A general strategy for heterogenizing olefin polymerization catalysts and the synthesis of polyolefins and composites

Polyolefins constitute roughly half of the over 380 million tonnes of plastics produced annually. Introducing polar functional groups into polyolefin backbones would broaden applications, yet industrial early transition-metal catalysts are poisoned by polar groups, making ethylene–polar monomer copolymerization a long-standing challenge. Late transition-metal catalysts (Ni, Pd) are more tolerant to polar groups and promising for such copolymerizations, but industry favors heterogeneous systems for morphology control and continuous processing. Conventional heterogenization routes for homogeneous catalysts on supports often lead to weak catalyst–support interactions, leaching, high cocatalyst demand, and exacerbated issues in the presence of polar comonomers. The research question is whether a general, simple heterogenization strategy can create strong, polar-tolerant catalyst–support interactions that enhance activity, stability, comonomer incorporation, and enable morphology control, high-temperature/scale polymerization, and direct composite fabrication.

Three common heterogenization routes are used: (a) immobilizing a precatalyst onto cocatalyst-pretreated supports; (b) adding cocatalyst to a precatalyst-pretreated support (often avoided due to side reactions); and (c) activating precatalyst with cocatalyst prior to support impregnation (unsuitable for sensitive species). These approaches suffer from catalyst leaching and interference by polar comonomers. Covalent tethering can mitigate leaching but requires complex synthesis and hinders characterization. Few late transition-metal heterogeneous systems exist for ethylene polymerization and even fewer for copolymerization with polar monomers. Reported systems include: clay/SiO2-supported phosphine-sulfonate Pd (moderate activity 142 kg mol−1 h−1, up to 4.3% incorporation, low Mw ~4700 g/mol); polystyrene-bound Pd systems with decreased activity/incorporation; α-diimine Ni on sulfated ZrO2 (low activity ~4 kg mol−1 h−1; 0.4% incorporation); related Pd on sulfated ZrO2 (14.9 kg mol−1 h−1; 0.46% incorporation); and anilinonaphthoquinone Ni on MMAO/SiO2 capable of select copolymerizations. These works highlight the need for stronger, more tolerant catalyst–support interactions and generalizable strategies.

Design of an ionic anchoring strategy (IAS): ligands were functionalized with hydroxyl groups to introduce ionic tags (e.g., −ONa) to facilitate strong interactions with oxide supports. - Ligand and complex synthesis: Hydroxyl-functionalized phosphinophenolate ligands were synthesized from 2-tert-butylhydroquinone via protection–lithiation–substitution–deprotection, then complexed with (pyridine)2NiMe2 to give Ni–OH and converted to Ni–ONa (also perfluorophenyl variants Ni-F–OH and Ni-F–ONa). A phenoxy–imine titanium system was prepared by condensing salicylaldehyde with 4-aminophenol to form the salicylaldimine, then treated with NaH (2 equiv.) and TiCl4 (0.5 equiv.) to form Ti–ONa. Analogous complexes without ONa (Ni, Ti) were synthesized for comparison. - Catalyst immobilization: Complex solutions were stirred with pretreated oxide supports (SiO2, TiO2, Al2O3, ZnO, MgO; powders pre-calcined at 600 °C for 6 h). Adsorption capacities were measured (e.g., Ni–ONa on SiO2: 77 μmol/g, exceeding Ni–OH and Ni), evidencing strong ionic tag–support interactions. Metal loading typically 20 μmol/g support. - Polymerization testing: Conducted in hydrocarbon solvents (hexanes or n-heptane). Ni catalysts were used without cocatalysts/scavengers; Ti catalysts employed Et2AlCl (500 equiv.). Conditions varied by study (ethylene pressure 2–30 atm; 30–170 °C; time 10–90 min; catalyst amounts 0.1–5 μmol for bench; scale-up in 2.5 L reactor with 5–30 μmol Ni). Homopolymerizations and copolymerizations with polar comonomers (e.g., methyl acrylate, tert-butyl acrylate, trimethoxyvinylsilane, allyl substrates, and long-spacer monomers like methyl 10-undecenoate) and α-olefins (1-hexene, 4-methyl-1-pentene) were conducted. - Active site counting: Poisoning with subequivalents of PMes; activity vs poison plotted and extrapolated to zero-activity to estimate fraction of active sites. - Characterization: Solid-state 1H MAS NMR to probe electronic environment; FT-IR to monitor C–O and C≡N shifts upon immobilization; DSC for Tm; GPC at 150 °C in trichlorobenzene for Mn/Mw/PDI; SEM for morphology; thermal diffusivity (laser flash); electrical conductivity (film resistance); cone calorimetry (ISO 5660) for flame behavior. - High-temperature stability: Time-on-stream studies at 100–170 °C, 30 atm, monitoring yield vs time and real-time ethylene consumption. - Scale-up: 2.5 L high-pressure reactor runs at 20 atm and 120–150 °C to assess productivity and polymer Mw. - Composite generation: In situ immobilization of Ni–ONa on fillers (glass fiber, Al2O3, graphene, expanded graphite, TiO2) followed by polymerization to generate composites with controlled filler loadings; comparisons to conventional melt-compounded HDPE composites prepared via twin-screw extrusion. - Photodegradation tests: PE/TiO2 composites (5 wt% TiO2) prepared via in situ polymerization and blending controls; UV irradiation (300–400 nm, 40 W/m2) for up to 192 h; Mw and PDI monitored to assess degradation; tested also on a polar-functionalized copolymer (ethyl ene–methyl 10-undecenoate) prepared with Ni–ONa–TiO2.

- Stronger immobilization via ionic tagging: ONa-tagged complexes exhibited substantially higher adsorption on SiO2 than non-ionic analogs (e.g., Ni–ONa 77 μmol/g vs Ni–OH and Ni much lower), supporting robust catalyst–support interactions. - Ethylene polymerization (Ni): Homogeneous Ni–ONa outperformed Ni and Ni–OH (activity up to 12.1×10^6 g mol−1 h−1; Mn 5.84×10^5 g/mol). Supported Ni–ONa catalysts showed dramatically enhanced activity and molecular weight, especially on more basic supports, with a clear trend MgO > ZnO > Al2O3 > TiO2 > SiO2. Representative data (10 min, 80 °C, 8 atm, 0.1 μmol Ni): • Ni–ONa–MgO: 52.8×10^6 g mol−1 h−1; Mn 2.41×10^6 g/mol; PDI ~2.3 (Entry 10). • Ni–F–ONa–MgO (30 atm): up to 109.2×10^6 g mol−1 h−1; Mn 3.265×10^6 g/mol at 80 °C (Entry 14); Mn up to 4.492×10^6 g/mol at 30 °C (Entry 15). • Ni–SiO2 (without ONa): inactive, highlighting necessity of ionic tag. Active site counting indicated ~95% of immobilized Ni centers were catalytically active. - Ethylene copolymerization (Ni): Supported catalysts improved activity, molecular weight, and comonomer incorporation vs unsupported counterparts. For E–MA at 80 °C, 8 atm (30 min, 5 μmol Ni): • Ni–ONa–MgO: activity 9.6×10^4 g mol−1 h−1; incorporation 2.5 mol%; Mc 5.6×10^4 g/mol (Entry 7). • Ni–F–ONa–MgO: up to 12.4×10^4 g mol−1 h−1 at 8 atm with 2.9 mol% incorporation (Entry 9); at 30 atm, activity 30.4–41.0×10^4 with 0.5–1.3 mol% incorporation and higher Mc (Entries 10–11, 21). • Broader monomer scope: acrylates, trimethoxyvinylsilane, allyl monomers, and long-spacer monomers (e.g., methyl 10-undecenoate) achieved high activity (up to 2.32×10^6 g mol−1 h−1; Entries 22–24) and high molecular weights (Mc up to 3.07×10^5 g/mol). - Ethylene polymerization and copolymerization (Ti): With Et2AlCl cocatalyst, Ti–ONa supported catalysts showed large boosts in activity and Mw. For homopolymerization at 30 °C, 8 atm (10 min): Ti–ONa–SiO2 achieved 38.9×10^6 g mol−1 h−1; Mn 2.897×10^6 g/mol (Entry 4), with further increases on Al2O3 and MgO. In E/1-hexene copolymerization, Ti–ONa–SiO2 increased activity (5.5×10^6) and Mw (Mn 5.69×10^5 g/mol) with moderate incorporation (2.5 mol%), vs Ti (2.6 mol% but much lower Mw and activity). - High-temperature stability: Supported Ni–ONa–MgO and Ni–F–ONa–MgO retained activity at 150–170 °C (30 atm), whereas homogeneous Ni rapidly deactivated. At 120–150 °C (10 min, 30 atm): activities remained high (e.g., 81.0 and 57.0×10^6 g mol−1 h−1) with Mn up to ~2.085×10^6 g/mol. - Large-scale runs: In a 2.5 L reactor at 20 atm and 120 °C using 5 μmol Ni–ONa–MgO, produced 150 g PE per run with activity 1.8×10^8 g mol−1 h−1 and Mn 1.62×10^6 g/mol. At 150 °C, Ni–F–ONa–MgO showed higher activity (1.5×10^8) and Mn 1.541×10^6 g/mol. - Morphology control: Supported catalysts generated free-flowing PE particles, avoiding reactor fouling observed with homogeneous catalysts. - In situ composites: Direct polymerization on fillers yielded composites with superior mechanical performance and dispersion versus melt-compounded controls. Examples: 10 wt% glass fiber composites exhibited 35.5 MPa tensile strength and ~970% elongation (vs <50% for extruded). Al2O3 composites showed improved thermal diffusivity at lower loadings (0.28/0.38 mm^2/s at 5/10 wt% vs 0.23/0.27 at 20/40 wt% for extruded). Graphene composites exhibited electrical conductivities orders of magnitude higher than extruded at similar loadings (e.g., 9.5×10^−4 vs 1.9×10^−8 S/m at 10 wt%). Expanded graphite composites showed improved flame retardancy metrics. - Photodegradation: In situ PE/TiO2 (5 wt%) composites with superior mechanical properties underwent significantly enhanced UV-induced Mw reduction over 192 h (from 1.152×10^6 to 1.14×10^5 g/mol) compared to blended controls. A polar-functionalized copolymer (0.6 mol% methyl 10-undecenoate, ~5 wt% TiO2) degraded from 3.07×10^5 to 2.1×10^4 g/mol, indicating polar groups further facilitate photocatalytic degradation.

The ionic anchoring strategy creates strong interactions between ionic tags (e.g., −ONa) on the catalyst ligands and oxide supports, stabilizing active sites and resisting leaching or poisoning by polar comonomers. The electronic influence of support basicity was evident: more basic supports (MgO, ZnO) increased electron donation to the metal center via the ionic tag, shifting NMR resonances and correlating with higher activities, molecular weights, and comonomer incorporation during copolymerization. This tuning enables single-site behavior with narrow PDIs and robust high-temperature operation. Supported Ni systems not only overcame typical trade-offs in polar-comonomer copolymerization (simultaneously enhancing activity, molecular weight, and incorporation) but also allowed operation at industrially relevant temperatures and scales. Morphology control inherent to heterogeneous catalysts was leveraged to prevent reactor fouling and to directly produce composites with uniform filler dispersion, translating to enhanced mechanical, thermal, electrical, and flame-retardant properties at relatively low loadings. Using photoactive supports like TiO2 enabled efficient photocatalytic degradation of otherwise persistent polyolefins; incorporating small amounts of polar comonomers further accelerated degradation, likely by introducing oxidizable sites and enhanced UV absorption, suggesting a path toward more sustainable end-of-life management.

A simple, general ionic anchoring strategy enables the heterogenization of late transition-metal olefin polymerization catalysts (e.g., phosphinophenolate Ni, phenoxy–imine Ti) on diverse oxide supports. The resulting catalysts exhibit greatly enhanced performance compared to homogeneous analogs: higher activity, stability (including at 120–170 °C), product molecular weight, and comonomer tolerance/incorporation. The approach affords morphology control and scalability (demonstrated at 2.5 L scale), and facilitates direct synthesis of polyolefin composites with superior property sets at low filler loadings due to excellent dispersion. Photoactive supports (TiO2) provide tunable photodegradation without compromising mechanical properties, pointing to strategies for addressing polyolefin disposal. Future work may explore broader catalyst classes and supports, systematic electronic tuning via support basicity, incorporation of photosensitizing comonomers, and optimization toward specific property profiles and end-of-life degradability.

- Support dependence: Catalyst performance is sensitive to support identity and basicity; certain combinations (e.g., Ni on SiO2 without ionic tag) were inactive, indicating the need for appropriate ionic tagging and support selection. - Trade-offs under high ethylene pressure: Increased pressure boosted activity and molecular weight but reduced polar comonomer incorporation in some cases. - Cocatalyst requirement for Ti systems: Titanium catalysts required Et2AlCl, adding process complexity compared to cocatalyst-free Ni systems. - Scope not fully exhaustive: While many monomers and supports were demonstrated, generality across all catalyst classes and industrial supports remains to be validated; long-term catalyst lifetimes and recyclability were not comprehensively assessed. - Pre-treatment needs: Supports required high-temperature calcination, and some fillers were pretreated with Et2AlCl, which may not be compatible with all processing environments.