This article is licensed under CC-BY 4.

[Link]/IC Article

Octahedral Zirconium Salan Catalysts for Olefin Polymerization:

Substituent and Solvent Effects on Structure and Dynamics
Anna Dall’Anese, Pavel S. Kulyabin, Dmitry V. Uborsky,* Antonio Vittoria, Christian Ehm,*
Roberta Cipullo, Peter H. M. Budzelaar, Alexander Z. Voskoboynikov, Vincenzo Busico, Leonardo Tensi,
Alceo Macchioni, and Cristiano Zuccaccia*
Cite This: Inorg. Chem. 2023, 62, 16021−16037 Read Online
See [Link] for options on how to legitimately share published articles.

ACCESS Metrics & More Article Recommendations *

sı Supporting Information

ABSTRACT: Group 4 metal-Salan olefin polymerization catalysts typically have

relatively low activity, being slowed down by a pre-equilibrium favoring a non-
Downloaded via [Link] on February 2, 2026 at 16:18:02 (UTC).

polymerization active resting state identified as a mer-mer isomer (MM); formation

of the polymerization active fac-fac species (FF) requires isomerization. We now
show that the chemistry is more subtle than previously realized. Salan variations
bearing large, flat substituents can achieve very high activity, and we ascribe this to
the stabilization of the FF isomer, which becomes lower in energy than MM.
Detailed in situ NMR studies of a fast (o-anthracenyl) and a slow (o-tBu) Salan
precursors, suitably activated, indicate that preferred isomers in solution are
different: the fast catalyst prefers FF while the slow catalyst prefers a highly
distorted MM geometry. Crystal structures of the activated o-anthracenyl
substituted complex with a moderately (chlorobenzene) and, more importantly,
a weakly coordinating solvent (toluene) in the first coordination sphere emphasize that the active FF isomer is preferred, at least for
the benzyl species. Site epimerization (SE) barriers for the fast catalyst (ΔS > 0, dissociative) and the slow catalyst (ΔS < 0,
associative) in toluene corroborate the solvent role. Diagnostic NMe 13C chemical shift differences allow unambiguous detection of
FF or MM geometries for seven activated catalysts in different solvents, highlighting the role of solvent coordination strength and
bulkiness of the ortho-substituent on the isomer equilibrium. For the first time, active polymeryl species of Zr-Salan catalysts were
speciated. The slow catalyst is effectively trapped in the inactive MM state, as previously suggested. Direct observation of fast
catalysts is hampered by their high reactivity, but the product of the first 1-hexene insertion maintains its FF geometry.

■ INTRODUCTION
In homogeneous-phase olefin polymerization catalysis, the vast
the latter on the single olefin insertion reaction in the metal−
carbon bond.16
Most group 4 post-metallocene catalysts are inherently more
majority of catalytic systems are highly electrophilic cationic
flexible and the corresponding activated species often speciate
complexes of group 4 metals.1−4 Since catalytic reactions are
into various possible geometrical isomers, some of which may be
conducted in low-polarity solvents, ion pairs are considered the
inactive.17 On the other hand, many post-metallocenes feature
true active species, and the strength of the cation-anion
an easier precatalyst synthesis and superior performance at high
interaction is known to strongly affect catalytic performance.5,6
temperatures.18 Rationalizing how the equilibria between
Typical metallocene and ansa-metallocene systems have the
isomers or conformers19 modulate the overall catalytic perform-
great advantage that the active species retain the pseudo-
ance is difficult. In some cases, the ligand framework may
tetrahedral (rigid) structure of the precursors.7,8 Although
additionally even undergo in situ modification by reaction with
synthetically challenging, the ligand(s) can be rationally
monomer(s)20−23 or other additives (e.g., main group metal
engineered, based on conceptually simple molecular descriptors,
alkyls).24
to optimize catalytic performance.9 Moreover, it is also possible
Bis(phenoxy-amine) complexes (“Salan-type”) of group 4
to analyze in detail how the interactions between the cation,
metals represent a typical example in this respect (Figure 1).
anion, and solvent modulate the structure, dynamics, and
reactivity of the active species.10 For example, using both
systems of industrial interest and models derived from them, it Received: June 28, 2023
has been possible, through nuclear magnetic resonance (NMR) Published: September 19, 2023
studies, to investigate the specific effect of the solvent on the
epimerization processes of the active site,11−14 as well as the
relationship between the thermodynamic tendency of ion pairs
to form higher-order ion aggregates15 and the kinetic effect of
© 2023 The Authors. Published by
American Chemical Society [Link]
16021 Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

leading to multiple energetically accessible isomers (fac-fac

(FF), fac-mer (FM), and mer-mer (MM), after the fac or mer
geometrical arrangement of the two [ONN] fragments, see
Figure 1).30,31
The C2-symmetric FF isomer is generally favored in neutral
precatalysts LMR2 (R = Bn,25,28,32 Me2CHO,27,33 M = group 4
metal), but solution NMR spectroscopy and density functional
theory (DFT) concluded that the MM isomer is the most stable
isomer for pentacoordinated active cations, adopting a distorted
square pyramidal geometry, independently of the nature of R1
and R2.31 Even with a “sticky counteranion” like [MeB-
(C6F5)3]−, an outer sphere ion pair (OSIP) is formed.30 As
the coordination vacancy and growing polymeryl chain are in a
Figure 1. Illustrations of the fac and mer isomerism for octahedral trans arrangement, the MM isomer is polymerization inactive.
[ONNO] Salan-type catalysts (top) and the possible geometries of For insertion to occur, the complex must rearrange to the active
[ONN] fragments giving rise to the fac and mer nomenclature FF isomer, likely passing through an FM isomer (Figure 1). If
(bottom). the FF isomer is higher in energy than the MM isomer, then the
MM ⇆ FM pre-equilibrium is of course an integral part of the
First introduced by Kol and co-workers,25 these octahedral effective propagation barrier, and DFT studies have shown that
systems have been of particular interest as molecular catalysts the energy difference between the MM and FF isomers
resembling, in some respect, the octahedral catalytic site of correlates well with activity as long as the anion is included
heterogeneous Ziegler-Natta (ZN) Ti catalysts.26 computationally.29,34 Non-specific solvent effects were included
Facile Salan synthesis by a one-pot Mannich condensation via the conductor-like screening model (COSMO).
between amines, formaldehyde, and substituted phenols allows Although usually considered as a bulk media affecting ion
easy structural amplification, especially in comparison to the dissociation by means of their polarity, weakly coordinating
often-complex metallocene synthesis. Most studies on Salan solvents like toluene can play a more specific role by
complexes focused on tuning polymerization performance by coordinating and stabilizing the cationic species, especially for
ligand design; a wide range of polypropylene properties can be “open” catalysts such as half-metallocenes35,36 and constrained-
accessed through variations of R1 and R2 substituents (Figure geometry catalysts (CGC).37 The notion that this is also
1).27,28 The tuning space with respect to activity is remarkable: occurring for relatively crowded cationic olefin polymerization
introduced originally as isospecific, slow and living polymer- complexes is fairly new. For instance, we have recently
ization catalysts (R1 = tBu), polymerization activity increases by investigated how the specific catalyst-solvent interaction
five orders of magnitude with R1 = anthracenyl.29 Studies on the regulates ion pair symmetrization (IPS) barriers in metallocene
structural analysis of these catalysts in solution are rare and complexes, i.e., how the mutual position exchange between the
greatly complicated by an inherently flexible ligand framework. coordinative vacancy and the alkyl group in the first
The [ONNO] ligand can wrap around the metal in several ways, coordination sphere of the metal is influenced by solvent

Scheme 1. Complexes Investigated in This Work and Simplified Labeling Scheme for NMR Assignment.

16022 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

coordination.14 Extending the approach to post-metallocenes, Scheme 2. Three-Step Synthesis of the 15N-Enriched
we now report an in-depth analysis of seven Zr-Salan complexes Complexes *3 and *4H
(Scheme 1), differing in the ortho-substituent (R1) on the
phenoxy-amine ring. These complexes mirror the broad tuning
range in propene polymerization in this catalyst class, ranging
from slow and living (1, 4) to very highly active (2, 3, 7), from
highly isotactic selective (1, 7) to poorly isotactic selective (2) to
chain-end controlled syndiotactic selective (3), and from (very)
low (1−6) to high molar mass capability (7).
1−7 were activated with trityl borate ( ([CPh3]+[B(C6F5)4]−,
TTB, Scheme 1) in both C7D8 and C6D5Cl and investigated by
in situ NMR spectroscopy, applying 1D and 2D NMR
techniques, in order to determine the solution structure of the
complexes. For the first time, solid-state structures of solvent-
coordinated activated complexes were obtained, further
supporting NMR findings in solution. NMR dynamics and
DFT studies elucidate the solvent role. Importantly, diagnostic
13
C-NMR chemical shift differences for the NMe atoms in
different isomers uncovered for LMBn+ complexes (L = ligand)
allow, for the first time, NMR identification of the predominant
resting state isomer for LM-polymeryl+ species.

■ RESULTS AND DISCUSSION

Synthesis and Characterization of Neutral Complexes
1−7. Complexes 1, 2, 3, 5, and 7 were synthesized according to
established literature protocols.38,39 The synthesis of complexes
4 and 6 is detailed in the Supporting Information.
The most available and convenient starting material for 15N-
labeled Salan ligands is ethylenediamine-15N2 dihydrochloride. Table 1. Propene Polymerization Results with Precatalysts
The standard approach suggests the synthesis of a Salan ligand 1−7 and TiBA/TTB as Activator/Scavenger at Tp = 60 °C
from DMEDA, paraformaldehyde, and a phenol in one step, and ppropene = 6.6 Bara
albeit with a low or moderate yield (e.g., 15% for the ligand for
338). Since the three-step synthesis of DMEDA-15N240 itself precatalyst Rpb σd Mne Mwe PDI
would lead to inevitable losses of the expensive labeled 1 0.04 0.9975 3.1 5.4 1.7
precursor, an alternative method for ligand synthesis was 2 21.3 0.80 1.3 1.8 1.5
c
developed. Thus, the 15N-enriched *3 could be synthesized in 3 16.7 1.5 2.3 1.5
45% yield over three steps (Scheme 2a). Yields for the tBu- 4 0.01 0.93 0.7 0.8 1.2
substituted 4 could not be sufficiently improved to allow 5 0.15 0.986 5.1 8.6 1.7
economical synthesis of the 15N-enriched complex. However, 6 0.07 0.974 3.0 4.9 1.6
the closely related para-H derivative *4H was synthesized in 7 19.8 ≥0.9998 69 143 2.1
a
43% yield over three steps (Scheme 2b). Other experimental conditions: 5 mL toluene, 5 μmol TiBA, [TTB]/
Propene Polymerization Using Complexes 1−7. [Zr] = 0.9. bIn kg mmolZr−1 h−1 cChain-end controlled syndiotactic
Complexes 1−7 were tested in propene polymerizations under polypropylene dEnantioselectivity of propene insertion, calculated
rigorously identical conditions, using tri-iso-butyl aluminum according to the enantiomorphic site model.47 eIn kDa. All datapoints
(TiBA) as scavenger and TTB as activator, at Tp = 60 °C and are averages of duplicate experiments, for further details see
Supporting Information.
ppropene = 6.6 bar, in a high-throughput experimentation setup
(Freeslate parallel pressure reactor setup with 48 reaction cells
(PPR48), fully contained in a glove-box).41−45 the structure and dynamics of the cationic species. Benzyl group
Polymerization procedures are detailed in the Experimental abstraction cleanly forms the mono-benzyl cationic species
Section. Complexes 4 and 6 were explicitly tested for the (Scheme 1); the solution geometry results from the interplay of
purpose of this study; the results for complexes 1−3, 5, and 7 are ligand and solvent coordination strength.36,48 The cationic
taken in part from ref 46. The main polymerization results are species 3+ (“fast”) and 4+ (“living”) show distinctly different
summarized in Table 1. polymerization behaviors (Table 1) and were therefore natural
The broad performance tuning range for Salan catalysts is starting points to investigate their solution structure. Their 15N-
evident in the results presented in Table 1. Productivities span enriched derivatives, *3 and *4H, were synthesized to facilitate
three orders of magnitude; polypropylene stereoregularity speciation of the active species.
covers the whole range, from chain-end controlled syndiotactic Activation of 3. 1H and 13C NMR spectra of 3+ in C6D5Cl
to near perfect site-controlled isotactic. Molar mass capability show extensive peak broadening at 298 K; however, at 233 K,
ranges from Mn values of 0.7 kDa (4) to 69 kDa (7). distinct 1H and 13C NMR resonances for each half of the
Structures of Mono-Benzyl Cationic Species from 3 tetradentate ligand can be resolved, revealing the loss of C2
and 4. Complexes 3 and 4 were activated with one equivalent of symmetry (Figure S17). The resonances of each half can be
TTB in C7D8 and C6D5Cl and analyzed with in situ NMR grouped following the 1H,1H and 1H,13C scalar correlations.
spectroscopy to investigate substituent and solvent influences on Two singlets at δH = 1.93 and 1.37 ppm, both integrating to three
16023 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

protons each, show scalar correlations with carbons at δC = 47.7 Scheme 3. Limiting Coordination Modes of the [ONN′O′]
and 45.7 ppm, respectively; these are assigned to the NMe Ligand in Alkyl Monocationic Group 4 Metals.
moieties. We arbitrarily indicate as Me2 the methyl group having
a signal at 1.93 (47.7) ppm and as Me2′ the one with resonances
at 1.37 (45.7) ppm (see Scheme 1 for the NMR labeling of
protons and carbons). Me2 (δC2 = 47.7 ppm) shows 1H,13C long-
range coupling with two carbons at δC = 65.0 ppm and 54.7 ppm,
which in turn shows a single bond correlation with two doublets
at 4.02 and 2.99 ppm, and at δH = 2.97 and 1.79 ppm,
respectively (Figures S18 and S19). The carbon resonance at
54.7 ppm does not display any additional long-range coupling
with aromatic protons and is therefore assigned to carbon C1 of
the ethylene bridge. Consequently, the signals at δH = 2.97 and
1.79 ppm belong to H1a and H1b. H1a appears as a doublet of
doublets with 2JHH ≈ 3JHH ≈ 13.6 Hz and, based on Karplus
relationships, is assigned to the proton in axial position. A long-
range coupling between the carbon at δC = 65.0 ppm and the
aromatic proton resonance at δH = 6.80 ppm allows assignment
of the former to C3 and the latter to H4. Thus, the signals at δH =
4.02 and 2.99 ppm are due to H3a and H3b. Starting from H4, the In the case of 3+ in C6D5Cl, both Me2 and Me2′ show dipolar
remaining resonances of the phenolate and anthracenyl moieties interaction of very similar intensity with their corresponding
are assigned by crossing the data from 1H,1H COSY, 1H,13C H3a, H3b and H3a′, H3b′ protons, thus indicating that each ligand
HSQC, and 1H,13C HMBC NMR experiments (Figures S18− half adopts a fac configuration. The single species observed in
S20). Extensive overlaps of aromatic resonances of the solution has FF geometry (Figure 2). This is corroborated by the
anthracenyl substituent prevent full assignment of this moiety. lack of NOE contacts between Me2/Me2′ with H3 protons on
Similarly, starting from the Me2′ resonance (δH = 1.37 ppm, δC = the other side of the ligand. Selective Ha/Ho and Hb/Ho′ NOE
45.7 ppm), the resonances of the other half of the ligand could contacts observed within the benzylic fragment agree with the
be assigned (H1a′ = 2.57 ppm; H1b′ = 1.44 ppm; C1′ = 51.7 ppm; proposed η2 coordination mode. The relative orientation of the
H3a′ = 2.04 ppm; H3b′ = 2.25 ppm; C3′ = 62.0 ppm; H4′ = 6.54 benzyl ligand can be inferred from the lack of dipolar
ppm, and so on; see Supporting Information). interactions between aromatic protons of the benzyl moiety
The benzyl ligand appears coordinated in η2 fashion as and aliphatic resonances of the ligand, indicating that the
indicated by the observation of seven different resonances: two aromatic ring points toward the formally vacant coordination
are expected for the CH2 benzyl moiety (δH = 2.16 ppm (Ha) site.
and 1.01 ppm (Hb); δCbenzyl = 71.2 ppm) based on symmetry Whether a C6D5Cl solvent molecule occupies the “vacant” site
considerations, but the observation of five distinct aromatic cannot be directly inferred from NMR data but is confirmed by
resonances (δH = 6.39 ppm, Hp, 5.89 ppm, Ho′; 5.73 ppm, Hm; X-ray analysis on a single crystal of 3+FF (vide infra). NMR
5.61 ppm. Hm′; 4.01 ppm, Ho) is diagnostic for a restricted structural analysis of 3+FF generated in C7D8 is more complex
because 1H NMR resonances of the benzyl moiety are not
rotation around the CH2−Cipso bond. The shift of Cipso from
detected, likely as a result of σ-bond metathesis (SBM) with the
154.3 ppm, in the neutral 3 precursor, to 128.8 ppm in 3+, and
deuterated solvent leading to a fully deuterated benzyl ligand.49
the relatively small 2JHH (∼7 Hz) between Ha and Hb, lend
Generating 3+FF in C7H8, isolating it by precipitation with n-
further support.
pentane, and redissolving it in C7D8 at 233 K, allows to observe
The coordination mode of the tetradentate [ONN′O′] ligand
aromatic benzyl resonances at chemical shifts similar to those in
around Zr, i.e., the geometry adopted in solution by complex 3+, C6D5Cl (Figure S21). Indicative for very low SBM barriers,
can be inferred by analyzing relevant dipolar interactions in the those resonances disappear quickly even at low temperatures,
1
H ROESY NMR spectrum with the help of DFT optimized hampering full NMR elucidation. Proton and carbon NMR
structures. For cationic group 4 metals, in general, the resonances of the ONN′O′ ligand of 3+FF do not differ
[ONN′O′] ligand can coordinate with three limiting modalities significantly from those observed in C6D5Cl. Both Me2 and
(Scheme 3); FM and FM′ isomers differ in the position of the Me2′ show dipolar interactions of very similar intensity with
benzyl ligand, trans to N′ (in FM) and trans to O (in FM′). their corresponding H3a, H3b and H3a′, H3b′ protons, implying
Selective Ha/H3a and Hb/Me2 dipolar interactions observed that the C7D8 complex also adopts the 3+FF geometry (see
in the 1H ROESY NMR spectrum (Figure 2) indicate that both Supporting Information). This is further supported by single
benzylic protons (Ha and Hb) are spatially close to protons of crystal X-ray analysis.
one half of the ligand, clearly inconsistent with both MM and 15
N NMR data of the labeled complex *3+FF in C6D5Cl and
FM′ geometries. C7D8 and related Lewis base stabilized cationic species, *3+FF
Distinguishing structure FF from the FM geometry is less -THF and *3+FF-Lut (Lut = 3,5-lutidine, see Supporting
straightforward. DFT structures reveal a general feature: when Information) in C6D5Cl, were recorded to further elucidate
the three NN′O′ (or N′NO) arms of the ligand are arranged in a coordination effects. (Table 2).50 The chemical shift of the
fac configuration, the methyl group on N (or N′) is equally nitrogen atom cis to the benzyl group is sensitive to the nature of
distant from its adjacent H3 protons, whereas if the three NN′O′ L (L = Lewis base or solvent) decreasing with increasing
(or N′NO) arms are arranged in the mer configuration, one of coordination strength of L: δN = 49.47, 51.35, 54.09, and 56.77
the H3 protons becomes antiperiplanar to the Me moiety and ppm for L = Lut, THF, C6D5Cl, and C7D8, respectively. The
thus much more distant (Scheme 4). chemical shift of N′, the nitrogen atom trans to the benzyl group,
16024 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

Figure 2. Three sections of the 1H ROESY NMR spectrum of 3+FF (C6D5Cl, 233 K) showing (A) the NOE contacts, of similar intensity, of Me2 with
both H3a and H3b; (B) the NOE contacts, of similar intensity, of Me2′ with both H3a′ and H3b′; (C) the NOE contacts between Hb and Me2 and
between Ha and H3a. Relevant dipolar interactions are summarized in D by blue arches.

Scheme 4. Simplified Newman Projection Showing the Table 2. 15N NMR Chemical Shifts of N and N′ (δN and δN′,
Spatial Relationship of NMe with respect to its adjacent CH2 Externally Referenced to Ammonia in ppm) for Selected
Protons in the fac or mer Arrangements of the Ligand. Cationic Complexes Derived from *3 and *4H and
Experimental (Black) and Computed (Red for “FF” and
Green for “MM” or “FM” Geometries) Chemical Shift
Difference between N and N′ (ΔδN‑N′)

is only minimally affected by the nature of L: δN′ = 49.75, 51.17,

51.09 and 51.86 ppm for L = Lut, THF, C6D5Cl, and C7D8,
respectively. These observations are fully consistent with the
hypothesis that a donor molecule remains coordinated to the
metal in complexes 3+FF -L (L = C7D8, C6D5Cl, THF and Lut).
Activation of 4. The solution structure of 4+, and that of the
related 15N labeled complex *4H+, were investigated using the
same methodological approach; aside from obvious differences,
both complexes show nearly superimposable NMR spectra. For
temperatures higher than 253 K, 4H+ displays several NMR
features that are different from those of 3+FF. Firstly, well-
resolved spectra are obtained already at 268 K, both in C6D5Cl
and C7D8, suggesting a lower degree of dynamicity. Secondly,
both Me2 and Me2′ show dipolar interactions (Figure 3) with a
At 273 K where a single species of *4H+ is dominating in solution
only one adjacent H3 proton, suggesting each half of the ligand (vide infra). bAt 218 K for the second species of *4H+ present in
preferentially adopts a mer configuration (Scheme 4). This is solution (vide infra).
further supported by the observation of a relatively strong
dipolar interaction between Me2′ on one side of the ligand and
H3b, on the other side (Figure 3). Finally, only three aromatic rotation around the Cipso−CH2 bond consistent with a η1
resonances are observed for the benzyl ligand, indicating a fast coordination modality.
16025 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

data trends suggest substantial structural modifications on

passing from *4H+ to *4H+-THF and *4H+-Lut (Table 2 and
Supporting Information). 15N chemical shifts for *4H+FF-THF
(δN = 52.79 ppm; δN′ = 52.73 ppm) and *4H+FF-Lut (δN = 51.37
ppm; δN′ = 52.33 ppm) are found at lower frequencies, with
values very similar to those observed for complexes *3+FF-THF
and *3+FF-Lut, indicating stronger Lewis bases trigger a switch
of geometry.
Solid-State Structure of *3+FF. Crystals of *3+FF suitable for
X-ray analysis were obtained by layering n-pentane on top of
concentrated solutions of *3+FF in both C7D8 and C6D5Cl.
Although single crystal X-ray structures for neutral Salan
complexes are known,25,33,52 to the best of our knowledge, this
is the first example of a solid-state structural analysis for a
cationic Salan complex.
The unit cell of *3+FF·C6D5Cl contains two enantiomeric
cations with coordination environments of the Zr centers
exhibiting minimal differences. The octahedral coordination
geometry is slightly distorted with the tetradentate [ONN′O′]
ligand coordinated in the fac-fac configuration (Figure 4). The
O−Zr−O′ angles amount to 168.9°; the O−Zr−N′ (93.4−
93.9°), and the O′−Zr−N (100.8−100.9°) angles are found
close to 90°. The benzyl ligand and a chlorobenzene molecule in
Figure 3. Section of the 1H ROESY NMR spectrum of *4H+MM (C7D8, cis relative position complete the coordination environment.
263 K) showing (a) the selective interactions between Me2′ and H3a′; Similarly to other Zr(IV) complexes,48 the solvent molecule is
(b) the selective interactions between Me2 and H3a; (c) the strong coordinated to the metal via the chlorine atom (Zr−Cl 2.779 Å).
interaction between Me2′ and H3b. Relevant dipolar interactions are Metrical parameters indicate η2 hapticity53 of the benzyl
summarized in the structure depiction by blue arches. ligand with the Ph ring pointing away from the ligand
framework, toward the coordinated chlorobenzene molecule.
N and N′ of *4H+ resonate at significantly higher frequencies Indeed, for both cations, the difference between the Zr−CH2
with respect to *3+FF (C7D8: δN = 61.41, δN′ = 55.16 ppm; distances (2.266(3) Å; 2.272(4) Å) and the Zr−Cipso distances
C6D5Cl: δN = 61.67, δN′ = 56.01 ppm), both in C7D8 or C6D5Cl. (2.522(3) Å; 2.540(4) Å) are lower than 0.5 Å, whereas the Zr−
The chemical shifts are scarcely affected by the nature of the CH2−Cipso angles (81.90 deg.; 82.77 deg) are much lower than
solvent, in agreement with the lack of solvent coordination. The 97 deg.53 Consistently, the Zr−O and Zr−O′ distances (trans to
higher values of δN can be attributed to the switch of N from each other) are identical within the experimental error
trans to solvent (as in *3+FF) to trans to oxygen. The values of (1.973(2) and 1.971(2) Å). The Zr−N (trans to the solvent)
δN′, instead, suggest that this atom could be in an intermediate distances, 2.408(3) and 2.404(3) are noticeably shorter than the
situation, somewhere between trans to benzyl (as in *3+FF) and Zr−N′ (trans to the benzyl) distances, 2.483(3) and 2.485(3)
trans to oxygen; both MM and FM limiting geometries (Scheme Å).
3) could contribute to the actual structure of cationic complex In the case of *3+FF·C7D8, the unit cell contains four cation-
*4H+MM in solution and, on average, the actual geometry may be anion pairs. Again, Zr centers in *3+FFC7D8 display a slightly
described as a distorted trigonal bipyramid having the N and O′ distorted octahedral geometry with the tetradentate [ONN′O′]
arms in the apical positions.31,51 ligand framework coordinated in fac-fac configuration. The
Alternatively, the actual geometry might be so distorted that coordination environment is completed by an η2 coordinated
limiting “ideal geometries” are a bad fit. Importantly, 15N NMR benzyl ligand and a toluene molecule in cis relative position. Zr−

Figure 4. X-ray structures of: (a) *3+FF·C7D8 (meta-C interaction), (b) *3+FF·C7D8 (para-C interaction), (c) *3+FF·C6D5Cl. Ellipsoids drawn at the
50% probability level; hydrogen atoms have been omitted for clarity.

16026 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

Figure 5. Key bond angles, geometry index τ, and relative Gibbs free energies at T = 298 K for selected FM and MM isomers of 4+.

Figure 6. Sections of the 1H,13C HSQC NMR spectra of 3 (C6D6, 298 K), *4H (C6D6, 298 K), 3+FF (C6D6Cl, 233 K) *4H+MM (C6D5Cl, 263 K) and
*4+FF-THF (C6D5Cl, 292 K) showing the relationship between the coordination geometry adopted by the [ONN′O′] ligand and the 13C NMR
chemical shifts of C2 and C2′.

Cipso distances are slightly longer with respect to *3+FF·C6D5Cl We interrogated the equilibria between the different isomers
(2.593(4), 2.583(4), 2.574(4), and 2.582(4) Å). The four of 3+ and 4+ computationally employing Grimme’s conformer
cations differ in the orientation of the toluene molecule, not in rotamer ensemble sampling tool (CREST)54 followed by final
absolute configuration: in three of them, toluene is interacting DFT optimization at the TPSSh-Dzero(PCM)/TZ//TPSSh/DZ
with one of the meta carbons (average Zr−Cmeta distance of 2.89 level of theory.55−65 Experimental evidence unequivocally
Å, Figure 4a), while in the fourth cation the para carbon is closer points to *3+FF(solvent) as the lowest conformer in toluene
to Zr (Zr−Cpara distance of 2.87 Å, Figure 4b). In line with and chlorobenzene. Including the solvent in the coordination
trends of 15N NMR chemical shifts, the Zr−N′ (trans to the sphere, we identified one FF conformer for *3+FF-C6D5Cl and
benzyl) distances are similar to those of *3+FF·C6D5Cl, whereas two nearly isoenergetic FF conformers for *3+FF-C7D8, differing
the Zr−N (trans to the solvent) are, on average, slightly longer in the toluene orientation (meta- or para-C binding), mirroring
(2.405(3), 2.416(3), 2.427(3), 2.422(3) Å). the X-ray findings. In the toluene case, binding is weak as an FF
Overall, the X-ray data of *3+FF confirms the NMR findings: isomer with toluene in the outer coordination sphere was found
to be only 1.3 kcal/mol higher in Gibbs free energy than the
the tetradentate [ONN′O′] ligand framework is coordinated in
isomer with toluene in the first coordination sphere. Predicted
the fac-fac configuration both in solution and the solid state; 15
N NMR chemical shifts differences match experimentally
coordination of a solvent molecule stabilizes this geometry.
observed ones nicely (see Table 2) and agree with the solvent-
Unfortunately, all attempts to obtain single crystals of 4+ suitable
coordinated complex. Gas phase sampling of the conformational
for X-ray analysis have been unsuccessful so far. space in the absence of solvent reveals four nearly isoenergetic
DFT Insights into the Salan Isomer Equilibria. The findings isomers: two FF, one FM and one MM isomers.
discussed above clearly indicate that the structure adopted in The situation is less clear-cut for 4+, both computationally and
solution by the active cationic species is not always the same. It experimentally. Gas phase sampling of the conformational space
was hypothesized that Salan complexes are affected by a MM/ indicates six non-solvent-coordinated isomers within a Gibbs
FF pre-equilibrium. This is the case for 4 in weakly coordinating free energy window of 5.2 kcal/mol, three FM, two MM, and
solvents, a slow catalyst also studied previously.29 However, the one FF isomer. The two low-energy isomers are MM (0.0 kcal/
results for 3 (and 4 in strongly coordinating solvents) indicate mol) and FM (0.9 kcal/mol); predicted 15N-NMR chemical
that this equilibrium can be shifted to FF and the actual solution shifts are more comparable with the FM isomer. All isomers are
structure is strictly related to the interplay between the significantly distorted from ideal geometries with Addison τ
substituents on the [ONN′O′] ligand and the nature of the parameters,66 ranging from 0.17 to 0.70, indicating significant
solvent. distortion from ideal pyramidal (MM, τ = 0) and trigonal
16027 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

bipyramidal (FM, τ = 1) geometries (Figure 5). It should be (and continuity with previous literature), we continue to refer to
noted that the less distorted MM isomer (τ = 0.17) is 5.0 kcal/ it as MM.
13
mol higher in energy than the more distorted one (τ = 0.28). To C NMR chemical shifts of C2 and C2′ for neutral and
identify general trends, we compared predicted 15N-NMR shift cationic complexes in different solvents are collected in Table
differences between the N and N′ atoms for solvent-coordinated S1; a summary is reported in Table 3. Similarly to 3+FF, complex
and non-coordinated isomers of 3+, 4+ and 7+. As observed
experimentally, predicted chemical shift differences (ΔδΝ‑Ν′). Table 3. Geometries of Activated Complexes Derived from
decrease with donor strength of the solvent in solvent- 1−7
coordinated FF complexes. However, similarly, ΔδΝ‑Ν′ becomes
complex geometry (toluene) geometry (C6D5Cl) Rpa
smaller in the MM isomers, the less distorted the isomer is +
1 FM/MM FF 0.04
according to the τ parameter. FM isomers also show highly
2+ FF FF 21.3
flexible shift differences.
3+ FF FF 16.7
Likewise, absolute shifts are also highly flexible. It appears that
4+ FM/MM FF < 248 K 0.01
the complex interplay between benzyl group orientation, agostic
5+ FM/MM FF < 298 K 0.15
interactions and relative orientations of backbone Ph, and linker
6+ FM/MM FF < 268 K 0.07
geometry precludes identification of general trends that would
7+ FM/MM FF 19.8
allow positive determination of the prevalent isomer solely based a
on 15N chemical shifts. In kg mmolZr−1 h−1.
NMe Carbon Chemical Shifts as Diagnostic Indicators for
[ONN′O′]MR+ Complex Geometry. Extension of the NMR 2+FF, bearing an N-carbazolyl substituent directly bonded to the
methodological approach described for 3+FF and 4+MM to the phenolate rings, adopts the FF geometry irrespective of the type
cationic species derived from the other precursors 1−2, 5−7 of solvent. In contrast, complexes 1+MM and 5+MM-7+MM, having
proved to be difficult; 1H NMR spectra often exhibit broad and an aliphatic substituent on phenolate rings, show δC2′ lower than
superimposed resonances preventing unambiguous NOE 40 ppm in C7D8, similarly to 4+MM.
contact detection. When passing from C7D8 to the more coordinating C6D5Cl,
13
C NMR chemical shifts of the NMe moieties appear to the behavior of cationic species derived from precursors 1, 4, 5−
follow a rather regular trend. C2 and C2′ resonate at δC values 7 becomes more intricate, depending on the nature of the
well above 40 ppm (Figure 6) for all neutral and cationic substituent and temperature. For instance, below 248 K in
complexes with FF geometry. In contrast, for both complex C6D5Cl, a new set of resonances appears in the 1H NMR
4+MM and *4H+MM, C2′ appears at frequencies lower than 40 spectrum of *4H+MM (the same holds true for 4+MM). Despite
ppm (δC2′ in the range 38.4−37.6 both in C7D8 and C6D5Cl for full NMR characterization being hampered by extensive
T > 253 K, see Table S1 and Figure 6). For 8+MM, a complex broadening and spectral overlaps, the new species features two
analogous to 4+MM but having an OMe substituent in para diagnostic singlets at δH = 1.98 and 1.80 ppm, which are scalarly
position of the phenolate rings and a Zr−Me instead of a Zr−Bn coupled with carbon atoms resonating at 47.1 ppm (Figure 7).
moiety, previously shown to adopt the MM geometry based on These resonances are assigned to Me2 and Me2′ moieties and
19 1
F, H HOESY NMR data and DFT calculations, δC2 and δC2′ allow us to conclude that the new species is *4H+FF, which forms
are observed at 42.7 ppm and at 37.6 ppm, respectively, allowing
to exclude anisotropic effects originating from the benzyl ligand.
Thus, 13C NMR chemical shifts of C2 and C2′, easily measurable
NMR parameters, may indeed discriminate FF from MM/FM
geometries.
DFT prediction of 13C-NMR NMe chemical shifts confirms
their diagnostic value. For 3+, 4+, and 7+, 24 non-solvent-
coordinated and 14 solvent-coordinated isomers were identified
overall. All identified FF isomers (with or without solvent) show
predicted 13C-NMR shifts >43 ppm, while all MM and FM
isomers show one <43 ppm, predominantly <40 ppm. In FF
isomers, both methyl groups on the nitrogens are oriented
toward the back of the catalyst, therefore pointing away from the
empty site. Rearrangement from fac to mer of one of the
fragments orients its methyl toward the empty site, accompanied
by a lowering of the chemical shift. In the MM isomer, only one
of the methyls is oriented toward the empty site; the other one
remains anti to the empty site, and the chemical shift is
unaffected. Therefore, the observation of a chemical shift <40
ppm is diagnostic for the complex not being in fac-fac geometry,
although exact determination of its geometry (MM or FM or
FM′, Scheme 3) is not possible solely based on 13C-NMR NMe
chemical shifts. Given the insight from experiments and
computations discussed above, it appears highly likely that the
prevalent geometry for the non fac-fac isomer is very distorted Figure 7. Section of the 1H,13C HSQC NMR spectrum of an
from the idealized situation depicted in Scheme 3. For simplicity equilibrium mixture of *4H+FF and *4H+MM (C6D5Cl, 243 K).

16028 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

from *4H+MM by solvent coordination. At 218 K, the chemical In C7D8, the activation enthalpy (ΔH‡) and entropy (ΔS‡) of
shift difference of ΔδN‑N′ = 2.63 ppm (Table 2) corroborates this the SE process for 3+FF are found to be 15.2 kcal mol−1 and 10 cal
assignment as it is (a) much lower than for *4H+MM and (b) mol−1 K−1, respectively (Figure 8, left). The small, but positive,
similar to 3+FF in the same solvent (ΔδN‑N′ = 3 ppm). We will ΔS‡ value confirms that 3+FF has solvent molecules coordinated
discuss the equilibrium details between *4H+MM and *4H+FF to the formally “vacant” coordination site and that SE occurs
later. through a dissociative interchange mechanism, where de-
On the other side, in C6D5Cl, complex 1+FF displays δC2 and coordination of coordinated toluene appears to be the rate-
δC2′ higher than 40 in the whole investigated temperature range limiting step, followed by rapid benzyl migration from one side
(223−298 K): possibly, the increase of the substituent steric to the other and re-coordination of toluene.
bulk is sufficient to destabilize the MM geometry to the point For all investigated temperatures, the SE process of 3+FF in
that binding of C6D5Cl and switching to FF geometry become C6D5Cl is much slower than in C7D8; for example, kSE = 0.10 s−1
thermodynamically more favorable. Similar to 4+, the Me2 and in C6D5Cl vs 59.9 s−1 in C7D8 at 253 K. Eyring analysis showed
Me2′ 13C NMR chemical shifts are temperature dependent for that the enthalpic contribution (ΔH‡ = 19.6 kcal mol−1) to the
cationic complexes 5+ and 6+ in C6D5Cl (See Supporting overall barrier is about 4.5 kcal mol−1 larger than in C7D8,
Information), indicating the prevalence of the solvent-stabilized whereas the ΔS‡ value (14.8 cal mol−1 K−1) is somewhat more
5+FF/6+FF isomers and the five-coordinated 5+MM/6+MM isomers, positive than in C7D8 (Figure 8, left). This implies stronger
at low and high temperatures, respectively. C6D5Cl binding to the metal center, but based on the similar
Dynamics of Cationic Complexes in Solution. In entropy contribution, a dissociative interchange mechanism is
solvent-stabilized ansa-metallocene cationic complexes, inver- also operative here. The results mirror previous findings for
sion of configuration at the metal center, i.e., site epimerization ansa-metallocenes.
(SE) or IPS, occurs via a mechanism where solvent For temperatures higher than 253 K, isomers 4+MM and
decoordination is rate-limiting.67−69 Cationic FF Salan com- *4H+MM are prevalent in solution and the kinetics of the SE
plexes show some structural similarity with ansa-metallocene process are remarkably different from that of 3+FF (Figure 8,
cations and solvent choice should affect these barriers as well. right). Firstly, kSE are much less sensitive to the nature of solvent:
Then again, how these processes occur for systems adopting for instance, at 253 K, kSE is equal to 0.08, 0.16, and 1.03 s−1 for
MM geometry and how they are related to the nature of the 4+MM in C7D8, *4H+MM in C7D8, and *4H+MM in C6D5Cl,
solvent is less clear. respectively. Secondly, the more coordinating C6D5Cl accel-
Rate constants of the SE process (kSE, in s−1) can be measured erates the SE process with respect to C7 D 8 . Finally,
by means of dynamic NMR methods, such as 2D magnetization deconvolution of enthalpic and entropic contributions to the
transfer experiments (EXSY) and line shape analysis, by kinetic barrier reveals (i) similar and relatively small ΔH‡ values
monitoring chemical exchange between magnetically inequiva- (13.6 kcal mol−1 for 4H+MM in C7D8, 12.2 kcal mol−1 for
lent nuclei belonging to the two halves of the coordinated *4H+MM in C7D8 and 12.2 kcal mol−1 for *4H+MM in C6D5Cl)
[ONN′O′] ligand.70,71 We studied complexes derived from 3 and (ii) similar and negative ΔS‡ values (−9.3 cal mol−1 K−1 for
and 4 (or *4H) in C7D8 and C6D5Cl and kSE values as a function 4H+MM in C7D8, −13.5 cal mol−1K−1for *4H+MM in C7D8 and
of temperature and solvent are summarized in Tables S2 and S3 −9.8 cal mol−1K−1for *4H+MM in C6D5Cl). In contrast to 3+FF,
(Supporting Information); the activation parameters, obtained the negative values of ΔS‡ imply a different SE mechanism,
from Eyring analysis (see Figure 8) are reported in Table 4. associative in nature, occurs (see below). Importantly, despite
MM complexes having no solvent coordinated, solvent evidently
plays a role in the SE mechanism; however, the modality by
which the solvent triggers SE for MM systems is less obvious.
Crucial information in this respect was obtained by further
analyzing the equilibrium between *4H+MM and *4H+FF and the
corresponding dynamic behavior in C6D5Cl at temperatures
below 248 K. The presence of separated NMR resonances for
Me2 and Me2′ moieties of both *4H+MM and *4H+FF, in the
218−248 K temperature range, not only allowed to separately
determine kSE values for *4H+MM and *4H+FF, but also allowed:
Figure 8. Eyring plots of ln(kSE/T) vs 1/T for complexes 3+FF (left) and (i) to experimentally determine the thermodynamic constant
*4H+MM (right). Data in C7D8 and C6D5Cl are shown in black and red, (Keq) for the equilibrium depicted in eq 1, by estimating the
respectively. concentration of the two species using an external NMR
reference, and (ii) to measure the rate constants at equilibrium
Table 4. Activation Parameters for Site Epimerization (SE) in for forward (k1) and backward (k−1) reactions by following the
Cationic Complexes Derived from 3, 4, and *4H selective intermolecular magnetization transfer between corre-
sponding Me2 (or Me2′) moieties of *4H+MM and *4H+FF
species solvent ΔH‡ ΔS‡ ΔG‡(298) ΔG‡(333) (Figure 9). Experimental data are collected in Table 5.
+
3 FF C7D8 15.2 10.0 12.2 11.9
3+FF C6D5Cl 19.6 14.8 15.2 14.7
4+MM C7D8 13.6 −9.3 16.4 16.7
*4H+MM C7D8 12.2 −13.5 16.2 16.7
*4H+MM C6D5Cl 12.2 −9.8 15.1 15.5
*4H+FF C6D5Cl 16.6 15.2 12.1 11.6

16029 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

magnetization transfer (k1) by the concentration of C6D5Cl, was

found to be little dependent on temperature (Figure 10,
middle), with ΔH‡, ΔS‡, and ΔG‡(233) values of 9.0 kcal mol−1,
−25.1 cal mol−1 K−1, and 14.9 kcal mol−1, respectively.
On the contrary, the rate constant for the unimolecular
backward reaction, kFF→MM, is equal to the microscopic rate
constant (k−1) and clearly increases with increasing temperature
(Figure 10, middle). The corresponding activation parameters,
ΔH‡ = 14.9 kcal mol−1, ΔS‡ = 6.5 cal mol−1 K−1, reflect the
dissociative nature of the process, resulting in ΔG‡(233) = 13.4
kcal mol−1.
Finally, kSE values relative to *4H+FF were much larger than
those measured for *4H+MM, the major species present in
solution at temperatures higher than 253 K (Figure 10, middle).
Comparison of activation parameters (ΔH‡ = 16.6 kcal mol−1
and ΔS‡ = 15.2 cal mol−1 K−1 for *4H+FF; ΔH‡ = 12.2 kcal
mol−1; ΔS‡ = −9.8 cal mol−1 K−1 for *4H+MM) confirms that the
change in geometry is accompanied by an apparent change of
the SE mechanism, involving specific interactions with the
solvent molecule.
Combining kinetic and thermodynamic experimental data in
the temperature range 218−248 K allows to delineate an
experimental semi-quantitative free energy profile (at 233 K) as
depicted in Figure 11.
Under these conditions, the highest barrier is related to the
conversion of *4H+MM into the solvent-stabilized *4H+FF
isomer. Since coordination of the solvent trans to the benzyl
moiety in the *4H+MM isomer is unlikely to produce any
intermediate relevant for the subsequent SE, it is reasonable to
assume that solvent coordination occurs on the fac-mer isomer
Figure 9. Section of the 1H EXSY NMR spectrum of a mixture of (*4H+FM). Whether there is a pre-equilibrium between *4H+MM
*4H+MM and *4H+FF (C6D5Cl, 243 K) showing the pattern of and *4H+FM, as softly supported by 15N chemical shifts and DFT
magnetization transfer relative to the interconversion between *4H+MM computations (see above), or the *4H+FM isomer is interacting
and *4H+FF as well as chemical exchange due to site epimerization with the solvent only in the transition state, is difficult to prove
within the same geometrical isomer. experimentally. Note that the *4H+FF is thermodynamically less
stable than the starting *4H+MM isomer for entropic reasons.
Concerning the equilibrium, ΔH0 and ΔS0 values were Once formed, the barrier of the SE process for the *4H+FF
derived from the van’t Hoff plot of ln Keq vs 1/T and found to be isomer is indeed slightly lower than the barrier to convert it back
−7.1 kcal mol−1 and −36 cal mol−1 K−1, respectively, resulting in to *4H+MM by solvent dissociation. In other words, these data
ΔG0(233) = 1.4 kcal mol−1 (Figure 10, left). Interaction of the qualitatively confirm that the solvent triggers the apparent SE
solvent with *4H+MM is exothermic, adding evidence that process of *4H+ MM by allowing the formation of the
*4H+MM must be formulated as a pentacoordinate species and as intermediate *4H+FF isomer.
expected, the reaction to form *4H+FF is entropically disfavored. Quantification of this effect at higher temperature is strongly
As far as rate constants at equilibrium are concerned, the complicated by the interplay between the position of the
macroscopic, second order rate constant of the forward reaction, thermodynamic equilibrium in eq 1 and coalescence of NMR
kMM→FF, computed by dividing the microscopic rate constant of resonances of *4H+MM and *4H+FF isomers. However, the

Table 5. Equilibrium Constant (Keq) Relative to Eq 1, Corresponding Rate Constants at Equilibrium (kMM→FF and kFF→MM), Rate
Constant of Site Epimerization Relative to *4H+FF (kSE (*4H+FF)) and Corresponding Thermodynamic Parameters
T Keq (M−1) kMM→FF (M−1 s−1) kFF→MM (s−1) kSE (*4H+FF) (s−1)
218 0.175 0.25
223 0.115 0.48
228 0.036 0.69 1.16
233 0.050 0.055 1.29 2.44
238 0.046 0.075 2.48 4.99
243 0.032 0.104 4.12 12.13
248 0.216 11.98 29.80
ΔH0 or ΔH‡ −7.1 9.0 14.9 16.6
ΔS0 or ΔS‡ −36 −25.1 6.5 15.2
ΔG0a or ΔG‡a 1.4 14.9 13.4 13.1

a
At 233 K.

16030 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

Figure 10. Left: Van’t Hoff plot of ln Keq (*4H+MM + S → *4H+FF) vs 1/T in C6D5Cl. Middle: Eyring plots, in C6D5Cl, of ln(k/T) vs 1/T for: kMM→FF
(empty black circles); kFF→MM (filled red circles); kSE for *4H+MM (empty red squares); kSE for *4H+FF (filled black squares). Right: Eyring plots in
C6D5Cl of ln(kSE/T) vs 1/T for 3+FF (empty black squares) and *4H+FF (filled black squares).

Figure 11. Experimentally-determined free energy profile vs reaction coordinates for the various processes involving *4H+. All ΔG0 and ΔG‡ values are
computed at 233 K. Given the highly distorted nature of FM and MM isomers and their minimal energy differences, we assume a fast, barrierless or near
barrierless FM/MM equilibrium.

experimental negative ΔS‡ values for the SE process between

*4H+MM diastereomers, as obtained from kSE in the 243−308 K
temperature range, still represent an indication that this process
must be triggered by an associative step, i.e., some kind of
interaction with the solvent molecules is needed.
Dissecting the specific role of the phenolate substituents on
the rate of the SE process is only possible in C6D5Cl at low
temperature, where the behavior of 3+FF and *4H+FF can be
compared. The corresponding Eyring plots of kSE are shown in Figure 12. Topologic steric maps of the dichloride precursors show
Figure 10, right. Clearly, kSE for 3+FF, bearing the aromatic significantly more steric bulk close to the metal for catalyst 4 than for 3,
anthracenyl substituent, is much slower than that of *4H+FF, indicating steric interference between the tBu groups and solvent affects
having aliphatic substituents. The activation parameters suggest solvent coordination strength.
that the observed difference is mainly due to the enthalpic
contribution to the activation barrier (ΔH‡ = 16.6 vs 19.6 kcal in the presence of a 1-alkene monomer that are resting states in
mol−1 for *4H+FF and 3+FF, respectively). Although electronic olefin polymerization catalysis.72 Thus, reactions of *3+FF and
factors cannot be excluded, sterics seems to play a major role. *4H+MM with 1-hexene were monitored at low temperatures in
With respect to the “flat” and aromatic anthracenyl substituent C6D5Cl by NMR spectroscopy.
in 3, which can adopt less sterically demanding orientations, the Addition of ∼30 equiv of 1-hexene to *3+FF at 233 K resulted
t
Bu groups in 4 shield the metal more effectively and thereby in the immediate consumption of the starting material with the
hinder the approach of solvent (Figure 12). The decreased concomitant formation of a main species, displaying 13C NMR
solvent-metal interaction favors the (partial) release of the chemical shifts of both NMe moieties above 40 ppm (45.6 and
solvent molecule, which is necessary to trigger SE. 47.0 ppm, respectively; Figure 13a,b). NMR characterization of
Preliminary Studies of Salan Complexes under True such species points to the formation of *3FF-CH2-CH(C4H9)-
Polymerization Conditions. The identification of an “easy” CH2Ph+, the product of the first insertion of 1-hexene into the
NMR indicator to determine the structure of Salan catalysts Zr−benzyl bond, that maintains the polymerization active fac-fac
offered the unique opportunity to investigate the geometry geometry observed for the starting cationic species *3+FF (see
adopted by the active species, i.e., polymeryl bearing complexes Supporting Information). Noteworthy, no NMe carbon
16031 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

or fac-mer) isomers of the cationic species. Understanding the

factors that regulate those equilibria, and then which isomer
dominates catalyst speciation, is experimentally difficult,
especially in the presence of monomer, i.e., under catalytic
conditions. Detailed in situ NMR studies of cationic benzyl
species representative of fast (3+, o-anthracenyl) and slow (4+,
o-tBu) Salan catalysts allowed to dissect the intricate interplay
between the nature of substituents and solvent on their structure
and dynamic behavior in solution.
Cation 3+, bearing the aromatic substituent, adopts the six-
coordinate fac-fac geometry both in C7D8 (low polar and weakly
coordinating) and C6D5Cl (slightly more polar and moderately
coordinating). A solvent molecule is coordinated cis to the
benzyl ligand as further demonstrated by single crystal X-ray
analysis. Consistently, the SE process mirrors those observed for
typical (and more rigid) metallocene cations, i.e., it has a
dissociative nature and the barrier is higher in C6D5Cl due to the
higher enthalpic cost for loosening cation-solvent interactions.
For temperatures above 253 K, cation 4+, bearing the aliphatic
substituent, adopts the five-coordinated mer-mer (or fac-mer)
geometry with no solvent coordination. Consistently, SE
minimally depends on the nature of solvent; noteworthy, the
barrier is characterized by a negative entropic contribution,
suggesting an associative nature of the process. Thermodynamic
and kinetic studies of the 4+MM + S ⇋ 4H+FF equilibrium in
C6D5Cl (218−243 K) allowed to experimentally demonstrate
that coordination of solvent to 4+MM is rate determining along
the pathways of SE of 4+MM, offering a rationale for its associative
nature.
Detailed analysis of NMR data in combination with DFT
computations allowed to individuate diagnostic, and easily
accessible, NMR parameters (13C of NMe moieties) capable of
discriminating between fac-fac and mer-mer (or fac-mer) isomers.
On one hand, this allowed to experimentally determine the
prevalent species present in solution for benzyl cations bearing
Figure 13. NMe regions in the multiplicity-edited 1H, 13C HSQC NMR different substituents; the results correlate nicely with catalytic
experiments (black cross peaks = CH or CH3 moieties, red cross peaks productivities. Salans bearing large but flat substituents (such as
= CH2 moieties) of: (a) *3+FF (233 K); (b) *3FF-CH2-CH(C4H9)-
CH2Ph+ (233 K); (c) *4H+MM (253 K); (d) *4H-Pn+MM.
3) lead to fast catalysis, not limited by the presence of MM
species. More spherical substituents cause a thermodynamic
preference for MM geometries and the resulting low
resonances below 40 ppm could be detected. Minor species concentration of FF active species leads to slow catalysis (as
having the NMe carbons resonating above 40 ppm can instead for 4). Salan 7, with its sp3 substituent atom but large and flat
be observed. Although their characterization is hampered by “body”, is probably a borderline case.
their very low concentration in solution, we speculate that they More importantly, these detailed NMR studies allowed, for
can arise from zirconium polymeryl species, *3FF-Pn+, retaining the first time, to determine the geometry of active polymeryl
the fac-fac geometry. species in the presence of a large excess of monomer. Similar to
The reaction of 1-hexene with *4H+MM at 253 K (at this the starting benzyl cation, “slow” 4-polymeryl+ is observed in its
temperature, the equilibrium between *4H+MM and *4H+FF is inactive mer-mer resting state, as previously suggested by DFT.
strongly shifted to the left) proceeds much slower than in the On the contrary, the “fast” 3+ cation is initially converted to 3FF-
case of *3+FF. After several minutes *4H+MM is almost CH2-CH(C4H9)CH2Ph+, the product of the first 1-hexene
completely consumed and the new species dominating in insertion, observed in its FF geometry. No signature for species
solution features NMe carbon resonances at 37.7 and 43.8 ppm having the mer-mer (or fac-mer) ligand arrangement could be
(Figure 13c,d), leading us to formulate it as the Zr-oligomeryl detected in solution, suggesting that also the propagating
species, *4HMM-Pn+. This insertion product, like the starting species, 3FF-Pn+, prefer a fac-fac geometry.
benzyl derivative, prefers the polymerization inactive mer-mer
geometry (see Supporting Information for additional NMR data
supporting this assignment). ■ EXPERIMENTAL SECTION
General Considerations. All manipulations and synthesis of air-
■ CONCLUSIONS
The productivities of group 4-Salan olefin polymerization
and moisture sensitive chemicals were performed under rigorous
exclusion of oxygen and moisture in flame-dried Schlenk-type glassware
interfaced to a high vacuum line (10−5 Torr), or in a nitrogen-filled
catalysts are strongly influenced by the nature of the ligand MBraun glove-box (<0.5 ppm O2 and H2O). Hydrocarbon solvents
substituents, which appear to control the conformational pre- used for synthesis were dried over 4 Å molecular sieves and degassed by
equilibrium between the active (fac-fac) and inactive (mer-mer bubbling with dry argon. Ethereal solvents used for synthesis were

16032 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

distilled from sodium/benzophenone. Molecular sieves (4 Å, MS) were All polymers were characterized by means of high-temperature gel
activated for 24 h at ca. 200−230 °C under dynamic vacuum. All permeation chromatography (GPC) and 13C NMR spectroscopy. GPC
solvents used in the studies of the activated complexes were freeze- curves were recorded with a Freeslate Rapid GPC setup, equipped with
pump-thaw degassed on the high vacuum line, dried over the a set of two mixed-bed Agilent PLgel 10 μm columns and a Polymer
appropriate drying agent (Na/K alloy for benzene, pentane, toluene, Char IR4 detector. Calibration was performed with the universal
benzene-d6 and toluene-d8; CaH2 for chlorobenzene-d5, 1,2-difluor- method, using 10 monodisperse polystyrene samples (Mn between 1.3
obenzene) and vacuum transferred into dry storage Schlenk flasks and 3700 kDa).
equipped with PTFE valves. [CPh3][B(C6F5)4] was obtained from Quantitative 13C NMR spectra were recorded using a Bruker Avance
Boulder Scientific Company and used as received. TiBA (tri-isobutyl III 400 spectrometer equipped with a high-temperature cryoprobe for 5
aluminum) was purchased from Sigma-Aldrich and used as received. mm OD tubes, on 45 mg mL−1 polymer solutions in tetrachloroethane-
High-resolution mass spectra (HRMS) were recorded on an Agilent d2 (with BHT added as a stabilizer, [BHT] = 0.4 mg mL−1). Acquisition
Technologies 6530 Q-TOF LC/MS system paired with Agilent 1260 conditions were: 45° pulse; acquisition time, 2.7 s; relaxation delay, 2.0
HPLC and using Agilent JetStream or APCI ion source. Ethene (Linde, s; 2 K transients. Broad-band proton decoupling was achieved with a
99.95%) and propene (Rivoira, 99.6%) were purified by flowing them modified WALTZ16 sequence (BI_WALTZ16_32 by Bruker).
through a column containing activated 4 Å molecular sieves and an DFT Calculations. Geometries of dichloride precursor complexes
activated Cu catalyst (BASF R0-11G). 1-Hexene (Sigma-Aldrich, 99%) and activated complexes with and without solvent were fully optimized
was purified by passing it through a mixed-bed column of the activated using the Gaussian 16 software package.55 The BOpt software package
Cu catalyst and 4 Å molecular sieves. Toluene (Romil) was dried using was employed for data collection.58 Following the protocol proposed in
an MBraun SPS-5 solvent purification unit. 1,2-Dichlorobenzene ref 59, the TPSSh60/cc-pVDZ(-PP)61,62,77 level of theory, using a small
(Romil, >99.8% isomeric purity) was used as received. core pseudo potential on Zr,64,65 was employed for optimization. This
NMR Spectroscopy Experiments. All samples for NMR protocol has been successfully used, in combination with TPSSh-Dzero/
measurements were prepared inside the glove-box; flame-dried NMR cc-pVTZ(-PP) single-point energy corrections, to address polymer-
tubes equipped with a PTFE valve (J-Young NMR tubes) were used. ization related problems in post-metallocene chemistry.78,79 The
One- and two-dimensional, homo- and hetero-nuclear NMR spectra density fitting approximation (Resolution of Identity, RI)80−83 and
standard Gaussian16 quality settings [Scf = Tight and Int(Grid =
were recorded on a Bruker Avance III 400 spectrometer equipped with
UltraFine)] were used at the optimization stage and for single-point
a smartprobe and using standard pulse sequences. Unless otherwise
energy (SP) calculations. All structures represent true minima (as
stated, referencing is relative to external TMS (1H and 13C), NH3 (15N)
indicated by the absence of imaginary frequencies). Grimme’s
and CCl3F (19F). Variable temperature 1H EXSY NMR measurements
dispersion correction84 with zero damping and the polarizable
were acquired by using the PFG version of the NOESY sequence
continuum model (PCM)85 for solvent corrections were employed at
(noesygptp), setting a relaxation delay of 1 s, with mixing time values the SP stage. Final energies were then combined with thermal
(τm) ranging between 2.7 and 800 ms depending on the rate of chemical corrections (enthalpy and entropy, 298 K, 1 bar) to obtain final free
exchange. Typically, a matrix of 1024 × 1024 data points was used for energies; entropy corrections were scaled by 0.67 to account for
acquisition and the raw data were processed using zero-filling to 2048 reduced freedom of movement in solution.86−88 Buried volume
data points in both spectral dimensions. The spectral window and the descriptors were calculated using the SambVca 2.1 program.89 The
number of transients were optimized depending on distribution, CREST program54 was used to sample the conformational space and
relevant resonances, and sample concentration. Typically, at least two generate guess structures for full DFT optimization.
experiments with different τm values were acquired for each Further information about the synthesis of the ligands and the
temperature, and the rate constant values were obtained from the complexes, NMR spectra, polymerizations, and DFT calculations is
average of all the values. For the IPS dynamical motions, rate constants reported in the Supporting Information section.
(kIPS, s−1) were evaluated by the method proposed by Perrin and
Dwyer,70 and were calculated from the integration of the 2D spectra by
using the EXSYCALC software.73 In some cases, exchange rate
constants were also estimated at higher temperatures, where
■ ASSOCIATED CONTENT
* Supporting Information
sı
exchanging resonances first coalesce and then narrow, using methods
of lineshape analysis and standard equations for two-sites exchange in
The Supporting Information is available free of charge at
the absence of scalar coupling.71 Activation parameters of dynamical [Link]
motions were estimated from the corresponding Eyring plots; errors on General experimental procedures; X-ray data for *3+FF·
activation enthalpy and activation entropy were determined from the
C7D8 and *3+FF·C6D5Cl, computational details and
quality of linear fitting and computed at 95% confidence interval. For
the 15N NMR experiments, standard sequences provided by Bruker results; NMR characterizations; experimental SE rate
were employed, in particular, the zgig sequence for 1D 15N NMR constants as a function of temperature (PDF)
spectra, and the hsqcgpph sequence for 2D 15N NMR, with an
optimization of the 90° pulse at 21 μs at 74 PLW1, with a long-range
Combined coordinate file for optimized DFT structures
constant of 7 Hz. To describe the multiplicity of the signals, the (ZIP)
following abbreviations are used: s, singlet; bs, broad singlet; d, doublet;
bd, broad doublet; dd, doublet of doublets; t, triplet; and m, multiplet. Accession Codes
Synthetic Procedures. Complexes 1 and 3,38 2 and 7,39 5,74 4 and CCDC 2272131−2272132 contain the supplementary crystallo-
6, 3-(tert-butyl)-2-hydroxybenzaldehyde,75 and 3-(anthracen-9-yl)-2-
46
graphic data for this paper. These data can be obtained free of
(methoxymethoxy)-5-methylbenzaldehyde76 were prepared according charge via [Link]/data_request/cif, or by email-
to the published procedures. Ethylenediamine-15N2 dihydrochloride ing data_request@[Link], or by contacting The Cam-
(98% atom 15 N, 99% (CP), CAS: 84050-98-6) was purchased from bridge Crystallographic Data Centre, 12 Union Road, Cam-
Aldrich. All other reagents were purchased from commercial sources
bridge CB2 1EZ, UK; fax: +44 1223 336033.
and used as received.
Polymerization Experiments and Polymer Characterization.
CCDC 2272131 and 2,272,132 contain the supplementary
Polymerization experiments were performed at 60 °C and propene crystallographic data for this paper. These data can be obtained
partial pressure of 6.6 bar in a high-throughput experimentation setup free of charge via [Link]/data_request/cif, or by
(PPR48, fully contained in a glove-box), according to protocols emailing data_request@[Link], or by contacting The
reported in refs 40−45. In all cases, TiBA was used as a scavenger and Cambridge Crystallographic Data Centre, 12 Union Road,
[CPh3][B(C6F5)4] (TTB) as an activator. Cambridge CB2 1EZ, UK; fax: +441,223 336,033.
16033 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

■ AUTHOR INFORMATION
Corresponding Authors
C.Z. also acknowledge the Università degli Studi di Perugia and
the MUR for their support within the project Vitality.
Dmitry V. Uborsky − Department of Chemistry, Lomonosov
Moscow State University, 119991 Moscow, Russia; DPI, 5600
AX Eindhoven, The Netherlands; [Link]/0000-0002-
■ REFERENCES
(1) Kaminsky, W. Metallocene Catalysts. In Kirk-Othmer Encyclopedia
0093-6867; Email: duborsky@[Link] of Chemical Technology; Wiley, 2015.
Christian Ehm − Dipartimento di Scienze Chimiche, Università (2) Gibson, V. C.; Spitzmesser, S. K. Advances in Non-Metallocene
di Napoli Federico II, 80126 Napoli, Italy; DPI, 5600 AX Olefin Polymerization Catalysis. Chem. Rev. 2003, 103, 283−316.
Eindhoven, The Netherlands; [Link]/0000-0002-2538- (3) Kirillov, E.; Carpentier, J. F. {Cyclopentadienyl/Fluorenyl}-
5141; Email: [Link]@[Link] Group 4 Ansa-Metallocene Catalysts for Production of Tailor-Made
Polyolefins. Chem. Rec. 2021, 21, 357−375.
Cristiano Zuccaccia − Dipartimento di Chimica, Biologia e
(4) Brintzinger, H. H.; Fischer, D.; Mülhaupt, R.; Rieger, B.;
Biotecnologie, Università degli Studi di Perugia, 06123 Waymouth, R. M. Stereospecific Olefin Polymerization with Chiral
Perugia, Italy; DPI, 5600 AX Eindhoven, The Netherlands; Metallocene Catalysts. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143−
[Link]/0000-0002-9835-2818; 1170.
Email: [Link]@[Link] (5) Zaccaria, F.; Sian, L.; Zuccaccia, C.; Macchioni, A. Ion Pairing in
Transition Metal Catalyzed Olefin Polymerization, 1st ed.; Elsevier Inc.,
Authors 2020; Vol. 73, DOI: 10.1016/[Link].2019.08.001.
Anna Dall’Anese − Dipartimento di Chimica, Biologia e (6) Macchioni, A. Ion Pairing in Transition-Metal Organometallic
Biotecnologie, Università degli Studi di Perugia, 06123 Chemistry. Chem. Rev. 2005, 105, 2039−2074.
Perugia, Italy; DPI, 5600 AX Eindhoven, The Netherlands (7) Wang, B. Ansa-Metallocene Polymerization Catalysts: Effects of
Pavel S. Kulyabin − Department of Chemistry, Lomonosov the Bridges on the Catalytic Activities. Coord. Chem. Rev. 2006, 250,
Moscow State University, 119991 Moscow, Russia; DPI, 5600 242−258.
AX Eindhoven, The Netherlands; [Link]/0000-0003- (8) Cobzaru, C.; Hild, S.; Boger, A.; Troll, C.; Rieger, B. “Dual-Side”
Catalysts for High and Ultrahigh Molecular Weight Homopolypropy-
2548-2117
lene Elastomers and Plastomers. Coord. Chem. Rev. 2006, 250, 189−
Antonio Vittoria − Dipartimento di Scienze Chimiche, 211.
Università di Napoli Federico II, 80126 Napoli, Italy; DPI, (9) Kulyabin, P. S.; Goryunov, G. P.; Sharikov, M. I.; Izmer, V. V.;
5600 AX Eindhoven, The Netherlands; [Link]/0000- Vittoria, A.; Budzelaar, P. H. M.; Busico, V.; Voskoboynikov, A. Z.;
0003-3178-7427 Ehm, C.; Cipullo, R.; Uborsky, D. V. Ansa-Zirconocene Catalysts for
Roberta Cipullo − Dipartimento di Scienze Chimiche, Isotactic-Selective Propene Polymerization at High Temperature: A
Università di Napoli Federico II, 80126 Napoli, Italy; DPI, Long Story Finds a Happy Ending. J. Am. Chem. Soc. 2021, 143, 7641−
5600 AX Eindhoven, The Netherlands; [Link]/0000- 7647.
0003-3846-1999 (10) Bochmann, M. The Chemistry of Catalyst Activation: The Case
Peter H. M. Budzelaar − Dipartimento di Scienze Chimiche, of Group 4 Polymerization Catalysts. Organometallics 2010, 29, 4711−
Università di Napoli Federico II, 80126 Napoli, Italy; DPI, 4740.
(11) Deck, P. A.; Beswick, C. L.; Marks, T. J. Highly Electrophilic
5600 AX Eindhoven, The Netherlands; [Link]/0000-
Olefin Polymerization Catalysts. Quantitative Reaction Coordinates for
0003-0039-4479 Fluoroarylborane/Alumoxane Methide Abstraction and Ion-Pair
Alexander Z. Voskoboynikov − Department of Chemistry, Reorganization in Group 4 Metallocene and “Constrained Geometry”
Lomonosov Moscow State University, 119991 Moscow, Catalysts. J. Am. Chem. Soc. 1998, 120, 1772−1784.
Russia; DPI, 5600 AX Eindhoven, The Netherlands (12) Beck, S.; Lieber, S.; Schaper, F.; Geyer, A.; Brintzinger, H. H.
Vincenzo Busico − Dipartimento di Scienze Chimiche, Anion Exchange in Alkyl-Zirconocene Borate Ion Pairs. Are Solvated
Università di Napoli Federico II, 80126 Napoli, Italy; DPI, Alkyl-Zirconocene Cations Relevant Intermediates? J. Am. Chem. Soc.
5600 AX Eindhoven, The Netherlands; [Link]/0000- 2001, 123, 1483−1489.
0001-7079-1651 (13) Song, F.; Lancaster, S. J.; Cannon, R. D.; Schormann, M.;
Leonardo Tensi − Dipartimento di Scienze Farmaceutiche, Humphrey, S. M.; Zuccaccia, C.; Macchioni, A.; Bochmann, M.
Università degli Studi di Perugia, 06123 Perugia, Italy; Synthesis, Ion Aggregation, Alkyl Bonding Modes, and Dynamics of 14-
Electron Metallocenium Ion Pairs [(SBI)MCH2SiMe3+...X−] (M=Zr,
[Link]/0000-0002-0966-6859
Hf): Inner-Sphere (X = MeB(C6F5)3) versus Outer-Sphere (X =
Alceo Macchioni − Dipartimento di Chimica, Biologia e B(C6F5)4) Structures and the Implications for “Continuous” or
Biotecnologie, Università degli Studi di Perugia, 06123 “Intermittent” Alkene Polymerization Mechanisms. Organometallics
Perugia, Italy; DPI, 5600 AX Eindhoven, The Netherlands; 2005, 24, 1315−1328.
[Link]/0000-0001-7866-8332 (14) Sian, L.; Dall Anese, A.; Macchioni, A.; Tensi, L.; Busico, V.;
Complete contact information is available at: Cipullo, R.; Goryunov, G. P.; Uborsky, D.; Voskoboynikov, A. Z.; Ehm,
[Link] C.; Rocchigiani, L.; Zuccaccia, C. Role of Solvent Coordination on the
Structure and Dynamics of Ansa-Zirconocenium Ion Pairs in Aromatic
Hydrocarbons. Organometallics 2022, 41, 547−560.
Notes (15) Rocchigiani, L.; Bellachioma, G.; Ciancaleoni, G.; MacChioni,
The authors declare no competing financial interest. A.; Zuccaccia, D.; Zuccaccia, C. Synthesis, Characterization, Interionic

■ ACKNOWLEDGMENTS
This research forms part of the research program of DPI, project
Structure, and Self-Aggregation Tendency of Zirconaaziridinium Salts
Bearing Long Alkyl Chains. Organometallics 2011, 30, 100−114.
(16) Sian, L.; Macchioni, A.; Zuccaccia, C. Understanding the Role of
Metallocenium Ion-Pair Aggregates on the Rate of Olefin Insertion into
#835. A.M. and C.Z. also thank the European Union - the Metal-Carbon Bond. ACS Catal. 2020, 10, 1591−1606.
NextGenerationEU under the Italian Ministry of University (17) Baier, M. C.; Zuideveld, M. A.; Mecking, S. Post-Metallocenes in
and Research (MUR) National Innovation Ecosystem grant the Industrial Production of Polyolefins. Angew. Chem., Int. Ed. 2014,
ECS00000041 - VITALITY for supporting this work. A.M. and 53, 9722−9744.

16034 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

(18) Boussie, T. R.; Diamond, G. M.; Goh, C.; Hall, K. A.; LaPointe, Polymerization Catalysis: Is Entropy the Key? J. Am. Chem. Soc. 2010,
A. M.; Leclerc, M. K.; Murphy, V.; Shoemaker, J. A. W.; Turner, H.; 132, 13651−13653.
Rosen, R. K.; Stevens, J. C.; Alfano, F.; Busico, V.; Cipullo, R.; Talarico, (35) Gillis, D. J.; Quyoum, R.; Tudoret, M.-J. J.; Wang, Q.; Jeremic,
G. Nonconventional Catalysts for Isotactic Propene Polymerization in D.; Roszak, A. W.; Baird, M. C. Synthesis and Characterization of the
Solution Developed by Using High-Throughput-Screening Technolo- Series of d0 Arene Complexes [Cp*MMe2(η6-Arene)] [MeB(C6F5)3]
gies. Angew. Chem., Int. Ed. 2006, 45, 3278−3283. (M = Ti, Zr, Hf). Organometallics 1996, 15, 3600−3605.
(19) Gao, Y.; Mouat, A. R.; Motta, A.; Macchioni, A.; Zuccaccia, C.; (36) Lancaster, S. J.; Robinson, O. B.; Bochmann, M.; Coles, S. J.;
Delferro, M.; Marks, T. J. Pyridylamido Bi-Hafnium Olefin Polymer- Hursthouse, M. Synthesis and Reactivity of New Mono-
ization Catalysis: Conformationally Supported Hf···Hf Enchainment (Cyclopentadienyl)Zirconium and -Hafnium Alkyl Complexes. Crystal
Cooperativity. ACS Catal. 2015, 5, 5272−5282. and Molecular Structure of [{C5H3(SiMe3)2}HfMe2(.eta.6-Toluene)]-
(20) Froese, R. D. J.; Hustad, P. D.; Kuhlman, R. L.; Wenzel, T. T. [BMe(C6F5)3]. Organometallics 1995, 14, 2456−2462.
Mechanism of Activation of a Hafnium Pyridyl-Amide Olefin (37) Jia, L.; Yang, X.; Stern, C. L.; Marks, T. J. Cationic Metallocene
Polymerization Catalyst: Ligand Modification by Monomer. J. Am. Polymerization Catalysts Based on Tetrakis(Pentafluorophenyl)Borate
Chem. Soc. 2007, 129, 7831−7840. and Its Derivatives. Probing the Limits of Anion “Noncoordination” via
(21) Zuccaccia, C.; Busico, V.; Cipullo, R.; Talarico, G.; Froese, R. D. a Synthetic, Solution Dynamic, Structural, and Catalytic Olefin
J.; Vosejpka, P. C.; Hustad, P. D.; Macchioni, A. On the First Insertion Polymerization Study. Organometallics 1997, 16, 842−857.
of α-Olefins in Hafnium Pyridyl-Amido Polymerization Catalysts. (38) Busico, V.; Cipullo, R.; Friederichs, N.; Ronca, S.; Talarico, G.;
Organometallics 2009, 28, 5445−5458. Togrou, M.; Wang, B. Block Copolymers of Highly Isotactic
(22) Busico, V.; Cipullo, R.; Pellecchia, R.; Rongo, L.; Talarico, G.; Polypropylene via Controlled Ziegler−Natta Polymerization. Macro-
Macchioni, A.; Zuccaccia, C.; Froese, R. D. J.; Hustad, P. D. “Uni et molecules 2004, 37, 8201−8203.
Trini:” In Situ Diversification of (Pyridylamide) Hafnium(IV) (39) Ye, X.; Atienza, C. C. H.; Holtcamp, M. W.; Sanders, D. F.; Day,
Catalysts. Macromolecules 2009, 42, 4369−4373. G. S.; Titone, M. E.; Cano, D. A.; Bedoya, M. S. Int. Pat. Appl.
(23) Zuccaccia, C.; Tensi, L.; Kuhlman, R. L.; Gies, A. P.; Macchioni, WO2017058388, 2017.
A. C-H Activation and Olefin Insertion as Sources of Multiple Sites in (40) Zhou, S.; Doni, E.; Anderson, G. M.; Kane, R. G.; Macdougall, S.
Olefin Polymerization Catalyzed by Cpalkyl Hf(IV) Complexes. ACS W.; Ironmonger, V. M.; Tuttle, T.; Murphy, J. A. Identifying the Roles
Catal. 2017, 7, 563−567. of Amino Acids, Alcohols and 1,2-Diamines as Mediators in Coupling
(24) Kawai, K.; Fujita, T. Discovery and Development of FI Catalysts of Haloarenes to Arenes. J. Am. Chem. Soc. 2014, 136, 17818−17826.
for Olefin Polymerization: Unique Catalysis and Distinctive Polymer (41) Busico, V.; Cipullo, R.; Mingione, A.; Rongo, L. Accelerating the
Formation. Top. Organomet. Chem. 2009, 26, 3−46. Research Approach to Ziegler-Natta Catalysts. Ind. Eng. Chem. Res.
(25) Tshuva, E. Y.; Goldberg, I.; Kol, M. Isospecific Living 2016, 55, 2686−2695.
Polymerization of 1-Hexene by a Readily Available Nonmetallocene (42) Vittoria, A.; Mingione, A.; Abbate, R. A.; Cipullo, R.; Busico, V.
C2-Symmetrical Zirconium Catalyst. J. Am. Chem. Soc. 2000, 122, High Throughput Experimentation Protocol for Quantitative Measure-
10706−10707. ments of Regioselectivity in Ziegler-Natta Polypropylene Catalysis. Ind.
(26) Busico, V.; Cipullo, R.; Ronca, S.; Budzelaar, P. H. M. Mimicking Eng. Chem. Res. 2019, 58, 14729−14735.
Ziegler-Natta Catalysts in Homogeneous Phase, 1. C2-Symmetric (43) Ehm, C.; Mingione, A.; Vittoria, A.; Zaccaria, F.; Cipullo, R.;
Octahedral Zr(IV) Complexes with Tetradentate [ONNO]-Type Busico, V. High-Throughput Experimentation in Olefin Polymerization
Ligands. Macromol. Rapid Commun. 2001, 22, 1405−1410. Catalysis: Facing the Challenges of Miniaturization. Ind. Eng. Chem. Res.
(27) Segal, S.; Goldberg, I.; Kol, M. Zirconium and Titanium Diamine 2020, 59, 13940−13947.
Bis(Phenolate) Catalysts for α-Olefin Polymerization: From Atactic (44) Ehm, C.; Vittoria, A.; Goryunov, G. P.; Izmer, V. V.; Kononovich,
Oligo(1-Hexene) to Ultrahigh-Molecular-Weight Isotactic Poly(1- D. S.; Kulyabin, P. S.; Di Girolamo, R.; Budzelaar, P. H. M.;
Hexene). Organometallics 2005, 24, 200−202. Voskoboynikov, A. Z.; Busico, V.; Uborsky, D. V.; Cipullo, R. A
(28) Busico, V.; Cipullo, R.; Pellecchia, R.; Ronca, S.; Roviello, G.; Systematic Study of the Temperature-Induced Performance Decline of
Talarico, G. Design of Stereoselective Ziegler−Natta Propene Ansa-Metallocenes for IPP. Macromolecules 2020, 53, 9325−9336.
Polymerization Catalysts. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, (45) Ehm, C.; Vittoria, A.; Goryunov, G. P.; Kulyabin, P. S.; Budzelaar,
15321−15326. P. H. M.; Voskoboynikov, A. Z.; Busico, V.; Uborsky, D. V.; Cipullo, R.
(29) Ciancaleoni, G.; Fraldi, N.; Cipullo, R.; Busico, V.; Macchioni, Connection of Stereoselectivity, Regioselectivity, and Molecular
A.; Budzelaar, P. H. M. Structure/Properties Relationship for Weight Capability in Rac-R′2Si(2-Me-4-R-Indenyl)2ZrCl2 Type
Bis(Phenoxyamine)Zr(IV)-Based Olefin Polymerization Catalysts: A Catalysts. Macromolecules 2018, 51, 8073−8083.
Simple DFT Model To Predict Catalytic Activity. Macromolecules 2012, (46) Antinucci, G.; Dereli, B.; Vittoria, A.; Budzelaar, P. H. M.;
45, 4046−4053. Cipullo, R.; Goryunov, G. P.; Kulyabin, P. S.; Uborsky, D. V.; Cavallo,
(30) Ciancaleoni, G.; Fraldi, N.; Budzelaar, P. H. M.; Busico, V.; L.; Ehm, C.; Voskoboynikov, A. Z.; Busico, V. Selection of Low-
Macchioni, A. Activation of a Bis(Phenoxy-Amine) Precatalyst for Dimensional 3-D Geometric Descriptors for Accurate Enantioselectiv-
Olefin Polymerisation: First Evidence for an Outer Sphere Ion Pair with ity Prediction. ACS Catal. 2022, 12, 6934−6945.
the Methylborate Counterion. Dalton Trans. 2009, 41, 8824−8827. (47) Busico, V.; Cipullo, R. Microstructure of Polypropylene. Prog.
(31) Ciancaleoni, G.; Fraldi, N.; Budzelaar, P. H. M.; Busico, V.; Polym. Sci. 2001, 26, 443−533.
Macchioni, A. Structure and Dynamics in Solution of Bis(Phenoxy- (48) Wu, F.; Dash, A. K.; Jordan, R. F. Structures and Reactivity of
Amine)Zirconium Catalysts for Olefin Polymerization. Organometallics Zr(IV) Chlorobenzene Complexes. J. Am. Chem. Soc. 2004, 126,
2011, 30, 3096−3105. 15360−15361.
(32) Gendler, S.; Zelikoff, A. L.; Kopilov, J.; Goldberg, I.; Kol, M. (49) Waterman, R. σ-Bond Metathesis: A 30-Year Retrospective.
Titanium and Zirconium Complexes of Robust Salophan Ligands. Organometallics 2013, 32, 7249−7263.
Coordination Chemistry and Olefin Polymerization Catalysis. J. Am. (50) Pazderski, L. 15N and 31P NMR Coordination Shifts in Transition
Chem. Soc. 2008, 130, 2144−2145. Metal Complexes with Nitrogen- and Phosphorus-Containing Heterocycles,
(33) Cohen, A.; Kopilov, J.; Goldberg, I.; Kol, M. C1-Symmetric 1st ed.; Elsevier Ltd., 2013; Vol. 80, DOI: 10.1016/B978-0-12-408097-
Zirconium Complexes of [ONNO′]-Type Salan Ligands: Accurate 3.00002-0.
Control of Catalyst Activity, Isospecificity, and Molecular Weight in 1- (51) Fish, H.; Hart, S.; Lamb, K. J.; North, M.; Quek, S. C. Z.;
Hexene Polymerization. Organometallics 2009, 28, 1391−1405. Whitwood, A. C.; Woods, B.; Wu, X. Structural Analysis of Five-
(34) Ciancaleoni, G.; Fraldi, N.; Budzelaar, P. H. M.; Busico, V.; Coordinate Aluminium(Salen) Complexes and Its Relationship to
Cipullo, R.; MacChioni, A. Structure-Activity Relationship in Olefin Their Catalytic Activity. Dalton Trans. 2021, 50, 587−598.

16035 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

(52) Cohen, A.; Kopilov, J.; Lamberti, M.; Venditto, V.; Kol, M. Same Catalysts: Influence on Ligand Exchange Dynamics and Catalyst
Ligand, Different Metals: Diiodo-Salan Complexes of the Group 4 Activity. J. Am. Chem. Soc. 2007, 129, 9282−9283.
Triad in Isospecific Polymerization of 1-Hexene and Propylene. (69) Henderson, L. D.; Piers, W. E. Ion Pair Symmetrization in
Macromolecules 2010, 43, 1689−1691. Metallocenium Cations Partnered with Diborane Derived Borate
(53) Rong, Y.; Al-Harbi, A.; Parkin, G. Highly Variable Zr-CH2-Ph Counteranions. J. Organomet. Chem. 2007, 692, 4661−4668.
Bond Angles in Tetrabenzylzirconium: Analysis of Benzyl Ligand (70) Perrin, C. L.; Dwyer, T. J. Application of Two-Dimensional NMR
Coordination Modes. Organometallics 2012, 31, 8208−8217. to Kinetics of Chemical Exchange. Chem. Rev. 1990, 90, 935−967.
(54) Pracht, P.; Bohle, F.; Grimme, S. Automated Exploration of the (71) Bain, A. D. Chemical Exchange in NMR. Prog. Nucl. Magn. Reson.
Low-Energy Chemical Space with Fast Quantum Chemical Methods. Spectrosc. 2003, 43, 63−103.
Phys. Chem. Chem. Phys. 2020, 22, 7169−7192. (72) For other examples of direct observation of M-polymeryl species
(55) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; in solution see: (a) Landis, C. R; Rosaaen, K. A; Sillar, D. R. Direct
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. Observation of Insertion Events at rac-(C2H4(1-indenyl)2)Zr(MeB-
A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.; (C6F5)3)−Polymeryl Intermediates: Distinction between Continuous
Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. and Intermittent Propagation Modes. J. Am. Chem. Soc. 2003, 125,
V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams Young, D.; Ding, F.; 1710−1711. (b) Makio, H.; Fujita, T. Observation and Identification of
Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; the Catalytically Active Species of bis(Phenoxy-Imine) Group 4
Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, Transition Metal Complexes for Olefin Polymerization using 1H NMR
W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, Spectroscopy. Macromol. Symp. 2004, 213, 221−234.
M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; (73) EXSYCALC. [Link]
Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, (74) Cipullo, R.; Busico, V.; Fraldi, N.; Pellecchia, R.; Talarico, G.
M. J.; Heyd, J. J.; Brothers, E. N.; Kudin, K. N.; Staroverov, V. N.; Keith, Improving the Behavior of Bis(Phenoxyamine) Group 4 Metal
T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A. P.; Catalysts for Controlled Alkene Polymerization. Macromolecules
Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, 2009, 42, 3869−3872.
M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, (75) Gisch, N.; Balzarini, J.; Meier, C. Enzymatically Activated
K.; Farkas, O.; Foresman, J. B.; Fox, D. J.. Gaussian 16, Revision A.03. CycloSal-D4T-Monophosphates: The Third Generation of CycloSal-
(56) Baker, J. An Algorithm for the Location of Transition States. J. Pronucleotides. J. Med. Chem. 2007, 50, 1658−1667.
Comput. Chem. 1986, 7, 385−395. (76) Chandran, R.; Madhaiyan, K.; Sridhar, R.; Mahalakshmi, S.
(57) Baker, J. PQS, 2.4; Parallel Quantum Solutions: Fayetteville, Synthesis and Characterization of Anthracene Functionalized Novel
Arkansas, 2001. Fluorescent Amino Acids: Interactions of the Fluorescent Amino Acids
(58) Budzelaar, P. H. M. Geometry Optimization Using Generalized, with Metal Ions. Curr. Org. Synth. 2016, 13, 646−650.
Chemically Meaningful Constraints. J. Comput. Chem. 2007, 28, 2226− (77) Pritchard, B. P.; Altarawy, D.; Didier, B.; Gibson, T. D.; Windus,
2236. T. L. New Basis Set Exchange: An Open, Up-to-Date Resource for the
(59) Ehm, C.; Budzelaar, P. H. M.; Busico, V. Calculating Accurate Molecular Sciences Community. J. Chem. Inf. Model. 2019, 59, 4814−
Barriers for Olefin Insertion and Related Reactions. J. Organomet. Chem. 4820.
2015, 775, 39−49. (78) Cuthbert, E. N. T.; Busico, V.; Herbert, D. E.; Budzelaar, P. H. M.
(60) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Climbing Formation and Activation of Zr/Hf Bis(Phenolate-Ether) Precatalysts.
the Density Functional Ladder: Nonempirical Meta−Generalized Eur. J. Inorg. Chem. 2019, 2019, 3396−3410.
Gradient Approximation Designed for Molecules and Solids. Phys. (79) Cuthbert, E. N. T.; Vittoria, A.; Cipullo, R.; Busico, V.; Budzelaar,
Rev. Lett. 2003, 91, No. 146401. P. H. M. Structure-Activity Relationships for Bis(Phenolate-Ether) Zr/
(61) Balabanov, N. B.; Peterson, K. A. Systematically Convergent Hf Propene Polymerization Catalysts. Eur. J. Inorg. Chem. 2020, 2020,
Basis Sets for Transition Metals. I. All-Electron Correlation Consistent 541−550.
Basis Sets for the 3d Elements Sc-Zn. J. Chem. Phys. 2005, 123, (80) Whitten, J. L. Coulombic Potential Energy Integrals and
No. 064107. Approximations. J. Chem. Phys. 1973, 58, 4496−4501.
(62) Balabanov, N. B.; Peterson, K. A. Basis Set Limit Electronic (81) Baerends, E. J.; Ellis, D. E.; Ros, P. Self-Consistent Molecular
Excitation Energies, Ionization Potentials, and Electron Affinities for Hartree-Fock-Slater Calculations I The Computational Procedure.
the 3d Transition Metal Atoms: Coupled Cluster and Multireference Chem. Phys. 1973, 2, 41−51.
Methods. J. Chem. Phys. 2006, 125, No. 074110. (82) Feyereisen, M.; Fitzgerald, G.; Komornicki, A. Use of
(63) Schuchardt, K. L.; Didier, B. T.; Elsethagen, T.; Sun, L.; Approximate Integrals in Ab Initio Theory. An Application in MP2
Gurumoorthi, V.; Chase, J.; Li, J.; Windus, T. L. Basis Set Exchange: A Energy Calculations. Chem. Phys. Lett. 1993, 208, 359−363.
Community Database for Computational Sciences. J. Chem. Inf. Model. (83) Vahtras, O.; Almlöf, J.; Feyereisen, M. W. Integral Approx-
2007, 47, 1045−1052. imations for LCAO-SCF Calculations. Chem. Phys. Lett. 1993, 213,
(64) Schwerdtfeger, P. The Pseudopotential Approximation in 514−518.
Electronic Structure Theory. ChemPhysChem 2011, 12, 3143−3155. (84) Grimme, S.; Steinmetz, M. Effects of London Dispersion
(65) Peterson, K. A.; Figgen, D.; Dolg, M.; Stoll, H. Energy-Consistent Correction in Density Functional Theory on the Structures of Organic
Relativistic Pseudopotentials and Correlation Consistent Basis Sets for Molecules in the Gas Phase. Phys. Chem. Chem. Phys. 2013, 15, 16031−
the 4d Elements Y-Pd. J. Chem. Phys. 2007, 126, 124101. 16042.
(66) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor, G. (85) Tomasi, J. Thirty Years of Continuum Solvation Chemistry: A
C. Synthesis, Structure, and Spectroscopic Properties of Copper(II) Review, and Prospects for the near Future. Theor. Chem. Acc. 2004, 112,
Compounds Containing Nitrogen-Sulphur Donor Ligands; the Crystal 184−203.
and Molecular Structure of Aqua[1,7-Bis(N-Methylbenzimidazol-2′- (86) Tobisch, S.; Ziegler, T. Catalytic Oligomerization of Ethylene to
Yl)-2,6-Dithiaheptane]Copper(II) Perchlorate. J. Chem. Soc., Dalton Higher Linear α-Olefins Promoted by Cationic Group 4 Cydopenta-
Trans. 1984, 1349−1356. dienyl-Arene Active Catalysts: A DFT Investigation Exploring the
(67) Mathis, D.; Couzijn, E. P. A.; Chen, P. Structure, Dynamics, and Influence of Electronic Factors on the Catalytic Properties by
Polymerization Activity of Zirconocenium Ion Pairs Generated with Modification of the Hemilabile Arene Functionality. J. Am. Chem.
Boron-C6F5 Compounds and Al2R6. Organometallics 2011, 30, 3834− Soc. 2004, 23, 4077−4088.
3843. (87) Zaccaria, F.; Ehm, C.; Budzelaar, P. H. M.; Busico, V. Accurate
(68) Alonso-Moreno, C.; Lancaster, S. J.; Zuccaccia, C.; Macchioni, Prediction of Copolymerization Statistics in Molecular Olefin
A.; Bochmann, M. Evidence for Mixed-Ion Clusters in Metallocene Polymerization Catalysis: The Role of Entropic, Electronic, and Steric

16036 [Link]
Inorg. Chem. 2023, 62, 16021−16037
Inorganic Chemistry [Link]/IC Article

Effects in Catalyst Comonomer Affinity. ACS Catal. 2017, 7, 1512−

1519.
(88) Zaccaria, F.; Cipullo, R.; Budzelaar, P. H. M.; Busico, V.; Ehm, C.
Backbone Rearrangement during Olefin Capture as the Rate Limiting
Step in Molecular Olefin Polymerization Catalysis and Its Effect on
Comonomer Affinity. J. Polym. Sci., Part A: Polym. Chem. 2017, 55,
2807−2814.
(89) Falivene, L.; Credendino, R.; Poater, A.; Petta, A.; Serra, L.;
Oliva, R.; Scarano, V.; Cavallo, L. SambVca 2. A Web Tool for
Analyzing Catalytic Pockets with Topographic Steric Maps. Organo-
metallics 2016, 35, 2286−2293.

16037 [Link]
Inorg. Chem. 2023, 62, 16021−16037