Isocyanide insertion reactions into the Ta - C bonds of cationic and zwitterionic tantalocenes

Article abstract of DOI:10.1139/v03-051

The insertion of isonitriles R′NC (R′ = t-Bu, C ₆H₁₁, CH₂C₆H₅) occurs exclusively into the Ta - CH₂ bonds of the zwitterionic compounds (C₅H₄R)₂Ta[CH₂B(C₆F ₅)₃]CH₃ (R = H, 1a; R = CH₃, 1b) at relatively slow rates at room temperature to form N-out isomers as the sole kinetic products. By comparison, insertion of the same isonitrile substrates into a Ta - CH₃ bond of the non-zwitterionic analogs [(C ₅H₄R)₂Ta(CH₃)₂][A] (R = H, A = B(C₆F₅)₄, 3a; R = CH₃, A = BF₄, 3b) occurs much more rapidly, again to form N-out isomers exclusively under kinetic conditions. The difference in rate is attributed to the presence of a ground state α-agostic interaction in the zwitterionic compounds, which is not featured in the dimethyl ion pairs. All of the N-out isomers formed undergo thermal and irreversible conversion to the corresponding N-in isomers at rates that are conveniently followed by ¹H NMR spectroscopy. The rates were studied at different temperatures to obtain activation parameters for each transformation. The rate and activation trends for this isomerization were analyzed as a function of the structural changes in the compounds. It was found that the rate decreased as the steric bulk of the isonitrile substitutent R′ increased and that the rates were faster for the series that incorporated the more electron-donating C₅H ₄CH₃ ancillary ligand. Furthermore, isomerization rates for the zwitterionic N-out compounds were faster than those found in the non-zwitterionic series. This study represents one of the more extensive kinetic analyses of the rate of N-out to N-in isomerization as a function of structural changes. The observations are consistent with the mechanistic picture that has been developed for this process, involving dissociation of the η² iminoacyl ligand, rotation about the M -C_iminoacyl bond and recoordination to the inside site of the metallocene wedge.

Full text of DOI:10.1139/v03-051

1137
Isocyanide insertion reactions into the Ta—C
bonds of cationic and zwitterionic tantalocenes1
Kevin S. Cook, Warren E. Piers, Brian O. Patrick, and Robert McDonald
Abstract: The insertion of isonitriles R′NC (R′ = t-Bu, C6H11, CH2C6H5) occurs exclusively into the Ta—CH2 bonds of
the zwitterionic compounds (C5H4R)2Ta[CH2B(C6F5)3]CH3 (R = H, 1a; R = CH3, 1b) at relatively slow rates at room
temperature to form N-out isomers as the sole kinetic products. By comparison, insertion of the same isonitrile sub-
strates into a Ta—CH3 bond of the non-zwitterionic analogs [(C5H4R)2Ta(CH3)2][A] (R = H, A = B(C6F5)4, 3a; R =
CH3, A = BF4, 3b) occurs much more rapidly, again to form N-out isomers exclusively under kinetic conditions. The
difference in rate is attributed to the presence of a ground state α-agostic interaction in the zwitterionic compounds,
which is not featured in the dimethyl ion pairs. All of the N-out isomers formed undergo thermal and irreversible con-
1
version to the corresponding N-in isomers at rates that are conveniently followed by H NMR spectroscopy. The rates
were studied at different temperatures to obtain activation parameters for each transformation. The rate and activation
trends for this isomerization were analyzed as a function of the structural changes in the compounds. It was found that
the rate decreased as the steric bulk of the isonitrile substitutent R′ increased and that the rates were faster for the se-
ries that incorporated the more electron-donating C5H4CH3 ancillary ligand. Furthermore, isomerization rates for the
zwitterionic N-out compounds were faster than those found in the non-zwitterionic series. This study represents one of
the more extensive kinetic analyses of the rate of N-out to N-in isomerization as a function of structural changes. The
observations are consistent with the mechanistic picture that has been developed for this process, involving dissociation
of the η2 iminoacyl ligand, rotation about the M—Ciminoacyl bond and recoordination to the inside site of the
metallocene wedge.
Key words: cationic metallocenes, isocyanide insertion, agostic interactions, tantalum.
Résumé : À la température ambiante, l’insertion d’isonitriles, R′NC (R′ = t-Bu, C6H11, CH2C6H5) se produit à des vi-
tesses relativement faibles et uniquement dans les liaisons Ta—CH2 des composés zwitterioniques (C5H4R)2Ta[CH2B-
(C6F5)3]CH3 (R = H, 1a; R = CH3, 1b) pour conduire à la formation d’isomères N-out comme seuls produits cinéti-
ques. Par comparaison, l’insertion des mêmes substrats isonitriles dans une liaison Ta—CH3 des analogues non zwitte-
rioniques [(C5H4R)2Ta(CH3)2][A] (R = H, A = B(C6F5)4, 3a; R = CH3, A = BF4, 3a) se produit beaucoup plus
rapidement, mais dans des conditions cinétiques il ne se forme encore une fois que les isomères N-out. La différence
de vitesse est attribuée à la présence, dans l’état fondamental des composés zwitterioniques, d’une interaction α-agos-
tique qu’on n’existe pas dans les paires d’ions diméthylées. Tous les composés N-out qui se forment subissent une
conversion thermique et irréversible vers les isomères correspondants N-in, à des vitesses qui peuvent être facilement
1
suivies par spectroscopie RMN du H. On a étudié les vitesses à différentes températures pour déterminer les paramè-
tres d’activation de chacune des transformations. On a analysé les tendances dans les vitesses et les activations en fonc-
tion des changements structuraux dans les composés. On a trouvé que la vitesse diminue avec une augmentation de
l’encombrement stérique du substituant R′ des isonitriles et que les vitesses augmentent pour les séries qui incorporent
le ligand ancillaire électrodonneur C5H4CH3. De plus, les vitesses d’isomérisation des composés N-out zwitterioniques
sont plus rapides que celles des séries non zwitterioniques. Cette étude représente l’une des analyses cinétiques les plus
extensives de la vitesse d’isomérisation N-out en N-in en fonction de changements structuraux. Les observations sont
en accord avec les idées mécanistiques qui ont été développées pour ce processus et qui impliquent une dissociation du
ligand η2-iminacayle, une rotation autour de la liaison M—Ciminoacyle et une recoordination vers le site intérieur de
l’extrémité du métallocène.
Mots clés : métallocènes cationiques, insertion d’isocyanure, interactions agostiques, tantale. Cook et al. 1148
Received 14 January 2003. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 28 May 2003.
Dedicated to Prof. John Harrod for his outstanding contributions to organometallic chemistry.
Kevin S. Cook and Warren E. Piers.2 Department of Chemistry, University of Calgary, 2500 University Drive N. W., Calgary,
AB T2N 1N4, Canada.
Brian O. Patrick. Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, BC V6T 1Z1, Canada.
Robert McDonald. X-Ray Structure Laboratory, Department of Chemistry, University of Alberta, Edmonton, AB T6G 2G2,
Canada.
1This article is part of a Special Issue dedicated to Professor John Harrod.
2Corresponding author (e-mail: wpiers@ucalgary.ca).
Can. J. Chem. 81: 1137–1148 (2003)
doi: 10.1139/V03-051
© 2003 NRC Canada

1138
Can. J. Chem. Vol. 81, 2003
carborane (5), borollide (6), or imide (7) donors (which are
isoelectronic to, for example, Cp2ZrMe2) with various
isonitriles have also been studied and shown to undergo sim-
ilar chemistry. In one of these systems — the tantalum(V)
carborane system I shown in Chart 1 and reported by Finn,
Grimes, and co-workers (5) — the N-out isomers formed
with a variety of isocyanides were observed to be somewhat
more stable than those of the group 4 metallocenes, allowing
for the direct observation of the interconversion of these iso-
mers. The N-out structure thermally converts to the N-in
isomer, which can in turn be driven back to the N-out spe-
cies photochemically. While no kinetic data were reported,
both N-out and N-in isomers of one particular compound
were structurally characterized for direct comparison.
By virtue of their dianionic Cp mimics, these Ta(V)
dialkyls are similar to the group 4 dialkyls in that they are
neutral. Another family of Ta(V) dialkyls isoelectronic to
Cp2ZrMe2 are the cationic compounds [Cp2TaMe2]+[A]–,
formed readily by abstraction of methide anion from the
neutral trimethyl precursor Cp2TaMe3 (8). A subgroup of
this series are the zwitterionic analogs of these ion pairs,
prepared by the reaction of Schrock’s methylidene methyl
complex Cp2Ta(=CH2)CH3 with the highly electrophilic bor-
ane B(C6F5)3 (9). We found that the zwitterions react slowly
with tert-butyl isonitrile to form an N-out iminoacyl com-
plex, resulting from regioselective insertion into the Ta—
CH2B(C6F5)3 bond. Surprisingly, isonitrile insertions involv-
ing ion pairs [Cp2TaMe2]+[A]– have not been reported, so to
compare the characteristics of these reactions with those ob-
served for zwitterions, we have prepared examples of com-
pounds [Cp2TaMe2]+[A]– and evaluated their reactivity with
RNC (R = t-Bu, C6H11, and CH2C6H5). In all cases we find
the kinetic N-out isomers to be isolable and stable towards
isomerization under ambient conditions. All, however, un-
dergo clean conversion to their thermodynamic N-in isomers
under controlled conditions, allowing for determination of
the kinetic and thermodynamic parameters of this isomeri-
zation for a variety of related compounds.
Introduction
The migratory insertion of CO or CNR into metal—car-
bon bonds is a key C—C bond-forming reaction in
organometallic chemistry (1). Its mechanism has been inves-
tigated in several classic studies (1), and stoichiometric in-
sertions into the metal—carbon bonds of early transition
metal, bent metallocenes such as Cp2ZrMe2 have figured
prominently in this body of research (2). For these
electrophilic organometallic substrates, the resulting acyl or
iminoacyl products adopt an η2 bonding mode where an O
or N lone pair also contributes electron density to the Lewis
acidic metal center. Consequently, the η2 acyl or iminoacyl
complexes can exist as “O/N-out” or “O/N-in” isomers
(Chart 1), and the latter has been shown to be the thermody-
namically favored isomer, while the former is the kinetically
preferred product. Better orbital overlap between the acyl–
iminoacyl ligand frontier orbitals with the available
metallocene orbitals in the N-in isomer accounts for the
greater overall stability of this isomer (3).
Results and discussion
As shown in eq. [1], the methylidene methyl tantalocene
′
compounds Cp2Ta(=CH2)CH3 and Cp2Ta(=CH2)CH3 (Cp =
C5H5; Cp′ = C5H4Me) react rapidly with the perfluoroaryl
borane Lewis acids B(C6F5)3 (10) and H3CB(C6F5)2 (11) via
electrophilic attack on the nucleophilic carbene ligand.
[1]
The O/N-out to O/N-in isomerization is often quite rapid,
and the O/N-out isomers are for the most part only fleet-
ingly observed, making detailed study of this process diffi-
cult. However, based on both experimental (2) and
computational (3, 4) studies, it is accepted that the
isomerization probably takes place via dissociation of O or
N to form an η1 acyl–iminoacyl complex that can undergo
rotation about the M—Cacyl–iminoacyl bond. Much qualitative
experimental evidence in support of this picture exists, but
few studies have reported quantitative kinetic and thermody-
namic data as a function of systematic structural variation to
delineate the factors influencing this isomerization process.
While group 4 metal-based metallocenes have played a
crucial role in the development of this chemistry, recently
the reactions of tantalocene mimics containing dianionic
The products, as detailed in our earlier communication
(9), are the zwitterionic tantalum dialkyl compounds 1a–b
and 2a and are characterized by a strong ground state α
agostic interaction involving one of the C—H bonds of the
borate-substituted alkyl ligand. Strong evidence for this
© 2003 NRC Canada

Cook et al.
1139
interaction is found in the solid state structure of compound
pared in order to carry out the kinetic studies described be-
low. Unlike the zwitterionic compounds, neither 3a nor 3b
exhibits any evidence for α agostic interactions in the
ground state. No low-frequency IR C–H stretches are ob-
served in the IR spectra, and 1JC-H coupling constants for the
Ta-CH3 groups are normal, although admittedly perturbation
in this value would not be expected to be large given the six
C—H bonds over which the agostic interaction would be av-
eraged.
The zwitterions 1 react slowly with isocyanides to form
new N-out iminoacyl zwitterions via selective insertion of
the isocyanide reagent into the Ta—CH2 bond; no products
resulting from insertion into the Ta—CH3 bond are observed
(eq. [3]) (16).
1a, and it persists in solution, as evidenced by the low aver-
1
age JC-H values of 82–100 Hz found for the CH2 group. In
all three compounds, the anionic charge associated with the
borate moiety is largely localized on the methylene carbon,
rendering the C—H bonds quite electron rich and capable of
strong donation to the cationic metal center. In so doing, the
agostic interaction utilizes the last available metallocene or-
bital and perhaps stabilizes the complex with respect to a
neutralizing -C6F5 transfer from boron to tantalum, a process
that is apparently quite facile in related, titanium-based zwit-
terions formed upon elimination of methane from ion pairs
of the general formula [Cp(L)TiCH3]+[H3CB(C6F5)3]– (12).
Non-zwitterionic dialkyl tantalocene ion pairs 3a–b are
trivially synthesized via methide abstraction from Cp2TaMe3
′
and Cp2TaMe3 using trityl salts (eq. [2]).
[3]
[2]
In order for the reaction to proceed cleanly, an excess of
the isocyanide reagent is required; with only 1 equiv, the re-
actions are very slow. Under pseudo-first-order conditions,
the reactions are clean and conveniently followed by NMR
spectroscopy. The products 4a–b and 5a are assigned3 as the
η2 N-out iminoacyl isomers on the basis of 13C NMR spec-
This reaction has been known for some time and is a key
step in the synthesis of Cp2Ta(=CH2)CH3 (8). Note that the
counteranions in compounds 3 are different; in 3a, the
counteranion is the tetrakis-perfluoroarylborate [B(C6F5)4],
while for the C5H4Me-stabilized metallocene cation 3b, the
[BF4] counteranion is used. The reasons for this discrepancy
have to do with the solubility properties of the various ion
pairs and the requirements for the kinetic studies described
below. The C5H5–BF4 combination gave an ion pair that was
sparingly soluble even in donor solvents, while the com-
pound [(C5H4Me)2TaMe2][B(C6F5)4] was too soluble, even
in non-donor solvents, for us to isolate in purity sufficient to
carry out reliable kinetic studies. The C5H5–B(C6F5)4 and
C5H4Me–BF4 combinations, however, had ideal solubility
properties. Since the [Cp2TaMe2]+ cation has been shown to
be very weakly coordinating towards perfluoroaryl borate
counteranions (13), and since even the relatively more coor-
dinating triflate anion shows little interaction with the re-
lated [(C5Me5)CpTaMe2]+ cation (14), we feel that any
effects due to the change in anion can be largely ignored in
the interpretation of the data presented below. Nevertheless,
this is an assumption and must be acknowledged as such.
While compound 3b has been reported previously (8), 3a,
1
troscopy, H ROESY experiments, and the X-ray structure
for N-out 4b, which has been discussed in detail elsewhere
(9). In sharp contrast to the sluggish pace of the reactions of
eq. [3], reaction of ion pairs 3a and 3b with 1 equiv of RNC
at –78°C results in a rapid reaction and provides the
iminoacyl compounds 6a–b (t-BuNC), 7a–b (C6H11NC), and
8a–b (C6H5CH2NC) as one isomer (eq. [4]).
[4]
1
with the [B(C6F5)4]– counteranion, is a new ion pair. The H
NMR spectrum of the cation is identical to that reported for
the [BF4]– salt, lending credence to the above assumption
that the anion has little effect on the properties of the cation.
The 19F NMR data for the [B(C6F5)4]– anion is typical for
this species (15). Analytically pure samples of 3a were pre-
These reactions are essentially complete upon mixing and
are quantitative by H NMR spectroscopy, as indicated in
particular by a downfield shift of one of the Ta-CH3 groups
by about 2 ppm upon iminoacyl group formation. The 13C
NMR chemical shifts for the iminoacyl carbon all appear
1
3 Although IR spectroscopy is a valuable analytical tool for iminoacyl compounds because of the C=N chromophore, the stretching frequency
does not correlate well with the nature of the bonding, i.e. 01 vs. 02 or N-out vs. N-in. Therefore, since the iminoacyl compounds reported
herein have been well characterized by NMR spectroscopy and X-ray crystallography, we have not acquired IR data for the new com-
pounds.
© 2003 NRC Canada

1140
Can. J. Chem. Vol. 81, 2003
Fig. 1. ORTEP diagram of the cation in N-out 6a; the anion has been omitted for clarity. Selected bond distances (Å): Ta(1)—N(1),
2.143(2); Ta(1)—C(1), 2.147(3); Ta(1)—C(7), 2.267(3); C(1)—N(1), 1.259(4); C(1)—C(2), 1.472(4); N(1)—C(3), 1.500(4). Selected
bond angles (°): C(7)-Ta(1)-C(1), 78.7(1); C(1)-Ta(1)-N(1), 34.1(1); N(1)-Ta(1)-C(7), 112.9(1); C(2)-C(1)-N(1), 134.4(3); Ta(1)-C(1)-
C(2), 152.8(2).
within the narrow range of 210.1–211.7 ppm and are charac-
teristic of η2-iminoacyl ligands (2a).
electrophilicity of the metal center (2a). That is, the more
electrophilic the metal center, the stronger is the dative M—
N bond and the lower is the value observed for ∆. For N-out
6a and 7a, the values of ∆ are 0.004 Å and 0.000 Å, respec-
tively, which are similar to the value of –0.002 Å reported
for the carborane N-out I isomer of Chart 1 (5). Given the
cationic nature of the Ta centers in N-out 6a and 7a, one
might expect a larger difference in ∆ compared with the neu-
tral (and relatively electron-rich) dicarbollide complex. It
should be noted, however, that the values of ∆ for η2
iminoacyl complexes have been, for the most part, deter-
mined only for N-in isomers. Since the N-out isomers are
thermodynamically less stable than the N-in isomers, owing
to poorer overlap of the nitrogen lone pair with metallocene
orbitals in this geometry (3, 4), ∆ is likely less affected by
electronic changes at the metal center and more prone to
steric effects in the N-out isomer. Consistent with this
As in the reactions involving the zwitterions, the N-out
isomers were formed exclusively under these conditions and
were stable enough to isolate and fully characterize. The
X-ray structures of N-out 6a and N-out 7a were carried out
to demonstrate this conclusively; ORTEP diagrams of the
cations for these compounds are given in Figs. 1 and 2 along
with selected bond distances and angles. Full details on
these structure determinations are available in the supporting
information.4 Given the similarity in the 13C NMR data, the
other compounds were also assigned to the N-out geometry.
As can be seen in the figures, the nitrogen of the iminoacyl
ligand is coordinated to the Ta center on the outside of the
metallocene wedge in both cases. For transition metal imin-
oacyl complexes, the parameter ∆ is defined as d(M—N) –
d(M—C) and correlates at least qualitatively with the
4 Supplementary data may be purchased from the Depository of Unpublished Data, Document Delivery, CISTI, National Research Council
Canada, Ottawa, ON K1A 0S2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml for information on ordering electronically). CCDC
210457 (6a) and 210458 (7a) contain the supplementary data for this paper. These data can be obtained, free of charge, via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge, U.K.; fax
+44 1223 336033; or deposit@ccdc.cam.ac.uk).
© 2003 NRC Canada

Cook et al.
1141
Fig. 2. ORTEP diagram of the cation in N-out 7a; the anion has
been omitted for clarity. Selected bond distances (Å): Ta—N(1),
2.130(6); Ta—C(3), 2.130(7); Ta—C(1), 2.287(7); C(3)—N(1),
1.245(8); C(2)—C(3), 1.496(10); N(1)—C(4), 1.508(9). Selected
bond angles (°): C(1)-Ta-C(3), 79.0(3); C(3)-Ta-N(1), 34.0(12;
N(1)-Ta-C(1), 112.6(3); C(2)-C(3)-N(1), 132.0(7); Ta-C(3)-C(2),
154.7(6).
donor, allowing for rapid insertion chemistry. The lower
steric demands of the dimethyl compounds probably also
contribute to the dramatic rate difference observed.
[5]
Iminoacyl compounds N-out 4–8 are unusual in that they
are essentially immune to isomerization to N-in isomers un-
der ambient conditions. All of them do, however, undergo
clean conversion to the thermodynamic N-in isomers under
thermal duress (60–110°C, eq. [5]).
The isomerization is signaled in the 1H NMR spectrum by
a slight upfield shift in the cyclopentadienyl resonances and
a downfield shift in the Ta-CH3 and iminoacyl CHn protons
(n = 2 for 4 and 5; n = 3 for 6, 7, and 8). In addition, the 13
C
chemical shift for the iminoacyl carbon shifts downfield
8–9 ppm into the ranges of 221.7–223.6 ppm for the zwit-
terions and 217.7–220.2 ppm for compounds 6–8. Thus, the
two isomers are easily distinguishable on the basis of their
13C NMR spectra.
Baseline separation in the signals for the two isomers in
the 1H NMR spectra allows for convenient monitoring of the
isomerization and determination of first-order rate constants
for the various compounds in the study; this data, along with
the thermodynamic activation parameters obtained from
Eyring plots, is collected in Table 1. The reactions were all
found to be first order in the N-out isomer for greater than 3
half lives, and the rate constants given are an average of two
separately determined values.
notion is the larger value of ∆ of 0.088 Å observed for the
N-in isomer of I, which is more in line with expectations for
a neutral and relatively electron-rich d0 complex (17).
The reason for the marked difference in reaction rates for
the insertion of RNC, involving zwitterions 1 vs. ion pairs 3,
has its roots in the different ground-state structures for these
compounds. Kinetic studies on the reaction of t-BuNC with
1a under pseudo-first-order conditions pointed to a mecha-
nism for insertion that involves dissociation of the α agostic
interaction prior to isocyanide coordination and insertion (9).
Most telling was the substantial inverse kH/kD of 0.84(4) at
30°C, observed when comparing the rates of insertion for 1a
and d5-1a (prepared from Cp2Ta(=CD2)CD3 and B(C6F5)3).
In the deuterated compound, the equilibrium between agos-
tic and non-agostic structures will be perturbed towards the
latter, allowing for faster rates of isocyanide coordination
(and therefore insertion) in the labeled compound. In com-
parison to the ion pairs 3, the necessity for dissociation of
the α agostic C—H bond in zwitterions 1, with its itinerant
charge separation, slows the overall rate of the reaction con-
siderably. Since the compounds 3 do not exhibit this interac-
tion, they have an orbital available to accept the isocyanide
There are few studies measuring the activation parameters
for the N-out to N-in isomerization process and no studies
employing a wide range of derivatives. Bullock and co-
workers have investigated the isomerization of the zir-
conocene
system
prepared
through
reaction
of
Cp(PMe3)2RuCH=CHZrClCp2 with t-BuNC, the first system
studied that showed irreversible rearrangement to the N-in
isomer (18). Activation parameters reported for this
zirconocene system were ∆H‡ = 19.9(6) kcal mol–1, ∆S‡ =
–3.4(22) eu, with a ∆G‡298 = 20.9 kcal mol–1. Measurement
of the barrier to N-out/N-in interconversion in fluxional sys-
tems has also been reported. For example, Adams and
Chodosh (19) have investigated the molybdenum system
© 2003 NRC Canada

1142
Can. J. Chem. Vol. 81, 2003
Table 1. Rate constants and activation parameters for N-out to N-in isomerization of η2-iminoacyl complexes.
Compounda
T (K)
10–4 k (s–1)
∆H‡ (kcal mol–1)
∆S‡b (eu)
∆G‡298 (kcal mol–1)
N-out 4b
N-out 4b
N-out 4b
N-out 5a
N-out 5a
N-out 5a
N-out 5a
N-out 6a
N-out 6a
N-out 6a
N-out 6a
N-out 6b
N-out 6b
N-out 6b
N-out 7a
N-out 7a
N-out 7a
N-out 7b
N-out 7b
N-out 7b
N-out 7b
N-out 8a
N-out 8a
N-out 8a
N-out 8a
N-out 8b
N-out 8b
N-out 8b
N-out 8b
356
346
334
356
345
336
324
391
380
372
360
358
347
335
359
348
338
355
343
333
323
369
358
346
335
357
345
334
323
4.49(7)
1.30(1)
0.35(2)
16.9(4)
6.10(9)
2.09(2)
0.45(1)
5.4(1)
26.9(7)
1(2)
26.5(7)
25.5(3)
34.5(8)
0(1)
25.5(3)
30.2(8)
14(3)
1.80(2)
0.436(7)
0.121(3)
1.05(1)
0.320(9)
0.068(3)
2.15(4)
0.795(7)
0.205(3)
2.95(4)
1.75(2)
0.507(4)
0.147(3)
5.65(9)
1.91(3)
0.498(3)
0.15(6)
6.04(3)
1.70(2)
0.461(7)
0.108(6)
26.9(1)
26.8(9)
21(2)
–2.0(4)
0(3)
27.5(1)
27.1(9)
26(2)
–15(6)
26.2(4)
26.4(1)
–3(2)
27.0(4)
26.3(1)
0.3(4)
aReliable kinetic data for compound N-out 4a could not be obtained because of the long reaction times required to effect isomerization. Higher
temperatures also gave poor data owing to undesired side reactions.
bThe range observed in ∆S‡ (e.g., 6a and 7b) likely is a reflection of the narrow temperature ranges we were restricted to experimentally.
Cp(CO)2Mo[η2-C(=NMe)Me] and found activation parame-
ters of ∆H‡ = 14.7(5) kcal mol–1, ∆S‡ = –1.7(15) eu, and
∆G‡298 = 20.9 kcal mol–1. Rothwell and co-workers (20) have
and N-out 8a, exhibit isomerization rate constants that in-
crease as the alkyl group on the nitrogen atom changes from
t-Bu (k360 = 1.21 × 10–5 s–1) to C6H11 (k358 = 2.15 × 10–4 s–1),
which is similar to that found for the C6H5CH2 isomeri-
zation (k358 = 1.96 × 10–4 s–1). The ∆G‡ values parallel the
trend in rate constants, and a similar phenomenon is ob-
served for the C5H4Me cations. The origin of this effect
likely resides in steric interactions between the N-iminoacyl
substituent and the Cp rings upon rotation about the Ta—
Ciminoacyl bond in the η1 structure.
Comparison of the isomerization rates of the unsubstituted
cyclopentadienyl cations N-out 6a–8a with those of the
methylcyclopentadienyl compounds N-out 6b–8b provides
further insight. In each cation pair the isomerization of the
methylcyclopentadienyl derivative proceeds at an elevated
rate compared with the Cp derivative. For example, the t-
BuNC insertion product N-out 6a isomerizes with a rate
constant of k360 = 1.21 × 10–5 s–1, nearly an order of magni-
tude slower than N-out 6b (k358 = 1.05 × 10–4 s–1). The pres-
ence of more electron-donating cyclopentadienyl rings for
the C5H4Me derivatives facilitates the slippage of the
iminoacyl from η2 bonded to the η1 intermediate.
reported a Gibbs free energy of activation of ∆G2‡33
=
11.4(5) kcal mol–1 for the observed rotation in (ArO)2Zr[η2-
C(=N(2,6-Me2C6H3))CH2Ph]2 (Ar = 2,6-C6-t-Bu2H3). The
less bulky derivative (Ar = 2,6-C6Ph2H3) gave a result of
∆G‡208 = 10.3(5) kcal mol–1. Lappert et al. have also ob-
served the interconversion of the two isomers of Cp2Zr(Cl)-
[η2-C(=N(p-C6H4CH3))CH(SiMe3)2] (16b). While our values
for ∆S‡ are comparable to those found in other systems —
the magnitudes consistent with an intramolecular process, as
can be seen in Table 1 — the ∆H‡ and ∆G2‡98 values are gen-
erally substantially larger in the present cationic tantalocene
systems than these previously reported examples. This can
be attributed to the stronger metal—ligand bonds for third
row transition metal complexes, as well as the increased
electrophilicity of the cationic tantalum center. Assuming
the mechanism of isomerization involves slippage of the η2-
iminoacyl to an η1-bound species, a stronger tantalum—
nitrogen dative bond must be dissociated to form an η1-
iminoacyl ligand.
A trend is apparent in the rate of isomerization based on
the steric bulk of the iminoacyl substituent on the nitrogen
atom. The C5H5 cationic complexes, N-out 6a, N-out 7a,
Interestingly, a comparison of the rates of isomerization
between analogous zwitterions and ion pairs (N-out 4b vs.
N-out 6b and N-out 5a vs. N-out 7a) reveals that the zwit-
© 2003 NRC Canada

Cook et al.
1143
terions undergo more rapid isomerization at comparable
temperatures. This seems counterintuitive given that, in the
ground state of the N-out isomers whose structures have
been determined, the bond distances from Ta to the
iminoacyl nitrogen are comparable for N-out 4b (2.121(5)
Å) to those for the ion pairs N-out 6a (2.143(2) Å) and
N-out 7a (2.130(6) Å). Indeed, the ∆ parameter for the
zwitterion is smaller (–0.061 Å), suggesting a more electro-
philic metal center in the zwitterions. It might also be ex-
pected that more steric hindrance would exist in the
iminoacyl rotation necessary for isomerization in the zwit-
terionic species. Nevertheless, the rates for zwitterions iso-
merization are about an order of magnitude higher than the
non-zwitterionic counterpart, suggesting that differences in
ion pairing in the zwitterionic vs. the non-zwitterionic ion
pairs may be important in the transition state.
Anal. calcd. for C30H15BF15Ta: C 42.28, H 1.77; found:
C 42.42, H 1.50.
′
Synthesis of Cp2 Ta[CH2B(C6F5)3]CH3, 1b
A procedure analogous to the synthesis of 1a, using
(C5H4Me)2Ta(=CH2)(CH3), was employed. The product was
isolated with 0.5 equiv of toluene by 1H NMR after extended
exposure to vacuum. Reagent amounts: B(C6F5)3 (690 mg,
1.35 mmol), (C5H4Me)2Ta(=CH2)(CH3) (500 mg, 1.36 mmol).
Yield of 1b, accounting for toluene: 1.183 g, 1.28 mmol,
1
95%. H NMR (THF-d8): 6.05 (m, 2H, C5H4), 5.83 (m, 2H,
C5H4), 5.62 (m, 2H, C5H4), 5.57 (m, 2H, C5H4), 2.24 (s, 6H,
C5H4-CH3), 1.91 (br, 2H, CH2), 0.43 (s, 3H, CH3). 13C NMR
1
(THF-d8): 139.9 (CH2, JCH = 100.1 Hz), 126.2 (C5H4-CH3),
114.2 (C5H4), 109.1 (C5H4), 109.0 (C5H4), 102.9 (C5H4),
1
27.7 (Ta-CH3, JCH = 124.9 Hz), 14.8 (C5H4-CH3). 11B NMR
(THF-d8) = –4.4. 19F NMR (THF-d8): –130.9 (2F, o-F),
–161.2 (1F, p-F), –164.8 (2F, m-F). Anal. calcd. for
C32H19BF15Ta·1/2C7H8: C 46.03, H 2.50; found: C 45.98, H,
2.42.
Conclusions
The results reported herein add to the body of research on
the mechanistic aspects of isonitrile insertion into M—C
bonds, providing corroborative evidence for the general
mechanistic picture of the insertion step and the isomeri-
zation of N-out isomers to N-in structures. In particular, we
have shown how ground state α agostic interactions can
hamper coordination of the isonitrile substrate, leading to
overall lower rates of insertion in comparison to non-agostic
reactions. These results also show that kinetic N-out isomers
are stabilized in highly electrophilic species, and that the
barriers to conversion to the thermodynamic N-in isomers
are influenced by a variety of factors, including the steric
bulk of the isonitrile substituent and the electron-donating
ability of the Cp ligands on the metallocene. Finally, the rate
of isomerization in the zwitterionic N-out compounds is
somewhat faster than the non-zwitterionic reactions, al-
though the origin of this phenomenon is less clear.
Cp2Ta[CH2B(C6F5)2CH3]CH3, 2a
In a dry box, benzene (10 mL) was added to a flask con-
taining H3CB(C6F5)2 (122 mg, 0.338 mmol) and
Cp2Ta(=CH2)CH3 (113 mg, 0.332 mmol). The solution was
stirred at 25°C for 1 h, and the light yellow precipitate was
isolated by filtration. The precipitate was washed with ben-
zene through back distillation, and the solvent was removed
under reduced pressure. Yield of 2a: 183 mg, 0.261 mmol,
1
79%. H NMR (CD2Cl2): 6.07 (s, 10H, C5H5), 2.46 (br. s,
2H, CH2), 0.42 (s, 3H, Ta-CH3), 0.36 (br. s, 3H, B-CH3). 13
C
1
NMR (CD2Cl2): 159.6 (C5H5), 108.0 (CH2, JCH = 82.7 Hz),
1
20.6 (Ta-CH3, JCH = 112.3 Hz), 14 (B-CH3). 11B NMR
(CD2Cl2): –7.0. 19F NMR (CD2Cl2): –133.7 (2F), –163.5
(1F), –166.0 (2F). Anal. calcd. for C25H18BF10Ta: C 42.89,
H 2.59; found: C 42.59, H 2.20.
Synthesis of [Cp2TaMe2]+[B(C6F5)4]–, 3a
Experimental section
CH2Cl2 (10 mL) was condensed into an evacuated
flask containing Cp2TaMe3 (1.513 g, 4.25 mmol). In sepa-
rate flask, CH2Cl2 (40 mL) was condensed over
[Ph3C]+[B(C6F5)4]– (3.918 g, 4.25 mmol). The solution of
trityl salt was added to the Cp2TaMe3 solution by cannula at
25°C with rapid stirring. An immediate light yellow precipi-
tate formed, and the solution was stirred for an additional
15 min. The precipitate was isolated by filtration and washed
once by back distillation of CH2Cl2. The solvent was then
removed under reduced pressure. Yield of 5a: 3.917 g,
General procedures
General procedures have been described in detail else-
where (21).
Synthesis of Cp2Ta[CH2B(C6F5)3]CH3, 1a
Toluene (15 mL) was condensed into an evacuated flask
containing B(C6F5)3 (214 mg, 0.418 mmol) and a slight ex-
cess of Cp2Ta(=CH2)CH3 (156 mg, 0.458 mmol) at –78°C.
The solution was warmed to 25°C and stirred for 30 min.
The toluene was removed under reduced pressure to leave a
yellow residue. Hexane (15 mL) was condensed into the ves-
sel, and the light yellow precipitate was isolated by filtra-
tion. The precipitate was washed by back distillation of
hexane, and the solvent was removed under reduced pres-
sure. Yield of 1a: 324 mg, 0.380 mmol, 91%. X-ray quality
crystals were grown by cooling a saturated 1,2-dichloro-
1
3.84 mmol, 90%. H NMR resonances are identical to those
previously reported for the cation (9, 13). 13C NMR (THF-
1
d8): 114.1 (C5H5), 55.6 (Ta-CH3, JCH = 125.6 Hz). 11B
NMR (THF-d8): –13.6. 19F NMR (THF-d8): –130.9 (2F,
o-F), –163.0 (1F, p-F), –166.5 (2F, m-F). Anal. calcd. for
C26H16BF20Ta: C 42.38, H 1.58; found: C 41.94, H 1.48.
Reaction of 1a with t-BuNC: Synthesis of N-out 4a
THF (10 mL) was condensed into an evacuated flask con-
taining 1a (315 mg, 0.370 mmol) at –78°C. Under a flow of
argon, an excess of tert-butylisocyanide (400 mL,
3.83 mmol) was syringed into the flask. The solution was
warmed to 25°C and stirred for 75 min. THF was removed
1
ethane solution to –25°C. H NMR (C6D6): 5.93 (s, 10H,
C5H5), 2.13 (br., 2H, CH2), 0.56 (s, 3H, CH3). 13C NMR
1
(C6D6): 150.3 (CH2, JCH = 98.7 Hz), 108.4 (C5H5), 21.4
1
(CH3, JCH = 129.1 Hz). 11B NMR (C6D6): –9.4. 19F NMR
(C6D6): –132.0 (2F, o-F), –159.7 (1F, p-F), –164.4 (2F, m-F).
© 2003 NRC Canada

1144
Can. J. Chem. Vol. 81, 2003
under reduced pressure to leave a white residue, which was
filtered after suspension in hexane (10 mL). The precipitate
was washed by back distillation using hexane, and the sol-
vent was removed under reduced pressure. Yield of N-out
in at –78°C, and the suspension was sonicated to produce a
white powder. The product was isolated by filtration,
washed with hexanes by distillation (3 × 5 mL), and dried in
1
vacuo. Yield of N-in 4b: 250 mg, 0.257 mmol, 71%. H
1
4a: 288 mg, 0.308 mmol, 83%. H NMR (THF-d8): 5.67 (s,
NMR (THF-d8): 5.34 (m, 2H, C5H4), 5.10 (m, 2H, C5H4),
5.06 (m, 2H, C5H4), 5.02 (m, 2H, C5H4), 3.85 (br, 2H, CH2),
1.99 (s, 6H, C5H4(CH3)), 1.38 (s, 9H, C(CH3)3), 0.55 (s, 3H,
CH3). 13C NMR (THF-d8): 223.3 (C=N-t-Bu), 125.5 (C5H4-
CH3), 108.1 (C5H4), 103.9 (C5H4), 103.6 (C5H4), 98.4
(C5H4), 66.0 (C(CH3)3, 35.2 (CH2), 30.3 (C(CH3)3, 14.6
(C5H4-CH3), 11.1 (Ta-CH3). 11B NMR (THF-d8): –9.9. 19F
NMR: –129.8 (2F, o-F), –160.6 (1F, p-F), –164.7 (2F, m-F).
10 H, C5H5), 3.75 (br., 2H, CH2), 1.35 (s, 9H, C(CH3)3),
0.39 (s, 3H, CH3). 13C NMR (THF-d8): 219.8 (C=N-t-Bu),
107.4 (C5H5), 62.9 (C(CH3)3, 34.5 (CH2), 29.2 (C(CH3)3, 5.0
(CH3). 11B NMR (THF-d8): –10.9. 19F NMR: –129.4 (2F, o-
F), –162.2 (1F, p-F), –165.7 (2F, m-F). Anal. calcd. for
C35H24NBF15Ta: C 44.95, H 2.59, N 1.50; found: C 44.69,
H 2.39, N 1.57.
Isomerization of N-out 4a to N-in 4a
Reaction of 1a with C6H11NC: Synthesis of N-out 5a
A procedure analogous to that employed for the synthesis
of N-out 4a was used to prepare this compound. Reagent
amounts: 1a (300 mg, 0.352 mmol), cyclohexyl isocyanide
(500 mL, 4.02 mmol). Yield of N-out 5a: 247 mg,
N-out 4a (170 mg, 0.18 mmol) was placed into a 25 mL
glass bomb and THF (10 mL) was condensed in. The bomb
was heated at 70°C for 10 days. The colourless solution was
cannula-transferred into a 25 mL round bottom flask with
filtering frit. The THF was removed to leave a sticky, colour-
less residue. Hexanes (10 mL) were condensed in at –78°C,
and the residue sonicated to produce a white powder. The
product was filtered off, washed with hexanes by distillation
(3 × 5 mL), and volatiles were removed in vacuo. Yield of
N-in 4a: 152 mg, 0.16 mmol, 89%. 1H NMR (THF-d8): 5.46
(s, 10 H, C5H5), 3.85 (br, 2H, CH2), 1.41 (s, 9H, C(CH3)3),
0.78 (s, 3H, CH3). 13C NMR (THF-d8): 221.7 (C=N-t-Bu),
106.0 (C5H5), 66.1 (C(CH3)3, 35.0 (CH2), 30.0 (C(CH3)3, 3.9
(CH3). 11B NMR (THF-d8): –10.0. 19F NMR: –130.0 (2F, o-
F), –160.3 (1F, p-F), –164.6 (2F, m-F).
1
0.256 mmol, 76%. H NMR (THF-d8): 5.66 (s, 10 H, C5H5),
3.60 (br. s, 3H, CH2 and CH, C6H11), 1.05–1.95 (m, 10H,
C6H11) 0.53 (s, 3H, CH3). 11B NMR (THF-d8): –9.3. 19F
NMR: –129.5 (2F, o-F), –161.7 (1F, p-F), –165.5 (2F, m-F).
Anal. calcd. for C37H26NBF15Ta·C7H8: C 50.16, H 3.25, N
1.33; found: C 50.36, H 3.07, N 1.39.
Isomerization of N-out 5a to N-in 5a
N-out 5a (221 mg, 0.230 mmol) was placed in a 25 mL
glass bomb and THF (15 mL) was condensed in at –78°C.
The bomb was sealed and placed in an oil bath at 65°C for
24 h. The light yellow solution was then cannula-transferred
to a 25 mL round bottom flask equipped with a filtering frit.
The THF was removed under reduced pressure to leave a
yellow oil. Hexanes (10 mL) were condensed onto the oil at
–78°C and the solution sonicated to produce light yellow
powder. The powder was filtered off, washed with hexanes
(3 × 5 mL) by back distillation, and dried in vacuo. Yield of
Reaction of 1b with t-BuNC: Synthesis of N-out 4b
THF (10 mL) was condensed into an evacuated flask con-
taining 1b (325 mg, 0.35 mmol) at –78°C. Under a flow of
argon, an excess of tert-butylisocyanide (400 mL,
3.83 mmol) was syringed into the flask. The solution was
warmed to 25°C and stirred for 2 h. THF was removed un-
der reduced pressure to leave a white residue, which was fil-
tered from hexane (10 mL). The precipitate was washed by
back distillation with hexane, and the solvent was removed
under reduced pressure. Yield of N-out 4b: 283 mg,
0.29 mmol, 84%. X-ray quality crystals were grown by lay-
1
N-in 5a: 153 mg, 0.159, 61%. H NMR (THF-d8): 5.63 (s,
10 H, C5H5), 3.85 (br. s, 2H, CH2 and ipso-CH, C6H11),
1.05–1.95 (m, 10H, C6H11) 0.75 (s, 3H, CH3). 13C NMR
(THF-d8): 223.6 (C=N), 105.9 (C5H5), 60.4 (ipso-C, C6H11),
32.3 (CH2), 31.6 (C6H11), 26.4 (C6H11), 25.9 (C6H11), 3.2
(Ta-CH3). 11B NMR (THF-d8): –11.2. 19F NMR: –129.7 (2F,
o-F), –160.7 (1F, p-F), –164.6 (2F, m-F).
1
ering a saturated CH2Cl2 solution with hexanes. H NMR
(THF-d8): 5.75 (m, 2H, C5H4), 5.61 (m, 2H, C5H4), 5.35 (m,
2H, C5H4), 5.01 (m, 2H, C5H4), 3.76 (br, 2H, CH2), 2.01 (s,
6H, C5H4(CH3)), 1.35 (s, 9H, C(CH3)3), 0.22 (s, 3H, CH3).
13C NMR (THF-d8): 220.7 (C=N-t-Bu), 121.4 (C5H4-CH3),
110.0 (C5H4), 109.7 (C5H4), 108.2 (C5H4), 94.4 (C5H4), 62.9
(C(CH3)3, 33.1 (CH2), 29.4 (C(CH3)3, 14.2 (C5H4-CH3), 9.4
(Ta-CH3). 11B (THF-d8): –9.0. 19F NMR (THF-d8): –129.3
(2F, o-F), –162.3 (1F, p-F), –165.7 (2F, m-F). Anal. calcd.
for C37H28NBF15Ta·CH2Cl2: C 43.54, H 2.88, N 1.34; found:
C 43.63, H 2.78, N 1.34.
Reaction of 3a with t-BuNC: Synthesis of N-out 6a
THF (20 mL) was condensed into an evacuated flask con-
taining 3a (454 mg, 0.45 mmol) at –78°C. Under a flow of
argon, an excess of tert-butylisocyanide (300 mL,
2.87 mmol) was syringed into the flask at –78°C. An imme-
diate colour change from yellow to colourless was noted and
the solution was warmed to 25°C. THF was removed under
reduced pressure and the white residue filtered after suspen-
sion in hexanes. The solid was washed once with hexanes
and dried under reduced pressure. Yield of N-out 6a:
450 mg, 0.41 mmol, 92%. X-ray quality crystals were grown
Isomerization of N-out 4b to N-in 4b
N-out 4b (350 mg, 0.360 mmol) was placed into a 25 mL
glass bomb and THF (10 mL) was condensed into the vessel.
The bomb was heated at 70°C for 2 days. The colourless so-
lution was cannula-transferred into a 25 mL round bottom
flask with filtering frit. The THF was removed to leave a
colourless, waxy residue. Hexanes (10 mL) were condensed
1
by cooling a saturated CH2Cl2 solution to –35°C. H NMR
(THF-d8): 5.87 (s, 10H, C5H5), 2.80 (s, 3H, C(=N-t-
Bu)CH3), 1.49 (s, 9H, C(CH3)3), 0.76 (s, 3H, Ta-CH3). 13C
NMR (THF-d8): 211.1 (C=N-t-Bu), 107.3 (C5H5), 62.9
(C(CH3)3), 28.7 (C(CH3)3), 15.3 (C(=N-t-Bu)CH3), 5.6 (Ta-
© 2003 NRC Canada

Cook et al.
1145
CH3). 11B NMR (THF-d8): –13.6. 19F NMR (THF-d8):
–130.9 (2F, o-F), –163.1 (1F, p-F), –166.6 (2F, m-F). Anal.
calcd. for C41H25NBF20Ta: C 44.63, H 2.28, N 1.27; found:
C 44.28, H 2.26, N 1.25.
(m, 2H, C5H4), 5.43 (m, 2H, C5H4), 5.33 (m, 2H, C5H4),
3.11 (s, 3H, C(=NC(CH3)3)CH3), 2.02 (s, 6H, C5H4CH3),
1.33 (s, 9H, C(CH3)3), 0.62 (s, 3H, Ta-CH3). 13C NMR
(CD3CN): 219.6 (C=N), 118.3 (C-Me, C5H4Me, under sol-
vent resonance), 107.6 (C5H4Me), 107.2 (C5H4Me), 106.1
(C5H4Me), 101.6 (C5H4Me), 65.4 (C(CH3)3), 29.9
(C(CH3)3), 23.7 (NC-CH3), 14.7 (C5H4-CH3), 8.8(Ta-CH3).
11B NMR (CD2Cl2): –1.0. 19F NMR (CD2Cl2): –152.2 (4F).
Isomerization of N-out 6a to N-in 6a
N-out 6a (250 mg, 0.226 mmol) was placed into a 25 mL
glass bomb and THF (15 mL) was condensed in at –78°C.
The vessel was backfilled with argon and placed in a 110°C
oil bath for 24 h. The solution was cooled and then cannula-
transferred to a 25 mL round bottom flask with a filtering
frit. The THF was removed in vacuo to leave a sticky
colourless oil. The oil was broken into a precipitate through
sonication with hexanes (15 mL). The hexanes were then re-
moved under reduced pressure to give a white powder. More
hexanes (10 mL) were condensed in to make a slurry, from
which the precipitate was filtered. The product was washed
with hexanes (2 × 5 mL) by back distillation and the
volatiles removed in vacuo. Yield of N-in 6a: 228 mg,
Reaction of 3a with C6H11NC: Synthesis of N-out 7a
A procedure analogous to that used to prepare N-out 6a
was used to prepare this compound. Reagent amounts: 3a
(366 mg, 0.359 mmol), cyclohexyl isocyanide (60 mL,
0.483 mmol). Yield of N-out 7a: 295 mg, 0.261 mmol, 73%.
1H NMR (THF-d8): 5.88 (s, 10H, C5H5), 4.03 (m, 1H, CH,
C6H11), 2.70 (s, 3H, C(=NC6H11)CH3), 2.01 (m, 4H, C6H11),
1.58–1.30 (m, 6H, C6H11), 0.78 (s, 3H, Ta-CH3). 13C NMR
(THF-d8): 210.5 (C=N), 107.5 (C5H5), 63.5 (CH, C6H11),
32.5 (C6H11), 26.0 (C6H11), 25.8 (C6H11), 14.9 (NC-CH3),
5.3 (Ta-CH3). 11B NMR (THF-d8): –13.6. 19F NMR (THF-
d8): –130.9 (2F, o-F), –163.1 (1F, p-F), –166.6 (2F, m-F).
1
0.206 mmol, 91%. H NMR (THF-d8): 5.85 (s, 10H, C5H5),
3.20 (s, 3H, C(=N-t-Bu)CH3), 1.37 (s, 9H, C(CH3)3), 0.82
(s, 3H, Ta-CH3). 13C NMR (THF-d8): 217.7 (C=N-t-Bu),
106.6 (C5H5), 65.2 (C(CH3)3), 29.2 (C(CH3)3), 23.1 (C(=N-
t-Bu)CH3), 3.8 (Ta-CH3). 11B NMR (THF-d8): –13.6. 19F
NMR (THF-d8): –130.9 (2F, o-F), –163.1 (1F, p-F), –166.6
(2F, m-F). Anal. calcd. for C41H25NBF20Ta: C 44.63, H 2.28,
N 1.27; found: C 44.52, H 2.36, N 1.56.
Isomerization of N-out 7a to N-in 7a
Prepared in the same manner as N-in 6a. Reagent
amounts: N-out 7a (276 mg, 0.244 mmol). Yield of N-in 7a:
240 mg, 0.212, 87%. 1H NMR (CD3CN): 5.71 (s, 10H,
C5H5), 4.01(m, 1H, CH, C6H11), 2.94 (s, 3H,
C(=NC6H11)CH3), 1.80–1.10 (m, 10H, C6H11), 0.76 (s, 3H,
Ta-CH3). 13C NMR (CD3CN): 217.8 (C=N), 106.0 (C5H5),
68.4 (ipso-C, C6H11), 31.9 (C6H11), 26.1 (C6H11), 25.8
(C6H11), 21.0 (NC-CH3), 3.7 (Ta-CH3). 11B NMR (THF-d8):
–13.6. 19F NMR (THF-d8): –130.9 (2F, o-F), –163.1 (1F, p-
F), –166.6 (2F, m-F). Anal. calcd. for C43H27NBF20Ta: C
45.73, H 2.41, N 1.24; found: C 45.47, H 2.45, N 1.37.
Reaction of 3b with t-BuNC: Synthesis of N-out 6b
[Cp2TaMe2]+[BF4]–, 3b (295 mg, 0.647 mmol), was
weighed into a 25 mL two-necked, round bottomed flask
with septum, and CH2Cl2 (10 mL) was condensed in at
–78°C to dissolve the solid. tert-Butylisocyanide (75 mL,
0.663 mmol) was syringed into the solution at –78°C and re-
sulted in an immediate colour change from yellow to colour-
less. The solution was stirred for 30 min at 25°C and the
CH2Cl2 was removed in vacuo to give a colourless oil. Hex-
ane (10 mL) was condensed into the vessel and the solution
sonicated to generate a white powder. The hexane was re-
moved under reduced pressure, and a second quantity of
hexane (10 mL) was condensed into the flask. The vessel
was sonicated to free the white powder from the wall, the
powder filtered, and volatiles removed in vacuo. Yield of N-
Reaction of 3b with C6H11NC: Synthesis of N-out 7b
A procedure analogous to the synthesis of N-out 6b was
used to prepare this compound. Reagent amounts: 3b
(298 mg, 0.653 mmol), cyclohexyl isocyanide (82 mL,
0.659 mmol). Yield of N-out 7b: 332 mg, 0.587 mmol, 89%.
1H NMR (CD2Cl2): 5.91 (m, 2H, C5H4), 5.68 (m, 4H, C5H4),
5.34 (m, 2H, C5H4), 4.05 (m, 1H, ipso-CH, C6H11), 2.62 (s,
3H, C(=NC6H11)CH3), 1.96 (s, 6H, C5H4CH3), 1.96 (m, 3H,
C6H11), 1.51 (m, 7H, C6H11), 0.62 (s, 3H, Ta-CH3). 13C
NMR (CD2Cl2): 211.0 (C=N), 119.6 (C-CH3, C5H4CH3),
112.9 (C5H4), 110.3 (C5H4), 103.9 (C5H4), 100.4 (C5H4),
63.3 (ipso-CH, C6H11), 32.4 (C6H11), 25.6 (C6H11), 25.6
(C6H11), 14.0 (C5H4CH3), 13.5 (C(=NC6H11)CH3), 7.3 (Ta-
CH3). 11B NMR (THF-d8): –1.0. 19F NMR (CD2Cl2): –152.2
(4F). Anal. calcd. for C21H30NBF4Ta: C 44.70, H 5.36, N
2.48; found: C 44.14, H 5.85, N 2.46.
1
out 6b: 299 mg, 0.555 mmol, 86%. H NMR (CD2Cl2): 5.92
(m, 2H, C5H4), 5.70 (m, 2H, C5H4), 5.66 (m, 2H, C5H4),
5.24 (m, 2H, C5H4), 2.73 (s, 3H, C(=NC(CH3)3)CH3), 2.04
(s, 6H, C5H4CH3), 1.49 (s, 9H, C(CH3)3), 0.62 (s, 3H, Ta-
CH3). 13C NMR (CD3CN): 211.7 (C=N), 118.3 (C-Me,
C5H4Me, under solvent resonance), 113.1 (C5H4Me), 111.1
(C5H4Me), 103.9 (C5H4Me), 100.7 (C5H4Me), 62.7
(C(CH3)3), 29.0 (C(CH3)3), 13.8 (C5H4-CH3), 13.6 (NC-
CH3), 7.6 (Ta-CH3). 11B NMR (CD2Cl2): –1.0. 19F NMR
(CD2Cl2): –152.2 (4F). Anal. calcd. for C19H29NBF4Ta: C
42.32, H 5.42, N 2.60; found: C 41.76, H 5.44, N 2.53.
Isomerization of N-out 7b to N-in 7b
N-out 7b (30 mg, 0.052 mmol) was weighed into a J.
Young NMR tube, and an NMR amount of CD3CN was
added by Pasteur pipette. The tube was placed in a 100°C oil
Isomerization of N-out 6b to N-in 6b
1
N-out 6b (32 mg, 0.060 mmol) was weighed into a J.
Young NMR tube, and an NMR amount of CD3CN was
added by Pasteur pipette. The tube was placed in a 110°C oil
bath for 16 h. H NMR (CD3CN): 5.62 (m, 2H, C5H4), 5.60
(m, 2H, C5H4), 5.45 (m, 2H, C5H4), 5.35 (m, 2H, C5H4),
4.05 (m, 1H, ipso-CH, C6H11), 2.97 (s, 3H,
C(=NC6H11)CH3), 1.99 (s, 6H, C5H4CH3), 1.81 (m, 2H,
1
bath for 16 h. H NMR (CD3CN): 5.64 (m, 2H, C5H4), 5.58
© 2003 NRC Canada

1146
Can. J. Chem. Vol. 81, 2003
C6H11), 1.65 (m, 3H, C6H11), 1.51 (m, 2H, C6H11), 1.12–
1.40 (m, 3H, C6H11), 0.62 (s, 3H, Ta-CH3). 13C NMR
(CD3CN): 219.1 (C=N), 122.1 (C-CH3, C5H4CH3), 108.7
(C5H4), 106.6 (C5H4), 105.0 (C5H4), 101.7 (C5H4), 59.3
(ipso-CH, C6H11), 32.5 (C6H11), 26.0 (C6H11), 25.9 (C6H11),
21.0 (C(=NC6H11)CH3), 14.5 (C5H4CH3), 8.2 (Ta-CH3). 11B
NMR (CD3CN): –1.0. 19F NMR (CD3CN): –152.2 (4F).
C22H27NBF4Ta: C 46.10, H 4.75, N 2.44; found: C 45.71, H
4.97, N 2.41.
Isomerization of N-out 8b to N-in 8b
N-out 8b (30 mg, 0.052 mmol) was weighed into a J.
Young NMR tube, and an NMR amount of CD3CN was
added by Pasteur pipette. The tube was placed in an 80°C oil
1
bath for 16 h. H NMR (CD3CN): 7.38 (m, 3H, C6H5), 7.27
Reaction of 3a with C6H5CH2NC: Synthesis of N-out 8a
Compound 3a (312 mg, 0.306 mmol) was weighed into a
two-necked, round bottom flask fitted with a filtering frit
and rubber septum. THF (10 mL) was condensed in at
–78°C to produce a yellow solution, and benzyl isocyanide
(37 mL, 0.306 mmol) was syringed in under a flow of argon.
The resulting white slurry was stirred for 45 min while
warming to room temperature and THF removed under re-
duced pressure to give a clear residue. Hexanes (20 mL)
were condensed into the flask and the solution sonicated to
break down the residue into a white powder. Solvent was re-
moved in vacuo, and to leave a fine white powder, a second
portion of hexanes (10 mL) was added, the resulting white
slurry filtered, and solvent removed in vacuo. Yield of N-out
(m, 2H, C6H5), 5.66 (m, 2H, C5H4), 5.56 (m, 2H, C5H4),
5.49 (m, 2H, C5H4), 5.22 (m, 2H, C5H4), 4.83 (s, CH2), 2.98
(s, 3H, C(=NCH2C6H5)CH3), 1.80 (s, 6H, C5H4CH3), 0.52
(s, 3H, Ta-CH3). 13C NMR (CD2Cl2): 220.2 (C=N), 135.4
(ipso-C, C6H5), 131.5 (C6H5), 131.2 (C6H5), 129.6 (C6H5),
121.6 (C-CH3, C5H4CH3), 110.8 (C5H4), 106.1 (C5H4),
103.6 (C5H4), 101.1 (C5H4), 48.6 (CH2), 21.0
(C(=NCH2C6H5)CH3), 13.9 (C5H4CH3), 7.7 (Ta-CH3). 11B
NMR (THF-d8): –1.0. 19F NMR (CD2Cl2): –152.2 (4F).
Kinetic measurements for isocyanide insertion into the
Ta—CH2 bond of zwitterions 1
In a glove box, a known amount of zwitterions 1a or 1b
was dissolved into an NMR amount of THF-d8 and the
NMR tube fitted with a rubber septum. Using a vacuum line
in a vented fumehood, dry isocyanide reagent (~10 equiv)
was syringed out of the glass bomb under a flow of argon.
The isocyanide was added to the zwitterion solution at
–78°C through the rubber septum, and the NMR sample was
then placed into the NMR probe equilibrated to the desired
temperature. Prior to each analysis the temperature of the
NMR probe was checked against a standard of methanol. Ki-
netic data was collected by following the disappearance of
the cyclopentadienyl proton resonance for 1a or the methyl
signal in 1b, as compared with an internal standard of
ferrocene. 1H NMR spectra were generated with eight pulses
(90°) with a 5 s relaxation delay. At higher temperatures, the
reaction remains first order in [1] for greater than 3 half-
lives, while the reaction remained linear throughout the ex-
periment time for lower temperatures. Typical reagent
amounts: 1a (23 mg, 0.027 mmol), ferrocene (5 mg,
0.027 mmol), THF-d8 (0.63 mL), and cyclohexyl isocyanide
(40 mL, 0.322 mmol).
1
8a: 300 mg, 0.263 mmol, 86%. H NMR (THF-d8): 7.49 (m,
5H, C6H5), 5.70 (s, 10H, C5H5), 5.05 (s, 2H, CH2), 2.64 (s,
3H, C(=NCH2C6H5)CH3), 0.75 (s, 3H, Ta-CH3). 13C NMR
(THF-d8): 210.1 (C(=NCH2C6H5)), 134.7 (ipso-C, C6H5),
130.9 (C6H5), 130.6 (C6H5), 107.5 (C5H5), 57.1 (CH2), 14.2
(C(=NCH2C6H5)CH3), 5.4 (TaCH3). 11B NMR (THF-d8):
–13.6. 19F NMR (THF-d8): –130.9 (2F, o-F), –163.1 (1F, p-
F), –166.6 (2F, m-F). Anal. calcd. for C44H23NBF20Ta: C
46.46, H 2.04, N 1.23; found: C 46.25, H 2.00, N 1.27.
Isomerization of N-out 8a to N-in 8a
N-out 8a (30 mg, 0.052 mmol) was weighed into a J.
Young NMR tube, and an NMR amount of THF-d8 was
added by Pasteur pipette. The tube was placed in an 80°C oil
1
bath for 16 h. H NMR (THF-d8): 7.36 (m, 3H, C6H5), 7.24
(m, 2H, C6H5), 5.69 (s, 10H, C5H5), 4.82 (s, 2H, CH2), 3.06
(s, 3H, C(=NCH2C6H5)CH3), 0.73 (s, 3H, Ta-CH3). 13C
NMR (THF-d8): 218.9 (C(=NCH2C6H5)), 134.9 (ipso-C,
C6H5), 130.6 (C6H5), 130.1 (C6H5), 129.7 (C6H5), 106.3
(C5H5), 49.4 (CH2), 20.7 (C(=NCH2C6H5)CH3), 5.1 (Ta-CH3).
11B NMR (THF-d8): –13.6. 19F NMR (THF-d8): –130.9 (2F,
o-F), –163.1 (1F, p-F), –166.6 (2F, m-F).
Kinetic measurements for N-out to N-in isomerization
of η2-iminoacyl complexes
In a glove box, the N-out isomer was weighed into an
NMR tube with a sealable Pyrex neck and secured to a vac-
uum line with a 180° needle valve. A known amount of
CD3CN or THF-d8 was vacuum-transferred into the tube.
The sample was kept at –78°C and sealed under ~0.9 atm of
argon. The NMR tube was placed in the NMR probe and
equilibrated to the analysis temperature, which was checked
using an ethylene glycol standard. Kinetic data were col-
lected by following the disappearance of the cyclo-
Reaction of 3b with C6H5CH2NC: Synthesis of N-out 8b
An analogous procedure to that employed for the synthe-
sis of N-out 6b was employed, except that pentane was used
in the isolation of the product rather than hexane. Reagent
amounts: 3b (310 mg, 0.680 mmol); benzylisocyanide
(83 mL, 0.681 mmol). Yield of N-out 8b: 306 mg,
1
0.534 mmol, 79%. H NMR (CD2Cl2): 7.45 (m, 5H, C6H5),
5.70 (m, 2H, C5H4), 5.53 (m, 2H, C5H4), 5.47 (m, 2H,
C5H4), 5.12 (m, 2H, C5H4), 5.03 (s, CH2), 2.55 (s, 3H,
C(=NCH2C6H5)CH3), 1.86 (s, 6H, C5H4CH3), 0.58 (s, 3H,
Ta-CH3). 13C NMR (CD2Cl2): 211.1 (C=N), 133.8 (ipso-C,
C6H5), 130.7 (C6H5), 130.2 (C6H5), 129.9 (C6H5), 118.8 (C-
CH3, C5H4CH3), 112.9 (C5H4), 109.2 (C5H4), 104.3 (C5H4),
100.5 (C5H4), 57.3 (CH2), 14.0 (C5H4CH3), 13.0
(C(=NCH2C6H5)CH3), 7.5 (Ta-CH3). 11B NMR (THF-d8):
–1.0. 19F NMR (CD2Cl2): –152.2 (4F). Anal. calcd. for
1
pentadienyl proton resonance or the Ta-CH3 signal in the H
NMR spectrum. For t-BuNC and benzylisocyanide deriva-
tives, ferrocene was used as an internal standard and
weighed directly into the NMR tube with analyte. For
cyclohexyl derivatives, an internal standard, toluene, was
used by spiking CD3CN (61 mg of toluene in 6.223 g of
1
CD3CN). H NMR spectra were generated with four pulses
(90°) with a 5 s delay. At higher temperatures, the reaction
© 2003 NRC Canada

Cook et al.
Table 2. Summary of data collection and structure refinement details for N-out 6a and N-out 7a.
1147
N-out 6aa
N-out 7ab
Formula
C41H25NF20BTa
C43H27BF20NTa·CH2Cl2
FW
1103.38
1214.34
Temperature (K)
Crystal size (mm3)
λ (Å)
Crystal system
Space group
173(1)
0.30 × 0.25 × 0.20
0.71069
Monoclinic
P21/n
193(1)
0.42 × 0.09 × 0.08
0.71073
Orthorhombic
Pbca
a (Å)
b (Å)
c (Å)
13.8132(5)
19.1799(6)
14.5419(5)
92.836(3)
3848.0(2)
4
16.5684(14)
22.7930(19)
23.0095(19)
β (°)
V (Å3)
Z
8689.4(13)
8
1.856
Dcalcd (mg m–3)
1.904
µ (mm–1)
2.984
2.775
4736
0.8085, 0.3887
52.86
F(000)
2144.00
1.000, 0.7608
55.7
Max, min transmission
2θmax (°)
Reflections
Unique reflections
No. of variables
Restraints
R1
32 197
7922
577
37 899
8917
601
0
0
0.022
0.065
0.054
0.127
wR2
GoF
0.95
–1.67, 1.23
1.019
–1.119, 1.689
Residual density (min, max) (e Å–3)
aR1 (calcd on F, I > 3σ(I)) = ∑ ͉͉Fo͉ – Fc / ∑ ͉Fo͉.
bR1 (calcd on F, I > 2σ(I)) = ∑ ͉͉Fo͉ – Fc / ∑ ͉Fo͉; wR2 (calcd on F2, all data) = [∑ (w(Fo – Fc2)2) / ∑ w(Fo )2]1/2
.
2
2
remains first order in the N-out isomer for greater than
3 half-lives, while the reaction remained linear throughout
the experiment time for lower temperatures. Typical reagent
amounts: N-out 6a (24 mg, 0.021 mmol), ferrocene (5 mg,
0.027 mmol), and THF-d8 or CD3CN (0.599 g). Reaction
rates in THF-d8 and CD3CN were identical within experi-
mental error.
Acknowledgments
Funding for this work came from the Natural Sciences
and Engineering Research Council of Canada (NSERC) in
the form of a Discovery Grant (to WEP), an E.W.R. Steacie
Fellowship (2001–2003) (to WEP) and Scholarship support
(PGSA and PGSB) to KSC. KSC also thanks the Alberta
Heritage Scholarship Fund for a Ralph Steinhauer Award of
Distinction Scholarship (2000–2001).
X-ray crystallography
N-out 6a
References
See supporting information4 and Table 2 for full experi-
mental details.
1. J.P. Collman, L.S. Hegedus, J.R. Norton, and R.G. Finke. Prin-
ciples and applications of organotransition metal chemistry.
University Science Books, Mill Valley, CA. 1987.
2. (a) G. Erker. Acc. Chem. Res. 17, 103 (1984); (b) L.D. Durfee
and I.P. Rothwell. Chem. Rev. 88, 1059 (1988); (c) G.
Gachinetti, C. Floriani, F. Marchetti, and S. Merlino. J. Chem.
Soc. Chem. Commun. 522 (1976); (d) G. Fachinetti, G. Fochi,
and C. Floriani. J. Chem. Soc. Dalton Trans. 1946 (1977).
3. K. Tatsumi, A. Nakamura, P. Hofmann, P. Stauffert, and R.
Hoffmann. J. Am. Chem. Soc. 107, 4440 (1985).
4. (a) P. Hofmann, P. Stauffert, and N.E. Schore. Chem. Ber. 115,
2153 (1982); (b) F. De Angelis, S. Sgamellotti, and N. Re.
Organometallics, 19, 4904 (2000); (c) F. De Angelis, S.
Sgamellotti, and N. Re. J. Chem. Soc. Dalton Trans. 1023
(2001).
5. E. Boring, M. Sabat, M.G. Finn, and R.N. Grimes. Organo-
metallics, 16, 3993 (1997).
N-out 7a
Data were collected on a Bruker P4/RA/SMART 1000
CCD diffractometer using Mo Kα radiation at –80°C. Unit
cell parameters were obtained from a least-squares refine-
ment of the setting angles of the number of reflections from
the data collection. An empirical absorption correction was
applied to the data through use of the SADABS procedure.
The structures were solved using the direct-methods pro-
gram SHELXS-86 (22) and full-matrix least-squares refine-
ment on F2 was completed using the program SHELXL-93
(23). Hydrogen atoms were assigned positions based on the
geometries of their attached carbon atoms and were given
isotropic thermal parameters 20% greater than the equivalent
isotropic displacement parameters of the attached carbons.
© 2003 NRC Canada

1148
Can. J. Chem. Vol. 81, 2003
6. G.C. Bazan, S.J. Donnelly, and G. Rodriguez. J. Am. Chem.
Soc. 117, 2671 (1995).
7. F. Amor, J. Sanchez-Nieves, P. Royo, H. Jacobsen, O. Blacque,
H. Berke, M. Lanfranchi, M.A. Pellinghelli, and A.
Tiripicchio. Eur. J. Inorg. Chem. 2810 (2002).
8. R.R. Schrock and P.R. Sharp. J. Am. Chem. Soc. 100, 2389
(1978).
16. (a) J. Jeffery, M.F. Lappert, N.T. Luong-thi, M. Webb, J.L.
Atwood, and W.E. Hunter. J. Chem. Soc. Dalton Trans. 1593
(1981); (b) M.F. Lappert, N.T. Luong-Thi, and C.R.C. Milne.
J. Organomet. Chem. 174, C35 (1979).
17. (a) C. Floriani, K. Schenk, A. Chiesi-Villa, and C. Rizzoli.
Organometallics, 15, 337 (1996); (b) R. Fandos, M.
Lanfranchi, A. Otero, M.A. Pellinghelli, M.J. Ruiz, and P.
Terreros. Organometallics, 15, 4725 (1996).
9. K.S. Cook, W.E. Piers, S.J. Rettig, and R. MacDonald.
Organometallics, 19, 2243 (2000).
18. F.R. Lemke, D.J. Szalda, and R.M. Bullock. Organometallics,
10. A.G. Massey and A.J. Park. J. Organomet. Chem. 2, 245
(1964).
11. R.E. v H. Spence, W.E. Piers, Y. Sun, M. Parvez, L.R.
MacGillivray, and M.J. Zaworotko. Organometallics, 17, 2459
(1998).
12. (a) J.D. Scollard, D.H. McConville, and S.J. Rettig. Organo-
metallics, 16, 1810 (1997); (b) M.G. Thorn, J.S. Vilardo,
P.E. Fanwick, and I.P. Rothwell. Chem. Commun. 2427
(1998); S. Zhang, W.E. Piers, X. Gao, and M. Parvez. J. Am.
Chem. Soc. 122, 5499 (2000).
11, 876 (1992).
19. R.D. Adams and D.F. Chodosh. Inorg. Chem. 17, 41 (1978).
20. L.R. Chamberlain, L.D. Durfee, P.E. Fanwick, L. Kobringer,
S.L. Latesky, A.K. McMullen, I.P. Rothwell, K. Folting, J.C.
Huffman, W.E. Streib, and R. Wang. J. Am. Chem. Soc. 109,
390 (1987).
21. (a) K.S. Cook, W.E. Piers, T.K. Woo, S.J. Rettig, and R. Mac-
Donald. Organometallics, 20, 3937 (2001); (b) K.S. Cook,
W.E. Piers, P.G. Hayes, and M. Parvez. Organometallics, 21,
2422 (2002).
13. E.Y.-X. Chen and D.A. Abboud. Organometallics, 19, 5541
(2000).
14. W.P. Schaefer, R.W. Quan, and J.E. Bercaw. Acta Cryst. Ser.
22. G.M. Sheldrick. Acta Crystallogr. A46, 467 (1990).
23. G.M. Sheldrick. 1993. SHELXL-93: Program for crystal struc-
ture determination [computer program]. University of
Göttingen, Germany.
C, 48, 1610 (1992).
15. E.Y.-X. Chen and T.J. Marks. Chem. Rev. 100, 1391 (2000).
© 2003 NRC Canada