Figure 3. Stereoerrors xstereo vs T for iPP samples made with catalyst
p
systems based on C1−C5. Data for C1 at 150 °C were not available.
The iPP average molar mass data in Table 1 deserve a careful
analysis. Whereas catalysts C4 and C5 were superior to C1−C3
Figure 4. (A) Visualization of the rigid active pocket of C4: topographic
map of steric bulk (SambVca 2.0, higher steric bulk denoted by darker
shades of red). (B) Major sources of steric interference of the ligand
with the BHT TS (shortest C−C contacts to 2-Me and triptycene
“paddle wheel”) and resulting severe ligand distortion (blue, indenyl
plane; red, plane through the triptycene axis carbons).
in this respect as well, the drop in M between 60 and 150 °C
61
n
(
from >1 MDa to ≤10 kDa) was dramatic for all of the systems
tested. In general, β-H elimination (BHE) becomes important
for zirconocene catalysts, including 2-R-substituted catalysts, at
T > 100 °C, and as a result M increases with increasing
p
n
monomer concentration. The experiments of Table 1 were
carried out at a “moderate” propene partial pressure (6.5 < pC3
bar); therefore, an additional set of experiments for C4, C5,
and C1 for comparison was run at 120 °C and pC3 ≈ 30 bar in a
<
8
with the desired and fixed 90° dihedral angle relative to the
indenyl plane; (c) the triptycene motif brings steric bulk back
for comparison to the distortion found in the BHT TS of C1).
The ligand wraps around the active metal center and defines a
rigid catalytic pocket in which the TSs of all undesired reactive
Under the new set of conditions (Table 2), the iPP samples
produced with C4 and C5 featured M values ≥350 kDa (to be
n
Table 2. Results of Propene Polymerization and iPP
32,60
events, including growing chain epimerization
and BHT,
Characterization for Catalyst Systems Based on C1, C4, and
a
can only be accommodated with severe distortions. Unlike most
traditional olefin polymerization catalysts, C4 and C5 are
essentially void of any rotational degrees of freedom, ensuring a
broad operational window and high performance even at high
temperatures.
C5 at 120°C and “High” Propene Pressure
b
ID
Rp
2,1 (%) 3,1 (%) 1-σ (%) M (kDa)
Đ
Tm (°C)
n
c
C1
C4
C5
12
209
575
0.35
0.59
0.44
0.24
0.02
0.03
0.28
<0.03
23
360
2.6
2.2
146.6
156.0
157.9
d
d
In conclusion, our integrated HTE/QSAR approach led us to
a
experimental conditions: 150 mL of heptane, activator/scavenger
identify two novel C -symmetric bis(1-indenyl) ansa-zircono-
2
cene catalysts with superior high-temperature performance in
isotactic-selective propene solution polymerization, thanks to a
ligand framework uniquely combining a smart design and a
rigidity that is rare in the “soft” world of organometallics. We
believe that this achievement represents a happy ending to an
almost 30-year-long story. Yet, we know well that, whenever a
story ends, a new one is about to start, and like the itsy-bitsy
spider we are prepared to climb up the spout again.
5
b
−1 −1
c
MAO, [Al]/[Zr] = (1.5−2.0) × 10 . R in kg mmol
of iPP (64%) and aPP (36%), stereo- and regioselectivity values refer
to iPP. The reaction temperature was not fully controlled (the
h . Mixture
p
Zr
d
activity was too high to avoid overheating). Therefore, R was likely
p
overestimated and the polymer regularity and average molar mass
were somewhat underestimated.
compared with 23 kDa for the sample made with C1) and Tm
values (DSC, second heating) of 155.9 and 157.9 °C,
respectively (146.6 °C for C1). Moreover, the catalyst
productivity of C4 and C5 was 20−50 times higher than that
ASSOCIATED CONTENT
sı Supporting Information
■
*
Synthetic procedures and characterization, detailed NMR
spectra, polymerization procedures, polymer character-
ization procedures, and polymer analytical character-
izations, enantioselectivity estimations, computational
details for QSAR models, predicted vs observed perform-
ance of C4 and C5, propene consumption profiles for C1,
C4, and C5 at 130 °C, and a comparison of ligand
distortions in C1 and C4 (PDF)
To our knowledge, C4 and C5 show the most extreme
stereoselectivities reported to date at 120 °C, reaching an
enantiodiscrimination of 6−7 kcal/mol: i.e., one misinsertion
every 4000 insertions! At the same time, the molar mass
capability remains high. Looking at Figure 4, one can speculate
why the active species of C4 and C5 perform so well: (a) the
steric bulk of the ligand frame is almost ideally distributed
between empty and occupied quadrants (Figure 4A); (b) the
“bidentate 4,5-substituent” locally mimics a 4-Ph substituent
Coordinates for DFT structures (XYZ)
7
644
J. Am. Chem. Soc. 2021, 143, 7641−7647