Metallocene-catalyzed olefin polymerizations using triphenylcyclopropenium tetrakis(pentafluorophenyl)borate as the activator

Article abstract of DOI:10.1139/v03-047

Triphenylcyclopropenium (TPCP) tetrakis(pentafluorophenyl)borate activates bis(cyclopentadienyl)dimethyl titanium resulting in a highly reactive initiating system for the polymerization of styrene. In contrast to triphenylmethyl tetrakis(pentafluorophenyl)borate which is quite active in the absence of the metallocene, TPCP borate shows no activity for styrene polymerization in the absence of bis(cyclopentadienyl)dimethyl titanium . TPCP is the most efiicient activator in the carbonium ion - borate class. We propose, based on ¹H NMR evidence that reaction of Cp₂TiMe ₂ and TPCP borate leads to the formation of the cationic Ti complex [Cp₂TiMe]⁺B(C₆F₅)₄ ^-. Evidence for the latter is also provided by UV-vis spectroscopy in that we found a bathochromic shift of the Cp₂TiMe₂ LMCT absorption band from 361 to 482 nm in CH₂Cl₂ and 487 nm in toluene, respectively. Thermal decomposition of the cationic complex [Cp₂TiMe]⁺B(C₆F₅)₄ ^- leads to less activity. The systems are good catalysts for ethylene polymerization as well, but are less active when using propylene. A conventional Ziegler-Natta coordination polymerization mechanism accounts for ethylene and propylene polymerization while a carbocationic polymerization mechanism is proposed for styrene.

Full text of DOI:10.1139/v03-047

7
58
Me talloce ne -catalyze d ole fin polyme rizations
us ing triphe nylcycloprope nium
te trakis (pe ntafluorophe nyl)borate as the activator
Huiying Li a nd Dougla s C. Ne c ke rs
Abstract: Triphenylcyclopropenium (TPCP) tetrakis(pentafluorophenyl)borate activates bis(cyclopentadienyl)dimethyl ti-
tanium resulting in a highly reactive initiating system for the polymerization of styrene. In contrast to triphenylmethyl
tetrakis(pentafluorophenyl)borate which is quite active in the absence of the metallocene, TPCP borate shows no activity
for styrene polymerization in the absence of bis(cyclopentadienyl)dimethyl titanium. TPCP is the most efficient activator
1
in the carbonium ion – borate class. We propose, based on H NMR evidence that reaction of Cp TiMe and TPCP borate
2
2
+
–
leads to the formation of the cationic Ti complex [Cp TiMe] B(C F ) . Evidence for the latter is also provided by UV–vis
2
6 5 4
spectroscopy in that we found a bathochromic shift of the Cp TiMe LMCT absorption band from 361 to 482 nm in
2
2
+
–
CH Cl and 487 nm in toluene, respectively. Thermal decomposition of the cationic complex [Cp TiMe] B(C F ) leads
2
2
2
6 5 4
to less activity. The systems are good catalysts for ethylene polymerization as well, but are less active when using propylene.
A conventional Ziegler–Natta coordination polymerization mechanism accounts for ethylene and propylene polymerization
while a carbocationic polymerization mechanism is proposed for styrene.
Key words: triphenylcyclopropenium tetrakis(pentafluorophenyl)borate, bis(cyclopentadienyl)dimethyl titanium, activator,
olefin polymerization.
Résumé : Le tétrakis(pentafluorophényl)borate de triphénylcyclopropénium (TPCP) active le bis(cyclopentadiényl)dimé-
thyltitane qui conduit à un système très réactif pour initier la polymérisation du styrène. Par opposition au tétrakis(pen-
tafluorophényl)borate de triphénylméthyle qui est très actif en l’absence de métallocène, le borate de TPCP ne présente
aucune activité vis-à-vis de la polymérisation du styrène en l’absence du bis(cyclopentadiényl)diméthyltitane. Le TPCP
est l’activateur le plus efficace de la classe borate – ion carbonium. On suggère, sur la base de données de RMN du
1
H, que la réaction du Cp TiMe et du borate de TPCP conduit à la formation du complexe cationique du Ti,
2
2
+
–
[
Cp TiMe] B(C F ) . Ce résultat est également appuyé par des mesures effectuées par spectrométrie UV–vis qui dé-
2 6 5 4
montrent un déplacement bathochrome de la bande d’absorption du LMCT du Cp TiMe de 361 nm vers 482 nm pour
2
2
le CH Cl et vers 487 nm pour le toluène, respectivement. Une décomposition thermique du complexe cationique
2
2
+
–
[
Cp TiMe] B(C F ) conduit à une perte d’activité. Les systèmes sont aussi de bons catalyseurs pour la polymérisation
2 6 5 4
de l’éthylène, mais ils sont moins réactifs pour le propylène. Un mécanisme de polymérisation conventionnel de Zie-
gler–Natta par coordination permet d’expliquer la polymérisation de l’éthylène et du propylène alors qu’on propose un
mécanisme de polymérisation carbocationique pour le styrène.
Mots clés : tétrakis(pentafluorophényl)borate de triphénylcyclopropénium, bis(cyclopentadiényl)diméthyltitane, activa-
teur, polymérisation d’oléfine.
[
Traduit par la Rédaction] Li and Neckers 763
Introduc tion
(1). Successful catalysts invariably consist of transition metal
complexes called catalyst precursors and activators also
called co-catalysts. The partners react forming an active spe-
cies that is comprised of a cation–anion ion pair. The struc-
Metallocene coordination polymerization catalyst systems
for olefins are an important research and development topic
Received 9 January 2003. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 23 June 2003.
It is a sincere pleasure to recognize the career-long contributions to the photochemical sciences of Professor Don Arnold on the
occasion of his retirement. Through the application of the principles of physical organic chemistry to the study of the emerging
field of reactions catalyzed by ultraviolet light, Professor Arnold and his students enhanced predictive power in organic
photochemistry. His colleagues are most grateful for his insights.
H. Li and D.C. Neckers.1,2,3 Center for Photochemical Sciences, Bowling Green State University, Bowling Green, OH 43403,
U.S.A.
1
Corresponding author (e-mail: neckers@photo.bgsu.edu).
Contribution number 500 from the Center for Photochemical Sciences.
Dedicated to Professor Dr. J.W. Neckers, on the occasion of his 101st birthday.
2
3
Can. J. Chem. 81: 758–763 (2003)
doi: 10.1139/V03-047
© 2003 NRC Canada

Li and Neckers
759
ture of the activator significantly influences catalytic activity,
the character of the polymerization, and polymer properties.
The number of viable co-catalysts for metallocene poly-
merizations is rather limited (2). Methylaluminoxane (MAO)
is an important industrial co-catalyst, but must be used in
huge excess to achieve high activity. Triphenylmethyl car-
Fig. 1. Structure of the activator and the catalyst.
–
benium ion (“trityl”) if paired with B(C F ) (1) has been
6
5 4
reported to be a highly efficient activator for group 4 di-
methylmetallocenes (3) at low co-catalyst to catalyst ratios
(
1:1). A number of other borates have also been developed
to improve properties such as solubility in organic solvents
and thermal stability (4).
UV-2401PC UV–vis recording spectrophotometer. Number-
and weight-average molecular weights (M and M ) and
n
w
polydispersity ratios (M /M ) were estimated by gel perme-
w
n
ation chromatography (GPC) on a Shimudzu HPLC system
equipped with a Plgel 5 µm MIXED-C 300 × 7.5 mm col-
umn (Polymer Laboratories), using THF as the eluent with a
–
1
flow rate of 1.0 mL min by polystyrene calibration, and a
RID-10A refractive index (RI) detector.
We have recently discovered that triarylcyclopropenium
salts may be used as cationic initiators for both the thermal
and photochemical polymerization of epoxides (5, 6). Tri-
phenylcyclopropenium (TPCP) cation is a strong electro-
phile (7) and when TPCP salts paired with weakly
nucleophilic anions tetrakis(pentafluorophenyl)gallate are
dissolved in ketones with active α-hydrogens such as cyclo-
hexanone we found that super acids (presumably H-
Ga(C F ) ) form. Super acids also result from irradiation of
Synthesis of TPCP tetrakis(pentafluorophenyl)borate (2)
To a suspension of TPCP chloride (11) (1.51 g, 5 mmol)
in acetonitrile (25 mL) was added a solution of potassium
tetrakis(pentafluorophenyl)borate (4.00 g, 5 mmol) in ace-
tonitrile (15 mL) and the reaction mixture stirred at room
temperature for 2 h. The solid thus formed was removed by
filtration and the filtrate concentrated to give a brown vis-
cous substance that was further purified using a silica gel
column with CH Cl as the eluent. After washing with
6
5 4
2
2
+
–
TPCP Ga(C F ) in various solvents (8).
pentane, the product was obtained as white crystals (62%),
6
5 4
1
3
In view of this new discovery we wished to explore fur-
ther the advantages of cyclopropenium salts in various poly-
merization reactions. We report herein results of olefin
polymerization catalyzed by bis(cyclopentadienyl)dimethyl
titanium (3) using TPCP borate (2) as the activator (Fig. 1).
mp 209 to 210°C. H NMR (CDCl ) δ: 7.88 (dd, J
= 7.6,
3
H-H
3
7.8 Hz, 6H, 3,5-H on phenyl), 8.06 (t, J
H on phenyl), 8.42 (d, JH-H = 7.6 Hz, 6H, 2,6-H on
phenyl). F NMR (CDCl ) δ: –133.12 (s, 2, 6-F on
= 7.8 Hz, 3H, 4-
H-H
3
1
9
3
C F ), –163.40 (sm, 4-F on C F ), –167.20 (s, 3,5-F on
6
5
6 5
C F ). Anal. calcd. for C H F B: C 57.11, H 1.60; found:
6
5
45 15 20
C 57.02, H 1.66.
Expe rim e nta l s e c tion
Materials
Polymerization of styrene
All chemicals were used as received from Aldrich unless
otherwise noted. Benzene and toluene were distilled over so-
dium under argon. CH Cl was freshly distilled over CaH
In a typical polymerization, 2 (0.03 mmol) was dissolved
in solvent (2 mL) (TPCP tetrakis(pentafluorophenyl)borate
dissolved in CH Cl and is partially soluble in toluene) and
2
2
2
2
2
under argon. α,α-Dichlorotoluene was purchased from Acros
Organics and used as received. Bis(cyclopentadienyl)tita-
nium dichloride was purchased from Strem Chemicals Inc.
the solution degassed (freeze–thaw) for three cycles. 3
(0.03 mmol, 36% in toluene) was also dissolved in styrene
(2 mL) and the solution degassed as above. The catalyst so-
lution in styrene was then transferred into the solution of 2
under dry argon at room temperature and the mixture stirred
for 30 min. Subsequently, the polymerization mixture was
quenched with MeOH (containing 1% HCl). The resulting
polymer was purified twice from CH Cl –MeOH or until a
white powder of the polymer was obtained. The polymer
was dried under vacuum overnight. The activity of the cata-
lyst system was calculated by the following:
Activity = weight of polymer (g)/(Ti (mol) monomer
(mol) time (h))
and used as received. Styrene was distilled over CaH at
2
5
0°C under vacuum and kept over molecular sieves (4 Å)
under argon in a refrigerator. Potassium tetrakis(pentafluoro-
phenyl)borate was prepared as reported (9). Bis(cyclopenta-
dienyl)dimethyltitanium (3) was prepared according to the
literature (10) and kept in a freezer as a solution (~36%) in
toluene.
2
2
Measurements
1H and 19F NMR spectra were recorded either with a
Varian Gemini 200 NMR or a Varian Unity plus 400 NMR
spectrometer. Chemical shifts are in ppm with TMS as the
Polymerization of styrene with preactivated initiating
systems
1
internal standard ( H NMR) or CFCl as the external stan-
3
1
9
dard ( F NMR). Melting points were determined with a
Thomas Hoover capillary melting point apparatus and were
uncorrected. UV–vis spectra were recorded on a Shimadzu
In a typical polymerization run, 3 (36% in toluene) was
mixed with CH Cl (2 mL), and the solution degassed using
2
2
freeze–thaw techniques for three cycles. This initiating solu-
©
2003 NRC Canada

7
60
Can. J. Chem. Vol. 81, 2003
tion in CH Cl was then transferred into the polymerization
CH Cl (containing traces of toluene) was degassed for three
2
2
2
2
tube (prevacuum evacuated) containing 2 under dry argon
the reactions are quite sensitive to spurious moisture) at
cycles. 2 was placed in a quartz cuvette and the cuvette de-
gassed with bubbling dry argon for 20 min. Catalyst solution
was introduced using a two-tipped needle and the vessel
sealed under dry argon. UV–vis spectra of the resulting solu-
tions were recorded at room temperature at various times.
(
room temperature and the mixture stirred (aged) for 10 min.
The mixture turned from yellow to dark red. Styrene
(
predegassed) was next introduced into the system, and the
polymerization mixture stirred at room temperature for 15–
0 min. It was next quenched by MeOH (containing 1%
HCl). The resulting polystyrene (PS) was purified twice
from CH Cl –MeOH or until a white powder of the polymer
3
Re s ults a nd dis c us s ion
2
2
Synthesis of TPCP borate
Anion exchange of TPCP chloride with KB(C F ) afforded
–
was obtained, and the polymer then dried under vacuum
overnight.
6
5 4
triphenylcyclopropenium tetrakis(pentafluorophenyl)borate 2
as a white solid, mp 209 to 210°C. 2 is stable under air at
room temperature and soluble in CH Cl but only slightly
Polymerization of ethylene
2
2
In a typical polymerization run, a solution of 3 (1.5–
soluble in toluene.
3
.0 mmol, 36% in toluene) in CH Cl (2 mL) was degassed
2 2
for three cycles and then transferred into a polymerization
tube (25 mL, prevacuumed) containing 2 (1.5–3.0 mmol) un-
der an ethylene atmosphere. The catalyst system was
preactivated for 10 min at room temperature and then ethyl-
ene at 1 atm (1 atm = 101.325 kPa) was bubbled through the
solution. Polyethylene (PE) precipitated from the solution
immediately. After 5 min, the polymerization reaction was
quenched with MeOH (containing 1% HCl, 5 mL). The
polymer was collected by filtration. After washing with
MeOH, the polymer was dried under vacuum overnight. The
resulting polymer did not dissolve in THF or in 1,1,2,2-
tetrachloroethane.
Polymerization of styrene
Styrene polymerizations were carried out in toluene or
CH Cl at 23°C under dry argon (Table 1). Control experi-
2
2
ments show neither 2 nor 3 had significant activity when
used alone in CH Cl or toluene. Following the prescribed
2
2
period of preactivation, 2 was among the most active activa-
tors for styrene polymerization (entry 6) of the type carbon-
ium ion – borate (12). The solvent had little effect on
initiation activity (entries 1 and 2), but polystyrene obtained
from toluene solution was of lower weight average molecu-
lar weight (M ), presumably because of the limited solubil-
w
ity of 2 and the active species. The activity of a preactivated
initiating system was much higher than that without pre-
activation. If the preactivation time was increased to 20 min
(entry 4), the activity of the initiating system was reduced by
about 50%. We presume this is due to the thermal decompo-
sition of the catalytic cationic complex at room temperature.
The efficiency of 2 in the polymerization of styrene is
compared to that of the trityl salt 1 in toluene. Titanocene 3
shows a high activity of 6.04 × 10 g (PS)/mol (Ti) × mol
(styrene) × h in the presence of 1 (entry 11), however, from
control experiments, 1 itself is a rather active initiator for the
polymerization of styrene with the activity of 5.60 × 10 g
(PS)/mol (Ti) × mol (styrene) × h (entry 12). This is not the
Polymerization of propylene
Polymerization of propylene resulted if a similar proce-
dure to that used for ethylene polymerization was employed.
Propylene was bubbled through the catalyst solution at 1 atm
(
1 atm = 101.325 kPa) for 30 min at room temperature. Af-
ter quenching with MeOH (1% HCl), the resulting poly-
propylene (PP) (viscous oil) was collected by carefully
decanting the solvent and subsequently dried under vacuum
overnight.
6
6
1H NMR studies
To complete the reaction of Cp TiMe with TPCP salts, a
case with 2.
2
2
mol ratio of 1.1:1.0 for TPCP:Ti was used in this experi-
ment. In a typical experiment, CD Cl was vacuum distilled
All polystyrene samples obtained in our reactions are sol-
uble in 2-butanone indicating only atactic polystyrene (a-PS)
is formed. This was confirmed by C NMR spectroscopy
(13).
2
2
1
3
over CaH , trapped at 78 K, and then degassed by freeze–
2
thaw techniques for three cycles. The catalyst (0.03 mmol,
3
6% solution in toluene) was dissolved in the above CD Cl
2 2
and the solution transferred to a vacuum evacuated tube con-
taining TPCPB (0.032 mmol) at –78°C. The solution was
subsequently warmed to room temperature and held there for
Polymerization of ethylene and propylene
Polymerization was carried out at room temperature by
bubbling of ethylene (1 atm, (1 atm = 101.325 kPa)) through
a preactivated solution of 4 (1.5–3.0 mmol) formed by reaction
of an equimolar concentration of 2 and 3 in CH Cl (2 mL)
(Table 1). PE immediately precipitated from the solution.
After 5 min, quenching the reaction mixture with acidic
5
min. During this period, the catalyst system turned from
yellow to dark red. The dark red solution was next trans-
ferred into an NMR tube that had been degassed by passing
dry argon through it for 30 min. The NMR tube was cooled
to –78°C again and taken to the NMR. The sample tempera-
ture was equilibrated for 1 h at 0°C and the NMR data col-
lected.
2
2
methanol yielded lightly yellow polymers with T 127 ~ 130°C.
The PE obtained was not soluble in organic solvents such as
m
CH Cl , THF, and 1,1,2,2-tetrachloroethane.
2
2
Bubbling of propylene into the catalyst system for 30 min,
followed by quenching of the polymerization mixture with
acidic methanol, gave polypropylene as a viscous oil of rela-
tively low molecular weight (4.0–7.2) × 10 g mol . Though
the quantity of catalyst used had little affect on the yield of
UV–vis studies
Absorption spectra of the catalyst systems were recorded
under dry argon in a quartz cuvette (1.0 cm path length). In
3
–1
–
4
a typical experiment, a solution of 3 (4.2 × 10 M) in
©
2003 NRC Canada

Li and Neckers
761
+
–
a
Table 1. Polymerization of styrene, ethylene, and propylene by Cp TiMe –TPCP B(C F ) in CH Cl .
2
2
6
5 4
2
2
Catalyst Polymerization
Yield
(g)
Activity
(× 10 )
Mw
(kg mol )
6
b
–1
M /M
Run (mmol)
time (min)
Monomer
w
n
1c
2
3.0
3.0
3.0
3.0
3.0
1.5
3.0
1.5
3.0
5.0
3.0
30
30
30
30
15
15
5
Styrene
Styrene
Styrene
Styrene
Styrene
Styrene
Ethylene
Ethylene
Propylene
Propylene
Styrene
Styrene
1.23
1.06
1.80
0.92
1.78
1.43
0.17
0.17
0.037
0.038
1.59
1.46
4.68
4.05
7.15
3.50
13.5
21.8
0.068
0.14
20.6
65.3
50.8
50.5
40.2
54.8
—f
2.13
1.93
2.31
2.14
2.38
2.27
—
—
1.46
2.68, 2.14
2.48
4.14
d
3
4
5
6
7
8
9
0
e
d
d
d
d
5
—
d
30
30
30
30
0.0035
0.0015
6.04
3.97
7.18, 5.56
6.3
d
1
1
1
c
g
1
c
h
2
—
5.60
8.3
a
Polymerization conditions: Styrene polymerization: temperature = 23°C, styrene = 2 mL, solvent = 2 mL,
[Styrene] = 4.4 M. Ethylene and propylene polymerization: temperature = 23°C, solvent = 2 mL, monomer
pressure = 1 atm (1 atm = 101.325 kPa).
b
Unit: g (PS)/(mol Ti × mol styrene × h); g (PE or PP)/(mol Ti × atm × h) (1 atm = 101.325 kPa).
c
Toluene was used as the solvent.
d
Preactivation time: 10 min.
e
Preactivation time: 20 min.
f
The polymers were not soluble in CH Cl , THF, and CHCl CHCl for M measurement.
2
2
2
2
w
g
h
Trityl tetrakis(pentafluorophenyl)borate was used as the activator.
Only trityl tetrakis(pentafluorophenyl)borate was used.
PE and PP, the use of a large amount of catalyst (5.0 mmol)
for propylene resulted in polymers that exhibited a bimodal
molecular weight distribution pattern.
UV–vis spectroscopy
There are a number of reports of spectroscopic studies of
the reactions of zirconocene precursors with MAO (16) and
+
–
PhMe NH B(C F ) (17) but few such reports with acti-
2
6 5 4
vated titanocene catalysts (18). In the zirconocene studies,
the activated complex formed is easily distinguished by
changes in the positions of the ligand-to-metal charge trans-
fer (LMCT) bands in the optical spectra. Though UV–vis
spectroscopy is not suitable for catalyst systems containing
trityl cation owing to the overlap of absorption of the cation
with the LMCT bands of the complexes formed in the acti-
vation process (Fig. 2), TPCP cation has no absorption be-
yond 320 nm. Thus, one may examine the LMCT bands
formed from catalyst systems containing TPCP cation using
UV–vis spectroscopy.
Formation of the cationic complex
The dimethyl Ti(IV) complex reacts with a stoichiometric
amount of TPCPB at room temperature to yield the cationic
+
–
titanium – monomethyl complex [Cp TiMe] [B(C F ) ] (4)
2
6 5 4
as a dark red solution. If the reaction is carried out at 0
or –23°C, it takes longer (>30 min) to form the red solution.
On the basis of its NMR spectrum taken at 0°C, the cationic
complex 4 exhibits resonances at δ 6.38 (s, 10H), 1.30 (s,
3
H) attributed to the Cp and Ti-Me hydrogens, respectively.
The spectrum of 4 differs from that of the parent Cp TiMe
2
2
1
(
H NMR in CD Cl at 25°C δ: 6.07 (s, 10H), –0.15 (s, 6H))
2 2
In the absence of 2, 3 shows a LMCT band at 361 nm in
and the integrated ratio of Cp to methyl evolved to 10:3 in
the activated complex as compared to a value for the parent
compound of 10:6. Bochmann et al. (14) reported a similar
complex was generated by reaction of the parent complex
with dimethylanilinium tetraphenylborate and that it exhibits
resonances at δ 6.28 (s, 10H), 1.26 (s, 3H). We presume that
the resonance differences from the two cationic complexes
are due to differences in ion pairing. There were also small
resonances in the ranges of 0.20–2.00 and 6.40–6.70 which
could not be assigned.
If the catalyst solution is warmed to 25°C and equilibrated
for 30 min, the resonances at δ 6.38 and 1.30 decrease
slightly, and the resonances at δ 0.22 and 6.56, 6.64, 6.72
grow presumably because of decomposition of the cationic
complex. One of the deactivation processes of the metallo-
both CH Cl and toluene. The LMCT absorption band of 3 is
2
2
shifted to 482 nm in CH Cl in the presence of 2 and to
2
2
4
87 nm in toluene, respectively, corresponding to a decrease
of electron density on the metal center (16a). This is consis-
tent with the formation of cationic Ti(IV) complex observed
by NMR spectroscopy. During styrene polymerization, we
found that with preactivation, a deep approximately dark red
solution of initiator was required in order for the system to
exhibit the highest activity. Yellow solutions showed little or
no activity as polymerization initiators. Therefore, the corre-
sponding species, which absorbs at 482 nm, must be the ac-
tual active species.
In the case of the 2, 3 couple, formation of the active spe-
cies 4 was followed by observing the absorption changes in
dichloromethane at 482 nm (Fig. 3). In the absence of mono-
mer at fixed concentrations of reagents, the maximum con-
centration of 4 was observed after 17 min (Fig. 3a). The
absorption then decayed presumably because of the thermal
–
cenium cation that may occur results from Cl abstraction
from the solvent if the reaction is carried out in CH Cl (15).
During our NMR studies no by-products such as CH -CD Cl
were observed.
2
2
3
2
©
2003 NRC Canada

7
62
Can. J. Chem. Vol. 81, 2003
Fig. 2. UV–vis spectra of catalyst Cp TiMe (3) and a mixture
Fig. 3. Variation of the absorption of 2 and 3 at 482 nm with
2
2
–
4
of 2 and 3 in CH Cl . Concentration of catalyst and activator:
time in CH Cl . [3] = 4.2 × 10 M, ratio of 2:3:styrene in mol
2
2
2 2
–4
[
3] = 4.2 × 10–4 M, [2] = 4.8 × 10 M; spectrum of the mixture
(a) in the absence of styrene; (b) in the presence of styrene; and
(c) styrene added at 17 min.
of 2 and 3 was taken after being mixed for 17 min.
Scheme 1. Electrophilic attack of 4 on styrene.
decomposition of the active cationic complex. Consistent
with results obtained in the polymerization of styrene in that
if the initiation system was preactivated for times greater
than 17 min (e.g., 20 min, entry 4, Table 1), one observes
decreased activity. Consistent with the lower activity ob-
served for the polymerization of styrene when the initiator is
prepared in its presence, the growing in of the absorption at
4
82 nm was much slower in the presence of styrene than in
its absence (Fig. 3b). This can be explained if styrene con-
sumes the active species during its formation. A rapid decay
of 4 also results from the addition of styrene to the initiation
system (Fig. 3c). We presume this to be from the reaction of
styrene at the Ti center resulting in a decreased concentra-
tion of the cationic complex 4 (Scheme 1).
Scheme 2. Coordination polymerization pathway.
The formation and decay of cationic complex 4 was stud-
ied by following the absorption changes at 482 nm. The for-
mation and decay lifetimes of the active species were
obtained by fitting the experimental data to a double expo-
nential decay function using ORIGIN 6.1 software. The spe-
cies grew in rapidly with τ of 7.96 min and then decayed
1
slowly with τ of 28.9 min.
2
Mechanism of olefin polymerization
The mechanism we propose for olefin polymerization is
that dimethyl Ti(IV) complex reacts with a stoichiometric
amount of TPCPB at room temperature to yield the cationic
merization of olefins (19) and that styrene may undergo
polymerization via both coordination and carbocationic
mechanisms. Using Cp*TiMe –B(C F ) as the catalyst and
if the polymerization is carried out at a low temperature,
only atactic polystyrene was produced. If the polymerization
temperature rises to above –10°C, highly syndiotactic poly-
styrene is obtained (20). At high temperature Ti(III) com-
plex formed by reductive decomposition was confirmed to
be the active catalyst for production of syndiotactic polysty-
rene (21). At lower temperatures, cationic Ti(IV) complex
+
–
titanium – monomethyl complex [Cp TiMe] [B(C F ) ] (4).
2
6 5 4
The system is active for nonfunctionalized olefins like ethyl-
ene, propylene, and styrene with polymerization of ethylene
and styrene being much faster in CH Cl at room tempera-
3
6 5 3
2
2
ture than the polymerization of propylene. That ethylene
polymerization is faster than that of propylene is consistent
with a coordination polymerization mechanism (Scheme 2).
It has recently been found that some well-characterized
Ziegler–Natta catalysts can also initiate carbocationic poly-
©
2003 NRC Canada

Li and Neckers
763
Scheme 3. Cationic polymerization of styrene.
Hustad, and S. Reinartz. Angew. Chem. Int. Ed. 41, 2236
(
2002).
2
3
4
. E.Y.-X. Chen and T.J. Marks. Chem. Rev. 100, 1391 (2000),
and refs. cited therein.
. J.C.W. Chien, W.-M. Tsai, and M.D. Rausch. J. Am. Chem.
Soc. 113, 8570 (1991).
. (a) V.C. Williams, G.J. Irvine, W.E. Piers, Z. Li, S. Collins, W.
Clegg, M.R.J. Elsegood, and T.B. Marder. Organometallics,
1
9, 1619 (2000); (b) F.A.R. Kaul, G.T. Puchta, H. Schneider,
M. Grosche, D. Mihalios, and W.A. Herrmann. J. Organomet.
Chem. 621, 177 (2001); (c) S.J. Lancaster, A. Rodriguez, A.
Lara-Sanchez, M.D. Hannant, D.A. Walker, D.H. Hughes, and
M. Bochman. Organometallics, 21, 451 (2002).
5
. (a) H. Li, K. Ren, W. Zhang, J.H. Malpert, and D.C. Neckers.
Macromolecules, 34, 2019 (2001); (b) H. Li, K. Ren, and D.C.
Neckers. Macromolecules, 34, 8637 (2001).
initiates a carbocationic process for the production of atactic
polystyrene.
+
Our spectroscopic studies indicate that cation [Cp TiMe]
is more stable than [Cp*TiMe ] but that it decomposes,
2
+
6. D.C. Neckers and W. Zhang. U.S. Patent 6 420 460, July 16, 2002.
7
2
. (a) R. Breslow. J. Am. Chem. Soc. 79, 5318 (1957); (b) R.
Breslow and C. Yuan. J. Am. Chem. Soc. 80, 5991 (1958).
. H. Li, K. Ren, and D.C. Neckers. J. Org. Chem. 66, 8556 (2001).
. F. Castellanos, J.P. Fouassier, C. Priou, and J. Cavezzan. J.
Appl. Polym. Sci. 60, 705 (1996).
though moderately, at room temperature on the polymeriza-
tion time scale. The decay lifetime of the cationic complex
is 28.9 min indicating that most of the cationic complex sur-
vived during the polymerization process. Each of the PS
samples prepared using our initiating system is soluble in 2-
butanone, suggesting no syndiotactic PS (s-PS) forms.
Therefore, we assume that the cationic Ti(IV) complex is the
actual active species in the system reacting as a carboca-
tionic initiator with styrene and producing atactic polystyrene
8
9
1
1
0. L.M. Dollinger, A.J. Ndakala, M. Hashemzadeh, G. Wang, Y.
Wang, I. Martinez, J.T. Arcari, D.J. Galluzzo, and A.R.
Howell. J. Org. Chem. 64, 7074 (1999).
1. In this procedure HCl(g) was used instead of HBr(g). R. Xu
and R. Breslow. Org. Synth. 74, 72 (1997).
(
Scheme 3). The relatively narrow polydispersities of the re-
12. G. Xu. Macromolecules, 31, 586 (1998).
sulting PS samples (1.92–2.56) is further evidence for a
carbocationic polymerization process.
In summary, TPCP borate efficiently reacts with
bis(cyclopentadienyl) dimethyltitanium to form cationic Ti
13. H.N. Cheng. Int. J. Polym. Anal. Charact. 2, 439 (1996).
14. M. Bochmann, A.J. Jahhar, and J. Nicholls. Angew. Chem. Int.
Ed. Engl. 29, 780 (1990).
15. (a) R. Gomez, M.L. Green, and J.L. Haggitt. J. Chem. Soc.
Dalton Trans. 939 (1996); (b) T. Guenca, M. Galakhov, G.
Jiménez, E. Royo, P. Royo, and M. Bochmann. J. Organomet.
Chem. 543, 209 (1997).
+
–
complex [Cp TiMe] B(C F ) (4). The formation of cati-
2
6 5 4
onic Ti complex 4 was confirmed by NMR spectroscopy.
The reaction was also studied using UV–vis spectroscopy
and observed the bathochromic shift of the Cp TiMe LMCT
main absorption band from 361 to 482 nm in CH Cl . The
kinetic profile showed the cationic species grew in rapidly
and then decayed slowly. Though the cationic complex
1
6. (a) P.J.J. Pieters, J.A.M. van Beek, and M.F.H. van Tol. Macromol.
Rapid Commun. 16, 463 (1995); (b) D. Coevoet, H. Cramail,
and A. Deffieux. Macromol. Chem. Phys. 199, 1451 (1998);
2
2
2
2
(c) D. Coevoet, H. Cramail, and A. Deffieux. Macromol. Chem.
+
–
Phys. 199, 1459 (1998); (d) J.-N. Pédeutour, D. Coevoet, H.
Cramail, and A. Deffieux. Macromol. Chem. Phys. 200, 1215
[
Cp TiMe] B(C F ) was observed to be thermally unstable
2 6 5 4
and its activity in the polymerization of styrene decreased
with time, the system shows generally high polymerization
activity for both ethylene and styrene. It is less active for the
polymerization of propylene. The cationic Ti(IV) complex is
proposed to be a conventional Ziegler–Natta coordinative
catalyst for ethylene and propylene polymerization, while it
acts as a carbocationic initiator for styrene polymerization.
(1999); (e) J.-N. Pédeutour, H. Cramail, and A. Deffieux. J. Mol.
Catal. A: Chem. 174, 81 (2001); (f) J.-N. Pédeutour, H. Cramail,
and A. Deffieux. J. Mol. Catal. A: Chem. 176, 87 (2001).
7. U. Wieser and H.H. Brintzinger. Organometallic catalysts and
olefin polymerization. Springer-Verlag, Berlin, Germany. 2001.
pp. 3–13.
8. (a) J. Kim, K.H. Kim, J.E. Cho, S. Kwak, K.U. Kim, W.H. Jo,
H.S. Yoon, and D.S. Lim. J. Polym. Sci. Part A: Polym. Chem.
36, 1733 (1998); (b) A. Anderson, H.G. Cordes, J. Herwig, W.
Kamisky, A. Merk, R. Mottweiler, J.H. Sinn, and H.-J. Vollmer.
Angew. Chem. Int. Ed. Engl. 15, 630 (1976).
1
1
Ac knowle dgm e nts
We would like to thank the United Soybean Board for fi-
nancial support. We also thank Dr. Kangtai Ren and Ms.
Priya Hewavitharanage for helpful discussions.
1
2
9. M.C. Baird. Chem. Rev. 100, 1471 (2000), and refs. cited
therein.
0. (a) Q. Wang, R. Quyoum, D.J. Gillis, M.-J. Tudoret, D. Jeremic,
B.K. Hunter, and M.C. Baird. Organometallics, 15, 693 (1996);
Re fe re nc e s
(
b) C. Pellecchia, D. Pappalardo, L. Oliva, and A. Zambelli. J.
1
. For recent reviews see: (a) J.A. Gladysz. Chem. Rev. 100, 4
2000); (b) Y. Imanishi and N. Naga. Prog. Polym. Sci. 26,
147 (2001); (c) G.J.P. Britovsek, V.C. Gibson, and D.F. Wass.
Angew. Chem. Int. Ed. 38, 428 (1999); (d) G.W. Coates, P.D.
Am. Chem. Soc. 117, 6593 (1995).
(
1
2
1. C. Pellecchia and A. Grassi. Top. Catal. 7, 125 (1999).
©
2003 NRC Canada