NMR evidence for B(C6F5)3 attack on the inward position of a highly hindered meso ansa-zirconocene

Article abstract of DOI:10.1016/j.jorganchem.2004.10.017

The reaction of the ansa-zirconocene meso-[C₂H ₄(4,7-Me₂Indenyl)₂]ZrMe₂ with B(C₆F₅)₃ afforded, besides the expected ion pair with outward MeB(C₆F₅)₃^- anion, also a minor isomer (molar fraction 0.043) with the bulky methylborate anion in the inward site. 2D ¹H EXSY experiments revealed exchange between the two isomers, at T > 320 K, through the migration of the MeB(C ₆F₅)₃^- anion between outward and inward sites. The presence of free B(C₆F₅)₃ fastened the exchange, suggesting B(C₆F₅)₃ attack on the zirconium bound methyl ligand of the two ion pairs.

Full text of DOI:10.1016/j.jorganchem.2004.10.017

Journal of Organometallic Chemistry 690 (2005) 640–646
www.elsevier.com/locate/jorganchem
NMR evidence for B(C6F5)3 attack on the inward position of a
highly hindered meso ansa-zirconocene
a
a,
a
b
Daniela Maggioni , Tiziana Beringhelli *, Giuseppe DꢀAlfonso , Luigi Resconi
a
` `
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, Facolta di Farmacia, Universita degli Studi di Milano,
Via Venezian 21, 20133 Milano, Italy
b
Basell Poliolefine Italia S.p.A., P.le Privato G. Donegani 12, 44100 Ferrara, Italy
Received 21 July 2004; accepted 7 October 2004
Abstract
The reaction of the ansa-zirconocene meso-[C2H4(4,7-Me2Indenyl)2]ZrMe2 with B(C6F5)3 afforded, besides the expected ion pair
with outward MeBðC6F5Þꢀ anion, also a minor isomer (molar fraction 0.043) with the bulky methylborate anion in the inward site.
3
ꢀ
1
2D H EXSY experiments revealed exchange between the two isomers, at T > 320 K, through the migration of the MeBðC6F5Þ
3
anion between outward and inward sites. The presence of free B(C6F5)3 fastened the exchange, suggesting B(C6F5)3 attack on
the zirconium bound methyl ligand of the two ion pairs.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Ansa-zirconocenes; Perfluoroarylboranes; Meso-isomers; Ion-pair; NMR; Dynamics
1. Introduction
sites, usually described as ‘‘inward’’ and ‘‘outward’’ with
respect to the bisindenyl pocket (Chart 1). As expected,
sterically demanding ligands are preferentially coordi-
nated in the less encumbered outward position. Theoret-
ical studies have previously shown that the inward
coordination was disfavored by ca. 5–6 kcal/mol, both
in the case of propene compared with methyl [10] and
in the case of isobutyl vs. propene [11]. Such energy
Group 4 chiral ansa-bisindenyl metallocenes have
been the object of intense research activity, since the pio-
neer reports [1,2] of their high catalytic activity for ole-
fins polymerization [3,4], upon activation with suitable
Lewis acids [5]. Differently from C2-symmetric ansa-
rac-bisindenyl metallocenes, which afford with good
selectivity isotactic polypropylene, the CS meso isomers
produce atactic polypropylene [3]. However, these iso-
mers allow efficient synthesis of ethylene-based polymers
with high molecular weights and narrow molecular
weight distributions [6]. Therefore, also these isomers
can have important technological uses [7–9], and the
mechanisms responsible for the growth of the polymer
chain on the metal site have been investigated [10,11].
In these meso ansa-bisindenyl metallocenes the ethy-
lenic loop forces the two indenyl ligands in an ‘‘eclipsed’’
conformation and generates two, achiral, diastereotopic
Zr
Me
Me
1
*
Corresponding author. Tel.: 39 02 503 14350; fax: 39 02 503 14405.
E-mail address: tiziana.beringhelli@unimi.it (T. Beringhelli).
Chart 1.
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.10.017

D. Maggioni et al. / Journal of Organometallic Chemistry 690 (2005) 640–646
641
difference implies (at least for these species, that contain
slightly but not negligibly different ligand sets with
respect to our ones) that the relative amount of a deriv-
ative containing the bulkier ligand in the inward posi-
tion should be much lower that 0.001.
The reactivity of the recently synthesized ansa-zir-
conocene meso-[C2H4(4,7-Me2Ind)2]Zr(Me)2 (1, Chart
1) [12] was in line with the above expectations: the reac-
tion with 2 equiv of t-BuOH gave the outward mono-
tert-butoxy derivative only, and the attack of B(C6F5)3
on the same compound was found to occur on the out-
ward methyl group, affording the ion pair [C2H4(4,7-
Me2Ind)2ZrMeinw]+ [MeB(C6F5)3]ꢀ [12].
A careful NMR analysis of these reaction mixtures
has now revealed the formation also of another, very
minor species, which has been identified as the isomer
in which the [MeB(C6F5)3]ꢀ anion is in the inward posi-
tion. We discuss here the characterization of this com-
plex and some aspects of its dynamics, showing that
the inward position of this meso form is more accessible
for B(C6F5)3 than expected.
which has been unambiguously identified as the contact
ion pair 2a, with B(C6F5)3 bound in the outward posi-
tion (Chart 2) on the basis of the dipolar correlations
shown by the methyl group bound to the boron atom
(Me–B, easily identified from the broadening due to
the quadrupolar boron isotopes) and by the zirconium
bound methyl group (Me–Zr).
Indeed on the basis of the assignment of the indenylic
resonances (Fig. 1), performed as detailed in the Exper-
imental section, in Fig. 2 we observe that Me–B displays
correlations with Me9 and H3 (and to a lesser extent
with H2), while Me–Zr correlates with both Me9 and
Me10, as well as with both the protons H5 and H6.
In these ansa-meso species a further evidence assists in
the identification of the inward or outward coordination
of methyl ligands, namely the chemical shift of the
methyl resonances, the inward one being shifted to high
field by the shielding effect of the aromatic p-electrons
(for instance d ꢀ2.12 vs. 0.02 in 1 [12]). In 2a the sharp
signal of Me–Zr is observed at ꢀ1.82 ppm (in agreement
with its inward position) and the broad Me–B resonance
is at 0.12 ppm.
However, a detailed analysis of dipolar and exchange
correlations in experiments at 320 K suggested the pres-
ence in solution also of a second very minor isomer (2b,
ratio about 4.5/100 with respect to 2a), in slow exchange
with 2a. Even if most of the signals of this species were
2. Results and discussion
The addition of 1 equiv of B(C6F5)3 to a toluene solu-
tion of 1 resulted in the formation of one main species,
9
5
3
6
2
10
8
(C6F5)3BMe
Zr
Me
Zr
Me
MeB(C6F5)3
2a
2b
Chart 2.
8b
8a
3
2
5
6
Me9
(a)
(c)
(b)
1. 84
1. 92
Me10
(ppm )
2. 88
2. 80
2. 72
2. 64
2. 56
8b
8a
2.80
6.48
6.40
6.32
6.24
6.16
6.08
6.00
Fig. 1. Selected regions of a 2D 1H NOESY of 2a (toluene-d8, 273 K, sm = 0.5 s) showing the correlations between the indenylic signals.

642
D. Maggioni et al. / Journal of Organometallic Chemistry 690 (2005) 640–646
Me10 Me9
chemical shifts (see the spectrum of Fig. 3): Me–B0 in 2b
2
3
5
6
lies at negative chemical shift, as observed for Me–Zr of
the major isomer 2a and for the inward methyl groups of
the free zirconocene 1.
(ppm)
Me-Zr
-1.80
Isomer interconversion might occur by two different
pathways, involving: (a) dissociation of B(C6F5)3 from
one methyl, followed by attack on the other one, or
-1.60
-1.6
-1.40
(b) separation of the whole MeBðC6F5Þꢀ anion, reorgan-
3
ization of the cationic part of the ion pair and then
recombination with the novel stereochemistry. These
two processes correspond to the two types of dynamiþc
processes that have been observed in ½Cp02ZrMeꢁ
0. 00
0.0
Me-B
0. 20
*
(ppm)
(ppm)
11..8844 1.76
6.40
6. 40
6.20
6. 20
6.00
6. 00
6. 60
2. 00 1. 92
1. 76
½MeBðArFÞ ꢁꢀ ion pairs (Cp0 = cyclopentadienyl like
3
Fig. 2. Selected regions of the sameH experiment in Fig. 1 showing the
correlations between the protons of the indenyl moiety and the Me–Zr
and Me–B signals. The asterisk indicates CH4 arising from some
decomposition [13].
ligand, ArF = aryl group with fluorine substituents)
[5,15–17], as shown in Chart 3 for the prototypal
[(Me2Cp)2ZrMe]+[MeB(ArF)3]ꢀ ion pairs [15]. The
terms dissociation-recombination (d–r) and ion pair
separation (ips) have been used to indicate the borane
or the anion exchangꢀe, respectively [15]. In the
½Cp0 ZrMeꢁþ½MeBðArFÞ ꢁ ion pairs the two methyl sites
barely detectable (see the spectra in Figs. 3 and 4), they
could be clearly identified by the exchange cross peaks
with the resonances of the major isomer 2a. In particular
Fig. 3 shows that Me–Zr of 2a exchanges with the Me–
Zr group of the minor species 2b (Me–Zr0) and that the
Me–B signal of 2a correlates with the Me–B signal of 2b
(Me–B0). The exchange correlations between the indeny-
lic protons of the two species (as well as some mutual
n.O.e. correlation between the signals of the minor spe-
cies 2b) are shown in Fig. 4.
These data, and the notably constancy of the 2b/2a
ratio in different samples, allowed to rule out the
hypothesis that 2b was merely an accidental impurity.
Attempts to isolate as a pure solid any of the two iso-
mers failed.
Further evidence supports the hypothesis that 2b was
the ion pair in which the [MeB(C6F5)3]ꢀ anion is in the
inward site of the ansa meso-zirconocene. In particular
Me–Zr0, in 2b, exhibits the same n.O.e correlations
shown by Me–B in the major species 2a (i.e., it correlates
with Me9 and with H3) [14]. Moreover, the exchange of
the inward and outward position of Me–Zr and Me–B
with respect to 2a is confirmed by the inversion of their
2
3
are equivalent and therefore both the processes are
degenerate (mutual exchanges). By contrast, in the pre-
sent case, due to the Cs symmetry of the rigid meso form,
the methyl sites are diastereotopic and the d–r and ips
dynamic processes become isomerization reactions.
The consequences of the two dynamic processes are
depicted in Fig. 5. The ‘‘ips’’ motion would bring the
outward Me–B of 2a into the inward position in 2b,
resulting in the following exchange pattern (the primes
refer to the minor isomer): (Me–B) M (Me–B)0,
(Me–Zr) M (Me–Zr)0. By contrast, the migration of
B(C6F5)3 from the outward to the inward methyl
(according to the ‘‘d–r’’ mechanism) would result in
the (Me–B) M (Me–Zr)0 and (Me–Zr) M (Me–B)0
exchanges.
Fig. 3 clearly shows that the exchange pattern ob-
served at T P 320 K agrees with the ips and not with
the d–r process. No evidence of the exchange pattern ex-
pected from the d–r process was obtained up to 340 K.
Therefore in the temperature range 320–340 K the anion
exchange (‘‘ips’’) mechanism was the only process active
Me-Zr
Me-Zr’
*
Me-B’
Me-B
(ppm)
-1. 60
(ppm)
0. 8
0. 0
-0. 8
-1. 6
1
Fig. 3. Selected region of a 2D H EXSY spectrum (toluene-d8, 320 K, sm = 0.5 s) showing exchange cross peaks between Me–Zr and Me–B of 2a
and the corresponding signals of 2b (magnified 12· in the insets). The asterisk indicates CH4 arising from some decomposition. The sharp weak signal
overlapping the broad resonance of Me–B of 2b is the low-field 13C satellite of the Me–Zr signal of the main isomer 2a.

D. Maggioni et al. / Journal of Organometallic Chemistry 690 (2005) 640–646
643
Fig. 4. Selected regions of a 2D 1H EXSY (full line cross peaks)/NOESY (dotted line cross peaks) spectrum (toluene-d8, 300 K, sm = 0.3 s), showing
the correlation between the indenyl resonances of the major isomer (2a) with the corresponding ones of the minor isomer 2b: the latter are identified
by a prime after the number. The F2 projection of panel d has been magnified. Please note also the inversion of the chemical shifts of protons 2 and 3
(2a) with respect to 273 K (Figs. 1 and 2) due to their different temperature dependence. The n.O.e cross-peaks are indicated by boxes.
in the time scale of the EXSY experiments. This is
attributable to the fact that B(C6F5)3 is a very strong Le-
wis acid: this favors methyl abstraction and hampers the
cleavage of the Me–B interaction (d–r). For a series of
related ½Cp0 ZrMeꢁþ½MeBðArFÞ ꢁꢀ ion pairs containing
it can be safely stated that the rates of both the ips
and d–r processes are much slower than in the corre-
sponding bis-indenyl derivative without the ansa bridge,
i.e., [(4,7-Me2Ind)2ZrMe]+[MeB(C6F5)3]ꢀ, which adopts
a rac-like conformation in solution [18]. Indeed, in the
latter species the rate of the ips process was measurable
at a temperature as low as 254 K and at room tempera-
ture both the ips and d–r processes were simultaneously
operating [18].
2
3
different B(ArF)3 acids, the ips process resulted kineti-
cally favored with respect to the d–r mechanism only
with the most strongly bound borane, i.e., B(C6F5)3
[15c].
The low intensity of the resonances of 2b prevented
reliable estimation of the kinetic parameters. Anyway,
In principle, the detection of the minor isomer does
not imply necessarily that B(C6F5)3 is able to attack

644
D. Maggioni et al. / Journal of Organometallic Chemistry 690 (2005) 640–646
3. Conclusions
-
R*A
d-r
This work indicates that the inward position in ansa-
meso-bisindenyl zirconocenes is relatively accessible,
even for bulky reactants like B(C6F5)3. Indeed the spe-
cies in which the [MeB(C6F5)3]ꢀ anion is into the inward
site of the meso complex has been found less unfavored
than expected (ca. 1.9 Kcal molꢀ1, on the basis of the
measured ratio between isomers 2a and 2b). Migration
of the [MeB(C6F5)3]ꢀ anion from the outward to the in-
ward position (the ips process) also occurs, even if with a
rate smaller than that of the migration of the
[MeB(C6F5)3]ꢀ anion in the related bisindenyl ion pair
without the ansa bridge. Moreover, the exchange pat-
tern of the methyl resonances in the 2D 1H EXSY/
NOESY maps recorded at 283 K in the presence of free
B(C6F5)3 suggests that B(C6F5)3 is able to attack even
the inward position of the ansa-bis(indenyl)ZrMe+ cat-
ion in ion pair 2a. This finding contrasts with previous
reports concerning related bis(cyclopentadienyl)-
dimethyl metallocenes [15a,20], for which the rate of
the d–r process resulted unaffected by an excess of
B(C6F5)3. Most likely, the occurrence of electrophilic at-
tack on this (and other ones [18]) (indenyl)2ZrMe+ cati-
ons is a consequence of the higher donor power of the
indenyl groups that makes these species softer with re-
spect to Cp2ZrMe+.
R*
R
+
M
M
+ A
R
R*
+
M
-
RA
-
RA
+
+
-
M
M
R*
+ AR
R*
ips
Chart 3.
(a)
(Me-Zr)
(Me-B)
(Me-Zr)’
(b)
(Me-B)’
0.6
0.2
-0.2
-0.6
-1.0
-1.4
(Me-Zr)
(Me-B)
4. Experimental
(Me-B)’
(Me-Zr)’
All manipulations were performed under nitrogen in
oven dried Schlenk-type glassware. B(C6F5)3 (Boulder
Scientific) was purified by extractions with refluxing
n-pentane. Deuterated toluene (C.I.L.) was dried on
0.6
0.2
-0.2
-0.6
-1.0
-1.4
Fig. 5. A picture of the exchange pattern between the methyl
resonances of 2a and 2b, according to (a) the ips or (b) the d–r
mechanism.
˚
activated molecular sieves (4 A). The NMR spectra were
acquired on Bruker AVANCE DRX-300. 19F NMR
spectra were referenced to external CFCl3 (d = 0 ppm)
and 11B NMR spectra to external Et2O Æ BF3 (d = 0
ppm). meso-[C2H4(4,7-Me2Ind)2ZrMe2] was prepared
as previously described [12].
The samples of meso-[C2H4(4,7-Me2Ind)2ZrMe]+
[MeB(C6F5)3]ꢀ were prepared in situ, as follows:
equimolar amounts of B(C6F5)3 and meso-[C2H4(4,
7-Me2Ind)2ZrMe2] (ca. 20 mg, 0.046 mmol) were sepa-
rately weighted under N2 in two NMR tubes and dis-
solved in weighted amounts of dried toluene-d8; then
the solution of B(C6F5)3 was added to that of the
zirconocene.
the inward position: indeed the fact that isomers equili-
bration occur via anion migration and not B(C6F5)3 dis-
sociation would be compatible with the hypothesis that
borane attacks the outward site only. However, in the
presence of free B(C6F5)3, the otherwise undetectable
(Me–B) M (Me–Zr)0 and (Me–Zr) M (Me–B)0 exchange
pattern, characteristic of the d–r process, was observed
at a temperature as low as 283 K (Fig. 6) [19]. Since
the presence of free B(C6F5)3 cannot affect the rate of
a truly dissociative process such as the d–r one, the ob-
served exchange pattern must arise from a bimolecular
process, whose results mimic those of the d–r mecha-
nism. We have therefore to admit that, in spite of the
steric hindrance, B(C6F5)3 is able to attack the (inward)
methyl site of 2a, leading to isomer interconversion
through a SE2-like mechanism, i.e., formation of an
unstable ansa-(Ind)2Zr2+. 2[MeB(C6F5)3]ꢀ ion pair, fol-
lowed by B(C6F5)3 extrusion from one methyl site.
4.1. meso-[C2H4(4,7-Me2Ind)2ZrMe]+[MeB(C6F5)3]ꢀ
(2a, major isomer)
1
Panel (b) of the 2D H NOESY experiment in Fig. 1
shows the correlation of indenyl methyl resonance at
1.91 ppm with one of the multiplets (centered at 2.83

D. Maggioni et al. / Journal of Organometallic Chemistry 690 (2005) 640–646
645
Fig. 6. Selected regions of 2D 1H EXSY (full line cross peaks)/NOESY (dotted line cross peaks) spectra (toluene-d8, 283 K, sm = 0.5 s) of ion pair 2,
in the absence (panels a and b), or in the presence (c and d) of free B(C6F5)3. Panels a and b: n.O.e. peaks between Me–Zr and Me–B of 2a only, no
exchange with 2b. Panels c and d: n.O.e. peaks between Me–Zr and Me–B of 2a, but also exchange with 2b according to a ‘‘d–r like’’ exchange
pathway (see Fig. 5(b)). The n.O.e cross-peaks are indicated by boxes.
ppm) of the ethylene ansa, so these resonances are as-
signed to Me10 and H8a (i.e., the methylenic proton
pointing the inward position), respectively. The low-field
component (6.50 ppm) of the AB system of the six mem-
bered ring is identified as H6 for its cross peak with
Me10, while the resonance at 6.14 ppm is assigned to
H3 for its correlation with Me9 at 1.83 ppm (Fig.
1(a)). The assignment of the doublet centred at 6.10
ppm to H2 is straightforward for its cross peak with
the ansa proton H8b, at ca. 2.65 ppm, (i.e., the methy-
lenic proton pointing the outward position) (Fig. 1(c)).
1H NMR: d 6.54 (d, H6), 6.32 (d, H5), 6.17 (d, H3),
6.05 (d, H2), 3.0–2.8 (m, H8a-inward), 2.73–2.56 (m,
H8b-outward), 1.90 (s, Me10), 1.82 (s, Me9), 0.120 (s
br, MeB), ꢀ1.82 (s, MeZr). 13C NMR: the attributions
were performed through 2D1H–13C HMQC and HMBC
experiments (300 K, toluene-d8 7.1 T). Some ambiguities
have been resolved by recording the experiments at dif-
ferent temperatures (273 and 300 K): (13C{1H}, 300 K):
d 148.6 (m, 1JC–F = 243 Hz, CF ortho or meta), 137.6 (m,
1
1JC–F = 251 Hz, CF meta or ortho), 139.5 (m, JC–F
=
249 Hz, CF para), 124.2 (C1), 115.9 (C2), 103.4 (C3),

646
D. Maggioni et al. / Journal of Organometallic Chemistry 690 (2005) 640–646
[5] E.Y.-X. Chen, T.J. Marks, Chem. Rev. 100 (2000) 1391.
[6] L. Resconi, F. Piemontesi, M. Galimberti, US Patent 5,585,448 to
Montell Tech. Company.
132.2 (C4), 127.1 (C5), 129.9 (C6), 133.4 (C7), 31.59
(C8), 17.28 (C9), 20.25 (C10), 128.7 (C3a), 126.2
(C7a), 51.69 (Me–Zr), 17.64 (Me–B). 11B NMR: d
ꢀ13.6 (Dm1/2 = 23 Hz). 19F NMR: (297 K) d ꢀ132.85
(m, F-ortho), ꢀ159.55 (pseudo t, F-para), ꢀ164.35 (m,
F-meta).
[7] P. Galli, G. Collina, P. Sgarzi, G. Baruzzi, E. Marchetti, J. Appl.
Polym. Sci. 66 (1997) 1831.
[8] L. Resconi, U. Giannini, T. DallꢀOcco, in: W. Kaminsky, J.
Scheirs (Eds.), Metallocene Catalysts, Wiley–Weinheim, New
York, 1999, p. 69.
[9] T. Uozomi, K. Miyazawa, T. Sano, K. Soga, Macromol. Rapid
Commun. 18 (1997) 883.
4.2. meso-[C2H4(4,7-Me2Ind)2ZrMe]+[MeB(C6F5)3]ꢀ
(2b, minor isomer)
´
[10] L.A. Castoguay, A.K. Rappe, J. Am. Chem. Soc. 114 (1992) 5832.
[11] G. Guerra, L. Cavallo, G. Moscardi, M. Vacatello, P. Corradini,
Macromolecules 29 (1996) 4834.
1H NMR: (300 K) d 6.38 (d, H3), 6.05 (pseudo:s H5
and H6), 5.37 (d, H2), 3.40 (m, H8a-inward), 2.73 (m,
H8b-outward), 2.32 (s, Me9), 2.10 (s, Me10), 1.00 (s,
MeZr), ꢀ1.61 (s br, MeB). The accidental overlap of
the 8b ansa resonances for 2a and 2b is confirmed by
the n.O.e. correlations with the related 8a and 2 hydro-
gens (Figs. 4(c) and (d)).
[12] D. Balboni, I. Camurati, G. Prini, L. Resconi, S. Galli, P.
Mercandelli, A. Sironi, Inorg. Chem. 40 (2001) 6588.
[13] We suppose that the weak methane signal usually observed in our
reaction mixtures arises from methyl protonation by the
B(C6F5)3 Æ H2O adduct, which is a strong Brønsted acid, formed
from adventitious water, present also after careful solvent
anhydrification. The hypothesis that methane arises from some
slow decomposition pathway can be ruled out, since its signal was
already observed immediately after B(C6F5)3 addition to the
zirconocene solution.
Acknowledgements
[14] No correlation could be found for Me–B because its signal,
besides being weak, was also very broad.
[15] (a) X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 116
(1994) 10015;
T.B. and G.D. thanks Basell Polyolefins for partial
funding of this work. Thanks are also due to Simona
Guidotti and Davide Balboni for providing the metallo-
cene sample, to Pasquale Illiano for NMR technical
assistance and to the Italian CNR (ISTM) for providing
facilities for low-temperature and inert atmosphere
manipulations.
(b) P.A. Deck, T.J. Marks, J. Am. Chem. Soc. 117 (1995) 6128;
(c) P.A. Deck, C.L. Beswick, T.J. Marks, J. Am. Chem. Soc. 120
(1998) 1772.
[16] A.R. Siedle, R.A. Newmark, J. Organomet. Chem. 497 (1995)
119.
[17] S. Beck, S. Lieber, F. Schaper, A. Geyer, H.H. Brintzinger, J.
Am. Chem. Soc. 123 (2001) 1483.
[18] T. Beringhelli, G. DꢀAlfonso, D. Maggioni, P. Mercandelli, A.
Sironi, Chem. Eur. J., in press.
References
[19] The presence of free B(C6F5)3 did not affect the rate of the ips
process, since the exchange pattern typical of this mechanism
became detectable only at 320 K, as in the absence of B(C6F5)3
excess. In Fig. 4 the 2a ! 2b isomer exchange is observable at
T = 300 K because of the presence of free B(C6F5)3 in the sample
used for this experiment. The same correlations albeit with a
worse S/N, due to the poor stability of the mixture at high
temperature [12], are observable at 320 K without free B(C6F5)3.
[20] X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 113 (1991)
3623.
[1] J. Ewen, J. Am. Chem. Soc. 106 (1984) 6355.
[2] W. Kaminsky, K. Kulper, H.H. Brintzinger, F.R.W. Wild,
¨
Angew. Chem., Int. Ed. Engl. 24 (1985) 507.
[3] L. Resconi, L. Cavallo, A. Fait, F. Piemontesi, Chem. Rev. 100
(2000) 1253.
[4] See H.H. Brintzinger, D. Fischer, R. Mulhaupt, B. Rieger, R.M.
¨
Waymouth, Angew. Chem., Int. Ed. Engl. 34 (1995) 1143, and refs
therein.