Access to this full-text is provided by Wiley.
Content available from Angewandte Chemie International Edition
This content is subject to copyright. Terms and conditions apply.
Polymer Chemistry
Aluminum Alkyl Induced Isomerization of Group IV meso
Metallocene Complexes
Tim M. Lenz, Ion Chiorescu, Fabrizio E. Napoli, Jin Y. Liu, and Bernhard Rieger*
Abstract: The synthesis of group IV metallocene preca-
talysts for the polymerization of propylene generally
yields two different isomers: The racemic isomer that
produces isotactic polypropylene (iPP) and the meso
isomer that produces atactic polypropylene (aPP). Due
to its poor physical properties, aPP has very limited
applications. To avoid obtaining blends of both poly-
mers and thus diminish the mechanical and thermal
properties of iPP, the meso metallocene complexes need
to be separated from the racemic ones tediously—
rendering the metallocene-based polymerization of
propylene industrially far less attractive than the
Ziegler/Natta process. To overcome this issue, we
established an isomerization protocol to convert meso
metallocene complexes into their racemic counterparts.
This protocol increased the yield of iPP by 400 % while
maintaining the polymer’s excellent physical properties
and was applicable to both hafnocene and zirconocene
complexes, as well as different precatalyst activation
methods. Through targeted variation of the ligand
frameworks, methoxy groups at the indenyl moieties
were found to be the structural motifs responsible for an
isomerization to take place—this experimental evidence
was confirmed by density functional theory calculations.
Liquid injection field desorption ionization mass spec-
trometry, as well as 1H and 29Si nuclear magnetic
resonance studies, allowed the proposal of an isomer-
ization mechanism.
Introduction
The discovery of the coordinative polymerization of
ethylene and propylene by Ziegler and Natta in the 1950s
marked arguably the most crucial breakthrough in polymer
research to this date.[1] 30 years later, Brintzinger and
Kaminsky established well-defined single-site, homogeneous
ansa-zirconocene complexes for the iso-selective polymer-
ization of propylene.[2] Using methylaluminoxane (MAO) as
a co-catalyst, they were able to produce polypropylene with
far higher molecular weights and more narrow molecular
weight distributions than the Ziegler/Natta systems based on
TiCl4/ID/MgCl2(ID =internal electron donor). Nowadays,
research regarding the homogeneous polymerization of
propylene still relies on this pioneering work. However,
maximizing the catalyst performance, as well as the stereo-
regularity and molecular weight of the produced polymer,
became the main focus, and novel techniques such as high-
throughput experimentation are being established to achieve
these goals.[3] When it comes to highly isotactic polypropy-
lene, indenyl-based metallocenes play a crucial role. Using
the ultra-rigid ansa-hafnocene complex rac-IHf, our group
was able to produce iPP with the highest melting transition
temperature ex reactor to this date.[4] One issue regarding
not only our benchmark catalyst but all indenyl-based
metallocene precatalysts is the laborious and sometimes
impossible separation of the racemic (rac) and meso isomers
of these complexes.[5] While the racemic metallocene species
yield isotactic polypropylene (iPP), their meso analogs are
known to produce undesired atactic polypropylene (aPP)
according to Ewen’s symmetry rules.[6] Since generally, both
isomers are formed upon attachment of the ligand to the
central metal atom, obtaining the racemic isomers in a very
pure manner is mandatory before any precatalyst can be
considered suitable for the iso-selective polymerization of
[*] T. M. Lenz, J. Y. Liu, Prof. Dr. B. Rieger
Wacker-Lehrstuhl für Makromolekulare Chemie
Catalysis Research Center
Technische Universität München
TUM School of Natural Sciences
Lichtenbergstraße 4, 85748 Garching
Garching bei München (Germany)
E-mail: rieger@tum.de
Dr. I. Chiorescu
Department Chemie
Technische Universität München
TUM School of Natural Sciences
Lichtenbergstraße 4, 85748 Garching
Garching bei München (Germany)
F. E. Napoli
Lehrstuhl für Anorganische und Metallorganische Chemie
Catalysis Research Center
Technische Universität München
TUM School of Natural Sciences
Lichtenbergstraße 4, 85748 Garching
Garching bei München (Germany)
© 2024 The Authors. Angewandte Chemie International Edition
published by Wiley-VCH GmbH. This is an open access article under
the terms of the Creative Commons Attribution Non-Commercial
NoDerivs License, which permits use and distribution in any med-
ium, provided the original work is properly cited, the use is non-
commercial and no modifications or adaptations are made.
Angewandte
Chemie
Research Article www.angewandte.org
How to cite: Angew. Chem. Int. Ed. 2024,63, e202406848
doi.org/10.1002/anie.202406848
Angew. Chem. Int. Ed. 2024,63, e202406848 (1 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
propylene. This purification usually consists of several frac-
tional crystallizations of the complexes and thus diminishes
the yield of the racemic isomer; furthermore, considerable
amounts of the precious ligands are wasted in the form of
discarded meso isomers. Considering the expensive and
time-consuming preparations of metallocene precatalysts,
this fact contributes to the industrial unprofitability of the
homogeneous polymerization of propylene. Even though
some protocols exist to convert meso metallocene complexes
into their racemic analogs, their applicability is limited as
they often reach an equilibrium state between both isomers,
and they add an additional, labor-consuming purification
step to the precatalyst syntheses.[7] Recently, we reported
the isomerization of hafnocene complex meso-IHf upon
conversion with 1.00 equivalent (eq.) of triisobutylaluminum
(TIBA).[8] The subsequent in situ activation using 200 eq.
TIBA and 5.00 eq. [Ph3C][B(C6F5)4] (TrBCF) afforded
perfectly isotactic polypropylene with macromolecular char-
acteristics analog to polypropylene produced by the pure
racemic isomer.
In this work, we optimized the isomerization conditions
regarding TIBA concentration, reaction temperature, and
time, thus achieving a more thorough isomerization. We
explored the structural motif responsible for the isomer-
ization and postulated a reaction mechanism based on
density functional theory (DFT) calculations, as well as
liquid injection field desorption ionization mass spectrome-
try (LIFDI-MS) and various nuclear magnetic resonance
(NMR) experiments. Currently, both research and industry
prefer zirconocene complexes for the production of poly-
propylene, while their heavier analogs experience far less
attention.[9] For this reason, we extended the scope of the
isomerization protocol to different zirconocene/MAO sys-
tems but also structurally assorted hafnocene complexes
proven by 1H and 29Si NMR studies.
Results and Discussion
Complex Syntheses
The isomerization behavior of complex meso-IHf upon TIBA
addition was investigated thoroughly in our previous work.[8]
To identify the structural motifs of the ligand framework
responsible for the isomerization to occur, the hafnocene
complexes meso-IHf–VHf, as well as their corresponding
zirconocene analogs meso-IZr–VZr were synthesized, and
their reactions upon TIBA addition were evaluated
(Scheme 1).
The syntheses of complexes I–Vafforded isomeric
mixtures ranging from rac/meso 2/1 to pure meso, depending
on each individual precatalyst. The presence of racemic
isomers did not affect the isomerization behavior of the
corresponding meso species. Therefore, we did not isolate
the pure meso isomers of all complexes. Complexes IIZr,
IIIHf, and IVHf have previously not been reported in the
literature. Their synthesis was carried out analog to the
literature-known analogs and afforded rac/meso
mixtures.[3b,10] The racemic isomers were separated through
fractional crystallization and thoroughly characterized using
1H, 13C, and 29Si NMR spectroscopy, as well as LIFDI-MS.
IIZr was additionally characterized using single-crystal X-ray
diffraction (SC-XRD).[11] Complex VHf has previously been
reported in the literature,[12] however, was never thoroughly
characterized to the best of our knowledge. Thus, we carried
out its characterization using NMR, MS, and SC-XRD.
Optimization of Isomerization Conditions
In our previous work, isomerization experiments were
carried out dissolving the metallocene complex and 1.00 eq.
TIBA in toluene with the mixtures being heated to 100 °C
for 16 hours for complete conversion from meso to rac. This
procedure, however, afforded small amounts of side prod-
ucts visible in the 1H NMR spectra of the isomerized
complexes.[8] While these side products did not affect the
polymerization performance of the activated catalysts, we
still aimed to achieve an entirely clean isomerization to
maximize the yield of the racemic isomers and, thus, the
later-produced isotactic polypropylene. We therefore opti-
mized the isomerization conditions and found that side
products could be avoided by reducing the amount of TIBA
to 0.50 eq. and increasing the reaction temperature to
120 °C; conversions were monitored using 1H NMR spectro-
scopy. Sixteen hours of reaction time were necessary for a
complete conversion of complex meso-IHf, while only one to
four hours were required for complexes meso-IZr,meso-IIHf,
and meso-IIZr. The absence of other signals besides the
racemic products in the 29Si NMR spectra proved an actual
isomerization of the complexes to the corresponding racemic
species took place—not just a degradation of the meso
isomers (Scheme 2). Direct comparison of the respective
metallocene complexes revealed that the isomerization of
zirconocene complexes tended to occur faster (3 h for meso-
IZr and 1 h for meso-IIZr) than the isomerization of the
corresponding hafnocene analogs (16 h for meso-IZr and 4 h
for meso-IIZr). Trimethylaluminum (TMA) was also suitable
Scheme 1. Syntheses of metallocene complexes I–V.
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (2 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
to isomerize these meso complexes; however, due to the
known fact that TMA severely hampers the activity of
hafnocene complexes regarding the polymerization of
propylene, it was only used for mechanistic studies and not
for polymerization experiments.[13]
In the absence of TIBA or TMA, no isomerization was
observed for complex meso-IHf,while partial isomerization
was observed for meso-IIHf,meso-IZr, and meso-IIZr. How-
ever, the formation of side products was predominant in all
of these control experiments. No isomerization was ob-
served for complexes meso-III,meso-IV, and meso-V(M:
Zr, Hf) under various conditions, proving that both the
methoxy groups of the metallocene complexes, as well as
the presence of an aluminum alkyl were crucial for an
isomerization to take place. The role of both will be
discussed later on.
Polymerization Experiments Using Isomerized Metallocene
Complexes
To compare the polymerization performances of isomerized
meso-Iand meso-II (M: Zr, Hf), as well as their racemic
analogs, propylene polymerizations were carried out. The
resulting polymers were characterized according to their
molecular weight, stereoregularity, and melting transition
temperature (Table 1).
Polymerizations involving hafnocene complexes were
conducted using the in situ activation method with TIBA/
TrBCF in analogy to our previous work. As evidenced by
Table 1, the isomerization of meso-IHf and meso-IIHf and
subsequent in situ activation yielded iPP with equal proper-
ties to iPP produced from the corresponding pure racemic
isomers, proving a complete isomerization. Contrary to
hafnocene complexes, which are usually insufficiently acti-
vated when MAO is used,[13] zirconocene precatalysts can
easily be activated with MAO; this activation is generally
more frequently employed.[9a,14] Thus, in addition to the in
situ activation of hafnocene complexes, we aimed to extend
our isomerization/polymerization protocol to zirconocene
complexes, subsequently activated with MAO. iPP obtained
from the isomerized zirconocene complexes meso-IZr and
meso-IIZr,which were activated with MAO, exhibited the
same macromolecular characteristics as iPP produced by the
direct MAO activation of the corresponding racemic
isomers. Evidently, the MAO activation worked just as well
for previously isomerized zirconocene complexes and was
additionally not hampered by residual TIBA used for the
isomerization. Many commercially available MAO solutions
are enhanced through the addition of TIBA, anyway.[15] In
our previous work,[8] we showed that the direct addition of
Scheme 2. Excerpts of the 29Si NMR spectra of a mixture of rac-IZr and
meso-IZr before and after isomerization according to the depicted
reaction using 0.50 eq. TIBA at 120°C for 3 h. Full spectra for all
isomerization experiments are provided in the Supporting Information.
Table 1: Conditions and results for the polymerization of propylene with complexes rac/meso-Iand II (M: Zr, Hf).[a]
entry precatalyst n[b] activation[c] TPolym[d] tPolym[e] [mmmm] [f] Mw[g] Ð[h] Tm[d] P[i]
1[7] rac-IHf 1.65 TIBA/TrBCF 30 30 >99 1600 1.6 165 6000
2[7] rac/meso-IHf (1/4) [j] 1.65 TIBA/TrBCF 30 30 >99 1300 1.6 165 5000
3rac-IIHf 1.30 TIBA/TrBCF 30 15 99 500 1.8 160 1200
4rac/meso-IIHf (1/4) [j] 1.30 TIBA/TrBCF 30 15 >99 700 1.6 161 1000
5rac-IZr 1.64 MAO 30[k] 15 >99 700 2.9 159 32000
6rac/meso-IZr (1/4) [j] 1.64 MAO 30 15 >99 800 2.5 159 30000
7rac-IIZr 1.45 MAO 60 15 98 600 1.6 156 10000
8rac/meso-IIZr (1/4) [j] 1.45 MAO 60[k] 15 98 700 2.1 155 15000
[a] Vtoluene =120 mL; p=pAr +ppropylene =4 bar, pAr =1.5 bar. [b] In μmol. [c] TIBA/TrBCF: initiator [Ph3C][B(C6F5)4]=5.00 eq., activator (TIBA) =
200 eq., scavenger (TIBA) =0.55 mmol; methylaluminoxane (MAO): scavenger =activator (MAO) =2000 eq. [d] In °C. [e] In min. [f] In %,
determined via 13C NMR spectroscopy assuming the enantiomorphic site model. [g] In kg mol1, determined absolutely via SEC-GPC in 1,2,4-
trichlorobenzene at 160 °C with dn/dc =0.097 mL g1. [h] Ð=Mw/Mn. [i] in kgPP [molcat h]1. [j] Pre-isomerized with 1.00 eq. (entry 2)/0.50 eq.
(entry 4, 6, 8) TIBA at 120°C. [k] �10 °C.
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (3 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
MAO to meso metallocene complexes, however, led to the
formation of atactic polypropylene which we attributed to
MAO’s ability to generate a free coordination site by itself.
This highlights the necessity of a complete isomerization
using either TIBA or TMA before MAO is employed.
One of the most considerable complications regarding
polymerizations relying on metallocene precatalysts is that
their syntheses yield meso and racemic isomers in ratios of
up to 4/1.[8] The herein established isomerization/polymer-
ization protocol circumvents the issue of wasting precious
metallocene complexes for both hafnocene and zirconocene
complexes, even for different activation methods, while
maintaining the desired polymer properties.
Structural Motifs Responsible for the Isomerization
While the complexes meso-Iand meso-II (M: Zr, Hf) were
readily isomerized at 120 °C after the addition of 0.50 eq.
TIBA, the complexes meso-III,meso-IV, and meso-V(M:
Zr, Hf) exhibited no isomerization at temperatures ranging
from 60 °C to 120 °C and TIBA concentrations ranging from
0.20 eq. to 10.0 eq. Moreover, no isomerization of the latter
complexes employing the sterically less hindered TMA
could be observed; in our previous work, we showed that
this reagent was also suitable to induce an isomerization of
meso-IHf.[8] The isomerization phenomenon thus seems to be
independent of the central metal atom—an observation that
can be attributed to the inherent chemical similarity of
hafnium and zirconium regarding their electronegativity, as
well as atomic and ionic radii.[16] The results presented,
however, suggest that the methoxy substituents at position 7
of the indenyl frameworks of the ligands are the crucial
factors for an aluminum alkyl induced isomerization to take
place. Independent of the steric demand of the aryl moieties,
only meso metallocene complexes bearing these structural
motifs underwent isomerization to their racemic analogs.
Monitoring the isomerization using 1H NMR spectroscopy,
we did not observe changes in the chemical shifts of these
methoxy groups—besides the differences between both
isomers and the intermediates discussed below. We interpret
these results as a hypothetical high-energy intermediate
having a shorter lifetime than the NMR timescale and, thus,
not being detectable via this analytical method.
Using DFT with the standard B3LYP functional (see
Supporting Information for details), the energies required
for ligands 2 and 3 to adapt the geometry they occupy in the
respective racemic and meso hafnocene complexes were
calculated (Figure 1).
40 kJ/mol are required to form each conformer of ligand
3bearing no methoxy groups, rendering no conformation,
rac-IIIHf or meso-IIIHf, energetically favored. On the other
hand, ligand 2 uses 40 kJ/mol to complexate Hf and yield
rac-IIHf, but 59 kJ/mol to adapt the geometry as in meso-IIHf.
The excess of energy is required to bend the aromatic
groups at the silicon atom by ~10°from the 105°in the free
ligand to the conformation with a central metal atom.
Consequently, the meso isomer is destabilized by 19 kJ/mol
relative to rac-IIHf, mostly as a result of the intramolecular
stress in the complexated ligand 2 (Table S1). Therefore, the
methoxy groups at position 7 of the indenyl framework are
ultimately responsible for the isomerization, with the driving
force being the relative thermodynamic stabilities of meso
and rac conformations of metallocene complexes bearing
these substituents. The reason for meso-Iand meso-II (M:
Zr, Hf) to be formed besides their racemic counterparts
during the metallocene syntheses at all thus seems to be of
kinetic origin. It is feasible that the di-lithiated species
obtained by deprotonation of the bis(indenyl) precursors
exhibit some lithium-oxygen chelation. Thus, a pre-orienta-
tion of both ligand sites before the transmetalation step
involving the group IV metal halide seems to occur, yielding
the thermodynamically less favored meso isomers by up to
80 %.
Isomerization Mechanism
While the energetic differences between the ligand con-
formations required for meso-IIHf and rac-IIHf to form
explain the driving force of the isomerization, they yet fall
short of elucidating the detailed reaction mechanism. Since
Figure 1. Energy diagrams of the transformation of ligands 2and 3into
the conformers required to form the respective racemic and meso
metallocene complexes.
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (4 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
no thorough isomerization at 120 °C was observed in the
absence of TIBA or TMA for any complex, an entirely
thermal isomerization can be ruled out, and it becomes clear
that the respective aluminum alkyl acts as a catalyst for the
isomerization in some way.
To gain further insight into the isomerization mecha-
nism, a kinetic 1H NMR study and LIFDI-MS experiments
of the conversion of meso-IHf with TMA were conducted.
Since the latter method is not yet quite widespread, a brief
overview of its functionality and its specific advantages in
elucidating our isomerization mechanism is given in the
Supporting Information. As visible via 1H NMR, the isomer-
ization proceeded via two different intermediates: A sym-
metric species formed from meso-IHf over the course of
90 min at 100 °C and an asymmetric one formed from this
symmetric intermediate within four hours at 100 °C. To a
certain extent, the latter species was then converted into
rac-IHf. This asymmetric species was stable under argon
atmosphere (>15 days), and removal of the overlaying
solution, as well as residual TMA, under vacuum, did not
lead to decomposition. In the 1H NMR spectrum, the
former, symmetric species exhibited a characteristic peak at
δ=1.28 ppm with a relative integral of 3, suggesting a Hf-
methyl moiety. We therefore concluded this species to be
the mono-methylated meso-IHfMe (Figure 2).
Besides the signal corresponding to rac-IHf, the 29Si NMR
spectrum of the final reaction products (Figure S45) ex-
hibited only one additional signal, proving that the men-
tioned asymmetric species was, in fact, one single species
and not two symmetric ones with equal relative integrals.
Moreover, the 1H NMR spectrum of this species exhibited a
signal at δ=0.40 ppm with a relative integral of 3,
suggesting another Hf-methyl moiety. We concluded this
species to be rac-IHfMe (Figure 2); LIFDI-MS confirmed this
assumption (Figure S50). Selected signals in the respective
1H NMR spectra of meso-IHfMe and rac-IHfMe are depicted
in Figure 2. The mentioned Hf-methyl groups, as well as the
cyclopentadienyl (Cp)-methyl groups are highlighted, re-
spectively.
The methyl groups at the Cp moieties of the metal
complexes proved suitable for differentiating each species’
relative amounts during the isomerization process. Excerpts
of the corresponding 1H NMR spectra of the kinetic NMR
experiment in this spectral region are summarized in Fig-
ure 3.
The relative amounts of meso-IHfMe,meso-IHf,rac-IHf,
and rac-IHfMe were determined through integration of the
signals highlighted in Figure 3. Plots of these amounts over
the course of 90 minutes and 22 hours, respectively, are
depicted in Figure 4.
We could also obtain rac-IHfMe through the direct
reaction of rac-IHf with TMA; in this case, however, meso-
IHfMe and meso-IHf were not detected as additional reaction
products. Similarly, analog methylated, asymmetric products
were also detectable in the respective 1H NMR spectra when
rac-IZr,rac-IIHf, and rac-IIZr were heated to 100 °C in the
presence of TMA (Supporting Information). LIFDI-MS
confirmed the molecular masses of the reaction products to
match the calculated values for the corresponding meth-
ylated species.
In addition to the kinetic 1H NMR experiment discussed
above, we used LIFDI-MS to in situ monitor the reaction of
meso-IHf with TMA. Excitingly, we observed a species with a
mass equivalent to the combined masses of meso-IHf and
TMA (Figure 5). Since field-desorption is a very soft
ionization method, this species is most likely no artefact
from the mass-spectrometric investigation, but an actual
Figure 2. Excerpts of the 1H NMR spectra of meso-IHfMe and rac-IHfMe.
The corresponding protons are highlighted in the depicted chemical
structures. Full spectra are provided in the Supporting Information.
Figure 3. Excerpts of the 1H NMR spectra of the conversion of meso-IHf
with 2.00 eq. TMA at 100 °C. Full spectra are provided in the Supporting
Information.
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (5 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
reaction product (see Supporting Information for a detailed
explanation).
We thought three different structures for the complex/
TMA interaction conceivable: The first and the one we
consider most likely is depicted in Figure 5 and involves the
interaction of TMA with the chloride substituents at the
central metal atom—either before or after a methyl
exchange reaction which will be discussed shortly. The other
two possible coordination sites of the electrophilic TMA are
the electron-rich Cp-rings, as well as the methoxy groups of
meso-IHf. According to DFT calculations on the complexes
IIHf and IIIHf (Table S2), all three interactions are feasible
with none being energetically significantly preferred over
another. However, the interaction with the chloride sub-
stituents is energetically at least slightly favored by up to
15 kJ/mol and 14 kJ/mol, respectively, compared to an
interaction with the Cp-rings and the methoxy groups.
Moreover, both other interactions do not significantly
change the geometry around the central metal atom which
would be a prerequisite for the respective species to be
involved in the isomerization mechanism, whereas coordina-
tion of TMA to a chloride substituent elongates the
respective HfCl bond. Therefore, the following discussion
will only focus on the TMA/chloride interaction.
The chloride/methyl exchange reaction of metallocene
dichlorides with TMA via a L2ClHf(μ-Cl)AlMe3(L =ligand)
species is reasonably well understood in the literature.[17] We
performed DFT calculations of this reaction for meso/rac-
IIHf and the analog complex lacking the methoxy groups
meso/rac-IIIHf (calculations were not conducted for IHf due
to the fact that IIHf contains significantly less atoms than IHf
and both exhibited the same behavior, experimentally). The
energetics of this reaction are summarized in Table S3 and
indicate the parent species L2ClHf(μ-Cl)AlMe3to be en-
ergetically slightly favored over the corresponding meth-
ylated one L2ClHf(μ-Me)AlMe2Cl in each case. Methylation
of meso-IIHf, however, is favored by 13 kJ/mol over meth-
ylation of its racemic analog. In comparison, complex IIIHf
behaves vice versa where methylation of the racemic isomer
is preferred by 11 kJ/mol over methylation of its meso
counterpart. Since only complex IIHf was experimentally
able to undergo isomerization and considering this energetic
discrepancy, we concluded the metal-chloride-TMA inter-
action to be a crucial step of the isomerization mechanism.
According to the data presented, we propose a mechanism
for the TMA-induced isomerization of methoxy-substituted
meso metallocene complexes as depicted in Scheme 3—
using meso-IHf as an example since the kinetic NMR data
were gathered from this complex.
Our proposed reaction sequence starts with the coordi-
nation of TMA to meso-IHf to form L2ClHf(μ-Cl)AlMe3and
the subsequent exchange of one methyl group to form
L2ClHf(μ-Me)AlMe2Cl. Although, as mentioned, DFT cal-
culations revealed the chloride-substituted species to be
energetically slightly favored over the methylated one, the
data gathered from the kinetic 1H NMR study clearly
indicate that the reaction proceeds via the methyl-bridged
species. The driving force of the reaction thus probably
Figure 4. Relative abundances of meso-IHfMe, meso-IHf, rac-IHf, and rac-
IHfMe during the conversion of meso-IHf with 2.00 eq. TMA at 100 °C.
For better visibility, plot a) is a zoomed depiction of the first 90 min of
plot b). Abundances were determined by integration of the Cp-methyl
groups.
Figure 5. Calculated isotopic distribution and detected masses of meso-
IHf +TMA.
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (6 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
originates from this species being withdrawn from the
L2ClHf(μ-Cl)AlMe3/L2ClHf(μ-Me)AlMe2Cl equilibrium.
The complex/TMA adduct was stable enough to be
detected in the corresponding LIFDI mass spectrum (Fig-
ure 5). This was an exciting observation considering the fact
these types of interactions are usually regarded as dynamic
and short-lived.[17] The detectability of this adduct via
LIFDI-MS indicates the μ-Me coordination to be quite firm.
We therefore hypothesized that the cleavage of the HfCp
bond necessary for an isomerization to take place might be
enabled by the coordination of a second substituent from
the Al atom to the Hf center to form a doubly bridged
LClHf(μ-Cl)(μ-Me)AlMe2high-energy intermediate as de-
picted in Scheme 3. The arising donation of additional
electron density to the central metal atom should partially
compensate the formally cationic charge upon detachment
of the Cp-ligand. From this intermediate, the ligand would
be able to freely rotate and to be subsequently reattached to
the central metal atom to form the thermodynamically
favored racemic isomer. A similar mechanism has been
proposed in the literature.[7a] DFT calculations proved the
possibility of such a high-energy intermediate to exist for
complex IIHf, its formation being endergonic by 160 kJ/mol.
Optimization trials of an analog doubly-bridged L2ClHf(μ-
Cl)(μ-Me)AlMe2species with both Cp-ligands still attached
to the central metal atom failed and yielded L2ClHf(μ-
Cl)AlMe3in all cases, substantiating the need for one Cp-
ligand to be detached. The geometry-optimized structure of
this high-energy intermediate is shown in Figure 6.
Methylation of the meso isomers of complexes III–V(M:
Zr, Hf) bearing no methoxy groups was also observed.
However, the reaction stopped at the methylated complexes
and no isomerization took place. In line with these
observations is the elevated endergonicity of 196 kJ/mol for
the analog high-energy intermediate of complex IIIHf to
form from its parent compound. The higher energy differ-
ence of 36 kJ/mol compared to IIHf highlights again the
necessity and role of the methoxy groups in the aluminum-
alkyl induced isomerization of the investigated group IV
metallocene complexes.
Conclusion
In summary, we improved and extended the scope of our
previously introduced isomerization of hafnocene complex
meso-IHf onto metallocene complexes with different ligand
frameworks and zirconocene/MAO systems. We were there-
fore able to utilize meso metallocene complexes—that are
generally considered waste—to produce isotactic polypropy-
lene with macromolecular characteristics equal to iPP
produced by the corresponding pure racemic isomers. This
allowed for an increase of iPP per amount of ligand used for
the metallocene syntheses by up to 400 %.
Scheme 3. Proposed TMA-induced isomerization mechanism of methoxy-substituted group IV metallocene complexes using meso-IHf as an
example.
Figure 6. Geometry-optimized structure of the high-energy intermediate
of IIHf upon reaction with TMA and detachment of one ligand.
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (7 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
By progressively varying substituents on the indenyl
moieties of the ligands, we were able to identify the methoxy
substituents at position 7 to be the crucial factor for an
isomerization to take place. DFT calculations revealed that
the energies required to adapt the respective ligand con-
formations in the presence of these substituents differ by as
much as 19 kJ/mol. Ligands bearing no methoxy groups
exhibited no energetic differences in both conformers, and
thus, it becomes evident that the driving force of the
isomerization lies in these energetic discrepancies. Based on
results gathered from 1H and 29Si NMR, as well as LIFDI-
MS experiments, we proposed a mechanism of the TMA-
induced isomerization. This mechanism involves an inter-
mediate stabilized by a μ-Me and μ-Cl interaction of TMA
with the central metal atom. From this intermediate, a
rotation of one indenyl ligand to form the corresponding
racemic isomer can take place. The proposed mechanism
should yield the methylated compounds meso-IHfMe and
rac-IHfMe as intermediary species and both were actually
detected experimentally. Furthermore, two other feasible
isomerization pathways were ruled out using DFT calcu-
lations, substantiating the proposed mechanism.
The facile and effective isomerization of 7-methoxy
substituted group IV metallocene complexes—in combina-
tion with their outstanding polymerization performances—
should steer the future design of this class of catalysts
toward such in the scope of an energy- and atom-efficient
polyolefin economy.
The authors have cited additional references within the
Supporting Information.[18–32]
Acknowledgements
The authors would like to thank Patricia Aufricht, Cara
Bommer, and Ruocheng Tang for help with the metallocene
syntheses. Furthermore, they would like to thank Dr. Lucas
Stieglitz and Magdalena Kleybolte for proofreading the
manuscript, as well as Dr. Sergei Vagin, Dr. Sven Krüger,
and Patrick Mollik for valuable discussions. The authors
also gratefully acknowledge the computational and data
resources provided by the Leibniz Supercomputer Center
(www.lrz.de). Open Access funding enabled and organized
by Projekt DEAL.
Conflict of Interest
The authors declare no conflict of interest.
Data Availability Statement
SC-XRD data is available from www.ccdc.cam.ac.uk/ prod-
ucts/csd with deposition numbers 2308439 (rac-VHf) and
2308440 (rac-IIZr).[33]
Keywords: density functional calculations ·isomerization ·
metallocenes ·olefin polymerization ·polypropylene
[1] a) G. Natta, Angew. Chem. 1956,68, 393–403; b) K. Ziegler, E.
Holzkamp, H. Breil, H. Martin, Angew. Chem. 1955,67, 541–
547.
[2] W. Kaminsky, K. Külper, H. H. Brintzinger, F. R. W. P. Wild,
Angew. Chem. Int. Ed. Engl. 1985,24, 507–508.
[3] a) P. S. Kulyabin, G. P. Goryunov, M. I. Sharikov, V. V. Izmer,
A. Vittoria, P. H. M. Budzelaar, V. Busico, A. Z. Voskoboyni-
kov, C. Ehm, R. Cipullo, D. V. Uborsky, J. Am. Chem. Soc.
2021,143, 7641–7647; b) N. M. R. Machat, D. Lanzinger, A.
Pöthig, B. Rieger, Organometallics 2017,36, 399–408; c) A.
Vittoria, G. Urciuoli, S. Costanzo, D. Tammaro, F. D.
Cannavacciuolo, R. Pasquino, R. Cipullo, F. Auriemma, N.
Grizzuti, P. L. Maffettone, V. Busico, Macromolecules 2022,
55, 5017–5026.
[4] A. Schöbel, E. Herdtweck, M. Parkinson, B. Rieger, Chem.
Eur. J. 2012,18, 4174–4178.
[5] P. S. Kulyabin, V. V. Izmer, G. P. Goryunov, M. I. Sharikov,
D. S. Kononovich, D. V. Uborsky, J. A. M. Canich, A. Z.
Voskoboynikov, Dalton Trans. 2021,50, 6170–6180.
[6] J. A. Ewen, J. Am. Chem. Soc. 1984,106, 6355–6364.
[7] a) R. M. Buck, N. Vinayavekhin, R. F. Jordan, J. Am. Chem.
Soc. 2007,129, 3468–3469; b) W. Kaminsky, A.-M. Schauwie-
nold, F. Freidanck, J. Mol. Catal. A 1996,112, 37–42.
[8] L. Stieglitz, T. M. Lenz, A. Saurwein, B. Rieger, Angew. Chem.
Int. Ed. 2022,61, e202210797.
[9] a) J. Karger-Kocsis, T. Bárány, Switzerland: Springer Nature
2019; b) V. V. Izmer, A. Y. Lebedev, D. S. Kononovich, I. S.
Borisov, P. S. Kulyabin, G. P. Goryunov, D. V. Uborsky,
J. A. M. Canich, A. Z. Voskoboynikov, Organometallics 2019,
38, 4645–4657; c) A. Vittoria, G. P. Goryunov, V. V. Izmer,
D. S. Kononovich, O. V. Samsonov, F. Zaccaria, G. Urciuoli,
P. H. Budzelaar, V. Busico, A. Z. Voskoboynikov, Polymer
2021,13, 2621.
[10] a) M. Nikulin, A. Tsarev, A. Lygin, A. Ryabov, I. Beletskaya,
A. Voskoboinikov, Russ. Chem. Bull. 2008,57, 2298–2306;
b) W. Spaleck, M. Antberg, J. Rohrmann, A. Winter, B.
Bachmann, P. Kiprof, J. Behm, W. A. Herrmann, Angew.
Chem. 1992,104, 1373–1376.
[11] M. Muhr, P. Heiß, M. Schütz, R. Bühler, C. Gemel, M. H.
Linden, H. B. Linden, R. A. Fischer, Dalton Trans. 2021,50,
9031–9036.
[12] K. P. Bryliakov, E. P. Talsi, A. Z. Voskoboynikov, S. J. Lancas-
ter, M. Bochmann, Organometallics 2008,27, 6333–6342.
[13] V. Busico, R. Cipullo, R. Pellecchia, G. Talarico, A. Razavi,
Macromolecules 2009, 1789–1791.
[14] W. Kaminsky, H. Sinn, Polyolefins: 50 years after Ziegler and
Natta II: Polyolefins by Metallocenes and Other Single-Site
Catalysts 2013, 1–28.
[15] R. Tanaka, T. Kawahara, Y. Shinto, Y. Nakayama, T. Shiono,
Macromolecules 2017,50, 5989–5993.
[16] a) R. D. Shannon, Acta Crystallogr. Sect. A 1976,32, 751–767;
b) R. T. Shannon, C. T. Prewitt, Acta Crystallogr. Sect. B 1969,
25, 925–946.
[17] a) S. Beck, H. H. Brintzinger, Inorg. Chim. Acta 1998,270,
376–381; b) U. Wieser, D. Babushkin, H.-H. Brintzinger,
Organometallics 2002,21, 920–923.
[18] H. H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger, R. M.
Waymouth, Angew. Chem. Int. Ed. 1995,34, 1143–1170.
[19] B. Coto, J. M. Escola, I. Suárez, M. J. Caballero, Polym. Test.
2007,26, 568–575.
[20] W. A. Herrmann, J. Rohrmann, E. Herdtweck, W. Spaleck, A.
Winter, Angew. Chem. Int. Ed. 1989,28, 1511–1512.
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (8 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
[21] W. Spaleck, F. Kueber, A. Winter, J. Rohrmann, B. Bachmann,
M. Antberg, V. Dolle, E. F. Paulus, Organometallics 1994,13,
954–963.
[22] a) D. Andrae, U. Haeussermann, M. Dolg, H. Stoll, H. Preuss,
Theor. Chim. Acta 1990,77, 123–141; b) F. Furche, R. Ahlrichs,
C. Hättig, W. Klopper, M. Sierka, F. Weigend, Wiley
Interdiscip. Rev.: Comput. Mol. Sci. 2014,4, 91–100.
[23] J. Tomasi, M. Persico, Chem. Rev. 1994,94, 2027–2094.
[24] J. Ho, A. Klamt, M. L. Coote, J. Phys. Chem. A 2010,114,
13442–13444.
[25] S. Miertuš, E. Scrocco, J. Tomasi, Chem. Phys. 1981,55, 117–
129.
[26] G. M. Sheldrick, Acta Crystallogr. Sect. A 2015,71, 3–8.
[27] G. M. Sheldrick, Acta Crystallogr. Sect. C 2015,71, 3–8.
[28] C. B. Hübschle, G. M. Sheldrick, B. Dittrich, J. Appl. Crystal-
logr. 2011,44, 1281–1284.
[29] International Tables for Crystallography, Vol. C (Ed.: A. J.
Wilson), Kluwer Academic Publishers, Dordrecht, The Nether-
lands 1992, Tables 6.1.1.4 (pp. 500–502), 4.2.6.8 (pp. 219–222),
and 4.2.4.2 (pp. 193–199).
[30] C. F. Macrae, I. J. Bruno, J. A. Chisholm, P. R. Edgington, P.
McCabe, E. Pidcock, L. Rodriguez-Monge, R. Taylor, J.
van de Streek, P. A. Wood, J. Appl. Crystallogr. 2008,41, 466–
470.
[31] D. Kratzert, I. Krossing, J. Appl. Crystallogr. 2018,51, 928.
[32] A. L. Spek, Acta Crystallogr. Sect. D 2009,65, 148–155.
[33] a) P. C. Möhring, N. J. Coville, Coord. Chem. Rev. 2006,250,
18–35; b) R. M. Shaltout, J. Y. Corey, N. P. Rath, J. Organo-
met. Chem. 1995,503, 205–212; c) Deposition Numbers
2308439 (for rac-VHf) and 2308440 (for rac-IIZr) contain the
supplementary crystallographic data for this paper. These data
are provided free of charge by the joint Cambridge Crystallo-
graphic Data Centre and Fachinformationszentrum Karlsruhe
Access Structures service.
Manuscript received: April 10, 2024
Accepted manuscript online: July 7, 2024
Version of record online: August 20, 2024
Angewandte
Chemie
Research Article
Angew. Chem. Int. Ed. 2024,63, e202406848 (9 of 9) © 2024 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH
