Synthetic and Mechanistic Investigations of Trimethylsilyl

Intermediates on the Road to a Dinuclear Tantalum Tetrahydride Derivative. ..... Complexes as Synthetic Entries to Low-Coordinate Phosphorus Deriv...
0 downloads 0 Views

J. Am. Chem. Soc. 1996, 118, 3643-3655


Synthetic and Mechanistic Investigations of Trimethylsilyl-Substituted Triamidoamine Complexes of Tantalum That Contain Metal-Ligand Multiple Bonds Joel S. Freundlich, Richard R. Schrock,* and William M. Davis Contribution from the Department of Chemistry 6-331, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 ReceiVed NoVember 14, 1995X

Abstract: [N3N]TadPPh ([N3N]3- ) [(Me3SiNCH2CH2)3N]3-) reacts with excess lithium metal in tetrahydrofuran to give “[N3N]TadPLi”, as judged by NMR studies and by reactions with RX at -35 °C to afford the phosphinidene complexes [N3N]TadPR (R ) Me, n-Bu, SiMe3, SiMe2Ph). [N3N]TaCl2 reacts with 2 equiv of LiN(H)R (R ) H, CMe3, Ph) to produce 1 equiv of RNH2 and imido complexes [N3N]TadNR and with 2 equiv of benzylmagnesium chloride or ((trimethylsilyl)methyl)lithium to afford the alkylidene complexes [N3N]TadCHR (R ) Ph or SiMe3). The ethylene complex [N3N]Ta(C2H4) is formed quantitatively upon addition of 2 equiv of ethylmagnesium chloride to [N3N]TaCl2. [N3N]Ta(C2H4) decomposes in a first-order manner in solution over a period of days at room temperature to give a complex in which a C-N bond in the TREN backbone has been cleaved. Alkylation of [N3N]TaCl2 with 2 equiv of RCH2CH2MgX (R ) CH3, CH2CH3, CH(CH3)2, C(CH3)3; X ) Cl or Br) produces a mixture of alkylidene and products derived from decomposition of the incipient olefin complex. When R ) t-Bu, only an alkylidene complex is formed as a consequence of a sterically disfavored β abstraction process. [N3N]TaCl2 reacts with 2 equiv of vinylmagnesium bromide to afford white crystalline [N3N]Ta(C2H2). An analogous benzyne complex can be prepared by refluxing [N3N]TaCl2 with 2 equiv of phenyllithium in toluene. [N3N]Ta(C2H4) reacts with a catalytic amount of phenylphosphine to afford [N3N]TadCHMe, while reactions with ammonia, aniline, or pentafluoroaniline yield [N3N]TadNR complexes. In contrast, excess Me3SiAsH2 reacts with [N3N]Ta(C2H4) to afford [N3N]TadCHMe first, and then what is proposed to be [N3N]TadAsSiMe3. [N3N]Ta(C2H4) reacts with dihydrogen to give [N3N]Ta(H)(C2H5) reversibly. [N3N]Ta(C6H4) reacts with ArNH2 (Ar ) Ph, C6F5) to give [N3N]TadNAr complexes, but [N3N]Ta(C2H2) is relatively unreactive. X-ray structures of [N3N]Ta(Me)Et and [N3N]Ta(C2H2) are included.

Introduction Recent efforts in these laboratories1-10 and others11-19 have focused on the preparation of complexes of metals in groups 4, 5, and 6 that contain tetradentate triamidoamine ligands, [(RNCH2CH2)3N]3-. When R is a bulky silyl group (usually SiMe3), rare types of complexes can be prepared, e.g., a tantalum(V) phosphinidene,3 a VdNH species,6 and molybdeX Abstract published in AdVance ACS Abstracts, April 1, 1996. (1) Cummins, C. C.; Schrock, R. R.; Davis, W. M. Organometallics 1992, 11, 1452. (2) Cummins, C. C.; Lee, J.; Schrock, R. R. Angew. Chem., Int. Ed. Engl. 1992, 31, 1501. (3) Cummins, C. C.; Schrock, R. R.; Davis, W. M. Angew. Chem., Int. Ed. Engl. 1993, 32, 756. (4) Kol, M.; Schrock, R. R.; Kempe, R.; Davis, W. M. J. Am. Chem. Soc. 1994, 116, 4382. (5) Freundlich, J. S.; Schrock, R. R.; Cummins, C. C.; Davis, W. M. J. Am. Chem. Soc. 1994, 116, 6476. (6) Cummins, C. C.; Schrock, R. R.; Davis, W. M. Inorg. Chem. 1994, 33, 1448. (7) Shih, K.-Y.; Schrock, R. R.; Kempe, R. J. Am. Chem. Soc. 1994, 116, 8804. (8) Shih, K.-Y.; Totland, K.; Seidel, S. W.; Schrock, R. R. J. Am. Chem. Soc. 1994, 116, 12103. (9) Schrock, R. R.; Shih, K.-Y.; Dobbs, D.; Davis, W. M. J. Am. Chem. Soc. 1995, 117, 6609. (10) Zanetti, N.; Schrock, R. R.; Davis, W. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 2044. (11) Christou, V.; Arnold, J. Angew. Chem., Int. Ed. Engl. 1993, 32, 1450. (12) Naiini, A. A.; Menge, W. M. P. B.; Verkade, J. G. Inorg. Chem. 1991, 30, 5009. (13) Plass, W.; Verkade, J. G. J. Am. Chem. Soc. 1992, 114, 2275. (14) Naiini, A. A.; Ringrose, S. L.; Su, Y.; Jacobson, R. A.; Verkade, J. G. Inorg. Chem. 1993, 32, 1290.


num and tungsten terminal phosphido complexes.10 Key features of complexes that contain a [(RNCH2CH2)3N]3- ligand include the sterically protected “pocket” formed by the bulky R groups and the presence of one σ-type and two orthogonal π metal orbitals directed toward the remaining (fifth) coordination site. This orbital arrangement is ideally suited for forming d0 complexes that contain a triple bond or pseudo triple bond between the metal and the ligand in the apical coordination site, a double and a single bond, or (sterically least feasibly) three single bonds. (The [(RNCH2CH2)3N]3- ligand itself is only a 12-electron donor (4σ,2π) since one of the three linear combinations of atomic orbitals constructed from the p orbitals on the three amido nitrogens is a ligand-centered nonbonding orbital.) In view of tantalum’s ability to form multiple bonds to C, N, or O,20 and because many starting materials are readily available, we chose to explore the chemistry of [(Me3SiNCH2CH2)3N]Ta(X) complexes that contain a multiple Ta-X bond. We first reported that [N3N]TaCl2 ([N3N]3- ) [(Me3SiNCH2CH2)3N]3-) reacts with 2 equiv of a lithium phosphide LiP(H)R (R ) Ph, Cy, t-Bu) to afford 1 equiv of RPH2 and gold crystalline d0 phosphinidene complexes, [N3N]TadPR, in moderate to high yields.3 An X-ray crystal structure of [N3N]TadPCy revealed (15) Plass, W.; Verkade, J. G. Inorg. Chem. 1993, 32, 3762. (16) Verkade, J. G. Acc. Chem. Res. 1993, 26, 483. (17) Duan, Z.; Verkade, J. G. Inorg. Chem. 1995, 34, 1576. (18) Schubart, M.; O’Dwyer, L.; Gade, L. H.; Li, W.-S.; McPartlin, M. Inorg. Chem. 1994, 33, 3893. (19) Hill, P. L.; Yap, G. A.; Rheingold, A. L.; Maatta, E. A. J. Chem. Soc., Chem. Commun. 1995, 737. (20) Schrock, R. R. In Reactions of Coordinated Ligands; Braterman, P. R., Ed.; Plenum: New York, 1986.

© 1996 American Chemical Society

3644 J. Am. Chem. Soc., Vol. 118, No. 15, 1996 Table 1.


Freundlich et al.

NMR Data and Yields for Phosphinidene Complexes



MeI n-BuBr Me3SiCl PhMe2SiCl

1a 1b 1c 1d


157 186 212b 203c


yield (%) 33 68 58 77

a Determined via 1H NMR with a (Me3Si)2O internal standard. b ∆ν1/2 ≈ 3600 Hz (23 °C), 400 Hz (-30 °C), 300 Hz (-60 °C). c ∆ν1/2 ≈ 3600 Hz (23 °C), 800 Hz (-30 °C), 400 Hz (-60 °C).

an essentially linear Ta-P-C angle and a Ta-P bond length (2.145 Å) consistent with a pseudo triple bond between Ta and P. We subsequently turned to the synthesis and reactions of alkylidene, olefin, and acetylene complexes.5 Full details and additional results in both areas are reported here. Results Synthesis of Tantalum Phosphinidene and Imido Complexes. We became interested in the possibility of preparing the parent phosphinidene complex, [N3N]TadPH, or the terminal phosphido complex, {[N3N]TatP}-. We found that [N3N]TadPPh reacts with excess lithium metal in tetrahydrofuran to give a species whose 31P NMR spectrum reveals a resonance at 575 ppm (∆ν1/2 ≈ 600 Hz). This species reacts at -35 °C with alkyl and silyl halides to yield the phosphinidene complexes [N3N]TadPR (R ) Me, n-Bu, SiMe3, SiMe2Ph), 1a-d (eq 1), according to 31P NMR data. The yields of the 1. 5 equiv of Li, THF

[N3N]TadPPh 9 8 2. 3 equiv of RX, THF, -35 °C [N3N]TadPR + PhR + 2LiX (1) 1a-d R ) Me (1a), n-Bu (1b), SiMe3 (1c), SiMe2Ph (1d) phosphinidene complexes, as judged by proton NMR versus an internal standard, are listed in Table 1. The NMR yields are modest, and isolated yields are poor (10-20%) as a consequence of the extreme solubility of the phosphinidene complexes in common organic solvents. The formulations of 1a-d are confirmed by reactions with pivaldehyde to yield [N3N]TadO and the corresponding trans-phosphaalkenes, which were identified by 1H and 31P NMR, a reaction that is known for several isolated tantalum phosphinidene complexes.3 We have also prepared complex 1b as shown in eq 2, although the isolated yield is again low (10%) as a consequence of the high solubility of 1b in common organic solvents. 2LiP(H)-n-Bu

[N3N]TaCl2 9 8 [N3N]TadP-n-Bu + n-BuPH2 (2) Et2O, -35 °C 1b A puzzling fact is that the phosphorus resonances in compounds 1 are broad, especially those in 1c and 1d. Alkyland arylphosphinidenes exhibit a 31P resonance with a halfheight width of 100-200 Hz at 25 °C, while the analogous resonances in 1c and 1d have widths of 300-400 Hz at -60 °C. We currently attribute the broadened phosphinidene phosphorus resonances to coupling to 181Ta, but why coupling is greater in the silyl-substituted phosphinidene complexes is unclear. Broadened resonances are not found in proton NMR spectra. The intermediate whose 31P NMR spectrum contains a resonance at 575 ppm (∆ν1/2 ≈ 600 Hz) we propose to be “[N3N]TadPLi”, rather than {[N3N]TatP}-. The primary reason is that the chemical shifts of the terminal phosphido

ligands in related neutral d0 complexes, [N3N]MotP,10 [N3N]WtP,10 and [(t-Bu)NAr]3MotP21 (Ar ) 3,5-Me2C6H3), range from 1080 to 1346 ppm. Therefore a chemical shift of 575 ppm, even though it is ∼400 ppm larger than that in a typical TadPR species (∼200 ppm), we believe to be too small to ascribe to a “{[N3N]TatP}-” species, i.e., one in which lithium is not bound to the phosphorus. Since bent phosphinidenes are characterized by a more downfield resonance (335 ppm in (t-Bu3SiO)3TadPPh22 and 600-800 ppm in Cp2MdPR complexes (M ) Mo, W, Zr)23-25), a chemical shift of 575 ppm would be more consistent with a “[N3N]TadPLi” species in which the Ta-P-Li bond angle is less than 180°. Unfortunately, we could find no way to separate the “[N3N]TadPLi” species from phenyllithium, the other product of the cleavage reaction, and so could not isolate and structurally characterize it. All attempts to prepare [N3N]TadPH so far have failed. For example, the reaction of [N3N]TaCl2 with 2 equiv of LiPH2 in 1,2-dimethoxyethane at -78 °C afforded intractable products, while quenching “[N3N]TadPLi” with proton sources such as HNMe3Cl or 2,6-lutidinium triflate led to complex mixtures in which no species could be identified. [N3N]TaCl2 reacts with 2 equiv of LiN(H)R (R ) H, CMe3, Ph) to produce 1 equiv of RNH2 and imido complexes 2a-c in 62-95% yield (eq 3). The synthesis of 2a is noteworthy, as parent imido complexes are relatively rare.6,26-29 NMR and IR spectra of 2a are similar to those for [N3N]VdNH, which has been structurally characterized.6 A notable difference is that the imido proton resonance is observed as a broad 1:1:1 triplet (1J14NH ) 50 Hz) in 2a whereas it is not seen in the 1H NMR spectrum of [N3N]VdNH, presumably as a consequence of additional coupling to 51V (I ) 7/2, 99.75%). Resolved coupling between the imido proton and 14N has also been observed in Cp*MMe3(NH) (M ) Mo, W) complexes27,28 and was attributed to a low electric field gradient about the imido nitrogen.30 The white crystalline imido complexes are stable when heated as ∼0.1 M solutions in toluene-d8 in sealed NMR tubes to 110 °C for several days. They do not react with benzaldehyde in benzene-d6 (∼0.1 M in Ta, 2 days) at ∼25 °C to yield known [N3N]TadO.3 2LiN(H)R

[N3N]TaCl2 9 8 [N3N]TadNR + RNH2 (3) Et2O (R ) CMe3, Ph) 2a-c THF (R ) H) -35 °C

R ) H (2a), CMe3 (2b), Ph (2c) Synthesis and Reactivity of Tantalum Alkylidene Complexes. [N3N]TaCl2 reacts with 2 equiv of ((trimethylsilyl)(21) Laplaza, C. E.; Davis, W. M.; Cummins, C. C. Angew. Chem., Int. Ed. Engl 1995, 34, 2042. (22) Bonanno, J. B.; Wolczanski, P. T.; Lobkovsky, E. B. J. Am. Chem. Soc. 1994, 116, 11159. (23) Ho, J.; Rousseau, R.; Stephan, D. W. Organometallics 1994, 13, 1918. (24) Hou, Z.; Breen, T. L.; Stephan, D. W. Organometallics 1993, 12, 3158. (25) Hitchcock, P. B.; Lappert, M. F.; Leung, W.-P. J. Chem. Soc., Chem. Commun. 1987, 1282. (26) Parkin, G.; van Asselt, A.; Leahy, D. J.; Whinnery, L.; Hua, N. G.; Quan, R. W.; Henling, L. M.; Schaefer, W. P.; Santarsiero, B. D.; Bercaw, J. E. Inorg. Chem. 1992, 31, 82. (27) Schrock, R. R.; Glassman, T. E.; Vale, M. G. J. Am. Chem. Soc. 1991, 113, 725. (28) Glassman, T. E.; Vale, M. G.; Schrock, R. R. Organometallics 1991, 10, 4046. (29) Chatt, J.; Choukroun, R.; Dilworth, J. R.; Hyde, J.; Vella, P.; Zubieta, J. Transition Met. Chem. 1979, 4, 59. (30) Mason, J. In Multinuclear NMR; Mason, J., Ed.; Plenum Press: New York, 1987; Chapter 2.

Triamidoamine Complexes of Tantalum

J. Am. Chem. Soc., Vol. 118, No. 15, 1996 3645

methyl)lithium or benzylmagnesium chloride to afford the alkylidene complexes 3a and 3b (eq 4) in >90% yield. We dialkylation

[N3N]TaCl2 9 8 [N3N]TadCHR + RCH3 Et2O, -35 °C 3a,b


R ) SiMe3 (3a; LiCH2SiMe3), Ph (3b; PhCH2MgCl) propose that dialkyl complexes are intermediates in these reactions for several reasons. First, if only 1 equiv of alkylating agent is employed, [N3N]Ta(CH2Ph)Cl and [N3N]Ta(CH2SiMe3)Cl can be isolated by fractional crystallization, and each is converted into the expected alkylidene upon reaction with an additional 1 equiv of the appropriate metal alkyl. Second, reaction of Li3[N3N] with Ta(CH2R)2Cl3 (R ) Ph,31 CMe332) gives 3b and 3c (eq 5) in yields of 86% and 56%, respectively. Complexes 3a-c show no signs of decomposition in toluened8 (∼0.1 M) after being heated for days in sealed tubes at 110 °C. They react rapidly with aldehydes to afford a mixture of [N3N]TadO and cis- and trans-isomers of the expected olefin. A variety of tantalum alkylidene complexes are known to react readily with aldehydes in this manner.20,33 Li3[N3N]

Ta(CH2R)2Cl3 9 8 [N3N]TadCHR + RCH3 (5) Et2O, -35 °C 3b,c R ) Ph (3b), CMe3 (3c) Complexes 3a-c all have NMR spectra consistent with 3-fold symmetry on the NMR time scale. HR resonances are found near 0 ppm in 1H NMR spectra, a region characteristic of alkylidenes that are highly “distorted” through an R-agostic34 CH interaction,20,35 and 13C NMR spectra show an alkylidene carbon resonance in the range 200-215 ppm. We were surprised by the unusually low values for 1JCHR (∼70 Hz), the lowest known for d0 alkylidene complexes.20 The alkylidene ligands are effectively pseudo triply bound to tantalum as 1σ,2π ligands. In spite of the fact that the π interactions in the apical position are of two distinct types, all alkylidene complexes show 3-fold symmetry on the NMR time scale down to -90 °C. Apparently only steric constraints would lead to a breaking of the dxz/dyz degeneracy and slowing of “rotation” of the alkylidene about the Ta-C bond. Such steric constraints would seem to be minimal, if the TadCRsCβ angle is relatively large. No crystals suitable for X-ray studies have yet been obtained, although an X-ray study of a related Et3SiTREN alkylidene complex of tantalum has recently confirmed that the TadCRsCβ angle is indeed large (>170°)36 and, therefore, that a large TadCRsCβ angle should be expected for any alkylidene complex in this category. Synthesis of a Tantalum Ethylene Complex. r versus β Abstraction Processes. An ethylene complex (4) is formed quantitatively upon addition of 2 equiv of ethylmagnesium chloride to [N3N]TaCl2 (eq 6). An alternate route to 4 consists of alkylation of [N3N]Ta(Me)OTf5 with 1 equiv of ethylmagnesium chloride. Proton and carbon NMR spectra of 4 are consistent with it being a 3-fold-symmetric complex on the (31) Messerle, L. W.; Jennische, P.; Schrock, R. R.; Stucky, G. D. J. Am. Chem. Soc. 1980, 102, 6744. (32) Schrock, R. R.; Fellmann, J. D. J. Am. Chem. Soc. 1978, 100, 3359. (33) Schrock, R. R. J. Am. Chem. Soc. 1976, 98, 5399. (34) Brookhart, M.; Green, M. L. H.; Wong, L. Prog. Inorg. Chem. 1988, 36, 1. (35) Schultz, A. J.; Williams, J. M.; Schrock, R. R.; Rupprecht, G. A.; Fellmann, J. D. J. Am. Chem. Soc. 1979, 101, 1593. (36) Freundlich, J. S.; Schrock, R. R.; Davis, W. M. Submitted for publication.


[N3N]TaCl2 9 8 [N3N]Ta(C2H4) Et2O, -35 °C 4


NMR time scale, even at -90 °C. The ethylene ligand is observed as a singlet at 2.15 ppm in the proton NMR spectrum and a triplet (1JCH ) 144 Hz) at 62.6 ppm in the 13C NMR spectrum. We propose that 4 has a structure similar to that of [N3N]Ta(C2H2) (vide infra) in the solid state, i.e., one in which the C-C axis of the ethylene is lined up with one of the Ta-N bonds, but that the ethylene rotates readily in solution to give a C3-symmetric species on the NMR time scale as a consequence of the degeneracy of the dxz and dyz orbitals. Reaction of [N3N]TaCl2 with only 1 equiv of ethyl Grignard reagent yields the yellow crystalline monoethyl derivative 5 in 72% yield (eq 7). 5 reacts immediately with 1 equiv of ethylmagnesium chloride to yield 4, and with 1 equiv of CD3CD2MgBr to afford a 5.7:1 mixture of [N3N]Ta(C2H4) and [N3N]Ta(C2D4). The labeling experiment suggests that the intramolecular isotope effect for β abstraction in the diethyl intermediate is 5.7. EtMgCl

[N3N]TaCl2 9 8 [N3N]Ta(Et)Cl Et2O, -35 °C 5


Although 4 is formed upon treatment of [N3N]Ta(Me)OTf with 1 equiv of ethylmagnesium chloride over a period of 1 day, [N3N]Ta(Me)Cl reacts with 1 equiv of ethyl Grignard under the same conditions to afford a 2:1 mixture of 4 and [N3N]Ta(Me)Et (6, eq 8). [N3N]Ta(Me)Et can be isolated as a yellow EtMgCl

8 [N3N]Ta(Me)Cl 9 Et O, -35 °C 2

[N3N]Ta(C2H4) + [N3N]Ta(Me)Et (8) 4 6 crystalline solid via fractional recrystallization. It shows 3-fold symmetry on the NMR time scale. Over a period of 1 day at ∼25 °C, 6 decomposes to yield 4 and methane. At 52 °C in toluene-d8, decomposition of 6 was shown to follow first-order kinetics with k ) 2.4(1) × 10-4 s-1 (two runs). [N3N]Ta(Me)Et is also formed in the reaction between [N3N]Ta(Et)Cl and 1 equiv of methylmagnesium chloride. An X-ray structure of 6 (Table 2; Figure 1) shows it to be a six-coordinate species with methyl and ethyl ligands in apical coordination sites that lie approximately in the N(2)-Ta-N(4) plane. (Relevant bond lengths and angles are listed in Table 3.) The smaller methyl group is pointed toward N(2). Consequently, the Ta-N(2)-Si(2) angle (136°) is somewhat larger than the other two Ta-N-Si angles (129° and 132°), but all are larger than the usual values of 125-126 °C in crystallographically characterized M[N3N] species.1-3,6,9,10,19 Since the ethyl ligand points away from the methyl group and between N(1) and N(3), the N(1)-Ta-N(3) angle opens to 133°, compared to 104° and 100° for the other two N-Ta-N angles. The distance between Cβ of the ethyl group and Ta is 3.14 Å, too far for any β agostic interaction; in any case there is no readily available orbital with which the methyl CH bond can interact when the ethyl group is oriented in the observed manner. Therefore we propose that β abstraction first involves rotation of the ethyl group past one SiMe3 group, possibly with concomitant “dissociation” of the amine nitrogen donor from the metal, followed by activation of Hβ through an agostic interaction with the remaining π orbital that lies in a plane approximately 90° to that containing Ta, C(9), and C(7). The

3646 J. Am. Chem. Soc., Vol. 118, No. 15, 1996 Table 2. Crystallographic Data, Collection Parameters, and Refinement Parameters for [N3N]Ta(Me)Et (6) and [N3N]Ta(C2H2) (8) empirical formula formula wt cryst color, habit cryst dimens (mm) cryst syst no. reflns used for unit cell determination (2θ range, deg) a (Å) b (Å) c (Å) β (deg) V (Å3) space group Z Dcalc (g/cm3) F000 µ(Mo KR) (cm-1) scan type temp (°C) total no. of unique reflns no. of observations with I > 3.00σ(I) no. of variables R Rw GOF



C18H47N4Si3Ta 584.80 yellow, plate 0.150 × 0.150 × 0.05 monoclinic 25 (14.0- 22.0)

C17H41N4Si3Ta 566.74 colorless, needle 0.280 × 0.120 × 0.120 orthorhombic 25 (15.0- 25.0)

10.0504(8) 15.010(1) 17.937(1) 95.79(1) 2692.1(6) P21/n 4 1.443 1192 41.75 ω-2θ -86 3679 2270

17.154(1) 16.756(1) 34.365(3) 90.0 9878(2) Pbca 8 1.525 4528 45.47 ω-2θ -72 9483 5137

235 0.041 0.035 1.25

439 0.048 0.053 3.42

Freundlich et al. Table 3. Selected Intramolecular Distancesa and Anglesb for the Non-Hydrogen Atoms of [N3N]Ta(Me)Et and [N3N]Ta(C2H2) Bond Lengths [N3N]Ta(Me)Et Ta-N(1) Ta-N(2) Ta-N(3) Ta-N(4) Ta-C(7) Ta-C(9) C(7)-C(8)


2.00(1) 2.071(9) 1.956(9) 2.444(8) 2.21(1) 2.21(1) 1.55(2)

Ta-N(1) Ta-N(2) Ta-N(3) Ta-N(4) Ta-C(7) Ta-C(8) C(7)-C(8)

2.07(1) 2.02(1) 2.04(1) 2.30(1) 2.09(1) 2.10(1) 1.26(2)

Bond Angles [N3N]Ta(Me)Et Ta-N(1)-Si(1) Ta-N(2)-Si(2) Ta-N(3)-Si(3) N(1)-Ta-N(2) N(1)-Ta-N(3) N(2)-Ta-N(3) Ta-C(7)-C(8) a

131.6(5) 136.1(5) 128.8(5) 99.6(4) 133.1(4) 103.7(4) 112.0(9)

[N3N]Ta(C2H2) Ta-N(1)-Si(1) Ta-N(2)-Si(2) Ta-N(3)-Si(3) N(1)-Ta-N(2) N(1)-Ta-N(3) N(2)-Ta-N(3)

126.1(6) 125.7(6) 126.4(6) 110.5(5) 110.6(4) 122.5(5)

In angstroms. b In degrees.

ethyl ligand are found at 1.93 and 1.46 ppm, respectively. Heating of a toluene-d8 solution of [N3N]Ta(C2D4) at 110 °C in a sealed tube yields a product analogous to 7b that contains a TaCD2CD2H group. At 70 °C kD ) 1.53(2) × 10-4 s-1 for a kH/kD of 0.89(2), consistent with a change in hybridization of the ethylene carbon atoms from sp2 to sp3 in the rate-limiting step.37 Proton and carbon NMR spectral data are similar to those for the product resulting from the thermolysis of [(Et3SiNCH2CH2)3N]Ta(C2H4), whose structure has been determined in an X-ray study.36 Thermolysis of [N3N]Ta(C2D4) (0.03 M

Figure 1. X-ray crystal structure of [N3N]Ta(Me)Et (6).

Ta-N(4) distance (2.444 Å) is comparable to that found in [N3N]TadTe (2.487 Å),11 but is somewhat longer than found in [N3N]Ta(HCtCH) (2.30 Å; see later). [N3N]Ta(Et)Cl reacts with 1 equiv of benzylmagnesium chloride to yield 4. If only 0.5 equiv of Grignard is used, the proton NMR spectrum shows no evidence for [N3N]Ta(CH2Ph)Cl formed by alkyl exchange. [N3N]Ta(CH2Ph)Cl similarly reacts with 1 equiv of ethylmagnesium chloride to afford 4. We propose that [N3N]Ta(Et)(CH2Ph) is the intermediate in each of these reactions. The benzylidene complex (3b) is not formed in either reaction. [N3N]Ta(C2H4) is not stable in solution. After a period of several days at ∼25 °C, solutions of 4 show signs of decomposition; the red color lightens, and NMR spectra consistent with formation of the yellow ethyl complex 7b (eq 9) are observed. Decomposition of a toluene solution of 4 ([4] ) 0.0059, 0.0089, 0.010, 0.012 M) was followed by UV/vis at 494 nm and shown to be first order in tantalum with k ) 1.37(1) × 10-4 s-1 at 70 °C. Most prominent in the 1H NMR spectrum of 7b are the vinyl resonances, a doublet of doublets at 6.49 ppm and a doublet at 4.25 ppm. (The latter is obscured by the 4.07 ppm resonance for the diastereotopic ligand methylene protons.) The triplet and quartet resonances for the

in toluene-d8) in the presence of 1 atm of ethylene produces only a TaCD2CD2H species. All of these data are consistent with decomposition of 4 by intramolecular β abstraction of a proton from the side chain of the amido ligand. The TREN backbone must turn and flex to a considerable degree, possibly after dissociation of the apical nitrogen donor atom, in order to present the C-Hβ bond to the metal for removal of Hβ and transfer to the ethylene ligand. Formation of 7b′ would constitute removal of a γ proton if the apical donor nitrogen were not coordinated at the time. [N3N]TaMe25 decomposes when heated above 60 °C to produce 7a, according to 1H and 13C NMR spectra (eq 10). Thermolysis of [N3N]Ta(CD3)2 produces CD3H and 7a that contains a CD3 ligand, according to 1H and 2H NMR spectra. Therefore we can rule out [N3N]Ta(CD2) as an intermediate. (37) Carpenter, B. K. Organic Reaction Mechanisms; John Wiley & Sons: New York, 1984.

Triamidoamine Complexes of Tantalum

J. Am. Chem. Soc., Vol. 118, No. 15, 1996 3647 Table 4. Percent Yields of Alkylidene and Decomposition Products Resulting from the Reaction between [N3N]TaCl2 and 2RCH2CH2MgX

([N3N]Ta(CH2) is still an unknown compound.) It is interesting to note that in [N3N]TaMe2, as in 4, only two of the three orbitals in the apical position are used for bonding to apical ligands, i.e., one orbital is empty. Other species discussed in this paper in which all three orbitals are involved in bonding to a ligand in the apical position do not decompose upon heating for days at 100 °C. Reactions between [N3N]TaCl2 and 2 equiv of RCH2CH2MgX (X ) Cl or Br) in which R is not a proton do not yield olefin complexes analogous to 4, but alkylidene complexes in yields that correlate with the bulk of the R group, and products whose NMR spectra are analogous to those of 7a and 7b. We propose that the latter products arise via facile decomposition of intermediate olefin complexes. For example, n-propylmagnesium chloride affords a mixture of a propylidene complex (3e) in 32% yield and a decomposition product (7c) in 66% yield, as determined by 1H NMR versus an internal standard (eq 11). The spectra of 3e are analogous to those of other alkylidenes described here; in this case HR is a triplet at -0.28 ppm in the proton NMR spectrum. We propose that 3e and 7c arise via competitive R and β abstraction, respectively, in a dipropyl intermediate.


% alkylidene

% olefin or dec prod.

H CH3 CH2Me CHMe2 CMe3

0 32 42 84 (76b) 83 (77b)

96b 66 54 15