crop the total yield was 828 mg (56%).

21. [(^tBu₃SiO)₂TaMe]₂(μ-N₂) (8-Me). To a 25 mL flask

charged with 100 mg of (0.075 mmol) [(silox)₂TaCl]₂N₂at

-78 ꢀC was added 10 mL of Et₂O via vacuum transfer. After

the solid dissolved, 0.08 mL of CH₃MgBr (2.0 M in Et₂O, 2.1

equiv) was added via syringe. The solution was allowed to warm

to 23 ꢀC over an 8 h period. After 30 min at 23 ꢀC, a white

precipitate formed, and the solution was filtered. The solution

was filtered, concentrated to 2 mL, and cooled to -78 ꢀC,

affording a light-yellow powder that was collected by filtration

and dried in vacuo (50 mg, 52%).

Single-Crystal X-ray Diffraction Studies. Upon isolation, the

crystals were covered in polyisobutenes and placed under a 173

K N₂stream on the goniometer head of a Siemens P4 SMART

CCD area detector (graphite-monochromated Mo KR radia-

˚

tion, λ = 0.71073 A). The structures were solved by direct

methods (SHELXS). All non-hydrogen atoms were refined

anisotropically unless stated, and hydrogen atoms were treated

as idealized contributions (Riding model).

23. (^tBu₃SiO)₃TadNPh (1dNPh). A colorless block (0.3 ꢀ

0.3 ꢀ 0.4 mm) was obtained from a hot heptane solution, and

was sealed in a 0.5 mm glass capillary. A total of 5077 reflections

NMR Tube Reactions. General Procedures. Solid (^tBu₃SiO)₃-

Ta (1) or (^tBu₃SiO)₃TaH₂(2-H), typically 15-20 mg (0.017-

0.025 mmol) was added to an NMR tube attached to a ground

glass joint. Any solid reagents were also added along with

8544 Inorganic Chemistry, Vol. 49, No. 18, 2010

Hulley et al.

were collected with 4665 being symmetry independent (R_int

=

pentane/hexamethyldisiloxane solution at -78 ꢀC over a

period of 4 h. A total of 58,495 reflections were collected with

12,659 being symmetry independent (R_int= 0.0686), and

8,578 were greater than 2σ(I). A semiempirical absorption

correction from equivalents was applied, and the refinement

utilized w^-1= σ²(F_o²) þ (0.0605p)²þ 73.4015p, where p =

0.0254), and 3371 were greater than 2σ(I). A semiempiri-

cal absorption correction from equivalents was applied, and

the refinement utilized w^-1= σ²(F_o²) þ (0.0549p)²þ 7.4307p,

where p = (F_oþ 2 F_c)/3. The ^tBu groups of one silox group

were disordered and modeled accordingly.

2

(F_oþ 2F_c)/3. The disorder in the ^tBu groups of one silox was

modeled appropriately. The electron density in the region

between the two tantalums did not refine well, and the nitro-

gen atoms were refined isotropically. The distances and angles

between the two tantalums are inadequately determined (see

text).

24. (^tBu₃SiO)₃TadCPPh₃(1dCPPh₃). A colorless block

(0.15 ꢀ 0.20 ꢀ 0.25) was obtained from cooling a saturated

pentane solution to -40 ꢀC. A total of 73,173 reflections were

collected with 17,745 being symmetry independent (R_int

=

0.0766) and 12,757 were greater than 2σ(I). A semiempirical

absorption correction from equivalents was applied, and the

refinement utilized w^-1= σ²(F_o²) þ (0.0260p)²þ 0.0173p,

Calculations. To obtain minima for these complexes, full

geometry optimizations, without any metric or symmetry re-

strictions, were performed using the Gaussian 03⁹⁰package, and

these employed density functional theory (DFT), specifically the

BLYP functional.⁹¹Atoms were described with the Stevens

effective core potentials (ECPs) and attendant valence basis sets

(VBSs).⁹²This scheme, dubbed CEP-31G(d), entails a valence

triplet zeta description for the transition metals, a double-ζ-

plus-polarization basis set for main group elements, and the

-31G basis set for hydrogen. This level of theory was selected on

the basis of previous work.^18,83

2

where p = (F_oþ 2 F_c)/3.

25. [(silox)₃MeTa](μ-η²-N,N:η¹-CNHNHCH₂CH₂CH₂)[Ta-

(k-O,C-OSi^tBu₂CMe₂CH₂)(silox)₂] (7). A colorless plate (0.03

ꢀ 0.10 ꢀ 0.20) was obtained from cooling a hot THF solution.

A total of 57,689 reflections were collected with 17,150 being

symmetry independent (R_int= 0.0762) and 11,852 were greater

than 2σ(I). A semiempirical absorption correction from equiva-

lents was applied, and the refinement utilized w^-1= σ²(F_o²) þ

2

(0.0624p)²þ 15.5864p, where p = (F_oþ 2F_c)/3. Disordered

solvent was SQEEZEd from the cell, and a disorder in the ^tBu

groups of one silox ligand was modeled appropriately.

26. [(^tBu₃SiO)₂TaCl]₂(μ-N₂) (8-Cl). A colorless block (0.15 ꢀ

Acknowledgment. P.T.W. thanks the NSF (CHE-0718030)

and Cornell University, and TRC the DOE (DE-FG02-

03ER15387) for financial support.

0.20 ꢀ 0.25) was obtained from slow evaporation of a 1:1

(90) Frisch, M. J. et al. Gaussian 03; Gaussian, Inc.: Pittsburgh PA, 1998.

(91) Parr, R. G.; Yang, W. Density-functional Theory of Atoms and

Molecules; Oxford University Press: Oxford, 1989.

(92) Stevens, W. J.; Krauss, M.; Basch, H.; Jasien, P. G. Can. J. Chem.

1992, 70, 612–630.

Supporting Information Available: CIF files for 1dNPh,

1dCPPh₃, 7, and 8-Cl. This material is available free of charge

via the Internet at http://pubs.acs.org.

Article Doi

Pnictogen-hydride activation by (silox)3Ta (silox = tBu3SiO); Attempts to circumvent the constraints of orbital symmetry in N2 activation

DOI: 10.1021/ic101147x

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ic101147x

Full text of DOI:10.1021/ic101147x

Products guided by the article

R&D Labs maybe for 1242420-60-5

Relevant to this article

Hot Product