Quantitative models in transition-metal catalysis are typically restricted to narrow catalyst families, largely because no universal featurization strategy exists for systems that differ in ligand architecture, metal, and/or coordination geometry. On the example of regioselectivity in propene polymerization, we introduce a simple approach allowing the design of custom ligand-independent 3D-steric descriptors guided by ChemDraw depictions of competing transition states. A simple multiple linear regression model trained on a rigorously curated high-throughput dataset (86 catalysts, including octahedral and tetrahedral geometries, 2 metals, 3 families, and 8 classes) achieves high accuracy (R 2 = 0.88) with only four 3D-steric descriptors and one electronic descriptor. Importantly, the model is fully interpretable: descriptors directly map onto specific transition-state interactions (and the ChemDraw depictions), enabling clear connections between ligand modifications and catalytic outcomes. This approach demonstrates how experimental intuition, encoded through ChemDraw depictions, can be translated into robust quantitative predictions, portable across catalyst classes.

Ziegler-Natta (ZN) catalysts for stereoselective propene polymerization originated as relatively simple mixtures of crystalline TiCl3 in a layered modification and an Al-alkyl activator. At a later stage, to enhance productivity with respect to Ti and decrease the amount of acidic Ti-Cl residues in the polymer, supported versions were introduced in which TiCl4 is adsorbed on a nanocrystalline MgCl2 matrix and subsequently alkylated and reduced by AlEt3. The addition of certain organic electron donors to the solid precatalyst ('Internal Donor', ID) and, in most cases, to AlEt3 ('External Donor', ED) is mandatory to achieve a high stereoselectivity, but how donor modification works in detail is not well understood. For several decades now the models of catalytic species for Mg-Ti systems have been inspired by the structural similarity between the layer lattices of MgCl2 and TiCl3, which led many to propose that the active site precursors are epitaxial TiCl4 adsorbates with coordinatively saturated octahedral Ti, and hence no room left for direct Ti-Donor bonding interactions. In the present communication, based on state-of-the-art Density Functional Theory calculations and topological analyses of the surface, we show that the hypothesis of non-epitaxial TiCl4-ID adducts with ID molecules directly bonded to the active Ti is in much better agreement with the experimental facts. Breaking the paradigm of epitaxy leads to redefine structure-properties relations for these important catalysts, and can change the strategy of donor design.

Catalytic stereoselective hydrogenation of unfunctionalized polyenes remains a major challenge because multiple stereogenic centers must be generated in the absence of directing groups. Here we investigate squalene, a flexible hydrocarbon containing six isolated trisubstituted double bonds, as a demanding substrate for assessing stereochemical bias in the complete hydrogenation of an unfunctionalized polyene. Evaluation of Rh- and Ir-based homogeneous catalysts identified one Ir-P,N complex that promotes complete conversion to squalane while inducing a highly non-statistical diastereomeric distribution, reaching up to 90% d.e. The stereochemical outcome was quantified directly by 13C NMR spectroscopy of crude reaction mixtures, without derivatization or chromatographic separation. Buried-volume analysis was used to compare the steric environment of the selective Ir catalyst with representative non-selective systems, providing a qualitative interpretation of the observed behaviour. To our knowledge, this work provides the first catalytic access to stereoregular squalane and establishes a practical NMR-based approach to evaluate stereochemical bias in hydrogenated isoprenoid hydrocarbons.

Ethylene/Vinylcyclohexane (E/V) copolymers are attractive materials combining polyethylene-like processability with enhanced thermal properties. Until now, establishing quantitative relationships between chain microstructure and material properties has been hampered by the incomplete assignment of the 13C NMR spectrum for samples with high V incorporation and consecutive V units, now accessible with certain non-metallocene catalysts. In this work, we address this limitation by means of natural-abundance 2D 13C–13C INADEQUATE NMR experiments, enabling a comprehensive assignment of the 13C NMR spectrum of regioregular E/V copolymers. This, in turn, enables quantitative analysis of comonomer sequence distribution at the triad level and determination of monomer reactivity ratios for different catalyst systems. Copolymers with comparable V contents but different sequence distributions, from alternation-enriched to more blocky microstructures, exhibit significant differences in thermal behavior despite comparable compositions, highlighting the important role of sequence distribution in governing the thermal behavior of E/V copolymers.

Polyolefins are unique among synthetic polymers because their wide application envelope originates from a finely controlled microstructure of hydrocarbon chains, lacking any distinctive functional groups. This hampers the methods of automated sorting based on vibrational spectroscopies and calls for much more complex 13C NMR elucidations. High-temperature cryoprobes have dramatically shortened the acquisition time of 13C NMR spectra, and few minutes are now enough for polyolefin classification purposes; however, conventional data analysis remains labor and time-consuming. In this paper, we introduce an instrument for automated fast determinations of the 13C NMR microstructure on polyolefin materials, implemented by integrating High-Throughput Experimentation and Data Science tools and methods. From the scientific standpoint, the main interest of the approach is the solution proposed to address the general problem how to rapidly characterize statistically distributed analytes, of which synthetic polymers are a most important case. In practical terms, the instrument represents the first automated tool for microstructural polyolefin analysis: it is readily applicable to monomaterials, whereas extension to multimaterials, including postconsumer streams, is feasible but still requires some work.

Group 4 metallocenes are competent catalysts for the oligomerization of higher α-olefins. Among the many chemical and physical variables of importance in the process, one is the choice of cocatalyst (activator). The impact of various activators on the performance of a representative catalyst, (nBuCp)2ZrCl2, in the oligomerization of 1-octene was thoroughly investigated; in particular, the molecular weight distribution (MWD) of the oligomers was determined by means of high-resolution high performance liquid chromatography (HR-HPLC). Unexpectedly, a bimodal MWD was highlighted when the precatalyst was activated with methylaluminoxane (MAO), whereas a single Schulz–Flory (SF) MWD was observed with borate salts. The presence of Al centers with different Lewis acidity in the complex and ill-defined structure of MAO is well known, and the broadening effects on the MWD of olefin polymerization products made with metallocene/MAO catalyst systems have been reported before. However, to the best of our knowledge, clear HR-HPLC evidence of two active species resulting from activation with MAO of one single zirconocene precursor, yielding two discrete SF product distributions, is unprecedented. By varying the polarity of the reaction medium, we managed to modulate the MWD of the oligomers from bimodal to monomodal, even with MAO, thus demonstrating that ion pairing effects are behind these unusual findings.

Well-defined Al-alkyl borate (AAB) salts {[iBu2(L)Al]2(μ-H)}+[B(C6F5)4]- (AlHAl_L) with L = N-donor ligands have been recently reported as promising "complete" cocatalysts for olefin polymerization. Herein, we explore structural variations of AlHAl_L going beyond the class of N-donors like the prototypical N,N-dimethyl aniline (DMA). Thirteen P-, O-, and C-donor ligands were screened, allowing isolation of AAB salts with mono- and bidentate phosphines, alkyl-, aryl-, and silyl-ethers, and a N-heterocyclic carbene. Except for the diphosphine with the longest spacer between the P atoms [bis(diphenylphosphino)hexane, DPPH], all donors gave well-defined tetracoordinate or tricoordinate molecular species, which were characterized in solution (NMR) and solid state (XRD), and tested as cocatalyst in ethylene/1-hexene copolymerization with an ansa-zirconocene catalyst [rac-Me2Si(2-Me-4-Ph-Ind)2ZrCl2]. The vast majority of novel AAB salts provided active catalytic systems, further demonstrating the broad tunability of these species. Consistent with previous studies, variability in productivity upon L variation is primarily related to the efficiency of precatalyst activation, determining the fraction of Zr active sites. Variations in polymer molecular weight and comonomer incorporation observed with some P-, O- and C-donor ligands indicate that also interactions between the L donors and the Zr active species might be relevant in determining catalytic performance in some cases.

Chromophore quench-labeling (CQL) is an elegant and effective method to count the fraction of active metal (x*) in olefin polymerizations mediated by molecular transition metal catalysts. In this study, the method was successfully applied for the first time to a heterogeneous Ti-based Ziegler-Natta catalyst of industrial relevance. CQL experiments using 1-hexene as the monomer ended up with a value of x* = 0.49 ± 0.09%, close to that measured for the same catalyst in the polymerization of propene under otherwise identical conditions using an alternative quenched flow (QF) approach. We ascribe such a low x* value to the fact that the catalytically active species are transient metastable surface Ti adducts, as proposed in the recent literature.

The phase behavior of ethylene/alk-1-ene statistical multiblock copolymers (OBCs), synthesized via chain shuttling copolymerization, is investigated. These copolymers consist in the alternation of amorphous (soft) blocks with high alk-1-ene content and crystalline (hard) blocks with low alk-1-ene content and are characterized by significant inter- and intra-molecular constitutional heterogeneity. The alk-1-ene content in the soft and hard blocks is approximate to 20 and less than 0.5 mol%, respectively. Samples incorporating various comonomers (hex-1-ene, oct-1-ene, 4-methyl-pent-1-ene, hexadec-1-ene) and with different soft-to-hard block weight ratios (80/20, 65/35, 50/50) exhibit diverse solid-state morphologies, influenced by hard block crystallization. Rheological analysis in the melt reveals mesophase separation through time-temperature superposition (TTS) principle failure at low frequencies. Six key descriptors are identified to characterize phase behavior: (i) average number molecular mass (M-n); (ii) molecular mass of the longest hard-soft units; (iii) crystallization temperature (T-c); (iv) segregation strength (chi N), given by the product of the interaction parameter chi of Flory and the number of monomers in the long hard-soft units N; (v) degree of morphological heterogeneity (D-h); (vi) the exponent (n) marking the low-frequency behavior of elastic modulus. These parameters are visualized in a Kiviat's diagram to assess mesophase separation, domain structure, and enable quantitative comparison across different OBC compositions. The approach can be more general, as it can be extended to other complex systems showing tendency toward mesophase separation, allowing for finding useful information to study structure-properties relationships.

The role of entanglements and co-crystallization in the compatibilization of iPP/PE blends with PP-based diblock copolymers (BCs) was investigated. A smart design of experiments was set up, based on the selection of two different BCs, iPP/HDPE and iPP/LDPE blends (30/70 wt%/wt%), and different processing conditions. The selected BCs were a hard-hard copolymer with crystalline iPP and PE blocks (BC1), and a hard-soft copolymer with iPP blocks linked to amorphous EP blocks (BC2). Both BCs significantly improved blend morphology by inducing a fine dispersion of the minority iPP phase in the PE matrix, even at low loads (3-5 wt%). However, they exert different compatibilization effectiveness. In iPP/HDPE blends, trapped entanglements are coupled with well-formed co-crystals of large lamellar thickness, and only 3 wt% of BC1 is sufficient to ensure good ductility and toughness. For iPP/LDPE blends, larger BC1 content (over 5 wt%) was needed due to formation of thinner co-crystals. In contrast, BC2, which does not co-crystallize with PE, only ensures good adhesion in rapidly cooled samples due to the "freezing" of trapped entanglements.

Cationic Salan Zr- and Hf-benzyl precatalysts for olefin polymerization were reacted with AlMe3 and ZnMe2 to probe their tendency to form stable heterobimetallic adducts and understand the role played by the ligand-based Lewis-basic functionalities in modulating the nature of such adducts. NMR studies in solution and DFT computations show that unlike typical metallocenes, Salan complexes bind AlMe3 and ZnMe2 to form only a single μ-Me interaction between the transition and the main-group metal. The latter is engaged in a further interaction with one oxygen atom of the ligand, leading to an unsymmetrical adduct. The formed heterobimetallic complexes are fluxional in solution, and their chemical exchange patterns were characterized by 1H EXSY NMR. In the case of AlMe3, exchange between bridging and terminal Al-Me moieties occurs preferentially via an intramolecular mechanism without the involvement of external AlMe3. In stark contrast, both bridging and terminal Zn-Me groups undergo chemical exchange with external ZnMe2 in the corresponding heterobimetallic adduct. Quantification of the activation parameters for Hf/ZnMe2 systems suggests a dissociative mechanism for the exchange involving the bridging methyl and an associative mechanism for the terminal methyl.

Automated High-Throughput Experimentation (HTE) workflows are increasingly used in catalysis to generate large and reliable databases of Quantitative Structure-Properties Relations (QSPR). Data-driven approaches integrating HTE and Artificial Intelligence (AI) tools such as Machine Learning (ML) and Deep Learning (DL), can be exploited to rapidly and thoroughly navigate complex variable hyperspaces and build models predicting catalyst performance. In a recent publication we highlighted the utilization of a custom-made HTE/AI workflow for the preparation, screening, and "black-box" QSPR modeling of a large library of "High-Yield" Ziegler-Natta (HY-ZN) catalyst formulations, with the ultimate goal of identifying Internal Donors (ID) specifically for tunable applications. In the present paper, we illustrate how a smaller but more homogeneous ID subset containing diesters only can be utilized for "clear-box" QSPR modeling also aiming at increased mechanistic insights. The study led to unconventional conclusions that challenge some long-standing hypotheses about the role of surface modification by electron donors in HY-ZN catalysis. In particular, evidence was achieved that the ID leaves a permanent footprint in the catalyst, which durably affects catalyst performance even in case the ID is reactive with the AlEt3 activator and is extensively removed from the solid phase during polymerization.

Linear Low-Density Polyethylene (LLDPE) is a versatile polyolefin made by copolymerizing ethene with minor amounts of a 1-alkene. The short side chain branches in the comonomer units partly hinder the ability of the polyethylene main chain to crystallize, thus providing a way to fine-tune material properties between the extremes of a thermoplastic and a moderate elastomer. In this function, higher 1-alkenes such as 1-hexene or 1-octene are more effective than shorter homologs like propene or 1-butene, because their alkyl substituents are fully incompatible with the polyethylene lattice. On the other hand, the former comonomers are also more expensive and, above all, poorly reactive with heterogeneous Ziegler-Natta (ZN) catalysts, the workhorses of the polyolefin industry; as a matter of fact, they can only be used with technologically more demanding molecular catalysts. The molecular kinetic factors governing this important and complicated catalytic reactivity are still poorly understood, and perusal of the literature led us to conclude that data reliability is often questionable due to experimental limitations in reaction equipment and protocols, particularly in academic laboratories. In this study, we made use of a state-of-the-art High-Throughput Experimentation workflow to measure the reactivity ratios with ethene of two representative higher 1-alkenes, namely 1-hexene and 1-decene, in the presence of a variety of well-defined molecular catalysts of metallocene and post-metallocene nature comparatively with a typical MgCl2/TiCl4 ZN catalyst for polyethylene application. We found that the two comonomers react almost identically with molecular catalysts, whereas a major decrease in reactivity for 1-decene compared with 1-hexene was observed idiosyncratically for the ZN catalyst. In our opinion, the overall results suggest that in the latter case, surface effects can be dominant over direct comonomer interactions with the coordination sphere of the active metal in dictating the observed molecular kinetic behavior.

Artificial Intelligence (AI) tools and methods are dramatically innovating the application protocols of most polymer characterization techniques. In this paper, we demonstrate that, with the aid of custom-made and properly trained machine learning algorithms, analytical Crystallization Elution Fractionation (aCEF) can be changed from an ancillary to a standalone approach usable to identify and categorize commercially relevant polyolefin materials without any prior information. The proposed protocols are fully operational for monomaterials, whereas for multimaterials, integration with AI-aided 13C NMR is a realistic intermediate step.

In olefin polymerization, even seemingly simple concepts like the influence of electronic effects have sometimes eluded qualitative understanding despite 70 years of continuing research in the field. Of course, the intimate coupling of electronic and steric effects that olefin polymerization is so famous for might be simply too complex, with data science approaches - which are rapidly gaining adoption in catalysis - being the only way to solve the puzzle. Data science relies on machine-readable features or descriptors that encode essential aspects of the catalysts, and the accuracy of models depends both on the quality of data and featurization. Here, we show that some of the basic assumptions used so far (in any kind of modelling) may be flawed to the extent that they prevent accurate evaluation (and separation) of steric and electronic effects. We undertake a comprehensive analysis of the suitability of different model structures for data science approaches and analyze the performance, reliability, and data spacing of common electronic descriptors determined thereof for several group 3 and group 4 metal complexes. The insight developed in this work points not only to the complexity of the underlying chemistry being problematic but also to the inefficiency of many commonly employed descriptors in properly capturing electronic effects relevant for olefin polymerization. Recognizing the strengths and weaknesses of various approaches may help researchers select appropriate features/descriptors and better understand the scope of models beyond the initial training data.

Organic electron donors are essential components of Ziegler-Natta (ZN) catalysts to produce isotactic polypropylene. In particular, aromatic or aliphatic diesters are widely used as ‘Internal Donors’ (ID) in MgCl2/ID/TiCl4 precatalyst formulations. Diesters are reactive with AlEt3 (by far the most common ZN precatalyst activator) and are partly removed from the solid phase in the early stages of the polymerization process; this is detrimental for catalyst functioning, and a surrogate donor (‘External Donor’ (ED), usually an alkoxysilane) is added to the system to restore performance. Recent studies, however, demonstrated that even in cases where most of the diester is extracted by AlEt3, the active sites retain a ‘memory’ of it in several aspects of the catalytic behavior (such as, e.g., the average productivity and the polydispersity index of the polymer produced). Considering that the residual diester is always in molar excess with respect to the active Ti, one may speculate that long-lasting interactions between the latter and diester molecules can occur. In turn, this should imply that the reactivity of AlEt3 is different with binary MgCl2/ID or ternary MgCl2/ID/TiCl4 mixtures. In this work, the latter hypothesis was explored for a library of diester IDs with large structural diversity. In line with the anticipation, the fractional amount of ID extracted by AlEt3 was generally lower for ternary mixtures, although to an extent exquisitely dependent on diester structure.

Formation of long chain branches (LCB) in polyethylene (PE), via incorporation of in situ generated vinyl macromonomers, is known to affect material properties dramatically, making their detection and quantification of primary importance. 13C NMR spectroscopy is the archetypal technique for the analysis of polymer microstructure, yet it suffers from major limitations in the analysis of LCB in polyethylene, primarily in terms of resolution. Herein, we propose a simple and effective methodology for detecting and quantifying LCB based on the analysis of C atoms in β-position with respect to the branching point. By analyzing model ethylene/α-olefin copolymers bearing methyl, ethyl, butyl, hexyl or tetradecyl chain branches, we show how the Cβ resonances can be used to discriminate between shorter or longer branches. Importantly, the proposed method allows the most critical discrimination between hexyl-type branches and LCB, with an up to three-fold detection enhancement with respect to previously proposed procedures based on the analysis of the methine carbons. The proposed approach is then tested on a representative industrial sample of HDPE, proving that it is suitable to detect very small amounts of LCB.

The aluminum-alkyl borate (AAB) salt {[iBu2(DMA)-Al]2(μ-H)}+[B-(C6F5)4]- (AlHAl_DMA; DMA = N,N-dimethylaniline) is able of fully activating dichloride precatalysts for olefin polymerization and serving as an impurity scavenger, thus deserving to be called a molecular cousin of the well-established methylaluminoxane (MAO). With respect to MAO, it offers the advantage of having a well-defined molecular structure, which was exploited herein to investigate its mechanism of action as a cocatalyst. Particularly, the reaction of the precatalyst (Me2SiCp2)-ZrCl2 with AlHAl_DMA and with stable [AliBu2(L)]+, modeling the putative abstracting species [AliBu2(DMA)]+, was studied. The latter reaction led to the isolation of a rare, singly bridged Zr-(μ-Cl)-Al heterodinuclear adduct (2), which is a plausible intermediate of chloride abstraction from the precatalyst. Addition of di-iso-butylaluminum hydride (DIBAL-H) to 2 yielded a mixture of several multinuclear Zr/Al adducts with bridging μ-Cl and μ-H fragments (3-6), which were fully characterized by in-depth 2D NMR spectroscopy. Analogous products were observed in the reaction between (Me2SiCp2)-ZrCl2 and AlHAl_DMA, reinforcing the hypothesis that they are intermediates of chloride/hydride exchange, which generates a polymerization-active Zr-H species. The solid-state structure of [(Me2SiCp2)-Zr]2(μ-H)-(μ-Cl)-(μ2 -iBu2AlH2) (5) was determined by single-crystal X-ray diffraction. The presence of the μ-H fragment in AlHAl_DMA appears to be relevant also for determining the excellent impurity scavenging properties of this cocatalyst, as it was found to react more rapidly than Al-iBu moieties upon exposure of solutions of this cocatalyst to atmospheric oxygen and moisture.

A set of metallocene olefin polymerization catalysts bearing triptycene moieties in either position 4-5 (complexes Ty1-Ty5) or in position 5-6 (complexes Ty6-Ty8) of the basic dimethylsilyl-bridged bis(indenyl) system has been tested in propene polymerization and in ethene/1-hexene copolymerization. Comparison of the results with QSPR (quantitative structure-property relationship) predictions not parametrized for these exotic ligand variations demonstrates that trends can still be identified by extrapolation. Interestingly, Ty7, upon suitable activation, provides a highly isotactic polypropylene with an exceptional amount of 2,1 regio-errors (8%). The previously developed QSPR type models successfully predicted the low regioselectivity of this catalyst, despite the fact that the catalyst structure differs significantly from the benchmark set.

Modern "high-yield" Ziegler-Natta (HY-ZN) catalysts for the isotactic polymerization of propene consist of MgCl2/TiCl4/electron donor/AlEt3 formulations which are much more complex and far less understood than the early TiCl3-based versions. Many important aspects, such as the structure and formation of the active sites and the details of their modification by means of organic electron donors, are largely unknown, and the resulting hybrid (organic-inorganic) catalysts therefore can be regarded as black boxes. For a long time, the similarity between the layered crystal lattices of "violet" TiCl3 and MgCl2 and the concept of epitaxial TiCl4 chemisorption on the latter have been set at the foundation of the mechanistic hypothesis on the inner functioning of HY-ZN systems. Here we provide experimental and computational evidence that metastable nonepitaxial MgCl2/TiCl4 adducts with dynamic character are involved in the activation of the precatalyst with AlEt3 and its deactivation with alkoxysilanes (competent surface modifiers in the catalytic state). These findings can have important consequences in the modeling of catalytic Ti species.