Preparation of A1(C6 F5)3 and its use for the modification of methylalumoxane

Article abstract of DOI:10.1016/S1381-1169(97)00277-X

Tris(pentafluorophenyl)aluminum was prepared from a metathesis between AlMe₃ and B(C₆F₅)₃ in a quantitative yield. Out of CN stretching frequency of various benzonitrile adducts, it is found that the title aluminum compound is a quite weaker Lewis acid than B(C₆F₅)₃ is, and has rather similar acidity as AlMe₃. When mixtures of methylalumoxane (MAO) arid Al(C₆F₅)₃ (toluene)(0.5) or B(C₆F₅)₃ were heated at 65°C, it is observed that the C₆F₅ group was exchanged with the methyl group of the MAO. Several MAOs were prepared with variable degrees of modification. Catalytic activity of pristine MAO and modified MAOs were tested for ethylene and propylene polymerization. It is found that the activity is enhanced with the modification, and the effectiveness is reduced as modification level is increased. In particular, 7-fold increase of activity for polyethylene is observed, when 5.5 x 10^-3 equimolar amount of Al(C₆F₅)₃ (toluene)(0.5) is used for the modification. The modification with B(C₆F5)₃ incorporates boron atoms into the oligomeric chain, and is less effective than that with the aluminum analogue.

Full text of DOI:10.1016/S1381-1169(97)00277-X

Ž
.
Journal of Molecular Catalysis A: Chemical 132 1998 231–239
ž
/
Preparation of Al C6 F5 3 and its use for the modification of
methylalumoxane
a
a
b
Choong Hoon Lee , Sang Jin Lee , Joon Won Park a,), Kyong Hee Kim ,
b
b
Bun Yeol Lee , Jae Seung Oh
a
Department of Chemistry, Center for Biofunctional Molecules, Pohang UniÕersity of Science and Technology, San 31 Hyoja-dong,
Pohang, North Kyongsang, South Korea
LG Chemical, Research Park, P.O. Box 108 Yu Song, Science Town, Taejon 305-343, South Korea
b
Received 1 September 1997; accepted 20 November 1997
Abstract
Ž
.
Ž
.
Tris pentafluorophenyl aluminum was prepared from a metathesis between AlMe3 and B C6F5 in a quantitative yield.
3
Out of CN stretching frequency of various benzonitrile adducts, it is found that the title aluminum compound is a quite
Ž
.
Ž
.
weaker Lewis acid than B C6F5 is, and has rather similar acidity as AlMe3. When mixtures of methylalumoxane MAO
3
Ž
.
Ž
.
Ž
.
3
and Al C6F5 3P toluene
or B C6F5 were heated at 658C, it is observed that the C6F5 group was exchanged with the
0.5
methyl group of the MAO. Several MAOs were prepared with variable degrees of modification. Catalytic activity of pristine
MAO and modified MAOs were tested for ethylene and propylene polymerization. It is found that the activity is enhanced
with the modification, and the effectiveness is reduced as modification level is increased. In particular, 7-fold increase of
3
activity for polyethylene is observed, when 5.5=10y equimolar amount of Al C6F5 3P toluene
is used for the
Ž
.
Ž
.
0.5
Ž
.
modification. The modification with B C6F5 incorporates boron atoms into the oligomeric chain, and is less effective than
3
that with the aluminum analogue. q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
Ž
.
Ž
.
Keywords: Al C6 F5 3; B C6 F5 3; Methylalumoxane MAO ; Olefin polymerization; Lewis acidity
1. Introduction
w x
w x
MAO has been widely used as a co-catalyst for Ziegler–Natta type polymerization 4–8 . Due to its
superb activity the aluminum catalyst has been a topic of intense research in the academic and
w x
Since the reports of Sinn and Kaminsky 1 , Sinn et al. 2 and Pasynkiewicz 3 , methylalumoxane
Ž
.
w
x
w
x
industrial domains 9–19 . Moreover, recent advances in homogeneous catalysts for the alkene
w
x
polymerization highlight its importance 4,20–23 .
Because of its inherent structural complexity MAO has escaped full characterization. Only recently,
w
x
its structure could be deduced from crystal structures of tert-butyl-substituted alumoxanes 24 ,
)
Corresponding author. Fax: q82-562-279-3399; e-mail: jwpark@vision.postech.ac.kr
1381-1169r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII S1381-1169 97 00277-X

(
)
232
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
w
x
w x
spectroscopic data 25 , and other analytical methods 26 . Also it is known that preparative methods
w
x
determine its molecular weight and structure, eventually catalytic activity 13–19 .
Effect of other substituents on the aluminum center has not been reported. Choice of the
substituents is very limited because the pendant group should be noncoordinating to the strongly
electrophilic metallic center in which a polymeric chain grows, otherwise the activity for the
polymerization is hampered.
Ž
.
w
x
Weakly coordinating B C6F5 4-anion 27–33 stimulates the use of C6F5 group for the modification
Ž
.
Ž
.
Ž .
from B C6F5 3, and
of MAO. Herein, we wish to report facile synthesis of Al C6F5 3 P toluene
0.5
modification of MAO with both aluminum and boron reagents.
2. Experimental
2.1. General
All manipulation of air- andror moisture-sensitive compounds was carried out with use of standard
Schlenk or vacuum-line technique. Argon was purified by passage through columns of Ridox oxygen
˚
Ž
.
scavenger R31-3, Fisher and Linde 4-A molecular sieves. Solids were transferred and stored in an
N2-filled Vacuum Atmospheres glovebox equipped with a HE-493 Dri-Train, a CS-40 Dri-Cold, and
Ž
.
an oxygen analyzer Model 315, Teledyne Analytical Instruments .
Benzene was stirred over CaH2 and then transferred to a sodiumrbenzophenone ketyl. Anhydrous
toluene, tetrahydrofuran, and diethyl ether were purchased from Aldrich, and transferred to a
sodiumrbenzophenone ketyl without prior treatment. Pentane was stirred over concentrated H2SO4,
dried over CaH2, and then transferred to a sodiumrbenzophenone ketyl. Dichloromethane was dried
over CaH2 and degassed by evacuation using freeze–pump–thaw cycles. Dried deoxygenated
solvents were vacuum-transferred to dry glass vessels equipped with a J-young valve and stored under
argon. Benzene-d6 and toluene-d8 were vacuum-transferred from purple sodiumrbenzophenone ketyl.
Benzonitrile was dried over anhydrous MgSO4, and then vacuum-distilled to a dry glass vessel, stored
˚
over Linde 4-A molecular sieves, and finally degassed through freeze–pump–thaw cycles. AlMe3
Ž
Ž
.
Ž
.
Ž
.
97%, Aldrich , n-BuLi 1.6 M in hexane, Aldrich , BCl3 1.0 M in hexane, Aldrich , MgCl2 P6H2O
.
Ž
.
Ž
. w x
99%, Aldrich , and C6F5Br )99%, Tokyo Kasei Kogyo were used as purchased. B C6F5 3 32
w
x
Ž
.
and MAO 13 were prepared as in the literatures. Ethylene polymer grade was purchased from
Matheson Gas Products, and used after passing through columns packed with Ridox oxygen
˚
scavenger and Linde 4-A molecular sieves.
NMR spectra were recorded on a Bruker DPX300 300 MHz, H; 121 MHz, C spectrometer.
Chemical shifts are reported in d, referenced to residual solvent signals. F H NMR and B H
NMR data are referenced externally to CF3COOH and B OH dissolved in D2O, respectively
downfield positive . IR spectra were recorded on an evacuable Bomem DA8 FT-IR spectrophotom-
1
13
Ž
.
19
11
1
1
Ä
4
Ä
4
Ž
.
3
Ž
.
eter. Spectra were taken while a sample chamber was evacuated in order to eliminate CO2
interference. Molecular weights of the alumoxanes were measured by isothermal distillation method
w
x
Ž
.
34 . A polystyrene of known molecular weight MWs2500, Aldrich was used as the standard.
Elemental analyses were performed at Galbraith Laboratory.
(
)
(
)
2.2. Preparation of tris pentafluorophenyl aluminumP toluene 0.5
Ž
.
Ž
.
Ž
.
When AlMe3 0.020 ml, 0.21 mmol was added to sublimed B C6F5 3 106 mg, 0.207 mmol
Ž
.
dissolved in toluene 3 ml at room temperature, bubbles were observed. The solution was stirred for

(
)
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
233
19
Ž
.
Ž
.
1 h, and drying in vacuo produced a white solid 126 mg, 0.203 mmol, 98% . F NMR C6D6 d
1
Ž
.
Ž
.
.
Ž
.
Ž
.
Ž
.
y46.4 d, 6F, ortho , y74.1 t, 3F, para , y83.9 t, 6F, meta ; H NMR C6D6 d 7.07 m, 5H ,
2.09 s, 3H . Calcd. for C18AlF15 C7H8 0.5: C, 44.97; H, 0.70. Found: C, 44.52; H, 1.00. For
recrystallization, the solid 30 mg was dissolved in THF 0.5 ml , and hexane 5 ml was layered on
Ž
.
Ž
Ž
.
Ž
.
Ž
.
the top of the solution. After 2 or 3 days, the solution produced transparent white crystals,
19
Ž
.
Ž
.
Ž
.
Al C6F5 3 PTHF, suitable for elemental analysis: F NMR C6D6 d y46.4 d, 6F, ortho , y74.5
1
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
t, 3F, para , y84.1 t, 6F, meta ; H NMR C6D6 d 3.42 t, 4H , 1.00 t, 4H ; Anal. Calcd. for
C22 H8AlF15O: C, 44.02; H, 1.34. Found: C, 43.52; H, 1.29.
(
)
(
)
2.3. Preparation of benzonitrile adducts of AlMe3 , Al C6 F5 3 , and B C6 F5 3
The aluminum and the boron reagents were added in benzene, dissolving an equimolar amount of
benzonitrile, and the mixtures were stirred for 2 h at room temperature. The solid benzonitrile adducts
were obtained by evaporating the solution in vacuo. IR spectra of the solids were recorded right after
the KBr pellets were made in the glove box.
2.4. Preparation of methylalumoxane
Ž
.
Ž
.
MgCl2 P6H2O 3.00 g, 14.8 mmol in toluene 8 ml was added to a 100-ml flask equipped with a
w
x
magnetic stirrer, a nitrogen inlet and a dropping funnel 13 . The suspension was cooled to 08C, then
AlMe3 4.70 ml, 49.3 mmol in toluene 20 ml was added dropwise at 08C for 1 h. Then, the reaction
Ž
.
Ž
.
mixture was gradually heated up to 508C and stirred at this temperature for 24 h. After filtration, the
1
Ž
.
solid-like MAO 0.84 g, Ys29% was obtained by evaporating the filtrate to dryness: H NMR
13
1
Ž
.
Ž
.
Ž .
Ž
.
Ž .
toluene-d8 d y0.01 br , y0.15 s ; C NMR toluene-d8 d y6.98 br . H NMR spectrum
shows a broad resonance as well as a sharp resonance, where the latter is attributed to bound AlMe3
to the MAO through a chemical association. Molecular weight of the oligomer was measured to be
2.1=103 from the isothermal distillation.
(
)
2.5. Modification of methylalumoxane with B C6 F5 3
Ž
.
w
x
Ž
.
Ž
.
Sublimed B C6F5 3 32 11.6 mg for MMAO-1 was added to the above MAO 250 mg dissolved
Ž
.
in toluene 10 ml at room temperature. The solution was gradually heated up to 60–658C and stirred
for 12 h at the temperature. After the solution was cooled to room temperature, glassy residue was
1
obtained by evaporating in vacuo Y)80% : H NMR toluene-d8 d 0.06 br . For 19F NMR
Ž
.
Ž
.
Ž .
Ž
Ž
. .
spectra, the reactants 28 mg of MAO, 25 mg of B C6F5 3 were added in an NMR tube, subsequently
Ž
.
toluene-d8 0.5 ml was vacuum-transferred, and finally the tube was flame-sealed to keep volatile
products from escape. After the NMR tube was heated at 658C for 12 h, 19F and 11B NMR spectra
were recorded. 19F NMR toluene-d8 d y49.7 br, ortho , y77.7 br, para , y86.6 br, meta . 11B
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
Ž .
NMR toluene-d8 : d 72.3 s . When the sample was rigorously evacuated to dryness and redissolved
in toluene-d8, no 11B resonance was observed at all. The dried sample was analyzed. Anal. Found: C,
30.13; H, 1.74; Al, 23.03; B: 0.20.
(
)
(
)
2.6. Modification of methylalumoxane with Al C6 F5 3 P toluene 0.5
Ž
.
Ž
.
Ž
13.5 mg for MMAO-2; 26.7 mg for MMAO-3; 53.4 mg
0.5
Certain amount of Al C6F5 3 P toluene
.
Ž
.
Ž
for MMAO-4 was added to the above MAO 250 mg, ca. 4.3 mmol of Al dissolved in toluene 10

(
)
234
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
.
ml at room temperature. The solution was gradually heated up to 60–658C and stirred for 12 h. After
the solution was cooled to room temperature, glassy residue was obtained by evaporating in vacuo
1
Y)80% : H NMR toluene-d8 d 0.04 br ; Only broad peaks are observed in 19F NMR
Ž
.
Ž
.
Ž .
19
Ž
.
Ž
.
Ž
.
Ž
.
spectroscopy. F NMR toluene-d8 d y48.5 br, ortho , y78.3 br, para , y86.9 br, meta .
2.7. Polymerization of ethylene
Ž
.
Ž
.
Out of toluene solution 20 ml dissolving zirconocene dichloride 2.0 mg , 1.0 ml of the solution
was taken, and it was transferred in a 500 ml heavy-wall glass pressure reactor. A MAO was weighed
to adjust the molar ratio of aluminum to zirconium to 1600:1. Together with the measured MAO, 10%
Ž
.
trimethylaluminum in toluene 1.0 ml and 250 ml of toluene were added in the reactor. The resulting
solution was stirred at 808C under 40 psi of ethylene for 15 min. The polymerization was quenched by
Ž
.
evacuation of ethylene and subsequent addition of ethanolic solution 300 ml of HCl. White
precipitate was filtered and washed with copious ethanol, and finally dried in a vacuum oven at 608C
for 12 h.
2.8. Polymerization of propylene
5
Ž
.
Ž
Ž
..
Out of toluene solution 20 ml dissolving rac- 1,2-ethylenebis h -indenyl zirconium dichloride
Ž
.
5.0 mg , 2.0 ml of the solution was taken, and it was transferred in a 500 ml heavy-wall glass
pressure reactor. Each MAO was weighed to adjust the molar ratio of aluminum to zirconium to
Ž
.
1600:1. Together with the measured MAO, 10% trimethylaluminum in toluene 1.0 ml and 250 ml of
toluene were added in the reactor. The resulting solution was stirred at 808C under 40 psi of propylene
for 1 h. The polymerization was quenched by evacuation of propylene and subsequent addition of
methanolic solution of HCl. After toluene was evaporated to reduce its volume to about 100 ml,
acetone was added to produce a white precipitate. The precipitate was filtered, and washed with
copious methanol and subsequently copious acetone, and finally dried in a vacuum oven at 608C for
12 h.
3. Results and discussion
(
)
3.1. Synthesis of Al C6 F5 3 PsolÕent
Ž
.
Ž
.
w
x
Preparation of Al C6F5 3 Pdiethyl ether and Al C6F5 3 PTHF has been reported previously 35,36 .
These compounds were heated under vacuum in order to eliminate the oxygen-containing solvents
from the Lewis acidic center, but the THF adduct only decomposed at elevated temperatures
Ž
.
Ž
.
)2008C . Special care was taken for the heating because there are reports that Al C6F5 3 andror its
w
x
adducts are explosive 36,37 .
Ž .
Ž
.
,
It was observed that as soon as trimethylaluminum l was contacted with solvent-free B C6F5 3
bubbles, later identified as BMe3 boiling pointsy208C , were formed. 19F NMR spectrum of
residual solid showed that new resonances were very similar to those of Al C6F5 3 PTHF Eq. 1 .
Ž
.
Ž
.
Ž
Ž ..
Ž
.
Ž
.
When toluene was used as a solvent for the better mixing, only a Al C6F5 3 P C7H8 0.5 was detected
1
Ž
Ž ..
after evaporating the reaction mixture in vacuo Eq. 2 . Presence of toluene is obvious from H
NMR spectrum, and the stoichiometry of toluene has been determined as 0.50 from elemental
analysis. The product is pure enough to use it without further purification steps. No necessity to use

(
)
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
235
Table 1
Stretching frequency of CN bond of benzonitrile and benzonitrile acid adducts
y1
a
y1
Ž
. Ž
.
Ž
Difference cm
.
Compounds
n CN cm
PhCN
PhCNPAlMe3
2230
b
b
Ž
.
Ž
.
2276 this work , 2271
46 this work , 41
Ž
.
PhCNPAl C6 F5
2280
2324
50
94
3
Ž
.
PhCNPB C6 F5
3
aAs KBr pellets, except benzonitrile.
b
w
x
Ref. 38 .
oxygen-containing solvents is one of advantages of the metathesis. Also, in our laboratory, impure
Ž
.
Ž
.
Ž
.
Ž
.
solvent-free Al C6F5 3, purified Al C6F5 3 P toluene 0.5, and Al C6F5 3 PTHF have been heated up to
2008C under nitrogen for the test. However, no incidence of explosion has been met in our laboratory.
It is interesting to notice that trimethylaluminum gives a clean metathesis, but triethylaluminum
w
x
Ž
.
does not. Pohlmann and Brinckmann 36 reported that B C6F5 3 and the latter aluminum reagent
gave only tarry products at 508C, and even detonated when heated to 708C. Out of NMR analysis we
Ž
.
found that the reaction with triethylaluminum produces the Al C6F5 3 along with numerous side
products at room temperature. Obviously, volatility of trimethylboron helps to drive the reaction
complete.
The product is soluble in benzene, toluene, and other hydrocarbon solvents, and too soluble to be
recrystallized in these solvents. The product was dissolved in a minimum amount of THF, and hexane
was layered on the top of the THF solution. In several days, white crystals of which spectral data are
1
Ž
.
w x
exactly same as those of Al C6F5 3 PTHF were obtained 35 .
Ž .
1
Ž .
2
(
)
(
)
3.2. Lewis acidity of AlMe3 , Al C6 F5 3 , and B C6 F5 3
Lewis acidity of the aluminum and the boron reagents was measured. It is known that stretching
w
x
frequency of CN bond of benzonitrile acid adduct is dependent on strength of Lewis acids 38 . The
stronger a Lewis acid is, the higher stretching frequency of the benzonitrile acid adduct is observed.
Ž
.
As can be seen in Table 1, Al C6F5 3 adducts absorb at slightly higher frequency than AlMe3, but
Ž
.
B C6F5 3 adduct resonates at much higher frequency than those two aluminum adducts. The
spectroscopic result clearly shows that the boron reagent is the stronger Lewis acid than the other two
aluminum reagents, and the pentafluorophenyl substituent is only marginally effective to increase
Lewis acidity.
1
Ž
.
Converted chemical shifts y123.4, y151.5, y161.1 ppm in reference to CFCl3 are in accord with the reported values.

(
)
236
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
3.3. Preparation of modified MAO
Facile metathesis between AlMe3 and B C6F5 3 implied that the methyl group of MAO can be
Ž
.
Ž
.
Ž
.
easily exchanged by C6F5 group of B C6F5 3, even of Al C6F5 3. While Michiels and Munoz-Escalona
39 observed the enhanced reactivity of MAO when B C6F5 3 was used as an additive, the
overlooked metathesis could be associated with the modulation of the activity. The process is
˜
w
x
Ž
.
economical because it does not lose precious C6F5 group. Significant portion of the group will be lost
Ž
.
Ž
.
as C6F5H if a mixture of AlMe3 and Al C6F5 3 or B C6F5 3 was hydrolyzed to form modified MAO.
Ž
.
When B C6F5 3 was mixed with the above prepared MAO in toluene-d8 and heated, it was indeed
found that there was a facile exchange. After heating, only broad resonances in 19F NMR spec-
troscopy were observed, and concurrent generation of a single sharp peak of BMe3 was observed in
11
B NMR spectroscopy 40 . The 19F NMR spectrum is very similar to that of MAO modified with
w
x
Al C6F5 3 vide infra . After evaporating the sample, the 11B resonance is totally disappeared.
However, elemental analysis of the dried sample showed that most of boron was remained in the
oligomeric chain. Therefore, it can be said that only small part of boron reagent transformed into
Ž
.
Ž
.
Ž
Ž ..
volatile BMe3, but most of boron is incorporated into the oligomeric chain Eq. 3 . Also, most of
C6F5 group are bound to the aluminum center rather than the boron center, presumably due to
statistical reason.
Ž
.
Ž
.
When Al C6F5 3 P toluene
0.5
is mixed with the MAO in toluene-d8 and heated, broad resonances
were appeared, and no sharp resonance attributed to Al C6F5 3 is remained in 19F NMR spectroscopy.
Ž
.
Ž
.
The positions of the broad resonances are exactly same as for the B C6F5 3 case. The spectroscopic
data support that there is a complete metathesis to generate AlMe3 together with modified MAO Eq.
Ž
Ž ..
4 .
Because all of MAO contains variable amount of AlMe3, there is also possibility that both boron
and aluminum reagents metathesize with the trimethylaluminum concurrently. However, the fact that
major portion of C6F5 group is incorporated into the MAO is evident from the broadening of 19F
NMR spectra. Also, very similar 19F NMR spectrum has been obtained from the direct partial
Ž
.
hydrolysis of Al C6F5 3
.
Ž .
3
Ž .
4
3.4. Catalytic actiÕity for ethylene polymerization
Catalytic activity of pristine MAO and modified MAOs are tested in the presence of Cp2 ZrCl2 in
Ž
.
toluene. Ratio of Cp2 ZrCl2 and MAO was maintained at 1:1600 mol of Zr vs. Al and is not
optimized. As can be seen in Table 2, the catalytic activity is increased significantly by substituting

(
)
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
237
Table 2
Catalytic activity of pristine MAO and modified MAOs for ethylene polymerization
Co-catalyst
Molar ratio MeAlO : Al C6 F5 3 P toluene
Activitya kg PErg cat h
Žw
x
Ž
.
Ž
.
.
Ž
.
0.5
pristine MAOb
MMAO-1c
MMAO-2
MMAO-3
MMAO-4
y
39
144
272
176
110
100:0.50
100:0.55
100:1.1
100:2.2
a The activity was measured under 40 psi of ethylene at 808C. Weights of the polyethylene was extrapolated from those obtained at 15 min
of the reaction.
b For comparison the MAO was heated at 658C for 12 h without addition of the aluminum or the boron reagents.
c
Ž
.
B C6 F5 is used for the case.
3
Table 3
Catalytic activity of MMAO-2 at various AlrZr ratios for ethylene polymerizationa
Co-catalyst
AlrZrb s1600
AlrZrs400
AlrZrs100
MMAO-2
272
95
18
a The activity was measured under 40 psi of ethylene at 808C. Weights of the polyethylene was extrapolated from those obtained at 15 min
of the reaction. Units are in kg PErg cat h.
bAlrZr represents the molar ratio of aluminum center of MMAO-2 to zirconocene dichloride.
Ž
.
some of the methyl group with C6F5 group. Modification with B C6F5 6 increases the activity by
Ž
.
Ž
.
3.7-fold. With Al C6F5 3 P toluene 0.5, further increase of the activity up to 7-times is achieved.
However, heavier substitution reduces the favorable effect of the fluorinated phenyl group.
While the sizable increase of the catalytic activity was observed for MMAO-2, the amount of
MMAO-2 was varied to elucidate its influence to the activity. As in Table 3, the molar ratio of
aluminum center of MMAO-2 to zirconocene dichloride has been reduced from 1600 to 400 and 100.
With the constant amount of trimethylaluminum as a scavenger, the activity changes from 272 kg
PErg cat h to 95 kg PErg cat h, and it decreases finally to 18 kg PErg cat h. Therefore, as for the
pristine MAO, the amount of the employed modified MAO greatly affects the catalytic activity.
It can be speculated that the new aluminum center having the noncoordinating substituent, i.e.,
pentafluorophenyl group is more effective for the abstraction of a methyl group from zirconocene.
Alternatively the new center increases the reactivity due to the reduced association between the
cationic catalytic center and the counter-ion. For the cases, thermodynamic favor of the new center
Ž
.
over the methyl aluminum center should be assumed Scheme 1 , because the enhanced reactivity is
evident even with the modification of the low extent. Also, the heavier modification of MAO as well
as the incorporation of boron atom could alter the desirable original structure, and the alterations must
result in the reduction of the effectiveness.
3.5. Catalytic actiÕity for propylene polymerization
Ž
Catalytic activity of pristine MAO and modified MAOs are tested in the presence of rac- 1,2-eth-
5
Ž
..
Ž
.
ylenebis h -indenyl ZrCl2 in toluene. Ratio of rac- EBI ZrCl2 and MAO was maintained at 1:1600
Scheme 1.

(
)
238
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
Table 4
Catalytic activity of pristine MAO and modified MAOs for propylene polymerization
Co-catalyst
Molar ratio MeAlO : Al C6 F5 3 P toluene
Activitya kg PPrg cat h
Žw
x
Ž
.
Ž
.
.
Ž
.
0.5
pristine MAOb
MMAO-2
MMAO-3
y
20
26
13
10
100:0.55
100:1.1
100:2.2
MMAO-4
a
Ž
.
Ž
.
The polymerization was carried out under 40 psi of propylene at 808C for 1 h using rac- EBI ZrCl2 AlrZrs1600 .
b For comparison the MAO was heated at 658C for 12 h without addition of the aluminum reagent.
Ž
.
mol of Zr vs. Al as in the polymerization of ethylene. As in Table 4, the catalytic activity is
modulated by substituting some of the methyl group of MAO with C6F5 group. The most effective
MAO for ethylene polymerization, that is, MMAO-2 produces the highest yield of polypropylene.
Therefore, the effectiveness of the substitution with C6F5 group is evident for both polymerizations,
Ž
.
even though the effectiveness is so reduced that only marginal increase of the activity 30% is
observed for propylene polymerization. Rather highly modified MAOs such as MMAO-3 and
MMAO-4 produce lower weight of polypropylene than that for the pristine MAO. The last
undesirable effect with the high modification is also observed for ethylene polymerization.
4. Conclusion
Ž
.
Ž
.
A facile metathesis between AlMe3 and B C6F5 3 produces tris pentafluorophenyl aluminum in a
quantitative yield. Thermal exchange between the methyl group of MAO and the pentafluorophenyl
Ž
.
Ž
.
group of both B C6F5 3 and Al C6F5 3 was observed. The modification of MAO with the noncoordi-
nating group modulates the catalytic reactivity for the olefin polymerization.
Acknowledgements
Ž
This work is supported by the Korea Foundation of Science and Engineering Program number
.
931-0300-003-1 .
References
w x
Ž .
H. Sinn, W. Kaminsky, Adv. Organomet. Chem. 18 1980 99.
H. Sinn, W. Kaminsky, H.-J. Vollmer, R. Woldt, Angew. Chem., Int. Ed. Engl. 19 1980 390.
1
w x
2
Ž .
w x
Ž .
S. Pasynkiewicz, Polyhedron 9 1990 429.
3
w x
Ž
.
4
G. Fink, R. Mulhaupt, H.H. Brintzinger Eds. , Ziegler Catalysts, Springer-Verlag, Berlin, 1995.
¨
w x
5
J. Boor Jr., Ziegler-Natta Catalysts and Polymerization, Academic Press, New York, 1979.
w x
Ž
.
6
W. Kaminsky, H. Sinn Eds. , Transition Metals and Organometallics as Catalysts for Olefin Polymerization, Spring-Verlag, Berlin,
1988.
w x
Ž .
P.J.T. Tait, in: S.G. Allen Ed. , Comprehensive Polymer Science, Vol. 4, Pergamon, Oxford, 1989, pp. 1–25.
7
w x
8
Ž .
P. Pino, R. Mulhaupt, Angew. Chem., Int. Ed. Engl. 19 1980 857.
¨
w x
Ž .
A.R. Barron, Organometallics 14 1995 3581.
9
w
w
w
x
Ž
.
10 S.S. Reddy, G. Shashidhar, S. Sivaram, Macromolecules 26 1993 1180.
x
11 M.S. Howle, MetCon ’93, Catalyst Consultant, Houston, 1993, pp. 246–266.
x
Ž
.
12 C.J. Harlan, S.G. Bott, A.R. Barron, J. Am. Chem. Soc. 117 1995 6465.

(
)
C.H. Lee et al.rJournal of Molecular Catalysis A: Chemical 132 1998 231–239
239
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
w
x
Ž
.
13 K. Peng, S. Xiao, J. Mol. Catal. 90 1994 201.
x
Ž
.
14 K. Peng, S. Xiao, Makromol. Chem. Rapid Commun. 14 1993 633.
x
Ž
.
15 E. Giannetti, G.M. Nicoletti, R. Mazzocohi, J. Polym. Sci., Chem. Ed. 23 1985 2117.
x
16 R. Davis, US Patent 4,968,827.
x
17 S. Bottelberghe, US Patent 4,908,463.
x
18 H. Sinn, D. Clausnitzer, H. Winter, US Patent 5,087,713.
x
19 S. Sangokoya, US Patent 5,041,583.
x
Ž
.
20 M. Bochmann, S.J. Lancaster, J. Organomet. Chem. 434 1992 C1.
x
Ž
.
21 G.W. Coates, R.M. Waymouth, J. Am. Chem. Soc. 115 1993 91.
x
Ž
.
22 A.M. Thayer, Chem. Eng. News 11 Sep. 1995 15.
x
Ž
.
23 G.W. Coates, R.M. Waymouth, Science 267 1995 217.
x
Ž
.
.
24 M.R. Mason, J.M. Smith, S.G. Bott, A.R. Barron, J. Am. Chem. Soc. 115 1993 4971.
x
Ž
25 T. Sugano, K. Matsubara, T. Fujita, T. Takahashi, J. Mol. Catal. 82 1993 93.
x
Ž
.
26 H. Sinn, Macromol. Symp. 97 1995 27.
x
Ž
.
27 S.H. Strauss, Chem. Rev. 93 1993 927.
x
Ž
.
28 L. Jia, X. Yang, A. Ishihara, T.J. Marks, Organometallics 13 1995 3135.
x
Ž
Ž
.
.
29 X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 116 1994 10015.
x
30 X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 113 1991 3623.
x
Ž
.
31 L. Jia, X. Yang, C. Stern, T.J. Marks, Organometallics 13 1994 3755.
x
Ž
.
32 P.A. Deck, T.J. Marks, J. Am. Chem. Soc. 117 1995 6128.
x
Ž
.
33 A.G. Massey, A.J. Park, J. Organomet. Chem. 2 1964 245.
x
34 D.F. Shriver, M.A. Drezdzon, The Manipulation of Air-Sensitive Compounds, Wiley, New York, 1986.
x
Ž
Ž
.
.
35 T. Belgardt, J. Storre, H.W. Roesky, M. Noltemeyer, H.-G. Schmidt, Inorg. Chem. 34 1995 3821.
x
36 J.L.W. Pohlman, T.E. Brinckmann, Z. Naturforsch., B: Anorg. Chem., Org. Chem. 20 1965 5.
x
Ž
.
37 J.A. Ewen, in: K. Soga, M. Terano Eds. , Catalyst Design for Tailor-Made Polyolefins, Elsevier, Amsterdam, 1994, pp. 405–410.
x
Ž
.
38 K. Starowieyski, S. Pasynkiewicz, M. Boleslawski, J. Organomet. Chem. 10 1967 393.
x
Ž
.
39 W. Michiels, A. Munoz-Escalona, Macromol. Symp. 97 1995 171.
˜
x
40 H. Noth, B. Wrackmeyer, Nuclear Magnetic Resonance Spectroscopy of Boron Compounds, Springer, Berlin, 1978.
¨