Insertion of H2C=CHX (X = F, Cl, Br, OiPr) into (tBu3SiO)3TaH2 and β-X-elimination from (tBu3SiO)3HTaCH2CH2X (X = OR): Relevance to Ziegler-Natta copolymerizations

Article abstract of DOI:10.1021/ja003315n

The insertion of H₂C=CHX (X = OR; R = Me, Et, ⁿPr, ⁱPr, CH=CH₂, Ph) into (^tBu₃SiO)₃TaH₂ (1) afforded (^tBu₃SiO)₃HTaCH₂CH₂X (2-CH₂CH₂X), which β-X-eliminated to give ethylene and (^tBu₃SiO)₃HTaX (3-X). β-X-elimination rates were inversely proportional to the size of R. An X-ray crystallographic study of (^tBu₃SiO)₃ HTaCH₂CH₂O^tBu (2-CH₂CH₂O^tBu) revealed a distorted trigonal bipyramidal structure with an equatorial plane containing the hydride and a -CH₂CH₂O^tBu ligand with a staggered disposition. erythro- and threo-(^tBu₃SiO)₃HTaCHDCHDOEt (2-CHDCHDOEt) are staggered in solution, according to ¹H NMR spectroscopic studies, and eliminated cis- and trans-HDC=CHD, respectively, helping verify the four-centered transition state for β-OEt-elimination. When X = F, Cl, or Br, 2-CH₂CH₂X was not observed en route to 3-X, signifying that olefin insertion was rate-determining. Insertion rates suggested that substantial positive charge on the substituted carbon was incurred. The reactivity of other H₂C=CHX with 1, and a discussion of the observations and their ramifications on the incorporation of functionalized monomers in Ziegler-Natta copolymerizations, are presented.

Full text of DOI:10.1021/ja003315n

4728
J. Am. Chem. Soc. 2001, 123, 4728-4740
Insertion of H2CdCHX (X ) F, Cl, Br, OiPr) into (tBu3SiO)3TaH2
and â-X-Elimination from (tBu3SiO)3HTaCH2CH2X (X ) OR):
Relevance to Ziegler-Natta Copolymerizations
Stephanie A. Strazisar and Peter T. Wolczanski*
Contribution from the Baker Laboratory, Department of Chemistry and Chemical Biology,
Cornell UniVersity, Ithaca, New York 14853
ReceiVed September 8, 2000
n
i
Abstract: The insertion of H2CdCHX (X ) OR; R ) Me, Et, Pr, Pr, CHdCH2, Ph) into (tBu3SiO)3TaH2
(1) afforded (tBu3SiO)3HTaCH2CH2X (2-CH2CH2X), which â-X-eliminated to give ethylene and (tBu3SiO)3-
HTaX (3-X). â-X-elimination rates were inversely proportional to the size of R. An X-ray crystallographic
study of (tBu3SiO)3HTaCH2CH2OtBu (2-CH2CH2OtBu) revealed a distorted trigonal bipyramidal structure with
an equatorial plane containing the hydride and a -CH2CH2OtBu ligand with a staggered disposition. erythro-
1
and threo-(tBu3SiO)3HTaCHDCHDOEt (2-CHDCHDOEt) are staggered in solution, according to H NMR
spectroscopic studies, and eliminated cis- and trans-HDCdCHD, respectively, helping verify the four-centered
transition state for â-OEt-elimination. When X ) F, Cl, or Br, 2-CH2CH2X was not observed en route to 3-X,
signifying that olefin insertion was rate-determining. Insertion rates suggested that substantial positive charge
on the substituted carbon was incurred. The reactivity of other H2CdCHX with 1, and a discussion of the
observations and their ramifications on the incorporation of functionalized monomers in Ziegler-Natta
copolymerizations, are presented.
Introduction
methylene spacers14-16 and through complexation to Lewis
acids17,18 when early transition metal ZN catalysts are used. A
few late transition metal catalysts will copolymerize standard
conjugated polar vinyl substrates,19-21 but less control over the
polymer microstructure and the resulting properties hampers
commercial applications.
Scheme 1 illustrates the generally accepted Cossee´ mechanism
of insertion and â-H-elimination in the polymerization of
ethylene and related monomers.22,23 As path A shows, the
insertion processsin this case shown for ethylenescan be
assigned a simple rate constant (ke) for the overall event, or
can be separated into microscopic olefin binding (ke1/ke(-1)) and
insertion (ke2) steps. Propagation rates (Rp) are ascribed to the
insertion of olefin and are often first order in monomer and
The inclusion of functionality into hydrocarbon-based poly-
olefins can dramatically alter their microstructure and proper-
ties.1,2 Many commercially important polymers and resins could
be supplanted, provided the incorporation of functionality can
be controlled and the material is inexpensive. Current post-
polymer modification technology and free-radical-based copo-
lymerizations do not meet these criteria and fail to deliver the
diversity of products needed to generate interest in expanding
this research direction.2-4
Single-site Ziegler-Natta (ZN) catalysts can polymerize
ethylene at relatively low temperatures (<100 °C) and pressures
(<4 atm) and can be used to generate random copolymers when
an R-olefin is an additional substrate.5-11 While the current
generation of ZN catalysts can be tuned to afford hydrocarbon-
based polymers with varying properties, they will tolerate only
a limited number of functionalities present on the comonomer.
Polar comonomers must be protected by steric bulk12,13 and
(12) Kesti, M. R.; Coates, G. W.; Waymouth, R. M. J. Am. Chem. Soc.
1992, 114, 9679-9680.
(13) Wile´n, C.-E.; Luttikhedde, H.; Hjertberg, T.; Na¨smen, J. N.
Macromolecules 1996, 29, 8569-8575.
(14) Klabunde, U.; Ittel, S. D. J. Mol. Catal. 1987, 41, 123-134.
(15) (a) Aaltonen, P.; Fink, G.; Lo¨fgren, B.; Seppa¨la¨, J. Macromolecules
1996, 29, 5255-5260. (b) Aaltonen, P.; Lo¨fgren, B. Macromolecules 1995,
28, 5353-5357.
(16) Schneider, M. J.; Scha¨fer, R.; Mu¨lhaupt, R. Polymer 1997, 38,
2455-2459.
(1) Boffa, L. S.; Novak, B. M. Chem. ReV. 2000, 100, 1479-1493.
(2) Padwa, A. R. Prog. Polym. Sci. 1989, 14, 811-833.
(3) Doak, K. W. In Encyclopedia of Polymer Science and Engineering;
Mark, H. F., Ed.; John Wiley and Sons: New York, 1986; Vol. 6, pp 386-
429.
(4) Hagman, J. F.; Crary, J. W. In Encyclopedia of Polymer Science and
Engineering; Mark, H. F., Ed.; John Wiley and Sons: New York, 1985;
Vol. 6, p 325.
(5) Coates, G. W. Chem. ReV. 2000, 100, 1223-1252.
(6) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt, R.; Rieger, B.; Waymouth,
R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143-1170.
(7) Alt, H. G.; Ko¨ppl, A. Chem. ReV. 2000, 100, 1205-1222.
(8) (a) Hlatky, G. G. Chem. ReV. 2000, 100, 1347-1376. (b) Hlatky, G.
G. Coord. Chem. ReV. 1999, 181, 243-296.
(17) Marques, M. M.; Correia, S. G.; Ascenso, T. R.; Ribeiro, A. F. G.;
Gomes, P. T.; Dias, A. R.; Foster, P.; Rausch, M. D.; Chien, J. C. W. J.
Polym. Sci., Part A: Polym. Chem. 1999, 37, 2457-2469.
(18) Novak, B. M.; Hiromitsu, T. Polym. Mater. Sci. Eng. 1999, 80, 45.
(19) Johnson, L. K.; Mecking, S.; Brookhart, M. J. Am. Chem. Soc. 1996,
118, 267-268.
(20) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem.
Soc. 1998, 120, 888-899.
(21) Ittel, S. D.; Johnson, L. K.; Brookhart, M. Chem. ReV. 2000, 100,
1169-1203.
(9) Resconi, L.; Cavallo, L.; Fait, A.; Piemontesi, F. Chem. ReV. 2000,
100, 1253-1346.
(10) Kaminsky, W.; Arndt, M. AdV. Polym. Sci. 1997, 127, 143-187.
(11) Bochmann, M. J. Chem. Soc., Dalton Trans. 1996, 255-270.
(22) Cossee´, J. J. Catal. 1964, 3, 80-88.
(23) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA, 1987.
10.1021/ja003315n CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/25/2001

Insertion of H2CdCHX into (tBu3SiO)3TaH2
J. Am. Chem. Soc., Vol. 123, No. 20, 2001 4729
Scheme 1
Scheme 2
event, â-H-elimination from a growing polymer chain produces
a metal hydride that is still an active polymerization catalyst
(path B), but â-X-elimination from a functionalized polymer
generates a species that is incapable of initiating further
polymerization (path D)30 unless it is reactivated. Chain transfer
by â-H-elimination can also occur from the functionalized
polymer chain, but if kâ-X[LnMP+] is swifter than or similar to
ke[LnMP+][CH2dCH2], only oligomers can be formed.
The reactivity of (tBu3SiO)3TaH2 (1)31 with simple polar vinyl
monomers (CH2dCHX, X ) OR, halide) has been examined.
Dihydride 1 is less electrophilic than a typical cationic ZN
catalyst, allowing relative rates of insertion and â-X-elimination
to be observed as a function of X. As a consequence, inferences
to propagation and termination events in true ZN systems can
be made.
Results and Discussion
Vinyl Ethers. 1. Insertion and â-OR-Elimination Products.
(tBu3SiO)3TaH2 (1)31 was treated with a series of vinyl ethers
(CH2dCHOR, R ) Me, Et, nPr, iPr, tBu, Ph) in toluene-d8 at 7
( 1 °C, as indicated in Scheme 2. 1,2-Insertion to afford the
ethyl â-ether complexes, (tBu3SiO)3HTaCH2CH2OR (2-
CH2CH2OR), was observed, and none of the alternative 2,1-
insertion products were evident by 1H NMR spectroscopy. With
the exception of 2-CH2CH2OtBu, the remaining 2-CH2CH2OR
complexes underwent an apparent â-OR-elimination to pro-
vide (tBu3SiO)3HTaOR (3-OR). Free ethylene was observed,
and some formation of the known ethyl hydride derivative,
(tBu3SiO)3HTaEt (4-Et),31 was noted. All species in Scheme 2
were observed by 1H NMR spectroscopy, and Table 1 lists these
data, along with 13C{1H} NMR spectra that were obtained for
selected compounds.
2. Verification of â-OR-Elimination. To verify that all
(tBu3SiO)3HTaOR (3-OR) complexes were generated according
to the insertion/â-OR-elimination pathway illustrated, some
mechanistic probes were utilized. Scheme 3 indicates an
alternative mechanism predicated on the direct exchange of OR
for a hydride of (tBu3SiO)3TaH2 (1), which would render the
stereochemistry of the byproduct olefin unchanged from the
vinyl ether. Treatment of 1 with a mixture of cis- and trans-
â-methoxystyrene resulted in no reaction (eq 1). Since the
second order overall. The olefin complex is taken as a fleeting
intermediate,24 and the steady-state approximation can be applied
e
to obtain Rp ) ke[LnMP+][CH2CH2] and ke ) ke1ke2/(ke(-1)
+
ke2). In most cases and under most conditions, olefin coordina-
tion is considered to be fast and reversible, and the insertion
step is considered rate-determining (i.e., ke ≈ ke1ke2/(ke(-1)). In
some late metal systems, the catalyst resting state is considered
to be the alkyl-olefin complex, resulting in chain growth rates
that are zero order in monomer.25 Several studies have uncov-
ered polymerization systems that show [monomer]n (n > 1),
where the propagation approaches second order in monomer.
Ystenes proposed a “trigger mechanism” where coordination
of two monomers is crucial,26,27 while Resconi et al. explained
similar results by invoking two rapidly interconverting forms
of a catalyst resting state, with one reacting much faster than
the other.9,28,29 For the purposes of discussion herein, the
straightforward mechanisms in Scheme 1 will suffice.
A similar propagation sequence describes the insertion of
CH2dCHX (path C), and in a copolymerization process, an even
distribution of comonomers results if ke[LnMP+][CH2dCH2] ≈
kX[LnMP+][CH2dCHX]. Comonomer concentrations can be
controlled to a certain extent (neat monomer being an obvious
limit for liquid substrates), but each substrate also has an
inherent reactivity with LnMP+ (ke vs kX). In a chain-transfer
(tBu3SiO)3TaH2
+ MeOCHdCHPh f no reaction (1)
(24) Grubbs, R. H.; Coates, G. W. Acc. Chem. Res. 1996, 29, 85-93.
(25) Johnson, L. K.; Killian, C.-M.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 6414-6415.
1
(26) Ystenes, M. J. Catal. 1991, 129, 383-401.
phenyl group is spatially removed from the carbon bearing the
(27) Chien, J. C. W.; Yu, Z.; Marques, M. M.; Flores, J. C.; Rauch, M.
D. Polym. Sci., Part A: Polym. Chem 1998, 36, 319-328.
(28) Fait, A.; Resconi, L.; Guerra, G.; Corradini, P. Macromolecules
1999, 32, 2104-2109.
(30) For a recent example of how â-Cl-elimination interferes with the
Ziegler-Natta polymerization of ethylene and vinyl chloride, see: Stock-
land, R. A., Jr.; Jordan, R. F. J. Am. Chem. Soc. 2000, 122, 6315-6316.
(31) 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.
(29) Resconi, L.; Fait, A.; Piemontesi, F.; Colonnesi, M.; Rychlicki, H.;
Ziegler, R. Macromolecules 1995, 28, 6667-6676.

4730 J. Am. Chem. Soc., Vol. 123, No. 20, 2001
Strazisar and Wolczanski
Table 1. 1H and 13C{1H} NMR Spectral Data for (tBu3SiO)3HTaCH2CH2(X/OR) (2-CH2CH2(X/OR)), (tBu3SiO)3HTa(X/OR) (3-(X/OR)),
(tBu3SiO)3HTaR (4-R), and Related Complexes in C7D8 unless Otherwise Noted
a-c
1H NMR (δ,a assignment, multiplicity, J (Hz))
13C{1H} NMR (δ,b assignment)
compound
((H3C)3C)3
TaHd
other
C(CH3)3
C(CH3)3
other
(tBu3SiO)3TaH2 (1)e
(tBu3SiO)3TaHD (1-d)e
2-CH2CH2OMe
1.23
1.23
1.27
21.91
21.97
22.35
(br)
29.86
24.14
2.04 (CH2, dt, 7.8, 3.0),
3.20 (CH3, s),
4.28 (CH2O, t, 7.6)
1.14 (CH3, t, 6.9)
3.44 (OCH2, q, 7.0),
4.25 (CH2O, t, 7.6)
0.89 (CH3, t, 6.0)
1.57 (CH2, m, 6.7)
2.14 (TaCH2, dt, 3.1, 7.0),
3.39 (OCH2, t, 6.6)
4.26 (CH2O, t, 8.6)
2.25 (CH2, m)
f
2-CH2CH2OCH2CH3
1.24
1.25
22.38
(t, 3.2)
2-CH2CH2OCH2CH2CH3
22.38
(br)
2-CH2CH2OPh
1.25
1.25
22.45
(br)
4.82 (CH2O, m)
6.6-6.7 (Ph, m)
2-CH2CH2OCHMe2
22.38
(t, 3.2)
1.13 (Me2, d, 6.1)
2.14 (CH2, m, 3.4)
3.60 (CH, sept, 6.0)
4.20 (CH2O, m, 8.4)
1.22 (Me3, s)
30.65
30.80
23.66
23.78
22.77 (Me2)
70.32
70.84
75.97 (OCH)
30.60 (Me3)
64.21
2-CH2CH2OCMe3
1.25
1.25
22.37
(br)
2.06 (CH2, m)f
4.11 (CH2O, m, 8.2)
72.53
76.88 (CMe3)
f
2-CH2CH2OCHdCH2
22.41
4.21 (CH2O, m, 6.1)
4.30 (dCHH, m, 11.9, 3.1)
6.57 (OCHd, dd, 10.8, 6.3)
4.15 (s)
3-OMe
3-OCH2CH3
1.28
1.25
22.86
21.60
1.17 (CH3, t, 7.0)
4.48 (CH2, q, 7.0)
0.86 (CH3, t, 6.4)
3-OCH2CH2CH3
1.27
21.16
1.59 (CH2, sext, 7.0)
4.46 (CH2O, t, 6.7)
1.14 (Me2, d, 6.1)
5.05 (CH, sept, 6.1)
7.1-7.2 (Ph, m)
4.06 (dCHH, d, 6.0)
4.49 (dCHH, d, 13.5)
5.77 (OCHd, dd, 10.8, 6.3)
0.82 (CH3, t, 7.3)
1.92 (CH2, sept, 4.0)
5.13 (CHdCH2, m)
5.70 (CH, quin, 8.9)
3-OCHMe2
1.28
21.19
30.78
30.78
23.71
23.65
26.47 (Me2)
76.92 (OC)
3-OPh
3-OCHdCH2
1.25
1.27
21.60
21.18
3-OCHEtCHdCH2
3-OCH2CHdCH2
1.28
21.26
9.37 (CH3)
31.43 (CH2)
87.46 (OC)
116.27 (dCH2)
140.89 (dCHs)
1.26
21.51
21.19
4.98 (OCH2, dq, 5.5, 1.2)
5.10 (dCHH, dq, 10.4, 1.2)
5.20 (dCHH, dq, 17.1, 1.6)
5.92 (dCHs, m)
1.14 (Me2, d, 6.1)
5.05 (CH, sept, 6.1)
3-OCHMe2
3-F
1.28
1.24
30.78
30.51
23.71
23.75
26.47 (Me2)
76.92 (CH)
21.96
(d, 95)
20.03
19.08
22.24
(t, 3.1)
22.30
(t, 3.2)
3-Cle
3-Br
4-CH2CH3
1.23
1.27
1.25
30.84
30.76
24.52
23.69
e,g
2.15 (CH3, t, 7.8)
1.76 (CH2, dq, 3.4, 7.8)
0.96 (CH3, t, 7.2)
1.80 (CH2, dt, 3.4, 7.8)
2.33 (CH2, m)
16.44 (CH3)
70.90 (CH2)
4-CH2CH2CH3
1.26
1.29
0.91 (CH3, t, 7.5)
31.04
23.90
11.77 (CH3)
f
(tBu3SiO)3TaOCHEtCH2CH2
(5-Et)
1.70 (CH2, m, 6.7)
2.66 (CHH, m, 14.4, 5.5)
3.48 (TaCH2, quin, 4.9)
4.58 (OCH, m)
30.50 (CH2Me)
42.96 (CH2)
68.90 (TaCH2)
89.14 (OC)
1.28
2.12 (TaCH2, t, 7)
(tBu3SiO)3TaOCH2CH2CH2
(5-H)
2.92 (CH2, quin, 6.7)
4.62 (OCH2, t, 6.2)
1.25 (CH3, t, 7.6)
4.68 (CH2, q, 7.1)
1.27 (CH3, t, 6)
1.78 (CH2, sext, 7.6)
4.58 (CH2, m, 8.3)
(tBu3SiO)3Ta(OCH2CH3)2
1.28
1.30
(6-(OEt)2)
(tBu3SiO)3Ta(OCH2CH2CH3)2
(6-(OnPr)2)
a Referenced to the residual methyl resonance of C7D7H at δ 2.09. b Referenced to methyl of toluene-d8 at δ 20.4 unless otherwise noted. c When
multiplets were observed, phenomenological couplings are given. d Singlet unless otherwise noted. e See ref 31. f Some resonances obscured or
t
partially obscured by solvent or BuSiO. g 13C {1H} NMR spectrum taken in benzene-d6 and referenced to C6D6 at δ 128.00.

Insertion of H2CdCHX into (tBu3SiO)3TaH2
J. Am. Chem. Soc., Vol. 123, No. 20, 2001 4731
Scheme 3
3. Solution Geometry of (tBu3SiO)3HTaCH2CH2OR (2-
CH2CH2OR). While the eclipsed geometry of the transition state
is required for â-OR-elimination, it is also possible that the â-OR
group is bound in the ground-state geometry of (tBu3SiO)3-
HTaCH2CH2OR (2-CH2CH2OR). Variable-temperature 1H NMR
spectroscopic studies on the insertion intermediates 2-erythro-
CHDCHDOEt and 2-threo-CHDCHDOEt were performed in
order to ascertain the likely solution conformation of the â-OEt-
alkyl. The 3JHH coupling constant was measured from 23 to -50
°C, and the Karplus relationship,34 which relates the coupling
between two vicinal protons to their dihedral angle (φ), was
used to determine the dominant conformation. For 2-threo-
CHDCHDOEt, Scheme 4 shows that φ ) 120° when Ta and
OEt are eclipsed and φ ) 60° in the staggered conformer. The
Karplus relationship predicts slightly smaller coupling when φ
3
) 60°; hence, JHH′ should decrease upon cooling if the
OMe substituent, a direct exchange should be relatively sterically
unencumbered. An insertion/â-OR-elimination pathway requires
the Ph group to occupy a sterically unfavorable R-position in
the incipient (tBu3SiO)3HTaCHPhCH2OMe species, and it is
plausible that this interaction obviates any reaction involving a
1,2-insertion. The opposite 2,1-insertion does not occur either
and may also be a consequence of the large methoxy (relative
to H) substituent on the R-carbon in the putative alkyl hy-
dride product. When excess 2-methoxypropene was added to
1, (tBu3SiO)3HTaOMe (3-OMe) did form (eq 2), but the
staggered conformer is at a lower energy, and only a minimal
3
change from JHH ) 4.9 to 4.3 Hz was observed.
In 2-erythro-CHDCHDOEt, φ ) 0° in the eclipsed conformer
versus φ ) 180° in the one where Ta and OEt are staggered.
Since coupling is greater when φ ) 180°, 3JHH′ should increase
upon cooling if the staggered geometry is the lower energy
3
conformer, and JHH increased slightly from 11.9 to 12.2 Hz.
3
For both isomers, JHH changed in the direction predicted by
the Karplus relationship if the population of the staggered
conformer increased, and the larger absolute magnitude of 3JHH′
for 2-threo-CHDCHDOEt versus 2-erythro-CHDCHDOEt was
also consistent with the proposed antiperiplanar geometry. Since
3JHH did not change much with temperature in either case, it is
likely that 2-CH2CH2OR exists predominantly in the staggered
geometry at 23 °C.
C7D8
(tBu3SiO)3TaH2
+ H2CdCMe(OMe) 23 °C8
1
(tBu3SiO)3HTaOMe (3-OMe) + H2CdCHMe +
(tBu3SiO)3HTanPr (4-nPr) (2)
4. Structure of (tBu3SiO)3HTaCH2CH2OR (2-CH2CH2-
OtBu). The single-crystal X-ray structure of (tBu3SiO)3HTaCH2-
CH2OtBu (2-CH2CH2OtBu) constitutes a relatively rare example
of a structurally characterized early transition metal alkyl
hydride. An abbreviated list of data collection and refinement
parameters is given in Table 2, and pertinent bond distances
and angles are also tabulated (Table 3). The molecular view in
Figure 1 reveals a severely distorted trigonal bipyramidal
geometry with an equatorial â-OtBu-CH2CH2 ligand that
exhibits a nearly staggered (i.e., dihedral/Ta(C1)C2O4 ) 160.7°)
conformation. The hydride was located in an equatorial position
with d(Ta-H) ) 1.72(4) Å, presumably a reflection of its strong
σ-donating capacity. The two axial siloxides (d(Ta-O2) )
1.926(3) Å, d(Ta-O3) ) 1.913(2) Å) are longer than the
equatorial one (d(Ta-O1) ) 1.874(3) Å) as expected,35 and
the alkyl occupies the remaining equatorial position with a
normal d(Ta-C1) of 2.176(4) Å. While it is somewhat
surprising that both axial positions are occupied by the large
siloxide ligands, strong σ-donors (e.g., H, alkyl) typically occupy
equatorial sites in trigonal bipyramids,35 and the more elec-
tronegative ligands are often axially disposed.36
intermediate â-OMe-propyl hydride was not noted. The half-
life for the appearance of 3-OMe was significantly longer (t1/2
≈ 12 h, 23 °C) than that for the corresponding methyl vinyl
ether reaction (t1/2 ≈ 0.75 h, 7 °C). The propyl hydride,
(tBu3SiO)3HTanPr (4-nPr), was also generated due to release of
propene upon â-OMe-elimination, and this reaction was inde-
pendently verified by the reaction of 1 and H2CdCHMe. From
these observations, steric factors appear more critical at the
R-carbon, but â-carbon substituents can also significantly
influence the overall rate of the insertion/â-OR-elimination
pathway.
A more informative set of experiments concerns the insertion/
â-OEt-elimination of 1,2-dideuterioethyl vinyl ethers illu-
strated in Scheme 4. cis- and trans-1,2-dideuterioethylene32 are
readily distinguished by bending vibrations (cis-CHDdCHD,
842 cm-1; trans-, 988, 726 cm-1) in their IR spectra.33
When (tBu3SiO)3TaH2 (1) reacted with each labeled ethylvinyl
ether, complete inversion of ethylene stereochemistry oc-
curred: 1 + cis-CHDdCDOEt-d2 gave trans-CHDdCHD, and
1 + trans-CHDdCDOEt-d2 gave cis-CHDdCHD. The stereo-
chemistry of the ethylene products is a natural consequence of
the standard four-center transition state that arises when the EtO
group of the pendent â-OEt-alkyl binds via the eclipsed
geometry. A direct exchange pathway would result in retention
of ethylene stereochemistry, as shown in Scheme 3 for trans-
CHDdCDOR-d2. On the basis of these findings and the
â-methoxystyrene experiment, and the direct observation of
(tBu3SiO)3HTaCH2CH2OR (2-CH2CH2OR), the insertion/
â-OR-elimination process in Scheme 2 was deemed verified.
The axial siloxides have a noticeable cant toward the hy-
dride ( H-Ta-O2 ) 81.3(12)°, H-Ta-O3 ) 81.3(12)°,
O2-Ta-O3 ) 154.63(11)°) and away from the more steri-
cally demanding siloxides ( O1-Ta-O2 ) 102.07(11)°,
O1-Ta-O3 ) 102.49(11)°). If the core were to be described
as a square pyramid, Ta-O1 would be apical ( H-Ta-O1 )
119.1(12)°, O1-Ta-C1 ) 112.76(15)°). In view of the rather
open H-Ta-C1 angle of 128.1(12)°, one can imagine a rotation
(34) (a) Karplus, M. J. Chem. Phys. 1959, 30, 11-15. (b) Karplus, M.
J. Am. Chem. Soc. 1963, 85, 2870-2871.
(35) Rossi, A. R.; Hoffmann, R. Inorg. Chem. 1975, 14, 365-374.
(36) Hoffmann, R.; Howell, J. M.; Muetterties, E. L. J. Am. Chem. Soc.
1972, 94, 3047-3058.
(32) Keul, H.; Choi, H.-S.; Kuczkowski, R. L. J. Org. Chem. 1985, 50,
3365-3371.
(33) Arnett, R. L.; Crawford, B. L., Jr. J. Chem. Phys. 1950, 18, 118-
126.

4732 J. Am. Chem. Soc., Vol. 123, No. 20, 2001
Strazisar and Wolczanski
Scheme 4
Table 2. Crystallographic Data for Monoclinic
(tBu3SiO)3HTaCH2CH2OtBu (2-CH2CH2OtBu)
corresponding barrier is substantial in this system. Finally, the
â-OR-elimination pathway was further complicated by the
reaction of ethylene with the starting dihydride (1) to produce
(tBu3SiO)3HTaEt (4-Et).31
formula
formula weight 929.40
C42H95O4Si3Ta
T
Z
173(2) K
4
space group
λ
a
b
c
vol
P21/n
Fcalc
1.214 g/cm3
2.266 mm-1
0.0396
Rate constants for â-OR-elimination (Table 4) were deter-
mined by allowing k1, k2, and k3 in Scheme 2 to be parameters
in a least-squares fitting of the concentration versus time data
of H2CdCHOR, (tBu3SiO)3TaH2 (1), (tBu3SiO)3HTaCH2-
CH2OR (2-CH2CH2OR), (tBu3SiO)3HTaOR (3-OR), and
(tBu3SiO)3HTaEt (4-Et). For each particular vinyl ether present
in excess, the hydride resonance of each species was monitored
over time by 1H NMR spectroscopy. Figure 2 reveals the
observed concentrations versus time data for the H2CdCH-
OnPr case plotted in conjunction with simulated data from the
calculated rate constants k1, k2, and k3. An excellent agreement
between actual and simulated data was obtained for each
substrate, indicating the proposed mechanism was at least
consistent with the concentrations versus time profile. In
principle, such a procedure should have provided k1 and k3 in
addition to k2, but experimental difficulties permitted accurate
determination of only the latter. As Figure 2 exemplifies, only
a fraction of the “rise time” of the intermediate (tBu3-
SiO)3HTaCH2CH2OR (2-CH2CH2OR) could be determined for
0.71073 Å
µ
13.2367(10) Å R [I > 2σ(I)]a
17.7186(13) Å wR2 [I > 2σ(I)]b 0.0774
22.0853(16) Å GOFc
5083.7(6) Å3
0.986
a R1 ) ∑||Fo| - |Fc||/∑|Fo|. b wR2 ) [∑w(|Fo| - |Fc|)2/∑wFo ]1/2
.
2
c GOF ) [∑w(|Fo| - |Fc|)2/(n - p)]1/2, n ) number of independent
reflections (11 575 (Rint ) 0.0635) of 44 146), p ) number of
parameters (455).
about the Ta-C1 and C1-C2 bonds that allows the â-OR
substituent to coordinate opposite O1 in a six-coordinate transi-
tion state for â-OR-elimination. The solid-state structure predicts
two inequivalent siloxide ligands in a 2:1 ratio, but only one
was observed in the 1H NMR spectrum, consistent with the usual
rapid fluxionality accorded d0 five-coordinate species.
5. Rates of â-OR-Elimination. The direct determination of
â-OR-elimination rates was not feasible. The steady-state
approximation was invalid since the concentration of
(tBu3SiO)3HTaCH2CH2OR (2-CH2CH2OR) was appreciable and
varied substantially as the reaction progressed. Additionally, no
pre-equilibrium between (tBu3SiO)3TaH2 (1) and vinyl ether
substrate was detected (i.e., k1 . k-1, k2 . k-1) via treatment
with the labeled ethyl vinyl ethers. No (tBu3SiO)3TaH(D)
n
R ) Me, Et, Pr, and Ph. As a consequence, and due to the
nature of second-order reactions, a variety of k1 values (typically
(1-5) × 10-3 M-1 s-1) gave acceptable fits. Inaccuracies in k3
arise simply from analysis of the small amount of (tBu3-
SiO)3HTaEt (4-Et) generated at any point in time.
1
(1-d), which has a Ta-H resonance in the H NMR spectrum
(tBu3SiO)3TaH2 (1) was also treated with divinyl ether, whose
two olefinic residues can participate in insertion chemistry. As
Scheme 5 illustrates, insertion to give (tBu3SiO)3HTaCH2CH2-
OCHdCH2 (2-CH2CH2OVy) was observed along with â-OVy-
elimination to afford (tBu3SiO)3HTaOCHdCH2 (3-OVy), but
the reaction had an additional complication: formation of the
double insertion product ((tBu3SiO)3HTaCH2CH2)2O ((2-CH2-
CH2)2O). In this case, k1-k4 were the parameters used to fit
upfield from 1,31 was observed during the course of this reaction
(eq 3). In view of previous estimates of the thermochemistry
C7D8
(tBu3SiO)3TaH2
+ EtOCDdCHD N
1
t
+ EtOCHdCHD (3)
( Bu3SiO)3TaH(D)
1-d
(37) Schaller, C. P.; Cummins, C. C.; Wolczanski, P. T. J. Am. Chem.
Soc. 1996, 118, 591-611.
(38) Bennett, J. L.; Wolczanski, P. T. J. Am. Chem. Soc. 1997, 119,
10696-10719.
of â-H-elimination in (tBu3SiNH)3ZrCH2CHMe2 (∆H°(â-H-
elim) g 16 kcal/mol)37 and (tBu3SiO)2(tBu3SiNH)TiCH2CH3
(∆H°(â-H-elim) ) 16.3 kcal/mol),38 it is not surprising that the

Insertion of H2CdCHX into (tBu3SiO)3TaH2
J. Am. Chem. Soc., Vol. 123, No. 20, 2001 4733
Table 3. Interatomic Distances (Å) and Angles (Deg) Pertaining to (tBu3SiO)3HTaCH2CH2OtBu (2-CH2CH2OtBu)
Ta-O1
Ta-O2
Ta-O3
Ta-C1
Ta-H
1.874(3)
1.926(3)
1.913(2)
2.176(4)
1.72(4)
O1-Si1
O2-Si2
O3-Si3
C1-C2
C2-O4
Si-Cav
1.678(3)
1.671(3)
1.671(3)
1.507(6)
1.400(5)
1.927(7)
C3-O4
C3-C4
C3-C5
C3-C6
C-Cav
1.430(6)
1.473(8)
1.500(11)
1.486(8)
1.540(8)
SiC-Cav
1.537(17)
H-Ta-O1
H-Ta-O2
H-Ta-O3
H-Ta-C1
Ta-O2-Si2
C2-O4-C3
O4-C3-C6
C5-C3-C6
Si-C-Cav
119.1(12)
81.3(12)
81.3(12)
128.1(12)
156.60(17)
118.1(4)
105.6(5)
117.6(7)
111.6(12)
C1-Ta-O1
C1-Ta-O2
C1-Ta-O3
Ta-C1-C2
Ta-O3-Si3
O4-C3-C4
C4-C3-C5
O-Si-Cav
112.76(15)
90.19(14)
86.16(13)
117.2(3)
163.82(18)
108.5(5)
104.5(8)
106.7(12)
107.2(10)
O1-Ta-O2
O1-Ta-O3
O2-Ta-O3
Ta-O1-Si1
C1-C2-O4
O4-C3-C5
C4-C3-C6
C-Si-Cav
102.07(11)
102.49(11)
154.63(11)
178.03(18)
109.1(4)
113.2(5)
107.1(5)
112.1(9)
C-C-Cav
Table 4. First-Order Rate Constants for â-OR-Eliminations from
(tBu3SiO)3HTaCH2CH2OR (2-CH2CH2OR)a
(tBu3SiO)3HTaCH2CH2OR
k2
T
∆Gq
(2-CH2CH2OR)
(×104 s-1) ((1 °C) (kcal/mol) E c (R)b
s
2-CH2CH2OVyc
2-CH2CH2OMed
2-CH2CH2OEtd
2-CH2CH2OnPrd
2-CH2CH2OPhd
28(3)
9.8(3)
3.9(3)
3.6(1)
3(1)
719.6(1) 19.6(1)
7
7
7
7
20.2(1) -1.24
20.7(1) -1.62
20.8(1) -1.91
20.9(2) -1.01e
-3.82f
2-CH2CH2OiPrd,g
0.095(1) 35
-2.32
0.31(1)
1.70(1)
7.8(2)
45
60
75
7
2-CH2CH2OtBuh
<0.002
-3.7
a Monitored by H NMR spectroscopy in C7D8. b Taft parameters
(ref 40) used as a measure of steric bulk of R. c Determined from
nonlinear, least-squares fitting of the differential forms of the rate
expressions to the concentrations vs time data with parameters k1-k4
of Scheme 5. d Determined from nonlinear, least-squares fitting of the
differential forms of the rate expressions to the concentrations vs time
data with parameters k1-k3 of Scheme 2. e “Thickness” of phenyl ring
(ref 40). f “Width” of phenyl ring (ref 41). g Values used in the Eyring
plot (35-75 °C). From a weighted nonlinear least-squares fit of the
data: ∆Hq ) 22.9(2) kcal/mol, ∆Sq ) -7(3) eu, k2(7 °C) ) 2.0 ×
10-7 s-1, ∆Gq(7 °C) ) 24.8 kcal/mol. h Estimated assuming <2% of
2-CH2CH2OtBu â-OtBu-eliminates in 24 h (7 °C). This is a generous
value because no OtBu-elimination was noted for prolonged periods at
23 °C.
1
Figure 1. Molecular view of (tBu3SiO)3HTaCH2CH2OtBu (2-CH2-
CH2OtBu).
the concentrations vs time profiles of all observables, and
â-OCH2CH2TaH(OSitBu3)3-elimination from the dinuclear com-
plex (2-CH2CH2)2O was not detected.
The data in Table 4 indicate that rates of elimination
decreased with increasing size of R, whose Taft E cs values39-42
are also included as objective steric factors for comparison. The
inverse rate dependence on the size of R is consistent with a
greater difficulty in achieving the eclipsed transition-state
geometry for â-OR-elimination from (tBu3SiO)3HTaCH2CH2-
OR (2-CH2CH2OR). The fastest to eliminate was the vinyl ether
2-CH2CH2OVy, whose sp2 carbons render this group arguably
smaller (note the small value for the “thickness” of the phenyl
substituent) than OMe, which is ∼3 times slower.41 A small
declination in ratessabout a factor of 3sis then observed for
the group OMe > OEt > OnPr, OPh. It is conceivable that the
2-CH2CH2OVy and 2-CH2CH2OPh manifest electronic contri-
butions to rates of â-OR-elimination, since the vinyl and phenyl
substituents are more electron-withdrawing than their alkyl
counterparts. However, given the rather minor overall variance
in rates, no phenyl substituent studies were undertaken.
There is certainly a major steric influence for the secondary
manifests â-OR-elimination at 7 °C within 10 h. â-OiPr-
elimination from (tBu3SiO)3HTaCH2CH2OiPr (2-CH2CH2OiPr)
was observed at elevated temperatures (35-75 °C), enabling
∆Hq ) 22.9(2) kcal/mol and ∆Sq ) -7(3) eu to be determined
from an Eyring plot. The data are consistent with the con-
strained geometry of the purported four-center transition state
for â-OiPr-elimination and reveal a substantially elevated barrier
(∆∆Gq ≈ 4 kcal/mol) in comparison to the OnPr case. In
summary, steric factors appear to dominate variations in â-OR-
elimination rates, as anticipated in view of the four-center
transition state in Scheme 4.
6. Insertion of H2CdCHOiPr. Since (tBu3SiO)3HTaCH2-
CH2OiPr (2-CH2CH2OiPr) was stable at 7 °C, it was possible
to determine the rate of insertion of H2CdCHOiPr into a hydride
ligand of (tBu3SiO)3TaH2 (1). This was a standard second-order
reaction, where the disappearance of the dihydride can be
expressed as -d[1]/dt ) k1[1][H2CdCHOiPr]. Since the initial
concentrations of dihydride and isopropyl vinyl ether were
known, k1 was obtained straightforwardly through a plot of ln-
([1]/[H2CdCHOiPr]) vs time. A straight line was obtained
whose slope is equivalent to ([1]o - [H2CdCHOiPr]o)k1,43
yielding a value of 1.9(3) × 10-3 M-1 s-1 for the insertion rate
constant, which is included in Table 5.
i
t
and tertiary substituents, R ) Pr and Bu, neither of which
(39) Taft, R. W., Jr. J. Am. Chem. Soc. 1952, 74, 3120-3128.
(40) Unger, S. H.; Hansch, C. Prog. Phys. Org. Chem. 1976, 12, 91-
118.
(41) Charton, M. J. Am. Chem. Soc. 1969, 91, 615-619.
(42) MacPhee, J. A.; Panaye, A.; Dubois, J.-E. Tetrahedron 1978, 34,
3553-3562.

4734 J. Am. Chem. Soc., Vol. 123, No. 20, 2001
Strazisar and Wolczanski
for the vinyl ether series, no further verification of pathway was
considered necessary for the vinyl halide set.
3. Rates of H2CdCHX Insertion. Since (tBu3SiO)3HTaCH2-
CH2X (2-CH2CH2X) could not be directly observed, a fitting
procedure such as that utilized in the vinyl ether cases could
not be employed. Two mechanistic scenarios are plausible, given
the smooth conversion of (tBu3SiO)3TaH2 (1) + H2CdCHX to
(tBu3SiO)3HTaX (3-X): rate-determining olefin insertion (k1)
or reversible insertion and rate-determining â-X-elimination
(k1k2/k-1). The reversibility of the vinyl halide insertion step
was monitored by treating (tBu3SiO)3TaD2 (1-d2) with vinyl
chloride (eq 5). No hydrogen (<3%) was incorporated into the
C7D8
2(tBu3SiO)3TaD
t
2 + H2CdCHCl
8
+
( Bu3SiO)3DTaCl
60 °C
1-d2
3-D/Cl
(tBu3SiO)3DTaEt-d2
4-D/Et-d2
(5)
Figure 2. Concentration vs time data for the H2CdCHOnPr case:
experimental points and simulated lines from the fit of k1, k2, and k3 of
Scheme 2.
starting dideuteride (1-d2)31 as the reaction progressed, and no
hydrogen (<3%) was incorporated into the product (tBu3SiO)3-
DTaCl (3-D/Cl). The ethyl hydride product contained no H in
the hydride site, and two deuteriums were distributed over the
CR (∼25% D expected) and Câ (∼50% D expected) sites of
the ethyl group, consistent with insertion of H2CdCHD into
the Ta-D bond. Quantitation of the site distribution was
precluded by overlapping resonances, but the spectra were
roughly consistent with the statistical distribution. Together these
analyses indicate that H2CdCHCl insertion is essentially
irreversible, thereby eliminating the second mechanistic scenario.
Direct measurements of the H2CdCHX insertion rates were
accomplished by monitoring the disappearance of (tBu3SiO)3-
TaH2 (1) with an excess of vinyl halide. Slightly greater than
one-half of the rate of disappearance of 1 was due to reaction
with vinyl halide (eqs 6-9);
Scheme 5
t
(tBu3SiO)3TaH2
x( Bu3SiO)3HTaX
3-X
+ H CdCHX f
+
2
1
(1 - x)(tBu3SiO)3HTaEt
4-Et
(6)
Vinyl Halides. 1. Insertion and â-X-Elimination Products.
(tBu3SiO)3TaH2 (1) was exposed to H2CdCHX (X ) F, Cl,
Br) at 60 ( 1 °C in toluene-d8 to afford (tBu3SiO)3HTaX (3-
X)31 and (tBu3SiO)3HTaEt (4-Et) according to Scheme 6. NMR
spectral data for the products are given in Table 1. Complete
disappearance of 1 at 60 °C upon treatment with vinyl halides
generally took longer than that with the vinyl ethers at 7 °C,
and no alkyl-â-X-insertion products (tBu3SiO)3HTaCH2CH2X
d[1]
dt
(-x)
) k1[1][H2CdCHX]
(7)
(8)
d[1]
[1]
) -(k1/x)[H2CdCHX] dt
where
z ) [3-X]/[4-Et] ) x/(1 - x)
and
x ) z/(z + 1)
1
(2-CH2CH2X) were observed in H NMR spectra during the
course of the reaction. An approximately even distribution of
tantalum-containing products was observed, consistent with a
relatively swift insertion of ethylene in comparison to that of
H2CdCHX.
ln([1]/[1]o) ) -kobs
t
(9)
where
kobs ) (k1/x)[H2CdCHX]o
2. Verification of â-X-Elimination. (tBu3SiO)3TaH2 (1) did
not react with â-bromostyrene (eq 4), in concert with the
insertion/â-X-elimination mechanism in Scheme 6, and incon-
the remainder was from its reaction with ethylene. In all cases,
(tBu3SiO)3HTaX (3-X) and (tBu3SiO)3HTaEt (4-Et) grew in at
a constant ratio (z ) [3-X]/[4-Et]) throughout the course of
reaction, and z was always >1, presumably due to some escape
of ethylene to the headspace of the NMR tube. At the end of
the runs, a small amount of ethylene can be detected in solution
by 1H NMR spectroscopy. Since only the amount of 3-X formed
reflected the loss of 1 due to vinyl halide insertion, a correction
factor that accounts for the stoichiometry must be employed.
(tBu3SiO)3TaH2
+ BrCHdCHPh f no reaction (4)
1
sistent with a direct exchange pathway related to Scheme 3. As
discussed for â-methoxystyrene (eq 1), a direct exchange
mechanism for production of 3-X should be relatively insensitive
to substitution on the carbon â to X. Given the precedent set
(43) Benson, S. W. The Foundations of Chemical Kinetics; McGraw-
Hill Co.: New York, 1960.

Insertion of H2CdCHX into (tBu3SiO)3TaH2
J. Am. Chem. Soc., Vol. 123, No. 20, 2001 4735
Table 5. Second-Order Rate Constants for H2CdCHX (X ) F, Cl, Br, OiPr) Insertion into (tBu3SiO)3TaH2 (1)a,b
k1(av)c
c
[H2CdCHX]
k1
e
H2CdCHX
H2CdCHF
[1] (M)
(equiv)
(×104 M-1 s-1
)
(×104 M-1 s-1
)
xd
σ+
F,X
0.0452
0.0452
0.0346
0.0561
0.0452
0.0482
0.0560
0.0161
0.0178
0.0482
0.0482
0.0533
0.0469
0.0452
0.0444
0.0561
0.0561
0.0573
0.0482
0.0452
0.0482
0.0346i
0.0346i
0.0346i
6
6
6
9
9
9
16
16
16
6
6
6
6
6
6
10
29.1(7)
18.0(3)
23.9(4)
31.0(3)
19.1(2)
29.7(5)
35(1)
26.9(2)
20.7(2)
1.58(1)
1.08(1)
1.53(1)
4.89(2)
6.74(2)
6.13(8)
6.13(2)
6.20(3)
5.6(1)
24(5)
27(5)
27(6)
1.4(2)
5.9(8)
6.0(3)
3.3(8)h
19(3)i
0.75
0.75
0.70
0.70
0.75
0.75
0.75
0.80
0.75
0.55
0.55
0.60
0.60
0.65
0.65
0.65
0.65
0.60g
0.65
0.70
0.70
-0.07f
H2CdCHCl
H2CdCHBr
0.11f
0.15f
10
10
19
19
2.57(4)
4.39(3)
2.86(3)
20.2(4)i
16(2)i
19
H2CdCHOiPri
0.0720i
0.0882i
0.0869i
-0.78j
20.9(1)i
no reaction
t
H2CdCCO2 Bu
0.48k
a Monitored by 1H NMR spectroscopy in C7D8 at 60 °C unless otherwise noted. b r2 values of g0.998 were obtained for all runs unless otherwise
noted. c Determined from pseudo-first-order plots as described in eqs 6-9 unless otherwise noted. d Defined in eq 6. e Hammett σ+ parameters
F
(refs 45, 46) taken to indicate the influence of an electron-withdrawing (σ+ > 1) or electron-donating (σ+ < 1) substituent on an electron-poor
F
F
transition state. f Hammett σI inductive parameters: F, 0.50; Cl, 0.47; Br, 0.45. g r2 ) 0.997. h Aberration from 6 and 10 equiv values perhaps a
consequence of H2CdCHBr insolubility (see text). i Determined at 7 °C via a second-order plot as described in the text; [H2CdCHOiPr] is expressed
in moles per liter. j Assumed the same as for OMe. k Assumed the same as CO2Et.
Scheme 6
slower than the insertion of H2CdCHBr. The vinyl bromide
insertion rate constant was ∼6 × 10-4 M-1 s-1 according to
the 6 and 10 equiv runs, placing it between vinyl fluoride and
vinyl chloride. Note that the 19 equiv run was significantly lower
than the others, but it is suspected that the number actually
reflects the limits of vinyl bromide solubility in toluene and is
inaccurate. In summary, there is a modest range (factor of ∼18)
of insertion rate constants for the vinyl halides, with X ) F >
Br > Cl.
Alternative Substrates. A selection of other substrates,
including vinyl acrylates and ketones, were also submitted with
(tBu3SiO)3TaH2 (1) in the hopes of obtaining data relevant to
the insertion processes of Ziegler-Natta polymerizations. With
these substrates, both reduction of the carbonyl moiety and
insertion into a Ta-H bond could occur, and a survey conducted
in sealed NMR tubes revealed that the former dominated, as
expected.
Since z ) x/(1 - x), then x ) z/(z + 1), and the correction
factor 1/x can be easily obtained from the product ratio (eqs 7
and 8). The integrated rate expression for the disappearance of
1 (eq 9), where kobs ) (k1/x)[H2CdCHX]o, was valid when the
reaction was run under pseudo-first-order conditions in vinyl
halide.43 Values of kobs were obtained from straight-line plots
of -ln([1]/[1]o) vs time for all runs, consistent with the
mechanism in Scheme 6 and the kinetics analysis in eqs 7-9,
and Table 5 lists the second-order rate constants ascertained
from the plots.
Experimental limitations restricted the amount of vinyl halides
utilized in the kinetics experiments, and despite the borderline
pseudo-first-order conditions in 6 equiv runs, the experiments
were generally well behaved. For the case of H2CdCHF at 60
°C, 6, 9, and 16 equiv were used, good agreement was ob-
tained, and the rate constants for insertion were approximately
2.5 × 10-3 M-1 s-1. Only 6 equiv runs could be obtained for
H2CdCHCl because the solubility of vinyl chloride in toluene
proved to be limited. The value of 1.4(2) × 10-4 M-1 s-1 was
significantly slower than that for the fluorine derivative and
As eq 10 shows, treatment of 1 with ethyl vinyl ketone at 23
°C afforded the 3-pentenoxy hydride (tBu3SiO)3HTa(OCH(Et)-
CHdCH2) (3-OCH(Et)CHdCH2). Interestingly, at elevated
C7D8
(tBu3SiO)3TaH2
+ H2CdCHCOEt 70 °C8
1
(tBu3SiO)3HTa(OCH(Et)CHdCH2)
(10)
3-OCH(Et)CHdCH2
temperatures (70 °C), 3-OCH(Et)CHdCH2 underwent rear-
rangement to an apparent oxytantallacyclopentane, (tBu3SiO)3-
TaOCHEtCH2CH2 (5-Et), via insertion of the olefinic residue
into the remaining hydride (eq 11). While the aggregation state
of 5-Et is not known for certain, only one isomer was noted,
and its solubility suggested a monomeric formulation.

4736 J. Am. Chem. Soc., Vol. 123, No. 20, 2001
Strazisar and Wolczanski
C7D8
Scheme 7
3-OCH(R)CHdCH2
R ) H, Et
8 (tBu3SiO)3TaOCHRCH2CH2
70 °C
R ) H, 5-H; Et, 5-Et (11)
When (tBu3SiO)3TaH2 (1) reacted with methyl acrylate, the
methoxy hydride (tBu3SiO)3HTaOMe (3-OMe) and allyloxy
hydride (tBu3SiO)3HTaOCH2CHdCH2 (3-OCH2CHdCH2) were
formed in a 1:1 ratio that was constant with time. Scheme 7
illustrates a mechanism based on initial insertion of the ester
carbonyl followed by â-OMe-elimination to afford acrolein,
which is rapidly scavenged by 1 to produce 3-OCH2CHdCH2.
When heated to 70 °C, 3-OCH2CHdCH2 also cyclometalated
according to eq 11. When tert-butyl acrylate was treated with
1, no reaction took place, indicating that steric protection of
C7D8
t
(tBu3SiO)3TaH2
+ H2CdCHCO2 Bu 23 °C8 no reaction (12)
1.9(3) × 10-3 M-1 s-1 (7 °C) appears to support this contention,
and comparisons of the individual H2CdCHOR runs suggest
that steric effects are less influential on insertion than on â-OR-
elimination; i.e., the “rise times” of the intermediates (tBu3-
SiO)3HTaCH2CH2OR (2-CH2CH2OR) are relatively similar.
In comparing the vinyl ethers and vinyl halides, the latter
are decidedly slower, rendering the overall insertion event rate-
determining. If the H2CdCHOiPr insertion rate at 7 °C is
estimated to be ∼6 × 10-2 M-1 s-1 at 60 °C,44 its insertions
probably one of the slowest of the vinyl ethers given the bulk
of the isopropyl groupsis predicted to be at least a factor of 20
faster than that of vinyl fluoride and ∼400 times faster than
1
the ester functionality can be accomplished. Since insertion of
the vinyl group did not occur, steric inhibition might also be
claimed, but recall that H2CdCHOtBu inserted rapidly into 1,
and it is unlikely that H2CdCHCO2 Bu is significantly larger.
The ester group CO2R is a powerful electron-withdrawing group,
which may disfavor olefin binding or hydride transfer.
In principle, the reduction of carbonyl functionalities by
(tBu3SiO)3TaH2 (1) can be used to synthesize (tBu3SiO)3-
HTaOCHR1R2 (3-OR) derivatives, but in practice the greater
electrophilicity of 3-OR relative to that of 1 often offsets its
increased steric hindrance, and dialkoxides are formed. For
example, treatment of 1 with 1 equiv of acetaldehyde resulted
in incomplete reaction and a mixture of (tBu3SiO)3HTaOEt (3-
OEt) and (tBu3SiO)3Ta(OEt)2 (6-(OEt)2) (eq 13, n ≈ 0.6). A
t
that of vinyl chloride. From the standpoint of Hammett σ+
F,X
values (Table 5),45,46 which are arguably a relevant measure of
subtituent electron-donating or -withdrawing capacity for the
sp2-vinyl substrates, the strongly electron-donating OR substit-
uents encourage insertion to a greater degree than the halides,
which possess intermediate values, and it becomes clearer that
C7D8
23 °C8
(n + 1)/2(tBu3SiO)3TaH2
R1R2CO
+
t
the inability of H2CdCHCO2 Bu to insert may have a significant
1
R1 ) H, R2 ) Me, Et
R1 ) R2 ) Me
electronic component, as reasoned above.
Among the halides, the X ) F > Br > Cl trend is consistent
with a decrease in rate of insertion as the substituent is more
electron-withdrawing and less able to π-donate electron density,
except it is somewhat unusual to find Br faster than Cl. If it
can be assumed that these halides operate more in an inductive
capacity (cf. σI(F) ) 0.50) in this instance, Cl is known to be
slightly more inductively withdrawing (σI ) 0.47) than Br (σI
) 0.45), and a rationale is in place.45,46 To a greater degree
than the other halides, a fluoride substituent prefers to be
attached to an sp3 carbon rather than to a more electronegative
sp2 carbon;47 hence, this factor may also contribute to the overall
insertion event.
Referring to path A of Scheme 1, it may be appropriate to
view insertion of H2CdCHX into the tantalum hydride as
occurring in two steps: olefin binding and the actual insertion
event. Bercaw’s seminal studies on olefin insertion of Cp2/Cp*2-
MH(H2CdCHR) (M ) Nb, Ta; R ) alkyl, C6H4-p-Y) clearly
show that in the transition state, a slight positive charge develops
on Câ, the carbon to which the hydride is migrating.48 Stabiliza-
tion of a similar developing positive charge on Câ in the related
n(tBu3SiO)3HTaOCHR1R2
3-OEt, 3-OnPr, 3-OiPr
+
(1-n)/2(tBu3SiO)3Ta(OCHR1R2)2
6-(OEt)2, 6-(OnPr)2, 6-(OiPr)2
(13)
similar mixture of (tBu3SiO)3HTaOnPr (3-OnPr) and (tBu3SiO)3-
Ta(OnPr)2 (6-(OnPr)2) was obtained from propanaldehyde (eq
13, n ≈ 0.6), but if the ketone or aldehyde is large enough,
some stoichiometric control is attained. Multiple insertion was
not noticed with ethyl vinyl ketone or methyl acrylate, and with
acetone 1 can be converted principally to (tBu3SiO)3Ta(OiPr)2
(6-(OiPr)2) on a preparative scale (eq 13, n < 0.01).
Conclusions
Rates of Olefin Insertion. From the observations and direct
measurements delineated above, the rates of H2CdCHX inser-
tion into the hydride bond of (tBu3SiO)3TaH2 (1) follow the
trend X ) H ≈ Me ≈ OR . halide. It is difficult to ascertain
whether ethylene inserts faster than the vinyl ethers, given the
excess of H2CdCHOR utilized in those experiments, the very
minor amounts of (tBu3SiO)3HTaEt (4-Et) formed, and the
corresponding inability to obtain a reliable value for k3 ((1-5)
× 10-3 M-1 s-1) in Scheme 2. From the range of k1 values
generated from the kinetics fits (k1 ≈ (1-5) × 10-3 M-1 s-1),
it would appear that the faster vinyl ethers are competitive with
ethylene, which is usually taken to be swifter than propylene.
The direct measurement of the insertion of H2CdCHOiPr of
(44) The rate was estimated by assuming a ∆Sq of -30 eu for the
insertion, calculating ∆Hq ) 11.5 kcal/mol at 7 °C from the ∆Gq of 19.9
kcal/mol, and recalculating ∆Gq at 60 °C to be 21.4 kcal/mol.
(45) Johnson, C. D. The Hammett Equation; Cambridge University
Press: Cambridge, England, 1973.
(46) Ritchie, C. D.; Sager, W. F. Prog. Phys. Org. Chem. 1964, 2, 323.
(47) Hine, J.; Mahone, L. G.; Liotta, C. L. J. Am. Chem. Soc. 1967, 89,
5911-5920.
(48) (a) Doherty, N, M.; Bercaw, J. E. J. Am. Chem. Soc. 1985, 107,
2670-2682. (b) Burger, B. J.; Santarsiero, B. D.; Trimmer, M. S.; Bercaw,
J. E. J. Am. Chem. Soc. 1988, 110, 3134-3146.

Insertion of H2CdCHX into (tBu3SiO)3TaH2
J. Am. Chem. Soc., Vol. 123, No. 20, 2001 4737
Figure 3. Views of H2CdCHX binding and the transition state for
insertion.
transition state for H2CdCHX insertion into (tBu3SiO)3TaH2
(1) may be operative, as illustrated in Figure 3.
Alternatively, olefin binding may be the critical step, and an
extreme zwitterionic view of olefin binding, in which a formal
negative charge is placed on tantalum and a positive charge is
given to the X substituent, is also shown in Figure 3. The frontier
orbital energies of H2CdCHX lend credence to olefin binding,49
because electron-donating groups raise the HOMO of the olefin
for a better energy match with the d0 tantalum center, and the
coefficient on the R-carbon increases to further desymmetrize
the olefin in preparation for the insertion event. The situation
herein differs significantly from the Bercaw study in that a d0
metal center is involved, and its accompanying electrophilicity
can greatly polarize the bound olefin and heighten the impor-
tance of olefin binding energetics. Electron-donating groups also
raise the LUMO, rendering hydride transfer less favorable, but
this is counteracted by an increase of the coefficient on the
â-carbon, which would facilitate H- transfer; hence, frontier
orbital arguments are more ambiguous in predicting the energet-
ics of the insertion transition state. These same rationalizations
have been applied to hydroboration, where electron-donating
substituents facilitate nucleophilic attack of an alkene on a
borane in insertion events that have clear parallels with those
of early transition metal hydrides.50
In summary, the trend toward swifter insertions for olefins
with electron-donating substituents may reflect enhanced bind-
ing, i.e., stabilization of an intermediate (tBu3SiO)3H2Ta(η-
H2CdCHX), or swifter insertion, i.e., stabilization of the
transition state for hydride migration to the olefin. A slight
preference for the former rationale may be inferred from frontier
orbital arguments. A concerted, one-step mechanism without a
formal intermediate binding event could be satisfied through a
combination of the above rationales.
Figure 4. Probable reaction coordinates (RC) for â-X-elimination (X
) H, -OR, or F, Cl, and Br) from (tBu3SiO)3HTaCH2CH2X (2-CH2-
CH2X); asterisks indicate values used from X-ray structural study of
(tBu3SiO)3HTaCH2CH2OtBu (2-CH2CH2OtBu).
the value measured for â-OiPr-elimination from (tBu3SiO)3-
HTaCH2CH2OiPr (2-CH2CH2OiPr).51 As such, â-F-elimination
is predicted to be ∼100 times faster than insertion under standard
state conditions, consistent with the absence of 2-CH2CH2F as
an observable. From the available data, while â-X-eliminations
(X ) halide) may be safely taken as g100 times their respective
H2CdCHX insertions, no actual comparison with â-OR-
eliminations can be made. At 60 °C, 2-CH2CH2OiPr eliminates
ethylene at a rate similar to the halide cases, but the steric factor
of the isopropyl group renders this comparison an oddity.
It is conceivable that â-X-elimination may be significantly
faster than â-OR-elimination, because of the lack of significant
steric factors and as a consequence of geometric factors. Figure
4 illustrates simplified reaction coordinates for transfer of a â-H,
-OR or F, Cl, and Br that were generated using geometric
parameters gleaned from the X-ray structural determination of
(tBu3SiO)3HTaCH2CH2OtBu (2-CH2CH2OtBu). The reaction
coordinate, here shown as a direct line from the â-substituent
to the tantalum, is only marginally dependent on the nature of
the atom to be transferred. As a consequence, the distance the
â-group must traverse to become bonded to the tantalum is
substantially less as the covalent radii increase;52 hence, it is
likely that Cl and Br may be much swifter in â-X-elimination
than F, which is comparable to OMe when the steric effects of
the methyl are ignored. The reaction coordinate is effectively
compressed for the larger atoms in the region of interest.
The thermodynamics of â-X-elimination (path D in Scheme
1) can be estimated provided critical bond strengths are known.
The reaction involves the breaking of primary metal-carbon
and secondary carbon-X bonds, and the formation of metal-X
Rates of â-X-Elimination. As previously discussed, the rates
of â-OR-elimination shown in Table 4 are simply rationalized
on the basis of sterics, and the change in rate-determining step
prevents their direct comparison with â-X-eliminations. Quali-
tatively, the halide eliminations are fast relative to insertion,
since the putative intermediates, (tBu3SiO)3HTaCH2CH2X (2-
CH2CH2X; X ) F, Cl, Br), were not detected. For a rough
quantitative comparison, assume that the rate of â-F-elimination
is most closely approximated by â-OMe-elimination from
(tBu3SiO)3HTaCH2CH2OMe (2-CH2CH2OMe), which was es-
timated at 60 °C to be ∼0.2 s-1 by assuming a ∆Sq of -7 eu,
and a carbon-carbon π-bond (∼65 kcal/mol), thus ∆H °rxn
)
{D(LnM-CH2R) + D(RHPC-X)} - {D(LnM-X) + D(π-
(51) The rate was estimated by assuming a ∆Sq of -7 eu for â-OMe-
elimination, calculating ∆Hq ) 18.3 kcal/mol at 7 °C from the ∆Gq of
20.2 kcal/mol, and recalculating ∆Gq at 60 °C to be 20.6 kcal/mol.
(52) Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell
University Press: Ithaca, New York, 1960.
(49) Houk, K. N. J. Am. Chem. Soc. 1973, 95, 4092-4094.
(50) Smith, K.; Pelter, A. In ComprehensiVe Organic Chemistry, Vol.
8.; Trost, B. M., Fleming, I., Eds.; Pergamon Press: New York, 1991; pp
703-729.

4738 J. Am. Chem. Soc., Vol. 123, No. 20, 2001
Strazisar and Wolczanski
(CC))}. It is tempting to conclude that greater exothermicity
would be accorded X ) F > Cl > Br on the basis of stronger
D(M-X), but the differences in metal-halide bond strengths
are quite substantial (based on CH3X: D(C-F) ≈ 110, D(C-
Cl) ≈ 84, D(C-Br) ≈ 70 kcal/mol)53 and may compensate.
The â-OR cases may manifest greater exothermicity because
the D(C-O) of ∼85 kcal/mol may be small compared to that
of an early metal-alkoxide bond. Aside from having confidence
in the exothermicity of â-X-elimination where X ) halide or
OR, and the endothermicity of â-H/R-elimination,37,38 better
calculational and experimental input is needed to increase our
understanding of the thermochemistry.
e10-6-10-9 M-1, if the values ascribed to the late transition
metals translate to early metal systems.
The speed of â-X-elimination relative to that of H2CdCHX
insertion appears to rule out formation vinyl halide homopoly-
mers and copolymers with substantial vinyl halide incorporation,
but does it also eliminate the possibility of some minimal
incorporation? To circumvent the â-X-elimination problem, it
is necessary to expediently move the substituent from the
â-position, which occurs when ethylene rapidly inserts following
a vinyl halide insertion. If 2 × 10-3 M-1 s-1 is used for the
ethylene insertion rate constant, it is only ∼30 (F)-450 (Cl)
times faster than the insertion of the halides, rendering
incorporation of vinyl halide infeasible unless the polymeriza-
tions are conducted at very high pressures. Polymerizations at
low temperature will also tend to enhance the competitiveness
of the more entropically disfavored second-order insertion
processes relative to the first-order â-X-eliminations.
Relevance to Ziegler-Natta Polymerization of Polar
Comonomers. 1. General. Before any comment is given on
the relevance of this model system to true Ziegler-Natta
polymerizations, a cautionary statement about the assumptions
inherent to such comparisons is necessary. The discussion below
will assume that all of the events scrutinized in the models
insertions and â-X/OR-eliminationssscale to the active cata-
lysts. This cannot be rigorously correct because all events in
Ziegler-Natta systems are occurring at rates much higher than
those observed herein, and as barriers for individual steps
become smaller and smaller, rate differences will naturally
decline as well. A second caveat concerns the model system
itself. Features such as its charge neutrality, the nature of group
5 compounds in comparison to those of group 4 that usually
constitute the class of Ziegler-Natta early metal catalysts, and
the lack of observed olefin insertions into metal-carbon bonds
may render the relevance of the model questionable. In defense
of this system, past chemistry clearly supports its characteriza-
tion as an electrophilic metal center,54 and one that exhibits
reactivity commensurate with many group 4 species, both
charged and neutral.
3. Vinyl Ethers. Vinyl ethers possess a distinct advantage
over the vinyl halides in that steric factors can be used to shut
down the â-OR-elimination process. Since the rates of insertion
of ethylene and H2CdCHOR (R ) small) are relatively similar
according to the minimal information gleaned from the fits to
Scheme 2, steric control of â-OR-elimination should provide
t
the means of generating copolymers where R ) Bu or larger.
i
Even the case of R ) Pr can be tuned, since kOR′/kâOR ≈ 104
M-1 at 7 °C. Assuming the insertion into a metal-carbon bond
is ∼104-107 slower,55 the true ratio of insertion to â-OR-
elimination is kOR/kâOR ≈ 1-10-3 M-1, and under the most
optimistic circumstances with high pressures of ethylene,
isopropyl vinyl ether should be incorporated. For smaller R,
the small discrepancy between insertion and â-OR-elimination
renders alkyl vinyl ether incorporation approaching that of vinyl
fluoride difficult.
Unfortunately, an added difficulty arises when vinyl ethers
are considered as comonomers. With sufficiently electrophilic
metal centers looms the prospect of competing cationic polym-
erization mechanisms that may obviate any Ziegler-Natta
pathways.56 Such polymerization mechanisms have been ob-
served with cationic group 4 catalysts, and designing catalysts
and curtailing impurities to minimize interference from cationic
pathways is difficult.
4. Summary. It is unlikely that vinyl halides may be
incorporated into polyethylene, even under conditions of
maximum [H2CdCHX] (X ) halide) and high pressures of
ethylene, which in principle can partially offset the propensity
of â-X-elimination from the growing polymer chain. The
situation is much better for vinyl ethers, where steric factors
can be introduced to inhibit â-OR-elimination and copolymer-
ization can be effected. Cases of H2CdCHOR where R is small
approach those of vinyl halides, and minimal incorporation of
the vinyl ether is the best hope, again under conditions of high
concentrations of vinyl ether and high pressures of ethylene.
Other types of H2CdCHX are equally difficult to copolymerize
when X is an electron-withdrawing functionality, because rates
of olefin binding and/or insertion are intrinsically low. When
reactive functionalities such as carbonyls are present, steric
protection or masking approaches must be used to guard against
problematic side reactions.
2. Vinyl Halides. The data and observations above do not
bode well for the copolymerization of ethylene and vinyl ha-
lides. In a growing polyethylene chain (assuming the simple
e
mechanisms of Scheme 1), the rate of propagation is RP
)
)
ke[LnMP+][CH2CH2], and the rate of chain transfer is RCT
e
kâH[LnMP+], providing a rough measure of average chain length
e
e
as RP /RCT ≈ (ke/kâH)[CH2CH2]. In a corresponding chain of
poly(vinyl halide), the average chain length is approximately
RPX/RCT ≈ (kX/kâX)[CH2CHX]. For the halides, since kX′/kâX
X
e 0.01 M-1, a maximum concentration of vinyl halide would
be needed to offset the intrinsic rate problem presented by the
large rate discrepancy favoring â-X-elimination. Assuming the
reactions could be carried out in neat vinyl halides, the ∼14 M
(X ) Br) to ∼18 M (X ) F) concentrations would not be
enough to offset the more rapid â-X-elimination rate. The
problem is compounded by an additional major factor. The rates
(kX′) examined herein refer to olefins inserting into hydrides,
whereas the true kX must describe a more energetically difficult
insertion into a metal-carbon bond. Brookhart’s studies of
ethylene insertion into [Cp*{(MeO)3P}MR]+ (M ) Co, Rh; R
) H, polyethylene chain) revealed a 6-10 kcal/mol greater
barrier for ethylene insertion into an alkyl, depending on the
metal,55 which roughly translates into a ∼104-107 rate ratio at
60 °C. As a consequence, the true kX/kâX rate discrepancy is
It is doubtful that any conditions can be found to permit the
highly desired copolymerization of ethylene and vinyl halides
using standard Ziegler-Natta catalysts. Changes that incorporate
a significantly higher barrier for â-X-elimination are necessary
for the synthesis of true copolymers, and it is unlikely that such
(53) Data taken from the NIST database, available at http://web-
book.NIST.gov.
(54) Wolczanski, P. T. Polyhedron 1995, 14, 3335-3362 and references
therein.
(55) (a) Brookhart, M.; Hauptman, E.; Lincoln, D. M. J. Am. Chem. Soc.
1992, 114, 10394-10401. (b) Brookhart, M.; Volpe, A. F.; Lincoln, D.
M.; Horva´th, I. T.; Millar, J. M. J. Am. Chem. Soc. 1990, 112, 5634-
5636.
(56) Baird, M. C. Chem. ReV. 2000, 100, 1471-1478.

Insertion of H2CdCHX into (tBu3SiO)3TaH2
J. Am. Chem. Soc., Vol. 123, No. 20, 2001 4739
3. (tBu3SiO)3HTanPr (4-nPr). A 10-mL round-bottom flask con-
nected to a calibrated gas bulb was charged with (tBu3SiO)3TaH2 (1,
161 mg, 0.195 mmol). The apparatus was evacuated, and 5 mL of
pentanes was added via vacuum transfer at -78 °C. Propylene (0.390
mmol) was added to the flask through the gas bulb at -196 °C. The
mixture was allowed to warm slowly to 23 °C. The pale yellow solution
was stirred for 12 h, after which the volatiles were removed in vacuo
to yield a white powder. Anal. Calcd for C39H89O3Si3Ta: C, 53.76; H,
10.30. Found: C, 54.00; H, 10.40.
4. (tBu3SiO)3HTaOEt (3-OEt) and (tBu3SiO)3Ta(OEt)2 (6-(OEt)2).
A 10-mL round-bottom flask connected to a calibrated gas bulb was
charged with (tBu3SiO)3TaH2 (1, 100 mg, 0.121 mmol). The apparatus
was evacuated, and 5 mL of hexanes was added via vacuum transfer
at -78 °C. Acetaldehyde (0.182 mmol) was added to the flask through
the gas bulb at -196 °C. The mixture was allowed to warm slowly to
23 °C. The pale yellow solution was stirred for 12 h, and the volatiles
were removed in vacuo to yield an off-white powder.
a requirement can be met with conventional catalysts. A different
mechanistic approach is mandated by this problem. For example,
consider a system where the 2,1-insertion of substituted olefins
was greatly favored over 1,2-insertion. A halide or alkoxy
functionality in an R-position on a growing polymer chain may
be kinetically stabilized relative to a â-X group, thereby
permitting further insertions. Popular among current approaches
is a switch to later metals where the thermodynamic impetus
for â-X-elimination is perhaps mitigated.19-21 Given the limita-
tions evident in this study, and the necessity of using model
complexes to obtain rates slow enough to observe, a calcula-
tional investigation of the â-X-elimination problem should prove
incisive.
Experimental Section
5. (tBu3SiO)3HTaOnPr (3-OnPr) and (tBu3SiO)3Ta(OnPr)2 (6-
(OnPr)2). A 10-mL round-bottom flask connected to a calibrated gas
bulb was charged with (tBu3SiO)3TaH2 (1, 100 mg, 0.121 mmol). The
apparatus was evacuated, and 5 mL of hexanes was added via vacuum
transfer at -78 °C. Propionaldehyde (0.182 mmol) was added to the
flask through the gas bulb at -196 °C. The mixture was allowed to
warm slowly to 23 °C. The pale yellow solution was stirred for 12 h,
after which the volatiles were removed in vacuo to yield an off-white
powder.
NMR Tube Reactions. General. An NMR tube attached to a 14/
20 ground glass joint was flame-dried and charged with (tBu3SiO)3TaH2
(1, typical ∼35 mg, 0.04 mmol) in a dry box. A flame-dried calibrated
gas bulb was attached to the adapter, and the assembly was degassed.
Toluene-d8 was vacuum-transferred into the tube at -196 °C. Reagent
was condensed into the tube at -196 °C through the gas bulb. The
tube was sealed with a torch under dynamic vacuum.
6. (tBu3SiO)3HTaOCH(Et)CHdCH2). An NMR tube was charged
with (tBu3SiO)3TaH2 (1, 32 mg, 0.03858 mmol) in a dry box. A 90.6-
mL gas bulb was attached to the adapter and the assembly degassed.
Toluene-d8 was vacuum-transferred into the tube at -196 °C. Ethyl
vinyl ketone (50.7 Torr, 0.250 mmol) was condensed into the tube at
-196 °C. The tube was sealed with a torch under dynamic vacuum.
The tube was warmed to 23 °C, and the conversion to product was
immediate.
General Considerations. All manipulations of air-sensitive materials
were performed using either glovebox or high-vacuum techniques.
Hydrocarbon and ethereal solvents were refluxed over sodium and
vacuum-transferred from sodium benzophenone ketyl. Several milliliters
of tetraglyme per liter was added to hydrocarbon solvents for solubility
purposes. Halogenated solvents were stirred over CaCl2 and then
distilled onto CaCl2. Acetone was refluxed over sodium and distilled
under Ar onto 4-Å molecular sieves. Acetaldehyde was distilled under
Ar. â-Bromostyrene was stirred over sodium and vacuum distilled
onto 4-Å molecular sieves. Toluene-d8 was sequentially dried over
sodium and 4-Å molecular sieves. All glassware was either baked in
an oven for a minimum of 3 h or flame-dried under dynamic vacuum.
(tBu3SiO)3TaH2 (1) and (tBu3SiO)3TaD2 (1-d2) were prepared according
to the literature procedure.31 cis- and trans-1,2-dideuterioethyl vinyl
ethers were obtained via the literature procedure32 from ethyl ethynyl
ether in hexanes purchased from Aldrich. Pure ethyl ethynyl ether and
ethyl ethynyl ether-d were recovered by first generating the lithium
salt with LiHNMe. The salt was filtered from hexanes and then
quenched with H2O or D2O. Isopropyl vinyl ether,57 divinyl ether,58
and phenyl vinyl ether59 were prepared via literature procedures. Methyl
vinyl ether and vinyl fluoride were purchased from Lancaster. Vinyl
chloride was purchased from Fluka. All other organic reagents were
bought from Aldrich.
1H and 13C{1H} NMR spectra were obtained using Varian XL-200,
XL-400, INOVA-400, and Unity-500 spectrometers. Chemical shifts
were reported relative to toluene-d7 (1H, δ 2.09) and toluene-d8 (13C-
{1H}, δ 20.4). Infrared spectra were recorded on a Nicolet Impact 410
spectrophotometer interfaced to a Gateway PC. Combustion analyses
were performed by Robertson Microlit Laboratories in Madison, NJ.
Procedures. 1. (tBu3SiO)3HTaCH2CH2OtBu (2-CH2CH2OtBu). A
25-mL round-bottom flask connected to a calibrated gas bulb was
charged with (tBu3SiO)3TaH2 (1, 200 mg, 0.241 mmol). The apparatus
was evacuated, and 5 mL of hexanes was added via vacuum transfer
at -78 °C. tert-Butyl vinyl ether (0.362 mmol) was condensed into
the flask through the gas bulb at -196 °C. The mixture was allowed
to warm slowly to 23 °C. The pale yellow solution was stirred for 6 h,
and the volatiles were removed in vacuo to yield an off-white powder.
X-ray-quality crystals were grown by slow cooling to -35 °C of a
concentrated hexanes solution. Anal. Calcd for C42H95O4Si3Ta: C,
54.27; H, 10.32. Found: 55.67; H, 10.75.
7. (tBu3SiO)3TaOCH(Et)CH2CH2 (5-Et). The sealed NMR tube
containing a solution of 3-OC(Et)HCHdCH2 and residual ethyl vinyl
ketone in toluene-d8 was heated at 70 °C for 12 h to generate 5-Et.
8. (tBu3SiO)3HTaOCH2CHdCH2 (3-OCH2CHdCH2) and
(tBu3SiO)3HTaOMe (3-OMe). An NMR tube was charged with
(tBu3SiO)3TaH2 (1, 35 mg, 0.0422 mmol) in a dry box. A 90.6-mL gas
bulb was attached to the adapter and the assembly degassed. Toluene-
d8 was vacuum-transferred into the tube at -196 °C. Methyl acrylate
(44.6 Torr, 0.220 mmol) was condensed into the tube at -196 °C. The
tube was sealed with a torch under dynamic vacuum. After the reaction
mixture warmed to 23 °C, the product was immediately observed, along
with an equimolar amount of 3-OMe.
9. (tBu3SiO)3TaOCH2CH2CH2 (5-H). The sealed NMR tube
containing a toluene-d8 solution of 3-OCH2CHdCH2 (along with
3-OMe and residual divinyl ether) was heated at 70 °C for 12 h to
form 5-H.
2. (tBu3SiO)3HTaOiPr (3-OiPr). A 10-mL round-bottom flask
connected to a calibrated gas bulb was charged with (tBu3SiO)3TaH2
(1, 100 mg, 0.121 mmol). The apparatus was evacuated, and 5 mL of
hexanes was added via vacuum transfer at -78 °C. Acetone (0.182
mmol) was condensed into the flask through the gas bulb at -196 °C.
The mixture was allowed to warm slowly to 23 °C. The pale yellow
solution was stirred for 12 h, and the volatiles were removed in vacuo
to yield an off-white powder. Anal. Calcd for C39H89O4Si3Ta: C, 52.78;
H, 10.13. Found: C, 51.10; H, 9.59.
General Kinetics Procedures. For the vinyl ethers, a measured
amount (typically ∼35 mg, ∼0.04 mmol) of (tBu3SiO)3TaH2 (1) was
added to a flame-dried 5-mm NMR tube attached to a 14/20 joint. The
adapted NMR tube was connected to a calibrated gas bulb, and 1 mL
of toluene-d8 was vacuum-transferred into the evacuated assembly.
Substrate was then condensed into the solution via the gas bulb at -196
°C, and the tube was sealed with a torch under active vacuum. For the
vinyl halides, a stock solution of 1 in toluene-d8 was prepared in a
10-mL volumetric flask in a dry box. An aliquot of this solution
(typically ∼0.60 mL, 0.03 mmol) was added to an NMR tube attached
to a 14/20 ground glass joint and a calibrated gas bulb. The assembly
was subjected to three freeze-pump-thaw degas cycles. The vinyl
reagent was submitted to the NMR tube through the gas bulb at -196
(57) Watanabe, W. H.; Conlon, L. E. J. Am. Chem. Soc. 1957, 79, 12828-
2833.
(58) Cretcher, L. H.; Koch, J. A.; Pittenger, W. H. J. Am. Chem. Soc.
1925, 47, 1173-1177.
(59) McClelland, R. A. Can. J. Chem. 1977, 55, 548-551.

4740 J. Am. Chem. Soc., Vol. 123, No. 20, 2001
Strazisar and Wolczanski
°C. The tube was sealed under dynamic vacuum. The rate at which the
concentration of a species changed with time was typically monitored
by measuring the area under its hydride resonance. Pre-acquisiton delays
of increasing time intervals were utilized to observe the completion of
each reaction in the NMR spectrometer probe. Sixteen transients were
taken each time a spectrum was acquired. The relaxation times (T1) of
hydride resonances were determined to be ∼1 s by inversion recovery
following calibration of a 90° pulse width for the signal. D1 was set to
6T1 ) 6 s. The temperature at which the reaction was monitored was
regulated by the NMR probe. An exact measurement of the reaction
temperature was obtained by running an internal calibration with a tube
of ethylene glycol. All runs were taken out to a minimum of four half-
lives. â-OR-elimination rates were obtained from nonlinear least-squares
fitting to the differential forms of the rate expressions. Least-squares
fitting was performed with a modified Powell algorithm to minimize
the sum-of-squares deviations between observed data and model
calculations. Multiple simulations were performed to the minimize 95%
confidence range.60 Vinyl halide insertion rates were obtained by
plotting the integrated rate expression for disappearance of 1 under
pseudo-first-order conditions to obtain a straight line; the slope of the
resulting line was directly related to the insertion rate constant.
Determination of Ethylene-d2 Stereochemistry. A sealed NMR
tube in which ethylene-d2 had been evolved was placed in a tube-cracker
connected to a 180° needle valve. This apparatus was evacuated. The
tube was cracked while the system was closed to active vacuum. After
5 min, the assembly was cooled to -78 °C. After 15 min, the volatiles
were condensed into an evacuated round-bottom flask at -196 °C. This
flask was warmed to 23 °C, and its contents were allowed to expand
into an evacuated IR cell. After 10 min, the IR cell was sealed and
removed to the IR spectrophotometer. Absorbance measurements were
taken after 16 scans at a resolution of 1 cm-1. trans-1,2-Dideuterio-
X-ray Structure Determination of (tBu3SiO)3HTaCH2CH2OtBu
(2-CH2CH2OtBu). A few milligrams of crystals (vide supra) were
suspended in Paratone oil on a glass slide. Under a microscope, a single
colorless block (0.20 × 0.10 × 0.040 mm3) was picked up in a 100-
µm rayon fiber loop epoxied to a metal fiber. X-ray diffraction data
were collected on a Siemens SMART system (λ ) 0.71073) employing
a 1 K CCD detector and an Oxford Cryostream set at 165 K. Observed
intensities were corrected for Lorentz and polarization effects, and a
semiempirical absorption correction was applied (SADABS). The
structure was solved by direct methods (SHELXTL). The hydride was
located and refined isotropically, while the remaining hydrogen atoms
were calculated and included in the model. The structure was refined
by full-matrix least-squares on F2 using isotropic thermal parameters
for all non-hydrogen atoms. The largest difference peak and hole were
0.988 and -1.045 e Å-3
.
Acknowledgment. We thank Emil B. Lobkovsky for ex-
perimental assistance, and Prof. Barry Carpenter for consultation.
We gratefully acknowledge contributions from the National
Science Foundation (CHE-9816134), The Dow Chemical Co.,
and Cornell University.
Supporting Information Available: Plots of the kinetics
fits of Schemes 2 and 5, and a summary of crystal data,
encompassing data collection and solution/refinement, atomic
coordinates, isotropic and anisotropic temperature factors,
hydrogen atom coordinates, bond lengths and bond angles for
(tBu3SiO)3HTaCH2CH2OtBu (2-CH2CH2OtBu) (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
ethylene has two strong out-of-plane vibrations at 987 and 725 cm-1
cis-1,2-Dideuterioethylene has one strong out-of-plane vibration at 842
cm-1 33
.
.
(60) MicroMathR SCIENTIST, Experimental Data Fitting; MicroMath,
Inc., 1986-1995.
JA003315N