Observation of zirconium allyl species formed during zirconocene-catalyzed propene polymerization and mechanistic insights

Article abstract of DOI:10.1016/j.jcat.2014.12.004

Oligomeric Cp₂Zr⁺-allyl species were detected in reaction mixtures of either [Cp₂ZrMe][MeB(C₆F₅)₃] or [Cp₂ZrMe][B(C₆F₅)₄] with propylene by a combination of ¹H NMR spectroscopy and electrospray ionization (tandem) mass spectrometry techniques. Conjointly, the data imply that formation of Cp₂Zr⁺-allyl species occurs via (re)coordination of alkene (propylene and vinylidene end groups of unsaturated polymer chains, respectively) to Cp₂Zr⁺R (R = CH₃, H, polymeryl), followed by intramolecular proton transfer from Cγ to R and release of RH. Analysis of the olefinic region of the ¹H NMR spectra of the reaction mixture obtained from [Cp₂ZrMe][MeB(C₆F₅)₃] and propylene reveals the presence of a triplet resonance at δ ～ 5.2, which was attributed to the unprecedented 3-propenyl end group. A plausible mechanism for the origin of the 3-propenyl end groups is discussed. Additionally, a mechanism for incorporation of stereodefects into stereoregular polymers is also discussed. The cationic Cp₂Zr⁺-allyl intermediates formed during polymerization may contribute to catalyst deactivation.

Full text of DOI:10.1016/j.jcat.2014.12.004

Journal of Catalysis 323 (2015) 112–120
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Observation of zirconium allyl species formed during zirconocene-
catalyzed propene polymerization and mechanistic insights
⇑
Mihaela Vatamanu
Department of Chemistry, Queen’s University, Kingston, ON K7L 3N6, Canada
a r t i c l e i n f o
a b s t r a c t
Oligomeric Cp2Zr+–allyl species were detected in reaction mixtures of either [Cp2ZrMe][MeB(C6F5)3] or
[Cp2ZrMe][B(C6F5)4] with propylene by a combination of 1H NMR spectroscopy and electrospray ioniza-
tion (tandem) mass spectrometry techniques. Conjointly, the data imply that formation of Cp2Zr+–allyl
species occurs via (re)coordination of alkene (propylene and vinylidene end groups of unsaturated poly-
mer chains, respectively) to Cp2Zr+R (R = CH3, H, polymeryl), followed by intramolecular proton transfer
Article history:
Received 15 September 2014
Revised 8 December 2014
Accepted 10 December 2014
from Cc
to R and release of RH. Analysis of the olefinic region of the 1H NMR spectra of the reaction mix-
Keywords:
Zirconocene
Allyl species
Mechanism
Stereoerror
ture obtained from [Cp2ZrMe][MeB(C6F5)3] and propylene reveals the presence of a triplet resonance at
d ꢀ 5.2, which was attributed to the unprecedented 3-propenyl end group. A plausible mechanism for
the origin of the 3-propenyl end groups is discussed. Additionally, a mechanism for incorporation of ste-
reodefects into stereoregular polymers is also discussed. The cationic Cp2Zr+–allyl intermediates formed
during polymerization may contribute to catalyst deactivation.
Polymerization
3-Propenyl end group
Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction
(2,1) insertions are also possible. Chain transfer generally involves
b-hydride elimination to produce a cationic zirconocene hydride
Metallocene catalysts of the type [Cp02ZrMe]+[X]ꢁ (Cp0 = substi-
tuted
5-cyclopentadienyl group and X = weakly coordinating
[Cp02ZrH]+, which is able to reinitiate the polymerization and
release a polymer chain with an olefinic end group (Scheme 1)
[6,7].
g
anion) have revolutionized the polyolefin industry, making new
polymeric materials possible through precise control of polymer
microstructure, composition, and molecular weights [1–5]. The
great amount of research conducted in this area to date has
resulted in a consensus regarding the main steps of the polymeri-
zation reaction [2,6,7]. Thus, it is now established that the active
species in these catalysts is the cationic zirconocene alkyl complex
[Cp02ZrMe]+, generated in situ via methide abstraction from the
catalyst precursor Cp02ZrMe2 by a strong Lewis acid. Common
Lewis acids include tris(perfluorophenyl)borane and ammonium
or trityl salts of [B(C6F5)4]ꢁ [8–12]. Initiation takes place by coordi-
nation of alkene to the vacant site at the metal center, followed by
Nevertheless, one aspect of zirconocene-catalyzed alkene poly-
merization, not yet sufficiently well understood, concerns those
incidents that take place during catalysis and impair the catalyst
activity and selectivity [3,13–17]. Such occurrences include regio-
irregular propylene insertion into the Zr–C bond to form secondary
(2,1) zirconium polymeryl species [18–21], formation of zirconium
allyl species [22–27], and chain epimerization [28–32].
In a prior publication, it was demonstrated that vinylidene-
terminated compounds such as 2,4-dimethyl-1-pentene and 2,
4-dimethyl-1-heptene react irreversibly with cationic complexes
of the type [Cp2ZrMe]+ to form mixtures of dynamic isomeric 3 zir-
g
migratory insertion of alkene into the metal–carbon
the polymerization is initiated, the polymer chain growth occurs
by successive alkene coordination and insertion into the Zr–C
r
-bond. Once
conium allyl compounds, which coexist in solution in rapid equilib-
rium with their g1 counterparts, along with methane [33].
Zirconium allyl complexes have long been presumed to form during
bond. For
a-alkene (e.g., propylene) polymerizations, insertion of
a-olefin polymerization and participate in premature deactivation
monomer into the Zr–C bond occurs predominantly in a 1,2-fash-
ion to give primary (1,2) Zr–polymeryl species, but secondary
of the catalyst. Recently, oligomeric zirconium allyl species could
actually be observed directly by 13C NMR spectroscopy in 1-13
C-1-hexene polymerization catalyzed by [rac-Me2Si(indenyl)2
ZrCH2SiMe3][B(C6F5)4] [34] and by 1H NMR spectroscopy as
byproducts in propylene polymerization by [Me2C(Cp)
⇑
Corresponding author. Present address: 100 City Centre Drive, P.O. Box 2225,
Mississauga, ON L5B 3C6, Canada.
(indenyl)Zr(Me)][MeB(C6F5)3] [35]. More recently, by using
a
E-mail address: vatamanumi@gmail.com
http://dx.doi.org/10.1016/j.jcat.2014.12.004
0021-9517/Ó 2014 Elsevier Inc. All rights reserved.

M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
113
in diethyl ether) was added dropwise over a period of 45 min to a
stirring suspension of zirconocene dichloride (1.75 g, 6 mmol) in
20 mL of dry diethyl ether under argon at ꢁ20 °C, and the reaction
mixture thus formed was stirred at this temperature for an addi-
tional 30 min. The reaction mixture was then warmed to ambient
temperature and the solvent was removed under reduced pressure.
The residue was extracted with hexane and filtered, the solvent
was evaporated in vacuo, and the product was purified by sublima-
tion at 70 °C. Alternatively, the product was purified by recrystalli-
zation from hexane. Dimethylzirconocene was obtained as a white
powder at a yield of 60.4%. 1H NMR (300 MHz, CD2Cl2): d 6.1 (s,
10H, Cp), ꢁ0.41 (s, 6H, Me) ppm. 1H NMR (400 MHz, C6D5CD3): d
5.7 (s, 10H, Cp), ꢁ0.21 (s, 6H, Me) ppm.
2.3. Synthesis of tris(pentafluorophenyl)boron
Scheme 1. General mechanism for propylene polymerization by metallocene
catalysts (Cp0 = 5-cyclopentadienyl type ligand, P = polypropenyl chain).
Tris(pentafluorophenyl)boron was prepared by a modified pro-
cedure of Massey and Park [40]. A solution of butyllithium (33 mL,
53 mmol, 1.6 M solution in hexane) was added dropwise over 3 h
under argon to a stirred solution of bromopentafluorobenzene
(6 mL, 48.1 mmol) in 300 mL dry hexane at ꢁ78 °C. The resulting
white suspension was stirred at ꢁ78 °C for an additional 3 h, after
which boron trichloride (16 mL, 16 mmol, 1 M solution in hexane)
was added all at once and the stirring was continued for another
1 h. After warming to ambient temperature, the reaction mixture
was filtered to remove LiCl and the solvent was removed under
reduced pressure. Tris(pentafluorophenyl)boron was obtained as
white crystals at a yield of 25%. 19F NMR (400 MHz, CD2Cl2): d
ꢁ128 (br s, 6F, o-F), ꢁ145 (t, 3F, p-F), ꢁ161 (m, 6F, m-F) ppm.
g
combination of NMR and UV–vis spectroscopic techniques, Zr–allyl
complexes have also been shown to form during 1-hexene poly-
merization by rac-[Me2Si(1-indenyl)2ZrMe2] activated with trityli-
um perfluorotetraphenylborate in the presence of trimethyl-
aluminum and to constitute about 90% of the total catalyst [36].
In this paper, the results of a series of in situ propylene poly-
merization reactions as catalyzed by Cp2ZrMe2/B(C6F5)3 [9,10]
and Cp2ZrMe2/[Ph3C][B(C6F5)4] [37], respectively, are described.
The aim of this study is to obtain new experimental evidence that
zirconium allyl intermediates form during zirconocene-catalyzed
propylene polymerization and to obtain further insights about
the chemistry involved in this and in other transformations that
might occur during polymerization. A detailed understanding of
such transformations is of great importance in the metallocene-
catalyzed polymerization of olefins.
2.4. In situ propylene polymerization by [Cp2ZrMe][MeB(C6F5)3]
A solution of Cp2ZrMe2 (40 lmole) in C6D5Cl (0.5 mL) was
added to a solution of B(C6F5)3 (1.1 equiv) in C6D5Cl (0.5 mL) and
the yellow ion pair solution thus formed was transferred to an
NMR tube. The NMR tube was sealed with a rubber septum and
Parafilm and was removed from the glove box, and the ion pair
solution was characterized by 1H NMR spectroscopy. After the 1H
NMR, the predetermined amount of propylene was injected into
the NMR tube via a syringe at ambient temperature and the NMR
tube was shaken vigorously, after which a new 1H NMR spectrum
was acquired.
2. Experimental
2.1. General considerations
All chemicals were purchased from Aldrich unless otherwise
stated. Deuterated chlorobenzene was purchased from Cambridge
Isotope Laboratories (>99% atom% D) and dried by vacuum distilla-
tion from CaH2, stored over molecular sieves, and handled in a
glove box. Chlorobenzene was dried by distillation under argon
from CaH2 prior to use. Polymerization-grade propylene
(99.5 wt% purity, liquid phase, Praxair) was dried by passage
through a column of activated 4 Å molecular sieves prior to use.
2,4-Dimethyl-1-pentene was purchased from ChemSampCo.
Cp2ZrMe2 and B(C6F5)3 were synthesized as described below.
Handling and storage of air-/moisture-sensitive organometallic
compounds was done using an MBraun LABmaster glove box.
NMR spectra were run on a Bruker AV 600 spectrometer, chem-
ical shifts being referenced using the residual proton signals of the
deuterated chlorobenzene. Probe temperatures were calibrated
using methanol (low-temperature) and ethylene glycol (high-tem-
perature) samples as references. 1H NMR spectra were acquired
with a 45° pulse and 1 s delay between pulses; 16 transients were
stored for each spectrum. ESI-MS/MS experiments were performed
on an MDS Sciex QSTAR XL QqTOF mass spectrometer in ESI posi-
tive mode and were run by Dr. Bernd O. Keller.
2.5. In situ propylene polymerization by [Cp2ZrMe][B(C6F5)4]
A solution of Cp2ZrMe2 (40 lmole) in C6D5Cl (0.5 mL) was
added to a solution of [Ph3C][B(C6F5)4] (1.1 equiv) in C6D5Cl
(0.5 mL) and the orange ion pair solution thus formed was trans-
ferred to an NMR tube. The NMR tube was sealed with a rubber
septum and Parafilm and removed from the glove box, and the
ion pair solution was characterized by 1H NMR spectroscopy. The
NMR tube was heated at 40 °C for about 30 min and a new 1H
NMR spectrum was acquired. The predetermined amount of pro-
pylene was injected into the NMR tube via a syringe at ambient
temperature and the NMR tube was shaken vigorously, after which
another 1H NMR spectrum was acquired.
2.6. In situ reaction between [Cp2ZrMe][B(C6F5)4] and 2,4-dimethyl-1-
pentene
A solution of Cp2ZrMe2 (40 lmole) in C6D5Cl (0.5 mL) was
added to a solution of [Ph3C][B(C6F5)4] (1.1 equiv) in C6D5Cl
(0.5 mL) and the orange ion pair solution thus formed was trans-
ferred to an NMR tube. The NMR tube was sealed with a rubber
septum and Parafilm and removed from the glove box. Before the
NMR spectrum was determined, 2,4-dimethyl-1-pentene (1.0–1.5
2.2. Synthesis of dimethylzirconocene
Dimethylzirconocene was prepared using a procedure similar to
that described in the literature [38,39]. Methyllithium (9 mL, 1.4 M

114
M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
equiv) was injected into the NMR tube at ambient temperature via
a microsyringe. The reaction products could not be isolated and
were characterized in solution by 1H NMR and correlation
spectroscopy.
via a syringe at ambient temperature. The vial was quickly shaken,
the reaction mixture was further diluted with an additional 1 mL of
dry chlorobenzene and shaken again. The resulted reaction mix-
ture was then fed continuously into the first quadrupole of a
QSTAR XL QqTOF mass spectrometer, under an atmosphere of
dry nitrogen, via a syringe. The flow rate could be varied between
2.7. NMR spectroscopic data of [Cp2Zr(g3
-
5 and 20 lL/min.
CH2C(CH2CHMe2)CH2)][B(C6F5)4]
1H NMR (600 MHz, C6D5Cl, 0 °C): d 5.71 (s, C5H5, 5H), 5.60 (s,
C5H5, 5H), 3.38 (s, allyl Ha, 2H), 2.56 (s, allyl Hs, 2H), 1.96 (d, CH2,
3. Results and discussion
3
3JHH = 6.5 Hz, 2H), 1.76 (m, CH, 1H), 1.01 (d, CH3, JHH = 6.5 Hz,
6H) ppm.
3.1. Reaction of propylene with [Cp2ZrMe][MeB(C6F5)3]
13C NMR (600 MHz, C6D5Cl, 0 °C): d 112.8 (C5H5, 1JC–H = 171 Hz),
1
1
d 111.2 (C5H5, JC–H = 171 Hz), 68.5 (allyl CH2, JC–Ha = 145 Hz,
The [Cp2ZrMe][MeB(C6F5)3] ion pair (1), generated in situ by
reacting Cp2ZrMe2 with a slight excess (1.1 equiv) of B(C6F5)3 in
chlorobenzene-d5 at ambient temperature, was reacted with
5–10 equiv of propylene. The 1H NMR spectra of the reaction mix-
tures recorded following addition of propylene reveal rapid and
quantitative conversion of propylene to short-chain atactic poly-
propylene oligomers, as indicated by the presence of the backbone
–[CH2CH(Me)]n– resonances in the aliphatic (d 0.7–2.2) region of
the spectra, while free propylene [42] resonances are missing
(Fig. 1). Small quantities of unreacted 1 (d 5.97 (Cp) and d 0.56
(Zr–Me)) are still present when 5 equiv of propylene are used,
which indicates that chain propagation is faster than initiation
(i.e., the first propylene insertion into the Zr–Me bond). The
Me–B (d 0.34) resonance is not evident any more, and it seems rea-
sonable that it overlaps with the backbone polypropylene oligomer
resonances after being shifted downfield toward a free [MeB
(C6F5)3]ꢁ (d 1.1–1.2) resonance [33]. The 1H NMR spectra of the
reaction mixtures also show the presence of a series of low-
relative-intensity resonances in the Cp region between d 5.9 and
6.4, in addition to the olefinic group resonances [43,44] present
between d 4.8 and 5.4. While no attempt to identify these reso-
nances was made, it seems reasonable that they belong to the Cp
hydrogens of various Cp2Zr+–polypropenyl species present in the
reaction mixture (vide infra).
1JC–Hs = 156 Hz), 163.6 (allyl C), 53 (CH2, JC–H = 135 Hz), 31.2 (CH,
1
1JC–H = 127 Hz), 23 (CH3, JC–H = 129 Hz) ppm.
1
2.8. NMR spectroscopic data of [Cp2ZrMe][MeB(C6F5)3]
1H NMR (600 MHz, C6D5Cl, 25 °C): d 5.97 (s, C5H5, 10H), 0.56 (s,
Zr–CH3, 3H), 0.34 (br.s, B–CH3, 3H) ppm.
1H NMR (600 MHz, C6D5Cl, 0 °C): d 5.93 (s, C5H5, 10H), 0.55 (s,
Zr–CH3, 3H), 0.33 (br.s, B–CH3, 3H) ppm.
2.9. General procedure for variable-temperature NMR experiments
The procedure used for variable-temperature NMR experiments
is similar to that described in the literature [33]. Samples of [Cp2-
Zr+( 3-C3H4)CH2CHMe2)][B(C6F5)4] used for variable-temperature
g
1H NMR studies were prepared as described above. Samples were
brought to the indicated temperature and allowed to equilibrate
for 7–10 min before the 1H NMR spectra were acquired. All of the
temperature-dependent changes are reversible; i.e., on cooling
the probe temperature, the reverse chemical shift changes occur.
Free energy of activation,
equation
D
Gà, was calculated in kJ/mol from the
D
Gà = R ꢂ T[23.76 ꢁ ln(k/T)], where R is the gas constant
(8.3145 J/mol K), T is the absolute temperature (K), and k is the rate
constant (sꢁ1) [41].
Notable, however, are the low-intensity, exchange-broadened
resonances detected in the allylic (d 2.5–4.8) and Cp (d ꢀ 5.7)
regions of the 1H NMR spectra and a sharp singlet at d 0.25 attrib-
utable to methane. Owing to the close similarity to the previously
reported 1H NMR spectra of related complexes used as model com-
pounds [33], these resonances were attributed to the allylic and Cp
hydrogens, respectively, of various oligomeric Zr–allyl intermedi-
2.10. General procedure for ESI-MS/MS experiments
A
solution of Cp2ZrMe2 (40 lmole) in dry chlorobenzene
(0.5 mL) was added to a solution of B(C6F5)3 (1.1 equiv) in dry chlo-
robenzene (0.5 mL) and the yellow ion pair solution thus formed
was transferred to a vial. The vial was sealed with a rubber septum
and Parafilm and removed from the glove box and a predetermined
amount of propylene (5–10 equiv) was injected into the NMR tube
ates of the type [Cp2Zr(g g
3-CH2C(CH2P)CH2)]+ and [Cp2Zr( 3-CH2
C(Me)CHP)]+ (P = short chain alkyl or polymeryl group) formed
during the propylene oligomerization reaction.
+
[Cp2ZrMe]
+
[Cp2ZrMe]
CH4
4.6 4.2 3.8 3.4 3.0 2.6
(a)
4.6 4.2 3.8 3.4 3.0 2.6
(b)
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
ppm
Fig. 1. 1H NMR spectra of in situ propylene oligomerization by 1 in chlorobenzene-d5 (ambient temperature, 600 MHz): (a) after addition of 5 equiv of propylene, (b) after
addition of 10 equiv of propylene. Insets show the allylic region.

M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
115
Fig. 2. Positive mode ESI-MS spectra of molecular ions of m/z 403 corresponding to [Cp2Zr(C3H6)4CH3]+ and m/z 429 corresponding to [Cp2Zr(C3H4)(C3H6)4H]+.
In any case, the allylic region of the 1H NMR spectra in Fig. 1
looks complicated because of the presence of different chain length
Cp2Zr+–allyl oligomers, in addition to the inter- and intramolecular
exchange processes [33] associated with these zirconium allyl
complexes in solution, and as a result, clear identification of partic-
ular zirconium allyl intermediates by 1H NMR spectroscopy is dif-
ficult. To identify individual oligomeric Cp2Zr+–allyl species
present in the catalytic system, the reaction of propylene with 1
was also investigated using electrospray ionization (tandem) mass
spectrometry. Unlike 1H NMR spectroscopy, this technique allows
the identification of individual Cp2Zr+-containing intermediates
present in a reaction mixture by mass-to-charge ratio (m/z).
In a typical experiment, 5 equiv of propylene was added to a
solution of 1 (4 ꢃ 10ꢁ2 M) in chlorobenzene at ambient tempera-
ture. The resulted reaction mixture was further diluted twofold
with chlorobenzene and then fed continuously into the first quad-
rupole of a QSTAR XL QqTOF mass spectrometer via an electrospray
ionization interface. The electrospray ionization technique is
known to release molecular ions preformed in solution with very
little or no fragmentation [45,46]. The electrospray ionization mass
spectrometry (ESI-MS) spectra of the reaction mixture obtained in
the positive ion mode by scanning the first quadrupole were com-
plex and of low quality because of many overlapping peaks of the
Cp2Zr+-containing species with only small differences in their
molecular ion masses.
(C3H4)(C3H6)nCH3 (n = 0–5) species, respectively, selected from
the first quadrupole, together with detailed electrospray ionization
tandem mass (ESI-MS/MS) spectra of two selected molecular ions.
As can be seen from the insets of Fig. 3, the selected molecular
ions of m/z 277, corresponding to Cp2Zr+(C4H9), and m/z 317, cor-
responding to Cp2Zr+(C7H13), show loss of the alkyl and allyl
ligands, respectively, to yield a signal of m/z 220 corresponding
to the [Cp2Zr]+ fragment ion. The [Cp2Zr]+ fragment ion is diagnos-
tic for identification of a zirconocene complex [46,47]. Also present
in the Cp2Zr+(C4H9) and Cp2Zr+(C7H13) tandem mass spectra are the
molecular ions of m/z 295 and 335, which could be attributed to
the [Cp2Zr(C4H9)(H2O)]+ and [Cp2Zr(C7H13)(H2O)]+ adduct ions,
respectively, and a molecular ion of m/z 237 corresponding to
the [Cp2ZrOH]+ complex [48]. These are byproducts formed in
the collision cell from the reaction of the corresponding zircono-
cene cations with traces of water present in the instrument,
½Cp ZrðC4H9Þꢄþ þH2O ! ½Cp ZrðC4H9ÞðH2OÞꢄþ
!
2
2
m=z 277
m=z 295
! ½Cp ZrOHꢄþ þC4H10
;
ð1Þ
2
m=z 237
½Cp ZrðC7H13Þꢄþ þH2O ! ½Cp ZrðC7H13ÞðH2OÞꢄþ
!
2
2
m=z 317
m=z 335
! ½Cp ZrOHꢄþ þC7H14
;
ð2Þ
2
Indeed, a typical ESI-MS spectrum of a cationic zirconocene
complex displays a distinct isotopic pattern of seven peaks, due
to the contributions of the five principal natural isotopes of zirco-
nium and along with the natural isotopes of carbon and hydrogen
[47]. Shown in Fig. 2 are the positive mode ESI-MS spectra for
peaks at m/z 403 and 429 with isotopic patterns corresponding
to [Cp2Zr(C3H6)CH3]+ and [Cp2Zr(C3H4)(C3H6)4H]+, respectively.
Nevertheless, due to the large variety of Cp2Zr+-containing reaction
intermediates that can form during the polymerization, which can
differ by an m/z less than 7, extensive overlapping of peaks
occurred. In addition, the complexity of the ESI-MS spectra was
further enhanced by some decomposition of the reaction interme-
diates due to the traces of water and other contaminants present in
the instrument. Therefore, direct identification of various oligo-
meric zirconocene intermediates present in the reaction mixture
on the basis of their natural isotope distribution pattern is difficult.
In order to identify the cationic Cp2Zr+ intermediates of interest,
specific m/z peaks corresponding to particular molecular zircono-
cene complexes were isolated from the first quadrupole and
further examined by collision-induced dissociation (CID) experi-
ments. A single m/z value, corresponding to the cationic Cp2Zr+-
containing species having the most abundant 90Zr isotope, was
selected for each particular zirconocene complex. Fig. 3 displays
the molecular ion distribution for the cationic zirconocene alkyl
Cp2Zr+(C3H6)nCH3 (n = 1–6) and cationic zirconocene allyl Cp2Zr+
m=z 237
and their presence in these spectra, although unfortunate, also
attests the presence of the cationic zirconocene species of interest.
The ESI-MS/MS spectra of all selected molecular ions are similar in
nature. Likewise, cationic Cp2Zr+(C3H6)nH (n = 1–5, m/z 263, 305,
347, 389, 431) and Cp2Zr+(C3H4)H (m/z 261) were also detected.
Note that traces of water and other contaminants present in the
instrument would inevitably react with both cationic Cp2Zr+–allyl
and Cp2Zr+–alkyl complexes, thus lowering their molecular ion
peak intensities in the acquired ESI-MS/MS spectra. On the other
hand, a small contribution to the intensity of oligomeric Cp2Zr+–
allyl peaks can come from a Cp2Zr+–polypropenyl species bearing
an internal unsaturation (vide infra). Hence, this investigation
focuses only on qualitative identification of the oligomeric zircono-
cene allyl and alkyl intermediates present in the reaction mixture
and does not provide quantitative information about these species.
Cationic Cp2Zr+–polymeryl intermediates have been observed
previously in gas-phase reaction of Cp2Zr+–Me with ethylene and
1-butene, respectively, by ESI-MS and ESI-MS/MS [46,49]. Also, cat-
ionic Cp2Zr+(allyl) complexes were formerly detected by tandem
mass spectrometry as the main reaction products of corresponding
Zr–polymeryl species in the CID experiments via loss of H2 [46].
However, the direct identification of oligomeric Cp2Zr+–polyprope-
nyl and Cp2Zr+–allyl intermediates preformed in solution under

116
M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
Fig. 3. The molecular ion distributions for (a) Cp2Zr+(C3H6)nCH3 (n = 1–6) and (b) Cp2Zr+(C3H4)(C3H6)nCH3 (n = 0–5). Insets in (a) and (b) show fragmentation of ions of m/z
277 and 317, respectively.
catalytic conditions by electrospray ionization (tandem) mass
spectrometry has not been reported previously.
addition of 10 equiv of propylene, here a majority (about 60% as
determined by 1H NMR spectroscopy using Ph3CMe, present in
the reaction solution, as an internal standard) of the initial catalyst
[Cp2ZrMe]+ remained unconsumed (Fig. 4), consistent with prior
observations that the propagation rate is higher when an anion
with a weaker coordinating ability is used [51,52].
3.2. Reaction of propylene with [Cp2ZrMe][B(C6F5)4]
The analogous reaction engaging [Ph3C][B(C6F5)4] as a cocata-
lyst was similarly investigated by 1H NMR spectroscopy. Cp2ZrMe2
was reacted with 1.1–1.2 equiv of [Ph3C][B(C6F5)4] in chloroben-
zene-d5 at ambient temperature to generate the corresponding
ion pair [Cp2ZrMe][B(C6F5)4] (2) and Ph3CMe. Small quantities of
As can be seen in Fig. 4, aside from the [Cp2ZrMe]+ (d 5.99 (Cp)
and 0.78 (Zr–Me)), Ph3CMe (d 2.13 (Me) and d 7.1–7.4 (Ph)) and the
expected backbone –[CH2CHMe]n– (d 0.9–2.0) resonances, the 1H
NMR spectra of the reaction mixtures obtained following addition
of propylene also show the presence of two low-intensity reso-
nances with an approximately 1:1 ratio of intensities at ꢀd 5.7
and 5.8, a series of weak resonances in the allyl region between d
2.5 and 4.7, and a resonance at d 0.25 ascribed to methane. On
the basis of comparison with literature precedents [33], in addition
to the above-described 1H NMR spectra (Fig. 1), the resonances
present in the Cp (ꢀd 5.7, 5.8) and allyl (d 2.5–4.7) regions were
attributed to the Cp and allyl hydrogens of oligomeric Cp2Zr+–allyl
species existing in the reaction mixture. However, the observation
methyl-bridged binuclear cations [(Cp2ZrMe)2(l
-Me)]+ were also
formed [8,50], and they were slowly converted to the mononuclear
species 2 and the catalyst precursor Cp2ZrMe2 by warming the
NMR tube at 40 °C for 30 min [50]. Treatment of a chloroben-
zene-d5 solution of 2 with 5–10 equiv of propylene at ambient
temperature resulted in rapid and complete conversion of propyl-
ene to low-molecular-weight polypropylene chains; however,
compared with the previous example where the initial cationic
zirconocene catalyst [Cp2ZrMe]+ was basically consumed after

M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
[Cp2ZrMe]+
117
Ph3CMe
Ph3CMe
[Cp2ZrMe]+
4.6
4.2
3.8
3.4
3.0
2.6
CH4
(a)
4.6
4.2
3.8
3.4
3.0
2.6
(b)
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5 ppm
Fig. 4. 1H NMR spectra of in situ propylene oligomerization by 2 in chlorobenzene-d5 (ambient temperature, 600 MHz): (a) after addition of 5 equiv of propylene, (b) after
addition of 10 equiv of propylene. Insets show the allylic region.
of two Cp resonances for the oligomeric Cp2Zr+–allyl intermediates
in the present case stands in marked contrast with the preceding
findings (vide supra) in which only one broad Cp resonance was
group of a polypropylene chain released in solution following
b-hydride elimination) to [Cp2ZrR]+ (R = Me, H, or polymeryl
group), followed by intramolecular proton transfer from C to the
c
observed, and implies a relatively static structure of the
g
3-allyl
R ligand to yield a cationic zirconocene allyl complex Cp2Zr+–allyl
and RH. Such reaction pathways find a precedent in alkene poly-
merization by organolanthanide complexes [22] as well as in reac-
tions of 1,1-disubstituted alkenes CH2@C(Me)(R) (R = Me, alkyl
chain) with cationic zirconocene catalysts [24,25,33,35]. In path-
way (c), which is supported by both computational [26,27] and
experimental [34] studies, the coordinated alkene of a cationic zir-
conocene(hydride)(alkene) complex, obtained following b-hydride
elimination of a cationic zirconocene–polypropenyl intermediate,
ligand of the oligomeric Cp2Zr+–allyl species with respect to the
Cp2Zr+ unit under these conditions. This difference in dynamics
of the allyl ligands of oligomeric Cp2Zr+–allyl complexes could be
related to the different coordination ability of the anions as well.
To determine whether the anion actually has such an effect on
the dynamic behavior of the Cp2Zr+–allyl species, the in situ reac-
tion between the catalyst 2 and 2,4-dimethyl-1-pentene (1.0–1.5
equiv), used as a model compound for a vinylidene-terminated
polymer chain, was also performed under otherwise identical con-
ditions. As expected, the 1H NMR spectrum of the resulted zircono-
does not dissociate but further transfers a
c-H to the hydride
ligand to form a cationic zirconocene allyl species and release a
cene allyl complex [Cp2Zr+( 3-C3H4)CH2CHMe2)][B(C6F5)4] (3), at
g
dihydrogen molecule.
ambient temperature, is similar to that of the [MeB(C6F5)3]ꢁ ana-
logue [33] except for the Cp resonances of 3, which appear as
two sharp equal-intensity resonances (d 5.76 and 5.65) and not
as only one broad Cp resonance (d 5.70) [33], as it was observed
in the latter case. These results confirm that the allyl ligand of 3
Here, the detection of methane in the 1H NMR spectra, along
with the observation of oligomeric Cp2Zr+–allyl complexes in both
the 1H NMR and electrospray ionization (tandem) mass spectra,
suggests that the Cp2Zr+–allyl species were formed by (re)coordi-
nation of alkene to the cationic zirconocene catalyst and subse-
is indeed
g
3-coordinated and relatively static at this temperature.
quent intramolecular c-H transfer from the coordinated alkene to
When the sample temperature is increased, the coalescence of the
Cp resonances occurs at about 60 °C. This translates into a free
energy of activation for the intramolecular exchange processes of
3 of 68.6 kJ/mol, approximately 9 kJ/mol higher than the value esti-
mated for the [MeB(C6F5)3]ꢁ analogue [33]. It seems likely that the
increased free energy of activation in this case can be correlated
with the increased electronegativity at the metal center, induced
by the more weakly coordinating, less nucleophilic anion
([B(C6F5)4]ꢁ vs. [MeB(C6F5)3]ꢁ) [53–55], which would cause the
R (pathways (a) and (b) in Scheme 2). However, an additional reac-
tion pathway in which b-H elimination of Cp2Zr+–polymeryl is fol-
lowed directly by C–H activation, i.e., without prior alkene
dissociation/recoordination (pathway (c) in Scheme 2), cannot be
ruled out.
Indeed, recent studies [33] have demonstrated that vinylidene-
terminated compounds, such as 2,4-dimethyl-1-pentene and 2,4-
dimethyl-1-heptene, do react with 1 to form methane and cationic
zirconocene allyl complexes. Remarkably, when 2 was employed
as a catalyst in the reaction with 2,4-dimethyl-1-pentene [56],
detailed low-temperature one- and two-dimensional 1H NMR
spectroscopy experiments revealed that the reaction occurs via
the transient [Cp2Zr(Me)(CH2@C(Me)CH2CHMe2)]+ species and
that the coordinated alkene of this species assumes a near-g1
structure (i.e., one in which Cb is almost carbocationic in nature)
rather than the conventional g2 structure. In addition, the study
also unveiled that the two terminal vinylidene hydrogens of the
coordinated alkene undergo intramolecular exchange. The latter
g
3-allyl ligand of 3 to bind more tightly to the metal center, thus
diminishing its fluxionality.
3.3. Formation of Zr–allyl species
Three plausible pathways for the formation of Cp2Zr+–allyl spe-
cies during propylene polymerization are depicted in Scheme 2.
The first two, pathways (a) and (b), are somewhat similar and con-
sist of coordination of alkene (propylene or the vinylidene end

118
M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
Scheme 2. Possible reaction pathways for formation of Cp2Zr+–allyl species during zirconocene-catalyzed propylene polymerization.
was suggested to occur through the carbocationic complex
Cp2Zr(Me)CH2C+(Me)CH2CHMe2.
Cp2Zr(Me)CH2C+(Me)(CH2CHMe2) also results in a reversal of the
alkene enantioface relative to zirconium. In the context of propyl-
ene polymerization by metallocene catalysts, this can constitute a
mechanism for chain end epimerization (see Scheme 3). Incorpora-
tion of stereoerrors within a polymer chain via the intermediacy of
a carbocationic intermediate is expected to become predominant
when the concentrations of propylene are reduced, consistent with
earlier observations that polypropylene stereospecificity decreases
when the polymerizations are run at low propylene concentrations
[28,31]. This mechanism for stereoerror formation was not consid-
ered previously.
These observations are of fundamental interest, as they indicate
that the vinylidene end groups of the polymer chains released in
solution following b-hydride elimination reactions, although steri-
cally unfavorable, can also recoordinate to the vacant site at the
metal center in
a
near g1 structure to form [Cp2Zr(R)
(CH2@C(Me)–polymeryl)]+. The coordinated vinylidene group of
the resulting metal–alkyl–alkene intermediate cannot insert, most
likely because of steric hindrance. However, the strong polarization
of the double bond in the near-g1 coordinated vinylidene group
allows rotation about the C –Cb bond via the carbocationic com-
A kinetic study [57] on formation of cationic zirconium allyl
species has shown that the reactions of 2,4-dimethyl-1-pentene
with either 1 or 2 to form Cp2Zr+–allyl species in chlorobenzene-
d5, follow a second order rate expression given by the equation
a
plex Cp2Zr(R)CH2C+(Me)–polymeryl; the protons of the groups
attached to Cb become strongly acidic, and therefore they can be
more easily transferred to the R group on the metal center to form
the corresponding Zr–allyl species and RH [56]. A corollary to this
is that formation of metal–allyl intermediates in metallocene-cat-
ꢁd½Cp2ZrþMeꢄ=dt ¼ k½Cp2ZrþMeꢄ½2; 4-dimethyl-1-penteneꢄ;
ð3Þ
alyzed polymerizations of higher a-olefins, i.e., those in which one
hydrogen atom on C is replaced by an alkyl group, would be more
difficult for both steric and electronic reasons.
Noteworthily, aside from the intramolecular exchange of the
two terminal vinylidene hydrogens of the coordinated alkene, the
where k is the temperature-dependent rate constant for the reac-
tion. The study indicated that the formation of Cp2Zr+–allyl species
occurs with an activation energy between 46.4 and 50.9 kJ/mol
[56,57]. Theoretical [58–60] and kinetic [61] studies on olefin poly-
merization by homogeneous metallocene catalysts, on the other
c
rotation about the C –Cb bond of the vinylidene group of [Cp2-
a
Zr(Me)(CH2@C(Me)CH2CHMe2)]+ through the carbocationic species
hand, showed that the insertion of olefin into Zr–C
r-bonds during
Scheme 3. Probable mechanism for incorporation of stereoerrors into the polymer chain (b-hydride elimination, rotation about C –Cb bond, reinsertion).
a

M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
119
the propagation polymerization occurs with an activation energy
between 25 and 30 kJ/mol. In this context, assuming that the pre-
exponential factors are of about the same magnitude, the kinetic
data suggest that the formation of Zr–allyl intermediates during
the polymerization under the present reaction conditions may
occur with a relative probability as often as one out of every
1000, as compared with olefin insertion, especially in systems
where noncoordinating anions are used.
between d 4.78 and 4.94 were attributed to the vinylidene end
groups and possibly internal vinylidene groups of different
lengths, short-chain atactic polypropylene oligomers. A terminal
vinylidene group arises from b-hydride elimination following a
primary (1,2) propylene insertion into Zr–alkyl chains, while an
internal vinylidene group is the result of propylene insertion into
a Zr–allyl bond [44,53,54]. The resonance at d ꢀ 5.2 was attrib-
uted to the vinylic proton of an internal unsaturation group of
the type –CH2CH2CH@C(CH3)– [44,66]. As noted by Resconi
et al. [44,46], a similar triplet resonance at d ꢀ 5.2 has been
observed previously in the 1H NMR spectra of isotactic polypro-
pylene. For the reason that this resonance was noticed to arise
only in conjunction with the cis-2-butenyl end group resonance
at d ꢀ 5.4 [65], it was proposed to be connected to the presence
of secondary (2,1) insertions [44,66]; yet no mechanism has been
suggested. A plausible mechanism that would account for the
occurrence of an internal olefinic group of the type –CH2CH2-
CH@C(CH3)– on the polymer chain is presented in Scheme 4
and would involve the presence of a Zr–allyl species in addition
to a secondary (2,1) Zr–alkyl intermediate.
3.4. Analysis of polymer chain unsaturation
Studies on the reactivity of Cp02Zr+–allyl (Cp0 =
g
5-cyclopentadi-
enyl type group) complexes conducted by Lieber et al. [25] and
Landis and Christianson [34] indicated that insertion of propylene
and 1-hexene, respectively, proceeds much more slowly into a Zr–
allyl bond than into a Zr–alkyl bond. Theoretical calculations by
Ziegler and co-workers [26,27] showed that insertion of ethylene
into a Zr–allyl bond can occur without substantial delay, and the
authors have argued that for higher
a-olefins the insertion could
be more difficult. Similar results were also obtained experimen-
tally by Temme et al. [62], who found that a zirconocene–allyl
borate–betaine system catalyzes the polymerization of ethylene
more efficiently than that of propylene, whereas van der Heijden
However, while isotactic polypropylene is known to incorporate
small amounts of regiodefects as a result of occasional secondary
propylene insertions [44], so far there are no reports indicating
the presence of secondary insertions in atactic polypropylene
[15]. An alternative mechanism that might explain the incorpora-
tion of an internal olefinic group of the type –CH2CH2CH@C(Me)(R)
(R = alkyl group) into the polymer chain is shown in Scheme 5.
Here, the cationic Cp2Zr+–CH2CH(Me)CH2CH2CH3 species, obtained
following two consecutive primary (1,2) insertions of propylene
into the Cp2Zr+–H bond, undergoes b-hydride elimination followed
by allylic C–H activation of the coordinated vinylidene chain end at
et al. [63] found that [(C5Me5)(C5H4CMe3)Zr(
is reactive toward ethylene but does not react with
g
3-CH2C(Me)CH2)]+
-olefins.
a
Subsequent propylene insertion into the Zr–allyl bond of an
oligomeric Cp2Zr+–allyl intermediate would invariably lead to
incorporation of internal unsaturation within the polymer chain
[64,65]. Therefore, an analysis of polymer chain unsaturation
would indicate whether the Cp2Zr+–allyl intermediates formed
during catalysis are reactive toward propylene.
the methylene group to form a cationic Cp2Zr+( 3-CH2C(Me)
g
The olefinic region of the 1H NMR spectra of the reaction mix-
tures obtained from propylene polymerization by 1 (Fig. 1) shows
the presence of multiple resonances between d 4.78 and 4.94 as
well as of a triplet resonance at d ꢀ 5.2 (JHH = 7 Hz). On the basis
of comparison with literature data [43,44], the resonances
CHCH2CH3) species. Subsequent propylene insertion into the new
Zr–allyl bond would result in a polymer chain with a CH3CH2
CH@C(Me)– end group. The presence of a 3-propenyl end group
on polypropylene chains has not been considered previously.
Scheme 4. Possible mechanism for formation of an internal olefinic group of the type –C(Me)@CHCH2CH2CH2–.
Scheme 5. Possible mechanism for formation of a CH3CH2CH@C(Me)– end group.

120
M. Vatamanu / Journal of Catalysis 323 (2015) 112–120
[5] W. Kaminsky, J. Chem. Soc., Dalton Trans. (1998) 1413
Thus, this latter mechanism shows that the presence of a triplet
[6] L. Resconi, L. Cavallo, A. Fait, F. Piemontesi, Chem. Rev. 100 (2000) 1253.
[7] M. Bochmann, J. Chem. Soc., Dalton Trans. (1996) 255
[8] E.Y.-X. Chen, T.J. Marks, Chem. Rev. 100 (2000) 1391.
[9] X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 113 (1991) 3623.
[10] X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 116 (1994) 10015.
[11] J.C.W. Chien, W.-M. Tsai, M.D. Rausch, J. Am. Chem. Soc. 113 (1991) 8570.
[12] X. Yang, C.L. Stern, T.J. Marks, Organometallics 10 (1991) 840.
[13] J.C.W. Chien, R. Sugimoto, J. Polym. Sci., Part A: Polym. Chem. 29 (1991) 459.
[14] M. Bochmann, Top. Catal. 7 (1999) 9.
[15] M. Vatamanu, B.N. Boden, M.C. Baird, Macromolecules 38 (2005) 9944.
[16] K.A. Novstrup, N.E. Travia, G.A. Medvedev, C. Stanciu, J.M. Switzer, K.T.
Thomson, W.N. Delgass, M.M. Abu-Omar, J.M. Caruthers, J. Am. Chem. Soc. 132
(2010) 558.
[17] B.M. Moscato, B. Zhu, C.R. Landis, Organometallics 31 (2012) 2097.
[18] V. Busico, R. Cipullo, J.C. Chadwick, J.F. Modder, O. Sudmeijer, Macromolecules
27 (1994) 7538.
[19] V. Busico, R. Cipullo, G. Talarico, L. Caporaso, Macromolecules 31 (1998) 2387.
[20] S. Lin, R. Kravchenko, R.M. Waymouth, J. Mol. Catal. A: Chem. 158 (2000) 423.
[21] V. Busico, R. Cipullo, S. Ronca, Macromolecules 35 (2002) 1537.
[22] G. Jeske, H. Lauke, H. Mauermann, P.N. Swepston, H. Schumann, T.J. Marks, J.
Am. Chem. Soc. 107 (1985) 8091.
[23] J.J.W. Eshuis, Y.Y. Tan, A. Meetsma, J.H. Teuben, J. Renkema, G.G. Evens,
Organometallics 11 (1992) 362.
[24] A.D. Horton, Organometallics 15 (1996) 2675.
resonance at d ꢀ 5.2 in the 1H NMR spectra of polypropylene can be
connected with the presence of Zr–allyl intermediates only and
does not necessarily need the presence of a secondary insertion.
Indeed, the 1H NMR spectra of in situ propylene polymerization
by 1 (Fig. 1) also show the presence of allylic resonances between
d 2.5 and 4.8, in addition to the olefinic resonances between d 4.78
and 4.94. Also, a closer examination of the 1H NMR spectra of iso-
tactic polypropylene in references [44,46] indicates the presence of
internal vinylidene groups, which were ascribed to the presence of
Zr–allyl species during the polymerization, in addition to the trip-
let resonance at d 5.2; yet no connection between the presence of
the triplet at d 5.2 and that of internal vinylidene groups was made.
No olefinic hydrogen resonances were detected in the 1H NMR
spectra of the in situ reaction of propylene with 2 (see Fig. 4). This
observation, coupled with the observation that allylic hydrogen
resonances are present between d 2.5 and 4.7, implies that the
insertion of propylene into the Zr–allyl bond in this case is more
difficult, consistent with the finding that zirconium–allyl interac-
tion is stronger in this case (vide supra).
[25] S. Lieber, M.-H. Prosenc, H.-H. Brintzinger, Organometallics 19 (2000) 377.
[26] P.M. Margl, T.K. Woo, T. Ziegler, Organometallics 17 (1998) 4997.
[27] C. Zhu, T. Ziegler, Inorg. Chim. Acta 345 (2003) 1.
4. Summary
[28] M.K. Leclerc, H.-H. Brintzinger, J. Am. Chem. Soc. 118 (1996) 9024.
[29] M.-H. Prosenc, H.-H. Brintzinger, Organometallics 16 (1997) 3889.
[30] L. Resconi, J. Mol. Catal. A: Chem. 146 (1999) 167.
[31] J.C. Yoder, J.E. Bercaw, J. Am. Chem. Soc. 124 (2002) 2548.
[32] D.R. Sillars, C.R. Landis, J. Am. Chem. Soc. 125 (2003) 9894.
[33] M. Vatamanu, Organometallics 33 (2014) 3683.
[34] C.R. Landis, M.D. Christianson, Proc. Natl. Acad. Sci. USA 103 (2006) 15349.
[35] A. Al-Humydi, J.C. Garrison, M. Mohammed, W.J. Youngs, S. Collins, Polyhedron
24 (2005) 1234.
[36] D.E. Babushkin, V.N. Panchenko, H.-H. Brintzinger, Angew. Chem., Int. Ed. 53
(2014) 9645.
[37] E.P. Talsi, J.L. Eilertsen, M. Ystenes, E. Rytter, J. Organomet. Chem. 677 (2003)
10.
[38] E. Samuel, M.D. Rausch, J. Am. Chem. Soc. 95 (1973) 6263.
[39] W.E. Hunter, D.C. Hrncir, R.V. Bynum, R.A. Penttila, J.L. Atwood,
Organometallics 2 (1983) 750.
[40] A.G. Massey, A.J. Park, J. Organomet. Chem. 2 (1964) 245.
[41] H. Günther, NMR Spectroscopy: Basic Principles, Concepts, and Applications in
Chemistry, second ed., Wiley, Chichester, UK, 1995.
[42] G.R. Fulmer, A.J.M. Miller, N.H. Sherden, H.E. Gottlieb, A. Nudelman, B.M. Stoltz,
J.E. Bercaw, K.I. Goldberg, Organometallics 29 (2010) 2176.
[43] R.M. Silverstein, F.X. Webster, Spectrometric Identification of Organic
Compounds, sixth ed., Wiley, 1998.
[44] L. Resconi, I. Camurati, O. Sudmeijer, Top. Catal. 7 (1999) 145.
[45] S.R. Wilson, Y. Wu, Organometallics 12 (1993) 1478.
[46] L.S. Santos, J.O. Metzger, Angew. Chem., Int. Ed. 45 (2006) 977.
[47] C.S. Christ Jr., J.R. Eyler, D.E. Richardson, J. Am. Chem. Soc. 112 (1990) 596.
[48] A.A. Aksenov, C.S. Contreras, D.E. Richardson, J.R. Eyler, Organometallics 26
(2007) 478.
[49] D. Feichtinger, D.A. Plattner, P. Chen, J. Am. Chem. Soc. 120 (1998) 7125.
[50] M. Bochmann, S.J. Lancaster, Angew. Chem., Int. Ed. Engl. 33 (1994) 1634.
[51] M.-C. Chen, J.A.S. Roberts, T.J. Marks, J. Am. Chem. Soc. 126 (2004) 4605.
[52] J. Zhou, S.J. Lancaster, D.A. Walker, S. Beck, M. Thornton-Pett, M. Bochmann, J.
Am. Chem. Soc. 123 (2001) 223.
[53] L. Jia, X. Yang, C.L. Stern, T.J. Marks, Organometallics 16 (1997) 842.
[54] Y.-X. Chen, M.V. Metz, L. Li, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 120 (1998)
6287.
[55] F. Song, S.J. Lancaster, R.D. Cannon, M. Schormann, S.M. Humphrey, C.
Zuccaccia, A. Macchioni, M. Bochmann, Organometallics 24 (2005) 1315.
[56] M. Vatamanu, G. Stojcevic, M.C. Baird, J. Am. Chem. Soc. 130 (2008) 454.
[57] M. Vatamanu, The activation parameters for Zr–allyl formation from 1 and 2,4-
Cationic Cp2Zr+–allyl species were detected in the reaction mix-
ture of Cp2ZrMe2 activated by either B(C6F5)3 or [Ph3C][B(C6F5)4]
and propylene by 1H NMR spectroscopy. Furthermore, short-chain
oligomeric Cp2Zr+–polymeryl and Cp2Zr+–allyl species were also
detected, for the first time, directly from the reaction mixture of
Cp2ZrMe2 activated by B(C6F5)3 and propylene by electrospray ion-
ization (tandem) mass spectrometry. The formation of Cp2Zr+–allyl
species occurs via coordination of alkene (propylene and vinyli-
dene end groups of unsaturated polymer chains, respectively) to
the active catalyst Cp2Zr+R (R = CH3, H, polymeryl chain), followed
by intramolecular proton transfer from C
c to R and release of RH;
however, an additional mechanism in which b-H elimination of
Cp2Zr+–polymeryl is followed directly by
c-H elimination to form
Cp2Zr+–allyl along with H2 cannot be discounted.
Analysis of the olefinic region of the 1H NMR spectra of the reac-
tion mixture obtained from 1 and propylene reveals the presence
of a triplet resonance at d ꢀ 5.2. This was attributed to the olefinic
proton of the unprecedented 3-propenyl end group. A plausible
mechanism for the origin of the 3-propenyl end groups of the
unsaturated polymer chains is discussed. Additionally, a mecha-
nism for the incorporation of stereodefects in stereoregular poly-
mers, via the intermediacy of the carbocationic species
Cp2Zr(H)(CH2C+(Me)–polymeryl), is advanced. The cationic Cp2Zr+–
allyl intermediates formed during polymerization may contribute
to catalyst deactivation.
Acknowledgments
Financial support from Queen’s University is gratefully
acknowledged. I thank Dr. Francoise Sauriol for her assistance with
NMR spectroscopy and Dr. Bernd O. Keller for his help with mass
spectrometry. I also thank Prof. Richard F. Jordan (University of
Chicago), Prof. Clark R. Landis (University of Wisconsin—Madison),
and Prof. Natalie M. Cann for helpful discussions.
dimethyl-1-pentene are
D
Hà = 48.3 kJ/mol and
D
Sà = ꢁ72.5 J/mol K, while
those for Zr–allyl formation from
2 and 2,4-dimethyl-1-pentene are
D
Hà = 44.4 kJ/mol and
D
Sà = ꢁ42.4 J/mol K (unpublished data).
[58] T. Yoshida, N. Koga, K. Morokuma, Organometallics 14 (1995) 746.
[59] T. Yoshida, N. Koga, K. Morokuma, Organometallics 15 (1996) 766.
[60] P. Margl, L. Deng, T. Ziegler, J. Am. Chem. Soc. 120 (1998) 5517.
[61] J.C.W. Chien, A. Razavi, J. Polym. Sci., Part A: Polym. Chem. 26 (1988) 2369.
[62] B. Temme, G. Erker, J. Karl, H. Luftmann, R. Fröhlich, S. Kotila, Angew. Chem.,
Int. Ed. Engl. 34 (1995) 1755.
References
[63] H. van der Heijden, B. Hessen, A.G. Orpen, J. Am. Chem. Soc. 120 (1998) 1112.
[64] G. Moscardi, L. Resconi, L. Cavallo, Organometallics 20 (2001) 1918.
[65] C.J. Schaverien, R. Ernst, P. Schut, T. Dall’Occo, Organometallics 20 (2001) 3436.
[66] L. Resconi, F. Piemontesi, I. Camurati, O. Sudmeijer, I.E. Nifant’ev, P.V. Ivchenko,
L.G. Kuz’mina, J. Am. Chem. Soc. 120 (1998) 2308.
[1] W. Kaminsky, Macromol. Chem. Phys. 197 (1996) 3907.
[2] G.W. Coates, Chem. Rev. 100 (2000) 1223.
[3] H.-H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger, R.M. Waymouth, Angew.
Chem., Int. Ed. Engl. 34 (1995) 1143.
[4] P.C. Möhring, N.J. Coville, J. Organomet. Chem. 479 (1994) 1.